Diagnosis of Gas Gangrene: Does a Discrepancy Exist Between Published Data and Practice PDF

# Diagnosis of Gas Gangrene: Does a Discrepancy Exist Between the Published Data and Practice ## Authors - Maryellen P. Brucato, DPM, AACFAS - Krupa Patel, DPM, AACFAS - Obinna Mgbako, DPM, AACFAS ## Affiliations - Atlantic Health System, Morristown, NJ - Foot and Ankle Specialists of New Jersey...

# Diagnosis of Gas Gangrene: Does a Discrepancy Exist Between the Published Data and Practice ## Authors - Maryellen P. Brucato, DPM, AACFAS - Krupa Patel, DPM, AACFAS - Obinna Mgbako, DPM, AACFAS ## Affiliations - Atlantic Health System, Morristown, NJ - Foot and Ankle Specialists of New Jersey, Rahway, Westfield, and Union, NJ ## Article Info - Level of Clinical Evidence: 4 - Keywords: - Clostridium - culture and sensitivity - diagnosis - infection - terminology ## Abstract The Infectious Disease Society of America defines gas gangrene as an infection caused by Clostridium species. However, in many clinical settings, soft tissue infections that produce subcutaneous gas have been diagnosed as gas gangrene without identification of the presence of Clostridium species. Instead, the diagnosis was based on clinical and radiographic findings. A chart review was performed of 25 consecutive patients treated at the Atlantic Health System for infections of the lower extremity that were diagnosed as gas gangrene. Wound cultures from the 25 patients grew 31 different species of bacteria. The most prevalent organism was *Staphylococcus aureus*, which accounted for 17 of 31 (54.84%) different organisms identified and 19.77% of 86 separate aerobic or anaerobic cultures. The most prevalent genus was *Streptococcus*, which accounted for 20 (64.52%) of the organisms identified and 23.26% of the cultures. The most prevalent anaerobic genus was *Peptostreptococcus*, which accounted for 8 (25.81%) of the organisms identified and 9.3% of the cultures. None of the cultures grew Clostridium species. These findings showed that gas-forming soft tissue infections of the lower extremity, commonly diagnosed as gas gangrene, are rarely caused by Clostridium species. ## Introduction The Infectious Disease Society of America practice guidelines for skin and soft tissue infections have defined gas gangrene as “a rapidly progressive infection caused by Clostridium” (1). Although the published data differentiate between gas gangrene owing to Clostridium and gas-forming soft tissue infections of nonclostridial origin, in practice, many physicians have used the diagnosis gas gangrene for any gas-forming soft tissue infection. Early reports have identified *Clostridium perfringens* as the primary offending organism in gas gangrene (2,3). Chiari (4), in 1893, was the first to describe a non-clostridial aerogenic infection when he described a diabetic patient who presented with a gas-producing infection caused by a colonic gram-negative rod. In 1941, Gillies (5) reported a case series in which 4 diabetic patients, including 1 with a foot infection, presented with infectious formation of gas in the soft tissues caused by gram-negative bacilli. In 1971, Altemeier and Fullen (6) reported the need to distinguish between gas gangrene and other gas-producing bacterial infections. They described true gas gangrene as clostridial myositis, which was characterized by its aggressively invasive course. In their experience, nonclostridial aerogenic infections were much more common than clostridial infections. In 1973, Weinstein and Barza (7) asserted that the diagnosis of clostridial gas gangrene could not be made according to the presence of soft tissue emphysema alone because nonclostridial organisms could produce air in the subcutaneous tissues. Nichols and Smith (8) reported a case series of crepitant soft tissue infections and cultured a variety of organisms, including *Peptostreptococcus, Eschericia coli, Bacteroides, Staphylococcus epidermidis, Streptococcus, diphtheroids, and Clostridium*. In 1975, Bessman and Wagner (9) described 48 cases of nonclostridial diabetic foot infections with radiographic evidence of subcutaneous gas. Currently, a general consensus does not seem to have been reached among many surgeons regarding the precise terminology for the diagnosis of nonclostridial gas-forming soft tissue infections, and the term gas gangrene seems to be commonly used whenever the clinical and radiographic findings indicate the presence of gas in the subcutaneous tissues. Although the Infectious Disease Society of America has clearly defined gas gangrene as being caused by Clostridium species, it has also recognized that nonclostridial organisms can produce aerogenic infections, which they have designated as nonclostridial myonecrosis. Hart et al (10) referred to this type of infection as gas-forming fasciitis, because necrotizing fasciitis is localized to the deep fascia and also to the superficial fascia and subcutaneous fat and is known to produce gas in many cases. In 2003, Finkelstein et al (11) acknowledged that in practice the term gas gangrene has been commonly used to describe any aerogenic soft tissue infection. At our institution, a lower extremity infection with tissue necrosis and clinical and radiographic evidence of subcutaneous emphysema will be given the diagnosis of gas gangrene without consideration given to the specific bacterial culture results. In an effort to determine whether the presence of Clostridium species was identified in any of the wound cultures, we reviewed a series of patients diagnosed with gas gangrene at our institution. Because it is generally known that nonclostridial gas-forming infections are commonly responsible for diabetic foot infections, we were also interested in determining the incidence of diabetes mellitus in the cohort of patients we reviewed. ## Patients and Methods A medical record review was performed of consecutive patients who presented to the podiatric surgery service of the Atlantic Health System during the 21-month period extending from January 1, 2009 to October 1, 2010, with a lower extremity soft tissue infection diagnosed as gas gangrene. The podiatric surgery service keeps a chronological record of patients, and one author (M.P.B.) used the descriptive diagnosis of gas gangrene to search the database. To be included, potentially eligible cases had to have documentation of clinical findings of lower extremity soft tissue necrosis (Figs. 1 and 2), including crepitus, corresponding radiographic findings of subcutaneous emphysema (Figs, 3 and 4), and wound culture results, all of which were recorded. All corresponding aerobic and anaerobic culture results were recorded, including any preoperative, intra-operative, and postoperative cultures. In addition to these variables, whether the patient had a history of diabetes mellitus was also recorded. Patients were excluded if they did not have subcutaneous emphysema confirmed on the radiographs, if they did not follow-up with our service, and if they did not have both aerobic and anaerobic cultures performed. The corresponding author (M.P.B.) reviewed the radiographs for subcutaneous emphysema independent of the radiology report, then confirmed that all of the authors concurred with this at the time of presentation. All of the authors concurred because we do not diagnose gas gangrene without the corresponding signs on the radiographs. ## Results A total of 27 potentially eligible patients were identified, of whom, 2 (7.4%) were excluded (1 was lost to follow-up after transfer to another service, and another did not have culture and sensitivity results). Thus, a total of 25 patients were included in our analyses, 22 (88%) of whom were diabetic and all (100%) of whom had radiographs available that showed the presence of subcutaneous emphysema. A total of 31 distinct organisms were identified in 86 separate cultures in the 25 consecutive patients (Table). The most prevalent organism was *Staphylococcus aureus*, which accounted for 17 of 31 different organisms (54.84%) and 19.77% of 86 separate cultures identified (Fig. 5). The most prevalent genus was *Streptococcus*, which accounted for 20 of the organisms (64.52%) and 23.26% of the cultures identified. The most prevalent anaerobic genus was *Peptostreptococcus*, which accounted for 8 of the organisms (25.81%) and 9.3% of the cultures identified (Fig. 6). None of the cultures grew Clostridium species. ## Discussion The results of the present observational investigation have made it clear that gas-forming soft tissue infections of the lower extremity that have been diagnosed as gas gangrene can be caused by a variety of organisms, none of which were *Clostridium* species. Furthermore, 88% of the patients in our series were diabetic, reinforcing the common understanding that gas-forming infections are particularly prevalent in patients with diabetes mellitus. It has also been our observation that, in practice, the diagnostic label of gas gangrene has been applied, perhaps loosely compared with the Infectious Disease Society of America standards, to any soft tissue infection with corresponding tissue necrosis that demonstrates subcutaneous emphysema clinically and on plain radiographic films. This was the case at our institution, and we presume it is likely the convention at other institutions. In the case series we have described, none of the wound pathogens were identified as *Clostridium* species, and this finding was in keeping with those from previous published reports (4-11). These infections might have been more appropriately diagnosed as non-clostridial myonecrosis, rather than gas gangrene, and we believe this has primarily been an issue of semantics and not an issue related to an inappropriate or improper clinical diagnosis. The distinction between clostridial and nonclostridial gas-forming infections has been discussed in published reports (11), and a key biologic difference between the 2 types of infection might have been based on the fact that Clostridium species secrete exotoxins that contribute to the rapid spread of the infection, whereas nonclostridial aerogenic infections are thought to typically have a slower onset and course (11). Clinicians should remember, however, that nonclostridial aerogenic infections, especially mixed, polymicrobial infections in diabetic patients, can be highly virulent, rapidly spreading, limb threatening, and potentially fatal (12,13). As long as the published data maintains a distinction between the presentation and course of 2 different types of infections (i.e., clostridial versus nonclostridial), our diagnostic terms and documentation should reflect that. Just as with most retrospective investigations, we realize that numerous methodologic shortcomings could threaten the validity of our conclusions. For instance, the high prevalence of gas-forming infections in diabetic patients was likely confounded by other factors that influence the foot and bring patients to our practice. Moreover, ascertainment of radiographic evidence of gas in the soft tissues was determined by a single coauthor who, in some cases, participated in the care of the study patients, and these factors might have biased our results. Despite these shortcomings, our observational findings support our concern regarding the clinical rationale for reserving the diagnosis of gas gangrene only for infections caused by Clostridium species. The case series we have described has highlighted the need for greater clarity among foot and ankle surgeons, and perhaps surgeons and clinicians in general, regarding the terminology related to the diagnosis of gas gangrene. Despite numerous reports demonstrating that nonclostridial gas-forming infections of the lower extremity are more prevalent currently than those caused by Clostridium perfringens, the published data appear to consistently imply that Clostridium species are the most common cause of infections that display soft tissue emphysema (14). The usefulness of making a semantic distinction between clostridial and nonclostridial gas gangrene has not been clearly articulated. Also, many surgeons, as a matter of practicality, have used the diagnosis of gas gangrene to refer to all gangrenous wounds that display subcutaneous emphysema. However, we believe that no advantage results from the distinction in regard to patient care. Thus, perhaps, the time has come to revisit the classification of aerogenic wound infections. It is our hope that our report will generate additional discussion regarding the semantic issues related to the diagnosis of gas-forming soft tissue infections in the lower extremities and elsewhere in the body. ## References 1. Stevens DL, Bisno Al, Chambers HF, Everett ED, Dellinger P, Goldstein El, Gorbach SL, Hirschmann JV, Kaplan El, Montoya JG, Wade JC. Practice guidelines for the diagnosis and management of skin and soft tissue infections. Clin Infect Dis 41:1373-1406, 2005. 2. Yachnin S. Infection of war wounds by Clostridium organisms (gas gangrene). Med Bull US Army Force Europe Theater Off Theater Chief Surg 29:19-27, 1945. 3. Stock AH. Clostridia in gas gangrene and local anaerobic infections during the Italian campaign. J Bacteriol 54:169-174, 1947. 4. Chiari H. Bacteriologie des septischen emphysemas. Prag Med Wochenschr 18:1-4, 1893. 5. Gillies C. Interstitial emphysema in diabetes mellitus due to colon bacillus infection. JAMA 117:2240-2242, 1941. 6. Altemeier WA, Fullen WD. Prevention and treatment of gas gangrene. JAMA 217:806-813, 1971. 7. Weinstein L, Barza M. Current concepts: gas gangrene. N Engl J Med 21:1129-1131, 1973. 8. Nichols RL, Smith JW. Gas in the wound: what does it mean? Surg Clin North Am 55:1289-1296, 1975. 9. Bessman AN, Wagner W. Nonclostridial gas gangrene: report of 48 cases and review of the literature. JAMA 233:958-963, 1975. 10. Hart G, Lamb R, Strauss M. Gas gangrene: a collective review. J Trauma 23:991-1000, 1983. 11. Finkelstein B, Kamble R, Ferdinando E, Mobarakai N. Autoamputation of the foot caused by untreated gas gangrene: a case report. J Foot Ankle Surg 42:366-370, 2003. 12. Bird D, Giddings AE, Jones SM. Non-clostridial gas gangrene in the diabetic lower limb. Diabetologia 13:373-376, 1977. 13. Brightmore T. Non-clostridial gas infection. Proc R Soc Med 64:36-37, 1971. 14. Takahira N, Shindo M, Tanaka K, Soma K, Ohwada T, Itoman M. Treatment outcome of nonclostridial gas gangrene at a level 1 trauma center. J Orthop Trauma 16:12-17, 2002.