Chen Haoran, Tang Xinyi, Li Xiaomin, Xie Yongpeng
Kangda College of Nanjing Medical University, Lianyungang, Jiangsu, China.
Department of Emergency and Critical Care Medicine, Lianyungang Clinical College of Nanjing Medical University, Lianyungang, Jiangsu, China.
Shock. 2025 May 1;63(5):688-694. doi: 10.1097/SHK.0000000000002528. Epub 2024 Dec 17.
Background: The relationship between the partial pressure of oxygen in arterial blood (PaO 2 ) and the prognosis of sepsis patients, and its potential variation over time, remains unclear. The optimal PaO 2 range for sepsis patients has always been a contentious issue, with no consensus. We aimed to explore the association between different levels of PaO 2 exposure over time and the 28-day mortality of sepsis patients, and to identify the optimal PaO 2 range for sepsis patients within a specific time frame. Methods: We retrieved data on adult patients diagnosed with sepsis within 24 h before or after intensive care unit (ICU) admission from the Medical Information Mart for Intensive Care IV (MIMIC-IV, version 2.2) database. We excluded patients who were not admitted to the ICU for the first time, those with ICU stay <24 h, and those without PaO 2 results during their ICU stay. We calculated the time-weighted average (TWA) of PaO 2 and used piece-wise exponential additive mixed models (PAMMs) to estimate the time-dependent changes in the association between TWA-PaO 2 and patient prognosis. Results: A total of 16,880 sepsis patients were included in the MIMIC cohort. Results indicated that patients' TWA-PaO 2 correlates with increased 28-day mortality after ICU admission in sepsis patients, and this association was mainly manifested in the early course of the disease. With a time window of the first 1-7 days after ICU admission, the optimal TWA-PaO 2 range for sepsis patients was ≥130 mmHg and ≤160 mmHg. Increased exposure time, proportion of exposure time, and exposure dose of high-risk PaO 2 outside the range were all associated with an increased risk of 28-day mortality. Conclusion: PaO 2 in sepsis patients should be closely monitored. During the first 1-7 days of ICU admission, PaO 2 should be maintained within the range of ≥130 mmHg and ≤160 mmHg. A dose-dependent relationship exists between high-risk PaO 2 outside the range and patient outcome.
动脉血氧分压（PaO₂）与脓毒症患者预后之间的关系及其随时间的潜在变化尚不清楚。脓毒症患者的最佳PaO₂范围一直是一个有争议的问题，尚未达成共识。我们旨在探讨脓毒症患者不同水平的PaO₂暴露随时间与28天死亡率之间的关联，并确定特定时间范围内脓毒症患者的最佳PaO₂范围。方法：我们从重症监护医学信息集市IV（MIMIC-IV，版本2.2）数据库中检索了在重症监护病房（ICU）入院前或入院后24小时内被诊断为脓毒症的成年患者的数据。我们排除了非首次入住ICU的患者、ICU住院时间<24小时的患者以及在ICU住院期间没有PaO₂结果的患者。我们计算了PaO₂的时间加权平均值（TWA），并使用分段指数加法混合模型（PAMMs）来估计TWA-PaO₂与患者预后之间关联的时间依赖性变化。结果：MIMIC队列共纳入16880例脓毒症患者。结果表明，脓毒症患者ICU入院后患者的TWA-PaO₂与28天死亡率增加相关，且这种关联主要表现在疾病早期。在ICU入院后的前1-7天时间窗口内，脓毒症患者的最佳TWA-PaO₂范围为≥130 mmHg且≤160 mmHg。该范围外高风险PaO₂的暴露时间增加、暴露时间比例和暴露剂量均与28天死亡风险增加相关。结论：应密切监测脓毒症患者的PaO₂。在ICU入院的前l-7天，PaO₂应维持在≥130 mmHg且≤160 mmHg范围内。该范围外的高风险PaO₂与患者预后存在剂量依赖关系。
