Serum concentrations of antidepressants, antipsychotics, and antiepileptics over the bariatric surgery procedure

Abstract

Abstract Purpose As a substantial proportion of bariatric surgery patients use psychotropic/antiepileptic drugs, we investigated the impact of this procedure on serum concentrations. Methods In a naturalistic, longitudinal, prospective case series, we compared dose-adjusted trough concentrations of antidepressants, antipsychotics, or antiepileptics in consecutive patients before and after bariatric surgery. Adherence to treatment over 2 weeks preceding each sampling was considered. Results In all, 85 participants were included (86% female, median age 45 years, median body mass index 42 kg/m2). They were being treated with 18 different psychotropic/antiepileptic drugs (7 substances: 6–17 individuals, 11 substances: 1–4 individuals) and contributed 237 samples over a median of 379 days after surgery. For four out of seven substances with pre-/post-surgery samples available from six or more individuals, the dose-adjusted concentration was reduced (sertraline: 51%, mirtazapine: 41%, duloxetine: 35%, citalopram: 19%). For sertraline and mirtazapine, the low-calorie-diet before surgery entirely explained this reduction. A consistent finding, irrespective of drug, was the association between the mean ratio of the post-/pre-diet dose-adjusted concentration and the lipophilicity of the drug (logD; correlation coefficient: −0.69, P = 0.0005), the low-calorie diet often affecting serum concentration more than the surgery itself. Conclusions Serum concentrations of psychotropic/antiepileptic drugs vary after bariatric surgery and can be hard to predict in individual patients, suggesting that therapeutic drug monitoring is of value. Conversely, effects of the pre-surgery, low-calorie diet appear generalizable, with decreased concentrations of highly lipophilic drugs and increased concentrations of highly hydrophilic drugs. Interaction effects (surgery/dose/concentration) were not evident but cannot be excluded.