Altered gastric anatomy and oral semaglutide: anatomy, pathophysiology and the case for the injectable route

Oral semaglutide is absorbed in the stomach rather than the small bowel, across a small area of mucosa immediately around the eroding tablet, and its absolute bioavailability of approximately 0.8% compares with 89% for the subcutaneous formulation. That single difference is what makes previous gastric surgery a route-selection question rather than a general caution: any operation that removes gastric mucosa, accelerates gastric emptying, excludes gastric tissue from the transit stream or removes the pylorus is predicted to reduce absorption of the tablet, while none of them affects a subcutaneous injection at all. The evidence base here is thin in a way that should be stated plainly rather than smoothed over: there is no pharmacokinetic study, case series or registry analysis of oral semaglutide in patients with prior bariatric or gastric surgery, the diabetes trial programme explicitly excluded such patients, and the product information states that no therapeutic experience exists. Our view is that patients who have had a sleeve gastrectomy, any form of gastric bypass, a duodenal switch procedure or a partial or total gastrectomy should be offered a subcutaneous glucagon-like peptide-1 receptor agonist rather than the tablet, that patients with mucosa-preserving and pylorus-preserving procedures such as an adjustable band sit in a genuinely uncertain middle group, and that the injectable route carries the additional advantage of direct trial evidence in exactly this population.

The pharmacological pivot

The mechanism by which oral semaglutide reaches the circulation was established by Buckley and colleagues in Science Translational Medicine, and it is unusual enough that it does not generalise from other oral drugs. Each tablet carries 300 mg of the absorption enhancer sodium N-(8-[2-hydroxybenzoyl]amino)caprylate. As the tablet erodes against the gastric wall, the enhancer raises the pH in the small volume of fluid around it, which protects semaglutide from pepsin because pepsin has negligible activity above approximately pH 5, promotes monomerisation of the peptide, and facilitates its transcellular passage across the gastric epithelium. Gamma-scintigraphy places the eroding tablet within the gastric cavity with absorption into the gastric venous drainage, and a readable clinical summary of the mechanism appears in Clinical Diabetes.

Three consequences follow, and they are what a clinician needs rather than the biochemistry. Absorption happens in the stomach and not further down the gut, so a tablet that leaves the stomach early has largely missed its opportunity. Absorption happens across a small patch of mucosa rather than the whole gastric surface, so the tablet needs somewhere to sit against intact epithelium. And the enhancing effect is concentration-dependent and transient, so the local environment has to be preserved for the minutes it takes the tablet to erode.

The measured result of all that is an absolute bioavailability modelled at 0.795% with a 95% confidence interval of 0.736 to 0.864 in the population pharmacokinetic analysis by Overgaard and colleagues. Subcutaneous semaglutide, by contrast, has an absolute bioavailability of 89%, recorded in the European product information, and reaches the circulation by a route that is indifferent to gastrointestinal anatomy. Tirzepatide and liraglutide share that indifference. The route of administration is therefore not a matter of patient preference in a post-surgical patient; it is the variable that determines whether the drug arrives.

Normal gastric anatomy and where the tablet actually works

The stomach divides into the cardia, fundus, body, antrum and pylorus, and the histology is not uniform across them, which matters for a drug absorbed locally. Parietal cells, which secrete hydrochloric acid and intrinsic factor, and chief cells, which secrete pepsinogen, concentrate in the glands of the fundus and body, while the antrum is dominated by mucous and endocrine cells; the surface is lined throughout by foveolar mucous cells producing the protective mucus layer, described in the normal gastric histology summary from Pathology Outlines and in the OpenStax account of gastric structure and function.

Two functional features shape where a swallowed tablet ends up. A relatively direct channel runs along the lesser curvature, referred to in the surgical and physiological literature as the magenstrasse or gastric canal, along which swallowed material can travel towards the antrum. And solids empty considerably more slowly than liquids, because the antrum grinds and sieves before the pylorus admits particles onward. A tablet taken with a small volume of water on an empty stomach therefore has both a surface to erode against and time in which to do it, and both of those are properties of an intact stomach rather than of the tablet.

The five features that absorption depends on

Setting out what the mechanism requires makes the effect of each operation predictable without needing operation-specific data.

• Mucosal surface area matters because the tablet needs intact epithelium to erode against and to cross, so procedures that resect mucosa or bury it reduce the available absorptive surface.
• Gastric residence time matters because absorption competes with emptying, so procedures that accelerate emptying shorten the window in which the tablet can work.
• The pylorus matters as a gatekeeper that holds solids back for grinding, so procedures that remove or bypass it allow a partly eroded tablet through to a segment of gut where the enhancement mechanism does not operate.
• Gastric pH matters because the enhancer works by locally raising a low pH, and a stomach that is already hypochlorhydric presents a different starting point, although whether that helps or hinders has not been established.
• Continuity of the transit stream matters because a bypass leaves gastric mucosa in the abdomen that a swallowed tablet will never reach.

Operation by operation

The table sets out each procedure with the anatomy created and the predicted direction of effect on oral semaglutide absorption; the predictions are mechanistic, because no procedure in this table has been studied with this drug.

The pattern that emerges is worth stating in prose because it is more useful than the individual rows. Procedures that resect gastric wall, remove the pylorus or divert swallowed material past gastric mucosa all point the same way, and the size of the predicted effect tracks how much functional gastric mucosa remains in the path of a swallowed tablet. Procedures that restrict without resecting sit in a different group, and gastric plication is the interesting case within it: nothing is removed, but the imbricated greater curvature is folded away from the lumen, so the functional absorptive surface falls even though the mucosa is anatomically present.

What the general post-surgical drug absorption literature shows

The wider literature on oral drug bioavailability after bariatric surgery is substantial and, read carelessly, gives false comfort. The systematic review by Angeles and colleagues in Obesity Reviews covered 22 studies across 32 drugs, mostly after Roux-en-Y gastric bypass, and found that the rate of absorption frequently increases after surgery with drug-specific effects on total exposure. A more recent review by Wigg and colleagues reaches the same conclusion of non-uniform, drug-dependent effects, as does the review in Current Obesity Reports.

The reason that literature does not transfer is instructive. A study after sleeve gastrectomy found paracetamol bioavailability increased rather than decreased, reported in the pharmacokinetic literature indexed in PubMed Central, and the explanation is that accelerated gastric emptying delivers paracetamol faster to the small bowel where it is absorbed. Almost every drug in that literature is absorbed in the small bowel, so faster gastric emptying helps it. Oral semaglutide is the opposite case: faster gastric emptying removes the tablet from the only site where its enhancement mechanism works. Applying the general finding to this drug therefore predicts the wrong direction of effect, which is why the mechanism has to lead the reasoning.

The formulation guidance carries a related trap. The perioperative nutritional and metabolic guideline from the American Association of Clinical Endocrinology and partner societies, alongside the ASMBS integrated health nutritional guidelines, advises using crushed or liquid rapid-release preparations in place of extended-release formulations to maximise absorption after surgery. That advice cannot be followed here, because the Rybelsus Summary of Product Characteristics and the United States prescribing information for oral Wegovy both require the tablet to be swallowed whole, since crushing destroys the controlled local erosion the mechanism depends on. A patient with altered anatomy is therefore given a tablet that cannot be adapted in the way post-surgical medication guidance would normally adapt it.

The evidence vacuum, stated plainly

No pharmacokinetic study, case series, cohort or registry analysis has examined oral semaglutide in patients with prior bariatric or gastric surgery. That is the central finding of this article and it should not be presented as caution in the absence of data; it is an absence of data.

The trial programme was designed to exclude these patients. The PIONEER 1 protocol registered as NCT02906930 excludes any history of major surgical procedures involving the stomach potentially affecting absorption of trial product, naming subtotal and total gastrectomy, sleeve gastrectomy and gastric bypass surgery. The public registry records for the obesity trials OASIS 1 and OASIS 4 publish abbreviated eligibility criteria that do not display a surgical exclusion, which means the registry cannot confirm one rather than that none existed. The Rybelsus Summary of Product Characteristics states that there is no therapeutic experience with semaglutide in patients with bariatric surgery, and the United States labels for both the diabetes and weight-management tablets carry no bariatric surgery warning, contraindication or dose adjustment at all, their only relevant statement being that the drug is not recommended in severe gastroparesis.

One surrogate finding offers partial reassurance and should not be stretched. The Overgaard analysis found no significant difference in oral semaglutide pharmacokinetics between subjects with and without upper gastrointestinal disease such as gastritis and reflux, with an exposure ratio of approximately 1.18 and a 95% confidence interval of 0.80 to 1.75. That tells us mucosal inflammation does not appear to disrupt absorption, and it tells us nothing about a stomach that has been resected, stapled or bypassed, because the mechanism at issue is anatomical rather than inflammatory.

Acid suppression and the non-surgical causes of altered gastric physiology

Proton pump inhibitor use is close to routine in this population, which raises a reasonable question about whether raising gastric pH pharmacologically interferes with an absorption mechanism that works by raising pH locally. In a French national cohort, continuous proton pump inhibitor use rose from 10.9% to 26.5% four years after sleeve gastrectomy or gastric bypass, reported in Surgery for Obesity and Related Diseases, and a Danish nationwide cohort of 17,740 patients found a markedly higher hazard of proton pump inhibitor initiation after sleeve gastrectomy than after Roux-en-Y bypass, at 7.06 with a 95% confidence interval of 6.42 to 7.77, alongside a higher hazard of reflux diagnosis at 1.93, reported in PubMed.

The interaction question has been answered directly, and reassuringly. Baekdal and colleagues randomised 54 healthy subjects to omeprazole 40 mg daily alongside oral semaglutide and found a slight and statistically non-significant increase in semaglutide exposure, with an area under the curve ratio of 1.13 and a 90% confidence interval of 0.88 to 1.45, and no dose adjustment is required. Acid suppression is therefore not the problem in a post-surgical patient; the anatomy is.

The non-surgical causes of altered gastric physiology deserve a mention because they raise the same mechanistic question without any surgical history to prompt it. Achlorhydria, atrophic gastritis, Helicobacter pylori infection and pernicious anaemia all change the gastric environment in which the tablet is designed to work, and no oral semaglutide data exist in any of them.

Why the injectable route is the stronger option in this group

The argument for a subcutaneous product in post-surgical patients rests on two separate legs, and the second is often forgotten. The first is the pharmacological one already made: 89% bioavailability by a route that gastrointestinal anatomy does not touch. The second is that the injectables have been studied in precisely this population, which no oral formulation has.

GRAVITAS30157-3), reported by Miras and colleagues in The Lancet Diabetes and Endocrinology, randomised 80 adults with persistent or recurrent type 2 diabetes at least a year after Roux-en-Y gastric bypass or vertical sleeve gastrectomy to liraglutide 1.8 mg or placebo, and found improved glycaemic control with a weight difference of -4.2 kg and a 95% confidence interval of -6.8 to -1.4 at 26 weeks. BARI-OPTIMISE, reported in JAMA Surgery, randomised poor responders after bypass or sleeve to liraglutide 3.0 mg or placebo and found body weight change of -8.82% against -0.54% at 24 weeks. A systematic review and meta-analysis in Obesity Surgery pooled eight studies and 557 patients and found liraglutide produced -6.0 kg against placebo at six months, with a 95% confidence interval of -8.66 to -3.33. Semaglutide 2.4 mg has now been tested in the same setting over 68 weeks in BARI-STEP, reported in Nature Medicine. Tirzepatide has observational and retrospective signals in post-bariatric patients and no dedicated randomised trial that this review located.

A caution belongs alongside that evidence. The cardiovascular outcomes data in the European product information for injectable semaglutide records more serious hypoglycaemia in participants with prior bariatric surgery, at 2.3% against none, which is a signal to note in patients on insulin or a sulfonylurea rather than a precaution against the drug class.

Gastroparesis, delayed emptying and the operated stomach

Every glucagon-like peptide-1 receptor agonist slows gastric emptying, and both the European and United States product information advise against use in severe gastroparesis and record reports of pulmonary aspiration under anaesthesia attributed to residual gastric contents. Adding a drug that slows emptying to a stomach that has been reshaped into a narrow tube, reduced to a small pouch or denervated is mechanistically uncomfortable, and the honest position is that no data quantify anastomotic or pouch complications attributable to this combination. What follows from that is a practical point rather than a prohibition: patients in this group who are listed for any procedure requiring sedation or general anaesthesia need their glucagon-like peptide-1 treatment declared to the anaesthetist, and the surgical history stated with it.

The practical framework

Reading the mechanism, the labelling and the trial evidence together supports a three-way split rather than a single rule; the middle group is genuinely uncertain and should be treated as such rather than resolved by preference.

What is not known

The gaps here are larger than the findings and should be named. There is no direct pharmacokinetic or clinical data on oral semaglutide after any bariatric or gastric procedure, and none in achlorhydria, atrophic gastritis, Helicobacter pylori infection or pernicious anaemia. The obesity trial registry records do not confirm whether patients with prior surgery were excluded. No completed randomised trial of tirzepatide in post-bariatric patients was located. And no expert consensus statement or editorial addresses the specific question of which post-surgical patients might reasonably be offered the tablet, so the framework above is reasoning from mechanism and labelling rather than from guidance. The European product information covering the tablet formulation should be read directly on the European Medicines Agency website as the consolidated document is updated, since it may add wording on gastric surgery that the diabetes tablet label does not carry.

Position

Previous gastric surgery is a question about route rather than a question about drug, and the answer follows from where oral semaglutide is absorbed: a tablet that works across a small patch of gastric mucosa in a short window cannot be relied on in a stomach that has been resected, stapled, bypassed or denervated, while a subcutaneous agonist reaching the circulation at 89% bioavailability is unaffected and has been tested in this population. The next step for any service prescribing oral semaglutide is to make prior gastric and bariatric surgery an explicit screening question with the procedure named rather than a general enquiry about abdominal operations, since the distinction between a band and a bypass changes the recommendation. Reviewed and current as at 27 July 2026.

Key questions

After gastric surgery, is oral semaglutide a drug question or a route question?

A route question. Oral semaglutide is absorbed across a small patch of gastric mucosa in a short window, so a stomach that has been resected, stapled, bypassed or denervated threatens the tablet, while a subcutaneous agonist reaching the circulation at about 89 per cent bioavailability is unaffected by gastrointestinal anatomy.

Which post-surgical patients should be offered the injectable rather than the tablet?

Sleeve gastrectomy, any gastric bypass, duodenal switch procedures, and partial or total gastrectomy: predicted loss of absorption combined with direct trial evidence for the injectable route. Mucosa- and pylorus-preserving procedures such as an adjustable band sit in a genuinely uncertain middle group.

Is there direct evidence for oral semaglutide after bariatric surgery?

No. There is no pharmacokinetic study, case series or registry analysis of oral semaglutide after bariatric or gastric surgery, the diabetes trial programme excluded such patients, and the product information states that no therapeutic experience exists. The injectables, by contrast, have randomised evidence in this population.