Porous PtCu Alloys Decode Plasma Metabolic Fingerprints for the Recognition of Severe Community‐Acquired Pneumonia

This article constructs porous PtCu alloys to effectively extract plasma metabolic fingerprints (PMFs) via laser desorption/ionization mass spectrometry (LDI MS), achieving rapid and accurate recognition of severe community-acquired pneumonia (CAP) by machine learning of PMFs, overcoming the clinical inefficiency for ATS/IDSA diagnosis process. The work affords a rapid and precise method for CAP management through metabolite analysis.

Abstract

Rapid and accurate recognition of severe community-acquired pneumonia (CAP) would facilitate the optimal intervention. Currently, the diagnosis of severe CAP is commonly based on the criteria established by Infectious Disease Society of America (IDSA)/American Thoracic Society (ATS), which include 2 primary and 9 secondary criteria, making the process cumbersome and time-consuming. Here, a porous PtCu alloy-assisted laser desorption/ionization mass spectrometry (LDI MS) is designed for the extraction of plasma metabolic fingerprints (PMFs), coupling with machine learning for the diagnosis of severe CAP. The PtCu alloys with optimal particle size exhibit excellent sensitivity, reproducibility, and universality for metabolite detection, due to the porous structure, promising photoelectric effect, and improved melting surface structure. Further, the nanoplatform successfully records the PMFs within seconds, using only 0.5 µL native plasma. Machine learning of PMFs on 69 individuals produces a diagnostic model with an area under curve (AUC) of 0.832. Particularly, a three metabolic biomarker panel demonstrates enhanced diagnostic efficiency (AUC of 0.846), outperforming reported biomarkers (AUC of 0.560–0.770). Notably, the diagnosis can be completed in ≈35 min. The work affords a rapid and precise method for CAP management through metabolite analysis.