UK trial: remove the gallbladder or wait?

Overview

Innes K et al.. Health Technol Assess 2024 · PMID 38943314 · DOI 10.3310/MNBY3104

Over 18 months general pain scores were similar, but about 30% of the 'wait' group needed surgery, 11.5% had gallstone complications before surgery, and gallstone-specific quality of life was better after surgery.

Does the study support the specific claim? Partly · Methodological quality: high

Original publication

Plain-language summary

Scientific in-depth review

1. Full reference

Authors

Innes K et al.

Journal

Health Technol Assess

Year

2024

Identifier

PMID 38943314 DOI 10.3310/MNBY3104

Verification

verified – PubMed efetch 2026-10-25

Retracted

no

Corrections

none

Extent checked

full text

Conflicts of interest and funding

NIHR HTA programme (14/192/71); core outcome set also NHS Grampian Endowment; MRC fellowship (one author). Declared interests: NIHR committee memberships, NIHR grants, professional society roles; no industry conflicts reported.

2. Research question

Cholecystectomy vs observation for uncomplicated symptomatic gallstones

3. Study design and population

Study design

Randomised controlled trial – Pragmatic, parallel-group, multicentre superiority RCT (C-GALL; ISRCTN55215960) with 1:1 minimisation (site, sex, age; 20% random element), embedded process evaluation, within-trial cost-utility and 10-year Markov model, and core outcome set development (NIHR HTA report)

Population

Adults with symptomatic uncomplicated gallstone disease (biliary colic or acute cholecystitis) referred to secondary care; mean age 51, predominantly female, 86% white

Country

UK

Period

Recruitment August 2016 - November 2019; follow-up up to 24 months

Sample size

2667 potentially eligible, 1369 eligible, 434 randomised (217 per group) from 20 UK centres; primary analysis 203 vs 205

Inclusion and exclusion criteria

Excluded: unable to consent, unfit for surgery, pregnancy, previous open major upper abdominal surgery, common bile duct stones or previous choledocholithiasis, history of acute pancreatitis, obstructive jaundice, empyema with sepsis, suspected gallbladder cancer, perforated gallbladder, haemolytic disease

4. Methods and endpoints

Exposure / intervention

Laparoscopic cholecystectomy

Comparator

Observation/conservative management (surgery allowed later if needed)

Endpoints

Primary: SF-36 bodily pain AUC over 18 months. Secondary: Otago gallstones condition-specific questionnaire (CSQ), other SF-36 domains, 24-month AUC, persistent symptoms, complications, further treatment, costs/QALYs

Statistical methods

Linear regression of AUC adjusted for minimisation covariates and baseline; multiple imputation for missing 18-month data; mixed models for repeated measures; Poisson models for complications and further treatment; sensitivity analyses for compliance, missing data and COVID-19

5. Actual results

Results

Surgery received by 24 months: 64 (29.5%) conservative vs 153 (70.5%) cholecystectomy group (median time to surgery 9.0 vs 4.7 months). Primary: SF-36 bodily pain AUC 46.8 in both, MD -0.0 (95% CI -1.7 to 1.7), p=0.996. CSQ favoured cholecystectomy: MD 6.6 (1.9-11.3) at 18 months and 9.0 (4.1-14.0) at 24 months (about 0.4 SD); persistent symptoms score MD 10.1 (4.2-16.0) at 24 months. Some SF-36 domains favoured surgery at 18 months but not 24 months. Participants with complications up to 24 months: 32 (14.7%) vs 46 (21.2%), RR 0.69 (0.44-1.09); pre-surgery complications (mostly cholecystitis/biliary colic, some pancreatitis) 25 (11.5%) vs 11 (5.1%); intraoperative 9 (4.1%) vs 24 (11.1%), mostly bile/stone spillage. Further treatment at 24 months 5.0% vs 7.9%, RR 0.62 (0.28-1.38). One cardiovascular death (cholecystectomy group). Cost difference -GBP 1033 for conservative management; QALY difference -0.019 (n.s.); ICER GBP 55,235 per QALY lost; sensitivity with longer-term QoL reduced savings to GBP 14,700 per QALY lost.

Absolute risks

29.5% of the conservative group had surgery within 24 months; pre-surgery gallstone complications 11.5% (conservative) vs 5.1%; overall complications 14.7% vs 21.2%; further treatment 5.0% vs 7.9%.

Authors' conclusion

In the short term (up to 24 months) observation/conservative management may be a cost-effective use of NHS resources in selected patients, but subsequent surgeries and QoL differences beyond 24 months could reverse this; long-term follow-up and identification of patients who should be offered surgery are needed.

6. Strengths

• Large pragmatic multicentre RCT with concealed minimisation
• Prespecified primary outcome and sensitivity analyses (compliance, missing data, COVID-19)
• Representative of UK gallstone population
• Economic evaluation and core outcome set
• No industry funding

7. Weaknesses

• Unblinded participants and surgeons; all outcomes patient-reported or unblinded (authors)
• Substantial crossover (29.5% conservative had surgery; 29.5% of surgery group not operated by 24 months)
• Long NHS waiting lists (median 4.7 months to surgery) dilute the contrast (authors)
• High questionnaire non-response at later time points (about 100/217 per arm at 24 months); non-responders younger with worse baseline pain (authors)
• Only 24 months follow-up
• Internal inconsistency between 18-month (10.1% vs 25.3%) and 24-month (14.7% vs 21.2%) complication percentages
• Clinical importance of CSQ difference uncertain (authors)

8. Risk of bias

Orienting tool

RoB 2

Type of assessment

structured assessment based on the full text by a single reviewer (no formal double review)

Full formal assessment

no

Overall risk of bias

some concerns

Rationale

Structured single-reviewer judgement from full text; a formal assessment would require two independent reviewers and the protocol.

Full-text review (25 October 2026): Added eligibility/exclusions, recruitment flow, CSQ and persistent-symptom results at 18/24 months, detailed complications (pre-surgery 11.5% vs 5.1%), further treatment, one death, cost-utility and modelling, funding/COI, limitations incl. non-response and internal inconsistency; RoB 2 domains added, overall unchanged (some_concerns).

9. Assessment of the statistical interpretation

Policy comparison (offer surgery vs observe) with diluted contrast; no difference in generic pain QoL, but gallstone-specific symptoms favour surgery by about 0.4 SD. Gallstone complications (cholecystitis, pancreatitis) did occur while waiting/observed (11.5%), and about 30% of the observed group needed surgery within 2 years.

10. Relevance to the specific claim

Shows watchful waiting is reasonable for some symptomatic patients (partial support to C-ORG-07's 'don't rush'), but that gallstones cause symptoms requiring surgery in about 30% within 2 years and that condition-specific QoL is better after surgery.

11. Independent contrary findings

12. Transferability to Germany and Europe

Directly European.

13. Overall methodological judgement

Key randomised evidence that observation is a reasonable short-term option for selected patients with uncomplicated symptomatic gallstones, while showing that gallstones do cause symptoms and complications and that gallstone-specific symptoms are better after surgery.

14. Open questions

Outcomes beyond 24 months; which patients benefit from surgery; applicability outside UK waiting-list settings.

15. Linked fact checks

• Adenoids, tonsils and the gallbladder are organs that should not be removed; correcting the diet should be the foundation of treatment. – Studies contradicting or limiting the claim
• Gallstones cause no symptoms and no harm, have nothing to do with pain after eating, and the gallbladder should not be removed. – Studies contradicting or limiting the claim