Étude anatomo-clinique du purpura : y a-t-il un intérêt à biopsier un purpura palpable ?

Abstract

For many physicians, palpable purpura is synonymous with vasculitis. However, a skin biopsy is almost always performed in common clinical practice in order to confirm the diagnosis. The aim of our study was to assess whether palpable purpura is always indicative of an inflammatory infiltrate in a vessel wall. Eighty-seven patients were included in this prospective monocentric study, 45of whom were presenting a palpable purpura. Patients were classified in two categories: "leukocytoclastic vasculitis" or "other diagnosis". The clinical and histopathological features of patients with a palpable purpura were studied. The mean age of patients presenting a palpable purpura was 69years. There were 26men and 19women. Of the 43patients biopsied, 37were included in the vasculitis group. The sensitivity, specificity, positive predictive value and negative predictive value for a diagnosis of vasculitis in patients with palpable purpura were respectively 82, 65, 86and 58%. The Odds ratio was 8.48 (95% CI, 2.52-31.80; P<0.05). Most of the palpable purpuras examined were indeed related to leukocytoclastic vasculitis. In the remaining cases, biopsy did not contribute to the diagnosis since it only showed purpura without vessel wall inflammation. In our opinion, a skin biopsy is thus not essential where the clinical presentation is typical.