Comparison of the FASTEST and the ZWOLLE risk scores for identification of very low-risk patients for all-cause mortality and MACE following primary PCI. (25th November 2020)

• Record Type: Journal Article
• Authors: Zaharijev, S; Mehmedbegovic, Z; Milasinovic, D; Jelic, D; Zobenica, V; Radomirovic, M; Vratonjic, J; Pavlovic, A; Djurosev, I; Vukcevic, V; Asanin, M; Stankovic, G
• Is Part Of: European heart journal. Volume 41:(2020)Supplement 2
• Publication Date: 2020-11-25
• Subjects: ST-Elevation Myocardial Infarction (STEMI); Cardiology -- Periodicals; Heart -- Diseases -- Periodicals; 616.12005
• DOI: 10.1093/ehjci/ehaa946.1772
• Languages: English
• ISSNs: 0195-668X

Abstract

Abstract: Background: Prior studies suggest that low-risk ST-segment-elevation myocardial infarction (STEMI) patients undergoing primary percutaneous coronary intervention (pPCI) can be considered for early discharge in order to reduce healthcare costs and improve resource utilization. Novel, simple, the FASTEST score, demonstrated additional prognostic value over guideline recommended ZWOLLE score in a derivation cohort, but robust data about external validation are lacking. Purpose: We aimed to compare overall predictive ability and discriminating power in identification of low-risk patients of novel FASTEST score compared to validated ZWOLLE score. Methods: From a high-volume, single-center, prospective registry, in a period from 2009–2019, we included STEMI patients who underwent successful pPCI in whom both, FASTEST (1 point added for: femoral access, age>65, LVEF <50, TIMI <3, creatinine >1.5 mg/dl; left main disease; and Killip≥2) and ZWOLLE (age, anterior infarct, Killip class, TIMI flow, ischemia time, 3 vessel disease) scores were both calculated. Predictive ability of scores for in-hospital, 30 days and 1 year mortality and hospital MACE was tested using ROC analysis and comparing AUC. Also, event rate was compared between low-risk patients as classified by FASTEST (score=0) or ZWOLLE (score≤3). Results: We included 5650 patients (age 60.8±11.4, male (71%), anterior STEMI (44%) and femoral approach (81%)). Overall, mortality rates were 2.1%, 3.1% and 8.1% for hospital, 30 days and one-year. As Low-risk subjects, ZWOLLE identified broader proportion of population compared to FASTEST (67% vs. 5.5%) mainly due to high prevalence of femoral approach (FASTEST low-risk 30% in radial approach subset), still, later had numerically lower mortality rates at hospital (0.7% vs. 0.3% (only 1 pt); p=0.62), 30 days (1.3% vs. 0.7%; p=0.39) and at one-year (4% vs. 2%; p=0.14). Both scores showed similar and very good predictive ability for in-hospital (AUC 0.81 vs. 0.81; p=0.66) and 30 days mortality (AUC 0.79 vs. 0.77; p=0.29), while at one-year, discrimination of crude mortality by FASTEST trended, but didn't reach statistical significance compared to ZWOLLE score, respectively (AUC 0.77 vs. 0.75; p=0.07). FASTEST showed better prediction for composite endpoint of in-hospital MACE - death, stroke, reinfarction and bleeding BARC class 3 or higher (AUC 0.71 vs. 0.67; p<0.000) (Figure 1). Conclusion: Both the FASTEST and the ZWOLLE scores showed very good discriminating power for in-hospital, 30 day mortality and one-year mortality, yet the FASTEST score offered comparative advantage for prediction of in-hospital MACE and could be used to identify selected patients where an early hospital discharge can be considered. Funding Acknowledgement: Type of funding source: None