Colchicine for Sweet's syndrome. A study of 20 cases

S ir, Corticosteroids have been used for the treatment of Sweet's syndrome (acute febrile neutrophilic dermatosis) since it was described in 1964.1 They represent the standard therapy and for some are included in the diagnostic criteria.2 In 1981, Suehisa and Tagami first proposed colchicine in this indication,3 but its use has rarely been reported in the literature. We present our experience of this treatment in 20 cases of Sweet's syndrome. The notes of patients with a diagnosis of Sweet's syndrome from 1984 to 1996 in our department were reviewed retrospectively. All of them were treated with colchicine. The diagnostic inclusion criteria were: the presence of typical skin lesions, i.e. painful, raised erythematous plaques; and of characteristic histopathological changes, e.g. a predominantly neutrophilic infiltration in the dermis without a leucocytoclastic vasculitis. All patients had a pyrexia > 38 °C, a raised white cell count composed of > 70% segmented nucleated neutrophils and an erythrocyte sedimentation rate of > 30 mm in the first hour (normal < 10). These constitute Su and Liu's minor criteria for the diagnosis for Sweet's syndrome.2 Twenty patients were identified (12 women, eight men), aged between 29 and 93 years (median 59). Colchicine was introduced 2–8 days after the first cutaneous signs at a dose of 1.5 mg/day (16 patients) or 1 mg/day (four patients). The dose of 1 mg/day was given to the first patients and was subsequently increased to 1.5 mg/day as recommended by Suehisa and Tagami,4 as it was well tolerated. The mean period of the treatment was 15 days (range 10–21). Aside from clinical signs required as inclusion criteria, seven patients had arthritic arthralgia, three conjunctivitis and two myalgia. In two patients liver function tests were deranged, with an aspartate aminotransferase level two to four times the upper limit of normal, and there was a two‐ or threefold elevation of the gamma glutamyl transpeptidase level in four patients. In one patient, pulmonary tuberculosis was simultaneously diagnosed. No association with malignancy was noted. The Sweet's syndrome responded to colchicine in only 18 cases. Apyrexia was obtained in 24–72 h, the cutaneous lesions were attenuated in 2–5 days and the arthralgia disappeared in 2–4 days. Laboratory tests returned to normal after treatment. This took 8–14 days in the case of the leucocytosis and 1 month in the case of the liver function tests. The erythrocyte sedimentation rate normalized by the end of the therapy in nine patients, and markedly decreased in the others. None of the patients suffered a relapse after discontinuation of colchicine. Colchicine was ineffective in two patients. After initial attenuation of lesions, new cutaneous lesions and pyrexia necessitated replacement of colchicine with oral corticosteroids (prednisone 1 mg/kg per day) in one patient and with potassium iodide (900 mg/day) in the other. After this change of therapy, the cutaneous lesions were attenuated in 1 and 4 days, respectively, and completely disappeared in 2–3 weeks. Colchicine was well tolerated except in one patient who had diarrhoea and vomiting on the third day of treatment, which was controlled with symptomatic therapies, without needing to stop colchicine. We contacted 16 of the 20 patients in 1998. None reported a relapse, a haematological problem or had developed cancer. The follow‐up was 2–10 years (median 8.5). The standard therapy for Sweet's syndrome is oral corticosteroids at an initial dose of 0.5–1 mg/kg per day. This therapy leads to a remission of the general malaise within hours, and of the skin lesions within 2–5 days. The treatment period varies from 1 to 12 weeks. Despite the good initial response, the disease is characterized by relatively frequent recurrences in 30% of cases, according to Kemmet and Hunter.5 On the other hand, Gunawardena et al.6 noted no recurrence, as in our study. Other therapies, principally including indomethacin and potassium iodide, have been proposed. In a study of 18 cases, the success of indomethacin was noted in 17, causing epigastric pain in two patients.7 Potassium iodide is also often used, at a dose of 900 mg/day.8 The efficacy of colchicine in Sweet's syndrome is well known, but rarely reported in the literature. The first reports were by Suehisa and Tagami in two articles in which good results were seen in one and three patients, respectively.3, 4 Recently, the efficacy of this therapy had been noted in a study of 55 cases where 33 patients were given colchicine at a dose of 1–1.5 mg/day.9 Our retrospective analysis confirms the success of colchicine with low doses (1–1.5 mg/day), independent of the severity of the disease. The response to therapy is relatively rapid in 2–5 days and a cure is obtained in 1–2 weeks. None of the patients had a relapse after discontinuation of colchicine and the tolerance was excellent. No association with malignancy was noted. However, idiopathic Sweet's syndrome does not seem to differ from that associated with a malignant pathology.10 We noted one case of pulmonary tuberculosis with Sweet's syndrome; this association has already been described (Palmer GD, Kheir S, Sams M, personal communication, 1982). Gastrointestinal problems, e.g. diarrhoea, nausea or vomiting, are the most frequent side‐effects of colchicine and are dose dependent. They were only observed in one patient and did not necessitate discontinuation of the treatment. Severe side‐effects have been infrequently reported at therapeutic doses: they occur exceptionally, generally in patients who suffer from renal or hepatic insufficiency. These severe side‐effects are similar to acute intoxication and can be gastroenteritis, haematological problems or myositis. These potential side‐effects must be compared with those of corticosteroids, potassium iodide or indomethacin. We thank Juliet Littlewood and Alex Kelly (Liverpool, U.K.) for technical assistance.