One-year clinical outcomes in patients with renal insufficiency after contemporary PCI: data from a multicenter registry

Background Chronic kidney disease (CKD) is highly prevalent in patients with coronary artery disease (CAD). Objective The outcome following revascularization using contemporary technologies (new-generation abluminal sirolimus-eluting stents with thin struts) in patients with CKD (i.e., glomerular filtration rate of < 60 mL/min/1.73m2) and in patients with hemodialysis (HD) is unknown. Methods e-Ultimaster is a prospective, single-arm, multi-center registry with clinical follow-up at 3 months and 1 year. Results A total of 19,475 patients were enrolled, including 1466 patients with CKD, with 167 undergoing HD. Patients with CKD had a higher prevalence of overall comorbidities, multiple/small vessel disease (≤ 2.75 mm), bifurcation lesions, and more often left main artery treatments (all p < 0.0001) when compared with patients with normal renal function (reference). CKD patients had a higher risk of target lesion failure (unadjusted OR, 2.51 [95% CI 2.04–3.08]), target vessel failure (OR, 2.44 [95% CI 2.01–2.96]), patient-oriented composite end point (OR, 2.19 [95% CI 1.87–2.56]), and major adverse cardiovascular events (OR, 2.34 [95% CI 1.93–2.83, p for all < 0.0001]) as reference. The rates of target lesion revascularization (OR, 1.17 [95% CI 0.79–1.73], p = 0.44) were not different. Bleeding complications were more frequently observed in CKD than in the reference (all p < 0.0001). Conclusion In this worldwide registry, CKD patients presented with more comorbidities and more complex lesions when compared with the reference population. They experienced higher rate of adverse events at 1-year follow-up. Graphic abstract One-year summary outcomes of contemporary PCI in renal insufficiency. CKD chronic kidney disease, POCE patient oriented composite endpoint, MACE major adverse cardiovascular events, TLF target lesion failure, TLR target lesion revascularization, ST stent thrombosis Electronic supplementary material The online version of this article (10.1007/s00392-019-01575-y) contains supplementary material, which is available to authorized users.


Methods
We analyzed the results of e-Ultimaster (NCT 02188355), a prospective, single-arm, multi-center, international registry with clinical follow-up at 3 months and 1 year (Fig. 1). This study sought to validate the safety and efficacy of an abluminal, sirolimus-coated stent with thin struts (Ultimaster DES) in unselected patients representing everyday practice. Secondary objectives were to evaluate the utilization of DES and the detection of rare events in a representative high-risk patient population such as patients with CKD (eGFR < 60 mL/min/1.73 m 2 ). The predefined objective was to enroll a representative population of patients (≥ 400) with impaired renal function. Additionally, we aimed to identify predictors of major adverse events, to assess access site utilization, vascular complications, procedural particularities, the duration of dual antiplatelet therapy (DAPT), and overall performance of the newest-generation DES (Ultimaster) across different patient and lesion subsets. The primary outcome measure was target lesion failure (TLF) defined as a composite of cardiac death, target vessel-related myocardial infarction (MI), and clinically driven target lesion revascularization (TLR) at 1 year. Patient-oriented composite end point (POCE) was defined as any cause of mortality, any MI or any coronary revascularization. Target vessel failure rate (TVF) was defined as cardiac death, target vessel-related MI, and target vessel revascularization. Additionally, major adverse cardiac events (MACE: cardiac death, any MI, clinically driven TVR, and emergent CABG) were documented. Stent thrombosis was documented and defined according to the Academic Research Consortium definitions. An independent clinical event committee reviewed and adjudicated all end point-related serious adverse events. Inclusion criteria were age ≥ 18 years, eligibility for PCI using DES, reference vessel diameter matches, available Ultimaster DES sizes, and written informed consent. The registry was conducted in accordance with the Declaration of Helsinki and country-specific regulatory requirements. All patients signed informed consent form, reviewed and approved by the institutional review board/ethics committee of each participating center. Patients with CKD were automatically allocated to this sub-study (Fig. 1). The data were collected using Electronic CRF (e-Capture). Patients were enrolled between October 2014 and June 2018 and follow-up is currently still ongoing. The current analysis includes all patients enrolled between October 2014 and November 2016, who had a 1-year follow-up visit completed or had died by the date of census (30 November 2016).

Statistical analyses
Patients' demographics, comorbidities, medical history, target lesion characteristics, and procedural characteristics are summarized with mean, standard deviation for continuous variables, and with frequencies and percentages for discrete variables ( Table 1, 2, 3, 4, supplement Table 1). All variables were tested for normal distribution with the Kolmogorov-Smirnov test. For non-normally distributed variables, the Wilcoxon-signed rank test was used. Fisher´s exact test or Chi-squared tests were used for categorical variables, as appropriate. Furthermore, medians with interquartile ranges (IQR) were reported where applicable. When appropriate, the multivariate logistic regression was used for dichotomous variables to adjust for the known risk and potential confounding factors identified with stepwise regression (multivariable p values thresholds to enter and stay in the model were 0.25 and 0.10, respectively), odds ratios and 95% confidence intervals. Variables considered for entry in the stepwise model include: age, gender, body mass index, smoking, diabetes, hypertension, hypercholesterolemia, renal failure, hemodialysis, family history of heart disease, history of MI, previous PCI, previous CABG, acute coronary syndrome, STEMI, multivessel disease, number of lesions identified, number of lesions treated, target vessel treated (right coronary artery, left main, left ascending coronary artery, circumflex artery, graft), type B2 and C lesions (according to classification of American College of Cardiology/American Heart Association/ [ACC/AHA]), bifurcation, moderate to severe calcification, chronic total occlusion, instent restenosis, ostial lesions, long lesions (≥ 25 mm), small vessels (≤ 2.75 mm), radial access, number of study stents implanted, total length of implanted study stents. Additionally, we included Kaplan-Meier estimates, Kaplan-Meier estimates adjusted for confounding using estimated propensity scores ("overlap weights" method), and hazard ratios for comparison of survival curves using a log-rank test where appropriate [19]. Patients were matched for lesion type C, , current smoker, diabetes mellitus, and age using the Xie, Liu method to control for possible confounding from these prognostic factors [19].

Subgroup of HD patients
Patients with CKD on HD (11.4% of all patients with CKD) were younger (p < 0.0001), had a lower body mass index (p < 0.0001), a higher prevalence of diabetes (p < 0.0001), and less frequently a family history of heart diseases (p = 0.05) compared with patients with CKD not requiring HD (Table 4). In CKD patients on HD, left main arteries were more frequently treated (HD: 12.0% vs. CKD: 5.9%, p = 0.007). Although multiple vessels were not more commonly diseased (HD: 58.7% vs. CKD: 57.7%, p = 0.80), they were more likely to be treated in HD patients (HD: 24.6% vs. CKD: 16.4%, p = 0.01). Debulking strategies were commonly utilized in HD patients, such as balloon pre-dilatation, atherectomy, microcatheters, and intravascular ultrasound. The preferred  access site in HD patients was femoral access, followed by radial or brachial, which is different compared with the CKD population (all; p < 0.0001; Table 4). Access site complications were comparable between the two groups. At 3 months, DAPT rates were significantly lower in patients with CKD on HD compared with non-HD patients with no significant difference at 1-year follow-up (  (Fig. 2b).

Discussion
In this large, contemporary, international, real-world registry, patients with CKD undergoing new-generation DES implantation have a significantly higher risk of 1-year adverse events including TLF, defined as the composite of cardiac death, target vessel-related MI, clinically driven TLR, and more bleeding events. Patients with CKD had a Chronic kidney disease is a worldwide growing health issue, affecting more than 1.5 million patients in Europe and the USA and is associated with worse outcome [1,2]. Cardiovascular disease, specifically CAD, is the leading cause of morbidity and mortality in patients with CKD. The optimal revascularization strategy is being debated and includes CABG and PCI, preferably with the use of DES in CKD [11-13, 20, 21]. While CABG was shown to be superior to PCI in patients with multivessel CAD and endstage renal disease regarding long-term survival, PCI was superior to CABG in short-term survival, stroke, and repeat revascularization [11]. However, data on the long-term performance of new-generation DES indicate no significant difference between PCI and CABG in patients with CKD [11][12][13][22][23][24]. Previous studies on revascularization have shown that patients with CKD after PCI and CABG are at increased risk of death and adverse events, correlating with the severity of renal insufficiency [23][24][25]. Herein, the 1-year event rates for patients with CKD were significantly higher when compared with the reference population. In contrast, the 1-year event rate for patients with preserved renal function was comparable to that in other recently published studies evaluating the safety and performance of new-generation DES indicating good performance of the Ultimaster stent. The prospective COMBO Stent registry, for instance, included 3614 patients and documented TLF in 3.9%, cardiac death in 1.6%, and definite stent thrombosis in 0.5% (versus 3.3%, 1.4%, and 0.4% herein) [25][26][27][28][29][30]. Although complication rates were relatively low in general, patients with CKD had a higher risk of major and minor bleeding as well as access site-related complications leading to prolonged hospitalization and higher rates of end points at discharge (supplement Table 1), which is in line with previous publications [28][29][30][31][32]. Bleeding occurred in 2.3% of all included patients, with twofold higher rates of minor and major bleeding in CKD. In this context, the choice of access site is of special importance. Interestingly, the rates of radial access were lower in CKD when compared with no-CKD patients (74% vs. 83%, p < 0.0001), implying that operators might have chosen different arterial accesses in CKD. One may speculate that this was due to an expectably higher complexity of PCI requiring debulking strategies, intravascular imaging, and maintaining radial arterial integrity to provide future dialysis vascular access. Of note, the use of radial access was associated with a 24% lower risk (OR 0.76, 95% CI 0.63-0.91, p = 0.003) of TLF (Fig. 4) and should also be considered as a bleeding avoidance strategy in patients with CKD.
The deleterious effect of CKD on the vasculature is reflected by significant anatomical and procedural differences (Tables 2 and 4). Patients with CKD have increased prevalence of medial calcification, which may impair response to PCI and pose several challenges [20][21][22]. This is supported by our findings showing a higher number of affected vessels, small vessel disease (≤ 2.75 mm), instent restenosis (primarily in DES), and severely calcified lesions, which were more commonly located in the left main artery. As a result, lesions in CKD patients were frequently treated using peri-interventional imaging (IVUS, OFDI) and debulking strategies, such as balloon pre-and post-dilatation, cutting balloon, atherectomy, and microcatheters. Although more lesions were identified in CKD patients, the number of treated lesions did not differ when compared with the reference population, which might be another indicator for the higher complexity in CKD patients leading to a higher rate of non-complete revascularization.
There is a linear relationship between cardiovascular mortality and decreasing GFR with the highest cardiovascular morbidity and mortality in patients on HD [5][6][7][8]. Patients with CKD on HD exhibit a high risk for cardiovascular events as well as an altered coagulation with both increased thrombotic and bleeding risks, which may be caused by denser clot structures [33]. Interestingly, HD patients were younger compared with CKD non-HD patients and, apart from a higher prevalence of diabetes (64%), did not have significantly more comorbidities. Patients with CKD on HD had higher rates of all-cause mortality (including noncardiac death) and TLF at 1 year compared with patients with CKD not requiring HD (Table 4). Of note, data on the 3-month and 1-year follow-up indicate a lower percentage of patients on DAPT in patients with CKD when compared to the reference population. This might be associated with higher bleeding rates in CKD suspected by patients and physicians, which were subsequently documented in this registry, although with a higher rate of patients on oral anticoagulation in patients with CKD in this registry.

Limitations
It is recognized that registry-based studies have limitations. Follow-up procedures were not standardized and, thus, may have influenced the reporting of safety and adverse events. This may have led to underreporting of adverse events, which may have occurred, due to the discretion of the investigator. The characteristics of the patients included in this study were representative of patients with CKD in everyday practice. However, the results could be influenced by confounders in patient characteristics such as age and comorbidities. Furthermore, there may be biological heterogeneity, e.g., due to different disease severity regarding categorical variables, different glomerular filtration rates within the groups, and underlying causes of CKD within the groups. Also, adherence to medication is often dynamic, and rigorous assessment with toxicological screening and type of DAPT was not reported.

Conclusion
In this large cohort of all-comer patients undergoing newgeneration DES implantation, CKD and in particular HD were associated with a higher risk of adverse events including TLF when compared with the reference population. Although higher rates of adverse events were observed, low stent thrombosis rates indicate good performance in this high-risk patient population. Additional research on this topic preferably in randomized controlled trials is required to validate these findings.

Clinical perspectives
CKD is a common comorbidity in everyday practice and patients with CAD. Moreover, these patients present with more comorbidities and complex lesions when compared with the reference population. Therefore, a high level of expertise is necessary to assure sufficient treatment. Randomized controlled trials are needed to validate these findings.