Abstract
Introduction
Robotic-assisted metabolic and bariatric surgery (MBS) is gaining popularity. Revisional MBS is associated with higher perioperative morbidity compared to primary MBS. The optimal surgical approach to minimize complications in these complex cases is unclear. The goal of this study was to assess robot utilization in revisional MBS and compare laparoscopic and robotic revisional MBS outcomes in the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP) database.
Methods
A retrospective review of the MBSAQIP database was performed identifying revisional sleeve gastrectomy (SG) and Roux-en-Y gastric bypass (RYGB) cases from 2015 to 2022. Primary MBS, open/emergent cases, cases converted to another approach, and combined cases other than esophagogastroduodenoscopy were excluded. 30-Day outcomes for laparoscopic and robotic cases were compared using multivariate logistic regression adjusting for patient demographics, comorbidities, and operative variables.
Results
41,404 Cases (14,474 SG; 26,930 RYGB) were identified. From 2015 to 2022, the percentage of revisional SG and RYGB cases performed robotically increased from 6.1% and 7.3% to 24.2% and 32.0% respectively. Laparoscopic SG had similar rates of overall morbidity, leak, bleeding, readmission, reoperation, and length of stay compared to robotic. Laparoscopic RYGB had significantly higher rates of overall morbidity (6.2% vs. 4.8%, p < 0.001, AOR 0.80 [0.70–0.93]), blood transfusion (1.5% vs. 1.0%, p < 0.05, AOR 0.74 [0.55–0.99]), superficial incisional SSI (1.2% vs. 0.4%, p < 0.001, AOR 0.30 [0.19–0.47]), and longer length of stay (1.87 vs. 1.76 days, p < 0.001) compared to robotic. Laparoscopic operative times were significantly shorter than robotic (SG: 86.4 ± 45.8 vs. 113.5 ± 51.7 min; RYGB: 130.7 ± 64.7 vs. 165.5 ± 66.8 min, p < 0.001).
Conclusion
Robot utilization in revisional bariatric surgery is increasing. Robotic surgery has lower postoperative morbidity and shorter length of stay in revisional RYGB when compared to laparoscopic. Robotic platforms may have the capacity to improve the delivery of care for patients undergoing revisional bariatric surgery.
Keywords: Revisional bariatric surgery, Robotic surgery, MBSAQIP, Sleeve gastrectomy, Roux-en-Y gastric bypass
Metabolic and bariatric surgery (MBS) has been established as the most effective treatment for obesity [1, 2]. MBS is gaining popularity as a result, with sleeve gastrectomy (SG) and Roux-en-Y gastric bypass (RYGB) being the most common procedures [3]. However, primary MBS is often complicated by weight regain, postoperative reflux, metabolic disorders, and other anatomic complications that can require a revision or conversion operation. For example, an estimated 10–20% of patients experience weight regain after primary MBS or fail to achieve significant weight loss altogether [4–6]. Revisional cases are the third most common type of MBS in the United States (US), with over 30,000 revisions performed each year on average since 2016 [3].
Revisional MBS can be technically challenging due to the presence of altered anatomy in reoperative fields with adhesions [7, 8]. Consequently, revisional MBS is generally associated with higher morbidity and longer operative times compared to primary MBS [5, 9–11]. The rise of robotic surgical platforms, with their improved dexterity and visualization, has led many to apply this new technology to revisional MBS in an effort to overcome the shortcomings of conventional laparoscopy [12–14]. The optimal surgical approach for these complex cases, however, remains unclear. Several case series, national database studies, and meta-analyses comparing laparoscopic and robotic revisional MBS have been performed with varying results [15–22].
The Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program Participant Use Data File (MBSAQIP PUF) is the largest, bariatric-specific, clinical dataset in the US. Prior studies of the MBSAQIP PUF comparing laparoscopic and robotic revisional MBS have been performed using data through 2020 [20]. The MBSAQIP has since been updated through 2022 with the addition of multiple new variables. To our knowledge, there have been no studies comparing the outcomes of laparoscopic and robotic revisional MBS with these new data. The goal of this study was to perform the largest retrospective analysis of the MBSAQIP database comparing perioperative outcomes between laparoscopic and robotic-assisted revisional SG and RYGB cases. We also sought to assess robot utilization in minimally invasive revisional MBS over the 8-year period of available MBSAQIP data.
Methods
Study design and data source
A retrospective analysis of the 2015–2022 MBSAQIP PUF was performed identifying revisional SG and RYGB cases performed laparoscopically and robotically. The MBSAQIP is a database created by the American College of Surgeons (ACS) and American Society of Metabolic and Bariatric Surgery (ASMBS). All nationally accredited metabolic and bariatric surgery centers in the United States report outcomes to the MBSAQIP. All outcomes recorded in the dataset are 30-day outcomes. Data from each center are collected by trained clinical reviewers and audited. The MBSAQIP is a deidentified database, and this study was therefore exempt from institutional review board approval. The ACS, MBSAQIP, and the centers participating in the MBSAQIP are the source of the data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors.
Case selection
The case selection algorithm for this study is shown in Fig. 1. Revision and conversion MBS cases were first selected by excluding all other case types. Cases involving concurrent procedures other than esophagogastroduodenoscopy (EGD) were then removed. All surgical approaches other than laparoscopic and robotic were excluded. Cases converted to another surgical approach and emergent cases were also removed. SG and RYGB cases were then selected by current procedural terminology (CPT) codes 43775, 43644, and 43645. Standard laparoscopic and robotic-assisted cases were separated. Finally, cases with incomplete 30-day follow-up and missing data were excluded.
Fig. 1.
Case selection algorithm identifying revisional laparoscopic (L-) and robotic (R-) sleeve gastrectomy (SG) and Roux-en-Y gastric bypass (RYGB) cases from the MBSAQIP database. MBSAQIP PUF Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program Participant Use Data File, EGD esophagogastroduodenoscopy
Patient demographics and operative data
Patient demographic data included age, sex, American Society of Anesthesiologists (ASA) classification, and body mass index (BMI) closest to surgery. Comorbidities included hypertension, hyperlipidemia, dependent functional health status, current smoker within one year, diabetes mellitus, chronic steroid/immunosuppression use, chronic obstructive pulmonary disease (COPD), pulmonary embolism (PE), deep vein thrombosis (DVT) requiring therapy, preoperative therapeutic anticoagulation, inferior vena cava (IVC) filter, obstructive sleep apnea (OSA), gastroesophageal reflux disease (GERD), history of myocardial infarction (MI), previous cardiac surgery, renal insufficiency, and dialysis. Operative variables included leak test, drain placed, and operative time.
Outcomes
30-Day outcomes included overall morbidity, anastomotic leak, bleeding, blood transfusion within 72 h of surgery, readmission, reoperation, reintervention, mortality, superficial incisional surgical site infection (SSI), deep incisional SSI, organ/space SSI, wound disruption, sepsis, septic shock, urinary tract infection (UTI), ventilator > 48 h, unplanned intubation, postoperative pneumonia, venous thrombosis requiring therapy, PE, stroke, unplanned admission to the intensive care unit (ICU), acute renal failure requiring dialysis, progressive renal insufficiency, cardiac arrest requiring cardiopulmonary resuscitation (CPR), myocardial infarction, and hospital length of stay. Robotic utilization as a percentage of total minimally invasive revisional MBS over the 8-year period was also analyzed.
From 2015 to 2019, there are no dedicated variables for anastomotic leak or bleeding. Beginning in 2020, the variables “Anastomotic/Staple Line Leak” (defined as a leak of endoluminal contents through an anastomosis or staple line from the MBS procedure) and “Gastrointestinal Tract Bleeding” (defined as bleeding of any portion of the gastrointestinal tract, which may be at a staple line, anastomosis, enterotomy, or ulcer) were included. We defined anastomotic leak for 2015–2019 as an aggregate of reoperation, readmission, and reintervention for suspected leak, drain present at 30 days, and organ/space SSI. For 2020–2022, leak was defined as an aggregate of reoperation, readmission, and reintervention for leak and the variable “Anastomotic/Staple Line Leak.” Bleeding for 2015–2019 was defined as an aggregate of reoperation, readmission, and reintervention for suspected bleeding. For 2020–2022, an aggregate of reoperation, readmission, and reintervention for gastrointestinal tract bleeding and the variable “Gastrointestinal Tract Bleeding” was used. This aggregate complication methodology for 2015–2019 data was previously described by Berger et al. [23].
Overall morbidity was defined as an aggregate of anastomotic leak, bleeding, blood transfusion within 72 h of surgery, superficial incisional SSI, deep incisional SSI, organ/space SSI, wound disruption, sepsis, septic shock, UTI, ventilator > 48 h, unplanned intubation, pneumonia, venous thrombosis requiring therapy, PE, stroke, unplanned admission to ICU, acute renal failure requiring dialysis, progressive renal insufficiency, cardiac arrest requiring CPR, and myocardial infarction.
Statistical analysis
Statistical analysis was performed in R (Version 4.4.1, Vienna, Austria). For univariate analyses, Pearson’s Chi-square test or Fisher’s exact test was used for categorical variables as appropriate. Independent two-sample t test was used for continuous variables. Categorical variables were reported as frequency and percentage, and continuous variables as mean ± standard deviation (SD). A p value of < 0.05 was considered statistically significant.
Multivariate logistic regression was used to analyze 30-day outcomes and adjust for demographic characteristics, comorbidities, and operative variables. Only variables with p < 0.05 in the univariate analysis were included in the adjustment. An adjusted odds ratio (AOR) and 95% confidence interval (CI) were reported for each outcome with laparoscopic cases used as the reference group.
Results
Patient demographics and operative data
A total of 41,404 cases (14,474 SG; 26,930 RYGB) were included in the analysis. Of the 14,474 revisional SG cases, 12,680 were laparoscopic (87.6%, L-SG) and 1,794 were robotic (12.4%, R-SG). Patients who underwent revisional L-SG were significantly younger (48.5 ± 10.8 vs. 49.1 ± 10.8 years, p < 0.05) than patients who underwent R-SG. There were significant differences in ASA class between groups as detailed in Table 1. Revisional L-SG patients also had a lower prevalence of hypertension, diabetes mellitus, OSA, and GERD compared to R-SG patients. For operative variables, L-SG patients were more likely to have a leak test performed (75.6% vs. 68.3%, p < 0.001) and a drain placed (16.5% vs. 11.0%, p < 0.001). Operative time for revisional L-SG was significantly shorter than R-SG (86.4 ± 45.8 vs. 113.5 ± 51.7 min, p < 0.001).
Table 1.
Patient demographics and operative variables for laparoscopic and robotic revisional sleeve gastrectomy
| Demographics | L-SG (N = 12,680) | R-SG (N = 1794) | p value |
|---|---|---|---|
| Age, mean ± SD (years) | 48.5 ± 10.8 | 49.1 ± 10.8 | < 0.05 |
| Female, n (%) | 10,633 (83.9) | 1497 (83.4) | 0.68 |
| ASA, n (%) | < 0.05 | ||
| 1–2 | 2997 (23.6) | 380 (21.2) | |
| 3 | 9356 (73.8) | 1352 (75.4) | |
| 4–5 | 327 (2.6) | 62 (3.5) | |
| BMI, mean ± SD (kg/m2) | 43.6 ± 7.9 | 43.7 ± 8.0 | 0.57 |
| Comorbidities, n (%) | |||
| Hypertension | 5740 (45.3) | 885 (49.3) | < 0.01 |
| Hyperlipidemia | 2818 (22.2) | 435 (24.3) | 0.06 |
| Dependent functional status | 163 (1.3) | 14 (0.8) | 0.09 |
| Smoking | 766 (6.0) | 95 (5.3) | 0.23 |
| Diabetes mellitus | 2182 (17.2) | 348 (19.4) | < 0.05 |
| Chronic steroid/immunosuppression | 291 (2.3) | 54 (3.0) | 0.08 |
| Chronic obstructive pulmonary disease | 130 (1.0) | 22 (1.2) | 0.51 |
| Pulmonary embolism | 226 (1.8) | 39 (2.2) | 0.29 |
| Deep vein thrombosis requiring therapy | 258 (2.0) | 35 (2.0) | 0.88 |
| Therapeutic anticoagulation | 424 (3.3) | 61 (3.4) | 0.96 |
| Inferior vena cava filter | 69 (0.5) | 8 (0.5) | 0.72 |
| Obstructive sleep apnea | 3803 (30.0) | 607 (33.8) | < 0.01 |
| Gastroesophageal reflux disease | 4084 (32.2) | 626 (34.9) | < 0.05 |
| History of myocardial infarction | 159 (1.3) | 19 (1.1) | 0.56 |
| Previous cardiac surgery | 161 (1.3) | 20 (1.1) | 0.66 |
| Renal insufficiency | 61 (0.5) | 8 (0.5) | 0.98 |
| Dialysis | 20 (0.2) | 6 (0.3) | 0.17 |
| Operative variables | |||
| Leak test, n (%) | 9589 (75.6) | 1226 (68.3) | < 0.001 |
| Drain placed, n (%) | 2086 (16.5) | 197 (11.0) | < 0.001 |
| Operative time, mean ± SD (minutes) | 86.4 ± 45.8 | 113.5 ± 51.7 | < 0.001 |
L-SG Laparoscopic sleeve gastrectomy, R-SG robotic sleeve gastrectomy, SD standard deviation, ASA American Society of Anesthesiologists, BMI body mass index
Bold values indicate statistical significance
Of the 26,930 revisional RYGB cases, 21,755 were laparoscopic (80.8%, L-RYGB) and 5,175 were robotic (19.2%, R-RYGB). Patients who underwent L-RYGB were less likely to be female (88.0% vs. 90.2%, p < 0.001) compared to R-RYGB. There were significant differences in ASA class between groups as detailed in Table 2. L-RYGB patients were less likely to have preoperative GERD (58.0% vs. 64.5%, p < 0.001) and more likely to have a drain placed during surgery (26.6% vs. 12.8%, p < 0.001). Operative time for revisional L-RYGB was significantly shorter than R-RYGB (130.7 ± 64.7 vs. 165.5 ± 66.8 min, p < 0.001).
Table 2.
Patient demographics and operative variables for laparoscopic and robotic revisional Roux-en-Y gastric bypass
| Demographics | L-RYGB (N = 21,755) | R-RYGB (N = 5175) | p value |
|---|---|---|---|
| Age, mean ± SD (years) | 46.7 ± 10.6 | 46.7 ± 10.7 | 0.97 |
| Female, n (%) | 19,148 (88.0) | 4669 (90.2) | < 0.001 |
| ASA, n (%) | < 0.05 | ||
| 1–2 | 5603 (25.8) | 1245 (24.1) | |
| 3 | 15,602 (71.7) | 3790 (73.2) | |
| 4–5 | 550 (2.5) | 140 (2.7) | |
| BMI, mean ± SD (kg/m2) | 41.5 ± 8.3 | 41.4 ± 8.2 | 0.92 |
| Comorbidities, n (%) | |||
| Hypertension | 8447 (38.8) | 2043 (39.5) | 0.40 |
| Hyperlipidemia | 4236 (19.5) | 1032 (19.9) | 0.45 |
| Dependent functional status | 158 (0.7) | 45 (0.9) | 0.33 |
| Smoking | 1097 (5.0) | 227 (4.4) | 0.05 |
| Diabetes mellitus | 3383 (15.6) | 822 (15.9) | 0.57 |
| Chronic steroid/immunosuppression | 504 (2.3) | 140 (2.7) | 0.11 |
| Chronic obstructive pulmonary disease | 263 (1.2) | 76 (1.5) | 0.15 |
| Pulmonary embolism | 385 (1.8) | 89 (1.7) | 0.85 |
| Deep vein thrombosis requiring therapy | 503 (2.3) | 107 (2.1) | 0.31 |
| Therapeutic anticoagulation | 652 (3.0) | 175 (3.4) | 0.16 |
| Inferior vena cava filter | 126 (0.6) | 19 (0.4) | 0.08 |
| Obstructive sleep apnea | 6010 (27.6) | 1409 (27.2) | 0.58 |
| Gastroesophageal reflux disease | 12,611 (58.0) | 3336 (64.5) | < 0.001 |
| History of myocardial infarction | 230 (1.1) | 53 (1.0) | 0.89 |
| Previous cardiac surgery | 197 (0.9) | 51 (1.0) | 0.65 |
| Renal insufficiency | 77 (0.4) | 27 (0.5) | 0.10 |
| Dialysis | 34 (0.2) | 7 (0.1) | 0.88 |
| Operative variables | |||
| Leak test, n (%) | 19,598 (90.1) | 4693 (90.7) | 0.20 |
| Drain placed, n (%) | 5776 (26.6) | 660 (12.8) | < 0.001 |
| Operative time, mean ± SD (minutes) | 130.7 ± 64.7 | 165.5 ± 66.8 | < 0.001 |
L-RYGB Laparoscopic Roux-en-Y gastric bypass, R-RYGB robotic Roux-en-Y gastric bypass, SD standard deviation, ASA American Society of Anesthesiologists, BMI body mass index
Bold values indicate statistical significance
Robot utilization
The total number of revisional R-SG cases increased by 5.3 times from 2015 (86 cases) to 2022 (453 cases). In contrast, the total number of revisional L-SG cases increased by 1.08 times from 2015 (1317 cases) to 2022 (1420 cases). The total number of revisional R-RYGB cases increased substantially by 22.4 times from 2015 (89 cases) to 2022 (1995 cases). The number of revisional L-RYGB cases increased by 3.7 times from 2015 (1130) to 2022 (4236). Taken together, the percentage of all revisional SG and RYGB performed robotically increased from 6.1% and 7.3% to 24.2% and 32.0%, respectively, over the 8-year period (Fig. 2).
Fig. 2.
Percentage of revisional sleeve gastrectomy (SG) and Roux-en-Y gastric bypass (RYGB) cases performed laparoscopically (L-) and robotically (R-) from 2015 to 2022
Outcomes
Outcomes for revisional SG are detailed in Table 3. Compared to R-SG, L-SG had a significantly lower risk of postoperative pulmonary embolism (0.08% vs. 0.28%, p = 0.03, AOR 3.72 [1.25–10.99]). There were otherwise no significant differences between L-SG and R-SG in terms of overall morbidity, leak, bleeding, readmission, reoperation, reintervention, mortality, length of stay, or any other postoperative complications.
Table 3.
30-Day outcomes for laparoscopic and robotic revisional sleeve gastrectomy
| Outcome, n (%) | L-SG (N = 12,680) | R-SG (N = 1,794) | p value | AOR (95% CI) |
|---|---|---|---|---|
| Overall morbidity | 408 (3.2) | 69 (3.8) | 0.19 | 1.23 (0.95–1.60) |
|
Aggregate leak Reoperation for leak Intervention for leak Readmission for leak Organ/space SSI Drain present at 30 days* Anastomotic/staple line leak** |
131 (1.0) 45 (0.35) 40 (0.32) 61 (0.48) 88 (0.69) 34 (0.41) 26 (0.59) |
15 (0.8) 4 (0.22) 3 (0.17) 6 (0.33) 16 (0.89) 5 (0.72) 4 (0.36) |
0.51 0.49 0.40 0.50 0.44 0.23 0.49 |
0.91 (0.53–1.57) 0.67 (0.24–1.87) 0.57 (0.18–1.84) 0.72 (0.31–1.67) 1.35 (0.79–2.32) – – |
|
Aggregate bleeding Reoperation for bleeding Intervention for bleeding Readmission for bleeding GI tract bleeding** |
45 (0.35) 29 (0.23) 5 (0.04) 8 (0.06) 18 (0.41) |
2 (0.11) 1 (0.06) 0 0 1 (0.09) |
0.12 0.17 0.87 0.61 0.15 |
0.30 (0.07–1.24) 0.24 (0.03–1.78) – – – |
| Blood transfusion | 62 (0.49) | 9 (0.50) | 0.99 | 1.05 (0.52–2.12) |
| Readmission | 451 (3.6) | 57 (3.2) | 0.45 | 0.89 (0.67–1.17) |
| Reoperation | 189 (1.5) | 25 (1.4) | 0.83 | 0.95 (0.62–1.44) |
| Reintervention | 154 (1.2) | 22 (1.2) | 0.99 | 1.05 (0.67–1.65) |
| Mortality | 10 (0.08) | 0 | 0.62 | – |
| Superficial incisional SSI | 51 (0.4) | 11 (0.6) | 0.28 | 1.54 (0.80–2.98) |
| Deep incisional SSI | 9 (0.07) | 3 (0.17) | 0.18 | 2.33 (0.63–8.66) |
| Wound disruption | 5 (0.04) | 0 | 0.87 | – |
| Sepsis | 27 (0.21) | 7 (0.39) | 0.18 | 1.77 (0.77–4.07) |
| Septic shock | 9 (0.07) | 2 (0.11) | 0.64 | 1.50 (0.32–7.01) |
| Urinary tract infection | 36 (0.3) | 9 (0.5) | 0.19 | 1.86 (0.89–3.88) |
| Ventilator > 48 h | 8 (0.06) | 3 (0.17) | 0.15 | 2.52 (0.66–9.61) |
| Unplanned intubation | 10 (0.08) | 4 (0.22) | 0.09 | 2.58 (0.80–8.28) |
| Pneumonia | 26 (0.21) | 7 (0.39) | 0.18 | 1.80 (0.78–4.18) |
| Venous thrombosis requiring therapy | 30 (0.24) | 9 (0.50) | 0.05 | 2.20 (1.04–4.67) |
| Pulmonary embolism | 10 (0.08) | 5 (0.28) | 0.03 | 3.72 (1.25–10.99) |
| Stroke | 1 (0.01) | 1 (0.06) | 0.23 | 7.55 (0.43–134.26) |
| Unplanned admission to ICU | 78 (0.6) | 10 (0.6) | 0.89 | 0.89 (0.46–1.73) |
| Acute renal failure requiring dialysis | 8 (0.06) | 0 | 0.61 | – |
| Progressive renal insufficiency | 11 (0.09) | 1 (0.06) | 0.99 | 0.57 (0.07–4.47) |
| Cardiac arrest requiring CPR | 7 (0.06) | 0 | 0.99 | – |
| Myocardial infarction | 2 (0.02) | 0 | 0.99 | – |
| Length of stay, mean ± SD (days) | 1.49 (1.41) | 1.49 (1.98) | 0.96 | – |
*variable only present in 2015–2019 datasets; **variable only present in 2020–2022 datasets
L-SG Laparoscopic sleeve gastrectomy, R-SG robotic sleeve gastrectomy, AOR adjusted odds ratio, CI confidence interval, SSI surgical site infection, GI gastrointestinal, ICU intensive care unit, CPR cardiopulmonary resuscitation
Bold values indicate statistical significance
Outcomes for revisional RYGB are detailed in Table 4. Compared to R-RYGB, L-RYGB had significantly higher rates of overall morbidity (6.2% vs. 4.8%, p < 0.001, AOR 0.80 [0.70–0.93]), blood transfusion (1.5% vs. 1.0%, p < 0.05, AOR 0.74 [0.55–0.99]), superficial incisional SSI (1.2% vs. 0.4%, p < 0.001, AOR 0.30 [0.19–0.47]), and longer length of stay (1.87 vs. 1.76 days, p < 0.001). Conversely, L-RYGB had lower rates of pulmonary complications, including ventilator > 48 h (0.15% vs. 0.29%, p < 0.05, AOR 1.98 [1.06–3.70]) and unplanned intubation (0.19% vs. 0.35%, p < 0.05, AOR 2.08 [1.18–3.67]). There were otherwise no significant differences between L-RYGB and R-RYGB in terms of leak, bleeding, readmission, reoperation, reintervention, mortality, or any other postoperative complications.
Table 4.
30-Day outcomes for laparoscopic and robotic revisional Roux-en-Y gastric bypass
| Outcome, n (%) | L-RYGB (N = 21,755) | R-RYGB (N = 5175) | p value | AOR (95% CI) |
|---|---|---|---|---|
| Overall morbidity | 1341 (6.2) | 246 (4.8) | < 0.001 | 0.80 (0.70–0.93) |
|
Aggregate leak Reoperation for leak Intervention for leak Readmission for leak Organ/space SSI Drain present at 30 days* Anastomotic/staple line leak** |
262 (1.2) 101 (0.46) 52 (0.24) 71 (0.33) 193 (0.89) |
54 (1.0) 25 (0.48) 12 (0.23) 21 (0.41) 49 (0.95) |
0.37 0.95 0.99 0.45 0.74 |
1.00 (0.74–1.35) 1.12 (0.72–1.75) 1.11 (0.59–2.10) 1.38 (0.84–2.26) 1.11 (0.81–1.53) – – |
|
Aggregate bleeding Reoperation for bleeding Intervention for bleeding Readmission for bleeding GI tract bleeding** |
207 (0.95) 76 (0.35) 48 (0.22) 73 (0.34) |
51 (0.99) 10 (0.19) 7 (0.14) 23 (0.44) |
0.88 0.10 0.29 0.29 |
1.09 (0.80–1.49) 0.61 (0.31–1.19) 0.64 (0.29–1.42) 1.37 (0.85–2.20) |
| Blood transfusion | 329 (1.5) | 54 (1.0) | < 0.05 | 0.74 (0.55–0.99) |
| Readmission | 1495 (6.9) | 386 (7.5) | 0.14 | 1.10 (0.98–1.24) |
| Reoperation | 685 (3.1) | 166 (3.2) | 0.86 | 1.07 (0.90–1.27) |
| Reintervention | 532 (2.4) | 112 (2.2) | 0.25 | 0.91 (0.74–1.12) |
| Mortality | 23 (0.11) | 9 (0.17) | 0.29 | 1.87 (0.85–4.13) |
| Superficial incisional SSI | 270 (1.2) | 19 (0.4) | < 0.001 | 0.30 (0.19–0.47) |
| Deep incisional SSI | 58 (0.27) | 7 (0.14) | 0.12 | 0.54 (0.25–1.20) |
| Wound disruption | 19 (0.09) | 3 (0.06) | 0.79 | 0.68 (0.19–2.32) |
| Sepsis | 70 (0.32) | 16 (0.31) | 0.99 | 1.01 (0.58–1.75) |
| Septic shock | 35 (0.16) | 15 (0.29) | 0.08 | 1.93 (0.98–3.58) |
| Urinary tract infection | 115 (0.53) | 19 (0.37) | 0.17 | 0.66 (0.41–1.08) |
| Ventilator > 48 h | 32 (0.15) | 15 (0.29) | < 0.05 | 1.98 (1.06–3.70) |
| Unplanned intubation | 41 (0.19) | 18 (0.35) | < 0.05 | 2.08 (1.18–3.67) |
| Pneumonia | 102 (0.47) | 27 (0.52) | 0.70 | 1.12 (0.73–1.72) |
| Venous thrombosis requiring therapy | 38 (0.17) | 13 (0.25) | 0.33 | 1.49 (0.79–2.82) |
| Pulmonary embolism | 45 (0.21) | 10 (0.19) | 0.98 | 1.05 (0.53–2.11) |
| Stroke | 1 (0.005) | 1 (0.02) | 0.35 | 5.96 (0.33–108.97) |
| Unplanned admission to ICU | 248 (1.1) | 54 (1.0) | 0.60 | 0.97 (0.72–1.30) |
| Acute renal failure | 13 (0.06) | 3 (0.06) | 0.99 | 1.16 (0.32–4.12) |
| Progressive renal insufficiency | 12 (0.06) | 4 (0.08) | 0.53 | 1.54 (0.48–4.87) |
| Cardiac arrest requiring CPR | 15 (0.07) | 5 (0.10) | 0.57 | 1.56 (0.56–4.37) |
| Myocardial infarction | 5 (0.02) | 2 (0.04) | 0.63 | 1.35 (0.26–7.03) |
| Length of stay, mean ± SD (days) | 1.87 (2.09) | 1.76 (2.17) | < 0.001 | – |
*variable only present in 2015–2019 datasets; **variable only present in 2020–2022 datasets
L-RYGB Laparoscopic Roux-en-Y gastric bypass, R-RYGB robotic Roux-en-Y gastric bypass, AOR adjusted odds ratio, CI confidence interval, SSI surgical site infection, GI gastrointestinal, ICU intensive care unit, CPR cardiopulmonary resuscitation
Bold values indicate statistical significance
Discussion
This study represents the largest retrospective comparison of laparoscopic and robotic revisional SG and RYGB outcomes in the literature, and to our knowledge, the first study reporting these outcomes from the 2022 iteration of the MBSAQIP database. As robot utilization increases across surgical disciplines, it is important to understand the role of robotics in MBS, particularly in revisional cases, as the technical advantages of robotic platforms may be most apparent in these challenging operations. In addition, complication rates for reoperative MBS are generally higher than primary MBS, highlighting the need for technical strategies to optimize patient outcomes [5, 11, 24].
Our study demonstrated a single statistically significant difference in 30-day outcomes between laparoscopic and robotic revisional SG cases, with R-SG having a higher rate of postoperative pulmonary embolism, though absolute rates in both approaches were low. This may be related to longer operative times or the higher rate of comorbidities in the R-SG group. Outcomes between L-SG and R-SG were otherwise equivalent. In contrast, there were several significant differences between laparoscopic and robotic revisional RYGB, with R-RYGB having lower overall morbidity, blood transfusion, superficial incisional SSI, and shorter length of stay. The shorter length of stay for R-RYGB may be a consequence of the lower overall morbidity, though additional factors not captured in this dataset such as postoperative pain may be contributors. R-RYGB did however have a higher risk of ventilator > 48 h and unplanned intubation, which may also be related to the longer operative time and anesthesia duration with robotic cases, though absolute rates of these complications were also low [25]. It may be argued that these respiratory complications present a more significant risk of morbidity to patients than SSI and blood transfusion, and attention to these respiratory complications is warranted in future studies. Overall, despite significantly longer operative times for robotic SG and RYGB, we found robotic revisional MBS to have an acceptable safety profile compared to laparoscopic. In addition, robotic complications for revisional RYGB were overall lower when compared to laparoscopic. The robotic platform may be advantageous in these more complex RYGB cases, as they often require careful and precise dissection of altered anatomy in hostile operative fields. It follows that the technical and ergonomic advantages of robotic platforms are uniquely suited to these procedures.
Several studies comparing laparoscopic and robotic revisional MBS have been performed with similarities and differences to our results. Acevedo et al. performed 1:1 case control-matching of revisional SG and RYGB cases using 2015–2016 MSBAQIP data and similarly found a higher transfusion rate with L-RYGB, while R-SG had a higher rate of postoperative sepsis [19]. Outcomes were otherwise similar between groups in their matched analysis. Nasser et al. analyzed revisional SG and RYGB cases using 2015–2017 MBSAQIP data with multivariate logistic regression and reported R-SG had higher rates of overall morbidity, reintervention, reoperation, ventilator > 48 h, organ space SSI, and sepsis [18]. They also found R-RYGB had a lower rate of pulmonary complications, any SSI, and blood transfusion. In contrast to these reports, we found L-SG and R-SG outcomes to be relatively similar. Smaller cases series, meta-analyses, and propensity-score matched MBSAQIP analyses have also shown no difference in outcomes between laparoscopic and robotic revisional MBS [17, 20–22, 26]. The reasons for the lack of a consistent body of evidence are likely multifactorial. Aggregate progression of surgeons across the robotic MBS learning curve in the newer MBSAQIP data may be a contributor. It is likely that many centers across the country have not yet completed the learning curve for robotic MBS. There may also be shifts in surgeon technique over time not captured in the MBSAQIP that could account for the improved R-SG outcomes in our study. The addition of more granular surgical technique data to the MBSAQIP would be beneficial for future retrospective outcome comparisons. Methodological differences in study design and outcomes reporting are important considerations as well [22]. Due to these factors, analyzing outcomes in robotic surgery is complicated, and randomized controlled trials with long-term follow-up are needed to compare operative approaches in revisional MBS. Nevertheless, the potential for improved outcomes with robotic surgery is promising.
Increasing utilization of robotics in MBS has been demonstrated in multiple studies [13, 27, 28]. A recent study analyzing over 1 million cases from 2015 to 2020 MBSAQIP data found that robotic RYGB cases increased from 6.8% to 16.7%, while robotic SG increased from 6.0% to 17.2% [13]. We found a continued increase in robotic use for revisional MBS through 2022. The widespread adoption of robotics in MBS, and surgery more broadly, remains controversial due to cost concerns with the robotic platform [29, 30]. However, there is evidence that the cost of robotic MBS is decreasing, and single-institution studies have reported no cost difference between the approaches [31–33]. It is well documented that operative times for robotic MBS cases are generally longer than conventional laparoscopy, resulting in longer anesthesia duration for patients [19, 26]. The difference in operative time between laparoscopic and robotic approaches may diminish as surgeons progress across the learning curve. Buchs et al. reported that after overcoming the learning curve for R-RYGB, there was no longer a difference in operative time between laparoscopic and robotic cases [16]. While we identified a clinical benefit of robotics for revisional RYGB despite longer operative times, it remains to be seen how the net cost of robotic platforms evolves with increased utilization nationally.
Determining the optimal surgical approach for revisional MBS will also be critical in the broader context of obesity management. With the recognition of obesity as a chronic, multifactorial disease and the rise of MBS worldwide, the need for revisional surgery is likely to increase. The increasing use of robotic platforms in these complex cases and associated outcomes should be monitored to help patients achieve optimal outcomes and make informed decisions. It is also important to recognize that surgery represents one treatment option in the multidisciplinary approach to obesity management. Pharmacologic therapy in the form of glucagon-like peptide 1 (GLP-1) receptor agonists has emerged as a promising treatment for weight regain [34]. There are also several emerging incretin-based therapies that are likely to influence the obesity management landscape [35]. Counseling patients on the risks and benefits of revisional surgery as part of shared decision-making discussions may be directly influenced by advances in surgical technique and potential patient safety benefits afforded by robotic platforms. The efficacy of revisional surgery will also need to be weighed against the risk of morbidity as medical therapies continue to advance.
There are several limitations to this retrospective study of a large national database. Many important and potentially confounding variables are not available in the MBSAQIP, such as years of surgeon experience with robotic MBS, hospital robotic case volume, anastomosis techniques (handsewn, stapled, or combination), and whether cases were totally or partially robotic. Before 2019, a robotic stapler was not widely available, and consequently, all stapling was performed laparoscopically by a bedside assistant. It is unclear how surgeon practices have changed with the introduction of robotic staplers and how these changes may have impacted patient outcomes. For 2015–2019 data, it is not possible to know what the index operation was that required a revision/conversion, which may directly influence case complexity, complications, and operative time. Technical details regarding the type of revisional surgery performed are also missing from current data. Clinical criteria for the choice of operative approach are not available, which may introduce selection bias in a surgeon’s preference to perform cases robotically. The MBSAQIP also only provides 30-day outcomes data, limiting the ability to assess long-term outcomes.
Conclusion
Robotic and laparoscopic revisional MBS are both safe and effective surgical approaches. Robot utilization in revisional bariatric surgery is increasing annually. Robotic surgery has lower postoperative morbidity and shorter length of stay in revisional RYGB when compared to laparoscopic. Robotic platforms may have the capacity to improve the delivery of care for patients undergoing revisional bariatric surgery.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Declarations
Disclosure
Dr. Broderick is a consultant for Stryker Corporation. Dr. Sandler is a consultant for Boston Scientific. Dr. Horgan is a consultant for Stryker Corporation, Fortimedix Surgical, and Alume Biosciences. Dr. Jacobsen is a consultant for Gore Medical and Viacyte. Drs. Spurzem, Kunkel, and Hollandsworth have no disclosures.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Eisenberg D, Shikora SA, Aarts E, Aminian A, Angrisani L, Cohen RV et al (2022) 2022 American society for metabolic and bariatric surgery (ASMBS) and international federation for the surgery of obesity and metabolic disorders (IFSO): indications for metabolic and bariatric surgery. Surg Obes Relat Dis 18(12):1345–1356. 10.1016/j.soard.2022.08.013 [DOI] [PubMed] [Google Scholar]
- 2.Gloy VL, Briel M, Bhatt DL, Kashyap SR, Schauer PR, Mingrone G, Bucher HC, Nordmann AJ (2013) Bariatric surgery versus non-surgical treatment for obesity: a systematic review and meta-analysis of randomised controlled trials. BMJ 347:f5934. 10.1136/bmj.f5934 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Estimate of Bariatric Surgery Numbers, 2011–2022. American society for metabolic and bariatric surgery. Published June 27, 2022. Accessed 24 April 2024. https://asmbs.org/resources/estimate-of-bariatric-surgery-numbers/
- 4.Pinto-Bastos A, Conceição EM, Machado PPP (2017) Reoperative bariatric surgery: a systematic review of the reasons for surgery, medical and weight loss outcomes, relevant behavioral factors. Obes Surg 27(10):2707–2715. 10.1007/s11695-017-2855-7 [DOI] [PubMed] [Google Scholar]
- 5.Brethauer SA, Kothari S, Sudan R, Williams B, English WJ, Brengman M, Kurian M, Hutter M, Stegemann L, Kallies K, Nguyen NT, Ponce J, Morton JM (2014) Systematic review on reoperative bariatric surgery: american society for metabolic and bariatric surgery revision task force. Surg Obes Relat Dis 10(5):952–972. 10.1016/j.soard.2014.02.014 [DOI] [PubMed] [Google Scholar]
- 6.Karmali S, Brar B, Shi X, Sharma AM, de Gara C, Birch DW (2013) Weight recidivism post-bariatric surgery: a systematic review. Obes Surg 23(11):1922–1933. 10.1007/s11695-013-1070-4 [DOI] [PubMed] [Google Scholar]
- 7.Cheung D, Switzer NJ, Gill RS, Shi X, Karmali S (2014) Revisional bariatric surgery following failed primary laparoscopic sleeve gastrectomy: a systematic review. Obes Surg 24(10):1757–1763. 10.1007/s11695-014-1332-9 [DOI] [PubMed] [Google Scholar]
- 8.Spyropoulos C, Kehagias I, Panagiotopoulos S, Mead N, Kalfarentzos F (2010) Revisional bariatric surgery: 13-year experience from a tertiary institution. Arch Surg 145(2):173–177. 10.1001/archsurg.2009.260 [DOI] [PubMed] [Google Scholar]
- 9.Mor A, Keenan E, Portenier D, Torquati A (2013) Case-matched analysis comparing outcomes of revisional versus primary laparoscopic Roux-en-Y gastric bypass. Surg Endosc 27(2):548–552. 10.1007/s00464-012-2477-z [DOI] [PubMed] [Google Scholar]
- 10.Inabnet WB, Belle SH, Bessler M, Courcoulas A, Dellinger P, Garcia L, Mitchell J, Oelschlager B, O’Rourke R, Pender J, Pomp A, Pories W, Ramanathan R, Wahed A, Wolfe B (2010) Comparison of 30-day outcomes after non-LapBand primary and revisional bariatric surgical procedures from the longitudinal assessment of bariatric surgery study. Surg Obes Relat Dis 6(1):22–30. 10.1016/j.soard.2009.10.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Shimizu H, Annaberdyev S, Motamarry I, Kroh M, Schauer PR, Brethauer SA (2013) Revisional bariatric surgery for unsuccessful weight loss and complications. Obes Surg 23(11):1766–1773. 10.1007/s11695-013-1012-1 [DOI] [PubMed] [Google Scholar]
- 12.Sheetz KH, Claflin J, Dimick JB (2020) Trends in the adoption of robotic surgery for common surgical procedures. JAMA Netw Open 3(1):e1918911. 10.1001/jamanetworkopen.2019.18911 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bauerle WB, Mody P, Estep A, Stoltzfus J, El Chaar M (2023) Current trends in the utilization of a robotic approach in the field of bariatric surgery. Obes Surg 33(2):482–491. 10.1007/s11695-022-06378-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Snyder B, Wilson T, Woodruff V, Wilson E (2013) Robotically assisted revision of bariatric surgeries is safe and effective to achieve further weight loss. World J Surg 37(11):1. 10.1007/s00268-013-1968-y [DOI] [PubMed] [Google Scholar]
- 15.Gray KD, Moore MD, Elmously A, Bellorin O, Zarnegar R, Dakin G, Pomp A, Afaneh C (2018) Perioperative outcomes of laparoscopic and robotic revisional bariatric surgery in a complex patient population. Obes Surg 28(7):1852–1859. 10.1007/s11695-018-3119-x [DOI] [PubMed] [Google Scholar]
- 16.Buchs NC, Morel P, Azagury DE, Jung M, Chassot G, Huber O, Hagen ME, Pugin F (2014) Laparoscopic versus robotic Roux-En-Y gastric bypass: lessons and long-term follow-up learned from a large prospective monocentric study. Obes Surg 24(12):2031–2039. 10.1007/s11695-014-1335-6 [DOI] [PubMed] [Google Scholar]
- 17.Buchs NC, Pugin F, Azagury DE, Huber O, Chassot G, Morel P (2014) Robotic revisional bariatric surgery: a comparative study with laparoscopic and open surgery. Int J Med Robot Comp 10(2):213–217. 10.1002/rcs.1549 [DOI] [PubMed] [Google Scholar]
- 18.Nasser H, Munie S, Kindel TL, Gould JC, Higgins RM (2020) Comparative analysis of robotic versus laparoscopic revisional bariatric surgery: perioperative outcomes from the MBSAQIP database. Surg Obes Relat Dis 16(3):397–405. 10.1016/j.soard.2019.11.018 [DOI] [PubMed] [Google Scholar]
- 19.Acevedo E, Mazzei M, Zhao H, Lu X, Edwards MA (2020) Outcomes in conventional laparoscopic versus robotic-assisted revisional bariatric surgery: a retrospective, case–controlled study of the MBSAQIP database. Surg Endosc 34(4):1573–1584. 10.1007/s00464-019-06917-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Seton T, Mahan M, Dove J, Villaneuva H, Obradovic V, Falvo A, Horsley R, Petrick A, Parker DM (2022) Is robotic revisional bariatric surgery justified? An MBSAQIP analysis. Obes Surg 32(12):3863–3868. 10.1007/s11695-022-06293-5 [DOI] [PubMed] [Google Scholar]
- 21.Clapp B, Liggett E, Jones R, Lodeiro C, Dodoo C, Tyroch A (2019) Comparison of robotic revisional weight loss surgery and laparoscopic revisional weight loss surgery using the MBSAQIP database. Surg Obes Relat Dis 15(6):909–919. 10.1016/j.soard.2019.03.022 [DOI] [PubMed] [Google Scholar]
- 22.Bertoni MV, Marengo M, Garofalo F, Volonte F, Regina DL, Gass M, Mongelli F (2021) Robotic-assisted versus laparoscopic revisional bariatric surgery: a systematic review and meta-analysis on perioperative outcomes. Obes Surg 31(11):5022–5033. 10.1007/s11695-021-05668-4 [DOI] [PubMed] [Google Scholar]
- 23.Berger ER, Clements RH, Morton JM, Huffman KM, Wolfe BM, Nguyen NT, Ko CY, Hutter MM (2016) The impact of different surgical techniques on outcomes in laparoscopic sleeve gastrectomies: the first report from the metabolic and bariatric surgery accreditation and quality improvement program (MBSAQIP). Ann Surg 264(3):464. 10.1097/SLA.0000000000001851 [DOI] [PubMed] [Google Scholar]
- 24.Zhang L, Tan WH, Chang R, Eagon JC (2015) Perioperative risk and complications of revisional bariatric surgery compared to primary Roux-en-Y gastric bypass. Surg Endosc 29(6):1316–1320. 10.1007/s00464-014-3848-4 [DOI] [PubMed] [Google Scholar]
- 25.Fernandez-Bustamante A, Frendl G, Sprung J, Kor DJ, Subramaniam B, Ruiz RM, Lee J, Henderson WG, Moss A, Mehdiratta N, Colwell MM, Bartels K, Kolodzie K, Giquel J, Melo MFV (2017) Postoperative pulmonary complications, early mortality, and hospital stay following noncardiothoracic surgery: a multicenter study by the perioperative research network investigators. JAMA Surg 152(2):157–166. 10.1001/jamasurg.2016.4065 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.El Chaar M, King K, Pastrana M, Galvez A, Stoltzfus J (2021) Outcomes of robotic surgery in revisional bariatric cases: a propensity score-matched analysis of the MBSAQIP registry. J Robotic Surg 15(2):235–239. 10.1007/s11701-020-01098-z [DOI] [PubMed] [Google Scholar]
- 27.Scarritt T, Hsu CH, Maegawa FB, Ayala AE, Mobily M, Ghaderi I (2021) Trends in utilization and perioperative outcomes in robotic-assisted bariatric surgery using the MBSAQIP database: A 4-Year analysis. Obes Surg 31(2):854–861. 10.1007/s11695-020-05055-5 [DOI] [PubMed] [Google Scholar]
- 28.Tatarian T, Yang J, Wang J, Docimo S, Talamini M, Pryor AD, Spaniolas K (2021) Trends in the utilization and perioperative outcomes of primary robotic bariatric surgery from 2015 to 2018: a study of 46,764 patients from the MBSAQIP data registry. Surg Endosc 35(7):3915–3922. 10.1007/s00464-020-07839-3 [DOI] [PubMed] [Google Scholar]
- 29.Barbash GI, Glied Sherry A (2010) New technology and health care costs—the case of robot-assisted surgery. NEJM 363(8):701–704. 10.1056/NEJMp1006602 [DOI] [PubMed] [Google Scholar]
- 30.Wright JD (2017) Robotic-assisted surgery: balancing evidence and implementation. JAMA 318(16):1545–1547. 10.1001/jama.2017.13696 [DOI] [PubMed] [Google Scholar]
- 31.Read MD, Torikashvili J, Janjua H, Grimsley EA, Kuo PC, Docimo S (2024) The downtrending cost of robotic bariatric surgery: a cost analysis of 47,788 bariatric patients. J Robotic Surg 18(1):63. 10.1007/s11701-023-01809-2 [DOI] [PubMed] [Google Scholar]
- 32.Chaar ME, Gacke J, Ringold S, Stoltzfus J (2019) Cost analysis of robotic sleeve gastrectomy (R-SG) compared with laparoscopic sleeve gastrectomy (L-SG) in a single academic center: debunking a myth! Surg Obes Rel Dis 15(5):675–679. 10.1016/j.soard.2019.02.012 [DOI] [PubMed] [Google Scholar]
- 33.Salem JF, Bauerle WB, Arishi AA, Stoltzfus J, El Chaar M (2023) Direct medical costs of robotic sleeve gastrectomy compared to laparoscopic approach in a single academic center. J Robotic Surg 17(1):49–54. 10.1007/s11701-022-01385-x [DOI] [PubMed] [Google Scholar]
- 34.Pratama KG, Nugroho H, Hengky A, Tandry M, Pauliana P (2024) Glucagon-like peptide-1 receptor agonists for post-bariatric surgery weight regain and insufficient weight loss: a systematic review. Obes Med 46:100533. 10.1016/j.obmed.2024.100533 [Google Scholar]
- 35.Müller TD, Blüher M, Tschöp MH, DiMarchi RD (2022) Anti-obesity drug discovery: advances and challenges. Nat Rev Drug Discov 21(3):201–223. 10.1038/s41573-021-00337-8 [DOI] [PMC free article] [PubMed] [Google Scholar]