Abstract
Objective
To provide novel insights for the diagnosis and management of patients with septic shock, this study analyzed the clinical characteristics of these patients and compared the predictive and combined value of blood lactate levels and gastric intramucosal pH (PHi) for in-hospital mortality.
Methods
Comparisons between groups were performed using independent sample t-tests. Logistic regression analyses were conducted using Firth's penalized likelihood method to calculate adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for gastric mucosal PH and lactate levels. A variance inflation factor (VIF) test was performed to rule out potential issues of multicollinearity. Receiver operating characteristic (ROC) curves were generated to assess predictive performance, and the area under the curve (AUC) was calculated.
Results
Patients with septic shock who were admitted to Tianjin Medical University General Hospital's Emergency Intensive Care Unit between November 2022 and March 2024 were recruited for a prospective case-control study. There were 72 patients in all. Within 24 h of admission, baseline data, blood lactate levels, PH values of the stomach mucosa, and Sequential Organ Failure Assessment (SOFA) scores were documented. Patients were divided into two groups based on their 28-day in-hospital mortality: the death group (n = 18) and the survival group (n = 54). Among the 72 patients, 54 (75.0%) survived and 18 (25.0%) died. Independent t-test results showed a significant difference in gastric mucosal PH between the two outcome groups (t = 4.558, p < 0.001), with lower PH levels observed in the death group. Lactate levels also differed significantly between groups (t = −3.782, p < 0.001), being higher in the death group. In the univariate Firth logistic regression model, gastric mucosal PH (OR = 3.40, p < 0.0001) was significantly associated with poor prognosis, whereas lactate (OR = 1.69, p > 0.05) did not reach statistical significance, as its 95% CI included 1.With ORs of 3.94 and 2.29, respectively, both variables remained significant in the bivariate model that included both, indicating a possible mutual attenuation effect and improved predictive power in comparison to univariate models. Lactate lost statistical significance (OR = 0.80, p = 0.782), whereas stomach mucosal PHi remained a independent predictor of death (OR = 10.03, p < 0.05) after controlling for covariates in the multivariate model. These results suggest that while lactate's predictive significance may be muddled, gastric mucosal PH offers reliable and independent prognostic information. The variance inflation factor (VIF) test showed that all included variables had VIF values less than 5 (specifically, 1.15 for lactate and 1.22 for the SOFA score), indicating no severe multicollinearity in the model. ROC curve analysis further confirmed these results. The AUC for gastric mucosal PH was 0.796 (p < 0.001), and for lactate, 0.648 (p = 0.035). When both indicators were combined, the AUC increased to 0.851 (p < 0.001), indicating that combined assessment markedly improves prognostic accuracy in septic shock.
Conclusion
Both gastric mucosal pH and blood lactate levels were associated with in-hospital mortality in patients with septic shock. While lactate did not achieve statistical significance as an independent predictor in the current model, PHi demonstrated an independent connection with prognosis in the multivariable analysis. These results imply that gastric mucosal pH may provide a potential regional perfusion viewpoint for evaluating the state, in contrast to lactate, which represents systemic metabolic status. Significantly, predictive accuracy increased when both markers were evaluated together, suggesting that systemic and regional indicators can be complementing in risk assessment.
Keywords
Get full access to this article
View all access options for this article.