Drug–Drug Interactions Still High in the Medical ICU

June 11, 2012 — Drug–drug interactions (DDIs) are frequent occurrences in the medical intensive care unit (MICU). The severity of the condition and the drug combinations used in the MICU are distinct from those of the DDIs published in other ICU settings (eg, cardiac ICU). Prevention of these DDIs is an important aspect of patient safety.

Pamela L. Smithburger, PharmD, from the Department of Pharmacy and Therapeutics, University of Pittsburgh School of Pharmacy, Pennsylvania, and colleagues presented the results of a 4-week, prospective, observational study in an article published online June 7 in the International Journal of Pharmacy Practice. This is the first study that evaluates drug interactions specifically in a MICU population with a variety of diagnoses and comorbid disease states. The authors conclude with the suggestion that patient characteristics and location should be considered when developing a DDI alerting system.

The study included 240 patient evaluations. Two interaction databases (Micromedex and Lexi-Interact) were used to identify DDIs and assign a severity rating to the identified DDIs. Of 457 DDIs identified, roughly 25% (114/457) were considered to be major interactions per at least 1 of the databases used. The rate of DDIs was 190.4/100 patients, and the most commonly involved medications were antihypertensive and anticoagulant/antiplatelet agents.

In accordance with requirements of the institutional review board and the health insurance portability and accountability act, the study did not collect patient identifiers, meaning that clinical outcomes could not be followed over time. Other limitations included lack of generalizability; the academic institution where the evaluation took place had particularly ill MICU patients as a result of transfers from community hospitals and also housed many posttransplant patients.

The authors identified 3 possible reasons for overlooking DDIs in the presence of an alerting system: pharmacist alert fatigue, judgment of an appropriate risk–benefit ratio for the patient receiving the interacting drugs, or clinical decision support software (CDSS) that did not include the DDIs identified by the drug interaction databases used in this study. The authors emphasize that use of computerized physician order entry and CDSS in the ICU setting should provide practitioners with a way to use technology at the point of order entry.

The authors also note that studies in cardiac ICUs have identified more DDIs per 100 patient days than were identified in this study. Moreover, a previously published study conducted in a mixed ICU found that 70% of patients had a DDI during the study period.

Altogether, these studies underscore the fact that drug interaction compendia need to be improved to better identify and guide clinical decision-making.

The authors have disclosed no relevant financial relationships.