Research Letter
Comparative Safety of Robotic-Assisted vs Laparoscopic Cholecystectomy in Contemporary Practice
Cody Lendon Mullens, Eunice Y. Lee, Jyothi R. Thumma, et al
JAMA Surg Published Online: May 27, 2026
doi: 10.1001/jamasurg.2026.1585
Robotic-assisted cholecystectomy has expanded rapidly in the US and is thought to now comprise one-quarter of cholecystectomies.1 Prior work has identified an association between robotic-assisted cholecystectomy and higher rates of bile duct injury (BDI), raising concerns about the safety of the robotic-assisted approach.2,3 However, other analyses have suggested that robotic-assisted and laparoscopic cholecystectomy have similar safety profiles and that robotic-assisted approaches may have some benefits such as lower conversion to open rates.4 One criticism of work that identified higher BDI rates is that the data may reflect the earlier adoption period of robotic-assisted surgery when many surgeons were still on their learning curve.5 In this context, we used contemporary data to evaluate trends in robotic-assisted cholecystectomy use and compared risk-adjusted outcomes with laparoscopic cholecystectomy among Medicare beneficiaries.
Methods
We conducted a retrospective cohort study of Medicare fee-for-service beneficiaries undergoing inpatient cholecystectomy between January 1, 2020, and December 31, 2023. We excluded beneficiaries with a diagnosis of gallbladder or hepatopancreaticobiliary cancer. Procedures were distinguished as robotic assisted or laparoscopic using procedure codes in Medicare claims. Patient demographics, Elixhauser comorbidities, and gallbladder disease diagnoses were identified from claims data. The University of Michigan Institutional Review Board waived the need for ethics review and informed consent due to the use of retrospective deidentified secondary data. The study followed the STROBE reporting guideline.
Descriptive statistics were used to compare patient characteristics between groups, with 2-sample t tests and Pearson χ2 tests where appropriate. The primary outcome was BDI requiring operative repair within 1 year using previously described methodologies.2 ,3,6 Secondary outcomes included biliary intervention, any postoperative complication, and serious complications. The eMethods in Supplement 1 present coding strategies and details.2,3Multivariable logistic regression models were used to estimate risk-adjusted outcome probabilities accounting for beneficiary age, sex, race and ethnicity, 29 Elixhauser comorbidities, biliary diagnosis, and year of surgery. Adjusted risks, risk differences, and relative risks (RRs) were estimated using predictive margins, with SEs clustered at the hospital referral region level. Two-tailed P < .05 was considered significant. Analyses were performed using Stata, version 19 (StataCorp LLC).
Results
Among 379 970 Medicare beneficiaries undergoing cholecystectomy between 2020 and 2023, 58 045 procedures (15.3%) were performed robotically. Robotic-assisted cholecystectomy use increased substantially during the study period, rising from 9.2% of cases in 2020 to 22.4% in 2023. Robotic-assisted cases were more often elective and occurred slightly more frequently among patients with multiple comorbidities (Table 1).
Table 1. Patient Characteristics of Medicare Beneficiaries Who Underwent Robotic-Assisted vs Laparoscopic Cholecystectomy, 2020-2023a
| Characteristic | Cholecystectomy type (N = 379 970) | P value | |
|---|---|---|---|
| Robotic assisted (n = 58 045) | Laparoscopic (n = 321 925) | ||
| Age, mean (SD), y | 72.7 (10.6) | 73.1 (10.7) | <.001 |
| Sex | |||
| Female | 29 427 (50.7) | 164 762 (51.2) | .03 |
| Male | 28 618 (49.3) | 157 163 (48.8) | |
| Race and ethnicityb | |||
| American Indian or Alaska Native | 314 (0.5) | 2214 (0.7) | <.001 |
| Asian or Native Hawaiian or Other Pacific Islander | 1246 (2.1) | 7607 (2.4) | |
| Black | 5205 (9.0) | 26 589 (8.3) | |
| Hispanic | 2683 (4.6) | 14 341 (4.5) | |
| White | 46 187 (79.6) | 257 270 (79.9) | |
| Otherc | 1224 (2.1) | 7183 (2.2) | |
| Unknown | 1186 (2.0) | 6721 (2.1) | |
| No. of Elixhauser comorbidities | |||
| 0 | 2462 (4.2) | 15 823 (4.9) | <.001 |
| 1-2 | 18 695 (32.2) | 106 593 (33.1) | |
| ≥3 | 36 888 (63.6) | 199 509 (62.0) | |
| Elective admission | 10 954 (18.9) | 26 673 (8.3) | <.001 |
| Diagnosis | |||
| Cholecystitis | 48 491 (83.5) | 282 168 (87.7) | <.001 |
| Other | 9554 (16.5) | 39 757 (12.3) | |
| Year of surgeryd | |||
| 2020 | 8326 (9.2) | 82 625 (90.8) | <.001 |
| 2021 | 11 901 (12.5) | 83 385 (87.5) | |
| 2022 | 15 266 (16.4) | 77 825 (83.6) | |
| 2023 | 22 552 (22.4) | 78 090 (77.6) | |
After risk adjustment, robotic-assisted cholecystectomy was associated with a higher risk of BDI compared with laparoscopic cholecystectomy (0.38% vs 0.18%; RR, 2.09 [95% CI,1.69-2.49]) (Table 2). There was no significant difference in biliary intervention between the 2 approaches (3.09% vs 2.93%; RR, 1.06 [95% CI, 0.98-1.13]). Robotic procedures were associated with slightly higher risks of any postoperative complication (27.04% vs 26.46%; RR, 1.02 [95% CI, 1.00-1.04]) and serious complications (10.49% vs 9.60%; RR, 1.09 [95% CI, 1.06-1.13]) (Table 2).
Table 2. Risk-Adjusted Outcomes After Robotic-Assisted vs Laparoscopic Cholecystectomy Among Medicare Beneficiaries, 2020 to 2023
| Outcomea | Rate, % | Absolute difference, % (95% CI) | Relative risk (95% CI) | |
|---|---|---|---|---|
| Robotic-assisted cholecystectomy (n = 58 045) | Laparoscopic cholecystectomy (n = 321 925) | |||
| Bile duct injury | 0.38 | 0.18 | 0.20 (0.13-0.27) | 2.09 (1.69-2.49) |
| Biliary intervention | 3.09 | 2.93 | 0.17 (−0.06 to 0.39) | 1.06 (0.98-1.13) |
| Any complication | 27.04 | 26.46 | 0.58 (0.06-1.10) | 1.02 (1.00- 1.04) |
| Serious complication | 10.49 | 9.60 | 0.89 (0.54-1.23) | 1.09 (1.06- 1.13) |
Discussion
This national analysis of Medicare beneficiaries undergoing cholecystectomy between 2020 and 2023 had 2 principal findings. First, the use of robotic-assisted cholecystectomy continued to increase substantially. Second, despite its growth, robotic-assisted cholecystectomy remained associated with higher risk-adjusted rates of BDI compared with laparoscopic cholecystectomy, while differences in overall complication rates between approaches were small.
These findings extend prior work that demonstrated higher rates of BDI in robotic-assisted cholecystectomy.2,3BDI remains the most feared complication of cholecystectomy because of its substantial morbidity and long-term consequences.7,8 Prior national analyses reported approximately 3-fold higher BDI rates among Medicare beneficiaries undergoing robotic-assisted compared with laparoscopic cholecystectomy.2,3 In this contemporary cohort, the relative difference was smaller but remained approximately 2-fold. This finding suggests that although outcomes may be improving as robotic surgery matures, meaningful differences between approaches persist. Although the magnitude of this difference is decreasing, parity with laparoscopic cholecystectomy may only emerge with substantially greater experience with robotic-assisted platforms and likely shifts in operative expertise away from laparoscopy.
The continued expansion of robotic cholecystectomy despite limited evidence of clinical advantage has raised important questions regarding its role in clinical practice.9 In addition to safety concerns, robotic-assisted cholecystectomy has been associated with higher procedural costs without improvements in patient outcomes.1 Although our study was not designed to evaluate causal mechanisms, these findings may reflect an ongoing national learning curve despite increasing case volumes or the use of cholecystectomy as a training procedure for surgeons adopting robotic-assisted platforms, highlighting the potential importance of structured support and feedback during early adoption.
This study has limitations, including data that were limited to the inpatient setting. Administrative claims lack detailed operative and clinical information (eg, conversion to open, lower-severity BDI that did not require operative repair). However, the outcomes we evaluated are reliably coded in claims data, and our findings are nationally generalizable. As robotic-assisted cholecystectomy continues to expand in clinical practice, greater scrutiny of its safety and value may be warranted.
