Notice of Retraction and Replacement. Gross et al. Selective vs Routine Cholangiography Across a Health Care - Summary - MDSpire
Coming Soon: Introducing MDSpire News. Learn more
Conexiant’s news site is now MDSpire News. Learn more

Retraction and Replacement Notice: Gross et al. on Selective Versus Routine Cholangiography Across a Health Care Enterprise

  • By

  • Toms Augustin

  • November 1, 2026

Share

Objective:

To correct the original analysis comparing perioperative outcomes among patients treated by surgeons using routine cholangiography (RC) or selective cholangiography (SC) during cholecystectomy.

Approach:
  • Study population: The study included 28,212 cholecystectomies for benign biliary disease performed from January 2015 through June 2023 across 18 hospitals and 9 ambulatory surgery centers.
  • Revised analysis: After review identified concerns about sparse events and inverse-frequency weighting, the authors designated Firth penalized logistic regression with hospital fixed effects as the primary model for major bile duct injury (BDI). Hierarchical mixed-effects model…
Key Findings:
  • The cohort included 10,244 RC and 17,968 SC cases, with 26 major BDIs and 105 minor BDIs.
  • In the primary revised model, RC was not significantly associated with major BDI (OR, 0.48; 95% CI, 0.18-1.31) or minor BDI (OR, 0.93; 95% CI, 0.56-1.54).
  • The revised conclusions state that RC did not reduce major or minor BDIs or perioperative ERCP use; it was associated with a clinically insignificant increase in operative time.
  • BDIs clustered among a small number of surgeons and hospitals: 5 of 134 surgeons accounted for more than half of major BDIs, and 1 hospital accounted for 10 of 26.
Interpretation:

The revised analyses changed the original primary findings, which had reported lower odds of major and minor BDI with RC. The authors requested retraction and replacement so the corrected analyses and conclusions would be presented consistently.

Limitations:
  • Major BDI events were sparse, with 26 events in the study cohort.
  • The notice reports substantial clustering of BDI events by surgeon and hospital.
Conclusion:

The article was corrected online, with revised text and supplements documenting the changes.

Sources:

Original Source(s)

Related Content