降钙素原与白蛋白比值在脓毒症并发急性呼吸窘迫综合征预后中的评估价值

Evaluation of the Prognostic Value of the Procalcitonin to Albumin Ratio in Patients with Sepsis-Induced Acute Respiratory Distress Syndrome

摘要: 目的：探讨降钙素原与白蛋白比值(PAR)在脓毒症并发急性呼吸窘迫综合征(ARDS)预后中的评估价值。方法：回顾性分析2019年9月~2022年9月入住益阳市中心医院呼吸内科重症监护室(Respiratory Intensive Care Unit, RICU)的131例脓毒症并发ARDS患者的临床资料，根据患者28天预后，进一步将脓毒症并发ARDS患者分成生存组(n = 71)和死亡组(n = 60)，比较两组降钙素原与白蛋白比值的差异，采用受试者工作曲线(ROC)评估PAR对脓毒症并发ARDS的预后价值；Spearman相关分析分析PAR与SOFA评分的相关性；Cox回归筛选脓毒症并发ARDS预后的独立危险因素；进一步比较不同亚组的PAR水平；最后采用Kaplan-Meier生存分析法比较不同PAR水平患者的生存率。结果：死亡组PAR水平显著高于生存组[0.85 (0.11, 2.84) vs. 0.19 (0.03, 1.01), P = 0.001]。ROC曲线分析表明，PAR判断脓毒症并发ARDS的预后性能最佳，曲线下面积(AUC)为0.725 (95% CI: 0.6141~0.800, P < 0.001)，截断值为1.74，敏感度为40.00%，特异度为98.55%。相关性分析表明PAR与SOFA评分成正相关(r = 0.365, P < 0.001)。多因素Cox回归分析表明PAR高水平是脓毒症并发ARDS预后的独立危险因素(HR: 1.316, 95% CI: 1.092~1.584, P = 0.004)。亚组分析表明革兰阴性杆菌组PAR水平显著高于革兰阳性球菌组[1.24 (0.20, 2.65) vs. 0.13 (0.03, 0.79), P < 0.001]。生存分析表明PAR > 1.74的脓毒症并发ARDS患者28-d预后更差(P < 0.001)。结论：PAR对脓毒症并发ARDS患者28 d死亡率有一定的预测价值，当PAR > 1.74时，脓毒症并发ARDS患者28 d死亡率更高。

Abstract: Objective: To investigate the prognostic value of serum procalcitonin to albumin ratio (PAR) for the patients with sepsis-induced acute respiratory distress syndrome (ARDS). Methods: To retrospec-tively analyze the clinical data of 131 patients with sepsis-induced ARDS admitted to the respirato-ry intensive care unit (RICU) of Yiyang Central Hospital from September 2019 to September 2022. According to the prognosis at 28 d, 131 sepsis patients were divided into survival group (n = 71) and non-survival group (n = 60). PAR levels were compared between different groups. The receiver operating characteristic (ROC) curve was used to evaluate the prognostic value of PAR for sep-sis-induced ARDS. Spearman correlation was used to test for an association between PAR and SOFA score. Independent prognostic factors of sepsis-induced ARDS were evaluated using the Cox regres-sion analyses. PAR levels amongst various subgroups were further compared. Finally, the Kaplan-Meier survival analysis was used to compare the survival status of patients with different PAR levels. Results: The PAR level of the non-survival group was significantly higher than that of the survival group [0.85 (0.11, 2.84) vs. 0.19 (0.03, 1.01), P = 0.001]. ROC curve analysis showed that PAR had the best prognostic performance in determining sepsis-induced ARDS, the area under ROC curve of PAR in predicting prognosis was 0.725 (95% CI: 0.6141~0.800, P < 0.001), and the cutoff value, the sensitivity, and the specificity were −1.74, 40.00%, 98.55%, respectively. Correlation analysis suggested a significant positive correlation between serum PAR levels and SOFA scores (r = 0.365, P < 0.001). Multivariate Cox regression analyses indicated that high-level PAR expression served as an independent prognostic factor for poor survival in patients with sepsis-induced ARDS (HR: 1.316, 95% CI: 1.092~1.584, P = 0.004). Subgroup analysis showed that PAR levels were sig-nificantly higher in the Gram-negative bacilli group [1.24 (0.20, 2.65)] than in the Gram-positive cocci group [0.13 (0.03, 0.79)]. Kaplan-Meier survival analysis showed that higher PAR levels (>1.74) were correlated with poor 28-day survival in patients with sepsis-induced ARDS (P < 0.001). Con-clusion: This study indicates that serum PAR is useful in predicting the 28-day mortality among pa-tients with sepsis-induced ARDS, especially those with serum PAR of more than 1.74.