Clinical pharmacist-led prevention of intravenous drug incompatibilities in an adult ICU: a prospective observational study.

Abstract

BACKGROUND: Patients in intensive care units (ICU) usually need multiple intravenous drugs (IV) simultaneously and have a limited number of venous routes. This makes IV incompatibility an important problem in these units. The aims of this study were to identify the frequency of possible drug incompatibilities in the ICU and to prepare specific pharmaceutical guidelines for the unit to the prevent them. METHODS: This cross-sectional and prospective study was conducted in Medical Intensive Care Unit (MICU) at Ankara University. Possible incompatibilities of IV drugs were assessed using the IBM Micromedex® and Stabilis 4.0 databases. RESULTS: Three hundred and thirty-eight drug request lists (DRL) of a total of 67 patients were included. At least one possible incompatibility was detected in 68.3% (95% CI: 63.3%–73.3%) of the drug request lists. First three drug groups that cause the most of incompatibilities belonged to the alimentary tract and metabolism, the anti-infectives for systemic use, systemic hormonal drugs and hormone regulators. The analysis of the type of IV drug administration showed that possible incompatibilities most commonly occurred between the continuously infused and bolus drugs (25%, n = 134/532). There was a moderate positive correlation between the number of anti-infective agents and possible incompatibilities (rho:0.41, p:0.001). Extreme drugs accounted for 40.6% (n = 432/1064) of the drugs that cause possible drug-drug incompatibilities. In terms of compatibility category, a poor degree of agreement was found between Micromedex and Stabilis (Cohen kappa = 0.174; 95% CI: 0.08–0.27; p < 0.001). Excluding “unknown” drug pairs slightly improved agreement between the databases (Cohen kappa increased from 0.174 to 0.293). A total of 567 recommendations were made during the study to prevent possible drug incompatibilities. All of which have been accepted and implemented. Of specific importance, all drug-solution incompatibilities were associated with anti-infective agents. A cross-table containing the Micromedex incompatibility data of the most frequently administered drugs with incompatibilities was prepared based on the data. There was no data for 33% (n = 438/1326) of the analyzed drug pairs in this cross-table. CONCLUSIONS: Interventions performed by clinical pharmacists contribute to preventing and reducing possible drug incompatibilities. CLINICAL TRIAL NUMBER: Not applicable.