The Sleeve Without a Scalpel: ESG Meets the Registry

Laparoscopic sleeve gastrectomy removes roughly four-fifths of the stomach and leaves a stapled tube. Endoscopic sleeve gastroplasty does something that looks similar on a fluoroscopy screen and nothing like it in an operating theatre: it runs a series of full-thickness sutures through the stomach wall from the inside, drawing the greater curvature in on itself until the organ is plicated into a narrow channel. No incisions, no staples, no resected tissue, no abdominal wall to heal.

The appeal is obvious, and the volumes show it. A paper published in Surgical Endoscopy on 15 October puts the two procedures side by side in the same national registry for the same year. It sits alongside a second, independent analysis of that file and a fresh meta-analysis — and together they describe a procedure whose safety case is better than its metabolic case.

First, the randomized evidence

Endoscopic sleeve gastroplasty, or ESG, was tested properly once. The MERIT trial, reported in The Lancet in 2022, randomised 209 adults with class 1 or class 2 obesity at nine US centres to ESG plus diet and activity support, or to diet and activity support alone.

At 52 weeks the ESG arm lost 49.2% of excess weight against 3.2% in the controls, and 13.6% of total body weight against 0.8%. Seventy-seven per cent of the ESG group reached at least 25% excess weight loss, versus 12% of controls. Metabolic comorbidities improved in 80% of ESG participants assessed and worsened in 12%; in the control arm, 45% improved and half worsened. At 104 weeks, 68% of the ESG group still held at least 25% excess weight loss. Serious adverse events attributed to the procedure occurred in three of 131 participants — 2% — with no deaths and no need for intensive care or surgery. The trial was funded by the device manufacturer and the Mayo Clinic.

Two limitations travel with that result. The comparator was lifestyle change, not an operation. And the population was class 1 and class 2 obesity — a lighter group than the one most registry data describe.

The registry answer at six months

The new comparison, from a Cleveland Clinic group, draws on the 2023 MBSAQIP file merged with its six-month follow-up data: 115,476 patients, of whom 114,967 had laparoscopic sleeve gastrectomy and 509 had ESG. Roughly one ESG for every 226 sleeves.

The two cohorts did not start equal. ESG patients were older (44.5 against 42.4 years), lighter at baseline (mean BMI 40.9 against 45.0) and less comorbid. At six months the sleeve produced more than twice the weight loss: a BMI reduction of 8.2 against 3.9 kg/m², and 22.6 against 11.1 kg of total body weight.

The comorbidity numbers diverge more sharply still. After sleeve gastrectomy the study records a near-halving of medication- or insulin-dependent diabetes, plus improvements in dyslipidaemia, hypertension and sleep apnoea. After ESG it records “no clinically meaningful comorbidity improvements from baseline to 6 months.”

The authors are careful about their own comparison: the ESG cohort was less sick to begin with, so there was less room to improve. They are also comparing 509 procedures against nearly 115,000.

A second pass, with matching

The same registry year has now been analysed by a University of Minnesota group in the Journal of the American College of Surgeons using target trial emulation — a method that asks what a randomized trial would have found by mimicking its design on observational data. Patients were matched 1:8 on age, sex, race, diabetes, hypertension, cardiovascular comorbidity and highest preoperative BMI, giving 3,753 patients: 417 ESG and 3,336 sleeve.

Six-month weight change from highest preoperative weight was −12.33% after ESG and −18.51% after sleeve, a gap of roughly seven percentage points that survived inverse-probability weighting. Comorbidity outcomes again favoured the sleeve.

Then the counterintuitive finding. Readmission was higher after the endoscopic procedure — 3.8% against 1.6% — as was reoperation, 1.0% against 0.3%. Both are low in absolute terms, and a registry cannot explain them. But the less invasive procedure carried the burden of surprise visits in this cohort, not the operation.

Where ESG looks better

The pooled literature tilts the other way on tolerability. A systematic review and meta-analysis in Surgical Endoscopy in October gathered ten studies and 7,442 patients — 3,583 ESG, 3,859 sleeve — with a mean follow-up of 32.1 months.

Weight loss again favoured the sleeve, by 8.3 percentage points of total body weight at six months, 12.6 at one year and 9.8 at two years. But ESG was associated with roughly half the overall adverse events (relative risk 0.54) and a third the rate of new-onset reflux disease (0.33), an outcome that matters because sleeve gastrectomy is a recognised reflux risk. Note what got pooled: mean age 35.5 years and mean BMI 34.6 — trial-shaped patients, not the practice population in either registry analysis.

The direction of travel

Volume growth has been steep from a small base. Analysts at Northwestern and Michigan counted 2,371 ESG procedures in MBSAQIP between 2020 and 2023 — an increase of 50.5% across the period, with the mean BMI of patients rising from 37.8 to 40.8 while serious adverse events stayed low and flat. The same group looked specifically at clot risk in Gastrointestinal Endoscopy: across the same 2,371 patients, 30-day deep vein thrombosis occurred in 0.3% and pulmonary embolism in 0.2%. Two-thirds of those who developed a clot had received no prophylaxis at all — and, the authors note, there are no established guidelines for periprocedural clot prevention after an endoscopic procedure, unlike for surgery.

The economics are unresolved and comparator-dependent. A UK Markov model in Obesity Surgery found conventional sleeve gastrectomy cost-effective against ESG for class II–III obesity at £10,593 per quality-adjusted life year, while ESG looked cost-effective against semaglutide for class I–II obesity at £7,267. A Singapore disease-progression model in SOARD compared ESG plus lifestyle with lifestyle alone rather than with surgery, and landed at SG$24,580 (about US$18,410) per QALY. Guidance is being written: the ASMBS Clinical Issues Committee published a PRISMA systematic review of endobariatric procedures in January, and Italy’s SICOB issued GRADE-based endobariatric guidelines in February. Patient-facing material lags the literature — NIDDK’s list of surgical options, last updated in April 2026, names sleeve gastrectomy, gastric bypass, the adjustable band and the duodenal switch, with endoluminal devices parked in a separate section and a caution that they have not been studied over the long term.

What this does not settle

No randomized trial has compared ESG with sleeve gastrectomy head to head; the randomized evidence sets ESG against lifestyle change. The registry comparisons inherit their selection — ESG patients were older, lighter and less comorbid, and no matching scheme fully repairs that. And the pooled meta-analysis describes a population some ten BMI points lighter than the registry cohorts.

For all the volume growth and the appeal of an incisionless option, the metabolic claim is the one still waiting for its evidence. Sleeve gastrectomy’s diabetes effect is the reason the operation is called metabolic as well as bariatric; six months of registry data after ESG do not show the same effect, and the trials that could show it do not exist. The LABS study, funded by NIDDK, remains the template for the kind of multi-year prospective cohort that would answer the question — and it was never designed with an endoluminal arm. Whether ESG belongs in the same conversation as surgery, or in the conversation about what happens to procedure volumes overall, depends on staging by BMI class and comorbidity, which is the work a BMI calculator can only begin.

None of this constitutes a recommendation for or against any procedure. Both are interventions with real risks, and the published record is what this archive reports.

niddklabs.org is an independent archive and is not affiliated with, endorsed by, or speaking for the NIH or NIDDK. This article reports published research and does not offer medical advice.