CLINICAL OUTCOMES OF EARLY REPATRIATION FOR PATIENTS WITH ST-SEGMENT ELEVATION MYOCARDIAL INFARCTION: A PROPENSITY-MATCHED ANALYSIS

Abstract

Given limitations on hospitals resources, patients with ST elevation myocardial infarction (STEMI) who undergo successful primary percutaneous coronary intervention (PCI) are often repatriated to non-PCI centers. However, the safety of this practice is not clear, given the transfer of care. Our objective was to evaluate the safety of early repatriation of stable STEMI patients post primary PCI, compared to ongoing treatment at the PCI center. Consecutive STEMI patients who received primary PCI at four PCI hospitals in Toronto, Canada between 2010 and 2012 were identified. Patients in cardiogenic shock or who died within 24 hours of presentation were excluded. In order to account for potential confounders, propensity score matching of repatriated vs. non-repatriated patients was performed. The primary outcomes were 1-year all-cause mortality and re-admission for recurrent myocardial infarction (MI). Secondary outcomes were 30-day all-cause mortality, 30-day re-admission for MI, mean first medical contact to balloon time (minutes) and mean length of stay (days). The overall cohort consisted of 1445 patients, of whom 792 (55%) were repatriated. Patients were repatriated within 1.5 days on average (median 1 day, 95% confidence interval (CI) 1.37-1.63 days). Using the propensity score, 427 well-matched pairs were identified. There was no significant difference between repatriated and non-repatriated groups in 1-year mortality (repatriated 6.6%; non-repatriated 5.6%; Hazard ratio (HR) 1.14, 95% CI 0.66-1.97, p= 0.629). The 1-year re-admission rates for MI were significantly greater for the repatriated group (repatriated: 11.5 %, non-repatriated: 5.9%; HR 1.97, 95% CI 1.22-3.19, p= 0.006). These findings were consistent in the 30-day outcomes (30-day mortality: 2.6% and 4.9% for repatriated and non-repatriated, p-value=0.072; 30-day re-admission for MI 5.6% and 1.9% for repatriated and non-repatriated, p-value =0.004). The mean contact to balloon time was comparable in both groups (repatriated 138 minutes, 95% CI 121-156; non-repatriated 150 minutes, 95% CI 129-172, p= 0.405). Similarly, the mean length of total hospital stay was comparable in both groups (repatriated 6 days, 95% CI 5-6; non-repatriated 5.5 days, 95% CI 4-7, p= 0.947). A strategy of early repatriation of STEMI patients was associated with an increase in re-admission for MI. Our study revisits the safety associated with an early repatriation strategy. Further research is needed to address challenges involved with transfer of care between medical institutions in this population.