Abstract:
The treatment of agitation during acute hospitalization is challenging. Behavior must be controlled sufficiently to ensure safety and physiologic stability, while side effects of the pharmacologic regimen need to be minimized. Neuroleptics (e.g. haloperidol) are commonly chosen due to their relatively quick therapeutic onset, but in addition to a multitude of potential side effects, there is some concern about potential long-term effects on recovery. Other choices available, such as benzodiazepines (e.g. lorazepam, alprazolam) and narcotics (e.g. morphine, oxycodone), also have the potential for significant side effects.

This naturalistic study has been designed to provide preliminary indications of the costs and benefits of alternate pharmacologic approaches, including relationships between medication choice and functional status and cognitive recovery. The use of agents to manage agitation were monitored. The three most commonly used classes of medications, neuroleptics, benzodiazepines, and narcotics, were then further investigated. Since patients were frequently administered medications from a combination of these classes, analysis were conducted based on the most common combinations (a) without considering whether an individual had received a benzodiazepine during the acute care stay, the group of individuals who had received both narcotics and neuroleptics were compared to those who had received narcotics but not neuroleptics; b) for those who were given benzodiazepines, the group of individuals who had received both narcotics and neuroleptics were compared to those who had received narcotics but not neuroleptics; and c) for those who were not given benzodiazepines, the group of individuals who had received both narcotics and neuroleptics were compared to those who had received narcotics but not neuroleptics).

First, the association between the use of these medications and injury severity (lowest GCS, initial pupillary response, and intracranial hypertension) was analyzed. It was found that there was no association, suggesting that medication choice was not associated with injury severity.

The association between use of medications from these three classes and outcomes (functional independence as measured by FIM, and length of PTA as measured by the GOAT) was then assessed. In each of the analyses, individuals who were given neuroleptics and narcotics (with or without benzodiazepines) tended to have a longer length of PTA than those who were not given neuroleptics. In some of the analyses the association was significant, in other analyses the relationship was weaker. No consistent associations with functional independence were found.

Interpretation of these results must be made with caution, as the associations found do not prove causality. Controlled experimental designs can assist with isolating the cause-effect relationship.