Pharmacotherapy following metabolic and bariatric surgery: From time-dependent pharmacokinetic changes to clinical decision support

Abstract

Metabolic and bariatric surgery (MBS) is increasingly performed worldwide as the most effective intervention for sustained weight loss in patients with severe obesity. MBS induces anatomical and physiological changes that affect drug pharmacokinetics (PK), raising concerns about pharmacotherapy. Despite growing clinical experience, evidence-based guidance following MBS remains limited. By enhancing our understanding of the PK changes after MBS, this thesis aims to provide practical recommendations for pharmacotherapy. MBS alters drug absorption, distribution, metabolism, and elimination, leading to time-dependent PK changes. Some drugs showed initial reductions in plasma levels, followed by normalisation or further alterations over time, with considerable variability between drugs. A nationwide survey among Dutch healthcare professionals showed that 98% acknowledged that MBS influences drug effects, with 23% reporting reduced efficacy and 21% noting increased adverse events. This survey indicates the need for increased awareness and enhanced training to optimise pharmacotherapy in patients following MBS. Analysis of pharmacovigilance reports from the Dutch (Lareb) and global (VigiBase) databases showed that most individual case safety reports (ICSRs) were categorised as “increased effect/adverse events” rather than “decreased effect or subtherapeutic”. Discrepancies were found in the number and classification of MBS-associated ICSRs between the two databases. While pharmacovigilance databases can help identify MBS-related drug issues, standardised data extraction and improved access to medical history in global databases are crucial for more reliable analysis. Following Roux-en-Y gastric bypass (RYGB), paracetamol (PCM) absorption was faster and more extensive. While first-pass metabolism remained comparable to pre-surgery, intrinsic hepatic clearances via all metabolic pathways declined with weight loss, and this decrease exceeded what would be expected from pre-surgery values and values in individuals without obesity. This additive effect was best described by incorporating relative weight loss. Over the first year after RYGB, PCM exposure increased, PCM-glucuronide slightly increased, PCM-sulfate exposure remained similar, and oxidative metabolites decreased substantially. These findings demonstrate that PK changes after MBS involve more than weight loss alone, potentially reflecting alterations in liver function and size. In 571 cancer patients with prior MBS, 30 oral anticancer drugs were identified, with varying risk of underdosing. Available TDM samples showed that 25% of post-MBS plasma levels were subtherapeutic. These routine practice data informed the development of a risk assessment flowchart to prioritise drugs for closer monitoring or therapy adjustments. This flowchart is intended as an initial tool for individual clinical decision-making. A case series of four chronic myeloid leukaemia patients treated with tyrosine kinase inhibitors following MBS demonstrated high inter- and intraindividual variability, requiring dose modifications or treatment switches. This study illustrates the clinical impact of MBS-induced PK alterations. Finally, evaluation of electronic health records revealed inadequate registration of MBS as a contraindication across healthcare settings, increasing the risk of inappropriate drug prescribing and dispensing. Better registration of MBS as a contraindication is essential to enable clinical decision support (CDS) recommendations. Overall, MBS induces time-dependent PK changes. Integrating structured PK research, extending advice in CDS systems, and improving patient documentation are essential to develop comprehensive pharmacotherapy guidelines for MBS patients, ensuring safe and effective medication use.