Predictive validity of the simplified Radiographic Assessment of Lung Edema score for mortality in critically ill patients with COVID-19.

内容摘要

Early and precise identification of COVID-19 patients at high risk of mortality is vital for guiding treatment decisions. This study evaluated the predictive validity of the simplified Radiographic Assessment of Lung Edema (RALE) score, derived from chest X-rays (CXRs), for hospital mortality among critically ill patients with Delta-variant COVID-19. We conducted a retrospective observational study of 105 critically ill adults with confirmed Delta-variant COVID-19 at an intensive care center in Ho Chi Minh City, Vietnam, between 30 July and 15 October 2021. We calculated the simplified RALE score (range 0-8) from admission frontal CXRs. We evaluated predictive performance using the area under the receiver operating characteristic curve (AUROC). We used multivariable logistic regression to assess associations between the simplified RALE score and hospital mortality, adjusting for baseline confounders (inter-hospital transfers, demographics, comorbidities, clinical and laboratory parameters, and gas exchange data) and mediators (respiratory support, adjunctive therapies, and in-hospital complications). The cohort averaged 60.5 years of age (40% male), and 79% died in the hospital. Bilateral lung opacities appeared in 99% of patients, and the median simplified RALE score reached 8.0 (IQR 6.0-8.0). The simplified RALE score discriminated mortality well (AUROC 0.747, 95% confidence interval [CI] 0.617-0.877), closely matching SOFA (0.747) and CURB-65 (0.776) scores. In baseline-adjusted models, both the simplified RALE (adjusted odds ratio [AOR] 2.170, 95% CI 1.353-3.482) and CURB-65 scores (AOR 4.316, 95% CI 1.684-11.059) independently predicted hospital mortality. Even after further adjusting for mediators, the simplified RALE score (AOR 1.934, 95% CI 1.220-3.066) remained an independent predictor. The simplified RALE score performed comparably to SOFA and CURB-65 scores in predicting hospital mortality and remained an independent predictor across all multivariable analyses. These findings support the use of the simplified RALE score as a practical tool for risk stratification in critically ill patients with Delta variant COVID-19, particularly in high-burden and resource-limited settings. Further studies in larger and more diverse populations are warranted to confirm these findings and facilitate broader clinical implementation. This study was previously presented in abstract form at: LIVES 2024, 37th Annual Congress of the European Society of Intensive Care Medicine; October 7-9, 2024; Barcelona, Spain. Do SN, Luong CQ, Nguyen MH, Pham DT, Khuat NH, Pham QT, et al. Predictive validity of the simplified Radiographic Assessment of Lung Edema score for mortality in critically ill COVID-19 patients with the B.1.617.2 (Delta) variant in Vietnam: a retrospective observational study. Intensive Care Med Exp. (2024) 12(Suppl 1):87. doi: 10.1186/s40635-024-00658-z.