Necrotizing fasciitis is an uncommon disease with a high incidence of morbidity and mortality. The disease was first described by Hippocrates, and since then, it has been known by many names.1 In 1952, Wilson2 called it necrotizing fasciitis, which until now most accurately describes the pathology of the disease. Necrotizing fasciitis potentially affects any part of the body. However, studies have shown that the mortality is highest when the disease involves the abdominal and perineal regions.3,4 On the other hand, survival appears to be highest in disease of the extremities.4 Regardless of location, however, the pillars of treatment have not changed, namely, early diagnosis, resuscitation, early aggressive surgical debridement, and broad-spectrum antibiotics.5
This study aims to review the presentation and the management outcome of patients with upper limb necrotizing fasciitis treated in our department and to identify methods in improvement of overall outcome of this dreadful disease.
MATERIALS AND METHODS
Twenty-nine cases of necrotizing fasciitis were treated in the Department of Hand Surgery, Singapore General Hospital, from 1998 to April 2005. All available case and surgical records and x-rays were reviewed retrospectively. Diagnosis was made based on intraoperative findings of grayish necrotic and avascular fascia which broke down easily with gentle finger dissection, and classic "dishwater" pus. Thirteen patients had histological samples taken during surgery; of which, 12 had histological findings consistent with necrotizing fasciitis, namely, vascular thrombosis, inflammatory infiltrate, and necrotic fascia.
The mean age was 65 years (range, 38-83 years). There were 14 male and 15 female patients. The average duration of stay was 22.5 days (range, 6-60 days). The average follow-up period was 6.3 months (range, 1-35 months). The left upper limb was involved in 9 patients, whereas the right was involved in 20 patients. Three patients had concurrent involvement of another site (2 in the lower limb and 1 in the perineum). Seventeen (58.6%) were diabetic patients, whereas 8 (27.6%) had other forms of immunosuppression, such as steroid therapy (2), leukemia, lymphoma, or other forms of blood dyscrasia on chemotherapy (4). There were 5 patients with ischemic heart disease, 3 with severe renal disease, and 3 with liver cirrhosis. Twelve of these patients had multiple comorbidities. None were intravenous drug abusers or patients with human immunodeficiency virus infection. Twenty (69%) had a known inciting event. These included trauma related to intravenous or intra-arterial line insertion (7), preceding history of tight bandaging or massage to injured extremity (3), trauma related to a marine creature (4), other forms of penetrating injury (4), and blunt trauma (2). Three patients sought traditional medical treatment before hospital presentation. The time of onset (patients who presented from the community) to hospitalization averaged 5.8 days (range, 1-21 days).
All patients presented with swelling, whereas 28 (96.6%) presented with both pain and erythema. Only 20 patients (69%) were febrile, 9 patients (31%) had ecchymosis, and 8 patients (27.6%) presented with bullae. Initial sites of involvement were at the level of the digits in 2 patients, the hand in 7 patients, the hand to forearm in 13 patients, the forearm alone in 5 patients, the forearm to arm in 1 patient, and the hand to the upper arm in 1 patient. Two patients had 1 upper limb with contralateral lower limb involvement. Only 2 (13.8%) of 14 available x-rays demonstrated soft tissue gas.
On presentation, only 35% (10) of patients were suspected to have necrotizing fasciitis from the outset. Other initial diagnoses included cellulitis (4), abscess (10), gangrene (2), and deep infection (1).
Twenty-five patients underwent debridement within 24 hours of admission. Three patients had surgery between 24 and 48 hours after admission. The mean time to surgery for these 28 patients was 13.2 hours (range, 2-48 hours). One patient's surgery was performed only after 120 hours because he refused treatment initially. Postoperatively, 8 patients were monitored in the intensive care or high-dependency unit. The duration of stay was 4.4 days on average (range, 1-13 days). The average number of wound debridements required was 1.7 (range, 1-5). In between debridements, the wounds were dressed with hydrocolloid gel dressings or gauze dressings. The average number of isolates was 1.4. The number of isolates per specimen was as follows: single isolate in 16 patients, double isolates in 6 patients, 3 isolates in 3 patients, and 4 isolates in a single patient. There was no bacterial growth in 3 patients. The organisms cultured from the intraoperative tissue samples are displayed in Table 1. Only 3 of 24 blood cultures were positive. Two grew group B streptococci, whereas one grew Staphylococcus aureus. Broad-spectrum antibiotics were started for all patients, and antibiotic therapy was subsequently modified according to tissue cultures and sensitivity.
Wound closure was achieved by healing by secondary intention in 1 patient as he refused further surgery for resurfacing, 3 patients underwent secondary suture, 14 patients had wound coverage by split skin grafting, whereas 3 patients required flap coverage of the wound. Seven patients (24.1%) underwent amputation (Table 2). One patient died of cardiac complications before any resurfacing of the wound was performed.
Postoperatively, there were 3 mortalities (10.3%; 2 from cardiac events postoperatively and 1 from overwhelming sepsis). Other complications included lung collapse (1), nosocomial pneumonia (1), urinary tract infection (1), cardiac ischemia (1), hepatic encephalopathy (1), anemia (1), lower limb cellulitis (1), bleeding gastrointestinal tract (1), and dry gangrene of toes secondary to septic emboli (1).
Necrotizing fasciitis is a severe infection that progresses rapidly, often leading to severe morbidity and mortality. However, mortality of upper limb necrotizing fasciitis appears to be less than other sites.5
At high risk for necrotizing fasciitis, in general, include diabetic patients, alcoholics, immunosuppressed patients, intravenous drug abusers, and patients with peripheral vascular disease.5 With regard to upper limb necrotizing fasciitis in particular, when we compared our series to that of Schecter et al,6 of whom 32 of 33 patients were drug users, none of our patients were intravenous drug users. However, 58.6% of our patients were diabetic patients, and 27.6% were patients on steroid therapy or with hemopoietic disorders on chemotherapy. Eleven other patients had ischemic heart disease, liver disease, or severe renal disease.
Most (69%) of our patients with upper limb necrotizing fasciitis had a known inciting cause leading to skin trauma, creating portals of entry for bacteria. Importantly, 11 patients (37.9%) had predisposing events related to medical treatment of a different problem, featuring prominently 7 patients with intravenous or intra-arterial cannulation leading to infection. This reminds us of the need to practice strict aseptic techniques during insertion of invasive lines in patients, even more so when patients are immunocompromised. The remainder of 9 patients had no known cause. It has been documented in the literature that up to 50% of cases may not have a known cause.7
In our series, the most common presenting findings of upper limb necrotizing fasciitis were that of swelling (100%), pain (96.6%), and swelling (96.6%). In the literature, the main symptoms in upper limb necrotizing fasciitis are generally pain, swelling, erythema, and fever (Table 3).3,7-9 However, although early diagnosis and early aggressive treatment are important in improving outcome, the difficulty lies in the diagnosis.3 Swelling, pain, and erythema with fever are nonspecific, and in our study, almost half (14) of our patients were initially diagnosed as either cellulitis or abscess of the upper limb. The diagnosis of necrotizing fasciitis was subsequently made intraoperatively. Although bullae are an important diagnostic clue, they were present only in 27.6% of patients. The proportion of patients presenting with bullae has ranged from 4% to 44.9%.3,7,9 The formation of bullae is also related to the time of patient presentation in the disease process and is often only present when the disease is fairly advanced. Early in the disease, when the patient presents only with pain, some swelling, and erythema, the disease is often mistaken as cellulitis.5 This, unfortunately, leads to inevitable delay in appropriate treatment of this rapidly progressing disease.
Hence, a high index of suspicion regarding the diagnosis of necrotizing fasciitis is required. Regular review of patient condition is needed because necrotizing fasciitis can often progress within the space of a few hours, and rapid advancement of the erythematous margins and indurated skin despite the commencement of broad-spectrum antibiotic therapy is an important diagnostic clue.7 The difficulty in early diagnosis is underlined by the many methods proposed by various authors, including the use of frozen section,11 fine-needle aspiration,12 "finger test,"13 ultrasound,14 magnetic resonance imaging,15 and laboratory tests. Although conventional radiography may demonstrate soft tissue gas, only 13.8% of our cases with available x-rays had positive findings (Fig. 1). The Laboratory Risk Indicator for Necrotizing Fasciitis score16 is one such scoring system with reported positive predictive value of 92.0% and negative predictive value of 96.0%. It is a scoring that is based on the leukocyte counts, C-reactive protein levels, hemoglobin, serum sodium, creatinine, and glucose. The main potential of this score is for use in patients who present primarily from the community setting with suspected necrotizing fasciitis. However, we find that in our series where more than one third of our patients were inpatients with multiple medical problems with deranged serum electrolytes or ongoing nosocomial infections (pneumonia and urinary tract infections), this scoring system has limited use. Hence, frequent serial review, enhanced with noninvasive tests such as ultrasound (if expertise is available), may be a good option to exclude necrotizing fasciitis when the diagnosis is uncertain, particularly so during the early stages of the disease. Invasive methods such as fine-needle aspiration and frozen section of tissue samples taken from the bedside have yet to receive widespread acceptance and practice.
Although necrotizing fasciitis is now recognized to be often a polymicrobial disease,17 upper extremity necrotizing fasciitis can often present with monomicrobial disease.18,19 In the literature where many of the series involved patients who have necrotizing fasciitis involving the trunk, perineum, and extremities, the average number of bacterial isolates per patient has ranged from 1.95 to 5.15.4,20,21 However, in our series, the average number of isolates was only 1.4, with 16 (55.2%) of 29 cases being monomicrobial. This was comparable to the series on upper limb necrotizing fasciitis by Schecter et al,6 where the average number of isolates was 2, and 19 (57.6%) of 33 cases were monomicrobial in nature. Gonzalez et al8 reported 3 (25%) of 12 cases with monomicrobial disease. Only one of our specimens had a positive anaerobic culture. This would have contributed to our overall lower average number of isolates as well. Brook and Frazier17 reported culture of anaerobic bacteria only in 22% of specimens, and mixed aerobic-anaerobic floras were recovered in 68% specimens. However, most of the anaerobic bacteria were cultured from the buttocks, trunk, neck, external genitalia, and inguinal areas. Our low anaerobic culture yield was partly contributed to proper collection and transport of specimens for anaerobic culture in only 15 cases. The need for proper collection and culture of tissue specimens of aerobic and anaerobic bacteria must be emphasized. This also indicates the need for adequate laboratory support and provision of suitable culture medium, especially outside of normal working hours, when much of the emergency surgery for necrotizing fasciitis is done. Fungal cultures should also be considered in patients with diabetes and other patients who are immunocompromised or have multiple comorbidity. This is particularly so as antibiotic therapy needs to be guided by cultures. Of concern is that antibiotic-resistant bacteria and fungi have also been reported to cause necrotizing fasciitis.7-10 We had 6 patients with methicillin-resistant S. aureus and 2 patients with Candida spp cultured from their tissue specimens.
Two patients with group B streptococcus necrotizing fasciitis were additionally treated with intravenous immunoglobulin. Intravenous immunoglobulin mode of action includes antigen neutralization, bacterial opsonization, and cytokine modulation and has been shown to have benefit in the treatment with potential on reducing mortality in streptococcal infections causing toxic shock syndrome and necrotizing fasciitis.22
We achieved wound closure of the upper limb soft tissue defect after debridement in 21 patients. Seven patients (24.1%) required amputation to control infection (Table 2). Reasons for amputation included ascending infection not controlled by repeat debridement, exposed joints (eg, wrist joint) after initial debridement, and patients in severe sepsis in whom we performed early amputation to control the septic foci and to reduce the number of surgical debridements and, hence, the risks associated with each associated anesthetic event. Schecter et al6 and Gonzalez et al8 reported upper limb amputation rates of 6% and 25%, respectively. Gonzalez23 emphasized the need for amputation to control a severe infection and when a limb lacked useful function after radical debridement, or shoulder disarticulation to prevent spread of upper extremity infection into chest wall to save life.
The role of hyperbaric oxygen therapy is still controversial. Until now, there are no prospective randomized trials demonstrating the efficacy of hyperbaric oxygen therapy in the treatment of necrotizing fasciitis.5 Most studies are retrospective and with small sample sizes. Brown et al24 suggested that hyperbaric oxygen therapy may be reserved for most ill patients, for whom this treatment modality may lead to improved outcome. Should it be used, however, this modality is an adjunct and must not preclude early and aggressive surgical debridement.6
Our mortality rate of 10.3% compares favorably to other series. One patient with a background of hepatitis and lymphoma died of postoperative cardiac collapse. The second patient died 1 month after discharge from a cardiac event. The third case was an elderly gentleman with concurrent upper and contralateral lower limb necrotizing fasciitis. Despite upper limb amputation, he unfortunately succumbed to overwhelming septicemia 11 days after admission. Reports of mortality rates resulting from necrotizing fasciitis, in general, range from 9.3% to 76%.3,19 A meta-analysis of 14 studies by Callahan et al,25 totaling 660 cases, revealed an overall mortality rate of 26%. Pessa and Howard4 noted that extremity infection had a mortality rate of 18%, compared with 44% and 33% in abdominal and perineal infections, respectively. Schecter et al6 reported a 21% mortality rate in upper limb necrotizing fasciitis. Multiple risk factors associated with increased mortality include advanced age, delays in diagnosis and treatment, diabetes mellitus, peripheral vascular disease, poor nutrition, immunocompromised state, obesity, intravenous drug abuse, and infections of the trunk and perineum.26 Using multivariatelogistic regression analysis, Wong et al7 found that the only factor that independently affects survival was a delay in admission to operative debridement of more than 24 hours (P < 0.05, relative risk = 9.4). However, in our 3 patients who died, all underwent surgery between 3 and 9 hours after consultation. On the other hand, the onset of their disease had started between 6 and 21 days before surgery. This long delay before treatment could have contributed to their poor outcome.
The treatment of established necrotizing fasciitis has not changed significantly over the years. Early diagnosis, resuscitation, broad-spectrum antibiotic therapy with subsequent modification according to cultures, and early and aggressive surgical debridement will always remain the foundations of treatment.5 The benefits of intravenous immunoglobulin have also been discussed.
However, there is potential to improve the survival and reduce the morbidity of this disease. Diabetic patients and immunocompromised patients are susceptible to infection, in general, and 58.6% of our patients were diabetic. Although most surgeries were performed within 24 hours of consultation, patients presented, on average, 5.8 days after disease onset. It is important that patients (especially diabetic patients and patients with chronically immunocompromised states) are educated by physicians and nurse practitioners regarding the importance of early and appropriate treatment of any penetrating injury. They should also be aware that even blunt trauma can cause microscopic breaks in the skin.5 Daily inspection of the extremities are also required as diabetic patients with neuropathy often do not realize that they have sustained injury during the course of the day. Particular to the Asian culture is the widespread acceptance of traditional medical therapy as appropriate primary treatment of disease especially in the elderly and rural populations.27,28 All these can lead to delay in appropriate treatment of necrotizing fasciitis.
Meanwhile, physicians must continue to be vigilant and maintain a high index of suspicion when patients with known risk factors present with nonspecific complaints of soft tissue swelling, pain, and erythema of the limbs. If necrotizing fasciitis cannot be ruled out, precautions must be made for serial review, and considerations be made for bedside diagnostic testing, if the expertise is available. Physicians and nurse practitioners should also maintain strict aseptic technique when inserting intravenous and intra-arterial lines, with subsequent monitoring of these sites for any early signs of infection. Patients who have diabetes or who are immunocompromised and require bandaging or splinting in the outpatient setting should also be given adequate advice regarding the affected limbs and should be instructed to seek early medical advice should significantly worsening pain, swelling, and erythema occur in the affected limb.
In conclusion, although necrotizing fasciitis has always been associated with significant morbidity and mortality, we believe that, with patient education, vigilance of medical personnel, as well as strict adherence to the pillars of treatment of necrotizing fasciitis, there is potential in reducing the incidence and improving the outcome of this dreadful disease.
The authors thank Dr Wong Chin Ho, Ms Jane Wong, and Ms Charmaine Tan for their assistance in the research of this article.
1. Patino JF, Castro D. Necrotizing lesions of soft tissues: a review. World J Surg
2. Wilson B. Necrotizing fasciitis. Am Surg
3. Fontes RA Jr, Ogilvie CM, Miclau T. Necrotizing soft-tissue infections. J Am Acad Orthop Surg
4. Pessa ME, Howard RJ. Necrotizing fasciitis. Surg Gynecol Obstet
. October 1985;161(4):357-361.
5. Green RJ, Dafoe DC, Raffin TA. Necrotizing fasciitis. Chest
. July 1996;110(1):219-229.
6. Schecter W, Meyer A, Schecter G, et al. Necrotizing fasciitis of the upper extremity. J Hand Surg [Am]
. January 1982;7(1):15-20.
7. Wong CH, Chang HC, Pasupathy S, et al. Necrotizing fasciitis: clinical presentation, microbiology, and determinants of mortality. J Bone Joint Surg Am
. August 2003;85-A(8):1454-1460.
8. Gonzalez MH, Kay T, Weinzweig N, et al. Necrotizing fasciitis of the upper extremity. J Hand Surg [Am]
. July 1996;21(4):689-692.
9. Elliott DC, Kufera JA, Myers RA. Necrotizing soft-tissue infections: risk factors for mortality and strategies for management. Ann Surg
10. Wai PH, Ewing CA, Johnson LB, et al. Candida
fasciitis following transplantation. Transplantation
11. Stamenkovic I, Lew PD. Early recognition of potentially fatal necrotizing fasciitis: the use of frozen section biopsy. N Engl J Med
12. Lille ST, Sato TT, Engrav LH, et al. Necrotizing soft tissue infections: obstacles in diagnosis. J Am Coll Surg
. January 1996;182(1):7-11.
13. Andreasen TJ, Green SD, Childers BJ. Massive soft-tissue injury: diagnosis and management of necrotizing fasciitis and purpura fulminans. Plast Reconstr Surg
14. Chau CLF, Griffith JF. Musculoskeletal infections: ultrasound appearances. Clin Radiol
. February 2005;60(2):149-159.
15. Schmid MR, Kossman T, Duewell S. Differentiation of necrotizing fasciitis and cellulitis using MR imaging. AJR
16. Wong CH, Khin LW, Heng KS, et al. The LRINEC (Laboratory Risk Indicator for Necrotizing Fasciitis) score: a tool for distinguishing necrotizing fasciitis from other soft tissue infections. Crit Care Med
. July 2004;32(7):1535-1541.
17. Brook I, Frazier EH. Clinical and microbiological features of necrotizing fasciitis. J Clin Microbiol
18. Chelsom J, Halstensen A, Haga T, et al. Necrotising fasciitis due to group A streptococci in western Norway: incidence and clinical features. Lancet
. September 1994;344:1111-1115.
19. McHenry CR, Piotrowski JJ, Petrinic D, et al. Determinants of mortality for necrotizing soft-tissue infections. Ann Surg
20. Giuliano A, Lewis F Jr, Hadley K, et al. Bacteriology of necrotizing fasciitis. Am J Surg
21. Sudarsky LA, Laschinger JC, Coppa GF, et al. Improved results from a standardized approach in treating patients with necrotizing fasciitis. Ann Surg
22. Norrby SR, Norrby-Teglund A. Infections due to group A streptococcus: new concepts and potential treatment strategies. Ann Acad Med Singapore
23. Gonzalez MH. Necrotizing fasciitis and gangrene of the upper extremity. Hand Clin
. November 1998;14(4):635-645.
24. Brown DR, Davis NL, Lepawsky M, et al. A multicenter review of the treatment of major truncal necrotizing infections with and without hyperbaric oxygen therapy. Am J Surg
25. Callahan TE, Schecter WP, Horn JK. Necrotizing soft tissue infection masquerading as cutaneous abscess following illicit drug injection. Arch Surg
. August 1998;133(8):812-827.
26. Voros D, Pissiotis C, Georgantas D, et al. Role of early and extensive surgery in the treatment of severe necrotizing soft tissue infections. Br J Surg
27. Wee HL, Ho HK, Li SC. Public awareness of diabetes mellitus in Singapore. Singapore Med J
28. Shahar S, Earland J, Abdul Rahman S. Social and health profiles of rural elderly Malays. Singapore Med J