MECHANICS OF SOFT TISSUES AND BODY LOADING: EXPERIMENTAL RESEARCH FOR THE PURPOSE OF PRESSURE INJURY PREVENTION AND MODELING By Justin D. Scott A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Engineering Mechanics Doctor of Philosophy 2021 ABSTRACT MECHANICS OF SOFT TISSUES AND BODY LOADING: EXPERIMENTAL RESEARCH FOR THE PURPOSE OF PRESSURE INJURY PREVENTION AND MODELING By Justin D. Scott Over 3 million people in the United States rely on wheelchairs, spending about 12 hours in a wheelchair every day to carry out tasks of daily living [1], [2] . A common and expensive side effect of wheelchair use is the formation of pressure injuries, which are a result of sustained normal and shear pressures on the body, such as those experienced in the b uttocks and thighs while seated [3] [6] . The prevalence of pressure injuries in wheelchair users in care facilities is as high as 47%, and treatment costs up to $120,000 per incident [7], [8] . Because pressure injuries are destructive, both i n terms of quality of life and financially, there is a clear need for strategies to prevent them. Despite advances in technology used to prevent pressure injuries and models that assess risk, both the new technology and models have limitations. Even with the implementation of new wheelchairs, wheelchair cushions, and other pressure injury prevention tools, pressure injury prevalence only declined by one third over the past decade [9] . From the modeling perspective, the most advanced buttock and thigh models still use limited data from humans, excluding critical data fr om wheelchair users, the population most at risk for developing pressure injuries [10] . As such, the goals of this dissertation were to 1) determine the material properties of - bodied people, 2) to design an articulating chair that can reduce the seated normal pressure on the buttocks, and 3) to evaluate the ability of several seat pan covers to reduce shear pressure on the buttocks while seated. The first task was to create a protocol to determine the material properties of the buttocks and thighs in a positio n representative of the seated position and accessible to wheelchair users. Force and deflection data of the buttocks and thighs of able - bodied people were obtained in seated, quadruped, and prone positions. It was found that the buttocks and thighs were s imilar in the seated and quadruped positions. As the quadruped position was accessible, it was used to test wheelchair users; and data collected from the buttocks to the middle regions of the thighs were softer in wheelchair users than able - bodied people, while the distal thighs of wheelchair users were stiffer. An accessible articulating chair was created with independently rotating parts, including seat pan, back, pelvic support, and thoracic support. The chair was used to collect seated interface pressur e data during induced back recline, seat pan tilt, and changes to back articulation in able - bodied individuals and wheelchair users. Recline increased pressure on the buttocks, while seat pan tilt decreased pressure on the buttocks, and changes in back art iculation changed pressure patterns on the back. Three seat pan covers were evaluated for their coefficients of friction when paired with materials used to make pants and their abilities to reduce shear pressure on the buttocks of seated individuals. A sl ed was used to determine the coefficients of friction of pants materials on vinyl, one - layer nylon, and two - layer nylon seat pan covers. Coefficients of friction were compared to values found using human participants, and finally it was shown that the two - layer nylon seat pan cover reduced shear on seated individuals in the articulating chair while undergoing back recline. The material property data from the buttocks and thighs and the interface pressure data collected represent advances in both pressure in jury risk assessment and prevention. The material property data from wheelchair users were the first of their kind and can be used in finite element models. Data collected from the articulating chair and seat pan covers yielded interface and boundary condi tions for models of the seated buttocks and thighs as well as strategies to reduce the normal and shear pressures on the buttocks and thighs, reducing the risk of pressure injuries. iv This work is dedicated to Alexander Spaulding Hobbs, a man who is more inspirational than he can know, who taught us that nobody should ever be left behind or forgotten and to treat everyone and everything with respect and dignity. His benevolence is an ideal for which we all can strive. Without him, this dissertation would not have been possible. Thank you. v A CKNOWLEDGEMENTS This dissertation is the result of a joint effort by so many individuals that it is difficult to give credit to everyone who deserves it. So many people have guided me in my development , allowed me into their spaces to do research , and helped me conduct th e research as well. To everyone who supported me in my journey to graduate school, this would not have been possible without you. My parents and friends from Rutgers have all helped me become literate in the language of academia, which opened up the oppor tunity that brought me to Michigan. Since I have been at Michigan State, Dr. Bush has been a constant source of inspiration and has helped me refine myself into a better person and researcher. Her contributions to my work cannot be overstated , and she has cultivated an environment in the Biomechanical Design Research Lab that fosters student growth . Neither can the contributions of my committee, including Dr. Grimm, Dr. Roccabianca, and Dr. Slade , be ignored . All have taught me and contributed to my growth , whether formally in the classroom, by asking the right questions in meetings , or by offering their expertise when needed. My experience has also been shaped by the people who have facilitated my research and become my friends in the process. Piotr Pasi in allowing me to participate in the Adaptive Sports Club has been a blessing. Brian Sheridan and the staff at Level 11 are some of the most competent and caring people I have met . And, of course, the u ndergraduate s I have had the privilege of working with have made this dissertation possible. Last, but not least, Ania and I have been partners through a lot. A lot of late nights and a lot of life development over the past few years. But we have given ea ch other a lot of support and a lot of love. I could not have done this without you. vi T ABLE OF CONTENTS LIST OF TABLES ................................ ................................ ................................ ......................... ix LIST OF FIGURES ................................ ................................ ................................ ........................ x INTRODUCTION ................................ ................................ ................................ .......................... 1 CHAPTER 1: A REVIEW OF THE STATISTICS, RISKS, PATHOPHYSIOLOGY, AND TREATMENT OF PRESSURE INJURIES IN THE BUTTOCKS AND THIGHS ...................... 4 Pressure Injury Definition ................................ ................................ ................................ ........... 4 Prevalence and Statistics of Pressure Injuries ................................ ................................ ............ 5 Financial Cost of Treating Pressure Injuries ................................ ................................ ............. 6 Common Locat ions for Pressure Injuries ................................ ................................ ................... 7 Lower Back, Buttock and Thigh Anatomy ................................ ................................ ................... 8 Anatomical Changes in the Buttock and Thigh Soft Tissue ................................ ...................... 11 Spinal Cord Injury ................................ ................................ ................................ ................. 11 Aging ................................ ................................ ................................ ................................ ..... 11 Risk Factors for Pressure Injuries ................................ ................................ .......................... 12 Mechanisms of PI Development ................................ ................................ ................................ 15 Shear Forces ................................ ................................ ................................ ............................. 17 Medical Attention for Pressure Injuries ................................ ................................ .................... 18 Treatment ................................ ................................ ................................ ............................... 18 Prevention ................................ ................................ ................................ .............................. 18 Experimental Methods for Pressure Reduction ................................ ................................ ........ 21 Pressure Relieving Cushions ................................ ................................ ................................ . 21 Finite Element Evaluations of Internal Stress ................................ ................................ ........... 22 Advances in Geometry ................................ ................................ ................................ .......... 23 Descriptions of Soft Tissue: Linear Elastic Models ................................ .............................. 24 Descriptions of Soft Tissue: Hyperelastic Models ................................ ................................ 25 Descriptions of Soft Tissue: Poro/Viscoelastic Models ................................ ........................ 26 Expe rimental Determination of Material Parameters ................................ ............................ 27 Animal Studies ................................ ................................ ................................ ...................... 28 Human Studies ................................ ................................ ................................ ....................... 28 CHAPTER 2: THE EFFECTS OF BODY POSITION ON THE MATERIAL PROPERTIES OF SOFT TISSUE IN THE HUMAN THIGH ................................ ................................ ................... 30 Introduction ................................ ................................ ................................ ............................... 30 Methods ................................ ................................ ................................ ................................ ..... 33 Participan t Recruitment ................................ ................................ ................................ ......... 33 Test positions ................................ ................................ ................................ ......................... 33 Test Regions ................................ ................................ ................................ .......................... 34 Tissue Testing Protocol ................................ ................................ ................................ ......... 35 vii Model Fitting ................................ ................................ ................................ ......................... 36 Results ................................ ................................ ................................ ................................ ....... 38 Force and Deflec tion Data ................................ ................................ ................................ ..... 39 Ogden Material Parameters ................................ ................................ ................................ ... 42 Discussion ................................ ................................ ................................ ................................ . 46 Conclusions ................................ ................................ ................................ ............................... 50 CHAPTER 3: T ISSUE MATTERS: IN - VIVO TISSUE PROPERTIES OF PERSONS WITH SPINAL CORD INJURIES TO INFORM CLINICAL MODELS FOR PRESSURE INJURY PREVENTION ................................ ................................ ................................ ............................. 52 Introduction ................................ ................................ ................................ ............................... 52 Methods ................................ ................................ ................................ ................................ ..... 55 Participants ................................ ................................ ................................ ............................ 55 Quadruped Position ................................ ................................ ................................ ............... 55 Tissue Testing Proto col & Material Parameter Estimation ................................ ................... 57 Statistical Analysis ................................ ................................ ................................ ................ 57 Results ................................ ................................ ................................ ................................ ....... 58 Demographic Data ................................ ................................ ................................ ................. 5 8 Force and Deflection Data ................................ ................................ ................................ ..... 58 Ogden Material Parameters ................................ ................................ ................................ ... 59 Discussion ................................ ................................ ................................ ................................ . 62 Supplementary Material ................................ ................................ ................................ ............ 66 Tissue Testing Protocol ................................ ................................ ................................ ......... 66 Ma terial Parameter Estimation ................................ ................................ .............................. 66 CHAPTER 4: SHIFTING LOADS AND PERFUSION RESPONSES AS A RESULT OF CHAIR ARTICULATION IN THE CONTEXT OF PRESSURE INJURIES ................................ .......... 69 Introduction ................................ ................................ ................................ ............................... 69 Methods ................................ ................................ ................................ ................................ ..... 71 Pa rticipant Recruitment ................................ ................................ ................................ ......... 71 Articulating Chair Design ................................ ................................ ................................ ...... 71 Back and Seat Pan Pressure Measurements ................................ ................................ .......... 73 Pressure Data Analysis ................................ ................................ ................................ .......... 73 Perfusion Measurements ................................ ................................ ................................ ........ 75 Perfusion Data Analysis ................................ ................................ ................................ ........ 76 Statistics ................................ ................................ ................................ ................................ . 76 Results ................................ ................................ ................................ ................................ ....... 76 Participants ................................ ................................ ................................ ............................ 76 Recline ................................ ................................ ................................ ................................ ... 77 Seat Pan Tilt ................................ ................................ ................................ ........................... 78 Articulation of the thorax and pelvic supports slouched, neutral, erect ............................. 80 Effects of Chair Articulation on Perfusion ................................ ................................ ............ 81 Discussion ................................ ................................ ................................ ................................ . 82 Conclusions ................................ ................................ ................................ ............................... 84 viii CHAPTER 5: KEY COMPONENTS RELATED TO PRESSURE INJURY FORMATION: UNDERSTANDING PRESSURE DISTRIBUTION AND BLOOD PERFUSION RESPONSES IN WHEELCHAIR USERS ................................ ................................ ................................ ......... 85 Introduction ................................ ................................ ................................ ............................... 85 Methods ................................ ................................ ................................ ................................ ..... 87 Participant Recruitment ................................ ................................ ................................ ......... 87 Articulating Chair Positions ................................ ................................ ................................ .. 87 Interface Pressure Data Collection & Analysis ................................ ................................ ..... 89 Regional Analysis of Pressures ................................ ................................ ............................. 90 Perfusion Measurements & A nalysis ................................ ................................ .................... 91 Statistics ................................ ................................ ................................ ................................ . 92 Results ................................ ................................ ................................ ................................ ....... 92 Participants ................................ ................................ ................................ ............................ 92 Articulation Effects on Pressure ................................ ................................ ............................ 93 Postural Effects on Perf usion ................................ ................................ ................................ 97 Discussion ................................ ................................ ................................ ................................ . 98 Limitations ................................ ................................ ................................ ........................... 100 CHAPTER 6: THE EFFECTS OF PANTS AND CUSHION COVER MATERIAL ON SHEAR FORCES EXPERIENCED WHILE SEATED ................................ ................................ ........... 102 Introduction ................................ ................................ ................................ ............................. 102 Methods ................................ ................................ ................................ ................................ ... 104 Overview ................................ ................................ ................................ ............................. 104 Articulating Chair Design ................................ ................................ ................................ .... 105 Coefficients of Friction with Human Participants ................................ ............................... 109 Evaluati ng the Effects of Seat Pan Cover on the Shear Forces While Seated ..................... 112 Statistics ................................ ................................ ................................ ............................... 113 Results ................................ ................................ ................................ ................................ ..... 113 Coefficients of Friction with the Simulated Buttocks ................................ ......................... 113 Effects of Seat Pan Cover on Shear Forces ................................ ................................ ......... 115 Discussion ................................ ................................ ................................ ............................... 116 CONCLUSIONS ................................ ................................ ................................ ......................... 121 APPENDICES ................................ ................................ ................................ ............................ 127 A PPENDIX A: REPEATABILITY OF TISSUE INDENTATION PROTOCOL ....................... 128 APPENDIX B: CONFIRMATION OF SYSTEM ACCURACY WITH VINYL COVER ON THE SEAT P AN ................................ ................................ ................................ ............................... 129 APPENDIX C: COEFFICIENTS OF FRICTION CALCULATED USING THE SLED AND HUMAN PARTICIPANTS ................................ ................................ ................................ ....... 131 REFERENCES ................................ ................................ ................................ ........................... 132 ix L IST OF TABLES Table 1.1 Material models and their strain energy functions, with their parameters and dependence on the material deformation listed [19 1 ] ... ... . 2 7 Table 3.1 Material parameters for each region for males in the SCI and AB groups with standard errors 61 Table 3.2 Material parameters for each region for females in the SCI and AB groups with standard errors ........................ 61 Table 4.1 Combinations of chair back recline, seat pan tilt, and back articulation in which pressure measurements were taken. Each back recline was used with each seat pan tilt and back articulation resulting in 27 different chair positions 72 Table 5.1 Combinations of chair back re cline, seat pan tilt, and back articulation in which pressure measurements were taken. Each back recline was tested with each seat pan tilt and back articulation 89 Table 6 .1 Material pairings for the pants materials and seat pan covers .... 107 Table C . 1 Static coefficients of friction for each pair of materials during the sled trials . 13 1 Table C . 2 Kinetic coefficients of friction for each pair of materials during the sled trials .. 13 1 Table C.3 Static coefficients of friction for each pair of materials during the human trials 13 1 Table C . 4 Kinetic coefficients of friction for each pair of materials during the human trials . 13 1 x L IST OF FIGURES Figure 1 .1 Left) Lateral view of a seated individual, with PI prone regions highlighted by red circles, including the calcaneus, popliteal fossa, ischial tuberosity, coccyx, sacrum, and scapulae. Top right) Lateral view of a supine individual with PI prone regions highlighted, including the occiput, scapulae, sacrum, ischial tuberosity, and calcaneus. Bottom right) Lateral view of an individual lying on their side with PI prone regions highlighted, including the ear, greater trochant er, and malleolus 7 Figure 1.2 Lateral view of the right half of the pelvis with bony prominences labelled within their respective regions. The ilium is on the top, the pubis is in the bottom right, and the ischium is in the bottom left. PIs commonly form around bony prominences, such as the ischial tuberosity 8 Figure 1.3 Buttock muscle origins and insertions. The regions around the ischial tuberosity, coccyx, greater trochanter, and sacrum, circled, carry the most load wh ile seated and are especially prone to PIs . . .. 9 Figure 1.4 Thigh muscle origins and insertions. The ischial tuberosities and greater trochanter (circled) are prone to PI formation and are covered by the thigh muscles, as shown 10 Figure 1.5 Sample Braden Scale assessment sheet. Lowers in any of the six categories indicate higher risk for developing a pressure injury. Categories shaded in blue are addressed in this research 1 3 Figure 2.1 Indentation test setups for every position. The seated position (left) has a segment of the chair removed to expose the middle thigh region, the quadruped position (middle) allows access to the entire thigh without obstruction, as do es the prone (right) position. In the seated position, the stand was placed beneath the participants, while it was placed behind them in the quadruped position, and above them in the prone position 3 3 Figure 2.2 Regions of by placing the indenting tool flush against skin. As the tissue was compressed a positive deflection was obtained ... 3 4 Figure 2.3 Cross - section of the thigh with measurements used to determine the undeformed tissue is denoted by 3 7 Figure 2.4 Average force - deflection results separated by thigh region. The data for the seated position is represented by black lines, the quadruped position by bl ue lines, and the prone position by red lines in the proximal thigh region (left), middle thigh region (middle), and distal thigh region (right). Dashed lines are 95% confidence intervals on the mean value of deflection . 41 xi Figure 2.5 The original seated position (left) removed the posterior segment of the seat pan to allow access to the proximal thigh. The participants sat on a support bar located in the white ellipse, which caused a dropping of the buttocks and additional tissue tensions as this region was stretched between the rear bar and front thigh support. The amended seated position (right) included the participant sitting on the edge of testing chair with a straight back ... 42 Figure 2.6 µ values for each thigh in each position, separated by sex. The µ values for the thigh regions in the prone position are smaller than those of the seated and quadruped positions, as indicated by the asterisks. Seated parameters are represented with black outlines, quadruped with blue ou tlines, and prone with red 4 3 regions in the prone position are larger than those of the seated and quadruped positions, as indica ted by the asterisks. Daggers indicate differences between thigh regions within a position for black outlines, quadruped with blue outlines, and prone with red ..... 4 3 Figure 3.1 Positioning protocol for users in the AB (left) and SCI (middle and right) groups. Participants with SCIs were helped onto a table and into the quadruped position by their therapists while the harnesses were secured (middle); and then supported in the testing position with an exercise ball beneath the torso and sand bags in front of the knees (right) . 5 5 Figure 3.2 a) Indenting tool with the six - axis load cell embedded beneath the indenting surface and line ar potentiometer used to obtain distance data, and b) Indenting tool as used during testing with the participant ... 5 7 Figure 3.3 Force and deflection data from the proximal thigh (top), middle thigh (middle), and distal thigh (bottom) regions of the SCI group. The solid black line is the average AB data, and the dashed black lines represent the confidence intervals of the AB group. The data shaded in light green (left of the average) has less deflection than the average AB, an d the data shaded in darker green has more deflection 5 9 able - bodied group are white. Significant differences between groups are indicated by asterisks (*). Between - region (i.e., proximal, middle, distal) differences within the SCI group are indicated by plus signs (+). Between - region differences in the able - bodied group are indicated by daggers ... 60 Figure 4.1 a) Pressure mats were also affixed to the chair during portions of testing, b) Schematic of the rotations of the articulating chair, with the back, thorax, pelvis and seat pan identified. The thorax and pelvic supports rot ated about their centers, and the back and seat pan rotated about the point labelled CoR . 71 Figure 4.2 Segmentation of the seat pan (left) and back (right) pressure mat readings into regions. The seat pan was split up into the buttocks and thigh regions. The back was split up into the upper, middle, and lower back regions. , the location of the maximum pressure, was measured in mm from the posterior edge of the pressure mat in the buttocks and thighs and from the inferior edge of the pressure mat for the upper, middle, and lower back regions . . 7 4 xii Figure 4.3 The effects of back recline on the magnitudes of the maximum pressures within each region. Columns are grouped by regions. Statistical differences between recline angles are indicated by asterisks (*). Increases in back recline angle consistently resulted in significant decreases in maximum pressures in the upper back and thighs and increases in maximum pressure in the lower back and buttocks .. 7 7 Figure 4.4 The effects of back recline on the locations of the maximum pressures within each region. Statistical differences between recline angles are indicated by asterisks (*). Increases in back recline angle resulted in the maxi mum pressure shifting inferiorly for the middle back and posterior shifts in the buttocks 7 8 Figure 4.5 The effects of seat pan tilt on the magnitudes of the maximum pressures within each region. Statistical differences between seat pan tilts are indicated by asterisks (*). Increases in seat pan tilt corresponded to increased maximum pressure in the upper back, middle back, and thigh regions and decreased maximum pressure in the lower back and buttocks regions .. 7 9 Figure 4.6 The effects of seat pan tilt on the locations of the maximum pressures within each region. Statistical differences between seat pan tilts are indicated by asterisks (*). Increasing seat pan tilt corresponded to superior shift of maximum pressures in th e middle and lower back regions and the anterior shift of maximum pressures in the buttocks region .. 7 9 Figure 4.7 The effects of back articulation on the magnitudes of the maximum pressures within each back region. Statistical differences be tween back articulations are indicated by asterisks (*). When changing articulation from slouched to erect, the maximum pressures in the upper back region increased, while the maximum pressures in the middle and lower back regions decreased 80 Figure 4.8 The effects of back articulation on the locations of the maximum pressures within each back region. Statistical differences between back articulations are indicated by asterisks (*). When changing articulation from slouch ed to erect, the maximum pressures in the upper and middle back regions shifted inferiorly, while the maximum pressures in the lower back regions shifted superiorly ... 81 Figure 4.9 Average buttock perfusion values for e ach back recline (left) and seat pan tilt angle (right). Statistical differences between back recline or seat pan tilt angles are indicated by asterisks (*). Perfusion decreased in the full back recline positions relative to the no back recline positions a nd increased with seat pan tilt, regardless of the back recline angle . 82 Figure 5.1 a) Schematic of the rotations of the articulating chair, with the centers of rotation for the back and seat pan (CoR), pelvic support, and thoracic support identified. b) Pressure mat placement on the articulating chair during testing. c) Wheelchair user sitting in the articulating chair during testing ... 8 8 Figure 5.2 Break down of the seat pan (left) and back (right) pr essure mat readings. The seat pan was split up into the buttocks and thigh regions with the zero reference point being at the most posterior location of the mat. The back was split into the upper, middle, and lower back region with the zero reference being at the most inferior point of the mat (near sacrum) . 91 xiii Figure 5.3 The effects of back recline on the magnitudes of the maximum pressures within each region. Increasing the back recline angle increased maximum pressures in the buttocks and lower back and decreased maximum pressures in the upper back and thighs. Differences between recline angles are indicated by asterisks (*) .. 93 Figure 5.4 The effects of back recline on the locations of the maximum pressures within each region. Increasing the back recline angle caused a posterior shift in the maximum pressure regions of the buttocks and thighs. Differences between recline angles are indicated by asterisks (*) 9 4 Figure 5.5 The effects of seat pan tilt on the magnitudes of the maximum pressures within each region. Increasing the seat pan tilt angle increased the maximum pressures in the upper and middle back and decreased maximum pressures in the buttocks. Differences between seat pan tilts are indicated by asterisks (*) ... 9 5 Figure 5.6 The effects of seat pan tilt on the locations of the maximum pressures within each region. Increasing seat pan tilt angle caused a superior shift in the maximum pressure regions of the middle and lower back and an anterior shift in the maximum pressure regions in the thighs. Differences between seat pan tilts are indicated by asterisks (*) . 9 5 Figure 5.7 The effects of back articulation on the magnitudes of the maximum pressures within each region. Articulating the back from slouched to erect increased maximum pressures in the upper back and decreased maximum pressures in middle back. Differences between back articulations are indicated by asterisks (*) .. 9 6 Figure 5.8 The effects of back articulation on the locations of the maximum pressures within each region. Articulating the back from slouched to erect caused an inferior shift in the maximum pressure regions in the upper back and a superior shift of the maximum pressure regions in the lower back. Differences between back articulations are indicated by asterisks (*) . 9 7 Figure 5.9 Normalized perfusion values for the entire cohort (left) and responders (right) at every back recline and seat pan tilt angle. Seat pan t ilt increased perfusion in the responder group (p = .0013) 9 8 Figure 6.1 Articulating chair setup for the sled testing. The markers on the front right (FR), front left (FL), back left (BL), and back right (BR) were used to determine the orientation of the seat pan ( ). The markers on the sled, S1 - (AA) and posterior (AP) markers on the chair arm were used to determine the back recline. All positions were given i n the coordinate system in the top left of the figure 10 6 Figure 6.2 Free body diagrams for the sled before (left) and after (right) it starts to slide. The friction before the sled starts to slide is determined by the static coefficient of friction ( ), and the friction after it starts to slide is determined by the kinetic coefficient of friction ( ... 10 8 Figure 6.3 Reflective marker placements on the lower body of a participants: a nterior superior iliac spines (ASISs), greater trochanters, lateral epicondyles, and lateral malleoli ... 1 10 xiv Figure 6.4 (a) Front view of a participant sitting on a flat seat pan, (b) side view of a participant sitting on a flat seat pan, with the arrow indicating the space between the participant and the chair back, (c) participant sitting on a tilted seat pan just before starting to slide . 1 10 Figure 6.5 Coefficients of friction for each of the material pairings, found using the sled t o simulate the buttocks. Brighter bars represent static coefficients of friction, while darker bars are kinetic coefficients of friction 1 13 Figure 6.6 Coefficients of friction for each of the material pairings, found us ing human participants. Brighter bars represent static coefficients of friction, while darker bars are kinetic coefficients of friction ... 11 4 Figure 6.7 Shear forces on the seat pan of each material pairing at each angle o f recline. All shear forces are in the anterior direction of the seat pan. Recline increased shear force on the seat pan, regardless of the material pairing. Asterisks (*) indicate significant differences in shear forces on each pair 11 5 Figure A.1 Force and deflection data from repeated indentation protocols on the proximal (left), middle (middle), and distal (right) thigh regions in all three body positions for one participant. Data from each region in each position are consistent, showing good repeatability 12 8 Figure B.1 Experimental and theoretical normal shear forces on the seat pan at tilt angles up to 30° for the 45 Newton sled (left) and 110 Newton sled (right) 1 30 1 I NTRODUCTION Pressure injuries (PIs) are a dire issue for populations who use wheelchairs, including people with spinal cord injuries, surgery patients, and elderly individuals [7], [11], [12] . The prevalence of PIs in these populations can be as high as 47% in persons with spinal cord injuries and 61% in elderly individuals [7], [11] . Not only are PIs prevalent across several populations, but the y are expensive to treat and can require a lengthy recovery. The average cost of treatment is approximately $35,000, and PIs can take up to 22 weeks to heal [13 ], [14] . These factors contribute to the $27 billion annual expenditure for PI treatment in the United States and motivate research efforts for methods to reduce the incidence of PIs [15] . Among the risk factors associated most intimately with PI formation are sustained normal and shear pressu res on soft tissues and the internal stresses and strains they cause [16] . Such pressures, stresses, and strains are common in the buttocks, thighs, and lower back while seated; and over half of PIs in wheelchair users occur around bony prominences in the buttocks, thighs, and lower back, such as the greater trochanters, ischia l tuberosities, and sacrum [17] . Therefore, efforts have been made to quantify and reduce pressure, tissue stresses, and tissue strains in those regions. Finite element models have been created to esti mate the internal stresses and strains in the buttocks and thighs under seated loading conditions [18] [20] . Meanwhile, common practices to reduce normal and shear pressures on the body while seated include repositioning strategies such as reclining the back of a chair, and the use of specialized cushions [2], [21] . In these ways, clinicians and researchers have been working to understand and reduce to risk of PI formation. Even with the progress in modeling research and clinical practices related to PI prevention, there are still consistent limitations in each area. Fro m a modeling perspective, several inputs are needed to determine the stresses and strains in the buttocks and thighs under seated loading. The 2 seated load itself, the geometry of the buttocks and thighs, and accurate material properties of the buttocks and thighs are required to estimate internal stresses and strains. Though models have attempted to estimate internal stresses and strains of humans in the seated position, the models used to do so included material properties from animals [18] [20] . While some studies reported material properties of the human buttocks and thighs, those studies investigated able - bodied people lying down [2 2], [23] . Thus, they represent able - bodied people lying down and not seated wheelchair users, a crucial difference for models of the buttocks and thighs. From a clinical perspective, incidence of PIs is still a significant issue. Specialized cushions are able to reduce, but not eliminate, the incidence of PIs, leaving room to improve patient outcomes [21 ] . Further, the evidence supporting the use of recline as a pressure relieving strategy is inconclusive [4] . Recline decreases the normal load on the seat pan overall. However, there is a need to determine if the overall reduction translates to a pressure reduction on the bony regions of the buttocks. Furthermore, back recli ne shifts loading from the seat pan to the back, which may increase the risk of PI formation on the lower back (sacrum). Finally, back recline can increase shear force on the buttocks, which is another risk factor for PI development [4] . This makes strategies to reduce shear load on the buttocks critical to reducing PI risk in reclined positions. Overall, this research aimed to develop and use novel technology to address the above limitations in clinical practices and more accurately characterize the mechanical responses of was done by achieving five goals, addressed in chapters 2 through 6. The motivation of this work is further explained in Chapter 1 , along with an in - depth literature review. Chapter 2 determines and compares the material properties of the soft tissues in t he buttocks and thighs of able - bodied people in the seated, quadruped, and prone positions. 3 Chapter 3 determines the material properties of the soft tissues in the buttocks and thighs of individuals with spinal cord injuries while in the quadruped position and compares them to the material properties of tissues in able - bodied people. Chapter 4 evaluates the ability of an articulating chair design to redistribute pressure on the body of able - bodied people using back recline, seat pan tilt, and changes in bac k articulation. Changes in blood perfusion over the ischial tuberosity of the pelvis are also investigated as an additional factor associated with PI risk. Chapter 5 evaluates the effects of back recline, seat pan tilt, and back articulation on interface p ressures and perfusion over the ischial tuberosities of wheelchair users. Chapter 6 determines the coefficients of friction between three seat pan covers (vinyl, one - layer nylon, and two - layer nylon) and common clothing fabrics using a sled, compares those coefficients of friction to those found with human participants, and finally evaluates the ability of the three seat pan covers to reduce shear l oad on the buttocks and thighs while seated. A synopsis of the impact of this work is presented in the Conclusions . As with many scientific writings, sections of this research either have been published or have been submitted to various academic journals. Chapters 2 and 3 have been published, and the citation information for the journal articles are included at the beginning of each chapter. Chapters 4 and 5 have been submitted and are awaiting review. Because each chapter is written as a standalone journa l article, the motivations of this work, including the statistics, risk factors, and prevention strategies with respect to PIs are stated within each chapter. The research included in this dissertation was funded by NSF Grant CBET - 1603646. 4 CHAPTER 1: A REVIEW OF THE STATISTICS, RISKS, PATHOPHYSIOLOGY, AND TREATMENT OF PRESSURE INJURIES IN THE BUTTOCKS AND THIGHS Pressure Injury Definition P I s are defined as localized damage to the skin and underlying soft tissue, usually over a bony prominence . Many times PI s are related to a medical devices, such as wheelchairs [24], [25] . P I s can either start in the skin or deeper tissue below the skin [26] [28] . If the P I starts in the skin, a stage 1 injury is defined as the presence of a redness that blanches when pressed on by a finger . Stage 2 of an inj ury includes partial skin - loss, exposing the dermis. A stage 3 injury includes full skin - loss, exposing subcutaneous fat. A stage 4 injury has full skin and tissue loss, exposing the underlying bone. If there is significant debris in the injury , it can be deemed unstageable [24], [25] . P I s that start in deeper tissue versus the skin, termed deep tissue pressure injuries (DPTIs), are difficult to detect because of the lack of an extern al wound. A potential indicator for a DPTI is non - blanching redness. The stages for DPTIs are defined similarly to injurie s that start in the skin, however the extent of the tissue damage is unknown until the wound opens. Thus , these injurie s tend to have more layers of tissue damage before the injury is detected. Injuries may start in the decreased resilience in hypoxic conditions [27], [29] . The most common areas for P I formation are the buttocks, sacrum, gre ater trochanter, foot, and heel [16] . However, P I s may also form in other regions of the body, includi ng the popliteal fossa, scapulae, and back of the head [30] [33] . 5 Prevalence and Statistics of Pressure Injurie s P I s are a stubborn medical problem in the United States healthcare system. Incidence in general hospital populations has been reported between 3% and 36% [34], [35] . In high risk populations, such as surgery patients post - operation or wheelchair using patients, the prevalence of P I s can be as high as 47% [7] . In addition to pervading the United States healthcare syste m , P I s are an issue all over the world. A recent global review showed that the prevalence of P I s in adults and children in hospital settings were shown to be 11 % and 8 % , respectively [36] . In Europe, P I prevalence rates can vary between 6% and 54% [30], [31], [37] [43] . Several countries in Asia have reported prev alence rates between 8% and 29% [44] [48] . Prevalence studies in South Africa, Brazil, and Australia reported 30% , 14% , and 12% prevalence as well [49] [51] . P I s have been shown to be an issue in every care setting on almost every continent, demonstrating the global nature of the problem. The prevalence of PI s in the United States has decreased since 2006, yet PI s have remained a significant issue for individuals in hospitals as well as those who use assistive devices [9], [12] . PI s affect as many as 10% to 27% of individuals in general hospitals and over a third of surgery patients [52], [53] . Depending on the type of facility at whic h the patient is being treated, the prevalence of PI s can vary. The prevalence of PI s is 15% of those in community care facilities, 19% of those in rehabilitative care, 27% of people in acute care facilities, and 47% in long - term acute care facilities acqu ire PI s [13], [54], [55] . Individuals with a spinal cord injury (SCI) are at a particularly high risk of developing PI s. The incidence of SCIs in the United States is about 50 people per million each year, leading to about 17,500 new cases of SCI each year [56] [58] . This is in addition to the estimated 285,000 individuals currently living with a SCI in the United States [57] . The incidence of PI s in patients 6 with SCIs over four years of care can be as high as 30% , wit h a prevalence as high as 47% for those in long - term care [6], [7] . Additionally, 40% of patients with SCIs develop a PI during post - injury rehabilitation [17], [ 56], [59] . These factors contribute to a 80% lifetime incidence of PI s in individuals with SCIs [60], [61] . Individuals in community living facilities or nursing homes also experience high rates of PI s, even without SCIs. As many as 61% of individuals in nursing homes have PI s [11] . Even with the use of pressure relieving cushions, the incidence of PI s is as high as 40% per year for elderly individuals using wheelchairs . Moreover, those who experience PI s tend to have higher peak pressures beneath their buttocks than those who do not experience them [62] . Increased pressure under the buttocks is important, because almost 25% of all PI s occur in the soft tissue in that region [16], [63] . Financial Cost of Treating Pressure Injurie s Not only are PI they are also expensive to treat. An average PI incurs total healthcare costs exceeding $35,000 per PI incident [13] . If a PI reaches stage 4 (bone exposed) , the total healthcare cost can be greater than $120,000 for a single PI [8] . If a P I is acquired while being hospitalized for a separate issue, the treatment of the injury adds over $8,500 to the total cost of treatment for the individual [64] . These treatment costs add up to a total of $26.8 billion annually in the United States [15], [16] . The cost to society is clearly high, and the goal of preventing PI s entirely is the best solution. 7 Common Locations for Pressure Injurie s PI s commonly develop in the regions around bony prominences . While supine or seated , PI s are common around the calcanei, malleoli, greater trochanters, ischial tuberosities, coccyx, sacrum, and scapulae ( Figure 1.1) [17], [21], [30] [33], [37], [50], [51], [65], [66] . Three of those bony prominences, the ischial tuberosities, coccyx, and sacrum, are located in the buttocks and lower back, where abou t half of PI s occur [6], [65] . These regions in or near the buttocks are regions that experience large amounts of external pressure while supine or seated [20], [67], [68] . B lood perfusion in the those regions decreases a s a result of the external pressure applied while in the supine or seated positions [69] . Additionally, PI s may form in or near the popliteal fossae of the knees while seated due to external pressure [70] [74] . As will be discussed later, external pressure and decreased blood perfusion are two factors that facilitate the development of PI s; and they ultimately contribute to the risk of PI development in tissues near the bony prominence in and around the buttocks. Figure 1 .1 Left) Lateral view of a seated individual, with PI prone regions highlighted by red circles, including the calcaneus, popliteal fossa, ischial tuberosity, coccyx, sacrum, and scapulae. Top right) Lateral view of a supine individual with PI prone regions highlighted, including the occiput, scapulae, sacrum , ischial tuberosity, and calcaneus. Bottom right) Lateral view of an individual lying on their side with PI prone regions highlighted, including the ear, greater trochanter, and malleolus 8 Lower Back, Buttock and Thigh Anatomy The lower back, buttocks and thighs are regions of the body that may experience PI s while seated. As mentioned in the previous section, PI s can occur near the sacrum, coccyx, ischial tuberosities, and near the popliteal fossae of the knees. Although PI s may form in other regions while lying down, these regions are at risk while seated. Because of the prevalence of PI s, it is important to understand the anatomy in these regions. The lower back and buttocks region s consist of the pelvis and inferior aspects of the sacrum an d coccyx encompassed by several layers of muscle, fat, and skin. The three regions of the pelvis are the ilium, pubis, and ischium. There are several bony landmarks on the pelvis. In the ilium, there is the anterior superior iliac spine (ASIS), anterior in ferior iliac spine (AIIS), posterior superior iliac spine (PSIS), and posterior inferior iliac spine (PIIS). In the pubis, there is the pubic tubercle. The ischial tuberosity is on the ischium ( Figure 1 . 2 ). Figure 1 . 2 Lateral view of the right half of the pelvis with bony prominences labelled within their respective regions. The ilium is on the top, the pubis is in the bottom right, and the ischium is in the bottom left . P I s commonly form around bony prominences, such as the ischial tuberosity 9 Origins of the muscles in the buttocks are on the ilium and ischium of the pelvis, as well as the sacrum. B uttocks muscles originat ing on the ilium include the gluteus maximus, medius, and minimus. The gluteus minimus and medius insert on the greater trochanter , while the gluteus maximus inserts on the posterior femur. Muscles deep to the gluteal muscles that originat e on the ischium (ca lled lateral hip rotators) are the superior and inferior gemelli, quadratus femoris, and obturator internus, which insert on the greater trochanter . The piriformis originates on the sacrum, ultimately inserting on t he greater trochanter as well ( Figure 1. 3 ). The gluteus maximus is the only muscle that overlays the ischial tuberosity and carries the most load while in the seated posture . Though there is adipose tissue as well, i t is the only layer of muscle separating bone and skin over the ischial tuberosit y [75] . The sacrum also has only a thin layer of muscle overlaying the bone. Figure 1. 3 Buttock muscle origins and insertions. The region s around the ischial tuberosity, coccyx, greater trochanter, and sacrum, circled, carr y the most load while seated and are especially prone to PI s Origins of the muscles in the thighs are on the ilium, pubis, ischium, femurs, and ischial tuberosities ( Figure 1. 4 ). Several of the muscles have multiple origins and/or insertions. As mentioned before, the gluteus medius and maximus originate on the ilium and insert on the femur. 10 Adductor magnus and gracilis originate on both the pubis and isch ium and insert on the femur and tibia, respectively. Vastus lateralis originates on the femur and inserts on the tibia, while one head of the biceps femoris originates on the femur and inserts on the fibula. The second head of the biceps femoris, the semit endinosus, the semimembranosus, and part of the adductor magnus originate on the ischial tuberosity and insert on the fibula, tibia (both semitendinosus and semimembranosus), and femur, respectively. The ischial tuberosities are bony prominences of the bu ttocks that are covered by or attached to muscles in the buttocks and thighs, and the surrounding tissues are prone to PI formation [17], [33] . Though there is not muscle tissue covering the popliteal fossae of the knees, the tendons of the gracilis, vastus lateralis, biceps femoris, semitendinosus, and semimembranosus all cover the region, where PI s may also occur [70] [74] . Figure 1. 4 Thigh muscle origins and insertions. The ischial tuberosities and greater trochanter (circled) are prone to PI formation and are covered by the thigh muscles, as shown 11 Anatomical Changes in the Buttock and Thigh Soft Tissue Spinal Cord Injury Anatomical changes to the tissue in the lower limbs (buttocks and thighs) have been documented for those who have a SCI . C hange s in skeletal muscle cross sectional area are rapid in patients with SCIs , whether they have complete or incomplete injuries. While both types of injuries cause muscle atrophy, it is exacerbated in people with complete SCIs, with as much as a 45% decrease in the area of skeletal muscle within the first six weeks post injury [76], [77] . These changes in cross sectional area have been confirmed using both MRIs and CT scans [78] . SCI also leads to adipose tissue infiltrating lower limb muscles, which can affect their ability to contract regardless of the classification of the SCI [79], [80] . Further changes to lower limb muscle tissue include a gradual change of type 1 (oxidative) muscle fibers to type 2 (glycolytic) muscle fibers post - injury [81] . All these tissue changes, especially the deterioration of muscle in terms of qu antity and quality, increase internal tissue stresses when the tissue is loaded and increase the risk of PI formation [82], [83] . Aging As people age, the soft tissue in their buttocks undergoes changes in structure. X - ray scans have show n t hat the cross - sectional area of skeletal muscles peaks between the ages of 30 and 50 , after which muscle size generally decreases due to decreases in hormonal production and physical activity [84], [85] . The skeletal muscle cross sectional area of elderly individuals is abou t 40% smaller than the area of individuals in the peak age range [86] . In addition to the muscles physically getting smaller as individua ls age, their ability to exert forces diminishes as well , which is in part a result of the reduction in size and number of type 2 muscle fibers with aging [87] [89] . Similarly 12 to individuals with SCIs, these tissue changes put the tissue at risk of PI formation due to increased tissue stresses when loaded. Risk Factors for Pressure Injurie s Scales have been created to assess the risk of PI development. These include the Norton, Waterlow, and Braden Q scales, but by far the most widely used tool is the Braden Scale [90], [91] . The Braden Scale has six risk categories, where lower scores in each of the categories indicate s a higher risk of PI development. Figure 1. 5 shows the six categories: sensory perception, mobility, activity, moisture, nutrition, and friction/shear [33], [66] . Research into the prevention of PI s has largely focused on reducing the risk in those six areas. Thus, reducing PI risk involves increasing sensory perception, mobility, activity, and nutrition levels while decreasing moisture and friction/shear. This section focuses on the mechanical aspects of the Braden Scale, namely mobility, activity, and friction/shear , which are directly linked to the research conducted for this dissertation . Moisture wil l be discussed briefly as part of current research into the effects of microclimate on PI development. Individuals with decreased mobility are at an increased risk of developing PI s. E lderly people are at an increased risk, particularly wheelchair users and the bedridden individuals [60] . Those with SCIs have an elevated risk of exp eriencing PI s as well; and the risk of developing a PI scales with the severity of the injury. S and that less mobility leads to higher risk of PI formation . 13 Sensory Perception ability to respond meaningfully to pressure - related discomfort 1. Completely Limited Unresponsive to painful stimuli, due to diminished level of consciousness 2. Very Limited Responds only to painful stimuli. Conveys discomfort only by restlessness 3. Slightly Limited Responds to verbal commands but cannot always convey discomfort 4. No Impairment Responds to verbal commands. Has no sensory deficit Moisture degree to which skin is exposed to moisture 1. Constantly Moist Skin is moist constantly by sweat, urine, etc. Dampness every time the patient is moved 2. Very Moist Skin is often, but not always moist. Linen must be changed at least once a shift 3. Occasionally Moist Skin is occasionally moist, requiring an extra linen change once a day 4. Rarely Moist Skin is usually dry, linen only requires changing at routing intervals Activity degree of physical activity 1. Bedfast Confined to bed 2. Chairfast Ability to walk severely limited or non - existent. Cannot bear own weight and/or must be assisted into chair or wheelchair 3. Walks occasionally walks occasionally during day, but short distances, with or without assistance. Spends most time in bed or ch air 4. Walks frequently Walks outside room at least twice a day and inside room at least once every two hours during waking hours. Mobility ability to change and control body position 1. Completely Immobile Does not make any changes in body position without assistance 2. Very Limited Makes occasional slight changes in body position but unable to make frequent changes 3. Slightly Limited Makes frequent though slight changes in body or extremity position i ndependently 4. No Limitation Makes major and frequent changes in position without assistance Nutrition usual food intake pattern 1. Very Poor Never eats a complete meal. Rarely eats more than ½ of any food offered. Eats 2 servings or less of protein per day. Does not take a liquid supplement 2. Probably Inadequate Rarely eats a complete meal and generally eats ½ of food offered. 3 servings o f protein per day. Occasionally takes a supplement 3. Adequate Eats over ½ of most meals. Eats 4 servings of protein per day. Occasionally will refuse a meal, but will usually take a supplement when offered 4. Excellent Eats most of every meal. Usual ly eats 4 or more servings of protein per day. Occasionally eats between meals. Does not require supplementation. Friction & Shear 1. Problem Needs at least moderate help to mov e . Cannot l ift without sliding against sheets. S lides in bed or chair, needing repositioning with help. Spasticity, contractures or agitation leads to constant friction. 2. Potential Problem Moves feebly or requires help. While moving, skin slides to some extent a gainst sheets, chair, etc. Maintains relatively good position in chair or bed most of the time but occasionally slides. 3. No Apparent Problem Moves in bed and in chair independently and has sufficient muscle strength to lift up completely during move. M aintains good position in bed or chair. 4. N/A Figure 1. 5 Sample Braden Scale assessment sheet. Lowers in any of the six categories indicate higher risk for developing a PI . Categories shaded in blue are addressed in this research 14 Activity level is another factor that affects the risk of an individual acquiring a PI . Reduced activity, such as that seen in individuals with chronic pain, can increase the likelihood of obtaining a PI [92] . Using an assistive device may also directly contribute to PI risk [24] . O ver half of elderly individuals and people with SCIs report chronic pain . Further, a quarter of elderly people use an assistive device to ambulate and experience some trouble walking , while almost every individual with a SCI uses an assistive device [93] [95] . Therefore, both chronic pain and assistive device use are factors that contribute PI risk in both populations via decreased mobility. This decrease in mobility has led to deceased activity levels in both populations , which contribute s to the high PI preva lence [96], [97] . Friction and shear force (force parallel to the skin) are cited as risk factors for the development of PI s in 1) the Braden scale, 2) literature that predates the Braden scale, and 3) in subsequent literature [98], [99] . This is in part due to the reduction in blood perfusion in the tissue when shear force is applied [28], [100] [102] . Yet both normal forces (forces perpendicular to the skin) and shear forces are capable of reducing the blood perfusion in the loaded tissue, so both should be considered as risk factors for PI development [100] . Animal research has shown that soft tissue s are able to sustain normal loads greater than 45 kilopascals as long as they are intermittently relieved, whereas loads as small as 9 kilopascals have been shown to cause tissue damage if sustained for two hours without relief [103] [107] . Pressures of 32 kilopascals can start causing tissue damage if they are sustai ned for longer than 15 minutes [106] . This value of pressure is clinically significant because the peak normal pressures on the buttocks of seated individuals is about 28 kilopascals, indicating a risk of tissue damage within the first two hours if the pressure is not relieved [108] [112] . 15 Microclimate is a relatively new area of study that describes the environment surrounding a PI , with particular emphasis on the humidity and temperature of the region around the wound. The microclimate around the wound has several effects on the tissue. For example, increasing either the temperature or humidity of the microclimate increases the coe fficient of friction between skin and fabric [113] [116] . In addition to increasing the coefficient of friction, increasing the temperature and humidity decreas es the stiffness and ultimate strength of skin. Because of these effects of temperature and humidity, there have been several attempts to control the microclimate around areas that are prone to PI formation. One such study investigated the use of Dartex, s pandex, and incontinence (moisture - blocking) covers to control temperature and moisture of the ischial region of the buttocks but found no clear benefits [117] . Another study investigated the use of an insert between the buttocks and cushion, which had limited effects on both temperature and humidity [118] . As there has yet to be any cushion , insert , or cover that significantly decreases temperature and moisture , there is still a need for equipment that keeps the ischial region cool and dry. Mechanisms of PI Development M ultiple theories have been proposed to explain the mechanism of tissue death. There is evidence that the origins of PI s within tissue s are related to the normal and shear pressures applied to the tissue . Shear pressures have been stated to cause superficial injuries in the skin, and normal pressures are believed to cause deep tissue injuries in the muscle [27] . Within the affected tissue, there are several mechanisms by which the pressures can cause necrosis. Occlusion of blood perfusion to soft tissue has been observed with normal pressures as small as 12 kilopascals [119] . Including shear pressure in addition to normal pressure augments 16 the effect of normal pressure on blood perfusion occlusion [101], [120] . Without blood perfusion , the tissue does not receive oxygen and thus undergoes anaerobic respiration to produce energy. Anaerobic respiration produces lactic acid as waste, which builds up in the tissue since there is no blood perfusion to carry the lactic acid away from the tis sue, lowering the pH of the tissue. This lack of oxygen and lowered pH causes tissue necrosis over time, even with relatively small applied pressures [ 121] . In the same way that blood perfusion may be occluded by sustained pressures , the lymphatic system is also affected. Cellular waste that is not transported through the cardiovascular system may be transported through the lymphatic system. However, this system is also disrupted by applied pressures . Interstitial fluid is the fluid in the body that is neither within the cells nor the cardiovascular system. The lymphatic system normally maintains the interstitial fluid when the soft tissue is not loaded. However, a 10% increase of interstitial fluid per unit of muscle volume has been seen in subjects with external load sustained on their soft tissue. The pressures on the tissue s result in reduced interstitial fluid flow, leaving more cellular waste in the tissue, eventually causing tissue damage [26] . PI development can occur quickly with direct compre ssion of the tissue as well . Large amounts (greater than 32 kilopascals) of normal pressure on the tissue can initiate changes in the tissue within as little as 15 minutes. When tissue is compressed to 40% of its original size for one hour, necrosis can oc cur ; and when tissue is compressed to 55% of its original size , necrosis can happen in less than five hours [26] . These results align with those previously reported, giving a time frame for cell death between one and two hours [105], [106] . 17 Regardless of the mechanism of tissue necrosis, the underlying cause of it is the presence of external loads. These loads result in internal stresses , which in turn cause occlusion of the cardiovascular and lymphatic systems as well as tissue deformation [26] . Shear Forc es While the effects of normal forces on the soft tissue have been widely studied with respect to the issue of PI s, shear forces have been examined in a relatively small number of studies. The idea that shear forces may have a role in PI formation has been around for over half a century, and has been reaffirmed by its inclusion in the Braden Scale and recent studies that suggest reducing shear force may be the most effective way to reduce PI s [91], [98], [101], [122], [123] . There were numerous studies that investigated the effects of external shear force on the perfusion and vasculature of the skin [28], [101], [120], [124] . However, none of these studies focused on the perfusion of thigh and buttock tis sue while seated. Studies investigating the effects of seated shear load on the buttocks and thighs are lacking. One study investigated the effects of posture on the shear forces on the buttocks and thighs while seated [12 5] . Yet this study only investigated the gross shear force on the buttocks and thighs together while seated, so the shear on individual regions could not be estimated. Even in the most advanced experimental chair designs used to quantify the seat pan - so ft tissue interaction have limitations with regards to their ability to measure shear force [126] . There has yet to be an investigation into the effects of shear forces on the blood perfusion in the buttocks and thighs while seated; and this includes shear forces in different postures. Posture change has been shown to affect both normal and shear forces, but the effects on blood perfusion due to those changes is not available in the literature [125] . Therefore, it is necessary to describe the interaction between 18 posture change, shear force measurements, and blood perfusion measurements and then leverage that interaction to prevent PI s. Medical Attention for Pressure Injurie s Treatment Several methods have been used clinically to treat PI s. Stage one injurie s heal the fastest (16.9 days on average), and stage four injurie s can take up to 22 weeks on average to heal [14], [16] . Regardless of the type of treatment, the wound area is always offloaded [127] . Wounds are first cleaned and debrided of any dead tissue [128] . After cleaning, antibiotics and several topical treatments are applied to reduce the amount of time needed for healing to occur [129] . Among them are collagen , hydrofoam, hydropolymer, and hydrocolloid dressings, which have been shown to decrease time to healing for injurie s [130] [132] . Alginate and si lver nano particle dressings have been used, although less frequently. Negative pressures and electric stimulation have also been applied topically to wounds to create a more hospitable healing environment [133] . In addition to surface treatments, nutritional supple ments like collagen peptides have shown some benefit to PI recovery [134] . In serious cases, surgery may be used to excise the injured tissue surrounding the PI , and the wound may be covered by a flap of healthy tissue from the surrounding area [128], [135] . Prevention There are several measures that are currently taken to prevent the occurrence of PI s , as prevention is more desirable than treatment . The cost of PI prevention in inpatient facilities is actually less expensive than treatment [136] . Further, prevention of a PI is more desirable than 19 treatment from a patient care perspective. In a clinical setting, p revention includes a daily risk assessment, nutrition assessment, consistent repositioning, the use of specialized support surfaces, prophylactic dressing, and moisture/incontinence control [129], [133], [137] [142] . Some of these prevention strategies were shown to lead to lower PI incidence and reduce d healthcare costs [90], [136], [143] . If carefully implemented, prevention measures have, in some cases, been shown to eliminate the incidence of PI s in inpatient settings [54], [144] . These strategies become especially important after su rgery, when the risk of PI development can be mitigated by properly dressing the area on which the operation occurred as well as areas that commonly experience large loads [145] . Pressure relieving equipment (such as an adjustable bed) has also been introduced in inpat ient settings, however the equipment is not as effective as increased time with nurses [138], [146] . Prevention not only reduces the cost of healthcare, but it saves the patients time in the hospital because they do not have to recover from a PI . Outpatient modalities of PI prevention have focused on pressure relieving wheelchair cushions, wheelchair s adjustments , and pressure - relieving movements carried out by the user. Specialized c ushions have shown promise in reducing the incidence of PI in wheelchair users but have not eliminated the issue altogether [21], [133], [147] . Among the innovations in pressure reducing seat cushions are developments in contoured cushions that are molded to the wheelchair or materials , and dynamic cushions with air bladders that inflate sections of the cushion to be loaded while deflating s ections of the cushion to be offloaded [148] [157] . Common measurements used to determine the effectiveness of cushions are contact area, mean pressure, and peak pressure [5], [111], [158] . Increasing area of contact and decreasing me an and peak pressure are generally considered desirable to reduce the risk of PI s . Pressure relieving cushions have been able to reduce the peak 20 pressures in the buttocks by up to 40% , but they have been unable to reduce the pressure on the buttocks to less than 32 millimeters of mercury, which is the intra - arteriolar pressure [159], [160] . Thus, these cushions have not eliminated the risk of blood vessel occlusion nor the risk of PI s . In general, chair adjust ments such as whole body tilt and back recline are the most utilized ways of relieving pressure on the buttocks [2] . Whole body tilt describes a simultaneous rotation of the chair back and seat pan that results in the seated individual facing upward. Tilting has been shown to decrease pressures and improve blood perfusion in the buttocks if the whole body is tilted by a t least 35° [161] . However, this angle of tilt is often impractical for everyday tasks and makes interactions with other people more difficult for the chai r user. Furthermore, it increases the pressure on the lower back, another location where PI s often occur [16], [63], [125] . Recline has been shown to reduce the normal force on the buttocks while seated, but has also shown an increase in shear force on the same area, thus decreasing one risk factor for PI development while increasing anot her [108], [ 162] . Load relieving movements carried out by the seated individual have also been evaluated for use in outpatient settings and daily life, provided that the person has enough strength and mobility to perform them . Movements like forward and lateral trunk tilt have been introduced as a means by which chair occupants may offload regions of their buttocks . Forward bending of the trunk has been shown to reduce peak pressures on the buttocks by approximately 10% [125] . Meanwhile, lateral trunk tilt offloads the contralateral side of the body, but tissues in the ipsilateral side of the body experience pressures that can put them at risk for PI development [20], [108], [163] . Other movements that completely offload the buttocks, such as the wheelchair user lifting themsel ves off of their cushion (a wheelchair pushup) , have been proposed. But the lift has to be 21 held for almost two minutes for blood perfusion to return to pre - loaded levels, and can thus be difficult to perform [164] . Experimental Methods for Pressure Reduction Existing PI prevention strategies have reduced the prevalence of PI s, however they have not been able to eliminate the risk of PI s altogether [9], [12] . As PI s are still a significant issue for many people , there are ongoing efforts to further diminish the risk of PI development. Many of these efforts are focused on modifications of the chairs on which people sit or changing the interfaces between the seated people and the c hairs. In many cases, these modifications are tantamount to updates to currently existing technologies and PI prevention strategies. Pressure Relieving Cushions Cushions affect the peak pressure on the buttocks and thighs while seated. P eak pressure on t he soft tissue in the buttocks and thighs while seated has been a n indicator of the risk for PI development , and t wo cushion - dependent mechanisms have been found to increase the peak pressure on those regions while seated [62] . If a cushion is too stiff, the cushion will not distribute the pressure on the buttocks and thighs, creating a pressure concentration near the bony prominences. If a cushion is too soft, bony areas sink into the cushion unt pressure over the compressed areas, once again creating a pressure concentration near bony prominences on the buttocks and thighs. Th erefore, there is an optimal stiffness for a cushion used in seating, where the cushion distributes pressure without bottoming out [165] . A limitation of the optimal stiffness found in this study is that all the participants in the study were able - bodied. This 22 limitation is significant because studies have shown individuals with SCIs have larger pressure concentrations on bony prominences and different anatomies while seated relative to able - bodied individuals [77], [ 79], [108], [160] . Because of this, it is imperative to optimize cushion material properties to reduce the risk of PI s in high - risk populations, such as those with SCIs. Shear pressure has also been noted as a risk factor for PI development, and there i s a need to reduce shear pressure on the buttocks and thighs while seat ed [91], [143], [166] . One avenue by which shear pressure can be reduced is by the introduction of a shear reducing cushion cover. Though reducing she ar with a cushion cover may be a viable strategy, research on new shear reducing cushion covers is limited . There have been studies on the effects of seated posture on shear [4], [167] . There have been studies on the effects of skin microcl imate on shear [113] [115] . Yet there have been very few studies on the effects of shear reducing cushion covers. To the extent that shear reducing cushion covers have been studied, one recen t cover mechanism was able to reduce shear pressure while adjusting to a reclined posture relative to a control cover [168] . However, the mechanism actually increased shear pressure while in the reclined posture relati ve to the control cover. Since seated individuals spend more time in a static posture than switching between postures, this mechanism is limited. Due to the lack of research in shear reducing cushion covers, there is a clear need to improve upon existing c ushion covers to reduce the shear, and thus the PI risk, in seated individuals. Finite Element Evaluations of Internal Stress Finite element models evaluate the effects of normal and shear forces by considering the entire model as a group of smaller, connected pieces that transmit load to one another. Each piece in the model deforms based on inputs, such as the geometry of the model, the material of the model, 23 and the normal and shear forces applied . Finite element models have b een used to evaluate PI prevention strategies. Suc h models investigated the effects of changing the material properties or shapes of the specialized cushions used to prevent PI s [169] [173] . T here have been vast improvements in some model inputs, like anatomically correct 3 - D model geometry . However, accurate in - vivo material properties of tissues are still needed . Advances in Geometry There have been improvements in the techniques used to create the geometry of finite element models , which have advanced from two - dimensional to three - dimensional models [18], [174], [175] . Two - dimensional models were used to show how slices of the thighs and buttocks respond ed while under seated load and evaluate d how changes in posture affect ed the load [20], [174] . However, it has been shown that material properties and internal str ess of the tissue differ when considering two dimensional slices and a full three - dimensional model [10], [19] . Patient - specific t hree - dimensional finite element models either used a homogeneous material to represent the soft tissue or utilized a more detailed geometry captured by MRI [18], [175] [182] . Regardless of the model geometry , there was a noted lack of realistic material properties for in - vivo human soft tissue s in the seated position . A relatively new consideration in finite element modeling has been to investigat e how external pressures affect ed internal anatomical structures, such as vasculature. Previously, the geometry of the vasculature had been ignored, but recently o ne area under investigation has been the skin and its blood vessels. Numerous finite element models investigated the change in cross sectional area and blood flow of blood vessels after normal stress was applied by using fractal patterns to model the geometry of the vasculature [183] [185] . Some models evaluated the effect 24 of shear force and friction on vess el size and blood flow [183] . However, authors of these studies noted that the prope rties of skin var ied greatly depending on the storage and testing conditions, highlighting again the need for accurate material propertie s . Despite the lack of appropriate material properties for skin, models provided insight into the amount of external pr essure that may cause blood flow occlusion. Descriptions of Soft Tissue: Linear Elastic Models The behaviors of materials, such as soft tissues, are described by material models when using finite element programs. The choice of material model s used to represent material s is of the utmost importance when the goal is to have accurate evaluations of internal stresses and deformations via finite element modeling. The simplest material model used is a linear elastic model, where the intern al stress is Th is type of material behavior was popular due to its simplicity. Because of its simplicity , investigators used linear elastic behavior to represent the tissue in the thighs and buttocks in models dating back to the 1970s [186] . Models used linear elastic representations of tissue to determine stresses in the buttocks while lying in the supine position [187] . Further , investigations using linear elastic material properties for tissues obtained results like those showing an increas e in the thickness of a cushion to more than eight centimeters d id not improve the pressure distribution abilities of the cushion [18 8] . More recent uses of linear elastic materials to represent the soft tissue took - ( Eq . 1.1) 25 time [189] . However, it was largely determined that soft tissue wa s not accurately represented by linear elastic material s , meaning that most recent finite element models d id not consider linear elastic materials to represent soft tissue [190] . Descriptions of Soft Tissue: Hyperelastic Models M ost researchers started using hyperelastic material models to represent soft tissues . The equations that represent common hyperelastic models can be found in Table 1. 1. To determine the relationship between the stress and the deformation for any hyperelastic material, the derivative of the potential function must be taken with respect to the stretch of the material. Eqs . 1. 2 and 1. 3 show how the str etch and the stress are determined, respectively: When considering soft tissues with these material models, the Yeoh, Ogden, and Martins models were shown to best fit the uniaxia l compression data from soft tissues , supporting their use for soft tissue representation [10], [191] . Despite their relatively poor performances with respect to the other material models, the Neo - Hookean and Mooney - Rivlin materials have been employed in several situations. Both Neo - ( Eq. 1. 2 ) (Eq. 1. 3) 26 Hookean and Mooney - Rivlin repre sentations of soft tissue were used to show that the highest stresses in the buttocks while seated were found in the muscle tissue near the ischial tuberosity [192] [194] . Neo - Hookean materials were further used to show that stresses in the buttock tissue were in general lower for individuals within a normal body mass index range versus those in the low or high range [195] . Modifications to Neo - Hookean materials included damage components to show how the stiffening of the soft tissue as it encounters injury can create further stress concentrations in the surrounding tissue [196] . The Odgen material model perform ed well in describing the behavior of soft tissue, and has also been used in various investigations into the response of soft tissue while seated [10], [197] . Ogden materials have been used for over twenty years, spanning models that have been two dimensional up until recent models that u se anatomically accurate three - dimensional geometries gathered from magnetic resonance images [18], [174] . A sensitivity analysis showed that the internal stresses evaluated by finite element models strongly rel ied on the material parameters used to inform the material [198] . This last study indicate d the importance of carefully selected material parameters based on accurate experimental data to achieve realistic internal stress values. Descriptions of Soft Tissue: Poro/Viscoelastic Models Poroelastic and viscoelastic material model s considered the time - dependent properties of material s . Like other types of materials, poroelastic and viscoelastic materials showed that the largest stresses in the buttocks and thighs while seated were in the muscle layers nea r the ischial tuberosit ies [199], [200] . Poroelastic and viscoelastic models also showed the benefits of using contoured versus flat cushions [201] , that individ uals with SCI experience larger stresses and 27 stretches in the buttocks while seated [82] , and were used to explore the response of muscle tissue to vibration [202] . Table 1. 1 Material models and their strain energy functions, with their parameters and dependence on the material deformation listed [191] Material Name Potential Function Material Parameters Dependence on Deformation Mooney - Rivlin Invariants of Right Cauchy Green Strain Tensor Yeoh c 1 , c 2 , and c 3 First invariant of Right Cauchy Green Strain Tensor Neo - Hookean c 1 First invariant of Right Cauchy Green Strain Tensor Ogden Principle stretches Humphrey Q Right Cauchy Green Stretch Tensor Martins a, b, c, and A Principle stretch and first invariant of Right Cauchy Green Stretch Tensor Veronda - Westmann c 1 , c 2, Invariants of Right Cauchy Green Stretch Tensor Experimental Determination of Material Parameters Regardless of the material model used to describe the soft tissue in the human body, experimental data we re needed to verify the material properties of the tissue used in the models. Several of the finite element studies in the previous section s noted the dependence of internal stress evaluations on the appropriate selection of material model and material properties [191], [195], [198] . Of particular importance to the research performed for this dissertation is the need for material properties that accurately represent in - vivo human tissue while seated. These properties 28 are critical to represent the internal stresses in wheelchair users , however current material propert y data were collected from animal s or human s in limited , specific situations. Animal Studies Material properties obtained from the mechanical testing of animal tissue were commonly used to inform finite element models th at evaluate stresses in human tissue. Particularly common were studies that determined the material properties of soft tissue in rats and pigs. Investigations into the material properties of the rat soft tissue were able to determine the properties of rat muscle tissue independently [177], [203] . Additionally, material properties from rat soft tissue were used in several finite element models [175], [198] . Material property data from pigs were also collected and used in some of the most recent finite element models [18], [195], [204] . These animal data have been used to inform finite element models. But it has been noted that changes in t he material parameter values, especially relative changes between fat and muscle, significantly impact ed stresses and deformations in the respective tissues [204] . Therefore , there is a need to determine the material properties of human soft tissue to improve the accuracy of evaluations made with finite element models. Human Studies Data on the force and deflection of human soft tissue has been collected using various methods . H owever, there have been limitations in terms of their usefulness in representing the buttocks and thighs while seated. A few investigations of the force and deflection responses of the buttock s occurred while participants were laying down. There were indentation tests of the buttocks as well as force and deflection measurements taken through digital image correlation 29 [23], [205] [209] . These data are limited in their usefulness in the seated position due to the fact that both the muscle thickness and tension in the gluteal muscles are different in the seated position with respect to lying down [210], [211] . An additional limita tion to the data collected through digital image correlation was that the data were collected included only a few points, meaning parameters. Given that tissue is a nonlinear material, numerous data points are required to accurately fit parameters. The most recent investigations into the deformation of soft tissue in the buttocks while seated used magnetic resonance images to collect soft tissue deflection while in the seated position while loaded versus unloaded. While the deflections seen were accurate, there were only two load states considered, and only six males were included in the study [212] . So, a major flaw in this investigation was its attempt to characterize a nonlinear material using two data points and the single gender represented by the data . To more accurately represent the soft tissue of the buttocks while seat ed, more participants are required, particularly including both males and females, and more data points are needed while people are in the seated position. More data points per individual will allow for more accurate recreations of force - deflection curves in soft tissue while seated. 30 CHAPTER 2: THE EFFECTS OF BODY POSITION ON THE MATERIAL PROPERTIES OF SOFT TISSUE IN THE HUMAN THIGH The contents of this chapter were originally published in the Journal of Mechanical B ehavior of Biomedical Materials . J. Mech. Behav. Biomed. Mater. , vol. 110, 2020, doi: 10.1016/j.jmbbm.2020.103964. Introduction The number of p eople who have experienced P I s is staggering, creating a substantial financial burden for both individuals and society at large. Prevalence rates of PI s have been shown to be extremely high, depending on the setting. Approximately 40% of patients with spinal cord injuries (SCI) experienced PI s during rehabilitation [17] . 47% of wheelchair users in long - term care facilities had PI s [7] . Furthermore, over 60% of individuals who were admitted into geriatric care facilities had a t least one pre - existing PI [11] . Because an average PI incurs over $35,000 in healthcare costs, PI s have added as much as $1.3 billion to medical expenses each year in the United States [13], [64], [111] . Half of these injurie s were on the thigh or buttock regio ns, meaning that understanding the conditions leading to PI s in those areas is critical to reducing the risk and the cost [6], [16], [62], [147] . While the cellular mechanism of PI development is still undetermined, it has been generally agreed upon that internal stresses in the tissue were factors in PI development [26], [106], [213], [214] . Such stresses were reported in the thighs and buttocks while seated [106], [213] . Over time, 31 these stresses may cause stress - induced necrosi s, vascular damage, reduced interstitial flow, and hypoxia due to a reduced volume of blood in the tissue. All of these potential mechanisms of PI formation are caused by internal stress [26], [27], [102], [215] . Finite element models have used the material properties of the buttocks and thighs to predict internal stresses that occur under seated loading conditions. Yet, determining appropriate material properties for use in these has been difficult. Some finite el ement models used tissue properties obtained from experiments on animals [191], [192], [198], [203], [216], [217] . More recently, in vivo indentation tests have been conducted to find the strains caused by forces applied to the human thighs and buttocks [207], [208] . However, there are significant limitations to these studies. Some used only two load states, unloaded and fully loaded, to obtain the material properties of the tissue. Two load states are not adequate because soft biological ti ssues are nonlinear materials, meaning that more than two load points are necessary for accurate material descriptions [191], [218] . However, there have been several models that have utilized material descriptions based on t wo loading states [18], [192], [216] . Other experimental studies have collected data with indiv iduals in the prone or supine position [23], [205], [208], [219] . These positions have different knee and hip articulations as c ompared to the seated position; and joint articulations have been shown to affect the tensions in the tendons and the muscles, so it is likely that in vivo material properties will differ with body position [211], [220] . Furthermore, tissue thickness in the buttock region depends on the hip articulation, meaning that the same external loading on the soft tissue will not result in the same inter nal loading pattern when seated versus laying down [210] . Because wheelchair users are at an elevated risk for PI development in their thighs and buttocks while seated, accurate in vivo material descriptions for these tissues are needed. Testing 32 in the seated position is ideal, yet significant challenges exist with respect to obtaining the material properties of wheelchair users in a seated position. The authors conducted force - deflectio n testing on the buttocks and thighs in an elevated lab chair [10], [221] . That protocol required removing sections of the seat pan to expose the regions of interest, which r educed the stability of the person in the chair and called for balance and core strength that not every wheelchair user has. In addition, repeatedly transferring into and out of the elevated lab chair, as required by the mechanical testing protocol, was pr oblematic for this population. The combination of those challenges makes testing in the seated position impossible for some wheelchair users, so developing a protocol for an alternative position is imperative. The supported quadruped (crawling) position is common in physical therapy for wheelchair users and has the same knee and hip articulations as the seated position, making it a candidate for the data collection from wheelchair users. This position has the thigh and buttocks regions exposed for taking me asurements, has a similar body configuration relative to the seated position, and many wheelchair users have experienced it while in physical therapy. Therefore, the supported quadruped position was included with the prone and seated positions for our expe rimental tests. A need exists for in vivo material property data representing the thigh and buttocks region - thighs and buttocks with able - bodied individua ls in the prone or supine positions, and these data have been used as an analog for the seated thighs and buttocks despite potential discrepancies [207], [208] . As such, the goals of this research were to determine and compare the material properties of the soft tissue in the thigh in the seated, quadruped, and prone positions. 33 Methods Participant Recruitment 20 able - bodied individuals ( 10 males and 10 females) volunteered to participate in this study, which was approved by the Biomedical and Health Institutional Review Board at Michigan State University. Exclusion criteria for this study included active wounds on the buttocks or thighs, osteoporosis, o ccult bone disease, musculoskeletal diseases, and pregnancy. Test positions All participants wore snug - fitting, but stretchy pants or shorts to permit testing. The participants were tested in each of three positions: seated, quadruped (crawling), and pro ne ( Figure 2.1). In the seated position, the participants sat on a chair with removable slats, each of which exposed a body region where the tissue testing was performed. In the quadruped position, participants faced the floor with their knees and hips fle xed at ninety degrees and a support under their chest. In this position, their thighs were also blocked by a padded vertical barrier. Figure 2. 1 Indentation test setups for every position. The seated position (left) has a segment of the chair removed to expose the middle thigh region, the quadruped position (middle) allows access to the entire thigh without obstruction, as does the prone (right) position. In the seated position, the stand was placed beneath the participants, while it was placed behind them in the quadruped position, and above them in the prone position 34 Test Regions The three test regions were all determined by their positions relative to the greater trochanter and lateral epicondyle of the femur ( Figure 2.2 a ). The femoral length was defined as the distance between the greater trochanter and lateral epicondyle. The proximal thigh region was 10% of the femoral length distal to the greater trochanter. This region was just distal to the ischial tuberosity of the pel vis, which has been shown to be one of the most common regions for PI development [70] . Although this region is near the ischial tuberosity and could be termed the buttocks region, this region will be referred to as the proximal thigh region throughout the paper. The middle thigh region was 50% of the femoral length distal to the greater tr ochanter. And the distal region was 90% of the femoral length distal to the greater trochanter, another region where PI s form [71] . Thus, three regions were tested in all participants and are referred to as the proximal, middle and distal thigh regions. Figure 2.2 Regi determined by placing the indenting tool flush against skin. As the tissue was compr essed a positive deflection was obtained 35 Tissue Testing Protocol Force and deflection measurements of the soft tissue were collected via a custom - built indenter. The indenter was designed to collect force and deflection data continuously. The indenting surface was embedded with a load cell that obtained the force measur es (AMTI, Watertown, MA). The tool was connected to a linear potentiometer, which collected linear displacement data (Honeywell, Morris Plains, NJ). These instruments were synchronized and used to generate a continuous force - deflection curve to characteriz e the non - linearity of the tissue response to external force . The surface area of the indenter was a five centimeter by five centimeter square. One end of the potentiometer was attached to the load cell, while the other was attached to a reference structur e that remained stationary throughout testing ( Figure 2. 1). To start each test, a reference distance measurement was taken. The indenting tool was positioned near the test region, and the load cell was zeroed. The reference measurement was the largest dis tance recorded by the potentiometer with no force recorded by the load cell ( Figure 2.2b). The indenter was then pressed into the thigh region using a constant loading rate until a physiological resistance was felt. Each person was first tested in the seat ed position in order to determine what this maximum applied force should be. The resistance was felt in the seated position when the tissue was compressed against the femur and would no longer deform. In other words, additional applied force would have lif ted the entire leg. This force was between 60 and 100 N, depending on the individual. Then this was used as the maximum applied in the quadruped and prone positions [221] . The load was then removed. This procedure was followed for each region, making sure that the indenter was perpendicular to the body region during each test. A five minute rest period for tissue recovery was provided before a second trial was conducted. Two trials 36 were completed for each region of the thigh in each position. An assessment of the repeatabili ty of this protocol is included in Appendix A . Model Fitting The force and deflection data were converted into stress and stretch data respectively. Stress values were calculated by dividing the force perpendicular to the indenter face by the area of the indenting face. Stretch values perpendicular to the axis of th e thigh were calculated using the undeformed thickness of the soft tissue in each thigh region (described below) and the deflection values collected during the indentations (Eq 2.1). Here, o is the undeformed tissue thickness and d the deflection of the tissue. To determine the undeformed tissue thickness for each individual thigh region, the ratio of the total thickness of the thigh (i.e., anterior to posterior, AP, distance) to the thigh ci rcumference was obtained for each region ( Figure 2 . 3 ). This ratio was approximately equal to 0.33, as determined through the use of existing data sets that included a 3D scan and anatomical measures of the thigh regions [10], [221] . thigh circumference to determine the total thicknesses for each test location. The portion of the soft tissue thickness in the thigh posterior to the femur has been reported to be 0.55 of the total thickness of the thigh [222] . Thus, the undeformed thickness of the tissue was given by Eq. 2 .2 where C is the thigh circumference, and the undeformed thickness is ( Eq. 2.1) ( Eq. 2. 2 ) 37 Figure 2.3 Cross - section of the thigh with measurements used to determine the undeformed The stresses and stretches of every thigh region of each participant in these positions were then modeled using an incompressible first order Ogden material [18], [192], [203], [223] . The strain energy function of this model is represented by Eq. 2. 3 . The stretches in the three principal directions were given by 1 2 , and 3 respectively, with 1 being the stretch in the direction perpendicular to the femur (i.e., loading direction); µ was a material parameter with the dimension of stress and was a dimensionless material parameter [18], [192], [203], [223] . The stress in any direction was defined as the derivative of the strain energy function with respect to the stretch in that direction multiplied by the stretch in that direction itself. Stress in the perpendicular direction was defined by Eq. 2. 4. With the condition of incompressibility, the volume of the thigh region must be preserved, 1 2 3 = 1 , at all times. Incompressibility can be applied due to the large fluid cont ent of tissue ( Eq. 2 .3 ) ( Eq. 2.4 ) 38 such as fat and muscle, and near incompressible behavior has previously been observed in human soft tissue [224] . The only forces acting in each transverse direction were the internal tissue forces, meaning that the stretches were the same in each transverse direction ( ). Considering these two conditions together, 2 3 = . This makes 1 the only inde pendent variable, which we replaced with the symbol . In the uniaxial loading of the indentation tests, the relationship between stress and stretch is described by Eq. 2. 5. In this case, is the stress perpendicular to the axis of the femur, and is the stretch using a Levenberg - Marquardt least squares algorithm to best describe the stress an d stretch data collected for each body position and thigh region for each participant. A two - way, repeated measures ANOVA [225] was used to determine the sign ificance of the µ and parameters within each sex. A post - hoc Tukey test was used to make all pairwise comparisons for the µ and parameters between regions and positions. Differences between sexes for were detected with t tests [226] . A Wilcoxon Signed - Rank Test was used to determine within - individual differences in the µ and parameters in the three body positions for each region. Results Ten males (average age 22.4 ± 2.2 years, average height 179.7 ± 10.8 cm, average weight 77.4 ± 6.4 kg) and ten females (average age 21 ± 1.3 years, average height 164.9 ± 6.0 cm, average weight 66.5 ± 12.6 kg) participated in this study. ( Eq. 2.5 ) 39 Force and Deflec tion Data Average force - deflection curves of all participants for each position and region were obtained and are presented in Figure 2. 4. The average force - deflection curves included both hile seated is 50 N (noted by a horizontal line on Figure 2. 4) [109], [159], [227] . Upon initial analysis of these data sets, the proximal thigh region while seated was much stiffer than any other region in any position. Since our data indicated a much higher stiffness for the proximal thigh in the seated position as compared to the other positions and regions, the authors investigated further and found that the support structure for the laboratory seat system had caused an additional tension near this thigh region ( Figure 2. 5, left). The participants sat on a support bar located at the rear of the buttocks which caused a dropping of the buttocks and additional tissue tensions as this region was stretched between the rear bar and front thigh support. The authors were able to succ essfully recruit half of the original cohort (five males and five females) to return for a follow - up test in a modified seated position that removed this load. This new positioning protocol had the participants sitting on the edge of the testing chair ( Fig ure 2. 5, right) so the buttocks load was on the back edge of the buttocks, not the thigh. All three leg regions were tested again. In this positioning protocol, the middle and distal thigh region data yielded the same results as those of the first protocol , however the force and deflection data from the proximal thigh region had a reduction in stiffness. With the modified protocol, the data from the proximal thigh in the seated position still indicated that it was stiffer than the middle and distal regions while seated, but it was no longer the stiffest region across all positions. Therefore, data reported for the proximal thigh in the seated position are from the modified protocol (n=10) while the other two regions include all participant data (n=20). The o ther two positions (quadruped and prone) included data from all 20 participants. 40 In the prone position (red curves in Figure 2. 4) , the proximal thigh showed the most deflection, and the distal thigh showed the least deflection. At 50 N of force, there was an average deflection of 31.9 ± 6.5 mm in the proximal thigh, 28.1 ± 4.0 mm of deflection in the middle thigh, and 21.3 ± 4.5 mm of deflection in the distal thigh. The prone position consistently produced less deflection than the seated and quadr uped positions, regardless of the amount of load ( Figure 2. 4). In the seated position (black curves in Figure 2. 4) , all three thigh regions displayed similar force - deflection behaviors. At 50 N of force, there was an average of 39.6 ± 8.4 mm of deflection in the proximal thigh, 39.4 ± 6.7 mm of deflection in the middle thigh, and 40.7 ± 9.0 mm of deflection in the distal thigh. The seated position consistently produced more deflection than the prone position and less deflection than the quadruped position for the same force ( Figure 2. 4). In the quadruped position (blue curves in Figure 2. 4), all three regions displayed similar force - deflection behavior. At 50 N of force, there was an average of 44.4 ± 6.9 mm of deflection in the proximal thigh, 44.5 ± 6.8 m m of deflection in middle thigh, and 45.8 ± 4.9 mm of deflection at 50 N for the distal thigh in the quadruped position ( Figure 2. 4). The quadruped position consistently produced the most deflection of all the positions with the same force. The deflection values in the seated position were consistently closer to those in the quadruped position than those of the prone position for all regions. 41 Figure 2. 4 Average force - deflection results separated by thigh region. The data for the seated position is represented by black lines, the quadruped position by blue lines, and the prone position by red lines in the proximal thigh region (left), middle thigh region (middle), and distal thigh region (right). Dashed lines are 95% confidence intervals on the mean value of deflection 42 Figure 2. 5 The original seated position (left) removed the posterior segment of the seat pan to allow access to the proximal thigh. The participants sat on a support bar located in the white ellipse, which caused a dropping of the buttocks and additional tissue tensions as this region was stretched between the rear bar and front thigh support. The amended seated position (right) included the participant sitting on the edge of testing chair with a straight back Ogden Material Pa rameters The Ogden model has been shown to successfully describe nonlinear materials using only two parameters [10], [191], [198] . All optimized Ogden material parameters for each thigh region in each position are reported for both males and females in Figures 2. 6 and 2. 7 . Comparisons of Body Positions compared to one another for both males and females. Overall, the µ values were the lowest in the prone position Figures 2. 6 and 2. 7, respectively). In the proximal thigh region overall, the average µ value for the prone position was lower than the average µ for the seated (p = . 022) and quadruped (p = . 003) positions. The difference between µ in the prone and quadruped positions were significant for both the groups of males and females, however the d ifference was not significant between prone and seated positions. In both the middle and distal thigh regions, the average µ value for the prone position was lower than the average µ value for the seated and quadruped positions. These differences were sign ificant for males and 43 Figure 2. 6 µ values for each thigh in each position, separated by sex. The µ values for the thigh regions in the prone position are smaller than those of the seated and quadruped positions, as indicated by the asterisks. Seated parameters are represented with black outlines, quadruped with blue outlines, and prone with red Figure 2. 7 thigh regions in the prone position are larger than thos e of the seated and quadruped positions, as indicated by the asterisks. Daggers indicate differences between thigh regions are represented with black outlines, quadruped with blue outlines, and prone with red 44 females, with p < . 0001 for all comparisons. There were no significant differences between the µ values of the seated and quadruped positions in any thigh region for males or females ( Figure 2. 6). values in the seated and quadruped positions. These differences were significant for both the groups of males and females, with p < . 0001 for all comparisons. In the proximal thigh for males quadruped position (p = . 0412) ( Figure 2. 7). From these data, we can see that there are no significant differences between the µ and values for seated and quadruped positions in any thigh region. However, the µ values for the thigh regions in the prone position were significantly smaller than those in the seated or quadruped ition were significantly greater than those for either the seated or quadruped position. These data indicate the values obtained in the prone position differed significantly from those in both the seated and quadruped positions, but the seated and quadrupe d positions were not significantly different. Comparisons of Thigh Regions within a Given Position differences between the thigh regions in each of the three positions. Males and females were evaluated separately. There were no overall differences between thigh reg ions for the µ parameter within any of the three positions ( Figure 2. 6). values of the more distal regions of th e 45 thigh were smaller than those of the more proximal regions in the seated and quadruped positions. values in the more distal regions of the thigh were higher than the proximal thigh regions while prone. Figure 2. 7 presents the relationships between regions. Specifically, in the seated position, . 0076) and the distal thigh in both males and females (p < . 0001 for both). The middle thigh in the . 0086) for males. In the quadruped . 0024) and the distal thigh in males and f emales (p < . proximal thigh was smaller than that of the middle thigh in females only (p = . 0315) and the distal thigh for both males and females (p < . h was also smaller than that of the distal thigh in the prone position for both males (p = . 0009) and females (p = . 0024). Comparisons of Thigh Regions Between Males and Females Comparisons were made between males and females for each thigh region and ea ch position. The µ values did not show any consistent trends between sexes, nor were there any Figure 2. 7) for the majority of comparisons and one statistically sig nificant difference was noted; males . 0205). 46 Within - Individual Differences Between Body Positions Within - individual differences between the Ogden parameters in each body position were analyzed for each thigh region using a Wilcoxon Signed - Rank Test. Using the Signed - Rank Test, significant differences were found between the µ values of the proximal thigh regions in the quadruped versus prone position (p = .000 6 ). Significan t differences were also found between the µ values of the middle thigh regions in the seated versus prone position (p = .0001) and quadruped versus prone position (p < .0001). Lastly, significant differences were found between the µ values of the distal th igh regions in the seated and quadruped position (p = .0251), seated and prone position (p < .0001), and the quadruped and prone position (p < .0001). Significant within - region. In t position and prone position (p = .0059) and the quadruped position versus the prone position (p < ent in the seated position and quadruped position (p = .0025), the seated position and prone position (p < .0001), and the quadruped position and the prone position (p < .0001). significantly different in the seat ed position versus the prone position (p < .0001) and in the quadruped position versus the prone position (p < .0001). Discussion The goals of this research were to determine and compare the thigh and buttock soft tissue properties in the seated, quadruped, and prone positions. Determining the most realistic material parameters for the seated position is crucial to improve mechanical models of seated individuals. 47 In situations where collecting data from individuals in the seated position is not possible, identifying the best analog to the seated position is also critical. Mechanical models have been used to understand internal tissue stresses of seated individuals, particularly wheelchair users, and to identify the risk of PI formation [189] . The accuracy of these simulations have depended on the material properties used in these models. Thus, having in vivo, human - based, material properties that reflect the body region and the position being modeled is imperative to yielding the most real istic simulations of internal tissue stresses and strains going forward. Our force and deflection data are unique in many ways: 1) they identify differences across three positions, 2) they represent multiple regions along the thighs and buttocks, 3) they are in vivo, and 4) they represent continuous force and deflection measures which are extremely important for the non - linear nature of soft tissues. These datasets show a clear trend indicating that the thigh soft tissue is stiffer in the prone position t han it is in either the seated or quadruped positions. For example, using data from the prone position at the distal region of the thigh underestimates the deflection in that region by approximately 20 - 25 mm while seated, decreasing the deflection by more than half. Because of this, it is imperative that body position be taken into account when selecting the material parameters for soft tissue in any model. We chose to use an Ogden model to represent the soft tissue for its relatively small number of parameters and accuracy of describing tissues [191] . In the Ogden model, the values of the material parameters of µ provide insight into the non - linearity of the stress - stretch curves. A higher is indicative of a quicker change from a soft to a stiff material as stretch increases. So, the differences identified through the Ogden model relate to material stiffnesses directly. All 48 seated position. Following these differences, our data indi cated that soft tissues in the thighs and buttocks in the prone position were stiffer for large deformations relative to the soft tissue while seated. Conversely, there were no significant group - level differences between the µ and values in the seated an d quadruped positions, meaning that the shapes of the quadruped stress - stretch females, with only one significant difference, in the middle thigh while quadru ped. This consistent trend may be due to males having larger amounts of muscle versus adipose tissue relative to females [228] . Because they demonstrated similar between - can be drawn between the seated and quadruped positions. For both the seated and the quadruped positions, the µ values of the proximal, middle, and distal thigh regions showed no overall to distal regions. This finding matched that of previous work conducted to obtain material properties of the seated thigh [221] . The opposite trend emerged in the prone position, where the The Wilcoxon Signed - Rank Test determined that there were some within - individual differences between the Ogden parameters in the seated position versus the quadruped position regions). However, there were no consistent differences between the two positions in either of the parameters. Further, because there we re no significant group - level differences in either parameter between the two positions, the data suggested that differences in parameters between the seated and quadruped positions were small relative to variations of the parameters within the positions . On the other hand, there were consistent group - level and within - individual differences between 49 the parameters in the prone position when compared to either the seated or quadruped positions. As such, the prone position should be considered to produce tissu e material properties distinct from the seated and quadruped. The position should determine data used for material properties. Both material parameters across thigh regions were different for the seated and prone positions. Meanwhile, the quadruped positio n matched for both material parameters across thigh regions of the seated position more closely than the prone position. Because of the similarity between the material properties in the seated and quadruped positions, parameters determined from the quadrup ed position can be used in finite element models that predict internal stresses in seated individuals, whereas the data from the prone position should not be used. This finding is important as we begin collecting in vivo data sets on the buttocks and thighs of wheelchair users. Limitations of this study were the estimation of the initial tissue thickness in each of the three test regions and assumption s used in the tissue modeling: small, but equal forces in the transver se directions and incompressibility . Measurements of the thigh circumferences of each participant were obtained in each region and used with ratios derived from literature to estimate initial tissue thicknesses. Future work in this area will investigate th e use of a portable ultrasound machine to obtain a direct measurement of the tissue thickness [229] . The assumption of incompressibility used in the material model may have cause d the values of the Ogden paramet ers to be underestimated if the internal stresses in the buttock and thigh tissues surrounding the testing regions were large (and resisted lateral expansion of the tissue being tested). However, because the lateral forces were assumed to be small, and the re were no other external forces applied to the tissues surrounding the testing region, the effects of incompressibility on the reported parameters was likely small. 50 Another limitation of this study was the population from which the force - deflection data w ere collected. All participants in this study were healthy, able - bodied individuals. Data from the elderly and wheelchair user groups, who have been deemed at high risk of developing PI s, are also necessary. Work has been conducted to describe the effects of aging and wheelchair use on muscle structure and function, yet there is a lack of research to describe changes in the tissue properties due to the change in structure [80], [86], [230] . Accordingly, obtaining data sets from the thighs and buttocks of wheelchair users and elderly individuals will be the next step in our research. Conclusions A common issue for performing rese arch focused on individuals with disabilities has been the lack of equipment adaptation to suit each individual. Having consulted with individuals with disabilities and rehabilitation specialists, we determined that the quadruped position was an acceptable position for this type of test. In this position, it is possible to compare able - bodied individuals and those with disabilities, breaking down a significant barrier to evaluating the stresses in wheelchair sitting. This research shows that the properties of the soft tissue in the thigh in the prone position do not represent the properties in the same regions when seated or in the quadruped position. Therefore , it is not appropriate to use data collected from the soft tissue in the thighs and buttocks whil e lying down to represent the seated thighs and buttocks. Instead , it is possible to use the quadruped position as an analog for the seated position if the seated position is not a possibility for testing. Using tissue properties obtained while individuals are in either the seated or quadruped position will provide modeling tools with more accurate inputs and ultimately increase their ability to assess PI risk while seated by making internal stress predictions more realistic. Our research 51 will yield better models, providing improved data to engineers and clinicians, with the eventual impact of a decrease in the prevalence of PIs in high - risk groups, such as wheelchair users and the elderly. 52 CHAPTER 3: TISSUE MATTERS: IN - VIVO TISSUE PROPERTIES OF PERSONS WITH SPINAL CORD INJURIES TO INFORM CLINICAL MODELS FOR PRESSURE INJURY PREVENTION The contents of this chapter were originally published in the Journal of Biomechanics. J. Scott, B. Sheridan, R. Andrus, N. Monday, A. Selby, In - vivo tissue properties of persons with spinal cord injuries to inform clinical models for pressure ulcer J. Biomechanics , vol. 120 (7), 2021, doi: 10.1016/j.jbiomech.2021.110389 . Introduction Individuals with spinal cord injuries (SCIs) commonly have secondary medical complications that result in subsequent hospitalizations post - injury. P I s have been reported as the second most common complication and are expensive to treat [6], [54], [231] . The prevalence of PI s in patients with SCIs has been estimated to be between 30% and 47%, with approximately 15% of SCI patients developing a PI within the first year post - injury [7], [54], [60], [232] . PI treatment costs between $20,000 and $30,000 on average, with some PI s costing $70,000 to treat [233] . Because of these costs, the annual expense for treatment in the U nited States has been estimated at 1.6 billion dollars [90] . This expense has prompted efforts to better understand PI formation and develop clini cal models to help in PI risk assessment. Internal tissue stresses in the buttocks and thighs have repeatedly been noted as a cause of PI development [26], [106], [121], [214] . Internal stress concentrations exist in the soft tissues, especially in those w ith a SCI; this is due to the large external loads that occur over extended time periods while seated [16], [62], [82], [147], [196], [234] . Almost half of the PI s affecting the SCI population are located in the buttock and thigh regions [6] . As such, numerous models of the 53 buttocks and thighs have been developed to help assess the risk of PI formation [21], [69], [148], [159], [163], [196], [235] . Finite element models are used to obtain a better understanding of the internal stresses seen in soft tissues and to inform device design. These models have relied on body geometry, external load, and material properties o f the soft tissues to make internal stress evaluations in seated individuals. However, persistent limitations exist in the material properties used in these models. The majority of finite element models used material properties determined from animal model s, such as rats or pigs [192], [196] . The limited research with humans has typically been conducted with able - bodied (AB) participants lying down [23], [208], [219] . Material properties from such studies are appropriate for models of people lying down. However, wheelchair us ers are seated. The tension and thickness of muscle tissue in the buttocks change while seated versus lying down [210], [211], [236] . Recent work found that material properties and tension affect pressure distributions experienced by a human limb analog [237] . Currently, ther e are only three studies with humans that included data from the soft tissue of the buttocks or thighs while seated. Two of them described the buttocks and thighs of AB individuals [221], [236] , while the third collected force and deflection data from the buttocks of wheelchair users in two load states [238] . The material properties of the tissues of wheelchair users have been explored in some studies [239] . However, models designed to evaluate internal stresses and strains in wheelchair users have yet to integrate in - vivo material properties, such as indentation moduli, from persons with disabilities. Muscle progressively atrophies and intramuscular fat increases in the buttocks and thighs of individuals with SCIs [79], [80], [240] . Thus, material properties from this population, which 54 individuals. Differences in material properties manifest as increased interface pressure, increased internal stress, and decreased perfusion in tissue under seated load [82], [83], [241] . This is reinforced by findings from several existing studies, which have shown that the higher levels of muscle atrophy and fat infiltration in individuals with SCIs are correlated with higher PI risk [78], [242] . As such, there is a need for in - vivo material properties of the buttock and thigh soft tissues from individuals who use wheelchairs, such as patients with SCIs. Material properties representative of individuals with SCIs will lead to improved human body models, which lead to improved PI risk assessment and better device designs. There are numerous challenges in obtaining material properties for soft tissues, particularly in the buttocks and thighs of seated individuals with SCIs. Doing so requires access to the buttock and thigh regions while seated. Access to those regions can be accomplished by using a specialized chair with removable supports. Using such a chair is likely to be difficult for individuals with SCIs, wheelchairs and into a lab - type chair adds another challenge. As an alternative to the seated position, the quadruped position is commonly used in the physical therapy of individuals with SCIs and produces similar joint angles to a seated posture. We previously showed tha t the material properties obtained in the quadruped position were similar to those in a seated position [236] . The aims of this research were to determine the material properties of soft tissue in the buttocks and thighs of individuals with SCIs while in the quadruped position and to compare these material properties to those of AB people. Realistic material properties in human body models will lead to advancem ents in the understanding of PI formation, further improvements in existing device designs, such as pressure relieving cushions [21], [175], [198], [201] , and be the basis for the development of new devic es. 55 Methods Participants Eleven individuals who were wheelchair users with a spinal cord injury and twenty AB individuals volunteered to participate in this study. All testing was approved by the Biomedical and Health Institutional Review Board at Michigan State University, and co nsent was obtained from all participants. Participants were excluded from the study if they did not feel they would be able to maintain a supported quadruped position for half an hour, if they were pregnant, had active wounds on the leg, osteoporosis, or o ccult bone disease. Participants with spinal cord injuries were required to have been using a wheelchair for at least six months. Quadruped Position AB participants wore athletic clothing and were placed in a quadruped (crawling) position that provided support at the torso and knees [236] . The heights of the supports were adjusted such hs of the participants were blocked with a support to ensure that their thighs did not move during testing, as seen in Figure 3. 1. Figure 3. 1 Positioning protocol for users in the AB (left) and SCI (middle and right) groups. Participants with SCIs were helped onto a table and into the quadruped position by their therapists while the harnesses were secured (middle); and then supported in the testing position with an exercise ball beneath the torso and sand bags in front of the knees (right) 56 Individua ls in the SCI group wore clothes that they regularly wore to physical therapy, and testing occurred prior to their physical therapy sessions at their rehabilitation clinic. With their exercise unit. Individuals were first positioned prone on the table with harness straps beneath their thorax and pelvis. The harnesses were attached to the universal exercise unit and tightened as the table was lowered, moving the participant into a quadr uped position with the hips and knees flexed at ninety degrees. A peanut - shaped exercise ball was used to assist in supporting the torso; and sandbags were positioned in front of the thighs and ball to stop their thighs from moving during the testing sessi on, seen in Figure 3. 1. Three body regions were evaluated: the proximal thigh/buttocks (proximal thigh), middle thigh, and distal thigh regions. These regions were identified based on a percentage length of the femur. The length of the femur was considere d as the distance from the greater trochanter to the lateral epicondyle, and the three regions were defined by their locations relative to the two bony landmarks. The respective distances of the proximal, middle, and distal thigh regions distal from the gr eater trochanter were 10, 50, and 90 percent of the length of the femur. For the region called (which was 5 cm by 5cm) that encompassed tissues over the ischial tu berosity as well as tissues just distal to the ischial tuberosity. This is a region in which PI s commonly develop. The distal thigh was just proximal to the popliteal fossa of the knee, a place where PI s have been also reported [70], [71], [73], [74], [243] . The middle thigh was between these two locations. Circumference buttocks. 57 Tiss ue Testing Protocol & Material Parameter Estimation A custom - made indenting tool previously reported was used to collect continuous force and deflection data for this study ( Figure 3. 2a) [236] . Force and deflection data were collected in each of the three testing regions during two indentations with a maximum applied load of between 30 and 40 kPa ( Figure 3. 2b). T hese force values were in the same ranges as those used during the testing of able - bodied participants in Chapter 2. These force and deflection data were converted into stress and stretch data using the geometry of the indenting tool and the circumference [10], [221], [236] . Stress and stretch data were mapped to an incompres sible first - order Ogden material model. A complete characterization of these methods has been included in the supplemental material. Figure 3. 2 a) Indenting tool with the six - axis load cell embedded beneath the indenting surface and linear potentiometer used to obtain distance data, and b) Indenting tool as used during testing with the participant Statistical Analysis measures ANOVA was used to differentiate between the values of parameters in the three regions (proximal, middle, and distal thigh regions). An ANOVA was used to differ entiate between the values of the parameters in each group (SCI versus AB). Post - hoc Tukey tests were used to detect 58 method. Results Demographic Data Nine males with spinal cord injuries (age 44.0 ± 5.2 years, height 180.6 ± 1.7 cm, mass 77.6 ± 2.9 kg, time in chair 9.8 ± 5.1 years), and two females with spinal cord injuries (age 45.5 ± 7.5 years, height 165.1 ±2.5 cm, mass 73.0 ± 22.2 kg, time in chair 3.25 ± 2.75 years) were recruited for this study. Additionally, ten AB males ( age 22.4 ± 2.2 years, height 179.7 ± 3.4 cm, mass 77.4 ± 2.0 kg ) and ten AB females ( age 21 ± 1.3 years, height 164.9 ± 1.9 cm, mass 66.5 ± 4.0 kg ) volunteered. The two groups will be referred to as the SCI and AB groups. Force and Deflection Data Force and deflection curves are shown for each region in Figure 3. 3. For each region, the average force and deflection data from AB individuals were plotted, including 95% confidenc e intervals. Deflections at 50 N of applied force (20 kPa of applied pressure) were compared across groups because it has been reported as a typical load on the buttocks and thighs while seated [227], [244] . Seven of the eleven in the SCI group had more deflection in their proximal thigh than the average AB individual at 50 N of force (denoted by the thick line in Figure 3.3 ), and two had less. In the middle thigh, seven in the SCI group had more deflection, and one had less. In the distal thigh, two members of the SCI group had more deflection and seven had less. Upon statistical analysis of these data, the SCI group had significantly more deflection at 50 N in both the proximal and middle thighs (8.5mm, p = .0222 and 7.2 mm, p = .0162, respectively). Although the deflection 59 in the distal thighs of the SCI group was 4 mm less than the deflection in the distal thighs relative to AB group, the difference was not significant (p = . 6173). Figure 3. 3 Force and deflection data from the proximal thigh (top), middle thigh (middle), and distal thigh (bottom) regions of the SCI group. The solid black line is the average AB data, and the dashed black lines represent the confidence intervals of the AB group. The data shaded in light green (left of the average) has less deflection than the average AB, and the data shaded in darker green has more deflection To ensure that the inclusion of both sexes did not change the results, statistical analyses were run twice. No significant differences were found between the deflection data of AB males and females. For the SCI group, one analysis included only males (the SCI group only had two females), and one included both sexes. As there were no differences in the results, the latter are reported here. Ogden Material Parameters To describe the material properties, the stress - stretch curves were mathematically fit usin g the Ogden model (Eq. 3. 5 in the Supplementary Material ). Values for the material parameters µ Tables 3. 1 and 3. e the shapes of the material 60 response curves for each region. The material parameters for each region in each group are presented with standard errors in Figure 3. to detect differences. The proximal, mi ddle, and distal thigh regions of both the SCI and AB groups were compared to find between - region differences within each group. Figure 3. and the able - bodied group are white. Significant differences between groups are indicated by asterisks (*). Between - region (i.e., proximal, middle, distal) differences within the SCI group are indicated by plus signs (+). Between - region differences in the able - bodied group are indicat There were no significant differences between the µ values of AB males and females in . 0205) for the middle thigh region only. Thus AB males and females were con sidered together. Although there were only two females in the SCI group, there were no significant between - sex differences in the deflection or either of the parameters of that group. Therefore, males and females were also considered together in the SCI gr oup. The µ values for the wheelchair users were significantly smaller than those of the AB individuals in the middle thigh (p = .0075) and distal thigh (p = .0349) regions. The µ values for ; and although they did not achieve significance (p = . 1831), a medium effect size of 0.50 was computed. For comparisons 61 .0001). Additionally, a medium effect size of 0.40 was found for the middle thigh region and a Between - region differences within each group were detected by comparing the proximal, middle, and distal thigh regions. In the SCI group, both the proximal thigh (p = .0094) and the the proximal (p = .0091) and distal (p < .0001) thigh regions of SCI group were greater than the and distal thigh regions in the SCI group. Table 3. 1 Material parameters for each region for males in the SCI and AB groups with standard errors Proximal Th igh Middle Thigh Distal Thigh µ (k P a) SCI 3.90 ± .54 4.54 ± .66 3.51 ± .52 AB 4.91 ± .38 5.35 ± .44 4.78 ± .59 SCI 5.89 ± .78 4.42 ± .69 6.37 ± .85 AB 5.72 ± .43 4.72 ± .38 3.84 ± .27 Table 3. 2 Material parameters for each region for females in the SCI and AB groups with standard errors Proximal Thigh Middle Thigh Distal Thigh µ (k P a) SCI 3.95 ± .22 2.91 ± .32 2.31 ± 1.32 AB 5.24 ± 1.14 6.65 ± 1.17 4.39 ± .42 SCI 6.13 ± 1.12 4.98 ± 1.27 6.40 ± 1.64 AB 5.53 ± .52 3.61 ± .28 3.26 ± .34 62 The µ values for the middle thigh regions of AB individuals were larger than those of the significantly greater than the values in the thigh regions were larger than the values in the distal thigh, but that difference was not significant (p = .4051). Discussion The motivation of this research was to determine the material propertie s of the soft tissue in the buttocks and thighs of individuals who had a SCI and compare them to those of AB individuals. Distinguishing between the material properties of individuals with SCIs and AB individuals is vital to understanding PI risk and forma tion. Additionally, developing human body models that are informed by in - vivo material properties of the population being studied will lead to improved engineering devices and designs, such as new wheelchairs and cushions. The research presented shows clea r differences between the tissue properties of AB individuals and people with SCIs. Published research concluded that there are optimal material properties for seat cushions to best distribute seated pressures [165] . These optimal cushion characteristics are dependent on the properties of the soft tissue of the occupant. Material property differences between AB people and those with SCIs may help explain why wheelchair users experience higher peak tissue stresses tha n AB individuals while seated on the same foam cushion [82] . Human body models used to identify int ernal stresses and strains have implemented tissue properties from AB individuals and animals [18], [180], [196] . Although these models provide initial fin dings that can be applied to those who use wheelchairs, these models will improve when material properties obtained from the population being studied are implemented. Using material 63 properties derived from individuals with SCIs will lead to more realistic internal tissue stress predictions and even personalized modeling. Internal stresses and strains have been linked to the development of PI s; therefore, it is imperative that they be quantified using realistic soft tissue properties so improved PI mitigatio n techniques can be developed. The dataset reported here is the first continuous force - deflection dataset for undeformed buttock and thigh tissue in individuals with SCIs and the only one to compare them to AB individuals. One published study collected an unloaded condition and two poin ts of force - deflection data from buttock tissues in wheelchair users directly on top of the ischial tuberosity while experiencing a seated load [238] . Differences in the testing protocol a nd location of testing between the previous study and the one presented here likely contributed to differences measured tissue responses between both studies. Here, continuous force and deflection data were obtained in the quadruped position, which is simi lar to the seated position. These data will support tissue property selection, particularly for design and modeling applications [236] . These force and deflect ion data show an increase in deflection of the proximal and middle thigh regions for participants in the SCI group relative to AB participants. In the distal thigh region, the SCI group experienced less deflection than AB individuals. The same trends seen in the force and deflection data were also reflected in the Ogden material parameters used to describe the tissue. and non - the stiffness of the tissue model, whereby an increase in µ inc reases the slope of the stress - stretch, - linearity of the curve. The µ parameter is a multiplicative 64 largely affects the large - stretch behavior of the model; and load bearing regions of the buttocks and thighs experience large compressive stretches while seated. middle have smaller µ values, indicating the softer tissue. This finding i s supported by previous work showing an increase in intramuscular fat and decrease in muscle in the buttocks and thighs, which would lead to softer tissue responses [78], [80], [245] . In the middle and distal thigh regions, the decreases in µ values were sig nificant for the SCI group. Additionally, the between - group comparison of the µ parameter had a medium effect size in the proximal thigh. Such an effect size suggests that a larger sample of participants would likely result in a significant difference betw een the SCI and AB groups. Moreover, parameters indicated that the middle thighs were the softest in the SCI group, while the distal thighs were softest in the AB group. Thus, there were differences in trends within the thighs of the SCI and AB groups. Ove rall, the proximal and middle thigh regions of those who had SCIs were softer than the same regions as AB individuals, while the distal thigh regions were stiffer. The Ogden model used to describe the tissues of AB group consistently had larger µ values th an the models of the SCI the models of the SCI group were larger than those of AB group, indicating the stiffer tissue. In the proximal and middle thigh relative to the AB group indicated softer tissues. Due to these variations in stiffness, it is critical to consider wheelchair users and AB individuals separately in design and modeling applic ations. Further, future studies may consider the varying degrees of SCIs, such as the level of the injuries 65 and whether they are complete or incomplete. Doing so will provide a more complete description of the material properties individuals with SCIs beca use it is known that complete SCI leads to more intramuscular fat than incomplete SCI, ultimately leading to higher risk of PI development [78], [241] . Limitations of this work included the ratio of males and females in the SCI group. The clinic we tested has a compa rable number of males and females over the course of a year, however more men volunteered for our study than women. Additionally, to convert the force - deflection data sets to stress and stretch, MRI or ultrasound measurements would have been ideal. Circumf erence measurements of the thighs, and ratios to relate the thigh circumferences to the posterior compartment thicknesses, were used instead of imaging or direct measurements of the anterior - posterior thicknesses of the thighs. Either of these methods woul d have provided a more direct measurement of initial tissue thickness and will be utilized in future work. Future work will identify approaches that can that identify other options for in - clinic measures. Lastly, the location of the proximal thigh region w as used to represent the regions around the ischial tuberosities, as they are prone to PI development. The indentation testing for the proximal thigh region occurred with the center of the indenter just distal to the ischial tuberosities, while the proxima l half of the indenter overlapped the ischial region. The tissue over the ischia region has a different underlying anatomical structure than the middle and distal thigh regions. Thus, variation within a given group due to region of testing could be attribu ted to anatomical differences. This anatomical difference was not addressed as part of this study, but is likely to be relevant for modeling. 66 Supplementary Material Tissue Testing Protocol A custom - made indenting tool previously reported was used to collect continuous force and deflection data for this study [236] . A five centimeter by five centimeter square indenting surface was mounted on a s ix - axis load cell ( Figure 3. 2a). A linear potentiometer was attached to the indenting tool to record deflections during tissue indentation. Before each indentation, the tool was positioned flush with the skin without deforming the tissue. At this point was a zero force value and the largest distance between the tool and a fixed reference pole [236] . This was defined as the undeformed thigh position for each region ( Figure 3. 2b). This was unique for each region of each individual and depended on body geometry. Distance measurements during indentation recorded by the tool were accurate within 2 mm, and force measurements were accurate to within 2 N. The tool wa s then pressed into the thigh using a constantly increasing load until the soft tissue no longer compressed, at approximately 40 kPa of pressure ( Figure 3. 2b). The load was then slowly removed. Deflection was calculated by subtracting the reference measure ment from the distance measurements. During each indentation, the indenting tool moved perpendicular to the long axis of the thigh and was positioned at the height of each test region (e.g. the proximal, middle, and distal thigh). Two trials were completed for each region; a five minute rest period for tissue recovery was provided after testing before a second trial was conducted. Material Parameter Estimation The experimental data were mapped to the Ogden material model and model parameters obtained. To m ap to the model, force and deflection data collected during the indentation tests were transformed into stress and stretch data. The force data were converted into stress by dividing 67 deflection data were every thigh region was determined through the use of previously published data and anthropometric measurements from each individual [10], [221] . Based on published literature, the ratio of the total anterior - posterior thickness of the thigh with respect to the thigh circumference was identified as 0.33 [10] . The portion of soft tissue thickness in the thigh posterior to the femur has been reported to be 0.55 of the entire anterior - posterior thickness of the thigh [221], [222] . Thus, the undeformed tissue length was computed by multiplying these two ratios with each ne the undeformed thickness of the thigh tissue was Eq. 3. 1. In Eq. 1, c t was the measured thigh region circumference, t AP was the anterior - posterior thickness, and t P was the thickness of the thigh tissue posterior to the femur. The undeformed thickness of each region and the deflection data were used to calculate stretch perpendicular to the long axis of the thigh using Eq. 3. 2 . In Eq. 3. 2, o was the undeformed thickness of the tissue thickness, and d was the deflection. The stress and stretch data were used to fit parameters in an incompressible first order Ogden material. The strain energy function of this material was given by Eq. 3. 3. ( Eq. 3.1 ) ( Eq. 3.2 ) ( Eq. 3.3 ) 68 1 2 3 1 specifically was the stretch in the direction was a unit - less nonlinearity parameter. The stress in any direction was define d as the product of the stretch itself and the derivative of the strain energy function with respect to the stretch in that direction. The stress perpendicular to the long axis of the femur was defined by Eq. 3. 4. Incompressibility was enforced by conserving the volume of the buttock and thigh tissue, with 1 2 3 = 1 at all times. Only internal forces, which were assumed to be the same in both directions due to the isotropic nature of the Ogden model, acted in the transverse directions; thus the stretches in the transverse directions were equal ( 2 3 1 - 1/2 ). The relationship between the stress and stretch during the uniaxial compression was described using Eq. 3. 5. 1 , respectively. The values individual. ( Eq. 3.4 ) ( Eq. 3.5 ) 69 CHAPTER 4: SHIFTING LOADS AND PERFUSION RESPONSES AS A RESULT OF CHAIR ARTICULATION IN THE CONTEXT OF PRESSURE INJURIES Introduction Pressure injuries (PIs) disproportionately affect wheelchair users, such as those with spinal cord injuries (SCIs). Prevalence rates are as high as 47% in some care facilities [7], [17] . PIs can take months to heal and incur an average cost of $35,000 per incident, with some PIs costing over $120,000 [8], [13], [14], [16] . Recurrence rates up to 53% indicate that PIs can become chronic, debilitating conditions that disengage people from their daily lives [246] . Further, the prevalence of PIs, cost of treatment, and recurrences result in annual costs estimated near $27 billion in the United States alone [15] . PIs are localized soft tissue injuries, usually around bony prominences [24] . However the mechanisms that cause PI formation have yet to be confirmed [24], [25] . One thought is that interface pressures on soft tissue occludes the blood perfusion to the affected area and deprives tissue of nutrients, resulting in necrosis [119], [213] . This theory was supported by studies reporting decreased blood perfusion in tissue experiencing normal and shear pressure [101], [120] . Another potential mechanism is that tissue dies due to direct mechanical compression [26] . Animal studies have shown that tissues experienced necrosis w ithin two hours with sufficient compressive pressure [104] [107] . Regardless of the mechanism, susta ined pressures have been linked to PI development. Thus, it is imperative to identify new approaches of decreasing pressures on high - risk areas to reduce PI incidence. Sustained pressure is especially salient for individuals with SCIs, as many use wheelc hairs and have limited sensation, making it difficult for them to determine when body regions experience such pressures. Those with SCIs have numerous regions of the body susceptible to PI formation, including the heels, greater trochanters, ischial tubero sities, sacrum, coccyx, and scapulae [17], 70 [30] [33], [51] . In particular, the soft tissue in the buttocks (beneath the ischial tuberosities, or ITs, and around the coccyx) and lower back (behind the sacrum) are high - risk areas, in part due to lack of a thick layer of soft tissue overlaying the bony prominences [16], [51], [182] . One PI prevention strategy is seated repositioning, either with back recline or whole - body tilt to change pressure patterns [2], [161], [247] . Whole - body tilt as implemented in current chairs is a simultaneous back and seat pan rotation where the angle between the back and se at pan remains constant at approximately 90°, resulting with the wheelchair user facing upwards. The effects of back recline on risk factors associated with PIs are mixed [108], [248] . And although whole - body tilt has shown promise for reducing factors related to PI incidence, ma ximum benefits in terms of pressure reduction are achieved at large tilt angles, which are the least popular because they disengage people from tasks of daily living [247], [249] . A robust study on the effects of both back recline and isolated seat pan tilt on seated pressure and blood perfusion has yet to be published. Understanding how these individual movements affect pressure and perfusion is essential to identifying seated positions that can reduce PI risk. There are limited investigations into how changes in interface pr essures on the buttocks due to wheelchair repositioning affect blood perfusion [69], [100], [247] . Some studies have probed the effects of normal pressure, shear pressure, and skin temperature on perfusion, but these have not replicated seated pressures [3], [100], [101] . Research to address the relationships between pressures and perfusion in various seated positions needs to be expanded for a more comprehensive assessment of PI risk. Therefore, the goals of this research were to evaluate the ability of an articulating c hair design with isolated back recline, seat pan tilt, and back articulations to 1) redistribute pressures on the body and 2) change blood perfusion. 71 Methods Participant Recruitment Twenty able - bodied adults, ten male and ten female, gave consent to par ticipate in this study. Approval for this study was granted by the Biomedical and Health Institutional Review Board at Michigan State University. Figure 4. 1 a) Pressure mats were also affixed to the chair during portions of testing, b) Schematic of the rotations of the articulating chair, with the back, thorax, pelvis and seat pan identified. The thorax and pelvic supports rotated about their centers, and the back and seat pan rotated about the point labelled CoR Articulating Chair Desi gn An articulating chair was designed with an independently rotating (tilting) seat pan, as well as independently rotating pelvic and thoracic supports. The pelvic and thoracic supports were mounted on a reclining back, as seen in Figure 4. 1. All three sup ports were attached to six - axis 72 load cells (AMTI, Watertown, MA) and covered by a thin layer of foam with a vinyl cover. A horizontal (0° tilt), and the pel vic and thoracic supports were parallel to the back (vertical). The back recline was able to move from vertical to 20° rearward of recline [2] . The front of the seat pan was able to tilt up to 30° with respect to horizontal, such that the front edge of the seat pan was higher than the back. The rotation center for the chai r back and seat pan (CoR in Figure 4. 1b) was located approximately at the hip joint center. Both the pelvic and thoracic supports (which rotated about their centers) were able to rotate 15° in either direction (i.e. upward facing 15° or downward facing 15° ). The thoracic support was rotated to face downward 15°, and the pelvic support was rotated to face upward 5°to create a concave surface for a slouched (kyphotic) back articulation. The opposite rotations were used to support an erect (lordotic) articulat ion [250] ( Figure 4. 1b). The articulating chair was used to collect data in two sequential sessions, as described in the following sections. Pressure data were collected in 27 seated positions. Based on the results of the pressure data during pilot testing, a second set of tests w ere conducted in nine positions to collect perfusion data. These nine positions were those chair configurations that had the largest effects on pressure in the regions of the ITs. Note, chair back articulations had the least effect on the IT pressures; the refore only back recline and seat pan tilt were used during the collection of perfusion data. Table 4. 1 Combinations of chair back recline, seat pan tilt, and back articulation in which pressure measurements were taken. Each back recline was used with eac h seat pan tilt and back articulation resulting in 27 different chair positions Back Recline Seat Pan Tilt Back Articulation No recline (0°) No pan tilt (0°) Slouched Half recline (10°) Half pan tilt (15°) Neutral Full recline (20°) Full pan tilt (30°) Erect 73 Back and Seat Pan Pressure Measurements Pressure measuring mats (Tekscan, Boston, MA) covered the seat pan and the pelvic and thoracic supports of the articulating chair ( Figure 4. 1a). Each mat was an array of 42 by 48 sensors. The pressure mats were secured to the seat pan and back supports to maintain consistent alignment. To quantify interface forces, the seat pan, pelvic, and thoracic supports were mounted on load cells located in the centers of their cross sections. All load cel ls were calibrated before the participant sat in the chair. Each participant was instructed to sit with their buttocks in contact with the pelvic support while the chair was in a neutral position (i.e. no back recline, no seat pan tilt, neutral back articu lation). The chair was then articulated into each of 27 positions (Table 4. 1) with a randomization strategy. A measurement of the pressure on the buttocks and back was obtained while in each position with simultaneous recordings from the load cells. Pre ssure Data Analysis The total force measured by the pressure mats (in Newtons) on the seat pan and back supports were compared to the perpendicular forces measured by the load cells (also in Newtons) beneath the supports [251], [252] . Uniform scaling factors (the force measured by the load cell(s) divided by the force measured by the pressure mat) were applied to the press ure measurements (in kPa) on each pressure mat to ensure the sums of the forces measured by pressure sensors equaled the total force obtained by the load cells, described by Eq. 4. 1. (Eq. 4.1) 74 The total back pressure mat force was compared to the sum of the pelvic and thoracic load cell forces, and the total seat pan pressure mat force was compared to the seat pan load cell force. The value of each sensor with a nonzero measurement of pressure (P i,j ) was then reported as the average value of nine sensors (the sensor itself and all eight adjacent sensors). This process is described by Eq. 4. 2. - 42) and column (1 - 48) of the individual pressure - - indicated in Eq. 4. 2. Figure 4. 2 Segmentation of the seat pan (left) and back (right) pressure mat readings into regions. The seat pan was split up into the buttocks and thigh regions. The back was split up into the upper, middle, and lower back regions. , the location of the maximu m pressure, was measured in mm from the posterior edge of the pressure mat in the buttocks and thighs and from the inferior edge of the pressure mat for the upper, middle, and lower back regions Regions of interest were determined by segmentation of the back and seat pan pressure mats. The back regions were determined by the segmentation of the back pressure mat into thirds from (Eq. 4.2) 75 superior to inferior (upper, middle, and lower). The seat pan pressure mat was segmented into posterior and anterior halves, lab eled the buttocks and thighs ( Figure 4. 2). The locations and magnitudes of the sensors recording the maximum pressures within each body region were recorded (i.e., the row of the measurement from the pressure mat). These rows were used to determine the sup erior - inferior location of the maximum pressures on the back and the anterior - posterior location of the maximum pressures on the seat pan. The locations of the maximum pressures were recorded as the distances from the posterior edge of the seat pan in mill imeters for the buttocks and thighs ( , respectively, larger values indicated more anterior); and locations of the maximum pressures in the back regions were recorded as distances from the inferior edge of the back ( is the upper back l ocation, is the mid back location, is the lower back location, and larger values indicated more superior), as seen in Figure 4. 2 . Perfusion Measurements An initial analysis of pressure data indicated that back recline and seat pan tilt had much larger effects on the magnitude and locations of pressure concentrations in the buttocks than back articulation. Therefore, perfusion data were collected on the ski n over the right IT in the nine combinations of back recline and seat pan tilt indicated in the first two columns of Table 4. 1, while the back articulation was maintained in the neutral position. All participants wore athletic shorts so a perfusion senso r could be attached directly to the skin over the right IT. Participants were instructed on how to identify the IT via palpation; and once identified, a researcher attached a laser Doppler flowmetry sensor (Perimed, Järfälla, Sweden) to the tissue with a d ouble - sided tape ring under and tape over the sensor to ensure good contact throughout the session. Participants sat in the position for two minutes to allow perfusion 76 to stabilize [164] . Perfusion measurements consisted of recording 30 seconds of data in each of the nine positions during which participants did not move. Perfusion Data Analysis The average values of perfusion over each recording period were calculated. Perfusion data were then normalized for each individual by dividing the average measurement in each position by the maximum average measurement recorded for that individual, meaning the values reported spanned between zero and one. Such normalization has previously been conducted to ensure that comparisons bet ween chair positions were not affected by inter - person differences in perfusion magnitude [247] . Statistics A repeated measures ANOVA, followed by a Tukey test, wa s conducted to determine the effects of the back recline, seat pan tilt, and back articulation on the magnitudes and locations of the maximum pressures in the three back regions, buttocks, and thighs. A second analysis was conducted to determine the effect s of the seat pan tilt and back recline on the blood perfusion in the tissue over the IT. Results Participants Ten males (average age 22.2 ± 3.0 years, average height 178.8 ± 4.2 cm, average mass 78.6 ± 11.6 kg), and ten females (average age 22.7 ± 3.6 years, average height 165.5 ± 5.4 cm, average mass 58.9 ± 9.3 kg) participated in this study. 77 Seated pressures and perfusion were analyzed within each sex. There were no differences in trends of pressure or perfusion between sexes with respect to recli ne or seat pan tilt . There were also no differences in pressure trends in response to changes in back articulation. Thus, the seated loading data and perfusion for both sexes were analyzed together and are presented in the results. Recline The effects of back recline on the magnitudes and locations of the maximum pressures in each region are reported in Figures 4. 3 and 4. 4, respectively. Reclining the chair back increased the maximum pressure in the buttocks (p < .0001) and lower back (p < .0001) regions, while it decreased the maximum pressures in the upper back (p < .0001) and thigh (p < .0001) regions. Reclining the bac k also caused a significant posterior shift of the maximum pressures in the buttocks (p < .0001) and inferior movement of maximum pressure in the middle back (p < .0001). Figure 4. 3 The effects of back recline on the magnitudes of the maximum pressures within each region. Columns are grouped by regions. Statistical differences between recline angles are indicated by asterisks (*). Increases in back recline angle consistently resulted in significant decreases in maximum pressures in the upper ba ck and thighs and increases in maximum pressure in the lower back and buttocks 78 Figure 4. 4 The effects of back recline on the locations of the maximum pressures within each region. Statistical differences between recline angles are indicated by asterisks (*). Increases in back recline angle resulted in the maximum pressure shifting inferiorly for the middle back and posterior shifts in the buttocks Seat Pan Tilt The effects of seat pan tilt on the magnitudes and locations of the maximum pressures in each region are reported in Figures 4. 5 and 4. 6, respectively. Increasing seat pan tilt decreased the maximum pressures in the buttocks (p < .0001) and lower back (p < .0001) regions and increased the maximum pressures in the thighs (p < .0001) and in the upper (p < .0001) and middle (p < .0001) back regions. The maximum pressures in the middle (p < .0001) and lower (p = .0005) back regions shifted superiorly when the seat pan was tilted. The maximum pressures in the buttocks shifted anteriorly with increas ing seat pan tilt (p < .0001). 79 Figure 4. 5 The effects of seat pan tilt on the magnitudes of the maximum pressures within each region. Statistical differences between seat pan tilts are indicated by asterisks (*). Increases in seat pan tilt corresponded to increased maximum pressure in the upper back, middle back, and thigh regions and decreased maximum pressure in the lower back and buttocks regions Figure 4. 6 The effects of seat pan tilt on the locations of the maximum pressures within each region. Statistical differences between seat pan tilts are indicated by asterisks (*). Increasing seat pan tilt corresponded to superior shift of maximum pressures in the middle and lower back regions and the anterior shift of maximum pressures in the butt ocks region 80 Articulation of the thorax and pelvic supports slouched, neutral, erect The effects of thorax and pelvic support articulations on the magnitudes and locations of the maximum pressures are reported in Figures 4. 7 and 4. 8, respectively. When ch anging articulation from slouched to erect, the maximum pressures in the upper back increased (p < .0001), while the maximum pressures in the middle (p < .0001) and lower (p = .0106) back regions decreased. Changing articulation from slouched to erect shif ted the maximum pressures inferiorly in the upper (p < .0001) and middle (p = .0012) back regions and superiorly in the lower back region (p < .0001). Figure 4. 7 The effects of back articulation on the magnitudes of the maximum pressures within each bac k region. Statistical differences between back articulations are indicated by asterisks (*). When changing articulation from slouched to erect, the maximum pressures in the upper back region increased, while the maximum pressures in the middle and lower ba ck regions decreased 81 Figure 4. 8 The effects of back articulation on the locations of the maximum pressures within each back region. Statistical differences between back articulations are indicated by asterisks (*). When changing articulation from slouched to erect, the maximum pressures in the upper and middle back regions shifted inferiorly, while the maximum pressures in the lower back regions shifted superiorly Effects of Chair Articulation on Perfusion The average perfusion values at each back recline and seat pan tilt angle are shown in Figure 4. 9. Perfusion values were observed to have stabilized in each position before/during each measurement (slope of - .01±.14 perfusion units/second over the duration of each recording period). Participants had less perfusion in positions with larger back recline angles than in positions with no recline (p = .0450). Increasing seat pan tilt increased blood perfusion in the buttocks regardless of back recline angle (p = .0006). Participants had gr eater perfusion in positions with the half tilt angle relative to positions with no tilt (p = .0265) and greater perfusion in positions with full tilt than positions with half tilt (p = .0158). 82 Figure 4. 9 Average buttock perfusion values for each back recline (left) and seat pan tilt angle (right) . Statistical differences between back recline or seat pan tilt angles are indicated by asterisks (*). Perfusion decreased in the full back recline positions relative to the no back recline positions and increa sed with seat pan tilt, regardless of the back recline angle Discussion The primary goal of this research was to investigate the effects of chair articulation on interface pressure and perfusion while seated. This goal was met by determining the relations hips between isolated chair articulations, seated pressures, and perfusion measurements. Of particular interest were the buttocks and lower back regions due to the prevalence of PIs in those areas. Our data suggested that implementing only back recline, wh ich is a commonly recommended approach for wheelchair users to reduce pressure on the ITs, may not be effective at relieving the pressure concentrations in those high - risk regions. On the contrary, our data indicated that the opposite occurred, and back re cline increased pressure on the ITs. However, tilting the seat pan independently of the back significantly decreased the maximum pressure in the buttocks and caused a shift in pressure concentrations to different body regions. Furthermore, an increase in blood perfusion was consistently seen with seat pan tilt, further s upporting the use of an isolated seat pan tilt for PI risk reduction. These findings indicated that with the correct combination of chair articulations, we can affect changes in parameters that are linked to PI formation. 83 Trends in these data showed that back recline and slouching increased the maximum pressures on the buttocks and lower backs, thus increasing the risk of PI formation. This is problematic because back recline is the most common pressure relief movement for wheelchair users [2] . Moreover, both recline and slouching moved the locations of the maximum pressur es towards bony landmarks around which PIs form (sacrum, ITs, and coccyx), further increasing risk in those areas [30], [51] . The data from the seat pan tilt articulation and erect back articula tion showed the opposite trends, indicating potential protective effects in the buttocks and lower back. These movements led to reduced pressures and, in the case of seat pan tilt, increased perfusion. Seat pan tilt shifted pressure from the buttocks onto the thighs and shifted the maximum pressure locations anteriorly in the buttocks region, away from the ITs and coccyx. In the back, the erect articulation decreased the maximum pressure in the lower back and shifted its location superiorly, offloading the bottom area of the lower back near the sacrum. These movements reduced risk factors associated with PI formation in the buttocks and lower back, both of which are PI prone areas. This research study provides new insights into how seated articulations affec t key PI risk factors. Previous research indicated that tissues can survive large loads, as long as they are applied intermittently [106], [253] . Thus, the work presented here indicates that one potential strategy to reduce PI risk would be a new wheelchair design that cyclically loads and unloads different regions of the body while seated. Not only would such a regimen intermittently unload regions of the body, our research indicates that it restores perfusion to areas that may experience decreased perfusion while seated, such as the ischium. Further research is needed to confirm these effects in wheelchair users, but these initial findings in able - bodied individuals are promising. 84 Conclusio ns This work conducted tests on able - bodied individuals in an articulating chair to investigate the effects of seated repositioning on interface pressure over the entire body and blood perfusion in the buttocks. Results indicated that 1) tilting the seat pan while seated reduced the pressure on the buttocks and lower back, 2) reclining the back reduced pressure on the upper back and t highs, and 3) reductions in pressure in the buttocks coincided with increased perfusion under the ITs. Based on these results, a seated repositioning regimen to cyclically reduce pressure on body regions and promote blood flow can be used, thus leading to a reduction of reduce pressure injury risk. Future work will investigate whether pressure and perfusion trends seen in able - bodied individuals are also the same in wheelchair users. 85 CHAPTER 5: KEY COMPONENTS RELATED TO PRESSURE INJURY FORMATION: UNDERSTANDING PRESSURE DISTRIBUTION AND BLOOD PERFUSION RESPONSES IN WHEELCHAIR USERS Introduction PIs have been shown to cause significant challenges for wheelchair users, including prevalence rates as high as 47% , hospital stays of up to 22 weeks and increased risk of infection [7], [14], [16], [254] . Furthermore, studies showed that recurrence rates were as high as 53% [246] . T reatment costs of PIs have also been reported to be significant with expenses of $120,000 per incident [8] . These factors contributed to the reported annual cost s of $27 billion for PI treatment in the United States [15] . The problem of PIs for wheelchair users is not trivial. The regions that have the largest incidence of PI formation were predominately found around b ony landmarks that experience large loads while seated [16] . Specifically, t he ischial tuberosities, coccyx, sacrum, and shoulder blades have been reported as having the highest incident of PI formation [16], [17], [21], [32] . Wheelchair users have been reported to spend 10 - 12 hours a day in their chair with limited movement; studies have shown that the seated position produced larger tissue stresses in their buttocks than able - bodied people [ 2], [82], [255], [256] . R educed capillary blood flow, or blood perfusion , combined with the increased internal tissue stresses for extended time in the seated position has been shown to result in tissue necrosis and PI formation [17], [27], [101] . Thus, loading and blood perfusion have been frequently been used by researchers as metrics for the risk assessment of PI development [28], [91], [120] . Several strategies have been utilized by clinicians and wheelchair users in an attempt to reduce the incidence of PI formation. These strategies have included specialized cushions, who le - body - lifts executed by the patient (e.g., using their arms to lift their buttocks off the seat pan) and seated repositioning [21], [257] [146] . S tudies reported that cushions were able to reduce, but not 86 eliminate, the occurrence of PIs [21], [147], [148] . Also, t he literature reported that whole - body - lifts requ ired more mobility than some wheelchair users possess [164] . T he most commonly used repositioning strategies in wheelchairs were back recline and whole body tilt, yet l imited evidence has been published to support their ability to reduce pressure s in high - risk regions [2], [108], [125], [248] . A recent study found that whole body tilt and back recline could improve perfusion in the buttocks , ho wever, this was true only if the tilt and recline angles were large [161] . T ilted positions also were shown to have some negative outcomes, such as increased pressure on the sacrum and disengaged patients from tasks of daily living [68], [249] . Thus, a need exists for PI prevention strategies that are accessible to all wheelchair users and have evidence to support their usage . Seated repositioning has been one strategy for PI prevention, but existing research on repositioning is limited b oth by the populations tested and the modalities of recorded data. Two published studies were located that included individuals with SCIs and evaluated the effects of isolated seat pan tilt and whole - body tilt on seated interface pressure. One study investigated isolated seat pan tilt, but it did not include any other movements of the chair, and interface pressure data were only collected on the seat pan [258] . The second study investigated pressure changes in the buttocks and back due to whole body tilt, but isolated back recline or seat pan tilt were not investigated [68] . Neither of these studies included isolated back recline, which is the most commonly used position for pressure relief in wheelchair users [2] . Existing work noted that peak loads on the buttocks of individuals with SCIs were greater than those of able - bodied individuals, deepening the nee d for an improved understanding of interface pressure data from wheelchair users using common repositioning strategies [21], [108], [160] . Pressure changes on tissue have been shown to affect perfusion in the tissue [100], [101], [119], [259] . While one stud y investigated how perfusion in the buttocks changes with seated repositioning and pressure changes, it focused 87 on able - bodied individuals [161] . Since pressures in wheelchair users are larger than able - bodie d individuals , there is a need to understand perfusion responses in the buttocks of that population as well . The research reported here was designed to address some of these gaps. Thus, the goals of this study were to evaluate the effects of back reclin e, seat pan tilt, and back articulation on the interface pressures and buttock perfusion across a sample of wheelchair users. Methods Participant Recruitment Ten adult male wheelchair users volunteered to participate in this study. Exclusion criteria included active wounds on the buttocks or thighs, vascular diseases, or an inability to transfer into or change position in the laboratory chair. Approval for this study was granted by the Biomedical and Health Institutional Review Board at Michig an State University. Articulating Chair Positions A custom designed articulating chair was used to collect data for this study, shown in Figure 5. 1. The articulating chair had independently rotating seat pan, pelvic, and thoracic supports; and the pelvic and thoracic supports were mounted on a reclining back. Both the seat pan and back rotated about a point approximately at the hip joint center of a seated individual (labeled Figure 5. 1). The seat pan, pelvic, and thoracic supports were each mounted to a six - axis load cell (AMTI, Watertown, MA) located in the middle of their cross sections and covered with a thin foam layer and vinyl. Angles of back recline, seat pan tilt, and rotation of the pelvic and thoracic supports are reported in Table 5. 1. The angle of the back was measured relative to the 88 vertical axis (up to 20° of recline), the seat pan relative to the horizontal axis (up to 30° of rearward tilt) ( Figure 5. 1). The rotation of the thoracic support face was downward 15° for the slouched back articulation, while the rotation of the pelvic support face was 5° upward with respect to the overall recline [250] . The opposite rotations were used for the erect back articulation. For the neutral back articulation, both supports were parallel to one another. Controls were positioned on the arm of the cha ir to adjust the motors that drove articulations. 27 combinations of back recline, back recline, no seat pan tilt, and a neutral back articulation. Figure 5.1 a) Schematic of the rotations of the articulating chair, with the centers of rotation for the back and seat pan (CoR), pelvic support, and thoracic support identified. b) Pressure mat placement on the articulating chair during testing. c) Wheel chair user sitting in the articulating chair during testing Based on the results of pilot work, it was found that back recline and seat pan tilt caused the largest changes in pressure (both magnitude and locations) under the buttocks and thus also affect ed perfusion the most. Therefore, a subset of positions with three back recline angles and three seat pan tilt angles (9 in total) were repeated with the addition of a perfusion probe under the right ischial tuberosity. 89 Table 5. 1 Combinations of chair bac k recline, seat pan tilt, and back articulation in which pressure measurements were taken. Each back recline was tested with each seat pan tilt and back articulation Back Recline Seat Pan Tilt Back Articulation No recline (0°) No pan tilt (0°) Slouched Half recline (10°) Half pan tilt (15°) Neutral Full recline (20°) Full pan tilt (30°) Erect Interface Pressure Data Collection & Analysis Pressure and force data were collected in all 27 chair positions. Two pressure mats (Tekscan, Boston, MA), consisting of 42 rows of 48 sensors, covered the seat pan, pelvic, and thoracic supports ( Figure 5. 1b). One mat covered the seat pan, and the other covered the pelvic and thoracic supports. Consistent alignment throughout the testing was maint ained by securing the pressure mats to their respective supports. The load cells mounted beneath each support were calibrated before participants transferred into the chair and were used to collected forces on each support. Participants transferred into th e chair while it was in a neutral position, and each participant sat with their buttocks contacting the back of the chair along the pelvic support. The chair was then articulated, using a randomization strategy to select the back recline angle, seat pan tilt angle, and back articulation until data were collected in all positions. The forces measured from the load cells were used to ca librate the pressure data. The total forces measured by each pressure mat were related to the forces perpendicular to the face of each support measured by each load cell [251], [252] . The magnitudes of each pressure sensor measurement on the seat pan were multiplied by the ratio of the perpendicular force measured by the load cell under that support divided by the total fo rce measured by the pressure mat on the seat pan . The magnitudes of every pressure sensor measurement on the back were multiplied by the ratio of the sum of the perpendicular forces measured by the load cells behind the pelvic and 90 thoracic supports divided by the total force measured by the pressure mat on the back . This process is described by Eq . 5. 1 : where and were used to represent the pressure magnitude (in kPa) and area (in mm 2 ) of each sensor in the mat, as the perpendicular force (in N) measured by either the load cell under the seat pan or the sum of the load cells behind the pelvic and thoracic supports, and represented the pressure adjusted so that the total force matched that of the load cell. After adjustment, the final pressure value for each sensor was determined to be the average value of the sensor itself and the eigh t surrounding sensors. This process is summarized by Eq . 5. 2 : where represented the final averaged pressure value for the sensor in the i th row (1 - 42) and j th column (1 - 48) . Regional Analysis of Pressures Each pressure mat was split into regions based upon the rows of sensors. The pressure mat on the seat pan was split into anterior and posterior halves, termed the thighs and buttocks. The pressure mat on the back was split into thirds from inferior to superior, termed the lower back, middle back, and upper back ( Figure 5. 2). The locations of the ischial tuberositi es and sacrum were in the buttocks and lower back regions, respectively. Both magnitudes and locations of the ( Eq. 5.2 ) ( Eq. 5.1 ) 91 largest pressure values in each region were recorded. Locations on the seat pan ( for the buttocks and for the thighs ) were measured as d istances in millimeters from the posterior edge of the seat pan, and locations on the back ( for the lower back, for the middle back, and for the upper back ) were measured as distances from the inferior edge of the pelvic support. Fig ure 5.2 Break down of the seat pan (left) and back (right) pressure mat readings. The seat pan was split up into the buttocks and thigh regions with the zero reference point being at the most posterior location of the mat. The back was split into the upper , middle, and lower back region with the zero reference being at the most inferior point of the mat (near sacrum) Perfusion Measurements & Analysis The right ischial tuberosity was located via palpation, and a laser Doppler flowmetry sensor (Perimed, Järfälla, Sweden) was attached to the soft tissue over the right ischial tuberosity with double - sided tape beneath the sensor and medical tape over top o f the sensor. Participants then sat in the articulating chair in a neutral position. With the participant seated and a neutral back articulation, the chair was then moved into one of the 9 combinations involving back recline and seat pan tilt. The particip ants sat in that position for two minutes before perfusion measurements 92 were obtained [164] . Thirty seconds of blood perfusion data were collected; and if the participa nt moved or encountered a spasm during a recording, the recording for that position was started over after the movement or spasm finished. The reported measurement for each position was the average value of the perfusion recording over the thirty seconds. Lastly, the perfusion measurements were normalized within individuals by dividing them by their largest perfusion measurement across all positions. This permitted comparisons of the changes caused by posture changes across individuals [247] . Statistics A repeated measures ANOVA was used to determine the effects of back recline, seat pan tilt, and back articulation on magnitudes and locations of the maximum pressure data across all positions. A second repeated measures ANOVA was used to determine the effects of back recline and seat pan tilt on perfusion measurements in the buttocks. Differences in the magnitudes and locations of the maximum pressures and per fusion data due to the articulations were detected with post - hoc Tukey tests. Results Participants Ten male wheelchair users (average age 37.1±7.89 years, average time using a chair 21.8±12.85 years, average height 174.8±13.21 cm, average mass 76.4±16.53 kg) participated in this study. Participants reported spending an average of 11.50±2.51 hours per day in their chair and had a range of conditions and injuries that resulted in wheelchair usage. These included spinal cord injuries, cerebral palsy, and spi na bifida. 93 Articulation Effects on Pressure Recline Back recline affected the magnitudes and/or locations of the maximum pressures in the upper back, lower back, buttocks, and thighs, shown in Figures 5. 3 and 5. 4. The maximum pressures in the buttocks and lower back regions increased with back recline (p < .0001 for both regions), while the maximum pressures in the upper back (p < .0001) and thighs (p = .0003) decreased. The locations of the maximum pressures shifted posteriorly in the buttocks (p = .0002) and thigh regions (p = .0136) with back recline. Figure 5.3 The effects of back recline on the magnitudes of the maximum pressures within each region. Increasing the back recline angle increased maximum pressures in the buttocks and lower back and decreased maximum pressures in the upper back and thighs. Differences between recline angles are indicated by asterisks (*) 94 Figure 5.4 The effects of back recline on the locations of the maximum pressures within each region. Increasing the back recline angle caused a posterior shift in the maximum pre ssure regions of the buttocks and thighs. Differences between recline angles are indicated by asterisks (*) Seat Pan Tilt Seat pan tilt affected the magnitudes and/or locations of the maximum pressures in the upper back, middle back, buttocks, and thighs, shown in Figures 5. 5 and 5. 6. The maximum pressures in the buttocks decreased (p < .0001) and maximum pressures in the mid dle (p < .0001) and upper (p < .0001) back regions increased with seat pan tilt . The locations of the maximum pressures in the middle (p < .0001) and lower (p < .0001) back regions shifted superiorly and locations shifted anteriorly in the thigh (p < .0001 ) regions with seat pan tilt. 95 Figure 5.5 The effects of seat pan tilt on the magnitudes of the maximum pressures within each region. Increasing the seat pan tilt angle increased the maximum pressures in the upper and middle back and decreased maximum pressures in the buttocks. Differences between seat pan tilts are indicated by asterisks (*) Figure 5.6 The effects of seat pan tilt on the locations of the maximum pressures within each region. Increasing seat pan tilt angle caused a superior shift in the maximum pressure regions of the middle and lower back and an anterior shift in the maxi mum pressure regions in the thighs. Differences between seat pan tilts are indicated by asterisks (*) 96 Back Articulation Back articulation affected the magnitudes and/or locations of the maximum pressures in upper, middle, and lower back regions, shown in Figures 5. 7 and 5. 8. No significant effects due to back articulation were detected in the magnitudes or locations of the maximum pressures in the buttocks and thighs. The maximum pressures increased in the upper back (p < .0001) and decreased in the middle back (p < .0001) regions when changing back articulation from slouched to erect. The locations of the maximum pressures shifted inferiorly in the upper back (p < .0001) and superiorly in the lower back (p < .0001) when changing back articulation from slou ched to erect. Figure 5.7 The effects of back articulation on the magnitudes of the maximum pressures within each region. Articulating the back from slouched to erect increased maximum pressures in the upper back and decreased maximum pressures in middle back. Differences between back articulations are indicated by asterisks (*) 97 Figure 5.8 The effects of back articulation on the locations of the maximum pressures within each region. Articulating the back from slouched to erect caused an inferior shift in the maximum pressure regions in the upper back and a superior shift of the maximum pressure regions in the lower back. Differences betwee n back articulations are indicated by asterisks (*) Postural Effects on Perfusion Mixed results were observed with respect to blood perfusion in the wheelchair users. P erfusion values were observed to have stabilized in each position before/during each me asurement (slope of .00±.20 perfusion units/second over the duration of each recording period). Participants fell into two groups: 1) those who experienced an increase in blood perfusion in the buttocks in response to seat pan tilt (termed responders), and 2) those who did not experience an increase in perfusion in response to seat pan tilt (termed non - responders). Four p articipants experienced an increase in perfusion with increasing seat pan tilt (p = .0013) and were designated the responder group. The no n - responder group included two participants who had no change to their buttocks perfusion with seat pan tilt and four participants who experienced a decrease in buttock perfusion with seat pan tilt (p < .0001). The perfusion values for the entire cohort an d responder group are reported in Figure 5. 9 . 98 Figure 5.9 Normalized perfusion values for the entire cohort (left) and responders (right) at every back recline and seat pan tilt angle. Seat pan tilt increased perfusion in the responder group (p = .0013) While both the responder and non - responder groups expe rienced the smallest maximum pressures in the buttocks in the positions with the maximum seat pan tilt, the magnitudes of the pressures differed between the two groups. The average maximum pressure in the buttocks of the responder group was 22.7 ± 3.6 kPa in positions with a full seat pan tilt. This average pressure was significantly smaller (p < .0001) than the average maximum pressure in the buttocks of the non - responder group in the same position (66.3 ± 4.3 kPa). Though there was a significant differenc e between pressures in the responder and non - responder group, there was not a significant difference between the masses of the responders (81.1 ± 7.6 kg) and non - responders (73.3 ± 6.5 kg) (p = . 50). Discussion This research was conducted to evaluate the effects of chair articulations on interface pressures and buttocks perfusion of wheelchair users while seated. This research filled a knowledge gap, permitting a detailed understanding of how changes in seating p osition of wheelchair users affect interface pressures and blood perfusion . 99 Wheelchair users experienced increased pressure in the buttocks and lower back with back recline . In addition, they experienced a posterior movement of pressure concentrations in the buttocks. These two findings combined indicated that back recline produces circumstances that negatively affect factors known to increase PI formation. This is specifically true for body regions where PIs are common, the buttocks and lower back. Howev er, one of the strategies wheelchair users are encouraged to implement frequently is to recline the common clinical advice being that reclining would reduce pressures in these regions [2] . Based on our findings, the clinical suggestion of recline as a pressure relieving mechanism for the buttocks (ischial tuberosities an d coccyx) and lower back (sacrum) should be re - evaluated. S eat pan tilt decreased pressure on the buttocks, which supports its use for the reduction of PI risk in the buttocks. The superior shift in maximum pressure of the lower back with seat pan tilt further suggests that seat pan tilt may reduce PI risk around the sacrum. However, few wheelchairs offer the ability to change the seat pan tilt as an option; many times tilting is done as a whole - body - tilt, where the seat back and pan tilt together a s a single motion. Our research also indicated changes in the back supports which created erect and slouched positions also showed the ability to shift pressure from one body region to another. An erect back articulation was observed to shift pressure conc entrations in the lower and upper back towards the middle of the back and away from the bony prominences of the sacrum and scapulae. The presented data show that chair articulations such as seat pan tilt and back articulation are able to affect factors tha t are known to cause PIs . Despite the consistent reduction of pressures in the buttocks with seat pan tilt, all participants did not exhibit a consistent change in perfusion. Some participants showed an increase in perfusion, while others showed no change , and some exhibited a decrease with seat pan tilt. This issue needs further exploration. These differences could be related to the anatomy of the individual 100 (less muscle mass and more adipose tissue), or the fact that the sample had varying types of injur ies/diseases, or that there was a minimal layer of foam on our chair as compared to a thicker cushion. T hose who did not experience an increase in buttock perfusion had approximately three times as large a pressure concentration in the buttocks as those wh o had increased perfusion , providing insight into the relationship between pressure and perfusion changes . While seat pan tilt alone did not increase the buttock perfusion of all the participants in our chair, it could be used in conjunction with other pre ssure reducing tools, such as specialized cushions, to further reduce pressure in the buttocks to a magnitude small enough that all wheelchair users are able to experience an increase in perfusion. Based on the findings of our research, seated repositioni ng has promise as a tool to reduce the incidence of PIs by shifting the magnitudes and locations of interface pressure concentrations , and for some individuals, by increasing perfusion. Pr ior research indicated that tissues were able to survive external pr essures for limited lengths of time [213] . Therefore, seat articulations such as the ones evaluated in this study could be used to cyclically load each region of the body , reduc ing the risk of PIs in any specific region by limiting exposure to pressure [106] . Doing so would be protective of the ischial tuberosities, coccyx, sacrum, and scapulae, all bony landmarks where PIs frequently occur [51] . Thus, a regimen of timed articulations presents one additional strategy in the prevention of PIs in wheelchair users. Limitat ions This work investigated the seated pressures under the buttocks and backs of wheelchair users. Although the cohort of wheelchair users in this study had ten people, there were no femal es who volunteered to participate. Additionally, we did not have a homogenous group in terms of 101 their injury or disease which required wheelchair use . Both of these limitations should be addressed in future studies . A limitation of the perfusion measurements was that the positions in which the maximum perfusion measuremen ts were taken were not consistent for all individuals. Ideally, the maximum measurements could have been obtained in a position with the buttocks unloaded for all individuals, but such a position was not feasible with this cohort. In future studies, access ible perfusion data collection protocols will be explored for a position with no load on the buttocks Such a position can be used as a consistent maximum perfusion position with which to normalize all perfusion data. Future work may also consider normalizing loads by the body weight of the participants as an alternative to comparing raw pressure values to describe trends in the load data. 102 CHAPTER 6: THE EFFECTS OF PANTS AND CUSHION COVER MATERIAL ON SHEAR FORCES EXPERIENCE D WHILE SEATED Introdu ction There are over 3 million wheelchair users living in the United States [1] . These wheelchair users are at an elevated risk of developing pressure injuries (PIs), with up to an 80% lifetime risk of experiencing one [61] . Incidence of a PI is highly disruptive, as treatment for PIs almost always involves consistent repositioning to offload the area around the wound, debridement of the wound itself, and dressings that need to be changed, among many other lesser us ed treatment strategies [260] . Further, treatment of PIs cost about $ 27 billion in the United States each year [15 ] . Thus, PIs are serious issues in how widespread, disruptive, and expensive they are. One of the most consistent factors cited in the development of PIs is the presence of shear loading on the tissue [91], [98], [122], [261] . Shear loading has b een implicated in the formation of PIs that start in superficial tissues and spread to deeper tissues [27] . Shear causes internal stress perfusion to the loaded tissues [3], [101], [120], [124], [175] . Both internal stress and reduced blood perfusion contribute to tissue necrosis over time, which can lead to PI formation. Soft tissues in the buttocks and thighs experience shear loading while seated, which creates a higher risk for PI development [1 01], [262] . Almost half of all pressure injuries occur in the buttocks and thighs, indicating a need to address the shear loading those regions experience [16] . Despite evidence showing that the buttocks and thighs experience increased shear load while sitting in reclined positions, reclined positions are the most popular positions for wheelchair users, and they are prescribed for pressure relief clinically [2], [108], [249], [263], [264] . The popularity of reclined positions while seated, paired with the prevalence of PIs in the buttocks and thighs, 103 underscore a need to reduce the shear loading on the buttocks and thighs while seated, especially in reclined positions. One method by which the shear load on the buttocks and thighs may be reduced is by selecting materials for the cushion covers that have low coefficients of friction. Standard cushion covers use materials such as vinyl, however thos e materials may have relatively high coefficients of friction, causing increased shear loading on the buttocks and thighs while seated. Another material, such as the nylon used in transfer sheets in hospitals, may provide a lower friction surface that help s lower the shear force and therefore reduce the risk of PIs. By using a material with a lower coefficient of friction for a cushion cover, the maximum amount of friction the cover can sustain is reduced, and that provides a mechanism by which shear on the buttocks and thighs is reduced. Friction is generated by contact between two surfaces; and thus the interaction between the cushion cover material and that of clothing is an important factor in order to understand shear force . Numerous studies have inves tigated the effects of moisture and temperature on friction between skin and textiles commonly used to make clothes [113], [115], [265] . These studies found that increasing moisture on the skin increased friction , correlating two factors that are thought to contribute to PI risk [91] . However, there is a dearth of literature describing the friction generated between cushion cover material s and materials commonly used to make pants while seated. As such, there is a need to evaluate the frictional properties of cushion covers when paired with clothing to reduce sh ear loading. Though there is a lack of information regarding the interactions of the material on the cushion cover and the clothing worn by the wheelchair user, this interaction is critical to the generation of shear force on the buttocks and thigh tissue s. Thus, it is necessary to determine the 104 friction and shear forces on these material pairings. As stated, reducing the coefficient of friction between pants and cushion cover reduces the maximum amount of shear that can be sustained by a pair of materials before sliding occurs, potentially reducing friction on the buttocks and thighs. Additionally, understanding the effects of a commonly used positioning method back recline on shear force is also important. Therefore, the goals of this study included using a tilting seat pan 1) to determine the coefficients of friction between three cushion covers (vinyl, one - layer nylon, and two - layer nylon) and two clothing fabrics commonly worn by able - bodied individuals and wheelchair users (cotton denim a nd cotton - polyester blend) using a laboratory system (called a sled) to simulate the buttocks, 2) to compare the coefficients of friction found using the sled to coefficients found with human participants, and 3) use a chair with back recline to evaluate the ability of the cushion covers to reduce shear loading on the buttocks and thighs while seated in reclined positions . Methods Overview Three specific hypotheses were tested in this work to address the three specific goals. The first hypothesis was that the coefficients of friction when pairing pant materials with seat pan covers would differ and that pairings of pant materials with covers made from nylon would have smaller coefficients of friction than pairings with vinyl covers. This was tested using a mechanical system that consisted of a weighted wooden sled, whose bottom was covered by the pant materials. This experimental work utilize d the tilting seat pan surface of the articulating chair designed for the work in chapters 4 and 5. 105 The second hypothesis was that coefficients of friction could be determined using humans instead of the sled and that the frictional coefficients would be the same with either the sled or the humans, even though humans have deformable soft tissue in their buttocks and thighs. To test this hypothesis, participants sat in the articulating chair while wearing pants made of the same material tested with the sle d, and the seat pan was tilted until the participants started to slide. This was done for each of the pant materials and each of the seat pan covers used in the sled testing. The third hypothesis was that the seat pan covers would affect the shear forces on the buttocks and thighs while seated in reclined positions , specifically that the nylon seat pan covers would reduce shear on the buttocks and thighs. To do this, motion capture data and force data were collected as participants sat in the articulating chair, and the chair was reclined. Shear force data were collected at three levels of recline, as determined by motion capture, and compared between the pairings of seat pan cover and pant materials for which the coefficients of friction were previously d etermined. Articulating Chair Design The custom - designed articulating chair designed for chapters 4 and 5 was also used for this research . The articulating chair had independent, motor - driven back recline and seat pan tilt movements. The back was able to recline rearward from vertical to nearly 20 degrees and the seat pan was able to tilt up to 40° such that the anterior edge of the seat pan was higher than the posterior edge. Two reflective markers were attached to the chair back, and four reflective mar kers were attached to the seat pan to record the orientations of the back and seat pan, respectively ( Figure 6.1). The seat pan was mounted on a six - axis load cell (AMTI, Watertown, MA) to collect normal and shear force data on the seat pan during testing. 106 Determination of Coefficients of Friction via a Sled to Simulate the Buttocks Static and kinetic c oefficients of friction were determined for the pairings of two pant materials (denim and cotton - polyester blend ) and three cushion covers (vinyl, one - layer nylon, and two - layer nylon) us ing a sled that simulated the buttocks ( Table 6.1 ). The sled had a square cross - sectional area of 225 cm 2 , was designed to hold 45 N of weight (weight carried by the ischial tuberosities while seated), and had interc hangeable fabrics attached to the bottom [5], [266] . The seat pan covers were attached to the seat pan with Velcro. For the one - layer covers, and for the bottom layer of the two - layer nylon cover, the material was attached with Velcro to the lateral and anterior edges of the seat pan. The top layer of nylon for the two - layer nylon cover was attached to the posterior edge of the seat pan to ensure that it could freely slide on top of the bottom layer. Four reflective motion capture markers were attached to the sled to determine its position in space (markers S 1 - 4 in Figure 6.1). Figure 6.1 Articulating chair setup for the sled testing. The markers on the front right (FR), front left (FL), back left (BL), and back right (BR) were used to determine the orientation of the seat pan ( ). The markers on the sle d, S1 - The anterior (AA) and posterior (AP) markers on the chair arm were used to determine the back recline. All positions were given in the coordinate system in the top left of the figure 107 Table 6.1 Material pairings for the pants materials and seat pan covers Seat p an cover material Sled bottom material Vinyl Denim (100% cotton) One - layer nylon Cotton (60%) - polyester (40%) blend Two - layer nylon M otion capture data were collected at 60 Hz to determine the static and kinetic coefficients of friction of the different material pairings using the position of the sled and the orientation of the seat pan. The position of the sled ( ) was the average position of the four marke rs on the sled (i.e. ) . The orientation of tilt of the seat pan was determined using two of the markers (Front left (FL) and back left (BL) in Figure 6.1 ) and Eq . 6.1. where was the position vector of marker FL, was the position vector of marker BL, and was the unit vector that represented the orientation of the seat pan. The angle of tilt of the seat pan was determined using vector and Eq . 6.2. where was the y - component of the seat pan orientation vector, was the z - component of the seat pan orientation vector, and was the angle of tilt of the seat pan. The position of the sled along the seat pan was calculated using the orientation of the seat pan and the positions of the sled and marker BL in space. This is described in Eq . 6.3. ( Eq. 6.1 ) ( Eq. 6.2 ) 108 where was the position of the sled along the seat pan. Figure 6.2 Free body diagrams for the sled before (left) and after (right) it starts to slide. The friction before the sled starts to slide is determined by the static coefficient of friction ( ), and the friction after it starts to slide is determined by the kinetic coefficient of friction ( ) The sled was placed on the seat pan while it had 0° of tilt. The seat pan was then tilted until the sled started to slide along the seat pan. Just prior to the sled starting to slide along the seat pan, the angle of tilt ( ) was held constant. The only forces acting on the sled were gravity, friction parallel to the seat pan, and the force normal to the seat pan ( Figure 6.2). As such, the static coefficient of friction ( ) for the material pairing was determined using Eqs . 6.4 through Eq . 6.6. All motion of the sled was along the seat pan orientation vector ( ). Thus, the acceleration ( ) of the sled along the seat pan was determined by using changes in the position of the sled along the seat pan ( ) to determine mean acceleration between frames. To deter mine velocity, the change in ( Eq. 6.4 ) ( Eq. 6.6 ) ( Eq. 6.5 ) ( Eq. 6.3 ) 109 position along the seat pan was tracked from frame to frame. Since the frequency of data collection was known (60 Hz), the change in position was divided by time between frames to determine the velocity of the sled during frame s when it slid. Changes in velocity between frames were then divided by the time between frames to determine the acceleration of the sled during frames when it slid. The acceleration while the sled was sliding was used to determine the kinetic coefficient of friction ( ) for the material paring using Eq . 6.7. where was the gravitational constant ( ). Ten trials were conducted for each material pairing, and the coefficients of friction were averaged among the trials. Coefficients of Friction with Human Participants The coefficients of friction for each material pairing were also computed with data sets collected on human participants. 10 able - bodied individuals (5 male, 5 female) volunteered , and t esting protocols were approved by the Biomedical and Health Institutio nal Review Board at Michigan State University . C onsent was obtained from all participants. As with the simulated buttocks, four reflective markers were attached to the seat pan. ior superior iliac spines, greater trochanters, lateral epicondyles, and lateral malleoli ( Figure 6.3). Motion capture data of all the markers were collected. Participants each wore two pairs of pants: denim jeans and cotton - polyester blend sweatpants, and three cushion covers were tested (Table 6.1). ( Eq. 6.7 ) 110 Figure 6. 3 Reflective marker placements on the lower body of a participants: anterior superior iliac spines (ASISs), greater trochanters, lateral epicondyles, and lateral malleoli Figure 6.4 (a) Front view of a participant sitting on a flat seat pan, (b) side view of a participant sitting on a flat seat pan, with the arrow indicating the space between the participant and the chair back, (c) participant sitting on a tilted seat pan just be fore starting to slide 111 Participants initially sat down on the seat pan with 0° of seat pan tilt, with the back of their buttocks 15 cm from the posterior edge of the seat pan, hands placed in their lap, and were instructed to not contact the back of the c hair with any part of their back ( Figure 6.4a and Figure thighs were fulling resting on the seat pan. The seat pan was then tilted until the participant started upper back made contact with the back of the chair, whichever came first ( Figure 6.4c). After the participant started to slide or contacted the back of the chair, the seat pan tilt was held constant. The orienta tion and angle of the seat pan tilt were calculated using Eqs . 6.1 and 6.2. The average position of the lateral epicondyles was recorded relative to the seat pan, similarly to position of the sled in the previous section. The average position of the marker s on the lateral epicondyles was used to determine when participants slid along the seat pan because the markers on the anterior superior iliac spines and greater trochanters were sometimes obscured . 6.8. where was the average position vector of the markers attached to the lateral epicondyles, and was the average position of lateral epicondyles along the seat pan. The acceleration of the knees along the seat was found using the same process that was used to obtain the acceleration of the sled, and the static and kinetic coefficients of friction wer e calculated using Eqs . 6.6 and 6.7, respectively. Three trials were conducted for each participant wearing each pair of pants on each cushion cover, and coefficients of friction were averaged across trials. In trials where the vinyl seat pan cover was use d, participants did not slide along the pan. A seat pan tile of over 30 degrees was required to start sliding and by the time the pan reached that angle, participants were unable to ( Eq. 6.8 ) 112 keep their upper backs from touching the back of the chair without leaning significantly forward, which would have affected results. Thus, coefficients of friction were not found for the material pairings including the vinyl cover with human participants. Evaluating the Effects of Seat Pan Cover on the Shear Forces While Seated When back recline happens, shear force on the buttocks and thighs of wheelchair users increases. To identify the effects of seat pan cover materials during back recline, a third set of tests was run. Motion capture data of the participant and chair , and shear force data on the seat pan, were collected while participants sat in the articulating chair at multiple recline angles for all the material pairings listed in Table 6.1. Participants sat on the articulating chair with the backs of their buttock s contacting the pelvic support, their hands in their laps, and their feet resting on the footrest. The back was reclined to 17° past vertical and then moved back to the 0° recline (vertical) position while the seat pan remained horizontal. Participants th en stood up out of the chair to prepare for the next trial. Three separate trials were conducted for each material pairing (for a total of 18 trials per participant), and motion capture and shear force data were collected during each trial. Motion capture and shear force data (collected at 60 Hz and 120 Hz, respectively) were used to determine the recline angle of the back and shear force on the seat pan, respectively. Markers AA and AP ( Figure 6.1 ) were used to determine the angle of back recline. The shear forces in the anterior - posterior direction of the seat pan were recorded when the angle of back recline was 5°, 10°, and 15° for each trial. A description of the procedure used to verify the accuracy of the shear force measurements on the seat pa n is included in Appendix B . 113 Statistics A two - way repeated measures ANOVA was used to determine the effects of recline angle and material pairing on the shear forces on the seat pan. Tukey tests were then used to identify significant differences between t he shear forces in different recline angles and material pairings. Results Coefficients of Friction with the Simulated Buttocks Figure 6.5 Coefficients of friction for each of the material pairings, found using the sled to simulate the buttocks. Brighter bars represent static coefficients of friction, while darker bars are kinetic coefficients of friction The coefficients of friction a s found from the weighted sled are reported in Figure 6.5 and Appendix C . The static and kinetic coefficients of friction were largest for the vinyl cover, regardless of pants material it was paired with; and the coefficients of friction were smallest for the two - layer nylon cover for both pant materials. Though the coefficients of friction for the pant materials on one layer of nylon were consistently smaller than the coefficients of friction of the 114 pants on the vinyl cover, the reduction in the coefficien t of friction was dependent on the type of fabric used in the pants. The reduction in both static and kineti c coefficients of friction when switching the cover from vinyl to the one - layer nylon cover was 10% for the denim pants material and 26% for the cotton - polyester blend. The kinetic coefficients were smaller than the static coefficients of friction by a rang e of 2 - 9%, with the smallest change occurring in the denim on two - layer nylon cover and the largest change occurring in the cotton - polyester blend on one - layer nylon cover. In summary, the vinyl seat pan cover had the largest coefficients of friction, and the two - layer nylon cover had the smallest for both pant materials. The cotton - polyester blend pant material had smaller coefficients than the denim pant material on the vinyl and one - layer nylon seat pan covers . However, the two pant materials had simila r coefficients of friction when interacting with the two - layer nylon seat pan cover. Coefficients of Friction with Human Participants Figure 6.6 Coefficients of friction for each of the material pairings, found using human participants. Brighter bars represent static coefficients of friction, while darker bars are kinetic coefficients of friction 115 Five able - bodied males (average age 22.4 ± 3.1 years, average height = 181.4 ± 5.0 cm, average weight = 754.7 ± 85.9 N) and five able - bodied females (average age 23.0 ± 6.2 years, average height = 162.6 ± 10.5 cm, average weight = 607.9 ± 39.3 N) participa ted in this study. The coefficients of friction as found from the human participants are reported in Figure 6.6. and Appendix C . As stated in the methods, coefficients of friction were not found with human participants when a vinyl seat pan cover was invol ved because the angle of tilt needed start the For both pairs of pants, the coefficients of friction were smaller on the two - layer nylon cover than on the one - laye r nylon cover. The kinetic coefficients of friction determined from the human trials were between 2.5 - 3.5% smaller than the static coefficients of friction. Effects of Seat Pan Cover on Shear Forces Figure 6.7 Shear forces on the seat pan of each material pairing at each angle of recline. All shear forces are in the anterior direction of the seat pan . Recline increased shear force on the seat pan, regardless of the material pairing. Asterisks (*) indicate significant differences in shear forces on each pair 116 The shear forces on the seat pan with all three covers and both pants materials at three recline angles are reported in Figure 6.7. Interaction effects between the recline angle and material pairing were investigated, but there were no significant findings. All shear forces reported are in the anterior direction of the seat pan (+y in Figure 6.1). The data indicated that increasing b ack recline, regardless of seat pan cover , increased the shear force on the seat pan (p < .0001). Though there were no significant differences between the shear forces on any of covers across pant materials, the two - layer nylon cover reduced shear force on the buttocks and thighs by about 20% relative to the vinyl cover. The results also showed that there were several material pairings that reduced shear force relative to others. Denim on vinyl had less shear force than denim on one - layer nylon (p = .0370) and cotton - polyester blend on vinyl (p = .0071). Denim on two - layer nylon (p = .0258) and cotton - polyester blend on two - layer nylon (p = .0487) also had less shear force than cotton - polyester blend on vinyl. Discussion This work was carried out to deter mine frictional properties of pairings of seat pan covers and pant materials. Additionally, this work investigated the effects different seat pan covers had on shear force as recline was varied. The methods used to determine frictional properties, the fric tional properties themselves, and the shear force results have implications that span experimental, computational, and clinical biomechanics, in particular as they relate to PI prevention. Coefficients of friction of the material pairs were found using a sled and tilted surface. This method was previously used to determine coefficients of friction between material samples of rocks, whose shape remains relatively constant during testing [266], [267] . However, this was the 117 first published work that applied the method to the human body, specifically the buttocks and thighs covered by pants materials. The results indicated that the nylon covers had smaller coefficients of friction than the vinyl cover. This is significant because the shear forces that the nylon covers can support are smaller than those that of the vinyl cover, providing one mechanism by which shear may be reduced on the buttocks and thighs of wheelchair users. This was determined using a seated position with a tilting seat pan. Understanding the effects a material surface has on shear force generation can help in reducing internal tissue shear as well as maintaining positioning of seated occupants. The coefficients of friction of the material pairs were found using the tilted surface method with a rigid mechanical system (sled) used to simulate the buttocks and again with human participants. The results of this study showed the sa me trends and similar numerical coefficients of friction with the rigid sled and the deformable human. T he coefficients of friction for the denim and cotton - polyester blend pants on the one - and two - layer nylon cushions were close to one another (within 10 %) between the simulated buttocks and human participants. It also means that using simulated buttocks may be an adequate substitution for human participants when determining coefficients of friction. Experiments with simulated buttocks are more repeatable than using human participants, as evidenced by the smaller standard deviations of the coefficients of friction found using the sled. Because simulated buttocks remove human variation, they provide a method to consistently determine coefficients of friction that may be useful for clinical settings (by using cushion covers with small coefficients of friction). One finding from this work that directly impacts current clinical care and best practices is that back recline increased shear force on the buttocks and thighs regardless of material pairing . This finding can be taken in conjunction with previous work that shows that normal forces on the 118 buttocks, specifically over the ischial tuberosities, increase with back recline as well. Together, this research do es not support the current practice of using back recline as a prevention method for PIs in the buttocks . Instead, back recline increases normal and shear pressure on the buttocks, which have been shown to increase tissue stress and decrease perfusion, the reby increasing the risk of PI formation. From a PI prevention perspective, it was essential to identify material pairings that could reduce shear force on the buttocks and thighs while seated. In this study, the three material pairings with the smallest amount of shear on the buttocks were denim on vinyl, denim on two - layer nylon, and cotton - polyester blend on two - layer nylon. The denim and cotton - polyester blend on two - layer nylon were both material pairings with low coefficients of friction as found wit h the simulated buttocks and human participants, and they indicated that the two - layer nylon cover may have the potential to yield small shear forces on the buttocks and thighs while seated. The two - layer nylon cover decreased the shear force on the buttoc ks and thighs by about 20% relative to the vinyl cover for the denim and cotton - polyester blend pants. The lack of a statistically significant difference was due to large standard deviations of the shear force measurements. Nevertheless, that decrease in s hear force could reduce tissue stresses in the buttocks and thighs, increase perfusion in the skin in those regions, and help protect tissues in those regions from PI formation. To implement a shear reducing cover of any kind in a wheelchair, precautions must be taken to ensure that chair occupants do not slide into an unintended position or even out of their chair. Most manual and powered wheelchairs have foot plates mounted in front of the seat pan that provide some protection against the chair occupant sliding forward, and lateral pads and support straps can be implemented as well. The combination of shear reducing covers and wheelchair supports can reduce the risk of PIs and unintended repositioning at the same time. 119 This work furthers the understanding of shear forces on the buttocks and thighs while seated, thereby aiding in the task of reducing PI incidence. The results from the tilted surface method of determining coefficients of friction suggested that it can be used to consistently det ermine coefficients of friction and yields results similar to humans with deformable tissues. Shear force measurements from human participants sitting in reclined positions showed that recline increases shear on the buttocks and thighs and that the two - lay er nylon cover may be able to reduce shear . F urther testing will be needed to determine the coefficients of friction for other commonly worn fabrics (such as polyester) on seat pan covers and other types of material with which the two - layer cushion works w ell . However, these results show that the two - layer nylon cover is a promising avenue for shear force reduction, reducing PI risk in seated individuals. A limitation to this work is that the coefficients of friction were not found for pant materials on th e vinyl cover using human participant s. This was because participants contacted the back of the chair at the large tilt angles needed to overcome static friction with the vinyl cover. The participants touching the back of the chair instead of sliding indic ated that the tilting surface method of determining coefficients of friction may not be used for humans and materials with high coefficients of friction. While the coefficients of friction are not limited for methods using a sled on a tilting surface, expe riments that use human participants need to limit the magnitude of the coefficients of friction tested. Additionally, because one of the seat pan covers tested consisted of two layers of nylon attached on different sides of the seat pan, the potential exis ted for wrinkling or creasing. This attachment approach was used to accommodate for the load cell mounted beneath the seat pan, which would not permit a complete wrapping around the seat pan of the fabric. However, if a two - layer cover is implemented on an actual wheelchair cushion, the top layer of the cover could enclose the entire cushion and bottom layer of the cover. Doing so would create 120 tension in the top layer and simulate a conveyor belt, where the occupant could slide on the cushion cover without the potential of the cover wrinkling. 121 C ONCLUSIONS The overall objective of this work was to use novel methods to generate a description of the soft tissues of wheelchair users that may be used in finite element models and to develop wheelchair technology to address limitations in the practices used to prevent PIs. Additionally, this work investigated pant and seat pan cover material pairings to reduce shear forces on the buttocks while seated, an area where there is a dearth of research. The com ponents of this research address several assumptions used in engineering applications, especially in designs meant for persons with disabilities. Accurate representations of soft tissues are integral to designs of engineering models and devices with which people interact. Though people spend much of their day in a seated position, and many wheelchair users develop PIs, the protocols used to obtain m aterial properties from the buttocks and thighs of humans have mostly been limited to able - bodied people while lying down. Therefore, it was imperative to determine if the differences between the hip and knee joint angles in the seated position versus the lying position affected the reported material properties of the buttocks and thighs. Further, creating an accessible protocol to determine the material properties of the soft tissues in the buttocks and thighs in a seated or seated - like position was a chal lenge that was overcome. Successfully completing this component of the research allowed for comparisons of material properties between able - bodied individuals and wheelchair users. Several challenges arose during the path to completion of this work. The fi rst challenge was to determine whether hip and knee joint angles affected the material properties in the buttocks and thighs. This was done by creating a protocol to collect and analyze force and deformation data from the buttocks and thighs of able - bodied individuals in the seated, quadruped, and prone positions. The seated and quadruped positions produced similar buttock and 122 thigh tissue responses, while the prone position elicited stiffer tissue responses, indicating that the buttock and thighs tissues w ere affected by hip and knee joint flexions. This meant that it would be appropriate to use the tissue response in the quadruped position, but not the prone position, to represent the seated position. Confirming that the seated and quadruped positions pro duced similar material responses from the buttocks and thighs was critical to being able to obtain and calculate material properties from those tissues in wheelchair users. There were numerous obstacles to evaluating wheelchair users. They included issues related to balance and support while transferring into and then sitting in a laboratory test chair. These challenges were addressed by conducting tests in the quadruped position. Thus, the protocol to collect force and deformation data from people in the q uadruped position was adapted to test people with SCIs in a rehabilitation clinic. The tissue responses were obtained and compared to those of able - bodied people. People with SCIs had softer tissue responses in the buttocks and more proximal regions of the thighs, while their distal thigh regions were stiffer than those of able - bodied people. These findings had implications for both experimental and computational biomechanics. First, the differences in tissue stiffness throughout the buttocks and thighs of people with SCIs relative to able - bodied individuals showed that it is vital to differentiate between groups as differences are significant. Secondly, the differences in stiffnesses themselves were considered in conjunction with the effects of body positio n on tissue responses in the buttocks and thighs. Together, these studies indicated that the buttocks and proximal thigh tissues of wheelchair users in a seated position were softer than those of able - bodied people lying down. Since the soft tissues in the buttocks and thighs are prone to PI development, it is important to accurately represent them in all engineering applications that evaluate PI risk and 123 prevent PI occurrence. The comparisons made and the data sets yielded by these new methods will benefit wheelchair users through device designs and human body models. Equipment designed for the purpose of PI prevention has yet to eliminate these injuries, and PIs continue to be a significant problem for wheelchair users. Most commercial wheelchairs, if able to reposition the occupant, are limited to back recline and whole - body tilt. The evidence to support the use of back recline and whole - body tilt for PI prevention was limited, and spending time in those positions affects the tasks occupants can conduct. A s such, it was important to determine whether those repositioning strategies actually reduced PI risk, and if not, to develop new strategies to do so. Therefore, an accessible articulating chair was built to collect interface pressure data and perfusion da ta from the skin over the ischial tuberosities of seated individuals. The chair was first tested on able - bodied individuals and had independently rotating seat pan, pelvic, and thoracic supports, of which the pelvic and thoracic supports were mounted on an independently reclining back. Interface pressure data were segmented into body regions and regional averages found over key areas in the thighs, buttocks, lower back, middle back, and upper back. Segmentation allowed for specific characterizations of the effects of seated repositioning. After analysis, it was found that back recline increased pressure on the buttocks and lower back, two regions where PIs are common. Back recline also decreased perfusion in the skin over the ischial tuberosities. However, s eat pan tilt decreased pressure on the buttocks and lower back and increased perfusion in the skin over the ischial tuberosities. Back articulation shifted pressure concentrations between back regions. The results of this study suggest that using back recl ine as a method to reduce pressure on the buttocks, though popular, may actually have the opposite effect. Instead, new repositioning strategies, such as seat pan tilt or changes in back articulation, may be 124 effective ways to protect the buttocks and lower back. Studying these movements required a custom, laboratory chair because these features are not available via any commercial chair. As with the studies on the factors that affect material properties of the buttocks and thighs, it was necessary to determ ine whether the results of seated repositioning were the same for wheelchair users as they were for able - bodied individuals. Similarly to able - bodied individuals, back recline increased pressure on the buttocks and lower back, and back articulation was abl e to shift pressure concentrations between back regions. Seat pan tilt decreased pressure on the buttocks, with mixed results on the lower back. Seat pan tilt only increased perfusion in wheelchair users whose maximum buttocks pressures were similar to tho se of able - bodied individuals. Wheelchair users who had larger maximum buttocks pressure did not experience an increase in perfusion. These results indicated that the use of specialized cushions, or other pressure distributing strategies, may be necessary with seat pan tilt to increase perfusion in wheelchair users with high buttocks pressures. Though the effects of the chair articulations on the interface pressures of wheelchair users were similar to able - bodied individuals, the differences in the perfusio n trends between the two groups were not predictable based on the able - bodied data alone. The data collected from the wheelchair users, while they still suggest that seat pan tilt and back articulation are viable strategies for PI prevention, illustrate th e need to include people with disabilities. Lastly, a study was conducted on the shear forces experienced while seated on various seat pan covers. Though shear loads on soft tissue have been repeatedly implicated in the formation of PIs, and there has been research on how to reduce shear on individuals while lying down, there is relatively little research on how to reduce shear on seated individuals. Recent attempts to create a shear reducing seat pan cover had limited success. 125 This work aimed to further u nderstand how seat pan covers may be used to reduce shear forces while seated. To do this, the coefficients of friction of vinyl, one - layer nylon, and two - layer nylon seat pan covers were determined experimentally when paired with materials commonly used t o make pants. The shear forces experienced while reclining in the articulating chair were then quantified for each of the three seat pan covers. It was found that the coefficients of friction between the one - and two - layer nylon covers were consistently sm aller than those of the vinyl seat pan cover, regardless of the pant material with which they were paired. Further, the two - layer nylon seat pan cover showed trends towards reducing shear forces on the buttocks and thighs while seated. Regardless of seat p an cover, shear force on the buttocks and thighs increased at higher back recline angles, indicating that again, recline may not be a viable seated repositioning strategy to reduce PI risk. The methods used to 1) determine the coefficients of friction betw een two materials using the simulated buttocks and 2) measure the shear force on the buttocks at multiple back recline angles can be used to evaluate other material pairings that may reduce shear forces on the buttocks and thighs while seated. The results of this research yielded useful material pairings for clinicians and input parameters for modelers investigating the effects of shear force on soft tissue. This research has implications for clinical, experimental, and computational biomechanics. This work included novel methods and experimental approaches, with associated computations to obtain these findings. This research yielded first - of - its - kind continuous, in - vivo material response data from people with SCIs. The articulating chair that was developed included chair articulations not available in commercial chairs and also evaluated common seated repositioning strategies, with the resulting data suggesting that the current methods for PI prevention need re - evaluation. Finally, the experiments to determi ne the coefficients of friction for material pairings that 126 commonly occur between pants and seat pan covers showed that seat pan covers can significantly affect the coefficients of friction and that simulated buttocks can be representative of human partici pants. Overall, this work contributes significantly to our understanding of components related to human in - vivo material properties, device design and frictional properties all which play significant roles in the formation of PIs. 127 APPENDICES 128 A PPENDIX A : REPEATABILITY OF TISSUE INDENT ATION PROTOCOL Repeatability of the indentation protocol was investigated by considering multiple trials per individual. Since there were two trials per testing region per body position for each individual, the force and deflection data from the first trial were compared to the second. A sample of force and deflection data from all three regions in all three positions for one participant are provided in Figure A.1. The data indicate that the protocol has very good repeatability. Figure A.1 Force and deflection data from repeated indentation protocols on the proximal (left), middle (middle), and distal (right) thigh regions in all three body positions for one participant. Data from each region in each position are consistent, showing ver y good repeatability 129 A PPENDIX B : CONFIRMATION OF S YSTEM ACCURACY W ITH VINYL COVER ON THE SEAT PAN The load cells used in this research have been shown to accurately measure normal and shear forces on a rigid surface attached to the load cell itself. However, it was important to ensure that shear forces measured on a deformable surface with a cover were accurate because wheelchair cushions are deformable and have covers; and thus testing for this research occurred on a thin deformable foam laye r whose thickness was approximately 1 cm. To do this, shear force measurements were obtained for the sled on the foam layer with the vinyl cover for seat pan tilt angles up to 30°. The vinyl cover was chosen because it could support the shear forces on the sled at large tilt angles. The same markers used in the protocol to determine the coefficients of friction using the and force data were collected by the load cell beneath the seat pan. The sled was placed on the seat pan while it was covered by the vinyl cushion cover and had a tilt angle of 0° to record the weight of the sled. The seat pan was then tilted to 30° while the sled was on the seat pan. Force and motion capture data were collected at 60 Hz throughout the tilting movement. This was done with the sled weighing 45 Newtons because that was weight of the sled during the trials to determine the coefficients of friction, and 110 Newtons because that weight represented the average contact pressure of a seated person over the area of the sled bottom [5] . Theoretical and normal shear forces on the seat pan were calculated for the sled at seat pan tilt angles using the free body diagram from Figure 6.2. The normal force of the sled on the tilted seat pan was , where was the normal force on the sled, and was 130 the weight of the sled on the flat seat pan. The shear force of the sled on the tilted seat pan was . Experimental motion and force data collected on the seat pan were compared to the theoretical calculations. The angle of the seat pan tilt was determined using , a nd the normal and shear forces on the seat pan were plotted against the angle of seat pan tilt in Figure 6.8. The largest difference between the experimentally measured and theoretical values of shear force were approximately 1.5 Newtons for the 45 Newton sled and 3.5 Newtons for the 110 Newton sled, which were both within the sensitivity of the load cell beneath the seat pan. Figure B.1 Experimental and theoretical normal s hear forces o n the seat pan at tilt angles up to 30° for the 45 Newton sled (left) and 110 Newton sled (right) 131 A PPENDIX C : COEFFICIENTS OF FRICTION CALCULATED USING THE SLED AND HUMAN PARTICI PAN TS Table C.1 Static coefficients of friction for each pair of materials during the sled trials Cover Material Vinyl One - Layer Nylon Two - Layer Nylon Pants Material Denim (100% cotton) 0.61 ± .016 0.55 ± .048 0.27 ± .006 Cotton (60%) - polyester (40%) blend 0.58 ± .022 0.43 ± .016 0.27 ± .011 Table C.2 Kinetic coefficients of friction for each pair of materials during the sled trials Cover Material Vinyl One - Layer Nylon Two - Layer Nylon Pants Material Denim (100% cotton) 0.58 ± .020 0.52 ± .034 0.27 ± .008 Cotton (60%) - polyester (40%) blend 0.53 ± .008 0.39 ± .009 0.26 ± .008 Table C.3 Static coefficients of friction for each pair of materials during the human trials Cover Material Vinyl One - Layer Nylon Two - Layer Nylon Pants Material Denim (100% cotton) -- 0.50 ± .044 0.29 ± .059 Cotton (60%) - polyester (40%) blend -- 0.44 ± .067 0.30 ± .038 Table C.4 Kinetic coefficients of friction for each pair of materials during the human trials Cover Material Vinyl One - Layer Nylon Two - Layer Nylon Pants Material Denim (100% cotton) -- 0.48 ± .064 0.29 ± .059 Cotton (60%) - polyester (40%) blend -- 0.43 ± .068 0.29 ± .037 132 R EFERENCES 133 REFERENCES [1] [2] D. Ding et al. - in - space, recline, and elevation seating functions in natural Journal of Rehabilitation Research and Development , vol. 45, no. 7. pp. 973 984, 2008, doi: 10.1682/JRRD.2007.11.0178. [3] C. Lachenbruch, Y. T Contributions of Interface Presssure, Shear Stress, and Temperature on Ischmic - induced, Skin - reactive Hyperemia in Healthy Volunteers: A Repeated Measures Laboratory Ostomy Wound Mana g. , vol. 61, no. 2, pp. 16 25, 2015. [4] inclination and the shearing force of the buttocks while seated in a wheelchair: Preliminary research in non - disabled individuals Assist. Technol. , pp. 1 7, 2018, doi: 10.1080/10400435.2018.1547333. [5] P. V. B. Mendes, L. C. C. Gradim, N. S. Silva, A. L. C. Allegretti, D. C. D. M. Carrijo, subj Disabil. Rehabil. Assist. Technol. , vol. 14, no. 6, pp. 555 560, 2019, doi: 10.1080/17483107.2018.1463399. [6] development of pressure sores in s Paraplegia , vol. 29, no. 4, pp. 261 267, 1991, doi: 10.1038/sc.1991.37. [7] L. J. Cowan et al. Spinal Cord Injury in Long - Adv. Ski. Wound Care , vol . 32, no. 3, pp. 122 130, 2019, doi: 10.1097/01.ASW.0000553109.70752.bf. [8] H. Brem et al. Am. J. Surg. , vol. 200, no. 4, pp. 473 477, 2010, doi: 10.1016/j.amjsurg.2009.12.021. [9] C. VanGilder, C. Lachenbruch, C. Algrim - pressure ulcer PrevalenceTMSurvey: 2006 - 2015: A 10 - year pressure injury prevalence J. Wound, Ostomy Cont. Nurs. , vol. 44, no. 1, pp. 20 28, 2017, doi: 10.1097/ WON.0000000000000292. [10] Biomech. Model. Mechanobiol. , no. 0123456789, 2019, doi: 10.1007/s10237 - 019 - 01212 - 7. [11] E. Jaul and R. Calderon - Int. Wound J. , vol. 12, no. 3, pp. 254 259, 2015, doi: 10.1111/iwj.12086. [12] Int. 134 Surg. J. , vol. 7, no. 4, p. 1229, 2020, doi: 10.18203/2349 - 2902.isj20201402. [13] - of - illness studies in chronic ulcers: a sys J. Wound Care , vol. 26, no. sup4, pp. S4 S14, 2017, doi: 10.12968/jowc.2017.26.Sup4.S4. [14] S. Brewer et al. - containing nutritional supplement on pressure J. Wound Care , vol. 19, no. 7, pp. 311 316, 2010, doi: 10.12968/jowc.2010.19.7.48905. [15] - acquired pressure Int. Wound J. , vol. 16, no. 3, pp. 634 640, 2019, doi: 10.1111/ iwj.13071. [16] S. D. Horn et al. - J. Am. Geriatr. Soc. , vol. 50, no. 11, pp. 1816 1825, 2002, doi: 10.1046/j.1532 - 5415.2002.50510.x. [17] J. H. M. Verschueren, M. W. M. Post, S. De Groot, L. H. V. Van Der Woude, F. W. A. in - Spinal Cord , vol. 49, no. 1, pp. 106 112, 2011, doi: 10.1038/sc.2010.66. [18] R. M. A. Al - D Ann. Biomed. Eng. , vol. 44, no. 9, pp. 1 12, 2016, doi: 10.1007/s10439 - 016 - 1560 - 3. [19] A. Macro n et al. Clin. Biomech. , vol. 71, no. October 2019, pp. 92 100, 2020, doi: 10.1016/j.clinbiomech.2019.10. 005. [20] the internal strains in the sacral region of bed ridden patients? - A contribution to pressure Clin. Biomech. , vol. 35, pp. 7 13, 2016, doi : 10.1016/j.clinbiomech.2016.03.009. [21] D. Brienza et al. J. Am. Geriatr. Soc. , vol. 58, no. 12, pp. 2308 2314, 2010, doi: 10.1111/j.1532 - 5415.2010.03168.x. [22] J. Biomech. , vol. 45, no. 7, pp. 1252 1258, 2012, doi: 10.1016/j.jbiomech.2012.01.037. [23] J. - ation of hyperelastic properties of passive thigh muscle under compression with an inverse method from a displacement J. Biomech. , vol. 48, pp. 4081 4086, 2015, doi: 10.1016/j.jbiomech.2015.10.007. 135 [24] L. E. Edsberg, J. M. Black, M. Go ldberg, L. McNichol, L. Moore, and M. Sieggreen, J. Wound, Ostomy Cont. Nurs. , vol. 43, no. 6, pp. 585 597, 2016, doi: 10.1097/WON.0000000000000281. [25] J. Black et al. 274, 2007. [26] Br. J. Nurs. , vol. 25, no. 15, pp. 840 843, 2016. [27] Arch. Phys. Med. Rehabil. , vol. 84, no. 4, pp. 616 619, 2003, doi: 10.1053/apmr.2003.50038. [28] M. Zhang, A. R. Turner - Proc. Inst. Mech. Eng. Part H J. Eng. Med. , vol. 208, no. 4, pp. 217 222, 1994, doi: 10.1243/PIME_PROC_1994_208_ 291_02. [29] J. Am. Acad. Dermatol. , vol. 6, no. 6, pp. 1014 1021, 1982, doi: 10.1016/S0190 - 9622(82)70085 - 4. [30] Z. Moore et al. s: A cross - sectional J. Wound Care , vol. 24, no. 8, pp. 333 339, 2015, doi: 10.12968/jowc.2015.24.8.333. [31] pressure ulce Hosp. Pract. (1995) , vol. 46, no. 1, pp. 30 36, 2018, doi: 10.1080/21548331.2018.1418139. [32] E. Park - Sta NCHS Data Brief , no. 14, pp. 1 8, 2009. [33] J. Am. Acad. Dermatol. , vol. 81, no. 4, pp. 881 890, 2019, doi: 10.1016/j.jaad.2018 .12.069. [34] hospital - acquired pressure ulcers - a population - Int. Wound J. , vol. 13, no. 5, pp. 809 820, 2016, doi: 10.1111/iwj.12386. [35] J. S. Arch. Phys. Med. Rehabil. , vol. 85, no. 8, pp. 1257 1264, 2004, doi: 10.1016/j.apmr.2003. 08.108. [36] - related pressure ulcers: A systematic review and meta - Int. J. Nurs. Stud. , vol. 92, pp. 109 120, 136 2019, doi: 10.1016/j.ijnurstu.2019.02.006. [37] patient - related risk factors in hospitals in the province of Erzurum: A cross - sectional J. Tissue Viability , vol. 27, no. 3, pp. 135 140, 2018, doi: 10.1016/j. jtv.2018.05.001. [38] J. Wound Care , vol. 27, no. 2, 2018. [39] A. Tubaishat, P. Papanikolaou, D. Ant Prevalence in the Acute Care Setting: A Systematic Review, 2000 - Clin. Nurs. Res. , vol. 27, no. 6, pp. 643 659, 2018, doi: 10.1177/1054773817705541. [40] M. W. van Leen, J. M. Schols, S. E. Hovius, and R. J. Longitudinal Prevalence Data to Determine the Use of Pressure Ulcer Preventive Measures in Dutch Nursing Homes, 2005 Ostomy Wound Manag. , vol. 63, no. 09, pp. 10 20, 2017, doi: 10.25270/owm.2017.09.1020. [41] M. I. González - Méndez, M. Lima - Serrano, C. Martín - Castaño, I. Alonso - Araujo, and J. S. Lima - J. Clin. Nurs. , vol. 27, no. 5 6, pp. 1028 1037, 201 8, doi: 10.1111/jocn.14091. [42] P. García - Molina, E. Balaguer - López, F. P. García - Fernández, M. de los Á. Ferrera - and risk factors in neonatal intensive care and inte Int. Wound J. , vol. 15, no. 4, pp. 571 579, 2018, doi: 10.1111/iwj.12900. [43] M. Artico et al. Palliat . Med. , vol. 32, no. 1, pp. 299 307, 2018, doi: 10.1177/0269216317737671. [44] J. Tissue Via bility , vol. 28, no. 2, pp. 70 74, 2019, doi: 10.1016/j.jtv.2019.02.001. [45] J. Pediatr. Neonatal Care , vol. 8, no. 5, 2018, doi: 10.15406/jpnc.2018.08.00350. [46] - term care: J. Wound Care , vol. 28, no. 11, pp. 702 709, 2019, doi: 10.12968/jowc.2019.28.11.702. [47] X. Deng, T. Yu, and A. - acquired pressure ulcers in Crit. Care Nurse , vol. 37, no. 4, pp. e1 e11, 2017, doi: 10.4037/ccn2017548. 137 [48] four Indonesian hospitals: prevalence, patient characteristics, ulcer characteristics, prevention Int. Wound J. , vol. 14, no. 1, pp. 184 193, 2017, doi: 10.1111/iwj.12580. [49] cal complications and risk factors for pressure ulcers after traumatic spinal cord injury during acute care in South Spinal Cord , vol. 54, no. 7, pp. 535 539, 2016, doi: 10.1038/sc.2015.189. [50] D. Becker et al. Incidence and clinical and Intensive Crit. Care Nurs. , vol. 42, pp. 55 61, 2017, doi: 10.1016/j.iccn.2017.03.009. [51] F. Coyer et al. non - intensive care patients: A state - Aust. Crit. Care , vol. 30, no. 5, pp. 244 250, 2017, doi: 10.1016/j.aucc.2016.12.003. [52] A. C. Kwok, A. M. Simpson, J. Willcockson, D. P. Donato, I. A. Goodwin, and J. P. their associations following the surgical repair of pressure Am. J. Surg. , vol. 216, no. 6, pp. 1177 1181, 2018, doi: 10.1016/j.amjsurg.2018.01.012. [53] Factor AORN J. , vol. 103, no. 3, pp. 271 281, 2016, doi: 10.1016/j.aorn.2016.01.008. [54] W. H. Donovan, R. Edward Carter, G. M. Bedbrook, J. S. Young, and E. R. Griffiths, nal cord injury: Patients in specialised, compared with non - Paraplegia , vol. 22, no. 5, pp. 282 290, 1984, doi: 10.1038/sc.1984.46. [55] 2009 International Pre ssure Ulcer Prevalence TM Survey and a3 - Year, Acute Care, Unit - Specific Ostomy Wound Manag. , no. November, pp. 39 46, 2009, doi: 10.1016/j.scijus.2009.06.001. [56] J. Spinal Cord Med. , vol. 34, no. 6, pp. 620 1, 2017, doi: 10.1179/204577212X13237783484262. [57] J. Spinal Cord Med. , vol. 37, no. 3, pp. 355 6, 2014. [58] N. B. Jain et al. - JAMA - J. Am. Med. Assoc. , vol. 313, no. 22, pp. 2236 2243, 2015, doi: 10.1001/jama.2015.6250. [59] alence in people with spinal cord injury: Age - period - Arch. Phys. Med. Rehabil. , vol. 86, no. 6, pp. 1208 1213, 2005, doi: 10.1016/j.apmr.2004.12.023. [60] - term medica l 138 Arch. Phys. Med. Rehabil. , vol. 80, no. 11, pp. 1402 1410, 1999, doi: 10.1016/S0003 - 9993(99)90251 - 4. [61] M. Hubli et al. Spinal Cord , 2020, doi: 10.1038/s41393 - 020 - 0522 - 7. [62] between pressure ulcer incidence and buttock - seat cushion interface pressure in at - risk Arch. Phys. Med. Rehabil. , vol. 82, no. 4, pp. 529 533, 2001, doi: 10.1053/apmr.2001.21854. [63] lts of Nine International Pressure Ostomy Wound Manag. , vol. 54, no. 2, 2008. [64] Surgery - related Hospital - Med. Care , vol. 54, no. 9, pp. 845 851, 2016, doi: 10.1097/MLR.0000000000000558. [65] Ann. Intern. Med. , vol. 123, no. 6, pp. 433 442, 1995, doi: 10.7326/0003 - 4819 - 123 - 6 - 199509150 - 00008. [66] J. Eval . Clin. Pract. , vol. 13, no. 2, pp. 227 235, 2007, doi: 10.1111/j.1365 - 2753.2006.00684.x. [67] Arch. Ph ys. Med. Rehabil. , vol. 84, no. 10, pp. 1466 1472, 2003, doi: 10.1016/S0003 - 9993(03)00269 - 7. [68] angles of wheelchair tilt on interface pressure among individuals with spi Spinal Cord , vol. 49, no. 7, pp. 827 831, 2011, doi: 10.1038/sc.2010.194. [69] M. Makhsous et al. J. Spinal Cord Med. , vol. 30, no. 5, pp. 497 507, 2007, doi: 10.1080/10790268.2007.11754584. [70] spinal cord - injured patients following a structured treatment programme: a 10 - year follow - Int. Wound J. , vol. 14, no. 2, pp. 355 359, 2017, doi: 10.1111/iwj.12609. [71] Int. Wound J. , vol. 11, no. 6, pp. 696 700, 2014, doi: 10.1111/iwj.12033. [72] K. Ozer, O. Colak, F. B. Goktas, N. Su Int. Wound J. , vol. 13, no. 2, pp. 287 288, 2016, doi: 10.1111/iwj.12257. [73] 139 by Ir. J. Med. Sci. , vol. 180, no. 2, pp. 601 602, 2011, doi: 10.1007/s11845 - 009 - 0384 - 7. [74] - the bilateral popliteal fossa: A case Oman Med. J. , vol. 27, no. 3, 2012, doi: 10.5001/omj.2012.65. [75] M. Makhsous, F. Lin, J. Bankard, R. W. Hendrix, M. Hepler, and J. Press, occupational low back pain BMC Musculoskelet. Disord. , vol. 10, pp. 1 11, 2009, doi: 10.1186/1471 - 2474 - 10 - 17. [76] spinal cord injury on skele tal muscle cross - sectional area within the first 6 months of Eur. J. Appl. Physiol. Occup. Physiol. , vol. 80, no. 4, pp. 373 378, 1999, doi: 10.1007/s004210050606. [77] P. K. Shah et al. - Extremity Muscle Cross - Sectional Area After Incomple te Spinal Arch. Phys. Med. Rehabil. , vol. 87, no. 6, pp. 772 778, 2006, doi: 10.1016/j.apmr.2006.02.028. [78] able - bodied and SCI individuals - J. Tissue Viability , vol. 22, no. 3, pp. 74 82, 2013, doi: 10.1016/j.jtv.2013.03.003. [79] C. D. M oore et al. - extremity muscle atrophy and fat infiltration after chronic J. Musculoskelet. Neuronal Interact. , vol. 15, no. 1, pp. 32 41, 2015. [80] ular fat Spinal Cord , vol. 45, no. 4, pp. 304 309, 2007, doi: 10.1038/sj.sc.3101968. [81] Neurologic Diseases: A Lexical o Pathophsiologica Eur. J. Phys. Rehabil. Med. , vol. 49, 2013. [82] E. Linder - Ganz, N. Shabshin, Y. Itzchak, Z. Yizhar, I. Siev - and stresses in sub - dermal tissues of the buttocks are greater in paraplegics than in healthy during si J. Biomech. , vol. 41, no. 3, pp. 567 580, 2008, doi: 10.1016/j.jbiomech.2007.10.011. [83] effects of user anatomy and wheelchair cushion type on tissue deformatio J. Tissue Viability , vol. 27, no. 1, pp. 42 53, 2018, doi: 10.1016/j.jtv.2017.04.001. [84] Neuroradiology , vol. 17, no. 3, pp. 127 136 , 1979. [85] C. M. Nascimento, M. Ingles, A. Salvador - Pascual, M. R. Cominetti, M. C. Gomez - Free Radic. 140 Biol. Med. , vol. 132, no. August 2018, pp. 42 49, 2019, doi: 10.1016/j.fre eradbiomed.2018.08.035. [86] J. Appl. Physiol. , vol. 95, no. 4, pp. 1717 27, 2003, doi: 10.1152/japplphysiol.00347.2003. [87] Whole 2016, doi: 10.1155/2016/6206959. [88] skeletal muscle with age in sedentar y males, age 22 Acta Physiol. Scand. , vol. 103, no. 1, pp. 31 39, 1978, doi: 10.1111/j.1748 - 1716.1978.tb06187.x. [89] M. T. Korhonen et al. - trained J. Appl. Physiol. , vol. 101, no. 3, pp. 906 917, 2006, doi: 10.1152/japplphysiol.00299.2006. [90] of evidenced - J. Burn C are Rehabil. , vol. 25, no. 5, pp. 388 410, 2004, doi: 10.1097/01.BCR.0000138289.83335.F4. [91] Nurs. Res. , vol. 36, no. 4, 1987. [92] B. G. (VU U. Raijmakers, M. G. (Amsterdam R. R. C. - R. Nieuwenhuizen, H. (VU U. Beckerman, and S. A. R. R. C. - R. Am. J. Phys. Med. Rehabil. , vol. 94, no. 2, pp. 101 113, 2014, doi: 10.1097/PHM.0000000000000206. [93] Pain , vol. 70, no. 1, pp. 3 14, 1997, doi: 10.1016/S0304 - 3959(96)03266 - 6. [94] Spinal Cord , vol. 39, no. 2, pp. 63 73, 2001, doi: 10.1038/sj.sc.3101116. [95] R. J. Hart J. Aging Health , vol. 10, no. 1, pp. 99 116, 1998. [96] I. K. Crombie et al. Age Ageing , vol. 33, no. 3, pp. 287 292, 2004, doi: 10.1093/ageing/afh089. [97] K. A. Martin Ginis et al. l Activity in a Population - Based Sample of People With Spinal Cord Injury Part I: Demographic and Injury - Arch. Phys. Med. Rehabil. , vol. 91, no. 5, pp. 722 728, 2010, doi: 10.1016/j.apmr.2009.12.027. [98] J. Am. Med. 141 Assoc. , vol. 166, no. 7, pp. 762 763, 1958, doi: 10.1001/jama.1958.62990070004010a. [99] review of risks Intensive Care Med. , vol. 28, pp. 1379 1388, 2002, doi: 10.1007/s00134 - 002 - 1487 - z. [100] transcutaneous oxygen level characterizations in human skin with changes in normal and shear loads - Clin. Biomech. , vol. 25, no. 8, pp. 823 828, 2010, doi: 10.1016/j.clinbiomech.2010.06.003. [101] external shear loads on arterial and venous blood flow: Implications for pressure ulcer Clin. Biomech. , vol. 28, no. 5, pp. 574 578, 2013, doi: 10.1016/j.clinbiomech.2013.04.001. [102] perfusion responses under normal and combined loadings: Comparisons between legs with venous stasis ulcers and Clin. Biomech. , vol. 30, no. 10, pp. 1218 1224, 2015, doi: 10.1016/j.clinbiomech.2015.08.001. [103] Archives of physical medicine and rehabilitation , vol. 42. pp. 19 29, 1961. [104] Arch. Phys. Med. Rehabil. , vol. 55, no. 4, pp. 147 152, 1974. [105] experimental study of some pressure effects on tissues, with reference to the bed - J. Pathol. Bacteriol. , vol. 66, no. 2, pp. 347 358, 1953, doi: 10.1002/path.1700660203. [106] E. Linder - ssure - time cell death J. Biomech. , vol. 39, no. 14, pp. 2725 2732, 2006, doi: 10.1016/j.jbiomech.2005.08.010. [107] E. M. H. Bosboom, C. V. C. Bouten, C. W. J. Oomens, H. W. M. Van Straaten, F. P. T. Med. Eng. Phys. , vol. 23, no. 3, pp. 195 200, 2001, doi: 10 .1016/S1350 - 4533(01)00034 - 0. [108] - seat J. Rehabil. Res. Dev. , vol. 29, no. 4, p. 21, 1992, doi: 10.1682/JRRD.1992.10.0021. [109] J. Berthold, B. E. Dicianno, and R pressure distributions and the potential risk for skin ulceration in a population of sledge Disabil. Rehabil. Assist. Technol. , vol. 8, no. 5, pp. 387 391, 2013, doi: 10. 3109/17483107.2013.769123. [110] 142 J. Biomech. , vol. 46, no. 2, pp. 240 247, 2013, doi: 10.1016/j.jbiomech.2012.11.006. [ 111] Different Types of Wheelchair Cushions for Pressure Ulcers in View of the Experimental Proc. LASTED Int. Conf. , pp. 164 167, 2017, doi: 10.2316/p.2017.852 - 043. [112] S. D. Darrah, B. E. Dicianno, J. Berthold, A. McCoy, M. Haas, and R. A. Cooper, Disabil. Rehabil. Assist. Technol. , vol. 11, no . 3, pp. 241 246, 2016, doi: 10.3109/17483107.2014.921939. [113] temperature on the in - Biotribology , vol. 6, pp. 21 28, 2016, doi: 10.1016/j.biotri.2016.03.003. [114] J. R. Soc. Interface , vol. 5, no. 28, pp. 1317 1328, 2008, doi: 10.1098/rsif.2008.0034. [115] Int. Wound J. , vol. 15, no. 6, pp. 866 874, 2018, doi: 10.111 1/iwj.12937. [116] Clin. Biomech. , vol. 59, no. September, pp. 62 70, 2018, doi: 10.1016/j.clinbiomech.2018.09.010. [ 117] G. J. Hodges, M. M. Mallette, A. Rigby, P. Klentrou, S. S. Cheung, and B. Falk, J. Tissue Viability , vol. 28, no . 3, pp. 144 151, 2019, doi: 10.1016/j.jtv.2019.04.003. [118] C. M. Olney et al. - based wheelchair seating systems J. Tissue Viability , vol. 27, no. 3, pp. 181 187, 2018, doi: 10.1016/j.jtv.2018.06.001. [119] E. C. Herrman, C. F. Knapp, J. C. Donofr surface pressure - J. Rehabil. Res. Dev. , vol. 36, no. 2, pp. 109 120, 1999, doi: 10.1086/250095. [120] E. Linder - cts of pressure and shear on capillary closure in Ann. Biomed. Eng. , vol. 35, no. 12, pp. 2095 2107, 2007, doi: 10.1007/s10439 - 007 - 9384 - 9. [121] duced Deep Tissue Ann. Biomed. Eng. , vol. 43, no. 2, pp. 297 305, 2015, doi: 10.1007/s10439 - 014 - 1202 - 6. [122] 143 pressure and shear on tissue viability of h uman skin in relation to the development of J. Tissue Viability , vol. 26, no. 3, pp. 157 171, 2017, doi: 10.1016/j.jtv.2017.04.003. [123] P. Kaewprag, C. Newton, B. Vermillion, S. Hyun, K. Huang, and R. Machiraju, edictive models for pressure ulcers from intensive care unit electronic health records BMC Med. Inform. Decis. Mak. , vol. 17, no. Suppl 2, 2017, doi: 10.1186/s12911 - 017 - 0471 - z. [124] ear forces externally applied to skin J. Biomed. Eng. , vol. 15, no. 6, pp. 451 456, 1993, doi: 10.1016/0141 - 5425(93)90057 - 6. [125] pres J. Phys. Ther. Sci. , vol. 28, no. 10, pp. 2830 2833, 2016, doi: 10.1589/jpts.28.2830. [126] anthropometric d Appl. Ergon. , vol. 73, no. December 2017, pp. 13 21, 2018, doi: 10.1016/j.apergo.2018.05.015. [127] Surg. (United Kingdom) , vol. 35, no. 9, pp. 505 510, 2017, doi: 10.1016/j.mpsur.2017.06.008. [128] Indian J. Plast. Surg. , vol. 48, no. 1, pp. 4 16, 2015, doi: 10.4103/0970. [ 129] J. Am. Acad. Dermatol. , vol. 81, no. 4, pp. 893 902, 2019, doi: 10.1016/j.jaad.2018.12.068. [130] J. F. Graumlich et al. loid: A J. Am. Geriatr. Soc. , vol. 51, no. 2, pp. 147 154, 2003, doi: 10.1046/j.1532 - 5415.2003.51051.x. [131] Br. J. Nurs. , vol. 25, no. 8, pp. 443 448, 2016, doi: 10.12968/bjon.2016.25.8.442. [132] Cochrane Database Syst. Rev. , vol. 2017, no. 10, 2017, doi: 10.1002/14651858.CD011332.pub2. [133] Spinal Cord , vol. 56, no. 3, pp. 186 198, 2018, doi: 10.1038/s41393 - 017 - 0054 - y. [134] the use of nutritional supplements containing collagen peptides to facilitate the healing of 144 J. Nutr. Intermed. Metab. , vol. 8, pp. 51 59 , 2017, doi: 10.1016/j.jnim.2017.05.001. [135] B. Smith et al. Ann. Intern. Med. , vol. 159, no. 1, 2013. [136] W. V Padula et al. er Care With Prevention: A Cost - Med. Care , vol. 49, no. 4, pp. 385 392, 2011. [137] hospital - acquired pressure ulcers prevention: a systematic r Int. Wound J. , vol. 16, no. 5, pp. 1087 1102, 2019, doi: 10.1111/iwj.13147. [138] pressure ulcers in a hospital setting: A pragmatic randomised controlled trial, Int. J. Nurs. Stud. , vol. 72, no. March, pp. 53 59, 2017, doi: 10.1016/j.ijnurstu.2017.04.007. [139] Community Wound Care , no. March, pp. 155 165, 2017, doi: 10.1002/9781118281796.ch12. [140] S. Mäki - Turja - Rostedt, M. Stolt, H. Leino - interventions for pressure ulcers in long - term older people care facilities: A systematic J. Clin. Nurs. , vol. 28, no. 13 14, pp. 2420 2442, 2019, doi: 10.1111/jocn.14767. [141 ] Br. J. Nurs. , vol. 27, no. 18, pp. 29 31, 2018. [142] Adv. Wound Care , vol. 7, no. 2, pp. 57 67, 2018, doi: 10.1089/wound.2016.0697. [143] Ann. Intern. Med. , no. 94, pp. 661 666, 1981. [144] V. Lopez et al. among patients in the acute aged care and rehabilitation units in the Australian Capital Territory: A best practice Int. J. Evid. Based. Healthc. , vol. 9, no. 4, pp. 429 434, 2011, doi: 10.1111/j.1744 - 1609.2011.00234.x. [145] J. Tissue Viability , vol. 24, no. 1, pp. 1 11, 2015, doi: 10.1016/j.jtv.2015.01.001. [146] valence of pressure ulcers between the Netherlands and Germany - Associations between risk, prevention and J. Clin. Nurs. , vol. 17, no. 9, pp. 1237 1244, 2008, doi: 10.1111/j.1365 - 2702.2007.022 25.x. [147] R. F. Edlich et al. 304, 145 2004. [148] Provide Long - term Protection Against Pressure - Ulcers for Indivi duals with a Spinal Cord Adv. Ski. Wound Care , vol. 28, no. 7, pp. 303 316, 2015, doi: 10.1097/01.ASW.0000465300.99194.27. [149] prevention of p J. Tissue Viability , vol. 17, no. 1, pp. 15 21, 2008, doi: 10.1016/j.jtv.2007.09.011. [150] J. Tissue Viability , vol. 27, no. 1, pp. 2 9, 2018, doi: 10.1016/j.jtv.2017.03.001. [151] S. P. Burns and K. Arch. Phys. Med. Rehabil. , vol. 80, pp. 566 571, 1999. [152] interface pressure distribution: A J. Phys. Ther. Sci. , vol. 28, no. 1, pp. 227 230, 2016, doi: 10.1589/jpts.28.227. [153] - Style Off - Loading Wheelchair Seat Cushion Reduces Interface Pressure Under Ischial Tuberosities and Sacroco ccygeal Arch. Phys. Med. Rehabil. , vol. 97, no. 11, pp. 1872 1879, 2016, doi: 10.1016/j.apmr.2016.04.004. [154] real - time pressure adjusting seat cushion to pr Proc. ASME Des. Eng. Tech. Conf. , vol. 6, pp. 1 7, 2018, doi: 10.1115/DETC2018 - 86210. [155] W. Carrigan et al. and modulating seat cushion for pressure ulcer preve Med. Eng. Phys. , vol. 69, pp. 17 27, 2019, doi: 10.1016/j.medengphy.2019.06.006. [156] Med. (Unite d States) , vol. 98, no. 4, 2019, doi: 10.1097/MD.0000000000014231. [157] - dimensional J. Text. Inst. , vol. 108, no. 11, pp . 1940 1948, 2017, doi: 10.1080/00405000.2017.1301019. [158] calibration by users with spinal cord injury and by occupational therapists, using a pressure mapping system. Diffe rences in cushion calibration by the user and occupational Eur. Int. J. Sci. Technol. , vol. 5, no. 6, pp. 77 85, 2016. [159] Am. J. Occup . Ther. , vol. 55, no. 4, pp. 470 475, 2001, doi: 146 10.5014/ajot.55.4.470. [160] Bull. Prosthet. Res. , pp. 129 143, 1971. [161] R. Ze - in - Space and Recline Functions: Influence on Sitting Interface Pressure and Ischial Blood Biomed Res. Int. , vol. 2019, 2019, doi: 10.1155/201 9/4027976. [162] J. Biomech. Eng. , vol. 129, no. 1, pp. 58 65, 2007, doi: 10.1115/1.2401184. [163] of the effect of trunk tilt on compressive soft tissue deformations under the ischial tuberosities using weight - Clin. Biomech. , vol. 25, no. 5, pp. 402 408, 2010, doi: 10.1016/j.clinbiomech.2010.01.019. [164] specialist seating assessment clinic: Changing pressure Spinal Cord , vol. 41, no. 12, pp. 692 695, 2003, doi: 10.1038/sj.sc.3101527. [165] Ergonomics , vol. 44, no . 10, pp. 901 921, 2001, doi: 10.1080/00140130110064685. [166] S. Horn et al. - Term Care Study: Pressure Ulcer Development in Long - J. Am. Geriatr. Soc. , vol. 52, no. 3, pp. 359 367, 2004, doi: 10.1111 /j.1532 - 5415.2004.52106.x. [167] X. Wang, M. Cardosso, and I. Theodora - Hum. Simul. Virtual Environ. Work With Comput. Syst. (WWCS), Process Control , no. June, pp. 743 753, 2019, doi: 10.1007/97 8 - 3 - 319 - 96077 - 7. [168] Rehabilitation: Assistive Technology An investigation into the effectiveness of a novel wheelchair seat - cover assembly for the reduction of forces exerted onto the but Disabil. Rehabil. Assist. Technol. , pp. 1 6, 2020, doi: 10.1080/17483107.2020.1780484. [169] the impact of cushion design on buttocks tissue deformation: An MRI J. Tissue Viability , vol. 27, no. 3, pp. 162 172, 2018, doi: 10.1016/j.jtv.2018.04.001. [170] D. A. Gayol - for wheelchair patients who are prone to develop pressure injuri Technol. Heal. Care , vol. 25, no. 4, pp. 749 760, 2017, doi: 10.3233/THC - 160803. [171] Ulcers: Biomechanical Modelization and Simulation of Human Seat Cushion Cont Proc. Int. Conf. Adv. Comput. Mech. , no. February, pp. 1271 1280, 2017, doi: 10.1007/978 - 981 - 10 - 7149 - 2. 147 [172] - cell - based cushion for pressure ulcer protection remarkably reduces tissue stresses in the seated buttocks with respect to foams: J. Tissue Viability , vol. 23, no. 1, pp. 13 23, 2014, doi: 10.1016/j.jtv.2013.12.005. [173] - loading wheelchair cushion versus air - cell - Int. Wound J. , vol. 14, no. 6, pp. 1327 1334, 201 7, doi: 10.1111/iwj.12807. [174] SAE Tech. Pap. Ser. , no. 412, pp. 135 142, 1997, doi: 10.4271/970593. [175] M. M. V erver, J. van Hoof, C. W. J. Oomens, J. S. H. M. Wismans, and F. P. T. Baaijens, Comput. Methods Biomech. Biomed. Engin. , vol. 7, no. 4, pp. 193 203, 2004, doi: 1 0.1080/10255840410001727832. [176] V. Luboz et al. - time pressure ulcer prevention in sitting J. Tissue Viability , vol. 27, no. 1, pp. 54 58, 2018, doi: 10.1016/j.jtv.2017.06.002. [177] W. A. Traa et al. d 3D finite element modelling to investigate deep tissue Comput. Methods Biomech. Biomed. Engin. , vol. 21, no. 14, pp. 760 769, 2018, doi: 10.1080/10255842.2018.1517868. [178] K. M. Moerman et al. and subject - specific finite element analysis for evaluation of in - Comput. Methods Biomech. Biomed. Engin. , vol. 20, no. 5, pp. 483 491, 2017, doi: 10.1080/10255842.2016.1250259. [179] NITE ELEMENT PREDICTION of SUB - DERMAL TISSUE STRESSES of the BUTTOCKS during WHEELCHAIR J. Mech. Med. Biol. , vol. 16, no. 4, pp. 1 15, 2016, doi: 10.1142/S0219519416500585. [180] he thigh - buttock Comput. Methods Biomech. Biomed. Engin. , vol. 5842, pp. 1 10, 2018, doi: 10.1080/10255842.2018.1466117. [181] A. Macron, H. Pillet, J. Doridam, evaluation of a new methodology for the fast generation of patient - specific Finite Element models of the buttock for sitting - J. Biomech. , vol. 79, pp. 173 180, 2018, doi: 10.1016/j.jbiomech.2018.08.001. [182] J. Tissue Viability , vol. 29, no. 2, pp. 69 75, 2020, doi: 10.1016/j.jtv.2020.01.004. 148 [18 3] model to demonstrate the effects of friction and pressure on skin in relation to pressure Wear , vol. 376 377, pp. 266 271, 2017, doi: 10.1016/j.wear.201 6.11.026. [184] analysis to model ischemia experienced in the development of device related pressure Proc. Inst. Mech. Eng. Part H J. Eng. Med. , vol. 233, no. 7 , pp. 745 753, 2019, doi: 10.1177/0954411919851387. [185] occlusion of a horizontally running microvessel in response to skin surface pressure: a finite element study of sacr Med. Biol. Eng. Comput. , vol. 55, no. 1, pp. 79 88, 2017, doi: 10.1007/s11517 - 016 - 1500 - 2. [186] J. Biomech. Eng. , vol. 100, no. 2, pp. 79 89, 1978. [187] - dimensional computer model of the human buttocks, J. R ehabil. Res. Dev. , vol. 31, no. 2, pp. 111 119, 1994. [188] - interface pressures on Arch. Phys. Med. Rehabil. , vol. 8 3, no. 6, pp. 872 875, 2002, doi: 10.1053/apmr.2002.32677. [189] E. Linder - Ganz, G. Yarnitzky, Z. Yizhar, I. Siev - - time finite element monitoring of sub - dermal tissue stresses in individuals with spinal cord injury: Toward preventio Ann. Biomed. Eng. , vol. 37, no. 2, pp. 387 400, 2009, doi: 10.1007/s10439 - 008 - 9607 - 8. [190] - Proc. IEEE , vol. 86, no. 3, 1998. [191] P. A. L. S. Martins, R. Several Material Models for Prediction of Hyperelastic Properties: Application to Silicone - Strain , vol. 42, no. 3, pp. 135 147, 2006, doi: doi:10.1111/j.1475 - 1305.2006.0 0257.x. [192] IEEE Trans. Neural Syst. Rehabil. Eng. , vol. 15, no. 4, pp. 517 525, 2007, doi: 10.1109/TNSRE.2007.906967. [193] Q. Sun et al. - Thigh of a Seated Individual Elucidated by a RESNA 28th Annu. Conf. , pp. 1 7, 2005. [194] M. Grujicic, B. Pandurangan, G. Arakere, W. C. Bell, T. He, a - cushion and soft - tissue material modeling and a finite element investigation of the seating comfort for passenger - Mater. Des. , vol. 30, no. 10, pp. 4273 4285, 2009, doi: 149 10.1016/j.matdes.2009.04.028. [195] R. Sopher, J. under the ischial tuberosities during sitting is elevated at abnormally high or low body J. Biomech. , vol. 43, no. 2, pp. 280 286, 2010, doi: 10.1016/j.jbiomech.2009 .08.021. [196] E. Linder - J. Biomech. Eng. , vol. 131, no. 1, p. 011003, 2009, doi: 10.1115/1.3005195. [197] S. Chen, World Congress of Biomechanics , 2018. [198] C. W. J. Oomens, O. F. J. T. Bressers, E. M. H. Bosboom, C. V. C. Bouten, and D. L. Comput. Methods Biomech. Biomed. Engin. , vol. 6, no. 3, pp. 171 180, 2003, doi: 10.1080/1025584031000121034. [199] G. W. Schmid - Schonbein, S. - Y. Woo, and B. W. Zweifach, Frontiers in Biomechanics . New York, Berlin, Heidelberg, Tokyo: Springer - Verlag, 1986. [200] November 1997, pp. 40 6 408, 2015. [201] E. L. Wagnac, C. - - IEEE Trans. Biomed. Eng. , vol. 55, no. 2 Pt 1, pp. 774 783, 2008, doi: 10.1109/TBME.2007.912640. [202 ] between Seat Cushion and Human Buttock - Engineering , vol. 02, no. 09, pp. 720 726, 2010, doi: 10.4236/eng.2010.29093. [203] E. M. H. Bosboom, M. K. C. Hessel ink, C. W. J. Oomens, C. V. C. Bouten, M. R. Drost, J. Biomech. , vol. 34, no. 10, pp. 1365 1368, 2001, doi: 10.1016/S0021 - 9290(01)00083 - 5. [204] S. Loerakker, L. R. Solis, D. L. Bader, F. P. T. Baaijens, V. K. Mushahwar, and C. W. J. Comput. Methods Biomech. B iomed. Engin. , vol. 16, no. 5, pp. 520 529, 2013, doi: 10.1080/10255842.2011.627682. [205] J. - J. Biomech. Eng. , vol. 136, no. 11, p. 111012, 2014, doi: 10.1115/1.4028490. [206] S. 19th Congr. Eur. Soc. Biomech. , no. Vi, p. 7337, 2012. 150 [207] J. - disp lacement and strain fields with ultrasound and Digital Image Correlation (DIC) IRBM , vol. 36, pp. 170 177, 2015, doi: 10.1016/j.irbm.2015.02.002. [208] C. Then et al. Tech nol. Heal. Care , vol. 15, no. 6, pp. 385 398, 2007. [209] R. M. A. Al - quasi - static response to loading of sub - dermal tissues in the human buttock using C lin. Biomech. , vol. 50, no. October 2016, pp. 70 77, 2017, doi: 10.1016/j.clinbiomech.2017.09.017. [210] M. Makhsous et al. Clin. Anat. , vol. 24, no. 5, pp. 638 645, 2011, doi: 10.1002/ca.21119. [211] Nature , vol. 299, pp. 6 8, 1982. [212] R. M. A. Al - Dirini, M. P. Re Clin. Biomech. , vol. 30, no. 7, pp. 662 668, 2015, doi: 10.1016/j.clinbiomech.2015.05.008. [213] ep Ann. Biomed. Eng. , vol. 43, no. 2, pp. 297 305, 2015, doi: 10.1007/s10439 - 014 - 1202 - 6. [214] R. Salcido et al. - controlled pressure sessions in a fuzzy rat mo Advances in skin & wound care , vol. 7, no. 5. pp. 25 34, 1980. [215] Biomech. Model. Mechanobiol. , vol. 18, no. 6, pp. 1947 1964, 2019, doi: 10.1007/s10237 - 019 - 01187 - 5. [216] new sitting concept designed for prevention of pressure ulcer on the buttock using finite el Med. Biol. Eng. Comput. , vol. 45, no. 11, pp. 1079 1084, 2007, doi: 10.1007/s11517 - 007 - 0261 - 3. [217] E. Linder - - induced pressure sores in rat hindlimb: muscle J. Appl. Physiol. , vol. 96, no. 6, pp. 2034 2049, 2004, doi: 10.1152/japplphysiol.00888.2003. [218] Y. C. Fung, Biomechanics: Mechanical Properties of Living Tissue . Springer - Verlag, 1993. [219] T. J. Vogl et al. in tissue engineering using Acad Radiol , vol. 17, no. 12, pp. 1486 1491, 2010, doi: 10.1016/j.acra.2010.08.010. 151 [220] H. M. Clarkson, Musculoskeletal Assessment: Joint Range of Motion and Manual Muscle Str ength . Lippincott Williams & Wilkins, 2000. [221] Int. J. Non. Linear. Mech. , vol. 107, no. July, pp. 77 85, 2018, doi: 10.1016/j.ijnonlinmec.2018.09.007. [222] J. Tissue Viability , vol. 24, no. 2, pp. 51 61, 2015, doi: 10.1016/j.jtv.2015.03.003. [223] on the correlation of theory Proc. R. Soc. London. Math. Phys. Sci. , vol. 326, no. 1567, pp. 565 584, 1972. [ 224] - parameter shear wave inversion in MR elastography of incompressible transverse isotropic media: Application to Magn. Reson. Med. , vol. 75, no. 4, pp. 1537 1545, 20 16, doi: 10.1002/mrm.25740. [225] [226] Massachusetts, USA, 2015. [227] S. D. Darrah et al. pressure ulceration in ice sledge hockey players Measuring static seated pressure Disabil. Rehabil. Assis t. Technol. , vol. 11, no. 3, 2016, doi: 10.3109/17483107.2014.921939. [228] distribution in 468 men and women aged 18 - J. Appl. Physiol. , vol. 89, pp. 81 88, 2000, doi: 10.1152/jappl.2000.89.1.81. [229] evaluation of a new methodology for the fast generation of patient - specific Finite Element models of the buttock for sitting - acquired deep tis J. Biomech. , vol. 79, pp. 173 180, 2018, doi: 10.1016/j.jbiomech.2018.08.001. [230] - related losses in strength on the ability to J. Biomech. , vol. 4 6, no. 1, pp. 13 18, 2013, doi: 10.1016/j.jbiomech.2012.08.046. [231] L. Demarré et al. Int. J. Nurs. Stud. , vol. 52, no. 11, pp. 1754 1774, 2015, doi: 10.1016/j.ijnurstu.2015 .06.006. [232] C. Wann - patients with hospital - acquired and pre - 152 J. Clin. Nurs. , vol. 17, no. 13, pp. 1718 1727, 2008, doi: 10.1111/j. 1365 - 2702.2008.02286.x. [233] Proc. IASTED Int. Conf. , pp. 164 167, 2017, doi: 10.2316/P.2017 .852 - 043. [234] J. Neuroeng. Rehabil. , vol. 13, no. 1, pp. 1 11, 2016, doi: 10.1 186/s12984 - 016 - 0208 - 8. [235] to cancer risk: A meta - J. Natl. Cancer Inst. , vol. 106, no. 7, 2014, doi: 10.1093/jnci/dju098. [236] J. Scott, S. Chen, S. Rocca J. Mech. Behav. Biomed. Mater. , vol. 110, 2020, doi: 10.1016/j.jmbbm.2020.103964. [237] perimental Evaluation and Analytical Model of the Pressure Generated by Elastic Compression Garments on a Deformable Med. Eng. Phys. , 2020, doi: 10.1016/j.medengphy.2020.05.015. [238] response to loading in men with spinal PLoS One , vol. 13, no. 2, pp. 1 15, 2018, doi: 10.1371/journal.pone.0191868. [239] J. R. Basford, T. R. Jenkyn, K. - N. An, R. L. Ehman, G. Heers, and K. R. Kaufman, Arch. Phys. Med. Rehabil. , vol. 83, no. 11, pp. 1530 1536, 2002, doi: https://doi.org/10.1053/apmr.2002.35472. [240] ology , Fracture Prediction , and Bone Loss and Muscle Atrophy in Spinal J. Spinal Cord Med. , vol. 29, no. 5, pp. 489 500, 2006, doi: 10.1080/10790268.2006.11753898. [241] D. P. Lemmer et al. Arch. Phys. Med. Rehabil. , vol. 100, no. 6, pp. 1042 1049, 2019, doi: 10.1016/j.apmr.2018.11.006. [242] llowing Spinal Cord Injury and Implications for Ostomy Wound Manag. , vol. 60, no. 2, pp. 34 45, 2014. [243] n Int. Wound J. , vol. 13, no. 2, pp. 287 288, 2016, doi: 10.1111/iwj.12257. [244] 153 Seating Interface of Custom - Arch. Phys. Med. Rehabil. , vol. 83, pp. 1151 1156, 2002, doi: 10.1053/apmr.2002.33987. [245] C. M. Modlesky et al. injury using dual - energy X - ray absorptiometry an J. Appl. , vol. 96, pp. 561 565, 2004. [246] A. Mehta et al. Am. J. Surg. , vol. 203, no. 3, pp. 303 307, 2012, doi: 10.1016/j.amjsurg.2011.10.007. [247] wheelchair tilt - in - space and recline angles on skin perfusion over the ischial tuberosity in people with spinal cord inju Arch. Phys. Med. Rehabil. , vol. 91, no. 11, pp. 1758 1764, 2010, doi: 10.1016/j.apmr.2010.07.227. [248] - Sized Men in Seated J. Biomech. Eng. Asme , vol. 129, no. 1, pp. 58 65, 200 7, doi: 10.1115/1.2401184. [249] - in - space wheelchairs increase occupational Disabil. Rehabil. Assist. Technol. , vol. 11, no. 1, pp. 3 12, 2016, doi: 10.3109/17483107.2014.93202 1. [250] J. Biomech. Eng. , vol. 133, no. 11, 2011, doi: 10.1115/1.4005400. [251] pressure Gait Posture , vol. 39, no. 1, pp. 224 228, 2014, doi: 10.1016/j.gaitpost.2013.07.010. [252] system for the measurement of plant ar forces and pressures during barefoot level walking J. Foot Ankle Res. , vol. 3, no. 1, pp. 1 9, 2010, doi: 10.1186/1757 - 1146 - 3 - 11. [253] Spinal Cord , vol. 49, pp. 136 141, 2011, doi: 10.1038/sc.2010.58. [254] I. A. Braga, C. colonization of pressure ulcers: Assessment of risk for bloodstream infection and impact J. Hosp. Infect. , vol. 83, no. 4, pp. 314 320, 2013, doi: 10.1016/j.jh in.2012.11.008. [255] - time J. Rehabil. Res. Dev. , vol. 53, no. 5, pp. 585 598, 2016, doi: 10.1682/JRRD.2015.07.0130. [256] S. E. Sonenblum, S. Sprigle, 154 Arch. Phys. Med. Rehabil. , vol. 89, no. 3, pp. 486 491, 2008, doi: 10.1016/j.apmr.2007.09.029. [257] M. Makhsous et al. itting Load to Decrease the Risk of Arch. Phys. Med. Rehabil. , vol. 88, no. 7, pp. 862 870, 2007, doi: 10.1016/j.apmr.2007.03.017. [258] Phys. Ther. , vol. 84, no. 3, pp. 255 261, 2004, doi: 10.1093/ptj/84.3.255. [259] ovascular collapse to tissue Biomech. Model. Mechanobiol. , vol. 18, no. 6, pp. 1947 1964, 2019, doi: 10.1007/s10237 - 019 - 01187 - 5. [260] J. A. Whitney et al. Wound Repair Regen. , vol. 14, no. 6, pp. 663 679, 2006, doi: 10.1111/j.1524 - 475X.2006.00175.x. [261] Adv. Sk in Wound Care , vol. 23, no. 1, pp. 21 24, 2010, doi: 10.1097/01.asw.0000363489.38996.13. [262] J. Biomech. Eng. , vol. 129, no. 1, p. 58, 2006, doi: 10.1115/1.2401184. [263] K. Kobara J. Phys. Ther. Sci. , vol. 26, no. 5, pp. 701 706, 2014, doi: 10.1589/jpts.26.701. [264] B. E. Dicia nno et al. Assist. Technol. , vol. 21, no. 1, pp. 13 22, 2009, doi: 10.1080/10400430902945769. [265] L. C. Gerhardt, N. Mattle, G. U. Schrade, N. D. Spencer, and - fabric interactions of relevance to decubitus: Friction and contact - pressure Ski. Res. Technol. , vol. 14, no. 1, pp. 77 88, 2008, doi: 10.1111/j.1600 - 0846.2007.00264.x. [266] Assist. Technol. , vol. 15, no. 1, pp. 49 57, 2003, doi: 10.1080/10400435.2003.10131889. [267] testing and basic frictio Rock Mech. Rock Eng. , vol. 45, no. 6, pp. 1023 1035, 2012, doi: 10.1007/s00603 - 012 - 0265 - 7. [268] Turkey under dry, wet and submerged Bull. Eng. Geol. Environ. , vol. 75, no. 4, pp. 1683 1699, 2016, doi: 10.1007/s10064 - 015 - 0828 - 4.