ENDE: :5 EHLSfiOfi ' i HRITEC HEPERT ' ENE? PER. Lofe'Pft-i 5583' W “W rs _ fim . . u __,:......n. ‘ $5.». fl1Il‘l'lilcll’i'llell‘lTUlWlflll‘l’il'flflfiili[llllil'fiflflilllHI 3 1293 10595 6225 LI BRA R Y‘L‘f Michigan State . University This is to certifg that the thesis entitled MESENTERIC HEMODYNAMICS OF PERINEPHRITIC HYPERTE NS ION IN DOGS presented by Geza Simon has been accepted towards fulfillment of the requirements for Ph.D. Physiology degree in I | Major pgfessor ( l ‘ - Date ;//}f/7 L/ 0-169 ‘ l av HOAE & SONS’ 800K BINDERY INC. LIBRARY BINDERS ABSTRACT MESENTERIC HEMODYNAMICS 0F PERINEPHRITIC HYPERTENSION IN DOGS By Geza Simon To extend the investigation of the regional hemodynamics of experimental renal hypertension, mesenteric (ileum) blood flow and pressure relationships and mesenteric venous pressure-volume relation- ships were measured in fifty dogs. In addition, femoral vein pressure- volume relationships were studied in twenty-two dogs. In ten dogs (group H-l), one kidney was wrapped in silk eleven days before study; in fifteen dogs (H-Z), one kidney was wrapped four weeks, and the other kidney was removed two weeks before study; in eight dogs (H-3), one kidney was wrapped and the other kidney was removed four to twenty-four weeks before study; thirty-three dogs were prepared as normotensive controls (C-l, C-2 and 0-3). A significant rise (P‘<0.05) in mean arterial blood pressure occurred in groups H-l, H-2 and H-3 in the unanesthetized state. Under pentobarbital anesthesia (35 mg/kg, iv), blood flow and intravascular pressures were measured in the collateral-free, innervated, naturally perfused loop of ileum. Venous pressure-volume relationships were measured in isolated segments of mesenteric vein j__vivo and in excised segments of the superior mesenteric and femoral veins jg_vitro. 65;) ’70 Geza Simon 0 Veins from hypertensive and normotensive dogs were examined histologically and analyzed for their water, Na and K content. In the combined group of hypertensive dogs (H-l plus H-2), mesenteric (ileum) blood flow and intravascular pressures were increased (P‘<0.05) and total and segmental vascular resistances were normal (P:>0.05), compared to controls (C-l plus C-2). The venous pressure-' volume curves of early, gggrkidney (H-2) and chronic, gngfkidney (H-3), but not of early twgrkidney (H-l), hypertensive dogs were shifted in the direction of the pressure axis (P <0.05). Calculated compliances were also decreased (P< 0.05) in groups H-2 and H-3. While there were no histological differences, the veins of hypertensive (H-2 plus H-3) dogs had an increased water (P< 0.05), Na (P<:0.0l) and K (P<:0.05) content. This study is the first report of increased blood flow, accom- panied by normal vascular resistances, to a regional vascular bed in early experimental hypertension. Normal mesenteric vascular resistances, despite elevated intravascular pressures, suggest decreased vascular distensibility in hypertension. Vascular distensibility is decreased on both the arterial and the venous side of the circulation. Decreased venous compliance in early, ongfkidney (H-2) and chronic, 9n§:kidney (H-3) hypertensive dogs appears to be due to factors other than veno- constriction. This impression is based on the analysis of venous pressure-volume curves and the findings of increased water and electrolyte content of veins from hypertensive dogs. MESENTERIC HEMODYNAMICS OF PERINEPHRITIC HYPERTENSION IN DOGS By Geza Simon A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Physiology l974 To J anha ii ACKNOWLEDGMENTS I thank Dr. H. w. Overbeck for giving me the opportunity to come to Michigan State University and for suggesting the idea for this thesis. I also thank the members of my guidance committee: Dr. F. J. Haddy for his generosity and wisdom, Dr. C. C. Chou for his friendship, Dr. S. R. Heisey for his rigor, Dr. J. 8. Hook for his skepticism and Dr. J. B. Scott for his realism. ***** iii LIST OF FIGURES TABLE OF CONTENTS LIST OF TABLES .......................... SURVEY OF LITERATURE ................... Introduction ...................... Human Essential Hypertension .............. Cardiac Output ................... Blood Volume .................... Regional Hemodynamics ............... Vascular Compliance ................ Summary ...................... Experimental Renovascular Hypertension ......... Cardiac Output ................... Blood Volume .................... Regional Hemodynamics ............... Vascular Compliance ................ Summary ...................... Human Renovascular Hypertension ............ Systemic Hemodynamics ............... Blood Volume .................... Regional Hemodynamics ............... Vascular Compliance ................ Summary ...................... Spontaneously Hypertensive Rats ............ Systemic Hemodynamics ............... Blood Volume .................... Regional Hemodynamics ............... Vascular Compliance ................ Summary ...................... PURPOSE OF THE INVESTIGATION ............... METHODS ......................... Induction of Hypertension and the Preparation of Controls ..................... iv CHAPTER Page Mesenteric (Ileum) Blood Flow and Pressure Relationships .................... 5l Mesenteric (Ileum) Blood Flow and Perfusion Pressure Relationships ................ 56 Mesenteric Vein Pressure-Volume (Compliance) Relationships Ig_Vivo ................ 57 Superior Mesenteric Vein Pressure-Volume (Compliance) Relationships In Vitro ......... 62 Femoral Vein Pressure-Volume (Compliance) Relationships In Vitro ................ 66 Venous Wall Water: Sodium and Potassium Content . . . . 67 Histological Examination of the Mesenteric Vascular Arcade and the Femoral Vein ............. 68 Additional Studies ................... 68 IV. RESULTS ......................... 69 Mesenteric (Ileum) Blood Flow and Pressure Relationships .................... 7l Mesenteric Vein Pressure-Volume (Compliance) Relationships In Vivo ................ 75 Superior MesenteFTc Vein Pressure-Volume Relationships Ig_Vitro ................ 89 Femoral Vein Pressure-Volume Relationships In_Vitro . . l05 Venous Wall Water, Na and K Content .......... lO7 Histological Examination of the Mesenteric Vascular Arcade and the Femoral Vein ............. lll V. DISCUSSION ........................ ll7 Mesenteric (Ileum) Blood Flow and Pressure Relationships .................... l19 Venous Pressure-Volume (Compliance) Relationships . . . 125 VI. SUMMARY AND CONCLUSIONS ................. - 133 APPENDIX ............................. 135 BIBLIOGRAPHY ........................... 138 TABLE LIST OF TABLES Page General observations in control (C-l, C-2 and C-3) and in hypertensive (H-l, H-2 and H-3) dogs. Group means i SEM of awake blood pressures, body weights, blood urea nitrogen, urine specific gravity and number of days (range) of hypertension and qualitative urinary albumin (range) ...................... 70 Mesenteric (ileum) blood flow and pressure relationships under pentobarbital anesthesia. Group means:tSEM of weight of ileum, blood flow, pressures and resistances . . 72 Group mean-+SEM of in vivo mesenteric vein pressure- volume relationships and compliance (AV/AP) in control (C- l and C- 2) and in hypertensive (H- l and H- 2) dogs, during Injection Phase .................. 8O Group mean-+SEM of in vivo mesenteric vein pressure- volume relationships and compliance (AV/AP) in control (C- l and C- 2) and in hypertensive (H- l and H- 2) dogs, during Withdrawal Phase .................. 81 Body weights and vein weights (mean +SEM) in groups subjected to the study of in vitro superior mesenteric vein pressure- -volume relatian Hips .......... . . 90 Group mean +SEM of in vitro superior mesenteric vein pressure- -volume relationships and compliance (AV/AP) in control (C- l and C— 2) and hypertensive (H-l and H- 2) dogs, during Injection Phase ............. 9l Group mean +SEM of in vitro superior mesenteric vein pressure- -volume relationships and compliance (AV/AP) in control (C- l and C- 2) and hypertensive (H-l and H-2) dogs, during Withdrawal Phase ............ 92 Group mean-+SEM of in vitro superior mesenteric vein pressure-volume relationships and compliance (AV/AP) in control (C- 2) and in hypertensive (H- 2) dogs, during infusion at constant rate of 9. 88 ml/min ......... lOO vi TABLE 10. ll. l2. l3. l4. Correlation between weight of superior mesenteric vein (mean tSEM) and total volume (mean:tSEM) injected to reach 40 mm Hg intravenous pressure in groups H-2 and C-2 ......................... Correlation between weight of superior mesenteric vein (mean tSEM) and compliance (mean:tSEM) in pressure range of 5-15 mm Hg, during injection phase, in groups H-2 and C-2 ....................... Body weights and vein weights (mean-+SEM) in groups subjected to the study of Ln vitro femoral vein pressure- -volume relationsh1ps .............. Group mean +SEM of Ln vitro femoral vein pressure- volume relationships and compliance (AV/AP) in control (C- 2, C- 3) and in hypertensive (H- 2, H- 3) dogs, during infusion at constant rate of 3. 88 ml/min ........ Mean-+SEM of venous wall water, Na and K content in early,one- kidney (H- 2), chronic, one- -kidney (H- 3) hypertensive and control (C- 2, C-§)_dogs ........ Group mean-+SEM of Ln vitro superior mesenteric vein pressure- -volume relationships and compliance (AV/AP) in control (C- l) and in hypertensive (H- 1) dogs, during infusion at constant rate of 9.88 ml/min ........ vii Page 102 102 105 FIGURE 1. LIST OF FIGURES The characteristic constricted (sigmoid) and dilated (convex toward the volume axis) venous distensibility pattern ......................... Venous distensibility pattern obtained during rapid continuous infusion (upper dashed line) and during stepwise injections at ten second intervals (lower dashed line). The solid line represents the con- tinuous record of pressure-volume relationships in the vein during stepwise injections ........... Schematic drawing of the innervated, collateral-free, naturally perfused loop of small intestine (ileum) preparation ....................... Schematic drawing of the Y-shaped (4-4-1 l/2 cm) segment of mesenteric vein used to study venous pressure-volume relationships ifl_VlV0 .......... Schematic drawing of apparatus used to study venous pressure-volume relationships in.vitro ......... Blood flow-pressure relationships in the pump-perfused vascular bed of ileum. Mean:tSEM perfusion pressure in control (C-l, C-2) and hypertensive (H-l, H-2) dogs . . . . Blood flow-resistance relationships in the pump- perfused vascular bed of ileum. MeanztSEM resistance in control (C-l, C-2) and hypertensive (H-l, H-2) dogs Mean:tSEM jn_vivo mesenteric vein pressure-volume relationships (injection phase) in sham-operated control (C-l), ____., and early, twgrkidney hyper- tensive (H-l), ---- , dogs ............... In vivo mesenteric vein pressure-volume relationships (Injection phase) in 15 unilaterally nephrectomized, sham-operated control (C-2) and l5 early, ong:kidney hypertensive (H-2) dogs. Each curve represents the average of 2 or 3 separate measurements in one animal ......................... Page l6 17 54 59 64 77 79 84 86 FIGURE 10. 11. 12. 13. 14. 15. 16. 17. Mean tSEM jn_vivo mesenteric vein pressure-volume relationships (1njection phase) in unilaterally nephrectomized, sham-operated control (C-2) and early ggg:kidney hypertensive (H-2) dogs. With- drawa curve,-.—~, for controls and hypertensives is also shown. *P<0.05; #P<0.02 ........... Mean tSEM jn_vitro superior mesenteric vein pressure- volume relationships (injection phase) in sham-operated control (C-1) and early, Ewgfkidney hypertensive (H-l) dogs Mean tSEM jg_vitro superior mesenteric vein pressure- volume relationships (injection phase) in unilaterally- nephrectomized, sham-operated control (C-l) and early, 93g:kidney hypertensive (H-2) dogs. Withdrawal curve for controls and hypertensives is also shown. *P <0.05; #P <0.02 ................... Mean tSEM in vitro superior mesenteric vein pressure- volume relationships during infusion at constant rate of 9.88 ml/min in unilaterally-nephrectomized, sham- operated control (C-2) and early, gng:kidney hypertensive (H-2) dogs. *P‘<0.05 ........... Scatter diagram for the weight (gm) of superior mesenteric vein and the total volume (ml) injected to reach 40 mm Hg intravenous pressure in ll control (C-2) and 10 hypertensive (H-2) dogs .......... Mean:tSEM in vitro femoral vein pressure-volume relationshTEs during infusion at constant rate of 3.88 ml/min in unilaterally nephrectomized, sham- operated control (C-2, C-3) and early (H-2) and chronic (H-3) gggrkidney hypertensive dogs. *P<0.05 ........................ Cross section through portion of a small artery (left) and its accompanying vein from the mesentery. Trichrome stain. x 80 .................. Cross section through portion of a small mesenteric vein. Trichrome stain. x 160 .............. ix Page 88 95 97 99 104 109 113 115 FIGURE Page 18. Analysis of venous pressure-volume data. A. Actual record of superior mesenteric vein pressure-volume relationships, injection and withdrawal phase, in vitro. Stepwise injection and withdrawal of 0.2 ml aliquots of Krebs-Ringer solution, horizontal axis, vs. intravenous pressure (mm Hg). vertical axis. Paper speed at 1 mm/sec. B. Plot of pressure-volume measurements, injection phase. Points were taken at 10 seconds following each injection. The calculation of venous compliance in the pressure ranges of 5-15, 15-25 and 25-35 mm Hg is also shown ........... 136 CHAPTER I SURVEY OF LITERATURE Introduction According to present concepts, the elevated blood pressure in most established forms of arterial hypertension is attributable pri- marily to an increase in total peripheral vascular resistance. It is not known with certainty whether or not the peripheral resistance increases uniformly in all vascular areas or whether the pattern of the early distribution of the abnormal peripheral resistance is the same or similar to that seen in chronic arterial hyptertension. Both of these considerations are important in understanding the pathogenesis of hypertension. In this review, four types of hypertensive disease are consid- ered: human essential hypertension, experimental renovascular hyper- tension, human renovascular hypertension and genetic hypertension in rats. The review of each type is subdivided to consider: cardiac output, blood volume, regional blood flows and resistances (total, parallel- and series-coupled) and autoregulatory phenomena, with special attention to the vascular bed of the mesentery. Human Essential Hypertension Cardiac Output Human essential hypertension is elevated arterial blood pressure occurring in man without an evident cause. Over 90% of cases of human hypertension fall in this category. In its established, uncomplicated form, human essential hypertension is characterized by a normal cardiac output and an elevated total peripheral vascular resistance (Pickering, 1972; Varnauskas, 1955; Frohlich gt_gl,, 1967). Relatively recent findings, however, seem to indicate that cardiac output, accompanied by normal or even decreased total peripheral vascular resistance, may be increased in the early, so-called borderline stage of human essential hypertension. According to Julius and Schork (1971), hemodynamic data on over 400 subjects with borderline hyperten- sion are available. A substantial portion of these subjects has an increased cardiac output, which is a fairly reproducible finding. A group of patients with borderline hypertension and increased cardiac output continued after 50 months to maintain a significant elevation, although there was a tendency toward a decrease of cardiac output and an increase in total peripheral vascular resistance (Eich gt_gl,, 1966). There is some controversy as to whether the output is increased by an elevated heart rate, or a high stroke volume. The mechanism for the elevation of the cardiac output in borderline hypertension is unexplained. In this regard, the relative role of intravascular volume and its distribution and the contribution of venous return should be explored. An increase in intravascular volume, in the absence of heart failure or increased vascular capacitance, leads to an increase in cardiac output. Similarly, decreased venous distensibility (capacitance function), in the absence of decreased blood volume, leads to increased venous return and cardiac output in turn. Blood Volume Absolute blood volume in human essential hypertension seems to depend on the severity of the hypertension. Plasma volume, measured from the volume of distribution of 131I- or 125I-labeled human serum albumin, was found to be normal in essential hypertensive men with supine diastolic blood pressure less than 105 mm Hg. In hypertensive subjects with a diastolic blood pressure greater than 105 mm Hg, plasma volume was found to be reduced and inversely related to the diastolic blood pressure (Bello et_al,, 1965; Tibblin gt_al,, 1966; Dustan gt_al,, 1973). The reasons for the reduction of plasma volume are not apparent. It has been suggested that in human essential hypertension the reduction of plasma volume may reflect increased capillary hydrostatic pressure, because total extracellular fluid volume, estimated with either °28r or 2"Na, was found to be normal, while the ratio of plasma volume to interstitial fluid volume was significantly reduced (Dustan gt_a1,, 1973). Regional Hemodynamics Indirect measurements of regional blood flow in human subjects with established, uncomplicated human essential hypertension suggest that the increased vascular resistance is uniformly distributed in the systemic circuit, except perhaps in skeletal muscle and kidneys. There is evidence that vascular resistance may be decreased and blood flow increased in the skeletal muscle vascular beds. There is also evidence that total blood flow and blood flow per functional tubular tissue usually decrease in the renal vascular bed, the result of disproportionately great increase in renal vascular resistance. Brod gt_al, (1960, 1962) measured vascular resistance simul— taneously in several vascular beds of men with essential hypertension. They estimated cardiac output by the congo-red and rose-bengal dye- dilution method and found it to be similar in hypertensives and in normotensives. They measured renal, splanchnic, skin and muscle blood flow by the p-amino-hippurate clearance, the rose-bengal disappearance, the heated thermocouple and the occlusion plethysmographic methods, respectively. Their results indicate that, compared to normals, essential hypertensives at rest have higher resistances in the renal, splanchnic and cutaneous vascular beds, while their muscle vascular resistance is lower. Thus, in hypertensives, blood flow appears to be shunted from the viscera and skin to muscle. Interestingly, the same distribution of cardiac output occurs in normal individuals during exercise and stress. The authors suggested that a disturbance of the autonomic nervous system may be responsible for the observed hemodynamic changes in human essential hypertension. Other studies of individual regional vascular beds support the findings of Brod et a1. Abramson and Fierst (1942) studied resting blood flow, by the venous occlusion plethysmographic technique, in the forearm (hand excluded) and leg (foot excluded) of hypertensive and normotensive subjects. They found that in hypertensives, the resting blood flow in the forearm and leg was significantly greater than that in normals. In addition, they noted a tendency toward a linear rela- tionship between systolic blood pressure and average blood flow to extremity in hypertension, implying a normal or near normal muscle vascular resistance. They, however, failed to analyze their data in terms of vascular resistance. Their reported data are insufficient to make such calculations in retrospect. Conway (1963), using water plethysmography, measured resting blood flow in the forearm (hand excluded) of essential hypertensive and normotensive subjects. He found resting blood flow to be increased and forearm vascular resistance to be normal in established human essential hypertension. Overbeck gt_gfl, (1969), using an indicator dilution technique, studied blood flow and pressure relationships in the entire forearm, with the hand included, of essential hypertensive and normotensive subjects. The authors found the blood flow to the forearm to be similar in the two groups. Mean forearm vascular resistance, on the other hand, was significantly greater in hypertensives. The discrepancy between the findings of Overbeck gt_al,, on the one hand, and Abramson, Fierst and Conway, on the other, can be explained on the basis of the relative composition of the forearm with the hand excluded and the forearm with the hand included. The forearm with the hand excluded is made up primarily of muscle and bone. The hand is made up principally of skin and bone. The findings of Overbeck 23:21:, therefore, suggest that, while the vascular resistance to muscle is normal or only slightly increased, skin vascular resistance is significantly increased in established hypertension. Amery gt_gl, (1969) studied blood flow to a single muscle, the tibialis anterior, using the 133Xe local clearance method in hyperten- sive and normal male subjects, both at rest and after ischemic exercise. Their findings show that at rest in hypertension, both muscle blood flow and resistance are increased, implying that the muscle vessels share in the increase of total peripheral vascular resistance, but to a lesser extent than the total circulation. Reduction in renal blood flow has been a consistent finding in human essential hypertension, as attested to by both the older clearance studies and the more recent radioactive inert gas washout or desatura- tion methods. What is more, a direct relationship between the severity of the hypertension and the degree of reduction in renal blood flow per unit mass of tissue can be discerned. In view of the arterial hyperten- sion, the reduced renal blood flow is indicative of a considerably elevated renal vascular resistance, which is in excess of that seen in other regional vascular beds (Goldring gt_al., 1941; Ladefoged gt_al,, 1969; Hollenberg gt_al,, 1969). In contrast to the elevated muscle blood flow and the reduced renal blood flow in human essential hypertension, investigators have found normal blood flow and increased vascular resistance in the brain (Kety gt_gl,, 1948; Mallett g£_al., 1963), the splanchnic bed (Culbertson g§_al., 1951; Wilkins gt_al,, 1952; Cohn gt_al,, 1962), the coronary bed (Rowe gt_al,, 1961), the hand (Stead gt_al,, 1940), the foot (Stead gt_gl,, 1940), the digit (Caliva gt_al,, 1963) and the skin (Brod gt_al, 1962). In comparison to the other regional vascular beds, few data are available concerning the splanchnic circulation in human essential hypertension. The splanchnic circulation is defined here as the vascular system that is perfused by the celiac, the superior mesenteric, and the inferior mesenteric arteries, and is drained by the hepatic veins. The reason for the scarcity of data concerning the splanchnic circulation is the relative inaccessibility of these vascular beds. Wilkins §t_al, (1951, 1952) measured hepatic blood flow in hypertensive patients and in normal subjects, utilizing the BSP extrac- tion method, first developed in their laboratory. The method is based on the ability of the liver to take up the dye rapidly. Although it was once considered to be a simple index of liver blood flow, the clearance of BSP from the blood is also influenced by the rate of transmembrane transport of the dye, the hepatic storage capacity and the blood-to-bile transfer maximum. Furthermore, the method requires catheterization of a branch of the hepatic vein, a process which in- troduces a venous sampling error, in that blood withdrawn from the catheter may not truly represent the concentration of the dye in the hepatic venous effluent. The average estimated hepatic blood flow, as measured by Wilkins et al., was 1303::50 ml/min in a group of 41 hypertensives and 1381:178 ml/min in a group of 21 normotensives. The difference in blood flow was not significantly different in the two groups. Unfortunately, only a small fraction of the two groups of subjects is characterized as to age, body weight, severity of hypertension and hepatic function. It is, therefore, difficult to determine how well the two groups were matched. Cohn §t_al, (1962) measured hepatic blood flow in hypertensives, without comparison to normotensives, using both the BSP clearance and the radioactive colloidal chromic phosphate uptake method. In a group of 8 hypertensives, the estimated hepatic blood flow averaged 1576 ml/min by the BSP and 1825 ml/min by the radioactive colloidal chromic phosphate method. The latter method is, however, based on the ability of the reticuloendothelial system of not only the liver, but also the spleen and the mesenteric lymph nodes to take up the colloidal material. The results obtained by the two methods are, therefore, not comparable, but they emphasize the difficulty that is inherent in the interpretation of this type of data. As far as the discrepancy between the mean esti- mated hepatic blood flow of 1303 ml/min in hypertensive patients, obtained by Wilkins, and that of 1576 ml/min, obtained by Cohn, is concerned, there is no apparent explanation. The calculated hepato- portal resistance, on the other hand, is invariably increased in human essential hypertension, but the extent of this increase in proportion to the other regional vascular resistances is uncertain. The estimation of hepatic blood flow is a crude measure of splanchnic blood flow for it lumps together blood flow through two regional vascular beds: the hepatic and the mesenteric-portal. The arteries supplying the two vascular beds are in parallel with respect to each other, but they share the same venous outflow (hepatic veins). Furthermore, the mesenteric-portal bed itself is composed of two vascular beds in series. To date, no method applicable to the study of human subjects has been devised that would allow us to study these three vascular beds separately from each other. Their relative par- ticipation in the increase of the total peripheral vascular resistance in human essential hypertension remains to be defined. Finally, blood flow to the spleen, although of minor importance in man, should be considered, too. Parallel-coupled resistances.--Few studies have been designed to explore the status of parallel-coupled resistances within an organ or tissue in established human essential hypertension, probably because of the technical difficulties involved. However, an alteration in parallel resistances in organs concerned with blood pressure regulation, such as the brain and the kidney, might well induce or modify the hyper- tension. To date, indicator-dilution techniques have been used to study the distribution of blood flow in the brain and in the kidney of hyper- tensive subjects. The distribution of blood flow in the brain appears to be normal to the white and gray matter of the cerebrum (Mallett g3_ 31,, 1963). Dickinson 95:21: (1960), on the other hand, raised the possibility that blood flow through the vertebral artery system to the 10 brain stem may be impaired early in the course of the hypertension, leading to an increased basal vasomotor tone and the perpetuation of the hypertension in turn. In post-mortem studies, they were able to demonstrate reduced flow and increased resistance of the cerebral arteries of hypertensive patients, with preferential and early in- volvement of the vertebral arteries. Unfortunately, there is no method at the present time that would allow one to measure blood flow to the brain stem selectively jg_yjyg_in human subjects, a necessary require- ment for testing Dickinson's hypothesis. Recent investigators (Logan gt_al., 1973) applied the inert gas-washout and the constant infusion, indicator-dilution techniques to the study of renal hemodynamics in hypertensive patients. The results indicate that the reduction of renal blood flow in human essential hypertension is due to a selective fall in renal cortical blood flow. In view of the arterial hypertension, the reduced cortical blood flow is indicative of a considerably elevated renal cortical vascular resistance. Medullary blood flow is preserved and medullary vascular resistance rises to a lesser degree than cortical vascular resistance. An increased pressure in the vasa recta because of transmission of elevated arterial pressure through a relatively nonvasoconstricted medullary vascular bed could inhibit sodium transport in the loop and account for the concentrating defect and exaggerated natriuresis observed in hypertensive subjects (Baldwin et al., 1965). 11 With regard to the splanchnic circulation, there is no report in the literature concerning compartmental blood flow and parallel- coupled resistances of the vascular bed of the gastrointestinal wall in human essential hypertension. The gastrointestinal wall consists of four layers: the mucosa, the submucosa, the muscularis and the serosa. The blood flow through these layers is not uniform. Certain stimuli, such as hypertension, may influence blood flow through one of these layers without altering blood flow through the others. Other stimuli, including hypertension, may increase mucosal flow and decrease muscle flow, the total flow being unchanged. An alteration of mucosal blood flow may bring about important changes in the absorptive and secretory function of the small intestine, possibly affecting the overall body water and electrolyte metabolism. Series-coupled resistances.--Series-coupled resistances have been more extensively studied in hypertension. The consideration of series-coupled resistances within any given organ is just as important from the pathogenetic point of view of hypertension as that of parallel- coupled resistances, especially if series-coupled resistances are altered early in the course of the hypertension. Normal blood flow and pressure relationships on the venous or low pressure side of the circulation, throughout the course of the hypertension, would suggest stimuli which act solely on the arterial side of the circulation. An early and selective alteration of the flow and pressure relationships of the venous side of the circulation would suggest pathogenetic stimuli which act primarily on the veins. 12 Nailfold and skin capillary pressure have been studied indirectly, by the color blanching (Ellis et_al., 1929), and directly, by the microinjection method (Eichna gt_al,, 1942), in hypertension and have been reported as normal. The actual mean values in the two studies, however, differ greatly, not to mention the wide range of readings in both normals and hypertensives. In view of the normal blood flow to these areas, the similarity of the cutaneous capillary blood pressure of normals and hypertensives suggests that the increased vascular resistance of hypertensive subjects is precapillary and presumably arteriolar. A possible exception to the finding of normal capillary pres- sure in hypertension is the renal vascular bed where an elevated filtration fraction, as measured by the clearance ratios of inulin and p-aminohippurate, is an early finding in chronic diastolic hyper- tension. Elevated filtration fraction may possibly reflect glomerular capillary hypertension, with relatively greater constriction of the efferent arteriole than of the afferent arteriole (Goldring gt_al., 1941). However, this conclusion about the distribution of resistance between the efferent and afferent arterioles has been disputed (Gomez g;t__a_l_. , 1951). What is the contribution of the larger arteries to the increased precapillary resistance in human essential hypertension? Oppenheimer and Prinzmetal (1937) attempted to answer this question by measuring the pressure gradient from the brachial to the digital artery of hyperten- sive subjects, assuming that the total forearm and hand blood flow is 13 normal. The authors reasoned that if the pressure gradient from large to small artery should become steeper with the rise of blood pressure in hypertension, it would point to an increased resistance in the larger arteries. They measured the brachial pressure by sphingomanometry and the digital artery pressure by Gartner capsule. The authors found that there is no increase in the arterial pressure gradient. They concluded that the arteries do not play an important role in the production of high peripheral resistance in human essential hypertension. Kettel, Overbeck gt_al, (1969) measured intravascular pressures directly, through the insertion of catheters, in the brachial artery (large artery), the radial artery (small artery), a vein on the dorsum of the hand (small vein), and an antecubital vein (large vein) in hyper- tensive and normotensive subjects. The authors found the pressure gradient from small artery to small vein to be increased in hyperten- sives. The pressure gradient from large artery to small artery and from small vein to large vein was similar in the two groups. Assuming that blood flow to the whole arm is also similar in the two groups, their findings suggest that not only the large artery resistance but also the venous resistance in the arm of hypertensives is normal. Therefore, the elevated upper extremity vascular resistance in hyper- tension must result from constriction of the smaller vessels. However, Caliva gt_al, (1963), using a venous occlusion plethysmographic tech- nique to measure indirectly venous pressure in the digits of patients with essential hypertension, found an elevated digitial venous pressure with normal digital blood flow and concluded that venous resistance is 14 elevated in hypertension, although to a lesser extent than small vessel resistance. As a result of the relative inaccessibility of the splanch- nic circulation in human subjects, there is a complete lack of data concerning series-coupled resistances in this vascular bed. Specif- ically, little or nothing is known about the contribution of the portal and hepatic pre- and postsinusoidal and of the mesenteric pre- and post- capillary resistances to the overall increase of splanchnic vascular resistance. It should be mentioned at this point that all the studies in this section of this review have dealt with the various regional hemo- dynamic features of established, uncomplicated human essential hyper- tension, characterized by normal cardiac output and increased total peripheral vascular resistance. The relatively recent finding of an increased cardiac output in borderline hypertension indicates the need to evaluate regional hemodynamics at the onset of hypertension, so that primary, pathogenetically important patterns may be distinguished from secondary, complicating features. To date, the distribution of the increased peripheral blood flow in subjects with borderline hypertension is essentially unexplored. Bello §t_al. (1967) indicate that renal blood flow may be elevated in borderline hypertension. Vascular Compliance General considerations (Spencer et al., 1963; Alexander et al., l963).--Compliance is a pr0perty of the vascular wall, arising from its distensibility. Each component of the vessel wall, smooth muscle. 15 elastin, collagen and ground substance contribute to compliance. Compliance (C) is expressed in terms of blood volume (V) in the segment and the attending pressure difference across the vascular wall (Pc)’ V C:— Pc In vitro studies of excised segments of arteries and veins revealed two types of tension-length or pressure-volume (compliance) curves (Figure 1). One was described as curvilinear with considerable convexity toward the length or volume axis, the other sigmoid in char- acter. At first, the latter was ascribed to an experimental artifact, until MacWilliam took the precaution of collecting fresh tissues and observed their behavior during the immediate post-mortem period. Shortly after the vessel was excised, it developed marked spasm. Tension-length or pressure-volume measurement at this time would reveal the characteristic sigmoid curve. When the state of contraction was eliminated by warming, the more conventional stretch curve was obtained. MacWilliam interpreted the sigmoid curve as a manifestation of the resistance to stretch of the smooth muscle. The curvilinear curve with convexity toward the length or volume axis, on the other hand, represents primarily the resistance to stretch of the connective tissue elements of the vessel wall, the smooth muscle elements being in a state of relaxation. Later, Alexander (1948, 1953, 1954, 1955), in his jn_ yiyg_study of the femoral and mesenteric veins, demonstrated the same distensibility patterns. A controversy exists as to whether the sigmoid distensibility pattern of constricted veins is or is not associated with an overall reduction in total distensibility. 16 40 , 3O .. 2" fig 20 —- g2 10 - O Volume (ml) Figure l. The characteristic constricted (sigmoid) and dilated (convex toward the volume axis) venous distensibility pattern. Another feature of vascular distensibility or compliance is the marked time dependency in elastic behavior, sometimes referred to as "elastic hysteresis," "delayed compliance" or "stress relaxation," the latter implying that pressure dissipates following sudden distention to a constant volume (Figure 2). As a result, the pressure-volume (com- pliance) curve observed on injection and the pressure-volume curve found on withdrawal form a loop. Alexander pointed out that the initial state of venoconstriction or venodilation has appreciably greater influence on the pressure-volume curve observed during injection than during with- drawal. The latter curve is more reproducible than the former, being subject to fewer stimuli. 40 - / / 30 ‘ / / / 2;; / 5: 20 r / wE / 32" // / 10 ’ / / / / / 0 / Volume (m1) + off Figure 2. Venous distensibility pattern obtained during rapid continuous infusion (upper dashed line) and during stepwise injections at ten second intervals (lower dashed line). The solid line represents the continuous record of pressure- volume relationships in the vein during stepwise injections. From the experimental point of view, especially, when one is trying to compare experimental data from one animal with those from another, these characteristics of vascular distensibility have important practical implications. First, the timing of the administration of anesthetics and of surgery should be standardized, so that the state of the vascular smooth muscle would be comparable in different animals. In the case of an i vitro study, the time elapsed between the excision of the tissue and the first jfl_vitro determination of vascular distensi- bility should be similar from one experiment to another. Second, since 18 pressure-volume data obtained from arteries and veins are strongly influenced by the exact conditions of vascular distension, identical rates and magnitudes of distension should be used. Third, the time interval between subsequent pressure-volume determination should be set. Stress relaxation implies that a second stretch curve differs greatly from an initial stretch unless sufficient time is allowed between measurements for a complete restoration of the resting dis- tensibility state. 'Studies have shown that this critical time period is between 20-30 minutes. Fourth, pressure-volume data have no absolute meaning unless related to a critical volume. Therefore, an attempt should be made to standardize the initial length and/or volume of the tissue. In the case of pressure-volume studies, the initial volume should be 0. There are obvious practical difficulties which interfere with accomplishing this objective, especially in the jg_yjyg_study of veins, where valve action may prevent complete emptying. Inter-groUp comparison of vascular pressure-volume (compliance) relationships is valid only if proper attention has been paid to these considerations. Vascular compliance in human essential hypertension.--Greene g; 31. (1966) studied pressure-volume (compliance) relationships directly in the isolated, temporarily occluded segment of the brachial artery of normotensive and hypertensive subjects. Following the isolation and occlusion of a 3 cm segment of the brachial artery, they emptied the segment of its blood content; 0.05 ml aliquots of normal saline at body 19 temperature were injected into the segment through an indwelling Cournand needle. An equilibration period of approximately ten seconds was allowed following each injection. Stepwise increments of fluid were added until intrasegmental pressure reached 150-200 mm Hg pressure. The authors found that the arterial pressure-volume curve of hyperten- sives is shifted in the direction of the pressure axis, implying decreased distensibility of the brachial artery in chronic human essential hypertension. This finding can be explained on the basis of structural changes, such as medial hypertrophy, elastic intimal thickening and increased water and electrolyte content, described by various investigators in both large and small arteries from hyperten- sive patients (Pickering, 1972). The question, then, arises as to whether these changes are primary or secondary to the increased intra- luminal pressure in hypertension. The consensus of opinion at the present time appears to be that these are secondary changes. There are no reports of direct measurements of venous compli- ance in essential hypertensive subjects. A number of indirect measure- ments are, however, available. Caliva g§_al, (1963) examined digital vascular compliance by rheoplethysmographic techniques (Burch, 1956), in hypertensive and normotensive subjects. Following occlusion of the venous outflow from the digit, the time course of the volume increase in the digit was measured until outflow pressure exceeded the occlusion pressure. The hypertensive pressure-volume curves showed small changes in volume associated with large pressure increases, that is, a steep pressure-volume curve. Since approximately two-thirds of systemic blood 20 is normally contained in veins, the authors concluded that, in addition to arterial compliance, digital venous compliance is probably also decreased in hypertensives. More recently, Walsh g§_al, (1969) studied forearm venous distensibility in human essential hypertension, using the electro- capacitance plethysmographic method, and came to a conclusion similar to that of Caliva gt_gl, The authors first determined the minimal occluding_pressure that resulted in a measurable increase in forearm volume, usually in the range of 3-10 mm Hg. Then they measured the distensibility curve during emptying ("elastic hysteresis" or "delayed compliance") rather than during filling. Vascular distensibility, defined as the percent increase in forearm volume at cuff pressure of 30 mm Hg above the minimal occluding cuff pressure, was significantly less in their hypertensive than in their normotensive subjects. Whether the decreased vascular distensibility is due to decreased arterial and/or decreased venous compliance cannot be determined by this method. Anderson (1954) studied total forearm vascular distensibility by a slightly different technique. He suddenly occluded the total circulation to the forearm, while he monitored venous pressure through a catheter inserted into one of the major superficial veins of the fore- arm. A period of 15-45 seconds was allowed for the intravascular pres- sure to come into equilibrium between the arterial and venous sides of the circulation. The highest reading in the vein, following the occlu- sion of the total circulation to the forearm, was defined as the intrinsic blood pressure. Intrinsic blood pressure in normals ranged 21 between 12-19 cm H20. Patients with chronic diastolic blood pressure showed a significantly elevated intrinsic blood pressure. Blood volume was not measured. Ulrych g§_al, (1964) studied the cardiac and renal responsive- ness of hypertensives and normotensives to acute volume expansion with 1200 ml/7O kg of iso-oncotic dextran in isosmotic normal saline. Control central venous pressure, cardiac output, glomerular filtration rate and urinary volume were similar in the two groups. Control total peripheral vascular resistance, on the other hand, was significantly higher and urinary Na excretion significantly lower in the hypertensive group. Fifty to seventy minutes after the infusion, the absolute in- crease in cardiac output, urinary sodium excretion and urinary flow was significantly greater in hypertensives than in normotensives, even though the absolute increase in central venous pressure was not dif- ferent. Blood pressure changes, if any, were not reported. The authors suggest that the excessive response of cardiac output in hypertension may possibly reflect a deficient ability of the capacity vessels to dilate in response to the sudden expansion of intravascular volume, which leads to increased venous return. If this is true, one would have expected to see a significantly greater absolute increase in cen- tral venous pressure as well. In addition, the hypertensive subjects demonstrated a significantly greater fall in total peripheral vascular resistance than normotensives. The possibility that the greater in- crease in cardiac output in hypertension is due to the more pronounced vasodilatory response of the resistance vessels of hypertensives to 22 infusion of dextran cannot be ruled out. Although blood volume was not measured, an initially increased blood volume in hypertensives is an unlikely explanation for the reported findings in light of other studies indicating normal or even decreased volume in essential hyper- tensive patients. These experiments of Ulrych g§_al, confirm the findings of earlier studies showing an exaggerated natriuresis in human essential hypertension, following saline load (Baldwin 23431., 1965). In contrast to the above studies, there is other evidence which seems to indicate that venous compliance is not elevated in hypertension. Two groups of investigators measured vascular distensibility, using water-plethysmographic techniques, of the forearm and the calf of hypertensives subjects. Abramson and Fierst (1942) studied the vascular distensibility of both the forearm and the calf during filling. The initial 20 mm Hg occluding pressure resulted in a volume increment of only 0.16 ml/lOO m1 of forearm volume. The maximum volume increment at 70 mm Hg occluding pressure amounted to 0.52 ml/lOO ml. When the results are plotted on a pressure-volume curve, the rapidly rising portion of a characteristic distensibility curve is never obtained. Thus, a difference in distensibility in the higher pressure ranges between hypertensives and normotensives could have been missed by the authors. The same observations applies to the experiments of Wood (1961), who studied forearm volume increments only up to 30 mm Hg occluding pressure. (An occluding pressure of 30 mm Hg applied to the surface of the arm may not represent an equivalent pressure in the deep veins of the arm). There was no demonstrable difference in the vascular 23 distensibility in hypertensives and in normotensives in this range of venous (occluding) pressure. There are no reports in the literature concerning the venous distensibility of other regional vascular beds in human essential hypertension, including the liver and the intestines. This is un- fortunate in view of the important blood storage function of the splanchnic circulation. Under normal physiologic conditions 14% of circulating blood volume is contained within the mesenteric veins (Texter g§_al,, 1968). If there is a disturbance of venous capac- itance function in human essential hypertension, the logical place to look for it is in the splanchnic vascular bed. As to the mechanism underlying the suggested reduction in vascular distensibility in established, uncomplicated human essential hypertension, the consensus of opinion appears to be that it is due to increased smooth muscle tone, that is to say, to arterial and/or venous constriction (Caliva gt_al,, 1963; Walsh §t_gl,, 1969). However, data that would bear out this suggestion are lacking. Walsh g§_al. (1969) repeated distensibility measurements in their hypertensive subjects after several weeks of antihypertensive therapy, but were unable to show a statistically significant increase in vascular distensibility. Finally, while decreased arterial compliance in chronic human essential hypertension seems to be established, the evidence for decreased venous compliance is indirect and incomplete. 24 Venous wall anatomy and function (Alexander, 1963; Bader, .1963).--In order to understand the mechanism that may underlie the suggested reduction in venous distensibility in human essential hypertension, one has to consider the normal anatomy and function of the venous circulation. The venous bed includes veins and venules, i.e., vessels which cannot constrict against an internal pressure greater than 70-80 mm Hg (Abramson g3_al., 1942). The capacitance function of the venous bed is about equally shared by the venules and the larger veins. In contrast to the arteries, the veins are very variable in their wall structures. Usually they have a larger percentage of collagen fibers, especially the large veins, but there are veins in which the muscular mass exceeds that of the collagen fibers, especially in the smaller veins and the veins of the lower extremities, which are subject to a large hydrostatic pressure. The inferior mesenteric vein of humans, for instance, is made up primarily of collagen and elastin 0>90%), while the skin vein of the foot is 50-60% smooth muscle. Another important difference between arteries and veins is that the veins are securely embedded in the surrounding tissues. Decreased venous distensibility, therefore, could be partially due to decreased interstitial tissue compliance, as had been suggested by Lucas g3_al. (1973). The elastic modulus defined as AP/AV/V, (where AP is pressure in dynes/cm2 and V is volume in ml), of the various vascular wall elements are as follows: elastic mOd”]us(collagen) = 300 x 106 dynes/ 2. o = 6. cm , elast1c mOd"l"S(elastin) 6 x 10 , elastic m°d"1”5(smooth muscle) 25 and elastic m°d"]”5(endothelium) = 0.1 x 106 (Wiederhielm, 1965). On this basis, collagen fibers are considered some 3000 times stiffer (or less distensible) than smooth muscle fibers. Any consideration of vascular distensibility, therefore, has to take into account the contribution of collagen fibers, especially, in the higer pressure ranges, along the rapidly rising portion of the pressure-volume curve, where the connective tissue elements of the vessel wall begin to be stretched. With regard to the proposed reduction in venous distensibility in human essential hypertension, pathologic data are completely lacking. In particular, data concerning the water and electrolyte and collagen content of the venous wall are needed. The composition of the inter- stitial tissue surrounding the veins should also be examined. Summary Chronic human essential hypertension is characterized by a normal cardiac output and an elevated total peripheral vascular resistance. Cardiac output, accompanied by a normal or even decreased total peripheral vascular resistance, may be increased in a large pro- portion of early human essential hypertension. Blood volume is either normal or decreased. The study of the regional hemodynamics of human essential hypertension is extensive, but confined to a great extent to the chronic stage of the hypertension, where secondary, complicating changes may obscure primary, pathogenetically important changes. Also, by necessity the experimental techniques that have been utilized are indirect and, 26 therefore, the information concerning the role of parallel- and series- coupled resistances is incomplete or completely lacking, as in the case of the splanchnic circulation. Although vascular distensibility seems to be decreased, the relative contribution of arterial vs. venous compliance has not been established. Experimental Renovascular Hypertension In 1934, Goldblatt g§_al. (1934) demonstrated that constriction of one or both renal arteries can lead to hypertension. In the dog, constriction of one renal artery, the other kidney being intact, pro- duces either no hypertension or only a mild, transient elevation of arterial blood pressure. Constriction of both renal arteries or con- striction of one renal artery plus removal of the opposite kidney, on the other hand, leads to chronic hypertension. There is, therefore, a definite difference in the severity of the hypertension produced by two-kidney and one-kidney Goldblatt hypertension. In 1939, Page (1939) introduced an alternative method of pro- ducing renal hypertension in experimental animals. He showed that wrapping one kidney in silk or cellophane, with or without contralateral nephrectomy, induces an inflammatory reaction around the kidney which, in turn, leads to hypertension. This type of experimental hypertension is referred to as perinephritic hypertension, and it is one of the most effective ways of producing hypertension in experimental animals. Whether the hypertension is due to compression of the kidney itself 27 or to the compression of the renal pedicle is still uncertain. So far as is known, the hypertension produced by perinephritis behaves in every respect like that due to renal artery constriction (Pickering, 1972).: The pathogenesis of both the Goldblatt and the perinephritic type of experimental renal hypertension remains uncertain despite the large body of research dedicated to this question since the introduction of these experimental models some 40 years ago. Cardiac Output In the animal models of experimental renovascular hypertension so far studied, cardiac output is increased in the early (less than four weeks) stage of the hypertension, followed by a gradual return of cardiac output to normal. Ledingham gt_al, (1964, 1967) monitored the cardiac output of unanesthetized rats with electromagnetic flowmeters in place around the arch of the aorta, both before and after the clipping of one renal artery and the removal of the untouched kidney. Following clipping, total peripheral vascular resistance and blood pressure rose within two hours and remained elevated throughout the rest of the experiment. Cardiac output after having fallen below control values during the first five days postoperatively, began to rise on the fifth day and remained elevated for as long as 35 days, when the experiment was terminated. Ferrario engl, (1970) monitored cardiac output with an electro- magnetic flowmeter implanted around the arch of the aorta in unanesthe- tized dogs before and after the induction of perinephritic hypertension. 28 Eight to 15 days following wrapping of one kidney in cellophane, with the other kidney untouched, they found the cardiac output to be signif- icantly elevated with slightly decreased total peripheral vascular resistance. Following removal of the untouched kidney, at two weeks, arterial blood pressure and, for the first time, total peripheral vascular resistance began to rise, while cardiac output rose further, reaching a maximum, amounting to 18% of control values, approximately two weeks postnephrectomy. By the fourth to sixth week postnephrectomy, cardiac output returned to normal, and an elevated total peripheral vascular resistance had become the predominant cause of the hypertension. Very recently, Ferrario (l974) performed similar studies in unilaterally nephrectomized dogs with Goldblatt hypertension. Following constriction of the renal artery of the sole remaining kidney in unanes- thetized dogs by an externally adjustable, chronically implanted clamp, cardiac output rose during the first week and remained elevated for a period of four weeks. During the first two weeks, the elevated cardiac output was solely responsible for the rise in mean arterial blood pressure, total peripheral vascular resistance being normal. An in- crease in heart rate and to a lesser extent an increase in stroke volume accounted for the increase in cardiac output. At three weeks after renal artery constriction, total peripheral vascular resistance began to rise, and by the fifth week an elevated total peripheral vascular resistance was the only cause of the hypertension. These two studies of Ferrario confirm the impression of previous investigators regarding the similarity between the hemodynamics of perinephritic and Goldblatt experimental hypertension. 29 Bianchi gt_al, (1970, 1972) noted a more transient rise in cardiac output and an earlier rise in total peripheral vascular resistance in conscious dogs with one-kidney (contralateral nephrectomy) and two-kidney (contralateral kidney intact), Goldblatt hypertension. In one-kidney hypertension, cardiac output, measured by dye dilution technique, was significantly elevated on days 4 and 7, following renal artery constriction. Total peripheral vascular resistance rose imme- diately and remained elevated for a period of 24 hours, thereafter, returning toward normal. In two-kidney hypertension, the rise in cardiac output was significant only on day 1 after renal artery con- striction. By 6-7 days, cardiac output was normal. Total peripheral vascular resistance rose immediately and remained elevated throughout the duration of the study (7 days). Coleman and Guyton (1969) studied hypertension caused by sub- total nephrectomy and salt loading in dogs and reported findings similar to those of Bianchi gt_al,, namely, an early transient rise in cardiac output, with decreased total peripheral vascular resistance, followed by a return to normal of cardiac output and a rise of total peripheral vascular resistance. With regard to an initial rise in cardiac output, followed secondarily by a rise in total peripheral vascular resistance, the "autoregulation theory" of the development of hypertension was proposed (Coleman and Guyton, 1969; Coleman gt_a1,, 1971). According to this theory, the increase in blood flow through all or some of the regional vascular beds gradually leads to an increase in total peripheral vascular resistance. 30 The autoregulatory function of the renal vascular bed is well known, renal blood flow being maintained at a relatively constant level in the arterial blood pressure range of 80-200 mm Hg. In the splanchnic circulation, autoregulation has been described in the liver (hepatic artery system), spleen and intestine. In the stomach, there is no autoregulation (Texter et_al,, 1968). The exact mechanism underlying autoregulation is not clear. Furthermore, our knowledge of autoregula- tion is confined to information derived from short-term animal experi- ments or clinical observations. Little is known about the time course and duration of autoregulatory response over the extended period of weeks or months. Therefore, it is uncertain how short-term regional autoregulation relates to the long-term total body autoregulation that has been suggested to occur in the course of the development of hypertension. Blood Volume Ledingham and Cohen (1964) measured extracellular fluid volume and plasma volume by the thiocyanate and Evans blue distribution method, respectively, in unilaterally nephrectomized rats following renal artery clipping. Although extracellular fluid volume and plasma volume tended to increase in both the clipped and the sham-operated control rats, the rise of extracellular fluid volume and plasma volume was significantly greater at three and seven days post-clipping of the renal artery. Extracellular fluid volume and plasma volume returned to normal by the fifteenth day following clipping. 31 Bianchi g£_al. (1970), using similar techniques, measured extracellular fluid volume and plasma volume in conscious, unilaterally nephrectomized dogs following renal artery constriction. The authors found a significant increase in extracellular fluid volume and plasma volume on days 1, 3-4 and 6-7 following renal artery constriction. By the twelfth to fourteenth day, extracellular fluid volume and plasma volume returned to normal. Therefore, a transient expansion of extra- cellular fluid volume and plasma volume appears to be a consistent accompaniment of the evolution of experimental renovascular hypertension in unilaterally nephrectomized animals following renal artery constriction. Bianchi g3_a1, (1972) repeated these same studies in conscious dogs following renal artery constriction, the contralateral kidney being left untouched. In these dogs, there was no significant change in extra- cellular fluid volume and plasma volume either during the first or the second week following renal artery constriction. Ferrario gt_gl, (1970) studied plasma and total blood volume, using the radioiodinated (1251) serum albumin method, in dogs, 15 days after wrapping one kidney in cellophane, 15 days after contralateral nephrectomy (30 days post-wrapping), and after 10-17 months of chronic, one-kidney, perinephritic hypertension. There was no significant change in plasma and total blood volume at any one time investigated. Extra- cellular fluid volume was not measured by these authors. In contrast, very recently, Ferrario (1974) found, by the lO-minute Evans blue dye- dilution technique, a small rise in plasma and total blood volume, 32 accompanied by a fall in hematocrit, during the first two weeks after renal artery constriction in unanesthetized, unilaterally nephrectomized dogs. Extracellular fluid volume was measured, using the constant infusion of mannitol method, by Grollman g§_al, (1953) in dogs with chronic, one-kidney, Goldblatt hypertension. Extracellular fluid volume was found to be significantly increased in comparison to that of control dogs. However, there is no information available concerning the renal and cardiac status of these animals. Uremia and/or congestive heart failure could possibly account for the extracellular volume expansion of hypertensive animals. Regional Hemodynamics In chronic hypertensive dogs with unilateral renal artery- stenosis and intact, contralateral kidney, blood flow per unit renal weight, measured directly, is reported normal in the non-stenotic kidney, and normal or decreased in the stenotic kidney, depending on the post-stenotic blood pressure. Renal vascular resistance, excluding the resistance offered by the stenosed artery, increases in both the stenotic and the non-stenotic kidneys, but the rise is greater in the stenotic kidney (Bounous gt_al,, 1962). In unilaterally nephrectomized, Goldblatt and perinephritic hypertensive dogs, in the early and chronic stages of hypertension, absolute renal blood flow, calculated from the phenol clearance and extraction percentage, and effective renal plasma flow, calculated 33 from the plasma diodrast clearance, are maintained at a normal level (Corcoran and Page, 1974). More recently, Ferrario and McCubbin (1973) studied renal blood flow with the aid of chronically implanted flowmeters in unanesthetized and unilaterally nephrectomized dogs, before and after renal artery constriction. After mild stenosis (c. 20%) of the renal artery, mean renal blood flow decreased during the first 1-8 days; thereafter, mean renal blood flow returned to, or even exceeded, control values. Pres- sure distal to the stenosis was low compared to the systemic pressure, but in view of the developing hypertension, it was above control values. Therefore, the renal vascular resistance, excluding the resistance offered by the stenosed artery, was initially elevated, but tended to return toward normal after the first week of renal artery stenosis. More severe renal artery stenosis (c. 45%) led to sustained reduction in mean renal blood flow and sustained rise in renal vascular resistance distal to the stenosis. Total coronary vascular resistance and coronary vascular resistance calculated on a unit myocardial weight basis have been reported as elevated in dogs with perinephritic experimental renovas- cular hypertension. These dogs had normal or slightly increased coronary blood flow per unit weight of myocardium (West g§_al., 1959). Brain blood flow, determined by the particle distribution method, has also been reported as normal in rats with experimental renovascular hypertension (Flohr et al., 1971). 34 More recently, hemodynamics in the vascular bed of the limb in dogs have been intensively investigated in the early (less than 4 weeks) and the chronic (more than 4 weeks) stages of perinephritic hypertension. Limb hemodynamics in the early stage of hypertension are apparently similar to those in the chronic stage, including normal blood flow and increased vascular resistance. The parallel resistances of limb skin and skeletal muscle vascular beds share equally in the general increase in limb vascular resistance. The elevated limb resistance in the early stage, as well as in the chronic stage, is primarily confined to the small vessel segment of the limb vascular bed, consisting of small arteries, arterioles, capillaries and venules, although large artery resistance to skin may also be elevated. Skin and muscle venous resistances were normal in both the early and chronic stage of peri- nephritic hypertension (Overbeck gt_al,, 1971; Overbeck, 1972). There is no report in the literature to date on the status of the splanchnic circulation in either the early or the chronic stages of experimental renovascular hypertension. Vascular Compliance Feigl gt_al, (1963) studied the mechanical properties of the femoral arteries in a group of dogs before and after the induction of chronic, bilateral, perinephritic renal hypertension. The authors monitored intrafemoral arterial pressure and vessel diameter simul- taneously with the help of transducers attached to the vessel wall in unanesthetized dogs. At the conclusion of the jg_vivo measurements, 35 a segment of the femoral artery was excised, and its wall thickness was determined. The authors expressed their findings in terms of the elastic modulus (Ep), defined as follows: Ep 2:: AP representing pulse pressure, Ae representing change in strain. a was, in turn, defined as: e = (r-rO/ro, r and r0 representing radius under stress and radius at zero stress, respectively). The elastic modulus represents the distensibility of the arterial segment. The elastic modulus of the femoral artery was significantly increased following the induction of chronic perinephritic hypertension. The authors suggest that the increased elastic modulus of the femoral artery probably rep- resents changes in the arterial wall material during hypertension. In this regard, the authors demonstrated that the water content of the hypertensive femoral artery was significantly increased over control values. A number of indirect studies are available to suggest that venous distensibility or compliance may also be reduced in experimental renovascular hypertension. Richardson gt_al, (1964) measured mgan_gjr- culatory pressure in ten unilaterally nephrectomized, Goldblatt hyper- tensive and in ten control dogs. Mean circulatory pressure was obtained by stopping the heart either by fibrillating it or by injecting acetyl- choline, and quickly pumping blood from the arterial to the venous side of the circulation, until an equilibrium of pressures was effected. The equilibrium pressure is termed mean circulatory pressure. Under 36 conditions of unaltered blood and interstitial tissue volume, mean circulatory pressure reflects the distensibility of the entire vascular bed. The authors found the mean circulatory pressure to be increased in hypertension, but they did not measure blood or extracellular fluid volume. Ferrario gt_al, (1970) confirmed the findings of Richardson g; 91, in dogs with early one-kidney perinephritic hypertension. In addi- tion, they measured the blood volume of these same animals and found it to be normal. Extracellular fluid volume was not estimated. Floyer g;_al. (1961) studied parabiotic rats. These are rats in pairs that are surgically united at a young age by grafting their skin together so that they have a capillary bed in common. In normo- tensive, parabiotic rats, the authors noted that the rat with the spontaneously higher blood pressure would show a higher plasma volume. The situation reversed itself after the induction of hypertension in one member of the pair by clipping one of its renal arteries and leaving the other kidney intact. Under these circumstances, although the plasma volume rose in both animals, the rise of plasma volume was less in the clipped rat. The authors inferred that the capillary hydrostatic pres- sure must have risen in the clipped or hypertensive member of the pair, and this resulted in a fluid shift into the other rat. To account for the rise in capillary hydrostatic pressure, the authors further suggest that, in view of the increased precapillary resistance in hypertension, the postcapillary resistance must have risen disproportionately. The increased postcapillary resistance may possible reflect decreased distensibility of the capacitance vessels. 37 The only direct evidence bearing on the subject of decreased venous distensibility in experimental renovascular hypertension comes from the work of Overbeck (1972). The author studied pressure-volume (compliance) relationships in the isolated, temporarily occluded segment of the femoral and jugular vein in the early (less than 4 weeks) stage of one- and two-kidney, perinephritic hypertension in dogs. He found the femoral but not jugular vein pressure-volume curve of the hyper- tensive dogs to be shifted in the direction of the pressure axis, suggesting decreased distensibility of the femoral vein in experimental renovascular hypertension. Systemic intravenous administration of pro- pranolol (a beta-blocker), diazoxide, or guanethidine (a ganglionic blocker) failed to alter the venous distensibility of hypertensive or normotensive dogs, therefore, the shift of the hypertensive femoral vein pressure-volume curve persisted. Furthermore, the shape of the pressure- volume curves tended to be convex toward the volume axis and were similar in hypertensives and normotensives. The author was never able to demon- strate the characteristic sigmoid configuration of the pressure-volume curves that would suggest venoconstriction. He, therefore, suggests that the decreased femoral venous compliance of hypertensive animals may be attributable to abnormal vascular wall water and electrolyte composition. This suggestion, however, has not been tested. Nor did the author offer an explanation for the normal pressure-volume rela- tionships in the hypertensive jugular veins, which may reflect a lower smooth muscle content of these veins. 38 Summary The earliest demonstrable hemodynamic changes in the various models of experimental renovascular hypertension are simultaneous increases of cardiac output and blood pressure, followed by the return to normal of cardiac output and the rise of total peripheral vascular resistance. A transient increase in blood volume may in part be responsible for the early increase in cardiac output, at least in one-kidney, Goldblatt hypertension. The distribution of the early rise in total peripheral blood flow and the distribution of the later rise in total peripheral vascular resistance among the various organs of the body remain, to a large extent, unexplored. With the exception of limb hemodynamics, no attempt has been made to study specific organ blood flow from the point of view of parallel- and series-coupled resistances and from that of the stage of the hypertension. In par- ticular, there is a complete lack of information concerning the role of the splanchnic circulation in the development and maintenance of experimental renovascular hypertension. Decreased venous distensibility, among others, has been suggested to account for the initial rise in cardiac output, but confirmatory studies are needed. The pathogenetic mechanisms underlying the possibily decreased venous distensibility should be explored. Human Renovascular Hypertension Goldblatt's demonstration that hypertension could be produced by renal artery stenosis led to a search for renal artery lesions in human hypertensives. Goldblatt himself showed that, in some cases of 39 human essential hypertension, the renal artery was obstructed. In 1954, six cases of severe hypertension with renal artery lesions were pre- sented, in which the hypertension remitted following removal of the affected kidney (Howard gt_al,, 1954). Since then, renovascular hypertension has become an established, thoughuncommon, form of human hypertension that is surgically correctable. Systemic Hemodynamics Frohlich §t_al, (1967) examined the systemic hemodynamics of 15 patients with hypertension secondary to non-atherosclerotic, fibros- ing, unilateral renal artery stenosis. The group comprised 12 females and 3 male patients, as this type of lesion occurs predominantly in young women. Patients with atherosclerotic renal artery stenosis were excluded, because antecedent human essential hypertension and unrecog- nized coronary arterial disease could not be ruled out in their case. The control group was matched for age and severity of hypertension, but its sex composition was different (13 male, 2 female patients). The results showed a significantly higher cardiac output and total periph- eral vascular resistance in the hypertensive group of patients. The results remained essentially unaltered when the series was extended to 29 patients, three with atherosclerotic renal artery stenosis (Frohlich gt_al., 1969). Brod g£_al, (1966) studied an older group of nine hypertensive patients with atherosclerotic renal artery stenosis. The authors found no difference between the cardiac output of hypertensives and of normo- tensive controls. 40 Blood Volume The total blood volume, measured by the radioiodinated human serum albumin distribution method, of Frohlich's patients (Frohlich gt_al,, 1967) (see above) was significantly lower than that of their controls. Therefore, the increased cardiac output, seen in patients with non-atherosclerotic renal artery stenosis, cannot be explained on the basis of increased intravascular volume. More recently, the finding of decreased intravascular volume in patients with human renovascular hypertension was confirmed in another series of patients by the same group of investigators (Dustan gt_a1,, 1973), using similar techniques. In addition, the reduction in plasma volume was found to be inversely related to the diastolic blood pressure. The red cell mass was normal in these patients. Regional Hemodynamics There is little information available concerning the regional hemodynamics of human renovascular hypertension. Conway (1963), using plethysmographic techniques, was unable to demonstrate statistically significant differences in the forearm (hand excluded) blood flow and vascular resistance of renovascular hypertensives and of normals, although both measurements tended to be higher in hypertensives. The forearm vascular bed, therefore, does not seem to participate propor- tionately in the increase of total peripheral vascular resistance in this form of hypertension. Brod gt_al, (1966), in addition to measuring cardiac output, compared the regional hemodynamics of 9 renovascular hypertensive 41 patients with those of control subjects, using techniques similar to the ones that they employed in the investigation of essential hyper- tensive patients (Brod, 1960; Brod gt_al., 1962) (see under Human Essential Hypertension). Renal, splanchnic and skin vascular resist- ances were significantly higher in hypertensives than in normotensives. Forearm (skin flow deducted) vascular resistance, on the other hand, was similar in the two groups. Since cardiac output was normal in this group of hypertensive subjects, the data indicate a redistribution of blood flow from the kidneys and the splanchnic bed to muscle. Additional regional flow studies in human renovascular hyper- tension were reported for the kidney. Blood flow to the affected kidney and the ratio of clearance of p-amino-hippurate (RPF) to Tm glucose are diminished, suggesting a decrease primarily in renal cortical blood flow (Traeger gt_al,, 1963). Stamey (1962) found an average decrease of 50 to 60% in renal plasma flow (PAH clearance) when the kidney with the renal artery stenosis was compared to the contralateral kidney. The reduction in renal plasma flow produced a proportional reduction in glomerular filtration rate (inulin clearance), i.e., the filtration fraction was nearly the same in the two kidneys. The contralateral kidney, however, was not evaluated with respect to the kidney of a normal subject. The findings of Stamey were confirmed by Kioschos g§_al. (1967), using the indocyanine-green dye dilution method across the kidney. The comparative study between the kidney with renal artery stenosis and its non-stenotic companion revealed a flow difference of 32%. 42 Somewhat different results were obtained by more recent investigators (Pederson g§_al,, 1969; Ladefoged g;_al., 1969) who studied renal blood flow per unit mass of tissue. In the stenotic kidney, renal blood flow per unit mass of tissue, measured by the 133Xe desaturation technique, appears to be reduced, in comparison to the mean value in normals, in all cases of human renovascular hypertension. There was a small but insignificant difference between the stenotic and the contralateral kidneys, renal blood flow being reduced in the non-stenotic kidney as well. In view of the systemic hypertension, the data indicate that renal vascular resistance was increased in both the stenotic (including the resistance offered by the stenosed artery) and the non-stenotic kidney, but the rise is greater in the stenotic kidney. Vascular Compliance There is only indirect evidence to suggest a change in vascular distensibility in human renovascular hypertension. Ulrych gt_al, (1969) calculated the cardiopulmonary blood volume of renovascular hypertensive patients from the product of the mean transit time of the indocyanine- green dye through the lungs and heart and the cardiac output per second. The authors found the cardiopulmonary blood volume of renal hypertensive patients to be significantly increased, while their total peripheral blood volume was decreased, in comparison with control values. In the absence of heart failure, the distribution of blood between the systemic and the cardiopulmonary segments of the circulation should reflect the relative capacitance function of the peripheral and pulmonary vascular beds. The shift of blood volume from the systemic into the pulmonary 43 vascular bed suggests decreased distensibility of the systemic vascular bed in human renovascular hypertension, due to arterial and/or venous abnormalities. Summary The systemic and regional hemodynamics of human renovascular hypertension are characterized by increased cardiac output, increased renal, splanchnic and skin vascular resistances and low or normal muscle vascular resistance. In contrast to human borderline essential hyper- tension, the increased cardiac output is accompanied by an increased total peripheral vascular resistance, and it is uncertain whether the increased cardiac output is an early, transient or a permanent feature of this type of hypertension. Blood volume is reduced and is inversely proportional to the diastolic pressure. The role of veins has not been directly evaluated. Spontaneously Hypertensive Rats During the past 15 years, a number of strains of genetically hypertensive rats have been developed, notably by Smirk and Hall (1958) in New Zealand, referred to as the New Zealand strain, and by Okamoto and Aoki in Japan (1963), referred to as the Okamoto-Aoki strain of hypertensive Wistar rats. Both strains develop severe and sustained hypertension spontaneously in practically 100% of the cases. When the New Zealand strain of hypertensive rats is compared to matched, normo- tensive, control rats, a difference in blood pressure is apparent by 44 two days of age (Jones g§_al,, 1970). It is generally held that these spontaneously hypertensive rats despite minor metabolic strain-related differences constitute the best model yet developed as an experimental counterpart for essential hypertension in man. The hypertension appears to be multigenetic in origin. There is little evidence for a Goldblatt- type mechanism like in experimental renovascular hypertension (Folkow et al., 1973). Systemic Hemodynamics Pfeffer and Frohlich (1973) measured, under ether anesthesia and thoracotomy, the aortic blood flow, via an electromagnetic flow- meter, and carotid arterial and central venous pressures in three age groups of paired normotensive and spontaneously hypertensive Wistar rats. In the youngest (9-12 weeks) group of spontaneously hypertensive rats, the authors found a significantly elevated arterial pressure accompanied by a hyperkinetic circulation, manifested by significantly increased heart rate and cardiac output. Total peripheral vascular resistance was normal in this stage of hypertension. In the older group of spontaneously hypertensive rats (18-33 weeks and 62-97 weeks), the elevated arterial pressure was associated with an increased total peripheral vascular resistance and a normal cardiac output. The results of Pfeffer and Frohlich were in part confirmed by Iriuchijima (1973). The author studied the systemic hemodynamics, under pentobarbital anesthesia, of spontaneously hypertensive rats and of matched controls, aged 12-23 weeks, using techniques similar to those 45 of Pfeffer and Frohlich. He found the cardiac output to be normal and blood pressure and total peripheral vascular resistance to be elevated in spontaneously hypertensive rats. Blood Volume Blood volume is reportedly normal in the Japanese strain of spontaneously hypertensive rats (Gresson gt_al,, 1973). Other investi- gators, however, report different findings in other strains. Nikodijevic et_al, (1972) performed direct, simultaneous mea- surements of plasma volume and erythrocyte volume by means of double isotope dilution technique (albumin 1251 and 51Cr) in spontaneously hypertensive rats and matched controls, weighing 200-250 gm. The authors found a significantly increased plasma and total blood volume in the hypertensive group of rats. Gresson gt_al, (1973) performed similar studies in male, albino rats, weighing 150-300 gm, of the New Zealand colony of spontaneously hypertensive rats. They used the normotensive parent strain of rats for weight-matched controls. They measured plasma volume, extracellular fluid volume and the total erythrocyte volume by the Evans blue, the standardized inulin (with the renal pedicles ligated) and the 51Cr labelled erythrocyte dilution technique, respectively. The hypertensive rats showed a significant reduction in all volumes measured, in comparison to values in controls. There is, therefore, evidence for normal, increased and decreased blood volumes in spontaneously hypertensive rats. The 46 discrepancies cannot be reconciled unless the pathogenesis of the hypertension in the various strains of spontaneously hypertensive rats is different, or we are dealing with strain- and age-dependent differ- ences that may or may not be related to the hypertension. Regional Hemodynamics Regional hemodynamics of hypertensive rats have been studied by various authors under constant flow conditions. Laverty and Smirk (1961) pump-perfused the intact hindlimb of 6-month-old, spontaneously hypertensive (New Zealand strain) and control rats with the rats' own blood at 1 ml/min, monitoring perfusion pressure. The mean hindlimb perfusion pressure was greater in the spontaneously hypertensive than in the control rats. Folkow gg_al. (1970) perfused, with an artificial solution, both hindlimbs of c. 6 month-old, spontaneously hypertensive and control Wistar rats, and monitored the resulting perfusion pressure. They found the hindlimb vascular resistance to be significantly higher in spontaneously hypertensive rats, almost in proportion to their raised blood pressure. Folkow §§_gl, (1971) also studied renal vascular resistance in 7-month-old spontaneously hypertensive rats and matched, normotensive controls, but at variable flow rates. At very low flow rates, they found renal vascular resistance to be similar in the two groups. At high flow rats, on the other hand, the renal vascular resistance of spontaneously hypertensive rats was significantly lower than that of controls, in contrast to findings in other major vascular beds. In 47 the discussion of their paper, the authors refer to some as yet unpublished data from their laboratory, indicating that, under natural flow, resting conditions, the renal vascular resistance of spontaneously hypertensive rats is similar or slightly higher than that of controls. McGregor gt_gl, (1968) perfused the superior mesenteric artery system of spontaneously hypertensive (New Zealand strain) and control rats, weighing 260-340 gm, at 1 ml/min constant flow, using the blood of donor rats or physiologic saline. They found higher perfusion pressures in the hypertensive rats. Vascular Compliance There is no report in the literature concerning vascular compliance in spontaneously hypertensive rats. Summar From the limited amount of data available, the spontaneously hypertensive rats appear to behave similarly to humans with borderline hypertension, with respect to an early, transient increase in cardiac output, followed by a sustained rise in total peripheral vascular resistance. Regional hemodynamic data suggest that in the chronic stage of spontaneous hypertension in rats, vascular resistance is increased in the limb and mesentery. The vascular resistance of the kidneys appears to be low in comparison to the rise in total peripheral vascular resistance, in contrast to other forms of hypertension. CHAPTER II PURPOSE OF THE INVESTIGATION The purpose of this investigation is twofold: (l) to study mesenteric (ileum) blood flow and pressure relationships in early perinephritic hypertension in order to determine the distribution of increased cardiac output at this stage of the hypertensive process, and (2) to evaluate venous function and to analyze venous wall compo- sition in early and chronic perinephritic hypertension. 48 CHAPTER III METHODS Healthy, conditioned, male mongrel dogs, weighing 18.1-27.2 kg were trained to lie quietly during femoral arterial punctures for blood pressure measurements. Resting arterial blood pressure less than 140 mm Hg mean pressure was documented on at least two occasions in each dog prior to surgery. Postoperatively, femoral arterial blood pressures and hematocrits were measured weekly until the time of the hemodynamic studies. Conditioning over a 2-4 week period included the examination of the stool for ova and parasites and vaccination against rabies, distemper, leptospirosis and hepatitis. During the entire study period, the dogs were maintained on a diet of standard dog chow (Wayne Dog Food, Allied Mills, Inc., Chicago, Ill. 60606) and water ad libitum. A total of 66 dogs were used, divided into the following groups: (Hypertensive)-1, 10 dogs; H-2, 15 dogs; H-3, 8 dogs; (Control)-l, 10 dogs; C-2, 15 dogs; and C-3, 8 dogs. Induction ofLHypertension and the Preparation of Controls Prior to surgery, the dogs were given procaine penicillin (600,000 units) and streptomycin (0.5 gm) intramuscularly, as a pro- phylactic measure against wound infection. A flank incision was made 49 50 under pentobarbital anesthesia (25 mg/kg, iv) and sterile conditions. In dogs of groups H-1, H-2 and H-3, I dissected one kidney free from its fat pad and wrapped it in silk to produce perinephritic hypertension. The silk-wrapped kidney, in turn, was wrapped in saran (Saran Wrap, The Dow Chemical Co., Midland, Mich. 48640), in order to minimize the amount of adhesions between the kidney and the surrounding tissues. In groups C-1, C-2 and C-3, I dissected one kidney free from its fat pad and restored the kidney to its normal position. Saran wrap was laid down along-side the kidney, and left in place in order to reproduce the experimental situation as closely as possible. In groups H-1 and C-1, the contralateral kidney was left intact. In groups H-2 and C-2, two weeks after the first operation, I performed contralateral nephrectomy under pentobarbital anesthetic (25 mg/kg, iv), following premedication with antibiotics (see above). In groups H-3 and C-3, contralateral nephrectomy was performed at the time of the first operation. I studied mesenteric hemodynamics in dogs of groups H-1 and C-1 9-11 days after kidney wrapping and sham surgery, respectively. I studied hemodynamics in dogs of groups H-2 and C-2 two weeks after nephrectomy. Finally, groups H-3 and C-3 were studied 4-20 weeks post- nephrectomy, after four consecutive weeks of arterial hypertension (>140 mm Hg mean pressure). The overall design of the experiments was as follows: Group H-l: Silk bag on one kidney 9-11 days hemodynamic studies. a 51 Group H-2: Silk bag on one kidney 2 weeks contralateral nephrectomy ——-+ 2 weeks hemodynamic studies. ———+ Group H-3: Silk bag on one kidney and contralateral nephrectomy 4-20 weeks hemodynamic studies. ————> Group C-l: Sham surgery on one kidney 8-10 days hemodynamic studies. ————+ Group C-2: Sham surgery on one kidney 2 weeks contralateral nephrectomy * 2 weeks hemodynamic studies. * Group C-3: Sham surgery on one kidney and contralateral nephrectomy 4-20 weeks hemodynamic studies. ——-———> In the course of my discussion, I will also refer to the H-1 group of dogs as early, Egg-kidney hypertension, to the H-2 group as early, one-kidney hypertension and to the H-3 group as chronic (four consecutive weeks of mean arterial blood pressure greater than 140 mm Hg), one-kidney hypertension. I will refer to the control animals (C-1, C-2 and C-3) as the appropriate controls. Mesenteric (Ileum) Blood Flow and *Pressure Relationships Dogs, fasted for 24 hours, were anesthetized with sodium pento- barbital, 35 mg/kg, iv. Supplemental doses of 50-100 mg iv were given later, as necessary, but n9§_until after the completion of mesenteric blood flow and pressure measurements. Following the induction of anesthesia, the dogs were intubated with a cuffed endotracheal tube 52 and artificially ventilated with a positive pressure respirator (model #607, Harvard Apparatus Co., Inc., Harvard, Mass.). Respiratory rate was set at twenty per minute and tidal volume was adjusted so that systemic arterial blood pH was at 7.39-7.43. Heparin (10,000 UPS units) was given intravenously for systemic anticoagulation. After an abdominal midline incision, the greater omentum was retracted, and the ileocecal junction of the intestines was identified. Starting from the ileocecal junction, the mesenteric vascular arcades were counted retrograde and numbered. The in_yjyg_hemodynamic studies were performed in a standardized segment consisting of a mesenteric vascular arcade, numbered between 5 and 10, and a portion of the small intestine corresponding roughly to the mid portion of the ileum. I used the innervated, collateral-free, naturally perfused loop of small intestine plus its mesentery for the study of mesenteric blood flow and pressures (Figure 3) (Scott gt_al,, 1964). Venous outflow from the loop drained through a single vein. With the artery and the extrin- sic nerves undisturbed, the vein was cannulated with polyethylene tubing (PE 240) for direct measurement of the venous outflow. The venous out- flow was directed into a reservoir and continuously pumped back to the dog via an external jugular vein. Through a T-tube arrangement, attached 2 cm distal to the insertion of the venous outflow catheter into the mesenteric vein, venous pressure was measured (large vein pressure, PLV)' Large vein pressure (PLV) was set at 7 mm Hg, by adjusting the vertical level of the venous outflow catheter. The purpose of this maneuver was to 53 Figure 3. Schematic drawing of the innervated, collateral-free, naturally perfused loop of small intestine (ileum) preparation. 55 standardize the preparation in order to allow comparison from animal to animal. I monitored small artery (PSA) and small vein pressure (PSV) through polyethylene catheters (PE 10) inserted into a small artery and retrograde into a small vein, near the junction of the mesenteric vascular arcade and the small intestine wall. Reflux of blood into either catheter was freely obtained. The small intestinal loop was tied at both ends and the mesentery cut to exclude collateral flow. The preparation was kept at 37°-39° by a heating lamp and moist by covering it with saran wrap. Finally, a femoral artery was cannu- lated (PE 240) for monitoring systemic blood pressure (large artery pressure, PLA). Intravascular pressures were detected by a Statham P 23Gb pressure tranducer (Hato Rey, Puerto Rico) and recorded on a Sanborn oscillographic recording machine. Beginning 90 minutes after the initial anesthesia in all dogs, I made repeated measurements of intravascular pressures and of mesen- teric venous outflow. Venous outflow was collected and measured in a calibrated cylinder, accurate to i 0.2 m1. Measurements were carried out in duplicate over two consecutive one minute periods and averaged. Duplicate readings were taken every five minutes for a minimum of three collection periods or until the averaged duplicate readings were within a 1.0 m1 of each other. Venous outflow for the experiment was calculated as the average of all duplicate readings. At the termination of the blood flow and pressure measurements, the loop of small intestine was excised and weighed on a top-loading balance, accurate to i 50 mg (Mettler, model P1200, Highstown, N.J.). 56 I expressed blood flow both as ml/min and as ml/min 100 g.1 of small intestine. I divided means of intravascular pressure gradients by the appropriate blood flow to calculate steady-state segmental vascular resistances as follows: 0 Mesenteric large artery resistance (RA) = (PLA-PSA)/blood flow/ 100 gm of small intestine. . Mesenteric small vessel resistance (RSV) = (PSA -PSV)/blood flow/ 100 gm of small intestine. . Mesenteric large vein resistance (RV) = (PSV-PLV)/blood flow/ 100 gm of small intestine. - Mesenteric total resistance (RT) = (PLA-PLv)/blood flow/100 gm of small intestine. Finally, I compared flows, pressures and calculated resistances in dogs of the hypertensive groups (H-1 and H-2) with those in dogs of the control groups (C-1 and C-2) by Student's t-test, rejecting the null hypotheses at probability values 5_0.05. Mesenteric (Ileum) Blood Flow and Perfusion Pressure Relationships At the conclusion of the mesenteric blood flow-pressure measure- ments in 3 H-1, 3 H-2, 3 C-1 and 3 C-2 dogs, I prepared an additional loop of small intestine for pump perfusion of its vascular bed. The venous outflow from the collateral-free, innervated small intestinal loop was directed into a reservoir and continuously pumped back to the dog via an external jugular vein. Large vein pressure (PLV) was moni- tored as described previously. A blood pump (Sigmamotor), pressure- independent to 300 mm Hg, was then interposed between the left femoral 57 artery and the single mesenteric artery supplying the loop of small intestine. A 16 gauge needle at the end of the pump tubing was care- fully inserted into the mesenteric artery, adjusted, and secured to prevent inflow obstruction. Flow to the loop of small intestine was measured by the direct collection of the venous outflow and varied from 3.9 to 45.1 m1/min. Inflow pressure (IP) was recorded via a needle inserted into the pump outlet tubing. The preparation was kept moist at 37°-39°C. At the end of the measurements, the loop of small intestine was excised and weighed as described previously. For each experiment, I plotted both the perfusion pressure (IP'PLV) (mm Hg) and the total mesenteric vascular resistance (mm Hg/ 1 ml min' 100 9'1) as a function of mesenteric blood flow (ml/min 100 g"). Mesenteric Vein Pressure-Volume (Compliance) Relationships In Vivo During the initial surgical preparation, a mesenteric vascular arcade, numbered between 5 to 10 and different from the ones selected for the flow and pressure studies (see above), was selected for the in_ yjyg_measurement of mesenteric vein pressure-volume relationships. The Y-shaped vein of the vascular arcade was standardized by measuring a 4-4-1 1/2 cm segment of its length in each dog (Figure 4). After the completion of the mesenteric blood flow and pressure studies the vein selected for mesenteric venous pressure-volume measure- ments was cannulated with polyethylene tubing (PE 200) at its downstream and and allowed to drain into a reservoir. The rest of the vein was left 58 Figure 4. Schematic drawing of the Y-shaped (4-4-1 1/2 cm) segment of mesenteric vein used to study venous pressure-volume relationships jg_vivo. 60 intact in its mesenteric covering. Small holes were made in the avascular portion of the mesentery on the two sides of the vein to permit occlusion of the venous segment (and adjacent arteries) by clamping with small vascular clamps at the time of the pressure-volume measurements. An additional vascular clamp was placed at the root of the vascular arcade to occlude arterial inflow. The extrinsic nerves were clamped along with the arteries. The preparation was kept at 37°-39°C with the aid of a heat lamp placed above the operating table. The exteriorized loop of small intestine was kept moist by periodically dripping physiologic saline solution at 39°C on it and by wrapping it in a sheet of saran. To study pressure-volume relationships in the mesenteric vein segment, the arterial inflow and venous branches were simultaneously occluded. The venous segment was allowed to drain of its blood content through the indwelling catheter until intrasegmental pressure was atmospheric. This catheter was used both for the injection and the withdrawal of blood. Intrasegmental pressure was monitored during the entire measurement of venous compliance through a T-tube arrangement attached 2 cm distal to the point of entry of the indwelling catheter into the vein. To produce stepwise increases in intrasegmental pres- sure, I injected 0.05 ml aliquots of the dog's own blood from a glass tuberculin syringe into the indwelling catheter. The blood that actually entered the venous segment had just drained from the vein and was contained in a 10-15 cm segment of the indwelling catheter overlying the abdomen, which was kept at 37°-39°C. Six to 19 0.05 ml 61 blood aliquots were injected, producing intrasegmental pressures of up to 50 mm Hg (injection phase). A ten-second pause after each injection was allowed to establish steady-state pressures. These steady-state intrasegmental pressures were detected with a Statham P 23Gb pressure transducer and recorded on a Sanborn oscillographic recording machine. I then withdrew 0.05 ml samples of blood at similar time intervals, again measuring the resulting pressures (withdrawal phase). The time required for each pressure-volume study, injection and withdrawal phase, was less than 5 minutes. Circulation through the venous segment was then restored by removing clamps. I repeated the same measurements after a waiting period of at least 25 minutes. Data were accepted if all injected blood was recovered from the segment, and intrasegmental pressure returned to atmospheric level i 2 mm Hg. Additionally, the total number of injections required to reach an intraluminal pressure of 40-50 mm Hg had to be the same during repeat measurements. Using the means of two or three series of measurements, a pressure-volume curve representing the injection phase and the with- drawal phase, respectively, was constructed for each dog. Total volumes, during injection and withdrawal, producing intrasegmental pressure of 5, 15, 25 and 35 mm Hg in hypertensive dogs were compared by Student's t-test with volumes in control normotensive dogs. Null hypotheses were rejected at a probability value of §_0.05. In addition, venous compli- ance, AV/AP, in the pressure ranges of 5-15, 15-25 and 25-35 mm Hg, was calculated for the averaged injection and withdrawal curve of each dog. Venous compliances of hypertensive dogs were compared by Student's t-test with venous compliances of controls. Again, null hypotheses were rejected at a probability value of 5_0.05. The analysis of pressure- volume data is illustrated in the Appendix (Figure 18). Superior Mesenteric Vein Pressure-Volume (Compliance) Relationships In Vitro At the conclusion of the 1n_yjyg_hemodynamic studies, I removed a segment of the superior mesenteric vein. The vein was exposed, and its caudal pancreatico-duodenal branch was identified. This branch of the vein serving as mid-point, a 5 cm-long segment of the superior mesenteric vein was dissected in situ, with the collaterals tied off and cut near their origin. The superior mesenteric vein segment was removed and transferred immediately to a chamber, containing Krebs- Ringer bicarbonate solution (NaCl, 118.3 mM; KCl, 4.7 mM; M9504, 1.2 mM; KH2P04, 1.2 mM; Ca gluconate, 2.5 mM; NaHCO3, 25.0 mM; and glucose, 11.1 mM) aerated with 95% O2 and 5% CO2 gas mixture (Matheson Gas Products, Joliet, 111.). The pH of the tissue bath was adjusted to 7.40-7.43, if necessary, by the addition of small aliquots (0.1-0.5 m1) of 0.1 N HCl. The temperature of the tissue bath was kept constant at 38°-39°C by circulating water through its double wall. Each end of the cut segment was secured to the tip of a thin- walled glass cannula with an outside diameter similar to the inside diameter of the vein. The preparation was placed horizontal in the chamber, and the in vitro length of the venous segment was adjusted to 4 cm in each experiment. The vein was perfused in the direction of flow in the intact animal, at constant flow (25 ml/min) with the 63 same Krebs-Ringer solution as used in the chamber (Figure 5) (Vanhoutte gt_al,, 1970). Pressure-volume measurements were made after an equilibration period of 10 minutes, as follows: perfusion was stopped and the inflow and the outflow tubes were clamped at their point of attachment to the glass cannulae. The intrasegmental pressure was adjusted to atm05pheric pressure through the side arm of one of the glass cannulae. Two-tenths m1 aliquots of Krebs-Ringer solution were injected through the side-arm of the other glass cannula. Pressure-volume measurements were repeated once after 25 minutes. Pressure-volume curves were constructed in a manner that is analogous to my jn_yjyg_technique. In addition, I also measured venous compliance by rapid con- tinuous infusion of Krebs-Ringer solution (Infusion-Withdrawal Pump, Harvard Apparatus, Model 600-000, Dover, Mass.). A 9.88 m1/min infusion rate was chosen so that 40-50 mm Hg intrasegmental pressure would be reached within 5-15 seconds. A rate of infusion of this kind was shown by others (Alexander, 1948 and 1963) to give rise to pressure-volume curves that are most sensitive to changes in vascular smooth muscle tone. In 3 H-2 and 3 C-2 dogs, I repeated the measurement of superior mesenteric vein compliance by rapid continuous infusion, following the perfusion of the segment with NaCN insaline (500 mg/liter) at 37°C, for 20 minutes. At the end of the experiment, the venous segment was removed, cleared of the suture ligatures, blotted dry and weighed on an analytical balance (Mettler, Type H 16, Highstown, N.J.) to within 1 0.001 gm. 64 Figure 5. Schematic drawing of apparatus used to study venous pressure-volume relationships in_vitro. 65 m mgamwd 5:05.225 ...... TlEovlt _ .325 (.45 1532.2 illillnw .liiliriiiilurllnitnil . u - N L 50:35: 0.— 66 The jn_vitro data were analyzed similarly to the jn_vivo pressure-volume data. Femoral Vein Pressure-Volume (Compliance) Relationships In Vitro The dogs of groups H-2, H-3 and C-2, C-3 were used. Under sodium thiamylal anesthesia (13.2 mg/kg, iv), a 5 cm segment of the femoral vein, proximal to the branching of the caudal femoral vein, was dissected in situ, with collaterals tied off and cut near their origin. The femoral vein was removed and transferred immediately to the in vitro chamber (Figure 5). Femoral vein compliance was studied in the same way as that of the superior mesenteric vein. During the initial measurement, I injected 0.1 m1 aliquots of the Krebs-Ringer solution at 37°C until an intrasegmental pressure of 40-50 mm Hg was reached to check the patency of the system. Thereafter, femoral vein compliance (infusion phase) was measured by rapid continuous infusion at a rate of 3.88 ml/min on two separate occasions, allowing 25-30 minutes between measurements. The withdrawal phase was not studied. At the end of the experiment, the venous segment was weighed to within i 0.001 gm. The data for the infusion phase were analyzed similarly to the 1 vivo pressure-volume data. 67 Venous Wall Water, Sodium and Potassium Content In a number of animals from groups H-2, H-3, C-2 and C-3, I removed segments of mesenteric and femoral veins and of Vena cava. The segments of vein were immediately dissected clean, opened longi- tudinally, blotted once with filter paper to remove surfaCe blood, placed in preweighed bottles and weighed on an analytical balance (Mettler, Type H 16, Highstown, N.J.) to the nearest 0.01 mg. Then the open bottles were placed in an oven at 100°C for 24 hours. Two hours following removal from the oven, the capped bottles were reweighed. The difference between the two weights represented water content, which was expressed as percent wet weight. The water contents of veins from hypertensive animals were compared to those from normotensive dogs by Student's t-test, rejecting the null hypothesis at a probability value of 5_0.05. For measurement of Na and K content, 5 milliliters of 0.1 N nitric acid were added to each bottle. The venous tissue was allowed to digest for a period of 2-4 weeks, following which the tissue content of Na and K was measured by flame photometry (Beckman Model 105 Flame Photometer). Ion content was expressed as mEq/kg tissue dry weight. The Na and K content of veins from hypertensive dogs was compared to that from control dogs by Student's t-test, rejecting the null hypoth- eses at a probability value of 5_0.05. 68 Histological Examination of the Mesenteric Vascular Arcade andlthe Femoral Vein In a number of animals from each group, a mesenteric vascular arcade and a segment of femoral vein were removed and placed in 10% buffered formalin for pathological examination, which was performed by John F. Dunkel, M.D., Asst. Prof., Dept. of Anatomy, Michigan State University. Serial sections of mesenteric vascular arcade and femoral vein were made. For each section of tissue the following stains were performed: hematoxylin and eosin, elastic, reticulum, trichrome, allochrome and AB PAS. Additional Studies l. Hematocrit. Arterial blood hematocrit was measured by the microhematocrit method, weekly pre- and post-operatively in each dog. 2. Blood urea nitrogen (BUN). At the time of the hemodynamic studies, immediately after the induction of anesthesia, 2-3 milliliters of blood were drawn from the cannulated jugular vein of each animal. The separated plasma was frozen and stored for the determination of BUN, expressed as mg/100 ml of plasma. 3. Urinanalysis. Prior to killing each animal, a sample of urine was obtained by direct puncture of the bladder. Urinary specific gravity was measured by hydrometer. Urinary protein was estimated by Uristix (Ames Co., Elkhart, Ind.). CHAPTER IV RESULTS General data on 66 dogs, 33 control and 33 hypertensive, are presented in Table l. The hypertensive dogs (H-1, H-2 and H-3) and their respective controls (C-1, C-2 and C-3) had similar preoperative body weights, awake mean arterial blood pressures and hematocrits. Following sham operation and nephrectomy in control animals, there was no significant change (P:>0.9) in the mean arterial blood pressure. Following the wrapping of one kidney in silk, all hyperten- sive groups (H-1, H-2 and H-3) showed a significant rise in mean arte- tial blood pressure. The mean arterial blood pressure of early, two- kidney hypertensives (H-l) rose 7% (P‘<0.05, paired Student's t-test). The rise of mean arterial blood pressure in groups H-2 and H-3 was 42% and 43% (P‘<0.001), respectively. One to five days prior to the hemo- dynamic studies, the awake mean arterial blood pressure of hypertensives was significantly higher than that of controls, including group H-l, 128.3 mm Hg, and C-1, 117.8 (P‘<0.05, unpaired Student's t-test). The average duration of hypertension, calculated from the time of induction to the time of the hemodynamic studies, in groups H-1, H-2 and H-3 was 11.2, 29.4 and 88.0 days, respectively. 69 70 .mwmmgummcm _muwnemaopcma Long: .mmwuaum qumcxcoEm; eo cam as» we nmgammmzo ”mmwuapm UPEeczuoEo; ecu op Lowca mAmv ml, :mxm» ”mesmemezmmms xpxmmz mo momem>wmcmugmaxn mo comwgmasoo Lee .Poo.oV.aaaa "No.oV.am "mo.ou.ma 2.: N72 Suz Suz , 80.3 E: 89038; 88.0... ,8; 89.038; 0.3268 2:88 8:: N72 212 muz muz A+F-ov o A+F-ov o A+N-ov o A+_-ov o ucwE:n_m meacmL: NT: S": T: 72 :28sz 33.2 33.2 8.0“ TS 8.03.: 88882 8.5 805 muz mnz Suz Buz 212 072 E 91.3.8 33.: 718.8 8.03.: 0.38.8 33.: $2888: .888 72 012 31.2 :12 Suz 072 :3 8.03.8 23.8 8.01.8 8.03.8 0.3.8 21.: $2888; 8?; mnz wuz Buz 2": 212 212 a: 503.8 @388 23.8 901.8 0.: SN 33.: $883 88 .888 T2 72 3",. Euz 212 S»: a: 903.8 0.0318 23.8 23.: 0.28.8 503.: $8.83 88 88¢ muz .m_uz OP": 8:- 3 8 3m- 3 8 $78 : 8:888: 8 £8 muz muz Euz 212 El. Suz 3:55 1:83;? 333: $5.322 723: .9332 21:: 2.2888 802 88 .888 mnz Tz Euz Biz 21.. 2": 3:5 as“ 05: on“ as: in as: 3... Z: 9N“ ma: mi 2: masses 8°: 88. .88.. m-: m-u N-= N-u 2.: _-u Ammcmgv cweznpm xgmcwcz o>wpmprmaa ecu cowmcmuewq»; we Ammcmcv whee we gangs: can xaw>mcm owewumam meme: .cwmogpm: owe: coop; .muzmmmz anon .mmgsmmmca coopn oxmzm mo 2mmn memos asocw .mmou Amuz use ml: .P-:v m>wmcwugmnxs cw can Amie cam ~-u .—-uv Foepcou cw mcovum>cmmno pmgmcmo .P mFQm» 71 The body weight of both hypertensive and control dogs remained constant throughout the observation period, up to the time of the hemodynamic studies. There was no significant change postoperatively in the hemato- crit of groups C-1, C-3, H-1 and H-3, on a paired basis. In contrast, the hematocrit of C-2 and H-2 dogs, which underwent two separate operations, fell significantly by 5% (P‘<0.001, paired Student's t-test) and 9% (P'<0.02), respectively. At the time of the hemodynamic studies, however, there was no significant difference in the hematocrit of any one hypertensive and appropriate control group. BUN, qualitative urinary albumin and urine specific gravity measured during the hemodynamic studies, under anesthesia, were similar in H-1 and H-2 hypertensive and C-1 and C-2 control groups, with one exception. In the early, Egg-kidney (H-l) hypertensive dogs, urinary specific gravity, 1.031, was significantly lower (P‘<0.02) than in their controls (C-l), 1.040. BUN, urinary albumin and urine specific gravity were not measured in H-3 and C-3 groups of dogs. Mesenteric (Ileum) Blood Flow and Pressure Relationships Table 2 presents mesenteric (ileum) blood flow and pressure data in the early, two-kidney (H-l), the early, one-kidney (H-2) hypertensive and the appropriate control (C-1, C-2) dogs. Table 2 also shows the combined data from groups H-l plus H-2 and C-1 plus C-2. In this and all subsequent tables, all the data obtained were included in the cal- culation of group means unless I state it otherwise. Mesenteric blood 2 7 .m: as N we uceemeee ewe: e maceewwwe meece> we eewuempwee uewgwe as eeeweueoe .eeecm Feeeeeo eueweeegeee new: eeegm m>wmeeugmex; we eemweeesee Lew ._oo.o vases ”No.o yew “mo.ou.es 293mm; :e.38_.o 8.32:5 05.3890 8.3:; 88.3825 58 8.5 :.3 :.N 2.383 :.33.~ 8.3m; 8:383 2.383 388 :28 8.38.0 8.38.0 2.38.0 2.385 2.38.0 8.32.0 :88 2.83 2.3 58 8.388 9.338 8.388 5.325 8.38: :58 AFTEm cow P-Cwe we\mx EEV museumwmmm E E E E E E 059, 88.. 8.03.2 «eff 23.392 939: 0.392 9352 Se, :25 383982 338: 3.8.3882 533:: 933‘: @3422 .38.: :95 $8.33: 93902 1.538% 8.38.2: _.3:S N838: 5.8.8 883 Am: EEV meemmmee 833.3 939:. 93 P8 N358 0.38.8 933:8 To 2: SEE 33.5 0.33: 23.8 5398 9395 833: 35:5 azo_w eoo_m N3 8.8 e3 52. _.3 N8 N3 98 31.: 93 0.: ES 8.588., we 23...: smuz amuz epuz epuz e_uz e_uz N4: 7: 3+ 3 3 E 7: Te ucmwez we 2mm“ memes eeecw .meeceumwmme ece mmeemmeee .zeww eeewe .Eeeww we .ewmecummce Fepweeeeeueme Leece mewgmcewpewmg mgemmmge ece zeww eeewe Azemwwv uwcmpcmmwz .m wreak 73 flow and pressure data were rejected from one H-2 hypertensive dog because of an unexplained drop in blood pressure during anesthesia (mean arterial blood pressure 117.5 mm Hg), prior to the blood flow and pressure measurements. By the time of the hemodynamic studies, this dog was blind, had a mean arterial blood pressure of 225 mm Hg and a hematocrit of 31%. Despite a normal BUN (19.0 mg/100 ml), the dog probably had malignant hypertension. His mesenteric (ileum) blood flow was 18.0 m1/min 100 g']. I also rejected the mesenteric blood flow and pressure data from one C-2 dog, because of atrial fibrillation during the hemodynamic measurements. This dog had a mesenteric (ileum) blood flow of 25.2 m1/min 100 g“. The weight of small intestine (ileum) was similar in all groups. Although there was a tendency for higher blood flows to the mesentery in 1, in hypertensive dogs, mesenteric (ileum) blood flow, ml/min 100 9' groups H-1, 53.9, and H-2, 56.1, were not significantly different (0.10=>P=>0.05) from corresponding values in their controls (C-1, C-2), 48.2 and 46.7. When I combined the flow data from the hypertensive animals (H-l plus H-2) and compared them to the combined flow data from controls (C-1 plus C-2), the mesenteric (ileum) blood flow in hyperten- sives, 55.2 ml/min 100 g" , became significantly higher (P‘<0.05) than that of controls, 47.3. In terms of mean values, the increase in blood flow in hypertensives (H-l plus H-2) was 17%. Because of minor varia- tions in the average weight of ileum studied among groups, the actual measured blood flow, m1/min, was similar (P >0.2) in hypertensives and controls, including the combined data from groups H-l plus H-2, 21.6, and C-1 plus C-2, 19.8. 74 Approximately 90 minutes after induction of pentobarbital anesthesia, the mean arterial blood pressure (large artery pressure) had increased by 35-40 mm Hg in groups H-1 and C-1 and by 25-30 mm Hg in groups H-2 and C-2, compared to values in the unanesthetized state. As a result of a greater rise in mean arterial (large artery) pressure, due to pentobarbital, in C-l dogs, their large and small artery pressure, 156.5 and 122.5 mm Hg, under anesthesia, were higher (P‘<0.05 and P‘<0.02) than the corresponding values, 145.9 and 106.8, in C-2 dogs. Intravascular pressures, large artery, small artery and small vein, under anesthesia, in early, twgrkidney hypertensives (H-l) were similar (P:>0.2) to those of their controls (C-l). The H-l hypertensive dogs, therefore, were not hypertensive under anesthesia. 0n the other hand, intravascular pressures in early, gflg:kidney hypertensives (H-2) and in the combined group of hypertensives (H-l plus H-2) were all significantly higher (P‘<0.05) than the corresponding values in controls (C-2 and C-1 plus C-2). Intravascular pressure rose on the venous side of the circulation too, in that the small vein pressure in H-2 dogs, 15.0 mm Hg, was significantly higher (P‘<0.02) than the small vein pressure in controls (C-2), 11.6. Mean large venous pressure, calcu- lated by adding small vein pressure and large vein pressure, and dividing by two, was 11.0:t0.6 (meanztSEM) mm Hg in hypertensives and 9.4:0.2 in normotensives (P<0.05). As compared to groups C-1 and C-2, the mesenteric (ileum) vascular resistances, total, large artery, small vessel and large vein of H-1 and H-2 hypertensive dogs were normal (P >0.05). The vascular 75 resistances of hypertensives remained normal, when data from groups H-1 and H-2 were pooled and compared to the combined data from C-1 and C-2 dogs. The combined total mesenteric (ileum) vascular resistance of 10 H-1 and 14 H-2 hypertensive dogs, 3.21:tO.17 mm Hg/ml min'1 100 g'1 (mean:tSEM) was practically identical to the combined total vascular resistance of 24 control dogs, 3.23::0.21. Figures 6 and 7 present the blood flow-pressure and blood flow— resistance relationships in the pump-perfused vascular bed of ileum of 6 hypertensive (3 H-1 and 3 H-2) and 6 normotensive control (3 C-1 and 3 C-2) dogs. These data extend the findings of the natural flow exper- iments by showing, over a wide range of flows, the similarity of perfu- sion pressure and total resistance in the mesenteric vascular bed of hypertensives and normotensives. No discernible shift or change of shape of the hypertensive flow-pressure and flow-resistance curve is apparent. Mesenteric Vein Pressure-Volume (Compliance) Relationships In’Vivo The study of mesenteric vein pressure-volume relationships was successful in every dog in which it was attempted. Tables 3 and 4 present the jg_vjvg_mesenteric vein pressure-volume data in early, Egg: kidney (H-1) and early, ongrkidney (H-2) hypertensives and the appro- priate controls (C-1 and C-2). (See Figure 18 in the Appendix for the analysis of pressure-volume data, including the illustration of an actual recording.) Both the means and standard errors of total volume (ml) which produced intravenous pressures of 5, 15, 24 and 35 mm Hg, 76 Figure 6. Blood flow-pressure relationships in the pump-perfused vascular bed of ileum. Mean tSEM perfusion pressure in control (C-1, C-2) and hypertensive (H-1, H-2) dogs. 77 o ecemwu :82 58:85: 803 :8: OO— Om 00 CV ON _ q H - q - a mmzmwmwmill mmuz N10 T I To Td a u. ..Aum. nu ob I. 0 H Lee. d . J a ab ab .en. m .8 .02 m H as Len... 78 Figure 7. Blood flow-resistance relationships in the pump-perfused vascular bed of ileum. Mean tSEM resistance in control (C-1, C-2) and hypertensive (H-1, H-2) dogs. 79 w ecemwu Tweefifizfigefi weeE Hem: e”: Mm owl e. e... _ Tle mwmuzwiwfl. I Joe; m muz WU T I . 1 C) q on L 004. [-3001 l111111 [III/3 H mm‘eoumstseg 80 .395 webcee muewgeegeem 5.5 eeegm mincetmeé we eemteeeee Lew .Neév A; ”medvew meee.euuwmoe.e weee.euupmoe.e Ppee.eunmmee.e weeo.ennw¢ee.e m: as mmnmm mwee.euuewee.e wwee.euueewe.e mmee.euumepe.e m—eo.ennmppe.e m: as mmumw «mpee.eeneepe.e mpee.euume~e.e mwee.eeueepe.e epee.euuem_e.o e: as m_-m Am: EE\FEV .a<\>< .mueawpaeoe mp uz mp uz ow uz opuz emcee «Lemme; mu: «.9 F»: Fuu $8.0 A See Sod.H emmd 256 A mmme God A move a... .5. mm $86 A mend mmeéw emmd Need A $36 mmed u. need a: 5:. mm «See A meme $65 .1. mmed owed n ewmd «Neda eerie a: E: 3 :86“ ~25 mwedn wmmd See.“ e26 Bed n 3:. a: E. m dwee ms=Po> mpuz mpuz epuz epuz 33393 N.: N- p- Flu meeee>eguem ewes; cewpeewcH mewgee .mmee AN-= use P-1V mswmcapcma»; cw new AN-e new _-ev Foepcou.mw Aa<\> ewseueemms 3.: 5 we Em“ gees geese .m e23 8'I .gzogm Fogucoo mumwgqogaam gum: aaogm m>wmcmugmnzg mo comtmaeou .8“. .35 v 3; “mod v A? Fooo.oH m..oo.o «moooo H 386 *m _.oo.o H mm Foo ¢ooooH Pmooo oooo.o H Neooo o..oo.o H eomoo mooo.o H oNoo.o oooo.o H Nmooo 886 H 356 m: EE\FE .¢<\>< .mucmwnmeau pooo.oH w..oo.o oooo.o H meooo mpoooH 886 a: as mm-m~ a: as mN-mF a: as m_-m mpuz m—uz opuz Suz mac? 338.5 mu... «.9 T: To HemoéH mmcd meodH wood ~36H Bmd godH eemé m: .5. mm 38.3 owed NeodH ammo oeod H :36 mgdH mmmd m: 5:. mm «mmodH Zed SBdH «$6 cmodH mmed SodH mwed m: E. 2 omod H mmmd mcod H Smd mmod H ommd «36 H ammo a: E: m AFEV ms=_o> mpuz mpuz Sn: 2...: 3:393 Nu: Nuu F- Fug mzo=m>mgH=H «was; szmgcgpwz mzwgzv .mmou 3.... EB 75 355:8? E 2.... 3-0 28 78 35:8 E. :32: 8522.8 can 3222:pr 2.33-8333 Em> 3,823.9: 3.: E “B EmH :8... 30.5 é ~32. 82 and the means and standard errors of compliance, AV/AP, in the pressure ranges of 5-15, 15-25 and 25-35 mm Hg, during the injection and with- drawal phase, are tabulated. To produce intravenous pressures of 5, l5, 25 and 35 mm Hg, during injection and withdrawal, total volumes were similar (P >0.5) in H-l and C-l dogs. Similarly, the calculated compliances in all pressure ranges were comparable and, as would be expected, compliance decreased as pressure increased in the two groups. In contrast, the total volumes required to produce intravenous pressures of 5, 15, 25 and 35 mm Hg during injection, and 15, 25 and 35 mm Hg during withdrawal, were significantly lower (P‘<0.05 at 5 and 15 mm Hg; P‘<0.02 at 25 and 35 mm Hg) in H-2 than in C-Z dogs. Calculated com- pliances were also lower in hypertensives (H-2) in the pressure ranges of 5-15 mm Hg (P‘<0.02), that is over the operating range, during injection, and in the pressure ranges of 5-15 (P‘<0.02) and 15-25 (P‘<0.05) mm Hg during withdrawal. Calculated compliances were similar in H-2 and C-2 dogs in the pressure ranges of 15-25 and 25-35 mm Hg during injection and in the pressure range of 25-35 mm Hg during withdrawal. Figures 8, 9 and lO illustrate the pressure-volume data in graphic form. Figure 8 illustrates the mesenteric vein pressure-volume relationships ig_vjvg_for early, twgykidney hypertensive (H-l) dogs and their controls (C-l). The two pressure-volume curves are practically identical. Mesenteric vein pressure-volume curves, injection phase, for individual early, one:kidney (H-2) hypertensive dogs and their controls (C-Z) are shown in Figure 9. Overall, curves in hypertensive 83 Figure 8. Mean :SEM in vivo mesenteric vein pressure-volume relationships {injection phase) in sham-operated control (C-l), ___. , and early, two-kidney hypertensive (H-l), ---- , dogs. 84 w mgamwd Emmficuwi ._E.GE:_O> 2. 8. cm. 3. on. an. Sin—6| 2...: _._..........._ 3. o 3 d J a a on n 1 .e w on w H 5 3 Figure 9. 85 In vivo mesenteric vein pressure-volume relationships (injection phase) in l5 unilaterally nephrectomized, sham-operated control (C—2) and 15 early, gg§;kidney hypertensive (H-2) dogs. Each curve represents the average of 2 or 3 separate measurements in one animal. It 86 S 8 [6H waamssaJd -____H-2 C-2 1.00 C” .70 .40 .so .60 Volume [ml] 30 .20 . .10 Figure 9 87 Figure 10. Mean tSEM jn_vivo mesenteric vein pressure-volume relationships (injection phase) in unilaterally nephrectomized, sham-operated control (C-2) and early gggrkidney hypertensive (H-Z) dogs. Withdrawal curve, ......, for controls and hypertensives is also shown. *P < 0.05; #P < 0.02. 88 o_ weaned 53232550523 05. 00. an. 0*. an. 0N. On. _U;O._Ur=_3olo o \\\ 9.2 «. 0| 2.2 a. .7--. .\ \. x .e. x N [3.1 o .. G) n .q; {CD 0') N (BH wwfilnssau O G 89 dogs are shifted in the direction of the pressure axis and are steeper over the low pressure range (O-lS mm Hg). Finally, Figure l0 illus- trates the mean:tSEM mesenteric vein pressure-volume relationships, during injection and withdrawal phase, for H-2 and C-2 dogs. Compared to the control curves, the shift toward the pressure axis of the hyper- tensive injection and withdrawal curve is statistically significant (P‘<0.05). For the sake of clarity, the standard error of mean volumes and the statistical analysis of the hypertensive and normotensive with- drawal curve are not shown. Superior Mesenteric Vein Pressure—Volume Relationships In Vitro I successfully completed the study of the superior mesenteric vein in 20 hypertensive (9 H-l and ll H-2) and in 20 control (9 C-l and ll C-2 dogs. Pressure-volume data were rejected in 3 hypertensive (l H-l and 2 H-2) and in 2 control (l C-l and 1 C-2) dogs because of venous leakage. Table 5 presents the body and vein weights of dogs in which the study of the superior mesenteric vein was successfully completed. The body weights of hypertensives were similar to those of controls. Although there was a tendency for the hypertensive veins to weigh less than their controls, this difference was not statistically significant (P>O.lO). Tables 6 and 7 present the jn_vitrg_superior mesenteric vein pressure-volume data in early, twg7kidney (H-l), early, gggrkidney (H-2) hypertensive and the appropriate control (C-l and C-2) dogs. 90 Table 5. Body weights and vein weights (mean::SEM) in groups subjected to the study of in vitro superior mesenteric vein pressure- volume relationships C-l H-l C-2 =1 H-Z N=9 N=9 N =1 l N l Body weight (kg) 22.4 $0.9 22.4 $1.0 23.0 $0.8 24.4 10.8 Weight of vein (gm) 0.270 $0.024 0.230 10.0l5 0.284 10.019 0.266 10.0183 aN=lO. These tables are organized similarly to Tables 3 and 4. The total volumes to produce intravenous pressures of 5, l5, 25 and 35 mm Hg and the compliances in all pressure ranges, during injection and withdrawal, were similar in groups H-l and C-l. In contrast, the total volumes required to produce intravenous pressures of 5, l5, 25 and 35 mm Hg, during injection and withdrawal, were significantly lower (P1<0.05) in H-2 than in C-2 dogs. In H-2 dogs, calculated compliance in the pres- sure range of 5-15 mm Hg, during injection, but not during withdrawal, was also lower (P‘<0.05) than the corresponding value in controls (C-2). Calculated compliances in the higher pressure ranges, 15-25 and 25-35 mm Hg, during injection and withdrawal, were similar in groups H-2 and C-2, with one exception. During the injection phase, in the pressure range of 25-35 mm Hg, the calculated venous compliance of controls (C-Z) was actually lower (P‘<0.05) than that of H-2 hypertensives. 91 .395 35:8 3330.83 5.; 35.3 $55282 mo commences .8» .Nodv A; ”modv n? mpoodH momod ¥o_.oo.oH oo Pod o~oo.oH Nm pod :oodH mo pod m: =2. mmumw owoodH eomod :oodH Smod :oodH oemod oNoodH momod o: E: mNump HowoodH omood ooood H Emod oooodH «mood omood H moood o: E: mpum Aw: EE\FEV .a<\>< .au=m_Fgeou .: u z 2 u z m u z m u z mono.» 95395 Nu: Nuo T: To «moodH mwoé SNdH mmmd mmFdH omod eode mgd o... .5. mm *mmodHSé; NoNdemod mN—dHNmo; deH Sud o: .5. mm *oordHooTF ooNde—oé mmdeNmoé oodezod o: .5. mp *ooodepmd deHmooé mNPdHoomd omFdH EN; 9. E: m H_ao waspo> :uz :uz muz muz mmgammmga N..: No 7: To 3053.55 omega cowuumncm mcweau .mmov Ami: new 75 032383; use Amuo new 78 35:8 5 Efipslflucmtasou new 3.2.... -5529. 8.32-3333 F2; “.2353... 8.233.. 95.; E .8 2mmH :35 395 d 033 92 .9395 35:8 323933 5.5 95.6 35:3..qu 98 282858 so; .modv 9. 88.0 H :86 386 H $8.9 moood H mused wooed H 386 m: E: mmumm 386 H mde good H mm 8.0 N506 H 88.0 NNOO.OH mm 5.0 m: .E. mmump wmocd H «9.56 88.0 H emmod N306 H mmcoé no 86 H 38.0 m: .E. m_...m a: E>é 32:2 £83358 Z nz Z uz muz muz mmcmg 8:39;. mu: «.9 T: To «mwodH mum; mmde NNN.N oSdH aco.~ omFdH Gv.~ m: .5. mm *Nwodemm; emp.onwP.N ode ~34 mmF.OHmwm.~ m: E: mm «deH Sm; me.onS.m mmdemmw; mmP.OHom~.~ m: .5. mp «deHNmmd cmFdHemmé NEGOH Km; oZdeumé m... Em AFEV waspo> :uz :uz mnz muz mmgzmmmga Nu: Nuo T: To £553.55 + mug: szmgcfimzactzu .38 «.1 new N-:v o>wmcmugoazg ucm Amuo vcm pnuv Fogpcou cw A¢<\> 3.82"..me 3.233 95.; .3 D6 EmH :8... 95.5 K mZE 93 Figure ll illustrates the mean:tSEM superior mesenteric vein pressure-volume relationships, injection phase, for the early, two: kidney hypertensive (H-l) and the appropriate control (C-l) dogs. Although there is a shift of the hypertensive pressure-volume curve in the direction of the pressure axis, this shift is not statistically significant (O.20:>P:>0.05). Figure 12 illustrates the meanztSEM pressure-volume relationships, during injection and withdrawal phase, for H-2 and C-2 dogs. There is a statistically significant (P‘<0.05) shift of the hypertensive injection and withdrawal curve in the direc- tion of the pressure axis. For the sake of clarity, the standard error of mean volumes and the statistical analysis of the hypertensive and normotensive withdrawal curve are not shown. In addition to stepwise injection and withdrawal of Krebs-Ringer solution, I also determined pressure-volume relationships in the same superior mesenteric veins by rapid (lo-15 second) infusion at a constant rate of 9.88 ml/min. Figure 13 illustrates the jfl_vit§g_superior mesenteric vein pressure-volume relationships at constant infusion in H-2 and C-2 dogs. Data on which Figure l3 is based are shown in Table 8. The shift of the hypertensive pressure-volume curve is sim- ilar to that in Figure 12. The total volume infused to reach a given intravenous pressure was, however, lower in both hyptertensives and normotensives. For instance, it required a total volume of 1.82 ml by stepwise injection, and 1.72 ml by rapid infusion, to reach 15 mm Hg pressure in normotensive veins. A similar relationship held for the individual hypertensive veins, but it is not apparent from the 94 Figure ll. MeanztSEM in vitro superior mesenteric vein pressure- volume relationships (injection phase) in sham-operated control (C-l) and early, twgykidney hypertensive (H-l) dogs. 95 00 II ll 2 z '7'"; I. I I I ' I~~ ~~~~ ‘s ‘\ \\‘ ‘\ \ \ \ Q Q 0 C * M N "‘ (6H unu) amssaid 2.0 2.4 Volume(ml),MeantSEM 1.6 1.2 Figure ll 96 Figure 12. Mean tSEM jn_vitro superior mesenteric vein pressure- volume relationships (injection phase) in unilaterally- nephrectomized, sham-Operated control (C-l) and early, one-kidney hypertensive (H-2) dogs. Withdrawal curve is? controls and hypertensives is also shown. *P < 0.05; #P < 0.02. 97 NN meamNN Emancooz._E.oE:.o> flu ed o.u 3 «.N a. _—uN¢z—ur.numb—NN¢eu.-uuuuo 245. ..N 0| =.z «. :i--. N N.. N.. N... o 3 M a S 5 8% w on w H 6 3 Figure 13. 98 Mean:tSEM jg_vitro superior mesenteric vein pressure- volume relationships during infusion at constant rate of 9.88 ml/min in unilaterally-nephrectomized, sham- operated control (C-2) and early, gggfkidney hypertensive (H-2) dogs. *P‘<0.05. 99 m— mcsmwm Emmncooi..E.oE:_o> «a an 0.. «A a. 2.2 «.ol 0...: «+7-... 0 n O G no N BH ww‘amssaad 8 100 Table 8. Group mean:tSEM of 1g_vitro superior mesenteric vein pressure- volume relationships and compliance (AV/AP) in control (C-2) and in hypertensive (H-2) dogs, during infusion at constant rate of 9.88 ml/min Intravenous C-2 H-2 pressures N = ll N = 9 Volume (ml) 5 mn Hg 0.028 $0.140 0.548 i0.08l* 15m Hg 1.723 $0.196 l.l8210.l05* 25m Hg l.972 $0.196 1.446i0.094* 35 mm Hg 2.l13:i0.192 l.604:t0.084* C-2 H-2 Pressure range N= ll N = 9 Compliance, mllmm Hg 5-15 mm Hg 0.0784 $0.007] 0.0634 i0.0044 15-25 rm] Hg 0.0264 $0.0013 0.0264 £0.0013 25-35 mm Hg 0.0140 i0.0006 0.0158 i0.0013 *P‘<0.05, for comparison of hypertensive group with control group. 10] group means for hypertensive veins because a smaller number of veins were studied by rapid infusion, 9, than by stepwise injection, ll. The superior mesenteric vein pressure-volume relationships by rapid infusion in H-l and C-l dogs (see Table 14 in the Appendix) was also similar to the pressure-volume relationships obtained during the slower injection (Figure 8). Finally, perfusing the jg_vitro vein with NaCN (500 mg/liter) for 20 minutes prior to the measurement of superior mesenteric vein pressure-volume relationships by rapid infusion failed to shift the pressure-volume curves of 3 H-2 and 3 C-2 dogs. Since the hypertensive superior mesenteric veins had a tendency to weigh less than the control veins (Table 5), I calculated the corre- lation between the superior mesenteric vein weights and the total volume required to reach 40 mm Hg intravenous pressure, on the one hand, and superior mesenteric vein weights and compliance in the pressure range of 5-l5 mm Hg, on the other hand, in both hypertensives and normoten- sives. Tables 9 and l0 show my analysis. There was no significant correlation (P:>0.05) between the weight of vein and the total volume injected and the weight of vein and the compliance in either hyperten- sives or normotensives. In Figure 14, the weight of superior mesenteric veins and the total volume injected to reach 40 mm Hg intravenous pressure are plotted for groups H-2 and C-2. 102 Table 9. Correlation between weight of superior mesenteric vein (mean i SEM) and total volume (mean tSEM) injected to reach 40 mm Hg intravenous pressure in groups H-2 and C-2 Height of vein Volume (ml) at Correlation (gm) 40 mm Hg coefficient (r) H-2, N=10 0.266i0.018 1.78:0.08 -0.02 C-2, N=11 0.284i0.019 2.23 $0.20 0.20 Table 10. Correlation between weight of superior mesenteric vein (mean i SEM) and compliance (mean tSEM) in pressure range of 5-15 mm Hg, during injection phase, in groups H-2 and C-2 Weight of vein Compliance, AV/AP Correlation (gm) ' (ml/mm Hg) coefficient (r) H-2, N =10 0.266 t 0.018 0.065 t 0.005 0.27 C-2, N=11 0.284i0.019 0.082 $0.007 0.44 103 Figure 14. Scatter diagram for the weight (gm) of superior mesenteric vein and the total volume (ml) injected to reach 40 mm Hg intravenous pressure in 11 control (C-2) and 10 hypertensive (H-2) dogs. 104 Aymax ]I-I]Il|mwr 9.2.m.._._. 5.2.N100 4N mesmNN 59.2903 Sc. 3». 8». Sn. ecu. 4‘ AV 105 Femoral Vein Pressure-Volume Relationships In Vitro I studied the femoral vein jg_vitro from 3 H-2, 8 H-3, 3 C-2 and 8 C-3 dogs. All experiments were successfully completed. Table 11 presents the body and vein weights of hypertensive and control dogs. While the body weight of the two groups was similar, the average weight was less in hypertensive veins, but the difference was not statistically significant (P >0.10). Table 11. Body weights and vein weights (mean tSEM) in groups subjected to the study of 13_vitro femoral vein pressure-volume relationships C-2, C-3 H-2, H-3 N=11 N=11 Body weight (kg) 27.0:0.4 26.6 220.6 Height of vein (gm) 0.136:t0.007 0.122:t0.007 Table 12 presents the group mean tSEM femoral vein pressure- volume relationships and compliance, AV/AP, at a constant infusion rate of 3.88 ml/min. The total volume to produce intravenous pressures of 5, 15 and 25 mm Hg was similar in hypertensives (H-2, H-3) and normotensives (C-2, C-3). The total volume required to produce 35 mm Hg intravenous pressure, on the other hand, was significantly lower (P‘<0.05) in H-2, H-3 dogs. Similarly, the compliance in the pressure ranges of 5-15 and 15—25, but not 25-35, mm Hg was significantly lower (P‘<0.05) in hyper- tensives than in normotensives. Pressure-volume relationships during withdrawal of Krebs-Ringer solution were not studied in these dogs. 106 Table 12. Group mean:$SEM of jg_vitro femoral vein pressure-volume relationships and compliance (AV/AP) in control (C-2, C-3) and in hypertensive (H-2, H-3) dogs, during infusion at constant rate of 3.88 ml/min Intravenous C-2, C-3 H-2, H-3 pressures N= 11 N = 11 Volume (m1) 5 nm Hg 0.241 $0.043 0.189 $0.046 15 mm Hg 0.537 $0.062 0.380 $0.051 25 nun Hg 0.649 $0.065 0.471 $0.048 35 nun Hg 0.709 $0.067 0.526 $0.047* C-2, C-3 H-2, H-3 Pressure range N = 11 N = 11 Compliance, m1/mn Hg 5—15 m Hg 0.0296 $0.0039 0.0175 $0.0017# 15-25 mm Hg 0.0112 $ 0.0011 0.0090 $0.0008# 25-35 nun Hg 0.0060 $ 0.0006 0.0055 $ 0.0005 *P <0.05°,. #P <0.02, for comparison of hypertensive group with control group. 107 Figure 15 illustrates the in vitro femoral vein pressure-volume data. There is a statistically significant (P<:0.05) shift of the hypertensive pressure-volume curve in the direction of the pressure axis at 35, but not at 5 (P >0.4), 15 (0.10>P>0.05) or 25 (0.10>P> 0.05) mm Hg pressures. Since the hypertensive femoral veins, like the hypertensive superior mesenteric veins, had a tendency to weigh less than the control veins, I calculated the correlation between femoral vein weight and total volume required to reach 40 mm Hg intravenous pressure, on the one hand, and femoral vein weight and compliance in the pressure range of 5-15 mm Hg, on the other hand, in both hypertensives and normoten- sives. There was no significant correlation between the weight of veins and the total volume infused in hypertensives (r==0.14) or between the weight of veins and compliance in hypertensives (r =-0.02) and normo- tensive controls (r=0.56). There was a slight correlation (r=0.70, P‘<0.05) between the weight of veins and the total volume infused in normotensive controls. Venous Hall Hater, Na and K Content Table 13 presents data on the water, Na and K content of mesenteric vein, femoral vein and vena cava in early, gngrkidney (H-2), chronic, ggg7kidney (H-3) hypertensive and control (C-2, C-3) dogs. Because of the small number of a particular vein that I have studied and the large experimental variation, the composition of hypertensive mesenteric veins, femoral veins and venae cavae was not significantly Figure 15. 108 Mean:$SEM jn_vitro femoral vein pressure-volume relationships during infusion at constant rate of 3.88 m1/min in unilaterall nephrectomized, sham- operated control (C-2, C-3) and early (H-2) and chronic (H-3) gggfkidney hypertensive dogs. *P <0.05. 109 mN meszN EmméuoE.NE.oE:.o> 8. an. 00. cm. =.z. 6.3-1 =.z want--1- 6.... 8. ON. an. a an .10 a «w .v~" .nu .m w .vmw mu H 5 CD 1' pmmoc om "moon mpu mmmoc e u a “mmou Na .azocm Nocpcou mpmwgqocanw cuNz asogm m>chmucwazg No comNNagEoo Low .No.oV.aea ”mo.oV.¢« 110 N39235: $523.3 1.7.48 5.... T: .3. . - . . - . . - . . chm> NNN 82 1.. 2 37.28 «7.38 5.23 3 3? ~52 38:2 238 mu: 3. 0 no wcm 99:38 9.3368 33.8 m-z 3 N, > 9&2”? 8.5158 273.8 mu: 9: . 92.. 3: 98.. :3 TN“ New mi. 3 59, $38.. «933$: 933.3,... 33.3 m": E 323.8 Sun :3 33.8 mu: 3 £2, 82358: NpcmNa: New mecmsv NpgmNm; New meamsv NpgmNmz pm: NV ucmucou x pcmucou oz acmucoo Loam: mmou.Nm-u .m-uv Nocucou new m>chmpgwaxg Nm-zv.zm:uqummm .zNme cN pcmpcou x ten 82 .swumz PPM: m=o=m> we zmmu cum: .mp mpnmh .oNcoccu .NN-xv xmchxummm 111 different from that of controls, with one exception. The Na content of hypertensive (H-3) femoral veins, 353.1 mEq/kg dry weight, was significantly higher (P-<0.05) than that of controls (C-3), 277.6 mEq/kg. The difference in femoral vein water content between hyperten- sives, 69.0%, and normotensives, 63.2%, was of borderline significance (0.10:>P:>0.05). 0n the other hand, when I combined the data on all the hypertensive and control veins, the water, Na and K content of hypertensive veins, 67.4%, 341.4 and 110.8 mEq/kg, respectively, became significantly higher (P‘<0.05) than corresponding values in controls, 64.0%, 272.4 and 70.9. Histological Examination of the Mesenteric Vascular Arcade and the Femoral Vein The mesenteric and the femoral veins of l H-1, 2 H-2 and l H-3 hypertensive dogs were examined and compared to veins from control (1 C-1, 2 C-2 and 1 C-3) dogs. The veins were studied with respect to the number of smooth muscle cell layers, the size and branching of smooth muscle cells, the content of neutral or acid polysaccharides, the ratio of vein wall thickness to adventitial collar of collagen, and the den- sity and size of collagen fibers. The amount of PAS material, edema and fibrosis in the mesentery was also evaluated. No significant differences between veins from hypertensive and normotensive dogs were identified. Figures 16 and 17 illustrate some of the histological features that Dr. Dunkel and I looked for. Figure 16 shows the relative thickness of the inner, darkly-stained, laminated smooth muscle cell 112 Figure 16. Cross section through portion of a small artery (left) and its accompanying vein from the mesentery. Trichrome stain. x 80. 113 Figure 16 114 Figure 17. Cross section through portion of a small mesenteric vein. Trichrome stain. x 160. 115 ‘ ~ ’ ‘ I'- Figure 17 ,‘r 159‘ . A ' , ‘. a I“ ‘ V‘- ( 13'. 116 layer and of the loose, adventitial collagen collar in a small mesenteric vein (on the right in the figure). Figure 17 is a close-up of the same vein showing the arrangement of the smooth muscle cells and the appearance of their nuclei. CHAPTER V DISCUSSION These studies show that, under pentobarbital anesthesia, the resistance to flow through the mesenteric vascular bed is normal, and flow is slightly but not significantly (0.10 >P >0.05) increased in early, twgfkidney (H-1) and early,ogg;kidney (H-2) perinephritic hypertensive dogs (Table 2). In the combined group of hypertensive dogs (H-l plus H-2), however, mesenteric (ileum) blood flow is increased (P‘<0.05). Mesenteric venous compliance, measured by 1g_vi!g_and in vit§g_pressure-volume studies, is decreased in early, gggfkidney, but ggt_in early,tw9:kidney perinephritic hypertension (Tables 3 and 6). Decreased venous compliance is also present in the more chronic stage of gggrkidney (H-3) perinephritic hypertension, as shown by the jg_vitro femoral vein pressure-volume studies (Table 12). Decreased venous com- pliance is accompanied by increased water, Na and K content of veins from gggrkidney perinephritic hypertensive dogs. The induction of unilateral perinephritis by wrapping one kidney in cellophane or silk, with the contralateral kidney intact or removed, is an effective way of creating hypertension in experimental animals. In this study, every dog with one of its kidneys wrapped in silk became hypertensive, thus confirming the experience of previous investigators (Page, 1939; Ferrario et al., 1970; Overbeck, 1972). Except for the 117 118 development of malignant hypertension (possibly, one dog in my series), the general health of dogs, following the induction of perinephritis, remained good. The small drop in hematocrit of early, ongrkidney controls (C-2) and hypertensives (H-2) was probably due to the fact that these dogs underwent two separate operations in the course of two to three weeks. The stable hematocrit of C-3 and H-3 dogs supports my suggestion. The latter underwent sham operation and wrapping, plus contralateral nephrectomy, in the course of one operation, instead of two. However, the possibility that the drop in hematocrit of H-2 dogs was due to plasma volume expansion cannot be ruled out. In this regard, Ferrario gt_gl, (1970) was unable to demonstrate plasma volume expansion in similarly prepared dogs. A possible disturbance in the water metab- olism of perinephritic dogs is also suggested by the decreased urinary specific gravity in early, twgfkidney hypertensive (H-l) dogs. The decreased urinary specific gravity could reflect an increased water intake. Alternately, the decrease in urinary specific gravity could be due to the perinephritis itself, with impairment of the concentrating ability of the wrapped kidney. The kidney function of hypertensive dogs, as measured by the blood urea nitrogen concentration was, however, normal. The rise of arterial blood pressure is a well-known effect of pentobarbital anesthesia in normal dogs, and was also observed by Over- beck (1972) in perinephritic hypertensive dogs. The significantly greater rise of mean arterial blood pressure under anesthesia in two: kidney controls (C-l) than in ong7kidney controls (C-2) is difficult 119 to explain, since this difference was not observed by Overbeck (1972) in similarly prepared control dogs. Although the C-2 controls had a significant drop in hematocrit after the preceding two operations (see above), the hematocrit of the two groups was identical at the time of the hemodynamic studies. Pentobarbital anesthesia, when combined with surgical trauma, in this case, laparatomy and manipulation of the small intestines, results in the secretion of renin via the activation of the sympathetic nervous system (Fray gt_gl,, 1974). It is conceivable that renin secretion was greater when both kidneys were present (C-l controls) than in the presence of only one kidney (C-2 controls). This difference could explain the greater rise of mean arterial blood pressure in the twgfkidney controls (C-l). Mesenteric (Ileum) Blood Flow and Pressure Relationships Regional hemodynamics in experimental renovascular hypertension have received little attention, with the possible exception of the kidney. The studies which have been performed indicate that in chronic, one-kidney renovascular hypertension, blood flow to the limbs (Overbeck gt_gl., 1971), heart (West gt_gl,, 1959), and brain (Flohr gt_gl,, 1971), and blood flow to the non-stenotic kidney in chronic two-kidney renovas- cular hypertension (Corcoran §t_al., 1942; Bounous gt_gl,, 1962) is normal, as is cardiac output in the chronic stage of the hypertensive process. The hypertension is accounted for by an increase in total peripheral vascular resistance, which appears to be uniformly distrib- uted to the various organs. 120 In contrast to the hemodynamics of chronic stages of experimental renovascular hypertension, there is now evidence that in the early stages (less than 4 weeks) of one-kidney Goldblatt hyperten- sion in rats, and of two- and one-kidney perinephritic and Goldblatt hypertension in dogs, cardiac output is elevated (Ledingham gt_gl,, 1967; Ferrario gt_gl., 1970; Bianchi gt_al,, 1970 and 1972; Ferrario, 1974). In dogs in the early stages of two-kidney perinephritic and one-kidney Goldblatt hypertension, the total peripheral vascular resistance is normal or decreased. Thus a change in systemic hemodynamics apparently occurs between the early and chronic stages of perirephritic and Gold- blatt hypertension. In perinephritic hypertensive dogs, Ferrario gt_gl, (1970) found that this transition took place between the fourth and sixth week: cardiac output, previously elevated, returned to normal levels and total peripheral vascular resistance, previously decreased, increased and became responsible for the sustained hypertension. In unilaterally nephrectomized dogs with Goldblatt hypertension, the tran- sition takes place somewhat earlier: during the first week, according to Bianchi §t_gl, (1970), between the second and third week, according to Ferrario (1974). With regard to an initial rise in cardiac output, followed by a rise in total peripheral vascular resistance, the "auto- regulatory theory" of the development of hypertension was proposed. According to this theory, the increase in blood flow through all or some of the regional vascular beds gradually leads to an increase in total peripheral vascular resistance (Coleman et al., 1971). 121 Overbeck (1972) previously investigated the possible "autoregulatory role" of the forelimb vascular beds in dogs with early, two- and one-kidney perinephritic hypertension, under pentobarbital anesthesia. His studies were performed at the stage of hypertension when Ferrario gt;gl, (1970) found cardiac output first elevated, and also at the stage when cardiac output is at its maximum. Contrary to his expectations, he did not find evidence of increases in blood flow. Actually, forelimb total blood flow per 100 9 weight was slightly, though not significantly, lower in early, two-kidney and early one- kidney hypertensives than in control dogs. As in the chronic stages of experimental one-kidney perinephritic hypertension, there were significant elevations of limb vascular resistance in the two hyper- tensive groups. The author concluded that the limb vascular beds probably do not participate in the change in systemic hemodynamics between the early and the chronic stage of perinephritic hypertension. Without evidence for increased blood flow through the forelimb vascular beds, even in the very early (less than twoweeks) stage of the develop- ment of hypertension, it was unlikely that the increase in forelimb vascular resistance was due to "autoregulatory response" to increased blood flow. It was, therefore, of great interest to me to determine how other vascular beds behave during the developmental stage of exper- imental renovascular hypertension. Specifically, I was interested in finding out which vascular bed receives the excess blood flow and whether "autoregulatory response“ is a plausible explanation for arteriolar constriction in that bed. I chose to study the mesenteric 122 vascular bed because it had not been previously investigated in experimental renovascular hypertension. I studied the very same stages of perinephritic hypertension in dogs that had been previously investigated by Ferrario gt_al. (1970) from the point of view of systemic hemodynamics, and by Overbeck (1972) from the point of view of limb hemodynamics. I used experimental techniques and anesthesia similar to those used by Overbeck in order to allow for comparison between his study and mine. In the early stages of perinephritic hypertension, the regional hemodynamics of the mesentery (ileum) (Table 2) appear to be different from those of the forelimb (Overbeck, 1972). There is, however, little difference in the mesenteric hemodynamics of early, twgrkidney (H-1) and of early, gngfkidney (H-2) perinephritic hypertension. The two groups, therefore, have been considered together. I found the mesen- teric (ileum) total blood flow per 100 gm weight in the combined group of hypertensives to be increased 17% above normal. Interestingly, the maximum increase in cardiac output reported by Ferrario gt_gl, (1970) in early perinephritic dogs was 18%. Total mesenteric (ileum) vascular resistance in hypertensives, on the other hand, ups practically identical to that of controls. Although the average small vessel segmental resistance was slightly lower and large vein segmental resistance slightly higher in hypertensives than in normotensive controls, these differences were not statistically significant. In view of the arterial hypertension, if the mesenteric (ileum) vascular bed of hypertensives behaved in a passive manner, I would have expected 123 the mesenteric total and segmental vascular resistances to be less than normal in hypertension. The normal total and segmental vascular resistances, despite elevated intravascular distending pressures (large artery, small artery and small vein), especially in gggfkidney (H-2) hypertensives, suggest decreased vascular distensibility in hyperten- sion. Decreased vascular distensibility is present not only on the arterial side, but also on the venous side of the circulation, as shown by the increased mean (large) venous pressure in early, gggrkidney (H-2) hypertensives, accompanied by normal large vein resistance. Decreased mesenteric vein distensibility was also shown to be present by the venous pressure-volume studies (see below). The decreased vascular distensibility of hypertensives could result from activation of vascular smooth muscle (washout of local vasodilator metabolites, Bayliss response, circulating vasoconstrictors, etc.) and/or from structural changes (muscle hypertrophy, edema, increased vascular wall collagen, etc.). Decreased distensibility of the hypertensive resist- ance vessels suggests the onset of a "regulatory response" to increased blood flow. However, to further support the "autoregulatory theory" of the development of hypertension, it would be necessary to demonstrate that mesenteric vascular resistance eventually becomes elevated and that mesenteric blood flow approaches normal in chronic, gggfkidney (H-3) hypertensive dogs. This remains to be done. Flow-pressure and flow-resistance curves in the same hyperten- sive dogs overlapped with those of controls. No discernible change in the shape of the curves occurred in hypertension. The increase in 124 pressure as a function of flow was similar in hypertensives and normotensives (Figure 6), as was the decrease in resistance as a function of flow (Figure 7). These findings are not what I would have expected on the basis of the natural flow-pressure studies. Under natural flow conditions, the evidence suggests decreased distensibility of hypertensive resistance vessels. If vascular distensibility were decreased in hypertension, at the same pressure, I would have expected the total vascular resistance of the pump-perfused mesenteric bed of hypertensives to be increased, instead of being the same as that of normotensive controls. I have no good explanation for this discrepancy in findings between flow-pressure relationships under natural flow con- ditions and during pump-perfusion. The discrepancy may be related to the small number of dogs studied during pump-perfusion. The six hyper- tensive (3 H-1 and 3 H-2) and the six normotensive (3 C-1 and 3 C-2) dogs selected at random for the pump-perfusion study were not repre- sentative of the total combined group of hypertensives (H-l plus H-2) and of normotensives (C-l plus C-2) (Table 2). Under natural flow conditions, the total mesenteric vascular resistance of the same six hypertensive dogs was lower than that of hypertensives in general, and the total mesenteric vascular resistance of the same six normotensive dogs was higher than that of normotensives in general. The lumping together of data from early, twgfkidney (H-1) and early, gngrkidney (H-2) hypertensives could also contribute to the discrepancy, since the evidence for decreased vascular distensibility in H-l hypertensive dogs is not conclusive. In this regard, under natural flow conditions, 125 there was no significant difference in vascular distending pressures and resistances between the total group of early, twgfkidney (H-l) hypertensives and the appropriate control (C-l) group (Table 2). Venous Pressure-Volume(Compliance) Relationships The compliance of large arteries is decreased in one-kidney perinephritic hypertensive dogs (Feigl §t_al., 1963) and in men with chronic essential hypertension (Greene gt_gl,, 1966). The decreased arterial compliance is felt to reflect structural changes which, in turn, may be secondary to sustained elevation of intraluminal pressure. The first direct indication that large vein compliance may be decreased in experimental renovascular hypertension came from the work of Overbeck (1972). He showed that the pressure-volume curve of the femoral vein, but not of the jugular vein, in early, one-kidney perinephritic hyper- tensive dogs is shifted in the direction of the pressure axis, suggest- ing decreased compliance of this vein in the early stage of the hyper- tensive process. The author suggested that decreased femoral vein compliance, like decreased arterial compliance, may be due to struc- tural changes, but offered no direct evidence to support his suggestion. Until the work of Overbeck (1972), the evidence suggesting an abnormality of venous function in hypertension was mainly indirect. Plethysmography of the extremities and of the digits in chronically hypertensive men showed decreased vascular distensibility in two studies (Caliva gt;al., 1963; Walsh gt_gl., 1969), but not in two earlier studies (Abramson gt;al,, 1942; Wood, 1961). Furthermore, the 126 possibility that the decreased vascular distensibility of the extremity is due to decreased arterial compliance cannot be entirely excluded by plethysmography. The same criticism applies to the work of Anderson (1954), showing increased mean forearm circulatory pressure in chron- ically hypertensive men, and to the work of Richardson gt_§l, (1964) and Ferrario gt_al, (1970), showing increased mean systemic circulatory pressure in dogs with experimental one-kidney renovascular hypertension. Finally, the demonstration by Ulrych gt_al, (1969) of an increased cen- tral and decreased peripheral blood volume in patients with renovascular hypertension could reflect decreased arterial, rather than decreased peripheral venous compliance, not to mention the possibility of impaired myocardial function in these hypertensive subjects. The finding of decreased venous compliance in early, one-kidney and, possibly, early, two-kidney perinephritic hypertension by Overbeck (1972) is of great interest from the pathogenetic point of view of hypertension. 0n the low pressure side of the circulation, increased intraluminal pressure is unlikely to account for the observed changes, leaving one with the consideration of humoral, nervous and local stimuli. With regard to the recent findings of increased cardiac output in the developmental stages of experimental hypertension and in early, borderline human hypertension, decreased venous distensibility has been proposed as one possible mechanism responsible for increased venous return and cardiac output in turn. It was, therefore, of great interest to me to extend the study of venous distensibility in experi- mental renovascular hypertension to the mesenteric veins and, at the 127 same time, to re-study the femoral vein at a later, more chronic stage of the hypertensive process. The important role of the mesenteric veins in storing blood is well known. Fourteen percent of the total blood volume, at any one time, is contained in these veins (Texter gt_ al,, 1968). The findings of my study indicate that the jn_vivo and the jg_vitrg_mesenteric vein compliance is decreased in the early, one: kidney (H-2), but not in the early, twgfkidney (H-l), stage of peri- nephritic hypertension in dogs. The decrease in venous compliance persists into the chronic stage (H-3) of gggrkidney hypertension, as demonstrated by the in vitro femoral vein pressure-volume studies. The similarity of the results between the jg_vjvg_and the jn_vjtrg_ experiments, using an artificial solution, rules out nervous and humoral factors as responsible for the observed changes in hypertension. One is left to consider structural and myogenic mechanisms. The generally convex shape of pressure-volume curves toward the volume axis in hypertensives and normotensives suggests that the mech- anism responsible for decreased compliance may not involve contraction of the venous smooth muscle. Venoconstriction characteristically changes the pressure-volume curve to a sigmoid configuration (Alexander, 1963). In the case of the mesenteric and the superior mesenteric vein, I have studied "delayed compliance," that is, following each injection, I have allowed ten seconds for equilibration of intravenous pressure. During the ten seconds, the smooth muscle that is stretched will tend 128 to relax. "Delayed compliance“ measurements are not very sensitive for demonstrating venoconstriction. For this reason, I have also studied the superior mesenteric vein by rapid intravenous infusion of Krebs-Ringer solution. There were no differences in the results obtained by step-wise injections and the results obtained by rapid infusion. This finding would suggest that venoconstriction did not contribute significantly to the shift of the hypertensive pressure- volume curves (Alexander, 1963). During withdrawal of injected fluid, the venous smooth muscle is considered to be relaxed. Withdrawal curves are always convex to the volume axis, even when the injection curve shows the characteristic sigmoid (vasoconstriction) configuration. During withdrawal, the shift of the mesenteric and superior mesenteric vein pressure-volume curve, in the direction of the pressure axis, persisted in early, 93g:kidney (H-2) hypertensive dogs. The shift of the hypertensive pressure-volume curves during withdrawal would suggest that structural changes were responsible for the shift (Alexander, 1963). In a few hypertensive (H-2) and control (C-2) superior mesen- teric veins, I attempted to shift the pressure-volume curve in the direction of the volume axis by abolishing venoconstriction with NaCN, a metabolic poison. There was, however, no shift of either the hyper- tensive or normotensive pressure-volume curves toward the volume axis, as one might have expected, again suggesting that venoconstriction did not contribute significantly to the shape of the pressure-volume curves. Despite the evidence favoring factors other than venoconstric- tion as the basis for decreased compliance in ong:kidney renovascular 129 hypertension, the contributing role of venoconstriction cannot be completely ruled out. Although the shift of the hypertensive (H-2) superior mesenteric vein pressure-volume curve in the direction of the pressure axis persisted during withdrawal, the calculated compliance in the pressure range of 5-15 mm Hg was no longer significantly differ- ent from that of controls. Furthermore, during injection, in the pressure range of 25-35 mm Hg, the calculated compliance of control (C-2) superior mesenteric veins was actually lower than that of hyper- tensives. The latter finding is difficult to explain, unless some degree of sigmoidicity of the hypertensive injection curve existed. The sigmoid character of the hypertensive injection curve is not readily apparent from simple inspection of Figure 12. Decreased compliance of the convex (vasodilated) pressure-volume curve, in comparison with the sigmoid curve, is, however, expected in the higher pressure ranges on the basis of Figure 1. What is apparent from the inspection of Figure 12 is that the hysteresis loop (the area enclosed by the injection and the withdrawal curve)is greater for the hypertensive veins. These findings suggest that venoconstriction may, to some degree, contribute to decreased venous compliance in hypertension. The findings of increased water, Na and K content of veins from early, gngrkidney (H-2) and chronic, Qgg:kidney (H-3) hypertensive dogs suggest the presence of structural changes. I have a broad definition of structural change in mind, according to which any alteration in the thermoelastic properties and/or physical dimensions of the vessel wall is considered a structural change (Friedman et al., 1967). The increase 130 in Na and K, 25% and 56%, respectively, far exceeds the increase, 5%, in water content. The extra Na and K of hypertensive veins, therefore, appear to be, to a large extent, osmotically inactive, and are probably bound to ion binding polysaccharides that constitute a large portion of the cellular and paracellular matrix of the vessel wall. The histological studies were apparently not sensitive enough to demonstrate the structural changes that are strongly suggested by the pressure-volume studies and the venous tissue water and electrolyte analysis in gggrkidney hypertensive dogs (H-2, H-3). Notably, there was no evidence for venous wall hypertrophy in hypertension. In addition, following 1g_vjtrg_perfusion, the 5 cm segments of the superior mesen- teric and of the femoral vein from hypertensive dogs weighed the same or slightly less than the 5 cm venous segments from normotensive controls (Tables 5 and 11). General hypertrophy of hypertensive veins is, there- fore, unlikely on the basis of weight measurements and of histological studies. However, vascular wall hypertrophy is neither a necessary consequence nor an invariable concomitant of decreased vascular dis- tensibility (Friedman gt_al,, 1971). In my study, for instance, there was no correlation between the weight of the vein and its compliance (Tables 9 and 10). Perhaps the structural changes, suggested by the findings of my study, are the kind that are hard to see--edema, for instance, or an increase in paracellular matrix (Friedman gt:11., 1967). An excessive paracellular matrix, binding Na and K, may account for the decreased arterial and venous distensibility in gggfkidney perinephritic hypertension. 131 The pressure-volume measurements of this study suggest that the structural changes in the veins of gggrkidney hypertensive (H-2 and H-3) dogs involve the smooth muscle elements of the venous wall. The decrease in compliance of hypertensive veins is over the lower, physio- logic pressure range, 5-15 mm Hg, where the resistance to stretch is due mainly to venous smooth muscle. In other words, the pressure- volume curves of ong:kidney hypertensives (H-2 and H-3) are steeper in the low pressure range (0-15 mm Hg) than those of controls. The shift of the hypertensive pressure-volume curves in the higher pressure ranges could be a cumulative effect on total volume, due to decreased smooth muscle compliance. In this regard, calculated compliance in the higher pressure range, 25-35 mm Hg, is the same or actually higher (see above) in hypertensives than in normotensives, i.e., the pressure-volume curves of hypertensives and normotensives run parallel to one another in this range. In the case of the mesenteric veins, which were studied jg_situ, as they were imbedded in surrounding tissue and covered by the mesentery, decreased interstitial space compliance, as suggested by Lucas gt_gl, (1973), could have contributed to the shift of the hypertensive (H-2) pressure-volume curve in the direction of the pressure axis. Finally, I would like to consider the possibility that the decreased compliance of hypertensive (H-2, H-3) veins was due to a greater initial, unstressed volume and, therefore, more apparent than real. This possibility cannot be entirely ruled out in the case of the mesenteric vein, where complete emptying of the veins could not be 132 ascertained at the time when the intravenous pressure was reduced to atmospheric level. It is possible that, at this point, the volume of blood in the hypertensive veins was greater than in normotensives. This criticism, however, cannot apply to the in vitro study of the larger superior mesenteric and femoral veins, where complete collapse of the veins, as intravenous pressure was reduced to atmospheric level, could be seen. Since venous compliance of the mesenteric and superior mesen- teric veins was found to be normal in early, twgfkidney (H-l) hyperten- sion, it is unlikely that these veins contribute to the initial rise in cardiac output described by Ferrario gt_gl, (1970) in dogs prepared in exactly the same way as my H-l group. Decreased compliance of the mesenteric veins, however, may be responsible, at least in part, for the rise in cardiac output, reported by Ferrario gt_al. (1970) to occur in early, one-kidney perinephritic hypertension (the stage of hyperten- sion that was reproduced by my H-2 group of dogs). Decreased venous compliance, as suggested by my study of hypertensive (H-3) femoral veins 1 vitro, seems to persist into the more chronic stage of the hyperten- sive process, while cardiac output returns to normal. Mechanisms other than the return to normal of venous compliance must account for the return to normal of cardiac output. CHAPTER VI SUMMARY AND CONCLUSIONS These studies show that, under pentobarbital anesthesia, the resistance to flow through the mesenteric (ileum) vascular bed is normal and that flow is slightly, but not significantly (0.10=>P:>0.05), increased in early, twgrkidney (H-1) and early, gngrkidney (H-2) perinephritic hypertensive dogs. In the combined group of hypertensive dogs (H-l plus H-2), however, mesenteric (ileum) blood flow is increased (P‘<0.05). The normal mesenteric (ileum) total and segmental vascular resistances, despite elevated intravascular pressures, especially in early, ggg:kidney (H-2) hypertensives, suggest decreased arterial and venous distensibility. Mesenteric venous compliance, measured by jg_vivo and jg_vitro pressure-volume studies, is decreased in early, gggfkidney (H-2), but not in early, twgrkidney (H-l) perinephritic hypertension. Decreased venous compliance is also present in the more chronic stage of on - kidney (H-3) perinephritic hypertension, as shown by the jn_vitro femoral vein pressure-volume studies. Decreased venous compliance is accompanied by increased water, Na and K content of veins from gggfkidney (H-2 and H-3) hypertensive dogs. Based on the analysis of the hypertensive and normotensive pressure-volume curves and the findings of increased water and electrolyte content of hypertensive 133 134 veins, it is suggested that structural changes rather than activation of vascular smooth muscle are mainly responsible for the decreased venous compliance in gngrkidney (H-2 and H-3) perinephritic hyper- tension. The structural changes appear to be associated with the smooth muscle elements of the venous wall, since the change in compliance occurs over the lower, physiologic pressure range, 0-15 mm Hg, where the smooth muscle is stretched. APPENDIX APPENDIX Table 14. Group mean:$SEM of jg_vitro superior mesenteric vein pressure- volume relationships and compliance (AV/AP) in control (C-1) and in hypertensive (H-l) dogs, during infusion at constant rate of 9.88 ml/min Intravenous C-l H-l pressures N = 9 N = 9 Volume (ml) 5 nm Hg 1.003 $0.099 0.891 $0.133 15 nm Hg 1.860 $0.171 1.510 $0.131 25 run Hg 2.081 $0.183 1.731 $0.129 35 nm Hg 2.208 $0.188 1.855 $0.131 C-l H-l Pressure range N =9 N = 9 5-15 mm Hg 15-25 mm Hg 25-35 mm Hg Compliance, ml/nm Hg 0.0856 $0.0135 0.0619 $0.0040 0.0223 i 0.0018 0.0221 $0.0124 0.0130 $0.0008 0.0124 $0.0008 135 Figure 18. 136 Analysis of venous pressure-volume data. A. Actual record of superior mesenteric vein pressure- volume relationships, injection and withdrawal phase, jn_vitro. Stepwise injection and withdrawal of 0.2 ml aliquots of Krebs-Rin er solution, horizontal axis, vs. intravenous pressure (mm Hg). vertical axis. Paper speed at 1 mm/sec. 8. Plot of pressure-volume measurements, injection phase. Points were taken at 10 seconds following each injection. The calculation of venous compliance in the pressure ranges of 5-15, 15-25 and 25-35 mm Hg is also shown. 137 E 952... 2518......» o." 3 N... a. c. a u u d n u n o. “ AM a: E... n Tn “ u. n 29 o. u m «M a: a... 8.3 _ u - ) u ........... 1----.. ..m o— _ H ~.>- an a... 84.... n n... 1|.- lllll lllllllllllllllllll- ,nn m