Materials and methods

The Department of Neuropathology in this Institute has for many years had an arrangement with the Procurators Fiscal and departments of pathology in this region whereby intact brains from patients known to have been managed in the Department of Neurosurgery for a head injury are sent for detailed neuropathological examination and subsequent report to the forensic pathologists. Although most of these brains come from patients who have died soon after injury we have, because of our research interest, encouraged the sending of brains from patients who have died long after their injury. For the present study we have identified from our data base of over 1300 cases of blunt head injury collected between 1968 and 1998—used in numerous previous studies—a group of patients who had recovered with only moderate disability before dying of any cause, often one remotely connected to their injury.

All brains were fixed in 10% formal saline for a minimum of 3 weeks before dissection, after which a full macroscopical and microscopical examination was undertaken in each case.4 In addition to conventional histological techniques, an immunocytochemical method was used to identify reactive microglia (CD68, 1: 200; Dako developed with vectastain ABC kit and visualised with diaminobenzidine). Blocks for microglial studies were taken from the cerebral hemispheres (bilateral parasagittal, bilateral internal capsules, and corpus callosum) and the brain stem. Contusions were assessed quantitatively using the total contusion index. This index takes into account the depth and extent of contusions in all regions of the brain: 0 means that there were no contusions, a contusion index of less than 9 is indicative of minimal contusions whereas one of more than 37 is indicative of severe contusions.5 Diffuse axonal injury, in which there is widespread damage to axons throughout the brain at the time of the original injury, was graded as previously described.6 7 In grade 3 (most severe) there are in addition focal lesions in the corpus callosum and in the dorsolateral segment(s) of the rostral brain stem; in grade 2 there is a focal lesion only in the corpus callosum; and in grade 1 there are no focal lesions. Because of the long period of survival in every case, the diagnosis of diffuse axonal injury depended on the microglial reaction rather than the identification of actual abnormalities in axons. Despite the long periods of survival, there was in every brain an obvious residual microglial response, particularly in the cerebral hemisphere where contusions were most severe and adjacent to ischaemic brain damage, and to the sites of intracranial haematomas. In relation to these haematomas, descending degeneration of wallerian type was often identifiable in the corresponding pyramidal tract. For the presence of diffuse axonal injury to be recorded we required there to be excessive bilateral (symmetric or asymmetric) proliferation of microglia associated with the presence of discrete clusters of microglia in the parasagittal white matter, in the internal capsule, and in long tracts in the brain stem.8

Ischaemic brain damage was graded severe if it was diffuse or multifocal or took the form of infarcts within entire specific arterial territories, and moderate when it was limited to arterial boundary zones alone or in combination with subtotal arterial territory infarcts.9 Pressure necrosis in one or both parahippocampal gyri was taken as evidence of there having been raised intracranial pressure.10 Only haematomas more than 35 ml were recorded.

The definition of moderate disability is that the person is sufficiently independent to live on his own but has some persisting physical or mental impairment. The classification of these patients into this category depended on review of the medical records by the neurosurgical author (BJ). Although these notes were often very limited, there was always clear evidence that the patient had been living independently.

Results

The reason that many of these cases reached us was because they had sudden deaths that led to investigation by the Procurator Fiscal (table 1). Post-traumatic epilepsy was a feature in 15 patients (75%) and this was the main disability. In 11 of these patients with epilepsy an acute intracranial haematoma had been evacuated. In seven of the cases of epilepsy the patient had been found dead at home and some of these could be regarded as cases of sudden unexplained death due to epilepsy (SUDEP).11-13 Of the five patients without epilepsy four had mental disabilities and one had hemiparesis. Alcohol misuse was a feature in 13 patients (65%) and in 10 patients there was both epilepsy and alcohol misuse.

The principal findings are given in table 2. There were 17 men in the group and survival ranged from 1 year to 47 years. The commonest cause of the injury was a fall (70%). Some clinical details are missing but of the 12 with a note about a lucid interval all but two were known to have talked soon after their injury, and nine (75%) of the 12 patients where information was available had a fracture of the skull.

Surface contusions were present in every case; in five they were minimal and in none were they severe. Diffuse axonal injury was identified in six cases—five were grade 1 and one was grade 2. There was a significant intracranial haematoma in 15 (75%) and all of these had been evacuated surgically. In five patients there was pathological evidence that the intracranial pressure had been high during life and in two of these (patients 2 and 5) there were small focal lesions in the brain stem of the type associated with shift and herniation of the brain. There were no cases where ischaemic brain damage was classified as severe but in three there was moderate damage of this type; in two it affected arterial territories and in one case there were bilateral lesions in arterial boundary zones. Abnormalities in the subcortical white matter were diffuse in the six patients with diffuse axonal injury, and small and focal in the three patients with ischaemic brain damage. In one of the patients with ischaemic damage there was a small focal abnormality in the thalamus. In no patient were there diffuse abnormalities in the thalamus. Brain weights were within normal limits and in five there was unilateral ventricular enlargement in relation to loss of ipsilateral brain tissue. In four patients there was bilateral ventricular enlargement and patient 4 was thought clinically to have normal pressure hydrocephalus. In patient 16 there were also the features of Wernicke's encephalopathy. In patient 15 who had grade 2 diffuse axonal injury there were small bilateral gliding contusions of the type often encountered in association with diffuse axonal injury.7

Discussion

The great majority of the 20 moderately disabled patients had injuries that were brought about by falls (14; 70%). As expected after such injuries most had talked immediately after their injury and there was a high incidence of intracranial haematoma (75%). Diffuse axonal injury was identified in six patients (30%) but in five of these it was grade 1 (the least severe form). Further evidence that the axonal injury was not severe was that in no case was there evidence by conventional microscopy of transneuronal changes in the thalamus—a standard feature in patients with the more severe grades of diffuse axonal injury who had survived for more than 2 or 3 months after their injury. Ischaemic brain damage occurred in only 15% and was bilateral only in patient 20, where the abnormalities in arterial boundary zones suggested that the patient had experienced an episode of hypotension and cerebral perfusion failure shortly after the injury. Brain weights were within normal and generalised ventricular enlargement was present in only four cases. The only patient with severe generalised hydrocephalus was thought clinically to have normal pressure hydrocephalus (patient 4). In the other five cases with an enlarged ventricle this was asymmetric and related to focal loss of brain tissue. The absence of any progressive fall in the weights of the brains or generalised increase in the size of the ventricular system indicates that progressive cerebral atrophy was not a feature in these patients.

There has been much interest in the long term outcome in those contact sports in which there is relatively minor but repeated head injury. These include boxers, a proportion of whom develop dementia pugilistica, often with parkinsonian features.14 The neuropathology in such patients comprises some features in common with Alzheimer's disease15 16 providing additional support for the significant association between a history of head injury with loss of consciousness, and the risk of subsequent Alzheimer's disease.17 If a head injury does precipitate Alzheimer's disease then a potent acute phase molecule such as IL1 and the upregulation of neuronal β-APP are necessarily involved18 probably in genetically susceptible (APOE ε4 positive) people.19 The case material in this report is part of a much larger study investigating the cellular and molecular relation between head injury and subsequent neurodegenerative disease, which will be reported when the study is complete.

These findings suggest that moderate disability after a head injury is related mainly to focal brain damage often associated with the surgical evacuation of an intracranial haematoma. In 75% of the patients in this series post-traumatic epilepsy was a major component of the disability and 11 of these 15 patients had had an intracranial haematoma. It seems from this study that epilepsy can be an important factor contributing to late but premature death in patients left moderately disabled after a head injury—some of these coming into the category of sudden unexplained death from epilepsy (SUDEP).11-13

One particular reason for these neuropathological studies was to establish whether or not the brains of patients who remained vegetative, severely disabled, or moderately disabled after a head injury showed a similar range of structural damage but of decreasing severity. Some types of brain damage were common to all, but diffuse damage (the severer grades of diffuse axonal injury and hypoxic brain damage) and thalamic damage predominated in vegetative survivors whereas focal damage (intracranial haematoma in particular) predominated in patients who remained severely or moderately disabled.3 But a third of the severely disabled patients had either severe diffuse axonal injury, or thalamic damage, or both.