Failed percutaneous transluminal renal angioplasty: Experience with lesions requiring operative intervention

Transcription

1 Failed percutaneous transluminal renal angioplasty: Experience with lesions requiring operative intervention Richard H. Dean, M.D., James T. Callis, M.D., Bruce M. Smith, M.D., and Patrick W. Meacham, M.D., Nashville, Tenn., and Winston-Salem, N.C. Percutaneous transluminal angioplasty (PTA) has received widespread pubficity as a safe, simple, and successful alternative to operation for the management of renal artery stenoses and renovascular hypertension (RVH). Although, in our institution, the primary management of RVH remains operative revascularization, with more than 750 such procedures having been done, we have had the opportunity to manage a spectrum of PTA failures in nine patients during the last 5 years. These include (1) acute dissection of atherosclerotic lesions and occlusion of the distal renal artery requiring emergency operation; (2) unilateral perforation and bilateral thrombosis of fibrodysplastic branch renal artery lesions requiring staged ex vivo reconstruction; (3) cholesterol embolization and recurrence to total occlusion of orificial atherosclerotic lesions with loss of excretory renal function; (4) chronic dissection from repeated "temporarily successful" PTAs of medial fibrodysplastic lesions; and (5) rapid recurrence and acceleration of hypertension in a 17-year-old girl with congenital renal artery stenoses. In each instance operation was compfieated by an intense perivascular inflammatory response from the previous PTA and required a more complex reconstruction than would have been needed originally. These sequelae argue for moderating enthusiasm for the use of PTA and for limiting its routine use to nonorificial atherosclerotic lesions and fibrodysplastic lesions restricted to the main renal artery. Orificial atherosclerotic lesions, branch fibrodysplastic lesions, and congenital stenoses have a high probability of failure, complications, or both when treated by PTA and should be considered for primary operative intervention. (J VAsc SuR~ 1987;6:301-7.) The preferred treatment of renovascular hypertension (RVH) is controversial. The introduction of increasingly specific antihypertensive medications has decreased the proportion of compliant patients not achieving adequate blood pressure control by drug therapy alone. Similarly, improved criteria for selection of operative candidates and techniques of operative intervention have provided increasingly accurate and safe means of permanently relieving renal ischemia and secondary RVH.1 However, the introduction in 1978 by Griintzig et al. 2 of a coaxial method of percutaneous transluminal angioplasty (PTA) has added a new option for treatment of RVH. From the Department of Surgery, Vanderbilt University Medical Center (Drs. CaUls, Smith, and Meacham) and the Department of Surgery, Wake Forest University Medical Center, Winston- Salem (Dr. Dean). Presented at the Eleventh Annual Meeting of the Southern Association for Vascular Surgery, Scottsdale, Ariz., Jan , Reprint requests: Richard H. Dean, M.D., Department of Surgery, Wake Forest University Medical Center, 300 South Hawthome Rd., Winston-Salem, NC PTA has gained widespread popularity, and its perceived ease, safety, and success have led m its preferential use for the treatment of RVH in many centers. In other centers, it is currently being employed as a diagnostic intervention without prior documentation of the physiologic importance of the lesion or its relationship to hypertension. PTA has been attempted for a wide range of anatomic lesions and clinical situations. Clearly, in the 9 years since its introduction, PTA has broadened enthusiasm within the medical community for correction of renovascular disease. Although the performance of PTA by experienced persons has provided early results similar to those achieved by surgical management, the liberal use of PTA has led to an array of complications that have required operative correction. 36 This article summarizes our experience with the operative management of the complications and failures of PTA. Evaluation of these cases and examination of the literature provide an opportunity for us to comment on the relative risk and value of PTA in the treatment of the various types of renovascular lesions. 301

2 302 Dean et al. Journal of VASCULAR SURGERY Table I. Clinical material in series of nine patients with failed PTA Angiographic lesions Pre-PTA blood Patient Sex Age (yr) pressure (ram Hg) Before PTA PTA procedure Consequence of PTA 1 F /120 Bilat FMD with branch Bilat PTA 2 F /120 Bilat FMD with branch Bilat PTA x 2 3 F /110 Bilat FMD with branch Right PTA 4 F /70 Bilat FMD with branch Bilat PTA 5 F /154 Bilat isolated congenital Left PTA x 2 stenoses 6 M /150 Ostial ather LRA, Left PTA patent RRA bypass 7 M /120 Ostial ather RRA Right PTA 8 M /140 Ostial ather RRA; total Right PTA LRA occlusion 9 F /120 Ostial ather RRA Right PTA RRA perforation, hemorrhage, total occlusion; LRA dissection, branch vessel occlusion; acceleration of hypertension; azotemia Recurrence of bilat lesions x 2; recurrence of hypertension at 2 and 1 mo Intimal dissection; residual stenosis; no change in hypertension Development of post-pta RRA aneurysm No change in stenosis x 2; no change in hypertension x 2 Infimal dissection, total occlusion; flank pain, tenderness; accelerated hypertension Restenosis at 2 mo; azotemia; loss of renal mass; recurrence of hypertension at 2 mo Increased azotemia; recurrence of lesion at 4 mo; recurrence of hypertension at 4 mo Azotemia; minimal change in lesion; no change in hypertension FMD = medial fibromuscular dysplasia; RRA = right renal artery; LRA = left renal artery; ather = atherosclerosis. PATIENTS Nine patients have required operative intervention by us for the consequences of PTA treatment of renovascular disease. Their ages ranged from 17 to 70 years. Three male patients and one female patient had atherosclerotic renovascular lesions, four female patients had medial fibromuscular dysplasia, and one female patient had congenital narrowing of the renal arteries. Eight of the nine patients had associated hypertension; their highest recorded diastolic blood pressures ranged from 110 to 154 mm Hg. The initial management of PTA in the four patients with fibromuscular dysplasia and the patient with congenital renal artery narrowing had been attempted at other medical facilities. Initial PTA treatment. Through a review of the patients' records from referring physicians and an examination of pre-pta angiographic studies, we could determine each patient's status before the initiation of the interventional therapy (Table I). Four patients were initially evaluated for PTA treatment of medial fibromuscular dysplasia. In each, the dysplastic disease extended distally to involve at least the origins of the renal artery branches as exemplified in Fig. 1. Bilateral disease was present in all four pa- tients, and PTA was attempted bilaterally in three. Indications for PTA were treatment of hypertension in three patients; the fourth patient was normotensive but was empirically treated with PTA at another medical facility. The congenital lesions were present in a 17-yearold girl and consisted of proximal stenosis of both the right and left renal arteries as well as tubular narrowing of the proximal infrarenal abdominal aorta (Fig. 2). Initial treatment with PTA was undertaken to manage accelerated hypertension. Four of the patients were initially treated for orificial atherosclerotic lesions of the renal arteries. Two patients had bilateral disease: one had undergone contralateral aortorenal bypass 6 years previously; the other had total occlusion of the artery to an atrophic, nonfunctioning contralateral kidney. Election of PTA as the treatment in all four patients was made on the basis of the difficulty of blood pressure control and the perceived increased risk of operative management because of advanced ischemic heart disease. Consequences of PTA. The initial and subsequent events after PTA in the nine patients are summarized in Table I. Immediate beneficial blood pressure responses to PTA were reported in one of the

3 Volume 6 Number 3 September 1987 Operative repair after failed PTA 303 Fig. 1. Preliminary arteriogram taken before attempted PTA demonstrates involvement of the distal main renal artery and proximal branch vessel. three hypertensive patients with medial fibromuscular dysplasia and in two of the four patients with atherosclerotic lesions. However, hypertension recurred in these patients at 1, 2, and 4 months, respectively. Repeat PTA was performed in one of these patients without improvement in hypertension or in the appearance of the lesion. Hypertension remained unchanged (three patients) or became worse (two patients) in the remaining hypertensive patients. Two patients with medial fibromuscular dysplasia had anatomic complications of PTA. One, who had undergone simultaneous bilateral PTA, had perforation of the right renal artery with development of a large retroperitoneal hematoma (Fig. 3) and bilateral dissections with total right renal artery occlusion and left branch renal artery occlusion (Fig. 4). The second, a normotensive patient with medial fibromuscular dysplasia (case 4) having undergone attempted empirical PTA, had a renal artery aneurysm (Fig. 5). The patient with congenital proximal renal artery stenoses had a clinical course marked by improved blood pressure control at the time of discharge from the hospital yet rapid worsening of hypertension immediately after discharge. During the PTA procedure, the angiographer had described a relative indistensibility of the stenotic areas. Repeat PTA attempted 3 months later was equally unsuccessful. One of the four patients with ostial atherosclerotic stenosis had immediate thrombosis of the vessel resulting from dissection of the plaque. Associated flank pain and tenderness prompted emergency op- Fig. 2. Preoperative arteriogram of a 17-year-old girl who had undergone two unsuccessful trials of PTA for treatment of congenital stenosis of the left renal artery. eration and renal revascularization. Each of the three remaining patients treated for atherosclerotic lesions had deteriorating renal function as a consequence of I~A. The first patient had a nonfunctioning contralateral kidney before PTA. The serum creatinine had been 2.3 mg/dl before PTA but increased to 3.3 mg/dl during the ensuing 5 months. The second patient had had 46% of total renal function in the involved kidney before PTA; 4 months later, the kidney had lost more than 1 cm in length and had only 7% of total excretory function when reassessed with isotope renography. The final patient had had serum creatinine of 1.2 mg/dl before PTA; 2 month later the level had risen to 2.3 mg/dl. In all three of these patients, repeat arteriography demonstrated severe stenosis or total occlusion of the previously dilated vessel. Operative management. The operative procedures performed on the nine patients are summarized in Table II. Twelve kidneys underwent revascularization: saphenous vein aortorenal bypass was used for nine reconstructions in seven patients; hypogastric arterial autografts were used for bilateral reconstructions in the patient with congenital lesions; and thromboendarterectomy with patch angioplasty and contralateral nephrectomy was employed in the ninth patient. Multiple branch reconstructions were required for revascularization of four kidneys. Ex vivo reconstruction for multiple branch repairs was used in one patient. Table II also summarizes the effect of the failed management by PTA on the type of operative procedure required to treat the resulting renovascular

4 304 Dean et al. Journal of VASCULAR SURGERY Fig. 3. Computed tomographic study obtained 2 weeks after right renal artery perforation in case 1 demonstrates an extensive hematoma from retroperitoneal hemorrhage. age two vessels with medial fibrommcular dysplasia when review of the pre-pta arteriograms and the operative findings suggested that distal bypass alone could have been used originally. The kidney requiring ex vivo reconstruction could have been treated by in situ repair before the failed PTA attempt. The patient who required emergency thrombectomy, distal intimal repair, and aortorenal bypass could have been managed before PTA by either bypass or thromboendarterectomy alone. Finally, all surgical procedures were prolonged by the increased difficulty encountered in safely exposing the involved vessels because of the presence of intense perivascular inflammation and scarring. Fig. 4. Arteriogram taken 2 weeks after attempted bilateral PTA for bilateral medial fibromuscular dysplasia (case 1). Note the main right renal artery occlusion and the inferior branch occlusions on the left as a consequence of dissection of the vessel wall. Ex vivo reconstruction was required to manage the left renal artery disease. disease. The complexity of renal reconstruction was significantly affected by PTA in 5 of the 12 procedures. Multiple branch repair was required to treat an iatrogenically produced renal artery aneurysm when otherwise no intervention would have been required to treat the medial fibromuscular dysplasia in this normotensive patient. Similarly, multiplebranch in sire reconstructions were required to man- RESULTS There was no perioperative morbidity or deaths associated with the surgical procedures, and followup evaluations ranged from 2 to 54 months. Classification of blood pressure response was made on the basis of previously described criteria.7 Two of the three hypertensive patients with medial fibromuscular dysplasia and the patient with congenital lesions were cured of hypertension by operative treatment. The remaining hypertensive patient with medial fibromuscular dysplasia and three of the four patients with atherosclerotic lesions had significant postoperative improvement in blood pressure control. The remaining hypertensive patient with an atherosclerotic lesion received no blood pressure benefit but

5 Volume 6 Number 3 September 1987 Operative repair a#er failed KFA 305 had undergone emergent operation to correct acute dissection, renal artery thrombosis, and threatened renal infarction. Interestingly, this patient had had nonlateralizing renal vein renin assays before undergoing PTA. The effect of PTA and operation on azotemia is seen in Table III. Five patients had had significant elevations of serum creatinine ranging from 1.7 to 3.6 mg/dl (mean 2.5 mg/dl) after PTA. Improvement in sertun creatinine was seen after operation in all five; mean postoperative value was 1.5 mg/dl. DISCUSSION "It is of use, from time to time, to take stock, so to speak, of your knowledge of a particular disease to see exactly where we stand in regard to k, to inquire to what conclusions the accumulated facts seem to point...,8 These words were spoken by Wiliam Osier s in his 1887 Gulstonian Lecture on malignant endocarditis. They would be no less appropriate if used today to examine the current status of renovascular disease and RVH. Although the experiment by Goldblatt et al. 9 that conclusively demonstrated a renovascular cause of hypertension was published 53 years ago and the first intervention to manage such hypertension was performed 4 years later, l little uniformity of opinion had evolved regarding any aspect of the disease. This controversy includes criteria for patient selection for evaluation, the preferred sequence of diagnostic studies, and the most appropriate method of treatment. The most recent contribution to this controversy was the introduction of coaxial PTA by Grfintzig et al.2 By this procedure, a stenosing lesion of the renal artery can be opened by a transluminal balloon catheter inserted percutaneously with the patient under local anesthesia, and no more than 2 to 3 days of hospitalization will be required. The potential merits of this technique include reduced patient discomfort and temporary disability, avoidance of the risks of general anesthesia and a major operation, and containment of escalating health care costs. Furthermore, proponents argue that if this technique is unsuccessful, one still has the option of more major intervention, that is, operative treatment. Clearly, the simplicity and favorable early results of PTA in selected patients has led to widespread enthusiasm for its value. During the 9 years since its introduction, PTA has been employed to manage a spectrum of anatomic lesions affecting the renal artery. Although exploration of the breadth of application of PTA has been appropriate, this exploration Fig. 5. Arteriogram of a 49-year-old normotensive woman demonstrates a distal renal artery aneurysm that appeared after empirical PTA for asymptomatic medial fibromuscular dysplasia of the right renal artery. has led to use of the procedure for lesions unlikely to be successfully dilated and in circumstances susceptible to the production of complications. Operative treatment has remained our primary method of intervention. This choice is based on our experience with the management of more than 750 patients with RVH and the current results of operation, which include an operative mortality rate of less than 1%, technically successful revascularization in more than 97% of patients, and a permanent beneficial reduction in blood pressure in more than 95% of patients. 1 Since our use of PTA has been limited by this choice, and since most of our patients undergoing operation after PTA were referred from a diverse group of physicians and radiologists, no calculation of complication rates or failure rates of PTA can be obtained from our experience. Nevertheless, through a review of reported experience with PTA and our observations from the operative management of unsuccessful PTA, we can help to fbrmulate recommendations regarding the indications for preferential use of PTA in the treatment of RVH. Our successful experience with the operative treatment of the failures and complications of PTA does not a priori rebut the contention that, if PTA is unsuccessful, operative treatment can be employed that will yield results similar to those obtained when operative treatment is employed as the primary, pro-

6 306 Dean et al. Journal of VASCULAR SURGERY Table II. Effect of failed PTA on operative selection Procedure feasible before failed Patient PTA Operation required after failed PTA 1 Bilateral saphenous vein graft 2 Bilateral saphenous vein graft 3 Right saphenous vein graft 4 No procedure required 5 Bilateral hypogastric autograft 6 Left saphenous vein graft 7 Right saphenous vein graft 8 Left nephrectomy; right endarterectomy 9 Right saphenous vein graft Right in situ branch reconstruction; left ex vivo branch reconstruction Right in situ branch reconstruction; left in situ branch reconstruction Right in situ branch reconstruction Right in situ branch reconstruction Right in situ branch reconstruction Left thrombectomy, distal intimal repair; saphenous vein graft Emergent right thrombectomy, distal inrimal repair; saphenous vein graft Right thromboendartcrectomy with patch angioplasty, distal intimal repair; saphenous vein graft; left nephrectomy Right thrombcctomy, distal intimal repair; saphenous vein graft Table III. Effect on intervention on renal function Serum creatinine (mg/ dl) Patient Pre-PTA Part-PTA Postoperative Mean cedure. However, prolongation of the procedure, increased difficulty of dissection, and the requirement for a more complex intervention were common experiences in our secondary procedures. When applied to a larger population of patients and surgeons, these characteristics of post-pta surgery unquestionably will lead to inferior results. Examination of the types of lesions managed in our post-pta group provides some insight into patient selection for initial PTA. Reported experience with PTA of fibrodysplastic lesions has results most similar to the operative results. Beneficial blood pressure responses have been reported to be as high as 100% ~ after PTA in properly selected cases; although vessel perforation, hemorrhage, and branch occlusions have been reported after PTA, 3-6 their incidence has been less than 5%32 One would anticipate that such complications would be most likely in patients with diffuse fibromuscular dysplasia affecting both the distal main renal artery and its branches. Certainly, this was the variety of disease represented in our post-pta surgical group. Similarly, the cure rate of fibromuscular dysplasia after PTA, even when the procedure is performed by experienced persons, varies from 37% 11 to 51%3 s In contrast, 77% of patients with fibrodysplastic lesions on whom we have operated have been cured of hypertension) Having routinely found that angiography underestimates the distal extent of fibromuscular dysplasia, we believe this difference in cure rates represents residual, inadequately managed disease in the PTA group. For these reasons, we believe that fibromuscular dysplasia extending to the branch level is best managed primarily by operation and that PTA should be reserved for the subgroup of medial dysplastic lesions clearly limited to the main renal artery. The 17-year-old patient referred for operation of failed PTA for congenital stenoses underscores the improbability of permanent benefit of PTA in this group. Such lesions occurring in children are usually discrete narrowings and therefore would appear ideal for PTA. However, the stenofic area is commonly a congenital narrowing of the entire vessel wall and is predominantly composed of elastic tissue. When such a vessel is subjected to PTA, a return to the original diameter follows dilatation, or if the vessel has been overdistended, rupture of the entire vessel wall is the likely end result. Therefore, we believe that PTA is an inappropriate method of intervention in this group and that success is best achieved by operative correction. The four patients who required operation after PTA for atherosclerotic lesions exemplify the problem with this technique when it is applied to orificial atherosclerotic lesions. In each of three instances of early benefit, follow-up was characterized by rapid recurrence of hypertension and recurrent stenosis. A review of series reporting results of PTA for atherosclerotic lesions dramatizes the frequency of this end result. Miller et al) 4 reported that only 45% ofostial and mixed lesions were improved after 6 months;

7 Volume 6 Number 3 September 1987 Operative repair after failed IYFA 307 Sos et al.13 reported only a 14% benefit rate when bilateral ostial lesions were treated. These results show that PTA has little value in the treatment of this variety of lesions, for one must accept the risks of cholesterol embolization, vessel thrombosis, and loss of renal function while expecting only a minimal chance for prolonged benefit. Certainly the patient in our series who lost almost all renal function in the involved kidney within 4 months of PTA underscores this potential outcome. Finally, the patient in whom a post-pta aneurysm developed requires special comment. Although the creation of a traumatic aneurysm by PTA is not peculiar to our experience, this patient most appropriately demonstrates a disturbing trend regarding diagnostic study and interventions. The perceived simplicity of PTA has led to its empirical implementation in many centers to correct anatomic disease without proof of the clinical or functional significance of that disease. We believe this to be an unfortunate consequence of the introduction of PTA, and such liberal application of the procedure to be unjustified. To summarize, experience with the liberal use of PTA has helped to clarify its role as one of the therapeutic options in the treatment of RVH, but data now accumulated argue for its selective application. In this regard, PTA of nonorificial atherosclerotic lesions and medial fibrodysplastic lesions limited to the main renal artery yields results comparable to the results of operation if carried out by those experienced in this technique. In contrast, the use of PTA for the treatment of congenital lesions, of fibrodysplastic lesions involving renal artery branches, and of ostial atherosclerotic lesions is associated with inferior results and increased risk of complications. For this reason, we believe that operation remains the initial treatment of choice for patients in these latter groups, and that the decision for therapy for RVH by PTA must be individualized. REFERENCES 1. Dean RH. Renovascular hypertension. In: Moore WS, ed. Vascular surgery: a comprehensive review. New York: Grune & Stratton, Inc, 1983: Griintzig A, Vetter W, Meier B, Kuhlmann U, Ltitolf U, Siegenthaler W. Treatment of renovascular hypertension with percutaneous transluminal dilatation of a renal artery stenosis. Lancet 1978;1: Puijlaert CBAJ, Mall WIrI'M, Rosenbusch G, van Straalen AM, Klinge J, Feldberg MAM. Delayed rupture of renal artery after renal percutaneous transluminal angioplasty. Radiology 1986;159: Perry MO. Intramural dissection of superior mesenteric artery. A complication of attempted renal artery balloon dilation. J VASe SURG 1985;2: Mills SR, Wertman DE Jr, Grossman SH. Renal cortical arteriovenous fistula complicating percutaneous renal angioplasty. AJR 1981;137: Laerum F, Castafieda-Zufiiga WR, Amplatz KA. Complications of transluminal angioplasty. In: Castaa3eda- Zufiiga WR, ed. Transluminal angioplasty. New York: Thieme-Stratton, Inc, 1983: Maxwell MH, Bleifer KH, Franklin SS, Varady PD. Cooperative study on renovascular hypertension: demographic analysis of the study. JAMA 1972;220: Osier W. The Gulstonian Lectures, on malignant endocarditis. Br Med J 1885;7: Goldblatt H, Lynch J, Hanzal RF, Summerville WW. Studies on experimental hypertension; production of persistent elevation of systolic blood pressure by means of renal ischemia. J Exp Med 1934;59: Leadberter WF, Burkland CE. Hypertension in unilateral renal disease. J Urol 1938;39: Tegtmeyer CJ, Kellum D, Ayers CA. Percutaneous transluminal angioplasty of the renal artery. Results and long-term follow-up. Radiology 1984;153: Martin LG, Casarella WJ, Alspaugh JP, Chuang VP. Renal artery angioplasty: increased technical success and decreased complications in the second 100 patients. Radiology 1986; 159: Sos TA, Pickering TG, Sniderman K, et al. Percutaneous transluminal renal angioplasty in renovascular hypertension due to atheroma or fibromuscular dysplasia. N Engl J Med 1983;309: Miller GA, Ford KK, Braun SD, et al. Percutaneous transluminal angioplasty vs. surgery, for renovascular hypertension. AJR 1985;144: