为了降低重症监护病房 (ICU) 住院期间和 ICU 出院后重症 COVID-19 患者的死亡率，测量其发生频率、确定预测因素并制定适当的 ICU 后随访策略至关重要。在这项多中心、前瞻性队列研究中，我们纳入了 586 名重症 COVID-19 患者。我们观察到 ICU 总体死亡率为 20.1% [95%CI：17.1% 至 23.6%] (118/586)，医院总体死亡率为 25.4% [95%CI：22.1% 至 29.1%] (149/586)。对于 ICU 幸存者，ICU 术后 30 天（早期）死亡率为 5.3% [95%CI：3.6% 至 7.8%] (25/468)，ICU 术后一年（晚期）死亡率为 7.9% [95%CI：3.6% 至 7.8%] (25/468) ：5.8％至10.8％]（37/468）。原有疾病/合并症被确定为 ICU 出院后死亡的主要独立预测因素：高血压和心力衰竭是早期死亡的独立预测因素；高血压、慢性肾病、慢性阻塞性肺病和癌症是晚期死亡率的独立预测因素。ICU 后早期和晚期死亡率最初出现激增（ICU 后前 30 天），随后随着时间的推移而下降。密切监测重症 COVID-19 重症监护室后幸存者，尤其是那些有既往病史的幸存者，对于预防不良后果、降低死亡率和制定适当的随访策略至关重要。版权所有：© 2023 Santos 等人。这是一篇根据知识共享署名许可条款分发的开放获取文章，允许在任何媒体上不受限制地使用、分发和复制，前提是注明原始作者和来源。

To mitigate mortality among critically ill COVID-19 patients, both during their Intensive Care Unit (ICU) stay and following ICU discharge, it is crucial to measure its frequency, identify predictors and to establish an appropriate post-ICU follow-up strategy.In this multicentre, prospective cohort study, we included 586 critically ill COVID-19 patients.We observed an overall ICU mortality of 20.1% [95%CI: 17.1% to 23.6%] (118/586) and an overall hospital mortality of 25.4% [95%CI: 22.1% to 29.1%] (149/586). For ICU survivors, 30 days (early) post-ICU mortality was 5.3% [95%CI: 3.6% to 7.8%] (25/468) and one-year (late) post-ICU mortality was 7.9% [95%CI: 5.8% to 10.8%] (37/468). Pre-existing conditions/comorbidities were identified as the main independent predictors of mortality after ICU discharge: hypertension and heart failure were independent predictors of early mortality; and hypertension, chronic kidney disease, chronic obstructive pulmonary disease and cancer were independent predictors of late mortality.Early and late post-ICU mortality exhibited an initial surge (in the first 30 days post-ICU) followed by a subsequent decline over time. Close monitoring of critically ill COVID-19 post-ICU survivors, especially those with pre-existing conditions, is crucial to prevent adverse outcomes, reduce mortality and to establish an appropriate follow-up strategy.Copyright: © 2023 Santos et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.