Prophylactic repair of renal artery stenosis is not justified in patients who require infrarenal aortic reconstruction☆☆☆★

Article Outline

• Abstract
• PATIENTS AND METHODS
  • Patients
  • Statistical analysis
• RESULTS
• DISCUSSION
• DISCUSSION
• References
• Copyright

Abstract 

Purpose: Simultaneous prophylactic repair of asymptomatic renal artery stenosis (ARAS) in patients who require infrarenal aortoiliac reconstruction is controversial. This study documents the natural history of ARAS in patients who require aortic reconstruction. Methods: Two hundred patients who required aortic reconstruction from 1985 to 1990 for indications other than hypertension or renal salvage were identified. ARAS was not repaired. Preoperative angiograms were available for 171 of 200 patients and were reviewed for renal artery stenosis. Patients were assessed for atherosclerotic risk factors, survival, preoperative and follow-up blood pressure, serum creatinine level, antihypertensive medication usage, and need for dialysis. Results: The mean duration of follow-up was 6.3 years. Twenty-four of 171 patients (14%) had preoperative unilateral 70% to 99% diameter reduction ARAS, and eight (5%) had bilateral 70% to 99% ARAS. Clinical features associated with ≥70% ARAS included coronary artery disease, increased age, and a diagnosis of hypertension (p < 0.05). Patients with ≥70% ARAS did not have a decreased 7-year survival rate (66% vs 84%; p = 0.10) but had higher systolic blood pressures (153 ± 25 vs 138 ± 30 mm Hg; p < 0.05) as well as increased numbers of antihypertensive medications at follow-up (1.1 ± 0.2 vs 0.7 ± 1; p < 0.05). The mean serum creatinine level (1.1 ± 0.3 preoperative vs 1.4 ± 0.8 mg/dl; p = NS) was not increased. One patient (0.58%) with polycystic kidney disease and minimal renal artery stenosis required dialysis. Conclusions: High-grade ARAS in patients who are undergoing infrarenal aortic reconstruction is associated at late follow-up with increased systolic blood pressure and a need for increased numbers of antihypertensive medications, but not decreased survival rate, dialysis dependence, or an increase in serum creatinine level. These data do not support renal artery repair in patients with ARAS who undergo infrarenal aortic reconstruction. (J Vasc Surg 1998;28:14-22.)

Management of asymptomatic renal artery stenosis (ARAS) that is incidentally discovered on an aortogram before aortic vascular repair remains controversial. Some studies have reported increased morbidity rates associated with prophylactic renal artery revascularization in patients who undergo infrarenal aortic reconstruction, whereas others have not.¹, ² Serial studies with duplex ultrasound have indicated a 40% progression rate of renal artery stenosis at 2 years.³ Tollefson et al.⁴ used serial renal angiography to document progression of renal artery stenosis in 50% of patients over 4.5 years with a 15% incidence of renal artery occlusion. Functional progression of ARAS in terms of hypertension and progressive renal failure is less well documented. This is especially important in patients who require infrarenal aortic reconstruction, in whom prophylactic repair of ARAS is often considered. The purpose of this study was to evaluate the functional outcome of ARAS that is detected on a routine preoperative aortogram in patients who require reconstruction of the infrarenal aorta for occlusive or aneurysmal disease.

Back to Article Outline

PATIENTS AND METHODS 

Patients 

Two hundred consecutive patients from the operative logs and vascular registry of the Portland Veterans Affairs (VA) Hospital who required infrarenal aortic reconstruction between 1985 and 1990 and whose records did not document an indication for renal artery reconstruction (suspected renovascular hypertension or renal insufficiency) were identified. Patients in whom renovascular disease was clinically suspected and who were treated for renal artery stenosis were excluded. Of the remaining patients, 171 had preoperative aortograms that were available for evaluation of renal artery stenosis. None of the 171 patients had renal artery stenosis treated at the Portland VA Hospital during the follow-up period. Charts were reviewed, and the patients' age, sex, type of operation, and indication for operation were recorded. Data were collected concerning preoperative blood pressure, antihypertensive medications, diabetes, creatinine level, tobacco use, coronary artery disease (history of angina or congestive heart failure or electrocardiographic evidence of previous myocardial infarction), angiographically demonstrated lower extremity arterial occlusive disease, and cerebrovascular disease (history of stroke or transient ischemic attacks). Follow-up was by chart review. Follow-up data were obtained on survival, serial systolic and diastolic blood pressures, creatinine level, the number and type of antihypertensive medications, and whether the patient had a need for renal dialysis.

Angiograms were reviewed separately, with the reviewer blinded to all preoperative and follow-up data. The degree of stenosis of each main and/or duplicated renal artery was recorded. Renal artery stenosis was measured with calipers comparing the width of the contrast column at the site of greatest luminal narrowing with the width of the contrast column in the distal renal artery at a site not obviously involved with poststenotic dilatation. Small accessory renal arteries were not considered.

Patients were grouped according to maximum renal artery stenosis present in any artery (0%, group A; 1% to 29%, group B; 30% to 69%, group C; 70% to 99%, group D [unilateral and bilateral]; and occluded, group E). Groups A and D plus E were compared for changes in preoperative and follow-up blood pressure, antihypertensive medications, and creatinine level.

Statistical analysis 

Multiple regression analysis was used to compare preoperative and postoperative variables in patients with varying degrees of ARAS. Survival analysis was performed using multivariate proportional hazards fit model. All statistical tests were performed with the statistics software package JMP (1995, SAS Institute, Cary, N.C.). Statistical services were provided through the General Clinical Research Center at Oregon Health Sciences University.

Back to Article Outline

RESULTS 

The mean duration of follow-up was 6.3 years (range, 1.0 to 10.6 years; median, 6.4 years) for the 171 patients enrolled in the study. Ninety-nine percent of the patients were male, and the mean age at aortography was 63 years. Preoperative demographic data are listed in Table I, A, and indications for operation in Table I, B.

Table I, A. Demographic features of 171 patients who required infrarenal aortic reconstruction

Table I, B. Indications for aortic reconstruction

*Includes infected graft.

The maximum percent diameter reduction of a main or duplicated renal artery in the 171 patients is listed in Table II.

Table II. Distribution of renal artery stenosis in the most severely narrowed renal artery in 171 patients who required infrarenal aortic reconstruction

Twenty-four patients had unilateral and eight patients bilateral 70% to 99% renal artery stenosis. The three patients with 100% renal artery stenosis had contralateral renal artery stenosis in the range of 50% to 70%. Survival data by group are demonstrated in a Kaplan-Meier curve in Fig. 1.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1. 

  Kaplan-Meier survival curve. Group A, no renal artery stenosis; group B, 1% to 29% stenosis; group C, 30% to 69% stenosis; group D + E, ≥70% stenosis. There was no significant difference between groups. Age and comorbidities are not addressed in this survival curve (see text and Table IX).

Table III lists the reconstructive procedures performed.

Table III. Reconstructive procedures in 171 patients with infrarenal aortoiliac aneurysmal or occlusive disease

One hundred twenty-seven of the procedures involved direct aortic reconstruction, and 37 were extraanatomic bypass grafting procedures. Other indications for operation include repair of pseudoaneurysms and removal of previously placed infected prosthetic grafts. The operative mortality rate was 1.2%; death was a result of myocardial infarction in both perioperative deaths.

Multiple regression analysis of the clinical features evaluated in this study indicated that coronary artery disease, a preoperative diagnosis of hypertension, and age were significantly associated with increased levels of renal artery stenosis at the time of aortography (Table IV).

Table IV. Association of renal artery stenosis present on preoperative angiogram with clinical features in 171 patients who required infrarenal aortic reconstruction

*By multiple regression analysis, adjusting for age, hypertension, diabetes, coronary artery disease, peripheral vascular disease, and cerebrovascular disease.

One-way analysis of variance between groups at the time of presentation indicated that groups did not vary in terms of diastolic blood pressure (DBP), creatinine level, and the number of antihypertensive medications taken (DBP, p = 0.566; creatinine level, p = 0.417; numbers of antihypertensives; p = 0.131). Systolic blood pressure, however, was different between groups at presentation (p = 0.024; Tables V to VIII).

Table VI. DBP before operation versus that at last follow-up in 171 patients who required infrarenal aortic reconstruction (mean follow-up, 6.3 years)

Data shown are mean ± standard deviation.

By multiple regression, p is 0.3149 for increased DBP over time and with increasing stenosis, considering clinical features of age, hypertension, coronary artery disease, cerebrovascular disease, diabetes, and peripheral vascular disease. See text for one-way analysis of variance between groups at preoperative time, last follow-up, and for change in DBP.

Table VII. Number of antihypertensive agents before operation versus that at last follow-up in 171 patients who required infrarenal aortic reconstruction (mean follow-up, 6.3 years)

Data shown are mean ± standard deviation.

By multiple regression, p is 0.0280 for increased use of antihypertensive medications over time and with increasing stenosis, considering clinical features of age, hypertension, coronary artery disease, cerebrovascular disease, diabetes, and peripheral vascular disease. See text for one-way analysis of variance between groups at preoperative time, last follow-up, and for change in use of antihypertensive medications.

There was no difference in patients' clinical variables between groups in terms of whether the indication for operation was aortoiliac occlusive disease or aneurysm. In addition, the distribution of renal artery stenosis in patients who underwent operation for aortoiliac occlusive disease or aneurysm did not differ.

Table V. Systolic blood pressure before operation versus that at last follow-up in 171 patients who required infrarenal aortic reconstruction (mean follow-up, 6.3 years)

SBP, Systolic blood pressure.

Data shown are mean ± standard deviation.

By multiple regression, p is 0.0236 for increased SBP over time and with increasing stenosis, considering clinical features of age, hypertension, coronary artery disease, cerebrovascular disease, diabetes, and peripheral vascular disease. See text for one-way analysis of variance between groups at preoperative time, last follow-up, and for change in SBP.

Table VIII. Creatinine level before operation versus that at last follow-up in 171 patients who required infrarenal aortic reconstruction (mean follow-up, 6.3 years)

Data shown are mean ± standard deviation.

By multiple regression, p is 0.1821 for increased creatinine level over time and with increasing stenosis, considering clinical features of age, hypertension, coronary artery disease, cerebrovascular disease, diabetes, and peripheral vascular disease. See text for one-way analysis of variance between groups at preoperative time, last follow-up, and for change in creatinine level.

At last follow-up, patients with ≥70% ARAS on their preoperative angiograms had higher systolic blood pressures (153.3 ± 25.2 mm Hg vs 148 ± 23 mm Hg; p = 0.0236) and used slightly more antihypertensive medications (1.54 ± 1.24 vs 1.14 ± 0.18; p = 0.0280) but had no increase in serum creatinine level (1.4 ± 0.8 vs 1.1 ± 0.3 mg/dl; p = 0.1821) or significant change in DBP (Table V, Table VIII). One-way analysis of variance between groups at last follow-up showed no significant difference in creatinine level and DBP (creatinine level, p = 0.6881; DBP, p = 0.2424). However, systolic blood pressure and numbers of antihypertensive medications were different between groups (systolic blood pressure, p = 0.0428; antihypertensive medications, p = 0.0353; Table V, Table VIII).

One of 171 patients followed-up became dialysis-dependent. That patient had polycystic kidney disease and minimal renal artery stenosis. None of the eight patients with bilateral 70% to 99% stenoses became dialysis dependent or had a worsening creatinine level or increased use of antihypertensive medications beyond the statistical means noted above.

Proportional hazards fit model identified diabetes and lower extremity arterial occlusive disease as clinical conditions that adversely influence the survival rate when adjusting for age and other clinical conditions. Increased renal artery stenosis was not associated with a decreased survival rate when other clinical features and age were considered (Table IX).

Table IX. Proportional hazards fit model to identify clinical features that predict survival in 171 patients who underwent infrarenal aortic reconstruction

Back to Article Outline

DISCUSSION 

One of the most debated issues regarding renal artery stenosis is how to address asymptomatic renal artery stenosis (ARAS) that is detected on a routine preoperative aortogram for patients who are undergoing aortic reconstruction. Management of ARAS is based on many factors, such as the anatomic natural history of renal artery stenosis, technical considerations that surround surgery, and ultimately, the clinical outcome as a result of renal artery stenosis progression with regards to renal insufficiency and hypertension. Some authors suggest that many “asymptomatic” lesions are actually mildly symptomatic and may contribute to the presence of hypertension. Indeed, we cannot exclude a renovascular component to the hypertension in our patients who had hypertension and renal artery stenosis; however, none were suspected of having renal artery stenosis before angiography, and other authors contend that even if a lesion is mildly symptomatic, given the attendant increased morbidity rate in those patients with significant comorbidities, repair should not be undertaken unless the patient has specific indications for renal artery bypass.¹, ⁵, ⁶

Cambria et al.¹ argue that patients with few comorbidities have equal morbidity and mortality rates when comparing aortic reconstruction alone with aortic reconstruction and renal artery repair. Cambria et al. refer specifically to patients with significant aortic disease and low cardiac morbidity. However, very few patients who require aortic reconstruction are likely to have low cardiac risk.¹ As many as 90% of patients who require abdominal aortic aneurysm repair have significant coronary artery disease as documented on coronary angiography; cardiac mortality rates range from 2% to 5% and morbidity rates range up to 11%.⁷, ⁸, ⁹ One report quotes mortality figures for combined aortic and renal construction at 5% to 31%.¹⁰ In addition, the majority opinion is clearly that in a typical patient who requires aortic reconstruction the addition of simultaneous renal artery repair results in overall increased perioperative morbidity rates.¹¹, ¹², ¹³, ¹⁴, ¹⁵ Given that the majority of patients are likely to have significant comorbidities that at least somewhat increase operative risk for simultaneous renal artery repair, one must carefully evaluate the merits of combining aortic reconstruction with renal artery repair for asymptomatic or perhaps minimally symptomatic renal artery stenosis.

Zierler et al.³ have provided some of the most specific information concerning the progression of anatomic renal artery stenosis detected by duplex ultrasound in patients who were evaluated for hypertension, decreased renal function, or both. Although such patients are not directly comparable with those described in this report, the results indicate that renal artery stenosis in such patients is often progressive. Seventy-six patients were followed-up for a mean of 32 months, and the cumulative incidence of progression from normal to ≥60% renal artery stenosis was 0% at 1 year, 0% at 2 years, and 8% at 3 years, whereas progression of a ≥60% renal artery stenosis was 30% at 1 year, 44% at 2 years, and 48% at 3 years. Seven percent of renal arteries with ≥60% stenosis progressed to occlusion. Progression occurred at a rate of 7% per year for all patients.³

Tollefson and Ernst⁴ also performed a retrospective analysis of 48 patients who were found to have renal artery stenosis on preoperative aortograms for aortic reconstruction. They evaluated follow-up aortograms to assess progression of renal artery stenosis and found a rate of progression of 4.6% per year, regardless of the initial degree of stenosis, and an overall renal artery occlusion incidence of 9% in patients with ≥60% stenosis initially. From this, the authors concluded that “preocclusive” renal artery stenosis (≥60%) should undergo “prophylactic” repair at the time of aortic reconstruction.⁴ However, because of the retrospective nature of this study, it is likely that these patients had indications for repeat aortography other than simply follow-up assessment of renal artery stenosis. Atherosclerotic disease in these patients may therefore be more progressive than in the general population of patients who require aortic reconstruction. In addition, Tollefson and Ernst noted changes in creatinine level and blood pressure did not correlate with stenosis progression.

Instead of following stenosis progression, our study focused on evaluating the clinical outcome in patients with untreated renal artery stenosis in whom renal artery disease was not suspected at presentation. It is our supposition that the central issue driving the decision to revascularize ARAS is whether stenosis leads to clinically significant future decline, that is, dialysis dependence, uncontrollable hypertension, or severe renal insufficiency. Whereas patients with high-grade renal artery stenosis may lose renal mass over time, this may not be of clinical significance in certain patient groups. It should be clearly demonstrated that ARAS leads to future uncontrollable hypertension, renal insufficiency, or dialysis dependence before repairing asymptomatic renal artery stenosis.

Our findings of a greater than 20% incidence of ≥70% renal artery stenosis in patients who undergo aortography before aortic reconstruction is consistent with other reports, as is the distribution of preexisting comorbidities.⁴ The distribution of procedures for our study group is also consistent with other reports; however, our patient population is strongly skewed to the male sex, reflecting the VA cohort.⁴ Statistical analysis using multiple regression, accounting for age, increased blood pressure, diabetes, coronary artery disease, lower extremity arterial occlusive disease, and cerebrovascular disease identified age, increased blood pressure, and coronary artery disease as being predictive for severity of renal artery stenosis. Other authors have remarked that multiple retrospective analyses have varied in identifying clinical features that predict stenosis, and most suggest that there is no reliable feature to predict stenosis.², ³, ⁴, ⁵, ⁶, ¹⁶

In long-term follow-up of blood pressure, creatinine level, and use of antihypertensive medications, we found by multiple regression analysis that took into account age, hypertension, diabetes, percent renal artery stenosis, lower extremity arterial occlusive disease, cerebrovascular disease, and coronary artery disease, that there is a significant increase in systolic blood pressure and the numbers of antihypertensive medications used in patients with ≥70% renal artery stenosis when compared with those who have no evidence of stenosis. It is important to note, however, that because the survival rate in this study was not affected by the presence of increased ARAS, the increase in systolic blood pressure and use of antihypertensive medications, albeit statistically significant (153 mm Hg vs 137 mm Hg, and 1.54 antihypertensives vs 0.92) likely does not have practical clinical significance in patients who require aortic reconstruction. Would the added risk of performing a combined renal artery revascularization and aortic reconstruction be worth reducing an already reasonably controlled systolic blood pressure? In addition, DBP and creatinine level did not change significantly during follow-up. Although patients with renal artery stenosis may lose renal mass without a change in creatinine level, the fact remains that only one patient in the study went on to require dialysis, and that patient had polycystic kidney disease with minimal renal artery stenosis.

It is important that the results of this study not be inappropriately extrapolated to other clinical settings. For instance, we have not addressed the issue of repair of symptomatic renal artery stenosis in conjunction with aortic reconstruction or repair of asymptomatic renal artery stenosis in the setting of suprarenal aortic surgery. The data presented here also do not imply that patients with ARAS will never have a deterioration in renal function, only that severe renal compromise must be quite rare in patients with ARAS who require aortic reconstruction who are observed for a median of 6.4 years. In addition, the small number of patients in this study who had bilateral high-grade renal artery stenosis or the combination of renal artery occlusion and high-grade renal artery stenosis precludes any definite conclusions regarding repair of asymptomatic renal artery stenosis in such patients who require aortic reconstruction. It is also possible that even longer follow-up of patients with high-grade renal artery stenosis may reveal more difficulty with renal insufficiency. One may argue that prophylactic transluminal angioplasty should still be considered in patients with asymptomatic renal artery stenosis who require aortic reconstruction; however, the benign natural history of ARAS in patients who require aortic reconstruction argues against any intervention for renal artery stenosis in such patients. Clearly, renal artery transluminal angioplasty of renal artery stenosis is not without added expense or potential complications. Overall, the data presented here and elsewhere do not support “prophylactic” repair of ARAS by any means in patients who undergo aortic reconstruction.

Back to Article Outline

DISCUSSION 

Dr. R. Eugene Zierler (Seattle, Wash.). Although the indications for renal revascularization in the setting of established renovascular hypertension generally are accepted, the role of intervention for renal artery stenosis in patients with controlled hypertension or mild degrees of renal insufficiency is not clearly established. This creates a therapeutic dilemma whenever renal artery lesions are discovered incidentally in patients who are evaluated for arterial disease elsewhere. Such situations are likely to arise frequently because the prevalence of significant renal artery stenoses in patients with abdominal aortic aneurysms or lower-extremity arterial occlusive disease is in the range of 30% to 40%. In the absence of a randomized clinical trial that compares early renal revascularization with medical therapy alone, clinical decisions must be on the basis of what is known about the natural history of renal artery stenosis. The critical issue is whether a high-grade but “asymptomatic” renal artery stenosis is likely to be associated over time with adverse consequences that could be avoided by early intervention.

This paper from the group at Oregon Health Sciences University sheds some light on this important issue. The authors conducted a retrospective review of 171 patients who underwent surgical procedures for abdominal aortic disease and had no compelling indications for renal revascularization at the time of their initial procedure. There were 32 patients with unilateral and 8 patients with bilateral 70% to 99% renal artery stenoses, which resulted in an overall prevalence for highgrade renal artery stenosis of 24%. In a mean follow-up period of 6 to 3 years, the patients with 70% to 99% renal artery lesions showed a trend toward decreased long-term survival compared with those with lesser degrees of renal artery disease. Those patients also had higher systolic blood pressures and required larger numbers of antihypertensive medications at the end of the follow-up period. However, there was no significant change in serum creatinine levels during the period of observation. The authors concluded that “prophylactic” repair of “asymptomatic” renal artery stenoses is not justified in patients who require surgical procedures for abdominal aortic disease.

At the University of Washington, we have an ongoing prospective study on the natural history of atherosclerotic renal artery stenosis that relies on serial duplex scanning to document the status of the renal arteries and kidneys. Patients are recruited from those screened for renovascular disease in the vascular laboratory. Our most recent published report was on the basis of 76 patients, with 132 eligible renal arteries that were followed for a mean period of 32 months. Progression of renal artery stenosis was relatively common, with an average rate of 7% per year for all categories of baseline disease combined and a cumulative incidence of 44% at 2 years for progression from less than 60% to 60% or more stenosis. The risk factors associated with renal artery disease progression were advanced age, increased systolic blood pressure, female gender, poor blood pressure control, and a history of carotid endarterectomy. We also have observed renal atrophy that occurs in patients with renal artery disease. On the basis of a decrease in kidney length of 1 cm or more, the cumulative incidence of renal atrophy at 2 years ranged from 6% for kidneys with normal renal arteries to 21% for kidneys with 60% or more renal artery stenoses. The risk factors associated with renal atrophy were elevated systolic blood pressure, high-grade renal artery stenosis, and decreased renal cortical blood flow velocities.

Although we are not ready to make any definitive recommendations, our natural history data suggest that a sub-group of patients with renal artery stenoses could benefit from early renal revascularization. Some of the factors that appear to be associated with adverse outcomes are high-grade renal artery stenosis, observed stenosis progression, renal atrophy, elevated systolic blood pressure, and decreased blood flow velocities in the renal parenchyma.

As the authors indicate, their patients and the patients in the University of Washington natural history study are not directly comparable. Their patients were selected because they were undergoing surgery for abdominal aortic disease, and our patients were identified through a screening program for renal artery stenosis. Consequently, our patients have a higher prevalence of hypertension, and the proportion of renal arteries with high-grade stenoses at baseline in our patient population is 46%; it is only 14% in their study. Our study population was also 47% male, and theirs was 99% male.

My only question for the authors of this paper concerns the way in which their follow-up data were analyzed and presented. Although it is stated that the mean follow-up interval is 6.3 years, no range or standard deviation is given, so the variability in follow-up among the patients is unknown. These patients were operated between 1985 and 1990, so it is possible that a few patients were followed for long periods and a larger number of patients had relatively short follow-up intervals. In our natural history study, the incidence of adverse events, such as end-stage renal disease, increased with the duration of follow-up. Therefore, the consistency and length of follow-up is a critical determinant of the reliability of the data. For the conclusions to be valid, a representative sample of patients must be followed for a sufficient period of time. In addition, the duration of follow-up must be equivalent for all patient subgroups. A life-table analysis is a more appropriate format for presenting this type of data.

With this in mind, can you tell us how much variability there was among the follow-up intervals for individual patients? Specifically, was the follow-up interval for patients in each of the renal artery stenosis categories equivalent? I would be particularly interested in the duration of follow-up for the patients with unilateral and bilateral high-grade renal artery stenoses because these are the patients that should be at the highest risk for ischemic nephropathy.

As a final comment, I would like to propose an alternative interpretation that leads to a different conclusion regarding the role of renal revascularization in this patient population. The authors noted decreased survival, increased systolic blood pressure, and the need for more antihypertensive medications in patients with high-grade renal artery stenoses who underwent abdominal aortic surgery, but they concluded that these findings did not support the need for concomitant renal artery repair. These observations, along with data from our own natural history study, indicate that patients with high-grade renal artery stenoses tend to have systolic hypertension that is both severe and difficult to treat pharmacologically. Systolic hypertension has emerged from numerous epidemiologic studies as one of the most prevalent and significant risk factors for cardiovascular disease. Even a modest degree of systolic hypertension is regarded as clinically important, and the threshold for defining hypertension has been lowered—the normal level is now less than 130/85 mm Hg. I offer the hypothesis that early renal revascularization in these patients would enable better blood pressure control over a long-term period and increased overall survival. Because death is a more frequent endpoint than renal failure in our natural history study, this increase in survival might allow patients to live long enough to develop ischemic nephropathy. So, do not prematurely give up the possibility that some of these patients still may benefit from early renal revascularization.

I compliment the authors on this very interesting and provocative study, and I appreciate the opportunity for discussion.

Dr. W. Kent Williamson. Thank you, Dr. Zierler, for your insightful comments. I would like to address these in order.

First, your question on variability and the follow-up in our study is an excellent point. The question actually brings up a fault and probably the shortcomings of a retrospective analysis primarily.

As I mentioned in the discussion, the follow-up years ranged from 1.1 to 10.6 years overall. I do not have specific data with regards to follow-up within each of the categories that I have outlined—groups A through D.

The best way to address this problem would be to increase the numbers in our study. We hoped that using 171 patients would help us attack that. However, it is obviously not perfect. The best way to address the problem may be to perform a prospective analysis overall.

With regard to your comments about the conclusion in the study, the bottom line or the hard end point of our study was, again, the need for dialysis. As I mentioned, only one patient or 0.6% of the study population became dialysis dependent, and that patient had minimal renal artery stenosis when first seen.

Some other issues needed to be discussed. You mentioned difficulty in controlling blood pressure, and that is also a matter of interpretation of the data. In my mind, an increased systolic blood pressure from around 137 mm Hg to a mean of 154 mm Hg may not necessarily represent a poorly controlled blood pressure. Again, though, that is a mean, and it does represent some high variability.

Also, the increased numbers of antihypertensive medications that were used may not have been clinically significant but were statistically significant. The overall number or mean number of antihypertensive medication used in group D, the 70% to 99% stenosis group at last follow-up, was around 1.5 to 1.7 antihypertensive medications.

So, at the time of initial aortic repair, is it really worth subjecting a patient to increased morbidity and mortality rates—which have been quoted as high as 31%—to save an extra antihypertensive medication?

Dr. Fred Weaver (Los Angeles, Calif.). This is a good paper, and the topic is important, particularly with the endoluminal grafting and the presence of concomitant renal artery stenosis.

Because the patient group that develops ischemic nephropathy usually has bilateral renal artery disease, did you look at and separate 10 patients who had a unilateral severe stenosis versus those patients who had a unilateral high-grade stenosis and contralateral renal artery disease greater than 50 percent? Did you analyze your data with regards to serum creatinine and hypertension?

Dr. Williamson. You make an excellent point. In fact, we did attempt to separate the groups on the basis of whether they had unilateral or bilateral 70% to 99% stenosis. We found the sample size, only 8 patients had bilateral 70% to 99% stenosis, was not enough to achieve statistical significance or power.

Therefore, we cannot make any reliable conclusions regarding the natural history of patients with bilateral greater than 70% to 99% stenosis.

Dr. Ron Stoney (San Francisco, Calif.). I enjoyed your paper, Dr. Williamson. I think the Oregon group is looking carefully at whether or not prophylactic renal artery repair is justified.

At the University of California at San Francisco, we have had a prophylactic group in each series of operations on the juxta- or pararenal aorta that we have reported. We saw no increase in morbidity rate in this group, compared with symptomatic patients, and yet, it is recognized that combined aortorenal reconstructions do increase risk for aortic reconstructions alone. Therefore, it is good to be selective.

I have one comment, which is similar to Dr. Zierler's comment. Some of these patients actually should be considered hypertensive and not asymptomatic on the basis of reported blood pressure elevations and the use of antihypertensive medications. Perhaps this could be woven into your reinterpretation of symptomatic and asymptomatic patients.

My other concern involves follow-up. In this population of patients, were you able to look at the angiographic status of disease in any patients over the time period that they were followed? Papers by Dr. Richard Dean and Dr. Calvin Ernst both describe progression of renal artery disease by imaging studies. If the disease does not progress, then how does symptomatic renal artery stenosis develop in patients?

Dr. Williamson. Both of your points are excellent. To address the first point about whether or not these patients are truly asymptomatic, the key feature of this study is that these patients were not identified with renal vascular disease or hypertension. In fact, at the time first seen, the patients with 70% to 99% stenosis did have higher systolic blood pressure and hypertension, but they were not felt to have indications for renal artery revascularization.

Other investigators have suggested that perhaps patients who are deemed asymptomatic are really symptomatic when they are placed under higher scrutiny. Does that necessarily mandate a revascularization at the time of surgery? We suggest that if it is that difficult to tell they have symptomatic renal vascular disease, then the increased risk of renal revascularization at the time of aortic resection is not justified.

With regard to your second question, we did not obtain or review follow-up aortograms. Dr. Towlson and Dr. Ernst are two investigators who have done a series of about 70 patients where follow-up aortograms were obtained for other reasons and then reviewed for progression of renal artery stenosis.

Those particular patient groups may not necessarily compare with our patient group because those patients are selected on the basis of the need for repeat aortography and, hence, may be a “sicker” group of patients.

We wanted to focus on the functional and clinical outcome rather than the anatomic outcome, which we think is actually the more important endpoint.

Back to Article Outline

References 

1. Cambria RP, Brewster DC, L'Italien G, Koustas G, Atamian S, LaMuraglia GM, et al.  Simultaneous aortic and renal artery reconstruction: evolution of an eighteen-year experience. J Vasc Surg. 1995;21:916–925
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (994 KB)
   • CrossRef
2. Stewart MT, Smith RB, Fulenwider JT, Perdue GD, Wells JO. Concomitant renal revascularization in patients undergoing aortic surgery. J Vasc Surg. 1985;2:400–405
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (520 KB)
   • CrossRef
3. Zierler RE, Bergein RO, Issacson JA, Strandness DE. Natural history of atherosclerotic renal artery stenosis: a prospective study with duplex ultrasonography. J Vasc Surg. 1994;19:250–258
   • View In Article
   • Abstract
   • Full Text
   • CrossRef
4. Tollefson DFJ, Ernst CB. Natural history of atherosclerotic renal artery stenosis associated with aortic disease. J Vasc Surg. 1991;14:327–331
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (429 KB)
   • CrossRef
5. Chaikof EL, Smith RB, Salam AA, Dodson TF, Lumsden AB, Kosinski AS, et al.  Ischemic nephropathy and concomitant aortic disease: a ten-year experience. J Vasc Surg. 1994;19:135–148
   • View In Article
   • Abstract
   • Full Text
   • CrossRef
6. Reilly JM, Rubin BG, Thompson RW, Allen BT, Flye MW, Anderson CB, et al.  Revascularization of the solitary kidney: a challenging problem in a high-risk population. Surgery. 1996;120:732–737
   • View In Article
   • Abstract
   • Full-Text PDF (578 KB)
   • CrossRef
7. Benjamin ME, Hansen KJ, Craven TE, Kieth DR, Plonk GW, Geary RL, et al.  Combined aortic and renal artery surgery: a contemporary experience. Ann Surg. 1996;223:555–567
   • View In Article
   • MEDLINE
   • CrossRef
8. Bunt TJ. The role of a defined protocol for cardiac risk assessment in decreasing perioperative myocardial infarction in vascular surgery. J Vasc Surg. 1992;15:626–634
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (733 KB)
   • CrossRef
9. Hertzer NR, Beven EG, Young JR, O'Hara PJ, Ruschhaupt WF, Graor RA, et al.  Coronary artery disease in peripheral vascular patients: a classification of 100 coronary angiograms and results of surgical management. Ann Surg. 1984;199:223–233
   • View In Article
   • MEDLINE
   • CrossRef
10. Brewster DC, Buth J, Darling RC, Austen WG. Combined aortic and renal artery reconstruction. Am J Surg. 1976;131:457–463
    • View In Article
    • MEDLINE
    • CrossRef
11. Yeager RA, Weigel RM, Murphy ES, McConnell DB, Sasaki TM, Vetto RM. Application of clinically valid cardiac risk factors to aortic aneurysm surgery. Arch Surg. 1986;121:278–281
    • View In Article
    • MEDLINE
12. O'Mara CS, Maples MD, Kilgore TL, McMullan MH, Tyler HB, Mundinger GH, et al. Simultaneous aortic reconstruction and bilateral renal revascularization: Is this a safe and effective procedure?. J Vasc Surg. 1988;8:357–366
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1717 KB)
    • CrossRef
13. McNeil JW, String ST, Pfiffer RB. Concomitant renal endarterectomy and aortic reconstruction. J Vasc Surg. 1994;20:331–337
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1315 KB)
    • CrossRef
14. Ernst CB. Surgical treatment of nonruptured infrarenal and juxtarenal aortic aneurysms. In: 3rd ed.  Ernst CB,  Stanley JC editor. Current therapy in vascular surgery. St. Louis: Mosby; 1995;p. 218–223
    • View In Article
15. Hansen KJ, Lundberg AH, Benjamin ME, Casey WJ, Craven TE, Bleyer AJ, et al. Is renal revascularization in diabetic patients worthwhile?. J Vasc Surg. 1996;24:383–393
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1243 KB)
    • CrossRef
16. Dean RH, Tribble RW, Hansen KJ, O'Neil E, Craven TE, Redding JF. Evolution of renal insufficiency in ischemic nephropathy. Ann Surg. 1991;213:446–456
    • View In Article
    • MEDLINE
    • CrossRef