Obesity Surgery (2019) 29:464–473 https://doi.org/10.1007/s11695-018-3532-1 ORIGINAL CONTRIBUTIONS Bariatric Surgery Did Not Increase the Risk of Gallstone Disease in Obese Patients: a Comprehensive Cohort Study Jian-Han Chen1,2,3 & Ming-Shian Tsai1,2,3 & Chung-Yen Chen1,2,3 & Hui-Ming Lee 3,4 & Chi-Fu Cheng5,6 & Yu-Ting Chiu5,6 & Wen-Yao Yin5,6,7 & Cheng-Hung Lee5,6 Published online: 11 November 2018 # Springer Science+Business Media, LLC, part of Springer Nature 2018 Abstract Purpose The aim of this study was to evaluate the influence of bariatric surgery on gallstone disease in obese patients. Materials and Methods This large cohort retrospective study was conducted based on the Taiwan National Health Insurance Research Database. All patients 18–55 years of age with a diagnosis code for obesity (ICD-9-CM codes 278.00–278.02 or 278.1) between 2003 and 2010 were included. Patients with a history of gallstone disease and hepatic malignancies were excluded. The patients were divided into non-surgical and bariatric surgery groups. Obesity surgery was defined by ICD-9-OP codes. We also enrolled healthy civilians as the general population. The primary end point was defined as re-hospitalization with a diagnosis of gallstone disease after the index hospitalization. All patients were followed until the end of 2013, a biliary complication occurred, or death. Results Two thousand three hundred seventeen patients in the bariatric surgery group, 2331 patients in the non-surgical group, and 8162 patients in the general population were included. Compared to the non-surgery group (2.79%), bariatric surgery (2.89%) did not elevate the risk of subsequent biliary events (HR = 1.075, p = 0.679). Compared to the general population (1.15%), bariatric surgery group had a significantly higher risk (HR = 4.996, p < 0.001). In the bariatric surgery group, female gender (HR = 1.774, p = 0.032) and a restrictive procedure (HR = 1.624, p = 0.048) were risk factors for gallstone disease. Conclusion The risk for gallstone disease did not increase after bariatric surgery, although the risk was still higher than the general population. The benefit of concomitant cholecystectomy during bariatric surgery should be carefully evaluated. Keywords Bariatric surgery . Validation . NHIRD . Gallstone disease . Acute cholecystitis . Obesity Wen-Yao Yin and Cheng-Hung Lee contributed equally to this work. Introduction * Wen-Yao Yin [email protected] Obesity is related to multiple comorbidities, including meta- * Cheng-Hung Lee bolic syndromes and malignancies [1], and is a reported risk [email protected] factor for gallstone formation. Bariatric surgery has been Jian-Han Chen shown to be a most effective and durable treatment of obesity [email protected] and associated co-morbidities [2]; however, bariatric surgery leads to rapid weight loss, which is also a risk factor for gall- 1 Bariatric and Metabolic Surgery Center, E-da hospital, stone formation. Kaohsiung, Taiwan The reported postoperative biliary complication rate requir- 2 Department of General Surgery, E-da hospital, Kaohsiung, Taiwan ing cholecystectomy is 0.9~7.5% after laparoscopic sleeve 3 School of Medicine, I-Shou University, Kaohsiung, Taiwan gastrectomy [3–5]and6–50% after laparoscopic gastric by- – 4 Department of General Surgery, E-Da Cancer Hospital, pass [6 9]. Therefore, some surgeons perform concomitant Kaohsiung, Taiwan cholecystectomies for patients who undergo bariatric surgery 5 Department of General Surgery, Buddhist Dalin Tzu Chi Hospital, without significantly increased operative times, length of hos- No.2, Minsheng Rd., Dalin Township, Chiayi County 622, Taiwan pital stay, and morbidity and mortality rates [7, 10, 11]. In 6 School of Medicine, Tzu Chi University, Hualien, Taiwan contrast, Warschkow et al. [12] conducted a meta-analysis involving concomitant cholecystectomies and reported signif- 7 Department of Surgery, College of Medicine, Tzu Chi University, Hualien, Taiwan icantly higher complication and mortality rates than bariatric OBES SURG (2019) 29:464–473 465 surgery alone. Moreover, Worni et al. [13] concluded that diagnosis code for overweight or obesity (ICD-9 codes concomitant cholecystectomies have higher postoperative 278.00–278.02 or 278.1) [18, 19] on admission between complications, including infections, gastrointestinal and pul- 2003 and 2010. The patients were separated into two groups monary complications, reintervention rates, and longer hospi- (bariatric surgery and non-surgical groups). tal stays. We defined bariatric surgery based on ICD-9 codes, as Thus, there is an ongoing debate regarding concomitant follows: malabsorptive procedures (high gastric bypass, cholecystectomy during bariatric surgery; however, most ICD-9-CM OP44.31; laparoscopic gastroenterostomy, ICD- studies have involved gallstone formation and cholecystecto- 9-CM OP44.38; and other gastroenterostomy without gastrec- my after bariatric surgery, and the difference between obese tomy, ICD-9-CM OP44.39) and restrictive procedures (lapa- patients who do and do not undergo bariatric surgery is un- roscopic sleeve gastrectomy, ICD-9-CM OP43.82; open and known. Moreover, selective concomitant cholecystectomies other partial gastrectomy, ICD-9-CM OP43.89; laparoscopic can be performed if risk factors for gallstone disease are pres- gastroplasty, ICD-9-CM OP44.68; other repair of stomach, ent. Until now, with the exception of age [14], no risk factors ICD-9-CM OP44.69; laparoscopic gastric restrictive proce- for gallstone disease after bariatric surgery have been identi- dure, ICD-9-CM OP44.95; and other operations on the stom- fied [5, 15]. Therefore, decisions regarding cholecystectomies ach, ICD-9-CM OP44.99) [20, 21]. during bariatric surgery vary, from routine prophylactic cho- lecystectomies [16] to cholecystectomies only in patients with Exclusion Criteria biliary symptoms [17]. The aim of this study was to determine the incidence of Patients were also excluded for the following reasons: (1) < 18 gallstone disease after bariatric surgery and identify the influ- or > 55 years of age, (2) had a history of cholelithiasis or ence of bariatric surgery on gallstone disease in obese patients. choledocholithiasis, (3) a history of cholecystectomy (before In addition, we attempted to identify the risk factors for gall- or concomitant with the index date), (4) a history of stonediseaseafterbariatricsurgery.Usinganationwide hepatobiliary cancer (ICD-9-CM code 155, 156), (5) died dur- population-based dataset, we analyzed sufficient patient data ing admission or within 30 days of the index admission [22], and performed comprehensive follow-up to answer these or (6) undetermined gender. questions. Patients who did not meet these criteria were given the option to self-pay for bariatric surgery. In the non-surgical group, we excluded patients admitted to the surgical ward to Method prevent this event. In the bariatric surgery group, we also excluded patients with a primary diagnosis other than obesity This retrospective study was fully evaluated and approved by to exclude those patients who underwent surgery for other the Institutional Review Board of Buddhist Dalin Tzu Chi reasons. Hospital (approval B10502007) and was conducted in accor- dance with the principles of the Helsinki Declaration. For this Covariates type of study, formal consent is not required. We extracted the dataset used in this study from the Taiwan National Health Several comorbidities, including diabetes (ICD-9 diagnostic Insurance Research Database (NHIRD; registered number code, 250.x), hypertension (ICD-9 diagnostic code, 401.x– NHIRD-103-246). The database was cross-checked, and med- 405.x) [23], hyperlipidemia (ICD-9 diagnostic code, 272.x), ical charts were validated by the Taiwan Bureau of National and gout (ICD-9 diagnostic code, 274.x). These patients were Health Insurance. For this analysis, we extracted the data from matched 1:1 based on age, gender, and comorbidities by pro- the inpatient expenditures by admission, the registry for ben- pensity score matching. eficiaries, and the Registry for Catastrophic Illness Patient Database from the NHIRD database (1996–2013). Validation Inclusion Criteria Although several validation studies have confirmed the reli- ability of this database [24–27], the validity of bariatric pro- In Taiwan, patients must meet several criteria in order for the cedure codes in the NHI database has not been well-docu- National Health Insurance (NHI) to pay for bariatric surgery. mented. We applied the inclusion criteria for bariatric surgery Specifically, the following criteria must be met: (1) between as detailed in the BInclusion Criteria^ section to patients who 18 and 55 years of age, (2) attempted conservative methods to underwent bariatric surgery covered by the NHI program at lose weight for at least 6 months, (3) BMI > 35 kg/m2 with Buddhist Dalin Tzu Chi Hospital (a 1000-bed regional hospi- comorbidities or > 40 kg/m2, and (4) no psychiatric disease. tal in Taiwan) from July 2001 to February 2016. Two general Thus, we included all patients 18–55 years of age with the surgeons (Drs. Cheng-Hung Lee and Chi-Fu Cheng) 466 OBES SURG (2019) 29:464–473 independently reviewed the clinical data, medical records, and End Point operative notes from all patients. Disagreements were re- solved by discussion until consensus was reached. The sensi- The primary end point was the development of biliary compli- tivity, specificity, positive predictive value (PPV), and nega- cations. Development of gallstone disease after bariatric sur- tive predictive value (NPV) of bariatric surgery procedure gery was defined as rehospitalization with a diagnosis of cho- codes were determined. lelithiasis (ICD-9-CM codes 574.x, 575.0, 575.1, and 575.6) after the index hospitalization. Patients with cholecystitis (ICD-9-CM codes 574.0, 574.1, 574.3, 574.4, 574.6, 574.7, General Population 574.8, 575.0, and 575.1) and choledocholithiasis ((ICD-9-CM codes 574.30-1, 574.40-1, 574.50-1, 574.60-1, 574.70-1, We randomly selected 100,000 adult patients (≧ 18 years of 574.80-1, and 574.90-1) were also identified.