Synchronous multifocal necrotizing fasciitis prognostic factors: a retrospective case series study in a single center

Necrotizing fasciitis (NF) is an uncommon deep soft tissue infection characterized by a rapidly spreading necrosis of superficial fascia and subcutaneous tissue [1]. Although typical NF involves primarily a single site, some recent studies report the concurrent involvement of multiple sites [2‐4]. Some reports [2, 5, 6] incidentally indicate that multifocal necrotizing fasciitis leads to a higher mortality rate than does monofocal NF. We found no reports that focused primarily on synchronous multifocal necrotizing fasciitis (SMNF). Thus, we evaluated the clinical characteristics and outcomes of SMNF.

This is the first report of a case series of SMNF and analysis of its predictive factors for mortality. In the current study, SMNF accounted for 5% of NF. The main characteristics of our patients with SMNF were male sex, lower limb involvement, and marine Gram-negative bacterial infection. The incidence of SMNF in noncontiguous sites is not clear in the literature. Park et al. [5], in a retrospective case review of 217 patients with NF in the coastal area of South Korea, reported that at least 73 (34%) patients had noncontiguous multifocal involvement with highly frequent isolation of Gram-negative marine bacteria. El-Khani et al. [2], in a case report and systematic review of 33 cases of multifocal NF, reported that most patients were male, that their lower limbs were involved, and that there were 18 Gram-positive bacteria, 6 Gram-negative bacteria, 5 anaerobic bacteria, 1 yeast, and 4 polymicrobial isolates in the microbiologic cultures. Similar to the findings of Park et al. [5], most of our cases were classified as type III NF (infection by monomicrobial pathogen, usually Gram-negative bacteria). Moreover, in 89% of our patients with SMNF, marine bacteria—Vibrio spp. in 78% and Aeromonas spp. in 11%—were most commonly isolated. In contrast, there were 12% type I NF (infection by polymicrobial bacteria) and 52% type II (infection by a monomicrobial pathogen, usually Gram-positive bacteria) in El-Khani et al. [2]. The different bacteria isolated might be attributable to the different geographical regions and epidemiology of the cases reported [7‐9]. Marine bacteria (e.g., Vibrio spp. and Aeromonas spp.) are often found in the warm coastal areas of Asia (e.g., Taiwan, Hong Kong, South Korea, Singapore, and Thailand), the Gulf of Mexico, South America, and Australia [10‐12]. Also, NF in patients with chronic liver diseases is usually a monomicrobial infection, caused primarily by Gram-negative bacteria [13, 14]. Our hospital is in the Tropic of Cancer in southwestern Taiwan, where many local inhabitants work in fisheries or consume seafood, and where alcoholism, chronic viral hepatitis, and liver cirrhosis are highly prevalent. Therefore, a specific epidemiology, i.e., a high prevalence of chronic liver diseases in a coastal area, would make marine Gram-negative bacteria the main causative pathogens of SMNF.

The overall mortality rate of our patients with SMNF was 67%. Most of the literature focuses only on monofocal NF and reports a cumulative average mortality rate of 34%. The survival rate for NF patients has risen to as high as 85% as the awareness of monofocal NF has increased and as more-frequent aggressive treatment and adjuvant therapy for the disease have increased. In contrast, multifocal NF remains highly fatal. Park et al. [5] reported that patients with infections in multiple sites had a mortality rate of 62%, significantly higher than the 30% in patients with a single-site infection. Similarly, Hua et al. [6], in a retrospective single-center cohort study, said that multifocal necrotizing soft tissue infection was an independent risk factor for mortality. SMNF might be caused by rapid metastatic septic embolization or simultaneous multifocal inoculation and the subsequent development of a necrotizing infection. The former implies fulminant septicemia with the hematogenous spread of bacteria, and the latter implies that more than one site has been concomitantly exposed to pathogenic bacteria. Both conditions can lead to overwhelming sepsis. Also, a large area of skin, subcutaneous tissue, fascia, and underlying necrotic destruction of muscle in multiple-site necrotic deep soft tissue will cause extensive complications and a high mortality rate. Like extensive NF that involves a large surface area of the body [15], an immediate fasciotomy and radical removal of necrotic tissue without a misdiagnosis is important when treating SMNF. The treatment protocol should consist of prompt antibiotic therapy, prolonged intensive care, nutritional support, serial surgical debridement, and reconstruction of soft tissue to reduce mortality and morbidity rates.

In the present study, non-survivors had a higher prevalence of involved bilateral lower limbs and thrombocytopenia than did survivors. There is a paucity of reports on the risk factors for SMNF mortality. However, the mortality rate in cases of monofocal NF with upper extremity involvement is not significantly different from that in cases with lower extremity involvement [16, 17]. El-Khani et al. [2] reported a significantly higher mortality rate in bilaterally affected lower limbs than in bilaterally affected upper limbs (60 vs. 0%), but the clinical mechanism of mortality related to lower extremity involvement is unclear. In the present study, chronic liver diseases were the most common comorbidities (61%) in SMNF, especially in non-survivors (75%). Some evidence [18‐20] indicates that atherosclerosis and cutaneous vasculitis in the lower limbs were highly associated with chronic liver diseases (e.g., viral hepatitis and alcoholic hepatitis). Also, extrahepatic cutaneous vasculitis [21], primarily of the lower limbs, leads to cutaneous lesions and to an insufficient barrier against bacterial invasions (Fig. 1). Atherosclerosis, vasculitis, and devitalized soft tissue are the results of a poor blood supply, which causes the clinical deterioration of necrotizing deep soft tissue infection [22]. These comorbidities might result in a higher prevalence of involved bilateral lower limbs in patients who die from NF (Fig. 2).

Consistent with our findings for multifocal NF, thrombocytopenia has also been a predictive factor for mortality in monofocal necrotizing deep soft tissue [23, 24]. Physicians should be aware of the grave prognosis for patients with SMNF of the bilateral lower limbs and who present with thrombocytopenia. Compared with survivors, non-survivors tended to have lower, but not significantly different, C-reactive protein (CRP) levels. In contrast, Park et al. [5] reported that the CRP level was one predictor of mortality in NF. Using the CRP level to predict mortality is controversial. Many studies [25, 26] have reported that CRP levels are not significantly different between survivors and non-survivors. Moreover, CRP levels reach peak values between 48 and 72 h after the onset of bacterial infection [27], and CRP does not precisely indicate the severity of bacterial infections at a single time point. In the present study, non-survivors had a relatively shorter onset of symptoms, which might indicate a lower CRP level on admission.

This study has some limitations. This is a case series study with a small number of patients, because NF is a rare disease and our inclusion criteria were restricted to peripheral NF (involvement of extremities). Central NF (involving head, neck, abdomen, and perineum) was excluded, because its etiology and surgical procedures are not similar to those of peripheral NF. NF of the head and neck often arises from an odontogenic infection with mixed aerobic and anaerobic abscesses [28, 29]. NF of the abdomen or perineum might be the result of urogenital or colorectal infections by enteric microorganisms [30]. A colostomy is usually required when treating abdominal or perineal NF [31‐33]. Moreover, we did not include preexisting chronic wounds, of which the etiology might be pyomyositis or osteomyelitis. Univariate analysis rather than multivariate analysis was used to determine predictive factors for mortality, because there were only a few patients in each group. Although this study indicated that the involvement of the bilateral lower limbs and thrombocytopenia were highly associated with mortality in SMNF, additional studies are needed to confirm this.

In conclusion, the incidence of SMNF of the extremities was 8% for all subtypes of NF. Male sex, bilateral lower limb involvement, and marine Gram-negative bacteria were the most frequently found factors in SMNF. The mortality of SMNF was as high as 67%, which was more frequently associated with initial thrombocytopenia and bilateral lower limb involvement. It is essential to detect a multifocal presence of NF in the extremities to justify an immediate fasciotomy and the radical removal of necrotizing tissue after prompt antibiotic therapy and sufficient resuscitation.