Predictors of Mortality and Amputation in Patients With Necrotizing Soft Tissue Infections of the Upper Extremity.

PURPOSE

Necrotizing soft tissue infection (NSTI) of the upper extremity is characterized by rapid progression, local tissue necrosis, systemic toxicity, and a high mortality rate. The negative consequences of debridement are balanced against preservation of life and limb. The primary objective of this study was to identify predictors of mortality in upper extremity NSTI. Secondary objectives were to identify predictors of amputation, final defect size, length of stay, and readmission within 30 days.

METHODS

An institutional registry for patients with NSTI was retrospectively queried from a single tertiary center covering a large referral population. Data on confirmed upper extremity NSTI were used to determine patient characteristics, infection data, and operative factors. Laboratory Risk Indicator for Necrotizing Fasciitis (LRINEC) and Quick Sequential Organ Failure Assessment (qSOFA) scores were calculated from primary data to provide a measure of clinical severity. Bivariate screening identifying possible predictors of mortality and multivariable regression was performed to control for confounding. Similar analyses were performed for amputation, final defect size, and readmission within 30 days.

RESULTS

A total of 99 patients met the study criteria. In-hospital mortality occurred in 12 patients, and amputation was performed in 7 patients. Etiology, causative organism, and clinical severity scores were variable. Logistic regression showed mortality to be independently predicted by vasopressor dependency outside of operative anesthesia. The relatively low number of case events, limited sample size, and multiple comparisons limited the evaluation of lesser predictor variables. The LRINEC score did not strongly predict amputation or death in this series.

CONCLUSIONS

Necrotizing soft tissue infection of the upper extremity carries risk of mortality and amputation, and effective treatment requires prompt recognition, early goal-directed resuscitation, and early debridement. The strongest independent predictor of in-hospital mortality was vasopressor dependence outside operative anesthesia. The LRINEC score did not strongly predict death or amputation in upper extremity NSTI.

TYPE OF STUDY/LEVEL OF EVIDENCE: Prognostic IV.