During the early phases of recovery from traumatic head injury, the level of functional cognition and the presence of agitation in patients appear to co-vary. However, it has been observed that there appears to be some temporal disassociation in the recovery of cognition and agitation. The purpose of this study was to investigate the degree to which attention accounts for the co-variation previously observed. Over a 1-year period, 130 patient-weeks of independent monitoring of cognition, agitation and attention were obtained from 20 head-injured patients in the acute phase of recovery. Weekly scores for measures of cognition, agitation and attention were each found to share approximately 50% of the variance when paired with one of the other two. When attention was extracted, only 7% of the variation in cognition was accounted for by agitation, and 40% of the variance could not be accounted for by either agitation or attention. These results support previous findings that cognition and agitation co-vary with most of the co-variance due to the effect of attention on each. Concomitantly, these results allow that significant portions of the variance in cognition and agitation may be temporally dissociated during the acute phases of recovery from traumatic head injury.

The case of a traumatic brain injury (TBI) patient with dramatic cognitive deterioration in the absence of medical aetiology other than simultaneous decline in serum sodium led to an investigation of the association between declines in sodium levels and cognitive status. In a population of 50 persons undergoing TBI rehabilitation, 12 (24%) had relative (3 mEq/L) decreases in serum sodium while five (10%) experienced absolute hyponatremia (136 mEq/L). Correlation with cognitive status was significant when the absolute hyponatremia group was compared with those whose sodium levels remained above 136 mEq/L. A case-matched study of the relative hyponatremia group yielded no significant association between sodium-level decreases and cognitive status. These data support previous conclusions indicating wide variation in individual responses to changes in serum sodium. The threshold for significant effects of hyponatremia may be higher in patients with TBI than in populations studied previously.

We have observed five individuals who appear to represent a unique subgroup of patients with traumatic brain injury (TBI). Because of the prominence of severe ataxia, this group has been labelled the 'ataxic subgroup'. These individuals are distinguished by both clinical course and outcome, including severe ataxia, prolonged coma and prolonged post-traumatic amnesia (PTA). They distinguish themselves from other severely impaired TBI patients in that they spend a relatively longer length of time prior to the establishment of volition, but progress rapidly through the period of confusion. We hypothesized that this group is unique in that they have suffered Grade III diffuse axonal injury (DAI) with no or minimal complications due to other primary or secondary brain damage. In order to investigate these hypotheses, a retrospective file review of a selected group of 72 patients was undertaken to determine the specificity and sensitivity of two diagnostic criteria. The existence of severe Grade III DAI without other primary or secondary brain damage was presumed if severe ataxia was present in conjunction with normal CT scans. Results of this review indicated that 33% of the population demonstrated severe ataxia, although only 11% also had normal CT scans. These dual criteria were neither adequately sensitive nor specific to define the five patients who comprised the 'ataxic subgroup'. When rate of clearing the confused period of PTA was added to the diagnostic criteria, specificity improved. Although this attempt to define this subgroup empirically was not entirely successful, further attempts to delineate this group are important in that prognosis for clearing PTA is good despite early indicators of poor outcome.(ABSTRACT TRUNCATED AT 250 WORDS)

A 2 × 3 × 3 factorial design was used to examine the recall of units of verbal information by head‐injured (n = 12) and hospitalized controls (n = 12) under three modes of presentation and three recall times. Presentation of three instruments commonly used in memory assessment was counterbalanced among overt passive (standard administration), covert active (silent reading), and overt active (reading aloud) modes. Information recall was taken immediately after presentation, 20 minutes, and 48 hours later. Analysis of variance revealed a significant main effect for group, but no main effect for either mode of presentation or recall time. Confirmatory nonparametric analysis supported initial results. Findings are discussed with regard to potential implications for clinical assessment on brain‐injured patients and further research in memory assessment on this population.