Incidence of clinically significant responses to zolpidem among patients with disorders of consciousness: a preliminary placebo controlled trial.

Objectives: The common hypnotic, zolpidem, has been reported to temporarily restore consciousness to individuals in the chronic vegetative state. In drug responders, repeated dosing appears to maintain consciousness. The frequency of such responses, however, is unknown and is important both to guide clinical use and to plan further research on the mechanisms underlying drug response. The objectives of this study were to obtain an estimate of the frequency of clinically significant responses among individuals with disorders of consciousness, to determine whether less obvious drug responses are present among "nonresponders," and to identify clinical features characteristic of zolpidem responders.
Design: Participants were individuals in the vegetative or minimally conscious state at least 1 month after brain injury. Each participant was studied individually in a double-blind, placebo-controlled, crossover design, once on zolpidem (10 mg per feeding tube) and once on placebo. Each assessment involved baseline administration of the Coma Recovery Scale-Revised, followed immediately by administration of the study drug, followed by 5 hourly readministrations of the Coma Recovery Scale-Revised. A replication pair of assessments was available for drug responders.
Results: : One of 15 participants (6.7%) demonstrated a clinically significant response, which altered his assessment from the vegetative state to the minimally conscious state, and this result was repeated in the replication assessment. The remaining 14 participants showed no evidence of a subclinical response to the drug.
Conclusion: These results confirm that clinically significant responses to zolpidem among individuals with disorders of consciousness do occur in a minority of patients and can be replicated. Failure to find a trend toward improved performance on zolpidem among nonresponders suggests a bimodal rather than a graded response to the drug. The fact that only one drug responder was identified in this small study prevents assessment of features characteristic of drug responders.