作者
Yan Zhou,Yong-wen Feng,Xiaomin Liang,Shuiqing Gui,Di Ren,Yuanzhi Liu,Jijia She,Xiaomei Zhang,Fei Song,Lina Yu,Yiwen Zhang,Jinping Wang,Zhi-ye Zou,Mei Jiang,Sha Wen,Mei Yang,Xinsi Li,Xuerui Tan,Ying Li
摘要
Abstract This study aimed to investigate whether changes in presepsin, procalcitonin, high-sensitivity C-reactive protein, and interleukin 6 levels predict mortality in septic patients in the intensive care unit. This study enrolled septic patients between November 2020 and December 2021. Levels of presepsin, procalcitonin, high-sensitivity C-reactive protein, and interleukin 6 were measured on the first (PSEP_0, PCT_0, hsCRP_0, IL-6_0) and third days (PSEP_3, PCT_3, hsCRP_3, IL-6_3). Follow-up was performed on days 3, 7, 14, 21, and 28 after enrollment. The outcome was all-cause death. The study included 119 participants, and the mortality was 18.5%. In univariable Cox proportional hazards regression analysis, ΔPSEP (= PSEP_3 – PSEP_0) > 211.49 pg/mL (hazard ratio, 2.70; 95% confidence interval, 1.17–6.22), ΔPCT (= PCT_3 – PCT_0) > –0.13 ng/mL (hazard ratio, 7.31; 95% confidence interval, 2.68–19.80), ΔhsCRP (= hsCRP_3 – hsCRP_0) > −19.29 mg/L (hazard ratio, 6.89; 95% confidence interval, 1.61–29.40), and ΔIL-6 (= IL-6_3 – IL-6_0) > 1.00 pg/mL (hazard ratio, 3.13; 95% confidence interval, 1.35–7.24) indicated an increased risk of mortality. The composite concordance index for alterations in all 4 distinct biomarkers was highest (concordance index, 0.83; 95% confidence interval, 0.76–0.91), suggesting the optimal performance of this panel in mortality prediction. In decision curve analysis, compared with the Acute Physiology and Chronic Health Evaluation II and Sequential (sepsis-related) Organ Failure Assessment scores, the combination of the 4 biomarkers had a larger net benefit. Interestingly, interleukin 6 was predominantly produced by monocytes upon lipopolysaccharide stimulation in peripheral blood mononuclear cells. ΔPSEP, ΔPCT, ΔhsCRP, and ΔIL-6 are reliable biomarkers for predicting mortality in septic patients in the intensive care unit, and their combination has the best performance.