Development of a nutrition- and inflammation-related biomarker score for predicting stroke-associated pneumonia after mechanical thrombectomy
Background Stroke-associated pneumonia (SAP) is a frequent and prognostically critical complication in patients with acute ischemic stroke due to large vessel occlusion (AIS-LVO) undergoing mechanical thrombectomy (MT). However, the potential value of integrating biomarkers related to nutritional status and systemic inflammation for SAP risk stratification has not been fully explored. We aimed to develop a nutrition- and inflammation-related biomarker score for improved risk stratification of SAP. Methods In this retrospective cohort study, AIS-LVO patients treated with MT were included. Machine learning algorithms (LASSO, SVM-RFE, and random forest) were applied to identify key biomarkers. A nutrition- and inflammation-related biomarker score was constructed and evaluated using Cox proportional hazards and Fine–Gray competing risk models. Model performance was comprehensively assessed using time-dependent receiver operating characteristic (ROC) curves, concordance index (C-index), decision curve analysis (DCA), net reclassification improvement (NRI), and integrated discrimination improvement (IDI), with comparison to established nutrition- and inflammation-related indices. Results Among 530 patients, SAP occurred in 38.3%. Prealbumin, D-dimer, and high-sensitivity C-reactive protein (hs-CRP) were consistently identified across machine learning algorithms and used to construct the Prealbumin-D-dimer-Hs-CRP (PDHS) score, which was visualized as a nomogram and implemented as an online calculator. The PDHS score was independently associated with SAP (HR, 2.25; 95% CI, 1.66–3.06; p < 0.001) and remained robust in competing risk analysis. Compared with established nutrition- and inflammation-related indices, PDHS showed relatively higher discrimination metrics and incremental value in risk stratification. Conclusion The PDHS score, comprising prealbumin, D-dimer, and hs-CRP, may provide a practical biomarker-based approach for SAP risk stratification in AIS-LVO patients undergoing MT.
Authors
- Junzhuo Li (ORCID: https://orcid.org/0000-0003-3111-699X)
- 連立明
- Wen Yang
- 童晓琼
Institutions
- Chongqing University (CN)
- Affiliated Hospital of North Sichuan Medical College (CN)
- Tianjin Medical University (CN)
Publication Details
- Journal
- Frontiers in Nutrition
- Published
- 2026-11-19
- DOI
- https://doi.org/10.3389/fnut.2026.1915173
- Primary Topic
- Dysphagia Assessment and Management
- Type
- article
- Field-Weighted Citation Impact
- 0.00