RESUMO
BACKGROUND: Despite emerging evidence suggesting that selected patients presenting with ST-segment elevation myocardial infarction (STEMI) treated successfully with primary percutaneous coronary intervention (PPCI) may be considered for early discharge, STEMI patients are typically hospitalised longer to monitor for serious complications. METHODS: We assessed the feasibility of identifying low-risk STEMI patients in our institution for early discharge using the Zwolle risk score (ZRS). We evaluated consecutive STEMI patients who underwent successful PPCI within the period 1 January 2016 to 31 December 2017. Low-risk was defined as ZRS≤3. Demographic, angiographic characteristics, length of stay (LOS), and 30-day major adverse cardiovascular events (MACE) defined as cardiac death, stroke, congestive cardiac failure, and non-fatal myocardial infarction, were recorded. RESULTS: There were 183 STEMI patients in our study cohort (mean age 62.0±12.2 years, 77.0% male). The median ZRS was 2 (interquartile range 1-4) with 132 (72.1%) patients classified as low-risk. The overall 30-day MACE and mortality rates were 10.4% and 3.3% respectively. None of the 35 (26.5%) low-risk patients who were discharged within 72 hours experienced MACE at 30 days. Low-risk STEMI patients had significantly shorter median LOS (86.3 vs. 93.2 hours, p=0.002), lower 30-day MACE (4.5% vs. 25.5%, p<0.0001) and mortality (0% vs. 11.8%, p<0.0001) compared to high-risk group (ZRS>3). Receiver operating characteristic (ROC) curve analyses for ZRS in predicting 30-day MACE and mortality yielded C-statistics of 0.79 (95%CI 0.68-0.90, p<0.0001) and 0.98 (95%CI 0.95-1.00, p<0.0001) respectively. CONCLUSION: Low-risk STEMI patients stratified by Zwolle risk score, who were treated successfully with PPCI, experienced low 30-day MACE and mortality rates, indicating that early discharge may be safe in these patients. Larger studies are warranted to evaluate the safety of ZRS-guided early discharge of STEMI patients, as well as the economic and psychological impacts.