Effects of Global Renal Artery Stenting on Chronic Renal Failure

Introduction

The natural history of atherosclerotic renal artery stenosis is characterized by progression, causing hypertension and chronic renal failure (RF). Up to 21% of patients with renal stenosis reducing luminal diameter by more than 60% progress to occlusion within 2 years. Renal stenosis is responsible for renal failure in 15% of adult patients who begin dialysis each year.^1–3 Renal angioplasty⁴ with renal artery stenting (RAS)^5–8 has become the procedure of choice in the treatment of stenosis. The technical success and low complication rates after percutaneous balloon angioplasty and stenting of renal stenosis have been widely published,^9,10 but despite these favorable reports, the ability of revascularization to improve the function of the treated kidney has not been clearly demonstrated.^11–13 As underlined by some researchers,^11,14–15 it is difficult to study the direct effect of treatment upon a single kidney because we lack simple methods to assess split function in this normally paired organ system. Serum creatinine is the more accepted method to evaluate RF, but reflects the function of the normal bilateral organs. Thus, in unilateral renal artery stenosis, an abnormal creatinine level suggests dysfunction of the contralateral “non-stenosed” kidney and obscures the direct effect of treatment of unilateral disease. To overcome this problem and isolate the direct effects of stenting, we studied a group of patients that presented stenoses of all remaining renal arteries (global renal ischemia). We evaluated the effects of stenting on overall RF during a medium-term period, using a longitudinal method to establish effective change in RF (i.e., slopes of reciprocal of creatinine over time).

Subjects and Methods Patient selection. All patients with renal artery stenosis, presenting through the nephrological and radiological departments, were evaluated for inclusion. Patients were considered eligible if they had a chronic renal impairment (serum creatinine > 1.5 mg/dL) and atherosclerotic renovascular disease with demonstrable stenosis (> 70% luminal diameter established by angiography with aid of a quantitative computerized software). Significant stenosis had to involve all renal arteries or, if the patients had unilateral stenosis, it must be in the setting of a solitary or single functional kidney. A nonfunctional kidney was defined as one being atrophic (pole-to-pole length 8 Patients with artery stenosis involving only one of two functioning kidneys were excluded because of the difficulty in establishing the cause for change in overall RF after unilateral intervention in these patients.

Procedure. Angiography, angioplasty and RAS were usually performed transfemorally (in two cases transaxillary). Heparin was administered (3000–5000 U) during the procedure, before angioplasty. To minimize contrast dose and reduce catheter manipulation, the renal arteries were localized by use of a pigtail catheter in the aorta, and flush injections of contrast diluted 1:1 with normal saline were used. During the stenting procedure, mean radiographic contrast dose was 48 ± 12 mL for the kidney. Only subocclusive stenoses were predilatated with balloon catheters. We usually performed direct stenting utilizing premounted stents (Corinthian and Genesis, Cordis Corporation, Miami Lakes, Florida or Express South Dakota, Boston Scientific Corporation, Maple Grove, Minnesota), with diameters ranging from 4–6 mm. In initial cases (n = 6), nonarticulated Palmaz stents (Cordis Endovascular, Warren, New Jersey) were mounted on appropriately-sized balloon catheters (5–7 mm) and deployed at 8–12 atm. After stenting was performed, aspirin (325 mg daily) was continued indefinitely.

Follow-up. The primary endpoint of the study was to assess the changes of RF before and after stenting by comparing the slopes of the reciprocal of serum creatinine versus time. Patients were evaluated at 1, 3, 6, 9 and 12 months, and every 6 months thereafter. Review involved clinical examination and measurement of serum creatinine. Follow-up was considered adequate only if at least 5 serum creatinine measurements were available during a post-stent 9-month period. Only patients with adequate prior documentation of the course of RF were included in this study, i.e., patients having at least 5 values of serum creatinine during the previous 6-month period. Renal color Doppler ultrasonography was performed at baseline at 1 and 6 months, and then yearly. Renal size (RS), assessed by ultrasonographic measurement of pole-to-pole length, and peak systolic velocity, measured along the trunk of renal artery and in the intra-stent segment, was evaluated. The effect of RS on renal dimension was assessed by comparing baseline with final follow-up kidney length. According to other authors,⁸ atrophy was defined as a reduction > 1 cm in kidney length during follow-up; a nonfunctional kidney was defined as one being atrophic (long diameter 16 Restenosis (> 70%) was suggested by a value of intra-stent or extra-stent velocity > 180 cm/s. In this case, angiography and angioplasty were performed. Statistical analysis. Lines of best fit for reciprocal of serum creatinine versus time were created retrospectively (before stent deployment) and prospectively (after stent deployment) for each patient by using a computer least-squares linear regression analysis.¹⁷ The degree of correlation between reciprocal of serum creatinine and time was assessed by calculation and testing significance of the angular coefficients.

In addition, we tested significant differences between the slopes of the regression lines before and after stenting for each patient by using a 2-tailed paired Student’s t-test. Comparisons of pole-to-pole kidney lengths and RI before and after stent deployment were also assessed by using a 2-tailed paired Student’s t-test. One-way analysis of variance was used to analyze possible predictive factors (pre-stent serum creatinine, kidney length and RI) with respect to the outcome variable; when significant, we used the Scheffè multiple comparison test to select different means. Plus minus values are mean ± 1 standard deviation, unless otherwise indicated.

Results screening and baseline characteristics. Between December 1996 and December 2004, our interventional radiology department performed a successful treatment of 93 stenosed renal arteries with stent deployment in 77 patients. This report includes data on 30 of these patients with successful stenting who showed chronic renal impairment and global obstructive atherosclerotic renovascular disease, and who completed the planned follow-up. These patients underwent bilateral RAS (12 patients) or unilateral stenting in the presence of a solitary (n = 3) or single-functioning kidney (n = 15). Three other patients did complete follow-up and were excluded from the analysis (these patients died within 6 months of the time of intervention, 1 from myocardial infarction and 2 from congestive heart failure). The remaining patients were excluded from the study because they underwent RAS only for malignant hypertension or because they had unilateral RAS in a setting of a non-diseased contralateral kidney. During follow-up, duplex sonography identified restenosis in 7 renal arteries in 5 patients. All patients who underwent renal angiography and restenosis (> 50% reduction in lumen diameter stenosis) were rapidly treated successfully with repeated balloon angioplasty. The mean creatinine level of the patients immediately preceding intervention was 3.11 ± 1.98 mg/dl (range 1.60–8.50). The median duration of follow-up was 24 months (range 9–108).

Outcomes. Before stent deployment, all patients exhibited a negative slope of regression lines, indicating progressive renal insufficiency. The mean slope was -0.0249 ± 0.0527 dL · mg-1 · month-1 before intervention and 0.0050 ± 0.0194 dL · mg-1 · month-1 after stent deployment.

Discussion

Renal artery stenosis is a progressive condition, and the disease itself can only worsen without treatment. Renal atherosclerotic obstructive disease progresses in 51% of cases within 5 years.¹⁸ Severe occlusive disease of the extraparenchymal renal artery results in chronic renal ischemia, and can lead to the development of an “ischemic atrophic nephropathy” and chronic renal insufficiency. Considering the progressive nature of renal stenosis,¹⁹ it is attractive to think that revascularization by stenting in such patients slows renal atrophy and progression to end-stage renal failure. For these reasons, RAS has been developed over the last 10 years. This technique shows a high grade of technical success, ranging from 95–100%. The present study isolated the effects of RAS on RF, other renal parameters (size and RI), and in a patient group with renal impairment and global obstructive renovascular atherosclerosis. The study of patients in whom all renal arteries were obstructed ensured the presence of “global” renal ischemia and isolated the potential effects of such intervention, without interference from the contralateral “healthy” kidney.¹⁵ The principal endpoint of this study was to evaluate the efficacy of RAS in respect to the natural progression toward terminal renal failure. The great percentage of previous studies that have assessed the effect of angioplasty and stenting of stenosed renal arteries on RF have used isolated creatinine measures before and after stent deployment.^{4–7,9,10,20,21} This single, static method presents a substantial percentage of error because it depends on many factors (i.e., the time of choosing the value, the particular value chosen, the inter and intra-assay variability of the parameter). These limitations can be improved by dynamically assessing RF on the basis of the slope of the regression line, created from the plot of the reciprocal of creatinine versus time. This method is more reliable than the evaluation of a single creatinine in detecting significant changes of RF.^8,17

However, this methodology depends on the assumption that the reciprocal of creatinine is linear with time in patients with chronic renal failure. In our study, using this technique, we have documented stabilization or improvement of RF in more than 76% of patients. These data are similar to those obtained by other authors.^20,21 According to previous studies, improvement or stabilization of RF is not universal in our study, perhaps reflecting the contribution of some patients with advanced parenchymal vascular disease. We do remember that renovascular disease is a complex and multi-factorial pathology which depends on patient age and on the presence of metabolic disease (i.e., diabetes mellitus, dyslipidemics), endothelial dysfunction, hypertensive nephroangiosclerosis and interstitial fibrosis, factors that can be aggravations or consequences of nephropathy, and are not always ameliorated after RAS.^11,22 Close to 24% of our patients showed a persistent decline of RF. It is interesting that a baseline serum creatinine is a negative predictor of improved RF after RAS, according to the results of other researchers.²¹ Ultrasound data demonstrated the preservation of renal size over time after stenting. As expected, decrease of renal length was observed in patients showing negative slopes. In our study, renal atrophy occurred in 4 of 7 cases in Group 3. We also considered pre-stent RI as a predictive parameter, to check the observation that a high RI is a bad predictor of stable renal function.²³ However, in our group, we were not able to demonstrate a significant association between pre-stent high RI and declining RF. Of course, after stenting, mean interlobar RI values increased in all groups. This result is not surprising; by eliminating the obstruction, we eliminate the phenomenon of pulsus tardus et parvus (i.e., delayed and reduced systolic peak) and observe an increase in the renal flow, while cortical resistances remain high. This means that a substratum of irreversible diffuse parenchymal angiosclerosis is often present and may explain the percentage of patients that do not respond to RAS. The present prospective, which is not a randomized study, has within it the limitations inherent in such a design; in particular, a control group is not present. However, we used each patient course before treatment as an individual internal control of untreated patients. We believe that it is necessary to plan randomized trials to establish the advantage of RAS over medical treatment in progression of renal failure. The only randomized trial²⁴ comparing balloon angioplasty versus medical treatment did not find any advantage of angioplasty with respect to RF. However, a high proportion of crossovers and the lack of using stenting hampered interpretation of that study. A new randomized study comparing RAS and medical treatment has been announced.²⁵

In conclusion, the treatment of stenosed renal arteries with RAS can be accomplished with favorable medium-term survival in patients with chronic renal impairment and global obstructive atherosclerotic renovascular disease. Revascularization stabilizes RF and preserves kidney size in a majority of patients. If we consider both stabilization and improvement in RF as indicators of clinical success, 76% of patients saw an advantage after revascularization.