Background: Patients suffered aneurysmal subarachnoid hemorrhage (aSAH) usually develop poor survival and functional outcome. Evaluating aSAH patients at high risk of poor outcome is necessary for clinicians to make suitable therapeutical strategy. This study is conducted to develop prognostic model using XGBoost (extreme gradient boosting) algorithm in aSAH. Methods: A total of 351 aSAH patients admitted to West China hospital were identified. Patients were divided into training set and test set with ratio of 7:3 to testify the predictive value of XGBoost based prognostic model. Additionally, logistic regression model was also constructed and compared with XGBoost based model. Area under the receiver operating characteristic curve (AUC), sensitivity and specificity were calculated to evaluate the value of XGBoost and logistic regression. Results: There were 74 (21.1%) non-survivors and 148 (42.1%) patients with unfavorable functional outcome. Non-survivors had older age (p=0.025), lower Glasgow coma scale (GCS) (p< 0.001), higher World Federation of Neurosurgical Societies WFNS score (p< 0.001), mFisher score (p< 0.001). The incidence of intraventricular hemorrhage (IVH) (p=0.025) and delayed cerebral ischemia (DCI) (p< 0.001) was higher in non-survivors than survivors. The AUC of XGBoost model for predicting mortality and unfavorable functional outcome were 0.950 and 0.958, which were higher than 0.767 and 0.829 of logistic regression model. Conclusion: XGBoost based model is more precise than logistic regression model in predicting outcome of aSAH patients. Using XGBoost prognostic model is helpful for clinicians to identify high-risk aSAH patients and therefore strengthen medical care.

Background: Traumatic brain injury (TBI) is a serious public health issue all over the world. This study was designed to evaluate the prognostic value of lactate to albumin ratio (LAR) on patients with moderate to severe TBI.Methods: Clinical data of 273 moderate to severe TBI patients hospitalized in West China Hospital between May 2015 and January 2018 were collected. Multivariate logistic regression analyses were used to explore risk factors and construct a prognostic model of in-hospital mortality in this cohort. A receiver operating characteristic (ROC) curve was drawn to evaluate the discriminative ability of this model.Results: Non-survivors had higher LAR than survivors (1.09 vs. 0.53, p < 0.001). Results of multivariate logistic regression analysis showed that Glasgow Coma Scale (GCS; odds ratio [OR] = 0.743, p = 0.001), blood glucose (OR = 1.132, p = 0.005), LAR (OR = 1.698, p = 0.022), subdural hematoma (SDH; OR = 2.889, p = 0.006), intraparenchymal hemorrhage (IPH; OR = 2.395, p = 0.014), and diffuse axonal injury (DAI; OR = 2.183, p = 0.041) were independent risk factors of in-hospital mortality in included patients. These six factors were utilized to construct the prognostic model. The area under the ROC curve (AUC) values of single lactate, albumin, and LAR were 0.733 (95% Cl; 0.673–0.794), 0.740 (95% Cl; 0.683–0.797), and 0.780 (95% Cl; 0.725–0.835), respectively. The AUC value of the prognostic model was 0.857 (95%Cl; 0.812–0.901), which was higher than that of LAR (Z = 2.1250, p < 0.05).Conclusions: Lactate to albumin ratio is a readily available prognostic marker of moderate to severe TBI patients. A prognostic model incorporating LAR is beneficial for clinicians to evaluate possible progression and make treatment decisions in TBI patients.

Purpose: To explore the prognostic value of albumin corrected anion gap (ACAG) within 24 hours of admission to the intensive care unit (ICU) for acute pancreatitis (AP). Patients and Methods: This was a retrospective cohort study. Adult AP patients admitted to ICU from June 2016 to December 2019 were included in the study, who were divided into three groups according to initial serum ACAG within 24 hours upon ICU admission: ACAG ≤ 14.87 mmol/L, 14.87 < ACAG ≤ 19.03 mmol/L, and ACAG > 19.03 mmol/L. The primary study outcome indicator was in-hospital mortality. Age, sex, Glasgow Coma Scale score, and Acute Physiology and Chronic Health Evaluation II (APACHE II) score were matched through propensity score matching (PSM) method to balance the baseline between the survivors and non-survivors. Multivariate Cox regression was used to determine the relationship between ACAG and in-hospital mortality. Results: A total of 344 patients (of them 81 non-survivors) were analyzed in this study. Patients with higher ACAG intended to present significantly higher in-hospital mortality, APACHE II score, creatine, lower albumin, and bicarbonate. Multivariate Cox regression analysis after matching demonstrated that white blood cell count, platelet count, and higher ACAG were independently associated with higher in-hospital mortality (ACAG ≤ 14.87 as a reference, 14.87 < ACAG ≤ 19.03 mmol/L with HR of 2.34 and 95% CI of 1.15– 4.76, ACAG > 19.03 with HR of 3.46 and 95% CI of 1.75– 6.84). Conclusion: Higher ACAG was independently associated with higher in-hospital mortality in patients with AP after matching the baseline between the survivors and non-survivors.

Background: The Endothelial Activation and Stress Index (EASIX) is a novel marker of endothelial injury and correlates with survival of various patients. The endothelial dysfunction plays an important role on the pathophysiological process of traumatic brain injury (TBI). This study was designed to explore the prognostic value of EASIX on TBI patients. Methods: 358 TBI patients hospitalized in the West China hospital between October 2018 and October 2022 were enrolled for this study. The EASIX was calculated based on the formula: lactate dehydrogenase (U/L) × creatinine (mg/dL)/platelets (10⁹ cells/L). The univariate and multivariate logistic regression with forward method was performed to explore the association between EASIX and mortality. A prognostic model was developed combining significant risk factors in the multivariate logistic regression. The receiver operating characteristic (ROC) curve was used to compare the predictive accuracy of the EASIX and the developed model. Results: The 30-day mortality of enrolled 358 TBI patients was 51.1%. Non-survivors had higher EASIX than survivors (p < 0.001). The multivariate logistic regression confirmed seven risk factors for mortality of TBI including injury mechanism (p = 0.010), GCS (p < 0.001), glucose (p < 0.001), EASIX (p = 0.017), subdural hematoma (p = 0.012), coagulopathy (p = 0.001). The AUC of EASIX, SOFA, GCS was 0.747, 0.748 and 0.774, respectively. The AUC of developed predictive model was 0.874 with the sensitivity of 0.913 and specificity of 0.686. Conclusions: The EASIX is a reliable marker for predicting mortality of TBI patients. The predictive model incorporating EASIX is helpful for clinicians to evaluate the mortality risk of TBI patients.

Background: Acute kidney injury (AKI) has been occurs commonly in the clinical management of traumatic brain injury (TBI) patients and is correlated with outcomes in these patients. We designed this study to investigate the incidence, duration, stage, and burden of AKI among these patients. Methods: A total of 419 TBI inpatients at our hospital were included in the study. We calculated the AKI burden, reflecting both stage and duration, and then analyzed associations among AKI occurrence, highest AKI stage, AKI duration, AKI burden, and outcomes with logistic regression analysis. Results: Incidence of AKI among TBI patients was 19.8%. These patients’ AKIs occurred mainly on the first day from admission (10.74%), and mostly developed stage 1 AKI (9.79%). Modes of AKI duration and burden in those with AKI were both 1. Multivariate logistic regression showed AKI occurrence (OR 3.792, p=0.004) and the highest AKI stage (OR 3.122, p< 0.001) was significantly associated with mortality. Neither AKI duration (OR 1.083, p=0.206) nor AKI burden (OR 1.062, p=0.171) were associated with mortality. Incorporating AKI occurrence or highest AKI stage did not improve the predictive value of the constructed prognostic model. Conclusion: The high-incidence period of AKI in TBI patients was the first 3 days after admission. AKI occurrence and highest AKI stage were associated with mortality, while AKI duration and AKI burden were not associated with mortality.

Background. The association between atherogenic index of plasma (AIP) and hyperuricemia remains indistinct. This study was aimed to examine the relationship between AIP and hyperuricemia among the middle-aged and the elderly Chinese population. Methods. Datasets were retrieved from the China Health and Retirement Longitudinal Study (CHARLS) survey conducted in 2011 and 2015. 13,021 participants in the CHARLS in 2011 and 7,017 participants involved both in 2011 and 2015 were included, respectively. The measurement of AIP and hyperuricemia was based on the test of fasting blood. Association between AIP and hyperuricemia was assessed by logistic regression, and the nonlinear association was examined by restricted cubic splines (RCS). The cutoff point of AIP was calculated using receiver operator curve (ROC). 1:1 propensity score matching (PSM) was adopted to further explore the relationship between AIP and hyperuricemia. Results. In the section of a cross-sectional study, a positive association between AIP and hyperuricemia was found. The odds ratios (ORs) of hyperuricemia were 1.00 (reference), 1.52 (1.102.10), 1.80 (1.312.47), and 3.81 (2.845.11). Nonlinear association was not detected using RCS analysis. There were 664 hyperuricemia cases during the four years follow-up. The hyperuricemia prevalence was 9.5. In the fully adjusted longitudinal analysis, the ORs for hyperuricemia across the quartiles of AIP were 1.00 (reference), 1.00 (0.741.37), 1.59 (1.202.11), and 2.55 (1.943.35), respectively. In the longitudinal analysis after PSM, the OR of hyperuricemia were 1.91 (1.45, 2.51) and 1.92 (1.45, 2.54) in the univariate and multivariate model, respectively. Conclusion. AIP can predict the prevalence of hyperuricemia in the Chinese middle-aged and elderly population.