men PRESSURE CHANG-ES ANDRENIN I r ' f ' ACTIVETY ASSOCIATED WITH RENAL ARTERIGVENOUS FISTULAE'INJ'HE DOG: V‘; .7 7. Thesis for‘the Degree of Ph. D. MICHIGAN STATE UNIVERSITY ROBERT BRUCE SPANGENBERG 1970 ’1." 3.3 Michigan State University This is to certify that the thesis entitled BLOOD PRESSURE CHANGES AND RENIN ACTIVITY ASSOCIATED WITH RENAL ARTERIOVENOU S FISTULAE IN THE DOG presented by Robert Bruce Spangenberg has been accepted towards fulfillment of the requirements for Ph.D. degree in Physiology Major professor Date November 10L197O 0-7639 BIN‘DING BY HUAE & suNS' 800K smnm me LiIMHY ”NOE“; m ABSTRACT BLOOD PRESSURE CHANGES AND RENIN ACTIVITY ASSOCIATED WITH RENAL ARTERIOVENOUS FISTULAE IN THE DOG BY Robert Bruce Spangenberg In humans, renal arteriovenous (a-v) fistulae have been associated with diastolic hypertension in h8%.of demonstrated antemortem cases, (Maldonado and Sheps, l966). The hemodynamic changes usually associated with a-v fistuiae allow for a small decrease in diastolic blood pressure. The question then arises why diastolic hypertension occurs with a-v fistulae at this particular anatomical site. Renovascuiar anomalies have long been known to cause hypertension. There is much evidence that this renovascular hypertension, to some degree, is mediated through the renin-anglotensin-aldosterone (RAA) system. Any large non-renal a-v fistula might also cause increased RAA act- ivity, but not hypertension. This study was undertaken to determine whether i) a communication between renal artery and vein could be successfully constructed, 2) whether this would produce diastolic hypertension in the dog, 3) whether there would be a rise in renin activity and if so, would there be a quantitatively greater rise than found with non-renal a-v fistulae. Surgical a-v fistulae between renal artery and vein were successfully constructed in ten dogs who were then followed for Robert Bruce Spangenberg -a four week period to observe any change in blood pressure or renin activity from the control period. Diastolic pressure rose an average 10 mm Hg from 95 mm Hg to l05 mm Hg. A greater rise was seen in the first two post-operative weeks with a subsequent lowering to near control values. Renin activity, measured by a rat blood pressure assay method, did not change significantly. The diastolic pressure rise was a small one, and by no means is it certain that the rise represented clinical hypertension. Eight dogs had fistulae constructed between the renal artery and spienic veins. This condition does not raise renal venous pressure as'in the previous series, but does decrease the perfusion pressure and flow to the kidney. There was no rise in blood pres- sure or renin activity. Four dogs had fistulae constructed between the femoral artery and vein. There was a slight but insignificant drOp in diastolic blood pressure and no change in renin activity. A sizable drop in perfusion pressure and flow in the distal arterial segment was demonstrated as well as a large increase in distal segment venous’ pressure. The pressure drop from arterial to venous circuit was greatly reduced. It is assumed that similar hemodynamic changes occur at the site of any a-v fistula. Surgical construction of a-v fistulae proved to be feasible. A small increase in diastolic pressure was seen only'with renal a-v fistulae, however renin activity did not increase nor was it elevated in the other two series of fistulae.» lschemia, necrosis, and loss of renal function were noted. This was more pronounced in the series where venous pressure was elevated. The effect of an Robert Bruce Spangenberg elevated renal venous pressure is implicated. When the renal vein pressure increase was prevented there was no rise in diastolic blood pressure. REFERENCES Maldonado, J. E. and S. G. Sheps. Renal arteriovenous fistula. Postgcgd, fie . #0: 263, l966. BLOOD PRESSURE CHANGES AND RENIN ACTIVITY ASSOCIATED WITH RENAL ARTERIOVENOUS FISTULAE IN THE DOG BY Robert Bruce Spangenberg A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Physiology I970 ACKNOWLEDGEMENTS The author wishes to express his sincere appreciation to his advisor, W. D. Collings, Ph.D., Professor of Physiology at Michigan State University, for his interest, encouragement and invaluable instruction during the course of the graduate program. The author is also indebted to the members of his guidance committee, Drs. L. F. Wblterink, J. B. Scott, J. B. Hook and R. M. Daugherty, Jr. for their advice and consultation. The author's gratitude is also extended to Drs. K. E. Sterner and H. Witt whose technical assistance and selfless interest made this project possible. ii INTRODUCTION 0 o o o o o o o o o o o o o o o o o o o o o 0 TABLE OF CONTENTS LITERATURE REVIEW 0 a o o o o o o o a a o o o o o o o a 0 Cardiovascular Changes Associated With Arteriovenous FIStUIae o o o a a a o o o a o a a a a a o 0 Renal Hypertension Measurement of Renin Activity and Angiotensin s Fistula: A Special Case Renal Arteriovenou Experimental Renal MATERIALS AND METHODS ANImaIS a o o a o a o a o o o o a a o a o A. Conditioning B. Pre-Operatlve care and observations C. Operative procedure: D. Operative procedure: fistula cont E. Operative procedure: Arteriovenous Fistula Renal a-v fistula Renal artery - spienic vein rOIS a o a o a a o a e a a a o a O Femoral a-v fistulae COUtrOIS o a e o o o o a a o a a a a a o o F. Post-operative care and observation . . . . G. Terminal and postmortem observations A. Preparation of plasma (Vallotton and Page) RCHIH ACtIVItY DeterMInatIOn o e a a a a a a o o B. Bioassay - Method of J. Gunneils (l967) . . RESULTS 0 O O o o a a e o o a o a o a a a a a o o a 0 Group i: Renal Artery to Renal Vein Anastomosis Group ii: Renal Artery to Spienic Vein Anastomosis Group lil: DISCUSSION SUMMARY AND CONCLUSIONS REFERENCES CITED 0 O O O O O O O O O O O O O O O O Femoral Artery to Femoral Vein AnaStOMOSIS o a a a o a o o a a a a a a a a 0 iii Page I2 l5 I7 19 20 20 20 20 2| 26 26 27 27 28 28 29 BI 35 ho #5 5] 52 APPENDICES . . . Appendix A Appendix B Appendix C iv Page 56 56 72 75 Table l. 2. 3. 4. 5. 9. IO. II. LIST OF TABLES Causes of renal a-v fistulae . . . . . . . . . . . . Diastolic blood pressure measurements in dogs with renal a-V fIStUIae o a a a o a e o o a a a a a a 0 Summary of renal pathologic features in dogs with renaI a-V fIStUIae o o a o a o a o o o o o o o a a 0 Pre- and post-operative renin activity values (ng percent*) in renal a-v fistulae . . . . . . . . . . Blood pressure measurement in dogs with renal SplenIC a'v fIStUIae o o a o o a a a a a a a a o a 0 Renal and spienic patholOgic features in dogs with renal SPICnIC a-V fIStUIae o o o o o o o o o o o o o Renin activity as ng percent In dogs with renal- SPIOUIC a-V fistulae a a e o a a e a a a a a a a a a Diastolic blood pressure measurements in dogs with femoral a'V fistulae a a a a a a a a a a a o a a a a Renin activity levels in one dog with a femoral a-v fistula expressed ng percent . . . . . . . . . . . . Blood pressure values following the opening of a femoral a-v fistula in mm Hg . . . . . . . . . . . . Blood flow values following the opening of a femoral a-V fIStUIa In mI/mIn o o o o o o a a a a a o o o a Page 18 32 36 37 38 39 40 hi hi Figure I. 2. LIST OF FIGURES Suture arrangement in the ostia which will form the fIStUIa O O O O O O O O O O O O O O O O O O O O 0 Photograph of a renal artery and vein with ostia being sewn together . . . . . . . . . . . . . . . . . Photograph comparing normal dog kidney to necrotic kidney resulting from renal a-v fistula . . . . . . . A typical a-v anastomosis showing direction of blood flow into three segments Page 23 25 3h “3 INTRODUCTION Since Goldblatt's classic experiment in l93h, alterations in renal hemodynamics have been implicated in renal hypertension. Among the known alterations which produce renal hypertension are renal arteriovenous (a-v) fistulae. Although a rare condition, and one . which is seldom diagnosed antemortem, it is associated with a high incidence of hypertension. Indeed, Maldonado and Sheps reviewing the world literature in l966 found an incidence of hypertension of h8%iwhen renal a-v fistula was suspected and subsequently demon- strated. The cause-effect relation in the etiology of this particu- lar kind of renal hypertension has not been established. Many in- vestigators have felt the renin-angiotensin-aldosterone (RAA) system is involved. Much evidence has accumulated over the years to suggest an intrinsic involvement of this hormonal system in other types of renal hypertension, although all the mechanics have not been estab- lished and are the subject of no small controversy. In renal a-v fistulae there are more complex hemodynamic changes than usually seen in other renovascular hypertensive situations. These hemodynamic changes in themselves may in some way account for elevation of blood pressure secondary to renal changes. The hypertension may be mediated 'eitherthrough the RAA system or occur independently. In one sense. this might be considered a paradoxical hypertension inasmuch as a-v fistulae usually cause a mild drop in diastolic pressure. In the presence of an a-v fistula there is a drop in peripheral resistance l 2 and a short circuiting of blood from the arterial to the venous compartment. The decrease in resistance is accompanied by a drop in diastolic blood pressure. This is sensed by the baroreceptors which reflexiy stimulate the heart to increase rate and stroke force. Venous return Increases because of the aforementioned short circuiting plus the action of the reflexiy mediated venoconstriction through sympathico-adrenal discharge. The net result is an increased cardiac output that is reflected in an increased systolic blood pressure and a return of diastolic pressure to near normal although a mild diastolic hypotension often remains. The pulse pressure is usually increased because of an increased stroke volume in the presence of a decreased total peripheral resistance. In the presence of large shunts or weak hearts, congestive heart failure of the high output variety can occur. It should be pointed out that congestive cardiac failure can occur with any aev fistula, renal or otherwise. This appears to be related to the size of the shunt, and obviously in the face of hypertension would be accelerated. When cardiac failure occurs, aldosterone levels are elevated. It is also possible to demonstrate aldosterone eIeVations in experimental a-v fistulae before frank failure has occurred. If it can be presumed that angiotensln is the major, if not the only, stimulant to increased secretion of aldosterone in the presence of a-v fistulae, the question arises as to the fundamental differences between non- renal and renal a-v fistulae. Perhaps in the former angiotensln only stimulates aldosterone secretion whereas in the latter it may also have a hypertension-producing effect. Quantitative differences are one of several considerations, and this possibility initially 3 prompted this study. The experimental canine model required suc- cessful surgical creation of a renal a-v shunt followed by demon- strable hypertension, and although early pilot studies suggested that dogs may react to renal a-v fistulae with elevations in blood pressure, as in humans, a later series of animals has cast considerable coubt on this premise. However, on the basis of the preliminary promising results, the decision was made to study renin activity levels In dogs with renal a-v fistulae as well as other hemodynamic parameters in an attempt to fix a cause for this unusual hypertension. LITERATURE REVIEW i ul h n l i h i nou i t . Since William Hunter first described an arteriovenous fistula in 1757, numerous clinical case reports have appeared in world medical lit- erature. However, it remained until the early i920's before the patho- physiolongwas explored clinically and experimentally. Emile Holman has been a prolific investigator of this anomaly, and in i923 he reported on his findings from 2i cases at the Johns HOpkins Hospital. He noted the characteristic blood pressure variations, viz., an elevated systolic and decreased diastolic pressure with resulting increased pulse pressure - all of which changed toward normal after surgical repair. He also noted the presence of cardiomegaly and the dilatation of the artery proximal to the shunt. He was also aware of an increase in cardiac output and attempted to explain this phenomenon. Lewis and Drury in I923 (cited by Cohen et,a1,, l9h8) had reported that there was no increase in cardiac output. Holman disputed this. He stated there is a necessary rise in venous pressure as an initiating.event. This, when later investigated, proved not to be the case. Still, according to Holman's observations there is an increase In cardiac output. He noted that ”the volume of blood which flows through the heart per unit time is obviously greatly in- creased when short circuiting occurs". He postulated that this causes in- creased myocardial work and hypertrOphy. Later physiologic observations shed more accurate light on the actual mechanisms of increased cardiac out- put and subsequent cardiomegaly. Holman also noted an increase In blood 5 volume based on the observation of low red cell counts which approached normal following surgical repair. His explanation was a Starling effect with the lowered blood pressure causing net reabsorption from the tissue fluids. This was not an unattractive explanation in l923 when methods for body fluid measurements were not devel0ped nor was there knowledge of aldosterone and Its role in producing hypervolemia under circumstances of cardiac compromise. Holman speculated on the nature of the dilatation proximal to the fiStula. He correctly observed that the final limiting factor determining the quantity of blood that will pass through a fistula is the resistance of the fistula itself. He suggested that the artery would dilate until its cross sectional area equals that of the arterio- venous communication. He stated that the dilatation will not occur if the arterial cross section is initially larger than the area of the shunt. In one of the first experimental arteriovenous fistulae, Holman reported that on opening the shunt there was an immediate rise in heart rate from lAB to 22Avwhlle systolic pressure drOpped from 204 to l76 and diastolic pressure dropped from 12h to 80. When the fistula was then quickly compressed, pressure returned to 208/llO. Reid (l925) reported observations on humans and dogs with experi- mental femoral arteriovenous fistulae. He noted the condition is a powerful stimulus to the devel0pment of collateral circulation. He observed that the artery proximal to the shunt becomes dilated and thin walled with degenerative changes while proximal veins dilate with thickening of the wall. He commented on ”grave cardiac disturbances” to the point of sudden death. Reid also noted that circulation in an affected limb distal to the shunt is impaired, but there is an increase in the capillary bed. 6 Reid (l920) had previously reported on local findings in experi- mental femoral and carotid fistulae. The presence of a thrill and a bruit were noted. He also commented on the proximal dilatation of the artery and also observed a distal narrowing which takes considerable time (up to 3 years) to devel0p. Cardiomegaly was incidentally noted, but the author admitted he does not know how this is functionally re- lated to the a-v fistula. Mates (I925) delivered a comprehensive lecture on the clinical findings associated with a-v fistulae. Although he presented little physiologic explanation, he did note that the degree of cardiac decompen- sation and the systemic findings depend on the size of the fistula, the caliber of the involved vessels, the proximity to the heart, the duration, and the volume of the arterial stream. Holman (l92h), following his report on the Johns Hopkins series reported on a series of experimental fistulae to verify and attempt to explain the findings in human patients. He was successful in creating a surgical fistula between the aorta and vena cava. He remarked on the striking thrill and bruit and noted both disappeared when the proximal vein was occluded. Holman's explanation for this is inconclusive. As previously mentioned, on Opening the shunt, heart rate increases and blood preSSUre falls. However, after a period of time the systolic pressure returns to normal although diastolic pressure remains low. The heart rate also slows. Holman suggested that these chronic changes occurred as a result of increase in blood volume. He measured blood volume (method of H00per) in four dogs with femoral a-v fistulae and found an increase after the shunt had been Open long enough to reverse the immediate changes. Holman also cites the increase in blood volume to explain the 7 changes which occur on closing the shunt, namely, a marked slowing of heart rate and increase in both systolic and diastolic blood pressure. Following closure, blood volume returns to normal as do heart rate and blood pressure. Venous pressure was elevated in the proximal segment near the fistula, but was normal in other peripheral veins if the heart was functioning well. In this same series of experiments, Holman re- ported dilation and hypertrophy of the heart and dilatation of the artery and vein proximal to the fistula. Ellis and Weiss (l929) reported on their findings in two patients with an a-v fistula. One had the classic findings of congestive heart failure. The authors were able to manually compress the fistulae and noted a drop in cardiac output with a decrease in heart rate and rise in blodd pressure. They speculated that there was an accumulation of blood in the venous segment followed by increased total blood volume. Local findings were also mentioned, specifically regional arteriolar dilatation, capillary pulsations and an increased skin temperature. _Cohen 33 pl. (l9h8) because of a conflict In the literature on changes in cardiac output In patients with arteriovenous fistulae felt further investigation was needed. Their observations were made on l2 male patients with fistulae resulting from wounds less than 2 years old. Cardiac output was determined by catheterlzation and blood flow In the forearm was determined by Venous occlusion plethysmography. These authors found an increase In cardiac output and the increase had some relation to the size of the fistula. Heart rate was also moderately increased. When the fistulae were closed by manual compression, there was a marked decrease In cardiac output along*with a decreased pulse and an increase“ in systolic blood pressure. Flow through forearms was normal, but in- creased on compression of the fistulae.' When the nerves to the forearm were blocked with procaine, the Increase in flow on compression was re- duced suggesting dilatation.as a cause of Increased flow rather than the rise In mean blood pressure. Van Loo gt, a1. (l949) examined the systemic effects of 2 an,” long femoral a-v shunts In 20 dogs. They, too, felt the problem needed re-exploration because of some conflicts in the early literature. Similar to other reports they noted a rise in cardiac output and heart rate on opening the fistula. There was also an immediate drOp In systolic, diastolic and mean blood pressure. Central venous pres- sure was unchanged. Elkin and warren (l9h7) published the first of several articles concerning their findings in the cardiovascular system in wounded veterans with a-v fistulae at the Ashford Hospital. In this first article, the authors reviewed local diagnostic and systemic signs. In a small group of patients, cardiac output was measured by critically damped low frequency ballistocardiography. 'More than half the patients had higher cardiac outputs than those found after surgical correction. There was generally a small depression of diastolic blood pressure with a wide pulse pressure. A small increase in blood volume was also noted. Warren gg_§l. (l95l) with a larger group of patients at the same facility again reported on cardiac output in patients with traumatic a-v fistulae using ballistocardiography. Predictably, most showed an increased cardiac output: The authors suggest a decrease in cardiac output may indicate a failing heart, i.e., the so called high output heart failure: In this series of A7 patients there was no evidence of 0., I frank congestive heart failure. The blood volume in Al of these patients was measured using Tl82h. Twenty-Three of these patients (56%) following surgical cor- rection showed a decrease in blood volume from preo-operative values of 200 ml or less, a value the authors felt was not significant. How- ever, l8 patients (hh%) showed a decrease of 200 to lOOO ml following surgery. In general, the patients with increased blood volume were those with functionally large fistulae. Nickerson e;_gj, (l95l) In work also done at the Astord Hospital examined venous pressures In six Hogs with experimental femoral a-v fistulae. They found Insignificant (0.5 mm Hg) pressure rises in the right atrium and peripheral veins on opening the shunt. However, venous pressure in the distal segment near the fistula is notably in- creased. Pressure In the artery proximal to the fistula falls on opening the fistula. Epstein 33,31. (l953) observed the effects of a-v fistulae in a variety of sites on renal hemodynamics in l7 Korean war casualties. Clearances of inulin and PAH (creatinine in some cases) were found to be normal. These did not change on closing the fistulae. How- ever, upon closure diastolic blood pressure rose an average of IS mm Hg. Sodium excretion also increased following closure of the fistula. Although an explanation for the increase in sodium ex- cretion*was not forthcoming, In light of present knowledge of aldos- terone this finding is predictable. Frank et_gl, (l955) measured several parameters in dogs with acute artificial shunts through tubing. They found an increased stroke volume and cardiac output In all experiments. When the a-v IO fistula accomodated less than 20%.of the cardiac output, the cardiac output increased by the full amount of fistula flow. In larger shunts, cardiac output did not increase to the full amount of fistula flow which can only mean a decrease in systemic capillary flow. The authors suggest that the factor which limits the degree of increase in cardiac output in the presence of large fistulae is Inadequency of venous return, and the limit to the rate of venous return seems to be the volume of blood available. Consequently, the increase in blood volume seen In chronic a-v fistulae may be compensatory and not an indication of congestive heart failure. The authors feel that in small shunts, the increased venous blood probably comes from blood reservoirs. In large ones, these are exhausted and there is a shift from the periphery. Muenster g3,§l, (i959) reported on six patients, two of whom were In congestive failure. The physiologic findings were the same as previously reported but the additional parameter of arteriovenous oxygen difference was measured. The A-Voz difference was markedly reduced in all cases. Central venous pressure was elevated in the two cases with clinical failure. The authors made the point that heart failure may be produced in the apparently healthy heart by a peripheral circulatory load. Hilton £3.31, (l955) studied the effects of an acute arterio- venous fistula on renal function. Filtration rate was normal or minimally reduced. However, renal blood flow always decreased, pro- ducing an increase in filtration fraction. This suggests a post- glomerular constriction and perhaps minimal afferent constriction. Since mean arterial pressure did not fall, there must have been ll increased renal vascular resistance. Dart gthgl. (l966) studied hemodynamic effects of femoral venous occlusion, before and after a-v shunting. Cardiac output was measured by placing an electromagnetic flowmeter probe on the ascending aorta. The femoral shunt was constructed using a rapidly polymerizing glue, and some were constructed*with sutures. Prior to opening the fistula, venous occlusion caused an average 7% drop In cardiac output. Femoral artery flow decreased by 2#%. There were no changes in blood pressure. When the shunt was Opened, cardiac output increased 3l%. Femoral artery flow increased 92% while mean blood pressure dropped l6 m Hg. Proximal venous pressure was unchanged, (a finding at variance with other investigators) while distal venous pressure increased to l/2 - 2/3 of arterial pressure. When the proximal venous segment was occlud- ed, cardiac output dropped l3%nwhlle femoral artery flow decreased 8#%. Mean femoral artery pressure rose i8%.and proximal and distal venous pressures Increased to near the femoral artery pressure. This was one of the few studies where flow was measured about the fistula. Holman (l952) examined pressure relations at the site of an a-v fistula and noted some findings concerning flow. He stated "the effects of a fistula depend in large measure upon Its size, but even more significantly upon the relationship of its size to the caliber of the vessels in*whlch the fistula lies”. He noted that fistulae of uniform size‘wili pour a greater volume of blood through them, the nearer the heart they lie. This paper deals with experimental femoral and iliac a-v fistulae and observations on flow and size of vessels after one of the three segments, vizs, distal artery, proximal vein and distal vein, Is ligated. He noted that banding the proximal l2 artery to prevent dilatation increased flow through the shunt via the distal artery and collateral bed. Formation of new collaterals always occurred after a-v fistulae were formed. Dye injections Indicated a reversal of flow in the distal arterlel segment through the shunt. This depends on the size of the fistula. Retrograde flow occurs only In large shunts when they are larger than the caliber of the artery. Pressure relations about the fistula in the above circumstance favored retrograde flow In the distal artery. Much of the flow available for this retrograde movement Is assumed to come from collateral circulation, and, indeed, in the case of older shunts, where such collateral vessels had time to deveIOp, the retrograde flow was greater. In a few experiments, the occurrence of a negative pressure In the proximal vein was noted. This was related to the high velocity of flow and was explained on the basis of the Venturi principle. Renal Hypertension The complicity of the kidney in hypertension has been a cone sideration of both clinician and physiélogist since l898 when Tigerstedt and Bergman demonstrated the presence of an extractable renal pressor substance which they named “renin“. However, it wasn't until l93h when Goldblatt with his classic experiments on reduction of renal blood flow with subsequent hypertension In the canine model that investigators began to grasp the potential importance of this excretory organ as an explanation for the etiology and pathophysiology of at least one form of blood pressure elevation. Work carried on simultaneously in Indianapolis by Page and Helmer (l9h0) and Braun-Menendez gt_§j,, in l3 Buenos Aires (I940) indicated that the active principle responsible for vascular pressor activity was not renin‘pe[_§g, but another com- pound named "angiotonln” by the forMer Investigators and ”hypertensin” by the latter. By international agreement, the compound is now known as "angiotensln". Work subsequently done in these two laboratories, as well as in others, demonstrated conclusively that renin, the sub- stance from the kidney, acts as a proteolytic enzyme which catalyzes the formation of a decapeptide known as anglotensinI from a sub- strate usually referred to as angiotensinogen. This substrate appears to originate in the liver and migrates electrophoretically as an “’2 globulin. AngiotensInI, once liberated, is further reduced in the presence of a'converting enzyme to angiotensin II - the active pressor substance. Page and Bumpus (l96l) as well as Braun-Menendez 31,31, (l9h6) have written excellent reviews of the properties, chemi- cal nature, synthesis, fate and activity of the various components of this system beginning with renin and ending with angiotensinlj'. The control of renin release has been recently reviewed by Vander (l967). The review suggests that reduction In perfusion pressure such as occurs in the Goldblatt experiment or in similar situations such as arteriovenous fistula can increase the renal output of renin. Increase in renin output has been the basis of numerous studies to-determine if this event will cause transient or sustained hyper- tension. With the advent of methods for quantitative determination of plasma renin levels, such studies became possible. One thing clearly has evolved from these studies, namely that there is no consistent evidence of an elevated circulating renin level under conditions ‘where hypertension of the renovascular (Goldblatt) type is suspected iA or known. Brown et.a1. studied several cases of surgically proved renal artery stenosis by assaying plasma renin concentration. They found many elevated, many normal and a very few below normal. Fasciolo at 31.. (1961+) studied plasma renin in dogs with uni- lateral clamped renal arteries. In six dogs they observed a rise in blood pressure with a subsequent return to normal. There was no rise In renin activity. When the contralateral kidnewaas later removed, blood pressure again rose, but, there was still no rise in renin. The authors concluded that canine renal hypertension is not caused by elevated circulating renin. Laragh (l962) reviewed the considerable evidence that angiotensin II is a direct stimulant to the adrenal cortical zona glomerulosa to release the salt-retaining hormone aldosterone. Such an interaction inspired considerable research into the possible role of aldosterone in renal or other types of hypertension. Davis £3.21: (l96h) created experimental aortic-caval a-v fistulae below the kidneys In l6 dogs. The dogs all devel0ped clinical signs of congestive heart failure along with the previously described sys- temic signs of a-v fistula. They found a marked hyper-secretion of aldosterone. This prompted the production of secondary aldosteronism In'a second group of dogs by constriction of the thoracic Inferior vena cava. Plasma renin levels were found to be elevated in all cases. This is Indirect evidence that arteriovenous fistulae cause elevated renin levels concurrent with a drOp In mean blood pressure. Davis 53,51, (l962) clamped both renal arteries in is dogs. Eight of these devel0ped a chronic benign hypertension*while the other sevenexhibited a malignant hypertension. The plasma renin l5 content In the former group was two times normal while that In the latter group was ten times normal. This later group also demonstrated a striking aldosteronism. This paper presents evidence In conflict ‘wlth studies showing no rise in renin levels In the presence of classic Goldblatt hypertension. Malrow (l96#) presented his findings in l6 patients with renal vascular lesions and hypertension. He was unable to demonstrate any significant rise In plasma renin activity. Cohen et,a1, (I965) studied nine cases of renovascular hypertension. They stressed the Importance of sampling blood while the patient Is In the upright position. Six of their nine cases had markedly elevated values (93l-3554 ng%) for renin activity. These are a few of the many studies on plasma renin activity in suSpected or proved clinical or experimental renal hypertension. The disparity of results and conclusions is provocative of skepticism and chagrin. Hypotheses are rampant and speculation concerning the actual role of renin In renovascular hypertension should yield a host of additional studies. A common criticism of the studies reported to date ls concerned with methodology for assay of renin activity. Standarization has been loose at best, but with-improved techniques, hopefully, this pitfall will be eliminated. o I I A chemical means to accurately assess circulating levels of renin and angiotensln Il'has been the subject of considerable research in the past 70 years. Braun-Menendez (l9h6) reviews much of the early efforts advanced toward this goal. Until recently, there was no accurate assay i6 for angiotensin II. A measure of renin activity has been available for a number of years. The actual measurement of renin has never been accomplished since this enzyme has never been isolated in pure form. Renin activity implies the generation of angiotensin II by incubation and Is measured by bioassay using a pressor response In an assay animal as the index of activity. Variations on three methods have been most‘widely accepted and used in the past decade. Helmer and Judson (l963) described a method of dialysing heparinlzed blood - a method*which removed other pressor agents but unfortunately dId not completely inactivate angiotensinases. The end product is assayed In a #8 hour nephrectomized pithed cat. Numerous modifications In various laboratories have Improved the validity of this assay. Boucher £1.31, Introduced the use of Dowex. This is a resin *which removes small peptideS‘while excluding other pressor agents and protects the angiotensln from proteolytic enzymes. EDTA was added to the plasma as an anticoagulant and, serendipitously was found to inhibit angiotenslnase. The plasma is usually Incubated at 37°C for exactly three hours. The angiotensin is eluted with diethlyamine, lypOphlezed, reconstituted in alcohol and assayed in the nephrectomized, pentolinium- Ized rat. Again, improvements In various laboratories since Introduction of this method have yielded a better assay. This method has been quite commonly used. Gunnells (i967) has Introduced a bioassay similar to Boucher's with modifications and Improvements. Details are described in the Methods Section elsewhere In this dissertation. l7 The Introduction of the radioimmunoassay has now made It possible to assay directly for angiotensin - a long awaited event. Vallotton and Page (i967) published a method requiring an extraction procedure from plasma‘whereas Grocke (1969) has simplified the technique by assaying from plasma directly. When the procedure is functioning well in the numerous laboratories where renin and angiotensin are being studied, it can be expected that much more consistency in re- sults from similar Investigations will be forthcoming. l tu ° i l 5 Renal arteriovenous fistulae are uncommon pathologic events, but they are of Interest because of the high Incidence of associated hypertension in humans. In l966, Maldonado and Sheps reviewed the world literature and found only 86 reported cases of this anomaly since the first case was described'by Varela in l928. In reviewing the reported cases, these authors found the incidence of diastolic hypertension (pressure more than 90 mm Hg) to be h8%. Fistulae following biopsy were not included since most were characterized by hypertension before the bIOpsy. Forty-six cases were considered. From the 68 cases reported where etiology could be determined, a classification evolved, (Table l). Maldonado, Sheps, Bernart, Deweerd and Harrison (I964) reported a case where radioisotope renography and split renal function studies demonstrated Ischemia distal to the fistula. Pathologic examination of the kidney in this case showed areas of Incomplete infarction In which juxtoglomerular cell counts were high. Tbblan (l966) states that this finding is characteristic of Goldblatt hypertension, and a Goldblatt mechanism l8 has been suggested as the cause of hypertension In some cases of renal a-v fistulae. Three additional cases of renal arteriovenous fistula have been reported since the above review. PapadOpoulous et,§1, (l967) reports on a 2A year old male with diastolic hypertension which was success- fully treated surgically. Sharif et a1, (l967) reported a case with borderline hypertension (l60/lOO) In a #8 year old female. Following surgery, blood pressure dropped to l30/90. These authors feel the hypertension results from an ischemlc kidney, l.e., a Goldblatt hypertension. Ditchek gt_§1, (l959) diagnosed and surgically corrected a fistula in a 50 year old female who was not hyper- tensive. TABLE i CAUSES OF RENAL A-V FISTULAE Category Number Per Cent of 68 Congenital 8 Idiopathic 23 3h Acquired hypernephronma l0 l5 trauma penetrating 9 I3 nonpenetrating 2 3 nephrolithotomy 2 3 Partial nephrectomy 2 3 Percutaneous biospy l6 24 Inflammation 3 h atherosclerosis _j_ _1_ TOTAL 68 IOO l9 Experimental Renal Arterioggnous fistula Lasher and Glen in I939 reported their findings from a series of l2 dogs In which theyihad surgically created renal a-v fistulae. The authors did not statistically analyze their data, but there was no apparent rise in blood pressure until a contralateral renal a-v fistula was created. Following this second procedure pressure rose only slightly and the dogs soon died of uremia. The authors point out that one major difference betWeen the Goldblatt experiment and their own is that in the Goldblatt studies there is no rise In venous pressure whereas In their own, there Is. They suggest that. the elevated venous pressure may in part be responsible for the altered physloi09y but propose no mechanism. In I969, Secrest reported the results of his Investigation of the effects of a renal a~v fistula on renal function and blood pressure. They found that after operation, there was a slight con- sistent widening of pulse pressure but no elevation of diastolic pressure. There was no rise In renin activity. One dog with a femoral a-v fistula was used as a control with no post-Operative change In blood pressure or renin activity. MATERIALS AND METHODS l. Animals. A. Conditioning. Fifty-five to eighty lb dogs of either sex were selected for ease of handling. They were vaccinated against canine distemper, leptospiro- sis and hepatitis; wormed with standard veterinary antihelminthics; examined for general good health and checked for anemia. The dogs were observed for a two week period while on standard dry dog chow and weight changes were noted. Daily the animals were brought to a quiet laboratory where they were trained to lie quietly on a table for blood pressure determination. The pressure was obtained by femoral arterial puncture with a 2i gauge needle attached to a Statham P236 pressure transducer and was recorded on a Gilson Ink-writing recorder (Model CH-CPBB, Gilson Medical Electronics, Middleton, Wisconsin). Every dog had at least two pressures In the acceptable range before being operated. Dogs whose resting pressure would not fall below l60/ll0 mm Hg were not used In the study. B. Pro-operative care and observations. Blood pressure was checked on the day of surgery. A pre- anesthetic l0 ml venous blood sample was drawn for renin activity determination. The dogs were Induced with sodium thyamylal (Parke Davis - Detroit, Michigan) 1:250, l ml/5 lbs body weight. They were intubated and maintained on methoxyflurane (Pitman Moore) gas anestheSia with an oxygen flow'of 350 ml/min. Respiratory assistance was not 20 2i required. They were closely shaved and the skin In the operative area was scrubbed with Septlsol (Vestal Laboratories - St. Louis, Mo.) fol- lowed by alcohol rinsing. Sterile procedure was followed throughout with standard draping. ' C. Operative procedure: Renal a-v fistula. A surgical incision was made along the left subcostal area from the lateral border of the paraspinous muscles to a point halfway to the venteromedial line. A muscle-splitting grid Incision was then made down to the peritoneum. The kidney was identified and freed from surrounding fascia. It was then retracted ventrally through the in- cision to expose the renal vessels. The kidney was wrapped In a warm sponge moistened with saline while the vessels were coated with warm parrafin oil. In some cases renal venous samples were taken for renin activity determination. The vessels were stripped of all fat and fascia for a 2 cm distance where possible. Carefully noting the time, the artery and vein were approximated and clamped proximally and distally with small serrafine clamps. A l cm longitudinal Incision was then made in both artery and vein so the lengths coincided. A stay suture of 5-0 silk was placed midway in the superior lip of each Incision to allow for retraction and visibility of the lumena. The ends of each Incision were then approximated, artery to vein, and tied with 6-0 mersilene suture. While retracting on the stay sutures the bottom lips were sewn together intraluminally with a continuous stitch. Upon reaching the opposite end of the incisions, the running stitch was tied to the previously-placed approximating suture. The superior lips of the Incisions were then sewn, stay sutures were removed and top-running stitch tied at the end where bottom suturing began (Figures l and 2). 22 Figure I. Suture arrangement in the ostia which will form the fistula. .————-—"—\ 23 {Renal artery ”stay suture 24 .cosuomou czom mc_oa o_umo no.3 c_o> ecu >couco .mcoc o to geocm0u05e .N mesa—m 25 26 Gelfoam (Upjohn Co., Kalamazoo, Michigan) was placed beneath and atop the sutured vessels and clamps were removed. Bleeding greater than lO-20 ml rarely occurred and was controlled in all cases. A palpable thrill and audible bruit were immediately apparent and test- Ified to the patency of the shunt. Duration of vessel occlusion for surgery ranged from 25 - #0 min. After lO minutes the kidney was replaced In the retroperitoneal fossa and wound layers were closed with h-O cotton. By completion of suturing, animals were showing signs of recovery from anesthesia. Two ml of Penstrep (Merck and Co. Inc., Rahway, N.J.) were administered I.m. and the animal returned to his cage. D. Operative procedure: Renal artery - spienic vein fistula controls. The procedure was carried out as In C to the point of strip- ping the renal artery. The renal vein was left intact. The animal's spleen was delivered onto the field and major vessels were identified. Considerable time was required to free-up and strip a suitable length (usually h-S cm) of vein. This was then approximated along side the renal artery and anastomosis was carried out as in C. Because of greater technical difficulty, clamping times ranged from #0 - 60 minutes. The procedure was completed as In C above. E. Operative procedure: Femoral a-v fistulae controls. An Incision was made through the femoral triangle to expose the femoral artery and vein. A l cm anastomosis was then carried out In the manner described in C taking care to allow sufficient dis- tance on both the proximal and distal limbs for electromagnetic flow- meter probe (Model 300, Carolina Medical Electronics, Inc., Winston Salem, N.C.) placement. Clamping time ranged from 20-25 minutes. 27 F0llowing hemostasis, flows were determined In the proximal and distal limbs of both artery and vein.’ (This had been attempted In some of the renal shunts, but the anatomy precluded obtaining satisfactory results). Pressures were then obtained In all four limbs with a 23.9 needle at- tached to a Statham P23G:pressure transducer and recorded on a Gilson ink-writer recorder. Where low pressures were anticipated, a Statham P23A low pressure transducer was used. The incision was closed with h-O cotton. From this point, procedure was as in Cyabove. F. Post-operative care and observation. There was no operative morbidity or wound Infection. All dogs were ambulatory and were observed to urinate on the first post- operative day.‘ Two dogs developed nonrelated Illnesses (probably viral) and were dropped from the study. Each dog's pressure was checked once or twice weekly for a total of eight times. Venous blood samples were drawn at the first sign of elevated blood pres- sure, and all dogs were sampled terminally whether hypertensive or not. Careful auscultation for flank bruit was done at time of pressure de- termination as an index of shunt patency. G. Terminal and postmortem observations. After A - 5 weeks of pressure determinations, the dogs were anesthetized and surgical wounds reopened. Kidneys and vessels were Inspected: The dogs were euthanized and the kidneys and vessels re- tained for gross and histopathological examination. 28 ll.' Renin Activity Determination. A. Preparation of plasma (Vallotton and Page). Ten ml of whole venous blood were withdrawn into a plastic tube containing 0.9 ml of 3.8% citrate solution and immediately chilled to 49C. The blood was then centrifuged at 4°C. for ten minutes at 2500 rpm. The plasma was Immediately withdrawn and frozen until prep- aration for assay. The plasma was just thawed to a point about 4°C. and maintained In an ice bath. The pH was adjusted to 5.5 with l N HCl. Following this, 3 drops of a 5% solution of diIsoprOpylflurOphosphate (DFP) ln isopropyl alcohol were added to eliminate the activity of angiotensinase.’ The samples were then incubated for exactly 3 hours at 37°C: and again placed in an Ice bath to StOp all reaction.‘ Two ml of the plasma were mixed with 2‘ml of a lO% suspension of Dowex 50 W‘XZ resin (Dow Chemical - Midland, Michigan) in .2M ammonium acetate at pH 6.0. The resin was prepared for use by cycling first through water, then .6N HCL, another water wash followed by 4N NaOH, then rewashlng and finally suspending In ammonium acetate at pH 6.0. This resin Is designed to pick up small polypeptides such as angiotensin II. The mixture was placed In a vitamin assay tube (Column) with a filter disc punched from standard filter paper. The resin was allowed to drain, then washed with 6xml of .2N ammonium acetate followed by 6 ml distilled water. These washings were discarded. The peptide (angio- tensin II) was then eluted from the resin with l ml of diethyiamine (Sigma Chemical, St. Louis, Mo.) followed by l ml of .2N NHu OH. .The eluant was then lyophylized and saved until time of assay'whereupon It was reconstituted with l ml of lM trhs.buffer pH 7.5 tO‘which had been added IOO mg of lysozyme (Sigma Chemical, St. Louis, M0.) for each IOO ml 29 of buffer. B. Bioassay - Method of J. Gunnells (l967). Male Sprague-Dawley rats weighing ISO-200 gm were bilaterally nephrectomized under ether anaesthesia l8-22 hours prior to assay. They were permitted neither food nor water during this interval. At the time of bioassay, the rats were anesthetized*with sodium pento- barbital, 30 mg/kg injected intraperitoneally. Anaesthesia was main- tained at a constant level by subcutaneous injectiOn of sodium pento- barbital, 50 mg/ml through a polyethylene catheter at Intervals of six minutes. The agent was delivered from a microsyringe at a constant volume of 2 ul. Ganglionic blockade was achieved by In- jection of l5-6O ul of pentolinium tartrate (Wyeth, Philadelphia, Pa.) (l0 ug/ml). The trachea was cannulated with PE 240 tubing. A carotid artery was cannulated*with PE 50 polyethylene tubing for determination of blood pressure. This was connected to a Sanborn low pressure trans- ducer (Sanborn, Boston, Mass.) and the tracing monitored on a Sanborn recorder (Series l7OOH.P., Sanborn, Boston, Mass.). The apposite jugu- lar vein and a femoral vein were Isolated. These were cannulated with PE l0 tubing, one being used for known quantity standards of angioten- sinll', the other for the unknown plasma samples. All tubes were filled with saline containing heparin (National Biochemicals, Cleveland, Ohio) (l0 units/ml). The commercial valine-5 angiotensinll' (Ciba, Summit, N.J.) was prepared daily In a dose of O.l ng/ul from a stock solution containing l0 ug/ml in saline prepared weekly. This was Introduced into the assay animal In Increments of 2 ul up to 10 ul to establish a stand- ard curve. The unknown plasma was injected In aliquots of 20 ul up to IOO ul depending on the degree of response. The recorder was adjusted 30 so that l mm Hg rise In pressure equaled at least a 4 mm rise of the recording sylus. A rise of 6 mm to an Injection of 2 ul of standard ‘was required to accept the animal for assay. The unknown samples were bracketed with known standards in doses to cause pressure to be recorded just above and below that observed for the unknown. The animal's blood pressure was maintained at 60-80 mm Hg with repeated doses of pentolinium tartrate. RESULTS 0 ° n r i n i . W. The presence of an. arteriovenous fistula between the renal artery and vein resulted In a significant post- operative rise in diastolic blood pressure, but not Into a range generally considered to be hypertensive. Ten dogs whose blood pressures were checked twice'weekly for four weeks showed an average rise of IO mm Hg. (I, - 95.0, N} I l05.6, S.E. I l.45, N - l0, (P<:.OOI).* FUrthermore, much of the rise occurred in the first two’ weeks after Opening the shunt with a subsequent return to near control values in most cases (Table 2). When a shunt sealed off or a thrombus occurred, such a return to normal might be predicted. The disappearance of the bruit In the flank region is the Index that such an event may have occurred. This situationwwas noted In animal I9B where the fistula spontaneously closed. A thrombus was present in animals 22B and 303; and, In the latter case, It Inter- ferred with the drainage of the left spermatic vein sufficiently to cause a large varicocoele. Although statistically highly signi- ficant, the average rise In blood pressure from a clinical standpoint was not impressive. *;l I Mean for controls, i2 - Mean for experimentals, S.E. I Stand- ard error of the mean difference, N I Sample size. 3i 32 Renal Pathology. Prior to sacrificing the animals, cannulation of the left ureter yielded insufficient urine flow to perform renal function studies. Postmortem examination revealed normal kidneys In 3 cases while the rest were shrunken or contracted to a fraction of the size of the contralateral kidney (Figure 3). Histopathologic changes (Table 3) varied from minimal to complete coagulation necrosis. Where there was sufficient Intact anatomy for assessment, tabular hyaline deposits and connective tissue Infiltration were common features. The complete pathology reports are in Appendiva. TABLE 2 DIASTOLIC BLOOD PRESSURE MEASUREMENTS IN DOGS WITH RENAL A-V FISTULAE . Post-Op. IKElJI__EEE:92a__l§£_£2§§2921__2fl§__3£§: 4th 95thésht_IUdL_£fldL_Jhuuzuuz 193 105 av 130 120 155 105 110 115 115 120 121 223 90 95 110 125 95 90 105 85 95 100 253 105 130 125 135 130 115' 110 105 121 273 100 115 115 115 110 100 110 100 120 111 283 90 95 100 115 115 95 100 90 85 99 303 85 130 105 85 90 85 100 95 100 99 323 95 95 120 95 105 100 100 90 100 101 353 100 105 95 90 100 105 110 95 100 101 #03 80 85 95 100 95 95 100 100 100 96 2c 100 110 1gp 115. 110 110 (35’ 100 105 107 Avg. 95.0 .1094 111_ 113 .106. 101 104 98 103 _105.6 53 .m_:pm_m >um —mcoc 50cm mc_u_:moc >oce_x u_u0cooc Op >oce_x mow .mEcoc mc_coaeoo camcmouo;m .m oczmwu 54 35 Benin Agtiyity Lgygls. The control values of renin activity for all peripheral venous samples were pooled and compared with the pooled post-operative values in each of the three treatment groups, renal a-v, spienic renal a-v, femoral a-v. Over the period of sampling following surgical creation of the shunt, there was no significant rise in renin activity. (i; I 5.70, N - 7, I} . 5.73, N I I6, S.E. I .4l3, values taken from Table 4 and converted to natural logarithms). Values in nanograms percent are recorded in Table 4. Group II. Rengl Artery to Spienic Vein Anastomosis. Diastolic Blood Pressure. Whenean a-v fistula was created between the renal artery and splenic vein, no significant rise In diastolic blood pressure could be found In the post-Operative period. (i1 . l04, ii I l07, S.E. - 3.22, N - 8). Values are presented In Table 5. Renal and Spienic Pathology. Histopathologic features are recorded In Table 6. In several cases, urine output on the shunted side was apparent, but quantity was too small for accurate renal function studies. Generally, the kidneys, In gross examination, appeared to be healthier than In the renal a‘v shunt series. The spleens all showed signs of congestion secondary to the elevation in venous pressure. Complete pathology reports are in Appendix A. Renin Activity Levels. The statistical analysis of data on renin activity*was handled as In the preceeding section. Details are In Appendix 8. Again, there was no significant rise In post- operative renin activity. (Y1 . 5.70, N - 7, 72 - 6.lO, N I 4, S.E. - .44l converted to natural logarithms). Values In Ng per- cent are recorded in Table 7. 56 u n u + ocoz .mEcoz mum m_ommu> Lm_~: cue—aeou c. .aEOccu o o o o co_um_=mmoo .uouuatucou mom ..,I.W ova—aeoo o + + + co_um_=mmou wouumtucou mmu o + + O 0H0 — 9:00 co_um—=mm00 vouomcucoQ . mmu u u + + covumpammou vouomcucoo mmm .> —mcoc :— massacgu .voumcav:_ + + + + co_um—3mmoQ .vouumcucoo mum co_umco _—ocm c_coxo_ch c_ccoum co mmflqmz: m_m0cooz mm o o Lm_:omm>0cu_z Lm_:mamu pawn—k o>_uuoccoQ cm_:n:h concede; «a: u nv.ucomam I I .ucomoca I.+ u<4=hm_m >u< 4¢