Platelet count predicts mortality in patients with sepsis: A retrospective observational study.

Abstract

Platelet count is a key component of sepsis severity score. However, the predictive value of the platelet count at admission for mortality in sepsis remains unclear. We designed a retrospective observational study of patients with sepsis admitted to our hospital from January 2017 to September 2021 to explore the predictive value of platelet count at admission for mortality. A total of 290 patients with sepsis were included in this study. Multivariate logistic regression analysis was used to evaluate the risk factors for mortality and construct a predictive model with statistically significant factors. Compared with survivors, nonsurvivors tended to be much older and had significantly higher acute physiology and chronic health evaluation II and sequential organ failure assessment scores (P < .001). The platelet count was significantly lower in the nonsurvivor group than in the survivor group (P < .001). Multivariate logistic regression analysis indicated that age (P = .003), platelet count (P < .001) and lactate level (P = .018) were independent risk factors for mortality in patients with sepsis. Finally, the area under the receiver operating characteristic curve of platelet count predicting mortality in sepsis was 0.763 (95% confidence interval, 0.709-0.817, P < .001), with a sensitivity of 55.6% and a specificity of 91.8%. In our study, platelet count at admission as a single biomarker showed good predictability for mortality in patients with sepsis.