Neuropathology in vegetative and severely disabled patients after head injury

Materials and methods.

These cases are drawn from a database of more than 1,300 cases of fatal blunt head injury covering the period 1968 to 1998, held in the Department of Neuropathology, Southern General Hospital (Glasgow, UK) and used in many previous publications. All brains were fixed in 10% formal saline for a minimum of 3 weeks before dissection, after which a full macroscopic and microscopic examination was undertaken in each case.6 Contusions were assessed quantitatively using the total contusion index (TCI), with an index of less than nine contusions considered minimal and more than 37 considered severe.7 Diffuse axonal injury (DAI) was graded as previously described.8,9⇓ In grade 1, there was diffuse axonal injury in the white matter without a focal lesion. In grade 2, a focal lesion was found in the corpus callosum. In grade 3, a focal lesion was also found in the dorsolateral segment of the rostral brain stem. Ischemic brain damage was graded severe if diffuse or multifocal or in the form of infarcts within entire specific arterial territories. It was considered moderate when limited to arterial boundary zones alone or in combination with subtotal arterial territory infarcts.10 Pressure necrosis in one or both parahippocampal gyri were taken as evidence of raised intracranial pressure having occurred.11 Only hematomas acting as a mass lesion (>35 mL) were recorded.

The clinical classification of outcome depended on a review of the medical records by the neurosurgical author (B.J.). Because some cases antedated the published descriptions and adoption of the labels “vegetative state” and “severe disability,” these specific terms were often not used in the contemporary notes. Consequently, these states had to be judged retrospectively, and only when outcome could be assessed with some confidence was a case included. For the severely disabled cases, three categories were recognized: mobile (M), bed bound, and minimally conscious state (MCS). The latter state has only recently been defined3 so that allocation to this category depended on evidence of more responsiveness than for vegetative patients, but no speech.

Findings.

The frequency of single features in the two groups are compared in table 1. In contrast to the VS cases, the severely disabled cases were older, with none aged 15 years or younger, and almost four times more patients were aged 65 years or more. Only half as many severely disabled patients survived for 12 months or more. The causes of injury were similar in the two groups. In the severely disabled group, three times as many had a lucid interval, a higher incidence of skull fracture (50 versus 34%) and a higher median total contusion index (TCI = 9.5 versus 4.0). Intracranial (IC) hematoma was much more common in the severely disabled group (70 versus 26%), all but two having been surgically evacuated. However, raised IC pressure was somewhat less common in the severely disabled group (57 versus 71%). The incidence of ischemic brain damage (IBD) was similar in the two groups (47 versus 43%).

Diffuse axonal injury was less common in the severely disabled group (50 versus 80%), and this difference was most marked for the most severe (grade 3) lesions (17 versus 63%). Likewise, structural abnormalities in the thalamus were much less common in the severely disabled patients (37 versus 80%); this was even more marked in those surviving longer than 3 months (37 versus 96%) by which time transneuronal thalamic degeneration subsequent to DAI could be identified microscopically.

Brain stem lesions and hydrocephalus were equally common in the two groups. Hydrocephalus was common even in those who survived less than 3 months in both groups. More brains were seen with low weights in the VS.

The differences between severely disabled and VS cases were significant (<0.05) for five of these features. Median comparisons were based on the Mann–Whitney test and comparison of proportions were based on the χ² test. No adjustment was made for multiple testing (for additional information, visit the Neurology Web site at www.neurology. org and scroll down the Table of Contents to find the title link for this article).

Table 2 shows that grade 2 or 3 DAI only and the combination of this with thalamic damage without IBD was less common in the severely disabled group (10 versus 20%, and 10 versus 34%). Thalamic damage with IBD, but without DAI, was similar in the two groups (20 versus 26%). The most striking difference, however, was that 15 (50%) of the severely disabled cases had neither DAI grade 2 or 3 nor thalamic damage, and 10 of these had no ischemic brain damage either. These combinations did not occur in a single one of the VS cases.

All but one of the 15 severely disabled cases with neither DAI 2 or 3 nor a thalamic lesion had an IC hematoma; nine had raised ICP; and five had a contusion index of 28 or above. A brain stem lesion was found in four cases. Moderate IBD was seen in five, two in the boundary zones, two in arterial territories, and one with focal lesions in the neocortex. Therefore, a considerable aggregate of focal damage was seen in these cases. Among the six cases with grade 1 DAI, one was without an IC hematoma or any cortical contusions, raised ICP, or IBD. Thus, it appears that DAI grade 1 alone can cause severe disability (for additional information, visit the Neurology Web site at www.neurology.org).

Among the VS cases, eight had DAI of 2 or 3, but no IC hematoma, raised ICP, IBD, or brain stem lesions; four of these had no contusions with the TCI ranging from three to six in the other four (for additional information, visit the Neurology Web site at www.neurology.org). Transneuronal thalamic damage was found in four; of those with no thalamic damage, three had survived less than 3 months. It is evident, therefore, that DAI 2 or 3 alone can result in the vegetative state. All three VS cases with DAI grade 1 had ischemic thalamic damage and raised ICP, two had IC hematoma, two had IBD, and two had brain stem lesions (for additional information, visit the Neurology Web site at www.neurology.org). Of the seven with no DAI, all had thalamic damage (diffuse in five) with raised ICP and IBD; five had IC hematoma and one a lateral brain stem lesion (for additional information, visit the Neurology Web site at www.neurology.org). In our study of 49 VS cases of either traumatic or nontraumatic origin, every one had widespread destruction of the white matter of the cerebral hemispheres or the thalamus, or both.1 In each causative group, some cases were seen with no structural abnormalities in the cerebral cortex.

The lesions with the three grades of outcome among the severely disabled cases are compared with each other and with VS in table 3. The lesions in the MCS cases were not as similar to those in the VS cases as might have been expected. No marked difference was seen between the three grades of severe disability in the incidence of hematoma, raised ICP, and ischemic damage. The median contusion index for mobile cases was closer to that of VS cases (three versus four), which is lower than the bed-bound and minimally conscious (17 and 11.5). In MCS cases, DAI 2 or 3 was more frequent than in the other severely disabled cases (42 versus 22%), but still much less than in the VS cases (71%). Thalamic lesions were found in 50% of MCS cases, less frequently than in VS (80%); however, as many as 44% of the mobile cases had thalamic damage. Hydrocephalus was equally common in the mobile and bed-bound cases (89%), which was more than in the VS cases (77%), and those in MCS (50%).

Discussion.

It might have been expected that the structural lesions in severely disabled cases would have been similar to those in the VS cases but less severe. However, in 15 of the severely disabled cases (50%), neither DAI grade 2 or 3 nor thalamic damage was seen, a combination that did not occur in a single one of the VS cases. The lesion underlying the severe disability in these cases appears to have been an aggregate of focal damage (from contusions, intracranial hematoma, ischemic brain damage, or brain stem lesion). Six severely disabled cases had grade 1 DAI, one of which had no other lesion at all. It seems, therefore, that even without the focal lesions of grade 2 or 3 the diffuse white matter damage of DAI can be extreme enough to result in severe disability.

Some severely disabled cases, however, had lesions similar to those in the VS cases; 30% had grade 2 or 3 DAI and 37% had thalamic lesions. Some severely disabled cases in each of the three outcome categories had lesions similar to those in VS. Of cases who became mobile, two had grade 2 or 3 DAI and four had thalamic damage; but there were seven cases who were minimally conscious or bed bound who had no DAI 2 or 3, IBD, or thalamic damage.

The only previous attempt to compare the lesions in vegetative and other severely brain damaged patients was a radiologic study.6 The lesions disclosed by MRI in patients in a VS 6 weeks after head injury were recorded, and comparisons made between 42 who were still in a VS 1 year after injury and 38 who had some recovery. The incidence of DAI grades 2 or 3 in patients still in a VS at 1 year was even higher than in our VS cases (98 versus 71%). Severe DAI was less common in the recovered cases, but 26% of them had grade 3 DAI (none with grade 2). This is similar to our finding of DAI grade 2 or 3 in 30% of severely disabled cases. We, therefore, confirm the overlap between the lesions in VS and severe disability, which makes it impractical to use the radiologic demonstration of severe DAI in the early stages after injury as reliable evidence that the patient will remain vegetative.12 A finding of severe DAI, however, does indicate a high likelihood of a poor outcome. Of the 34 cases with DAI 2 or 3, in our study, 25 became vegetative and five MCS (88%); only two (6%) became mobile. In the radiologic study, more of the recovered cases had thalamic lesions (53%) than did those who remained vegetative (40%). By contrast, we found more than twice as many VS cases with thalamic damage as in the severely disabled cases.

Regarding those severely disabled cases that appeared to have similar lesions to the VS cases, this may reflect the fact that this study recorded lesions only as present or absent. It is likely that the differences in outcome in cases with apparently similar lesions depend on quantitative variations in the amount of damage in various structures. We must await more subtle and perhaps quantitative analysis of the brain damage after head injury before we can explain differences in outcome in terms of these pathologic lesions.