Article Text

This article has a correction. Please see:


Hypothalamic amnesia and frontal lobe function disorders after Langerhans cell histiocytosis
  1. L Manning1,
  2. F Sellal2
  1. 1LN2C (UMR 7521), Université Louis Pasteur, 12 rue Goethe, 67000 Strasbourg, France,
  2. 2Service de Neurologie, Hôpitaux Universitaires, Strasbourg, France
  1. Correspondence to:
 Dr L Manning; 

Statistics from

Langerhans cell histiocytosis (LCH), a rare disease previously known as histiocytosis X, is characterised by abnormal cell proliferation. If the CNS is invaded, the hypothalamus is the typical site location. There are virtually no neuropsychological data on hypothalamic LCH sequelae. Memory disorders in the context of posterior but also anterior hypothalamic lesions, regardless of the aetiology, have in most cases been attributed to the involvement of the mammillary bodies (MB). However, Ptak et al reinterpreted a case, acknowledging that “Damage of the anterior hypothalamus, rather than the mammillary bodies, may [ . . .] have been responsible for [the observed] confabulatory amnesia” (page1600).1 An interesting question is whether neuropsychological deficits are secondary to hypothalamic damage in itself or to a disconnection syndrome. The latter is based on the bilateral hypothalamic nucleus—widespread brain complex connections. Two subsystems are of particular interest in the present case, the hippocampus-fornix-hypothalamus-MB circuit, and the amygdala-stria terminalis plus caudate nucleus-hypothalamus circuit.

We report the case of a patient who presented with hypothalamic LCH and underwent thorough neuropsychological, radiological, and metabolic assessments.

Case report

A right handed woman (date of birth 1931), was admitted (March 2000) to the Hôpitaux Universitaires’ Neurology Unit (Strasbourg, France) complaining of memory deficits. In 1990, bronchopulmonary biopsies had led to the diagnosis of LCH. In 1992, she was diagnosed as having hepatic and hypothalamic LCH, the first being confirmed by biopsy, the seond suggested by diabetes insipidus. Endocrinological assessment showed signs of anterior pituitary dysfunction. Standard biological and physical examinations were normal. MRI showed a bilobulated hypothalamic tumour, extending from the optic chiasm to the posterior part of the third ventricle floor and towards the pituitary stalk. It displaced the left thalamus very slightly and compressed both MB, which became non-identifiable. There was no abnormal signal in the thalami and the mesial temporal regions were morphologically normal. Cortical atrophy was normal for her age. MRI follow up (1998, 2000, 2001), showed stable lesions (20×18×14 mm; fig 1). An 18F fluorodeoxyglucose resting positron emission tomography (PET) scan (December 2000) revealed small hypometabolic regions in the ventromedial prefrontal cortices, left superior frontal gyrus, parietal lobe, caudate nucleus and upper brain stem, plus a pronounced hypothalamic hypermetabolism (fig 2).

Figure 1

T1 weighted coronal MR image showing a circumscribed hypothalamic tumour.

Figure 2

PET scan. Frontal, parietal, caudate nucleus, and upper brain stem hypometabolism, and hypothalamic hypermetabolism. Conditions were contrasted with 12 normal subjects (p<0.001).

Neuropsychological investigation (July 2000)

Written informed consent was obtained. The patient was disorientated for time only. Verbal IQ (90) and Performance IQ (96), language, constructional praxis, visuoperceptual, and spatial abilities were normal. Performance on anterograde and retrograde (episodic and semantic) memory tests was severely impaired. Likewise, scores on six tests sensitive to frontal lobe dysfunction were uniformly defective. Moreover, she had recently become “hostile and irritable”. Consequently, her husband completed a personality change scale (J Barrash, et al, 25th INS meeting, Orlando, Florida, February 1997). The considerable difference between her past (63) and present (110) behavioural characteristics was compatible with a frontal lobe disorder.


Secondary to hypothalamic LCH, our patient developed amnesia, dysexecutive syndrome, and personality alteration, in an otherwise normal context.

Ptack et al described a similar case (amnesia and frontal lobe dysfunction) in the context of circumscribed hypothalamic lesions.1 The postmortem diagnosis of sarcoidosis did not exclude the possibility of LCH. As the patient’s lesion involved mostly the medial hypothalamus, the authors accounted for the symptoms in terms of the strong reciprocal connections with the anterior limbic structures. Further studies have demonstrated connections from the medial prefrontal cortex to the anterior and ventromedial hypothalamus and those from the orbital region to the lateral hypothalamus.2

In this case, the large lesion (optic chiasm, pituitary stalk, and MB), restricted to the hypothalamus, renders the findings particularly interesting in terms of hypothalamic connections. Given the evidence that LCH in this case is not an infiltrating but an encapsulated form of the disease, the small cortical zones of hypometabolism are not thought to be directly caused by the histiocytosis condition. They could be accounted for in terms of disconnections resulting from the hypothalamic lesion. The deficit in the memory consolidation process could reflect a dysfunction of the hippocampus-fornix-MB circuit. Failure of retrieval processes could reflect the prefrontal-hypothalamic connections. Moreover, the patient’s false recognitions and confabulations showed the frontal involvement suggested by Ptak et al.1 However, for the time being, we are unable to account for the parietal and upper brain stem hypometabolism.

The hypothalamic hypermetabolism is interpreted in terms of an index of the active ongoing disease process.3

In conclusion, we suggest that our patient’s neuropsychological impairments, with their catastrophic consequences for daily life, require abnormality beyond the MB and might reflect the “superadditive” effect of damage at different, strategically important loci.4 Indeed, theoretical considerations based on the mere “addition” of hypothalamic lesions plus a few very small hypometabolic zones at various sites in the brain would most probably not have predicted our patient’s present cognitive status. Finally, given the paucity of hypothalamic LCH cases reported in the literature and the virtual absence of neuropsychological examination in those rare cases, the possibility of an as yet unknown specific effect of LCH on the CNS cannot be ruled out.


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.

Linked Articles