Fournier's gangrene: Evaluation of 68 patients and analysis of prognostic variablesHR Unalp1, E Kamer1, H Derici1, K Atahan1, U Balci2, C Demirdoven3, O Nazli1, MA Onal1
1 General Surgery Clinic, Ataturk Training and Research Hospital, Izmir, Turkey
2 Urology Clinic, Ataturk Training and Research Hospital, Izmir, Turkey
3 Plastic and Reconstructive Surgery Clinic, Ataturk Training and Research Hospital, Izmir, Turkey
Context: Fournier's gangrene (FG) is a rapidly progressing acute gangrenous infection of the anorectal and urogenital area. Aims: The objectives of this study were to investigate patients with FG and to determine risk factors that affect mortality. Settings and Design: Retrospective clinical study. Materials and Methods: Clinical presentations and outcomes of surgical treatments were evaluated in 68 patients with FG. Statistical Analysis Used: Chi-square, Student's t -test, and logistic regression test. Results: Mean age of patients was 54 and female-to-male ratio was 9:59. Among the predisposing factors, diabetes mellitus (DM) was the most common ( n =24, 35.3%), and sepsis on admission was detected in 31 (45.6%) and 15 (22.1%) patients, respectively. Seven (10.3%) patients died. Using logistic regression test, Fournier's Gangrene Severity Index (FGSI) >9, DM and sepsis on admission were found as prognostic factors. Conclusions: FG has a high mortality rate, especially in patients with DM and sepsis. An FGSI value >9 indicates high mortality rate.
Keywords: Diabetes mellitus, Fournier′s gangrene, infection, mortality, sepsis
Fournier's gangrene (FG) is a necrotizing infection caused by synergistic aerobic and anaerobic microorganisms with fulminant prognosis, which initiates from the anorectal or urogenital area and progresses towards the thighs and abdominal wall through fascial plains. , Most patients suffer from one or more predisposing factors such as diabetes mellitus (DM), alcoholism, malignancies, liver and kidney diseases. ,, The success of therapy depends mainly on early diagnosis, aggressive radical surgical debridement(s), and appropriate medication. However, the reported mortality rates in FG patients (ranging 3-67%) are high despite early diagnosis and aggressive treatment. ,,,,
In this study, we aimed to investigate patients with FG and to identify risk factors that affect mortality.
In this retrospective study, demographics, clinical and laboratory findings, surgical treatments, and their outcomes for 68 FG patients who received treatment between 01.01.2000 and 01.05.2007 were recorded by scanning the archives. Patient age, gender, infection source, existence of predisposing factors, duration between symptom onset and hospital admission, clinical findings, laboratory results, number of surgical debridements, whether or not colostomy was opened, and Fournier Gangrene Severity Index (FGSI) were used as individual variables to estimate the severity of FG; the effects of these factors on mortality were also investigated.
Univariate analyses (Chi-square and Student's t -test) were used for comparisons. Logistic regression tests were used in variables, which showed significant differences in univariate analyses and, hence, prognostic factors were determined. A P -value of <0.05 was considered statistically significant.
Mean age of 68 patients included in the study was 54.7±15.6 (range 25-92); 59 (86.8%) of these patients were males. Anorectal and urogenital regions were the source of infection in 41 (60.3%) and 27 (39.7%) patients, respectively. Anorectal abscess ( n =32, 78%) or primary scrotal abscess ( n =19, 70.4%) was the most common cause in FG originating from the anorectal or urogenital region, respectively. One or more predisposing factors such as DM ( n =24, 35.3%), malignancy ( n =7, 10.3%), chronic alcoholism ( n =3, 4.4%), hemiplegia ( n =2, 2.9%), chronic renal failure ( n =2, 2.9%) or immune failure ( n =1, 1.5%) were detected in 31 (45.6%) patients [Table - 1]. The most common symptom was localized pain ( n =60, 88.2%). Sepsis existed in 15 (22.1%) cases on admission [Table - 2]. Mean duration between symptom onset and hospital admission was 3±16 (range 1-8) days.
Seven patients (10.3%) with FG died; five out of the seven (71.4%) were males ( P =0.230). Mean age of patients who died was 66.7±5.6 years, while mean age of survivors was 53.3±15.8 years ( P =0.095). Mortality was observed in two (7.4%) and five (12.2%) patients with infection foci of anorectal and genitourinary regions, respectively ( P =0.420) [Table - 3]. Mortality rate was significantly higher in patients with predisposing factors ( P =0.003) [Table - 4].
Diabetes mellitus was the most common predisposing factor ( n =24, 35.3%, P <0.0001). No significant differences were found between diabetic and non-diabetic patients in terms of source of infection ( P =0.054) and duration between symptom onset and hospital admission ( P =0.138). All patients who died were diabetic ( P =0.000) and the mortality rate was 29.1% in diabetic patients. Higher rates of FGSI>9 and sepsis on admission were detected in diabetics compared with non-diabetic patients ( P =0.007 and 0.006, respectively) [Table - 5].
Mean duration between symptom onset and hospital admission was significantly longer in patients who died (5.3±1.7 days, range 3-8) compared with that in patients who survived (2.9±2.2 days, range 1-8) ( P =0.047). None of the patients who were admitted to the hospital within the first 48h of symptom onset died ( P =0.014).
Mean number of debridements was 2±0.82 (range 1-4). Mean number of debridements in patients who died and in those who survived was 2±0.8 (range 1-4) and 1.9±0.8 (range 1-4), respectively ( P =0.689). Diverting colostomy was performed in 22 (32.4%) patients. Mortality rate in patients with or without colostomy was 4.5% ( n =1) or 13% ( n =6) ( P =0.268).
Mean FGSI for all patients was 5.2±3.2 (range 1-16). Mean FGSI for patients who died or survived was 11.7±2.9 (range7-16) or 4.4±2.2 (range 1-12), respectively ( P =0.002). Mortality was observed in only one patient (1.7%) with FGSI≤9, while six out of nine (66.7%) patients with FGSI>9 died ( P =0.000).
Univariate analyses revealed high mortality rates in the following conditions: age>60 years ( P =0.013), hospital admission later than 48h of symptom onset ( P =0.007), existence of predisposing factor(s) ( P =0.003), serum albumin concentrations <3.5mg/dl ( P =0.002), DM ( P <0.0001), existence of sepsis on admission ( P =0.000), and FGSI>9 ( P =0.000). Factors affecting mortality in logistic regression test were DM ( P =0.030), existence of sepsis on admission ( P =0.012), and FGSI>9 ( P <0.0001) [Table - 6].
Mean duration of hospitalization was 23.4±11.4 (range 5-54) days. On an average, mortality occurred on day 19.8±16.4 (5-54). We found a mortality rate of 10.3% in patients with FG.
Various prognostic factors, such as advanced age, ,,,, primary anorectal infections,  DM,  delayed treatment, , systemic sepsis on admission, ,,, breadth and deepness of the disease, , opening of diverting colostomy, ,, low hematocrit and albumin levels,  elevated blood urea-nitrogen, ,, high serum creatinine,  elevated alkaline phosphatase  and high FGSI score,  have been reported for FG.
Existence of one or more of such co-morbidities as DM, alcoholism, neurological diseases, malignancy, immunosuppression, and, liver and kidney diseases can affect morbidity and mortality rates in cases with FG. ,, DM is the most common predisposing factor with an incidence of 46-76.9% ,, and mortality has been reported in 36-50% of these patients. , Similarly, 35.3% of all patients in our study had DM, and 29.1% of these patients died. It is also worth noting that all patients who died had DM. On the other hand, some authors reported that DM was not associated with mortality and morbidity. , However, consistent with the report of Korkut et al. ,  we found that DM was a prognostic factor for mortality in patients with FG. Infection in diabetic patients spreads relatively easily leading to systemic sepsis in a shorter period of time. In our study, we have found significantly higher rate of sepsis on admission in patients with DM compared with those without DM. In addition, we have found that like DM, the existence of sepsis on admission is a prognostic factor in FG. In good agreement with this finding, the presence of sepsis on admission has been reported to be associated with high mortality rates in FG. ,,,
None of the patients in this study who were admitted to the hospital within the first 48h of symptom onset died. These data support several previous studies, which concluded that mortality rates in FG could be reduced by early diagnosis (before sepsis onset) and immediate initiation of treatment. ,
Treatment must be initiated immediately after diagnosing FG. , Early and aggressive debridement including all non-viable tissue and administration of multiple wide-spectrum antibiotics chosen for the possible causative agent are the gold standard for decreasing the mortality and morbidity. , Since inefficient debridement may cause progression of gangrene as fast as 2.5cm 2 /h,  damaged and necrotic tissues must be debrided regardless of how big the defect might become until normal and healthy tissues are reached. Debridements must be repeated with the same aggressive approach when necessary. Multiple debridements were required in 70.6% of all cases in our study. Several studies failed to find a direct correlation between number of debridements and rate of mortality. ,
Our data are different from previous studies, which reported that mortality rates were higher in patients with the anorectal region as the source of infection.  We believe that the risk of mortality in FG is principally associated with the aggressive character of the disease, patient demographics, and the response obtained after aggressive treatments, rather than source of the infection.
Fournier Gangrene Severity Index, which was developed by Laor et al. , to assign a numerical score that describes the acuity of the disease is used in patients with FG.  However, only a limited number of studies investigated the association of FGSI with the mortality rate on an adequate number of patients. , Mean FGSI has been reported in the literature as 6.2-6.9, 4.6-8.6, or 11.5-14.4 in all patients, in surviving patients, or in patients who died, respectively. , Mean FGSI in our study was 5.2, 4.4, or 11.7 in all patients, in surviving patients, or in patients who died, respectively. In our study, using the definition of Laor etal. , we found mortality rates of 66.7 and 1.7% in patients with FGSI≥9 and FGSI<9, respectively. Multivariate analysis revealed that FGSI≥9 was a prognostic factor in patients with FG in our study. We believe, depending on the present data, that FGSI is a useful scoring system, which can predict both the severity of the disease and the mortality risk of patients, as was concluded by Lin et al .  FGSI may be helpful in predicting those patients who need extra aggressive initial debridement and intensive care therapy.
In conclusion, FG is a rapidly progressing fulminant disease, which is diagnosed mainly from clinical features. Since the infection is distributed secretly and rapidly through fascial plains, clinical findings in early period may be considered the "tip of the iceberg". Although FG is a rare disease, it has a high mortality rate. Early diagnosis, immediate radical surgical debridement(s), use of appropriate antibiotics, and long-term hemodynamic support when necessary are required to reduce FG's high mortality rate. Factors indicating poor prognosis include existence of sepsis on admission and DM. An FGSI value greater than 9 indicates a high mortality rate, which represents the bottom of the iceberg.
[Table - 1], [Table - 2], [Table - 3], [Table - 4], [Table - 5], [Table - 6]