- Email: [email protected]
Renal Artery Stent Placement for Restoration of Renal Function in Hemodialysis Recipients with Renal Artery Stenosis
Renal Artery Stent Placement for Restoration of Renal Function in Hemodialysis Recipients with Renal Artery Stenosis Mallik Thatipelli, MBBS, Sanjay Misra, MD, C. Michael Johnson, MD, James C. Andrews, MD, Anthony W. Stanson, MD, Haraldur Bjarnason, MD, and Michael A. McKusick, MD
PURPOSE: Renal artery stent placement to treat renal artery stenosis (RAS) in patients undergoing hemodialysis is not performed commonly. We present our outcomes of discontinuation of hemodialysis after treating patients with RAS by stent placement on acute (<30 days of hemodialysis) and chronic hemodialysis (>30 days). MATERIALS AND METHODS: A retrospective study was conducted in 16 patients (nine men) with an average age of 74.6 years ⴞ 10.6 (range, 49 – 86 y) who underwent treatment of 22 RAS for acute renal failure (n ⴝ 8) or uncontrolled hypertension with chronic hemodialysis (n ⴝ 8). The average follow-up was 448 days ⴞ 450 (median, 363.5 d; range, 6 –1,583 d). The primary outcome consisted of discontinuation of hemodialysis, death, and transplantation. Secondary endpoints included restenosis, changes in blood pressure (and use of medications), and estimated glomerular filtration rate (eGFR). RESULTS: After the stent procedure, eight patients were able to discontinue hemodialysis and remained free from dialysis over a mean period of 564 days ⴞ 533. The predictors of discontinuation of hemodialysis were 24-hour proteinuria, eGFR before renal artery stent placement, and size of the kidney on ultrasound studies (P < .05 for all three). There was no difference in patients who were undergoing acute versus chronic hemodialysis. There were three minor complications, and one patient died 6 days after the procedure because of multiple cardiovascular problems. CONCLUSIONS: Renal artery stent placement for the treatment of RAS in patients receiving hemodialysis can result in discontinuation of hemodialysis in patients with low proteinuria level and adequate kidney size and eGFR. J Vasc Interv Radiol 2008; 19:1563–1568 Abbreviations: eGFR ⴝ estimated glomerular filtration rate, ESRD ⴝ end-stage renal disease, NS ⴝ not significant, RAS ⴝ renal artery stenosis, RI ⴝ resistive index
ATHEROSCLEROTIC renovascular disease is an increasingly important cause of renal failure, the prevalence of which depends on the study group and the screening method used. It is
From the Division of Cardiology, Department of Medicine (M.T.); and Division of Vascular and Interventional Radiology, Department of Radiology (S.M., C.M.J., J.C.A., A.W.S., H.B., M.A.M.), Mayo Clinic College of Medicine, 200 First Street Southwest, Alfred 6460, Rochester, MN 55902. Received February 5, 2008; final revision received August 5, 2008; accepted August 11, 2008. Address correspondence to S.M.; E-mail: [email protected] None of the authors have identified a conflict of interest. © SIR, 2008 DOI: 10.1016/j.jvir.2008.08.016
estimated that 12%–18% of patients with end-stage renal disease (ESRD) have ischemic nephropathy caused by renal artery stenosis (RAS) (1,2). Percutaneous renal artery angioplasty with stent placement has become the cornerstone therapy for the treatment of atherosclerotic RAS. The reported technical success rates are excellent, at 98%–100%. However, the improvement in renal function is difficult to predict, with only 30%– 40% of patients showing improvement, 30%– 40% showing stabilization, and 20%–30% showing deterioration after renal artery stent placement (3,4). There have been several small studies in the literature describing percutaneous or surgical revascularization to allow the discontinuation of hemo-
dialysis in patients with RAS that resulted in ESRD. These studies have shown that renal artery revascularization is a viable option and can lead to discontinuation of dialysis in 40%–70% of patients (5– 8). The purpose of the present retrospective study is to review our experience in patients who were receiving hemodialysis and had RAS treated with renal artery angioplasty or stent placement.
MATERIALS AND METHODS Institutional review board approval was obtained for this retrospective study conducted from June 1996 to October 2006. All patients who were receiving hemodialysis with RAS and
1563
1564
•
Renal Artery Stent Placement for Restoration of Renal Function
underwent renal artery angioplasty with stent placement in a single institution comprise the study population. The primary endpoints of the study were death, dialysis, or renal transplantation at the time of follow-up, which were established through our medical records ending in June 2007. The secondary endpoints were restenosis, changes in blood pressure and number of antihypertensive medications, and estimated glomerular filtration rate (eGFR). Definitions RAS was defined as a diameter reduction of greater than 50% by visual estimate compared with its immediate distal nondilated main renal arterial segment (9). An RAS was defined as ostial if it was located within 5 mm from the renal artery origin from the aorta and defined as nonostial if it was located more than 5 mm away from the renal artery origin. Technical success of the stent procedure was defined by the presence of less than 30% residual stenosis by visual estimation on completion angiography after deployment of the stent. Twenty-four– hour predicted proteinuria level was estimated from the randomly collected preinterventional urine samples according to the following formula (9): Twenty-four– hour predicted proteinuria (in g/L) per 1.73 m2 body surface area ⫽ (urine protein level ⫻ 0.088)/urine creatinine level All complications were recorded according to the guidelines published by the Society of Interventional Radiology (10). Responders were defined as those patients who were able to discontinue hemodialysis after renal artery stent placement and remain dialysis-free for at least 30 days after stent placement. Nonresponders were defined as those patients who could not discontinue hemodialysis after renal artery stent placement. Acute renal failure was defined as occurring in patients who were receiving hemodialysis for no more than 30 days. Chronic hemodialysis was defined as occurring in patients who were receiving hemodialysis for more than 30 days.
Renal Duplex Ultrasonography We used renal duplex ultrasonography (US) for screening and quantification of severity of RAS (10). Baseline and follow-up duplex US studies were performed with a 2.5– 4.5-MHz phasedarray transducer (128/XP; Acuson, Mountain View, California; or Sequoia 512; Siemens, Erlangen, Germany). RAS or restenosis was defined on renal duplex US by the presence of at least 60% stenosis if the angle-corrected peak systolic velocity was at least 180 cm/sec or the peak systolic velocity–to–aortic velocity ratio was 3.5 at a Doppler insinuation angle of less than 60° (with aortic velocity measured near the renal artery origin from the aorta). In-stent restenosis suspected on US was confirmed by angiography and defined to be more than 50% stenosis by visual estimate. Resistive indices (RIs) were obtained by interrogation of the upper and lower intrarenal segmental arteries. RIs from the middle segmental arteries were used whenever available. An average of the two or three intrarenal segmental artery RIs was used to calculate the mean RI. Renal Artery Stent Placement Patients were prehydrated with oral fluids and intravenous crystalloid agents and 600-mg dose of n-acetylcysteine (Mucomyst) at 12 hourly intervals for 24 hours periprocedurally to reduce the risk of contrast agent nephropathy. Before renal angiography, informed consent was obtained from the patient for potential renal angioplasty and stent placement. Only balloon-expandable stents were used; the choice of stent was based on the interventionalist’s preference given the length and diameter (range, 5–7 mm) of the reference renal artery and the lesion length. An intravenous bolus of unfractionated heparin (50 U/kg body weight) was given during the procedure, and dual antiplatelet therapy was advised for at least 1 month after stent placement. A detailed description of the interventional procedure can be found elsewhere (11–13). Typically, no abdominal aortogram was obtained to limit the use of contrast. A “minimal touch technique” was followed during the procedure in which the guide catheter was placed just outside the renal ostium and the stenosis was crossed with only the guide wire. Selective angiography of the renal artery
November 2008
JVIR
was performed with 4 – 8 mL of Visipaque 320 (GE Healthcare, Princeton, New Jersey) or 4 – 8 mL of gadopentetate dimeglumine (Magnevist, Berlex, Wayne, New Jersey) to determine the extent of stenosis at the discretion of the proceduralist. The use of gadolinium has been discontinued at our institution because of the increased risk of nephrogenic systemic fibrosis in patients undergoing hemodialysis. It was used frequently before the Food and Drug Administration warning on gadolinium as a contrast agent. Study Group Sixteen patients underwent 16 procedures requiring the treatment of 22 cases of RAS with 22 balloon-expandable stents. The technical success rate of stent placement was 100%. Balloonexpandable stents used included Palmaz (n ⫽ 11; Cordis, Miami, Florida), Omnilink (n ⫽ 1) or Herculink (n ⫽ 3; Abbott Vascular, Abbott Park, Illinois), and Omniflex (n ⫽ 1; Angiodynamics, Queensbury, New York) stents. Five patients (31%) had bilateral RAS, and one of the five had three RAS that were all treated. Fourteen of the 16 patients had available preinterventional renal duplex US data. The mean combined kidney size was 10.53 cm ⫾ 1.3 (SD) and the mean RI was 0.76 ⫾ 0.07. The mean amount of blood pressure medications before the procedure was 2.1 ⫾ 1.1. Before renal artery stent placement, the average systolic blood pressure was 147 mm Hg ⫾ 26 and the average diastolic blood pressure was 66 mm Hg ⫾ 12. The average serum creatinine level and eGFR measured before initiation of hemodialysis were 3.89 mg/dL ⫾ 1.49 and 11.43 mL/min/24 h ⫾ 6.88, respectively. The average predicted 24-hour proteinuria was 2.7 g/24 h ⫾ 3.8. Statistical Methods JMP software (version 5.1.2; SAS, Cary, North Carolina) was used for the statistical analyses. Categoric variables were expressed as ratios and percentages. Continuous variables were expressed as means ⫾ SD or medians with ranges. Statistical comparison for continuous variables between groups was performed with use of the Wilcoxon signed-rank test, and for categoric variables the Pearson 2 test was used. Statistical significance for all tests was P ⱕ
Volume 19
Number 11
Thatipelli et al
•
1565
Table 1 Demographics and Cardiovascular Risk Factors for Responders and Nonresponders Characteristic
Data Available
Overall
Responders
Nonresponders
P Value
No. of patients Male sex Mean age at intervention (y) Family history of CAD Smoking (current or past) Past history of CAD Hypertension Diabetes Dyslipidemia Peripheral arterial disease Carotid disease Baseline antihypertensive BP medications Statin use Serum creatinine before HD eGFR before HD (mL/min/1.73 m2 BSA) Baseline median proteinuria (g/24 h)
16 16 16 16 14 16 16 16 16 16 16 16 16 12 12 15
16 9 (56) 74.6 ⫾ 10.6 7 (44) 7 (44) 10 (63) 15 (94) 5 (31) 11 (69) 6 (38) 9 (56) 2.08 ⫾ 1.17 10 (63) 3.89 ⫾ 1.49 11.43 ⫾ 6.88 2.7 ⫾ 3.8
8 5 (63) 75.5 ⫾ 5.5 4 (50) 4 (50) 5 (63) 7 (88) 2 (25) 7 (88) 4 (50) 5 (63) 1.9 ⫾ 1.4 6 (75) 3.98 ⫾ 1.44 10.13 ⫾ 5.43 0.4 ⫾ 0.3
8 4 (50) 73.7 ⫾ 14.5 3 (38) 3 (38) 5 (63) 8 (100) 3 (38) 4 (50) 2 (25) 4 (50) 2.5 ⫾ 0.6 4 (50) 3.83 ⫾ 1.63 12.35 ⫾ 8.03 4.8 ⫾ 4.5
NS NS NS NS NS NS NS NS NS NS NS NS NS NS .03
Note.—Values in parentheses are percentages. Values presented as means ⫾ SD where applicable. BP ⫽ blood pressure, BSA ⫽ body surface area, CAD ⫽ coronary artery disease, HD ⫽ hemodialysis. *Wilcoxon signed-rank test.
.05. A Cox proportional-hazards model was used to calculate the hazard ratios of preinterventional clinical, US, and angiographic variables to predict the endpoint with a univariate model. Variables with statistically significant P values were analyzed again with a multivariate method, adjusting for age, patient sex, and other cardiovascular variables. Results of these analyses were summarized as a hazard ratio with a 95% CI. A Kaplan-Meier survival model was used to calculate the 1-year survival rates for freedom from dialysis, and comparison among the subgroups was performed with a log-rank test.
RESULTS Study Group Sixteen patients with a mean age of 74.6 years ⫾ 10.6 (range, 49 – 86 y) underwent 16 procedures requiring the treatment of 22 cases of RAS (Table 1). The mean follow-up period was 448 days ⫾ 450 (range, 6 –1,583 d). The cardiovascular risk factors included hypertension (n ⫽ 16), diabetes (n ⫽ 5), dyslipidemia (n ⫽ 11), history of tobacco use (n ⫽ 7), and history of coronary artery disease (n ⫽ 10). The indication for the procedure was acute renal failure requiring dialysis (n ⫽ 8) and uncontrolled hypertension in eight patients undergoing chronic hemodialysis (n ⫽
Table 2 Renal Duplex US Data Finding Treated kidney size (cm) Baseline RI
Data Available
Overall
Responders
Nonresponders
P Value*
14
10.53 ⫾ 1.3
11.22 ⫾ 0.8
9.88 ⫾ 1.37
.04
0.79 ⫾ 0.11
NS
14
0.76 ⫾ 0.07
0.73 ⫾ 0.01
Note.—Values presented as means ⫾ SD where applicable. * Wilcoxon signed-rank test.
8). The mean time period during which study patients were receiving hemodialysis before renal artery stent placement was 164 days ⫾ 383 (median, 7 d; range, 3–1,033 d). Subgroups A subgroup analysis of the responders and nonresponders was performed. There were eight responders (mean age, 75.5 y ⫾ 5.5; 62% men) and eight nonresponders (mean age, 73.7 y ⫾ 14.5; 50% men; P value not significant [NS]). There were no differences between groups with respect to cardiovascular risk factors including hypertension, dyslipidemia, diabetes mellitus, history of tobacco use, and history of coronary artery disease. The serum creatinine and eGFR before renal artery stent placement were 3.98 mg/dL ⫾ 1.44 and 10.13 mL/min ⫾ 5.43 among responders and 3.83
mg/dL ⫾ 1.63 and 12.35 mL/min ⫾ 8.03 (P ⫽ NS) among nonresponders. However, the mean predicted 24-hour proteinuria was significantly different (P ⫽ .03) between responders and nonresponders, at 0.4 g/24 h ⫾ 0.3 and 4.8 g/24 h ⫾ 4.5, respectively. US Fourteen of the 16 patients had available preinterventional renal duplex US data. The mean combined kidney size was 10.53 cm ⫾ 1.3 and the mean RI was 0.76 ⫾ 0.07. The mean combined kidney sizes in responders and nonresponders were 11.22 cm ⫾ 0.08 and 9.88 cm ⫾ 1.37, respectively (P ⫽ .04), and the mean RIs were 0.73 ⫾ 0. 01 and 0.79 ⫾ 0.11, respectively (P ⫽ NS; Table 2). There were no differences between patients undergoing acute versus chronic hemodialysis.
1566
•
November 2008
Renal Artery Stent Placement for Restoration of Renal Function
Treatment with Renal Artery Stent Placement Sixteen patients had 22 cases of RAS treated with 22 balloon-expandable stents. Five patients (31%) had bilateral RAS, and one of these five had three RAS that were all treated. The technical success rate of stent placement was 100% with use of a variety of balloon-expandable stents such as Palmaz (n ⫽ 11; Cordis), Omnilink (n ⫽ 1) or Herculink (n ⫽ 3; Abbott Vascular), and Omiflex (n ⫽ 1; Angiodynamics) stents. Hypertension Response The mean amount of blood pressure medications before the procedure was 2.1 ⫾ 1.1 per patient and increased slightly after the procedure to 2.4 ⫾ 1.1 (P ⫽ NS). Before renal artery stent placement, the average systolic blood pressure was 147 mm Hg ⫾ 26 and the average diastolic blood pressure was 66 mm Hg ⫾ 12. At last follow-up, the average systolic blood pressure was 130 mm Hg ⫾ 36 and the average diastolic blood pressure was 59 mm Hg ⫾ 18 (P ⫽ NS). There were no differences in blood pressure response in patients who were undergoing acute versus chronic hemodialysis. Kidney Function Before the initiation of hemodialysis, the average serum creatinine level was 3.89 mg/dL ⫾ 1.49 and the average eGFR was 11.43 mL/min/24 h ⫾ 6.88. The average predicted 24hour proteinuria was 2.7 g/24 h ⫾ 3.8. There were eight patients who were able to discontinue hemodialysis over a median period of 3.5 days (range, 0 – 84 d), among whom three patients discontinued hemodialysis within 7 days of the procedure, one within 30 days, and the remainder within 90 days. There was no difference in the duration of hemodialysis dependency before renal artery stent placement between the responders (150 d ⫾ 352) and nonresponders (171 d ⫾ 426; P ⫽ NS; Table 3). Multivariate analysis using a Cox proportional-hazards model with adjustments for age, sex, and cardiovascular risk factors was performed for the outcome of discontinuation of hemodial-
JVIR
Table 3 Hemodialysis-free Patients and Survival Data Characteristic (n ⫽ 15)
Overall
Responders
Nonresponders
No HD at 7 d No HD by 30 d No HD by 90 d HD at last follow-up or death Mean follow-up (d) HD duration before stent (d) Death One-year cumulative HD survival rate (%)
3 1 4 7 448 ⫾ 450 161 ⫾ 378 9 65 (52–78)
3 1 4 — 595 ⫾ 548 150 ⫾ 352 4 71 (54–88)
— — — 7 301 ⫾ 288 171 ⫾ 426 5 60 (52–88)
Note.—No significant differences were found between responders and nonresponders in any category with the Wilcoxon signed-rank test or log-rank test as applicable. Values in parentheses are ranges. Values presented as means ⫾ SD where applicable. HD ⫽ hemodialysis. Table 4 Hazard Ratios Derived in Cox Proportional-hazards Model Univariate
Multivariate
Variable
HR
P Value
HR
P Value
Male sex Age Family history of CAD Smoking Refractory/accelerated hypertension Diabetes Dyslipidemia History of CAD Baseline antihypertensive medications Statin use Serum creatinine before HD eGFR before HD 24-h predicted proteinuria Mean treated kidney size Indication for intervention Unilateral or bilateral intervention
1.66 (0.73–4.48) 0.99 (0.94–1.06) 1.45 (0.28–6.82)
NS NS NS
1.55 (0.51–4.8) 0.92 (0.72–1.13) 10.65 (0.07–11.76)
NS NS NS
0.69 (0.25–1.74) 0.47 (0.17–2.1)
NS NS
2 (0.5–12.95) 0.24 (0.03–1.6)
NS NS
0.75 (0.29–1.58) 1.79 (0.75–7.74) 1.25 (0.57–3.33) 0.86 (0.44–1.71)
NS NS NS NS
0.95 (0.29–3.48) 2.9 (0.74–16.47) 0.48 (0.13–1.41) 5.12 (0.03–525)
NS NS NS NS
1.66 (0.68–4.36) 1.42 (0.71–2.91)
NS NS
1.02 (0.22–5.03) 2.36
NS NS
0.91 (0.76–1.04) 1
NS ⬍.001
0.14 1
⬍.001 ⬍.001
1.88 (1.01–4.37)
.05
5 (1.2–123)
.03
0.75 (0.32–1.61)
NS
0.7 (0.08–2.57)
NS
1.25 (0.61–2.57)
NS
0.56 (0.04–5.92)
NS
Note.—Hazard ratios presented with 95% CIs. CAD ⫽ coronary artery disease, HD ⫽ hemodialysis, HR ⫽ hazard ratio. *Adjusted for age, sex, and other cardiovascular risk factors.
ysis. This analysis showed that the average preinterventional kidney size on US (P ⫽ .03), baseline eGFR (P ⱕ .01) and 24-hour predicted proteinuria (P ⱕ .01) were predictors of a successful outcome after renal artery stent placement with discontinuation of hemodialysis (Table 4). There were no differences in
discontinuation of dialysis between patients who had been undergoing acute versus chronic hemodialysis. Follow-up The average study follow-up period was 448 days ⫾ 450. Only four
Volume 19
Number 11
patients had follow-up renal Doppler US studies, among whom one responder had in-stent restenosis 295 days after renal artery stent placement. The mean follow-up period for the responders was 595 days ⫾ 548, and their mean dialysis-free period after renal artery stent placement was 564 days ⫾ 533. Nine patients died during the follow-up period, with one patient dying of cardiovascular complications on the sixth day after the procedure, yielding a 30-day mortality rate of 6.3%. On Kaplan-Meier analysis, the 1-year cumulative survival rates based on deaths from all causes were 65% (range, 52%–78%) for the entire study sample, 71% (range, 54%– 88%) for responders, and 60% (range, 52%– 88%) for nonresponders (P ⫽ NS; Figure; Table 3). Contrast Used Iodinated contrast medium was used in five patients, gadolinium in six, and a combination of the two in four (Table 5). Data were not available for one patient. There was no incidence of nephrogenic systemic fibrosis. Complications Two patients had groin hematomas and one patient one had a non–flowlimiting dissection during the stent placement, which was managed conservatively.
DISCUSSION In the present study, we have shown that renal artery revascularization with stents is a viable option in patients undergoing hemodialysis with RAS and can provide a benefit in 50% of patients such that hemodialysis can be discontinued after the procedure. This study highlights that several important factors are potential predictors of a successful outcome, including baseline eGFR, 24-hour predicted proteinuria, and average renal size determined on duplex US imaging. The United States Renal Data System Annual Data Report showed that 6.7% of the total Medicare budget ($18.1 billion) in 2003 was spent on the care of patients diagnosed with ESRD (14). There are currently more than
Thatipelli et al
•
1567
Figure. Kaplan-Meier life-table survival analysis for responders and nonresponders.
Table 5 Procedural Data Variable
Data Available
No. of Cases
Grade of stenosis (⬎70%) IV hydration and n-acetylcysteine Contrast medium used Gadolinium Iodinated Both Stent size (mm) 5 6 7 Complications Groin hematoma Non–flow-limiting dissection
16 16 15
16 (100) 16 (100) 6 (40) 5 (33) 4 (27)
14 5 (31) 8 (63) 1 (6) 2 1
Note.—Values in parentheses are percentages.
400,000 patients with ESRD in the United States, with recent projections forecasting that, by 2030, the hemodialysis population will exceed 2 million (14). It is estimated that 12%–18% of patients with ESRD have RAS and ischemic nephropathy as the etiology of ESRD. There are several small series of articles discussing the role of percutaneous intervention (6,15,16) or surgical treatments (7,17–19) to allow discontinuation of dialysis in patients with RAS. In a series of four dialysis-dependent patients with solitary functioning kidneys, Shannon et al (8) reported discontinuation of hemodialysis after percutaneous revascularization; baseline patient characteristics and mortality of the dialysis-dependent subgroup were not reported separately,
and only 5% of the study group had diabetes (in contrast to 38% of the patients in the present study). In a separate study (15), 28 patients with chronic kidney disease and RAS were treated with percutaneous revascularization. Of the 28 patients, 11 had severe renal insufficiency (eGFR ⱕ10 mL/min). After renal artery intervention, there were two deaths (18%) within 30 days, and 64% of patients’ kidney function worsened or remained unchanged whereas only two (18%) had improvement in renal function (15). The present study, which involved a similar cohort, showed much better results, with only one death (6%) within 30 days after intervention and 50% of patients discontinuing their need for dialysis at an average follow-up period of 564 days ⫾ 533.
1568
•
Renal Artery Stent Placement for Restoration of Renal Function
Previously published studies have grouped and analyzed dialysis-dependent and non– dialysis-dependent patients together, meaning their conclusions and applicability to dialysisdependent patients alone are difficult to determine (5,6,16). Several previous articles have highlighted the importance of proteinuria, glomerular filtration rate, and severity of RAS as predictors of successful outcome after renal artery revascularization (20,21). In the present study, we observed significant differences in proteinuria between responders and nonresponders. In addition, our study draws attention to the prognostic role of proteinuria as a marker for microvascular injury before renal artery stent placement. This was a small retrospective study with inherent limitations. This is not a population-based study with subjects who constitute a selective group of patients referred to our tertiary care center. Therefore, our results are associated with selection bias and cannot be generalized. The power of our conclusions is limited by the small sample size. Despite these limitations, the study highlights the potential role of percutaneous renovascular revascularization in dialysis-dependent ischemic nephropathy in improving dialysis-free survival. Percutaneous renovascular revascularization in patients with dialysisdependent ischemic nephropathy is associated with a high technical success rate and low complication rates. In our series, there were several potential predicators of a successful outcome of dialysis discontinuation after renal stent placement, including baseline average kidney size, 24-hour predicted proteinuria, and renal function before dialysis. There was no survival benefit with revascularization in this patient cohort, potentially because of the short follow-up and small group of patients. Future studies need to be
performed to confirm the validity of these observations. References 1. Hansen KJ. Prevalence of ischemic nephropathy in the atherosclerotic population. Am J Kidney Dis 1994; 24: 615– 621. 2. Greco BA, Breyer JA. The natural history of renal artery stenosis: who should be evaluated for suspected ischemic nephropathy? Sem Nephrol 1996; 16:2–11. 3. Dorros G, Jaff M, Jain A, Dufek C, Mathiak L. Follow-up of primary Palmaz-Schatz stent placement for atherosclerotic renal artery stenosis. Am J Cardiol 1995; 75:1051–1055. 4. Dorros G, Jaff M, Mathiak L, et al. Four-year follow-up of Palmaz-Schatz stent revascularization as treatment for atherosclerotic renal artery stenosis. Circulation 1998; 98:642– 647. 5. Dean RH, Tribble RW, Hansen KJ, O’Neil E, Craven TE, Redding JF II. Evolution of renal insufficiency in ischemic nephropathy. Ann Surg 1991; 213: 446 – 455. 6. O’Donovan RM, Gutierrez OH, Izzo JL Jr. Preservation of renal function by percutaneous renal angioplasty in high-risk elderly patients: Short-term outcome. Nephron 1992; 60:187–192. 7. Wengrovitz M, Healy DA, Diamond JR, Atnip RG. Renal revascularization in patients on dialysis. J Cardiovasc Surg 1995; 36:241–246. 8. Shannon HM, Gillespie IN, Moss JG. Salvage of the solitary kidney by insertion of a renal artery stent. AJR Am J Roentgenol 1998; 171:217–222. 9. Xin G, Wang M, Jiao LL, Xu GB, Wang HY. Protein-to-creatinine ratio in spot urine samples as a predictor of quantitation of proteinuria. Clin Chim Acta 2004; 350:35–39. 10. Rundback JH, Sacks D, Kent KC, et al. Guidelines for the reporting of renal artery revascularization in clinical trials. J Vasc Interv Radiol 2003; 14(suppl):S477– S492. 11. Ivanovic V, McKusick MA, Johnson CM III, et al. Renal artery stent placement: complications at a single tertiary
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
November 2008
JVIR
care center. J Vasc Interv Radiol 2003; 14:217–225. Misra S, Sturludottir M, Matthews V, Bjarnason H, McKusick M, Iyer V. Treatment of complex stenoses involving renal artery bifurcations using drug-eluting stents. J Vasc Interv Radiol 2008; 19:272–278. Misra S, Thatipelli MR, Howe PW , et al. Preliminary study of the use of drug-eluting stents in atherosclerotic renal artery stenoses 4 mm in diameter or smaller. J Vasc Interv Radiol 2008; 19:833– 839. Collins AJ, Kasiske B, Herzog C, et al. Excerpts from the united states renal data system 2003 annual data report: atlas of end-stage renal disease in the united states. Am J Kidney Dis 2003; 42:A5–A7. Korsakas S, Mohaupt MG, Dinkel HP, et al. Delay of dialysis in end-stage renal failure: prospective study on percutaneous renal artery interventions. Kidney Int 2004; 65:251–258. Paulsen D, Klow NE, Rogstad B, et al. Preservation of renal function by percutaneous transluminal angioplasty in ischaemic renal disease. Nephrol Dial Transplant 1999; 14:1454 –1461. Ascer E, Gennaro M, Rogers D. Unilateral renal artery revascularization can salvage renal function and terminate dialysis in selected patients with uremia. J Vasc Surg 1993; 18:1012– 1018. Hansen KJ, Thomason RB, Craven TE, et al. Surgical management of dialysis-dependent ischemic nephropathy. J Vasc Surg 1995; 21:197–209. Kaylor WM, Novick AC, Ziegelbaum M, Vidt DG. Reversal of end stage renal failure with surgical revascularization in patients with atherosclerotic renal artery occlusion. J Urol 1989; 141: 486 – 488. Cheung CM, Wright JR, Shurrab AE, et al. Epidemiology of renal dysfunction and patient outcome in atherosclerotic renal artery occlusion. J Am Soc Nephrol 2002; 13:149 –157. Makanjuola AD, Suresh M, Laboi P, Kalra PA, Scoble JE. Proteinuria in atherosclerotic renovascular disease. Q J Med 1999; 92:515–518.
PURPOSE: Renal artery stent placement to treat renal artery stenosis (RAS) in patients undergoing hemodialysis is not performed commonly. We present our outcomes of discontinuation of hemodialysis after treating patients with RAS by stent placement on acute (<30 days of hemodialysis) and chronic hemodialysis (>30 days). MATERIALS AND METHODS: A retrospective study was conducted in 16 patients (nine men) with an average age of 74.6 years ⴞ 10.6 (range, 49 – 86 y) who underwent treatment of 22 RAS for acute renal failure (n ⴝ 8) or uncontrolled hypertension with chronic hemodialysis (n ⴝ 8). The average follow-up was 448 days ⴞ 450 (median, 363.5 d; range, 6 –1,583 d). The primary outcome consisted of discontinuation of hemodialysis, death, and transplantation. Secondary endpoints included restenosis, changes in blood pressure (and use of medications), and estimated glomerular filtration rate (eGFR). RESULTS: After the stent procedure, eight patients were able to discontinue hemodialysis and remained free from dialysis over a mean period of 564 days ⴞ 533. The predictors of discontinuation of hemodialysis were 24-hour proteinuria, eGFR before renal artery stent placement, and size of the kidney on ultrasound studies (P < .05 for all three). There was no difference in patients who were undergoing acute versus chronic hemodialysis. There were three minor complications, and one patient died 6 days after the procedure because of multiple cardiovascular problems. CONCLUSIONS: Renal artery stent placement for the treatment of RAS in patients receiving hemodialysis can result in discontinuation of hemodialysis in patients with low proteinuria level and adequate kidney size and eGFR. J Vasc Interv Radiol 2008; 19:1563–1568 Abbreviations: eGFR ⴝ estimated glomerular filtration rate, ESRD ⴝ end-stage renal disease, NS ⴝ not significant, RAS ⴝ renal artery stenosis, RI ⴝ resistive index
ATHEROSCLEROTIC renovascular disease is an increasingly important cause of renal failure, the prevalence of which depends on the study group and the screening method used. It is
From the Division of Cardiology, Department of Medicine (M.T.); and Division of Vascular and Interventional Radiology, Department of Radiology (S.M., C.M.J., J.C.A., A.W.S., H.B., M.A.M.), Mayo Clinic College of Medicine, 200 First Street Southwest, Alfred 6460, Rochester, MN 55902. Received February 5, 2008; final revision received August 5, 2008; accepted August 11, 2008. Address correspondence to S.M.; E-mail: [email protected] None of the authors have identified a conflict of interest. © SIR, 2008 DOI: 10.1016/j.jvir.2008.08.016
estimated that 12%–18% of patients with end-stage renal disease (ESRD) have ischemic nephropathy caused by renal artery stenosis (RAS) (1,2). Percutaneous renal artery angioplasty with stent placement has become the cornerstone therapy for the treatment of atherosclerotic RAS. The reported technical success rates are excellent, at 98%–100%. However, the improvement in renal function is difficult to predict, with only 30%– 40% of patients showing improvement, 30%– 40% showing stabilization, and 20%–30% showing deterioration after renal artery stent placement (3,4). There have been several small studies in the literature describing percutaneous or surgical revascularization to allow the discontinuation of hemo-
dialysis in patients with RAS that resulted in ESRD. These studies have shown that renal artery revascularization is a viable option and can lead to discontinuation of dialysis in 40%–70% of patients (5– 8). The purpose of the present retrospective study is to review our experience in patients who were receiving hemodialysis and had RAS treated with renal artery angioplasty or stent placement.
MATERIALS AND METHODS Institutional review board approval was obtained for this retrospective study conducted from June 1996 to October 2006. All patients who were receiving hemodialysis with RAS and
1563
1564
•
Renal Artery Stent Placement for Restoration of Renal Function
underwent renal artery angioplasty with stent placement in a single institution comprise the study population. The primary endpoints of the study were death, dialysis, or renal transplantation at the time of follow-up, which were established through our medical records ending in June 2007. The secondary endpoints were restenosis, changes in blood pressure and number of antihypertensive medications, and estimated glomerular filtration rate (eGFR). Definitions RAS was defined as a diameter reduction of greater than 50% by visual estimate compared with its immediate distal nondilated main renal arterial segment (9). An RAS was defined as ostial if it was located within 5 mm from the renal artery origin from the aorta and defined as nonostial if it was located more than 5 mm away from the renal artery origin. Technical success of the stent procedure was defined by the presence of less than 30% residual stenosis by visual estimation on completion angiography after deployment of the stent. Twenty-four– hour predicted proteinuria level was estimated from the randomly collected preinterventional urine samples according to the following formula (9): Twenty-four– hour predicted proteinuria (in g/L) per 1.73 m2 body surface area ⫽ (urine protein level ⫻ 0.088)/urine creatinine level All complications were recorded according to the guidelines published by the Society of Interventional Radiology (10). Responders were defined as those patients who were able to discontinue hemodialysis after renal artery stent placement and remain dialysis-free for at least 30 days after stent placement. Nonresponders were defined as those patients who could not discontinue hemodialysis after renal artery stent placement. Acute renal failure was defined as occurring in patients who were receiving hemodialysis for no more than 30 days. Chronic hemodialysis was defined as occurring in patients who were receiving hemodialysis for more than 30 days.
Renal Duplex Ultrasonography We used renal duplex ultrasonography (US) for screening and quantification of severity of RAS (10). Baseline and follow-up duplex US studies were performed with a 2.5– 4.5-MHz phasedarray transducer (128/XP; Acuson, Mountain View, California; or Sequoia 512; Siemens, Erlangen, Germany). RAS or restenosis was defined on renal duplex US by the presence of at least 60% stenosis if the angle-corrected peak systolic velocity was at least 180 cm/sec or the peak systolic velocity–to–aortic velocity ratio was 3.5 at a Doppler insinuation angle of less than 60° (with aortic velocity measured near the renal artery origin from the aorta). In-stent restenosis suspected on US was confirmed by angiography and defined to be more than 50% stenosis by visual estimate. Resistive indices (RIs) were obtained by interrogation of the upper and lower intrarenal segmental arteries. RIs from the middle segmental arteries were used whenever available. An average of the two or three intrarenal segmental artery RIs was used to calculate the mean RI. Renal Artery Stent Placement Patients were prehydrated with oral fluids and intravenous crystalloid agents and 600-mg dose of n-acetylcysteine (Mucomyst) at 12 hourly intervals for 24 hours periprocedurally to reduce the risk of contrast agent nephropathy. Before renal angiography, informed consent was obtained from the patient for potential renal angioplasty and stent placement. Only balloon-expandable stents were used; the choice of stent was based on the interventionalist’s preference given the length and diameter (range, 5–7 mm) of the reference renal artery and the lesion length. An intravenous bolus of unfractionated heparin (50 U/kg body weight) was given during the procedure, and dual antiplatelet therapy was advised for at least 1 month after stent placement. A detailed description of the interventional procedure can be found elsewhere (11–13). Typically, no abdominal aortogram was obtained to limit the use of contrast. A “minimal touch technique” was followed during the procedure in which the guide catheter was placed just outside the renal ostium and the stenosis was crossed with only the guide wire. Selective angiography of the renal artery
November 2008
JVIR
was performed with 4 – 8 mL of Visipaque 320 (GE Healthcare, Princeton, New Jersey) or 4 – 8 mL of gadopentetate dimeglumine (Magnevist, Berlex, Wayne, New Jersey) to determine the extent of stenosis at the discretion of the proceduralist. The use of gadolinium has been discontinued at our institution because of the increased risk of nephrogenic systemic fibrosis in patients undergoing hemodialysis. It was used frequently before the Food and Drug Administration warning on gadolinium as a contrast agent. Study Group Sixteen patients underwent 16 procedures requiring the treatment of 22 cases of RAS with 22 balloon-expandable stents. The technical success rate of stent placement was 100%. Balloonexpandable stents used included Palmaz (n ⫽ 11; Cordis, Miami, Florida), Omnilink (n ⫽ 1) or Herculink (n ⫽ 3; Abbott Vascular, Abbott Park, Illinois), and Omniflex (n ⫽ 1; Angiodynamics, Queensbury, New York) stents. Five patients (31%) had bilateral RAS, and one of the five had three RAS that were all treated. Fourteen of the 16 patients had available preinterventional renal duplex US data. The mean combined kidney size was 10.53 cm ⫾ 1.3 (SD) and the mean RI was 0.76 ⫾ 0.07. The mean amount of blood pressure medications before the procedure was 2.1 ⫾ 1.1. Before renal artery stent placement, the average systolic blood pressure was 147 mm Hg ⫾ 26 and the average diastolic blood pressure was 66 mm Hg ⫾ 12. The average serum creatinine level and eGFR measured before initiation of hemodialysis were 3.89 mg/dL ⫾ 1.49 and 11.43 mL/min/24 h ⫾ 6.88, respectively. The average predicted 24-hour proteinuria was 2.7 g/24 h ⫾ 3.8. Statistical Methods JMP software (version 5.1.2; SAS, Cary, North Carolina) was used for the statistical analyses. Categoric variables were expressed as ratios and percentages. Continuous variables were expressed as means ⫾ SD or medians with ranges. Statistical comparison for continuous variables between groups was performed with use of the Wilcoxon signed-rank test, and for categoric variables the Pearson 2 test was used. Statistical significance for all tests was P ⱕ
Volume 19
Number 11
Thatipelli et al
•
1565
Table 1 Demographics and Cardiovascular Risk Factors for Responders and Nonresponders Characteristic
Data Available
Overall
Responders
Nonresponders
P Value
No. of patients Male sex Mean age at intervention (y) Family history of CAD Smoking (current or past) Past history of CAD Hypertension Diabetes Dyslipidemia Peripheral arterial disease Carotid disease Baseline antihypertensive BP medications Statin use Serum creatinine before HD eGFR before HD (mL/min/1.73 m2 BSA) Baseline median proteinuria (g/24 h)
16 16 16 16 14 16 16 16 16 16 16 16 16 12 12 15
16 9 (56) 74.6 ⫾ 10.6 7 (44) 7 (44) 10 (63) 15 (94) 5 (31) 11 (69) 6 (38) 9 (56) 2.08 ⫾ 1.17 10 (63) 3.89 ⫾ 1.49 11.43 ⫾ 6.88 2.7 ⫾ 3.8
8 5 (63) 75.5 ⫾ 5.5 4 (50) 4 (50) 5 (63) 7 (88) 2 (25) 7 (88) 4 (50) 5 (63) 1.9 ⫾ 1.4 6 (75) 3.98 ⫾ 1.44 10.13 ⫾ 5.43 0.4 ⫾ 0.3
8 4 (50) 73.7 ⫾ 14.5 3 (38) 3 (38) 5 (63) 8 (100) 3 (38) 4 (50) 2 (25) 4 (50) 2.5 ⫾ 0.6 4 (50) 3.83 ⫾ 1.63 12.35 ⫾ 8.03 4.8 ⫾ 4.5
NS NS NS NS NS NS NS NS NS NS NS NS NS NS .03
Note.—Values in parentheses are percentages. Values presented as means ⫾ SD where applicable. BP ⫽ blood pressure, BSA ⫽ body surface area, CAD ⫽ coronary artery disease, HD ⫽ hemodialysis. *Wilcoxon signed-rank test.
.05. A Cox proportional-hazards model was used to calculate the hazard ratios of preinterventional clinical, US, and angiographic variables to predict the endpoint with a univariate model. Variables with statistically significant P values were analyzed again with a multivariate method, adjusting for age, patient sex, and other cardiovascular variables. Results of these analyses were summarized as a hazard ratio with a 95% CI. A Kaplan-Meier survival model was used to calculate the 1-year survival rates for freedom from dialysis, and comparison among the subgroups was performed with a log-rank test.
RESULTS Study Group Sixteen patients with a mean age of 74.6 years ⫾ 10.6 (range, 49 – 86 y) underwent 16 procedures requiring the treatment of 22 cases of RAS (Table 1). The mean follow-up period was 448 days ⫾ 450 (range, 6 –1,583 d). The cardiovascular risk factors included hypertension (n ⫽ 16), diabetes (n ⫽ 5), dyslipidemia (n ⫽ 11), history of tobacco use (n ⫽ 7), and history of coronary artery disease (n ⫽ 10). The indication for the procedure was acute renal failure requiring dialysis (n ⫽ 8) and uncontrolled hypertension in eight patients undergoing chronic hemodialysis (n ⫽
Table 2 Renal Duplex US Data Finding Treated kidney size (cm) Baseline RI
Data Available
Overall
Responders
Nonresponders
P Value*
14
10.53 ⫾ 1.3
11.22 ⫾ 0.8
9.88 ⫾ 1.37
.04
0.79 ⫾ 0.11
NS
14
0.76 ⫾ 0.07
0.73 ⫾ 0.01
Note.—Values presented as means ⫾ SD where applicable. * Wilcoxon signed-rank test.
8). The mean time period during which study patients were receiving hemodialysis before renal artery stent placement was 164 days ⫾ 383 (median, 7 d; range, 3–1,033 d). Subgroups A subgroup analysis of the responders and nonresponders was performed. There were eight responders (mean age, 75.5 y ⫾ 5.5; 62% men) and eight nonresponders (mean age, 73.7 y ⫾ 14.5; 50% men; P value not significant [NS]). There were no differences between groups with respect to cardiovascular risk factors including hypertension, dyslipidemia, diabetes mellitus, history of tobacco use, and history of coronary artery disease. The serum creatinine and eGFR before renal artery stent placement were 3.98 mg/dL ⫾ 1.44 and 10.13 mL/min ⫾ 5.43 among responders and 3.83
mg/dL ⫾ 1.63 and 12.35 mL/min ⫾ 8.03 (P ⫽ NS) among nonresponders. However, the mean predicted 24-hour proteinuria was significantly different (P ⫽ .03) between responders and nonresponders, at 0.4 g/24 h ⫾ 0.3 and 4.8 g/24 h ⫾ 4.5, respectively. US Fourteen of the 16 patients had available preinterventional renal duplex US data. The mean combined kidney size was 10.53 cm ⫾ 1.3 and the mean RI was 0.76 ⫾ 0.07. The mean combined kidney sizes in responders and nonresponders were 11.22 cm ⫾ 0.08 and 9.88 cm ⫾ 1.37, respectively (P ⫽ .04), and the mean RIs were 0.73 ⫾ 0. 01 and 0.79 ⫾ 0.11, respectively (P ⫽ NS; Table 2). There were no differences between patients undergoing acute versus chronic hemodialysis.
1566
•
November 2008
Renal Artery Stent Placement for Restoration of Renal Function
Treatment with Renal Artery Stent Placement Sixteen patients had 22 cases of RAS treated with 22 balloon-expandable stents. Five patients (31%) had bilateral RAS, and one of these five had three RAS that were all treated. The technical success rate of stent placement was 100% with use of a variety of balloon-expandable stents such as Palmaz (n ⫽ 11; Cordis), Omnilink (n ⫽ 1) or Herculink (n ⫽ 3; Abbott Vascular), and Omiflex (n ⫽ 1; Angiodynamics) stents. Hypertension Response The mean amount of blood pressure medications before the procedure was 2.1 ⫾ 1.1 per patient and increased slightly after the procedure to 2.4 ⫾ 1.1 (P ⫽ NS). Before renal artery stent placement, the average systolic blood pressure was 147 mm Hg ⫾ 26 and the average diastolic blood pressure was 66 mm Hg ⫾ 12. At last follow-up, the average systolic blood pressure was 130 mm Hg ⫾ 36 and the average diastolic blood pressure was 59 mm Hg ⫾ 18 (P ⫽ NS). There were no differences in blood pressure response in patients who were undergoing acute versus chronic hemodialysis. Kidney Function Before the initiation of hemodialysis, the average serum creatinine level was 3.89 mg/dL ⫾ 1.49 and the average eGFR was 11.43 mL/min/24 h ⫾ 6.88. The average predicted 24hour proteinuria was 2.7 g/24 h ⫾ 3.8. There were eight patients who were able to discontinue hemodialysis over a median period of 3.5 days (range, 0 – 84 d), among whom three patients discontinued hemodialysis within 7 days of the procedure, one within 30 days, and the remainder within 90 days. There was no difference in the duration of hemodialysis dependency before renal artery stent placement between the responders (150 d ⫾ 352) and nonresponders (171 d ⫾ 426; P ⫽ NS; Table 3). Multivariate analysis using a Cox proportional-hazards model with adjustments for age, sex, and cardiovascular risk factors was performed for the outcome of discontinuation of hemodial-
JVIR
Table 3 Hemodialysis-free Patients and Survival Data Characteristic (n ⫽ 15)
Overall
Responders
Nonresponders
No HD at 7 d No HD by 30 d No HD by 90 d HD at last follow-up or death Mean follow-up (d) HD duration before stent (d) Death One-year cumulative HD survival rate (%)
3 1 4 7 448 ⫾ 450 161 ⫾ 378 9 65 (52–78)
3 1 4 — 595 ⫾ 548 150 ⫾ 352 4 71 (54–88)
— — — 7 301 ⫾ 288 171 ⫾ 426 5 60 (52–88)
Note.—No significant differences were found between responders and nonresponders in any category with the Wilcoxon signed-rank test or log-rank test as applicable. Values in parentheses are ranges. Values presented as means ⫾ SD where applicable. HD ⫽ hemodialysis. Table 4 Hazard Ratios Derived in Cox Proportional-hazards Model Univariate
Multivariate
Variable
HR
P Value
HR
P Value
Male sex Age Family history of CAD Smoking Refractory/accelerated hypertension Diabetes Dyslipidemia History of CAD Baseline antihypertensive medications Statin use Serum creatinine before HD eGFR before HD 24-h predicted proteinuria Mean treated kidney size Indication for intervention Unilateral or bilateral intervention
1.66 (0.73–4.48) 0.99 (0.94–1.06) 1.45 (0.28–6.82)
NS NS NS
1.55 (0.51–4.8) 0.92 (0.72–1.13) 10.65 (0.07–11.76)
NS NS NS
0.69 (0.25–1.74) 0.47 (0.17–2.1)
NS NS
2 (0.5–12.95) 0.24 (0.03–1.6)
NS NS
0.75 (0.29–1.58) 1.79 (0.75–7.74) 1.25 (0.57–3.33) 0.86 (0.44–1.71)
NS NS NS NS
0.95 (0.29–3.48) 2.9 (0.74–16.47) 0.48 (0.13–1.41) 5.12 (0.03–525)
NS NS NS NS
1.66 (0.68–4.36) 1.42 (0.71–2.91)
NS NS
1.02 (0.22–5.03) 2.36
NS NS
0.91 (0.76–1.04) 1
NS ⬍.001
0.14 1
⬍.001 ⬍.001
1.88 (1.01–4.37)
.05
5 (1.2–123)
.03
0.75 (0.32–1.61)
NS
0.7 (0.08–2.57)
NS
1.25 (0.61–2.57)
NS
0.56 (0.04–5.92)
NS
Note.—Hazard ratios presented with 95% CIs. CAD ⫽ coronary artery disease, HD ⫽ hemodialysis, HR ⫽ hazard ratio. *Adjusted for age, sex, and other cardiovascular risk factors.
ysis. This analysis showed that the average preinterventional kidney size on US (P ⫽ .03), baseline eGFR (P ⱕ .01) and 24-hour predicted proteinuria (P ⱕ .01) were predictors of a successful outcome after renal artery stent placement with discontinuation of hemodialysis (Table 4). There were no differences in
discontinuation of dialysis between patients who had been undergoing acute versus chronic hemodialysis. Follow-up The average study follow-up period was 448 days ⫾ 450. Only four
Volume 19
Number 11
patients had follow-up renal Doppler US studies, among whom one responder had in-stent restenosis 295 days after renal artery stent placement. The mean follow-up period for the responders was 595 days ⫾ 548, and their mean dialysis-free period after renal artery stent placement was 564 days ⫾ 533. Nine patients died during the follow-up period, with one patient dying of cardiovascular complications on the sixth day after the procedure, yielding a 30-day mortality rate of 6.3%. On Kaplan-Meier analysis, the 1-year cumulative survival rates based on deaths from all causes were 65% (range, 52%–78%) for the entire study sample, 71% (range, 54%– 88%) for responders, and 60% (range, 52%– 88%) for nonresponders (P ⫽ NS; Figure; Table 3). Contrast Used Iodinated contrast medium was used in five patients, gadolinium in six, and a combination of the two in four (Table 5). Data were not available for one patient. There was no incidence of nephrogenic systemic fibrosis. Complications Two patients had groin hematomas and one patient one had a non–flowlimiting dissection during the stent placement, which was managed conservatively.
DISCUSSION In the present study, we have shown that renal artery revascularization with stents is a viable option in patients undergoing hemodialysis with RAS and can provide a benefit in 50% of patients such that hemodialysis can be discontinued after the procedure. This study highlights that several important factors are potential predictors of a successful outcome, including baseline eGFR, 24-hour predicted proteinuria, and average renal size determined on duplex US imaging. The United States Renal Data System Annual Data Report showed that 6.7% of the total Medicare budget ($18.1 billion) in 2003 was spent on the care of patients diagnosed with ESRD (14). There are currently more than
Thatipelli et al
•
1567
Figure. Kaplan-Meier life-table survival analysis for responders and nonresponders.
Table 5 Procedural Data Variable
Data Available
No. of Cases
Grade of stenosis (⬎70%) IV hydration and n-acetylcysteine Contrast medium used Gadolinium Iodinated Both Stent size (mm) 5 6 7 Complications Groin hematoma Non–flow-limiting dissection
16 16 15
16 (100) 16 (100) 6 (40) 5 (33) 4 (27)
14 5 (31) 8 (63) 1 (6) 2 1
Note.—Values in parentheses are percentages.
400,000 patients with ESRD in the United States, with recent projections forecasting that, by 2030, the hemodialysis population will exceed 2 million (14). It is estimated that 12%–18% of patients with ESRD have RAS and ischemic nephropathy as the etiology of ESRD. There are several small series of articles discussing the role of percutaneous intervention (6,15,16) or surgical treatments (7,17–19) to allow discontinuation of dialysis in patients with RAS. In a series of four dialysis-dependent patients with solitary functioning kidneys, Shannon et al (8) reported discontinuation of hemodialysis after percutaneous revascularization; baseline patient characteristics and mortality of the dialysis-dependent subgroup were not reported separately,
and only 5% of the study group had diabetes (in contrast to 38% of the patients in the present study). In a separate study (15), 28 patients with chronic kidney disease and RAS were treated with percutaneous revascularization. Of the 28 patients, 11 had severe renal insufficiency (eGFR ⱕ10 mL/min). After renal artery intervention, there were two deaths (18%) within 30 days, and 64% of patients’ kidney function worsened or remained unchanged whereas only two (18%) had improvement in renal function (15). The present study, which involved a similar cohort, showed much better results, with only one death (6%) within 30 days after intervention and 50% of patients discontinuing their need for dialysis at an average follow-up period of 564 days ⫾ 533.
1568
•
Renal Artery Stent Placement for Restoration of Renal Function
Previously published studies have grouped and analyzed dialysis-dependent and non– dialysis-dependent patients together, meaning their conclusions and applicability to dialysisdependent patients alone are difficult to determine (5,6,16). Several previous articles have highlighted the importance of proteinuria, glomerular filtration rate, and severity of RAS as predictors of successful outcome after renal artery revascularization (20,21). In the present study, we observed significant differences in proteinuria between responders and nonresponders. In addition, our study draws attention to the prognostic role of proteinuria as a marker for microvascular injury before renal artery stent placement. This was a small retrospective study with inherent limitations. This is not a population-based study with subjects who constitute a selective group of patients referred to our tertiary care center. Therefore, our results are associated with selection bias and cannot be generalized. The power of our conclusions is limited by the small sample size. Despite these limitations, the study highlights the potential role of percutaneous renovascular revascularization in dialysis-dependent ischemic nephropathy in improving dialysis-free survival. Percutaneous renovascular revascularization in patients with dialysisdependent ischemic nephropathy is associated with a high technical success rate and low complication rates. In our series, there were several potential predicators of a successful outcome of dialysis discontinuation after renal stent placement, including baseline average kidney size, 24-hour predicted proteinuria, and renal function before dialysis. There was no survival benefit with revascularization in this patient cohort, potentially because of the short follow-up and small group of patients. Future studies need to be
performed to confirm the validity of these observations. References 1. Hansen KJ. Prevalence of ischemic nephropathy in the atherosclerotic population. Am J Kidney Dis 1994; 24: 615– 621. 2. Greco BA, Breyer JA. The natural history of renal artery stenosis: who should be evaluated for suspected ischemic nephropathy? Sem Nephrol 1996; 16:2–11. 3. Dorros G, Jaff M, Jain A, Dufek C, Mathiak L. Follow-up of primary Palmaz-Schatz stent placement for atherosclerotic renal artery stenosis. Am J Cardiol 1995; 75:1051–1055. 4. Dorros G, Jaff M, Mathiak L, et al. Four-year follow-up of Palmaz-Schatz stent revascularization as treatment for atherosclerotic renal artery stenosis. Circulation 1998; 98:642– 647. 5. Dean RH, Tribble RW, Hansen KJ, O’Neil E, Craven TE, Redding JF II. Evolution of renal insufficiency in ischemic nephropathy. Ann Surg 1991; 213: 446 – 455. 6. O’Donovan RM, Gutierrez OH, Izzo JL Jr. Preservation of renal function by percutaneous renal angioplasty in high-risk elderly patients: Short-term outcome. Nephron 1992; 60:187–192. 7. Wengrovitz M, Healy DA, Diamond JR, Atnip RG. Renal revascularization in patients on dialysis. J Cardiovasc Surg 1995; 36:241–246. 8. Shannon HM, Gillespie IN, Moss JG. Salvage of the solitary kidney by insertion of a renal artery stent. AJR Am J Roentgenol 1998; 171:217–222. 9. Xin G, Wang M, Jiao LL, Xu GB, Wang HY. Protein-to-creatinine ratio in spot urine samples as a predictor of quantitation of proteinuria. Clin Chim Acta 2004; 350:35–39. 10. Rundback JH, Sacks D, Kent KC, et al. Guidelines for the reporting of renal artery revascularization in clinical trials. J Vasc Interv Radiol 2003; 14(suppl):S477– S492. 11. Ivanovic V, McKusick MA, Johnson CM III, et al. Renal artery stent placement: complications at a single tertiary
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
November 2008
JVIR
care center. J Vasc Interv Radiol 2003; 14:217–225. Misra S, Sturludottir M, Matthews V, Bjarnason H, McKusick M, Iyer V. Treatment of complex stenoses involving renal artery bifurcations using drug-eluting stents. J Vasc Interv Radiol 2008; 19:272–278. Misra S, Thatipelli MR, Howe PW , et al. Preliminary study of the use of drug-eluting stents in atherosclerotic renal artery stenoses 4 mm in diameter or smaller. J Vasc Interv Radiol 2008; 19:833– 839. Collins AJ, Kasiske B, Herzog C, et al. Excerpts from the united states renal data system 2003 annual data report: atlas of end-stage renal disease in the united states. Am J Kidney Dis 2003; 42:A5–A7. Korsakas S, Mohaupt MG, Dinkel HP, et al. Delay of dialysis in end-stage renal failure: prospective study on percutaneous renal artery interventions. Kidney Int 2004; 65:251–258. Paulsen D, Klow NE, Rogstad B, et al. Preservation of renal function by percutaneous transluminal angioplasty in ischaemic renal disease. Nephrol Dial Transplant 1999; 14:1454 –1461. Ascer E, Gennaro M, Rogers D. Unilateral renal artery revascularization can salvage renal function and terminate dialysis in selected patients with uremia. J Vasc Surg 1993; 18:1012– 1018. Hansen KJ, Thomason RB, Craven TE, et al. Surgical management of dialysis-dependent ischemic nephropathy. J Vasc Surg 1995; 21:197–209. Kaylor WM, Novick AC, Ziegelbaum M, Vidt DG. Reversal of end stage renal failure with surgical revascularization in patients with atherosclerotic renal artery occlusion. J Urol 1989; 141: 486 – 488. Cheung CM, Wright JR, Shurrab AE, et al. Epidemiology of renal dysfunction and patient outcome in atherosclerotic renal artery occlusion. J Am Soc Nephrol 2002; 13:149 –157. Makanjuola AD, Suresh M, Laboi P, Kalra PA, Scoble JE. Proteinuria in atherosclerotic renovascular disease. Q J Med 1999; 92:515–518.