Pneumocystis cariniipneumonia in a patient with pyoderma gangrenosum treated with cyclosporin

Sir, We read with interest the recent report on Pneumocystis carinii pneumonia (PCP) in dermatology patients1 and would like to report a case of PCP in a patient with aggressive pyoderma gangrenosum treated with systemic corticosteroids followed by cyclosporin. A 69‐year‐old caucasian woman presented with rapid onset of several ulcerating lesions on the limbs. Full clinical examination was normal. An incisional skin biopsy was consistent with the clinical diagnosis of pyoderma gangrenosum. Haematology, renal and liver function, urinalysis, autoantibodies including perinuclear and cytoplasmic immunofluorescent pattern antineutrophil cytoplasmic antibodies, complement levels and cryoglobulins were normal or negative. Serum electrophoresis showed a monoclonal IgA band but further investigations ruled out a diagnosis of multiple myeloma. Computed tomographic scanning of the thorax, abdomen and pelvis was normal. She was commenced on a reducing dose of oral prednisolone (from 0·5 mg kg−1 daily) with an initial improvement in her skin lesions. Three months later, she developed multiple painful, necrotic lesions on the trunk and limbs. There was no evidence of systemic vasculitis or infection. The dose of oral prednisolone was increased, but new lesions continued to appear. Because of the aggressive nature of the disease, oral cyclosporin (3 mg kg−1 daily) was started in view of recent reports of its efficacy2,3 in pyoderma gangrenosum. Unfortunately, she continued to deteriorate, despite a subsequent increase in the dose of cyclosporin to 5 mg kg−1 daily. At this stage, she was apyrexial with an unremarkable physical examination except for her skin lesions. In particular, clinical examination of the respiratory system was normal. Biochemical tests showed acute renal failure and the cyclosporin was immediately stopped. Chest X‐ray showed bilateral interstitial alveolar shadowing (Fig. 1) highly suggestive of PCP, and arterial blood gas sampling showed significant hypoxia. The diagnosis of PCP was confirmed by bronchoalveolar lavage aspirate showing the typical organisms on silver staining. High‐dose intravenous sulphamethoxazole‐trimethoprim (Septrin®, Glaxo Wellcome, Uxbridge, U.K.) and continuous positive airway pressure ventilation was immediately instituted. Despite intensive supportive therapy she died 5 days after the diagnosis of PCP was made. A request for a postmortem examination was declined by the relatives. Chest X‐ray showing bilateral alveolar shadowing sparing the apices, with a ground glass appearance typical of Pneumocystis carinii pneumonia. This case illustrates the rapidly fatal course of PCP in an immunosuppressed dermatology patient. While we initially suspected that she may have had an underlying malignancy, this was not confirmed on extensive investigation and the only associated finding was of monoclonal IgA gammopathy which is a well‐recognized association of pyoderma gangrenosum.4 She had no risk factors for human immunodeficiency virus (HIV) infection and a HIV test or CD4 count was not performed. We concluded that she was immunosuppressed solely due to her treatment with prednisolone and cyclosporin. While we agree with Raychaudhuri and Siu1 that there is a dearth of literature on PCP in dermatology patients, this is presumably because PCP occurring outside the setting of HIV infection, organ transplantation or malignancy is very rare. Our view is supported by the authors' own data on only four patients with dermatological diseases who developed PCP over a 16‐year period in their region of Northern California. Suggested risk factors for the development of PCP based on our own case include combined immunosuppressive regimens involving cyclosporin, multiple sequential immunosuppressive agents, and the development of a relative or absolute lymphopenia. It has been suggested that dermatologists should have a policy on PCP prophylaxis as adopted in the HIV and postorgan transplant population.1 There is evidence that the mortality associated with PCP in the HIV‐negative population is greater than in those who are HIV positive,5 particularly in those individuals with underlying malignancy.6 In addition, the risk of developing PCP may be higher in those non‐HIV‐positive patients who have been immunosuppressed with cyclosporin.7 However, the risk of PCP in the dermatology population must be weighed against the risk of adverse effects from prophylactic regimens; for example, there is a high incidence of adverse effects, especially rashes, with Septrin®. Also, methods of tailoring the regimen for PCP prophylaxis to the population at risk while avoiding undue adverse effects are still debated.8 Therefore, at present we would not advocate PCP prophylaxis for dermatology patients on immunosuppressive therapy. However, we would suggest that those using immunosuppressive agents for patients with difficult dermatological conditions should have a high index of suspicion. PCP may present with non‐specific symptoms such as malaise or confusion rather than the classical dry cough, fever and breathlessness. In addition, auscultation of the chest is often normal, even with widespread alveolar shadowing on the chest X‐ray, as in our patient. Because respiratory examination may be unhelpful and even falsely reassuring, chest X‐ray is mandatory with urgent referral to a respiratory unit for full assessment if indicated.