Pantelie Nicolcescu, Alina Bereanu, Ionela Mihai, Manuela Arbune, Mihai Sava, Nicolae Grigore, Adrian Hasegan
Background: Fournier's gangrene (FG) is a severe necrotizing infection requiring urgent resuscitation, empirical broad-spectrum antimicrobials, and radical debridement. We evaluated established severity scores, Sepsis-3-defined organ dysfunction, and malignancy status in a single-center cohort with surgically confirmed FG. Methods: This retrospective observational study analyzed 21 consecutive adult patients treated between 1 January 2019 and 31 January 2025 at a Romanian tertiary referral center. Pre-treatment baseline FGSI, SFGSI, LRINEC, NLR, SOFA, and qSOFA were calculated. All patients received intravenous antimicrobials within 90 min and underwent surgical debridement within 6 h. In-hospital mortality was assessed via ROC curves, Fisher's exact test, Haldane-Anscombe odds ratios (ORs), and Firth penalized logistic regression. Results: In-hospital mortality was 19.0% (4/21). Admission Sepsis-3 sepsis (SOFA ≥ 2) was present in 7/21 patients (33.3%) and accounted for all four deaths, whereas zero deaths occurred among the 14 patients without organ dysfunction (Fisher's exact p = 0.006; OR 37.33, 95% CI 1.60-866.90). All nine qSOFA-positive patients without SOFA dysfunction (42.9%) and five patients without organ dysfunction (23.8%) survived. FGSI showed the highest discrimination (AUC 0.934, 95% CI 0.76-1.00), followed by SFGSI (AUC 0.919, 95% CI 0.78-1.00) and LRINEC (AUC 0.860, 95% CI 0.68-0.99); NLR showed poor discrimination (AUC 0.574, 95% CI 0.13-1.00). Metastatic solid malignancy was present in 2/4 non-survivors versus 0/17 survivors (Fisher's exact p = 0.029; OR 35.00, 95% CI 1.27-961.40). All four non-survivors met Sepsis-3 criteria (two with and two without metastases). Seven patients required ICU admission (all four non-survivors and three survivors). Conclusions: Admission Sepsis-3-defined organ dysfunction was strongly associated with in-hospital mortality. FGSI and SFGSI demonstrated the highest point-estimate discrimination, LRINEC also showed good discrimination, and NLR demonstrated limited standalone discriminative value. Metastatic malignancy occurred exclusively among non-survivors but overlapped with Sepsis-3-defined sepsis. These findings support the combined assessment of Sepsis-3-defined organ dysfunction and established severity scores for early risk stratification, warranting validation in larger cohorts.