-_ -.v . 11. 7... “iv..—.-—' . “III. r ,,,.- ..1 .g.—~ _ l > l""| ""II “‘l‘u I' I"‘ . III 7"”, It; 0 II ‘ ‘II.I"‘ "V IL: .‘I- “IZ’ I‘?'? ';p leg .I'."2 '1‘)" .I' “.1, I. I‘ll . . :I l ‘1’" b, J“ I, 'l" I ' v1." » |.; ' .' ., at I. H. . 7- I‘g gIl; '.’ ‘ ”I: ;‘,._"§ IIII I], II'? III I: IfiJII‘ " I III‘II II‘I I ‘I'IIH I? Ifél'IIl I": I.” 'I “Ilnggkw ‘1':| I l " I ‘ ‘.l ‘ II '7 v .- I ‘ ‘ . IIITI'I' ‘-.H.' r' I'-I,‘I,‘I;|-Iv,«IlI “7v i‘k’fixm‘rq 1:; I; Igl ”II-“gun'g>,l,Y.;. II I: I "III. “III” “III Ii'lr'ulzlll”WI «"IIII'IW. M IIl.II'.vI'IH‘,g " . . .r; 4r . IIII’IIIIII'IILIIIIIII’IIIFII“ 4‘ I IIIIIIIINIIIII IEIIJ Iglg‘blj'u‘ ‘1‘“; _ '.‘I'mu' .I'IIMII ’II I ' 'l ". III I4' -, «I I: 5‘4 ‘- 35"" ,,| l.’ II g I h . I III’ 3 "I fit“ tn.» '1' ‘ ‘13:". ’;If' {Iril‘ III :H. ' I . I I‘ I I “I. ‘ ‘W;‘. 3: «1‘.th " I U" ‘_ 4— -- _._-:....¢_.... saw—"'— . ggngg yv"l_/RAYIJI‘IEFIET-fif$1!W¥.’T’W‘fii?tW‘gwfi, L...“ . _ I g . I '.' I”: I v I...“ ,' u I" ‘ ;....-"~1_. I it}: 1:“ ‘3’ “‘Il" I I \ I“ '.‘ . - I V '1' Mm -. 5. 'g ' ‘ -°'I"‘I’ 2* ~~ ." "' ~ l .g' .' I -' 4' ‘ 0 ‘."l ‘ 'é " . I "IIIIIIII‘ '0 ‘|-.‘,- I:,‘Il‘l ' ‘lgngI - ':' lug ‘I II. .'l f . I ' a u 1 ul- ‘ 0 fl 0 .‘. ’ .‘. — . f l . - _ . “'11: ,;"’," " .2 , I' . . - ‘5 ‘ . '1. :I' I”: I 1' n' 'n' 3: :. I . , - I I ‘I’N '3'. ‘3‘“... ' u c ' . I I t ' ~ I _ |' . _ n I o a g o .‘. g" I, ‘ q' I ‘ ‘ o ‘ r I n ‘....'J I. .Il: 3337i », ' :-‘.’gw’. . iJHII ’I‘I!‘ VI .III'I‘If’I IIII H .- 12% _'..' - It... .1! m1: ;_: :u; {1; : '.- r'- I!?"‘r {I'I iIl, fit 11“”?th I I 0' 4|]. I ‘1. - ._ pf ' I, ‘IIlll‘I." ",‘v .‘I'H’: ': “I.“I‘II Ilfl'..'.I":' 1'lg.» 5‘4. - II'IiI I 5:9,. 'll;l In Ill! IIIIIU'III film}: Iggy.) , 3 I II.,:,;." II IIIUI . “I" IIIIII‘II.III'II‘gz'fgifgIII IIII '51gi II‘I:I:III\'I|.1':I"IIO g MK. 363”“, . .{iiglq‘i1 7 .. .- . - 1"" ):I-' 2.“; '4 g ‘8” ““4“" w 13%? 1' , 4N3; *E‘ 7:.) L}: f 33'“ ’W‘h‘v ~_—.~ l l 7'l1i’IV'lIIIIIIIIIJIII. . III": ...: “b 1‘ “1‘: l I I"I .EII' III' II III: I lIIJ'II. ’ , £51359: M , I ”Int 1"? ,dAII . 1 \(J ~( I ‘ I .; Will}; W 1‘“ ll :‘ ‘ My ' ling}; RlI-‘k; llifil‘fif '31? I II,’ \I “I“ I II 1+” ”I W") ‘ IIIIIVIM I ' 51:; Ii‘XNALyg \ ,7. I ..~. I I III III III” I fréizfi‘siwfia qill I I “ht-In}: EEI .t‘g. . ' . ,. ' . ._- o . . - - _ i. .l . ‘ . . . n ‘ . ~ v ' _ ‘.*.¢.~ . ’ .~ up ..——-.~7}¢ ' ‘ . o b . 4 _ . ~ . I 4 . ‘ - ‘ . a . . ,- ~ .' .- ~ - - ‘ A “ a y A I: . -n s- -: . .2.-. .. ‘ ., .. . -. . II”. 'l’ IIEII ‘ lIlI JIII.II1‘II I . W I "IIIg‘ lll'l"IIr I. "III [WI “ “I l 'II'g ‘1. lI‘I III“; {III I I.,l I g ”I IIgIII 'III I‘ IIIIggIII. II III III“ -W *. ow _ ‘25,, - 4.-..n- l «~.. III ’ n | 6 I' ',I g ' "4/4 ‘1)“ I; I III . g; I‘ , IIII'IIII " gar 4:: 33III"I’-.III.‘ H m l .. l1, 3' .”I-ellel-IIglI'll'l Il . ‘ 'I' .1 « :~'»”I=:l-Illllulu . 3 .’ II: . ‘ " ‘I ‘r‘..’“"v‘-' WWII '“ , II‘V “ lllnlilmiul I‘j}}:',"'”IM. .llu 3‘ ' II ‘ Ilg' gI “If: IuI'I .I o. .g, ”I‘LII J luv II fix I: - ~,' III‘. :l‘I III"IIII‘ ‘21:” .‘I'I‘ III‘III‘I. I g IIIlI’l IIII‘I‘Y 'II'g'“ ,. .‘II‘ . ”i {'1‘ ~ VIII . II II .IIIIIIIII INIIIIIIIIIIH'I‘. {VIII ‘ I ?":N‘: III IV I' "III' § III!“ .III‘ ‘II' ”I“ ”III”. I; II I II ""II If!” :.I‘|IIII"L}[IQ ,. :7 I I! I II ‘J an)”. III I”, rl ”i. JII'III I:II :IWIIJ‘. ”:3le! m4!!! III,” I. :llj III I: l; IIIII""‘IIEII"-I 'I." . I‘ 5 I II II.” I II J Lynn Any 31m: tkl' ) MIC-019;“. Jiatc University This is to certify that the thesis entitled INCREASED RESISTANCE AND IMPAIRED VASODILATION IN THE NORMOTENSIVE HINDQUARTERS OF RATS WITH COARCTATION HYPERTENSION presented by David Robert Bell has been accepted towards fulfillment of the requirements for M. S . degree in Physiology Aaffl/W Major professor Date Mfg/77,7 0-7639 7’? ‘- INCREASED RESISTANCE AND IMPAIRED VASODILATION IN THE NORMOTENSIVE HINDQUARTERS OF RATS WITH COARCTATION HYPERTENSION By David Robert Bell A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Physiology 1977 ABSTRACT INCREASED RESISTANCE AND IMPAIRED VASODILATION IN THE NORMOTENSIVE HINDQUARTERS OF RATS WITH COARCTATION HYPERTENSION By David Robert Bell The role of pressure in structural vascular changes in coarctation hypertension was examined in isolated, innervated, pump-perfused hindlimbs of rats with 4 weeks of abdominal aortic coarctation hypertension (Group A); normotensive control rats (Group B); and normotensive abdominal aortic coarcted rats with hindquarters atrophy (Group C). Hindquarters intravascular pressures were always normotensive in Group A. In the hindlimbs of Group A com- pared to either groups B or C, pressure-flow curves were dis- placed toward the pressure axis (p < 0.01) while resting resistance, resistance after maximal vasodilation, and plasma renin concentrations were increased (p < 0.05). Thus, in normotensive beds of rats with coarctation hypertension: 1) Resistance is elevated; 2) Structural vascular changes indicated by impaired vasodilation may contribute to this elevation; 3) The changes are not attributable to bind- quarters atrophy but may be related to plasma renin; 4) Ele- vated resistance and impaired vasodilation were not caused by high intravascular pressures. DEDICATION To my wife Pamela whose constant patience and support was always felt and greatly appreciated. My greatest debt of thanks belongs to you. ii ACKNOWLEDGEMENTS I would like to acknowledge the help and advice of Drs. H. Overbeck L. Wolterink, J. Scott, and B. Selleck during my research and the preparation of this thesis iii TABLE OF CONTENTS List of Tables List of Figures Introduction Literature Review Methodology Results Discussion Appendix A: Problem Solving and the Development of Experimental Protocol Appendix B: Reasons for Limb Weight Normalization Appendix C: Data and Results of Statistical Analyses Bibliography iv Page v Page Page Page Page Page Page Page Page Page Page vii 3O 45 72 94 105 108 144 10. ll. 12. 13. 14. 15. 16. LIST OF TABLES General Parameters Average Systolic Caudal Arterial Pressure lst Measure- ment Average Systolic Caudal Arterial Pressure 2nd Measure- ment Average Systolic Caudal Arterial Pressure 3rd Measure- ment Average Systolic Caudal Arterial Pressure Total Measure- ments Week-to-Week Comparisons of Average Systolic Caudal Arterial Pressures Direct Mean Carotid Arterial Pressure Direct Mean Femoral Arterial Pressure Resting Resistance x Opposite Limb Wet Weight Resting Flow/gram Opposite Limb Wet Weight Resistance after Acute Denervation x Opposite Limb Wet Weight; Flow = 1.0 cc/minute Resistance after Sodium Nitroprusside x Opposite Limb Wet Weight; Flow = 1.0 cc/minute Average Pressure x Opposite Limb Wet Weight; Flow 0.125 cc/minute Average Resistance x Opposite Limb Wet Weight; Flow 0.125 cc/minute Average Pressure x Opposite Limb Wet Weight; Flow = 0.25 cc/minute Average Resistance x Opposite Limb Wet Weight; Flow 0.25 cc/minute 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. Average Pressure x Opposite Limb Wet Weight; Flow 0.5 cc/minute Average Resistance x Opposite Limb Wet Weight; Flow 0.5 cc/minute Average Pressure and Resistance x Opposite Limb Wet Weight; Flow = 1.0 cc/minute Average Pressure x Opposite Limb Wet Weight; Flow cc/minute Average Resistance x Opposite Limb Wet Weight; Flow 1.5 cc/minute Average Pressure x Opposite Limb Wet Weight; Flow 2.0 cc/minute Average Resistance x Opposite Limb Wet Weight; Flow 2.0 cc/minute Body Weights (grams) Opposite Limb Wet Weight/Body Weight Heart Weight/Body weight Left Kidney Weight/Body Weight Right Kidney Weight/ Body Weight Plasma Creatinine mg% Plasma Sodium mEq/L Plasma Potassium mEq/L Plasma Ca1cium.mEq/L Plasma Magnesium mEq/L Hematocrits Plasma Renin Concentrations (ng/m1)/hr Angiotensin I Heart Weight/Body Weight; for rats used for plasma renin concentrations vi (”MGM-$5005)?“ 10. 11. LIST OF FIGURES Pressure Drop across Pump Outflow Tubing and Cannula Opposite Limb Wet Weight/Body Weight Resting Resistance x Limb Wet Weight Resting Flow/gm Limb Wet Weight Average Pressure versus Flow Average Resistance versus Flow Average Resistance versus Average Pressure Resistance after Acute Denervation x Opposite Limb Wet Weight at a Flow of 1.0 cc/minute Resistance Times Opposite Limb Wet Weight at 1.0 cc/minute after Supramaximal Injection of Sodium Nitroprusside Plasma Renin Concentration Theoretical Vessel Radii and Circumference at Maximal Vasodilation and at Rest in the Rat Hindlimb vii Introduction In chronic arterial hypertension the systemic arte- rioles have thickened walls and narrowed lumens. The wall thickening has been attributed to medial hypertrophy and an increased wall water content. Both have been assumed to result from increased intravascular pressure. Animals with coarctation hyper- tension provide an interesting model for examining the role of pressure in the development of such vascular changes. In this type of hypertensive animal the arterial regions above the coarctation exhibit an increased intra- vascular pressure. However, arterial pressure in the regions below the coarctation are usually normal or near normal. Therefore, any vascular changes seen in these normotensive vascular beds could not be attributed to high intravascular pressures. In an earlier investigation, we observed that increased vascular wall water content occurs in the normotensive vascular beds of rats with coarctation hypertension. Therefore, we believe the assumption that vascular wall thickening in hypertension results from increased intra- vascular pressure is in question. The purpose of the present investigation was to deter- mine if resistance changes accompany this vascular wall edema, by examining the isolated, innervated blood-pump perfused hindlimb of rats with coarctation of the 1 abdominal aorta above the origin of both renal arteries. Consequently, the experiments were designed to measure resistence at rest and over a range of flows, in a normo- tensive vascular bed of rats with coarctation hypertension. The pressure-flow studies were intended to measure the effects of passive distention in the hindlimb vascu- lature. Also, resistance was to be measured in the acutely' denervated hindlimb during maximum active vaso- dilation induced by sodium nitroprusside. Measurements of both passive and active vasodilation were intended to give the investigator some insight into the actual structure of the vessel walls in the hindlimb. An elevation in the resistance measurements, (especially after maximal vasodilation) found in the normotensive hindlimb of rats with coarctation hyper— tension would be evidence that increased vascular resistance and structural alterations in the vessels in hypertension are 393; a secondary adaptation to an elevated intravascular pressure, but, instead, might somehow be related to the cause of the hypertension itself. Literature Review One of the major features of chronic arterial hyper- tension is an increase in peripheral vascular resistance, which, in turn, is responsible for the maintenance of the elevated blood pressure. Since blood viscosity, the sympathetic nerves, a variety of humoral stimuli, myogenic influences and the structure of arterial vessels can all influence vascular resistance, any or all of these factors then could, theoretically, be implicated in the etiology of hypertension. Pickering, in his early text on high blood pressure, reviewed previous investigations which showed that a generalized wall thickening of the arterioles occurs in chronic arterial hypertension. (40) Although recent his- toligical studies by Bevan et a1. support these results (3, 4), very often it is hard to detect true differences in the wall-to-lumen ratio between two vessels. One can not tell, for example, if he is observing a vessel with a truly thickened wall, or, a thin walled vessel that had contracted to a thicker state at the time of tissue fixation. Therefore, some other method is needed for examining the structure of the arterial vessels in hypertension. Smirk first suggested, and then Folkow and his co-workers investigated the hemodynamic consequences of vascular wall thickening; an area of which, by itself, 3 4 little was known at the time of his initial studies. Folkow postulated that different types of wall thickening could exist in hypertension. He also felt that the structural characteristics of the resistance vessels which in turn are affected by wall thickness, are best displayed when all smooth muscle activity in the vasculature is abol- ished; that is, when the vasculature has been brought to the state of maximal relaxation. (46, 10) Folkow hypothesized that, depending on the type of vessel thickening, three values for vascular resistances could be expected during a state of maximal vasodilation. Firstly, the vascular wall thickening could expand the vessel wall primarily in an outward direction so that, at maximal dilation, the lumen (and the resultant resis- tance to flow) of the hypertensive vessel would be the same as a normotensive vessel under the same conditions. Secondly, the wall thickening could progress in a primarily inward direction thus encroaching on the lumen of the hypertensive vessel even at maximal vasodilation. This, then, would result in an increase in resistance during maximal vasodilation. Finally, the vascular wall thick- ening could proceed in such a manner as to expand both the vessel wall and the lumen in an outward direction. In this latter case, the hypertensive vessel would show a decreased resistance at maximum vasodilation. More importantly, the thickened vessel wall, in and of itself, would affect vascular resistance in a much different manner than a "normal" thin-walled vessel, as soon as the smooth muscle of these vessels begins to con- tract. Folkow reasoned that the thicker vessel wall would have more tissue mass situated inside the line of force of smooth muscle contraction. Consequently, in a thick- ened vessel, more tissue mass would be forced into the lumen at any given level of smooth muscle shortening com- pared to a vessel with a normal wall. It follows also, then, that the resultant increase in resistance from the shortening of the muscle in the thickened wall would be considerably greater than that for a normal vessel. This exaggerated increase in vascular resistance for a given level of smooth muscle shortening has been termed "struc- tural hyperresponsiveness" and would occur totally apart from any variation in vascular smooth muscle sensitivity. (10, 12, 14, 16, 45) Moreover, Sivertsson has shown that this phenomenon will be observed in any comparison of a thick walled vessel with a thinner walled vessel by his study of A-V shunts in cats. (45) The forelimb vasculature of the cat contains two structurally different types of arterial circuits. (he cir- cuit is primarily involved with regulation of nutritional blood flow to the forelimb tissue, while the other "shunt" circuit is involved with the thermoregulation of the linb. The resistance vessels of this latter circuit have thicker walls compared to the resistance vessels involved in regulation of the nutritional flow. Sivertsson observed a vascular hyperresponsiveness to sympathetic nerve stimulation and intra-arterial norepinephrine in the thick walled vessels of the shunt circuit compared to the relatively thin walled vessels of the nutritional circuit. Consequently, vascular hyperresponsiveness would be expected in hypertension if, indeed, the vessel walls were thickened. As a first step in investigating these theories in hypertension, Folkow and his co-workers examined the fore- arm circulation of humans with established, benign, essen- tial hypertension. They found that resistance was still increased in the forearms of patients with essential hyper- tension, even during metabolically induced maximal vasodila- tion. (10) Also, in a separate study, Conway had shown that the diastolic pressure of hypertensive rabbits was still elevated, compared to normotensive controls during maximal vasodilation with nitroglycerine. (6) These findings at maximal vasodilation, suggested, but did not absolutely prove that a primarily inwardly directed vascular wall thickening occurs in the vascula- ture of hypertensives, i.e., there was some structurally based component to the increased resistance seen in hyper- tension. This structural component would raise the base- line from which all vascular changes would take place, and thus, as theorized by Folkow, cause an exaggerated increase in resistance at any given level of smooth muscle shorten- ing. (In addition, Sivertsson and Folkow have stated that an increased resistance at complete vascular smooth muscle relaxation, apart from any thickening of the vessel wall, will also raise the baseline from which all vascular changes take place and, therefore, also cause an exagger- ated (though not as great as if the vessel wall was thick- ened too) increase in resistance at any given level of smooth muscle contraction. (12, 45) ) The finding of evidence suggestive of vascular wall thickening in hypertension is very important because 05138 implication to the development and especially the maintenance of the elevated arterial pressure in hypertension. Theoretically, such a structural vascular change could elevate peripheral resis— tance and arterial blood pressure even though all other factors (such as neurogenic or humoral based smooth muscle vasoactivity) remained unchanged. Therefore, structural changes in the vessel wall could contribute greatly or even account for all the increase in resistance seen in chronic arterial hyperten- sion. Because of this possibility, considerable attention has been given to finding evidence for structural vascular changes in chronic arterial hypertension (3, 5, 6, 10, 11, 12, 14, 15, l6, 17, 24, 25, 26, 28, 30, 33, 34, 35, 36, 38, 40, 45, 46, 49, 50, 51, 52, 53). Folkow and his associates have conducted subsequent studies in experimental hypertension in this regard. These studies support the theory that vascular wall thickening occurs in chronic arterial hypertension. The experimental model of hypertension employed by Folkow and his associates consisted of rats with a genetic predisposition toward hypertension. These rats naturally become hypertensive early in their deve10pment without any experimental inter- vention by the investigator. Thus, these animals are often referred to as spontaneously hypertensive rats. (11, 12, 14, 15, 16, 27, 28, 36, 38, 39, 44, 53) In these spontaneously hypertensive rats perfused with an artificial medium, Folkow's group has found that vascu- lar resistance of the hindlimb is increased at maximal vasodilation, and, in addition that the slopes of norepine- phrine dose-response curves for these rats are significantly increased compared to normotensive controls, even though the thresholds to norepinephrine remained the same. (12, 16) In other words, this vascular bed showed the type of vascular hyper-reactivity that would be expected from vessels with thickened walls. This same vascular bed also showed an increased maximal response to vasoconstric- tor agents. (An increased maximal vasoconstrictor response and slope of the dose-response curve has also been demonstrated by Folkow in the renal vascular beds of spontaneously hypertensive rats. (14)) An increased maximal vasoconstrictor response is indicative of an increase in the amount of contractile tissue mass in the vessel. This increased maximal vasoconstrictor response, along with the increased resistance at maximal dilation and increased slope of the dose response curve, together then, can according to Folkow, only be explained 9 on the basis of a primarily medial hypertrophy of the smooth muscle in the resistance vessels of the hindlimbs of these rats. (12, 16) In addition, Folkow states that the hemo- dynamic consequence of the proposed structural change (cal- culated from a model based on the data concerning the "hyper- trOphied" vessel) can largely alone account for the raised resistance at rest in the SHR without necessitating any increased smooth muscle activity. (12) Other investigators have now provided a variety of data that is supportive of vascular wall thickening in hypertension. Sivertsson, in 1968, not only reproduced the attenuated vasodilation in the forearm of humans with essen- tial hypertension, but also reported a significant increase in forearm vascular response to norepinephrine. (45) As in Folkow's study with spontaneously hypertensive rats, no increased sensitivity to norepinephrine was demonstrated. Also, Lais and Brody have reported significant vascu- lar hyperresponsiveness in spontaneously hypertensive rats that they felt was at least, in part, due to vascular wall thickening. (27, 28) Phelan reported a significantly increased resistance in the pump perfused hindlimbs of chronic two-kidney Goldblatt hypertensive rats at maximal vasodilation, (38) as did Overbeck in the pump perfused forelimb of dogs with Chronic one-kidney perinephritic hypertension. (34) In Overbeck's study, passive distention of the vasculature, caused by increasing the perfusion pressure, could not reduce resistance in the hypertensive animals' forelimb 10 to a level attained in normotensive dogs at a distending pressure half as great as that in the hypertensive animals' limb. This impaired passive distention has also been demon- strated in the renal and general systemic circuit of spontaneously hypertensive rats by Folkow. (ll, 14) Importantly, impaired passive distention is another char- acteristic of vessels with thickened walls.(1l, 14, 35, 50) Medial hypertrophy is not the only structural vascu- lar change seen in hypertension. Tbbian and Binion, in 1952, reported that the renal arteries and psoas muscle of hypertensive patients contained significantly elevated amounts of water and sodium compared to normotensive controls. Such tissue "waterlogging", could substantially elevate peripheral resistance by increasing the wall-to- lumen ratio of the resistance vessels. Since structural vascular changes do apparently occur in chronic arterial hypertension, and have profound hemo- dynamic lmplications,a great deal of attention has been given to the etiology of these changes. Specifically, a large number of studies have centered on the role that pressure plays in such changes. These studies are gener- ally directed at examining the vascular effects of arti- fically produced hypotension, or at the comparison of vascular beds of patients, or anflmals, with aortic lcoarctation hypertension. Aortic coarctation provides a unique tool for ll examining the effects of pressure on vascular character- istics in that the vascular beds above and below the coarctation are exposed to two quite different lateral pressures. Arterial pressures above the coarctation are elevated, usually into the hypertensive range, while those below the coarctation are usually normal or near normal. (1, 3, 4, 5, 17, 23, 33, 36, 41, 43, 51, 52) Bevan and his associates measured the wall thickness and internal circumference of arterial strips taken from rabbits with two weeks of coarctation of the abdominal aorta above the renal arteries. They found a signifi- cant positive correlation between the wall thickness of’ arterial strips taken from the hypertensive regions, above the coarctation, and the carotid (hypertensive) arterial blood pressure of the animal. They did not find any correlation between carotid arterial blood pressure and the wall thickness of strips taken from the normo- tensive regions below the coarctation. They did find, however, an occasional correlation between the wall thick- ness of strips from the normotensive regions and the femoral arterial blood pressure of the rabbit. There was no apparent correlation between the internal circumference of any arterial strips and any blood pressure. (3) In addition, Bevan reported that the maximum norepinephrine contractile response of the "hypertensive" arterial strips was positively correlated with the carotid arterial blood pressure, while no such correlation was 12 found for the "normotensive" arterial strips. Again, there was an occasional positive correlation for these strips and the femoral arterial pressure. Furthermore, in a later study, Bevan found this same relationship concerning contractile response, using stimulation of the sympathetic nerves instead of norepinephrine as the stimulus. (5) Of significant importance in these findings is that the hyperresponsiveness of the hypertensive strips was shown to be due to the increased wall mass of the vessel and not due to an increased sensitivity to norepinephrine, since the norepinephrine ED50 was not altered by changes in arterial pressure. In patients with thoracic coarctation hypertension, Samanek and his associates have recently studied the hemo- dynamics of two vascular regions, above and below, the site of coarctation, an average of 11.5 years Eggs; successful surgical correction of the coarctation. (43) Even after this extended period of time, these investigators found that resting blood flows were significantly reduced in the upper extremities (forearm) compared to the vascular regions of the calf muscle, while blood flow in this latter region did not differ significantly from normotensive controls. These differences between upper and lower extremities remained significantly different during maxi- mal reactive hyperemia and during maximal vasodilation with amyl nitrate. Subsequent calculations revealed that the reduced 13 blood flow in the upper extremities during all three manuevers was attributable to an elevated resistance in the upper extremity, which, according to Samanek, was probably due to the structure of the vessels, as evidenced by the increased resistance at maximum vasodilation. In studies on regional hypotension, Sivertsson (45) found that the vascular bed of the hindlimb with 3 to 5 weeks partial occlusion of the supplying artery in the cat exhibited a significantly decreased responsiveness com- pared to normal hindlimbs. This vascular hyporesponsive- ness was revealed as a decrease in the vascular response of the occluded bed to graded sympathetic nerve stimu- lation and norepinephrine infusions. Threshold responses to norepinephrine were the same for both experimental groups. Similar results were later reported by Martin and Conrad in the hindlimb of the dog. (30) Six weeks after ligation of the supplying artery these investigators found that the resting resistance of the limb was significantly lower than that of acutely occluded controls. This lower resistance could not be attributed to alteration of the blood gas tension of carbon dioxide or oxygen. However, histomeric analysis revealed that the vessel walls of the hypotensive vasculature were significantly thinner than those from the control animals. Folkow and his co-workers produced hypotension in the hindlimbs of young spontaneously hypertensive and l4 normotensive control rats by ligating the aorta below the renal arteries. In contrast to finds in the unligated spontaneously hypertensive or normotensive control rats, the hindquarters vasculature, pump-perfused with an artificial medium, of the ligated animals exhibited a decreased resis- tance at maximal dilation, a decrease in the slope of the norepinephrine dose-response curve, and also a decreased maximal response to vasoconstrictors. Since the threshold to norepinephrine was not shown to be changed in the ligated animals, these results, together, were explained on the basis of a thinning of the vessel walls. What is of further interest, is that these same vari- ables, when measured in the unligated normotensive rats, yielded values that fell in between those of spontaneously hypertensive rats and those of the ligated animals. Thus, three separate sets of resistance measurements were ob- tained, each seemingly directly related to the level of pressure in the hindquarters. In another study, Folkow was able to reverse evidence for vascular structural changes in young spontaneously hypertensive rats by treating them for twelve weeks with the hypotensive drugs guanethidine and hydralazine. (15) In these treated animals, the slope of the norepinephrine dose-response curve was decreased as was the maximal vasoconstrictor response, while there was no alteration in the threshold to norepinephrine. 15 Weiss and Hallback recently studied the time course of regional hypotension in the hindlimbs of adult spon- taneously hypertensive rats, pump-perfused with an arti- ficial medium, by ligating the abdominal aorta below the renaliarteries.(53) After only three days of ligation the hindlimb vasculature already began to show signs of regression of the hypertensive vascular thickening. By three weeks the vessels had completely reversed their thickened state by exhibiting a widened lumina and a decrease in the wall mass. However, even after 17 weeks of hypotension these hindlimbs of adult rats would not adapt to the level seen in similarly treated ygung spon- taneously hypertensive rats. (The implication of this finding is discussed in the Discussion Section of this thesis.) Studies on vessel water and salt content in coarcta- tion hypertension have also been conducted. Hollander and his co-workers, in their study on thoracic coarcta- tion hypertension in dogs, reported that the sodium, chloride, and water content of the hypertensive portion of the thoracic aorta was increased compared to the same portion in normotensive dogs, while the water and elec- trolyte content of the normotensive portion of the aorta was virtually identical to a portion from the same area in control dogs. (26) In addition, Hollander reported an increase in the amount of ion-binding mucopolysaccharides in the coarcted dog in the hypertensive portion of the 16 aorta only. He then reasoned that the increased ion and water content of the hypertensive aorta was due to the increased amount of ion-binding mucopolysaccharides whose rate of synthesis was, in turn, positively influenced by an increased intravascular pressure. Villamil later reported similar results to Hollander's both in a preliminary investigation on experimental thoracic coarctation in dogs, (51) and also in a more complete study on the same experimental model with Matloff. (52) Tissue samples from the aorta proximal to the coarctation, taken four weeks after coarctation, showed significant increases in magnesium, calcium, water and intracellular sodium, compared to the same portion of aorta in normo- tensive controls, or to the distal segments of the aorta of the coarcted animal. Carotid artery samples, taken four weeks after coarctation showed significant increases in total and non-inulin sodium as well as increases in calcium. Importantly the ionic composition of the femoral artery which was not exposed to an elevated pressure, was not affected by the coarctation. The studies just mentioned on ionic composition, as well as the hemodynamic and histomeric studies on regional hypotension and coarctation hypertension, all seem to suggest that pressure is involved in, and may even be responsible for bringing about the structural vascular changes seen in hypertension. Because of these results, the investigators believe that the structural vascular 17 changes in hypertension are merely a secondary adaptive alteration to the increase in intravascular pressure. (3, 4, 5, 10, ll, 12, 14, 16, 40, 45, 50, 53) This, of course, by no means precludes the involve- ment of other factors, such as neural and humoral influ- ences, in the pathogenesis and maintenance of the struc- tural changes, their invOlvement being direct, or indirect. Indeed, there is a body of evidence that seems to suggest that an increase in intravascular pressure is not necessary for bringing about certain vascular alterations in hyper- tension. This becomes partially evident when one examines a recent study by Hansen and Bohr. (25) These investigators produced a hypotensive vascular bed by partially occluding the external iliac artery in only one hindlimb of groups of spontaneously hypertensive rats, rats with deoxycorticosterone acetate (DOCA) hyper- tension, and normotensive control rats. They then com- pared the contractility (maximum contractile force) and threshold response (to a variety of agonists) of helical strips from the "low" and "high" pressure femoral arteries. To complete the analysis these strips were also compared with respective strips taken from the normotensive control rats. In general, it was found that both the "high" and "low" pressure strips of either group of hypertensive rats exhibited a significantly increased sensitivity to some type of agonist, when compared to "high" and "low“ pres- sure strips of normotensive rats. Furthermore, it was shown that there was no difference in the sensitivity of "high" and "low" pressure strips in any group of rats. 18 Therefore, lowering the transmural pressure of a previously hypertensive limb could not reverse the changes in sensi- tivity seen in hypertension; and also, protecting a vascu- lar bed from an induced hypertension could not prevent this same altered sensitivity from occuring. It was also found that the contractility of both types of strips was significantly reduced in both the spontane- ously hypertensive and DOCA hypertensive rats compared to the appropriate normotensive controls. What is of further interest, is that, in all three groups of rats, the "low" pressure strips exhibited a significantly reduced con- tractility. Hansen and Bohr, therefore, concluded that functional changes in the sensitivity of a vessel are not dependent on the level of a transmural pressure. Extra- polating this to the development of hypertension in general, these investigators have postulated that struc- tural changes in hypertension are secondary to an increase in intravascular pressure, while changes in sensitivity are not. From a hemodynamic study done in 1963, Nolla-Panandes published a report on experimental coarctation in the rat. (33) In rats with aorta coarcted above the origin of both renal arteries, forelimb arterial pressure was elevated, while hindlimb arterial pressure was normal compared to animals clipped below the renal arteries or to sham oper- ated controls. These latter two groups had normal fore and hindlimb pressures. However, when the hindlimbs of 19 these rats were pump perfused with an artificial medium the following results were obtained. Even though the hindlimbs were not exposed to an elevated intravascular pressure, perfusion pressures generated in the hindlimbs, at a constant flow, in rats ‘with coarctation hypertension were significantly elevated compared to the other control groups. In addition, the mean pressure rise in response to a single dose of nor- epinephrine was significantly greater in the rats with coarctation hypertension compared to the control animals. There were no significant differences between the sham controls and the rats coarcted below the renal arteries. Also, for rats coarcted above the renal arteries only, Nolla-Panandes found a significantly positive regression of hindlimb norepinephrine responses on the initial hindlimb perfusion pressure. According to the author, the findings in the rats coarcted above the renal arteries could not be attributed to differences in the size of the animal, or to effects of coarctation on the development of the hindlimb vascu- lature, as body and limb weights were similar in all the groups of rats. Also, he reported that the vasculatures showed no residual muscular tone. He concluded that the changes seen obviously could not be attributed to an elevated intravascular pressure and instead might in some way be related to the mechanism.that produced the hyper- tension! 20 In a more recent study Pamnani and Overbeck examined the electrolyte and water content of rats coarcted about the aorta in the same manner as the models used by Nolla- Panandes. (36) Blood pressure recordings revealed that the rats coarcted above the renal arteries had elevated carotid arterial pressures compared to the sham operated normotensive controls, while there was no difference between the femoral arterial pressures for these two groups, as these pressures remained normotensive. In contrast to the findings of Hollander in thoracic coarctation in the dog, Pamnani and Overbeck found that the water, sodium and potassium content was significantly elevated in the abdominal aorta and veins, as well as the thoracic aorta of rats coarcted above the renal arteries, compared to controls. In addition, there were no signifi- cant'differences in ionic or water content between the hypertensive and normotensive portions of the aorta, nor was there any cerrelation between the magnitude of the vascular changes in the aorta, or veins, and the level of the carotid arterial pressure. These results, then, suggest that altered levels of vascular ions and water in hypertension, are not necessarily directly caused by an elevated intravascular pressure. Although these last three studies do not totally exclude the possibility that an increased intravascular pressure plays a role in determining vascular wall struc- ture, and function in hypertension, it would now seem 21 that increased intravascular pressure is not the sole determinant of vessel wall structure and function. As an additional determinant, the role of the kidney, especially through its renin—angiotensin system, has also been implicated in the pathogenisis of hypertension. This implication is based on the fact that the renin-angioten- sin system is known to be able to effect changes in blood pressure through the vasoconstrictive nature of angiotensin. II, and through the system's effect on sodium reabsorption. (51, 18, 40) This system is stimulated by a decrease in the perfusion pressure in the afferent arteriole of the kidney. Such a situation could arise from a decreased blood flow to the kidney. It has been seen that coarctation hypertension has been used in examining the effects of pressure on blood vessels, and that, this type of hypertension involves a constriction of the aorta above the renal arteries. Since the constriction could affect blood flow to the kidneys, one could postulate the renal system as an additional factor besides pressure in the development of the hyper- tension, in this experimental model. There has been general evidence since the 1930's that renal factors might be involved in the development of hypertension due to coarctation of the aorta. This was first demonstrated by Goldblatt who found that he could produce hypertension in dogs by coarcting the aorta above the renal arteries but not below. (21) 22 Similar results have since been obtained by Habib et al in dogs (23) and, as previously stated, by Nolla- Panandes in rats. (33) Both of these investigators have provided additional evidence in support of a renal factor in coarctation hypertension. Habib has found a generalized increase in blood pressure in coarctation of the thoracic aorta, not just an increase in arterial pressure above the coarctation. Also, he has found that transplanting a kidney to a vas- cular region above the coarctation will substantially (but not totally) reduce arterial pressure above and below the level of coarctation. Besides showing that coarcting the abdominal aorta would only produce hypertension if instituted above the renal arteries, Nolla-Panandes analyzed the time course of the blood pressure changes as a further indication of renal involvement. He observed, that immediately after coarcting the abdominal aorta above both renal arteries in rats, the fore and hindlimb blood pressures drop. After this point in time, both pressures rise within a day in a parallel manner until the forelimb pressure is hypertensive and the hindlimb pressure is normotensive again at the end of two weeks. Nolla- Panandes stated that this type of time course in the development of the hypertension could not be totally explained by a gradual change in the severity of the coarctation itself (i.e. such as a gradual sclerotic 23 narrowing of the coarctation). If this alone was the only development in the days following the coarctation, then as the aorta narrowed, pressure below the constric- tion would decrease as pressure above the constriction increased. More direct evidence for renal involvement in coarctation hypertension has been provided by Ribeiro and Krakoff. (41) In a preliminary study these investi- gators found an elevated plasma renin activity in human coarctation hypertension but not in essential hyperten- sion. Neither group of patients had any type of reno- vascular disease. Additionally, it was found that 30 minute infusions of saralasin, a specific angiotensin antagonist, would substantially reduce the blood pressure in the patients with coarctation of the aorta, while patients with essential hypertension were not affected by the same infusion. Bagby and her associates have found a significant decrease in renal plasma flow and the glomerular filtra- tion rate in salt-restricted dogs with thoracic aortic coarctation hypertension, compared to salt-restricted control dogs. The coarcted dogs also showed a signifi- cantly increased plasma renin activity and reactivity. (l) Timmis and Gordon have also implicated the renal pressor system in the pathogenesis of coarctation hyperten- sion and have suggested that this system is stimulated by 24 the coarctation's effect of dampening the pulsatile nature of the renal blood flow. (48) Furthermore, the renin-angiotensin system may be involved in the development of the structural cardiovas- cular changes that have been postulated to be largely responsible for the maintenance of the elevated arterial pressure in hypertension. Sen and her co-workers have reported an increase in ventricular weight and plasma renin activity in young spontaneously "hypertensive" rats before the animal actually becomes hypertensive. (44) The ventricular weight and plasma renin activity were positively correlated in these rats. In addition, anti- hypertensive drug therapy using methyldopa or hydralizine ‘was able to reduce blood pressure in the hypertensive rats. However, ventricular weight was reduced by methyl- dopa only. Interestingly, methyldopa reduced plasma renin activity in the rats while hydralizine increased the plasma renin activity. Although these findings suggest a possible hyper- trophic effect of the renin-angiotensin system.on cardiac ‘muscle tissue, it is not known whether or not this system can affect vascular smooth muscle in a similar manner. In this regard, Villamil (51) has shown that chronic exposure of arteries to angiotensin.II significantly increases their total sodium, magnesium, and calcium.content. Exposure to angiotensin also increased the artery's permeability to sodium and decreased the inulin space of the artery. 25 Importantly, the increased permeability seems to be specifically related to the angiotensin. In contrast to the findings that the renin-angiotensin system is involved in hypertension, Markiewicz et al reported decreased plasma renin activities in humans with coarctation hypertension and, in addition, could not demon- strate a correlation between the plasma renin activity and the level of arterial blood pressure. (29) However, there was no renal underperfusion present in these patients at the time of the study. Also, in Bevan's study on abdominal coarctation in rabbits, the investigators stated that their experimental model had, not to their knowledge been associated with an elevated plasma renin activity (3), and this study did not reveal any arterial alterations below the site of the coarctation. A humoral factor such as the renin-angiotensin system, would largely be expected to effect the entire vascular system of the animal. Whether or not such a; system was brought into play during a certain type of hypertension might explain why there is conflicting data in animal models primarily designed to study the effects of different intravascular pressures (i.e. coarctation hypertension). Aside from-humoral factors, neurogenic factors have been another area of investigation in the pathogenesis of hypertension. Bevan et al have reported significant 26 increases in the norepinephrine content, release, and neuronal uptake of the terminal adrenergic nerve plexus in the "hypertensive" ear artery of rabbits with coarc- tation of the abdominal aorta. They have also reported an increase in the vessels' permeability to norepinephrine. (4, 5) These changes were not seen in the "normotensive" vasculature or the veins. However, extraneuronal uptake of the norepinephrine as well as the activity of nor- epinephrine degradating enzymes 'were also shown to be increased in the ear artery. Coupled with the fact that, compared to controls, the media of these arteries were thickened, these last two factors concerned with norepinephrine disposition would tend to decrease the effects of sympathetic nerve activity, thus, according to Bevan, negating the effects of the increased activity in the terminal adrenergic nerve plexus. Lais and Brody have reported a selective increase in the sensitivity to norepinephrine in the pump perfused hindlimbs of spontaneously hypertensive rats. (28) They correspondingly concluded that altered sensitivity and vascular wall thickening were involved in the vascular hyperresponsiveness seen in these rats. It should be noted that, in contrast to Folkow's perfusion studies, this investigation used blood as the perfusate; or to use the vernacular of the investigators, the experiments were performed in a vascular bed exposed to a "full 27 humoral complement". Therefore, Lais and Brody stated that the altered sensitivity may in some way be related to cir- culating hormones. In this respect, Goldberg has shown that angiotensin II, can release norepinephrine from the nerve terminals of veins. (20). Turning to other neurogenic influences, although the baroreceptors appear to be "reset" in hypertension, (17, 27) it has been believed that actual sympathetic discharge is nevertheless normal. (17, 40, 45) Recent evidence by Lais and Brody would appear to support this belief. (27) In summary, then, the increased peripheral resistance seen in chronic hypertension appears to result, at least in part, from the structural characteristics of the resis- tance vessels, although there is evidence that the smooth muscle cells may be hypersensitive to certain vasocon- strictors. The studies suggest that the smooth muscle lining these vessels is hypertrOphied and that the vessel walls contain elevated amounts of salt and water and fibrous tissue. These factors tend to encroach on the lumens of the vessels thus elevating their resistance to flow, even at maximal vasodilation. A large body of evidence has suggested that these structural vascular changes are a secondary adaptive development to an increased intravascular pressure. How- ever, other studies, while not excluding pressure as a 28 possible influence on these changes, seem to suggest that other factors may be involved. A likely candidate as an additional factor in the development of the structural changesuey'be the renin- angiotensin,system. Hypothesis One of the characteristics of chronic arterial hypertension is an increase in total peripheral vascu- lar resistance. This increased resistance is, to a large extent, based on an underlying thickening of the walls of the resistance vessels which encroach upon their lumens, even at maximal vasodilation, thus raising the baseline from.which all resistance changes take place. It has been believed that this vascular wall thickening is a secondary adaptation of the resistance vessels to a primary increase in blood pressure. To test this hypothesis, resistance 'measurements were made in the normotensive vascular beds of rats with coarctation hypertension. Any evidence for increased resistance and/or vascular wall thickening in these normotensive vascular beds of a hypertensive animal could not be a secondary adaptation to high intravascular pressure, but instead, then, may be related to the etiology of the hypertension itself. 29 Methodology Pre-experimental Male, outbred, Sprague-Dawley rats, weighing between 150 and 200 grams were used in the study. Under sterile conditions and ether anesthesia, a midline/abdominal longi- tudinal incision was made in each animal. The rats were then randomly divided into four groups. In one group, hereafter to be called Group A, a small, circular silver clip, approximately 1 to 2 mm wide and 0.813 mm in interior diameter, was placed around the abdominal aorta above the origin of both renal arteries to produce coarctation hypertension. In a second group, hereafter to be called Group B, a silver clip, with an internal diameter of 1.48 umior 1.70 mm (See Appendix A) was placed around the abdominal aorta above the origin of both renal arteries. This group of rats served as sham operated normotensive controls, for the clip was too large to cause significant constriction of the abdominal aorta. A third group of rats, hereafter called Group C, were designed as controls for the hindlimb atrophy that might occur in Group A. In this group a constricting silver clip, with an internal diameter of 0.610 mm was placed around the abdominal aorta below the origin of both renal arteries. This type of aortic coarctation reduces intravascular 30 31 pressure and flow in the vascular regions below the clip, but does not produce hypertension in the vascular regions above the clip. Finally, two-kidney Goldblatt hypertension was created in a fourth group of rats, hereafter called Group D, by constricting the renal artery of the left kidney with a silver clip with an interior diameter of 0.406, 0.419, or. 0.434 am (See Appmdix A) and leaving the other kidney intact. The purpose of this final group of rats was to ensure that elevated hind- limb resistance could be detected in the perfusion experi- ments described below. After c1ipping,the incision was closed. The animal's weight in grams was recorded, after which it was placed in a plastic cage, three or four rats per cage, and allowed to recover from the effects of the anesthesia. The rats were then kept in a temperature controlled room.(approximately 24° C) in the Laboratory Animal Care Service at Michigan State University for a post-operative/ pre-experimental period of not less than four, and no greater than five weeks. During this period the animals were maintained on a diet of standard rat chow and tap water. No drugs were administered at any time during this four week period and with few exceptions, all the animals remained healthy. The mortality rate was less than 10% for all the groups of rats. Blood pressure and body weights were recorded during the second, third, and fourth weeks after clipping. Blood 32 pressures were taken in rats lightly anesthetized with ether by tail plethysmography, as described by Phelan. (37) TWO readings of caudal arterial systolic pressure were measured and recorded. These pressures were used to represent arterial pressures in the regions below the site of coarctation. Experimental Methodology The experiments in this study were designed to measure hemodynamics in the isolated innervated, blood pump perfused hindlimb of the four groups of rats just described. Experi- ments on the rats were always performed at least four weeks, and always less than five weeks, after a set of rats were clipped. For an experiment on a particular day, three rats clip- ped on the same day of the same group were selected so that their body weights would be as close to 350 grams as pos- sible. The majority of rats attained this weight sometime during the fourth week after they were clipped. Once selected, the rats were moved to the room where all the experiments were performed during the study. Here the rats were weighed to the nearest gram and the weights were recorded. Of the three rats selected, one rat was to be per- fused, another was to be used as a blood donor for the first, and the third rat was used as a backup for either the perfused rat or the blood donor. On days when two 33 experiments were to be performed, the same blood donor was used for both rats. To prepare a rat for an experiment it was first placed in an anesthesia chamber containing a few drops of ether. Once fully anesthetized (no whisker twitch) the rat was removed from the chamber and layed supine on a small surgical table where its limbs were extended and taped down. Anes- thesia was continued using a nose cone with an ether soaked cotton plug. A small midline neck incision was then made separating the lobes of the thyroid gland, and the sternothyroid muscle thus exposing the animalfistrachea which was then isolated. Next a 2 cm inguinal incision was made in the animal's right groin to expose the right femoral artery and vein. This vein was then cannulated in a proximal direction and secured using a small L-shaped glass cannula connected to a piece of rubber uterine tubing. The ether nose cone was then removed, and 70 mg/Kg in 0.7 cc of normal saline of alpha-chlorolose was given to the animal I.V. through the femoral vein cannula as in initial anesthesia. (Note: the initial dose of anes- thesia was not sufficient for the purposes of the experi- ments if less than 50 mg/Kg was given to the rat. Insuf- ficient anesthesia was considered due cause to invalidate data gathered on such an animal. See Appendix A) With the rat anesthetized, a small incision was then 34 made in the trachea allowing the insertion of a short ( of I.D.; 0.095 inches 0.D.). This "tracheal tube" was I! 1.5 cm) piece of polyethylene tubing (0.066 inches then secured with the sutures already present in the area and allowed one to aspirate any tracheal mucous and other fluids out of the lungs. The right carotid artery was then isolated and packed with gelfoam (Thrombin in plastic foam; Upjohn Co.) and 00 black silk suture was carefully slipped underneath it, to aid in its retraction. Procedures were then directed to isolating the rats left hindlimb which was to be used in the perfusion experi- ments for this study. Briefly, this hindlimb was isolated from the body by severing (with cautery) and ligating skin and muscle connections and by dislocating the hip from the pelvis with a heavy ligature. The femoral artery, vein, and nerve were left undisturbed as was the sciatic nerve and its medial branch. Thirty minutes after the last surgical procedure, heparin was administeredto the animal (400 USP units, WOlins Industries). During this thirty minute interval,allowed for clot- ting, the skin incision was closed, with the exception of approximately 1 1/2 cm directly over the femoral vein, artery and nerve. This area was gently packed with normal saline soaked cotton to keep structures moist,and a small amount of Gelfoam for hemostasis. The limb was gently extended and secured by connecting a hemostat to one toe, 35 then securing the hemostat to the table with tape. Also during this thirty minute interval the donor rat was then anesthetized with ether,laparotomized,and aortic blood was drawn into a 12 cc heparinized syringe. This blood was quickly transferred to a heparinized beaker and then swirled by hand to insure complete mixing with the heparin (and thus avoid the formation of micro- clots). The beaker was then covered with Parafilm to prevent evaporation and reduce the chance of contamin- ation. An autopsy was then performed on the donor rat. If anything proved wrong with this rat such as a lung infection or renal infarction,its blood was discarded and a new donor was used. The total amount of time used to prepare the experi- mental animal for perfusion including the 30 minutes waiting period before administering the heparin was moni- tored for each rat. If this time exceeded 2 hours and 30 minutes in duration, the animal would not remain suffi- ciently anesthetized during the full length of the per- fusion experiment and thus any data from animals of this type were discarded. By staying under the 2 hour, 30 minute level, it was assured that all animals studied would be equally sedated during the entire length of the experiment. (See Appendix A) At this time the pump used for this study was filled with normal saline and allowed to run at maximum.flow. This pump was a modification of a pump previously described (9); this pump provides a nonpulsatile flow, is pressure 36 independant to at least 350 mm Hg, and produces negligible hemolysis. For this study, side branches of the pump's inflow and outflow tubing could be connected to Statham P23Gb pressure transducers. These pressures could then be measured on a Hewlett Packard 77023 dual channel recorder- amplifier and a H.P. Moseley 7100B strip chart recorder. The pump was enclosed in a plexiglas cover, used to keep the pumping mechanism at a temperature of 37° C. The entire pump was on a platform that could be rotated 360 degrees. Before the perfusion of each animal, heparinized cannulae were inserted in a proximal direction into the right femoral and carotid arteries and secured with sutures and by means of adjustable clamps on the side of the surgical table. This table was then raised to the level of the Statham strain gauges and by means of a Silastic, normal saline filled, tube connected to one of the strain gauges, mean carotid and femoral arterial pressures could be directly recorded alternately. These pressures were used to repre- sent arterial pressures in the regions above and below the clip, respectively. At least two mean carotid and mean femoral arterial pressures were recorded for each experi- ‘ment. If each pair of carotid or femoral pressures did not agree within 5 mm Hg of each other, another carotid and femoral pressure was obtained until two successive pairs of pressures agreed in the prescribed manner. The cannulae were then clamped and left in place. With this segment of the experiment completed, the 37 Silastic tubing was disconnected and the pump inflow and outflow tubing was connected to the strain gauges via their side tubing. All the pump's tubing was flushed with normal saline and the entire pump was run at maximum flow while being rotated to 900 and 1800 to remove any and all trapped air. At this time, supplemental anesthesia was given to the rat as an IV injection of pentobarbital (J.T. Baker Indus- tries) 16 mg/Kg body weight and alpha-chloralose (25 mg/ Kg) (See Appendix A). With the supplemental anesthesia given to the rat, a heparinized cannula was inserted into the left femoral artery (the lone artery to the isolated limb) in a downstream (di- stal) direction and secured in a nonobstructive position with sutures and a side clamp connected to the surgical table. Before the limb could be perfused the pump was primed with the heparinized donor blood at 1cc per minute. (cc used interchangably with ml throughout paper; flow was actually measured in ml's) When the pump was completely filled.with blood the pump inflow tubing was connected to the carotid artery cannula, being sure that no air got into the system during the connection of the pump tubing to the cannula. With this connection made, the clamp on the carotid artery cmnula was removed and the pump was started at 1 cc per minute. Tube and cannula resistance produced an immed- iate drop in the inflow pressure of about 60 mm Hg. If a much greater drop in pressure occurred at this time, it was taken to mean that the carotid artery was in some manner 38 obstructed. The pump was stopped, and the cannula reposi- tioned. If no occlusion. was indicated, the pump outflow tubing was connected to the left femoral artery via the oarmula and the isolated limb was then perfused at 1 cc per minute for 15 minutes to establish a steady state. Blood leaving the isolated left hindlimb was allowed to freely return to the animals body via the left femoral vein. Only the right carotid artery was used to supply blood to the pump during the experiment. The right femoral arterial and venous cannulae were not included in the pump circuit nor were they used for any other measurements during the perfusion of the animals left hindlimb. Pump outflow (perfusion) pressure and pump inflow pressure were both recorded continuously on the strip chart recorder. Pump flow was adjusted by a digital pump rate control unit connected to the pump motor. After the 15 mdnute steady state period was completed, pump fIOW'WaS then adjusted so that steady state perfusion pressure (minus the pressure drop across the tubing and cannula(see below) ) was similar to the rat‘s femoral arterial pressure. The pressure and pump flow was recorded and designated "resting limb pressure" and "resting limb blood flow." "Resting limb resistance" was calculated using these values. Once resting flow and resistance were obtained, the effects of passive distension of the hindlimb arteries was 39 studied. To accomplish this, pump flow was first adjusted to 0.125 cc/minute and allowed to remain there for at least 5, but no longer than 9 minutes. In this manner enough time was allotted for achievement of a steady state at this flow. Flow was then adjusted, in order, to 0.25, 0.5, 1.0, 1.5, and 2.0 cc/minute. As in the case of the flow at 0.125 cc/mdnute, a steady state was first obtained atzeach flow before moving on to the next flow, providing steadyifiate was attained again before the ninth minute. (See Appendix A) After this set of flows was complete, the pump was stopped momentarily in order to monitor the carotid blood pressure of the animal as a rather rough indicator of the general condition of the animal. Also, the drop in limb perfusion pressure, once the pump was stopped, was watched to see if it dropped below 5-7 mm Hg. If this was the case, it was considered evidence for complete isolation of the hindlimb. Then the same set of flows were repeated and generated pressures again recorded. After completion of the two sets of flows, flow was returned to 1.0 cc/minute. Once a steady state was achieved, (or in 9 minutes) first the femoral then the sciatic nerves were cut. A drop in perfusion pressure after cutting the femoral and sciatic nerves was consid- ered evidence that they were viable. Ten minutes after severing the femoral nerve, the steady state perfusion pressure was recorded. The resistance calculated was thus termed resistance after acute denervation. 40 With the limb denervated, and the pump still at 1 cc/ minute, a supramaximal dose of the vasodilator sodium nitroprusside (0.6 mg/Kg in 1 cc normal saline/Kg) was injected rapidly into pump tubing upstream to the pump. This agent is believed to act directly on the vascular smooth muscle to produce relaxation of the smooth muscle. (22) Maximal vasodilation of the hindlimb vasculature was tested by successively doubling the dose of the nitro- prusside until no further decrease in hindlimb perfusion pressure could be detected. Perfusion pressure 4 minutes after the final injection was recorded; calculated resis- tance was designated "resistance after maximal sodium nitro- prusside vasodilation." With the pump still running at l cc/minute the cannula in the perfused limb was out free from the femoral artery ending the experiment. Two heparinized capillary tubes were quickly filled with the free-flowing blood. The rest of the animal's blood was pumped into a beaker. At this time, in all rats, the pressure gradient across the out- flow tubing and cannula was measured at each flow rate used during the experiment (Figure 1). This value was subtracted from the respective perfusion pressure recorded “M value was used to calculate all limb resistances. Also, for these calculations femoral venous pressure in the isolated hindlimb was taken to be constant during each portion of the experiment. The capillary tubes were used to determine the blood hematocrit of the rat while the beaker of blood 41 was used for determining plasma creatinine, sodium” potas- sium, calcium, and magnesium. The rat was autopsied, and the animal's clip type was verified. The animal's kidneys and heart were excised and weighed. Both hindlimbs were removed and weighed. These limbs were then dried in an oven at about 86° C and weighed not less than one week later. Resistance was expressed as net perfusion pressure/ cc minute gram-1 perfused limb wet and dry weight and also in terms of the opposite limb wet and dry weight (for discussion, see Appendix B). Sixteen additional rats of groups A and B were pre- pared after the end of the study. Samples were taken from these rats for plasma renin concentrations 4 1/2 weeks after clipping, by decapitating the animal and collecting the first two seconds of blood flow from the severed aorta. Blood was prepared by technicians from Dr. M. Bailie's laboratory at Michigan State University. The animals were then autopsied and the hearts were weighed. Body weights were also recorded at the 4 1/2 week mark. The renin activity was determined by radioimmunoassay procedures and expressed as ng angiotensin I/ml/hour. Plasma creatinines were analyzed with an autoanalyzer and expressed as mg percent while plasma Na+ and.K+ levels (m Eq/L) were analyzed on the Beckman Flame Photometer. Calcium and magnesium levels(m.Eq/L) were analyzed on the Perkin-Elmer Atomic Absorbtion Spectrometer. Calculated 42 resistances were analyzed with the one-way analysis of variance and the means were compared with the Student- NewmaneKuels test. Other experimental variables were analyzed among groups in the same manner, while comparisons within groups were done with the Student paired t-test. In addition to the analysis of variance, the pressure-flow and resulting resistance-flow curves were analyzed with profile analysis. (31) As a final note to the reader, it must be mentioned that the experimental protocol used in this study was as much a result of solutions to problems that occured during the study as it was a result of planned protocol. These problems in many cases dictated the "hows" and "whys" of our experimental protocol. These problems were fairly numerous and, therefore, it was felt that to mention them at each point in the methodology would only serve to make the methodology disjointed and thus hard for the reader to follow. Therefore, two appendices have been included in this paper in an effort to put in a concise form how our experimental protocol came to be. It is hoped that this will give the reader a better understanding of why certain methods had to be used in this study, as well as serve as an aid to problem solving for future students in this field. 43 Figure 1. Pressure drop across pump outflow tubing and cannula mm.Hg = pressure in millimeters of mercury cc/minuteaflow in cubic centimeters per minute 44 ohm H ouawfim meazgxoo. o._ o._ no . 3m med aims": mh:z_s_\oo 30...“. mammm> <432<0 Qz< 0253... >>O.._u_._.30 nzzzn. mmomo< aomo mmzmmmma 0. ON on on on om Oh om 0m 00. o: 5H mm 45 Results 1. General Parameters and Blood Pressures Body weights, blood pressures, hematocrits, plasma creatinines and electrolytes, etc., are listed in Table 1. As can be seen the groups of animals did not differ sig- nificantly in their body weight, nor was there any differ- ence in their plasma electrolytes. Similar hematocrits that were also within normal ranges suggest that no significant bleeding occurred during the experimental procedure in any group of rats. Although kidney mass was slightly reduced in Group A rats compared to Group B (significant for the left kidney only, p < 0.05), plasma creatinines were similar in all four groups of rats. Heart weight expressed in terms of body weight was significantly increased (p < 0.01) in hypertensive rats of Groups A and D compared to sham controls(by 44 and 21% respectively). There was also a slight (:f‘lZ) increase in heart weight/body weight in Group C rats compared to sham controls (p;< 0.05). One of the most important features of the above data is the fact that tail systolic pressures remained normotensive and similar in Groups A and B throughout the four week pre-experiment period. Note also that there were no significant week-to-week variations of the pressures within groups. This similarity Was further verified by the 45 46 direct measurements of blood pressure taken just prior to the start of the experiments. These direct measurements show that the mean femoral arterial pressures of these two groups did not differ significantly and remained normoten- sive,although mean carotid blood pressure was significantly higher in Group A rats compared to Group B (p < 0.01). Of additional interest is the fact that, the mean femoral blood pressure of Group C was significantly hypotensive (p < 0.01), although the mean carotid blood pressures for Groups B and C were normotensive and virtually identical. Also,as expected, Group D rats were hypertensive both in the carotid and femoral arteries (p < 0.01 for both vs. the respective sham control), as well as during the 4 weeks prior to the experiments. For clarity purposes the following figures list only the statistical comparisons that are most important for the particular item being depicted. A statistical summary of all intergroup comparisons for the data in table 1 (and also for all subsequent data) is given in Appendix C. II. Limb Weight Opposite limb wet weight, expressed in terms of body weight, was slightly but significantly reduced in rats of Groups A and C (6.0 and 5.3 percent respectively, p < 0.01), indicating that slight limb atrophy had occured in these two groups of rats (Figure 2). Consequently, all flow and 47 TABLE 1 GENERAL PARAMETERS All values represent means :_standard errors of the means. Group A - rats with coarctation hypertension; Groupl3- normotensive sham operated control rats; Group C - rats with coarctation of the abdominal aorta below the origin of both renal arteries; Group D - rats with two-kidney Goldblatt hypertension. wt - weight; mg% 8 milligrams percent; mEq/L = milliequivelents per liter; (* significant differ- ence, p < 0.05; ** significant difference p < 0.01;*** significant difference p < 0.001 all compared to Group B). ( ) = n, number of experimental units // p < 0.01 within group / p < 0.05 within group 1113 TABLE 1 Group A Group 8 Group C Group 0 Iody weight grace 352.8 1 4.86 (16) 353.1 1 2.93 (21) 358.9 1 3.72 (15) 358.7 1 10.13 3!:521£§.°::‘3§ ""“°‘ 18‘ “DOB 109.2 1 4.74 (7) 110.9 1 1.56 (10) 49.8 1 5.93 (3) 1.73 1 3.12 (4) 3'9 U903 108.5 1 4.57 (10) 109.6 1 2.13 (13) 56.0 1 4.90 (6) 182.2 1 7.60 (6) 4th week 112.8 1 2.68 (12) 112.7 1 2.18 (13) 56.7 1 4.82 (16) 173.3 1 11.74 (6) 3 week everege 110.5 1 2.19 (29) 111.1 1 1.17 (36) 25.1 1 2.99 (15) 176.6 1 5.12 (16) Carotid Arterial Pressure 152.8 + 5.96 (13) 110.3 + 2.46 (20) 110.2 1 3.84 (13) 149.1 1 8.88 (9) u. “I so ' ' so Fe-ors1 Artertel Pressure 104.9 + 2.49 (20) R; ' leer: Weight/Body Weight 112.3 1 3.32 (13) 33.0 1 3.01 (131 1:6.7 1 9.2 (9) 0.0033 1 0.00006 (21) 010034 1 0.00007 (15) 2‘0040 1 0.00014 (9) 0 0.0048 1 0.00013 (16) to Left Kidney Ut./lody 9:. 2.00326 1 0.00011 (16) 0.00374 1 0.00010 (21) 0.00370 + 0.00010 (15) e— 0.00273 1 0.00028 (19) t. It. Kidney Ut.llody Wt. 0.00340 1 0.00014 (16) 0.00370 1 0.00012 (21) 0.00370 1 0.00012 (13) 0.00439 1 0.00027 (9) so Creatinine set Sedu- seq/1. 1.11 1 0.11 (12) 130.3 1 1.05 (11) 0.91 1 0.12 130.9 1 2.05 (11) 1.24 1 0.17 (s) 131.0 1 2.32 (a) 1 23 1 0.20 (s) 134.2 1 2.33 (6) Poussin! seq/1. 4.26 1 0.20 (11) 4.46 1 0.15 (11) 4.39 1 0.22 (7) 4.19 1 0.23 (4) Gelctul seq/L 3.28 1 0.15 (11) 3.28 1 0.09 (11) 3.37 1 0.12 (7) 3.25 1 0.09 (4) le'nesiu- leg/L 2.55 1 0.11 (10) 2.72 1 0.14 (11) 2.45 1 0.8 (7) 2.42 1 0.06 (4) lelutocrtts 44.2 1 0.77 (13) 44.4 1 0.70 (14) 43.5 1 0.81 (12) 44.4 1 0.87 (7) Flee-e Renin Concentrs- 21.86 + 2 12.10 + 0.75 clone nz/ml/hr so ‘ - Angiotensin 1 leer: Wt./8ody Wt. 0.0047 + 0.00015 0.0033 + 0.00006 for recs used for *ffi - - P1esne Renin Concen- tretioos ::r2ueed limb H20 content 14.10 1 0.18 (14) 14.22 1 0.19 (20) 13.90 1 0.20 (15) 14.83 1 0.42 (9) ::pos1te 11.6 H20 content 13 10 1 0.21 (14) 13.89 1 0 15 (20) 13.46 1 0.22 (15) 14.10 1 0.39 (9) i I (I I 49 resistance data were expressed in terms of limb weight as detailed in Appendix B. III. Resting Resistance and Flow Resting resistance and resting flow, both expressed in terms of opposite limb wet weight, are depicted in Figures 3 and 4 respectively. .As expected resting resis- tance in the hindlimb vascular bed was elevated in rats with two-kidney Goldblatt hypertension (67%; p < 0.01). Hindlimb resistance was also elevated in Group A rats that were coarcted above the renal arteries; i.e., rats that had hypertensive forequarters and normotensive hindquarters. This elevation in resistance in Group A rats is statisti- cally significant when compared with either Group B or C (p < 0.01; 58% and 119% elevation, respectively). Resting hindlimb resistance, expressed in terms of opposite limb wet weight was reduced by almost 29% in rats of Group Clcompared to sham controls, (p < 0.01) using the two-tailed Students t-test. Resting flows per gram opposite limb wet weight were reduced by 29% in Group A rats compared to sham controls using the Student-Newman-Kuels test (p < 0.05). Resting flows were significantly reduced in Groups C and D compared to sham controls (p < 0.05, 17 and 22 percent, respectively) when the differences between the means were tested with the Student's two-tailed t-test. (Note: comparisons of flow Figure 2. SO Opposite Limb Wet Weight/Body Weight Coarc Ab = Group A = rats with abdominal aortic coarctationabove the origfi1 of both renal arteries Sham = Group B = normotensive sham operated control rats Coarc Bl = Group C = rats with abdominal aortic coarctation below the origin of both renal arteries Renal = Group D = rats with two kidney Goldblatt hypertension gm = gram; p = probability that the differences between means of the groups occured by chance n = number of experimental units .IO .09 .02 .Ol 51 Opposite Limb Wet Weight I Body Weight * Group A vs. Group B J Group C vs. Group B n=14 n=20 ":15 [1:19 * p<0.0| r=- Jp<0.0l .1. F:- —"'1 CoorcAb Sham Coochl Renal Group A Group B Group C GroupD Figure 2 Figure 3. 52 Resting Resistance x Limb Wet Weight Coarc Ab = Group A = rats with abdominal aortic coarctation above the origin of both renal arteries Sham = Group B = normotensive sham operated control rats Coarc Bl = Group C = rats with abdominal aortic coarctation below the origin of both renal arteries Renal - Group D = rats with two kidney Goldblatt hypertension p = probability that the differences between means of the groups occured by chance mm Hg = pressures in millimeters of mercury ml min’1 gm'1 = milliliters per minute per gram n = number of experimental units mm Hg/tml min.l gm-') 3500 .5 5 g .5 53 Resting Resistance 11 Limb Wet Weight x/GroupA vs. Group C it Group A vs. Group B n=10 - Jp Samoan. 8N Figure 6. 60 Average Resistance versus Flow Coarc Ab = Group A = rats with abdominal aortic coarctation above the origin of both renal arteries Sham = Group B = normotensive sham operated control rats Coarc B1 = Group C = rats with abdominal aortic coarctation below the origin of both renal arteries p = probability that the differences between means of group occured by chance Goldblatt hyper = rats with two kidney Goldblatt hypertension 1 mm Hg/ml min' = resistance n = number of experimental units cc/minute = cubic centimeters per minute 61 0 «Human 23...... \oo «1.28... 293863261 _ m 96.0 .2. 4 96.0 t 0 9.20 .9 4 960\ ou:..00000.063..0l.l 20......206380 ll $2.34 9.80 .4 9.9.0 2... 32.... .m> 00:20.00". l.ugul |u1/6H 111w Figure 7. 62 Average Resistance versus Average Pressure Coarc Ab= Group A = rats with abdominal aortic coarctation above the origin of both renal arteries Sham = Group B = normotensive sham operated control rats Coarc Bl = Group C - rats with abdominal aortic coarctation below the origin of both renal arteries Renal = Group D = rats with two kidney Goldblatt hypertension ‘mm Hg/ml min 1 gm'1 - resistance expressed as pressure in milliliters of mercury divided by flow expressed as milliliters per minute per gram apposite limb wet weight n = number of experimental units 63 n onsmwm 01...... 08 On. 00. o d 3 — a _ n _ .nxum 00.....an 2630.00 5 96.0 .ll _ «05.06.30 696.0 li _ Eu: .520 .m 96.0 .ll _ $2.54 6.80 .4960 1.... _ HI. .ooo. _ _ u .89 .I .608 .003 .000» .. «Swami .9. 85.6.4.3. Loomm (,-u16 I.ugu1 gun/6H 111w 64 reduce resistance to a level attained in Group B at a level of only 125 mm Hg. V. Acute Devervation After acute denervation, resistance at a flow of 1.0 cc/minute (not normalized for limb weight) remained elevated by 17% in Group A compared to Group B rats. With the opposite limb wet weight normalization Group A rats exhibited a significantly higher resistance compared to Group C only (52%, p < 0.01). Groups C and D also exhibited significant differences after acute denervation compared to sham controls (p < 0.01) being reduced 28% and increased 58% respectively. These results are shown graphically in Figure 8. VI. Resistance after Infusion of Sodium Nitroprusside Residual resistance in the denervated hindlimb of the four groups of rats after injection of supramaximal doses of the vasodilator sodium nitroprusside are presented in Figure 9. All groups differ significantly from all other groups (p < 0.01; p < 0.05 Group A versus Group B). Of interest is the fact that resistance remains elevated in rats of Group A compared to the sham-control group, presumably after all smooth muscle activity had been abol- ished. Figure 8. 65 Resistance after Acute Devervation x Limb Wet Weight; Flow = l cc/minute Coarc Ab = Group A = rats with abdominal aortic coarctation above the origin of both renal arteries Sham - Group B = normotensive sham operated control rats Coarc B1 = Group C - rats with abdominal aortic coarctation below the origin of both renal arteries Renal = Group D = rats with two kidney Goldblatt hypertension mm Hg/ml min'1 gm,”1 = resistance expressed as pressure divided by flow, expressed as milli- liters per minute per gram opposite limb wet weight n - number of experimental units mm Hg 11ml min-' gm") 66 Resistance Acute Denervation x Limb Wet Weight 4000 - \/ Group A vs. Group C 3500 - 3000 - 2500 - 2000 r n=12 ~/p<0.01 I500 - FL n=10 ._:1'_1 |000 - p; 500 - Coarc Ab Sham Coarc Bl Group A Group B Group C Figure 8 Renal Group D Figure 9. 67 Resistance Times Opposite Limb Wet Weight at 1.0 cc/minute after Supramaximal Injection of Sodium Nitroprusside Coarc Ab = Group A = rats with abdominal aortic coarctation above the origin of both renal arteries Sham.= Group B = normotensive sham operated control rats Coarc Bl = Group C = rats with abdominal aortic coarctation below the origin of both renal arteries Renal a Group D = rats with two kidney Goldblatt hypertension urn Hg/ml min'1 gm"1 =- resistance expressed as pressure divided by flow, expressed as milli- liters per minute per gram opposite limb wet weight NaNP = Sodium nitroprusside n = number of experimental units mm Hg/(ml min"I gm") IOOO ii 21' § § § ‘2? 68 Na NP Resistance x Limb Wet Weight \/ Group A vs Group C * Group A vs. Group 8 n=6 J'— n=12 x/p<0.0l *p<0.05 _:-T 11:10 Coarc Ab Sham Coarc Bl Renal Group A Group B Group C Group D Figure 9 69 VII. Plasma Renin Concentration The results of the measurements of plasma renin concentration in rats of Group A and B are presented in Figure 10. Plasma renin concentration was found to be increased by 80% in Group A rats relative to Group B (p < 0.001) when compared with the Student's two-tailed t-test. These data are listed in Appendix C. Figure 10. 70 Plasma Renin Concentration ng/ml/hour angiotensin I = nanograms per millili- ter per hour of angiotensin I Coarc Ab = rats coarcted above the renal arteries around the abdominal aorta (coarctation hyperten- sive) Sham.= normotensive sham.cperated control rats n = number of experimental units 35- 30- 25- ES 8 ng/ml/ hour angiotensin 1 IO- 71 Plasma Renin Concentration 0:16 Coa rc Ab Sham Figure 10 Discussion To test the hypothesis that the vascular wall thick- ening in hypertension results from increased intravascular pressure, hindlimb pressure-fhwvrelationships were measured in rats with experimental aortic coarctation hypertension. As is the case in most animals with this type of hypertension, the hindlimb vascular bed of the rats with coarctation hypertension in this study was 'not exposed to increased intravascular pressures. On the day of the perfusion experiments, and in the prior four weeks, arterial pressure in the regions distal to the coarctation did not differ significantly from the corresponding area in the sham operated normotensive control animals, these vascular regions remaining normotensive. It was not possible, in this study, to measure pressure weekly in the arterial region above the level of the coarc- tation. However, direct readings of mean carotid arterial pressure, on the day of each experiment, showed a signifi- cantly elevated, hypertensive pressure in rats coarcted above the renal arteries. Also, the presence of a cardiac hypertrophy in these animals (as well as those with two kidney Goldblatt hypertension) would suggest that a hyper- tensive condition was present before the conduction of the experiments as well. Cardiac hypertrophy is a common find- ing in hypertension and is approximately proportional to the level of the hypertension. (17, 33, 40) 72 73 Concerning the hemodynamic portion of the present study there are three major findings from the data in the normotensive hindlimb vasculature of Group A compared to the data from Group B. These are: 1) Resistance is significantly elevated, both at rest and over a range of flows while resting flow is reduced. 2) Pressure-flow relationships suggest impaired pass- ive dilath1of the resistance vessels. 3) Residual vascular resistance after maximal vaso- dilation is significantly increased. These findings are intriguing because, in the majority of previous studies by other investigators, vascular beds of hypertensive animals that are not exposed to an elevated intravascular pressure do not exhibit an increased resis- tance at maximal vasodilation, or at any level of smooth muscle contraction, and do not display evidence for the vascular wall thickening found in hypertensive vascular beds. (3, 5, 10, ll, 12, 13, 14, 15, 16, 24, 26, 28, 37, 38, 40, 45, 49, 53) In addition, early studies on coarctation hypertension state that resting flow and presumably resis- tance are normal in the vascular regions distal to the coarctation. (17) The increased hindlimb vascular resistance in Group A, as manifested by the pressure-flow relationships (Figures 5 and 6) and by resistance values at rest (Figure 3), could theoretically be attributed to neural, humoral,or structural 74 factors or a combination of these factors. With regards to humoral factors it was found in the present study that the plasma renin concentration of Group A rats was elevated by 81% compared to rats of Group B. However, basal tone (resistance after acute denervation] resistance after NaNP) in these two groups was almost iden- tical. Therefore, it is unlikely that elevated levels of plasma angiotensin, or any other circulating vasconstric- tors directly contributed to the elevated resistance via their effect on smooth muscle contraction. (45) The present study was not designed to investigate the role of the sympathetic nerves in hypertension. However, since hindlimb resistance in the hypertensive animals coarcted above the renal arteries was still significantly increased after the presumed abolishment of all smooth muscle activity.it can be concluded that a portion of the increased resistance was not due to the immediate effects of neural factors (but neural factors operating chroni- cally could have changed the vessel structurally). Plots of Resistance versus Pressure (Figure 7) suggest that passive dilation of the vasculature is impaired in Group A compared to Group B. Impaired passive dilation of the arteries is a characteristic of hypertension and has been used by previous investigators to argue for the ex- istance of a vascular wall thickening and/or an increase in the wall-to-lumen ratio of the resistance vessels. (ll, l4, 17, 34) Totally aside from any implication of vessel wall 75 structure, the existence of a hypertensive characteristic (impaired passive dilation) in the hindlimb vasculature of Group A is significant in that this vasculature was not hypertensive during the study. One might argue that passive distention is not even occuring in the Groups during the pressure-flow studies because resistance increases or stays constant between the 3rd and 4th flows (0.5 & 1.0 cc/minute) (2nd and 3rd for Group D, 0.25 & 0.5 cc/minute) rather than decreases. (Figures 6 & 7) This then would suggest that some auto- regulatory phenomenon is being displayed by the pressure- flow studies. Therefore, it can not be denied that smooth muscle con- traction could have influenced vascular resistance during the flow sequences. However, one can not rule out passive distention as a major determinant of vascular resistance during the flow sequences. It should be noted that even though resistance increases for example in Groups A & B between the third and fourth flows, resistance in both groups at the fourth flow (1.0 cc/minute) is still 2212! that at the initial flow (0.125 cc/minute) for both groups. Were it not for a considerable distending force at the fourth flow resistance would be even higher than at the initial flow of 0.125 cc/minute due to both metabolic and myogenic factors. Besides, resistance again continued to decrease in both groups at flows higher than 1.0 cc/minute. According to the law of LaPlace (T = P (r/w)) the 76 internal radius of the vessels in all groups would have to decrease, and thus resistance increase, beyond the initial values (at a flow of 0.125 cc/minute) in order to com- pletely negate the distending force of the increase in pressure at each successive flow. (13) Since, relative to the initial value, this never occurs in any group it may be said that the vessels were for the most part being distended over the flow range studied. It may then be postulated that distention of the vasculature was impaired in Group A, since resistance remained elevated over these flows in Group A compared to Group B. Whether or not this impaired distention is due to an underlying increase in vessel wall thickness (thus increas- ing the wall-to-lumen ratio) in Group A relative to Group B cannot be determined for certain from the pressure-flow study. Profile analysis revealed that the level of the pres- sure-flow curves for Group A was greater than the curves for Group B, but that the curves were not parallel to each other. Although this would be the case if the hindlimb vessels of Group A had structurally based greater wall- to-lumen ratios than Group B, one cannot tell for certain if this is the case in the present study, since smooth muscle activity was not abolished in any of the groups during the pressure-flow experiments. In explanation, one 77 cannot tell if the implied increase in wall-to-lumen ratio was due to an underlying thickened vessel wall or, instead, due to an increased amount of vascular smooth muscle contraction or both. (13) Nevertheless, it can at least be said that the hind- limb arteries of the intact animals with coarctation hypertension are less distensible than those for normoten- sive controls. One might suggest that more conclusive data may have been obtained if the pressure-flow studies were done dur- ing complete vascular smooth muscle relaxation. However, such a study has its own set of experimental problems. Firstly, if the vessels are completely relaxed in the intact animal, blood pressure is markedly reduced so that the life support of the animal is severely compromised. Thus, the hindlimbs (if that is to be the vascular region studied) have to be completely separated (nerves and vessels severed) from the animal and the animal killed if a pressure-flow study is to be performed. (11) Also, in this type of study with the hindlimbs separated from the animal's body an oxygenated plasma substitute has to be used as a perfusate. (11, 12, 14, 15, 16) Thus, data on an intact animal (full neural and humoral factors present) cannot be obtained. Secondly, with the vessels completely relaxed the formation of edema is considerably facilitated in the hindlimbs aside from what might be normally expected from high perfusion pressures and flows. In fact, Folkow, in this type of pressure-flow experiment stated that intra- vascular pressures had to be kept between 10 and 40 mm Hg in order to avoid edema formation in the hindlimb. (11) This is hardly within the normal physiological pressure range for the animal. As can be seen, whatever extrapola- tion is made between data gathered in this manner (no vascular tone, low pressures, artificial media, etc.) and the intact animal under normal physiological conditions is tenuous at best. There apparently seems to be no "good" way to conduct pressure-flow studies on the hindlimb vasculature. In the present study, in the hindlimbs of Group A compared to either Groups B or C, resistance at maximal active vasodilation was significantly increased. It is unlikely that this difference is due to an increase in extravascular pressure through tissue edema because similar water contents were found in the perfused limbs of all groups. (Table l) This impaired active vasodilation suggests that some structural component is involved in the general elevated hindlimb resistance seen in the rats with coarctation hypertension. Concerning structural factors, the changes in resistance in the hindlimbs of Group A cannot be explained on the basis of a decreased number of vessels due to the slight hindquarters atrophy that occurred after 79 clipping, because there was similar limb atrophy in the normotensive rats with aortas clipped below the renal arteries (Group C); and, in these latter rats hindlimb resistance was reduced, passive vasodilation was enhanced, and residual resistance after maximal vasodilation was reduced, compared to the sham operated control rats. In other words, aortic coartation alone (Group C) would tend to cause changes in the resistance of the hindlimb vessels in the opposite direction from those that were observed in the rats with coarctation hypertension. Indeed, the findings in the control group of rats coarcted below the renal arteries add increased significance to the obser- vations made in the rats with coarctation hypertension. Consequently, from this logic it is not surprising that resistance at maximal vasodilation in Group A is increased by "only" 10% compared to Group B. One may wonder why resistance measurements in Group A & B differ by 10% at maximal vasodilation yet differ by 57% at rest. These anomalies are depicted in Figure 11 which depicts theoretical radii and internal circumfer- ences for all the groups at rest and maximal vasodilation. These values are derived from Poiseuillc's equation which 31:2 Wr4N made within each group (which will be the case below), states that resistance = (13) If comparisons are 8Ln can be an arbitrary constant. For purposes of the ‘WN model, the internal vessel radius of Group B at maximal Figure 11. 80 Theoretical vessel radii and circumferences at maximal vasodilation and at rest in the rat hindlimb Group A rats with coarctation hypertension Group B - sham operated normotensive control rats Group C = rats coarcted about the aorta below both renal arteries Group D = rats with two-kidney Goldblatt hypertension A% Maximal Vasodilation to rest = percent difference in circumference from.maximal vasodilation to rest 81 ..0Nl véml hmm...» 0mm..m 0 V0000 mmmfiw N000. 0 130mm 0 anomo .H «Human 0.4.0.1 mwmbé Nmmué 0000. m anomo m..mt mONNé 0.31.0 4 anomo 0.0mm 0... 292.53.; .. 1.43.3322 mtz: >m 4452x442 ...< mmozmmmu. 1.23050 024 :04m 1.42me2_ 4mmmm> 1.40....mm0m10. 82 dilation is set at 1.0. This, then sets the constant of proportionality, 8Ln, relating resistance to the internal u radius of the vessel at 498.4. Using the conservative case by which one sets the wall—to-lumen ratio of each vessel equal at maximal vasodilation, one can calculate the per- cent of vessel contraction necessary to elevate flow resis- tance in each group from the level of maximal vasodilation to the level at rest. As can be seen from Figure 11 a contraction of roughly 24% is required to elevate resistance from maximal dilation to resting values in Groups B & C. However, a contraction of 29% is required in Group D, and a still higher contraction of slightly over 31% is required in Group A, to obtain resting resistance values for these two groups. These last two values pose an interesting, though purely speculative, problem in that the maximum amount of vessel contraction believed to be obtainable in a vessel with normal wall structure is about 27 and not more than 30 percent. (12, 13) Consequently, from this point of view, the increase in resistance in Groups B and C from maximal vasodilation to rest could be accounted for solely on the basis of an increase in the amount of vessel con- traction. However, since the amount of contraction in the vessel, by definition, cannot exceed the maximum amount attainable, then resistance at rest in Group D and espe- cially Group A cannot be explained simply by a certain 83 amount of vessel contraction. With this the case, then the only way Group A could attain its resistance value at rest would be for it to have an increased amount of vessel wall mass displaced into the vessel lumen for each given level of vascular contraction compared to Group B. In other words, Group A would have to have an increase in vessel wall thickness relative to Group B. Of course, this statement is not proof positive for vessel wall thickening and depends on the fact that the resis- tance vessels cannot constrict more than 27 to 30 percent. However, coupled with the fact that the increase in resis- tance at maximal vasodilation in Group A relative to the normotensive sham controls (Group B) cannot be simply explained by tissue edema or limb atrophy, the possibility that this increased resistance at maximal vasodilation in Group A reflects an underlying vessel wall thickening now seems somewhat more plausible. Whatever the case may truly be, of prime importance related to finding elevated resistance measurements in the hindlimb vasculature of rats with coarctation hyper- tension (Group A) is that this vasculature was 331 exposed to an elevated intravascular pressure. Therefore, high intravascular pressure cannot be postulated as being responsible for the resistance changes seen in these rats. The findings of impaired vasodilation in the hind- limbs of these rats would especially appear to contradict 84 the hypothesis that structural changes in vascular walls of hypertensive animals (which can account for almost all the elevated resistance) are the result of increased intravascular pressure. (3, 5, lo, 11, 12, 15, 16, 26, 30, 45, 51, 52, 53) There are a number of reasons why this study's results apparently contradict this hypothesis. First are the interpretation of data and types of experimental models used by some investigators to advance the opinion that structural changes in hypertension result from increased intravascular pressure. In separate studies, Folkow, Sivertsson, Weiss, and Martin have shown that hypotension can reduce resistance by causing an atrophy of the arterial vessel walls in the hypotensive region. (15, 16, 45, 53, 30) These investi- gators then used a "reverse" logic to postulate that an increased pressure would, consequently, lead to a thick- ening of the vessel wall, as is the case in hypertension. Indeed, using the results, in the present study, in animals with coarctation of the abdominal aorta below the renal arteries, and those with two-kidney Goldblatt hypertension, one might come to the same conclusion. In explanation, the hindquarters of the rats with Goldblatt hypertension were exposed to an elevated intravascular pressure. These hind- limbs exhibited a significantly increased resistance at maximal vasodilation compared to the groups of rats with normotensive hindlimbs. (A & B) Conversely, the 85 hindquarters of rats coarcted below the renal arteries were exposed to a decreased intravascular pressure and these hindlimbs showed a decreased resistance at maximal vasodil- atian. Concerning this latter point, however, finding that hypotension can cause vessel wall thinning does not neces— sarily prove the opposite situation can be true. It is possible that other factors may come into play in a hypotensive vascular bed besides the reduced intra- luminal pressure. As suggested by Weiss and Hallback (53), and Folkow (16) the underperfusion of the vascular region exposed to hypotension may decrease the "nutritional" supply to the vessels, thus resulting in an atrophy of the vessel wall. Therefore, the influence of blood pressure on vascular changes may not be strictly a pressure phenom- enon when low levels of pressure are present. Also, it is important to note that, unlike the present study, Folkow's studies never examined a normotensive vas- cular bed of a hypertensive animal. Instead, Folkow exam- ined the "protected" (hypotensive) vascular bed of spontan- eously hypertensive rats. Data from these beds, compared to normotensive control rats, provided evidence, through reverse logic, that structural changes in the resistance vessels result from increased intravascular pressure. However, the appropriate control for a "protected" vascular bed in a hypertensive animal may, instead of a normotensive bed, be a similarly "protected" bed in a normotensive animal. Using this type of comparison, Weiss 86 and Hallback's data in "protected" beds of adult spontan- eously hypertensive rats provide evidence that vessel wall thickness remains greater in the hypertensives. Since decreased pressure could not totally reverse the structural change, this result then suggests that other factors besides pressure may be involved in the vascular structural changes. It would also seem, then, that a better test for the hypothesis that, structural changes are secondary to changes in pressure alone, might be provided by examining a normo- tensive, rather than a sub-normotensive vascular bed of a hypertensive animal. This type of model was used in the present study and in the studies of Nolla-Panandes (33) and, Pamnani and Overbeck (36), and these studies have provided evidence for elevated resistance and structural vascular changes in the normotensive vascular beds of rats with coarctation hypertension. Therefore, it seems that some factor may be present in this type of experimental model that is not present in other models used to investi- gate the role of pressure in the deve10pment of structural vascular changes and an elevated resistance. It has already been noted that the rats with coarc- tation hypertension in this study (Group A) had elevated levels of plasma renin compared to normotensive sham controls (Group B), but that it was unlikely that this factor directly caused the elevated resistance seen in Group A. However, it is possible that the elevated renin 87 levels may have affected vascular resistance through effects on vascular wall structure. Sen and her associates have correlated plasma renin activity in spontaneously hyperten- sive rats with ventricular hypertrophy. (44) Also, exposure to elevated levels of renin or angiotensin may have altered the ionic composition of the arterial walls in rats with coarctation hypertension, as had been demonstrated by Villamil. (51) Using the findings of Villamil (51) and, Pamnani and Overbeck, (36) one could form a hypothesis concerning why increased hindlimb resistance was found in this study's rats with coarctation hypertension. In short, Villamil found that chronic exposure to angiotensin increases the salt content and sodium.perme- ability of arterial strips. Pamnani and Overbeck found increases in the sodium,potassium and water content of arteries taken from the normotensive vascular beds of rats with coarctation hypertension. In this same experimental model, the present study has shown that plasma renin concentrations are increased as is hindlimb vascular resistance at maximal vasodilation. Therefore, increased plasma renin concentrations in the present study's rats with coarctation hypertension could have increased the amount of salt and water content in the rats' hindlimbs (as in Pamnani's and Overbeck's study) thus causing a thickening of the vessel walls. This in turn would have resulted in an increased hindlimb resistance 88 even at maximal vasodilation, which indeed, was truly the case in the present snidy's rats with coarctation hyper— tension. With respect to other investigators' work in coarc- tation hypertension, the involvement of the renin-angio- tensin system in coarctation hypertension has both been supported and denied. Whether or not, and to what extent, this system is brought into play in coarctation hyper- tension may explain some of the differences in the results of studies conducted by different investigators. In this regard, most studies that have shown that no vascular alterations occur in the vascular regions distal to the coarctation have been on thoracic aortic coarctation. It is possible that the differing results may be due to dif- ferent sites of coarctation (thoracic or abdominal) and the possible resulting difference in renin-angiotensin involve- ment , Aortograms in dogs with thoracic coarctation hyperten- sion have shown that this type of coarctation hypertension is associated with an extensive amount of collateral circu- lation (23). It is possible that this collateral circula- tion could reduce the degree of involvement of the renal pressor system in thoracic coarctation through its effect on renal perfusion (1). lbcall that, in the present study coarc- tation hypertension was produced by coarcting the abdominal aorta just above the origin of the renal arteries, which 89 obviously brought the renin-angiotensin system into play. (Figure 10) It is true, however, that in another study of abdominal coarctation hypertension Bevan et al. (3) did not find ele- vated levels of salt and.water in either the hypertensive or the normotensive arterial regions. Additionally, Bevan et al. found that the vessel wall thickness of only the ar- teries from hypertensive regions could be correlated with the level of hypertension. Bevan's study implies that high intravascular pres- sure may be responsible for increasing the wallthickness of the arteries. This implication, and his results on water and ions, differ from the implications of the present study and results reported by Pamnani et al. (36), respectively, both of whom, like Bevans, used animals with abdominal coarctation as the experimental model. This apparent di- cotomy between data on similar experimental models may be attributable to a large variety of factors. Firstly, the model of hypertension studied by Bevan et al. apparently does not have elevated levels of plasma renin in contrast to the present study. Also, it appears that Bevan et al. did not produce as severe a coarctation as was produced by the present study, which then may have reduced the in- volvement of the renin-angiotensin system. The pressure gradient between the average carotid and femoral arterial pressures was only 26 mm Hg in the rabbits (127 mm Hg-lOl mm Hg), compared to 40 mm Hg in the rats of the present study (152 mm Hg-llZ mm Hg). Also, the duration of the coarctation was only two weeks in the rabbits studied by Bevan, compared to four weeks in the rats used by Pamnani et al. and the present 90 investigator. Finally, and most importantly, Bevan et al. studied the arterial vessels in vitro and did not make direct com- parisons between hypertensive and control animals. In summary, certainly, the results of the present study do not exclude the possibility that elevated intra- luminal pressure may alter vessel wall structure. Indeed, hindlimb resistance after maximal vasodilation was signifi- cantly higher in rats with Goldblatt hypertension (which were exposed to higher mean arterial pressures) than the rats with coarctation hypertension. The plasma renin con- centrations of the rats with Goldblatt hypertension were not measured, however. (See Appendix A) Therefore, the possi- bility remains that a greater degree of vascular altera- tion was brought about by elevated levels of plasma renin and not by elevated levels of intravascular pressure in Group D. Two kidney Goldblatt hypertension has been pre- viously associated with elevated plasma renin levels (40), but this cannot be demonstrated in the present study. Consequently, the results of the present study (Group A versus B, and, Group D versus A) suggest that, although elevated intravascular pressure may cause structural changes in the resistance vessels, it is not the only factor that can cause these changes and therefore is not necessary for bringing about the structural vascular changes seen in hypertension. In addition the results of Group A versus B further suggest that structural vascular changes (indicated by impaired.maximal vasodilation) and 91 elevated vascular resistance are not necessarily a second- ary vascular adaptation to a primary increase in intra- vascular pressure but instead may be related to the cause of hypertension itself. Additional avenues of investigation should be pursued to determine if the resistance changes seen in the rats with coarctation hypertension were actually due to an increase in the wall-to-lumen ratio of the hindlimb vessels, for example, by conducting norepinephrine dose-response experiments from a level of maximum dilation to maximum constriction, as prescribed by Folkow and his coworkers (12). Also, pressure-flow studies, done on the vasculature during complete smooth muscle relaxation, although difficult, would help in this regard. Similar studies could also be done on rats with coarc- tation hypertension where the effects of neural and renal humoral factors are abolished to determine what role these components may play in the vascular alterations. This might be accomplished by severing the sympathetic nerve supply to the hindlbmb vessels in the coarcted rats, or by creating chemical sympathectomy (and adrenal, demedullation) before doing the hemodynamic studies. It would also be interesting to block angiotensin II in these rats (e.g. with Saralysin). In conclusion, then, the results of the present study state: 1) Nommotensive vascular beds of rats with coarctation 92 hypertension display an elevated resistance. 2) Structural vascular changes, indicated by impaired vasodilation, may contribute to this elevated resistance. 3) These changes are not attributable to hindquarters atrophy, but may be related to elevated levels of circulat- ing renin; and, 4) In contrast to previous theories, the elevated resistance and impaired vasodilation were not caused by high intravascular pressures or increased limb blood flows. APPENDI CES The following appendices are designed to serve as a guide to the development of experimental.protocol and as a reference for statistical analyses. 93 APPENDIX A PROBLEM SOLVING AND THE DEVELOPMENT OF EXPERIMENTAL PROTOCOL When dealing with the perfusion of a small vascular bed, such as the rat hindlimb, the possibility of embolizing the bed becomes quite high. With such a vascular bed, even the smallest amounts of air or foreign material, amounts that can easily pass through the perfusion circuit virtually unnoticed, can significantly occlude the perfusion circuit at the level of the small blood vessels. During early attempts at conducting experiments for this study, the rats' hindlimb frequently became embolized, presumably by air, as soon as the perfusion started. Since this condition prevented all portions of the experiment from being completed, the occurrence of embolization was a serious problem. (As stated previously, as a general means of preventing this problem, the transparent pump tubing leading to the hindlimb was carefully watched for the first eleven minutes of every perfusion so that potential emboli could be re- moved. However, watching the tubing did nothing to prevent embolization immediately as the perfusion started. 94 95 Since immediate, as well as early, embolization could be caused by small blood clots, and general particulate matter, as well as air, changes were made in the experi- mental preparation so that the occurrence of these potential emboli might be eliminated. Consequently, as opposed to early experimental attempts, all beakers of normal saline used for flushing canuli, etc., were then covered with parafilm to prevent their exposure to particulate matter. To further prevent blood clots from forming, heparin was to be placed inside the syringe and beaker used for collection of donor blood, instead of just the beaker, as had been the case during the early experiments. Also, whenever an unsuccessful attempt at cannulating a femoral or carotid artery occurred, the resultant air space in the cannula was to be flushed out with heparin instead of saline. To aid in flushing out trapped air in the cannula after their insertion into an artery, the rubber part of the cannula was to be trimmed back so that a needle, attached to a syringe of saline, could be inserted all the way to the bottom of the cannula. In this manner one could insure that air could be removed from the entire length of the cannula. By taking special care during the pre-perfusion prep- arations, and by implementing these changes, the problem of embolization was greatly, if not entirely, eliminated. During the period of time that the embolization prob- lem was being worked on, a Sigmamotor finger pump was used 96 for perfusing the rat's hindlimb. This pump possessed its own unique problem in that it produced pulse pressures in excess of 150 mm Hg (at a flow of 1.0 CC/min) when con- nected to a non-embolized limb. Since this pulse pressure was well outside the physiological norm for the rat, steps had to be taken to try to reduce the pulsations to a more acceptable level. As a first attempt at solving this problem, a saline filled "Windkessel" was connected to the downstream pump tubing. This significantly reduced the pulsations during the initial part of the perfusion. However, the gradual increase in hindlimb resistance due to reactive hyperemia caused a large back-up of blood into the "Windkessel". This then produced two undesirable effects. Firstly, this sig- nificantly reduced the air space in the Windkessel thereby increasing the pulse pressure. Secondly, this caused a displacement of blood from the animal's circulation which in turn further increased resistance by activating the baroreceptors. This then established a positive feedback loop which prevented a steady state from ever being estab- lished in the preparation. The use of duel Windkessels could not prevent this situation from occurring. As an alternative to the Windkessel concept, attempts were made to reduce the pulse pressure by reducing the stroke output of the pump and alternatively by increasing the pump downstream capacitance. Increasing the pump's 97 capacitance by increasing the volume available in the pump circuit proved unsatisfactory. Although this alteration significantly reduced the pulsations it diaplaced too much blood from the rat's circulation. Consequently, many types of pump tubing of varying diameters were tested to see if the large pulse pressure could be reduced by reducing the stroke output of the pump. This too proved unsatisfactory because pump tubing of a small enough diameter to adequately reduce the stroke output could not deliver flows to the hindlimb in amounts greater than 0.75 cc/min. Conversely, pump tubing of a diameter large enough to deliver higher flows did so at unacceptable pulse pressures. After much searching for tubing of an optimum diameter it was found that the "best" tubing still produced pulse pressures in excess of 100 mm Hg at the upper levels of flows to be used in the experiments. Since such pulsations were still abnormally high for the rat hindlimb the Sigma motor pump had to be abandoned altogether. A duel chamber hydrolic pump was then employed as the blood pump for the perfusion studies. This pump was de- signed such that pulse pressures were never higher than 8 mm Hg. However, this pump had a drawing pressure of 60 mm Hg at 1.0 ec/min. In the preparation prior to the use of this pump, blood was drawn through the pumping mechanism from the rats right femoral artery. In rats clipped below the renal arteries, 98 mean arterial pressure in this region was approximately 75 mm Hg. Consequently, whenever flow was increased above 1.0 cc/min., a negative pressure was induced in the inflow circuit of the pump. This cannot only damage the pump but also facilitate the passage of air into the pump tubing. Therefore, the surgical procedure on the rats had to be altered so that the carotid artery could be isolated and used to supply blood to the pump. By incorporating the isolation of the carotid artery into the pre-perfusion preparations, it was then possible to measure arterial pressures above and below the site of the coarctation. In a previous study by Pamnani et.al. (36), on the same type of rats, these same pressures were measured with the rats under light ether anesthesia. To simulate this condi- tion using chlorolose and nembutol anesthesia, the pre- viously used anesthesia combination had to be altered. This altered combination had to leave the rat "light” enough for blood pressure measurements yet deep enough to keep the animal completely calm for the remainder of the experiment. Before it was decided that both carotid and femoral arterial pressures were to be measured, the rat was given its total dosage of both anesthetics early in the surgical preparation. This dosage regimen kept the animal too deep for the pur- poses of measuring the arterial pressures. After experi— menting with various sequences of dosage administration, the best dosage sequence was the one that has been 99 previously mentioned in the methodology. The hydraulic pump used for the perfusion studies con- tained about 1 cc of fluid space outside of the rats circu- lation. Early in the study this space was filled with normal saline which was then pumped into the animal at the start of each experiment. The resulting displacement of 1 cc of blood from the rat‘s circulation helped reduce arterial pressure to the extent that negative pressures were occasionally incurred in normotensive rats. For this reason it was decided to prime the pump with blood taken from an - appropriate donor. Therefore allowances had to be made for yet another addition to the pre-experimental protocol. Since the rats used for the perfusion at this time were from a highly inbred colony transfusion reactions could be discounted. Unfortunately at this time the colony of rats contracted pneumonia. This problem initially caused the elimination of a number of experiments due to the discovery of the disease upon autopsy of the animal. However, this problem became so severe that rats had to be obtained from outside the colony to continue the study. . Due to financial considerations the new rats had to be obtained from an outbred colony. Consequently, these rats had to be tested for transfusion reactions before they could be employed in the experiments. Five pairs of rats were randomly selected and paired from the outbred colony. Using the standard testing 100 procedures with washed cells and serum from each rat, compatibility tests were then commenced on all rats. The tests showed no signs of hemolysis or agglutination either macrosc0pica11y or microscopically for any pair of rats. Therefore, the rats of the outbred colony were deemed com- patible. During subsequent experiments, no signs of trans- fusion reactions were ever noted. This new strain of rats, however, was associated with some quite unexpected problems. Because this strain was outbred, the rats were much heartier than inbred rats. Consequently, the new, outbred strain grew to be an average of 100 grams larger than the rats used by Pamnani (36), over the same four week pre-experimental period. The clip sizes for the normotensive sham controls then became relatively too small for the new larger breed of rats. Consequently, this ”control” group started to develop coarctation hyper- tension. New clips were then constructed to be approximately the same percentage larger in internal diameter as the new rats were in general body size. The first set of new clips (1.48 mm 1.0.) maintained the carotid-femoral arterial pressure gradient at less than 6 mm Hg. One other set of clips, with an overly large interior diameter (1.70 mm) were then employed to see if this gradient could further be reduced. There was virtually no difference seen in the gradient between animals with either sham clip, but as an 101 exercise in caution the larger clip was used for the sham controls for the remainder of the study. A more serious problem with the new colony of rats was revealed during the surgical isolation of the hindlimb of the rats. Either for reasons of a variation in the hindlimb vasculature or for lack of a good clotting capacity, the out- bred rats bled greatly with little or no clotting when the muscles were teased apart, as had been the prescribed method for isolating the hindlimb. This bleeding never stopped during the experiment because heparin had to be used in the blood to prevent it from clotting in portions of the pump. Consequently, the rats were invariably in some stage of_ shock during an experiment, and, therefore, the results of such experiments could not be considered valid. As a first measure to combat this problem the time between the completion of the hindlimb isolation and admin- istering heparin to the rat was extended 15 minutes above the normal 30 minute period. However, waiting an extra 15 minutes let the animals become so light, that it was impos- sible to get readings of the carotid and femoral arterial pressures. More importantly this maneuver did little to stop the bleeding from the hindlimb. Subsequently, additional ligatures and gelfoam were employed to stop the bleeding but these measures proved to be ineffectual. Finally, it was decided upon to use cautery to isolate the limb. By sealing the ends of severed 102 muscles and small blood vessels this method proved highly effective in preventing any bleeding from occurring during the experiment. What experiments were completed, before cautery was used to isolate the hindlimb, seemed to reveal that there were no differences in the resistances of rats-with coarctation hypertension compared to the normotensive sham controls. This could have simply been related to the hemorrhaging of the hindlimb but there was no way that could be determined for certain. Therefore, to determine if the results in the rats with coarctation hypertension were artificial or truly valid, a fourth group of rats were to be included in the study. This group consisted of rats with two-kidney Goldblatt hypertension and were to be used to determine if the experimenter could detect resistance changes in hypertension. Three different renal clip sizes were used before the perfect size was determined. The smallest clip size totally occluded the kidney, which then caused either malignant hypertension or no hypertension at all (presumably since the damaged kidney could no longer produce renin). Conversely, the largest clip apparently did not cause a severe enough reduction of renal flow, as the majority of the rats prepared in this manner did not develop a substan- tial hypertension. Nevertheless, regardless of the clip used, any resultant hypertensive rat was used to comprise 103 the fourth group of experimental animals. Through all the aforementioned problems and their solu- tions a series of criteria were devised to serve as a set of guidelines for rejecting any data obtained from completed experiments. These criteria are as follows: 1) 2) 3) 4) 5) 6) 7) 8) 9) excessive bleeding and/or hematocrit below 40%; lung infection of any type; infarcted kidney(s) in any group except renal hypertensives; creatinine values outside two standard deviations of the mean; I carotid arterial-femoral arterial pressure gradient above 10.0 mm Hg in sham operated animals; carotid arterial pressure below 130 mm Hg in rats coarcted above the renal arteries; carotid arterial pressure sham rats or rats co- arcted below above 130 mm Hg in the renal arteries under conditions of normal sedation; surgery lasting longer than 2 hrs 30 mins (surgery defined as time animal is placed in ether until the start of perfusion; includes 30 minutes wait for heparin, taking carotid and femoral arterial pres- sure twice, and pump set up); animal getting less than 0.5 cc of chloralose (normal dosage was 0.7 cc chloralose); 10) ll) 12) 13) 14) 104 animal significantly embolized, i.e., NaNP resis- tance greater than 2 standard deviations of the mean, or sharp increase ( > 30 mm Hg rise in 12 seconds or less) in perfusion pressure without a return to pre- increase levels, : S‘mm Hg, after corrective measures were taken; differences between flow sets of greater than 30% where there is evidence for embolization before differences; acute denervation and/or NaNP resistance segments of experiment are to be eliminated if the animal constantly moves and tugs against cannula during entire measurement period; no decrease in perfusion pressure after acute denervation thus signifying a non-vital nerve supply to the hindlimb; experimental values beyond 1 3 standard deviations of the mean APPENDIX B REASONS FOR LIME WEIGHT NORMALIZATION When perfusing any extremity (such as the rat's hind limb) at a constant flow, one of the determinants of vas- cular resistance can be the size of the extremity; the lo- gic being, that limbs of different sizes contain different numbers of vessels arranged in parallel. Differences in limb size can add to the amount of variability already pre- sent in the resistance measurements in each experimental group, thus making it all the more difficult to pick up true difference in the values of each group. Consequently, it is common practice to express resistance measurements in terms of the amount of tissue mass (weight) in which the resistance was determined. Of course, if these weights were the same in all the experimental groups studied, there would be no need for such a ”normalization” of the data. In this study a dicotomy existed in the data on the weights of the perfused hindlimbs. The limb weights, ex- pressed in terms of body weight, of the rats with coarcta- tion hypertension were not different from those of sham controls. -However, both these limb weight/body weight ratios were significantly greater than either of the ratios for rats coarcted below the renal arteries or for the rats with two-kidney Goldblatt hypertension. Therefore the question arose as to in what manner the data should be ex- pressed. 105 106 Presumably there should be some form of inverse rela- tionship between the size of the limb and the resistance to flow in the limb; all other factors being constant. Conse- quently a linear regression was done of resistance during maximal vasodilation versus the limb weight of each group of rats. In all groups, whether the perfused limb wet weight or dry weight was used, or the Opposite limb's wet or dry weight was used, or whether a linear or exponential curve was being fitted to the data, there was absolutely Egysig- nificant regression or correlatiOn of any type present in the data. However, negative results do not necessarily mean that no relationship truly exists. Therefore, other means had to be used to determine whether or not a limb weight normalization was to be used on the data. In a subsequent analysis, it was found that the per- fused limbs of all the groups of rats contained more water, presumably from edema due to the perfusion, than did the Opposite non-perfused limb. Therefore, the perfused limb wet weights of the rats may have been different from the true limb weights of the rats (before perfusion). There- fore, these weights could not be used to normalize the re- sistance data. It was also found that the dry limb weight data, on either limb, had higher coefficients of variability than the other limb weights, so much so that resistance data 107 corrected with dry limb weight actually contained more variability than the resistance data alone. Consequently, dry limb weights could not be used to normalize the data. This left only the Opposite limb wet weights as the closest estimate to the true limb weight of the animal. These weights expressed in terms of body weight, were sig- nificantly less in either coarcted group Of rats compared to sham controls. In addition, resistance data expressed in terms Of these weights had the lowest amount of vari- ability compared tO the same data expressed with or without the other limb weights. In spite of the results Of the regression analysis, since the Opposite limb weight/body weight ratios signified some significant atrOphy in the hindlimbs Of coarcted rats, it was felt that all the resistance data should be expressed in terms of some type of limb weight. Since the Opposite limb wet weights were shown to best reduce the variability in the data, they were chosen for the limb weight normali- zation of the data. APPENDIX C DATA AND RESULTS OF STATISTICAL ANALYSES 108 TABLE 2 AVERAGE SYSTOLIC CAUDAL ARTERIAL PRESSURE First Measurement Group A B C D n 7 10 3 4 Mean 109.21 110.95 49.83 173.38 is.n. 12.55 4.93 10.23 6.24 VAR 157.49 24.30 105.58 38.90 SEM 4.74 1.56 5.93 3.12 S-N-K TEST Comp Range Diff. LSR 0.05 Sig 0.05 D vs C 4 123.6 18.46 * D vs A 3 64.2 13.69 * B vs C 3 61.1 14.38 * D vs B 2 62.4 10.66 * B vs A 2 1.74 8.88 NO A vs C 2 59.4 12.43 * GROUP Mean S.D. VAR SEM Comp D vs D vs B vs D vs B vs A vs OFUOFO AVERAGE SYSTOLIC CAUDAL ARTERIAL PRESSURE A 10 108.5 14.44 208.56 4.57 Range NNNUU 109 TABLE 3 Second Measurement B C 13 6 109.62 56 7.69 13.16 59.13 144.25 2.13 4.90 S-N-K TEST Diff LSR 0.05 126.2 19.99 73.7 16.21 53.6 15.50 72.6 13.43 1.1 10.94 52.5 13.43 D 6 182.17 18.61 346.27 7.60 Sig 0.05 * * NO Group Mean S.D. VAR SEM Comp V8 V8 V8 V8 V8 caverns V8 AVERAGE SYSTOLIC CAUDAL ARTERIAL PRESSURE (3&5an A 12 112.83 9.27 85.97 2.68 Range NNNWUub 110 TABLE 4 Third Measurement B 13 112.69 7.86 61.73 1 2.18 S-N-K TEST Diff 116.5 60.6 56.1 60.4 00.14 55.9 C 6 56.75 11.81 39.58 4.82 LSR 0.05 22.0 17.04 17.26 14.31 11.46 14.13 D 6 173.25 28.76 827.38 11.74 Sig 0.05 No Group Mean S.D. VAR SEM Comp D V8 V8 V8 V8 3 U! I: w V8 nzvmovo AVERAGE SYSTOLIC CAUDAL ARTERIAL PRESSURE A 29 110.47 11.78 138.73 2.19 Range NNNWU-b 111 TABLE 5 Total Measurement B C 36 15 111.10 55.07 7.03 11.58 49.45 134.21 1.17 2.99 S-N-K Test Diff LSR 0.05 121.6 11.54 66.2 9.10 56.0 8.98 65.5 7.32 0.63 6.08 55.4 7.75 D 16 176.63 20.46 418.65 5.12 Sig 0.05 NO Group A Group B Group C Group D 112 TABLE 6 WEEK-TO-WEEK COMPARISONS OF AVERAGE SYSTOLIC CAUDAL ARTERIAL PRESSURE lst vs. 2nd 2nd vs 3rd 3rd vs lst Not sig Not sig Not sig p>0.8 p>0.4 p>0.4 Not Sig Not Sig Not Sig p>0.6 p>0.3 p>0.5 Not Sig Not Sig Not Sig p>0.4 p>0.9 p>0.4 Not Sig Not Sig Not sig p >0.3 p >0.5 p:>0.9 Group Mean S.D. VAR SEM Comp V8 V8 V8 V8 V8 in U 3' C! :P e vs OWUOWO 113 TABLE 7 DIRECT MEAN CAROTID ARTERIAL PRESSURE A B C D 13 20 13 9 152.75 110.29 110.19 149.11 21.50 10.98 13.84 26.63 462.31 120.66 191.66 709.31 5.96 2.46 3.84 8.88 S-N-K TEST Range -Diff LSR 0.01 Sig 0.01 LSR 0.05 Sig 0.05 4 42.55 23.16 * ' ' 3 42.46 21.57 * ' ' 3 38.92 22.07 * ‘ ' 2 3.64 20.92 NO 15.69 NO 2 38.82 19.36 * ‘ - 2 0.10 17.19 NO 12.89 NO Group Mean S.D. VAR SEM Comp D VS VS VS VS w 3’ ‘3 > VS > C) (D O I'D 114 TABLE 8 DIRECT MEAN FEMORAL ARTERIAL PRESSURE B 20 104.88 11.14 124.20 2.49 A 13 112.28 19.9 395.99 5.52 Range Diff 4 71.3 3 41.79 3 36.91 2 34.39 2 7.4 2 36.91 S-N-K TEST 24.27 20.91 20.44 19.82 16.29 16.29 LSR0.01 Sig 0.01 * * C D 13 9 75.37 146.67 11.32 27.48 128.07 775.02 3.61 9.16 LSR 0.05 Sig 0.05 12.22 NO Group Mean S.D. VAR SEM Comp vs vs vs vs VS 00 3’ U U W U VS 0 CD 3’ CI! 0 O 115 TABLE 9 RESTING RESISTANCE X OPPOSITE LIMB WET WEIGHT A B C D 10 11 9 6 2447.2 1561.18 1119.33 2609.5 693.17 395.25 122.78 564.98 480,481 156,221 15075.0 319,200 219.20 119.17 40.93 230.65 S-N-K TEST Range Diff LSR 0.01 Sig 0.01 LSR 0.05 Sig 0.05 4 1490.17 868.0 * - - 3 1327.87 704.2 * - - 3 1648.32 688.9 * - - 2 162.30 689.9 NO 512.96 NO 2 886.02 583.7 * - - 2 441.85 600.5 *(t-test)446.47 NO 116 TABLE 10 RESTING FLOW (CC/Minute)/GRAM OPPOSITE LIMB WET WEIGHT Group A B C D n 10 ll 9 6 Mean 0.04896 0.06895 0.05713 0.05343 S.D. 0.01355 0.01470 0.00580 0.01097 VAR 0.00018 0.00022 0.00003 0.00012 SEM 0.00429 0.00443 0.00193 0.00447 S-N-K TEST Comp Range Diff LSR 0.05 Sig 0.05 B vs A 4 0.01999 0.01730 * B vs D 3 0.01552 0.01822 NO C vs A 3 0.00817 —— No B vs C 2 0.01182 0.01337 (t-fest) C vs D 2 0.00370 —- No D vs A 2 0.00447 -—- No Group n Mean S.D. VAR SEM Comp D VS DVS 117 TABLE 11 RESISTANCE AFTER ACUTE DENERVATION X OPPOSITE LIMB WET WEIGHT FLOW = 1.0 CC/MINUTE A 12 1457.75 230.21 52997.5 66.46 Range Diff 4 1137.89 3 768.03 3 501.31 2 636.58 2 131.45 2 369.86 B 10 1326.3 131.42 17271.8 41.56 5 S-N-K TEST LSR 0.01 357 326 278 275 235 253 Sig * * NO 956.44 74.86 603.3 24.95 0.01 2094.33 330.91 109499.5 135.09 LSR 0.05 Sig 0.05 175.06 No 118 TABLE 12 RESISTANCE AFTER SODIUM NITROPRUSSIDE X OPPOSITE LIMB WET WEIGHT Flow = 1.0 cc/Minute Group A B C D n 12 10 Mean 545.0 498.4 366.11 660.33 S.D. 41.34 57.27 44.97 44.68 VAR 1709.27 3280.27 2022.61 1996.67 SEM 11.93 18.11 14.99 18.24 S-N-K TEST Comp Range Diff LSR0.01 Sig 0.01 LSR 0.05 Sig 0.05 D vs C 4 294.22 84.42 * - D vs B 3 161.93 76.99 * ‘ - A vs c 3 178.89 65.74 * ' - D vs A 2 115.33 65.00 * ’ ~ A vs a 2 46.6 55.67 NO 41.12 * B vs C 2 132.29 59.73 * ' ' 119 TABLE 13 AVERAGE PRESSURE x OPPOSITE LIMB WET WEIGHT Group Mean S.D. VAR SEM Comp DVS DVS A vs DVS A vs BVS OW’OWO Flow - 0.125 cc/Minute A B C 11 10 9 397.27 270.30 258.67 102.94 57.08 56.14 10597 3259 3151 31.04 18.05 18.71 S-N-K TEST D 6 400.83 70.78 5010 28.90 Range Diff LSR0.01 Sig 0.01 LSR 0.05 Sig 0.05 4 142.16 135.59 * 3 130.53 123.60 * 3 138.60 107.62 * 2 3.56 105.95 No 2 126.97 91.20 * 2 11.67 95.91 No 78.76 71.30 NO NO AVERAGE RESISTANCE X OPPOSITE LIMB WET WEIGHT 0.125 cc/Minute Group Mean S.D. VAR SEM Comp D vs V8 V8 V3 U 3' ID > V8 OW>OWO A 11 3179 823 678538 248 Range 120 TABLE 14 Flow .73 .37 Diff 1142.4 1049.6 1110.9 31.5 1018.1 92.8 B 10 2161 455.00 207031 143.89 203036 S-N-K TEST D 6 3211 571.38 326480 233.27 LSR0.01 Sig 0.01 LSR 0.05 Sig 0.05 1086 990 852 848 730 768 t * * NO NO 630.8 No 571.1 NO 121 TABLE 15 AVERAGE PRESSURE X OPPOSITE LIMB WET WEIGHT Flow - 0.25 cc/Minute Group A B C D n 11 10 9 6 Mean 730.00 511.75 445.60 . 673.30 S.D. 185.1 94.12 92.20 75.63 VAR 34261 8859 8501 5720 SEM 55.81 29.76 30.73 30.88 S-N-K TEST Comp Range Diff LSR0.01 Sig 0.01 LSR 0.05 Sig 0.05 A vs C 4 284.4 194.0 * - - A vs B 3 218.25 175.6 * - - D vs C 3 227.7 212 * - - A vs D 2 56.7 178 No 132 No D vs B 2 161.3 181 NO 134 * B vs C 2 66.4 161 NO 120 NO 122 TABLE 16 AVERAGE RESISTANCE X OPPOSITE LIMB WET WEIGHT Flow - 0.25 cc/Minute Group A B C n 11 10 Mean 2918.9 2053 1771 S.D. 740.9 370.5 350.8 VAR 549007 137302 123092 SEM 223.4 117.17 116.9 S-N-K TEST Comp Range Diff LSR0.01 Sig 0.01 LSR A vs C 4 1147.9 767.17 * A vs B 3 865.9 694.3 * D vs C 3 922 837 * A vs D 2 225.9 703 No D vs B 2 640 715 NO B vs C 2 282 636 NO 2693 302.7 91599 123.6 0.05 Sig 0.05 523 532 473 NO NO Group Mean S.D. VAR SEM Comp w W U V U 0 VS V8 V8 VS V8 V8 nu: S’OWO 123 TABLE 17 AVERAGE PRESSURE X OPPOSITE LIMB WET WEIGHT Flow - 0.5cc/Minute A B C D 11 10 9 6 1256.9 794 617.6 . 1426 307.08 94.52 71.55 399.57 94298 8933 5119 159659 92.59 29.87 23.85 163.13 S-N-K TEST Range Diff LSR0.01 Sig 0.01 LSR 0.05 Sig 0.05 4 808.4 430 * - - 3 632 392 * ' ‘ 3 639.4 341 * - - 2 169.1 336 NO 250 NO 2 462.9 289 * - - 2 176.4 304 NO 226 NO 124 TABLE 18 AVERAGE RESISTANCE X OPPOSITE LIMB WET WEIGHT Group Mean S.D. VAR SEM Comp VS VS VS VS CO 3’ U 3' 0 VS 0 O U! 3’ O 00 Flow - 0.5 cc/Minute A B C D 11 10 9 6 2513.18 1587.9 1235 ~ 2853 614.13 189.27 143.32 798.7 377156 35822 20541 637904 185.17 59.85 47.77 326 S-N-K TEST Range Diff LSR0.01 Sig 0.01 LSR 0.05 Sig 0.05 4 1618 859 * - - 3 1265 783 * ~ ~ 3 1278 682 * - - 2 339.8 671 NO 499 No 2 925.3 578 * - - 2 352.9 608 No 452 No 125 TABLE 19 AVERAGE PRESSURE AND RESISTANCE X OPPOSITE LIMB Group Mean S.D. VAR SEM Comp VS VS VS VS VS 00 > U 3’0 0 O 01 3’0 (:00 VS WET WEIGHT A 11 2580 395. 156572 119. Range 4 7 3 Diff 1928.4 1270.9 1396.7 531.7 739.2 657.5 D 6 3111.7 833.4 694566 340.2 Sig 0.01 LSR 0.05 Sig 0.05 Flow - 1.0 cc/Minute B C 10 9 1840.8 1183.3 281.07 71.2 79003 5070 88.88 23.73 S-N-K TEST LSR0.01 757 * - 691 * - 601 * - 591.7 NO 509.4 * - 536 * - AVERAGE PRESSURE X OPPOSITE LIMB WET WEIGHT Group Mean S.D. VAR SEM Comp D vs w >' C, 3> V8 V8 V8 V8 V8 11 3275 310 96169 93. Range 4 3 3 2 2 2 126 TABLE 20 Flow - 1.5 cc/Minute .1 5 Diff 2263 1375.1 1504.9 758 617.1 887.8 B 10 2657.9 1770.1 457 208818 31369 144.5 S—N-K TEST LSR0.01 806 * 735 * 646 * 63o * 542 * 570 * 177.1 Sig 0.01 LSR 0. D 6 4033 05 839 703884 342.5 Sig 0.05 127 TABLE 21 AVERAGE RESISTANCE X OPPOSITE LIMB WET WEIGHT Comp D vs VS V8 V8 w >’ C! 3> VS OW>OWO Flow = 1.5cc/Minute A B C D 11 10 9 6 2183 1773.9 1190.6 2688.7 206.6 307.1 115.3 559.2 42682 94318 13290 312755 62.29 97.1 38.43 228.3 S-N-K TEST Range Diff LSR0.01 Sig 0.01 LSR 0.05 Sig 4 1498.1 540 * - - 3 914.8 493 * - - 3 992.4 429 * - - 2 505.7 422 * - - 2 409.1 363.4 * - - 2 583.3 382 * - - .05 128 TABLE 22 AVERAGE PRESSURE X OPPOSITE LIMB WET WEIGHT Flow - 2.0 cc/Minute Group A B C D n 11 10 9 6 Mean 3547 3197.8 2164 4566 S.D. 348.4 525.1 270.2 725 VAR 121374 275737 72996 526266 SEM 105.0 166.1 90.06 296 S-N-K TEST Comp Range Diff LSR0.01 Sig 0.01 LSR 0.05 Sig 0.05 D vs C 4 2402 828 * - D vs B 3 1368 754 * - A vs C 3 1383 657 * - D vs A 2 1019 646 * - No; sig A vs B 2 340. 557 No 414 w/O limb weight B vs C 2 1033. 585 * - - 129 TABLE 23 AVERAGE RESISTANCE X OPPOSITE LIMB WET WEIGHT Group Mean S.D. VAR SEM Flow = 2.0 cc/Minute A B C D 11 10 9 6 1773.5 1599.2 1082 2283 174.1 262.6 135.1 363 30313 68938 18244 131639 52.49 82.03 45.02 148 S-N-K TEST Range Diff LSR0.01 Sig 0.01 LSR 0.05 Sig 0.05 4 1201 414 * - - 3 683.8 377 * - - 3 691.5 328 * - _ 2 509.5 323 * - - . NO;Sig 2 174.3 278 NO 207 W/e limb weight 2 517 293 * - 7 Group Mean S.D. VAR SEM A 16 352.8 19.42 377.3 4.86 130 TABLE 24 BODY WEIGHTS (grams) B 21 353.1 13.41 179.7 2.93 C 15 358.9 14.41 297.6 3.72 NO significant differences among groups 358.7 30.40 821.3 10.13 Group Mean S.D. VAR SEM Comp A vs A vs VS VS (3 t: > VS WOUWOW A 16 HEART WEIGHT/BODY 0.00481 0.00059 3.52 x 10‘7 7.3lx 10‘8 132 TABLE 26 WEIGHT B C 21 15 0.00333 0.00337 0.00027 0.00025 6.35 x 10‘ 0.00006 0.00007 S—N-K TEST LSR0.01 Sig 0.01 0.00043 * 0.00043 * 0.00048 * 0.00044 * 0.00044 * 0.00036 * 0.00015 Range Diff 4 0.0148 3 0.0144 3 0.0070 2 0.0078 2 0.0068 2' 0.0004 8 D 9 0.00403 0.00041 1.65 x10- 0.00014 7 Comp 3’ 0 WC 0000 V8 V8 V5 U >‘ (3 I: p U A 16 0.00326 0.0042 1.77 x 10‘ 7 0.00011 Range 4 3 Diff 0.00101 0.00048 0.00097 0.00004 0.00044 0.00053 133 TABLE 27 LEFT KIDNEY WEIGHT/BODY WEIGHT B c 0 21 15 9 0.00374 0.00370 0.00273 0.00045 0.00039 0.00083 2.00 x 10'7 1.51 x 10'7 6.8 x 10’7 0.00010 0.000010 0.00028 S'N‘K TEST LSR0.01 Sig 0.01 LSR 0.05 Sig 0.05 0.00065 0.00050 0.00064 0.00045 0.00048 0.00055 * NO * NO NO NO 0.00040 0.00035 0.00037 0.00042 * NO 134 TABLE 28 RIGHT KIDNEY WEIGHT/BODY WEIGHT Group A B C D n 16 21 15 9 Mean 0.0034 0.0037 0.0037 , 0.0044 S.D. 0.00055 0.00056 0.00046 0.00081 -7 -7 -7 VAR 3.07 x 10 3.16 x10 2.14 x:10 6.56 x 10-7 SEM 0.00014 0.00012 0.00012 0.00027 S-N—K TEST Comp Range Diff LSR0.01 Sig 0.01 LSR 0.05 Sig 0.05 D vs A 3 0.00099 0.00073 * - - D vs B 2 0.00069 0.00062 * - - D vs C 2 0.00069 0.00065 * - - B vs A 2 0.00030 0.00051 NO 0.00038 NO C vs A 2 0.00030 0.00056 NO 0.00042 NO GROUP Mean S.D. VAR SEM 135 TABLE 29 PLASMA CREATININE ng A B C 12 11 8 13.70 10.00 9.90 1.14 0.91 1.24 0.366 0.408 0.486 0.134 0.166 0.236 0.106 0.123 0.172 No Significant differences among groups .25 .25 .456 .208 .204 GROUP Mean S.D. VAR SEM 136 TABLE 30 PLASMA SODIUM mEq/L A B C 11 ll 8 1433.5 1484.5 1053.0 130.3 134.9 131.6 6.23 6.81 7.13 38.91 46.42 50.77 1.88 2.05 2.52 No Significant differences among groups 805 134.2 6.99 48.87 2.85 137 TABLE 31 PLASMA POTASSIUM mEq/L GROUP A B C D n 11 ll 8 6 X 46.85 49.10 35.10 25.15 Mean 4.26 4.46 4.39 4.19 S.D. 0.65 0.50 0.62 0.57 VAR 0.43 0.25 0.39 0.32 SEM 0.20 0.15 0.22 0.23 No significant differences among groups GROUP Mean S.D. VAR SEM PLASMA CALCIUM mEq/L A 10 32.775 3.28 0.480 0.230 0.152 No significant differences among groups 138 TABLE 32 B 11 36.025 3.28 0.297 0.088 0.090 .37 .309 .100 .117 13.0 3.25 0.175 0.03 0.090 Group Mean S.D. VAR SEM No 139 TABLE 33 PLASMA MAGNESIUM mEq/L A B C 10 11 7 25.15 29.935 17.14 2.552 2.721 2.449 0.341 0.462 0.204 0.116 0.213 0.042 0.108 0.139 0.077 Significant differences among groups 9.665 2.416 0.124 0.015 0.062 GROUP Mean S.D. SEM VAR 140 TABLE 34 HEMATOCRITS A B 13 14 574.25 622 44.17 44.43 2.76 2.61 0.766 0.697 7.629 6.802 NO significant differences among 12 521.75 43.48 2.81 0.811 7.892 groups 310.75 44.39 2.30 0.869 5.289 141 TABLE 35 PLASMA RENIN CONCENTRATION (ng/m1)/hr ANGIOTENSION I GROUP A B Mean 21.9 12.1 S.D. 7.209 2.708 VAR 51.977 7.331 SEM 2.000 0.751 n 13 13 Significant difference 2-tailed t-test p < 0.001 142 TABLE 36 HEART WEIGHT/BODY WEIGHT FOR RATS USED FOR PLASMA RENIN CONCENTRATIONS Group A Group B n 13 15 Mean 0.00468 0.00330 S.D. 0.00054 0.00025 VAR 2.91 x 10'7 6.15 x 10"8 SEM 0.000149 0.000064 Means are statistically significantly different with students two-tailed t-test, p < 0.001. GROUP GROUP GROUP GROUP mean SQD Var. SEM. S-N-K test comp range Diff LSR 0.01 LSR 0.05 Sig 0.01 Sig 0.05 * No t test mEq/L 143 APPENDIX C KEY rats with coarctation of the abdominal aorta above the origin of bOth renal arteries (coarctation hypertension) normotensive sham operated control rats rats with coarctation of the abdominal aorta below the origin of both renal arteries rats with two-kidney Goldblatt hypertension number of rats in each group average group value for the parameter listed standard deviation of the group group variance standard error of the mean of the group. Student-NewmaneKuels test intergroup comparison being made number of group means the comparison encompasses difference between the means of the two groups being compared (Least Significant Range) smallest difference needed between the means of the two groups for the difference to be statistically significant, p < 0.01 least significant range, p < 0.05 significant difference at p < 0.01 significant difference at p < 0.05 a statistically significant difference exists no statistically significant difference exists no need for statistical comparison test used was students two-tailed t-test milliequivelents per liter 10. 11. 12. BIBLIOGRAPHY Bagby, S.P., W.J. McDonald, et al. Abnormalities of Renal Perfusion and the Renal Presser System in Dogs With Chronic Aortic Coarctation. Circ Res 37: 615-620, 1975. Berne, R.M. and M.N. Levy. Cardiovascular Physiology. C.V. Mosby CO., Saint Louis, 1972. Bevan, J.A., R.D. Bevan, et al. Analysis Of Changes in Reactivity of Rabbit Arteries and Veins Two Weeks after Induction of Hypertension by Coarctation of the Abdomi- nal Aorta. Circ Res 37: 183-190, 1975. Bevan, R.D. Nerve-Muscle Relationships in Hypertension. Blood Vessels 12(6): 331-332, 1975. Bevan, J.A., R.D. Bevan, et a1. Changes in the Contrac- tile Response of Arteries and Veins from Hypertensive Rabbits to Sympathetic Nerve Activity: Assessment of some Postsynaptic Influences. Blood Vessels 13: 167- 180, 1976. —— Conway, J. Vascular Reactivity in Experimental Hyper- tension Measured After Hexamethonium. Circulation 17: 807-810, 1958. Davisohn, I. and J.B. Henry. Clinical diagnosis by Laboratory Methods. Saunders, Philadelphia, 1969. Farris, E. J. and J. O. Griffith. The Rat in Laboratory Investigation. J.B. Lippincott CO., Philadelphia, 1949. Field, L.W. et a1. Constant Output Blood Perfusion Pump for Use in Cross Transfusion Experiments in Small Labo- ratory Animals. J. Appl. Physiol. 12: 142-144, 1969. Folkow, B., G. Grimby, et a1. Adaptive Structural Changes of the Vascular Walls in Hypertension and Their Relation to the Control Of the Peripheral Resistance. Acta physiol. scand. 44: 255-272, 1958. Folkow, B., M. Hallback, et a1. Structurally Based In- crease Of Flow Resistance in Spontaneously Hypertensive Rats. Acta physiol. scand. 79: 373-378, 1970. Folkow, B., M. Hallback, et a1. Background Of Increased Flow Resistance and Vascular Reactivity in Spontaneously Hypertensive Rats. Acta physiol. scand. 80: 93-106, 1970. 144 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. Folkow, B. and E. Neil. Circulation. Oxford University Press, New York, 1971. Folkow, B., M. Hallback, et a1. Renal Vascular Resis- tance in Spontaneously Hypertensive Rats. Acta physiol. scand. 83: 96-105, 1971. Folkow, B., M. Hallback, et al. The Effect of Intense Treatment with Hypotensive Drugs on Structural Design of the Resistance Vessels in Spontaneously Hypertensive Rats Acta physiol. scand. 83: 250-252, 1971. Folkow, B., M. Gurevich, et al. The Hemodynamic Conse- quences of Regional Hypotension in Spontaneously Hyper- tensive and Normotensive Rats. Acta physiol. scand. 83: 532-541, 1971. Freis, E.D. Hemodynamics of Hypertension. Physiol. Re- view 40: 27-54, 1960. '.— Ganong, W.F. Review of Medical Physiology. Lange Medi- cal Publications, Los Altos, 1971. Garbus, S.B. Renin and Hypertension Post Grad. Med. 57 (1): 133-138, 1975. Goldberg, M.R., P.D. Joiner, et al. Unusual Venoconstric- tor Effects of Angiotensin II. Proc Soc Exp Biol Med 149: 707-713, 1975. Goldblatt, H., J.R. Kahn, et a1. Studies on Experimental Hypertension IX. The Effect on Blood Pressure of Con- striction of the Abdominal Aorta Above and Below the Site of Origin of Both Main Renal Arteries. J. Exper. Med. 69: 649-671, 1939. ’Aj Goodman, L.S., and A. Gillman, Ed. The Pharmacological Basic of Therapeutics. Macmillan Publishing Co., Inc., New York, 1975. Habib, W.K., E. C. Nanson, et al. The Cause of Hyperten- sion in Coarctation of the Aorta. Annals of Surgery 168: 771-778, 1968. i— Haddy, F.J., et a1. Peripheral Vascular Resistance. Annual Rev. Med. 19: 167-1974, 1968. Hansen, T.R., D.F. Bohr. Hypertension, Transmural Pres- sure, and Vascular Smooth Muscle Response in Rats. Circ Res 36: 590-598, 1975. Hollander, W., J.M. Kramsch, et a1. Arterial Wall Meta- bolism in Experimental Hypertension of Coarctation of the Aorta of Short Duration. J. Clin Inves 47: 1221-1229, 1968. 145 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. Lais, L.T., R.A. Shaffer, et a1. Neurogenic and Humoral Factors Controlling Vascular Resistance in the Spontane- ously Hypertensive Rat. Circ Res 35: 764-773, 1974. Lais, L.T. and M.J.Brody. Mechanism of Vascular Hyper- Responsiveness in the Spontaneously Hypertensive Rat. Circ Res 36-37 (Suppl): 1-216 - 1-222, 1975. Mackiewicz, A., D. Wojczuk, et a1. Plasma Renin Acti- vity in Coarctation of the Aorta Before and After Surgery. Brit. Heart J. 37: 721-725, 1975. Martin, J.B. and M.C. Conrad. Adaptation of Resistance Vessels to Chronic Local Hypotension in the Dog. Micro- vas. Res 7: 170-180, 1974. Morrison, D. Multivariate Statistical Methods. McGraw Hill Pub., New York, 1976. Nachev, P. The Effect of Aortic Coarctation on the Con- centration of Water and Electrolytes in Cardiac Muscle. Proc Soc Exp Biol Med 147: 137-139, 1974. Nolla-Panades, J. Hypertension and Increased Hindlimb Re- activity in Experimental Coarctation of the Aorta. Circ Res 29: 12: 3-9, 1963. Overbeck, H.W., B.T. Swindall, et a1. Experimental Renal Hypertension in Dogs: Forelimb Hemodynamics. Circ Res 29: 51-62, 1971. Overbeck, H.W. Hemodynamics of Early Experimental Renal Hypertension in Dogs: Normal Limb Blood Flow, Elevated Limb Vascular Resistance, and Decreased Venous Compliance. Circ Res 31: 653-663, 1972. Pamnani, M.B. and H.W. Overbeck. Abnormal Ion and Water Composition of Veins and Normotensive Arteries in Coarc- tation Hypertension in Rats. Circ Res 38(5): 375-378, 1976. ' Phelan, E.L. The New Zealand Strain of Rats with Gen- etic Hypertension. NZ Med Journal 67: 334-344, 1968. Phelan, E.L. The Peripheral Resistance of the Blood- Perfused Hindlimbs of Rats with Genetic or Renal Hyper- tension. Proc Univ Otago Med Sch 48(3) 73-75, 1970. Phelan, E.L. and L.C.K. Wong. Sodium, Potassium, and Water in the Tissues of Rats with Genetic Hypertension and Constricted Renal Aftery Hypertension. Clin. Sci. 35: 487-494, 1968. 146 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. Pickering, George. Ribeiro, A.B. and L.R. Krakoff. High Blood Pressure. J. and A. Churchill Ltd., London, 1968. Coarctation of the Aorta. N. 148-150. Eng. J. Med. 295 (3): Rohlf, J.F. and R.R. Sokal. Freeman and Co., San Francisco, 1969. Samanek, M., J. Goetzova, et a1. Angiotensin Blockade in Statistical Tables. W. H. Differences in Muscle Blood Flow in the Upper and Lower Extremities of Patients after Correction of Coarctation of the Aorta. tion 54(3): 377-381, 1976. Sen, 8., R.C. Tarazi, et a1. taneously Hypertensive Rats. Cardiac HypertrOphy in Spon- Circ Res 35: 775-781, 1974. Circula- Sivertsson, R. Hemodynamic Consequences of Vascular Changes in Essential Hypertension. Supp. 343; 3-56, 1970. Acta physiol. scand. Smirk, F.H. Pathogenisis of Essential Hypertension. Brit Med J 791-799, 1949. Sokal, R.R., and F.J. Rohlf. and Practice of Statistics in Biological Research. Freeman and Co., San Francisco, 1969. Timmis, G.C. and S. tension Due to Coarctation of the Aorta. Gordon. Med. 270(16): 814-818. Tobion, L. Jr. and J.T. in Arterial Hypertension. 1952. Binion. Biometry: The Principles W.H. A Renal Factor in Hyper- N. Eng. J. Tissue Cations and Water Circulation 5: 754-758, Tobion, L. Jr. How Soium and the Kidney Relate to the Hypertesnive Arteriole. Fed. Proc. 33: 138-142, 1 974. Villamil, M.F. Angiotensin, Hypertension, and Vascular Ionic Composition. 57-62, 1972. Medicina (B Aires): 32 (Supp. Villamil, M.F. and J. Matloff. Changes in Vascular Ionic Content and Distribution Across Aortic Coarcta- tion in the Dog. Am. J. Physiol. 228(4), 1975. Weiss, L. and M. Hallback. 1): Time Course and Extent of Structural Vascular Adaptation to Regional Hypotension in Adult Spontaneously Hypertensive Rats (SHR) Acta physiol. scand. 91: 365-373, 147 1974. M'CIITI'IGUWHTIWITHW iilNH‘IIVIWVSIW “WW“ 3 1293 03057 6684