Renovascular hypertension: Predicting surgical cure with exercise renography Thomas H u p p , M D , John H. Clorius, M D , and Jens R. Allenberg, M D ,
Heidelberg, Germany Renal artery stenosis with resultant renovascular hypertension has attracted clinical attention because the disease is potentially curable and because numerous diagnostic and therapeutic modalities compete for clinical acceptance. An exercise-mediated disturbance of renal hippurate transport was recently described, and has been implicated as having a role in nephrogenic fixed hypertension. To predict the final course of renovascular hypertension before operation we carried out a prospective study with the goal of verifying the predictive value of exercise hippurate scintigraphy. The study was to test the hypothesis that patients with disturbance of renal hippurate transport (pathologic renogram) induced by exercise would have stabilized hypertension and would continue to be hypertensive after operation. Thirty-one patients with hypertension who had unilateral or bilateral renovasoalar stenosis documented on angiography were referred to rest and exercise hippurate scintigrams before operation. The results of the examinations at rest served as standard and were compared with the exercise scintigrams. In 19 of the 31 (61%) patients a disntrbance oftransrenal hippurate transport evolved during exercise, whereas 12 (39%) patients failed to respond to exercise with altered hippurate kinetics. Twenty-six patients went on to renovascular operations; five had percutaneous transluminal angioplasty. Revascularization results differed markedly when the blood pressure response of patients with positive results on exercise (abnormal) and patients with negative results on exercise (normal) were compared. Ten of 12 patients with hypertension who had normal exercise renograms were cured. In comparison, blood pressure values were tittle influenced by therapy in patients with an abnormal response, where 17 of 19 patients continued to have hypertensive disease after therapy. It is most important to note that the transitory tracer transport abnormality is easily recognized on scintigraphy. The study shows that exercise scintigraphy helps identify patients with curable renovascular hypertension and thus predicts which patients will benefit most from operation. (J Vasc S t ~ 1991;14:200-7.)
Renovascular hypertension has attracted extensive clinical attention because of the presence of physiologic models, its epidemiology, and the numerous diagnostic and therapeutic modalities competing for clinical acceptance. This extensive clinical interest is directed to a mere subgroup of patients with hypertension. Renovascular hypertension is found in only 1% to 6% of all patients with high blood pressure (BP). 1-6 Nevertheless, 60 to 360 thousand patients in Germany alone require therapy. The interest in renovascular hypertension also increased after the introduction of percutaneous transFrom the Department of Surgery, Division of Vascular Surgery (Drs. Hupp and AUenberg),Universityof Heidelberg and the German Cancer Research Center (Dr. Clorius), Heidelberg. Reprint requests: Thomas Hupp, MD, Division of Vascular Surgery,Departmentof Surgery,the Universityof Heidelberg, Im NeuenheimerFeld 110, D-6900 Heidelberg, Germany. 24/1/29810 200
luminal angioplasty (PTA), which greatly expanded the available therapeutic options. It would obviously be desirable to identify those patients who would benefit from PTA or surgical revascularization. At present it is not possible to securely predict BP response before PTA or operation. Only 60% to 80% of treated patients respond with cure or improvement of BP values after operation. Thus PTA, or surgical revascularization, fails to improve the hypertension of approximately 30% of treated patients. Iodine-123 (1-123) or 1-131 hippurate has transrenal kinetics comparable to paraaminohippurate. 7 The scintigraphic visualization of the tracer therefore contains information on renal cortical blood flow? It was recently recognized that ergometric exercise can provoke a disturbance oftransrenal hippurate transport. 9 The transport abnormality normalizes when exercise is terminated. This transitory and abnormal, orthoiodohippurate (1-131)
Volume 14 Number 2 August 1991
Exercise renography to predict surgical cure of renovascular hypertension
transport is considered indicative of disturbance of renal cortical perfusion. A posture or exercisemediated disturbance of renal hippurate transport has been documented only in patients with hypertension. The present study included patients with hypertension referred for surgical therapy of renovascular lesions. During a presurgical scintigraphic examination we identified those patients in whom the exercise-induced transrenal hippurate transport abnormality developed. The hypothesis was tested that those patients with hypertension- and an exerciseinduced hippurate transport abnormality had a form of renal hypertension not curable by operation or PTA. Revascularization could serve to preserve organ function and possible progression of the arteriosclerotic stenosis, but it would fail to cure hypertensive disease. In short, we sought to predict which paticnts would respond to surgical therapy with BP normalization. METHOD
One hundred fifty-one renal vascular reconstructions were required by 117 patients referred to this center during a 7-year period lasting from January 1983 to January 1990. We sought to include those patients in the prospective blind study who met the following criteria: unilateral or bilateral renal artery stenosis with 75% or greater lumen reduction verified on angiography, hypertensive BP values (Table I), and found suitable for surgical treatment. The patients had to be capable of ergometric exercise and show willingness to participate in the examinations. The mean BP value of the patients with an abnormal exercise scintigram was 186/105 mm Hg, whereas those patients with a scintigram not influenced by exercise had a mean BP value of 170/98 mm Hg, both values obtained with the patients under drug therapy (Tables I and II). Forty-two of the 117 patients were identified as meeting the selection criteria, and were referred to the scintigraphic examinations. Nine of these patients were excluded from final analysis because of incomplete scintigraphic data, after interruption of exercise when systolic BP values rose to 250 mm Hg, or diastolic pressure exceeded 125 mm Hg, or when exercise was terminated because of leg pain resulting from peripheral arterial vascular disease. It was thus possible to obtain a complete scintigraphic examination in 33 patients. The scintigraphic examinations were evaluated by one observer (I.H.C.), who was blinded with respect to the clinical data. Exercise scintigrams were compared with an examination at rest and were classified as normal or abnormal.
201
The final dinical evaluation included documentation of preoperative and postoperative antihypertensive medication, BP values both before and after surgery, and the compa:rison ofangiographic findings before and after therapy. Two of the 33 patients were dropped from the study when the surgical outcome was compromised by postoperative complications, so that 31 patients remained in the study. Postoperative data were collected at the time of discharge and during a follow-up examination scheduled on average 20.5 months after operation. The earliest follow-up examination was after 6 months, the latest after 78 months. The final clinical examination was generally made at this hospital's vascular-disease outpatient service. Pertinent information on the BP status and antihypertensive drugs was occasionally obtained from the family physician when patients failed to appear for follow-up examination. The clinical data were then evaluated by a surgeon (T.H.) who had no immediate access to the preoperative scintigraphic data. After collection and evaluation of the clinical data, we compared the final postoperative BP status with the preoperative scintigrams. To classify BP development we used slightly modified criteria established by the U.S. cooperative trial of surgical treatment of renovascular hypertension. s Hypertension was considered cured if the diastolic BP remained below 90 mm Hg without antihypertensive medication. Surgery was judged to have improved the disease when the diastolic BP was reduced by at least 15% with medication. Nonresponders were defined as those patients with hypertension who had a BP reduction of less than 15% or who required increasing doses of antihypertensive medication or a combination of three or more drugs. Each subject had renography in a prone position and in the upright (sitting) position during ergometric exercise. Both examinations were obtained within 1 week, generally within 72 hours. To obtain adequate hydration, patients were asked to drink 400 ml of fluid during the 30 minutes before renography. All patients received careful oral instructions about the examinations. The renogram in the prone position was begtm within minutes after patient positioning. The exercise protocol was used to document the influence of exercise on hippurate kinetics. Patient comfort was emphasized during the test. The workload was considered inadequate if it could be continued indefinitely and too strenuous if it resulted in cxhaustion. Patients sitting in front of the gamma camera on a bicycle ergometcr were asked to sit straight-backed so that the kidney-to-camera distance was kept small. Pulse and BP were noted during
Journal of. VASCULAR SURGERY
202 Hz~pp, Clorius, and Allenberg
Table I. Comparison of the BP-status of patients with renal artery stenosis ~stopera~e
Preoperative Patient No.
Stenosis
Normal renogramm 1 U 2 U 3 U 4 U 5 U 6 U 7 U 8
U
9
U BI 11 U 12 U Abnormal renogramm 13 U 14 BI 15 U 16 U 17 BI BI 18 U i9 U 20 BI 21 U 22 BI 23 BI 24 U 25 U 26 U 27 BI 28 U 29 U 30 U 31 10
BP (ram Hg)
No, of drugs
BP (mm Hg) at discharge
BP (ram Hg) follow-up
Month postop.
No. of drugs
180/100 160/85 200/120 170/110 170/100 160/80 170/100 170/90 150/100 155/90 180/110 175/100
2 2 3 0 0 3 3 4 0 4 4 5
130/75 130/80 130/90 150/80 150/100 130/85 120/70 150/80 140/80 150/70 140/85 150/70
100/60 160/80 140/90 150/95 135/70 160/90 130/80 150/80 140/80 140/80 140/80
31 13 39 6 8 6 7 78 9 36 16 8
0 0 3 0 0 0 0 0 0 0 2 0
180/110 240/140 200/120 240/100 180/90 200/110 120/80 120/65 220/120 200/130 210/140 190/100 120/80 200/100 195/110 200/100 205/I00 140/105 190/110
2 3 3 1 0 5 2 4 3 1 2 I0 1 6 3 5 3 6 2
130/90 160/80 140/80 140/70 140/80 160/70 150/90 130/80 160/100 200/100 140/80 170/90 140/80 165/85 160/95 160/90 180/90 160/100 180/95
145/75 160/80 130/80 190/90 150/90 170/90 I35/95 145/85 140/80 190/100 140/90 i90/90 180/90 150/80 150/100 160/90 180/80 I60/100 180/105
28 7 25 10 7 6 41 6 7 52 20 30 48 62 32 18 78 9 9
3 1 2 1 0 3 1 3 2 3 0 3 1 1 1 2 1 6 3
150/80
U, Unilateral stenosis; BI, bilateral stenosis; BP, blood pressure.
Table II. Exercise renography in 31 patients with hypertension was used to group patients and predict curabifity of disease; comparison of BP values and the mean number of drugs required before and after revascularization therapy Scintigraphy Hippurate transport Tracerin disturbed bladder (rain) Prone Normal scintigram (n =
12)
Abnormal scinfigram (n = 19)
Preoperative
Postoperative
Exercise Prone Exercise Mean BP (ram Hg) No of drug (2) Mean BP (ram Hg)
No of drags (~)
0
0
3.3
3.7
170/98
2.50
142/88
0.5 (10 took more)
0
19
3.3
8.0
I86/105
3.26
160/89
2.0 (2 took none)
exercise and were monitored at 5-minute intervals. Ergometric resistance was set at 60 watts for women and 80 watts for men. After 1 to 2 minutes of exercise the pulse rate change in response to exercise was noted. Renography was begun only after the pulse rate had increased at least 20 beats/rain. After
radiotracer injection, the patients continued to exercise. Ergometric resistance was adjusted according to the whishes of the patients. Radionuclide renography was begun after intravenous injection of either 7 t*Ci (I-131) or 6 b~Ci (I-123) o-iodohippurate per kilogram body weight.
Volume 14 Number 2 August 1991
Exercise renography to predict surgical cure of renovascular hypertension
A 15-inch gamma c:maera equipped with a generalpurpose medium-energy (360 keV) parallel-hole collimator was used for all studies, with a window setting at 20%, centered over the photopeak of the tracer. One-minute images were obtained, beginning with the injection, and continued at 1, 2, 3, 4, 7, 9, 14, and 19 minutes. To identify the appearance of the tracer in the bladder, which was recognized on the sequential scans, we extended the uninterrupted 1-minute image sequence past the fourth minute when required. The ,examination was terminated after 20 minutes. Data were stored on magnetic tape and were analyzed by minicomputer. Regions of interest were placed over each kidney to determine single-kidney function. No attempt was made to exclude the renal pelvic system. Background regions of interest were placed automatically around each kidney by use ofa 2-pixe]Lwidth. Single-kidney hippurate uptake, expressed as a percentage of total uptake of both kidneys, was determined. Uptake was taken to be proportional to the gradient of the renogram between 24 and 120 seconds. The third curve segment of the renogram was analyzed by inspection only. Results of prone and exercise renography and serial scintigrams were compared. RESULTS
Thirty-one of 117 patients with renovascular hypertension were included in the final evaluation. Twenty-three of 31 patients with hypertension had a unilateral stenosis, eight had bilateral disease. Treatment was as follows: 26 of the 31 patients had revascularization surgery, 7 of them bilaterally (Table III). In five patients the vascular stenosis was able to be treated with PTA, once bilaterally. Surgical intervention was required in 26 patients with 32 reconstructions. Included were 25 aortorenal bypass operations, 6 renal an:ery reimplantations after resection of the stenosed segment, and 1 renal artery endarterectomy. Function impairment of a poststenotic small kidney required a nephrectomy in one patient. Eleven of the 26 patients operated on also required surgical intervention distant from the treated renal vasculamre (simultaneous aortoiliac reconstruction). Five needed a resection of an abdominal aneurysm of the aorta, six had a Y-bifurcation bypass for obliterative arteriosclerosis, and two had a nephrectomy of a small kidney contralateral to the revascularization procedure. All patients had preoperative rest and exercise renography (Fig. 1). A disturbance of bilateral hippurate transport developed in 19 of 31 (61%) patients in response to exercise (abnormal scintigram) (Table II). This resulted in delayed excretion
203
Table III. Thirty-one patients with hypertension were included in the study; 26 had surgical revascularization, 5 had transluminal renal angioplasty (PTA) Patients
26
No. ofprocedures
33 ~ ~ 6
Surgicalprocedure S Dacron bypass Vein bypass Reimplantation Endarteriectomy Nephrectomy PTA
(No.) (16) (9) (6) (1) (1)
ofhippurate from the renal tissue. On average, tracer appeared in the bladder after 3.3 minutes when scintigraphy was performed with the patient at rest, and after 8 minutes during ergometric exercise (Table II). The hippurate transport abnormality was particularly severe in three patients in whom we noted the first radiotracer accumulation in the bladder after 19.5 minutes, 25 minutes, and 46 minutes after intravenous injection of the activity. A 20-minute value was used to calculate the mean appearance time of the tracer in the bladder, to avoid having these highly, abnormal results magnify the extent of the transport disturbance. Twelve of 31 patients with hypertension (39%) failed to respond to exercise with a bilateral hippurate transport abnormality (normal scintigram). Exercise resulted in a slight delay of radiohippurate excretion into the bladder. At rest the average appearance time of hippuram in the bladder was after 3.3 minutes, during exercise it was after 3.7 minutes (Table II). Normal as well as abnormal exercise scintigrams were observed in patients with balanced as well as with dissimilar single kidney function. Patients with hypertension who had an abnormal exercise scintigram had higher mean BP values before revascularization than those whose renogram failed to be influenced by exercise (Table II). The results achieved with surgical revascularization (n = 25), nephrectomy (n = 1), and renal artery dilation (n = 5) were clearly different when patients with exercise positive and exercise negative scintigrams were compared (Table IV). Ten of 12 patients with hypertension who had a normal exercise renogram were cured. Therapy improved the BP values of the other two patients (No. 3, No. 11) (Table I) even though they continued to require antihypertensive drugs. In comparison, systemic BP values were little changed by therapy in patients with abnormal exercise renograms (Table IV). Hypertension was cured in two out of the 19 patients. Both had bilateral vascular lesions. One of these patients (No. 17)
Iournal of VASCULAR SURGERY
204 Hupp, Clorius, and Allenberg
M~N
MtN
Fig. 1. Sequentialscintigramsofa60-year-oldfemalepatient, obtained with the patient in the prone position at rest and during exercise on a bicycle ergometer while sitting upright. The patient had a vascular stenosis of the right kidney. The examination at rest documents the reduction of single kidney function on the right side, as well as timely tracer excretion into the renal pelvis and bladder (fourth minute), and a regular tissue clearance of the tracer. The abnormal exercise renogram demonstrates a bilateral transrenal hippurate transport disturbance identified by the elevated excretory segment of the renogram, by hippurate retention in the renal tissue, and delayed excretion ofhippurate into the bladder (twentieth minute). Note: unilateral vascular disease is associated with a bilateral function abnormality during exercise.
(Table I) with bilateral stenosis demonstrated compromise of unilateral left kidney function (left 30%, right 70%). Exercise resulted in a bilateral disturbance of hippurate transit, with delayed excretion of activity into the bladder. Whereas the radiotracer was excreted in the fourth minute while the patient was at rest, it appeared in the bladder during the eighth minute during exercise. The second patient's (No. 23) (Table I) left kidney had less than 5% of total function after vascular occlusion after PTA, whereas the right organ had a documented, but functionally silent, vascular stenosis. Accordingly, the left kidney was barely visualized during scintigraphy. Exercise induced a hippurate transport disturbance in the right kidney. Bladder activity was noted after 4 minutes during rest, and after 7 minutes during exercise. The i9 patients with hypertension who had abnormal exercise renograms included two who were cured, eight patients with improved BP values, and nine who failed to respond to therapy (Table IV). DISCUSSION Those who suffer from renovascular hypertension require surgical intervention more often than do patients with other forms of high BP. The referral rate for revascularization surgery decreased after the introduction of PTA. Nevertheless, revascularization
continues to represent approximately 25% of the referrals seen by vascular surgeons. 10-12Postoperative BP values of these patients improve or return to normal 60% to 80% of the time. ~°'~1"1318 Sixty-three percent of the patients with hypertension referred for renovascular surgery in this institution remain normotensive without medication after surgery. The specific determinants that separate patients who respond from those who do not respond to therapy remain unknown. Accordingly it is impossible to identify those patients who can be cured of renovascular hypertension before therapy. Renin secretion is widely used to predict postoperative benefit. Renin determinations, however, will not permit the prediction of curable disease, since the approach is unable to differentiate between curable and improvable hypertension. 19"2° As a result, in evaluating the predictive value of renin, cure and improvement of hypertension are generally grouped together as a positive response to therapy. ~9-24 An advantage of exercise renography is its ability to identify curable renovascular hypertension. It should be noted, however, that patients who respond partially to revascularization and benefit by responding with lower BP, or who require fewer drugs, fail to be recognized with the procedure. Luscher et al. 23 examined 95 patients with hypertension and unilateral renal vascular dis-
Volume 14 Number 2 August 1991
Exercise renography to predict su~ical cure of renovascular hypertension 205
ease. The authors reported a sensitivity and predictive value of 92% for the renin secretion index, whereas the plasma renin activity ratio had a sensitivity of 69% and a predictive value of 89%. The specificity for both procedures was just over 40%. Couch et al.2s using the renal vein renin ratio, correctly predicted benefit that included[ cure and improvement in 76% of his patients. Pickering et al. 2~ determined the renal vein renin activity ratio before therapy, and achieved a sensitivity of 74% and specificity of 100% in predicting improvement of hypertensive disease. Delin et al.26reported results with the basal renal vein renin ratio method. In this study 26 patients with renal vein renin determinations not indicative of renovascular hypertension were identified. Sixteen were successfully treated with operation. Of 38 patients with positive results, 16 were successfully treated and 22 were failures. Renin stimulation resulted in either false-positive or false-negative test results in 15 of 64 patients. These representative studies demonstrate that renin determinations can be used to advantage for selecting patients for surgery. They also show the inherent disadvantage of this approach: the inability to predict cure, and the high rate of false-positive and false-negative results. Our data suggest that the exercise renogram will serve 1:o identify curable hypertensive disease. Results of surgery were reviewed to determine cure rate in hypertensive patients with positive and negative exercise results. Although cured and improved patients with hypertension are generally grouped together when evaluating therapeutic success, both being considered therapy responders, we only attempted to predict cure. Exercise renography will not predict improvement of hypertensive disease. Accordingly, in predicting curability, exercise renography achieved the following results: 10 true positives, 17 true negatives, 2 false-positives, and 2 false-negatives (Table 1V). When curability was predicted in patients with unilateral stenosis, we had 9 true positives, 12 true negatives, 2 false-positives, and no false-negatives (Table V). The exercise scintigram is more accurate in predicting cure in patients with unilateral stenosis, since stenosis itself may cause tracer retention in the poststenotic kidney during upright exercise. Renal transplants with vascular stenosis have also been known to develop a hippurate transport disturbance dependent on posture when the patient stands, aa Note that transplants lack direct sympathetic innervation. This should not be confused with the abnormal exercise response under investigation, which is by definition bilateral in a two organ individual mren when only a unilateral
Table IV. Preoperative exercise renography compared to results of therapy Hypertension Exercise renogram
Cured
Improved
No response
Normal (n = 12) Abnormal (n = 19)
lL0 2
2 8
0 9
Table V. Preoperative exercise response versus postoperative BP status in 23 patients with unilateral stenosis
Normal scintigram (n = ]1) Abnormal scintigram (n = 12)
Hypertension cured
Hypertension not cured
9 0
2 12
stenosis is present. We would also like to postulate that an abnormal (bilateral) exercise renogram may occur when two separate stenoses each individually cause a unilateral hippurate transport disturbance. This scintigraphic pattern would of course be indistinguishable from the :function disturbance under investigation. In patients with hypertension exercise can induce a transient renal hippurate transport abnormality recently described by Clorius et al.9"27 The author presented a theoretic model that suggests that transient tracer retention in the tissue of the kidney indicates presence of an exercise-induced cortical perfusion disturbance. The abnormal, exercisemediated accumulation of hippurate within the renal tissue is considered the result of tracer entrapment within pro3dmal tubular cells. Hippurate is actively pumped into these cells, which creates an interstitiell concentration gradient that promotes the passive outflow of the tracer across the cell membrane into the tubulus lumen. 28-3°Cellular isotope accumulation is probable when a reduced glomerular filtration rate results in slowed washout of hippurate from the tubular lumen. This event disrupts the transit of hippurate from the cell into the tubulus lumen. Thus preglomerular and postglomerular pressure changes, which decrease the glomerular filtration rate, can cause parenchymal tracer entrapment. New, as yet unpublished clearance data from an ongoing study, has shown that the abnormal exercise response is associated with a pronounced reduction in glomerular filtration rate, and a massive reduction in the filtration fraction. These data give experimental support to the hypothesis of an exercise-induced
206
Hupp, Clorius, and Allenberg
cortical perfusion disturbance. We consider a vas afferents contraction, caused by elevated sympathetic tone during exercise, to be the immediate cause of the renal cortical perfusion disturbance. We interpret our data as pointing to the central nervous system as in large part responsible for the observed abnormal exercise renogram in the patient with renal hypertension resistant to therapy. We postulate that the mechanisms involved are comparable to those considered to exist in essential hypertension. 9'31'32 Sympathetic stimulation may be augmented by hyperreactive renal vessels because of deficient vasodilator mechanisms, or because of increased vasoconstrictor stimuli. 3I Excessive vasoconstriction may also be due to a modified metabolism, release of greater quantifies of vasoconstrictors, or an enhanced vascular response to normal quantities of vasoactive substances, as postulated by Hollenberg et al.32 It is stressed that the postulated involvement of the central nervous system will result in a bilateral perfusion disturbance, even in the presence of unilateral stenosis. Indeed, a disturbance of transient unilateral tracer transport in a poststenotic kidney should not be confused with an abnormal exercise renogram. Similarly, a disturbance of bilateral tracer transport not caused by exercise is likewise not to be equated with the discussed response. Exercise-induced renal tracer retention (abnormal scintigram) indicates that revascularization therapy will not cure hypertension. At present we cannot identify the patient who will benefit from surgical intervention by responding with a reduced need for medication. Scintigraphic evidence of incurable renovascular hypertension does not in itself rule out the need for revascularization. Surgical intervention will continue to enlarge the therapeutic spectrum in the patient with incurable renovascular hypertension when normotension fails to be achieved with a well tolerated drug regimen, since the need for drugs was lowered in some of our patients who failed to be cured. Surgical revascularization will continue to be a therapy of choice when progression of stenosis or organ preservation motivate intervention. A negative or normal exercise scintigram is associated with curable hypertensive disease. Reliable prognostic data for these patients should permit a change in the present treatment strategy. We expect that recognition of curable disease will result in fewer patients being dependent on lifelong antihypertensive drug treatment.
Journal of VASCULAR SURGERY
REFERENCES 1. Beck K, Hilden T. The frequency of secondary hypertension. Acta Med Scand 1975;197:65-9. 2. Greminger R, Vetter W, Zimmermann K, Beckerhoff K, SiegenthalerW. Prim~ireund sekundiireHypertonie in einem poliklinischen Patientengut. Schweiz Med Wochenschr 1977; 107:605-9. 3. Kennedy AC, Luke RH, Briggs JD, Stirling AB. Detection of renovascular hypertension. Lancet 1965;2:963-8. 4. SwalesJD. The hunt for renal hypertension. Lancet 1:1976; 577-9. 5. MaxwellMH, Bleifer ICH, Franklin SS, Varady PD, Deegan C. Cooperative study of renovascular hypertension: demographic analysis of the study. JAMA 1972;220:1195-1204. 6. Mann IFE, Allenberg JR. Renovascukire Hypertonie. In: Ganten D, Ritz E, eds. Lehrbuch der Hypertonie. Stuttgart, Schattauer Verlag, 1985:503-15. 7. Stadalnik RC, Vogel JM, Jansholt A-L, et al. Renal clearance and extraction parameters of ortho-iodohippurate (I-123) compared with OIH (1-131) and PAIl. J Nucl Med 1980;21:168-70. 8. BlaufoxMD, Potchen EJ, MerrillJP. Measurement of effective renal plasma flow in man by external counting methods. J Nucl Med 1967;8:77-85. 9. Clorius IH, Schmidlin P. The exercise renogram: a new approach documents renal involvement in systemic hypertension. J Nucl Med 1983;24:104-9. 10. Novick AC. Evaluation andpreparation for surgical treatment of renal artery disease. Am Vase Surg 1988;2:150-4. 11. Allenberg JR, Hupp T. Indikationsstellung zur operativen Nierenarterienrekonstmktion und Ergebnisse. In: Maurer PC, D6rder J, Sommoggy S, eds. Gef~13chirurgie im Fortschritt. 1st ed. Stuttgart/New York: Thieme Verlag, I991:229-41. 12. Allenberg lR, Hupp T. Selection for surgical reconstruction of renal artery stenoses by PTA: technique and results. Angioarchiv 1984;7:69-72. 13. Sos TA, Picketing TG, Sniderman K, et al. Beneficial effects of percutaneous transluminal renal angioplasty in patients with renovascular hypertension due to atheroma or fibromuscular dysplasia. N Engl J Med 1983;309:274-9. 14. Marks LS, Maxwell MH. Renal vein renin. Value and limitations in the prediction of operative results. Urol Clin North Am 1975;2:311-25. 15. Lawrie GM, Morris GC Jr, Glaeser DH, DeBakey ME. Renovascular reconstruction: factors affecting tong-term prognosis in 919 patients followed up to 31 years. Am_ J Cardiol 1989;63:1085-92. 16. Stanley JC, Whitehouse WM Jr, Zelenock GB, Graham LM, Cronenwett JL, Lindenauer SM. Reoperation for complications of renal reconstructive surgery undertaken for treatment of renovascular hypertension. J Vasc SURG 1985;2: 133-44. 17. Novick AC, Straffon R, Stewart B, et al. Diminished operative morbidity and mortality in renal revascularization. JAMA 1981;246:749-53. 18. Schweiger H, Raithel D, Seyferth W, Zeider E. Surgical treatment of renal artery occlusive disease: long-term results. J Cardiovasc Surg 1984;25:111-4. 19. Pedersen EB, Danielsen H, Fjeldborg O, Kornerup HJ, Madsen B. Renovascular hypertension. Ability of renal vein
Volume 14 Number 2 August 1991
20.
21.
22.
23.
24.
25.
26.
Exercise renography to predict surgical cure of renovascular hypertension 2 0 7
ratio to predict the blood pressure level 18-24 months after surgery. Nephron 19:86;44:29-31. Stanley JC, Whitehouse WM, Graham LM, Cronenwett JL, Zelenock GB, Lindenauer SM. Operative therapy of renovascular hypertension. Br J Surg 1982;69(suppl):63-6. Ernst CB, Bookstein JJ, Montie J, Baumgartel E, Hoobler SW, Fry WJ. Renal vein renin ratios and collateral vessels in renovascular hypertension. Arch Surg 1972;104:496-502. McCready RA, Daugherty MA, Nighbert EJ, Hyde 6L, Freedman AM, Ernst CB. Renal vascularization in patients with a single functioning ischemic kidney. J VASC SVRG 1987;6:185-90. Luscher TF, Greminger P, Kuhlmann U, Siegenthaler W, Largiader F, Vetter W. Renal venous renin determinations in renovascular hypertension. Diagnostic and prognostic vahae in unilateral renal artery stenosis treated by surgery or percutaneous transluminal angioplasty. Nephron 1986;44: 17-24. Picketing TG, Sos A, Vaughan ED, et al. Predictive value and changes of renin secretion in hypertensive patients with unilateral renovascular disease undergoing successful renal angioplasty. Am J Med 1984;76:398-404. Couch NP, Sullivan J, Crane C. The predictive accuracy of renal vein renin activity in the surgery of renovascular hypertension. Surgery 1976;79:70-6. Delin K, Anrell M, Granerus G. Preoperative diagnosis of
27.
28.
29. 30.
31.
32.
33.
renovascular hypertension: the use of acute stimulation of renin secretion. Acta Med Scand 1984;215:363-9. Clorius JH, Allenberg JR, Hupp T, et al. Predictive value of exercise renography for presurgical evaluation of nephrogenic hypertension. Hypertension 1987;10:280-6. DeGrazia JA, Scheibe PO, Jackson PE, et al. Clinical application of kinetic model of hippurate distribution and renal clearance. J Nucl Med 1974; 15:102-14. Foulkes EC. IZinetics of p-aminohippurate secretion in the rabbit. Am J Physiol 1963;205:1019-24. Wedeen RP, Jernon HI. Autoradiographic study of cellular transport of Hippuran-~2sI in the rat nephron. Am J Physiol 1968;214: 776-85. Case DB, Casarella WJ, Laragh JH, et al. Renal cortical blood flow and angiography in low- and normal-renin essential hypertension. Kidney Int 1978;13:236-44. Hollenberg NK, Borucki LJ, Adams DF. The renal vasculature in early essential hypertension: evidence for a pathogenic role. Medicine 1978;57:167-78. Clorius JH, Dreikorn K, Zelt J, Schmidlin P, Raptou E, Georgi P. Posture-induced disturbance of pertechnetate flow and 1-123 iodohippurate transport in some renal graft recipients with hypertension. I Nucl Med 1980;21:829-34.
Submitted Aug. 13, 1990; accepted Mar. 27, 1991.
Heidelberg, Germany Renal artery stenosis with resultant renovascular hypertension has attracted clinical attention because the disease is potentially curable and because numerous diagnostic and therapeutic modalities compete for clinical acceptance. An exercise-mediated disturbance of renal hippurate transport was recently described, and has been implicated as having a role in nephrogenic fixed hypertension. To predict the final course of renovascular hypertension before operation we carried out a prospective study with the goal of verifying the predictive value of exercise hippurate scintigraphy. The study was to test the hypothesis that patients with disturbance of renal hippurate transport (pathologic renogram) induced by exercise would have stabilized hypertension and would continue to be hypertensive after operation. Thirty-one patients with hypertension who had unilateral or bilateral renovasoalar stenosis documented on angiography were referred to rest and exercise hippurate scintigrams before operation. The results of the examinations at rest served as standard and were compared with the exercise scintigrams. In 19 of the 31 (61%) patients a disntrbance oftransrenal hippurate transport evolved during exercise, whereas 12 (39%) patients failed to respond to exercise with altered hippurate kinetics. Twenty-six patients went on to renovascular operations; five had percutaneous transluminal angioplasty. Revascularization results differed markedly when the blood pressure response of patients with positive results on exercise (abnormal) and patients with negative results on exercise (normal) were compared. Ten of 12 patients with hypertension who had normal exercise renograms were cured. In comparison, blood pressure values were tittle influenced by therapy in patients with an abnormal response, where 17 of 19 patients continued to have hypertensive disease after therapy. It is most important to note that the transitory tracer transport abnormality is easily recognized on scintigraphy. The study shows that exercise scintigraphy helps identify patients with curable renovascular hypertension and thus predicts which patients will benefit most from operation. (J Vasc S t ~ 1991;14:200-7.)
Renovascular hypertension has attracted extensive clinical attention because of the presence of physiologic models, its epidemiology, and the numerous diagnostic and therapeutic modalities competing for clinical acceptance. This extensive clinical interest is directed to a mere subgroup of patients with hypertension. Renovascular hypertension is found in only 1% to 6% of all patients with high blood pressure (BP). 1-6 Nevertheless, 60 to 360 thousand patients in Germany alone require therapy. The interest in renovascular hypertension also increased after the introduction of percutaneous transFrom the Department of Surgery, Division of Vascular Surgery (Drs. Hupp and AUenberg),Universityof Heidelberg and the German Cancer Research Center (Dr. Clorius), Heidelberg. Reprint requests: Thomas Hupp, MD, Division of Vascular Surgery,Departmentof Surgery,the Universityof Heidelberg, Im NeuenheimerFeld 110, D-6900 Heidelberg, Germany. 24/1/29810 200
luminal angioplasty (PTA), which greatly expanded the available therapeutic options. It would obviously be desirable to identify those patients who would benefit from PTA or surgical revascularization. At present it is not possible to securely predict BP response before PTA or operation. Only 60% to 80% of treated patients respond with cure or improvement of BP values after operation. Thus PTA, or surgical revascularization, fails to improve the hypertension of approximately 30% of treated patients. Iodine-123 (1-123) or 1-131 hippurate has transrenal kinetics comparable to paraaminohippurate. 7 The scintigraphic visualization of the tracer therefore contains information on renal cortical blood flow? It was recently recognized that ergometric exercise can provoke a disturbance oftransrenal hippurate transport. 9 The transport abnormality normalizes when exercise is terminated. This transitory and abnormal, orthoiodohippurate (1-131)
Volume 14 Number 2 August 1991
Exercise renography to predict surgical cure of renovascular hypertension
transport is considered indicative of disturbance of renal cortical perfusion. A posture or exercisemediated disturbance of renal hippurate transport has been documented only in patients with hypertension. The present study included patients with hypertension referred for surgical therapy of renovascular lesions. During a presurgical scintigraphic examination we identified those patients in whom the exercise-induced transrenal hippurate transport abnormality developed. The hypothesis was tested that those patients with hypertension- and an exerciseinduced hippurate transport abnormality had a form of renal hypertension not curable by operation or PTA. Revascularization could serve to preserve organ function and possible progression of the arteriosclerotic stenosis, but it would fail to cure hypertensive disease. In short, we sought to predict which paticnts would respond to surgical therapy with BP normalization. METHOD
One hundred fifty-one renal vascular reconstructions were required by 117 patients referred to this center during a 7-year period lasting from January 1983 to January 1990. We sought to include those patients in the prospective blind study who met the following criteria: unilateral or bilateral renal artery stenosis with 75% or greater lumen reduction verified on angiography, hypertensive BP values (Table I), and found suitable for surgical treatment. The patients had to be capable of ergometric exercise and show willingness to participate in the examinations. The mean BP value of the patients with an abnormal exercise scintigram was 186/105 mm Hg, whereas those patients with a scintigram not influenced by exercise had a mean BP value of 170/98 mm Hg, both values obtained with the patients under drug therapy (Tables I and II). Forty-two of the 117 patients were identified as meeting the selection criteria, and were referred to the scintigraphic examinations. Nine of these patients were excluded from final analysis because of incomplete scintigraphic data, after interruption of exercise when systolic BP values rose to 250 mm Hg, or diastolic pressure exceeded 125 mm Hg, or when exercise was terminated because of leg pain resulting from peripheral arterial vascular disease. It was thus possible to obtain a complete scintigraphic examination in 33 patients. The scintigraphic examinations were evaluated by one observer (I.H.C.), who was blinded with respect to the clinical data. Exercise scintigrams were compared with an examination at rest and were classified as normal or abnormal.
201
The final dinical evaluation included documentation of preoperative and postoperative antihypertensive medication, BP values both before and after surgery, and the compa:rison ofangiographic findings before and after therapy. Two of the 33 patients were dropped from the study when the surgical outcome was compromised by postoperative complications, so that 31 patients remained in the study. Postoperative data were collected at the time of discharge and during a follow-up examination scheduled on average 20.5 months after operation. The earliest follow-up examination was after 6 months, the latest after 78 months. The final clinical examination was generally made at this hospital's vascular-disease outpatient service. Pertinent information on the BP status and antihypertensive drugs was occasionally obtained from the family physician when patients failed to appear for follow-up examination. The clinical data were then evaluated by a surgeon (T.H.) who had no immediate access to the preoperative scintigraphic data. After collection and evaluation of the clinical data, we compared the final postoperative BP status with the preoperative scintigrams. To classify BP development we used slightly modified criteria established by the U.S. cooperative trial of surgical treatment of renovascular hypertension. s Hypertension was considered cured if the diastolic BP remained below 90 mm Hg without antihypertensive medication. Surgery was judged to have improved the disease when the diastolic BP was reduced by at least 15% with medication. Nonresponders were defined as those patients with hypertension who had a BP reduction of less than 15% or who required increasing doses of antihypertensive medication or a combination of three or more drugs. Each subject had renography in a prone position and in the upright (sitting) position during ergometric exercise. Both examinations were obtained within 1 week, generally within 72 hours. To obtain adequate hydration, patients were asked to drink 400 ml of fluid during the 30 minutes before renography. All patients received careful oral instructions about the examinations. The renogram in the prone position was begtm within minutes after patient positioning. The exercise protocol was used to document the influence of exercise on hippurate kinetics. Patient comfort was emphasized during the test. The workload was considered inadequate if it could be continued indefinitely and too strenuous if it resulted in cxhaustion. Patients sitting in front of the gamma camera on a bicycle ergometcr were asked to sit straight-backed so that the kidney-to-camera distance was kept small. Pulse and BP were noted during
Journal of. VASCULAR SURGERY
202 Hz~pp, Clorius, and Allenberg
Table I. Comparison of the BP-status of patients with renal artery stenosis ~stopera~e
Preoperative Patient No.
Stenosis
Normal renogramm 1 U 2 U 3 U 4 U 5 U 6 U 7 U 8
U
9
U BI 11 U 12 U Abnormal renogramm 13 U 14 BI 15 U 16 U 17 BI BI 18 U i9 U 20 BI 21 U 22 BI 23 BI 24 U 25 U 26 U 27 BI 28 U 29 U 30 U 31 10
BP (ram Hg)
No, of drugs
BP (mm Hg) at discharge
BP (ram Hg) follow-up
Month postop.
No. of drugs
180/100 160/85 200/120 170/110 170/100 160/80 170/100 170/90 150/100 155/90 180/110 175/100
2 2 3 0 0 3 3 4 0 4 4 5
130/75 130/80 130/90 150/80 150/100 130/85 120/70 150/80 140/80 150/70 140/85 150/70
100/60 160/80 140/90 150/95 135/70 160/90 130/80 150/80 140/80 140/80 140/80
31 13 39 6 8 6 7 78 9 36 16 8
0 0 3 0 0 0 0 0 0 0 2 0
180/110 240/140 200/120 240/100 180/90 200/110 120/80 120/65 220/120 200/130 210/140 190/100 120/80 200/100 195/110 200/100 205/I00 140/105 190/110
2 3 3 1 0 5 2 4 3 1 2 I0 1 6 3 5 3 6 2
130/90 160/80 140/80 140/70 140/80 160/70 150/90 130/80 160/100 200/100 140/80 170/90 140/80 165/85 160/95 160/90 180/90 160/100 180/95
145/75 160/80 130/80 190/90 150/90 170/90 I35/95 145/85 140/80 190/100 140/90 i90/90 180/90 150/80 150/100 160/90 180/80 I60/100 180/105
28 7 25 10 7 6 41 6 7 52 20 30 48 62 32 18 78 9 9
3 1 2 1 0 3 1 3 2 3 0 3 1 1 1 2 1 6 3
150/80
U, Unilateral stenosis; BI, bilateral stenosis; BP, blood pressure.
Table II. Exercise renography in 31 patients with hypertension was used to group patients and predict curabifity of disease; comparison of BP values and the mean number of drugs required before and after revascularization therapy Scintigraphy Hippurate transport Tracerin disturbed bladder (rain) Prone Normal scintigram (n =
12)
Abnormal scinfigram (n = 19)
Preoperative
Postoperative
Exercise Prone Exercise Mean BP (ram Hg) No of drug (2) Mean BP (ram Hg)
No of drags (~)
0
0
3.3
3.7
170/98
2.50
142/88
0.5 (10 took more)
0
19
3.3
8.0
I86/105
3.26
160/89
2.0 (2 took none)
exercise and were monitored at 5-minute intervals. Ergometric resistance was set at 60 watts for women and 80 watts for men. After 1 to 2 minutes of exercise the pulse rate change in response to exercise was noted. Renography was begun only after the pulse rate had increased at least 20 beats/rain. After
radiotracer injection, the patients continued to exercise. Ergometric resistance was adjusted according to the whishes of the patients. Radionuclide renography was begun after intravenous injection of either 7 t*Ci (I-131) or 6 b~Ci (I-123) o-iodohippurate per kilogram body weight.
Volume 14 Number 2 August 1991
Exercise renography to predict surgical cure of renovascular hypertension
A 15-inch gamma c:maera equipped with a generalpurpose medium-energy (360 keV) parallel-hole collimator was used for all studies, with a window setting at 20%, centered over the photopeak of the tracer. One-minute images were obtained, beginning with the injection, and continued at 1, 2, 3, 4, 7, 9, 14, and 19 minutes. To identify the appearance of the tracer in the bladder, which was recognized on the sequential scans, we extended the uninterrupted 1-minute image sequence past the fourth minute when required. The ,examination was terminated after 20 minutes. Data were stored on magnetic tape and were analyzed by minicomputer. Regions of interest were placed over each kidney to determine single-kidney function. No attempt was made to exclude the renal pelvic system. Background regions of interest were placed automatically around each kidney by use ofa 2-pixe]Lwidth. Single-kidney hippurate uptake, expressed as a percentage of total uptake of both kidneys, was determined. Uptake was taken to be proportional to the gradient of the renogram between 24 and 120 seconds. The third curve segment of the renogram was analyzed by inspection only. Results of prone and exercise renography and serial scintigrams were compared. RESULTS
Thirty-one of 117 patients with renovascular hypertension were included in the final evaluation. Twenty-three of 31 patients with hypertension had a unilateral stenosis, eight had bilateral disease. Treatment was as follows: 26 of the 31 patients had revascularization surgery, 7 of them bilaterally (Table III). In five patients the vascular stenosis was able to be treated with PTA, once bilaterally. Surgical intervention was required in 26 patients with 32 reconstructions. Included were 25 aortorenal bypass operations, 6 renal an:ery reimplantations after resection of the stenosed segment, and 1 renal artery endarterectomy. Function impairment of a poststenotic small kidney required a nephrectomy in one patient. Eleven of the 26 patients operated on also required surgical intervention distant from the treated renal vasculamre (simultaneous aortoiliac reconstruction). Five needed a resection of an abdominal aneurysm of the aorta, six had a Y-bifurcation bypass for obliterative arteriosclerosis, and two had a nephrectomy of a small kidney contralateral to the revascularization procedure. All patients had preoperative rest and exercise renography (Fig. 1). A disturbance of bilateral hippurate transport developed in 19 of 31 (61%) patients in response to exercise (abnormal scintigram) (Table II). This resulted in delayed excretion
203
Table III. Thirty-one patients with hypertension were included in the study; 26 had surgical revascularization, 5 had transluminal renal angioplasty (PTA) Patients
26
No. ofprocedures
33 ~ ~ 6
Surgicalprocedure S Dacron bypass Vein bypass Reimplantation Endarteriectomy Nephrectomy PTA
(No.) (16) (9) (6) (1) (1)
ofhippurate from the renal tissue. On average, tracer appeared in the bladder after 3.3 minutes when scintigraphy was performed with the patient at rest, and after 8 minutes during ergometric exercise (Table II). The hippurate transport abnormality was particularly severe in three patients in whom we noted the first radiotracer accumulation in the bladder after 19.5 minutes, 25 minutes, and 46 minutes after intravenous injection of the activity. A 20-minute value was used to calculate the mean appearance time of the tracer in the bladder, to avoid having these highly, abnormal results magnify the extent of the transport disturbance. Twelve of 31 patients with hypertension (39%) failed to respond to exercise with a bilateral hippurate transport abnormality (normal scintigram). Exercise resulted in a slight delay of radiohippurate excretion into the bladder. At rest the average appearance time of hippuram in the bladder was after 3.3 minutes, during exercise it was after 3.7 minutes (Table II). Normal as well as abnormal exercise scintigrams were observed in patients with balanced as well as with dissimilar single kidney function. Patients with hypertension who had an abnormal exercise scintigram had higher mean BP values before revascularization than those whose renogram failed to be influenced by exercise (Table II). The results achieved with surgical revascularization (n = 25), nephrectomy (n = 1), and renal artery dilation (n = 5) were clearly different when patients with exercise positive and exercise negative scintigrams were compared (Table IV). Ten of 12 patients with hypertension who had a normal exercise renogram were cured. Therapy improved the BP values of the other two patients (No. 3, No. 11) (Table I) even though they continued to require antihypertensive drugs. In comparison, systemic BP values were little changed by therapy in patients with abnormal exercise renograms (Table IV). Hypertension was cured in two out of the 19 patients. Both had bilateral vascular lesions. One of these patients (No. 17)
Iournal of VASCULAR SURGERY
204 Hupp, Clorius, and Allenberg
M~N
MtN
Fig. 1. Sequentialscintigramsofa60-year-oldfemalepatient, obtained with the patient in the prone position at rest and during exercise on a bicycle ergometer while sitting upright. The patient had a vascular stenosis of the right kidney. The examination at rest documents the reduction of single kidney function on the right side, as well as timely tracer excretion into the renal pelvis and bladder (fourth minute), and a regular tissue clearance of the tracer. The abnormal exercise renogram demonstrates a bilateral transrenal hippurate transport disturbance identified by the elevated excretory segment of the renogram, by hippurate retention in the renal tissue, and delayed excretion ofhippurate into the bladder (twentieth minute). Note: unilateral vascular disease is associated with a bilateral function abnormality during exercise.
(Table I) with bilateral stenosis demonstrated compromise of unilateral left kidney function (left 30%, right 70%). Exercise resulted in a bilateral disturbance of hippurate transit, with delayed excretion of activity into the bladder. Whereas the radiotracer was excreted in the fourth minute while the patient was at rest, it appeared in the bladder during the eighth minute during exercise. The second patient's (No. 23) (Table I) left kidney had less than 5% of total function after vascular occlusion after PTA, whereas the right organ had a documented, but functionally silent, vascular stenosis. Accordingly, the left kidney was barely visualized during scintigraphy. Exercise induced a hippurate transport disturbance in the right kidney. Bladder activity was noted after 4 minutes during rest, and after 7 minutes during exercise. The i9 patients with hypertension who had abnormal exercise renograms included two who were cured, eight patients with improved BP values, and nine who failed to respond to therapy (Table IV). DISCUSSION Those who suffer from renovascular hypertension require surgical intervention more often than do patients with other forms of high BP. The referral rate for revascularization surgery decreased after the introduction of PTA. Nevertheless, revascularization
continues to represent approximately 25% of the referrals seen by vascular surgeons. 10-12Postoperative BP values of these patients improve or return to normal 60% to 80% of the time. ~°'~1"1318 Sixty-three percent of the patients with hypertension referred for renovascular surgery in this institution remain normotensive without medication after surgery. The specific determinants that separate patients who respond from those who do not respond to therapy remain unknown. Accordingly it is impossible to identify those patients who can be cured of renovascular hypertension before therapy. Renin secretion is widely used to predict postoperative benefit. Renin determinations, however, will not permit the prediction of curable disease, since the approach is unable to differentiate between curable and improvable hypertension. 19"2° As a result, in evaluating the predictive value of renin, cure and improvement of hypertension are generally grouped together as a positive response to therapy. ~9-24 An advantage of exercise renography is its ability to identify curable renovascular hypertension. It should be noted, however, that patients who respond partially to revascularization and benefit by responding with lower BP, or who require fewer drugs, fail to be recognized with the procedure. Luscher et al. 23 examined 95 patients with hypertension and unilateral renal vascular dis-
Volume 14 Number 2 August 1991
Exercise renography to predict su~ical cure of renovascular hypertension 205
ease. The authors reported a sensitivity and predictive value of 92% for the renin secretion index, whereas the plasma renin activity ratio had a sensitivity of 69% and a predictive value of 89%. The specificity for both procedures was just over 40%. Couch et al.2s using the renal vein renin ratio, correctly predicted benefit that included[ cure and improvement in 76% of his patients. Pickering et al. 2~ determined the renal vein renin activity ratio before therapy, and achieved a sensitivity of 74% and specificity of 100% in predicting improvement of hypertensive disease. Delin et al.26reported results with the basal renal vein renin ratio method. In this study 26 patients with renal vein renin determinations not indicative of renovascular hypertension were identified. Sixteen were successfully treated with operation. Of 38 patients with positive results, 16 were successfully treated and 22 were failures. Renin stimulation resulted in either false-positive or false-negative test results in 15 of 64 patients. These representative studies demonstrate that renin determinations can be used to advantage for selecting patients for surgery. They also show the inherent disadvantage of this approach: the inability to predict cure, and the high rate of false-positive and false-negative results. Our data suggest that the exercise renogram will serve 1:o identify curable hypertensive disease. Results of surgery were reviewed to determine cure rate in hypertensive patients with positive and negative exercise results. Although cured and improved patients with hypertension are generally grouped together when evaluating therapeutic success, both being considered therapy responders, we only attempted to predict cure. Exercise renography will not predict improvement of hypertensive disease. Accordingly, in predicting curability, exercise renography achieved the following results: 10 true positives, 17 true negatives, 2 false-positives, and 2 false-negatives (Table 1V). When curability was predicted in patients with unilateral stenosis, we had 9 true positives, 12 true negatives, 2 false-positives, and no false-negatives (Table V). The exercise scintigram is more accurate in predicting cure in patients with unilateral stenosis, since stenosis itself may cause tracer retention in the poststenotic kidney during upright exercise. Renal transplants with vascular stenosis have also been known to develop a hippurate transport disturbance dependent on posture when the patient stands, aa Note that transplants lack direct sympathetic innervation. This should not be confused with the abnormal exercise response under investigation, which is by definition bilateral in a two organ individual mren when only a unilateral
Table IV. Preoperative exercise renography compared to results of therapy Hypertension Exercise renogram
Cured
Improved
No response
Normal (n = 12) Abnormal (n = 19)
lL0 2
2 8
0 9
Table V. Preoperative exercise response versus postoperative BP status in 23 patients with unilateral stenosis
Normal scintigram (n = ]1) Abnormal scintigram (n = 12)
Hypertension cured
Hypertension not cured
9 0
2 12
stenosis is present. We would also like to postulate that an abnormal (bilateral) exercise renogram may occur when two separate stenoses each individually cause a unilateral hippurate transport disturbance. This scintigraphic pattern would of course be indistinguishable from the :function disturbance under investigation. In patients with hypertension exercise can induce a transient renal hippurate transport abnormality recently described by Clorius et al.9"27 The author presented a theoretic model that suggests that transient tracer retention in the tissue of the kidney indicates presence of an exercise-induced cortical perfusion disturbance. The abnormal, exercisemediated accumulation of hippurate within the renal tissue is considered the result of tracer entrapment within pro3dmal tubular cells. Hippurate is actively pumped into these cells, which creates an interstitiell concentration gradient that promotes the passive outflow of the tracer across the cell membrane into the tubulus lumen. 28-3°Cellular isotope accumulation is probable when a reduced glomerular filtration rate results in slowed washout of hippurate from the tubular lumen. This event disrupts the transit of hippurate from the cell into the tubulus lumen. Thus preglomerular and postglomerular pressure changes, which decrease the glomerular filtration rate, can cause parenchymal tracer entrapment. New, as yet unpublished clearance data from an ongoing study, has shown that the abnormal exercise response is associated with a pronounced reduction in glomerular filtration rate, and a massive reduction in the filtration fraction. These data give experimental support to the hypothesis of an exercise-induced
206
Hupp, Clorius, and Allenberg
cortical perfusion disturbance. We consider a vas afferents contraction, caused by elevated sympathetic tone during exercise, to be the immediate cause of the renal cortical perfusion disturbance. We interpret our data as pointing to the central nervous system as in large part responsible for the observed abnormal exercise renogram in the patient with renal hypertension resistant to therapy. We postulate that the mechanisms involved are comparable to those considered to exist in essential hypertension. 9'31'32 Sympathetic stimulation may be augmented by hyperreactive renal vessels because of deficient vasodilator mechanisms, or because of increased vasoconstrictor stimuli. 3I Excessive vasoconstriction may also be due to a modified metabolism, release of greater quantifies of vasoconstrictors, or an enhanced vascular response to normal quantities of vasoactive substances, as postulated by Hollenberg et al.32 It is stressed that the postulated involvement of the central nervous system will result in a bilateral perfusion disturbance, even in the presence of unilateral stenosis. Indeed, a disturbance of transient unilateral tracer transport in a poststenotic kidney should not be confused with an abnormal exercise renogram. Similarly, a disturbance of bilateral tracer transport not caused by exercise is likewise not to be equated with the discussed response. Exercise-induced renal tracer retention (abnormal scintigram) indicates that revascularization therapy will not cure hypertension. At present we cannot identify the patient who will benefit from surgical intervention by responding with a reduced need for medication. Scintigraphic evidence of incurable renovascular hypertension does not in itself rule out the need for revascularization. Surgical intervention will continue to enlarge the therapeutic spectrum in the patient with incurable renovascular hypertension when normotension fails to be achieved with a well tolerated drug regimen, since the need for drugs was lowered in some of our patients who failed to be cured. Surgical revascularization will continue to be a therapy of choice when progression of stenosis or organ preservation motivate intervention. A negative or normal exercise scintigram is associated with curable hypertensive disease. Reliable prognostic data for these patients should permit a change in the present treatment strategy. We expect that recognition of curable disease will result in fewer patients being dependent on lifelong antihypertensive drug treatment.
Journal of VASCULAR SURGERY
REFERENCES 1. Beck K, Hilden T. The frequency of secondary hypertension. Acta Med Scand 1975;197:65-9. 2. Greminger R, Vetter W, Zimmermann K, Beckerhoff K, SiegenthalerW. Prim~ireund sekundiireHypertonie in einem poliklinischen Patientengut. Schweiz Med Wochenschr 1977; 107:605-9. 3. Kennedy AC, Luke RH, Briggs JD, Stirling AB. Detection of renovascular hypertension. Lancet 1965;2:963-8. 4. SwalesJD. The hunt for renal hypertension. Lancet 1:1976; 577-9. 5. MaxwellMH, Bleifer ICH, Franklin SS, Varady PD, Deegan C. Cooperative study of renovascular hypertension: demographic analysis of the study. JAMA 1972;220:1195-1204. 6. Mann IFE, Allenberg JR. Renovascukire Hypertonie. In: Ganten D, Ritz E, eds. Lehrbuch der Hypertonie. Stuttgart, Schattauer Verlag, 1985:503-15. 7. Stadalnik RC, Vogel JM, Jansholt A-L, et al. Renal clearance and extraction parameters of ortho-iodohippurate (I-123) compared with OIH (1-131) and PAIl. J Nucl Med 1980;21:168-70. 8. BlaufoxMD, Potchen EJ, MerrillJP. Measurement of effective renal plasma flow in man by external counting methods. J Nucl Med 1967;8:77-85. 9. Clorius IH, Schmidlin P. The exercise renogram: a new approach documents renal involvement in systemic hypertension. J Nucl Med 1983;24:104-9. 10. Novick AC. Evaluation andpreparation for surgical treatment of renal artery disease. Am Vase Surg 1988;2:150-4. 11. Allenberg JR, Hupp T. Indikationsstellung zur operativen Nierenarterienrekonstmktion und Ergebnisse. In: Maurer PC, D6rder J, Sommoggy S, eds. Gef~13chirurgie im Fortschritt. 1st ed. Stuttgart/New York: Thieme Verlag, I991:229-41. 12. Allenberg lR, Hupp T. Selection for surgical reconstruction of renal artery stenoses by PTA: technique and results. Angioarchiv 1984;7:69-72. 13. Sos TA, Picketing TG, Sniderman K, et al. Beneficial effects of percutaneous transluminal renal angioplasty in patients with renovascular hypertension due to atheroma or fibromuscular dysplasia. N Engl J Med 1983;309:274-9. 14. Marks LS, Maxwell MH. Renal vein renin. Value and limitations in the prediction of operative results. Urol Clin North Am 1975;2:311-25. 15. Lawrie GM, Morris GC Jr, Glaeser DH, DeBakey ME. Renovascular reconstruction: factors affecting tong-term prognosis in 919 patients followed up to 31 years. Am_ J Cardiol 1989;63:1085-92. 16. Stanley JC, Whitehouse WM Jr, Zelenock GB, Graham LM, Cronenwett JL, Lindenauer SM. Reoperation for complications of renal reconstructive surgery undertaken for treatment of renovascular hypertension. J Vasc SURG 1985;2: 133-44. 17. Novick AC, Straffon R, Stewart B, et al. Diminished operative morbidity and mortality in renal revascularization. JAMA 1981;246:749-53. 18. Schweiger H, Raithel D, Seyferth W, Zeider E. Surgical treatment of renal artery occlusive disease: long-term results. J Cardiovasc Surg 1984;25:111-4. 19. Pedersen EB, Danielsen H, Fjeldborg O, Kornerup HJ, Madsen B. Renovascular hypertension. Ability of renal vein
Volume 14 Number 2 August 1991
20.
21.
22.
23.
24.
25.
26.
Exercise renography to predict surgical cure of renovascular hypertension 2 0 7
ratio to predict the blood pressure level 18-24 months after surgery. Nephron 19:86;44:29-31. Stanley JC, Whitehouse WM, Graham LM, Cronenwett JL, Zelenock GB, Lindenauer SM. Operative therapy of renovascular hypertension. Br J Surg 1982;69(suppl):63-6. Ernst CB, Bookstein JJ, Montie J, Baumgartel E, Hoobler SW, Fry WJ. Renal vein renin ratios and collateral vessels in renovascular hypertension. Arch Surg 1972;104:496-502. McCready RA, Daugherty MA, Nighbert EJ, Hyde 6L, Freedman AM, Ernst CB. Renal vascularization in patients with a single functioning ischemic kidney. J VASC SVRG 1987;6:185-90. Luscher TF, Greminger P, Kuhlmann U, Siegenthaler W, Largiader F, Vetter W. Renal venous renin determinations in renovascular hypertension. Diagnostic and prognostic vahae in unilateral renal artery stenosis treated by surgery or percutaneous transluminal angioplasty. Nephron 1986;44: 17-24. Picketing TG, Sos A, Vaughan ED, et al. Predictive value and changes of renin secretion in hypertensive patients with unilateral renovascular disease undergoing successful renal angioplasty. Am J Med 1984;76:398-404. Couch NP, Sullivan J, Crane C. The predictive accuracy of renal vein renin activity in the surgery of renovascular hypertension. Surgery 1976;79:70-6. Delin K, Anrell M, Granerus G. Preoperative diagnosis of
27.
28.
29. 30.
31.
32.
33.
renovascular hypertension: the use of acute stimulation of renin secretion. Acta Med Scand 1984;215:363-9. Clorius JH, Allenberg JR, Hupp T, et al. Predictive value of exercise renography for presurgical evaluation of nephrogenic hypertension. Hypertension 1987;10:280-6. DeGrazia JA, Scheibe PO, Jackson PE, et al. Clinical application of kinetic model of hippurate distribution and renal clearance. J Nucl Med 1974; 15:102-14. Foulkes EC. IZinetics of p-aminohippurate secretion in the rabbit. Am J Physiol 1963;205:1019-24. Wedeen RP, Jernon HI. Autoradiographic study of cellular transport of Hippuran-~2sI in the rat nephron. Am J Physiol 1968;214: 776-85. Case DB, Casarella WJ, Laragh JH, et al. Renal cortical blood flow and angiography in low- and normal-renin essential hypertension. Kidney Int 1978;13:236-44. Hollenberg NK, Borucki LJ, Adams DF. The renal vasculature in early essential hypertension: evidence for a pathogenic role. Medicine 1978;57:167-78. Clorius JH, Dreikorn K, Zelt J, Schmidlin P, Raptou E, Georgi P. Posture-induced disturbance of pertechnetate flow and 1-123 iodohippurate transport in some renal graft recipients with hypertension. I Nucl Med 1980;21:829-34.
Submitted Aug. 13, 1990; accepted Mar. 27, 1991.
