Chronic idiopathic thrombocytopenic purpura: primum non nocere.

The vast majority of children with idiopathic thrombocytopenic purpura, dramatic though their signs may be at presentation, remit within weeks, with or perhaps despite medical attention. Such children, when referred to the Queen Elizabeth Hospital, Hackney Road by their family doctor, or appearing in the busy casualty department, have a careful history taken with particular emphasis given to the length of history, antecedent infections, and drugs. They are examined for evidence of bleeding, injury, enlargement of lymph nodes, liver, and spleen. A full blood count and coagulation screen are performed and the film is carefully examined. If there is the slightest doubt about the diagnosis or if specific treatment is contemplated a bone marrow aspiration is performed. Treatment involves a period of relative rest and the avoidance of aspirin. The parents are told that this is not leukaemia, that the risk of serious bleeding is very low, and that most children improve spontaneously within weeks to months. A follow up appointment is given to facilitate further discussion. Further active treatment is rarely indicated. The debate about steroid treatment in acute idiopathic thrombocytopenic purpura)1 2 has to a large extent been supplanted by that about the benefits of intravenous immunoglobulin,3 4now alas widely advertised as desirable treatment for acute idiopathic thrombocytopenic purpura. As the mortality of acute idiopathic thrombocytopenic purpura is less than 1% it would take a controlled trial of at least 14 000 patients to show a significant reduction in mortality,5 and estimation of morbidity is difficult when symptoms usually remit before any alteration in platelet count. It is thus regrettable that a recent clinical trial comparing intravenous immunoglobulin and steroids did not include a control arm.4 6 While it may be justifiable to give steroids or immunoglobulin (or both) as a short term measure to a child with serious bleeding or extensive mucous membrane haemorrhage at presentation, there is no firm evidence that either is beneficial or, as has been claimed, decreases the risk of chronic thrombocytopenia. Intravenous immunoglobulin is extremely expensive (approximately £30 per gram or £720 per course for a 15 kg child) and although heat treated, carries the theoretical risk of transmission of viral infections, particularly non-A non-B hepatitis.7