Fournier's Gangrene: Presentation and Predictors of Mortality in Zaria, Nigeria

Abstract

Background:

Fournier's gangrene is an infectious urological emergency with associated morbidity and varying rates of mortality in the world. Various predictors of mortality such as advancing age, Fournier's Gangrene Severity Index (FGSI), anatomical extent of the disease, or presence of risk factors have been studied in the literature, though with conflicting results.

Aim:

The aim of the study was to determine the presentation and predictors of mortality in our environment, Nigeria.

Patients and Methods:

A review of medical records of all the patients managed from April 2012 to December 2018 at a tertiary referral center in Nigeria was conducted. Data on clinical presentation, FGSI, management, and outcome were retrieved and analyzed.

Statistical Analysis:

Descriptive studies using mean and standard deviation were used for continuous variables, Fischer's exact test was used to compare categorical variables among survivors and nonsurvivors, and logistic regression analysis was used to describe the relationships of these variables with mortality.

Results:

The mean age of the 31 patients was 60 ± 12 years. All were men, with 9 (29.0%) patients without clinical evidence of immunosuppression or predisposing factor (idiopathic). Fourteen (45%) had documented evidence of immunosuppression. All the patients had a polymicrobial infection; however, Escherichia coli was the most common organism cultured seen in 26 (83.9%) patients. The initial empirical antibiotic regimen of choice was a combination of intravenous ceftriaxone and metronidazole in 26 (83.8%) patients and intravenous ciprofloxacin and metronidazole in 5 (16.1%) patients. Mortality was recorded in three patients representing a rate of 9.6%. Anatomical extent of the disease, anemia requiring blood transfusion, severity of infection, and FGSI were all found to be the statistically significant variable of mortality in these patients using the Fischer exact test. Furthermore, on regression analysis only the FGSI and blood transfusion were significant with P < 0.05.

Conclusion:

Fournier's gangrene is a disease of the older men with a higher mortality rate when the FGSI is >9 or anemia requiring blood transfusion is present.

INTRODUCTION

Fournier's gangrene as initially described in five male patients by Alfred Fournier in 1883 is a sudden onset, fulminant idiopathic gangrene of the scrotum and penis.[1] Since the description, this clinical entity has been studied and now includes gangrene of the perianal, perineum, and genitalia. It is a disease of middle-aged and elderly men worldwide with a reported male-to-female ratio of 10:1.[2] Although not entirely idiopathic as initially described, it is a synergistic necrotizing polymicrobial infection of the external genitalia from organisms in the anorectal or urogenital systems.[3] Aerobic bacteria cause platelet aggregation, complement fixation, and thrombosis of dermal vessels resulting in decreased oxygen supply, while the microaerophilic bacteria and anaerobes produce collagenase and hyaluronidase that cause tissue destruction and further spread of infection.[4] The infection results in necrotizing fasciitis of the scrotum with variable involvement of the penis and perineal skin. The spread of the infection involves the Colles fascia and, by extension, the Campers fascia. This eventually leads to endarteritis of the superficial branches of the external pudendal and internal pudendal arteries that supply these regions of the body.[1,5] Not surprisingly, the testes supplied by a branch of the aorta, the glans corpus spongiosum, and carvenosum are rarely involved in Fournier's gangrene, while the overlying skin and subcutaneous tissues of the external genitalia slough off from wet gangrene.[6] Immunosuppression from various causes including diabetes and HIV has been implicated as predisposing factors.[7,8,9,10] Diabetes mellitus has been reported as the most common risk factor though not an independent predictor of mortality.[6,11,12]

Although the pathogenesis of the disease has been extensively studied, the mortality rates have been found to vary from the developed to the developing world with a higher rate consisting of 20%–30% in the developed world.[3] However, the principles of management consisting of diagnosis, resuscitation, hemodynamic support, debridement, use of broad-spectrum antibiotics, wound care, and skin cover are the same. Advancing age, background immunosuppression, presence of sepsis on admission, and progression to multiple organ failures have been associated with poor prognosis.[10] Besides, prognostic tools such as the Fournier's Gangrene Severity Index (FGSI), the Laboratory Risk Indicator for Necrotizing Fasciitis (LRINEC), and renal function have also been used to predict the outcomes of management.[13,14] In FGSI, the temperature, heart rate, respiratory rate, serum sodium, potassium, bicarbonate creatinine as well as hematocrit and leukocyte counts are measured at presentation. The deviation from normal is graded on a scale of 0–4. A 75% risk of mortality with an FGSI score of >9 was noted by Laor et al.[15] Subsequent studies on this subject have similar or different results on its predictive value of mortality.[16,17] We hereby review the presentation and outcomes of our patients to determine the mortality rate of Fournier's gangrene in our institution and the factors associated with it.

PATIENTS AND METHODS

A review of medical records of all the patients managed from April 2012 to December 2018 in the division of urology of the hospital for Fournier's gangrene was conducted. Patients with incomplete data were excluded from the study. We obtained approval from our Institutional Health Research Ethics Committee before the commencement of this study.

All the patients presented to the emergency room with swelling and/or ulceration of the external genitalia with/without extension to the adjacent skin. These patients were admitted, clinically evaluated, resuscitated with intravenous fluids broad-spectrum antibiotics, and subsequently had surgical wound debridement in the operating room. The number of debridement depended on the extent of the disease, the most subsequent debridement was done at the bedside as required. Blood transfusion and organ support were given to those presenting with severe anemia or sepsis. Wound care was done using hypertonic saline and povidone-iodine dressings. Skin cover varied from secondary wound closure to skin grafting based on the extent and site of infection. Patients with an identifiable cause of immunosuppression were further evaluated and managed accordingly.

Data on clinical presentation including indices of sepsis, duration of onset before the presentation, predisposing factor, source of infection, microorganism cultured from wound swabs, the extent of involvement of the external genitalia, number of debridement done, urinary or fecal diversions, blood transfusions, and duration of hospital stay were retrieved using a study pro forma. Furthermore, the nine parameters of FGSI (the temperature, heart rate, respiratory rate, serum sodium, potassium, bicarbonate, creatinine, hematocrit, and leukocyte counts) were retrieved and the FGSI score was calculated.

Statistical analysis

Descriptive analysis was used to illustrate the frequencies of variables on clinical presentation and management, Fischer exact test was used for comparison of the categorical variables among survivors and nonsurvivors, while binary logistic regression analysis was used to identify independent predictors of mortality, with P < 0.05 statistically significant. The analysis was done using SPSS version 24.0 (IBM, Armonk, NY, USA).

RESULTS

We identified a total of 31 patients with complete records. All the 31 patients were male, with a mean age of 60 ± 12 years. Nine (29.0%) patients were found without a predisposing factor or evidence of immunosuppression (idiopathic), as shown in Figure 1.

Figure 1.

Distribution of identifiable predisposing factors in patients with Fournier's gangrene

Twenty-five (80.6%) patients presented within the first 2 weeks of the onset of symptoms. Twenty-one (67.7%) had the infection limited only to the scrotum, 5 (16.1%) scrotal and penile infection, 4 (12.9%) scrotal penile and perineal involvement, and a patient (3.2%) extension to the anterior abdominal wall. Only a patient (3.2%) had severe sepsis at presentation, while 3 (19.7%) and 27 (87.1%) had sepsis and localized infections, respectively. All the patients had a polymicrobial infection from cultures of wound swabs using blood and MacConkey agar; however, Escherichia coli was the most common organism cultured in combination with other organisms in 26 (83.9%) patients. Other organisms cultured from the wound swab include Klebsiella in 12 (38.7%), Proteus in 8 (25.8%) patients, and Staphylococcus and Streptococcus in 16 (51.6%) and 4 (12.9%), respectively, with sensitivity patterns to either cephalosporins, aminoglycosides, or quinolones. The initial empirical antibiotic regimen of choice was a combination of intravenous ceftriaxone and metronidazole in 26 (83.8%) patients and intravenous ciprofloxacin and metronidazole in 5 (16.1%) patients. No patient had fecal diversion, however, 5 (16.1%) of patients had suprapubic cystostomies to divert the urine. Six (19.4%) patients were transfused during the management of the infection with mortality in 3 (9.6%) patients recorded. Table 1 shows the summary of the distribution of the proportions of clinical variables of the patients as related to mortality and their significance on analysis.

Table 1.

Clinical parameters of the patients as related to mortality

	Number of patients (%)	Mortality		P
		Yes	No
Duration of disease prior to presentation (weeks)
<1	4 (12.9)	0	4	0.671
1-2	25 (80.6)	3	22
>2	2 (6.5)	0	2
Anatomical extent of the disease
Scrotal	21 (67.7)	2	19	0.016
Scrotal and penile	5 (16.1)	0	5
Scrotal penile and perineum	4 (12.9)	0	4
Involvement of anterior abdominal wall	1 (3.2)	1	0
The severity of infection at presentation
Localized infection	27 (87.1)	0	27	0.001
Sepsis	3 (9.7)	2	1
Severe sepsis	1 (3.2)	1	0
Surgical debridements
1-2	24 (77.4)	2	22	0.116
3-4	5 (16.1)	0	5
>4	2 (6.5)	1	1
Blood transfusion
Yes	6 (19.4)	3	3	0.001
No	25 (80.6)	0	25
Reconstruction
Secondary wound closure	21 (67.7)	1	20	0.611
Skin grafting	10 (32.3)	1	9
Duration of hospital stay (weeks)
2	2 (6.5)	0	2	0.186
3-4	19 (61.3)	2	17
5-6	8 (25.8)	0	8
>6	2 (6.5)	1	1
FGSI
>9	11 (35.5)	3	8	0.014
<9	20 (64.5)	0	20

FGSI=Fournier’s Gangrene Severity Index

Anatomical extent of the disease, anemia requiring blood transfusion, severity of infection, and FGSI were all found to be the statistically significant variable of mortality in these patients using univariate analysis, as shown in Table 1. Furthermore, on regression analysis, as shown in Table 2, only the FGSI and blood transfusion were significant with P < 0.05.

Table 2.

Regression analysis of the independent variables of mortality in patients with Fournier’s gangrene

Model	Unstandardized coefficients		Standardized coefficients		Significant	95.0% CI for B
	Beta	SE	Beta	SE		Lower bound	Upper bound
Extent	0.337	0.284	0.113	1.185	0.247	−0.248	0.922
Transfusion	0.085	0.123	0.113	0.690	0.010	−0.168	0.338
FGSI	0.020	0.079	0.033	0.259	0.035	−0.142	0.183
Infection	0.519	0.093	0.785	5.585	0.100	0.328	0.710

^aDependent variable: Mortality. CI=Confidence interval, FGSI=Fournier’s Gangrene Severity Index, SE=Standard error

DISCUSSION

Fournier's gangrene is not an uncommon urological emergency, though the prevalence has been on the decline over the years. The studies on Fournier's gangrene have evolved from earlier reviews on its etiology, predisposing factors, presentation management, and outcomes to recent researches on the risk stratification and predictors of mortality. The epidemiology of the disease is still largely that of a disease that commonly afflicts elderly men as revealed in this study, in which the mean age was 60 years; however, patients with diabetes mellitus and HIV tend to present earlier in the third and fourth decades of life. The disease is typically a polymicrobial synergistic infection and E. coli remains the most common organism cultured. This observation may not be farfetched as it is a coliform implicated in urogenital and anal infections that share proximity with the genital and perineal skin.

We found a significant percentage (29%) of the patients with unknown predisposing factors following clinical evaluation, although 14 (45%) patients had documented evidence of immunosuppression. The number of patients with an idiopathic predisposing factor in this study is worthy of note as it varies widely in literature on Fournier's gangrene. Karthikeyan and Kumarasenthil found 17% as idiopathic,[7] Sockkalingam et al. in Coimbatore, India, found 29.8% as idiopathic,[18] and Kavya et al. also in India found an identifiable predisposing factor in 75% of 30 cases reviewed.[19]

Most of the infections were limited to the scrotum in 67.7% with the preservation of the testes. The infection may spread to the perineum and penis, with the former seen mainly with perianal abscesses and the latter in urethral strictures. Patients with background immunosuppression may have also spread to the abdominal wall.[20]

Although >80% of the patients presented with a localized infection, there was a pattern of delay in presentation of the disease to our hospital with many patients not presenting until 1–2 weeks from the onset of symptoms and others with even a more delayed presentation. This delay was due to reluctance to present with complaints on the external genitalia or prior attempt at the use of herbal medications. These observations were also noted by Ugwumba et al. in Enugu, Nigeria, and Neto et al. in Brazil.[21,22] This may ultimately contribute to the morbidity of prolonged hospital stay associated with the management of the infection but not statistically significantly related to mortality as found in this study.

Over 70% of these patients had 1–2 surgical debridement during management with a few requiring more debridement. This is like reports in Turkey by Unalp et al. in a review of 68 patients.[10]

We found the use of hypertonic saline sitz baths and povidone-iodine dressing rewarding in our patients with Fournier's gangrene. The hypertonic saline through its high osmotic gradient controls exudation and inhibits bacterial proliferation, while the povidone-iodine is bactericidal with antimicrobial activity against Gram-positive and Gram-negative bacteria including fungi and protozoa. These suffice for wound care with good granulation tissue obtained within a few days following debridement. Chalya et al. and Ghnnam in Tanzania and Egypt, respectively, noted similar favorable outcomes with the use of povidone-iodine dressing in patients with Fournier's gangrene.[23,24]

Similar to the experiences by Aliyu et al. and Chalya et al., who reported an average inhospital stay of 4 weeks in their experience with management of this disease,[24,25] the mean duration of stay in the hospital for this study is 22 days, though most of the patients (61%) were on admission for 3–4 weeks. This is because we achieve wound closure on the same admission before the discharge of the patient home. This is usually by tension-free secondary wound closure once there is healthy granulation tissue for scrotal and perineal infections, as shown in Figure 2. Furthermore, skin grafting of the phallus may be indicated when the penis is involved with the infection such as in the case depicted in Figure 3.

Figure 2.

Healthy granulation tissue following hypertonic saline dressing

Figure 3.

Scrotal and penile presentation of the disease

A mortality rate of 9.6% was noted in this series. This was in an elderly man with diabetes and two other patients who presented with severe sepsis on admission. This mortality rate is similar to 9.5% noted by Efem in Port Harcourt, Nigeria.[1] However, Ugbumba in Enugu, Nigeria, noted a 3.6% mortality rate.[25] The causes of death in their study were septic shock, diabetic ketoacidosis, and renal failure. This is similar to the findings of this review. In other climes such as Switzerland, Turkey, and Spain, the mortality rate of 15%–40% has been reported.[3,26,27] The higher mortality rate noted in these parts of the world may be due to the difference in presentation with more patients presenting with severe infections. For example, in Switzerland, Wetterauer et al. found a 15% mortality rate in a 5-year retrospective review, 75% of the patient managed for this disease presented with features of severe sepsis necessitating intensive unit care.[28] Aridogan et al. in Turkey and Medina et al. in Spain also reported 29.6% and 34% mortality rates, respectively, with >15% of their patients presenting with severe infections.[26,27] However, in our hospital, only 3.6% of our patients presented with severe sepsis. This observation of less aggressive disease in our environment was also noted in Enugu.[22]

In our series, the independent predictors of mortality are FGSI >9 and low hematocrit, necessitating the need for blood transfusion. Both the variables are, by extension, an assessment of host response to the infection and therefore a reliable predictor of mortality. Both the parameters have also been reported by other authors though with various cutoff for the FGSI in the determination of prognosis.[29,30,31] Conversely, Shukla et al. found FGSI not a good predictor of mortality in a study of the validity of the score.[32] Although in the same study, Shukla et al reported that the anatomical extent of the disease was a strong predictor of mortality in their series and interestingly we also found the anatomical extent of the disease to be significantly related to mortality in this study however only on univariate but not on logistic regression analysis.

CONCLUSION

Fournier's gangrene is predominantly a disease of the elderly and may be associated with immunosuppression. Most of the patients presented with infections localized to the external genitalia and this may be the reason for the lower mortality in our series unlike the western world. Higher FGSI value and low hematocrit requiring blood transfusions are the determinants of mortality.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.