Frontiers in cardiovascular medicine · 2026
Performance of predictive scores for in-hospital mortality in surgical infective endocarditis: a retrospective cohort study.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Pavia-Velandia Paula Ximena, Torres-Gómez Paula Liliana, Molano-González Nicolás, Rodríguez-Villegas Adelaida, Castillo-Valderrama Tomás, Guaque-Camargo Fabián Andrés, Hidalgo-Suárez Sara Mariana, Morales-Pertus Carlos, Rodríguez-Lima David René
Аннотация
BACKGROUND: Infective endocarditis (IE) carries high surgical mortality, particularly in Latin American populations where validated prediction models are scarce. Although EuroSCORE II, RISK-E, and APORTEI have shown acceptable performance in European cohorts, external validation in middle-income country settings remains limited.
METHODS: We conducted a retrospective cohort study of adult patients who underwent cardiac surgery for IE at a high-complexity tertiary-care center in Bogotá, Colombia, between 2014 and 2025. The primary outcome was all-cause in-hospital mortality. Discrimination was assessed using Harrell's C-statistic and calibration was assessed via intercept, slope, Spiegelhalter's z-test, Brier score, mean absolute error (Eavg), and maximum absolute error (Emax).
RESULTS: Seventy patients were included (mean age 51.9 ± 14.8 years; 70% male); in-hospital mortality was 30% (21/70). RISK-E demonstrated the highest discrimination (C = 0.741, 95% CI 0.61-0.85), followed by EuroSCORE II (C = 0.718, 95% CI 0.58-0.83) and APORTEI (C = 0.676, 95% CI 0.53-0.80). All three scores systematically underestimated mortality risk, as evidenced by positive calibration intercepts (range 1.14-1.75). Calibration was closest to ideal for RISK-E (slope=0.962), while APORTEI showed the greatest deviation (slope=1.425). Spiegelhalter's z-test indicated significant miscalibration for all three models (all p < 0.001).
CONCLUSIONS: RISK-E showed numerically superior discrimination and a calibration slope closest to unity, although pairwise DeLong's tests showed no statistically significant difference between any two scores, and these findings should be interpreted in the context of the limited sample size (70 patients, 21 events). All three scores systematically underestimated in-hospital mortality, suggesting that recalibration or development of locally derived models is needed for reliable risk stratification in Latin American IE surgical programs.
METHODS: We conducted a retrospective cohort study of adult patients who underwent cardiac surgery for IE at a high-complexity tertiary-care center in Bogotá, Colombia, between 2014 and 2025. The primary outcome was all-cause in-hospital mortality. Discrimination was assessed using Harrell's C-statistic and calibration was assessed via intercept, slope, Spiegelhalter's z-test, Brier score, mean absolute error (Eavg), and maximum absolute error (Emax).
RESULTS: Seventy patients were included (mean age 51.9 ± 14.8 years; 70% male); in-hospital mortality was 30% (21/70). RISK-E demonstrated the highest discrimination (C = 0.741, 95% CI 0.61-0.85), followed by EuroSCORE II (C = 0.718, 95% CI 0.58-0.83) and APORTEI (C = 0.676, 95% CI 0.53-0.80). All three scores systematically underestimated mortality risk, as evidenced by positive calibration intercepts (range 1.14-1.75). Calibration was closest to ideal for RISK-E (slope=0.962), while APORTEI showed the greatest deviation (slope=1.425). Spiegelhalter's z-test indicated significant miscalibration for all three models (all p < 0.001).
CONCLUSIONS: RISK-E showed numerically superior discrimination and a calibration slope closest to unity, although pairwise DeLong's tests showed no statistically significant difference between any two scores, and these findings should be interpreted in the context of the limited sample size (70 patients, 21 events). All three scores systematically underestimated in-hospital mortality, suggesting that recalibration or development of locally derived models is needed for reliable risk stratification in Latin American IE surgical programs.