rss
J Neurol Neurosurg Psychiatry 57:1334-1342 doi:10.1136/jnnp.57.11.1334
  • Research Article

Herpes simplex encephalitis: long term magnetic resonance imaging and neuropsychological profile.

  1. N Kapur,
  2. S Barker,
  3. E H Burrows,
  4. D Ellison,
  5. J Brice,
  6. L S Illis,
  7. K Scholey,
  8. C Colbourn,
  9. B Wilson,
  10. M Loates
  1. Wessex Neurological Centre, Southampton, UK.

      Abstract

      The first comprehensive in vivo documentation of the long term profile of pathological and spared tissue is described in a group of 10 patients with a diagnosis of herpes simplex encephalitis, who were left with memory difficulties as a major residual sequel of their condition. With a dedicated MRI protocol, which included high resolution images of temporal lobe and limbic system areas, data are provided on structures that have recently gained importance as anatomical substrates for amnesia. The major features of the lesion profile were: (1) unilateral or bilateral hippocampal damage never occurred in isolation, and was often accompanied by damage to the parahippocampus, the amygdala, specific temporal lobe gyri, and the temporal poles; (2) the insula was always abnormal; (3) neocortical temporal lobe damage was usually unilateral or asymmetric. It never occurred in isolation, and was invariably associated with more medial pathological changes; (4) anterior and inferior temporal lobe gyri were damaged more often and more severely than posterior and superior temporal lobe gyri; (5) pronounced abnormality was often present in the substantia innominata (region of the basal forebrain/anterior perforated substance); (6) there was evidence of significant abnormality in the fornix; (7) there was evidence of damage to the mammillary bodies; (8) thalamic nuclei were affected in around 50% of cases, with damage usually unilateral; (9) frontal lobe damage was present in a few patients, and affected medial areas more than dorsolateral areas; (10) there was some involvement of the striatum, although this was usually unilateral and mild; (11) there was usually limited involvement of the cingulate gyrus and of the parietal and occipital lobes; (12) the cerebellum and brain stem were never damaged. Lesion covariance analysis indicated a close relation between the presence of abnormalities in temporal lobe and limbic-diencephalic regions. Unlike severe head injury, lesions in the temporal pole were not associated with the presence of lesions in the orbitofrontal cortex. Long term neuropsychological impairments were characterised by a dense amnesia in 60% of cases, and a less serve but noticeable anterograde memory impairment in the others. Naming and problem solving deficits were found in a small number of cases. Only two patients were able to return to open employment. Severity of amnesia showed a significant relation with severity of damage to medical limbic system structures such as the hippocampus, with bilateral damage being particularly important. By contrast, there was a minimal relation between memory loss and severity of damage to the thalamus, to lateral temporal lobe areas, or to the frontal lobes.