Effect of the Anabolic Steroid Oxandrolone on the Rate of Catabolism in Acute Necrotizing Fasciitis
Disclosures: Dr. Demling is on the speakers bureau of BTG Pharmaceuticals Inc.
Necrotizing fasciitis is defined as a necrotizing nonclostridial soft-tissue infection spreading along fascial planes.[1–6] The thrombosis of local blood vessels leads to devascularization and necrosis of overlying subcutaneous tissue and skin. The infection is usually caused by a combination of Gram-positive and Gram-negative aerobic and anaerobic organisms.[1–6]
After early aggressive surgical debridement has been performed, a large open wound results, which often takes several weeks to months to close. The combination of a large wound and residual soft-tissue inflammation should lead to a profound catabolic state comparable to a severe burn resulting in a loss of lean body mass. Lost body protein can lead to an immune deficiency state, weakness, and impaired wound healing.[7–10] Although healthy patients can develop this disorder, a comorbid factor, such as diabetes, is often present.
The anabolic steroid, oxandrolone, has been shown to decrease catabolism and significantly attenuate lost lean mass after burn injury and other severe wounds.[12–14] There are currently no studies on the degree of the catabolic response to necrotizing fasciitis or on the effect of adding an anabolic steroid to this process.
The authors’ purpose was twofold. The authors first wanted to determine the degree and time course of catabolism, weight loss, and wound healing rate in patients with necrotizing fasciitis. Secondly, the authors wanted to determine the effect of adding the anabolic steroid, oxandrolone (Oxandrin®, BTG Pharmaceuticals, Inc., Iselin, New Jersey), to nutritional support and aggressive debridement on parameters measuring the rate of catabolism.
Study protocol. Patients over 18 years of age with the diagnosis of acute necrotizing fasciitis were eligible for the study. Age, sex, and presence of diabetes or other chronic disease were identified. Also identified was the initial site of the necrotizing fasciitis. On admission, patients were randomized into a standard care or standard care plus oxandrolone group. All patients were initially fluid resuscitated, begun on imipenem-cilastatin 1gm every six hours, underwent an initial surgical debridement, and then were begun on nutritional support. Caloric intake was determined using indirect calorimetry, and protein intake was determined to be 1.5g/kg/day. Nutrition was provided by the oral route using a small nasogastric feeding tube to provide the high-calorie, high-protein intake. All patients were given their assessed nutrient needs.
On Day 4 when nutritional support had been initiated, patients were treated either with standard care or standard care plus oxandrolone 20mg/day in two divided doses until wound closure was complete. Exclusion criteria for this Institutional Review Board-approved study was the presence of an elevated prostate-specific antigen (PSA), as prostate cancer is considered a contraindication for the use of oxandrolone.
The microbiology of the wound was identified through biopsies obtained on the initial and subsequent debridement, as well as the site of involvement. Gram stains of the biopsies, which included the involved soft tissue and fascia, were also obtained. The need for further debridement of involved tissue was based on daily clinical assessment of the open wound.
Local wound care consisted of daily dressing changes consisting of moist gauze dressings on the wound surface. Debridement was performed with the changes of the gauze, as well as local sharp debridement. This approach was used until all necrotic tissue was removed and there was no evidence of remaining infection. The clean wound was then managed using a closed drainage system (vacuum-assisted closure, V.A.C.®, Kinetic Concepts, Inc., San Antonio, Texas) until the wound was ready for closure. Closure was performed using a combination of delayed primary closure, split-thickness skin grafts, and soft-tissue flaps.
Measurements. Metabolic rate was measured using indirect calorimetry (kcal/M2/hr). The degree of hypermetabolism was defined as the percent increase above resting energy expenditure, which was calculated by the Harris Benedict equation.
The degree of catabolism was defined by measuring body weight and also nitrogen balance. Nitrogen intake was calculated from the nutrients provided, and nitrogen loss was determined by urinary losses using 24-hour urine collections. Catabolic and metabolic rates at Days 7, 14, and 21 were compared between groups. These times were selected, as peak catabolic changes are usually present by Day 7, and the process is usually beginning to resolve by Day 21.
Since the value of weight loss was determined at discharge when the wounds were closed, other factors, such as wound exudative loss during the acute period, did not affect the measurement. Weight loss was defined as weight on admission if this preceeded resuscitation, or weight by history minus weight at discharge to home or to a rehabilitation center. If resuscitation had begun, then the initial body weight determined by history was used. Wound healing rate was determined by the complete reepithelization of the first split-thickness skin donor site used to cover the open wound. The donor site was standardized to 0.012 inches deep and covered with a transparent dressing (Tegaderm, 3M Health Care, St. Paul, Minnesota) until healed. The first skin graft usually occurred between 6 and 10 days after initial debridement. The fasciitis wounds were sufficiently different from patient to patient, especially location and size, that the wound closure rate could not be used as a systemic marker of healing rate and compared between groups. In addition, final wound closure was performed using a combination of skin grafts and flaps and not by wound contraction alone.
Complications. The complication that was monitored was liver dysfunction. Mild dysfunction was defined as up to a three-fold increase in the liver enzymes aspartate amino transferase (AST) and alanine aminotransferase (ALT), while severe dysfunction was considered as greater than a 10-fold increase in these enzymes plus a bilirubin exceeding 3mg/dL.
Funding. The study was funded by the Burn Trauma Center fund at the Brigham and Women’s Hospital.
Statistical analysis. Within-group and between-group differences were analyzed using ANOVA. The Dunnett’s t-test was used to compare different time periods. A p value of
Twenty-one patients were studied between 2000 and 2002. Ten had standard care provided, and 11 were also given oxandrolone. There were no demographic differences between the groups. Sixty percent of the patients were men. Mean age was 44±8 years. Sixty percent had adult-onset diabetes. All patients were admitted within 36 hours of onset of clinical infection, and initial surgical debridement was carried out within 24 hours of admission. The site of origin of the infection was the perineum in 60 percent of cases, and extremities were the second most common site at 25 percent (Table 1). All patients required two surgical debridements to remove all involved necrotic or infected tissue from the wound site. Further extension was not evident after the second procedure.
The microorganisms involved in 70 percent of cases were a mix of aerobic and anaerobic bacteria, including enterobacter, Escherichia coli, non-group A streptococcus, and Staphylococcus aureus. Group A streptococcus was present in 25 percent of cases. Clostridia perfringens was isolated in two patients but was not considered to be the predominant organism in either case. Primary clostridial infections were not studied, as by definition these infections are not considered under the diagnosis of necrotizing fasciitis. Typically, wound inflammation is not seen, and myonecrosis is present with a clostridial infection. Myonecrosis is characteristically not seen in necrotizing fasciitis. There were three deaths, two in the control and one in the oxandrolone group. A group A streptococcus infection was involved in all of the deaths. All deaths were the result of progressive multisystem organ failure.
There were no differences in nutritional intake in the two groups. Mean and daily caloric intake was 34±4cal/kg in control and 35±3cal/kg in the oxandrolone and control groups, respectively. Mean daily protein intake was 1.7g/kg and 1.6g/kg in the control and oxandrolone groups, respectively. There was significant weight loss in all patients during the hospital course. In addition, there was significant net nitrogen loss at the measurement periods of 7, 14, and 21 days after onset. Values in controls were 15±3, 16±2, 13±3g/day at the measured periods, while patients receiving oxandrolone had values of 5±2, 4±2, and 3±1g/day, respectively.
Measurements. There was a significant increase in metabolic rate in all patients. The increase in metabolic rate was approximately 70 percent above calculated resting energy expenditure (REE) and was the same for both groups (Table 2, Figure 1). Weight loss and nitrogen loss was significantly lower in the oxandrolone group compared to control (Figure 2). Debridement and reconstruction procedures were not different between groups. Donor site healing rate was significantly increased in the oxandrolone group compared to controls (Table 2).
Complications. Mild transient liver function changes, i.e., less than three-fold increase in AST and ALT, occurred in approximately one third of patients of both groups. Three cases of severe liver dysfunction were evident, all of which occurred in patients who died.
Necrotizing fasciitis is a form of severe necrotizing soft-tissue infection characterized by necrosis, gangrene, and gas present in the subcutaneous space.[1–6] The infection spreads along the fascial planes but usually does not affect the underlying muscle. The result is a local tissue infection and generalized inflammation characterized by fever, pain, hypovolemia, and, in one third of patients, shock state upon presentation. The site of infection usually takes several days to become clinically evident with the typical organisms being mixed aerobic and anaerobic bacteria. If group A streptococcus is involved, the course is usually more virulent. Although this process can occur in a healthy patient, a history of diabetes is common.[1,4,7] The time course to recovery is usually weeks to months due to the typically large area of tissue necrosis and surrounding wound inflammation.
The standard treatment is an early aggressive surgical debridement of all necrotic subcutaneous tissue and skin, along with systemic antibiotics, fluid resuscitation, and aggressive nutritional support.[1–4] Mortality rate has been reported to range between 20 to 40 percent in patients with early diagnosis and treatment. Mortality is much higher if diagnosis is delayed. All patients in the authors’ study were diagnosed early in the course, and mortality rate was 15 percent.
The authors found that necrotizing fasciitis has a metabolic response very similar to burn injury, as tissue necrosis and infection are present in both.[8,9,11] The presence of a large wound, often open for several weeks, and residual tissue inflammation are responsible. This large wound would be expected to lead to a prolonged stress response comparable to a severe burn. Elevations of catecholamines, glucagon, and cortisol and decreases in levels of the anabolic agents testosterone and human growth hormone result in the severe catabolic hypermetabolic response.[16,17] The resulting hypercatabolic state leads to a large loss of lean body mass and body protein.[10,11,15,17,18] The authors found that the increase in metabolic rate was about 70 percent above resting energy expenditure between Days 7 and 14 in all patients. Net nitrogen loss was approximately 15 grams per day in controls, which corresponds to approximately one pound of muscle loss per day. Optimum nutrition alone has been reported to decrease nitrogen loss by about 50 percent after burns and traumatic injury.[12–14] The authors noted a greater than 10-percent loss of body weight in their study caused by the soft-tissue infection.
Oxandrolone is an anabolic steroid used for restoring lost body weight and lean mass after injury or infection. This agent is 10-fold more potent as an anabolic agent then testosterone and also has negligible androgenic or masculinizing effects. Since the kidney clears oxandrolone rather than the liver like other steroids, liver dysfunction is rare.[20–22] Oxandrolone has also been shown to decrease nitrogen loss in catabolic states, such as burns and AIDS patients.23,24 In addition, wound healing rate, as measured by the reepithelization rate, has been shown to be increased with oxandrolone after major burn injury compared to nutrition alone.[11–13] Oxandrolone can produce modest initial water retention. However, this effect is very transient and would not have an effect on body weight at discharge.
The authors noted that the addition of oxandrolone significantly decreased the degree of weight loss and lean mass loss measured as nitrogen loss in patients with an acute necrotizing fasciitis. In addition, wound healing rate, measured as reepithelization of a split-thickness donor site, was significantly increased. Although these variables strongly suggest an overall improvement in outcomes when using oxandrolone, the authors are unable to state that the time to wound closure was decreased with the anabolic steroid. A much larger series would be needed in order to assess this variable due to the differences in wound size and location.
There were also insufficient patient numbers to be able to assess the effect of oxandrolone on mortality. However, mortality in necrotizing fasciitis usually occurs from an initial uncontrolled sepsis, a process that would likely not be affected by an anabolic steroid.
In summary, the authors demonstrated that necrotizing fasciitis produces a severe catabolic state with loss of body weight and lean mass even in the presence of optimum care. The addition of the anabolic steroid, oxandrolone, significantly decreases the degree of lost body protein as well as improves wound-healing rate.