Author + information
- Received November 25, 2003
- Revision received November 1, 2004
- Accepted November 30, 2005
- Published online September 6, 2005.
- Krishna Rocha-Singh, MD⁎,⁎ (, )
- Michael R. Jaff, DO†,
- Kenneth Rosenfield, MD†,
- ASPIRE-2 Trial Investigators
- ↵⁎Reprint requests and correspondence:
Dr. Krishna Rocha-Singh, Vascular Medicine Program, Prairie Heart Institute, PO Box 19420, Springfield, Illinois 62794-9420
Objectives This study sought to define the safety and durability of renal stenting after suboptimal/failed renal artery angioplasty in patients with suspected renovascular hypertension.
Background Few prospective multicenter studies have detailed the safety, efficacy, and long-term clinical benefits of renal artery stent revascularization in hypertensive patients with aorto-ostial atherosclerotic renal artery lesions.
Methods This non-randomized study enrolled 208 patients with de novo or restenotic ≥70% aorto-ostial renal artery stenoses, who underwent implantation of a balloon-expandable stent after unsuccessful percutaneous transluminal renal angioplasty (PTRA), which was defined as a ≥50% residual stenosis, persistent translesional pressure gradient, or a flow-limiting dissection. The primary end point was the nine-month quantitative angiographic or duplex ultrasonography restenosis rate adjudicated by core laboratory analysis. Secondary end points included renal function, blood pressure, and cumulative incidence of major adverse events and target lesion revascularization at 24 months.
Results The stent procedure was immediately successful in 182 of 227 (80.2%) lesions treated. The nine-month restenosis rate was 17.4%. Systolic/diastolic blood pressure decreased from 168 ± 25/82 ± 13 mm Hg (mean ± standard deviation) at baseline to 149 ± 24/77 ± 12 mm Hg at 9 months (p < 0.001 vs. baseline), and 149 ± 25/77 ± 12 mm Hg at 24 months (p < 0.001 vs. baseline). Mean serum creatinine concentration was unchanged from baseline values at 9 and 24 months. The 24-month cumulative rate of major adverse events was 19.7%.
Conclusions In hypertensive patients with aorto-ostial atherosclerotic renal artery stenosis in whom PTRA is unsuccessful, Palmaz (Cordis Corp., Warren, New Jersey) balloon-expandable stents provide a safe and durable revascularization strategy, with a beneficial impact on hypertension.
Atherosclerotic renal artery stenosis is a common disorder often associated with coronary artery disease (1), aorto-iliac and infra-inguinal arterial disease (2), impaired renal function (3), and hypertension (4). Despite the proven efficacy of surgical revascularization (5,6), endovascular therapy has emerged as the preferred strategy for treatment. Percutaneous transluminal renal angioplasty (PTRA) and endoluminal stenting are being performed at an increasing rate despite a paucity of well-controlled multicenter data with long-term clinical follow-up confirming their efficacy. Several investigators who conducted single and multicenter studies and registries support the use of PTRA and stenting, especially in patients with the most complex form of the disease (7–10).
We report the results of the first multicenter study performed in patients who underwent the implantation of balloon-expandable stents immediately after unsuccessful PTRA, followed up prospectively by independent core angiography and duplex ultrasound laboratories to monitor the treatment safety, vessel patency, and long-term effects on blood pressure and renal function.
This prospective longitudinal non-randomized study was conducted at 23 U.S. medical centers between December 1997 and May 1999. The ethical review committees of participating institutions approved the study, and all patients signed informed consent before enrollment. Patients enrolled had uncontrolled hypertension, serum creatinine concentrations ≤3.0 mg/dl, ≥70% de novo or restenotic renal artery atherosclerotic stenoses, and persistent peak-to-peak translesional pressure gradient of ≥20 mm Hg, flow-limiting dissections, or residual ≥50% stenoses after PTRA attempts. Renal artery stenoses were unilateral or bilateral, within 10 mm of the aorto-renal artery border. Patients with accessory (polar) renal arteries were eligible for inclusion in the study if the arterial diameter was ≥4 mm. Pre-procedural hypertension, defined as a blood pressure >140/90 mm Hg, had been refractory to combined therapy with more than two antihypertensive agents administered in appropriate doses. Patients treated with a single antihypertensive drug were also eligible if they had a ≥24 mm Hg increase in systolic or a ≥10 mm Hg increase in diastolic blood pressure within four months before enrollment.
Criteria for exclusion from the study included a successful renal angioplasty, sequential stenoses in a single renal artery, a renal artery diameter <4 mm or >8 mm, an occluded renal artery, the need for more than two stents, a major vascular complication after PTRA, stenosis of a transplant or bypass graft anastomosis, non-atherosclerotic disease, serum creatinine ≥3.0 mg/dl, kidney length <8.0 cm, intolerance to aspirin, a life expectancy of fewer than two years, known hemorrhagic diathesis or hypercoagulable state, contraindication to receiving heparin, myocardial infarction within 30 days, an abdominal aortic aneurysm measuring >4.0 cm in diameter, current pregnancy, inability to grant informed consent, or patient refusal to undergo surgery to repair the renal artery or vascular access site in the event of a complication.
Pre-procedural laboratory testing consisted of routine screening blood and urine examinations, including a detailed lipid profile, abdominal and renal ultrasound studies, and a nonselective aorto-renal angiogram. Brachial blood pressure was measured in the sitting position according to American Heart Association guidelines (11).
Revascularization procedure and follow-up
Aspirin, 81 to 500 mg, was administered at least one day before undergoing PTRA, and continued thereafter at the discretion of the investigator in a daily dose of 81 to 500 mg. Intra-arterial heparin, 3,000 to 10,000 U bolus, was administered during PTRA. The procedure was performed in a standard fashion via the transfemoral or brachial approach, with angiographic images obtained to best visualize the origin of the renal artery. When the operator determined that PTRA had failed based on the criteria described, the patient was eligible for enrollment in the study and implantation of the stent.
A Palmaz (Cordis Corp., Warren, New Jersey) balloon-expandable stent (P104, P154, P204) was inserted at minimum balloon inflation pressures of 6 to 13 atm. The final diameter of the stent was determined by the diameter of the fully inflated delivery balloon. Proper positioning of the stent was verified in multiple fluoroscopic views, and was considered optimal when located within the renal artery ostium, entirely covering the lesion, and extending by ≤2 mm into the aorta. The procedural end points consisted of the smallest achievable residual stenosis and pressure gradient, obtained as simultaneous measurements. Final measurements included minimal lumen diameter, percent residual stenosis of the stented region, and mean translesion pressure gradient after device implantation. The Brigham and Women’s Hospital Cardiovascular Research Institute Angiographic Core Laboratory, Boston, Massachusetts, analyzed all angiographic images.
Patient follow-up included visits at 30 days and at 6, 9, and 24 months. At each visit, in addition to the collection of blood and urine for routine laboratory screening, special attention was paid to adverse event surveillance and measurements of resting systolic and diastolic blood pressure, as described earlier, and the concurrent antihypertensive drug regimen.
In the first 65 patients, nine-month renal angiography was performed to evaluate for in-stent restenosis. In the remainder of the population, restenosis was ascertained by a nine-month renal artery duplex ultrasound examination, because investigators thought that subjecting potentially asymptomatic patients to an invasive procedure was not justified. Renal artery duplex ultrasonography has been shown to be highly accurate in predicting native renal artery stenosis with a confirmed 91% sensitivity and a 97% specificity (12). Duplex ultrasonography of the renal artery and stent, recorded in real time, was performed according to a standardized protocol from the Washington Hospital Center Vascular Ultrasound Core Laboratory. The duplex ultrasound study was interpreted by the core laboratory as indicative of restenosis when the renal/aortic ratio within the stent was ≥3.5, or when the absolute peak systolic velocity within the stent was >200 cm/s with post-stenotic turbulence. A confirmatory angiogram was performed if the duplex ultrasound results suggested restenosis. Unscheduled renal angiograms were performed in case of interim loss of blood pressure control, deteriorating renal function without other systemic etiology, or other clinical manifestations pointing to the development of significant renal artery restenosis. The assigned core laboratories reviewed all angiographic, hemodynamic, and duplex studies obtained during long-term follow-up.
Study end points
The primary end point of A Study to evaluate the safety and effectiveness of the Palmaz balloon expandable stent In the REnal artery after failed angioplasty (ASPIRE-2) was the incidence of in-stent renal artery restenosis at nine months determined by duplex ultrasound, as defined earlier, or by angiography, defined as ≥50% diameter stenosis. Secondary end points included: 1) acute procedural success, defined as <30% residual stenosis immediately after stent deployment as determined by the Core Laboratory and ≤5 mm Hg residual mean translesion gradient; 2) technical success, defined as the successful placement of the stent at the lesion site; 3) worsening of renal function at 30 days, 6 months, 9 months, and 24 months, defined by a ≥50% increase in serum creatinine if the baseline level was ≤2.0 mg/dl, or a 1 mg/dl increase if the baseline level was >2.0 mg/dl; 4) treatment benefit, estimated from changes in systemic blood pressure and/or concurrent antihypertensive regimen at any time point of the follow-up; 5) absence of major adverse events (MAE) at 30 days, 3 months, 6 months, 9 months, and 24 months, defined as in-hospital procedure- or device-related death or Q-wave myocardial infarction, target lesion revascularization or significant systemic athero-embolic events resulting in end-organ damage (e.g., unanticipated kidney/bowel infarct, lower extremity ulceration or gangrene, loss of renal function).
The ASPIRE-2 study was designed as a non-randomized, non-inferiority study. As such, renal artery stenting after failed/suboptimal PTRA was compared with routine PTRA by establishing an objective performance criteria (OPC). This reflects the end point rate (i.e., renal artery restenosis), established by a literature review, as the standard for comparison with a new technique or technology (i.e., renal artery stenting). A 200-patient sample size was calculated using a 40% post-PTRA OPC restenosis rate to compare with renal stent restenosis with 80% power, a 10% loss to follow-up rate, and an estimated ≤31% nine-month restenosis rate associated with the Palmaz renal stent delivery system.
Results are presented as means ± standard deviation. Data were analyzed at 9 and 24 months on an intention-to-treat basis, including all patients in whom treatment with the balloon-expandable stent was attempted and 95% confidence intervals were calculated. All procedural complications and MAE are reported descriptively. The cumulative rate of MAE is presented in a Kaplan-Meier survival analysis. Computations were performed with the SAS statistical analysis software package (SAS Institute Inc., Cary, North Carolina).
The main baseline characteristics of the study population are presented in Table 1.Approximately two-thirds of the patients were women, with 68% current or previous smokers, 53% treated for blood lipid abnormalities, and 26% diabetic patients. Treatment of 252 lesions was attempted in 165 patients with single unilateral lesions and in 43 patients with bilateral lesions. One patient with bilateral disease underwent stenting of three separate lesions.
Immediate and nine-month angiographic and duplex ultrasound results
The baseline characteristics of 244 lesions with quantitative angiographic analyses available are shown in Table 2.Renal artery stenoses were evenly distributed between right and left renal arteries, in an ostial position in nearly 90%, and concentric in three-fourths of cases. The mean percent diameter stenosis of the target vessel after PTRA was 35.5 ± 20.1, versus –2.2 ± 17.5 immediately after stent implantation. The cumulative frequency distribution of percent diameter stenosis by quantitative angiographic analysis immediately before and after stent implantation is shown in Figure 1.Acute procedural success was observed in 182 of 227 lesions treated (80.2%).
At nine months, angiographic and/or duplex ultrasound follow-up examinations were available in 153 patients (74%) and 184 lesions treated (73%); 55 patients (26.4%) died, withdrew consent, refused follow-up testing, or were lost to follow-up. Of the 153 patients with nine-month testing, 89 patients (58%) underwent ultrasound only and 64 (42%) had either angiography and/or ultrasound. Restenosis was observed in 17.4% of lesions, well below the OPC-derived restenosis rate of 40%. The number of patients who underwent “confirmatory” renal angiography after having an abnormal duplex ultrasound was not explicitly tracked in this trial. Among multiple clinical and angiographic characteristics examined, a history of diabetes (p = 0.03), a smaller pre-procedure reference vessel diameter (p = 0.04), and smaller post-procedure in-lesion (p = 0.04) and in-stent (p = 0.005) minimum lumen diameters (MLD) were predictors of restenosis.
The clinical status was ascertained in 194 (93%) and 164 (79%) patients at 9 and 24 months, respectively. At two years, 85.9% of patients remained free of target lesion revascularization.
Table 3presents the impact of renal artery stenting on systolic and diastolic blood pressure measurements and the number of antihypertensive drugs prescribed before PTRA and at follow-up. Compared with baseline measurements, significant decreases were observed in the mean systolic and diastolic blood pressures at all time points. The Kaplan-Meier cumulative probability of treatment benefit is presented in Figure 2.
In the overall population, the mean serum creatinine concentration increased from 1.36 ± 0.52 mg/dl at baseline to 1.40 ± 0.61 mg/dl at 9 months (p = NS vs. baseline) to 1.46 ± 0.81 mg/dl at 24 months (p = 0.04 vs. baseline). In the subgroup of patients with baseline serum creatinine concentrations ≥1.5 mg/dl, the mean concentration was 1.94 ± 0.39 mg/dl at baseline, 1.87 ± 0.58 mg/dl at 9 months (p = NS), and 1.93 ± 0.71 mg/dl at 24 months (p = NS). Of a total of 63 patients with a baseline serum creatinine ≥1.5 mg/dl, 5 (7.9%) had worsening of renal function at nine months. At 24 months, the condition of 4 of 53 patients (7.5%) with abnormal baseline renal function who were re-evaluated had worsened. However, in this cohort with progressive renal dysfunction, no patients required temporary or permanent hemodialysis.
Adverse clinical events
The cumulative rate of MAE after 24 months of follow-up was 19.7% (Table 4).The Kaplan-Meier analysis of the MAE-free survival is presented in Figure 3.Other serious adverse clinical events, consisting of stent thrombosis, major hemorrhage, and access site complications, occurred in 1%, 1.4%, and 4.8% of patients, respectively. Among all clinical and angiographic variables examined, none were a significant predictor of MAE.
Unilateral versus bilateral stenting
At nine months, angiographic or duplex ultrasound data were available for 65 of the 87 lesions (75%) treated among the 43 patients with bilateral renal artery stenoses, and for 119 lesions (72%) among the 165 patients with single unilateral stenoses. The restenosis rate was 16.9% (11 of 65 lesions) in the subgroup of patients with bilateral disease, versus 16.8% (20 of 119 lesions) among patients with unilateral stenoses (p = NS). A separate analysis of the 43 patients who underwent stenting for stenoses in both renal arteries showed the same degree of treatment effect on systolic blood pressure and hypertensive medications as in the overall cohort, although the effect on diastolic pressure had mostly dissipated by the end of follow-up. No significant difference was observed in the overall probability of a treatment benefit among the 43 patients who underwent bilateral versus the 165 patients who underwent unilateral stenting. Furthermore, as observed in the group of patients with unilateral disease, bilateral renal stenting had no measurable effect on long-term renal function (data not shown).
Our large, prospective multicenter evaluation shows that use of stainless steel balloon-expandable stents for aorto-ostial renal artery stenosis is a safe and effective therapy after unsuccessful balloon angioplasty. Additionally, this study is the first to define the pre-intervention and post-intervention angiographic characteristics and nine-month stent restenosis rate as determined by independent core angiographic and duplex ultrasound laboratories. This study also establishes a favorable impact of stent implantation on blood pressure throughout the 24-month follow-up.
Several single-center observational trials (7,13–14) have established renal stenting as an attractive adjunct to conventional balloon angioplasty for atherosclerotic aorto-ostial stenosis because of its utility in resolving elastic recoil, dissection flaps, and residual translesional pressure gradients that may occur after angioplasty. However, small patient numbers, short follow-up durations, the potential for single-center bias, and the inconsistent use of core laboratories to define procedural success and stent patency limited these studies. Furthermore, early renal stent studies reported a troubling incidence of major complications, including technical failure, stent thrombosis, renal and systemic athero-embolization, major hemorrhage, and death (15,16). Other studies also raised a concern over the incidence of late in-stent restenosis, which ranged from 11% (7) to 44% (17). Our multicenter evaluation of 208 patients quantified the efficacy of renal stenting to resolve a post-PTRA residual stenosis, reducing it from 35.5 ± 20% post-PTRA to −2.2% after stenting (Fig. 1). Our experience also confirmed an excellent 94.9% technical success rate and high procedural safety as we noted no perforations, no renal artery ruptures, and no device or in-hospital procedure related deaths. Importantly, we observed a 1.4% rate of clinically evident in-hospital episodes of athero-embolization that were not associated with an increase in post-stent serum creatinine values or with mortality. By comparison, surgical revascularization, although effective in controlling hypertension and salvaging renal function, is associated with a higher procedural risk of morbidity and mortality (5,6,18,19). Our renal artery duplex-defined nine-month restenosis rate is comparable to the 11% rate reported by Blum et al. (7) and is significantly better than the 44% rate noted by Tullis et al. (17). The significant improvement in MLD after successful stenting compared with balloon angioplasty is not merely cosmetic, because the greater final MLD was associated with reduced restenosis and confirms the importance of a stent deployment strategy to safely maximize the post-deployment MLD. Our study corroborates this previous observation (14) and the known association between smaller pre-procedure reference vessel diameter and increased nine-month restenosis rate.
Presently, there is no standardized method for the quantitative interpretation of renal angiograms. The presence of an aorto-ostial stenosis often results in post-stenotic dilatation, making the determination of a “normal” main renal artery reference diameter difficult, and may result in the visual over-estimation of the severity of the renal artery stenosis. In this study, inclusion criteria required a ≥70% renal artery stenosis as assessed by visual estimation. However, our independent core angiographic laboratory defined a pre-intervention mean lesion severity of 61.5%. Similarly, a post-PTRA ≥50% visually estimated residual stenosis was required for enrollment. However, the angiographic core laboratory defined the mean residual stenosis as 35%. These visual over-estimations of lesion severity by the angiographer may ultimately impact the clinical decision to proceed with intervention for a physiologically insignificant stenosis (20) and may explain the variation in the hypertension response noted in various studies. Although contrast angiography remains the gold standard for the determination of renal artery lesion severity, our investigation suggests that a higher percent diameter threshold or use of quantitative computer angiography (QCA) should be adopted to confirm a critical renal artery stenosis in patients with suspected renovascular hypertension.
Stent implantation was associated with a significant overall lowering of mean systolic and diastolic blood pressure and a reduction of the number of anti-hypertensive medications (Table 3). Blood pressure improvement was observed within 24 h of successful stent implantation and persisted through 24 months of follow-up. However, despite a lowering of blood pressure in the cohort in general, only 47% of all patients experienced a “cure” or “improvement” in blood pressure at 24 months. This lower rate of clinical efficacy in our trial compared with other reports may reflect differences in the definitions of cure and improved used in our study, the absence of a minimum systolic/diastolic blood pressure criterion, or the lack of secondary confirmation of a critical renal artery stenosis by either duplex ultrasonography or QCA before study enrollment. Other investigators have reported an improvement in blood pressure in 62% (7) and 76% of patients (10) at one-year follow-up. Notably, both studies required duplex ultrasonography or QCA documentation of a critical renal stenosis before study enrollment. Therefore, it seems that many of our patients may have had primary (essential) hypertension, because they derived no demonstrable blood pressure benefit from successful renal stenting. This inconsistent blood pressure response to renal stenting highlights the importance of appropriate patient selection, particularly in light of the observed 19.7% major adverse event at two-year follow-up. Notably, we found no clinical, angiographic, or procedure-related predictor of improved blood pressure control. Although the mean impact on diastolic blood pressure was more durable in the group with unilateral disease, our patients, whether they were treated for global ischemic nephropathy (bilateral stenosis or stenosis to a solitary functioning kidney) or for unilateral renal stenosis, had the same overall likelihood of improved blood pressure control.
An important concern surrounds the safety of stenting in patients with atherosclerotic renovascular disease and renal dysfunction. In 26 patients with baseline renal dysfunction (mean serum creatinine, 2.4 ± 0.25 mg/dl) we observed no significant decline in renal function at 24 months after stent implantation (serum creatinine, 2.3 ± 0.8; p = NS). Furthermore, in patients with bilateral renal artery stenoses and renal dysfunction, stenting of both arteries at the same setting was performed routinely without a deleterious effect on renal function. This patient cohort with bilateral renal atherosclerosis and renal dysfunction is of particular clinical interest because natural history studies suggest that they may be at increased risk for renal atrophy and possible progression to end-stage renal disease (21,22). Unfortunately, there are only limited reports of stent revascularization to stabilize or slow progressive renal dysfunction in patients with ischemic nephropathy (23–25). Although these studies suggest a potential role for renal stenting in this highly selected population, they are limited by the small number of patients studied; limited clinical follow-up; inconsistent definitions of renal function improvement, stabilization, and decline; and lack of an optimally medically treated control cohort. Our patient population with presumed ischemic nephropathy was small (n = 40), and therefore, this pertinent question was not addressed in our study.
Several potential limitations of this investigation must be considered in interpreting the results. First, our study was not a randomized comparison with a medical therapy or balloon angioplasty control group, and therefore direct comparisons are not possible. Second, as the first prospective trial of renal artery stenting to use a core laboratory to adjudicate adherence to angiographic inclusion criteria and define angiographic success, we observed a disparity between the visually assessed renal artery lesion severity and the core laboratory measurements. This disparity may reflect the inherent inaccuracy of visual estimation, operator inexperience, or technique. The inclusion of patients with non-critical renal lesions may, in part, explain the relatively low percentage of patients who experienced a favorable blood pressure response and emphasizes the potential risk of exposing patients to an unnecessary invasive therapy. Furthermore, it suggests the need to consider a higher percent diameter stenosis threshold for intervention when visual estimation is used or the more consistent use of on-line QCA. Finally, we encountered a relatively low percentage of patients (74%) who returned for the nine-month duplex Doppler or angiographic assessment for in-stent restenosis. As such, we cannot exclude a potentially higher restenosis rate in this cohort.
This prospective, multicenter investigation clearly establishes the safety and efficacy of aorto-ostial renal artery stenting as an adjunct to balloon angioplasty stenosis, and defines an acceptable nine-month restenosis rate of 17.4% while highlighting the challenges in identifying those patients who may maximally benefit from stent revascularization. Although not a randomized trial, the low complication rate and associated long-term clinical efficacy should establish stenting as the initial consideration in patients requiring renal revascularization. Although near-term risks and mortality are higher with surgery, the long-term clinical durability and stent patency must be defined by additional studies. In the near future, new technologies, including drug-eluting renal stents and distal embolic protection devices, may be routinely used to reduce restenosis and renal athero-embolization to optimize clinical outcomes in selected patient cohorts.
For a list of the investigators and institutions that participated in the ASPIRE-2 study, please see the online version of this article.
This study was supported by a grant from Cordis Corporation, a Johnson & Johnson Company. Drs. Jaff, Rosenfield, and Rocha-Singh are presently consultants for Cordis Corporation.
- Abbreviations and Acronyms
- major adverse events
- minimum lumen diameters
- objective performance criteria
- percutaneous transluminal renal angioplasty
- quantitative computer angiography
- Received November 25, 2003.
- Revision received November 1, 2004.
- Accepted November 30, 2005.
- American College of Cardiology Foundation
- Pickering T.G.,
- Laragh J.H.
- Zeller T.,
- Frank U.,
- Muller C.,
- et al.
- Perloff D.,
- Grim C.,
- Flack J.,
- et al.
- White C.J.,
- Ramee S.R.,
- Collins T.J.,
- et al.
- Waston P.S.,
- Hadjipetrou P.,
- Cox S.V.,
- et al.
- Beutler J.J.,
- Van Ampting J.M.,
- Van De Ven P.J.,
- et al.
- Rocha-Singh K.,
- Ahuja R.,
- Sung C.H.,
- et al.