Interações Medicamentosas Potenciais em Gestantes de Alto Risco: uma Análise de Rede

Abstract

Introduction: High-risk pregnant women are more prone to complications and premature births, often requiring hospitalization and pharmacotherapy to control blood pressure and glucose levels, among other conditions. The physiological changes associated with pregnancy increase vulnerability to adverse events such as those resulting from drug interactions (DIs). Network analysis provides an effective approach to identifying complex patterns in health, such as the occurrence of DIs. Objective: To characterize the occurrence of DIs, the main drugs involved, and associated outcomes of severe DIs through network analysis. Methods: A prospective cohort study (September 2019 to June 2022) included 297 hospitalized pregnant women monitored daily at a teaching maternity hospital. DIs were classified using the Micromedex database into mild, moderate, and severe categories. Descriptive analysis was conducted using Stata© v15, and network analysis was performed using Gephi© 0.9.1. The Yifan Hu algorithm was employed in the network analysis, and betweenness centrality was used to identify drugs most implicated in severe interactions. Results: The patients had a mean age of 30.4 ± 6.6 years and were at a mean gestational age of 30.9 ± 7.0 weeks, with predominant diagnoses of gestational hypertension (32%) and gestational diabetes (35.7%). The average duration of treatment was 5.6 ± 4.2 days, and the mean total number of prescribed medications per patient was 7.4 ± 2.9. Approximately 85% of patients experienced at least one DI during the study period, with an incidence rate of 146.3 DIs per 1000 patient-days. The most frequent DI pairs were hydralazine/methyldopa (10.6%) and hydralazine/dipyrone (5.6%), both classified as moderate. Regarding severe DIs, network analysis identified three clusters: a) levomepromazine associated with depressant and anticholinergic drugs, b) dipyrone associated with anti-inflammatory and anticoagulant drugs, and c) iron sulfate with drugs with potential chelation properties. Conclusion: Drug interactions in hospitalized pregnant women are frequent, with severe DIs associated with risks such as cardiopulmonary depression, upper gastrointestinal bleeding, and anemia. The application of network analysis highlights complex drug interaction patterns, aiding in the understanding and management of pharmacotherapy in this vulnerable population.