Pharmacokinetic considerations for patients with a history of bariatric surgery

KEYWORDS:

• Obesity
• bariatric surgery
• pharmacokinetics
• therapeutic drug monitoring

1. Introduction

Bariatric surgery is a well-known option for obese subjects to achieve and sustain weight loss with a rapidly increasing number of surgeries performed worldwide over the last two decades [1]. Bariatric procedures can be restrictive, malabsorptive, a combination of the two, and/or alter the process of digestion in bypassing a part of the small intestine. All procedures substantially alter the anatomy of the gastrointestinal tract. Thus, alterations in the absorption and disposition of medications can be expected.

2. Weaknesses in the research done in this field

Research performed in this field suffers from several weaknesses. First, most research data is related to malabsorptive procedures, expected to be more highly responsible for alterations of drug pharmacokinetics than restrictive procedures alone [2]. Indeed, most studies have been performed on patients with Roux-en-Y gastric Bypass (RYGB). Many have demonstrated changes in drug absorption parameters that are certainly related to the smaller gastric pouch after surgery and the achlorydric state associated, and few showed change in overall exposure [3–6]. Thus, the consequences of restrictive procedures merit the same attention as time to maximal concentration (T_max) and maximal concentration (C_max) may change and have clinical consequences for some drugs. They may help to distinguish the respective role of the smaller gastric pouch and those of the diversion of the biliopancreatic system in the changes of pharmacokinetic parameters of absorption and/or drug exposure. Second, until recently, sample sizes of published studies were small and much of the available evidence included old case reports and case series. Results based on such a small number of subjects cannot be generalized as most drugs display interindividual variability that must be considered as well [7]. Third, most pharmacokinetic studies have been performed in parallel patient cohorts consisting of those who had bariatric surgery a few months prior and their controls, often matched based on body mass index, age, and sex [8,9]. This design does not allow measuring the effect of surgery and weight loss for a given individual, nor the effects of other factors of interindividual variability of drug disposition, such as drug metabolism phenotypes. Furthermore, the ideal control group for patients with surgery-induced weight loss, in terms of drug metabolism or body composition, may not be those matched for body mass index. Hence, these studies do not provide information on dose adaptation between preoperative and postoperative periods, nor on the effect of the baseline drug metabolism phenotype.

Finally, more recent and better designed studies, on a larger number of patients than those performed during the last several years, have shown that changes in drug disposition are drug-specific and difficult to predict [2]. The pharmacokinetic changes reported are also not always interpreted based on their clinical significance. While clinicians responsible for the follow-up of these patients know that they should consider the risk of alterations in drug plasma levels, they have no straightforward evaluation tool to meet the needs of clinical practice if blood levels or some other surrogate marker cannot be monitored.

Physiologically based pharmacokinetic models have been validated for predicting oral drug exposure and dose adjustments after surgery, based on drug solubility and permeability, which forms the basic concept of the Biopharmaceutical Classification System [10]. They showed good accuracy in predicting metoprolol pharmacokinetics after RYGB [10], but the application of these models still needs to be validated for other drugs. It is unclear whether they can accommodate the complex interplay of various factors linked to the drug, such as solubility, permeability, metabolic enzymes, and transporters, and those linked to the patient, such as changes over time in body weight and composition, gut microbiota, and low-grade inflammation, among others. Conversely, such factors may only play a minor role in oral drug disposition, but this has not yet been investigated. Finally, current models are difficult for prescribers to use in daily practice.

3. Potential and challenges of this area of research

This area of research is of great interest to pharmacologists and toxicologists. Appropriate clinical studies could substantially improve our knowledge in several areas of clinical pharmacology, to the benefit of the overall population, including the role of the stomach and intestinal barrier in determining drug bioavailability and pharmacokinetics, the influence of morbid obesity and its associated metabolic disorders, and the influence of weight loss on drug disposition. The study of the same individuals before and after surgery limits the impact of interindividual variability, well known to influence drug metabolism. It makes it possible to focus on the effect of the new anatomy of the gastrointestinal tract and/or the effect of surgery-induced weight loss. Furthermore, surgery itself allows the collection of intestinal or liver biopsies with limited risk to the individuals [5,11]. Most research on intestinal enzyme and cytochrome content has been performed on samples collected from various sites throughout the intestine from patients with severe intestinal diseases. Surgery allows the collection of homogeneous intestinal and liver tissues and, for example, testing the correlation between the content of drug metabolizing enzymes and drug pharmacokinetics [12], even if performed in morbidly obese patients, with potential metabolic diseases.

Pharmacologists and toxicologists can benefit from such an informative model to explore changes in drug pharmacokinetics, but an equally important issue is to distinguish the effect of weight loss and that of surgery on these changes (Table 1). An ideal pharmacological study would include pharmacokinetic analysis before and immediately after surgery, a few months later after the patient experienced weight loss, and ideally when the weight stabilized. We conducted a pharmacokinetic study in obese patients before, immediately after (8–15 days), and six months after surgery (OBEMO study) [4]. It was very difficult to conduct the study immediately after surgery. Some patients were extremely tired and others presented with early surgical complications (for example, bleeding, intestinal obstruction, and peritoneal infections), thus reducing the number of patients for this period. Conducting a pharmacokinetic study in patients who have just undergone a bariatric procedure and are still hospitalized (for example, two days after) would represent a possible solution, within the limit that anesthetic drugs might influence the activity of drug metabolizing (i.e. transporters, phase I, and phase II enzymes) and that abnormalities in motility are present early postoperatively. Hence, studies that aim to improve our pharmacological knowledge may be difficult to perform because the respective effects of weight loss and the novel anatomy of the gastrointestinal tract would be difficult to distinguish. Nevertheless, such studies should be strongly encouraged.

Table 1. Research priorities for the clinician and the pharmacologist.

Research priorities
From the point of view of the clinician	From the point of view of the pharmacologist
To know whether drug dosing of bariatric surgery patients should be different from that of the general population	To help the prescribers while interpreting the degree of change of several pharmacokinetic parameters in bariatric subjects
To know whether drug dosing should be different between the preoperative and postoperative periods	To improve knowledge of drug pharmacology using the surgery model
To have clear recommendations for dose adaptation strategies in clinical practice for several drugs	To identify markers of drug variability in the overall population
To know the circumstances where therapeutic drug monitoring is useful for dose adaptation	To find the appropriate methodology that would distinguish the effect of surgery from that of weight loss

4. The ultimate goal of the research in this field

From the point of view of the prescriber, the biggest challenge remains the patients themselves (Table 1). An increasing amount of information for drug dosing is required to guide clinical practice, especially as the growing population of patients with bariatric surgery will age and require medication. All surgeries should benefit from clinical studies. Clinical studies that compare drug pharmacokinetics in patients who underwent bariatric surgery a few months or years before to matched controls present several limits but would be straightforward to perform and provide an efficient means to gather data for optimizing clinical practice. Various drugs, having several metabolic pathways, galenic forms, and circumstances of administration should be considered. The first step would be to list the most commonly employed drugs in such population, and priority should be given to drugs that do not have a clinical or biological surrogate for dose adaptation. Outcome of vitamin, iron, or calcium absorption should be specifically studied as well for the different surgical procedures.

5. Needs to achieve the ultimate goal of the research

The risk of toxicity and side effects should be foreseen and the risk of inefficacy must be avoided. In a clinical study, we demonstrated that doses of oral morphine sulfate solution should be reduced in patients with RYGB to limit the risk of adverse events, as RYGB clearly increased the rate at which oral morphine sulfate is absorbed [4]. It is also important to know whether a standard dose is appropriate when the effect is generally evaluated over a span of 4 weeks or even months. It seems prudent to initiate therapeutic drug monitoring for certain drug classes until appropriate pharmacokinetic and/or pharmacodynamic surrogates are identified. For example, therapeutic drug monitoring remains a matter of debate for psychotropic drugs, but may be appropriate in this situation to avoid unsuitable concentrations that could partly explain a lack of effect before the diagnosis of failure. Drug monitoring of immunosuppressive agents is even more important in subjects after surgery than usual. Tamoxifen-guided dose escalation should be also considered in patients with bariatric surgery, especially as genetic factors also strongly affect its concentration and efficacy [13].

The study of metoprolol pharmacokinetics is of great interest as it is a CYP2D6 probe substrate, but this beta-blocker has a pharmacodynamic marker of efficacy [6]. In another example, it would be predicted that the dose of metformin, which is normally poorly absorbed, would need to be increased after RYGB, based solely on the fact that the duodenum is bypassed, but a recent pharmacokinetic study shows that this is not case [14]. This drug also benefits from a biological marker of efficacy making it easy to monitor. Bariatric surgery also has a positive effect on metabolic outcome, and glycosylated hemoglobin levels must be interpreted taking into account the effect of the surgery on metabolism throughout the postoperative period.

Studies should focus on the pharmacokinetic parameters of interest to propose dose adaptation strategies: maximal concentrations for drugs with short half-lives, such as several analgesics, and steady-state concentrations for drugs with prolonged half-lives, such as antidepressants. Studies should include clear advice for dose adaptation when possible.

In clinical practice, there is not only a significant risk of therapy failure because of subtherapeutic plasma levels for some drugs, but also an increased risk of toxicity due to lower intestinal drug metabolism or increased maximal concentration for others. Predicting the specific pharmacokinetic change associated with bariatric surgery and its clinical importance is challenging. Bibliographic sources should be systematically screened, but they do not currently resolve most clinical situations. Clinicians should be aware that individualized drug optimization is required, taking into account changes in the physiology and comorbidities of patients throughout the postoperative period, as much as possible. They should also consider interindividual variability of drug pharmacokinetics and pharmacodynamics due to the surgery itself, along with any well-known intrinsic variability.

Further research is clearly needed due to the unique characteristics of the bariatric population and each procedure. Available funding clearly limits the ability to perform all needed studies. As the population of bariatric patients grows, pharmaceutical manufacturers may contribute to this research, as none currently propose recommendations for dose adaptation.

6. Expert opinion

The sample sizes of studies performed to date are small and we cannot presume that results based on small numbers of subjects can be generalized. Pharmacokinetic studies are mostly performed in parallel patient cohorts consisting of those who underwent bariatric surgery a few months prior and their controls, often matched based on body mass index, age, and gender. Moreover, it appears that changes in drug disposition are drug-specific and difficult to predict

This area of research is of great interest to pharmacologists and toxicologists. Appropriate clinical studies could substantially improve the knowledge in several areas of clinical pharmacology, to the benefit of the overall population, including the role of the stomach and intestinal barrier in determining drug bioavailability and pharmacokinetics, the influence of morbid obesity and its associated metabolic disorders, and the influence of weight loss on drug disposition.

From the point of view of the prescriber, the biggest challenge remains the patients themselves. An increasing amount of information for drug dosing is required to guide clinical practice, especially as the growing population of patients who have undergone bariatric surgery will age and require medication.

All surgeries, not only malabsorptive, should benefit from clinical studies. Various drugs, having several metabolic pathways, galenic forms, and circumstances of administration, should be considered. Studies should include clear advice for dose adaptation, when possible.

Further research is clearly needed due to the unique characteristics of the bariatric population and each procedure. Priority should be given to drugs that do not have clinical or biological surrogates for dose adaptation. Available funding clearly limits the ability to perform all needed studies. As the population of bariatric patients grows, pharmaceutical manufacturers may contribute to this research, as none currently propose recommendations for dose adaptation.

Declaration of interest

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Additional information

Funding

This paper was not funded.