Assessment of agents for the treatment of head injury - Problems and pitfalls in trial design

Over the past decade many neuroprotective agents have been developed with t he hope of being able to improve outcome in patients with acute cerebral di sorders, such as stroke, head injury and subarachnoid haemorrhage. Unfortun ately, in the field of head injury none of the phase III tri als performed have convincingly demonstrated efficacy in the overall population. A common misconception is that consequently these agents are ineffective. Such has, however, not been proven, and some trials show evidence of efficacy in sub groups of the population studied. The negative results, as reported in the overall population, may in part be caused by specific aspects of the head i njury population, as well as by aspects of clinical trial design and analys is. The present manuscript addresses these issues, and critically analyses the relevance of preclinical evidence and pharmacokinetic studies. It appears that decisions about initiating phase III trials in head injury were strongly influenced by experience in related disorders, such as subara chnoid haemorrhage and stroke. The available evidence from experimental stu dies in the field of head injury is limited and uncertainties exist about w hether the pathophysiological mechanisms at which neuroprotective agents we re targeted really occurred in all patients studied. Moreover. for many age nts aspects of pharmacokinetics and penetration into the brain were insuffi ciently investigated before proceeding into phase ill trials. The heterogeneity of the head injury population may cause specific problems , such as risk of imbalances between placebo and treated groups. but also, causes problems when it possible treatment effect is evaluated in relation to the prognostic, effect present. It is concluded that trials of neuroprot ective agents should be targeted first of ail to a population in which the mechanism at which the agent is directed is likely to be present. and secon dly to a population in which the chances of demonstrating efficacy Lire rea listic. i.e. patients with an intermediate prognostic The aim in most trial s of trying to demonstrate a 10% absolute improvement in favourable outcome in patients with head injury is over-optimistic and unrealistic in relatio n to the heterogenous patient population. Specific problems incurred by the use of the dichotomised Glasgow Outcome Scale as the primary outcome measu re an discussed. and the risk highlighted that even a substantial change in outcome distribution may not be adequately detected. The complexity of problems occurring in clinical trial design and analysis in head injury is such that a strong and sustained input and effort is requ ired from all experts involved in the field of neurotrauma.