Zhu Ting, Tian Biao, Wang Lei
Department of Critical Care Medicine, South China Hospital of Shenzhen University Shenzhen 518100, Guangdong, PR China.
Am J Cancer Res. 2024 Dec 15;14(12):5839-5850. doi: 10.62347/NKOL2327. eCollection 2024.
This study investigated the predictive value of combining peripheral blood indicators with procalcitonin clearance rate (PCTc) to assess mortality risk in cancer patients with sepsis, aiming to develop a more sensitive and specific clinical tool. A retrospective analysis was conducted on 393 cancer patients with sepsis admitted to South China Hospital of Shenzhen University from January 2019 to January 2024. Collected data included clinical demographics, laboratory indicators such as white blood cell count, neutrophil count (NEUT), platelet count (PLT), lymphocyte count (LYC), C-reactive protein, procalcitonin (PCT), alanine aminotransferase, and the ratio of arterial oxygen partial pressure to inspired oxygen fraction, as well as functional scores like Acute Physiology and Chronic Health Evaluation II (APACHE II) and Sequential Organ Failure Assessment. Multivariate logistic regression and receiver operating characteristic curves assessed the predictive ability of these factors for 28-day survival. Results showed significantly higher NEUT (P<0.001) and lower PLT and LYC (P<0.001) in the death group, while APACHE II score (area under the curve (AUC) = 0.776) and PCT 24h (AUC = 0.723) demonstrated strong predictive value for mortality risk. The joint projection model's AUC reached 0.966, significantly outperforming individual indicators, indicating that combining multiple indicators offers a more accurate prediction of survival versus mortality risk. Additionally, 24h LCR and 24h PCTc were notably lower in the death group compared to the survival group, reinforcing the advantage of combined indicators for prognosis. Overall, using both peripheral blood indicators and PCTc significantly improves the accuracy of mortality risk assessment in cancer patients with sepsis, enhancing prognostic evaluation and supporting optimized clinical decision-making.
本研究探讨了外周血指标与降钙素原清除率（PCTc）相结合对癌症合并脓毒症患者死亡风险的预测价值，旨在开发一种更敏感、特异的临床工具。对2019年1月至2024年1月在深圳大学附属华南医院住院的393例癌症合并脓毒症患者进行回顾性分析。收集的数据包括临床人口统计学资料、实验室指标，如白细胞计数、中性粒细胞计数（NEUT）、血小板计数（PLT）、淋巴细胞计数（LYC）、C反应蛋白、降钙素原（PCT）、丙氨酸氨基转移酶以及动脉血氧分压与吸入氧分数之比，以及功能评分，如急性生理与慢性健康状况评分系统II（APACHE II）和序贯器官衰竭评估。多因素逻辑回归和受试者工作特征曲线评估了这些因素对28天生存的预测能力。结果显示，死亡组的NEUT显著升高（P<0.001），PLT和LYC降低（P<0.001），而APACHE II评分（曲线下面积（AUC）=0.776）和PCT 24小时（AUC = 0.723）对死亡风险具有较强的预测价值。联合预测模型的AUC达到0.966，显著优于单个指标，表明多个指标联合能更准确地预测生存与死亡风险。此外，死亡组的24小时淋巴细胞计数与C反应蛋白比值（LCR）和24小时PCTc显著低于生存组，进一步证明了联合指标对预后的优势。总体而言，外周血指标与PCTc联合使用可显著提高癌症合并脓毒症患者死亡风险评估的准确性，加强预后评估并支持优化的临床决策。
