Chronic kidney disease (CKD) increases the risk of death and other poor outcomes in patients with cardiovascular diseases. This study investigated the relation between the institutional CKD percutaneous coronary intervention (PCI) volume and in-hospital clinical outcomes in patients with CKD. Among 1,199,901 patients who underwent PCI in 2014 to 2018 from the Japanese nationwide registry, we analyzed 220,509 patients with CKD. Patients were classified into quartiles (Q) according to the mean annual institutional CKD-PCI volume (Q1 <42 PCIs/year, Q2 <74 PCIs/year, Q3 <124 PCIs/year, Q4 ≥125 PCIs/year). The primary outcome was a composite of in-hospital death and periprocedural complications. The mean age of patients was 73 ± 10 years, and 36% (n = 78,332) were on dialysis. PCI was more likely to be performed with rotational atherectomy devices in high-volume institutions. Contrast volume was lower, the rate of radial access PCI was higher, and door-to-balloon time (for ST-elevation myocardial infarction) was shorter in the highest quartile institutions. Primary outcomes were observed in 6,539 patients (3.0%). The crude rate of the primary outcome was lowest in institutions with the highest PCI volume (Q1 3.4%, Q2 3.0%, Q3 3.0%, Q4 2.4%, p <0.001); higher PCI volume was associated with reduced frequency of the primary outcome (odds ratio [95% confidence interval] relative to Q1:Q2, 0.89 [0.83 to 0.96]; Q3 0.90 [0.84 to 0.97]; and Q4 0.76 [0.84 to 0.97]). In conclusion, the procedural characteristics and outcomes of PCI differed significantly by institutional volume in patients with CKD. When considering revascularization among these patients, institutional CKD-PCI volume needs to be incorporated in decision-making.
Cardiovascular disease is the leading cause of death in patients with chronic kidney disease (CKD). Coronary revascularization is frequently performed in patients with CKD. Although this could imply that patients with CKD have much to gain from coronary revascularization, strong evidence of this is lacking. Indeed, a recent large-scale randomized control trial demonstrated that in patients with stable ischemic heart disease (SIHD) with CKD, an initial invasive strategy, compared with an initial conservative approach, did not reduce the risk of death or nonfatal myocardial infarction (MI) over a median follow-up of 2.2 years. Therefore, the risk and benefit of revascularization procedures must be carefully weighed in the decision-making regarding the management of patients with CKD. Previous studies have shown that institutional percutaneous coronary intervention (PCI) volume plays a crucial role in the clinical outcomes of patients who underwent PCI. , A better understanding of the volume-outcome relation of PCI may aid in improving the clinical outcomes of patients with CKD. This study aimed to investigate the relation between institutional CKD-PCI volume and in-hospital outcomes in patients with CKD using a nationwide PCI registry in Japan (J-PCI registry).
Methods
The J-PCI registry is a prospective and multicenter Japanese nationwide PCI registry which is operated by the Japanese Association of Cardiovascular Intervention and Therapeutics, designed to collect data on the clinical characteristics and in-hospital outcomes of patients who underwent PCI. , Since 2013, the J-PCI registry has been incorporated into the National Clinical Data System, a nationwide prospective web-based registration system. Each hospital has a data manager responsible for collecting PCI data and submitting it to a dedicated on-line database. The Cardiovascular Intervention and Therapeutics has an annual meeting of data managers to ensure adequate data collection and performs random site visits to validate the submitted data. The protocol of the J-PCI registry was approved by the Institutional Review Board Committee at the Network for Promotion of Clinical Studies in a specified nonprofit organization affiliated with the Osaka University Graduate School of Medicine. The need for written informed consent was waived because of the retrospective study design.
The present study included 1,199,901 patients in the J-PCI registry between January 2014 and December 2018. Patients who were <20 or >100 years of age and those with missing data for gender, clinical presentation, CKD, and dialysis status were excluded (n = 15,948). Among patients with complete data (n = 1,183,953), 220,509 patients with CKD (18.6%) who underwent PCI in 1,133 Japanese institutions were analyzed. Institutional CKD-PCI volume was divided into the first quartile (Q1 <42 PCIs/year, 757 institutes), second quartile (Q2 <74 PCIs/year, 197 institutes), third quartile (Q3 <124 PCIs/year, 120 institutes), and fourth quartile (Q4 ≥125 PCIs/year, 59 institutes) institutions from low- to high-volume institutions according to the annual number of PCI procedures during the study period ( Figure 1 ). The annual number of institutional CKD-PCI correlated roughly linearly with that of total PCI (Supplementary Figure 1).
The definitions of J-PCI variables were summarized in a previous report. In this registry, CKD was defined as the presence of proteinuria, serum creatinine ≥1.3 mg/dl, or an estimated glomerular filtration rate ≤60 ml/min/1.73 m 2 by referring to the guideline of the Japanese Society of Nephrology. In this database, it was not possible to perform an analysis using the definition of CKD based on Kidney Disease: Improving Global Outcomes. ST-segment elevation MI (STEMI) was defined as acute MI with electrocardiographic findings of ST-elevation or new-onset complete left bundle branch block or pure posterior MI. Non–ST-segment elevation acute coronary syndrome (NSTE-ACS) was defined as non-STEMI and unstable angina. Unstable angina was defined as acute coronary syndrome (ACS) without elevated cardiac biomarkers, including new-onset or increased-severity angina pectoris within 1 month of previous symptoms, angina pectoris at rest, and post-infarction angina pectoris. SIHD included stable angina, old MI, and silent ischemia. Door-to-balloon time was defined as the time from the patient’s arrival at the hospital to the first balloon inflation during PCI.
The primary outcomes were in-hospital complications, including in-hospital death within 30 days after PCI, cardiac tamponade, cardiogenic shock requiring mechanical and/or inotropic support, stent thrombosis, emergent cardiac surgery, and major bleeding complications. Patients presenting with cardiogenic shock were included in the present analysis. However, as the primary end point, cardiogenic shock that developed during or after PCI was included. Stent thrombosis was defined as definite stent thrombosis according to the Academic Research Consortium definition. Major bleeding complications were defined as periprocedural and postprocedural bleeding requiring blood transfusion from access and nonaccess sites.
Categoric variables were presented as numbers with relative percentages and compared using the chi-square test. Continuous variables are presented as mean ± SD and compared using analysis of variance. We performed multivariable logistic regression analysis to adjust for patient characteristics. We also constructed multivariable models in which we used penalized thin plate regression splines to visualize the association between institutional CKD-PCI volume and in-hospital outcomes. Variables entered into the multivariable analysis for statistical adjustment included age, male gender, hypertension, diabetes, hyperlipidemia, smoking, dialysis, chronic obstructive pulmonary disease, peripheral artery disease, history of MI, history of heart failure (HF), history of PCI, history of cardiac bypass grafting surgery, clinical presentation, cardiac arrest within 24 hours, acute HF within 24 hours, angina symptoms within 1 month, radial access PCI, number of diseased vessels, left main trunk (LMT) disease, and institutional CKD-PCI volume. We additionally evaluated the adjusted odds ratios of institutional CKD-PCI volume quartiles for the primary outcome in the subgroups according to dialysis status (dialysis and nondialysis CKD) and clinical presentation (SIHD, NSTE-ACS, and STEMI). Analyses were performed using the same variables as previously mentioned, except for dialysis and clinical presentation. Differences were considered significant for 2-sided p <0.05. All data were analyzed using R statistical software version 4.0.2 (Free Software Foundation, Inc., Boston, Massachusetts).
Results
Overall, 220,509 patients with CKD who underwent PCI performed by 6,934 operators in 1,133 institutions in Japan were analyzed in this study. Baseline patient characteristics are shown in Table 1 . Patients who were on dialysis were less frequent in the highest volume group, and hyperlipidemia and LMT disease were more frequent in the higher volume groups. In terms of procedural characteristics, PCI was more likely to be performed with rotational atherectomy devices in the higher volume groups. Contrast volume was lower, the rate of radial access PCI was higher, and door-to-balloon time (for patients with STEMI) was shorter in the highest quartile institutions. Table 2 shows the in-hospital clinical outcomes in each PCI volume group. The crude rate of the primary outcome was the lowest for the highest volume group. The crude rate of in-hospital death was higher in the lowest volume group than in the other groups. Access site bleeding was lowest in the highest volume group. The rates of cardiogenic shock requiring mechanical and/or inotropic support and stent thrombosis decreased as the PCI volume increased. We also confirmed that penalized thin plate spline regression analysis supported an inverse relation between these clinical outcomes and the annual number of CKD-PCI ( Figure 2 ).
| Variable | Overall | Quartile | P value | |||
|---|---|---|---|---|---|---|
| (n = 220,509) | 1 (n = 54,762) | 2 (n = 54,815) | 3 (n = 55,219) | 4 (n = 55,713) | ||
| Age (years) | 73 ± 10 | 73 ± 11 | 72 ± 10 | 72 ± 10 | 73 ± 10 | <0.001 |
| Male | 170,557 (77%) | 42,651 (78%) | 42,584 (78%) | 42,979 (78%) | 42,343 (76%) | <0.001 |
| Hypertension | 179,892 (82%) | 43,518 (79%) | 44,464 (81%) | 45,656 (83%) | 46,254 (83%) | <0.001 |
| Diabetes mellitus | 126,074 (57%) | 31,556 (58%) | 32,018 (58%) | 31,977 (58%) | 30,523 (55%) | <0.001 |
| Hyperlipidemia | 133,011 (60%) | 30,411 (56%) | 31,962 (58%) | 33,538 (61%) | 37,100 (67%) | <0.001 |
| Smoker | 58,949 (27%) | 14,165 (26%) | 15,602 (28%) | 14,150 (26%) | 15,032 (27%) | <0.001 |
| Dialysis | 78,332 (36%) | 20,728 (38%) | 21,316 (39%) | 20,893 (38%) | 15,395 (28%) | <0.001 |
| COPD | 6,824 (3.1%) | 1,648 (3.0%) | 1,639 (3.0%) | 1,797 (3.3%) | 1,740 (3.1%) | 0.04 |
| PAD | 35,994 (16%) | 7,571 (14%) | 8,636 (16%) | 10,237 (19%) | 9,550 (17%) | <0.001 |
| Prior MI | 59,517 (27%) | 13,807 (25%) | 14,710 (27%) | 15,823 (29%) | 15,177 (28%) | <0.001 |
| Prior HF | 59,652 (28%) | 15,166 (28%) | 14,620 (27%) | 15,454 (28%) | 14,412 (27%) | <0.001 |
| Prior PCI | 119,776 (55%) | 27,960 (51%) | 29,875 (55%) | 30,860 (56%) | 31,081 (57%) | <0.001 |
| Prior CABG | 15,547 (7.1%) | 3,307 (6.1%) | 4,104 (7.5%) | 4,243 (7.7%) | 3,893 (7.1%) | <0.001 |
| Clinical presentation | <0.001 | |||||
| SIHD | 147,387 (67%) | 34,264 (63%) | 36,410 (66%) | 37,544 (68%) | 39,169 (70%) | |
| NSTE-ACS | 44,805 (20%) | 12,828 (23%) | 11,506 (21%) | 10,682 (19%) | 9,789 (18%) | |
| STEMI | 28,317 (13%) | 7,670 (14%) | 6,899 (13%) | 6,993 (13%) | 6,755 (12%) | |
| Cardiac arrest within 24 h | 4,585 (2.1%) | 1,025 (1.9%) | 1,206 (2.2%) | 1,249 (2.3%) | 1,105 (2.0%) | <0.001 |
| Cardiogenic shock within 24 h | 9,675 (4.4%) | 2,469 (4.5%) | 2,517 (4.6%) | 2,586 (4.7%) | 2,103 (3.9%) | <0.001 |
| Acute heart failure within 24 h | 13,351 (6.1%) | 3,730 (6.8%) | 3,337 (6.1%) | 3,244 (5.9%) | 3,040 (5.6%) | <0.001 |
| Angina symptom within 1 month | 152,834 (69%) | 38,665 (71%) | 36,955 (67%) | 37,377 (68%) | 39,837 (72%) | <0.001 |
| No. of diseased vessels | <0.001 | |||||
| 1 | 117,229 (53%) | 29,721 (54%) | 29,329 (54%) | 28,764 (52%) | 29,415 (53%) | |
| 2 | 58,933 (27%) | 14,813 (27%) | 14,626 (27%) | 15,106 (27%) | 14,388 (26%) | |
| 3 | 32,347 (15%) | 7,901 (14%) | 7,922 (14%) | 8,188 (15%) | 8,336 (15%) | |
| LM disease | 12,000 (5.4%) | 2,327 (4.2%) | 2,938 (5.4%) | 3,161 (5.7%) | 3,574 (6.4%) | |
| Access Cite | <0.001 | |||||
| Radial | 99,225 (45%) | 24,930 (46%) | 23,960 (44%) | 22,851 (41%) | 27,484 (49%) | |
| Femoral | 100,549 (46%) | 24,699 (45%) | 26,460 (48%) | 27,011 (49%) | 22,379 (40%) | |
| Others | 20,733 (9.4%) | 5,133 (9.4%) | 4,395 (8.0%) | 5,356 (9.7%) | 5,849 (10%) | |
| DES use | 178,327 (81%) | 44,247 (81%) | 44,282 (81%) | 44,225 (80%) | 45,573 (82%) | <0.001 |
| Drug-coated balloon use | 30,659 (14%) | 6,632 (12%) | 7,614 (14%) | 8,771 (16%) | 7,642 (14%) | <0.001 |
| Rotational atherectomy | 15,760 (7.1%) | 1,545 (2.8%) | 3,639 (6.6%) | 5,046 (9.1%) | 5,530 (9.9%) | <0.001 |
| Contrast volume (ml) | 121 ± 71 | 122 ± 64 | 128 ± 82 | 125 ± 72 | 109 ± 61 | <0.001 |
| Door-to-balloon time in STEMI patients (min) | 91 ± 60 | 95 ± 62 | 90 ± 56 | 96 ± 68 | 83 ± 54 | <0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
