The clinical application of metagenomic next-generation sequencing in pathogen identification of postoperative spinal implant infection.

BACKGROUND CONTEXT: Postoperative spinal implant infection (PSII) poses significant diagnostic challenges and can lead to serious clinical outcomes. Traditional microbial culture methods are limited by prolonged turnaround times and low sensitivity, which have long hindered a comprehensive understanding of the pathogen spectrum in PSII. Metagenomic next-generation sequencing (mNGS) offers rapid, broad-spectrum, and highly sensitive pathogen detection, providing a promising solution to the diagnostic difficulties associated with PSII.

PURPOSE: This study aimed to evaluate the clinical utility of mNGS for the pathogenic diagnosis of PSII.

STUDY DESIGN/

SETTING

Retrospective cohort study.

PATIENT SAMPLE: A retrospective analysis was conducted on clinical data from 122 patients with suspected surgical site infection following spinal implant-related surgery between January 2020 and July 2025.

OUTCOME MEASURES: Identification of pathogenic bacteria in patients.

METHODS

According to the National Healthcare Safety Network (NHSN) criteria, patients were classified into the PSII group (n=89) and the nonPSII group (n=33). Pathogen spectrum characteristics of PSII detected by mNGS were analyzed, and the diagnostic performance of mNGS was compared with that of traditional microbial culture, laboratory tests, and imaging studies.

RESULTS

The pathogen detection rate of mNGS was significantly higher than that of traditional culture (79.78% vs 38.20%, p <.001). mNGS successfully detected pathogens in 42.70% (38/89) of culture-negative PSII cases. The main pathogens identified by mNGS included Staphylococcus aureus and coagulase-negative staphylococci (20.00% each, 15/75), Mycobacterium tuberculosis (18.67%, 14/75), and anaerobic bacteria (14.67%, 11/75). Among the 33 patients positive by both mNGS and culture, species-level concordance was 69.70%. No significant difference in the distribution of major pathogens was observed between early-onset (≤3 months) and late-onset (>3 months) infection groups. Antibiotic appropriateness evaluation revealed that only 32.58% (29/89) of patients had initial antibiotic regimens that provided complete coverage against the detected pathogens. The overall diagnostic performance of mNGS for PSII was superior to that of culture, with significantly higher sensitivity (79.78% vs 38.20%), specificity (100.00% vs 93.94%), positive predictive value (100.00% vs 94.44%), negative predictive value (64.71% vs 36.05%), accuracy (85.25% vs 53.28%), and AUC (0.8989 vs 0.6607).

CONCLUSION

mNGS significantly improves the detection of difficult-to-culture pathogens such as M. tuberculosis and anaerobic bacteria in PSII. It is recommended to combine mNGS with conventional methods, thereby improving diagnostic accuracy, guiding rational antibiotic use, and enhancing treatment outcomes.