Drug-drug interactions in cardiac and cardiothoracic intensive care units: an analysis of patients in an academic medical centre in the US.

BACKGROUND: Mortality and morbidity are increased in patients experiencing drug-drug interactions. Unfortunately, there is a paucity of literature describing clinically significant drug-drug interactions occurring in the intensive care unit (ICU). Knowing the clinically significant drug-drug interactions allows the opportunity for prevention through knowledge and computer-assisted programmes.
OBJECTIVE: To identify significant potential drug-drug interactions occurring in the cardiovascular ICU (CCU) and the cardiothoracic ICU (CTICU). STUDY
DESIGN: Prospective, observational study conducted over a total of 8 weeks in February and March 2009.
SETTING: CCU and CTICU in a major academic medical centre (Presbyterian Hospital, University of Pittsburgh Medical Centre). PATIENTS: All adult patients (>or=18 years of age) admitted during 1 month in each ICU. INTERVENTION: Micromedex and Lexi-Interact interaction databases were used to screen each patient's medication profile daily for the presence of potentially interacting drug pairs that would be considered a potential drug-drug interaction. A severity assessment using these databases was completed after a potential drug-drug interaction was identified. PRIMARY OUTCOME MEASURE: The frequency of significant drug-drug interactions, including those that were considered major or contraindicated, according to two commercially available interaction databases.
RESULTS: Evaluations of 400 patient medication profiles were conducted, resulting in 225 profiles possessing one or more potential drug-drug interactions. A total of 1150 potential interactions were identified, resulting in 287.5 potential interactions per 100 patient-days. Of the 1150 potential drug-drug interactions, 458 were unique interacting drug pairs; 5-9% of the potential interactions were considered major or contraindicated. Many of the significant and frequent potential interactions involved blood coagulation modifiers, potential interactions that could result in QTc prolongation, and cytochrome P450 inhibition. Micromedex and Lexi-Interact agreed on the severity ratings in 20.5% of the potential interactions.
CONCLUSIONS: Significant potential drug-drug interactions occur in the CCU and CTICU, highlighting the need for active surveillance to potentially prevent patient harm. Clinicians should also consider using two references for identifying interactions, due to the lack of congruence between sources.