1Division of Acute Care Surgery, Department of Surgery, Yonsei University Wonju College of Medicine, Wonju, Korea
2Department of Surgery, Soonchunhyang University Seoul Hospital, Seoul, Korea
3Division of Trauma and Critical Care Surgery, Department of Surgery, Keimyung University Dongsan Hospital, Keimyung University School of Medicine, Daegu, Korea
4Department of Surgery, Yongin Severance Hospital, Yonsei University College of Medicine, Yongin, Korea
5Health Check-Up Center, Wonju Severance Christian Hospital, Wonju, Korea
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Fournier gangrene (FG) is a life-threatening condition characterized by rapid tissue necrosis in the perineal and genital regions. Early identification of risk factors is essential for improving prognosis and reducing mortality rates. This study aimed to identify predictive factors associated with mortality in patients with FG.
Methods
A multi-institutional retrospective study was conducted across four tertiary care centers in Korea, including patients diagnosed with FG between January 2014 and December 2023. Data on demographic characteristics, laboratory findings, and clinical outcomes were collected. Independent risk factors for mortality were identified using multivariate logistic regression analysis, and optimal cutoff values for these predictors were determined.
Results
A total of 133 patients were included. The findings showed that age, initial lactate level, hemoglobin level, platelet count, and albumin level were significant predictors of mortality. Age was associated with an odds ratio of 1.105 (95% confidence interval [CI], 1.038–1.177; P=0.002), while initial lactate level exhibited an odds ratio of 1.820 (95% CI, 1.335–2.480; P<0.001). The optimal cutoff values identified were 64.5 years for age, 3.15 mmol/L for lactate, 8.65 g/dL for hemoglobin, 208×109/L for platelet count, and 3.05 g/dL for albumin, with varying sensitivity and specificity.
Conclusion
The study concluded that age, and initial levels of lactate, hemoglobin, albumin, and platelet count are independently predictive of mortality in patients with FG. These findings underscore the importance of aggressive management for patients presenting with abnormal serum values at admission to improve clinical outcomes. Further research is warranted to validate these results and to refine management strategies for FG.
Fournier gangrene (FG) is a severe and potentially fatal condition characterized by rapid necrosis of tissues in the perineal and genital areas, most often resulting from polymicrobial infection [1,2]. First described by the French surgeon Jean-Alfred Fournier in 1883, FG has since attracted significant medical interest because of its high morbidity and mortality rates when not promptly recognized and treated [3]. The pathophysiology of FG involves a complex interplay of local tissue destruction, bacterial invasion, and host immune response. Several risk factors have been identified, including diabetes mellitus, immunocompromised status, and prior trauma; however, FG can also occur in individuals with no predisposing conditions [4,5]. Due to its heterogeneous presentation, effective management typically requires a multidisciplinary approach involving urologists, surgeons, infectious disease specialists, and critical care physicians.
Despite advances in medical and surgical management, FG remains a major clinical challenge due to its rapid progression and potential for systemic complications [6]. Delayed diagnosis or intervention is strongly correlated with unfavorable outcomes [7]. Therefore, early identification of patients at risk for rapid progression or mortality is critical to ensuring timely and appropriate treatment. Numerous studies have proposed biomarkers for predicting adverse outcomes in FG, yet most have been limited by small sample sizes. Our group has previously reported mortality predictors based on single-institution cohorts [8,9].
Objectives
This study aims to identify clinical and laboratory predictors of in-hospital mortality among patients with FG treated at Korean tertiary hospitals. Specifically, it seeks to determine independent factors associated with in-hospital mortality using multivariable logistic regression analysis and to evaluate the discriminative performance of key variables and the multivariable model through receiver operating characteristic (ROC) curve analysis and calculation of the area under the curve (AUC).
METHODS
Ethics statement
This study was approved by the institutional review boards of Wonju Severance Christian Hospital (No. CR324044), Yongin Severance Hospital (No. 9-2024-0080), Keimyung University Dongsan Medical Center (No. DSMC 2024-01-017), and National Medical Center of Korea (No. NMC-2024-02-027). The requirement for informed consent was waived due to the use of deidentified data and the retrospective nature of the study. The study was conducted according to the tenets of the Declaration of Helsinki.
Study design and setting
This was a multi-institutional retrospective study. Electronic medical records of patients with FG treated between January 2014 and December 2023 from four tertiary care centers, including Wonju Severance Christian Hospital (Wonju, Korea), Keimyung University Dongsan Medical Center (Daegu, Korea), Yongin Severance Hospital (Yongin, Korea), and National Medical Center (Seoul, Korea) were searched on January 17, 2024. The medical records and initial computed tomography scans of these patients were retrospectively analyzed. None of the patients included in this study were transferred during treatment or withdrew from care.
Participants
Patients diagnosed with FG or genital and perineal necrotizing fasciitis who underwent debridement and reconstruction of the scrotum, genitalia, or perineum were identified using the International Classification of Diseases, 10th Revision codes. Only those with a confirmed diagnosis of FG were included after review of the medical charts. The diagnosis of FG was primarily based on clinical findings and the following criteria [10]: (1) soft tissue infections involving the scrotum, perineum, or perianal region; (2) presence of air infiltration in subcutaneous tissue identified by physical examination or imaging; (3) surgical evidence of gangrenous or necrotic tissue; and (4) histological confirmation of necrotizing fasciitis. The exclusion criteria comprised patients younger than 18 years of age and pregnant women. In total, 133 patients with FG were included: 82 from Wonju Severance Christian Hospital, 22 from Keimyung University Dongsan Medical Center, 12 from Yongin Severance Hospital, and 17 from the National Medical Center.
FG treatment involves both surgical and medical interventions [11]. Surgical management included radical wide excision of necrotic tissue, irrigation, and drainage. Medical management involved empirical broad-spectrum antibiotic therapy initiated before culture results were available, followed by targeted antibiotic therapy based on sensitivity testing. Initial fluid resuscitation and vasopressor therapy were administered to patients presenting with septic shock.
Variables
The primary outcome variable was in-hospital mortality, defined as death from any cause during the index admission for FG. Patients discharged alive were classified as survivors.
Exposure (predictor) variables were selected a priori based on biological plausibility and prior literature on necrotizing soft-tissue infections and FG. These included the following:
(1) Demographic factors: age (years, continuous; additionally explored with a data-driven cutoff) and sex.
(2) Anthropometric factor: body mass index (BMI; kg/m2).
(3) Comorbid conditions: hypertension, diabetes mellitus, liver disease, renal disease, and pulmonary disease (all coded as yes/no).
(4) Initial clinical status at presentation: systolic and diastolic blood pressure, body temperature, pulse rate, presence of septic shock, and presence of acute kidney injury (AKI), each defined in the “Data sources/measurement” section. Septic shock and AKI were treated as binary variables.
(5) Initial laboratory variables: lactate; white blood cell, neutrophil, and lymphocyte counts; hemoglobin; hematocrit; platelet count; albumin; international normalized ratio (INR); C-reactive protein (CRP); procalcitonin; creatinine; glucose; and bicarbonate (HCO3–). For each patient, the first laboratory values obtained in the emergency department or at admission were used for analysis.
(6) Disease severity and nutrition-related indices: Fournier’s Gangrene Severity Index (FGSI), Uludag FGSI (UFGSI), Laboratory Risk Indicator for Necrotizing Fasciitis (LRINEC), prognostic nutritional index (PNI), and albumin to globulin ratio (AGR), all calculated as described in the protocol.
Data sources/measurement
Medical charts were retrospectively reviewed. Demographic variables of interest included patient age and sex. Clinical variables included BMI, underlying diseases, blood pressure, body temperature, pulse rate, presence of septic shock or AKI at initial presentation, initial laboratory findings (lactate level; white blood cell, neutrophil, lymphocyte, hemoglobin, hematocrit, and platelet counts; albumin level; INR; CRP, procalcitonin, creatinine, glucose, and HCO3– levels), hospital length of stay, intensive care unit (ICU) length of stay, and overall mortality. Initial laboratory results were obtained from the first tests performed at the time of patient presentation, regardless of the route of admission. The FGSI and UFGSI scores were calculated. FGSI is a numerical index derived from clinical and laboratory parameters, including body temperature, heart rate, respiratory rate, serum electrolytes, creatinine, and hematocrit levels. UFGSI is an extension of FGSI that additionally incorporates patient age and the extent of disease involvement [12]. The LRINEC score was also calculated. LRINEC is a diagnostic scoring tool used to differentiate necrotizing fasciitis, a severe soft tissue infection, from less aggressive infections such as cellulitis [13].
The PNI is a scoring system used to evaluate a patient’s nutritional and inflammatory status and to predict clinical prognosis. It is calculated from serum albumin concentration and total lymphocyte count obtained from blood tests [14]. The AGR is another laboratory parameter that compares the concentrations of two major serum protein groups, albumin and globulin, providing information about liver function, nutritional condition, and immune system activity [15]. The above indices were analyzed collectively, as they were expected to correlate with patients’ nutritional status and overall prognosis.
AKI was defined by any of the following criteria [16]: (1) an increase in serum creatinine ≥0.3 mg/dL within 48 hours; (2) an increase in serum creatinine ≥1.5 times the baseline level, known or presumed to have occurred within the previous 7 days; or (3) urine output <0.5 mL/kg/hour for 6 consecutive hours.
Septic shock was defined as a subset of sepsis characterized by the need for vasopressor therapy to maintain a mean arterial pressure ≥65 mmHg and a serum lactate concentration >2 mmol/L in the absence of hypovolemia [17].
Bias
To minimize selection bias, all consecutive patients diagnosed with FG during the study period at the four tertiary hospitals were included, using consistent diagnostic definitions and discharge codes. Readmissions for the same disease episode were not considered new cases. To reduce measurement bias, only laboratory results obtained at the time of first presentation or admission were used as “initial” values, and all variables were abstracted from electronic medical records using a predefined case report form. The primary outcome, in-hospital mortality, was selected because it is an objective measure that is unlikely to be misclassified. To limit confounding, clinically plausible predictors, including age, lactate, hemoglobin, platelet count, albumin, and major comorbidities, were prespecified and included in the multivariable logistic regression model.
Study size
No formal sample size estimation was performed, as all eligible patients meeting the inclusion criteria during the study period were enrolled.
Statistical analysis
Continuous variables were tested for normality using the Shapiro-Wilk test. Normally distributed variables were expressed as the mean±standard deviation and compared between survivor and nonsurvivor groups using the Student t-test. Non-normally distributed variables were expressed as the median (interquartile range [IQR] and compared using the Mann-Whitney U-test. Categorical variables were analyzed using the chi-square test or Fisher exact test, as appropriate. To compare clinical characteristics among the four participating centers, continuous variables were analyzed using one-way analysis of variance or the Kruskal-Wallis test, depending on the data distribution and homogeneity of variance. Categorical variables were compared using the chi-square test or Fisher exact test, as applicable. Multivariate analysis was performed using logistic regression to identify independent risk factors for mortality. Results were expressed as odds ratios (ORs) with corresponding 95% confidence intervals (CIs). Missing values were addressed using mean imputation during multivariate analysis. Variables showing significant differences between survivor and nonsurvivor groups, along with those deemed clinically relevant by the investigators, were entered into the multivariable logistic regression model. Model selection was based on the Akaike information criterion (AIC). The final model, which yielded the lowest AIC value, included the following variables: age, sex, BMI, liver disease, systolic blood pressure, body temperature, lactate, hemoglobin, hematocrit, platelet count, albumin, INR, creatinine, CRP, bicarbonate, FGSI, and UFGSI. To assess multicollinearity among the independent variables, the variance inflation factor (VIF) was calculated for the final model, and all VIF values were below 5, indicating no significant multicollinearity. An ROC curve was constructed, and the Youden index was applied to determine optimal cutoff values for predicting mortality. All statistical analyses were performed using R ver. 4.1.0 (R Foundation for Statistical Computing). Statistical significance was set at P<0.05.
RESULTS
Participant characteristics
A total of 133 patients were included: 101 in the survivor group and 32 in the nonsurvivor group. The survivor group was younger than the nonsurvivor group (57.7±14.5 years vs. 66.9±13.9 years, P=0.002). Although both groups consisted predominantly of male patients (survivor group: 90 of 101, 89.1%; nonsurvivor group: 23 of 32, 71.9%), mortality was significantly higher among female patients (P=0.036). The mortality rate was also higher among patients with AKI than among those without it (40.0% [32 of 80] vs. 64.3% [18 of 28]). The nonsurvivor group included significantly more patients with liver disease compared with the survivor group (15.8% [16 of 101] vs. 37.5% [12 of 32]). Initial laboratory results revealed significant differences between the two groups. Compared with survivors, nonsurvivors had higher levels of lactate (1.9 mmol/L [IQR, 1.2–2.2 mmol/L] vs. 4.5 mmol/L [IQR, 2.0–7.1 mmol/L], P<0.001), procalcitonin (1.1 ng/mL [IQR, 0.3–6.1 ng/mL] vs. 14.0 ng/mL [IQR, 5.4–23.7 ng/mL], P=0.05), creatinine (1.0 mg/dL [IQR, 0.7–1.6 mg/dL] vs. 2.2 mg/dL [IQR, 1.1–3.6 mg/dL], P=0.015), and INR (1.1 [IQR, 1.1–1.2] vs. 1.3 [IQR, 1.1–1.5], P=0.013), while having lower hemoglobin (11.6±2.6 g/dL vs. 9.9±2.7 g/dL, P=0.002), hematocrit (34.8%±7.2% vs. 30.7%±7.9%, P=0.007), albumin (3.3 g/dL [IQR, 2.6–3.8 g/dL] vs. 2.4 g/dL [IQR, 2.0–2.9 g/dL], P<0.001), platelet count (259×109/L [IQR, 193–343×109/L] vs. 157×109/L [IQR, 62–278×109/L], P<0.001), and HCO3– levels (22.5 mmol/L [IQR, 20.4–25.2 mmol/L] vs. 19.3 mmol/L [IQR, 14.3–23.2 mmol/L], P=0.002). Disease severity indices also differed significantly between groups. The nonsurvivor group demonstrated higher FGSI scores (4 [IQR, 3–8] vs. 8.5 [IQR, 6–12], P<0.001), UFGSI scores (7 [IQR, 4–10] vs. 11.5 [IQR, 9–15.5], P<0.001), and lower PNI scores (37.6 [IQR, 33.4–43.0] vs. 28.9 [IQR, 22.9–32.0], P=0.025). The AGR was also significantly lower in nonsurvivors than in survivors (138.5±48.1 vs. 83.2±32.9, P=0.002). Table 1 summarizes the comparison of baseline and clinical characteristics between the survivor and nonsurvivor groups. Interhospital variations in baseline characteristics and outcomes were also assessed, and the results are presented in Table S1.
Independent predictive factors of mortality and their optimal cutoff values
The following factors were independently associated with mortality: age (OR, 1.105; 95% CI, 1.038–1.177; P=0.002), initial lactate level (OR, 1.820; 95% CI, 1.335–2.480; P<0.001), hemoglobin level (OR, 0.678; 95% CI, 0.492–0.934; P=0.018), platelet count (OR, 0.994; 95% CI, 0.989–0.999; P=0.021), and albumin level (OR, 0.220; 95% CI, 0.076–0.639; P=0.005) (Table 2). The AUC values were 0.681 (95% CI, 0.569–0.792) for age, 0.790 (95% CI, 0.676–0.904) for initial lactate level, 0.668 (95% CI, 0.554–0.781) for hemoglobin, 0.695 (95% CI, 0.585–0.805) for platelet count, and 0.776 (95% CI, 0.690–0.862) for albumin level. The optimal cutoff values were 64.5 years for age (sensitivity, 65.6%; specificity, 67.3%), 3.15 mmol/L for lactate (sensitivity, 68.8%; specificity, 92.1%), 8.65 g/dL for hemoglobin (sensitivity, 40.6%; specificity, 92.1%), 208×109/L for platelet count (sensitivity, 65.6%; specificity, 70.3%), and 3.05 g/dL for albumin (sensitivity, 87.5%; specificity, 60.4%). Detailed results are presented in Table 3 and Fig. 1.
DISCUSSION
Key results
This multicenter study of Korean patients with FG demonstrated that older age, elevated initial serum lactate, and reduced initial hemoglobin, platelet, and albumin levels were independently associated with in-hospital mortality. Among these factors, lactate showed the highest specificity and AUC, whereas albumin exhibited the highest sensitivity and AUC, suggesting complementary predictive roles. In contrast, commonly used FG severity indices, including FGSI and UFGSI, were not significantly associated with mortality in this cohort, despite previous reports of strong predictive performance. These findings suggest that simple, routinely available admission variables may outperform traditional composite scores in contemporary, heterogeneous FG populations.
Interpretation
These results reinforce the concept that death from FG is largely a consequence of sepsis, tissue hypoperfusion, and limited physiological reserve. Elevated lactate at admission likely reflects early circulatory failure and anaerobic metabolism, which are central features of severe sepsis and necrotizing soft-tissue infection [18–20]. Low albumin, hemoglobin, and platelet counts point to baseline frailty, chronic comorbidity, or malnutrition, all of which decrease tolerance to septic insult and extensive debridement [21–23]. Thus, patients who are older, hemodynamically/metabolically compromised (high lactate), and nutritionally or hematologically depleted (low albumin, hemoglobin, platelets) constitute a high-risk subgroup requiring earlier source control, faster hemodynamic optimization, and active nutritional support. The lack of significance of FGSI and UFGSI in this study may indicate that scores built on older intensive-care practices do not fully capture present-day management pathways (early antibiotics, damage-control debridement, rapid imaging, and organ-support bundles) and that their components may not be weighted appropriately for patients with profound sepsis or malnutrition in FG.
Comparison with previous studies
Our findings on age are in line with several studies reporting an age-related increase in FG mortality [24–26] and contradict studies in which age was not an independent predictor [8,9]. The discrepancies can reasonably be explained by differences in setting (ICU-only cohorts [9]), sample size, covariate selection (e.g. inclusion of delta neutrophil index and other sepsis markers [8]), and single-center vs multicenter design. Because the present study covers multiple Korean tertiary hospitals and a broader catchment, it likely provides a more generalizable estimate of the effect of age.
Similarly, the strong association of lactate with mortality parallels evidence from sepsis cohorts and necrotizing infections [18–20,27], confirming that lactate should be regarded as an early triage variable in FG as well. Our observation that low albumin and thrombocytopenia accompany poor outcomes is also consistent with reports linking malnutrition, systemic inflammation, and sepsis-related coagulopathy to mortality [21–23,28].
In contrast, we did not replicate the good performance of FGSI, UFGSI, or other composite scores reported in smaller series and in studies that used broader prognostic panels (e.g., Charlson Comorbidity Index, quick Sequential Organ Failure Assessment [qSOFA], LRINEC) [12,29,30]. Differences in the time of scoring, clinical environment, and variable composition (e.g., inclusion of bacterial growth, number of comorbidities, or novel inflammatory indices) probably contributed to this gap. This suggests that existing scores may require recalibration for contemporary multicenter practice.
Suggestions for further studies
First, future work should aim to develop and externally validate an FG-specific mortality model that combines easily obtainable admission variables (age, lactate, albumin, hemoglobin, platelet) with sepsis- or nutrition-related markers that showed promise in single-center studies (e.g. delta neutrophil index, INR) [8,9], using prospective, multicenter data. Second, because hypoalbuminemia and anemia appear to be more than innocent bystanders, interventional studies should test whether early nutritional optimization, albumin correction, and transfusion guided by sepsis-bundle principles can improve outcomes in high-risk FG patients [23,28]. Third, updated scoring systems should be reconstructed or recalibrated to reflect modern intensive-care management (early source control, vasopressors, organ support) and then compared head-to-head with FGSI/UFGSI in large datasets. Finally, subgroup analyses (by age, extent of fascial involvement, or initial lactate stratum) would clarify which patients benefit most from aggressive early resuscitation and combined surgical approaches.
Limitations
This study was initially designed as a multicenter investigation to obtain a larger dataset for FG; however, the number of eligible patients treated at other institutions was significantly lower than that at our main center. Moreover, the inclusion of additional variables was limited by differences in facility capabilities and laboratory parameters across institutions. The absence of a standardized treatment protocol likely led to minor variations in diagnostic and therapeutic approaches, which may have affected statistical consistency. Furthermore, although all hospitals were located in Korea, regional differences in age distribution, socioeconomic status, healthcare accessibility, and health awareness may have influenced disease prevalence and treatment experience. Another limitation is the use of mean imputation for missing data, which could underestimate variance and bias associations. Despite these limitations, this work remains significant as, to the best of our knowledge, it represents the first multicenter study of FG in Korea. Larger multicenter investigations are warranted to further identify high-risk FG patients and validate predictive models for mortality.
Conclusions
Age, initial lactate, hemoglobin, albumin, and platelet levels independently predict mortality in patients with FG. Among these, serum albumin showed the highest sensitivity, while lactate and hemoglobin demonstrated the highest specificity. Age and platelet count exhibited moderate predictive performance. These findings suggest that more aggressive management should be considered for patients presenting with high lactate levels and low albumin, hemoglobin, or platelet counts upon admission.
ARTICLE INFORMATION
Author contributions
Conceptualization: all authors; Data curation: ISS, SWJ, CHP, JWL, HJB; Formal analysis: ISS, KK; Funding acquisition: ISS; Methodology: all authors; Supervision: KK; Writing–original draft: ISS, KK; Writing–review & editing: all authors. All authors read and approved the final manuscript.
Conflicts of interest
The authors have no conflicts of interest to declare.
Funding
This study was funded by the Korean Society of Acute Care Surgery / Korean Society of Trauma & Acute Care Nursing.
Acknowledgments
The authors sincerely thank the members of the Korean Society of Acute Care Surgery / Korean Society of Trauma & Acute Care Nursing for selecting this topic for our research.
Data availability
Data analyzed in this study are available from the corresponding author upon reasonable request.
Receiver operating characteristics curves for age, initial levels of lactate, hemoglobin, albumin, and platelet count as predictors of mortality in patients with Fournier gangrene. Lactate (area under the curve [AUC], 0.790; cutoff, 3.15 mmol/L; sensitivity, 68.8%; specificity, 92.1%) and albumin (AUC, 0.776; cutoff, 3.05 g/dL; sensitivity, 87.5%; specificity, 60.4%) showed the highest discriminatory power. The AUCs for other predictors were as follows: platelet (AUC, 0.695; cutoff, 208×109/L), age (AUC, 0.681; cutoff, 64.5 years), and hemoglobin (AUC, 0.668; cutoff, 8.65 g/dL).
Table 1.
Comparison of patient characteristics between survivors and nonsurvivors (n=133)
Variable
Survivor (n=101)
Nonsurvivor (n=32)
P-value
Age (yr)
57.7±14.5
66.9±13.9
0.002
Sex
0.036
Male
90 (89.1)
23 (71.9)
Female
11 (10.9)
9 (28.1)
Body mass index (kg/m2)
23.2 (21.2–26.0)
23.1 (20.0–24.7)
0.067
Underlying disease
Hypertension
50 (49.5)
13 (40.6)
0.501
Diabetes mellitus
55 (54.5)
14 (43.8)
0.394
Liver disease
16 (15.8)
12 (37.5)
0.018
Renal disease
15 (14.9)
2 (6.2)
0.334
Pulmonary disease
12 (11.9)
5 (15.6)
0.803
Systolic blood pressure (mmHg)
120 (107–143)
100 (85.5–113)
0.001
Diastolic blood pressure (mmHg)
73 (60–82)
56 (48–65.5)
0.064
Body temperature (℃)
37.2 (36.6–38.0)
36.9 (36.1–37.3)
0.009
Pulse rate (bpm)
98 (84–111)
97 (87–114.5)
0.695
Septic shock
15 (14.9)
12 (37.5)
0.012
Acute kidney injury
32 (40.0)
18 (64.3)
0.046
Hospital LOS (day)
34 (18–49)
11 (3–41.5)
0.110
ICU LOS (day)
0 (0–4)
5 (1.5–10)
0.108
Initial laboratory finding
Lactate (mmol/L)
1.9 (1.2–2.2)
4.5 (2.0–7.1)
<0.001
WBC count (×109/L)
15.20 (11.18–20.60)
10.71 (47.25–18.78)
0.072
Neutrophil count (×109/L)
13.41 (90.70–18.05)
94.10 (38.70–15.24)
0.086
Lymphocyte count (×109/L)
1.03 (0.64–1.30)
5.85 (3.25–8.85)
0.998
Hemoglobin (g/dL)
11.6±2.6
9.9±2.7
0.002
Hematocrit (%)
34.8±7.2
30.7±7.9
0.007
Platelet count (×109/L)
259 (193–343)
157 (62–278)
<0.001
Albumin level (g/dL)
3.3 (2.6–3.8)
2.4 (2.0–2.9)
<0.001
INR
1.1 (1.1–1.2)
1.3 (1.1–1.5)
0.013
Procalcitonin level (ng/mL)
1.1 (0.3–6.1)
14.0 (5.4–23.7)
0.050
CRP level (mg/dL)
17.5 (9.9–24.1)
21.1 (7.9–29.1)
0.332
Creatinine level (mg/dL)
1.0 (0.7–1.6)
2.2 (1.1–3.6)
0.015
Glucose level (mg/dL)
144 (107–240.5)
139.5 (101.5–216.5)
0.083
HCO3– level (mmol/L)
22.5 (20.4–25.2)
19.3 (14.3–23.2)
0.002
Disease severity and nutrition-related index
FGSI score
4 (3–8)
8.5 (6–12)
<0.001
UFGSI score
7 (4–10)
11.5 (9–15.5)
<0.001
LRINEC score
5.9±3.1
6.9±3.0
0.089
PNI score
37.6 (33.4–43.0)
28.9 (22.9–32.0)
0.025
AGR
138.5±48.1
83.2±32.9
0.002
Values are presented as mean±standard deviation, number (%), or median (interquartile range).
bpm, beats per minute; LOS, length of stay; ICU, intensive care unit; WBC, white blood cell; INR, international normalized ratio; CRP, C-reactive protein; FGSI, Fournier's Gangrene Severity Index; UFGSI, Uludag Fournier's Gangrene Severity Index; LRINEC, Laboratory Risk Indicator for Necrotizing Fasciitis; PNI, prognostic nutritional index; AGR, albumin to globulin ratio.
Table 2.
Multivariate analysis of the predictors of mortality in Fournier gangrene
Variable
OR (95% CI)
P-value
Age (yr)
1.105 (1.038–1.177)
0.002
Lactate (mmol/L)
1.820 (1.335–2.480)
<0.001
Hemoglobin (g/dL)
0.678 (0.492–0.934)
0.018
Platelet count (×109/L)
0.994 (0.989–0.999)
0.021
Albumin level (g/dL)
0.220 (0.076–0.639)
0.005
OR, odds ratio; CI, confidence interval.
Table 3.
Performance of the predictive factors of mortality in Fournier gangrene
Variable
Optimal cutoff value
Sensitivity (%)
Specificity (%)
AUC (95% CI)
Age (yr)
64.5
65.6
67.3
0.681 (0.569–0.792)
Initial lactate (mmol/L)
3.15
68.8
92.1
0.790 (0.676–0.904)
Initial hemoglobin (g/dL)
8.65
40.6
92.1
0.668 (0.554–0.781)
Initial platelet count (×109/L)
208
65.6
70.3
0.695 (0.585–0.805)
Initial albumin level (g/dL)
3.05
87.5
60.4
0.776 (0.690–0.862)
AUC, area under the curve; CI, confidence interval.
REFERENCES
1. Ruiz-Tovar J, Córdoba L, Devesa JM. Prognostic factors in Fournier gangrene. Asian J Surg 2012;35:37–41.
2. Orhan E, Şenen D. Using negative pressure therapy for improving skin graft taking on genital area defects following Fournier gangrene. Turk J Urol 2017;43:366–70.
3. Sorensen MD, Krieger JN, Rivara FP, Klein MB, Wessells H. Fournier's gangrene: management and mortality predictors in a population based study. J Urol 2009;182:2742–7.
5. Belinchón-Romer I, Ramos-Belinchón A, Lobato-Martínez E, Sánchez-García V, Ramos-Rincón JM. National study of Fournier gangrene in Spain (2016-2021): gender/sex differences in mortality and risks. Medicina (Kaunas) 2024;60:1600.
6. Cochetti G, Paladini A, Lepri L, et al. Enhanced patient recovery with early extensive surgical deb-ridement in Fournier's Gangrene: evaluation of perioperative outcomes in a multicentric experience. Arch Ital Urol Androl 2025;97:13207.
7. Birben B, Akkurt G, Akın T, Surel AA, Tez M. Predictive efficacy of delta neutrophil index in diagnosis of acute and complicated appendicitis. Cureus 2021;13:e14748.
8. Shin IS, Gong SC, An S, Kim K. Delta neutrophil index as a prognostic factor for mortality in patients with Fournier's gangrene. Int J Urol 2022;29:1287–93.
9. Shin IS, Gong SC, An S, Kim K. Biomarkers to predict mortality in patients with Fournier's gangrene admitted to the intensive care unit after surgery in South Korea. Acute Crit Care 2023;38:452–9.
10. Kuo CF, Wang WS, Lee CM, Liu CP, Tseng HK. Fournier's gangrene: ten-year experience in a medical center in northern Taiwan. J Microbiol Immunol Infect 2007;40:500–6.
11. Noegroho BS, Siregar S, Mustafa A, Rivaldi MA. Validation of FGSI scores in predicting Fournier gangrene in tertiary hospital. Res Rep Urol 2021;13:341–6.
12. Üreyen O, Acar A, Gökçelli U, Atahan MK, İlhan E. Usefulness of FGSI and UFGSI scoring systems for predicting mortality in patients with Fournier's gangrene: a multicenter study. Ulus Travma Acil Cerrahi Derg 2017;23:389–94.
13. Wong CH, Khin LW, Heng KS, Tan KC, Low CO. The LRINEC (Laboratory Risk Indicator for Necrotizing Fasciitis) score: a tool for distinguishing necrotizing fasciitis from other soft tissue infections. Crit Care Med 2004;32:1535–41.
14. Pinato DJ, North BV, Sharma R. A novel, externally validated inflammation-based prognostic algorithm in hepatocellular carcinoma: the prognostic nutritional index (PNI). Br J Cancer 2012;106:1439–45.
15. Lv GY, An L, Sun XD, Hu YL, Sun DW. Pretreatment albumin to globulin ratio can serve as a prognostic marker in human cancers: a meta-analysis. Clin Chim Acta 2018;476:81–91.
17. Shankar-Hari M, Phillips GS, Levy ML, et al. Developing a new definition and assessing new clinical criteria for septic shock: for the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016;315:775–87.
19. Filho RR, Rocha LL, Corrêa TD, Pessoa CM, Colombo G, Assuncao MS. Blood lactate levels cutoff and mortality prediction in sepsis-time for a reappraisal? A Retrospective Cohort Study. Shock 2016;46:480–5.
20. Varis E, Pettilä V, Poukkanen M, et al. Evolution of blood lactate and 90-day mortality in septic shock. A post hoc analysis of the FINNAKI Study. Shock 2017;47:574–81.
22. Venkata C, Kashyap R, Farmer JC, Afessa B. Thrombocytopenia in adult patients with sepsis: incidence, risk factors, and its association with clinical outcome. J Intensive Care 2013;1:9.
23. Zengin A, Alagas G, Bag YM, et al. Biomarkers to predict 30-day mortality in patients with Fournier's gangrene disease: a retrospective study. ANZ J Surg 2025;95:433–9.
28. Saravi B, Goebel U, Hassenzahl LO, et al. Capillary leak and endothelial permeability in critically ill patients: a current overview. Intensive Care Med Exp 2023;11:96.
29. Tufano A, Dipinto P, Passaro F, et al. The value of Fournier's gangrene scoring systems on admission to predict mortality: a systematic review and meta-analysis. J Pers Med 2023;13:1283.
30. Arıkan Y, Emir B, Tarhan O, et al. Comparative analysis of scoring systems for predicting mortality in Fournier gangrene: single center, 15 years experience. Updates Surg 2024;76:2683–92.
Predictors of in-hospital mortality in Fournier gangrene at four Korean tertiary hospitals: a multicenter retrospective cohort study
Fig. 1. Receiver operating characteristics curves for age, initial levels of lactate, hemoglobin, albumin, and platelet count as predictors of mortality in patients with Fournier gangrene. Lactate (area under the curve [AUC], 0.790; cutoff, 3.15 mmol/L; sensitivity, 68.8%; specificity, 92.1%) and albumin (AUC, 0.776; cutoff, 3.05 g/dL; sensitivity, 87.5%; specificity, 60.4%) showed the highest discriminatory power. The AUCs for other predictors were as follows: platelet (AUC, 0.695; cutoff, 208×109/L), age (AUC, 0.681; cutoff, 64.5 years), and hemoglobin (AUC, 0.668; cutoff, 8.65 g/dL).
Graphical abstract
Fig. 1.
Graphical abstract
Predictors of in-hospital mortality in Fournier gangrene at four Korean tertiary hospitals: a multicenter retrospective cohort study
Variable
Survivor (n=101)
Nonsurvivor (n=32)
P-value
Age (yr)
57.7±14.5
66.9±13.9
0.002
Sex
0.036
Male
90 (89.1)
23 (71.9)
Female
11 (10.9)
9 (28.1)
Body mass index (kg/m2)
23.2 (21.2–26.0)
23.1 (20.0–24.7)
0.067
Underlying disease
Hypertension
50 (49.5)
13 (40.6)
0.501
Diabetes mellitus
55 (54.5)
14 (43.8)
0.394
Liver disease
16 (15.8)
12 (37.5)
0.018
Renal disease
15 (14.9)
2 (6.2)
0.334
Pulmonary disease
12 (11.9)
5 (15.6)
0.803
Systolic blood pressure (mmHg)
120 (107–143)
100 (85.5–113)
0.001
Diastolic blood pressure (mmHg)
73 (60–82)
56 (48–65.5)
0.064
Body temperature (℃)
37.2 (36.6–38.0)
36.9 (36.1–37.3)
0.009
Pulse rate (bpm)
98 (84–111)
97 (87–114.5)
0.695
Septic shock
15 (14.9)
12 (37.5)
0.012
Acute kidney injury
32 (40.0)
18 (64.3)
0.046
Hospital LOS (day)
34 (18–49)
11 (3–41.5)
0.110
ICU LOS (day)
0 (0–4)
5 (1.5–10)
0.108
Initial laboratory finding
Lactate (mmol/L)
1.9 (1.2–2.2)
4.5 (2.0–7.1)
<0.001
WBC count (×109/L)
15.20 (11.18–20.60)
10.71 (47.25–18.78)
0.072
Neutrophil count (×109/L)
13.41 (90.70–18.05)
94.10 (38.70–15.24)
0.086
Lymphocyte count (×109/L)
1.03 (0.64–1.30)
5.85 (3.25–8.85)
0.998
Hemoglobin (g/dL)
11.6±2.6
9.9±2.7
0.002
Hematocrit (%)
34.8±7.2
30.7±7.9
0.007
Platelet count (×109/L)
259 (193–343)
157 (62–278)
<0.001
Albumin level (g/dL)
3.3 (2.6–3.8)
2.4 (2.0–2.9)
<0.001
INR
1.1 (1.1–1.2)
1.3 (1.1–1.5)
0.013
Procalcitonin level (ng/mL)
1.1 (0.3–6.1)
14.0 (5.4–23.7)
0.050
CRP level (mg/dL)
17.5 (9.9–24.1)
21.1 (7.9–29.1)
0.332
Creatinine level (mg/dL)
1.0 (0.7–1.6)
2.2 (1.1–3.6)
0.015
Glucose level (mg/dL)
144 (107–240.5)
139.5 (101.5–216.5)
0.083
HCO3– level (mmol/L)
22.5 (20.4–25.2)
19.3 (14.3–23.2)
0.002
Disease severity and nutrition-related index
FGSI score
4 (3–8)
8.5 (6–12)
<0.001
UFGSI score
7 (4–10)
11.5 (9–15.5)
<0.001
LRINEC score
5.9±3.1
6.9±3.0
0.089
PNI score
37.6 (33.4–43.0)
28.9 (22.9–32.0)
0.025
AGR
138.5±48.1
83.2±32.9
0.002
Variable
OR (95% CI)
P-value
Age (yr)
1.105 (1.038–1.177)
0.002
Lactate (mmol/L)
1.820 (1.335–2.480)
<0.001
Hemoglobin (g/dL)
0.678 (0.492–0.934)
0.018
Platelet count (×109/L)
0.994 (0.989–0.999)
0.021
Albumin level (g/dL)
0.220 (0.076–0.639)
0.005
Variable
Optimal cutoff value
Sensitivity (%)
Specificity (%)
AUC (95% CI)
Age (yr)
64.5
65.6
67.3
0.681 (0.569–0.792)
Initial lactate (mmol/L)
3.15
68.8
92.1
0.790 (0.676–0.904)
Initial hemoglobin (g/dL)
8.65
40.6
92.1
0.668 (0.554–0.781)
Initial platelet count (×109/L)
208
65.6
70.3
0.695 (0.585–0.805)
Initial albumin level (g/dL)
3.05
87.5
60.4
0.776 (0.690–0.862)
Table 1. Comparison of patient characteristics between survivors and nonsurvivors (n=133)
Values are presented as mean±standard deviation, number (%), or median (interquartile range).
bpm, beats per minute; LOS, length of stay; ICU, intensive care unit; WBC, white blood cell; INR, international normalized ratio; CRP, C-reactive protein; FGSI, Fournier's Gangrene Severity Index; UFGSI, Uludag Fournier's Gangrene Severity Index; LRINEC, Laboratory Risk Indicator for Necrotizing Fasciitis; PNI, prognostic nutritional index; AGR, albumin to globulin ratio.
Table 2. Multivariate analysis of the predictors of mortality in Fournier gangrene
OR, odds ratio; CI, confidence interval.
Table 3. Performance of the predictive factors of mortality in Fournier gangrene
AUC, area under the curve; CI, confidence interval.
E-SUBMISSION
