éffié WWIIWWWWINHWHIIIHWHHIWWHHI

THESIS

      
     

LIBRARY
Michigan Sm:

University

This is to certify that the

thesis entitled

THE MECHANICAL PROPERTIES OF THE
RABBIT CERVIX THROUGHOUT PREGNANCY

presented by
ANDREA J. DANAS

has been accepted towards fulfillment
of the requirements for

Masters Mechanics

degree in

W

Major professor

Date To“; ‘0 [38‘

0-7639

 

 

MSU

LIBRARIES
n

RETURNING MATERIALS:

 

 

 

Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

THE MECHANICAL PROPERTIES OF THE
RABBIT CERVIX THROUGHOUT PREGNANCY

By

Andrea Joyce Danas

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Metallurgy, Mechanics and Materials Science

1981

ABSTRACT

THE MECHANICAL PROPERTIES OF THE
RABBIT CERVIX THROUGHOUT PREGNANCY

By

Andrea Joyce Danas

The purpose of this research was to study the mechanical
properties of the rabbit cervix throughout various stages of pregnancy
and to gain a better understanding of similar properties in the
human. White New Zealand Albino Rabbits were impregnated and their
cervixes removed at designated stages of pregnancy. Three types of
mechanical tests were employed on the excised tissue. The first set
of tests were loading and unloading cycles to a specified strain level
at various constant strain rates. The second series of tests were
sinusoidal strain tests conducted at various frequencies. And
finally, the third test was a stress—relaxation test held for a
specified period of time. Histological examinations were performed
on representative tissues at different stages of pregnancy to monitor

the changes in the connective tissue composition and structure.

ACKNOWLEDGMENTS

The author expresses her deepest appreciation to Dr. Robert Wm.
Little for his continual guidance and support throughout this project.
I especially thank him for his friendship throughout my career.

I would also like to thank Dr. Robert Hubbard and Mr. Robert
Schaffer for their technical assistance with the equipment needed for
this project.

I sincerely thank Mrs. Jane Walsh for her preparation of the
histological examinations for this study.

I would like to thank Mr. Robert Konopactz for his preparation
of the electron microscopy studies.

I would like to thank the Departments of Biomechanics and
Metallurgy, Mechanics and Materials Science for their support and
assistance.

I especially thank all of my friends and family without whose

continued support and encouragement this would not have been possible.

ii

 

TABLE OF CONTENTS

ACKNOWLEDGEMENTS

LIST OF TABLES

LIST OF FIGURES

Introduction

Histology of the Rabbit Cervix
Experimental Protocol

Results and Discussion
APPENDIX A

List of References

20
31
33
62
68

Table 1.
Table 2.

Table 3.

LIST OF TABLES

Average A values (N/m2 x 104)
Terminal slope values (N/m2 x 104)

Hysteresis values (N/m2 x 104)

37
38
38

Figure

Figure

Figure

Figure

Figure

Figure

Figure
Figure
Figure
Figure

Figure

Figure
Figure

Figure

Figure
Figure
Figure

Figure

10.
ll.

12.
13.
14.

15.
l6.
17.

18.

LIST OF FIGURES

Hexacrome stained cross—section of the nonpregnant cervix
(10X).

Hexacrome stained cross-section of the nonpregnant cervix
(50X)

Hexacrome stained cross—section of the nonpregnant cervix
(100x)

Electron microscopy of the nonpregnant cervix (23.4X)

Photomicrograph of a cervixal fold from the inner ring.
(78X)

Photomicrograph of a cervical fold from the inner ring
(546X)

Photomicrograph of the middle ring. (780x)
Photomicrograph of the outer ring. (7OOX)
Hematoxylin—eosin stained 23 day pregnant rabbit cervix.
Cross—section of a 33 day (term) cervix.
Hematoxylin—eoxin stained pregnant/nonpregnant rabbit
cervix. Left horn 1 hour post partum; right horn
nonpregnant.

1 Hour post partum left horn of the rabbit cervix.
Nonpregnant right horn of the rabbit cervix.

Hematoxylin-eosin stained 48 hour post partum rabbit
cervix.

A typical stress—strain curve.
Strain rate effects on the stress—strain curve.
Strain rate effects on the stress—strain curves.

The effects of strain rate on the hysteresis loops.

22

22

23

23
25

25

26
26
27
27
29

29
3O
30

34
35
36
39

vi

Figure 19. A typical relaxation plot of normalized stress vs log t. 41

Figure 20. 0lo of peak stress vs. number of cycles for the non— 44
pregnant rabbit cervix at frequencies of 1, 0.1, and
0.01 Hz.

Figure 21. 0/o of peak stress vs. number of cycles for the 20 day 46
pregnant rabbit at frequencies of 1, 0.1 and 0.01 Hz.

Figure 22. 0[0 of peak stress vs. number of cycles for the 24 day 48
pregnant rabbit at frequencies of l, 0.1, and 0.01 Hz.

Figure 23. 0/o of peak stress vs. number of cycles for the 23 day 50
(term) rabbit at frequencies of l, 0.1 and 0.01 Hz.

Figure 24. 0/o of peak stress vs. number of cycles for the 48 hour 52
post partum rabbit at frequencies of l, 0.1, and 0.01 Hz.

Figure 25. 0/o of peak stress vs. number of cycles for a 54
frequency of 0.01 Hz for all stages of pregnancy.

Figure 26. 0[0 of peak stress vs. number of cycles for a 56
frequency of 0.1 Hz for all stages of pregnancy.

Figure 27. 0/o of peak stress vx. number of cycles for a 58
frequency of 1 Hz for all stages of pregnancy.

Figure A-l. Description of the Cervix 63

INTRODUCTION

The definition of the cervix as an independent organ has been the
subject of much debate. The classical description of the cervix
defines it as the lowest segment of the uterus but recently, it has
been considered an organ in its own right aiding in the birth process.

The cervix undergoes remarkable changes throughout pregnancy from
a highly collagenous undistensible structure capable of resisting
pressure and gravity, and holding the nondeveloped fetus, to a soft
highly distensible structure permitting passage of the fetus at term.
In order for this to occur, the cervix must undergo major changes in
its connective tissue properties, the nature of which are not
completely understood. It is believed that there is a breakdown or
dissociation in the collagenous framework enabling the cervix to
become more easily dilated, but the exact nature of this breakdown is
unknown.

Several scientists have attempted to determine the amount of
dilatation the cervix can undergo at various stages of pregnancy
without incuring damage. They have attempted to measure tensile
strength and the elastic modulus throughout pregnancy to help quantify
these changes. Other scientists have attempted mathematical and
geometrical models of the cervix in hopes of more fully understanding
it's mechanical characteristics.

This research examined the mechanical properties and histological
changes of the cervix throughout the various stages of pregnancy as

1

2

well as the nonpregnant and post partum cervix. In order to do so, it
was necessary to understand the composition of the cervix. The
anatomical discussion of the cervix is included in Appendix A.

The function of the cervix during pregnancy is to help retain the
conceptus by contributing to the closing of the uterine cavity. Its
function during labor is to transmit increased pressure and tension
resulting in increased dilatation in response to contractions of the
corpus, thereby permitting the delivery of the fetus. (1,2)

Until the eighth week of pregnancy, the configuration of the human
uterus is little changed from the nonpregnant uterus. From eight to
twelve weeks of pregnancy, hypertrophy occurs with a lengthening and
thickening of the isthmus. After twelve weeks, the conceptus fills
the uterine cavity and further growth causes a gradual unfolding of
the isthmus until the isthmus reaches the fibromuscular junction. At
this time, further growth of the conceptus causes a thinning of the
uterine walls (2). During this unfolding of the isthmus, the muscle
fibers at the isthmocervical junction rearrange into a circular
configuration which has no sphincterlike activity but does transmit a
pull on the upper segment. During the sixth month of pregnancy, the
isthmus and lower part of the corpus are drawn upward and the lower
uterine segment is formed. This lower uterine segment is a cone of
muscle which is continuous with the inner interlacing layer of muscle
in the corpus, with an apex two—thirds of the way down the isthmus.
The muscle fibers of the upper and lower uterine segments contract and
relax drawing the lower uterine segment and the unfolding cervical
canal upward, causing the canal to progressively shorten, bringing

about effacement. The muscle cone works with the cervix to prevent

3

dilatation. The cone dilates and is drawn up in late pregnancy
allowing the fetal head to descend into the lower uterus and the
conceptus to be supported by the cervix. The cervix becomes softer
preparing for dilatation and is more responsive to the stronger
contractions of the corpus. Cervical dilatation is controlled by
uterine activity and mobility as well as the resistance of the lower
uterine segment and the circular musculature in the lower part of the
corpus.

Accoring to Philpott (2), Friedman analyzed the rate of dilatation
of the cervix and compared the time rate of change of dilatation in
the cervix with fetal head descent. Dilatation of the cervix is
completed in the first stage of labor, which is divided into two
phases, the latent and active phases. The latent phase is the time
from the onset of labor when contractions become regular until the
time when there is a noticeable increase in the rate of progress of
cervical dilatation. The active phase then continues through cervical
dilatation until it is completely dilated. The active phase is
further subdivided into three stages, the acceleration phase, the
middle phase where most of the dilatation takes place, and the
deceleration phase.

There is only minimal head descent during the latent and
acceleration phases. During the middle phase of cervical dilatation,
fetal head descent begins. Maximal rate of head descent occurs during
the deceleration phase and is completed during the second stage of
labor (2).

During uterine contractions, the increase in amniotic fluid

pressure causes the fetus to be pressed towards the cervix at which

ll

4
point dilatation begins. Lingren (1) shows that friction occurs

between the descending head and the uterine wall with a coefficient of
0.2. Maximum resistance from the cervix occurs at the base of the
head. He also determined that uterine contractions when the cervix
has effaced occur with a frequency of 11 — 12 per hour with a pressure
of 25 — 30 mm Hg. Amniotic fluid pressure changes also from 30
initially to 45 mm Hg when the cervix is fully dilated.

Studies have been made of the microscopic changes which take place
during cervical effacement. Since the cervix is mainly connective
tissue, the changes must be explained in terms of connective tissue.
Although there are changes in all connective tissue structures in the
body during pregnancy, the cervix has the most pronounced of these
changes. The cervix becomes softer and increasingly more dilatable.
During labor, the cervix dilates to a diameter of 10 cm (3). One week
after delivery it begins to regain its former contours and within one
month to six weeks, the cervix returns to normal.

In the nonpregnant cervix, collagen appears as dense tightly woven
bundles of interlacing fibers running parallel to one another (3),
whereas in the pregnant cervix, the collagen fibers dissociated into
fine fibrillar components which form branches (3). Fibrillar
dissociation is the change that permits the fibers to loosen and slide
upon one another consequently allowing the cervix to soften and dilate
without injury. There is an increase in the dissociation of collagen
with the increase of cervical ripening. In late pregnancy collagen
fibers are similar to those of the nonpregnant cervix except for the
intorduction of spaces between the bundles (3). Danforth, Buckingham,

and Roddick (4) observed that the loosening of connective tissue

5
occurs during the last stage of pregnancy. The fibrous dissociation

of collagen is completed during labor, when the fibers separate and
branch in many directions rather than running parallel to each other
(3). Collagen concentrations in the cervix decline during pregnancy,
however the maximum drop accompanied with an increase in water content
of the cervix occurs prior to labor and remains the same throughout
labor.

Ichijo, et al (5) describes five grades of collagen dissociation.
First, the fibers tend to lie straight, but many fibers are slightly
wavy; second, the fibers show a more wavy course, and begin to
dissociate; third, the fibers branch into fibrillar components;
fourth, the dissociation process progresses further and results in
many fine branchings; and finally all the fibers are branched with an
increase in dissociation with cervical ripening. He claims that
during effacement, the collagen fibers become immature as they
dissociate. Consequently, either collagen fibers transform into
reticular fibers or collagen fibers depolymerize and reticular fibers
appear in their place. The basis for this idea lies with the fact
that with persistant edema or inflammation, collagen fibers change to
reticular fibers. Also, in the embryonic development process,
collagen fibers evolve from reticular fibers, making it conceivable
for this mutual transformation to take place in the cervix. An
increase in the acid mucopolysaccharides in the cervical ground
substance in late pregnancy may provide the conditions necessary for
the transformation.

Prior to Ichijo, Danforth and Buckingham (6) questioned whether

the ground substance was a factor in collagen dissociation or whether

6

the trauma of labor could produce this loosening of the fibers. They
noted that during and after pregnancy, the characteristic banding
period of the collagen (640 A) is similar to the nonpregnant cervix,
except the fibers may be further apart. They subjected a nonpregnant
cervix to the most ”violent mastication in the colloid mill" to see if
the trauma of labor could produce loosening of the fibers. Although
the fibers were broken, they found that they remained tightly held
together. Therefore a possible cause could in fact be a change in the
ground substance. (7,8)

The ground substance mass increases significantly during
pregnancy, reaching a maximum at term and immediately after delivery,
as evidenced by the change in water concentrations (3). The
accumulation of water during pregnancy amounts to an increase of
70/0 maximum, indicating that it seems inadequate to constitute a
total explanation for the loosening and separation of the collagen
fibrils. However, there is also an increase in the
mucopolysaccharides in the ground substance which may affect the
collagen bonds. According to Danforth and Buckingham (6) the
metabolism of mucopolysaccharides is rapid compared to other
macromolecules; therefore, this change in the acid mucopolysaccharide
concentration may secondarily affect the changes in the collagen of
the cervix.

Another aspect Danforth and Buckingham (6) investigated was
collagen solubilities. Since old collagen is almost insoluble and
newly formed collagen is readily soluble in a neutral salt or dilute
citric or acidic acid, testing to see if there is any difference in

the amount of soluble or insoluble collagen should indicate an

7

increase in new or old collagen respectively. He found that in the
fraction of collagen which is soluble in a dilute acid, the identity
of the fibrils is lost and the banding period vanishes. Essentially,
collagen has gone into solution. However, neutralization of the acid
reconstituted the fibrils making them indistinguishable from the
originals. The question arises as to whether this phonomena happens
after labor.

More possibilities for causes of cervical effacement have been
considered. Danforth, et al. (13) isolated an enzyme, collagenase,
which specifically attacks collagen. It is the only enzyme isolated
thus far that causes the degradation of collagen. According to
Danforth, et al., Silbert(9) proposed that the hormonal activation of
collagenases had caused fibrillar dissociation. Also in 1973,
Groschel-Stewart, Herman and Schwalm(10) isolated a ground substance
protein resembling anyloid from the nonpregnant cervix. They
suggested that it may have some physiologic changes during pregnancy
and labor.

Some changes between the collagen of the cervix and the collagen
of the body were noted by Danforth and Buckingham (8). Amino acid
contents in collagen in nonpregnant cervixes were similar to those in
standard human collagen with a few exceptions. Hydroxyproline values
were only 820/0 of those in the standard; therefore, collagen
represents only 820/0 of the total protein in the nonpregnant
cervix. Also, two acidic acids, asparic acid and glutomic acid, were
increased by 1170/0 and 1270/0 respectively of there initial
nonpregnant values.

Exploring the changes in concentrations of amino acids during

8

pregnancy, Danforth and Buckingham discovered an absolute decline in
collagen and an increase in noncollagenous protein. The increase in
water content during pregnancy may result in a corresponding increase
in proteoglycan content. Other changes in the collagen included a
drop of hydroxyproline to 520/0 of the standard when complete
dilatation of the cervix was reached. There was also a considerable
increase in glycosaminoglycans and acidic amino acids.

Danforth and Buckingham indicate that glycan constitutes about
one—third of the collagen, therefore the drop in this amino acid
confirms that hydroxyproline changes. They also found high levels of
tyrosine and phenylaline which are normally present in small amounts
suggesting the presence of some other protein. They were unable to
identify the 180/0 of the proteins present in the nonpregnant and
470/0 present in the post partum cervix. It is known that they are
noncollagenous and do not originate from either muscle or serum.
Danforth and Buckingham proposed two possible solutions. The first is
that it is a constituent of the ground substance in the form of
proteoglycan core protein evidenced by the increase in this amino acid
which composes a major part of the ground substance. The second
possibility is that the new protein represents an increase in the
acidic structural protein of Timpl, Wolff, and Weiser.

The nature of the bond which holds the collagen fibrils together
in dense tissues or allows them to separate in looser tissues is not
known. One solution is that the molecular configuration may shift to
produce or prevent aggregation of the fibrils (6). A second idea
postulated by Hughesdon is that the hormone relaxin may be responsible

for the cementing of collagen together (11). Finally, another

9

solution, and the one most pursued, is the influence of the ground
substance in causing collagen fibers to bind together or slide apart
(8). The ground substance may affect the intermolecular reactions
causing either cohesion or dissociation.

Other changes in the cervix during pregnancy should also be
noted. Changes occur in the reticulin fibers during pregnancy as
well. In the nonpregnant cervix, the reticular fibers appear as short
segments irregularly dispersed throughout collagen bundles (3). In
the pregnant cervix, the reticular fibers are more robust and
luxuriant and can be traced in thin wavelike branchings which tend to
parallel the collagen fibers. The structure of the reticular fibers
post partum appears to be much the same as in the cervix at term
except they are more widely separated and somewhat thinner.

Secondly, some changes in the cervix can be accounted for by
consideration of the cervical musculature. According to Danforth (4),
Hughesdon claims that the degree of dissociation of collagen decreases
as it approaches the internal 05. The connective tissue surrounding
the intrinsic muscle bundles does not dissociate during pregnancy as
do the collagen fibers intermixed within them. Since there is more
muscle higher up in the cervix, softening of the cervix is relatively
less.

At three to four months of pregnancy, the intrinsic muscle fibers
of the cervix are more prominent, making the cervix appear more
muscular (11). However, the distribution of muscle is variable with
more bundles of muscle in the central portions of the cervix. Since
they are heavily interspersed with fibrous tissue, sphincteric effects

are an impossibility. Therefore, changes in the cervix due to

10

pregnancy are not due to muscular alterations.

During pregnancy there are connective tissue changes throughout
the body. Cervical collagen is qualitatively similar to that of the
anterior rectus sheath, where marked changes occur enabling the
abdomen to accomodate the enlarging uterus without tenseness (3).
Other changes in connective tissue occur in the skin where there is a
loss of tensile strength during pregnancy (12).

Since there is a marked increase in blood volume, vascular changes
occur in the body as well. The circulatory system may cause some of
the changes the body undergoes during pregnancy. An active pancreatic
enzyme, elastase, has been isolated which causes the breakdown of
elastin and quantities of an elastase inhibitor have been discovered
in late pregnancy two to three times the normal values (6). The
significance of the increase is unknown but, it may tend to limit some
vascular changes which are considered part of normal pregnancy. Other
vascular changes include an increase in vericose veins, changes in
both the aorta and peripheral vasculature. Further changes include
the loosening of the joints or the pelvis and lower back, the mobility
of the teeth and eye refraction. All of these are results of changes
in connective tissue and may be caused by some factor carried by the
circulation during pregnancy (3).

At this point, a brief mention should be made about cervical
incompetency. In a study by Hafez (13), it was discovered that the
percentage of muscle was higher in a group of patients with cervical
incompetency than in a similar control group immediately following
labor. Therefore, they believed that excessive amounts of muscle may

be a factor in the production of cervical incompetency. Since several

11

cervixes did not conform to this idea, cervical incompetency must be a
result of many factors rather than just the deviation in cervical
composition.

In order to determine the mechanical properties of the cervix
correlating with the histological changes, several researchers have
worked in various areas of mechanical testing of the cervix. One such
area was to design an apparatus which dilated the cervix and recorded
the amount of dilatation the cervix could undergo without damage to it.

T. Bakke (14) developed an instrument which measured the
fibroelasticity or consistency of the cervix. The instrument
measures, in vivo, the consistency of the cervix by measuring the
depth and shape of an impression while recording the force needed to
make this impression. He defined cervical consistency as the angle
between the walls of the depressed area divided by the applied force.
The instrument consisted of a wing shaped end which would conform to
the shape of the cervix, attached to a sliding tube which would
audibly register when the desired force is applied. Thus the angle
between the walls of the cervix is measured by the spread of the
wings, the force being set as desired , with the ratio of these two
quantities being the fibroelasticity. This instrument can give
quantitative measurement of the amount of cervical dilatation present
in vivo.

Three studies have attempted mathematical models of the cervix.
The first, Lui et al. (15) attempted to model the cervix as a
thick—walled cylinder. They subjected, in vitro, the cervixes from
multi-parous women to a uniform axial force by using a specially

designed force measuring dilator and formulated a stress vs. strain

12

relationship. From these curves, they were able to obtain a
quasi-static moduli for the different degrees of dilatation. They
also obtained theoretical distributions of the radial and
circumferential stress across the cervical wall and found a yielding
of cervical tissue at 9-11 mm dilation. The stress—strain curve for
the cervix is characteristic of a viscoelastic tissue. The
quasi—static modulus, a measure of the gradiant of a stress—strain
relationship, increased with each stage of dilation.

Lui, et al. assumed the human cervix to be an elastic, isotropic,
incompressible, homogeneous cylinder with a diameter and length of 25
mm. They also accounted for frictional resistance and the unknown
recovery rate after stretching. They employed the Lame equations for
a thick walled cylinder to obtain the quasi—elastic modulus for each
dilatation. Values of radial and circumferential stresses were
calculated for a range of dilations at increasing radii across the
cervical wall. They found that both radial and circumferential stress
decayed rapidly and nonlinearly towards the outer boundary of the
cervical wall. For radial stress, the limiting value was zero.

In 950/0 of the experimental tests, a radial tear beginning from
the inside of the cervical wall was observed at dilatations of 13—14
mm diameter after an initial microscopic yielding at dilatations of
9—11 mm diameter. Therefore, they concluded that to avoid tissue
damage 10 mm is a limitimg value of tissue compliance.

Rice and Yang (16,17) did two studies in which they mathematically
modeled the cervix. In their first study, the cervix was modeled as
an axisymmetric ring located at the external 05, for which they

qualitatively related intrauterine pressure to cervical dilatation

13

during the first stage of labor. In the second study they developed a
more mathematically detailed finite element nonlinear viscoelastic
axisymmetric membrane model for the human cervix.

Two theories have been proposed to explain cervical dilatation.
One concerned radial forces from the lower uterine ligaments. The
other, the one which Rice and Yang supported, concerned forces caused
by the wedging action of the fetal head driven by intrauterine
pressure. Rice and Yang refuted the theory that uterine ligaments
took part in dilatation
because the relative movement of these ligaments was so small.

Several assumptions for the two models were made: 1)the system
consisting of the uterus, cervix and fetal head were axisymmetric and
coaxial; 2) functional and inertial forces were negligible; 3) the
cervix was passive; and 4) intrauterine force caused by uterine
contractions was the only driving force.

Rice and Yang (16,17) had many reasons for their assumptions. The
fetus excluding its head was completely passive and pliable so that
the uterine contents were regarded as a homogeneous mass through which
hydrostatic pressure was transmitted. During uterine contraction,
head—to-cervix pressure in the axial direction summed over the cervix
balances the axial thrust of the intrauterine pressure on the fetal
head. Rice and Yang treated the uterus, uterine contents and cervix
as an isolated mechanical system.

Frictional effects were ignored for two reasons. Firstly, there
is a lubrication between the fetus and the structural system in the
form of vernix caseosa, amniotic fluid, blood and other secretions.

This lubrication significantly lowers friction effects. Secondly, a

14

plot of intrauterine pressure, which is proportional to the axial
force, versus dilatation was used as a rough indicator of relative
motion or force since a direct measure of friction was impractical.
The behavior of the curve indicated a lack of friction. Since the
acceleration of mass for inertial forces was insignificant, inertial
forces were also neglected.

In their first model, Rice and Yang (I6) depicted the fetal skull
as an ellipsoid and the cervix as an axisymmetric ring, the cervical
wall being tangent to the fetal head at the equator of the head. The
model, they attempted to design was one that quantitatively related
cervical dilatation to intrauterine pressure using a model based only
on the birth system geometry and tissue mechanisms, and a hydraulic
mechanism neglecting friction and inertial effects.

Briefly, Rice and Yang developed their model using force balances
and geometric relationships. In their formulation they determined
whether the cervical tissue was linearly or exponentially elastic.
The data of the cervix best fit an exponentially elastic curve, which
was also shown to be linearly viscous. Linear regression formulae
were used to determine various parameters.

The model was found to correspond well with the data gathered,
therefore Rice and Yang concluded that the hypothesis of a hydraulic
mechanism causing cervical dilatation was accurate.

In their second study, that of the axisymmetric membrane model
(19), Rice and Yang also assumed that the cervix followed the
following conditions: 1)the membrane was orthotropic with zero stress
across its thickness; 2)there was no shear strain along the tissue

axes; and 3)the tissue was incompressible (1) =0.5).

 

 

15

The results obtained were similar to those of the ring model,
except they were more specific; 1)the cervix did dilate after every
contraction showing elastic stretch and some viscous strain during
each contraction. This was attributed to the viscoelastic properties
of the cervix; 2) it was observed that dilatation was proportional to
the strength of contraction, implying a monotonic relation which the
model shows; 3)the larger the total dilatation, the larger the elastic
contraction at any given strength; 4)the time delay between
contraction and subsequent dialtation decreased as the inlatation
increased (the model however, does not indicate this result); and
5)oxytocin was found to cause greater contractions and correspondingly
faster dilatation. Greater pressure also resulted in faster
dilatation, which the model did show.

The work presented in this dissertation parallels some of the work
done by Harkness who performed several studies on rats to determine
the changes mechanically and histologically that the cervix undergoes
during and after pregnancy. In one such study, Harkness and Moralee
(18) attempted to determine the route loss of collagen from the rat's
uterus post partum. They wanted to determine whether the
disappearance of collagen from the uterus after parturition was a
result of internal reabsorption of the material, or, whether collagen
was shed into the lumen of the uterus and lost through the vagina.

In another study, Harkness and Harkness (l9) attempted to
determine if physical properties of the cervix change after
parturition. Stretching the cervix between two parallel steel rods,
one through each canal, the tissues were initially loaded with a 50 9
weight for 25 minutes. The load was then increased 25 9 every 15

seconds until it ruptured. It is important to note that these tests

 

16
applied load across the junction between the two canals and
histological studies show that fibers run circumferentially around the
canals and only the outer fibers surround both canals.

Harkness (20) has determined that the changes after parturition
fall into two phases. In the first phase immediately after
parturition, the reduction in circumference, was due to the
contraction of the smooth muscle followed by a consolidation of the
tissue due to the changes in the collagenous framework. He found that
the cirCUmference of the cervix dropped rapidly in the first 8 hours
to about 1/3 of its initial value at parturition and believed that the
rapidity of the change make it unlikely that there is a removal of
collagen and replacement with a "newly formed framework of a smaller
size" but, involves slip between collagen fibrils or fibers.

Harkness found that there was a decrease in weight of the cervix
after parturition with a smaller decrease in collagen resulting in an
overall rise in the concentration of collagen. After 24 hours, a
slower decline in circumference occured and was accompanied with a
decrease in weight and total collagen content. By the eighth day
after parturition, the circumference of the cervix droped from 1/3 its
initial value to 1/8 that of parturition taking until the 16th day to
fall to its normal size.

Collagen, on the other hand, did not begin to be lost after
parturition for about 24—48 hours after which there was an exponential
decay in collagen content for approximately two days at which time it
began to slow down but decreased to a below normal value for the
nonpregnant cervix. Harkness concluded that collagen was lost by
reabsorption which was in keeping with the observation that the

epithelial layer of the endometrium almost completely covered the

17
inside of the uterus within 24 hours of parturition. He noted that
this change in collagen content resembled the contraction of wound
tissue. The load required to break the tissues rose after parturition
and then fell. Harkness believed this was due to the fact that the
decline in circumference of the cervix proceeded faster than the loss
of tissue at first, so the walls became thicker needing a greater
breaking load. Later there was a loss of tissue , the walls becoming
thinner, consequently less load was needed from a breaking tension.
Breaking load per unit cross—sectional area of collagen varied little
through the whole period after parturition (20).

Harkness and Harkness (21) performed three different experiments
using rats to examine the physical properties of the cervix during
pregnancy. One consisted of applying a constant load of 509 for 30
minutes at 22°C at various stages of pregnancy. The second involved
subjecting the tissues to a load until they ruptured at both 22°C and
37°C. Finally, in the third experiment, the tissues were subjected to
various constant loads for periods longer than 50 minutes.

In the nonpregnant rat, the cervix was only slightly distensible
even by relatively large forces acting for long periods. At the end
of pregnancy, however, the cervix was very viscous distending slowly
and progressively under low forces. The change in the physical
properties of the cervix did not begin until the 11 — 12th day of
pregnancy after which progressive changes took place resulting in a
decrease in the amount of smooth muscle present and a fall in the
concentration of collagen.

In their experiments, Harkness and Harkness found that temperature
had no effect on the tissues. The circumference of the cervixes from

21-day—pregnant rats was 4-5 times larger than the earlier stages but

18

differed very little otherwise. They related load to change in
circumference and found rupture strengths of 0.5 MPa for non-pregnant
animals to 0.3 MPa at term with the rupture strength higher at
midpregnancy. The stiffness of the cervix was found to be 120 MPa for
nonpregnant decreasing to 40 MPa at term. These values are higher by
a factor of 100 than the values found in this research.

When Harkness and Harkness (19) loaded the tissue for a prolonged
period of time, they found that in the nonpregnant and 12—day animals,
the creep rate of the cervix diminished continuously until the end of
the experiment. They found the creep to be approximately a linear
function of the natural logarithm of time. On the other hand, with
cervixes from 21—day—pregnant rats the creep rate did not diminish
continuously with time but became a constant. Therefore, the two
major differences they found between the pregnant and nonpregnant
cervix were; firstly, the tissue had a larger circumference, and
secondly, the pregnant cervix behaved in a purely viscous manner.

Harkness and Harkness tried to model the response of the cervix by
using a rheological model with a single viscous and elastic element in
parallel, combined in series with a viscous element. They found
however, that this model did not fit the curves well.

Other mathematical models of the cervix should be briefly
mentioned. Kriewall (22) did a study on the uterine work in
parturition. He related work expended by the uterus to the
intrauterine pressure and cervical dilatation. Since contractions
during labor dilate the cervix through the mechanics of retraction, to
accomplish this work must be expended by the uterus. Considering
cervical contractions to be a quasi-equilibrium process, the work of

dilating the cervix can be calculated using the integral of

19

pressure-volume, PdV, where pressure was measured and the change in
volume was calculated from the change in cervical dilatation. Some of
the energy of contractions which was not used to dilate the cervix was
lost to hysteresis effects, in rearranging the molecular structure and
to the nonelastic characteristics of the cervical tissue. He
concluded that the net amount of work expended by the uterus to dilate
the cervix equals zero because the elastic forces of the cervix
exerted the same amount of work to restore the uterine volume as was

needed to decrease the volume.

HISTOLOGY OF THE RABBIT CERVIX

The rabbit cervix is similar to the human cervix except that it is
a double cervix. It is composed of glandular connective tissue and
muscle fibers with epithelium lining the canals. The nonpregnant
rabbit cervix will be considered followed by a look at the changes the
cervix undergoes throughout pregnancy. Figures 1—3 are
cross—sectional slides of the nonpregnant cervix at magnifications of
10X, 50X, and lOOX respectively. The tissues were stained with a
Hexachrome stain; the connective tissue staining yellow-green, the
muscle fibers light pink, and the epithelium dark pink.

Unlike the human cervix with its longitudinally orientated muscle
fibers, there are two distinct orientations of the muscle fiber
bundles in the rabbit cervix. These can be seen in the light pink
areas of Figures 1—3. The outermost muscle fiber bundles are
longitudinal in orientation running from the uterus to the vagina,
parallel to the cervical canal, similar to that of the human cervix.
Just inside this ring of muscle is a second ring of muscle fibers
circumferentially orientated. The muscle fiber bundles are surrounded
with loose connective tissue which is more dense around the muscle
bundles, becoming looser in the cervical folds. This is in accordance
with Danforth and Hughesdon's descriptions of the human cervix. The
cervical folds are devoid of all muscle.

Like the human cervix, the connective tissue of the rabbit cervix

20

21

consists mostly of loosely intermixed collagen fibers having no
apparent preferred orientation. Figures 1—3 show the yellow—green
collagen fibers intermixed among the muscle bundles as well as in the
cervical folds.

The dark elastic fibers are scarce, found mainly between muscle
bundles and in the walls of blood vessels. Reticular fibers are also
present in small quantities. They are found in association with
muscle bundles, particularly around the boundaries between the
connective tissue and other tissue types. They are continuous with
the collagen fibers, with a gradual transition from one to the other.

Electron microscopy studies were also performed on the rabbit

 

cervixes to examine tissue morphology. All specimens were fixed in a
modified Karnofsky's fixative medium of 2.50/0 Paraformaldehyde,

30/0 glutanaldehyde in a 0.1 M phosphate buffer rinse and then put
through a 10 step dehydration series from 100/0 to 1000/0 Ethanol;
freeze fracturing for some samples was done at this time. After
Critical Point Drying, the samples were gold coated and examined with
a 131 Super III Scanning Electron Microscope.

Figure 4 is a cross-section of the nonpregnant cervix at a
magnification of 23.4X. The cervical canal is located at the lower
right with the orientation of the canal running into the micrograph.
Three rings appear in the photograph; the inner ring including the
cervical folds consists of connective tissue and an epithelium lining;
the second middle ring consists of circumferentially orientated muscle
bundles intermixed with collagen fibers; and finally there is an outer
most ring, in the upper left hand corner, consisting of longitudinal

muscle fiber bundles intermixed with collagen fibers. Figures 5-8 are

 

 
 

 

 
 

 

22

 

 

Figure l. Hexacrome stained cross—section of the nonpregnant cervix
(10X)

 

Figure 2. Hexacrome stained cross-section of the nonpregnant cervix
(50X)

23

 

Figure 3. Hexacrome stained cross—section of the nonpregnant cervix
(lOOX)

 

Figure 4. Electron microscopy of the nonpregnant cervix (23.4X)

24

magnifications of these rings.

Figures 5 and 6 show a cervical fold from the inner ring. The
randomness in orientation of the collagen fibers is apparent. Figure
7 shows the middle ring. Again the randomness of the collagen fibers
is apparent. It is difficult to distinguish between the collagen and
muscle fibers in these photomicrographs. They can be distinguished
readily, however, on the histological photographs. Figure 8 shows the
outermost ring with similar randomness of collagen fibers.

According to many scientists, during pregnancy the cervix becomes
increasingly edemic. The collagen fibers dissociate, branching into
smaller intermixed fibers. With the dissociation of collagen, the
amount of muscle appears to increase although in fact, the quantity of
muscle doesn't change. The next set of figures shows the cervix at
different stages of pregnancy.

Figure 9 is a slide from a rabbit cervix 23 days pregnant. It is
a hematoxylin-eosin stain. The mucus secreting epithelium stained
dark brownish-black and is seen around the edges of the cervical
folds, lining the cervical canal. The increase in edema can be seen
in the increased yellow spaces between the brownish collagen fibers.
The muscle bundles are stained reddish-brown. Circumferential bundles
can be seen in the upper right-hand and lower left—hand corners.

Figure 10 is a cross-sectional slide of a rabbit cervix at 33 days
(term) pregnant. The collagen fibers are very dissociated, with very
few muscle bundles present.

The next series of figures is taken from a 1 hour post partum
cervix. One horn of the cervix was pregnant and the other was not

enabling us to make a visual comparison of the effects of pregnancy.

25

 

Figure 5. Photomicrograph of a cervical fold from the inner ring.(78X)

 

Figure 6. Photomicrograph of a cervical

fold from the inner ring.(546X)

26

 

Figure 7. Photomicrograph of the middle ring. (780X)

 

Figure 8. Photomicrograph of the outter ring. (7OOX)

27

 

Figure 9. Hematoxylin-eosin stained 23 day pregnant rabbit cervix.

 

Figure 10. Cross—section of a 33 day (term) cervix.

28

Figure 11 is a hematoxylin-eosin stain showing the two horns; the
nonpregnant horn is on the right. The difference in size between the
pregnant and nonpregnant horn is obvious, as well as the dissociation
of the collagen fibers. Figure 12 shows the degree of dissociation of
collagen fibers in the pregnant horn. Figure 13 is a section from the
nonpregnant horn; the bright green collagen fibers are much denser.

Figure 14 is a slide taken from a rabbit cervix 48 hours post
partum. The collagen fibers have undergone considerable
reconstruction.

Although the rabbit cervix structure is not quite the same as the

 

human cervix, it undergoes the same changes throughout a term of
pregnancy. From a very compact nonpregnant state, through a very
dissociated pregnant state, and then back to a compact post partum
state, the cervix is able to support and deliver without undergoing

massive destruction.

29

 

Figure 11. Hematoxylin-eosin stained pregnant/nonpregnant rabbit cervix.
Left horn 1 hour post partum; right horn nonpregnant.

"37/
v? ' '
. _ 5‘ 4'...»

L ..
1:, ,..

 

Figure 12. 1 Hour post partum left horn of the rabbit cervix.

30

1
Ti“

13‘ r‘.
t“ , 1.. .1.

 

Figure 13. Nonpregnant right horn of the rabbit cervix.

 

Figure 14. Hematoxylin—eosin stained 48 hour post partum rabbit cervix.

EXPERIMENTAL PROTOCOL

In order to determine the mechanical properties for the rabbit
cervix, the specimens were subjected to three different types of
tests. Relaxation tests were conducted at various levels of
deformation. Loading and unloading the specimen at constant
deformation rates from 0.5 to 100 Olo strain/second was used to
examine hysteresis and strain rate effects. Finally, the responses of
the cervix to sinusoidal extension tests were measured.

Female New Zealand rabbits from six months to a year—old were
used. Cervixes were tested from nonpregnant rabbits, pregnant - 20
days, pregnant - 24 days, and pregnant-33 days (term), and 48 hours
post partum. The rabbits were sacrificed with chloroform and
dissection performed immediately. The rabbit was shaved and a lower
abdominal incision used for entry. The cervix was located, removed at
the isthmus and all extra tissue surrounding the cervix was dissected
away. The tests were run on one side of the cervix with the other
side intact. A drip of 0.90/0 saline solution was used to keep the
tissues moist throughout testing.

Two yoke—shaped grips were used to mount the cervix in the testing
machine. Two semi—cylindrical pins, 0.048 inches in diameter were
slid through the cervical canal and mounted in the slots of the
grips. Care was taken not to damage the tissue and the grips were

positioned so that there was no initial circumferential extension of

31

32
the cervical canal. The initial dimensions of the cervix were
measured with a micrometer before it was mounted in the testing
fixture and the initial length was measured when a 0.1 N load was put
on the tissue.

An Instron Materials Testing Machine was used for the tests with
an Interface Model SSM-25 load cell. Specimen displacement was
measured with a Linear Variable Differential Transformer within the
hydraulic actuator of the Instron. Both load and displacement
transducer signals were conditioned by amplifiers in the Instron
instrumentation. The load and displacement of the specimens were
recorded on a Nicolet 2090 Explorer Digital Oscilloscope, and load vs.
time, displacement vs. time and load vs. displacement cervis were
plotted using a Varian F-80 X-Y recorder.

The following test protocol was followed. Initially the cervix
was 'preconditioned' using a loading and unloading triangle function
at a rate of 0.5 Olostrain/second. Three such cycles resulted in a
repeatable response in the tissue. After preconditioning, the
specimens were loaded and unloaded using the single cycle triangle
wave function at the same level of deformation varying the rates from
0.5 to 100 0/o strain/second to study the rate dependence of the
specimen. The second type of test, sinusoidal extension, was
performed at the same level of deformation for rates from 0.01 to 1.0
Hz. The last test was a relaxation test in which the specimen was
loaded to the same level of deformation at 100 0lostrain/second and

held for 100 minutes.

RESULTS AND DISCUSSION

A typical stress-strain curve is shown in Figure 15. This curve may
be divided into three regions. The first, is where the low slope of
the stress—strain curve represents straightening and alignment of
fibers. The second region is the transition region before the third
region where the response is linear with a higher tangent modulus.

The effect of strain rate on the shape and magnitude of the curves
is shown in Figures 16 and 17. As the strain rate increases, the toe
region decreases and the curves shift towards the stress axis.
Viscoelastic effects account for this phonomena; at the slower strain
rates, the tissue stress levels have a chance to relax resulting in a
longer toe region and lower peak stress values.

A computer routine to emperically fit the stress-strain response
curves in the form<5eA8?'was used. The constant A was determined for
each varying strain rates throughout pregnancy. Table 1 shows a
comparison of the average values of A at different strain rates for
each stage of pregnancy. Since the tissue was preconditioned at
0.50/0 strain/sec for three cycles, the first three columns indicate
the A values for the three preconditioning cycles. Two trends exist;
the first involves the rate dependency of the tissue. With the
increase in strain rates from 0.50/0 strain/sec to 1000/0

strain/sec there is an increase in the parameter A for all stages of

33

34

 

ao>eso :Hmeumlmmoeum Hmowmxp < .mH oesmflm

-OH x :pmmuw om om OH

>

N

 

O
l\
O
\O
0
“Ln
0
“V

_ . l n _

 

 

:m
x E z mmoepm
fleofi N \ V

 

.o>H:o cemeum mmoepm exp :0 muoommo ouwe :Hmaum .ofi mesmflm

NIOH x :Hmeum on 00 om ow om om 0H

wcooom\:flmeum wm.o

ncooom\:flmeum woofi

m
x E WWoeu
fleofi N \zv m

 

36

.mo>H:o :flmepmummoepm ocu co mpoommo puma :«meum .NH mesmflm

 

 

 

 

 

NIOH x Cflmhum om cm on 00 om ow om ON OH
x F. i. T + a .i e w u ” Immwg-

LT H

oam.o I. N
wm
on
we.

m

 

oom\efla paw oofl neofi x Ne\zv mmaabm

37

pregnancy, indicating that the strain rate at which the tissue is
loaded affects the loading characteristics of the tissue. The cervix
has a much stiffer response at the faster strain rates. The second
trend in the data corresponds to the changes in the value of A with
the change in the stage of pregnancy. As the rabbit nears term,
approximately 30-33 days of its pregnancy, the value of A decreases
indicating a loss in the stiffness of the cervix. However, 48 hours
post partum the cervix regains some of its initial nonpregnant
qualities, increasing its abiblty to carry higher loads at lower

levels of strain.

 

Table 1: Average A values (N/m2 x 104)

Stage of O/oStrain/Second
pregnancy 0.5 0.5 0.5 1.0 2.0 10.0 50.0 100.0

Non 1.90 1.80 1.75 2.15 2.80 2.60 3.45 4.00
24 days 1.63 1.20 1.10 1.23 1.43 1.45 1.58 1.75
30 days(term) 1 35 0.95 0.85 1.05 1.50 1.25 1.25 1.35
48 hours 2 5 1.90 1.75 1.75 2.07 1.20 2.50 3.25

post partum

To further investigate this change, terminal slopes were
evaluated on the loading cycle of the stress—strain curve. Table 2
shows a comparison of these values for each stage of pregnancy at
varying strain rates including the preconditioning cycles. At the
same strain rate, there is a decrease in the value of the terminal
slope with the increase in pregnancy followed by a recovery of the
slope in the post partum cervix. The slope values at 1.0 0/o
strain/sec decrease from 11.11 x 104 N/m2 in the nonpregnant

rabbit to 1.9 x 104 N/m2 in the cervix at term, with a recovery in

the post partum cervix to 4.33 x 104 N/mZ. These are strong

 

indications that pregnancy results in a loss of stiffness in the

38

cervix. Harkness and Harkness (21) reported a similar loss but their
stiffness values are much higher. This may be due in part to the
manner in which the loading was applied across the two canals.

Table 2: Terminal Slope values (N/m2 x 104 )
Stage of 0/oStrain/Second

pregnancy 0.5 0.5 0.5 1.0 2.0 10.0 50.0 100.0

Non 16.67 14.00 22.50 11.11 15.00 16.43 21.00 37.50
20 days 15.38 13.89 12.06 10.83 17.33 19.55 22.73 ——-—
24 days 3.93 2.07 2.98 2.11 3.34 2.59 3.78 5.07

33 days(term) 4.00 2.62 2.23 1.90 1.59 4.59 2.14 2.00
48 hours 2.67 3.54 2.38 4.33 5.00 6.67 7.50 10.00

post partum

Table 3: Hysteresis values (N/m2 x 102 )
8: energy absorbed during loading C=energy dissapated in hyst. loop

 

Stage of O/oStrain/Second
pregnancy 0.5 0.5 0.5 1.0 2.0 10.0 50.0 100.0
Non B 14.82 8.23 7.65 7.53 9.40 13.29 12.35 14.23
C 26.94 20.69 13.53 12.00 16.68 21.17 22.34 22.35
20 days 8 5.36 9.65 4.12 10.00 4.82 4.88 6.35 8.18
C 12.07 17.54 8.94 4.65 10.88 11.99 13.82 18.06
24 days 8 2.06 1.77 1.83 3.06 2.77 2.35 3.18 5.71
C 5.86 4.30 6.89 5.89 6.06 6.34 6.76 7.83
30 days 8 1.77 .826 2.47 3.12 2.82 3.06 4.24 4.38
(term) C 4.18 4.01 6.96 5.60 7.74 8.24 5.06 6.96
48 hours B 13.06 6.47 9.18 5.52 9.76 7.88 8.59 14.59
post partum C 22.59 11.29 17.65 9.16 16.11 16.11 18.24 18.24

 

.mmooH memoaoumxc cap :0 ope» :Hmpum mo muuomwo ose

N-OH x efiaabm OR OO Om

.wH ohzwfim
ow om om

_

OH

.aam\:flaapmw m.O
3

.Oam\:flmhpmw OOH

m
X E mmoh
AOOH N \zO pm

40

With the decrease during pregnancy in stiffness in the loading
curves, there is also an effect on the hysteresis response of the
tissue. Hysteresis is affected by both strain rate and pregnancy.
Figure 18 shows a variation between two hysteresis curves. Table 3
compares the energy absorbed during loading, B, and the absolute energy
dissapated in the hysteresis loop, C, at varying strain rates and the
different stages of pregnancy. Comparing the different strain rates
for each stage of pregnancy further indicates the rate dependency of
the tissue. Pregnancy as well as strain rate affects the response of
the tissue. There is a decrease in both the input energy and the

absolute energy dissapated in the cervix with increasing stages of

 

pregnancy. The decrease in stored energy is consistant with the loss
of stiffness in the tissue with pregnancy. The post partum cervix
indicates a gain in both energies B and 0 consistent with an increase
in stiffness of the tissue. The ratio of C to B remained constant at
approximately 500/0 for different stages of pregnancy and different
strain rates.

Tables 1,2 and 3 are indicative of the trends in the cervix as a
result of pregnancy and strain rate. Values have been averaged between
rabbits at a particular stage of pregnancy to show trends in the data.
Variations in the data are within experimental scatter in biological
tissues.

Relaxation tests have been evaluated for the cervix at various
stages of pregnancy. A typical relaxation plot of normalized stress
vs. log. of time is shown in Figure 19. Stress values after 6 seconds
of relaxation and the slope of the relaxation over 100 minutes.
(Acrfist/Alog t were measured for all stages of pregnancy. Variations

in this parameter between stages of pregnancy were within scatter of

41

.u .moH .m> mmoepm woNHHmEHoc mo Hon :oHummeoH Hmonxp < .mH oesmHm

HaeHHO .OOH

 

OOOO OOOH OOH OH H H.
1111* "3 ”"H_r .H 3314.1. H IHIHHH H 1::wa a

.N.O

.O.O

i©.O

.O.O

O.H

 

Hume\ eoO maaHHm OONHHaEHoz

42
the data and therefore not meaningful. An average value of the slope
of the relaxation curve was —2.8 x 104/log sec . Changes in this
viscous parameter as a result of pregnancy were not significant.

Cyclic data shows definite trends with pregnancy. In Figures 20
— 27, average peak stress for each cycle versus number of cycles was
plotted keeping frequency and stage of pregnancy constant. In figures
20 —24 a comparison of the frequency is made for each stage of
pregnancy. There is a more rapid decrease in peak stress for each
cycle in the lower frequency tests. This is not necessarily an
indication of frequency or strain rate dependence. If the response of
the cervix is a time phonomena and these curves were compared on an
absolute time basis instead of number of cycles, the higher frequency
data would collapse on the lower frequency data. The rate of decay of
the peak stress in the high frequency data is incorporated in the
beginning cycles of the lower frequency data.

In Figures 25—27, each stage of pregnancy is compared at a
constant frequency. There is a decrease in the stress amplitude with
the number of cycles. At later stages of pregnancy this decrease is
greater than in the earlier stages of pregnancy and in nonpregnant
rabbits under cyclic deformation. The peak stress decay in the post
partum cervix resembles that of the nonpregnant cervix indicating a
recovery from the response of the pregnant cervix.

Using this test protocol, changes in the cervix due to pregnancy
were determined. Constant strain rate loading and unloading tests
showed a strain rate dependency of the tissue at each stage of
pregnancy. Terminal slopes of the loading curves indicated a decrease
in stiffness of the cervix as it neared term followed by a recovery in

stiffness post partum. Hysteresis area was also found to decrease with

 

43

Figure 20. 0/o of peak stress vs. number of cycles for the non
pregnant rabbit cervix at frequencies of 1, 0.1 and 0.01 Hz

44

ow

moHoxu mo Hosesz

+-

.ON OHOOHO
OH O H

om

00

on

em

.N: Ho.o

 

.Nz H.O
.N: H , .

 

 

 

 

mmoepm xmom mo

 

o\°

45

Figure 21. 0/o of peak stress vs. number of cycles for the 20 day
‘pregnant rabbit at frequencies of 1, 0.1, and 0.01 Hz.

46

 

moHoxu mo eonssz

.HN OHOMHO

 

 

ow mm om mm om mH 0H m H
T11 I +1 H H + a 2
1-0m
to
4dn
.N: Ho.o

iOO

N: H.O

.N: H
lxom
mmohum xmom mo w

47

Figure 22. 0/o of peak stress vs. number of cycles for the 24 day
pregnant rabbit at frequencies of 1, 0.1, and 0.01 Hz.

 

48

 

 

 

moHoxu mo HofiEdz .NN oeswwm
ow . mm om mN ON mH OH m H
H H H H T n __ w
tom
Ioo
ion
.NI Ho.o
NE H 0 low
.N
I H Tom

mmoepm HOBO

 

m

 

49

Figure 23. 0/o of peak stress vs. number of cycles for the 33 day
(term) rabbit at frequencies of 1, 0.1 and 0.01 Hz.

50

CV

 

 

 

 

moHoxu mo Honezz .mm ohswwm

mm om mm om mH 0H m H
1.0m
TOO
.N: Ho.o , 1-05

.N: H.O
.N: H

2-8
Irom
mmohum xmom mwzw

 

 

 

 

51

Figure 24. 0/o of peak stress vs. number of cycles for the 48 hour
post partum rabbit at frequencies of 1, 0.1 and 0.01 Hz.

52

0%

 

 

 

 

moHoxu mo Honesz .VN oestm

mm om mm om mH 0H m H
43
ice
t:

.N: HO .O .
row

.N: H.p.
.N

I H t¢m

 

mmoepm xmom Ho &

53

Figure 25. 0/o of peak stress vs. number of cycles for a frequency
of 0.01 Hz. for all stages of pregnancy.

54

ow

 

moHoxu mo Honesz mm OHSMHH

v

 

 

 

r mm om mm (Om OH OH m H
1.0m
1.00
ucwcwoem mxmw mm :11 {on

pcmcmoem mxmw om ucmcmo m mxmw vm

ESHHOQ pmom meso: we cm

L

pcmcmoem c.‘

:om

 

mmoppm xmoa my w

55

Figure 26. 0/o of peak stress vs. number of cycles for a frequency
of 0.1 Hz for all stages of pregnancy.

56

 

 

 

 

 

 

 

 

 

 

moHoxu mo Honesz .om oesmHm
ov mm om mm om mH 0H m H
T) H w H H H H H
1-0m
1-00
ucwcmoem mxwm mm ion
passwoem mxmw vm l-ow
Hcmcmoem mxmv om
Eduewm pmom mesa: we

Hcmcwonm :oz ..om
mmoepm Hmom mo w

57

Figure 27. 0/0 of peak stress vs. number of cycles for a frequency of
1 Hz for all stages of pregnancy.

 

moHoxu Ho Honesz .ON oHSme

 

 

 

 

 

 

ow mm om mm om mH oH m H
T) I H H w H H H
+Om
too
00
5
:on
ucwzmoem mxww mm
L.om
ezpemm umom Hcmcmoem mxww om
meson we
pcwcmwmm mxww vm tom

 

ucmamoem :oz

 

mmoHpm xmom mo

o
Ac

59

the increase in pregnancy and recover post partum. Relaxation tests,
however, showed very little change due to pregnancy indicating that
viscosity of the cervix would not change with pregnancy. Finally,
cyclic deformation tests did not show a frequency dependence but very
definitely showed a decrease in stress amplitude with the increase in
pregnancy followed by a recovery with the post partum cervix.

Unlike Harkness and Harkness (18, 18, 19) who show an increase
in the viscous parameters in the cervix, the relaxation data reported
here does not indicate an increase in viscosity during pregnancy but
cyclic loading does show increases in the viscoelastic response of the

cervix. The maximum tangent modulus data indicates a loss in stiffness

 

during pregnancy as discussed earlier. A creep test was not run and it
is difficult to calculate reciprocal comparisons in nonlinear tissues
between creep and relaxation tests. However, the lack of change due to
pregnancy in the relaxation data would indicate that one might not see
changes in creep response with pregnancy. The cyclic data were
deformation cycles and not force cycles but if the tissue were
subjected to force cycling, one would expect to see the deformation
amplitude increase giving an increase in the compliance.

Several problems were encountered during this research which
made the data not as conclusive as it could have been. A combination
of five load cell set ups were used to measure very low loads in
attempting to obtain an acceptable degree of accuracy. The sensitivity
of the Instron Materials Testing Machine varied with different load
cells and a comparison of data from different specimens was not always
possible.

Another problem encountered was in the measurement of the

initial length, 10, of the specimen. There was a problem detecting

60
initial load on the tissue and it was decided that lO would be
measured when a 0.1 Newton load was placed on the tissue. As the stage
of pregnancy increased, so did l0. Since the cervix behaves with a
strongly viscous component, lO had to be continually increased in
order to hold the initial load. This variation in TO could affect
the initial linear range on a stress-strain curve possibly making a
more compliant tissue appear stiffer than a less compliant tissue.

The amount and orientation of the components of the cervix
contribute to the response of the cervix to mechanical testing. The
rabbit cervix is mainly composed of connective tissue and muscle
fibers. Collagen is thought to be the main load bearing component of
the cervix. The muscle fibers are imbedded in the collagen fibers
therefore probably do not serve as a significant load—bearing component
in the cervix. Since elastin exists in such small quantities it is
believed to have no roll in the mechanical response of the cervix both
in the pregnant and nonpregnant rabbit. The contribution of the
reticular fibers is neglibible compared to that of the collagen fibers.

The changes the cervix experiences histologically as a result of
pregnancy suggest reasons for changes in the mechanical properties of
the cervix due to pregnancy. In the nonpregnant state, the cervical
collagen fibers of the rabbit are coarse and intermixed having no
preferred direction. In the pregnant cervix, however, they become fine
and highly branched indicating that the amount of load or the stiffness
the cervix withstands during pregnancy would diminish as the rabbit
nears term due to this change in cross—linking within the individual
collagen fibers. The increasing edema and glycosaminoglycans in the
cervix with pregnancy also suggests an increase in the viscoelastic

response of the cervix.

61

The three mechanical tests in this study indicate changes in the
response of the tissue with pregnancy which are consistant with the
microscopic changes in the cervix. Constant strain rate loading and
unloading tests indicate both a strain rate dependence and a loss in
stiffnes due to pregnacy followed by a recovery in the 48 hour post
partum cervix. Terminal slopes also indicate both a strain rate
dependence and a loss in stiffness due to pregnancy followed by a
recovery in the 48 hour post partum cervix. Hysteresis indicates a
strain rate dependence and a decrease in area with pregnancy,the area
recovering in the 48 hour post partum cervix. The relaxation tests
show very little change with pregnancy indicating little change in the
viscous parameter in the tissue. Finally, the cyclic data shows no
frequency dependence, a decrease in the stress amplitude of each cycle
with pregnancy and an increase in the viscoelastic response of the

cervix with pregnancy.

APPENDIX A

 

APPENDIX A

The classical description of the uterus divided it into several
parts: the fundus being the uppermost portion; the body or corpus
lying above the internal 05; the isthmus lying between the internal 0s
and the external (histological) es; and the cervix lying below the
external 03 (see Figure A—l). The internal os is the constriction

where the corpus meets the isthmus, and the external os is the

 

transition between endocervical and endometrial epithelium. Danforth
(23) eliminated the isthmus as a separate structure indicating that
the corpus was chiefly the muscular part of the uterus and the cervix
was the fibrous part of the uterus, the isthmus or internal 05 being
the junction between the two.

The nulligravida cervix is a cylinder—shaped organ tapering at
both ends about 2.5 — 3.0 cm in diameter. The uterus is superior to
the cervix and is located in the pelvic cavity above the bladder and
in front of the rectum. The cervix is positioned slightly downward
and backward from the uterus. Since it is less freely moveable than
the uterus, the long axis of the cervix is seldom in line with that of
the corpus. At the transition between the corpus and cervix, the
isthmus or internal os , the uterus has the form of a sphincter.

The cervical canal crosses the cervix and communicates between the
uterine cavity above the internal os and the vaginal canal below the
external 05. It is broader in the middle than at either end. (12, 24,

25)
62

 

 

 

 

64

The human cervix, like the uterus, consists of an endometrium
(mucosa) and myometrium; the mucosa rests upon the myometrium which is
predominantly composed of fibrous connective tissue with smooth muscle
present in small quantities and elastic tissue is virtually
nonexistent. The corpus of the uterus is composed primarily of
muscular tissue. There is only about a 5 — 10 mm space from which the
fibrous cervix transcends from the muscular corpus. Smooth muscle
comprises about 150/0 of the cervix and elastic tissue is present
only on the walls of blood vessels (23).

In the cervix there are two types of epithelium present which line
the endometrium, stratified squamous and columnar. The epithelium
lining the mucosa of the cervix differs according to the cavity it
faces. The portio vaginalis is covered by stratified squamous
epithelium whereas the mucosa of the endocervical canal is covered by
a mucous membrane composed of tall columnar cells, some of which are
ciliated. The cells are uniform, arranged in single rows and actively
secrete mucin. The squamocolumnar junction, or external OS, is the
boundary between the squamous epithelium and columnar epithelium,
occuring over a region of 1 — 10 mm. (12,24,25)

Glands, also present in the cervix, extend from the surface of the
cervical canal to the underlying connective tissue. They too are
lined with columnar epithelium. (12,24,25)

The major thickness of the cervix, the myometrium, consists of
fibrous connective tissue with moderate amounts of smooth muscle
interwoven. The arrangement of the smooth muscle and connective
tissue is important in order to understand the mechanical response of

the tissue. The amount and orientation of the smooth muscle in

 

 

65

relation to the connective tissue varies throughout the cervix being a
subject of much debate. The most distal portion of the cervix, the
portio vaginalis, is devoid of muscle, consisting mainly of connective
tissue; the middle portion contains the terminal fibers of central
longitudinal uterine muscle; and the upper portion contains 50 —

600/0 of the smooth muscle (23,12). Danforth (26) and Hughesdon

(11) described strands of smooth muscle in the cervix that were
continuous with the muscle of the corpus superiorly, and inferiorly
with the walls of the vagina. They believed that the cervix consisted
of an outer one—fourth which is mainly muscular and an inner
three—fourths which is predominantly connective tissue. From saggital
sections, Hughesdon (11) noted a high muscle content of immature
muscle fibers in a submucous layer in which the intrinsic muscle is
present in radial and longitudinal bundles which lie in a mass of
dense collagen. Hughesdon concluded that the cervix consisted of an
outer contractile layer of muscle embedded in a mass of collagen
making contraction difficult, and an inner noncontractile mass. He
also noted a mass of circular smooth muscle at the internal 05
arranged in a spiral pattern. It has not been established whether or
not this mass of muscle has sphincteric capabilities.

Danforth (26) questioned the ability of the cervix to contract or
whether the contractions were transmitted from the corpus or were
otherwise extracervical. He tested the ability of the cervix to
contract and relax by suspending excised strips of cervical tissue in
a bath and recording their activity. He found the ability of the
cervix to contract was negligible when compared to that of the corpus.

The major portion of the cervix involved in mechanical activity is

66

connective tissue. The most dominant constituent of the connective
tissue of the cervix is collagen. Collagen undergoes the major
changes within the cervix during pregnancy and these changes will be
considered later. Collagen fibers in the nonpregnant cervix appear as
dense tightly woven bundles of interlacing fibers running parallel to
one another (3).

Collagen posseses hydroxyproline, an amino acid which is unique to
it alone. The hydroxyproline concentration in collagen is constant at
13 or 140/0 dry weight regardless of the physical characteristics
from which the collagen was extracted (6). The amount of
hydroxyproline found in a tissue may be used to determine the amount
of collagen present in the tissue. In the case of the collagen in the
cervix, hydroxyproline can also be measured in the urine.

The second major fibrous component of connective tissue is
elastin. Elastin fibrils are thinner than collagen fibrils, measuring
only about 70 A in diameter (6). The amount of elastin in the cervix
is insignificant, comprising a fraction of 10/0 of the total fibrous
tissue of the cervix. They are present, sparsely scattered randomly
throughout the cervix, being predominantly found in the walls of
larger blood vessels. Because of the insignificant amounts of elastic
fibers, their contribution to the mechanical response of the cervix is
negligible.

The other component of cervical connective tissue, the reticular
fibers, are fine threadlike structures similar to immature collager‘
fibers which are found around the bundles of collagenous fibers af“j

around the smooth muscle cells. Some studies have shown reticular

 

67
fibers to be continuous with collagen fibers (6). In the nonpregnant
cervix, the reticular fibers appear as short segments, very
irregularly despersed or clumped in a haphazard relation to the
collagen bundles. In the pregnant cervix, reticular fibers are more
robust and, in general, more luxuriant and tend to parallel the

collagen fibers (4).

 

LIST OF REFERENCES

10.

11.

LIST OF REFERENCES

Lingren, L. (1973) Biodynamics of the Cervix during Pregnancy
and Labor. The Biology of the Cervix., pg. 385-415.

 

Philpott, R.H. (1976) Biodynamics of the Cervix. The Cervix.,
pg. 217—250.

 

Buckingham, J.C., Selden, R., and Danforth, D.N. (1962)
Connective Tissue Changes in the Cervix during Pregnancy and
Labor. Annals, New York Academy of Sciences, 97, pg 733-742.

Danforth, D.N., Buckingham, J.C., and Roddick, J.W. (1960)
Connective Tissue Changes Incident to Cervical Effacement. Am.
J. Obst. and Gynec., 80, 5, pg 939—945.

Ichijo, M., Shimizu, T. and Sasai, Y. (1976) Histological
Aspects of Cervical Ripening. Tohoku J. Exp. Med., 118,
pg 153-161.

Danforth, D.N. and Buckingham, J.C. (1963) Connective Tissue
Mechanisms and their Relation to Pregnancy. Obst. a Gynec.
Survey, pg 715-732.

Danforth, D.N., Veis, A., Breen, M., Weinstein, H.G.,
Buckingham, J.C. and Manalo, P. The effect of pregnancy and
labor on the human cervix: Changes in collagen, glycoproteins,
and glycosaminoglycans. Am. J. Obst. a Gynec., 120, 5,

pg 641—651.

Danforth, D.N. and Buckingham, J.C. The Effects of Pregnancy
and Labor on the Amino Acid Content of the Human Cervix. .Ihe
Biology of the Cervix, pg 351.

 

Silbert, J.E., In Prenez-Tamayo, R. and Rojkind, M. editors:
Molecular Pathology of Connective Tissues, New York 1973, Marcel
Dekker, Inc. pg 391ff23.

Groschel — Stewart, U., Hermann, U. and Schalwm, H., Br. J. of
Exp Pathol. 54: 192, 1973.

Hughesdon, P.E. (1958) The Fibromuscular Structure of the
Cervix. J. Obst. and Gynec., 65, pg 763—766.

68

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.
26.

69

Singer, A. and Jordan, J.A. (1976) The Anatomy of the Cervix.
The Cervix., pg. 13-35

Hafez, E.S.E. (1973) Comparative Anatomy of the Cervical
Stroma. The Biology of the Cervix. pg. 40

 

Bakke, T. (1974) Cervical Consistency in Women of Fertile Age
Measured with a New Mechanical Instrument, Acta. Obstet. Gynec.
Scand. 53, pg 293—302.

Lui, D.T., et al. (1975) Dilatation of the Parous Non—Pregnant
Cervix. BR. J. of Obst. and Gynec., 82, pg 246—251.

Rice, D.A., Yang, T.Y. and Stanley, P.E. (1975) A Simple Model
of the Human Cervix during the First Stage of Labor. J.
Biomechanics 9, pg 153-163.

Rice, D.A., Yang, T.Y. (1976) A Nonlinear Viscoelastic Membrane
Model Applied to the Human Cervix. J. Biomechanics 9,
pg 201-210.

Harkness, R.D. a Moralee, B.E. (1956) The Time Course and
Route Loss of Collagen from the Rat's Uterus during Post-partum
Involution. J. Physiol. 132, pg 502—508.

Harkness, M.L.R. a Harkness, R.D. (1958) Changes in the Disten—
sibility of the Uterine Cervix of the Rat After Parturition. J.
Physiol. 141, P9 34

Harkness, M.L.R. a Harkness, R.D. (1959a) Effect of enzymes on
mechanical properties of tissues. Nature, Lond., 183, pg 1821

Harkness, M.L.R. a Harkness, R.D. (1959b) Changes in the
physical properties of the uterine cervix of the rat during
pregnancy. J. Physiol. 148, pg 524—547.

Kriewall, T.J. (1976) Uterine work in parturition. J. Applied
Physiology, 41, 3, pg 316—322.

Danforth, D.N. (1947) The Fibrous Nature of the Human Cervngd
and its Relation to the Isthmic Segment in Gravid and NongraV1
Uteri. Am. J. Obst. and Gynec., 53, pg 541—559.

Kranz, K.E. (1973) The Anatomy 0f the Human Cervix, Gross an(j
Microscopic. The Biology of the Cervix., pg. 57—70.

 

Grays Anatomy. ”The Cervix Uteri". pg. 1357

Danforth, D.N. (1954) The Distribution and Functicnial Acti i‘tj/
of the Cervical Musculature. Am. J. Obst. and Gyniec. , 68. 5’
pg 1261—1269.

 

 

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

ll)ll11111211111)[111111111111(HIHIIIIH

3070 8527