HONOLULU — Disappointing clinical trial results should not suggest that outcomes cannot be improved in traumatic brain injury, only that study methodologies may need to be refined, Dr. D. James Cooper said during a plenary address at the annual congress of the Society of Critical Care Medicine.
To be sure, various interventions have had hopeful improvements in animal models, but not in human trials.
But the heterogeneity of the traumatic brain injury (TBI) population and “huge differences” in the specific trauma suffered may make study results look unfairly pessimistic, said Dr. Cooper, deputy director of the intensive care unit at Alfred Hospital, Melbourne.
Experimental treatments may be initiated too late because of logistical and informed-consent dilemmas, and older patients may be so unlikely to benefit that they skew results. And follow-up assessment periods may be too brief, because it appears that Glasgow Outcome Scale scores improve greatly over time, but very slowly, he said.
A number of lessons have indeed been learned, even from negative clinical trials, and several promising approaches are currently under review.
Serious doubt has been cast on the efficacy of early high-dose steroids, for example, following the curtailment of the 10,000-patient randomized controlled MRC-CRASH (Corticosteroid Randomisation After Significant Head Injury) trial in the United Kingdom after excess deaths were reported in the steroid arm.
“It seems clear from the study that the use of an agent that has been very widely used, particularly in the developing world, clearly and unambiguously increases mortality, accounting for an absolute number of 3% excess deaths. I think it's abundantly clear … [that the] use of high-dose steroids should cease,” said Dr. Cooper, associate director for Australia's National Trauma Research Institute.
Because they lower vasopressor requirements in TBI patients, lower-dose steroids are used commonly in the intensive care environment, he noted. “There are no randomized controlled trials at all in this area, and it's clear to me, [based on the MRC-CRASH results,] there needs to be … a reevaluation” of this practice, said Dr. Cooper.
Another unexpected finding stemmed from the Australian SAFE-TBI (Saline Versus Albumin Fluid Evaluation-Traumatic Brain Injury) study, in which Dr. Cooper participated. That study of nearly 500 patients confirmed that albumin is independently tied to mortality in TBI patients when it is used for intravascular fluid resuscitation in the first 28 days. In contrast, saline was associated with lower mortality and better neurologic outcomes in patients with moderate to severe TBI.
The reasons remain unclear, although Dr. Cooper hypothesized that albumin may increase brain edema, prompting the use of other agents that could contribute to mortality; that it may increase bleeding or cause more coagulopathy; or that it may be the result of hemodilution.
The possibility remains that albumin's negative effect on survival may be a class effect of colloids, he said.
A recent analysis of data from both the SAFE-TBI study and the earlier ATBIS (Australasian Traumatic Brain Injury Study) “[adds] to our strong feeling that saline alone might be worthwhile,” he said.
As a final note, Dr. Cooper outlined two ongoing international clinical trials of early decompressive craniectomy to reduce intracranial pressure.
The absolute risk of mortality was halved with early decompressive craniectomy versus medical therapy alone in a recent, 38-patient French study; but the trial was concluded early because of slow recruitment.
Dr. Cooper's DECRA (Early Decompression Craniectomy in Patients With Severe Traumatic Brain Injury) trial at 21 sites is enrolling only patients younger than 60 years with blunt diffuse brain injuries—strict criteria that may be more conducive to interpreting results, he said.
So far, 112 patients have been enrolled of 165 anticipated, already more than the largest study ever conducted of early decompressive craniectomy, Dr. Cooper noted.
Among the first 42 patients who received surgery, the complication rate has been less than 10%, he said.