Systemic immune inflammation index and peripheral blood carbon dioxide concentration at admission predict poor prognosis in patients with severe traumatic brain injury

• DOI: 10.3389/fimmu.2022.1034916
• Publication Date: 2023-01-09T11:18:51Z
• Authors: Li Chen; Shaohuai Xia; Yi Zuo; Yinghong Lin; Xianshen Qiu; Qizuan Chen; Tianshun Feng; Xuewei Xia; Qixiang Shao; Shousen Wang
• Keywords: Glasgow Outcome Scale

Background Recent studies have shown that systemic inflammation responses and hyperventilation are associated with poor outcomes in patients severe traumatic brain injury (TBI). The aim of this retrospective study was to investigate the relationships between immune index (SII = platelet × neutrophil/lymphocyte) peripheral blood CO 2 concentration at admission Glasgow Outcome Score (GOS) 6 months after discharge TBI. Methods We retrospectively analyzed clinical data for 1266 TBI three large medical centers from January 2016 December 2021, recorded GOS discharge. receiver operating characteristic (ROC) curve used determine best cutoff values SII, , neutrophil lymphocyte ratio (NLR), (PLR), monocyte (LMR), chi-square tests were evaluate among basic characteristics Multivariate logistic regression analysis independent prognostic factors Finally, ROC curve, nomogram, calibration decision analyses value SII coSII-CO2 predicting prognosis And we multifactor method build CRASH model IMPACT model. included age, GCS score (GCS, Coma Scale) Pupillary reflex light: one, both, none. includes motor Results curves indicated PLR, NLR LMR 2651.43×10 9 22.15mmol/L, 190.98×10 9.66×10 1.5×10 respectively. high low significantly poorer than those . revealed systolic pressure (SBP), pupil size, subarachnoid hemorrhage (SAH), serum potassium [K + ], calcium [Ca 2+ international normalized (INR), C-reactive protein (CRP) co-systemic combined carbon dioxide (coSII-CO ) (P < 0.001) In training group, C-index 0.837 0.860 coSII-CO external validation 0.907 0.916 Decision confirmed a superior net benefit rather most cases. Furthermore, probability showed better agreement observed results when based on nomogram. According machine learning, ranked first importance followed by then SII. Conclusions predictive performance NLR, PLR LMR. can be as new, accurate objective predictors, combining improve accuracy prediction