Methods

Study population

The present study is a cohort data analysis of patients who were enrolled prospectively in the ACSIS from 2000 to 2016. Details of ACSIS have been previously reported (6). In brief, this survey is conducted biennially over a 2-month period among ACS patients admitted to coronary care units and cardiology wards in all 27 public hospitals in Israel.
Patient management was at the discretion of the attending physicians. The diagnosis of AMI and ST-segment elevation MI (STEMI) or non- ST-segment elevation MI (NSTEMI), was made by the attending cardiologist according to pre-specified survey criteria based on accepted guidelines. In total 15,200 patients were enrolled in our study. The patients were grouped according to the occurrence of ventricular arrhythmias during the course of index AMI. Data included baseline demographics, medical history, hospital course, procedures, complications, and outcome based on hospital’s charts. Additional data regarding health status, adverse events and medication adherence were collected from the patients via follow-up calls and from outpatients’ medical records one month after hospital discharge. Mortality rates were determined for all patients from hospital and out-patient charts and by matching their identification numbers with the Israeli National Population Register. In- hospital, one year, and five-year all-cause mortality was assessed for the absence of VTA, early VTA and late VTA according to the periods of presentation.

Definitions and endpoints

The diagnosis of VTA, sustained ventricular tachycardia or ventricular fibrillation was based on electrocardiogram (ECG) or continuous monitor strips. Sustained ventricular tachycardia was defined as a series of abnormally shaped QRS complexes longer than 120 milliseconds, with a rate 100-250 beats/minute, lasting longer than 30 seconds or requiring electrical cardioversion due to hemodynamic collapse. Ventricular fibrillation was defined as rapid, very irregular rhythm with indiscernible P waves or QRS. Patients were classified into three groups: no VTA; early VTA (≤48h of onset) and late VTA (>48h of onset). Survey periods were divided to early decade (2000-2006) vs. late decade (2008-2016). Primary endpoints were 30-day, 1-year and 5-year mortality rate. Secondary endpoints included 30-day major adverse cardiac events (MACE) (all-cause mortality rate, re-infarction, stent thrombosis and urgent revascularization). Left ventricular ejection fraction (LVEF) was graded according to echocardiography performed at admission in each center.

Statistical methods

Demographic, clinical features and the use of in-hospital therapies were reported for all study participants. Multiple logistic regression adjustments for age, gender, diabetes mellitus, chronic renal failure and heart failure and for survey periods was performed to assess the adjusted odds ratio (OR) and 95% confidence interval (CI) of in and out of hospital complications and 30-day, 1 and 5-year mortality rate associated with VTA at any time during and after the admission for ACS. The cumulative probability of all-cause mortality rate was graphically displayed using the Kaplan–Meier method and compared using the log-rank test. In order to evaluate the independent association of VTA with the 5-year all-cause mortality rate outcome, a Cox proportional hazards model was constructed adjusting for survey period. We employed an interaction-term analysis to examine whether the risk of 5-year mortality rate associated with VTA was different in the early period as compared with the late period. Statistical significance was accepted for a 2-sided p<0.05. Statistical analysis was performed using the R Statistical Package (R Foundation for Statistical Computing, Vienna, Austria).