Creatinine variation improves Zwolle score in selecting low-risk patients after ST-segment elevation myocardial infarction.

BACKGROUND The Zwolle score is recommended to identify ST-segment elevation myocardial infarction (STEMI) patients with low-risk eligible for early discharge. Our aim was to ascertain if creatinine variation (Δ-sCr) would improve Zwolle score in the decision-making of early discharge after primary percutaneous coronary intervention (PCI). METHODS AND RESULTS A total of 3296 patients with STEMI that underwent primary PCI were gathered from the Portuguese Registry on Acute Coronary Syndromes. A Modified-Zwolle score, including Δ-sCr, was created and compared with the original Zwolle score. Δ-sCr was also compared between low (Zwolle score ≤3) and non-low-risk patients (Zwolle score >3). The primary endpoint is 30-day mortality and the secondary endpoints are in-hospital mortality and complications. Thirty-day mortality was 1.5% in low-risk patients (35 patients) and 9.2% in non-low-risk patients (92 patients). The Modified-Zwolle score had a better performance than the original Zwolle score in all endpoints: 30-day mortality (area under curve 0.853 versus 0.810, P < 0.001), in-hospital mortality (0.889 versus 0.845, P < 0.001) and complications (0.728 versus 0.719, P = 0.037). Reclassification of patients lead to a net reclassification improvement of 6.8%. Additionally, both original Zwolle score low-risk patients and non-low-risk patients who had a Δ-sCr ≥0.3 mg/dl had higher 30-day mortality (low-risk: 1% versus 6.6%, P < 0.001; non-low-risk 4.4% versus 20.7%, P < 0.001), in-hospital mortality and complications. CONCLUSION Δ-sCr enhanced the performance of Zwolle score and was associated with higher 30-day mortality, in-hospital mortality and complications in low and non-low-risk patients. This data may assist the selection of low-risk patients who will safely benefit from early discharge after STEMI.