Gut microbiota and SCFAs as predictors of postoperative infection site in neurocritical care patients
A six-genus fecal microbial panel collected within 24 h post-surgery discriminated pulmonary from intracranial infection with an AUC of 0.901, supporting microbial profiling as a risk stratifier — but only in a single-center cross-sectional study of 90 patients.
| Population | Neurocritical care adults undergoing neurosurgical procedures (n=90), stratified into intracranial infection (IC, n=30), pulmonary infection (PI, n=30), and non-infected controls (CON, n=30) |
|---|---|
| Exposure | 16S rRNA gene sequencing and GC-MS quantification of SCFAs from fecal samples collected within 24 h post-surgery |
| Comparator | Three-group comparison (IC vs PI vs CON) |
| Outcome | Infection site discrimination (6-genus panel); Alpha microbial diversity (IC vs CON and PI vs CON); Fecal butyrate concentration (IC vs PI); Fecal propionate concentration (PI vs IC); Relative abundance of Proteobacteria (PI vs IC and CON); Branched-chain SCFA proportion (IC vs PI and CON); Predicted functional metagenomic profiles (metabolic and pathogenicity pathways) |
Summary of findings
| Outcome | Effect | 95% CI | Certainty | Clinical relevance | Notes |
|---|---|---|---|---|---|
| Infection site discrimination (6-genus panel) | AUC 0.901, 95% CI not reported | — | Low | — | 1 studies |
| Alpha microbial diversity (IC vs CON and PI vs CON) | Significant reduction in both infected groups vs CON; exact values and CI not reported | — | Low | — | 1 studies |
| Fecal butyrate concentration (IC vs PI) | Higher in IC vs PI; absolute values and CI not reported | — | Low | — | 1 studies |
| Fecal propionate concentration (PI vs IC) | Higher in PI vs IC; absolute values and CI not reported | — | Low | — | 1 studies |
| Relative abundance of Proteobacteria (PI vs IC and CON) | Expanded in PI vs IC and CON; exact values and CI not reported | — | Low | — | 1 studies |
| Branched-chain SCFA proportion (IC vs PI and CON) | Elevated in IC vs PI and CON; absolute values and CI not reported | — | Low | — | 1 studies |
| Predicted functional metagenomic profiles (metabolic and pathogenicity pathways) | Distinct pathways among groups by PICRUSt prediction; in the quantitative effect sizes reported | — | Low | — | 1 studies |
Context
Pulmonary and intracranial infections are the most frequent postoperative complications in neurosurgery, with incidences of ~40% and 0.7–8%, respectively. Antimicrobial management differs fundamentally between the two sites. No early biomarkers to guide targeted prophylaxis currently exist in clinical practice.
What the study showed
Both infected groups showed significantly reduced microbial diversity compared with controls. The IC group exhibited enrichment of Enterococcus, Staphylococcus, and Clostridium, higher butyrate levels, lower propionate levels, and elevated branched-chain SCFA proportions. The PI group showed Proteobacteria expansion, enrichment of Klebsiella and Escherichia-Shigella, higher propionate and lower butyrate concentrations. The six-genus panel (Enterococcus, Faecalibacterium, Dialister, Escherichia-Shigella, Agathobacter, Blautia) achieved an AUC of 0.901; 95% CI, sensitivity, and specificity were not reported in the available text.
How it was done
Single-center cross-sectional study of 90 post-neurosurgical patients. Fecal samples collected within 24 h after surgery. Analysis by 16S rRNA amplicon sequencing and GC-MS for SCFAs. Balanced groups (n=30 each); infection diagnosis based on standard clinical and microbiological criteria.
Effect magnitude
AUC 0.901 for the six-genus panel in discriminating IC vs PI vs CON; 95% CI for AUC, sensitivity, and specificity not reported in available text.
Risk of bias
Cross-sectional design precludes establishing causality or temporality between microbial profile and infection development. Small sample (n=30/group) from a single center limits generalizability. No formal risk-of-bias tool (RoB 2 or ROBINS-I) was applied. No external validation of the predictive panel. Confounders (preoperative antibiotics, diet, prior probiotic use, surgical duration) were not fully controlled.
What this study does NOT prove
This study does not prove that gut microbiota causes the type of infection; the association may be reverse or confounded by preoperative treatments. The predictive panel has not been externally validated and cannot be transposed to clinical use.
In clinical practice
Clinicians should not alter management based on this study alone. Findings support the hypothesis that early fecal profiles differ between patients who develop PI versus IC, which may inform future prospective validation studies. No therapeutic or diagnostic recommendation is directly derivable from this work.
Limitations
Cross-sectional design precludes establishing causality or temporality between microbial profile and infection development. Small sample (n=30/group) from a single center limits generalizability. No formal risk-of-bias tool (RoB 2 or ROBINS-I) was applied. No external validation of the predictive panel. Confounders (preoperative antibiotics, diet, prior probiotic use, surgical duration) were not fully controlled.
What is still missing
Prospective multicenter study with external validation of the six-genus panel, including AUC 95% CI, multivariate analysis controlling for antibiotics and surgical duration, and assessment of the clinical impact of early stratification.
Technical appendix
Version history
- 1.0 · 2026-10-13 — Auto-generated under Evidence Standard v1.0