CASE REPORT

A 58 year old Japanese woman, who had been receiving oral prednisolone treatment for SLE for 20 years, was admitted to our hospital because of a swollen and erythematous left leg. She had taken 5 mg prednisolone daily for more than 10 years and no other immunosuppressive agents. She had had liver cirrhosis due to hepatitis C virus infection five years earlier. In December 1999 she had small wound in her left ankle. There was widespread violaceous discolouration of the skin and scattered bullae over the left knee and thigh with oedema extending from the ankle to mid-calf (fig 1A). On admission, she was in septic shock and had multiple organ failure, including hepatic and renal dysfunction. Her vital signs on arrival were blood pressure 82/62 mmHg, body temperature 38.0°C, pulse rate 80 beats/min, and respiratory rate 18/min. She had jaundice without ascites. She was alert and answered questions properly. Laboratory investigations showed a white blood cell count of 21.6×10⁹/l with 54% segmented and 29% bands, a haemoglobin level of 131 g/l, a platelet count of 133×10⁹/l, C reactive protein (CRP) 130 mg/l, blood urea nitrogen 17.9 mmol/l of urea, creatinine 216 μmol/l, aspartate aminotransferase 135 IU/l, alanine aminotransferase 90 IU/l, and total bilirubin 136 μmol/l. Blood glucose was 5.8 mmol/l, and glycohaemoglobin was 0.052. Immunoserological analysis showed C3 0.4 g/l, C4 0.3 g/l, CH₅₀ 18.4 U/ml, and anti-double stranded DNA antibody 5 U/ml. Blood culture was negative for bacteria or fungi. Antibodies to hepatitis C virus and its RNA were both positive. There was no cryoglobulinaemia.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

(A) Preoperative appearance of the swollen and erythematous left leg with purpura, skin ulcer, and bullae formation. (B) Light micrograph of skin biopsy specimens including superficial fascia. The arrow indicates thrombosis of subdermal blood vessels. Arrowheads denote epidermal detachment. Blisters show massive inflammatory cell infiltration in subcutaneous tissue and superficial fascia. (Haematoxylin and eosin, original magnification ×200.)

Emergency debridement of the anterior, posterior, and lateral compartments of the left leg was performed. Methicillin sensitive Staphylococcus aureus and Serratia marcescens were recovered from necrotic tissue culture. Fosfomycin 4 g/day and sultamicillin tosilate 1.125 mg/day were substituted for the initial broad spectrum antibiotics. A skin biopsy specimen including subcutaneous tissue showed extensive infiltration of inflammatory cells along the superficial fascia, a characteristic of necrotising fasciitis (fig 1B). Over the ensuing days her vital signs and renal function improved once. The CRP levels decreased to 30–50 mg/l, but her liver dysfunction worsened and total plasma exchange was started on the 16th hospital day. She died of disseminated intravascular coagulation and multiple organ failure including hepatic failure eight weeks after the admission to hospital. There was no lupus flare up during the treatment of necrotising fasciitis.

DISCUSSION

Infection is a major cause of death in patients with SLE.² Active SLE itself contains many forms of immunological dysfunction, which may account for the increased susceptibility to infection. The abnormalities include acquired and inherited complement deficiencies, immunoglobulin deficiency, defects in chemotaxis, phagocytic activity, and delayed hypersensitivity. Even in clinical remission, an increased tendency to develop infections has been reported among patients with SLE.³ The decreased complement level in our patient might have been caused by consumption due to severe infection because there were no clinical symptoms of lupus flare up before and after admission to hospital.

Treatment with corticosteroids and other immunosuppressive agents may further have affected the increased susceptibility to infection. Corticosteroids decrease inflammatory response, effector cell response in cell mediated immunity, and immunoglobulin synthesis.⁴ Necrotising fasciitis is uncommon but might occur in SLE during immunosuppression due to high dose corticosteroid treatment.^5,6 In this condition Serratia marcescens might be a potential pathogen causing necrotising fasciitis in susceptible hosts.⁶ In our patient, prolonged steroid administration might have increased the risk of infection and caused the fragility of skin tissue, which in turn allowed easy access of bacteria into the circulation. Liver cirrhosis may be associated with necrotising fasciitis caused by an infection of enteric bacterium such as Escherichia coli or Klebsiella species through ascitic fluid.⁷ This possibility is unlikely in our patient because she had no ascites while in hospital. Liver failure instead affected the decreased complement production in our patient.

The case presented here suggests that long term administration of corticosteroids even in a low dose may be a risk factor for fatal infection such as necrotising fasciitis in patients with SLE.