The NCVD-PCI registry is an on-going collaboration between the Ministry of Health Malaysia and the National Heart Association of Malaysia. The data of patients underwent PCI from 2007 to 2014 in 15 participating hospitals (13 public and 2 private) across Malaysia was captured using standardized case report forms. The list of participating hospitals can be found in the Annual Report of the NCVD-PCI registry year 2013–2014 [14]. A unique national identification number was assigned to each patient to avoid duplication and maintain anonymity. Patient’s baseline characteristics, risk factor profile, extent of coronary artery disease, revascularization methods and estimated glomerular filtration rate (eGFR) were recorded. Follow-up was done at 1 year after hospital discharge via phone call or when the patient came to the clinic for review. Verified data will be entered using an established electronic data acquisition form with built-in plausibility checks [14].

The patients were categorized into two groups; CKD and non-CKD. CKD is defined as GFR of < 60 mls/min/1.73m² as determined by Modification of Diet in Renal Disease (MDRD) formula [15‐17]. In this registry, CKD and ESRD were combined as a single group. Therefore, we are unable to perform separate analysis for both conditions. STEMI was defined as persistent ST segment elevation ≥1 mm in two contiguous electrocardiographic leads, or the presence of a new left bundle branch block in the setting of positive cardiac biomarkers.

Data from this registry depends heavily on patients self-reporting for baseline characteristics and co-morbidities (self-recall, previous hospital’s discharge letter or list of medications). Apart from that, the information was cross-checked with patient’s medical records, laboratory results and pre-procedural notes. Malaysia is a unique multi-ethnic country with diversified races and religions. The 3 major ethnic groups are Malay, Chinese and Indian. The rest of the study population (approximately 5%) is categorized into others; these include other native people, non-native Malaysian, foreigner and unknown status. Single vessel disease is defined as lesions > 50% stenosis in only 1 major epicardial vessel, whereas multi-vessels disease is defined as lesions > 50% in 2 or more epicardial vessels. Lesion type is divided according to the American Heart Association / American College of Cardiology (AHA/ACC) classification [18]. Since this registry only enrolled patients who underwent PCI, data on thrombolytic therapy for STEMI was not captured. However, patient who underwent rescue PCI may represent most of the patients who might have received thrombolytic therapy as first-line treatment. The procedural complications were defined as per the NCVD data definition form document [14]. For the pharmacological treatment, only information from the hospital’s discharge document is recorded.

The study populations were STEMI patients stratified by presence or absence of CKD. Continuous variables were described as mean (SD) if normally distributed and compared using the Student’s t-test or as median (interquartile range) if skewed and compared using the Mann-Whitney U test. Categorical variables were described as numbers (percentages). Comparisons of categorical data were analyzed using the chi-square test or Fisher’s exact test. To evaluate the association between CKD and mortality within 1 year, their respective multivariable-adjusted hazard ratios (HR) were calculated using Cox proportional-hazards regression models. Variables included in the model were chosen by separate univariate analyses; those with p-value of < 0.05 were included in the final model. The variables were entered stepwise into the model using the forward likelihood ratio method with p-in: 0.05 and p-out: 0.10. Multicollinearity between the included variables was examined using standard error of b coefficient. All tests were two sided and a p-value of less than 0.05 was considered to be statistically significant. The assumption of proportional hazards for each covariate was reviewed separately by the means of log-minus-log survival plots. Hazard ratios were reported together with the 95% confidence interval (CI) values. All statistical analyses were performed using SPSS version 23. Missing data were assumed to be missing completely at random (MCAR) based on the Little’s MCAR test p-value of more than 0.05. These missing data were omitted automatically from the analysis by list-wise deletion.

The NCVD-PCI is registered in the National Medical Research Register of Malaysia (ID: NMRR-07-20-250) and received ethical approval from the Ministry of Health Medical Research and Ethics committee. Consent from individual patients was not necessary as the data collected were anonymized. Each patient will receive a unique identification number recorded in the registry.

A total of 6563 patients (almost 80% of total number of STEMI patients treated with PCI during study period) were included in the final analysis, 5765 (87.8%) men and 798 (12.2%) women. Patients with CKD were numerically older than their counterpart without CKD. Ethnic distribution depicted general Malaysian population with Malay being the most dominant ethnic group. In terms of co-morbidities, patients with CKD tend to have more conventional cardiovascular risk factors except for cigarette smoke exposure. Baseline characteristics of study population further elaborated in Table 1.

STEMI patients with CKD have more extensive coronary artery disease with higher rate of multi-vessels disease and more complex coronary lesions (type B2 / C and left main-stem involvement). The use of drug eluting stents (DES) which is regarded as the benchmark in PCI is lower in CKD (Table 2).

There was notably higher percentage of CKD patients with STEMI present in severe acute left ventricular dysfunction (Killip class 3 and 4; 24.9% in CKD vs 8.7% in non-CKD). Higher percentages of rescue PCI in both arms suggested that thrombolytic therapy was the default mode of treatment of STEMI in this population. Primary PCI, which is the preferred revascularization strategy was only performed in 20.7% among CKD patients and 15.5% among the non-CKD patients (Table 3).

Table 4 showed the medications prescribed at discharge for this study cohort. Usage of anti-platelets therapy was almost similar between the 2 groups, although the percentage of patients with CKD who were prescribed aspirin was lower. Combination of aspirin and clopidogrel remained the most commonly prescribed dual anti-platelet regime. It is obvious that the use of more potent newer generation of anti-platelets such as ticagrelor and prasugrel was low, below 10% of the population. Although CKD patients had more co-morbidity, they were given less of evidence-based medications. Except beta-blocker, the use of statin, angiotensin converting enzyme inhibitor (ACE-I) and angiotensin receptor blocker (ARB) was consistently lower in CKD patients in comparison to their non-CKD counterparts.

Patients with CKD are more likely to develop procedural related complications (Table 5). Vascular complications include bleeding, access site occlusion, loss of distal pulse, dissection and pseudoaneurysm. Major adverse cardiovascular events (MACE) included periprocedural MI, emergency PCI, bailout CABG, cardiogenic shock, arrhythmia, transient ischemic attack/stroke, cardiac tamponade and heart failure. Death was analyzed as a separate outcome.

Table 6 shows the odd ratios of developing in-hospital vascular complications, MACE and death in patients with CKD. After adjustment of the covariates, patients with CKD were more likely to develops in-hospital death (OR: 4.5, 95% CI 3.11–6.65) and in-hospital MACE (OR: 3.42, 95% CI 2.39–4.90) compared to non-CKD patients. Figure 1 showed the Kaplan Meier survival curves between CKD and non-CKD patients up to 1 year of follow up, with the CKD group had significantly lower cumulative survival after PCI compared to the non-CKD group. The early phase mortality contributes the most in the difference. Further analysis using multivariate Cox proportional hazard regression was done to adjust for the significant covariates, and it can be seen that patients with CKD had significantly higher risk of 1 year mortality (HR: 3.79, 95% CI 2.84–5.07) compared to the non-CKD group (Table 7).