Diagnostic performance of gram-stained soft tissue infectious fluid in necrotizing soft tissue infection: a cohort study with prospective validation

Soft tissue infectious fluid may be obtained for microbiological culture in severe skin and soft tissue infections. However, its diagnostic value in distinguishing necrotizing soft tissue infection (NSTI) from severe non-necrotizing skin and soft tissue infection (SSTI) remains unclear. This study evaluated the diagnostic performance of qualitative and semi-quantitative Gram staining of this specimen. We conducted a retrospective cohort study (April 2019–April 2023) with prospective validation (May 2023–April 2025). Adult patients presenting to the emergency department with suspected limb NSTI who underwent ultrasound-guided aspiration of soft tissue infectious fluid for Gram staining and culture were included. Patients were classified as having NSTI or non-necrotizing SSTI according to final discharge diagnoses. Seventy-six patients met the inclusion criteria, of whom 10 were excluded due to abscesses identified by ultrasound-guided aspiration. In the retrospective cohort, 22 patients had NSTI and 14 had non-necrotizing SSTI; in the prospective cohort, 19 had NSTI and 11 had non-necrotizing SSTI. Bacterial presence on Gram-stained smears was significantly more frequent in patients with NSTI in both the retrospective ( p = 0.011) and prospective cohorts ( p = 0.002). Semi-quantitative Gram-stain scores were significantly higher in patients with NSTI across cohorts, and culture positivity was also higher in NSTI ( p < 0.001). The presence of bacteria on Gram-stained smears demonstrated fair discrimination for NSTI (AUC 0.714–0.789), with high specificity (92.9%–100%) and positive predictive value (91.7%–100%). A semi-quantitative Gram-stain score ≥ 2 achieved maximal specificity and positive predictive value (100%) across cohorts. Qualitative and semi-quantitative Gram staining of soft tissue infectious fluid demonstrated high specificity and positive predictive value for NSTI and may serve as a rule-in diagnostic adjunct. Given its simplicity and rapid availability, it may support clinical decision-making during initial evaluation.