Same 30 Patients, Same Slides, Three Answers on Whether Cirrhosis Reversed

A Cleveland Clinic paired-biopsy series reports histologic cirrhosis regression after metabolic surgery in 33%, 40% or 60% of patients depending on which staging system scores the slides. The endpoint liver trials actually require is not a biopsy at all.

bariatric surgerylong-term outcomesevidence reviewmetabolic surgery

Thirty adults with biopsy-proven compensated MASH cirrhosis had metabolic surgery at the Cleveland Clinic between 2004 and 2024, and a second liver biopsy a median of 6.2 years later. The results, published on 2 November in Annals of Surgery, test hepatology’s oldest working assumption — that a cirrhotic liver stays cirrhotic — and return a number that depends on which ruler the pathologist picks up. On the NASH Clinical Research Network scale, 33% of the patients had regressed from stage 4 to stage 3 or lower. On modified Ishak staging, 40% had. On the Beijing classification, 60% showed features of regressive remodelling.

The patients are identical in all three counts. That is the part worth sitting with: “did the cirrhosis reverse” is not one measurement, and a single paired-biopsy series can report a third or three-fifths of the same cohort as improved without anyone having misread a slide.

What the cohort actually held

Thirty people, 22 of them women, median age 54.2, median baseline BMI 43.7 kg/m². Median weight reduction was 31.8 kg and median total weight loss 23.9% — the sort of loss a working bypass or sleeve produces and a diet rarely does. Alongside the stage shifts, steatosis, lobular inflammation and hepatocyte ballooning all improved. The authors’ own verb is careful: compensated cirrhosis “may be reversible after sustained metabolic intervention.”

The design underneath that verb is a retrospective single-centre series with no control group. Every patient in it had a repeat biopsy, over an interval whose interquartile range runs from 2.0 to 9.2 years. Two years is a short window in which to declare scar tissue permanent or gone.

The denominator is a second needle

A paired-biopsy study can only describe people who agreed to be needled twice: patients who survived, stayed in follow-up, and had a clinician who judged a second liver puncture worth the risk. How many cirrhotic patients the same surgeons operated on and never re-biopsied does not appear in the abstract. This is not a flaw unique to Cleveland Clinic — it is the structural weakness of the whole histologic literature, and it is why 33% should be read as “of those who came back for a second look,” not “of those who had the operation.” This archive covered the blood-test version of the same problem in its article on FIB-4 after sleeve gastrectomy, where the index moved and the biopsy was never taken.

What operating on a scarred liver costs

The safety numbers come from somewhere else, and they are an average rather than a comparison. A single-arm systematic review and meta-analysis in Obesity Surgery (October 2025; 26 studies, 22,489 participants) pooled 3.33% mortality (95% CI 2.06–5.35) and 14.0% early postoperative complications (7.56–24.49) among patients with well-compensated cirrhosis. The heterogeneity statistics are the honest part: I² of 96.7% for mortality, 90.3% for complications. An average drawn across series that disagree that violently describes a literature; it is not an estimate you can hand to anyone. The same review found 26.43% total weight loss at one year and a median stay of 3.77 days — real metabolic benefit, in a population selected by whoever decided these patients were well enough to be operated on.

The endpoint the field actually accepts

Here the story leaves surgery. A methodological review of MASH-cirrhosis trials in Drugs (May 2026) sets out what regulators now expect from a phase 3 programme in compensated cirrhosis: a composite of hepatic decompensation, transplant and death, followed for three to five years, in patients chosen for clinically significant portal hypertension and liver stiffness above 6.5 kPa, with expected effect sizes of 0.70 to 0.85. Histologic reversal of cirrhosis appears only as a possible basis for accelerated approval — the SYMMETRY phase 2b trial of efruxifermin reached F4 regression within 96 weeks.

The review’s authors disclose research funding and consulting fees from several of the companies running those trials, and their selection levers deserve a close read: a ceiling on stiffness, platelets above 110,000/µL, no significant portal hypertension, pre-specified FIB-4 thresholds. Those are plausibly the variables that decide who regresses. A surgical series reporting 33% without them is describing an unselected population against a drug literature that is now preselecting its own.

What would settle it

Not another paired-biopsy series from a high-volume centre. The missing instrument is a registry that records everyone screened: who was offered surgery, who was turned down, what their platelets and stiffness were, and how many went on to decompensate, be transplanted or die — biopsied or not. The LABS study, funded by NIDDK, was built on that principle for 30-day safety, and its design is described in the LABS archive pages.

What this paper adds is precise and modest. A third of a small, self-selected group of cirrhotic livers looked better under the microscope roughly six years later, and that share moves by nearly thirty percentage points depending on the classification used to score the same tissue. That is a hypothesis with unusually good histology behind it. It is not yet an outcome.

This archive is independent of NIH and NIDDK and reports published research only; nothing here is medical advice.

Sources

  1. Cirrhosis Regression After Metabolic Bariatric Surgery in Patients With Compensated MASH Cirrhosis: A Long-term Paired Biopsy Study (Annals of Surgery, 2 November 2026)
  2. Efficacy and Safety of Bariatric Surgery in Well-Compensated Liver Cirrhosis: A Systematic Review and a Single-Arm Meta-analysis (Obesity Surgery, October 2025)
  3. Metabolic Dysfunction-Associated Steatohepatitis (MASH)-Cirrhosis Clinical Trials: Lessons Learned and Future Directions (Drugs, May 2026)
From the blog

New research notes