ADVANCING PRECISION MEDICINE IN SURGICAL INFECTIONS THROUGH MOLECULAR DIAGNOSTICS, MICROBIOME ANALYSIS, AND ARTIFICIAL INTELLIGENCE: A SYSTEMATIC REVIEW

Abstract

Background: Surgical infections remain important causes of postoperative morbidity, prolonged hospitalization, reoperation, implant failure, antimicrobial exposure, and increased healthcare expenditure. Conventional microbial culture remains central to diagnosis but may fail after antimicrobial exposure or in infections involving biofilms, low microbial burdens, anaerobes, fastidious organisms, or complex polymicrobial communities. Molecular diagnostics, microbiome analysis, and artificial intelligence provide complementary methods for improving pathogen detection, identifying endogenous sources of infection, predicting risk, and supporting postoperative surveillance. Objective: This systematic review evaluated the clinical applications, diagnostic contributions, limitations, and future role of molecular diagnostics, microbiome analysis, and artificial intelligence in precision management of surgical infections. Methods: MEDLINE/PubMed, Embase, Scopus, Web of Science, and the Cochrane Library were searched for studies published from January 2000 through February 2026. Search terms combined surgical-site infection, postoperative infection, implant infection, periprosthetic joint infection, molecular diagnosis, polymerase chain reaction, next generation sequencing, metagenomics, microbial cell-free DNA, microbiome, machine learning, deep learning, and wound imaging. Primary clinical studies that evaluated these technologies for infection prediction, diagnosis, prevention, surveillance, or treatment guidance were included. Reviews, laboratory-only studies, case reports, non-surgical studies, and reports without clinically relevant outcomes were excluded. Eleven primary studies were included in the qualitative synthesis. Owing to heterogeneity in surgical settings, specimen types, diagnostic platforms, reference standards, and computational outcomes, statistical pooling was not undertaken. Results: Four studies primarily evaluated molecular detection of established or suspected infection, three investigated perioperative or implant-associated microbial communities, three evaluated artificial-intelligence-assisted postoperative surveillance, and one evaluated sequencing-guided antimicrobial prophylaxis. Broad-range 16S ribosomal RNA polymerase chain reaction identified bacterial DNA in 53 of 97 conventionally culture-negative surgical-site infection specimens, producing an additional detection rate of 54.6%. Sequencing and microbial cell-free DNA increased pathogen identification in selected periprosthetic joint infections but generated uncertainty concerning contamination, organism viability, and clinical significance. In a prospective spinal-fusion cohort, 86% of surgical-site infections were attributable to strains carried by patients before surgery, supporting the importance of endogenous microbial reservoirs. Colorectal microbiome studies demonstrated differences in intraoperative microbial communities before clinical infection developed. Artificial-intelligence systems using wound images and patient-reported outcomes achieved clinically promising discrimination and substantially reduced simulated clinical-review workload. A multicentre wound-image model achieved 94% accuracy for incision recognition and an area under the receiver-operating-characteristic curve of 0.81 for infection detection. Conclusions: Molecular diagnostics currently provide the clearest clinical contribution in culture-negative, antimicrobial pretreated, polymicrobial, and implant-associated infections. Microbiome analysis has improved understanding of endogenous infection sources and may eventually support individualized prevention. Artificial intelligence can facilitate remote monitoring, risk stratification, and surveillance but requires external validation, calibration assessment, equitable datasets, and clinical governance. These technologies function best as additions to high-quality specimen collection, conventional culture, histopathology, surgical source control, and antimicrobial stewardship.