To analyze the value of mean platelet volume/platelet count ratio (MPR) and lactate/albumin ratio (LAR) combined for predicting the prognosis of patients with severe pneumonia.

A total of 119 patients with severe pneumonia admitted to our hospital from January 2020 to December 2024 were retrospectively selected as study subjects. According to the survival status of patients, they were divided into good prognosis group (survival) and poor prognosis group (death). Clinical data and MPR and LAR values of the two groups were collected and compared. Pearson correlation analysis was used to explore the correlation between MPR, LAR and Acute Physiology and Chronic Health Evaluation Ⅱ (APACHE Ⅱ) scores in patients with severe pneumonia. Multivariate logistic regression analysis was used to identify the influencing factors for poor prognosis in patients with severe pneumonia. Receiver operating characteristic (ROC) curves and decision curve analysis were drawn to evaluate the value and clinical net benefit of MPR, LAR, and their combined application for predicting poor prognosis in patients with severe pneumonia.

Among 119 patients with severe pneumonia, 88 cases survived and 31 cases died after treatment. The APACHE Ⅱscore[(19.76±4.01)vs.(25.59±3.38)], D-D[(3.89±1.06)μg/ml vs.(4.35±1.19)μg/ml], CRP[(102.75±26.15)mg/L vs.(114.56±30.12)mg/L], PCT[(7.23±1.56)ng/ml vs.(8.91±2.64)ng/ml], MPR[(4.58±0.90)vs.(5.73±1.15)], LAR[(0.05±0.01)vs.(0.09±0.03)] of survivors and the dead, with statistically significant differences were found in comparison(P<0.05). Pearson correlation analysis showed that MPR and LAR were both positively correlated with APACHE Ⅱ scores in patients with severe pneumonia (P<0.05). Logistic regression analysis showed that after adjusting for other confounding factors, MPR (OR=2.962, 95%CI: 1.674~5.242) and LAR (OR=13.045, 95%CI: 3.899~43.649) were still independent risk factors for poor prognosis in patients with severe pneumonia (P<0.05). ROC curves showed that MPR and LAR had clinical significance in predicting the prognosis of patients with severe pneumonia, and the AUC of combined prediction was higher than that of single indicator prediction (P<0.05). Decision curve analysis results showed that MPR combined with LAR prediction could obtain maximum clinical benefit when the threshold probability was 0.30~0.65.

MPR and LAR are closely related to the condition and prognosis of patients with severe pneumonia, and their combined detection has higher efficacy in predicting poor prognosis of patients.