Drug-drug interactions and kidney disease in hospitalized patients

PROJECT SUMMARY As polypharmacy becomes increasingly common in patients with kidney disease, drug-drug interactions (DDIs), the phenomenon of one drug altering the effect of another drug, grow as potential sources of preventable harm in this vulnerable population. The clinical impacts of DDIs are poorly understood, particularly in the hospital setting, where up to half of patients have acute or chronic kidney disease. Further, the potential for bidirectional relationships between kidney disease and DDIs—with kidney disease functioning as both an outcome of DDIs and a baseline factor that alters DDI severity—complicates epidemiologic study and necessitates novel methodologic approaches. This proposal aims to improve the safety of pharmacotherapy in patients with kidney disease through the following broad objectives: 1) examine whether commonly used antibiotics in the hospital setting interact to increase the risk of acute kidney injury (AKI) and subsequent chronic kidney disease (CKD); 2) examine whether changes in kidney function in turn produce variability in DDI severity; 3) train the applicant in renal pharmacoepidemiology, prospective research design and analysis, and pharmacokinetic simulation methods, which will enable a novel translational approach to DDI research; and 4) provide the preliminary data and mechanistic insights for future R01 funded studies that quantify the health impacts of DDIs in the kidney disease population and develop preventive interventions. The applicant will integrate the results from large scale pharmacoepidemiologic studies, prospective cohort studies, and simulation models to achieve the stated objectives. The applicant will first determine the impact of a potential DDI between two common antibiotics on inpatient AKI risk and subsequent CKD using a large-scale health system database and prospective cohort study (Aim 1). The applicant will then examine the effect of kidney disease on the magnitude of a DDI mediated by hepatic CYP450 inhibition using two distinct, yet complementary approaches: developing a pharmacokinetic simulation model of the interaction, followed by a retrospective cohort study of the same interaction (Aim 2). Integration of methods will provide insights into the effects of kidney disease on CYP inhibition that wouldn't be possible with either method in isolation. This proposal will train the candidate in renal pharmacoepidemiology, advanced statistical modeling, and pharmacokinetic simulation, and have important implications for drug safety in the 35 million hospital admissions each year. This research will generate new knowledge extending beyond the specific drugs under study and foster the candidate's progression towards becoming an independent investigator in renal pharmacoepidemiology.