Comparison of the Predictive Value of Neutrophil Percentage-to-Albumin Ratio and Modified Glasgow Prognostic Score for the Risk of Stroke-Associated Pneumonia Among Stroke Patients

To assess the predicting value of neutrophil percentage-to-albumin ratio (NPAR) and modified Glasgow Prognostic Score (mGPS) for Stroke-Associated Pneumonia (SAP) occurrence among stroke patients. We recruited stroke patients (aged 18 years) hospitalized at Tianjin First Central Hospital from January 2022 to February 2023 for this retrospective cohort study. NPAR was categorized into four groups by considering the quartiles: Q1 (<1.38), Q2 (≥1.38 and <1.62), Q3 (≥1.62 and <1.87), Q4 (≥1.87). SAP incident was the primary outcome in this study. Univariate and multivariate logistic regression models were employed to explore the association between NPAR, mGPS and SAP occurrence among individuals with stroke. Besides, we compared the predicting value of NPAR and mGPS for SAP occurrence by the receiver operating characteristic (ROC) curve. Our study encompassed 851 patients with stroke. One hundred and forty-seven patients (17.27%) developed SAP. After accounting for confounding factors, we observed significant positive association of high NPAR with SAP occurrence [(for the third quartile: odds ratio (OR)=2.35, 95% confidence interval (CI): 1.01-5.47; for the fourth quartile: OR=3.35, 95% CI: 1.44-7.77)]. Additionally, the results also indicated that mGPS 1 (OR=2.26, 95% CI: 1.25-4.08) and mGPS 2 (OR=7.37, 95% CI: 2.63-20.70) were related to the increased probability of SAP, respectively. ROC analysis demonstrated that both the NPAR [area under the curve (AUC)=0.729, 95% CI: 0.687-0.771] and mGPS (AUC=0.671, 95% CI: 0.627-0.716) exhibited good predictive power for SAP occurrence. Based on the DeLong test, the predictive value of NPAR for SAP may be significantly superior to that of mGPS (P<0.05). Our findings suggest that both NPAR and mGPS serve as reliable biomarker for assessing SAP risk in stroke patients, with NPAR demonstrating superior predictive value for SAP compared to mGPS.