Description of the condition
Gallstones (solid calculi), which comprise gallbladder stones (cholelithiasis) and bile duct stones (choledocholithiasis), result from imbalances in the chemical composition of bile, usually due to a surplus of cholesterol or bilirubin. Other pathophysiological risk factors include gallbladder hypomotility and destabilisation of bile due to kinetic protein factors (Maurer 2009). Additionally, intestinal hypomotility can contribute to the development of lithogenic bile via the increased formation of the secondary bile salt, deoxycholate, hence increasing the risk of gallbladder stones. In fact, people with gallbladder stones have been reported to have both increased concentrations of bacterial colonic deoxycholate and slow intestinal transit (Portincasa 1996). Gallstones can be further classified as cholesterol stones and pigmented stones. Cholesterol stones primarily comprise cholesterol and occur in over 90% of people with gallstones. Pigmented stones comprise calcium bilirubinate and are also referred to as black or brown pigmented stones (Schafmayer 2006).
Gallbladder stone prevalence is 10% to 20% of the population in Europe and America, which is predicted to increase due to the obesity epidemic and the increasing ageing population (Everhart 1999; Völzke 2005; Go 2013). A seven-fold risk of gallbladder stones has been reported in morbidly obese people when compared to normal-weight populations. This was demonstrated in a longitudinal study comprising 90,302 women (Stampfer 1992). In addition, rapid weight loss and weight cycling increase the risk of gallbladder stone formation (Weinsier 1993; Tsai 2006), and there also appears to be a dietary-related risk factor. For example, large epidemiological studies have reported associations between gallbladder stone occurrence and high-calorie, high-carbohydrate, and low-fibre diets (Tsai 2004a; Tsai 2005a; Tsai 2005b). Physical inactivity is also reported to confer an increased risk of cholelithiasis (Banim 2010). All of these factors, which are synonymous with Westernised lifestyle habits, can promote penetrance of lithogenic genes, which is important because the risk of gallbladder stones is also, to some extent, genetically determined (Stinton 2010; Stokes 2011).
Though the majority of gallbladder stones remain asymptomatic, approximately 25% become symptomatic and cholecystitis is a common complication. In addition, bile duct stones may cause cholangitis and pancreatitis (Friedman 1993). Symptomatic gallbladder stones are typically characterised by pain in the upper abdominal quadrant (biliary colic) (Kraag 1995), and upon complications, usually lead to hospital admission where laparoscopic cholecystectomy is the gold-standard treatment (Peery 2012). In the US, over 700,000 cholecystectomies are performed annually, making cholelithiasis one of the most costly diseases from all the gastroenterological disorders (Everhart 2009). The risk of operative mortality with cholecystectomy is low (< 0.6%); however, it is not risk free as damage to the bile ducts can ensue (Rosenmüller 2007). Moreover, symptom recurrence can occur in 40% of post-cholecystectomised patients (Keus 2010), and includes a wide range of symptoms (biliary and extra-biliary). One Swedish population-based cohort study reported a weak association between oesophageal adenocarcinoma and people with cholecystectomy (Lagergren 2011a). The study did not control for confounders, such as obesity. However, this observed association may be attributed to increased concentrations of bile in gastric fluid, which consequently may come into contact with the oesophagus during gastro-oesophageal reflux. The study also found an association between hepatocellular carcinoma and cholecystectomy (Lagergren 2011b). Consequently, alternative preventive and treatment options have been investigated (New Reference).
Description of the intervention
The link between nutrition and gallbladder stones is longstanding. Several studies have reported an increased risk with the consumption of refined carbohydrates (Tsai 2005a; Tsai 2005b), or with fat intake, though this is controversial and may depend on dietary lipid composition. For example, an increased gallbladder stone incidence was reported with a higher saturated and trans fat intake in some prospective follow-up studies (Tsai 2005c; Tsai 2008), but reduced with a high polyunsaturated and monounsaturated fat intake (Tsai 2004b). In fact, high nut consumption has also been linked to a decreased risk of cholecystectomy (Tsai 2004c). Moreover, dietary vegetable protein was associated with a reduced risk of cholecystectomy in women as was increased fruit and vegetable consumption (Tsai 2004d). Additionally, one large cross-sectional study in Italy, the Multicenter Italian Study on Epidemiology of Cholelithiasis (MICOL) study, reported a positive correlation between higher protein intake and gallbladder stone risk (Attili 1998). High calorie diets have also been positively associated with increased cholelithiasis risk in some studies, though this risk may be confounded by the development of obesity (Ortega 1997; Caroli-Bosc 1998; Tsunoda 2004).
Consequently, several studies have investigated the manipulation of various modifiable risk factors, namely dietary-related interventions such as lower calorie intake not exceeding individual needs or moderately reduced for slow steady weight loss, and modification of specific food groups (i.e., fat intake) during weight loss (Gebhard 1996; Heshka 1996; Festi 1998). Some of these studies show promise with the potential to assist in gallbladder stone prevention during weight loss.
How the intervention might work
Rapid weight loss and weight cycling increase the likelihood of gallbladder stone formation, in part by reduced biliary bile salt secretion and gallbladder stasis (Weinsier 1993; Tsai 2006). Dietary interventions thus may help prevent gallbladder stones through their beneficial effect on gallbladder motility or by preventing the bile from becoming lithogenic. For example, both a higher fat intake and physical activity are suggested to increase intestinal motility, which can help lower cholelithiasis risk (Gebhard 1996; Festi 1998; Storti 2005; Banim 2010).
Why it is important to do this review
Because surgical prevention and the treatment of gallbladder stones is both invasive and expensive, less-invasive alternatives to cholecystectomy would be desirable. Moreover, evidence-based guidelines are needed to identify interventions that could be feasible to use in practice, particularly in high-risk individuals, for instance, in obese individuals undergoing weight loss as cholecystectomy is sometimes advocated in people undergoing bariatric surgery for weight loss. To date, no Cochrane review has evaluated dietary interventions for the primary prevention of gallbladder stones.
To evaluate the benefits and harms of dietary interventions in the primary prevention of gallbladder stones in adults.
Criteria for considering studies for this review
Types of studies
We plan to include all randomised clinical trials irrespective of blinding, language, sample size, or publication status. We will only include controlled clinical studies where quasi-randomisation methods have been used, such as day of the week, date of birth, medical record number, and other non-randomised studies if obtained through the searches for randomised clinical trials for the reported data on harm. We will include only data from the first period of cross-over trials (as if they were parallel trials), since the carry-over effect of the intervention may be conducive to confounding. We will exclude trials where ultrasonography is not primarily used to diagnose gallbladder stones, as other diagnostic methods are lower in sensitivity and specificity (e.g., intravenous cholangiogram).
Types of participants
This review will include both hospital- and community-based men and women (older than 18 years) from all ethnicity groups who do not have gallbladder stones as confirmed by ultrasonography. We will include studies using self reported outcomes when relating specifically to cholecystectomy, as they are deemed reliable. In terms of age groups, this review will only focus on the adults. However, we will include studies conducted in children to assist with the extraction of data on harm.
Types of interventions
We will consider trials for inclusion when at least one study group has been allocated to receive a dietary intervention following a standard (within trial) protocol for prophylaxis against gallstone formation. This may include the following interventions (single or multiple per trial): calorie restriction, dietary fat, carbohydrate, protein or fibre modification, micronutrient supplementation.
Settings: we will consider both hospital- and community-based interventions.
Duration: we will set the duration of therapy to a minimum of four weeks because studies have shown gallbladder stones to develop after this time frame (e.g., Liddle 1989; Gebhard 1996). We will exclude single-meal studies as these are deemed to have a minimal impact on prevalence of gallbladder stones.
Acceptable comparator groups will include:
- inactive control intervention such as placebo or no treatment.
- active control intervention such as a variant of the same intervention or other dietary treatments.
Types of outcome measures
We will include the following primary and secondary outcome measures based on the specifications provided in the Cochrane Hepato-Biliary Group module (Gluud 2012).
- All-cause mortality.
- Morbidity (formation of ultrasonically verified gallbladder stones and symptomatic gallbladder stones). We will not include studies that do not report on gallstone formation as an outcome.
- Number and type of adverse events during or within a reasonable post-intervention time frame; for example, participant and clinician reporting of side effects, poor quality of life scores, morbidity, and mortality. The International Conference on Harmonisation (ICH) define an adverse event as any untoward medical occurrence that does not have a causal relationship with the particular treatment and can include any unfavourable and unintended sign (e.g., abnormal laboratory findings), symptoms, or disease, which is temporarily associated with the use of the respective medicinal product being assessed, regardless of whether it is related to this medicinal product (ICH-GCP 1997). The ICH describe a severe adverse event as that which results in death; is life threatening; requires hospitalisation or prolongation of existing hospitalisation; results in persistent or significant disability/incapacity or any medical event that may jeopardise the participant or require an intervention to prevent it (ICH-GCP 1997). We will not only seek evidence of adverse events from randomised clinical trials but also from open studies and case reports.
- Quality of life measures.
- Weight loss (reduction in body weight assessed in kilograms or using body mass index).
Search methods for identification of studies
We will restrict search strategies to randomised clinical trials in adults. We will identify duplicate publications by comparing publications of the same authors with respect to study populations, date, location, and follow-up time of the study. The time frame when examining the literature will include searching electronic databases from the time of their inception to present.
We will search the Cochrane Hepato-Biliary Group Controlled Trials Register (Gluud 2012), the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, and Science Citation Index Expanded. These are reported to yield approximately 95% or more of the relevant studies (Royle 2003). Consequently we may modify the search strategies as the review progresses; however, preliminary search strategies are given in Appendix 1, which encompasses a broad search taking into account studies relevant to both dietary and pharmacological interventions (for the complementary Cochrane review on pharmacological interventions to prevent primary gallbladder stones).
Searching other resources
We will scan the reference lists of relevant articles and conference proceedings for further trial references. We will contact the principal authors of identified studies and other experts (authors of reviews) to elicit information regarding additional trials known to them. We will also scan trial registries in two search portals: the US National Institutes of Health (www.clinicaltrials.gov) and the World Health Organization (WHO) International Clinical Trial Registry Platform (www.who.int/ictrp/search/en/).
Data collection and analysis
We will perform this systematic review following the instructions in the Cochrane Handbook for Systematic Reviews of Intervention (Higgins 2011), and the Cochrane Hepato-Biliary Group Module (Gluud 2012). We will perform the analysis using Review Manager 5 (RevMan 2012).
Selection of studies
Two review authors will independently review titles and abstracts in order to select potentially relevant trials. We will assess the full text of these potentially relevant trials to determine if the inclusion criteria are met. We will resolve any differences in opinion by discussion and, if necessary, with a third review author. We will list excluded studies, with reasons for their exclusion, in the 'Characteristics of excluded studies' table.
Data extraction and management
At least two review authors will independently extract the data listed below from identified randomised clinical trials (although this list is not yet exhaustive). We will contact the authors of individual trials for any unclear or missing information.
- Published, unpublished.
- Year and language of publication (if published).
- Year of study (conduct of trial).
- Duplicate publication.
- Duration (and duration of follow-up period).
- Concealment of allocation.
- Checking of blinding.
- Length of intervention.
- Main intervention: for example, low-calorie diet, diet provided, dietary manipulation.
- Co-interventions/comparison intervention (as in point 2. above).
- Control group intervention.
- Duration of follow-up.
- Population characteristics (means, standard deviations, ranges, or a combination of these for: age, sex, body mass index, health condition, etc. will be extracted, where possible).
- Diagnostic criteria used to define presence/absence of gallbladder stones.
- Inclusion and exclusion criteria.
- Sample size (total, per compared groups).
- Similarity of groups at baseline.
- Other medication/therapy used.
- Assessment of compliance.
- Withdrawals/losses to follow-up.
- Main outcomes specified in study.
- Incidence of symptomatic gallbladder stones.
- Incidence of cholecystectomy.
- Bile lithogenicity changes.
- Weight loss during the trial.
- Adverse events reported.
Results for outcomes and times of assessment (including a measure of variation) intention-to-treat/per-protocol analysis
Methodological quality/risk of bias
Assessment of risk of bias in included studies
Trials with inadequate methodological quality increase the risk of overestimating beneficial intervention effects (Gluud 2006). Therefore, review authors will independently assess the methodological quality/risk of bias in the included trials based on instructions given in the Cochrane Handbook for Systematic Reviews of Intervention (Higgins 2011), and the Cochrane Hepato-Biliary Group Module (Gluud 2012). The process will be conducted without masking the trial names. We will seek unavailable information that is deemed necessary from the authors of the relevant trials.
The following domains will be assessed to determine the extent of systematic error and the risk of bias for each trial (Schulz 1995; Moher 1998; Kjaergard 2001; Wood 2008; Lundh 2012; Savović 2012a; Savović 2012b), and will be presented under the 'Characteristics of included studies'.
Allocation sequence generation
- Low risk of bias: sequence generation was achieved using computer random number generation or a random number table. Drawing lots, tossing a coin, shuffling cards, and throwing dice were adequate if performed by an independent person not otherwise involved in the trial.
- Uncertain risk of bias: the method of sequence generation was not specified.
- High risk of bias: the sequence generation method was not random. Quasi-randomised studies, those using dates, names, or admittance numbers in order to allocate participants were inadequate and will be excluded for the assessment of benefits but not for harms.
- Low risk of bias: the participant allocations could not have been foreseen in advance of, or during, enrolment. Allocation was controlled by a central and independent randomisation unit. The allocation sequence was unknown to the investigators (e.g., if the allocation sequence was hidden in sequentially numbered, opaque, and sealed envelopes).
- Uncertain risk of bias: the method used to conceal the allocation was not described so that intervention allocations may have been foreseen in advance of, or during, enrolment.
- High risk of bias: the allocation sequence was likely to be known to the investigators who assigned the participants.
Blinding of participants, personnel, and outcome assessors
- Low risk of bias: blinding was performed adequately, or the assessment of outcomes was not likely to be influenced by lack of blinding.
- Uncertain risk of bias: there was insufficient information to assess whether blinding was likely to induce bias on the results.
- High risk of bias: no blinding or incomplete blinding, and the assessment of outcomes was likely to be influenced by lack of blinding.
Incomplete outcome data
- Low risk of bias: missing data were unlikely to make treatment effects depart from plausible values. Sufficient methods, such as multiple imputation, were employed to handle missing data.
- Uncertain risk of bias: there was insufficient information to assess whether missing data in combination with the method used to handle missing data were likely to induce bias on the results.
- High risk of bias: the results were likely to be biased due to missing data.
Selective outcome reporting
- Low risk of bias: all outcomes were pre-defined and reported, or all clinically relevant and reasonably expected outcomes were reported.
- Uncertain risk of bias: it is unclear whether all pre-defined and clinically relevant and reasonably expected outcomes were reported.
- High risk of bias: one or more clinically relevant and reasonably expected outcomes were not reported, and data on these outcomes were likely to have been recorded.
For a trial to be assessed with low risk of bias in the selective outcome reporting domain, the trial should have been registered on the www.clinicaltrials.gov web site or a similar register, or there should be a protocol (e.g., published in a paper journal). In the case when the trial was run and published in the years when trial registration was not required, we will carefully scrutinise all publications reporting on the trial to identify the trial objectives and outcomes. If usable data on all outcomes specified in the trial objectives were provided in the publications results section, then we will consider the trial a low risk of bias trial in the 'Selective outcome reporting' domain.
- Low risk of bias: the trial appeared to be free of industry sponsorship or other type of for-profit support that may manipulate the trial design, conductance, or results of the trial.
- Uncertain risk of bias: the trial may or may not be free of for-profit bias as no information on clinical trial support or sponsorship was provided.
- High risk of bias: the trial was sponsored by the industry or had received other type of for-profit support.
Other risk of bias
- Low risk of bias: the trial appeared to be free of other components (e.g., academic bias) that could put it at risk of bias.
- Uncertain risk of bias: the trial may or may not be free of other components that could put it at risk of bias.
- High risk of bias: there were other factors in the trial that could put it at risk of bias (e.g., authors have conducted trials on the same topic, etc.).
We will consider trials assessed as having 'low risk of bias' in all of the above specified individual domains as trials with 'low risk of bias'. We will consider trials assessed as having 'uncertain risk of bias' or 'high risk of bias' in one or more of the specified individual domains as trials with 'high risk of bias'. If disagreements among review authors' evaluation occur, we will resolve them by discussion.
Measures of treatment effect
Where study outcomes are reported as dichotomous data (as opposed to continuous variables), we will present odds ratios (OR) or risk ratios (RR), 95% confidence intervals (CI), and P values for significance level. We may also include the risk difference, if this is presented, in order to illustrate the above results in absolute terms (e.g., number of participants who develop gallbladder stones), which may then be converted to provide a number needed to treat for an additional beneficial outcome (NNTB) estimate.
Unit of analysis issues
The risk of unit of analysis related errors in the meta-analysis increases when participants undergo more than one intervention in trials. If possible, we will combine groups from a single trial to create a single pair-wise comparison, as recommended by the Cochrane Handbook for Systematic Reviews of Intervention (Higgins 2011). Alternatively, we may apply a Bayesian framework so that all interventions in the analysis can be ranked using probabilistic methods.
Moreover, there may be multiple observations for the same outcome across studies (e.g., the follow-up assessments for presence of gallbladder stones at different time points may require that multiple ultrasonographies are performed for instance, at six months, one year, and 18 months). In this case, we may need to compute an effect measure for each individual participant, which incorporates all time points (e.g., trend over time).
Timing of outcome measurements being assessed post intervention may differ between studies. For studies with differing time frames (e.g., for follow-up), the following intervals will be used: short-term (three months or less than three months), medium-term (four to 11 months) and long-term (12 months or more than 12 months).
Dealing with missing data
We will seek data on all participants randomised to allow intention-to-treat analyses including all participants irrespective of compliance or follow-up and we will perform available-case analysis. Where data on the outcome of excluded participants are missing, we will contact trial authors for the original data. We will collect dropout rates together with reasons for drop-out, as reported by trial authors. We will document this information in the 'Characteristics of included studies' table. The analyses will further evaluate the importance of: missing data by performing the following sensitivity analyses: poor outcome analyses assuming that losses to follow-up are treatment failures and good outcome analyses assuming that losses to follow-up are treatment successes.
Assessment of heterogeneity
Assessment of reporting biases
Where possible, we will perform a funnel plot using Review Manager 5 to investigate for presence of bias by illustrating the study size against the treatment effect (RevMan 2012). We will use linear regression approach to determine funnel plot asymmetry, as described by Egger 1997.
We plan to perform the meta-analysis and display the results according to The Cochrane Collaboration recommendations (Higgins 2011). We will group data based on similarities of interventions implemented. We will present dichotomous data (e.g., presence of gallbladder stones) using pooled OR or RR with 95% CI, with their corresponding log transformed values. For continuous data, we will use the mean difference and 95% CI to estimate intervention effect (or the standardised mean difference, if diverse scales are used). We will derive precise P values for all comparisons where possible. We will perform analyses, wherever possible, using the intention-to-treat principle (e.g., for all randomised patients in the randomised trials); otherwise, we will conduct an 'available case analysis'.
If we identify a small number of trials, or if sample sizes are small, we will consider using the Mantel-Haenszel method for pooling data (Mantel 1959). The Mantel-Haenszel method assumes a fixed-effect meta-analysis and is used for dichotomous data. It is reported by the Cochrane Handbook for Systematic Reviews of Intervention as the best method for combining trials with small sample sizes (Higgins 2011). For continuous data, we will use standardised or mean differences to pool results. In the absence of heterogeneity between study results, we will synthesise (where similarities exist) and analyse data using the fixed-effect model meta-analysis (DeMets 1987). Alternatively, we will apply the random-effects model in the presence of heterogeneous results, and perform subgroup/sensitivity analyses to assess for potential causes (DerSimonian 1986). The review authors will consider conducting analyses with both models (fixed-effect and random-effects) to determine whether there is a discrepancy between their results.
Trial sequential analysis
Cumulative meta-analyses are at risk of producing random errors due to sparse data and multiple testing of accumulating data (Wetterslev 2008; Brok 2009; Thorlund 2009); therefore, trial sequential analysis can be applied as specified in the Cochrane Hepato-Biliary Group module to assess this risk (CTU 2011; Thorlund 2011). The required information size (i.e., the number of participants needed in a meta-analysis to detect or reject a certain intervention effect) can be calculated in order to minimise random errors (Brok 2008; Wetterslev 2008; Brok 2009; Thorlund 2009; Wetterslev 2009; Thorlund 2010). The required information size takes into account: the event proportion in the control group; assumption of a plausible relative risk reduction, or the relative risk reduction observed in the included trials with low risk of bias and the assumed heterogeneity or diversity of the meta-analysis (Brok 2008; Wetterslev 2008; Brok 2009; Thorlund 2009; Wetterslev 2009; Thorlund 2010). Trial sequential analysis enables testing for significance to be conducted each time a new trial is included in the meta-analysis. On the basis of the required information size, trial sequential monitoring boundaries can be constructed. This enables one to determine statistical inference concerning cumulative meta-analysis that has not yet reached the required information size (Wetterslev 2008). Firm evidence may be established if the trial sequential monitoring boundary is crossed before reaching the required information size, in which case further trials may be superfluous. In contrast, if the boundary is not surpassed, one may conclude that it is necessary to continue with further trials before a certain intervention effect can be detected or rejected. Firm evidence for lack of a postulated intervention effect can also be assessed with trial sequential analysis. This occurs when the cumulative Z-score crossed the trial sequential beta spending monitoring boundaries and enters the area of futility.
Subgroup analysis and investigation of heterogeneity
We will perform subgroup analysis, where appropriate, to assess for possible sources of heterogeneity. Specifically, if enough trials are found to justify subgroup analysis, we will use it to identify differences or similarities among results of trials that have common features; for example, trials could be stratified by duration of therapy or by participant subgroup. Moreover, we may carry out subgroup analyses to help answer specific questions about particular participant groups, types of intervention, or types of studies.
The following subgroups could be investigated further for effect modification.
- Intervention type: dietary (in combination or alone, differing doses, etc.).
- Intervention type based on participants receiving bariatric surgery versus those not receiving bariatric surgery.
- Intervention type based on participants actively trying to lose weight versus those not trying to lose weight.
- Outcome of symptomatic gallbladder stones or cholecystectomy, or both versus asymptomatic gallbladder stones.
- Quality of bias control (e.g., high compared to low).
- Treatment duration/duration of follow-up.
- Sex - gallbladder stones are more prevalent among females (Portincasa 2006).
- Different geographical locations or ethnicity, or both due to the fact that both ethnic and geographical disparities exist regarding the prevalence of gallbladder stones.
If we identify a sufficient number of randomised trials, we will perform sensitivity analysis to explore the impact of study quality. Specifically, if appropriate (e.g., where meta-analysis has shown heterogeneity among trial results), we will conduct a sensitivity analysis to explore the influence of the following factors on effect size.
- Publication status (e.g., excluding unpublished studies).
- Methodological quality and risk of bias (e.g., excluding studies of low quality, i.e., with missing dropout rates).
- Blinding (e.g., removal of unclear or inadequate).
- Allocation concealment (e.g., removal of unclear or inadequate allocation procedure).
- Duration (e.g., impact of short and long intervention duration).
- Language of publication.
Though a full economic evaluation in this review is predicted to be unfeasible, we will discuss the cost and cost-effectiveness of the interventions, and if data availability permits, a preliminary reflection of efficient use of resources with respect to the efficacy of interventions, for example, quality of life (as assessed by a validated questionnaire), and length of hospital stays.
'Summary of findings' tables
If possible, we will also assess the quality of evidence across trials at the outcome level using the GRADE approach; we will include the three primary outcomes in the 'Summary of findings' table (GRADE table) plus two of the secondary outcomes: treatment with cholecystectomy and bile lithogenicity (ims.cochrane.org/revman/other-resources/gradepro).
To Sarah Klingenberg, Trial Search Co-ordinator, for preliminary searching the electronic databases and to the Cochrane Hepato-Biliary Group for their support.
Peer reviewers: Hanns-Ulrich Marschall, Sweden; Marc Besselink, the Netherlands.
Contact editor: Vanja Giljaca, Croatia.
Appendix 1. Search strategies for identification of studies
Contributions of authors
C Stokes is the author and drafted the protocol, M Casper provided input and is the co-author of the protocol. F Lammert and LL Gluud provided guidance and are the co-authors of the protocol. All authors agreed to the published version.
Declarations of interest
Sources of support
- None Known, Other.
- None Known, Other.
Due to the complexity and the heterogeneity of the interventions, we decided to split the published protocol 'Non-pharmacological and pharmacological interventions for primary prevention of gallbladder stones in adults' (Stokes 2012) into two separate protocols for systematic reviews, i.e., "Diets for primary prevention of gallbladder stones in adults" and "Pharmacological interventions for the primary prevention of gallbladder stones in adults".