There are several known risk factors for gallstone formation in the general public, such as age, female gender, obesity, and parity [1, 2]. One well known risk factor for the development of gallstones and subsequent biliary disease is the period of rapid weight loss found in the first three years following roux-en-y gastric bypass (RYGB) [3,4,5]. The incidence of gallstone formation after RYGB is as high as 38%. Unfortunately, the mechanism of stone formation in bariatric patients is not fully understood but multiple theories have been proposed such as: increased cholesterol saturation as a result of mobilization from adipose tissue, gallbladder stasis, increased secretion of mucin and calcium, and increased presence of prostaglandins and arachidonic acid promoting cholesterol crystal precipitation [6].

The optimal management of biliary disease in bariatric patients has remained controversial. In the current era of minimally invasive bariatric surgery, the trend has been to avoid prophylactic or concomitant asymptomatic cholecystectomy (CCY) due to the low rate of biliary disease following bariatric surgery [7]. A recent review by the ASMBS (American Society of Metabolic and Bariatric Surgeons) Foregut Committee recommends concomitant CCY only in symptomatic patients undergoing primary RYGB [8]. These recommendations were based on the increased risk associated with interval cholecystectomy and the low risk of developing biliary disease [8]. Previous studies have been inconsistent on the risk of concomitant CCY [9]. The largest study of the NSQIP (American College of Surgeons National Quality Improvement Program) database demonstrated a higher rate of major adverse outcomes in patients undergoing RYGB with concomitant CCY [10]. A recent study evaluating a single year of the MBSAQIP database, however, demonstrated no increase in morbidity in patients undergoing concomitant CCY [11].

Advancements in robotics have significantly increased the utilization of robotic bariatric surgery. The safety and efficacy of robotics in bariatric surgery has been previously demonstrated [12,13,14]. On the other hand, studies, including Pokala et al. have evaluated the perioperative outcomes of robotic assisted CCY, finding an increase in the 30 day morbidity and length of stay [15, 16]. In review of the literature, the short-term outcomes of laparoscopic compared to robotic bariatric surgery with concomitant CCY have not been previously evaluated. The primary objective of this study was to compare 30 day outcomes of patients, in a propensity-matched cohort, undergoing minimally invasive (MIS)/RYGB versus MIS/RYGB with concomitant CCY. A secondary objective was to compare the 30-day outcomes for laparoscopic RYGB (LRYGB) versus robotic RYGB (RRYGB) with concomitant CCY.

Materials and methods

Prior to the initiation of research, this project was exempt by the Geisinger Institutional Review Board. This study uses the de-identified data from the MBSAQIP database, without the need for individual written consent. The Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP) database participant use files from 2015–2017 were queried for primary procedure CPT codes for laparoscopic and robotic RYGB (43,644, 43,645) alone and with the addition of laparoscopic and robotic CCY (47,562, 47,563). Emergency, revision procedures, and records missing data relevant for analysis were excluded. A total of 117,939 patients were identified as undergoing a MIS/RYGB from 2015–17. After matching 10:1 a SMD (Standard Mean Difference) below 0.01 was considered well matched for all variables. Variables selected for matching included Age, Body Mass Index (BMI), Sex, Race, first assist training level, American Society of Anesthesiologists (ASA) Score, and comorbidities included in the MBSAQIP database. After propensity matching 36,260 patients with MIS/RYGB and 3626 MIS/RYGB with concomitant CCY are identified in Table 1.

Table 1 Pre- and post-propensity match variables
Full size table

The 30-day outcomes of MIS/RYGB with and without CCY were compared, and these individual outcomes are noted in Table 2 and Table 3. The post matched cohort of MIS/RYGB with CCY was then separated into 213 LRYGB and 413 RRYGB cases for a second comparison. Outcomes were compared using the Chi-squared or Fisher’s exact test for categorical variables and the Wilcoxon rank-sum test for continuous variables. Results are reported as frequency (%) and median and interquartile range unless otherwise specified.

Table 2 Pre-match complications
Full size table
Table 3 Post-match complications
Full size table

Results

Query of the MBSAQIP participant use date file from 2015–2017, a total of 117,939 patients were identified as having a MIS/RYGB. A total of 36,260 patients who underwent an MIS/RYGB and 3,626 who underwent a MIS/RYGB with concomitant CCY were compared after propensity matching demonstrated in Table 1. As seen in Table 4, the mortality rate was low in both groups (0.2% vs. 0.1%; p = 0.209). Both groups had similar morbidity rates (4.6% vs. 4.6%; p = 0.892) as well as need for reoperations or interventions. The length of stay (LOS) and readmission rates were similar for both groups. The addition of concomitant CCY did, however, significantly increase the operative time (111 min vs. 139 min; p < 0.0001).

Table 4 Minimally invasive RYGB with concomitant cholecystectomy versus alone
Full size table

When comparing the surgical approach, the laparoscopic group was associated with significantly shorter operative times compared to robotic group (135 min vs. 177 min; p < 0.0001) demonstrated in Table 5. The robotic and laparoscopic groups demonstrated no significant difference in LOS, readmission, reoperation, or intervention. The morbidity and mortality were similar between the laparoscopic and robotic approach.

Table 5 Laparoscopic versus Robotic RYGB with concomitant cholecystectomy
Full size table

Discussion

The rapid development of gallstones after bariatric surgery is well documented in over one-third of patients. Previous studies have demonstrated concomitant CCY is safe; however, the perioperative morbidity has been inconsistently documented in published studies [9, 11, 17]. The current recommendations are to avoid cholecystectomy in bariatric patients with asymptomatic cholelithiasis. The ASMBS recommendations were based on 1. low incidence of biliary disease following MIS/RYGB and 2. the increased morbidity of interval cholecystectomy [8].

The NSQIP study is currently the largest published series of bariatric patients undergoing concomitant cholecystectomy demonstrated a higher morbidity with concomitant cholecystectomy group [10]. Our study is the largest and most comprehensive study on the perioperative outcomes of concomitant cholecystectomy in patients undergoing primary MIS/RYGB. The authors believe using the MBSAQIP PUF (Participant user files) has several advantages over the previous NSQIP study. The MBSAQIP PUF is a comprehensive database of all cases performed at MBSAQIP centers as opposed to NSQIP, which relies on random patient sampling. In addition, MBSAQIP centers are required to follow patients for 1 year so admissions and procedures performed at outside institutions should be captured during routine follow-up.

The current study demonstrates that the addition of concomitant CCY did significantly increase the operative time as would be expected. The length of stay was also increased by 0.3 days in the concomitant cholecystectomy group; however, the authors believe this is not clinically significant. Unlike the NSQIP data, our study demonstrated concomitant CCY was not associated with a significant increase in readmission, reoperation, or intervention [17].

The ASMBS guidelines also cited the low incidence of biliary disease as the second justification against asymptomatic cholecystectomy. The ASMBS guidelines document the risk of developing biliary disease as 6.8% [8]. In review of the supporting articles for the recommendation, there were two studies and one meta-analysis cited. One study reported an interval cholecystectomy rate of 4.9% in 1050 patients, which reported that 78% were followed up at 2 years [18]. The other study included both RYGB and Sleeve gastrectomy patients. The study included 146 patients with an interval cholecystectomy rate of 3.4% at 12-month follow-up [19]. The meta-analysis was more varied in findings. The interval cholecystectomy rate was between 2.3–18.6% [18, 20,21,22,23,24,25,26,27,28,29,30,31]. The studies included in the meta-analysis were small studies with half the studies less than 200 patients. The studies also had relatively short follow-up with only one study reporting the percent of patients followed up at 2 years [18, 20,21,22,23,24,25,26,27,28,29,30,31]. The risk of biliary disease, however, has been previously demonstrated to be increased for 3 years after bariatric surgery (5). In addition, most (67%) of the studies included in the meta-analysis performed concurrent cholecystectomy in asymptomatic patients with cholelithiasis. The interval cholecystectomy rate may also be underestimated since up to 75% of patients may have their interval cholecystectomy at a different institution [32].

The optimal technical approach for primary MIS/RYGB with concomitant CCY has not been previously studied. The robotic platform for primary bariatric surgery has demonstrated similar perioperative outcomes to the laparoscopic approach [12,13,14]. Robotic CCY has also been shown to be safe, however, with a potentially higher overall complication rate [16]. Our study demonstrates RRYGB with concomitant CCY can be safely performed with similar perioperative outcomes to the laparoscopic approach. The robotic platform, however, had a significantly longer operative times compared to the laparoscopic approach. Prior studies have also found significant cost differences in the two techniques [15, 16]. Strosberg et al. not only demonstrated increased cost with robotic CCY but also decreased hospital revenue when compared to the laparoscopic approach [15].

We recognize that this study is a retrospective review of a prospectively maintained national database which has inherent limitations of observer and reporter bias. The MBSAQIP PUF does not collect information regarding preoperative biliary symptomatology or clinical decision making on the technical approach to MIS/RYGB. Furthermore, we do not have long-term outcomes outside of 30 days postoperatively.

To date this is the largest and most comprehensive study comparing perioperative outcomes of primary MIS/RYGB with concomitant CCY. The current study demonstrates that concomitant cholecystectomy is associated with minimal impact to morbidity, mortality, readmissions or interventions compared to primary MIS/RYGB. The authors believe the current recommendations against cholecystectomy in asymptomatic patients is too strongly worded given the low risk and poor evidence on the rate of interval cholecystectomy. The authors believe the laparoscopic and robotic platform have similar perioperative outcomes, however, the robotic platform is limited by the significant increase in operative time and cost.