Sylke Haal

This thesis contributes to the knowledge of the pathogenesis of cholesterol gallstone formation after bariatric surgery and whether UDCA prophylaxis should be prescribed after bariatric surgery to prevent symptomatic gallstone disease, and to whom.

In chapter 2, we analysed the biliary lipid composition of 18 bariatric gallstone patients and 17 nonbariatric gallstone patients. The concentrations of the common biliary lipids (bile acids, cholesterol, and phospholipids) were all significantly higher in bariatric gallstone patients, while the cholesterol saturation index did not significantly differ between the two groups. In the lipidomics analysis, enhanced amounts of several lipid species were found in the gallbladder bile of nonbariatric gallstone patients. Most remarkable was a higher amount of triglyceride. After finding a concomitant increase of apolipoprotein B in nonbariatric gallstone patients, we hypothesised that biliary triglyceride-rich lipoprotein secretion might contribute to gallstone formation in this patient group. The high concentration of common biliary lipids and the large number of small stones found in the gallbladders of bariatric gallstone patients, implies that impaired gallbladder emptying favours gallstone formation after bariatric surgery.

In chapter 3a, the protocol of the UPGRADE trial is presented. The UPGRADE trial was a multicentre, double-blind, randomised, placebo-controlled superiority trial in three hospitals in the Netherlands. The aim of the UPGRADE trial was to assess the efficacy of UDCA in preventing symptomatic gallstones disease among patients undergoing bariatric surgery. Sample size calculation (n=980) was based on the expected effect of UDCA to reduce the occurrence of symptomatic gallstone disease from 11% to 5.5%. Eligible patients were those with an intact gallbladder scheduled for laparoscopic Roux-en-Y gastric bypass (RYGB) or sleeve gastrectomy. Patients with asymptomatic gallstones before surgery were included because in current clinical practice, no preoperative gallbladder screening is done in general unless patients are symptomatic. Patients were randomly assigned (1:1) to receive 900 mg UDCA daily for six months or placebo. Randomisation was done before surgery and stratified by the presence of asymptomatic gallstones at baseline and type of scheduled surgery. The primary endpoint was the proportion of patients with symptomatic gallstone disease within 24 months of follow-up.

In chapter 3b, the statistical analysis plan (SAP) of the UPGRADE trial is described. The SAP includes a brief description of the trial design. Furthermore, two important changes to the methods after trial commencement are reported. During the initial course of the trial, the bariatric procedure sleeve gastrectomy was added to the inclusion criteria, because of increasing use worldwide, in Europe, and in the participating centres. Second, the accepted start of the study drug was extended from two weeks to eight weeks, in case a patient experienced difficulties taking the study drug within the first two weeks after surgery. Moreover, the SAP describes how six different analysis populations are defined to estimate the effectiveness and efficacy of UDCA. Last, the methods that will be used to compare the primary and secondary endpoints between the UDCA and placebo group are described, and any additional analyses.

In chapter 4, the results of the UPGRADE trial are reported. 959 patients were included in the modified intention-to-treat population and available for primary endpoint assessment at 24 months. One-fifth (189 [20%]) of the patients was diagnosed with asymptomatic gallstones at baseline and most patients (92%) received a laparoscopic RYGB. We found a beneficial effect of UDCA in patients without gallstones at baseline (RR 0.47; 95% CI 0.27-0.84; p=0.008), and no effect in patients with asymptomatic gallstones at baseline (RR 1.22; 95% CI 0.61-2.47; p=0.57). The beneficial effect was even stronger in patients without gallstones at baseline undergoing RYGB (RR 0.37; 95% CI 0.20-0.71; p=0.002), whereas the subgroup of patients undergoing sleeve gastrectomy was too small to draw clear conclusions. Adverse events were rare.

Following the clinical results of the UPGRADE trial, in chapter 5 the prophylactic prescription of UDCA was evaluated economically for the subgroup of patients without gallstones before surgery, who received a RYGB, and who had data available for primary endpoint assessment at 24 months (n=705). The economic evaluation was performed as cost-effectiveness and cost-utility analysis from a healthcare and societal perspective with a time horizon of 2 years. UDCA prophylaxis resulted in a significant higher proportion of patients remaining free of symptomatic gallstone disease and in a surplus of nearly 0.05 in QALYs gained from baseline. The mean cost savings of -€1392 was non-significant, but indicative of sufficient return on investment to compensate for the initially higher costs upfront of patient selection and prophylactic treatment.

In chapter 6, potential predictors and reasons for poor adherence to the study drug of the UPGRADE trial are presented. Poor adherence was defined as the usage of less than 300 out of 364 pills within 8 months after bariatric surgery. 357 out of 967 patients (37%) were considered poor adherers. Younger age, foreign origin, unemployment, and a sleeve gastrectomy were associated with poor adherence in the multivariable analysis. Furthermore, a difference in adherence status was also noted for the centres of surgery. The two most common reported reasons for poor adherence were the lack of motivation to use the study drug (58%) and difficulties in taking the study drug (35%). The questionnaires showed that mainly the large size of the tablets and problems with swallowing hampered taking the study drug. Poor adherers reported that smaller pills, capsules instead of tablets, and one pill a day instead of two pills could help improve their adherence.

In chapter 7 we investigated associations between patients characteristics and gallstone disease, and gallstone formation after bariatric surgery using data from the UPGRADE trial. Associations were evaluated using logistic regression analysis. Risk factors for symptomatic gallstone disease were the presence of a pain syndrome and asymptomatic gallstones prior to surgery. Advanced age was protective, and UDCA prophylaxis did not reach statistical significance. No risk factors were identified for gallstone formation, whereas advanced age, the use of statins, and UDCA prophylaxis all reduced the risk.

Chapter 8 includes a retrospective case-control study in which we investigated the association between several (risk) factors and cholecystectomy after bariatric surgery. Female gender, Caucasian ethnicity, higher percentage total weight loss at 12 months, and preoperative pain syndrome were significantly associated with an increased risk for cholecystectomy, while older age and preoperative statin use were both associated with a decreased risk. A dose-effect relationship was found between the intensity of preoperative statin and risk for cholecystectomy, although only significant in the univariate model. We found no significant associations between three female-associated factors (fertility, offspring, and use of birth control pills) and cholecystectomy.

GENERAL DISCUSSION AND FUTURE PERSPECTIVES

The majority of patients with gallstones remain asymptomatic and do not require treatment. However, in case of symptoms, the therapeutic options are limited. The current gold standard is laparoscopic cholecystectomy. Laparoscopic cholecystectomy is a safe surgical procedure with a low morbidity and mortality rate, and one of the most frequently performed surgical procedures worldwide. Although humans can easily live without a gallbladder, it is not just an accessory organ of the gastrointestinal tract which stores and concentrates bile. The gallbladder also regulates the composition of bile, and controls the flow of bile into the intestine, and thereby the enterohepatic circulation of bile acids. Besides their known function as lipid solubilizers, bile acids act as signaling factors that regulate metabolism and inflammation. Accordingly, recent evidence has suggested that cholecystectomy has unfavourable metabolic effects, and may be less innocuous than we have always presumed. Alternatives for laparoscopic cholecystectomy include oral bile acid dissolution therapy with UDCA and extracorporeal shock wave lithotripsy. These options, however, are not recommended anymore because of the low cure and high recurrence rate. Hence, successful primary prevention strategies are warranted.

For the development of new prevention strategies, the expansion of our knowledge about the pathophysiology of cholesterol gallstones is needed. After a literature search revealed that the biliary lipid composition of bariatric and nonbariatric gallstone patients had never been compared, the observational study described in chapter 2 was designed. The study yielded two novel findings: a higher concentration of all three major biliary lipids and thus a higher total lipid concentration in bariatric gallstone patients, and in particular a higher amount of triglyceride (TG) in nonbariatric gallstones patients. These observed differences suggest that gallstone formation follows a different trajectory in bariatric patients compared to nonbariatric patients. The high total lipid concentration and the large number of small stones in bariatric patients has led to the hypothesis that impaired gallbladder emptying is the primary driving factor for gallstone formation during rapid weight loss. Others have demonstrated that gallbladder emptying is indeed impaired in patients on very-low-calorie diets, or after RYGB. Gallbladder stasis is the result of impaired gallbladder emptying, which provides the time and condition for the nucleation of cholesterol crystals and their growth into stones. The regulation of gallbladder motility is closely connected to digestion and mediated by neurohormonal signals. Drug affecting gallbladder motility might be a valuable option to prevent gallstone formation during rapid weight loss. To confirm our hypothesis, a future study should be performed in which gallbladder bile is obtained during and after bariatric surgery. Although time-consuming, ideally gallbladder emptying should be measured too. Even more important, gallstone-free bariatric patients should be included to verify that we are not just looking at alterations imposed by bariatric surgery, but at alterations that have facilitated gallstone formation. Although the work presented in this thesis has primarily focused on bariatric patients, in chapter 2 a new hypothesis for gallstone formation in nonbariatric patients is also presented. Based on the higher amount of TG and concomitant increase of apolipoprotein B, we hypothesised that biliary triglyceride rich lipoprotein (TRL) secretion might contribute to the formation of gallstones in nonbariatric patients. If our findings can be reproduced in a larger cohort with a control group, than the proposed mechanism should be further examined. Perhaps, biliary TRL secretion will become a potential novel target in the prevention of cholesterol gallstone disease in the general population. Yet, primary non-pharmalogical prevention would be more desirable in the general population, since cholesterol gallstone disease is a ‘mass’ disease with many asymptomatic cases. Several modifiable dietary and lifestyle factors facilitate gallstone formation. Accordingly, a healthy diet, regular physical activity, and the maintenance of an ideal body weight, are expected to prevent gallstone formation. A healthy lifestyle should thus be promoted, even though the quality of evidence of most lifestyle interventions is low. Their beneficial effects are simply too difficult to estimate in prospective studies.

Selected subjects high at risk for cholesterol gallstones are eligible for primary pharmalogical prevention, for example subjects with morbid obesity undergoing bariatric surgery. For a long time, the lack of high level evidence has withhold clinicians to prescribe UDCA prophylaxis after bariatric surgery. We conducted the UPGRADE trial - a multicentre, double-blind, randomised, placebo-controlled superiority trial - to finally answer the question whether UDCA should be given as prophylactic treatment. The UPGRADE trial has provided clear evidence that UDCA prophylaxis reduces the occurrence of symptomatic gallstone disease in patients undergoing RYGB without gallstones before surgery (chapter 4). We did not observe a beneficial effect of UDCA in patients undergoing sleeve gastrectomy, nor in patients with asymptomatic gallstones before surgery. However, one must keep in mind that the UPGRADE trial was not powered to evaluate the efficacy of UDCA prophylaxis in these two subgroups.

Nonetheless, in patients undergoing sleeve gastrectomy, a similar beneficial effect of UDCA on symptomatic gallstone disease is still possible, given the wide 95% confidence interval(s), and expected, given the following knowledge. First, the postoperative weight loss, which predisposes bariatric patients to gallstone formation, is almost comparable between RYGB and sleeve gastrectomy. Second, similar incidences of symptomatic gallstone disease are reported. Third, previous studies have shown a similar protective effect of UDCA on gallstone formation. However, a prospective study adequately powered to assess the efficacy of UDCA after sleeve gastrectomy is needed. Especially, since sleeve gastrectomy is now the most common type of bariatric procedure worldwide. In the meantime, the decision to prescribe UDCA after sleeve gastrectomy should be carefully considered by the treating physician.

A substantial part of the patients scheduled for bariatric surgery have asymptomatic gallstones before surgery. In the UPGRADE trial, preexistent asymptomatic gallstones were identified in 20% of the patients scheduled for surgery. In particular, these patients form a special subgroup. To begin with, they do not seem to benefit from UDCA prophylaxis, and secondly, they are at increased risk for symptomatic gallstone disease after bariatric surgery compared to those without (chapter 7). As a consequence, routine gallbladder screening should be included in the preoperative workup for bariatric surgery to select patients for prophylactic treatment with UDCA. This routine screening, however, has two major downsides. First, it fosters the risk that patients who are told to have asymptomatic gallstones are more prone to attribute abdominal symptoms after bariatric surgery to the presence of gallstones. Second, although ultrasound imaging technology has improved over the last decades, the transabdominal ultrasound has lower sensitivity to detect gallbladder pathology in patients with morbid obesity due to increased amounts of fat. As a result, some bariatric patients will receive UDCA prophylaxis unnecessary. A promising future option to protect patients with preexistent asymptomatic gallstones might be a combination of gallbladder-preserving surgery (cholecystolithotomy) during and prophylactic treatment with UDCA after bariatric surgery. Up to now, this approach has only been investigated in a pilot setting. Yet, future research assessing the value of this minimally invasive approach is desirable.

In the Netherlands, we have a high quality and accessible healthcare system. However, the sustainability of our healthcare system is under pressure since our healthcare expenditure is increasing faster than our economic growth. Annually, the government decides which healthcare is covered by the standard health insurance package. Health economic evaluations are increasingly being used to provide relevant information on the cost-effectiveness of (new) interventions and to provide guidance in the decision-making on how to spend our healthcare budget. Following the clinical results of the UPGRADE trial, we evaluated the prescription of UDCA prophylaxis economically for the subgroup of patients without gallstones before surgery, who received a RYGB, and who had data available for primary endpoint assessment at 24 months (chapter 5). We performed a cost-effectiveness and cost-utility analysis of UDCA prophylaxis from a healthcare and societal perspective with a time horizon of 2 years. We have shown that UDCA prophylaxis is an efficient drug treatment strategy in at least 87 of every 100 patients without gallstones scheduled for RYGB surgery.

As earlier mentioned in the introduction section of this thesis. Poor adherence to preventive treatment is a common problem, and a concern in patients undergoing bariatric surgery. In the UPGRADE trial, we observed a moderate overall adherence rate of 63%. Hence, there is room for improvement. We observed no difference in adherence rate between the UDCA group and placebo group, but did find that patients with a sleeve gastrectomy were more often nonadherent compared with patients with a RYGB. Since poor adherence reduces the efficacy of UDCA prophylaxis, we further investigated drivers of poor adherence. The findings described in chapter 6 help to identify those patients who may benefit from extra education and guidance in medication intake after bariatric surgery.

The studies in chapter 7 and 8 identified associations between patient characteristics and gallstone disease, gallstone formation, and cholecystectomy. Following these two predictor finding studies, a next step forward would be the development of a clinical prediction model. If a useful and accurate model can be developed which can estimate the probability of gallstone disease, gallstone formation, and cholecystectomy in the individual bariatric patient, maybe less patients will have to be treated to prevent symptomatic gallstone disease. Furthermore, such a model can assist clinicians and patients in their shared decision-making to start or not start UDCA prophylaxis after bariatric surgery.

In conclusion, cholesterol gallstone disease is a prevalent, multifactorial, and costly disease. The studies enlisted in this thesis have made a modest contribution to the understanding of the pathophysiology of cholesterol gallstones, and provided clear evidence for the efficacy, safety, and cost-effectiveness of UDCA as prophylactic treatment in patients undergoing RYGB without gallstones before surgery. Yet, much work remains to be done to tackle this widespread disease.