RESUMO
BACKGROUND: Although several risk indices have been developed to aid in the diagnosis of NSTIs, these instruments suffer from varying levels of reproducibility and failure to incorporate key clinical variables in model development. The objective of this study was to derive and validate a clinical risk index score - NECROSIS - for identifying NSTIs in emergency general surgery (EGS) patients being evaluated for severe skin and soft tissue infections. METHODS: We performed a prospective study across 16 sites in the US of adult EGS patients with suspected NSTIs over a 30-month period. Variables analyzed included demographics, admission vitals and labs, physical exam, radiographic, and operative findings. The main outcome measure was the presence of NSTI diagnosed clinically at the time of surgery. Multivariate analysis was performed to identify independent predictors for the presence of NSTI using the Hosmer-Lemeshow test and the Akaike information criteria. RESULTS: Of 362 patients, 297 (82%) were diagnosed with a NSTI. Overall mortality was 12.3%. Multivariate analysis identified 3 independent predictors for NSTI: systolic blood pressure ≤ 120 mmHg, violaceous skin, and WBC ≥15 (x103/uL). Multivariate modelling demonstrated Hosmer-Lemeshow goodness of fit (p = 0.9) with a c-statistic for the prediction curve of 0.75. Test characteristics of the NECROSIS score were similar between the derivation and validation cohorts. CONCLUSION: NECROSIS is a simple and potentially useful clinical index score for identifying at-risk EGS patients with NSTIs. Future validation studies are warranted. LEVEL OF EVIDENCE: Diagnostic Tests or Criteria, Level III.