Drug-Eluting Stents in Patients With ESRD on Dialysis: A Small Step Forward

Cardiovascular disorders represent the single most significant comorbid condition for patients with end-stage renal disease (ESRD) on dialysis therapy, leading to approximately 45% of deaths in this patient population.1 At the time of initiation of dialysis therapy, obstructive coronary artery disease (CAD) is present in greater than 50% of patients.2 One-year mortality after an acute myocardial infarction in this population is strikingly high (up to 60%).3 Despite these facts and the major advances in medical therapeutics and percutaneous and surgical revascularization for CAD during the last 2 decades, dialysis patients have been systematically excluded from clinical trials evaluating and establishing these advances.4, 5 Thus, high-quality data to support one approach or another in the management of CAD in dialysis patients are scarce, and treatment algorithms frequently are based on data from small retrospective studies, subgroup analyses from larger trials, and clinical experience. The study by Yachi et al6 published in this issue of the journal sheds some light on the value of drug-eluting stents (DESs), the most important recent innovation in percutaneous coronary revascularization, in the dialysis patient population.

In addition to the high prevalence and poor prognostic implication, CAD lesions in dialysis patients are frequently complex, heavily calcified and involve multiple territories,7 prompting frequent referral for revascularization. The choice between surgical and percutaneous revascularization approaches cannot be based on solid data, simply because none exist. Moreover, extrapolation of accepted traditional algorithms established for patients without ESRD can be misleading because the risk-benefit ratio in this patient population differs significantly from that in patients with normal or mildly impaired kidney function. The frequently encountered multivessel CAD in this patient population would suggest that they are best served by surgical revascularization, namely coronary artery bypass grafting (CABG). Even with the availability of DESs and their marked impact on restenosis and need for target-vessel revascularization, the most contemporary published randomized clinical trial shows that anatomically high-risk or complex coronary lesions and high-risk patients have a reduced risk of repeated revascularization with CABG than with DESs.8 However, in this and many other cardiovascular trials, kidney function and the presence or absence of ESRD is not addressed in the overall results.5 The extension of such data to the dialysis patient population is hampered by their greater risk of perioperative complications. The late mortality advantage of CABG over angioplasty and stent placement in patients with multivessel disease is relatively unproven in dialysis patients, who have very high short- and midterm mortality.1 Moreover, numerous registries have shown that CABG operative mortality for patients with ESRD is greater than 3-fold that of patients with normal kidney function.9, 10, 11 The risk of perioperative stroke in dialysis patients undergoing CABG is doubled, and the risk of perioperative infection is increased significantly.9

A similar dilemma exists with stent-based percutaneous coronary intervention (PCI) in patients with ESRD on dialysis therapy. Nonrandomized data have shown that outcomes of elective PCI in dialysis patients have improved with the introduction of bare-metal stents (BMSs) in the mid 1990s, leading to high procedural success and effective relief of angina. However, the efficacy of BMSs is markedly diminished in this population because of the greater risk of restenosis and recurrence of ischemia within months after PCI.1, 12 The availability of first- and second-generation DESs has the potential to shift the fault lines between PCI and CABG. Although there have been fluctuations in levels of DES use, these devices have become the most commonly used devices in PCI today, with use exceeding 60% in most institutions. The primary advantage of DESs is attributed to the dramatically decreased risk of restenosis; up to 60% relative risk reduction in preapproval phase 3 trials.13, 14, 15 DESs appear to be a very attractive revascularization option in patients with ESRD: they provide a much lower risk of procedural death or stroke compared with CABG and are expected to reduce restenosis and the need for repeated procedures compared with traditional PCI using BMSs. However, similar to CABG and BMS data, clinical trials comparing BMS with DES systematically excluded patients with ESRD.5 Data for DES use in dialysis patients are scarce, observational in nature, and based on retrospective analyses of small cohorts.16, 17, 18, 19, 20, 21, 22

In this issue of the journal, Yachi et al6 compared outcomes of DESs and BMSs in a small cohort of dialysis patients. Similar to previously published retrospective studies (Table 1),16, 17, 18, 19, 20, 21, 22 the procedural success rate is similarly high with both modalities despite more complex lesions in the DES cohort. Patients treated with DESs experienced fewer major adverse cardiac events and had lower all-cause mortality than patients treated with BMSs (a historic control group). The investigators also examined angiographic restenosis and its pattern. As expected, patients who received DESs had lower in-stent late lumen loss at follow-up. Although the incidence of binary in-stent restenosis was lower with DESs, the difference did not reach statistical significance. In addition, in-stent restenosis is more focal in DESs and more diffuse in BMSs.

Table 1. Published Reports of Adverse Cardiac Outcomes With Drug-Eluting Stent Use in Dialysis Patients

		Hassani et al18	Mishkel et al19	Nakazawa et al20	Das et al16	Halkin et al17	Yachi et al6	Aoyama et al22
	Control group	Patients with mildly or moderately decreased kidney function receiving DES			Patients on dialysis receiving BMS
	No. of dialysis patients receiving DESs	64	23	31	24	33	56	88
	Country of origin	US	US	Japan	US	US	Japan	Japan
	Follow-up (mo)	6	24	8	9	12	9	12
	Mortality (%)	14	24⁎	6.5	25	18	4.4†	3.4
	Myocardial infarction (%)	9	—	0	8	10	0	2.4
	Target vessel/lesion revascularization (%)	8	4.5	32.3	4	8	21‡	17‡
	Major adverse cardiac events (%)	23	33	38.7	33†	25†	25†	—

Abbreviations: BMS, bare-metal stent; DES, drug-eluting stent.

⁎ Eighteen percent at 1 year. † Statistically significant decrease compared with the control group. ‡ Target-lesion revascularization, including asymptomatic patients in whom restenosis is identified during scheduled follow-up angiography.

It is difficult to draw conclusions and change practice patterns based on such small studies. However, examining the study by Yachi et al6 in conjunction with other reports of DESs in dialysis patients shows some common features16, 17, 18, 19, 20, 21, 22 (Table 1). First, in all reports comparing DES with BMS use, there was a significant reduction in adverse events, primarily driven by reduction in the need for repeated revascularization. Second, there was no evidence of increased risk of myocardial infarction with DES use, although infarction rates varied among the different reports. This is important because it suggests that the risk of stent thrombosis, a feared complication in patients with ESRD with heightened inflammatory and thrombotic states, does not seem to be greater than for patients without ESRD or with BMS use. Third, in all reports, there was a dismally high rate of adverse events in a relatively short follow-up. In the Japanese reports, the high adverse event rate is driven by a high revascularization rate, which can be explained at least in part by the study design mandating angiographic follow-up. However, in US reports, adverse events mostly are caused by high early mortality. Reasons for the discrepancy in early mortality between the Japanese- and US-based studies are unclear, but more likely are related to patient and disease characteristics than to the revascularization procedure itself. The reasons underlying improved all-cause mortality with DESs reported by Yachi et al6 are not clear. Although it can simply be a statistical coincidence given the small number of patients, it also may reflect improved medical care and/or advances in periprocedural and postprocedural adjunctive therapy. A decrease in coronary events and possibly death also may be secondary to the extended dual antiplatelet regimens followed after DES (and not BMS) implantation. It is worth mentioning that decreased mortality after DES implantation has been reported by several large registries that included patients with chronic kidney disease.23, 24 The exact mechanisms underlying this finding in these larger registries also are unclear.

Similar to other reports in the literature, this study is limited by its size and design: the small number of patients included and the nonrandomized nature of the study weaken its conclusions. Historic control studies reportedly have favored the new therapies,25 and although the investigators constructed multivariate models to adjust for confounders, even the most sophisticated models fall short of resembling randomization. Other method limitations include the discrepancy in angiographic follow-up rates among groups and use of target-lesion versus target-vessel revascularization, all of which may have influenced the study conclusion.

Nonetheless, the study by Yachi et al6 and similar reports provide important insights into the current PCI practice in patients with ESRD. Use of DESs in this high-risk cohort known for its complex coronary lesions appears to be feasible and safe. It is reasonable to conclude that the beneficial effects of DESs in reducing restenosis and repeated revascularization appear to carry over to patients with ESRD, thus providing a therapeutic option to a patient population in dire need of a safe and effective revascularization strategy.6, 16, 17, 18, 19, 20, 21, 22 However, available data to date are short of concluding the superiority of one revascularization strategy over the others. Randomized and adequately powered studies with appropriate follow-up are desperately needed to examine the different coronary revascularization options in patients with ESRD. What is probably more important than defining the optimal revascularization strategy is reaching a comprehensive understanding of the pathophysiological mechanisms and identifying targets for therapy that can effectively reduce the extremely high rate of adverse cardiac events in dialysis patients.