Article Text


Acute ataxia complicating Langerhans cell histiocytosis
  1. J Pritchard
  1. Royal Hospital for Sick Children, Edinburgh, UK; drjpritchard{at}

Statistics from

Some of the statements in the interesting short report by A Polizzi et al1 can be challenged. It is incorrect to suggest that cerebellar ataxia has been reported “only occasionally” in children and that it is commoner in adults with Langerhans cell histiocytosis (LCH). Diabetes insipidus is the only CNS complication that is more common than cerebellar disease and though the precise relative incidence of cerebellar ataxia in children and in adults is unknown, because all published series are institution based, there is no reason to suspect that proportionately more adults suffer this complication. It is also misleading to suggest that the patient described by Polizzi et al represents a “unique” occurrence. Cerebellar ataxia may be present at diagnosis or appear during follow up and may be progressive or static. More details of the clinical and pathological spectrum of CNS involvement by LCH can be found in a recent review.2

As the authors point out, pituitary-hypothalamic axis involvement is caused by direct infiltration of these structures by pathological Langerhans cells (“LCH cells”) and accompanying inflammatory cells. In patients who develop ataxia, cerebellar biopsy usually reveals only gliosis and demyelination, but CD1a-positive cells have been demonstrated in a few instances. Therefore, it is likely that the cerebellar lesions are caused by “LCH cell” infiltration followed by cytokine and chemokine mediated neural damage. The same sequence, with fibrosis as the end point, occurs in the liver and lungs of other LCH patients. Immune mechanisms may also be involved, as suggested by Polizzi et al, because CD8-positive T cells are also found in the cerebellar biopsies (Grois NG, personal communication). It is unlikely, however, that they represent the primary pathogenetic process. In other words, it is improbable that cerebellar involvement represents a “paraneoplastic syndrome” (ie an autoimmune disorder), as suggested by Polizzi and colleagues, a view supported by the fact that cerebrospinal fluid (CSF) “anti-neural” antibodies have not been detected in two studies (Grois N et al and Donadieu J et al, unpublished observations).

The combination of reticulo-nodular pulmonary shadowing and ataxia, as in this case, makes the diagnosis of LCH a real possibility. Given the onset of fatal pneumothorax soon after the child’s discharge from hospital, a lung CT scan would almost certainly have shown cystic change and led to tissue diagnosis via lung biopsy or bronchial lavage. With chemotherapy, the prognosis for pulmonary LCH is usually favourable, even if complicated by pneumothorax.3

It is generally recommended that patients with suspected LCH are referred to paediatric oncologists and/or managed on the international evidence-based trials of the Histiocyte Society, which include recommendations for the management of CNS complications.2–4 More details of the Histiocyte Society’s activities and their contact address can be found on the Society’s website:


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.