Necrotizing Fasciitis: How Reliable are the Cutaneous Signs?

Necrotizing fasciitis (NF) is a surgical emergency. It is often aggressive and characterized by the rapidly progressive inflammatory infection of the fascia that causes extensive necrosis of the subcutaneous tissue and fascia, relatively sparing the muscle and skin tissue. As the disease progresses, thrombosis of the affected cutaneous perforators subsequently devascularizes the overlying skin. The course indeed can be a fulminant one. The diagnosis of NF, especially in the early stages, is extremely challenging, and it can be very close in presentation to other skin and subcutaneous tissue infections. The primary site of the pathology is the deep fascia. Necrosis of the tissues and fascia may manifest as erythema without sharp margins, swelling, warmth, shiny, and exquisitely tender areas. Pain out of proportion to physical examination findings may be observed. The subcutaneous tissue may be firm and indurated such that the underlying muscle groups cannot be distinctly palpated. Eventually, as the overlying skin is stripped of its blood supply, skin necrosis ensues and hemorrhagic bullae form. Bacteremia and sepsis invariably develop when the infection is well established. This paper discusses some of issues related to the cutaneous signs found in NF and also provides a review the current, available literature on the subject matter.

INTRODUCTION

Necrotizing fasciitis (NF) is a surgical emergency. It is often aggressive and insidiously advancing, characterized by the rapidly progressive inflammatory infection of the fascia that causes extensive necrosis of the subcutaneous tissue and fascia, relatively sparing the muscle and skin tissue. This is primarily due to the better blood supply of the muscles as compared to the fascia. As the disease progresses, thrombosis of the affected cutaneous perforators subsequently devascularizes the overlying skin. The course indeed can be a fulminant one.[12]

The primary site of the pathology is the deep fascia. Necrosis of the tissues and fascia may manifest as erythema without sharp margins, swelling, warmth, shiny, and exquisitely tender areas. Pain out of proportion to physical examination findings may be observed. The subcutaneous tissue may be firm and indurated such that the underlying muscle groups cannot be distinctly palpated. Eventually, as the overlying skin is stripped of its blood supply, skin necrosis ensues and hemorrhagic bullae form. Bacteremia and sepsis invariably develop when the infection is well established.[134]

The stages and types of necrotizing fasciitis

Early diagnosis is important for treatment and saving the life of the patient. The diagnosis is primarily based on the clinical findings. Clinical characteristics of NF are usually classified in three stages. The early stage of NF may not be distinguished clinically from other soft-tissue infections such as erysipelas and cellulitis, but there are other helpful clinical features such as poorly defined and indistinct margins and tenderness beyond the involved area, warmth, swelling, and induration. In the intermediate stage, there is often blisters or bullae formation with serous fluid. Purpura may or may not be found but does not preclude the diagnosis of NF. The areas surrounding these lesions tend to be more erythematous as the advanced stage sets in with skin discoloration, duskiness, necrosis and at times, eschar tissue formation.[124]

The cutaneous signs pointing toward NF in the early stages are easily confused with cellulitis and other skin and subcutaneous infections. NF can be suspected if the followings are seen:[34]

Rapid progression of the cutaneous lesion

Poor therapeutic response to usual modality of treatment and drugs

Blistering necrosis sets in

Dusky and cyanosis of the tissue

Extreme local tenderness, more exaggerated than what it tends to appear

Associated systemic signs such as high temperature, sepsis, tachycardia, hypotension, and altered mental state presentation.

NF is typically classified based on bacterial morphology; Type I NF and Type II NF. Type I is characterized by the presence of polymicrobial bacteria involving both aerobic or anaerobic bacteria. In Type I infection, at least one anaerobic species of NF is also interesting. For example, this takes place when Group A streptococcus infection happens in combination with one or more facultative anaerobic streptococci (other than group A) and members of the Enterobacteriaceae (eg, Escherichia coli, Enterobacter, Klebsiella, Proteus).[56]

Conversely, Type II NF is characterized by monomicrobial bacteria. It is most commonly caused by Group A Streptococcus (GAS). In cases with no clear portal of entry, the pathogenesis of infection likely consists of hematogenous translocation of GAS from the throat (asymptomatic or symptomatic pharyngitis) to a site of blunt trauma or muscle strain. Other common organisms include Aeromonas hydrophila and Vibrio vulnificus, which being marine bacteria, cause NF in association with freshwater and seawater injuries, respectively. Type III NF is also known as clostridial myonecrosis or gas gangrene. This type of skeletal muscle infection may also be related to recent surgery or trauma. Type IV NF, the more recently described type, is fungal NF. Candida species may be the etiological agent but at times can occur in combination with a bacterial etiology.[46789101112]

The pathophysiology of NF is also interesting. For example, when Group A strep adheres and infects host cells, it will deliver 2 types of streptolysin toxins into these cells. These toxins impair the mechanism for quality control of protein synthesis. It also triggers a defense stress response which increases the production of asparaginase, which alters the gene expression profile and rate of proliferation which can be deadly in the host.[13]

There is also a phenomenon described as subacute NF which runs a more indolent course. In these cases, there may be complaints of areas of festering soft-tissue infection with minimal pain or discomfort. This can fester for weeks to months, before a sudden deterioration taking on a more fulminant course happens. This makes the diagnosis of NF even more challenging, especially if it takes on this more subacute course of progression. Some have considered the presentation and clinical signs too mild in these early stages entertain the possibility of NF. In retrospect, there have been cases whereby the progression is certain and delay in diagnosis can result in higher morbidity and mortality.[1415]

Objective

The objective of this review is to determine the reliability of cutaneous signs in NF and to provide an update on what the latest literature has with regard to the aforementioned topic. It will also collate some of the more important observations and descriptions that have been made.

METHODOLOGY

Search engines such as PubMed, UpToDate, and Google Scholar were used in this literature review. These search engines were utilized in a systematic literature search from January 2000 to March 2017. The keywords used with “necrotizing fasciitis,” “cutaneous signs,” “diagnosis,” “retrospective studies,” and “reliability.” Results were filtered to include only retrospective studies undertaken in humans and published in English.

In the initial phase of search, it was noted that several literature reviews encompassing the similar topic had already been carried out. Of note, the most extensive, complete, and recent literature review was a prominent study: “Early diagnosis of necrotising fasciitis” by Goh et al. This study reviewed articles from January 1980 to May 2013 and incorporated a total of nine case studies. However, seven out of the nine studies included patients only till 2005.[16] This study is heavily quoted with more than seventy citations in much of the more recent literature or research surrounding NF. Thus, it was determined that an updating of the literature was necessary.

Studies were included if they met the following: (1) studies published after 2013; (2) studies beyond the time frame of the earlier literature review; and (3) studies that recorded the incidence of presenting signs and symptoms.

Data extraction and assessment of methodological quality

Data extraction and analysis were undertaken by two reviewers independently. Disagreements on the quality of the study chosen were resolved by consensus. Baseline data collected were year of publication, period of the study, country of study, and diagnosis of NF. Specific outcome data collected were incidence of signs and symptoms at presentation. This was based on the specific parameters provided by the earlier literature review by Goh et al. They include erythema, warmth, pain/tenderness, swelling, bullae, crepitus, and skin necrosis.[16]

DISCUSSION OF RESULTS

Reliability of cutaneous signs

Based on the earlier literature review by Goh et al., the top three presenting clinical features in NF were swelling (80.8%), pain (79.0%), and erythema (70.7%). After reviewing the recent literature, the results corroborated closely, with the most common initial signs seen remaining the same: swelling (79%), pain (76.9%) and erythema (69.6%) [Tables 1 and 2].[6161718192021222324252627282930313233]

Table 1

Literature review from Goh et al.

Table 2

Updated literature review

Classically, patients with NF present with the triad of pain, swelling, and erythema. Yet, other diagnoses present similarly, namely, soft-tissue infections such as cellulitis, impetigo, erysipelas, and early abscess. The most consistent feature of early NF is that the pain is out of proportion to the swelling or erythema. Other distinct features may serve as diagnostic clues: Tenderness extending beyond the apparent involved area due to the spread of the infection along the fascia below the surface of the skin (the spread of enzymes and toxins causes irritation to its contact surfaces); margins of tissue involvement being indistinct; lymphangitis being rarely seen in NF because the infection is in the deep fascia and not in the skin; and the rapid progression of NF despite antibiotic therapy. Repeated physical examination and clinical assessment of the patient with a visual pain score and marking of the spread of the infection are useful. After 2 or 3 days of constitutional symptoms, supralesional vesiculation and bullae formation may occur, with serosanguineous fluid leak (intermediate stage of NF).[124283435]

The progression of NF is marked by the development of clear blisters/bullae. This is the clearest way to distinguish the intermediate stage between early NF (apart from the cutaneous features mentioned above) and late NF. Late stage of NF is characterized by skin necrosis and crepitus formation. It is noted that this corresponds with the review's findings; bullae formation occurs in 44.5% while skin necrosis and crepitus occur less frequently 23.5% and 4.9%, respectively. Painless ulcers may spread along tissue planes. Black necrotic eschar may also be seen at the borders of the affected areas. In some cases, metastatic cutaneous plaques may occur. Other lesions include purpura with or without bullae formation which does not preclude the diagnosis of NF. NF has also been observed to develop after skin biopsy, at needle puncture sites, after frost bites, open fracture sites, and even insect bite sites [Table 2].[1416353637]

As the disease progresses, virulent organisms and toxins are released into the bloodstream from infected soft tissue. This results in the initiation of a systemic toxic reaction which can result in hypotension, disseminated intravascular coagulation, and eventually multi-organ failure.[5131619]

There are some limitations of this article in terms of assessing the reliability of the cutaneous signs. First, the authors are unable to ascertain the exact nature of the pain as quoted by the various cases studies, i.e., if it was just mild tenderness or pain out of proportion as is classical in NF. Second, while the literature review includes the actual number of patients with NF with the relevant signs, it is unable to elicit the sensitivity and specificity of that particular sign as there are many differential diagnoses to the same sign. Sensitivity and specificities of the various clinical signs were also not included in any study or in any preliminary search.

Comparison between Eastern and Western studies

While analyzing the various articles from the preliminary search, the authors also noted the different pathophysiology of NF based on the geographical locations of the relevant case studies. Of significance, would be countries situated geographically next to oceans or water-bodies compared with countries that are considerably inland such as the USA or Europe.

In addition to recording the incidence of cutaneous signs in patients with NF, Park et al. (a South Korean study) determined that the microbiology of the disease has distinctive features reflective of its surrounding community. In his case, South Korea is a country with extensive sea coasts where the average water temperatures in the summer are >15°C.[25] Vibrio spp. have been found in warm coastal waters with temperatures ranging from 9°C to 21°C. A. hydrophila is isolated from freshwater or seawater in the late summer/early autumn when temperatures are approximately 20°C–25°C. Stonefish stings have also been implicated in NF.[253839]

Likewise, the study also noted that marine bacteria such as Vibrio spp. have also been detected in other warm coastal regions, including Asia (Thailand, Taiwan, and Singapore), the Gulf of Mexico, South America, and Australia. In coastal areas, diverse marine bacteria should be considered as the major etiologic organisms causing NF after exposure to seafood and/or seawater.[253839]

Conversely, a study undertaken in the Netherlands noted that there is a lack of marine-related bacteria (Vibrio spp. or A. hydrophila) from bacterial cultures in patients with NF. The causative organisms in their study were mainly Gram-positive bacteria, in particular the GAS. It is believed that this difference in pathogen could influence the course of the disease and the outcome of patients with NF (260).

According to Tsai et al.,[38] Staphylococcus aureus and Streptococcus spp. (Gram-positive bacteria) were the most common microorganisms in the USA and Europe, and monomicrobial Gram-negative microorganisms were most common in Asia. Some authors conclude that clinical characteristics of Gram-negative infections are more fulminant than of Gram-positive infections.[20252628303940]

Use of imaging and laboratory results as diagnostic tools

The authors of this paper noted how studies would supplement the clinical diagnosis of NF with laboratory investigations and imaging modalities such as ultrasound scans, X-ray, computer-tomography (CT) scans, and magnetic resonance imaging (MRI).[414243444546]

Studies that incorporate laboratory values into their studies often quote the Laboratory Risk indicator for Necrotizing Fasciitis (LRINEC) score. This laboratory score was a tool developed by a Singaporean-based study lead by Wong et al.,[6] for distinguishing NF from other soft-tissue infection. The paper determined that the cutoff value for the LRINEC score was 6 points with a positive predictive value of 92.0% and negative predictive value of 96.0%. Thus, a value of 6 or more should rule in NF. However, a value <6 is unable to rule it out.[4142434445]

However, there are papers such as Sigt et al. and Tsai et al. that stated the score is unsatisfactorily examined and remains invalidated for larger studies.[2638] Despite its high specificity and negative predictive value, since the score was published, subsequent data have demonstrated its limited sensitivity.

Imaging modalities are also useful tools. Ultrasound scans can be used on an emergency basis to diagnose NF based on several key features such as subcutaneous thickening, air, and fascial fluid.[43]

Radiographic imaging studies, such as soft-tissue radiographs, CT scanning, and MRI, are most helpful if gas is identified in the tissue; this is seen most frequently in Type I NF or gas gangrene caused by clostridia. Asymmetric fascial thickening, fat stranding, and gas tracking along fascial planes are important imaging findings on CT and MRI scans. CT scans are estimated to have a sensitivity of 80% for detecting necrotizing soft-tissue infections.[444546] According to Schmid et al., the sensitivity of MRI is 100% with a specificity of 86%. It is particularly helpful in the investigation of areas that have disproportionately severe tenderness and when nonspecific signs of sepsis are present.[47]

CONCLUSION

This literature review has demonstrated that the most common (cutaneous) presenting symptoms of NF are swelling, pain, and erythema. While other differential diagnoses may also present with such similar results, there are diagnostic clues such as the nature of the pain (i.e., pain out of proportion to physical findings or extending beyond the area of involvement) as well as the further progression into the intermediate and later stages of NF that makes cutaneous signs even more reliable. The observation of the constellation of signs as the disease evolves is also important as the progressive changes become more apparent. The pairing of serial close inspection and a high index of suspicion cannot be overemphasized. Even when there are other adjuncts such as radiological investigations and LRINEC scores available to better predict the possibility of the NF, it has to be stressed that NF is still a clinical diagnosis and clinical acumen is necessary. Such tools should only be used primarily for patients in whom the diagnosis is doubtful as NF is a surgical emergency and necessitates immediate treatment.

Final recommendations

One recommendation that this literature review proposes would be to reanalyze the raw data from each case study, to produce the eventual sensitivity and specificity of each cutaneous sign for NF; taking into account the differential diagnosis of soft-tissue infections which may also present as such.

The second recommendation would be to carry out a meta-analysis of the various case studies' results and of the adjunct diagnostic tools. This will not only establish a more definitive answer to the reliability of cutaneous signs in the diagnosis of NF but also allow researchers to compare and rate the difference in the sensitivity and specificity between the different modalities – clinical, laboratory, and imaging – in the diagnosis of NF.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.