Unal Cetin Ece, Kurtkulagi Ozge, Kamis Fatih, Das Murat, Simsek Esen, Cetin Adil Ugur, Demirer Aydemir Ferhan, Beyazit Yavuz
Department of Internal Medicine, Faculty of Medicine, Çanakkale Onsekiz Mart University, Çanakkale, Turkey.
Department of Emergency Medicine, Faculty of Medicine, Çanakkale Onsekiz Mart University, Çanakkale, Turkey.
Medicine (Baltimore). 2025 Jul 4;104(27):e43097. doi: 10.1097/MD.0000000000043097.
Evaluating disease severity and predicting adverse outcomes using various risk prediction tools in early disease stages is essential to reduce sepsis-related mortality. Unfortunately, there is still no clear consensus on the best score. The present study aimed to develop and validate a multivariable risk prediction model for 30-days mortality by combining the lactate-to-albumin (L/A) ratio, Modified Early Warning Score (MEWS), and quick Sequential Organ Failure Assessment (qSOFA) in sepsis patients admitted to the intensive care unit (ICU). This retrospective study included ICU patients with suspected sepsis. We computed L/A ratio, MEWS, and qSOFA within 24 hours of ICU admission. Patients were followed until either death/hospital discharge or 30 days, whichever came first. The predictive performance of each scoring system and their combinations was assessed using logistic regression and receiver operating characteristic curve analyses. A total of 130 patients with sepsis admitted to the ICU were included in the study. The mortality rate was 63.07% (82/130). A higher L/A ratio, MEWS, and qSOFA were found to be associated with mortality in ICU sepsis patients. A statistically significant difference in terms of predicting mortality was demonstrated in the pairwise comparison after combining the L/A ratio with both the qSOFA and MEWS (difference between areas: -0.098, P = .011 and difference between areas: -0.098, P = .013 respectively). Mortality models combining L/A ratio with selected clinical variables have improved mortality prediction performance compared with models that use MEWS and qSOFA alone. The L/A ratio at ICU admission provide valuable prognostic information for predicting 30-days mortality in sepsis patients. Combining these ratio with MEWS and qSOFA improves the accuracy of predicting mortality in patients with sepsis.
在疾病早期阶段使用各种风险预测工具评估疾病严重程度并预测不良结局对于降低脓毒症相关死亡率至关重要。不幸的是，对于最佳评分仍没有明确的共识。本研究旨在通过结合入住重症监护病房（ICU）的脓毒症患者的乳酸与白蛋白（L/A）比值、改良早期预警评分（MEWS）和快速序贯器官衰竭评估（qSOFA），开发并验证一个用于预测30天死亡率的多变量风险预测模型。这项回顾性研究纳入了疑似脓毒症的ICU患者。我们在ICU入院后24小时内计算L/A比值、MEWS和qSOFA。对患者进行随访，直至死亡/出院或30天，以先到者为准。使用逻辑回归和受试者工作特征曲线分析评估每个评分系统及其组合的预测性能。本研究共纳入130例入住ICU的脓毒症患者。死亡率为63.07%（82/130）。发现较高的L/A比值、MEWS和qSOFA与ICU脓毒症患者的死亡率相关。将L/A比值与qSOFA和MEWS两者结合后的两两比较中，在预测死亡率方面显示出统计学显著差异（面积差异：-0.098，P = 0.011；面积差异：-0.098，P = 0.013）。与单独使用MEWS和qSOFA的模型相比，将L/A比值与选定临床变量相结合的死亡率模型具有更好的死亡率预测性能。ICU入院时的L/A比值为预测脓毒症患者30天死亡率提供了有价值的预后信息。将该比值与MEWS和qSOFA相结合可提高脓毒症患者死亡率的预测准确性。
