Chinese medicinal herbs for cholelithiasis

Abstract

Background

Cholelithiasis is a common disease of the biliary tract. Chinese medicinal herbs are being used widely as an alternative treatment in people with cholelithiasis, but their beneficial or harmful effects have not been assessed systematically.

Objectives

To assess the beneficial and harmful effects of Chinese medicinal herbs in people with cholelithiasis.

Search methods

We conducted searches in the Cochrane Hepato‐Biliary Group Controlled Trials Register, the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library, MEDLINE, EMBASE, Science Citation Index Expanded, Chinese Medicine Conference Disc, and Chinese Bio‐Medicine Disc to January 2013. We handsearched four Chinese journals. No language or year of publication restrictions were applied.

Selection criteria

Randomised clinical trials studying Chinese medicinal herbs for treatment of cholelithiasis.

Data collection and analysis

Two review authors (SJ, TG) independently extracted data. For dichotomous data, we estimated the risk ratio (RR), and for continuous data, we calculated the mean difference. We also calculated 95% confidence intervals (CI).

Main results

Eleven randomised trials with 1205 participants with asymptomatic or mild‐to‐moderate cholelithiasis were included. None of the randomised clinical trials compared a single Chinese medicinal herb with a Western medicine or with surgery. No placebo‐controlled trials were identified. In the trials comparing one Chinese herbal medicine (Gandanxiaoshi tablet) versus another (Aihuodantong tablet), there was no significant difference in the improvement of upper abdominal pain after the end of treatment (RR 1.21; 95% CI 0.71 to 2.05), and the heterogeneity among trials was not substantial. No other outcomes could be assessed. The remaining trials of Chinese medicinal herbs (Qingdan capsule, Danshu capsule, Paishi capsule, Rongdanpaishi capsule), did not offer specific data on symptoms, signs, or change in gallstones that would permit assessment of significant differences in curative effects between the treatment and control groups. No serious adverse events were reported.

Authors' conclusions

This review reveals no strong evidence that the analysed Chinese medicinal herbs have any beneficial effects on asymptomatic or mild‐to‐moderate cholelithiasis. Definitive conclusions will require much better designed randomised trials to reduce risk of bias and allow detailed assessment of clinical outcomes.

Plain language summary

Chinese medicinal herbs for cholelithiasis

Cholelithiasis (gallstone disease) causes significant morbidity, severe complications, and economic burden worldwide. Though current surgical therapies are effective for gallstones, some types of gallstones (e.g., calcium bilirubinate calculi) cannot easily be eradicated by surgery and they often recur in the bile duct system, putting people at high risk of severe complications. The objective of this review was to evaluate the benefits and harms of Chinese medicinal herbs for people with cholelithiasis. Though some Chinese medicinal herbs appear to be safe for people with asymptomatic, mild, or moderate disease, they have not been conclusively shown to have curative effects on gallstones due to the low methodological quality (high risk of bias) of the included trials. Thus, randomised clinical trials with low risk of bias should be conducted to assess the effects of Chinese medicinal herbs before they can be used widely in the clinic.

Background

Cholelithiasis, which refers to the presence of gallstones in the biliary tract, is a relatively common disease, with an incidence mean rate of 9% to 15% in adults worldwide (Shipeng 2000). The frequency of gallstone occurrence differs greatly across continents and countries, varying from 6% to 70% (Jorgensen 2000). The disease is caused by multiple risk factors, including increasing age, ethnic and hereditary factors, female sex, multiple pregnancies, obesity, diabetes, liver cirrhosis, long‐term intravenous nutrition (total parenteral nutrition), and certain types of operations for peptic ulcers (Mendez‐Sanchez 1991). In people with a mild form of the disease, no symptoms can be observed (silent stones), but gallstones may be found in the biliary system using such imaging methods as abdominal ultrasound, cholecystography, and computed tomography (Yirong 1994). In symptomatic patients, cholelithiasis can induce biliary colic, bile duct infection, cholecystitis, or even suppurative cholangitis and pancreatitis (Wu 2000). Severe complications of gallstone disease, such as suppurative cholangitis or pancreatitis, always occur due to migration of the gallstone into the common bile duct from the gallbladder. These stones may enter the common bile duct and cause occlusion of the lumen, causing obstructive jaundice; pancreatitis due to obstruction of the pancreatic ducts; or ascending cholangitis as a result of diminished bile flow (Veedfald 2011). These complications are the main causes of death related to cholelithiasis (Jieping 2000). Gallstone disease is usually chronic, and symptomatic gallstone disease adversely affects quality of life. Furthermore, gallstone disease is associated with high medical costs, and causes an economic burden in the form of healthcare expenses.

Currently cholecystectomy, predominantly performed via laparoscopy, is the standard treatment for symptomatic cholecystolithiasis (Keus 2010). It is estimated that more than 500,000 cholecystectomies are carried out annually for symptomatic gallstone disease in the USA (Fischer 2004). If the gallstone in the biliary system must be cleared, choledochoscopy and endoscopic papillotomy are invasive therapeutic options (Ronglai 2001). However, some types of gallstone recur easily and are quite difficult to eradicate in a single surgical or non‐surgical procedure (Nakavama 1980; Ha 2011); one such type of gallstone is calcium bilirubinate calculus, a sandy or massive stone in the bile duct caused by biliary tract infection or parasite infestation (Xiaosi 1994; Sackmann 1997). People with these types of gallstones can be a challenge to treat because the people are reluctant to undergo surgery, or because the presence of comorbidities can preclude the multiple surgeries needed to resolve the condition.

Several non‐invasive and pharmacological treatment options have been used as an alternative to surgery. These include lithotripsy (Neuhaus 1998), bile acid administration (chenodeoxycholic acid, ursodeoxycholic acid) (Veedfald 2011), disodium edetate, or methyl tert‐butyl ether (Patino 1998).

Chinese medicinal herbs are commonly defined as a series of therapeutic plants that are compatible with one another, which means that some are used as the main therapeutic ingredients for the disease, some are used as accessorial intensifiers for therapeutic ingredients, and others are used as neutral ingredients to control side effects. When an environmental factor is involved in the occurrence of a disease, that factor can interact with an underlying imbalance in the body and thereby lead to the disease. Thus, a physician will treat the underlying imbalance, not only the disease itself. Herbal medicines are abundant tools for re‐establishing the underlying balance, thereby restoring health to the body. Several medicinal herbs have been commonly held to be effective against cholelithiasis (Ziliang 2001). These herbs include Lysimachia christinae Hance, Scutellaria baicalensis Georgi, Citrus aurantium L, Artemisia capillaris Thunb, Forsythia suspensa (Thunb) Vahl, Lonicera japonica Thunb, and Crataegus pinnatifida Bge.var.major N.E.Br.

Chinese medicinal herbs are used to treat cholelithiasis in China (Hong 1997). Some clinical trials have been performed to assess the benefits or harms of these herbs on people with gallstones (Zhang 2001b; Binghui 2002). However, we did not find any Chinese systematic reviews until 2010 (Chen 2010); the review provided only an incomplete appraisal of methodological quality in the included trials, preventing definitive conclusions.

Objectives

To assess the beneficial and harmful effects of Chinese medicinal herbs for cholelithiasis when compared with placebo, no intervention, another therapeutic agent (e.g., ursodeoxycholic acid), or surgical treatment.

Methods

Criteria for considering studies for this review

Types of studies

All parallel‐group randomised clinical trials and the first phase of randomised cross‐over trials testing Chinese medicinal herbs for treatment of cholelithiasis. The trials could be published in any language.

Types of participants

People with cholelithiasis, confirmed by cholecystic or biliary operation, or clinically diagnosed by radiological imaging, including abdominal ultrasound, cholecystography, computed tomography, or other techniques. Cholelithiasis could be defined based on the presence of any cholecystic symptoms according to the criteria in the books Practical Internal Medicine (Lin 1993), Chinese Traditional Medicine Trial and Instruction Principle in Clinic (Wang 1993), and Chinese Traditional Sign and Diagnosis (Deng 1987). Diagnostic standards included the following definitions: 1) plus 2), or 1) plus 3), or 1) plus 2) plus 3):

1. A typical history of cholelithiasis, with or without positive physical signs.

2. Positive results of certain auxiliary examinations that can confirm the presence of gallstones in the biliary system (Downs 1996). These examinations include B‐mode ultrasonography (B‐ultrasound), computed tomography, magnetic resonance imaging, percutaneous transhepatic cholangiography, and endoscopic retrograde cholangiopancreatography/magnetic resonance cholangiopancreatography (Downs 1996).

3. Gallstones in the gallbladder or the biliary ducts, or both as confirmed by a surgical procedure.

Types of interventions

Chinese medicinal herbs (a single medicinal herb, i.e., patent herbal medicines or a medicinal herbal compound, i.e., self‐made herbal compounds) were compared with placebo or no treatment, ursodeoxycholic acid, or another Western medicine (a product made by chemical synthesis or extraction in a pharmaceutical factory), and surgery.

Trials with Chinese medicinal herbs combined with other Western medicine (the commonly used medicines included in Cecil Textbook of Medicine (Cecil 1996) andSabiston Textbook of Surgery (Sabiston 2012)) versus the same type of Western medicine in the intervention group were also considered for the review.

Types of outcome measures

We planned to evaluate the following outcomes.

Primary outcomes

1. Mortality.
 2. Number of people without relief of symptoms (i.e., upper abdominal pain, biliary colic, nausea, and vomiting or icterus).
 3. Number of people without reduction, excretion, or dissolution of gallstones.
 4. Adverse events defined as any untoward medical occurrence not necessarily having a causal relationship with the treatment but resulting in a dose reduction or discontinuation of treatment (ICH‐GCP 1997).

Severe adverse events are defined as any event that would increase mortality; is life‐threatening; requires inpatient hospitalisation; results in a persistent or significant disability; or any important medical event that may jeopardise the person or require intervention to prevent it. All other adverse events are considered non‐serious.

Non‐serious adverse events are defined as symptoms or abnormality of biochemical results caused by Chinese medicinal herbs at any dose:

• Vertigo or nausea.

• Abdominal pain/biliary colic.

• Diarrhoea.

• Other symptoms, such as rash, bitter taste, abdominal distension, constipation, etc.

• Abnormality of liver function or kidney function, or other biochemical results.

These adverse effects could disappear by the expectant treatment, or discontinuation of the therapies, or by reduction in the dosage of the Chinese medicinal herbs.

Secondary outcomes

5. The number of people undergoing biliary surgical interventions.
 6. Quality of life.
 7. Patients' expenditures for hospitalisation, medication, and medical examination. (In China, people may be responsible for paying some or all treatment costs depending on the type of hospitalisation, medication, and medical examination.)

Search methods for identification of studies

Electronic searches

We searched The Cochrane Hepato‐Biliary Group Controlled Trials Register (Gluud 2013), the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library, MEDLINE, EMBASE, Science Citation Index Expanded, Chinese Medicine Conference Disc (CMCD), and Chinese Bio‐Medicine Disc (CBMD) (Royle 2003). The search strategies with the time spans of the searches are given in Appendix 1.

Searching other resources

We also searched the following journals for the relevant trials from the first publication date onwards: Chinese Journal of Integrated Traditional and Western Medicine of Zhejiang (1991 to November 2012), Chinese Journal of Beijing University of Traditional Chinese Medicine (1980 to November 2012), Chinese Surgical Journal of Integrated Traditional and Western Medicine (1994 to November 2012), and Chinese Journal of Integrated Traditional, and Western Medicine (1980 to November 2012). Conference proceedings not published in journals, such as Chinese hepatobiliary conference proceedings, were also searched.

Further trials were sought by scanning of the reference lists of the relevant articles. In addition, the authors of the trials and the pharmaceutical companies producing the drugs involved in the review were contacted for information on additional published or unpublished randomised clinical trials, but no responses were received.

Data collection and analysis

We performed the review and meta‐analyses following the recommendations of The Cochrane Collaboration (Higgins 2011) and The Cochrane Hepato‐Biliary Group Module (Gluud 2013). The analyses were performed using Review Manager 5.2 (RevMan 2012).

Selection of studies

Two review authors (SJ, TG) independently selected the trials to be included in the review according to the above‐mentioned eligibility criteria. Disagreements were resolved by discussion. In case of disagreement, a third review author (YW) was consulted. Decisions on included trials or excluded studies were based on the entire text of the publication whenever available.

Data extraction and management

Data extraction was conducted up to and including November 2012 by two review authors (SJ, JC). Data were extracted from the included trials using standardised extraction sheets: study type and methodology; number and description of participants; details of type, dosage, and time schedule/duration of intervention; and type and measurement method of outcomes.

The authors of the trials were queried to specify any data not sufficiently reported in the publication. The contents of data extraction included information on participants, study design, characteristics of intervention formulae (CMS) and comparators, any adverse effects, and baseline data (Characteristics of included studies).

Assessment of risk of bias in included studies

Methodological quality is defined as the confidence that the trial design and report will restrict bias in the intervention comparison (Moher 1998). According to empirical evidence (Schulz 1995; Moher 1998; Kjaergard 2001; Wood 2008; Lundh 2012; Savovic 2012a; Savovic 2012b), bias risk assessment was achieved through the following domains:

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 adjudicator.

• Uncertain risk of bias: the trial was described as randomised, but the method of sequence generation was not specified.

• High risk of bias: the sequence generation method was not, or may not have been, random. Quasi‐randomised studies, those using dates, names, or admission numbers in order to allocate patients, are inadequate and were excluded for the assessment of benefits but not for assessing harms.

Allocation concealment

• Low risk of bias: allocation was controlled by a central and independent randomisation unit; sequentially numbered, opaque and sealed envelopes; or similar so that intervention allocations could not have been foreseen in advance of or during enrolment.

• Uncertain risk of bias: the trial was described as randomised, but 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: if the allocation sequence was known to the investigators who assigned participants or if the study was quasi‐randomised. Quasi‐randomised studies were excluded for the assessment of benefits but not for assessing harms.

Blinding of participants, personnel, and outcome assessors

• Low risk of bias: blinding was performed adequately, or the outcome measurement was not likely to be influenced by lack of blinding.

• Uncertain risk of bias: there was insufficient information to assess whether the type of blinding used was likely to induce bias on the estimate of effect.

• High risk of bias: no blinding or incomplete blinding, and the outcome or the outcome measurement was likely to be influenced by lack of blinding.

Incomplete outcome data

• Low risk of bias: the underlying reasons for missing data were unlikely to make treatment effects depart from plausible values, or proper methods were employed to handle missing data.

• Uncertain risk of bias: there was insufficient information to assess whether the missing data mechanism in combination with the method used to handle missing data were likely to introduce bias in the estimate of effect.

• High risk of bias: the crude estimate of effects (e.g., complete case estimate) was clearly biased due to the underlying reasons for missing data, and the methods used to handle missing data were unsatisfactory.

Selective outcome reporting

• Low risk of bias: pre‐defined or clinically relevant and reasonably expected outcomes were reported on.

• Uncertain risk of bias: not all pre‐defined or clinically relevant and reasonably expected outcomes were reported on, or were not reported fully, or it was unclear whether data on these outcomes were recorded or not.

• High risk of bias: one or more clinically relevant and reasonably expected outcomes were not reported on; data on these outcomes were likely to have been recorded.

The clinically relevant and reasonably expected outcomes include the data of mortality and morbidity, symptoms, physical signs, and image results of reduction, excretion, or dissolution of gallstones, or other results which the original investigators intended to assess.

Other bias

• Low risk of bias: the trial appeared to be free of other factors that could put it at risk of bias. 

• Uncertain risk of bias: the trial might or might not have been free of other factors 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.

Other bias include profit involvement, or academic bias (i.e., authors have conducted trials on the same topic, etc.).

Trials assessed as having 'low risk of bias' in all of the specified individual domains were considered trials with 'low risk of bias'. Trials assessed as having 'uncertain risk of bias' or 'high risk of bias' in one or more of the specified domains were considered trials with 'high risk of bias'.

Measures of treatment effect

For binary outcomes, such as the presence or absence of symptoms, the impact of the intervention was expressed as a risk ratio (RR) together with 95% confidence intervals (CI). For continuous outcomes, means and standard deviations (SD) were used to summarise the values in each group. The mean differences or the standardised mean differences for continuous outcomes were used, depending on whether the continuous data used the same scale or not.

Unit of analysis issues

Participants in the randomised trials. We planned to use participants' data from the first period of the randomised cross‐over trial if such trials were identified.

Dealing with missing data

The intention‐to‐treat analysis was performed for dichotomous and continuous outcomes if complete data were available. If complete data were not available at the end of the trial, for dichotomous data, a 'worst‐case' scenario was employed in which participants with missing or incomplete data were considered to be treatment failures in both groups; for continuous data, a similar scenario was also employed in which the baseline data at the time of group allocation were used as the final data for analysis.

Assessment of heterogeneity

We tested statistical heterogeneity using the Chi² test with a P value < 0.1 set as the cut‐off (Higgins 2002), if these existed.

Assessment of reporting biases

Whenever possible (i.e., if there were more than 10 included trials), we used a regression analysis to assess funnel plot asymmetry (Egger 1997).

Data synthesis

We used the Cochrane Review Manager 5.2 software (RevMan 2012) for meta‐analyses. All meta‐analyses were performed using both the random‐effects and fixed‐effect models. If the two models gave different results, both sets of results were reported; if the two models gave similar results, only the results of the fixed‐effect model were reported.

We planned not to carry out meta‐analysis of trials involving completely different types of interventions. The trials comparing herbs versus herbs were to be analysed separately from trials comparing herbs versus no treatment (with or without placebo). We planned to perform descriptive analysis using the Chi² test (P value < 0.05) to compare dichotomous outcomes between two treatment groups. If the theoretical frequency was < 1 or if fewer than 40 participants were involved in the comparison during the Chi² test, Fisher's exact test would then be used for further evaluation (Cochran 1954). Under theoretical frequency, it is to be understood a distributional frequency that would result if the data followed a theoretical distribution law rather than the actual observed frequencies. If we had only one trial for each comparison, Students' t‐test (P value < 0.05), as implemented in SPSS version 13.0, would be applied to continuous data in order to assess the efficacy of Chinese medicinal herbs to treat cholelithiasis.

Subgroup analysis and investigation of heterogeneity

We planned to compare the trials with low risk of bias versus those with high risk of bias for the primary outcome measures. We had also planned to perform further subgroup analyses (if P value < 0.1) to explore the association between dose and duration of experimental interventions and effect as well as the association between disease severity at entry and effect.

Results

Description of studies

Our search in January 2013 identified 336 articles. Seven publications came from the search in The Cochrane Hepato‐Biliary Group Controlled Trials Register, 48 from the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library, 26 from MEDLINE, 55 from EMBASE, 25 from Science Citation Index Expanded, 133 from CBMD, 29 from CMCD, and 13 from handsearching. Duplicate publications were excluded, leaving 133 publications. Of the 133 publications, we excluded 122 after reading the abstracts or full text, because they were not focused on people with cholelithiasis or on the objectives of this review. The reasons for exclusion are listed in the Characteristics of excluded studies table. The remaining 11 publications of randomised controlled trials (RCTs) were chosen for further assessment.

Trial design

All 11 publications described randomised trials with parallel design, involving 1205 people with asymptomatic, mild, or moderate cholelithiasis. No quasi‐randomised trials and cohort studies were identified for detailed assessment of harm. Trials comparing herbs versus placebo or no treatment were not identified, and no cross‐over trials were found. All 11 trials were performed in China. Five trials compared Chinese medicinal herbs versus other Chinese medicinal herbs (Dai 2000;Zhang 2001a;Zhang 2001b; Deng 2005;Su 2012), and the others compared Chinese medicinal herbs (or Chinese medicinal herbs plus Western medicines) versus Western medicines.

Sample size

Thirty‐four participants took part in the trial of herbs plus one antibiotic versus the same antibiotic (She 1981), while 60 to 200 participants (mean 110) took part in the other trials comparing Chinese medicinal herbs versus different types of herbs or Western medicines.

Setting

Four trials included outpatients (Zhao 2004; Xing 2009; Shi 2012; Su 2012). Although another six trials did not define the status of the patients (Zhu 1997; Dai 2000; Zhang 2001a; Zhang 2001b; Deng 2005; Huang 2005), we assume that they were outpatients based on the patient's condition and treatment duration of at least eight weeks. The last trial (She 1981), which compared Chinese medicinal herbs plus one antibiotic versus the same antibiotic alone, did not specify clearly the status of the patients, but we assume them to be inpatients since patients with cholecystitis requiring intravenous infusion are usually admitted as inpatients in Chinese hospitals.

Participants

Ten trials included adults only (She 1981; Zhu 1997; Dai 2000; Zhang 2001a; Zhang 2001b; Deng 2005; Huang 2005; Xing 2009; Shi 2012; Su 2012), and one trial included adult, children (the youngest was eight years old), and teenagers (Zhao 2004), but the ratio of the children and teenage patients in the two groups was not reported. The male to female ratio was approximately 1:2 in eight trials (275 men/520 women) (She 1981; Dai 2000; Zhang 2001a; Zhao 2004; Deng 2005; Huang 2005; Xing 2009; Shi 2012), 1.6:1 in one trial (Su 2012), and undefined in the remaining two trials (Zhu 1997; Zhang 2001b). The age of the participants in ten of the trials ranged from 8 to 91 years old (She 1981; Zhu 1997; Dai 2000; Zhang 2001a; Zhang 2001b; Zhao 2004; Deng 2005; Xing 2009; Shi 2012; Su 2012). The trial by Huang 2005 did not explicitly report age. The average age range in seven trials varied between 42 and 51 (Zhang 2001a; Zhao 2004; Deng 2005; Huang 2005; Xing 2009; Shi 2012; Su 2012). The trials of She 1981; Zhu 1997; Dai 2000; and Zhang 2001b did not provide the mean age of the participants.

Interventions

The Chinese medicinal herbs used in the intervention group of the 11 trials were either patent herbal medicines (i.e., proprietary formulations made by a pharmaceutical company) or herbal compounds prepared in a pharmacy at the discretion of the pharmacist. As a primary supportive treatment for symptoms, such as upper abdominal pain, biliary colic, nausea, and vomiting, trial participants in three trials received an antibiotic, spasmolytics, or analgesics. An antibiotic was given in the trial comparing herbs plus antibiotic versus the same antibiotic alone (She 1981), and the spasmolytics as well as analgesics were given in two trials (She 1981; Xing 2009). From the publication of the She 1981 trial, it is clear that the same antibiotic was administered to the control and intervention groups, but the identity and dosage were not provided (Table 1).

1. Administration of Chinese medical herbs (herbs plus Western medicine (an antibiotic) versus the same Western medicine (an antibiotic).

Number of trials	Herbs with Western medicine in the intervention group	Route of administration in the intervention group	Period of administration in the intervention group	Other herbs with Western medicine in the control group	Route of administration in the control group	Period of administration in the control group	Randomised trial and method of blinding	Observation of follow‐up
She 1981	Cai hu cheng qi decoction without specific dosages (Radix Bupleuri, Natrii Sulphas, Radix Et Rhizoma Rhei, Sterculis euosma W.W.Smith, Herba Lysimachiae, Herba Artemisiae Capillaris, Fructus Aurantii, Radix Aucklandiae, Radix Linderae, Radix Paeoniae Alba, Herba Taraxaci, Fructus Meliae Toosendan, Endothelium Corneum Gigeriae Galli, Radix Curcumae) (self‐produced herbal compounds) + transfusion and antibiotics without specific dosages and medicine	Taken orally	The period of treatment was unclear	Transfusion and antibiotics without specific dosages and medicine (same with the intervention group)	Intravenous drip	The period of treatment is unclear	Randomised trial without blinding	No follow‐up

Five trials compared Chinese medicinal herbs (538 people) with three types of Chinese medicines: Chinese medicinal herbs (Tongfupaishizhitong decoction) versus medicinal herbs (Anetholi trithionum) in one trial (Deng 2005); Chinese medicinal herbs (Paishi capsule) versus Chinese medicinal herbs (Lidanpaishi capsule) in one trial (Su 2012); and Chinese medicinal herbs (Gandanxiaoshi tablet) versus medicinal herbs (Aihuodantong capsule) in three trials (Dai 2000; Zhang 2001a; Zhang 2001b). Five trials compared Chinese medicinal herbs (633 people) versus two types of Western medicines: Chinese medicinal herbs versus ursodeoxycholic acid in four trials (Zhu 1997; Zhao 2004; Huang 2005; Shi 2012); and Chinese medicinal herbs versus chenodeoxycholic acid in one trial (Xing 2009) (Table 2; Table 3). One trial compared Chinese medicinal herbs plus Western medicine (antibiotic) (34 people) versus the same Western medicine (antibiotic) (She 1981) (Table 1). None of the included trials compared a single Chinese medicinal herb versus Western medicine or versus surgery.

2. Administration of Chinese medical herbs (herbs versus other herbs or Western medicine).

Number of trials	Herbs in regimen in the intervention group	Route of administration in the intervention group	Period of administration in the intervention group	Western medicine in the control group	Route of administration in the control group	Period of administration in the control group	Randomised clinical trial and method of blinding	Observation of follow‐up
Huang 2005	Danshu capsule without specific dosages and herbs (patent herbal medicine)	Oral	3 months	Ursodesoxycholic acid	0.2 g t.i.d.	3 months	Randomised without blinding	No follow‐up
Zhao 2004	Rongpaidanshi capsule (Pericarpium Citri Reticulatae Viride 10 g, Pericarpium Citri Reticulatae 10 g, Radix Aucklandiae 5 g, Fructus Aurantii Immaturus 5 g, Radix Curcumae 12 g, Radix Paeoniae Alba 10 g, Endothelium Corneum Gigeriae Galli 20 g, Alumen 10 g, Nitrum 21 g, Radix Et Rhizoma Rhei 10 g, Fructus Crataegi 5 g, Radix Scrophulariae 10 g, Radix Glycyrrhizae 5 g, Herba Artemisiae Capillaris 10 g, Flos Lonicerae 10 g, Herba Lysimachiae 10 g, Radix Codonopsis Pilosulae 10 g, Radix Angelicae Sinensis 10 g, Evodia rutaecarpa (Juss.) Benth 3 g, Radix Scutellariae 6 g) (self produced herbal compounds)	Oral	6 months	Ursodesoxycholic acid	500 mg q.n.	6 months	Randomised without blinding	6 months to 1 year
Zhu 1997	Qingdan capsule (No specific ingredients reported) (patent herbal medicine)	Oral	3 months	Ursodesoxycholic acid	10 mg/kg/day Q.n.	3 months	Randomised without blinding	No follow‐up
Xing 2009	Jinzhihuang capsule (Fructus Aurantii, Endothelium Corneum Gigeriae Galli, Rheum Officinale Baill, Radix Paeoniae Alba, Herba Lysimachiae, Rhizoma Corydalis, Radix Glycyrrhizae)	Oral	12 weeks	Chenodeoxycholic acid	0.5 g t.i.d.	12 weeks	Randomised without blinding	6 months
Shi 2012	Jinjidanshi decoction (Herba Lysimachiae 10 g, Endothelium Corneum Gigeriae Galli 15 g, Rhizoma Polygoni Cuspidati 30 g, Radix Bupleuri 10 g, Fructus Aurantii 10 g, Vaticamangachapoi Blauco 10 g, Rhizoma Cyperi 15 g, Herba Artemisiae Capillaris 15 g, Taraxacum officinale 15 g, Radix Curcumae 10 g, Rhizoma Corydalis Fructus 10 g) (self‐produced herbal compounds)	Oral	6 months	Ursodesoxycholic acid	0.3 g b.i.d.	6 months	Randomised without blinding	No follow‐up

b.i.d.: twice a day; t.i.d.: three times a day.

3. Western medicine or herbs used in the control group.

Number of trials	Western medicine by venous transfusion	Western medicine or herbs by oral administration
She 1981	Type of antibiotic without specific dosages and name	‐
Dai 2000	‐	(H) Aihuodantong capsule 0.1 g t.i.d. for 90 days
Deng 2005	‐	(H) Anetholi trithionum 25 mg t.i.d. for 8 weeks
Huang 2005	‐	(W) Ursodesoxycholic acid 0.2 g t.i.d. for 3 months
Zhang 2001a	‐	(H) Aihuodantong capsule 0.1 g t.i.d. for 3 months
Zhang 2001b	‐	(H) *Aihuodantong capsule 0.1 g t.i.d. for 3 months
Zhao 2004	‐	(W) Ursodesoxycholic acid 500 mg q.n. for 6 months
Zhu 1997	‐	(W) Ursodesoxycholic acid 10 mg/kg/dayq.n. for 3 months
Xing 2009	‐	(W) Chenodeoxycholic acid 0.5 g t.i.d. for 12 weeks
Shi 2012	‐	(W) Ursodesoxycholic acid 300 mg b.i.d. for 6 months
Su 2012	‐	(H) Lidanpaishi Capsule 3 # t.i.d. for 20 days

b.i.d.: twice a day; H: herbs; t.i.d.: three times a day; W: Western medicine.

All of the participants in the included trials were asymptomatic or had mild‐to‐moderate symptoms, including mild upper abdominal pain, biliary colic, nausea, vomiting, or icterus. In one trial, an antibiotic was administered for three to five days due to complications of mild acute cholecystitis, and the acute cholecystitis symptoms disappeared in less than one week (She 1981).

In most trials, herbs were given orally in dosages of 10 to 50 g for periods ranging from 20 days to six months (median, one to three months). Herb dosage was not reported in some trials to protect commercial or proprietary interests. Across all 11 trials, more than 30 individual Chinese medicinal herbs were used, with the most frequent being the 7 to 15 herbs listed in Table 2. The following herbs were used most frequently: Herba Lysimachiae, 54.5% (6 trials); Fructus Aurantii, 54.5% (6 trials); Radix Paeonia Alba, 45.5% (5 trials); Radix Aucklandiae, 45.5% (5 trials); Rheum Offcinale Baillon, 36.4% (4 trials); Rhizoma Corydalis Fructus, 36.4% (4 trials); Radix Curcumae, 36.4% (4 trials); Herba Artemisiae Capillaris, 36.4% (4 trials), and Radix Glycyrrhizae, 27.3% (3 trials). Each trial used different formulations and dosages of Chinese medicinal herbs. The constituents and the dosage of the Chinese medicinal herbs were different in all the trials except for the three trials by Dai 2000; Zhang 2001a; and Zhang 2001b.

Outcomes

The trial authors did not report some specific results anticipated by our protocol; instead, most trials reported general clinical effects based on a general clinical curative index that aggregates changes in symptoms and signs, as well as reduction, excretion, or dissolution of gallstones. In this index, complete curative effects were defined as disappearance of all symptoms and signs, and complete disappearance of gallstones based on imaging tests. Partial curative effects were defined as the disappearance of some symptoms, signs, and gallstones, and reduction (but not total disappearance) of gallstones based on imaging tests. No curative effects were defined as the disappearance of some or all symptoms, and an increase or no change in gallstones based on imaging tests. Some trials did not report symptoms and signs of interest to this review (see Types of outcome measures), but they did provide data on other symptoms, such as anorexia, belching, dry throat with bitter taste, and constipation. Some trials recorded some symptoms of adverse events, such as diarrhoea.

Risk of bias in included studies

The risk of bias of the included trials was high (Figure 1; Figure 2).

1.

Methodological quality graph: review authors' judgements about each methodological quality item presented as percentages across all included studies.

2.

Methodological quality summary: review authors' judgements about each methodological quality item for each included study.

The number of people at follow‐up and the reasons for the failures at follow‐up were not reported in any of the trials. All participants completed their respective treatment courses in all included studies. However, only three trials (27.3%) reported complete clinical outcome data (Zhang 2001a; Zhao 2004; Xing 2009); the other eight had incomplete results or incomplete follow‐up. Among the trials comparing Chinese medicinal herbs versus Western medicine, only two trials (40.0%) reported follow‐up lasting from six months to one year (Zhao 2004; Xing 2009; see Table 2); among the trials comparing certain Chinese medicinal herbs versus other herbs, only one trial (20.0%) reported follow‐up lasting from six months to one year (Zhang 2001a; see Table 4).

4. Administration of Chinese medical herbs (herbs versus other herbs).

Number of trials	Herbs in regimen in the intervention group	Route of administration in the intervention group	Period of administration in the intervention group	Herbs in the control group	Route of administration in the control group	Period of administration in the control group	Randomised clinical trial and method of blinding	Observation of follow‐up
Dai 2000	Gandanxiaoshi tablet (OX bile Infusion, Nitrum, Endothelium Corneum Gigeriae Galli, Fructus Crataegi, Rhizoma Cyperi, Pericarpium Citri Reticulatae Viride, Flos Inulae, Rhizoma Coptidis, Baked Rhizoma Atractylodis Macrocephalae, Fructus Aurantii, Rhizoma Galangae, Radix Aucklandiae, Massa Fermentata Medicinalis) without specific dosages (patent herbal medicine)	Oral	90 days	Aihuodantong capsule (extractive of Glechoma hederacea + Sunflower oil without specific dosages) (patent herbal medicine)	0.1 g t.i.d.	90 days	Randomised without blinding	No follow‐up
Deng 2005	Tongfupaishizhitong decoction (Herba Lysimachiae 30 g, Herba Artemisiae Capillaris 30 g, Spora Lygodii 20 g, Radix Codonopsis Pilosulae 20 g, Radix Atragali seu Hedysari 20 g, Radix Paeoniae Alba 15 g, Endothelium Corneum Gigeriae Galli 10 g, Radix Et Rhizoma Rhei 10 g, Fructus Aurantii 10 g, Radix Glycyrrhizae 10 g, Radix Pittospori 10 g, Fructus Meliae Toosendan 10 g, Rhizoma Corydalis 10 g, Radix Aucklandiae 5 g) (self produced herbal compounds)	Oral	8 weeks	Anetholi trithionum (2‐dithiole‐3‐thione)	1# t.i.d.	8 weeks	Randomised without blinding	No follow‐up
Zhang 2001a	Gandanxiaoshi tablet (OX bile Infusion, Nitrum, Endothelium Corneum Gigeriae Galli, Fructus Crataegi, Rhizoma Cyperi, Pericarpium Citri Reticulatae Viride, Flos Inulae, Rhizoma Coptidis, Baked Rhizoma Atractylodis Macrocephalae, Fructus Aurantii, Rhizoma Galangae, Radix Aucklandiae, Massa Fermentata Medicinalis) without specific dosages (patent herbal medicine)	Oral	3 months	Aihuodantong capsule (extractive of Glechoma hederacea + Sunflower oil without specific dosages) (patent herbal medicine)	0.1 g t.i.d.	3 months	Randomised without blinding	6 months to 1 year
Zhang 2001b	Gandanxiaoshi tablet (OX bile Infusion, Nitrum, Endothelium Corneum Gigeriae Galli, Fructus Crataegi, Rhizoma Cyperi, Pericarpium Citri Reticulatae Viride, Flos Inulae, Rhizoma Coptidis, Baked Rhizoma Atractylodis Macrocephalae, Fructus Aurantii, Rhizoma Galangae, Radix Aucklandiae, Massa Fermentata Medicinalis) without specific dosages (patent herbal medicine)	Oral	3 months	Aihuodantong capsule (extractive of Glechoma hederacea + Sunflower oil without specific dosages) (patent herbal medicine)	0.1 g t.i.d.	3 months	Randomised without blinding	No follow‐up
Su 2012	Paishi Capsule (Rhizoma Phei, Fructus Aurantii Immaturus, Natrii Sulfas Exsiccatus) without specific dosages	Oral	20 days	Lidanpaishi capsule (Herba Lysimachiae 25 g, Herba Artemisiae Capillaris 25 g, Radix Scutellariae 7.5 g, Radix Aucklandiae 7.5 g, Radix Curcumae 7.5 g, Rhizoma Phei 7.5 g, Semen Arecae 12.5 g, Fructus Aurantii Immaturus 12.5 g, sodium sulfate decahydrate 2.5 g, Magnoliaofficinalis Rehd.EtWils 5.0 g )	3 # t.i.d.	20 days	Randomised without blinding	No follow‐up

t.i.d.: three times a day.

No randomised clinical trials reported using an intention‐to‐treat analysis to evaluate their data. Information on baseline differences between the control and intervention groups was not reported sufficiently clearly in two of the trials (22.2%) (Zhu 1997; Dai 2000). The other trials reported no baseline differences between the two groups. See information in Table 2 and Table 4 (Chinese medicinal herbs versus Western medicine).

Allocation

All included trials were described as randomised clinical trials. Only one trial described the method used to generate the allocation sequence (random number table) (Zhang 2001a), and another trial described the method of allocation concealment (Zhu 1997).

Blinding

None of the study authors described that blinding was used in their trials.

Incomplete outcome data

Information on adverse events and follow‐up was incomplete, despite the fact that all participants completed the study and there were no treatment withdrawals and no changes within the trial groups.

Selective reporting

Trial protocols were not available. The trial publications left the impression that all outcomes that the investigators aimed to study were reported. However, outcomes important to participants and practitioners alike, such as mortality, biliary operation, quality of life, and medical costs, were not reported.

Other potential sources of bias

Unclear or not found in the trials.

Effects of interventions

Mortality

No mortality data were reported in any of the 11 trials.

Number of participants without relief of symptoms (i.e., upper abdominal pain, biliary colic, nausea, vomiting, and icterus)

Three trials comparing one Chinese herbal medicine versus another Chinese herbal medicine could be included in a meta‐analysis (Dai 2000; Zhang 2001a; Zhang 2001b). In these trials, Gandaoxiaoshi tablet, a compound of Chinese medicinal herbs, was given to the intervention group (200 people) at a dose of 3.0 g three times per day for three months, and Aihuodantong capsule at 0.1 g three times per day was given to the control group (90 people) for three months. The specific symptom of upper abdominal pain was mentioned in only two trials (Dai 2000; Zhang 2001a), and statistical analysis of the outcomes of the two trials are presented in Analysis 1.1. Authors of these trials did not report data for other symptoms and signs listed in the protocol, such as biliary colic, nausea, vomiting, and icterus, so they could not be assessed.

1.1. Analysis.

Comparison 1 Trials of Gandanxiaoshi tablet (herbs) versus Aihuodantong capsule (herbs), Outcome 1 Upper abdominal pain.

Upper abdominal pain

Of the three trials, only two presented specific data (Dai 2000; Zhang 2001a). The results of these trials showed no significant improvement on hypochondriac pain between the intervention group of 120 people (106 of whom reported upper abdominal pain before treatment) and the control group of 60 people (47 of whom reported upper abdominal pain before treatment) (RR 1.21; 95% CI 0.71 to 2.05) (Analysis 1.1).

In the trial of She 1981, the intervention group (16 people) received Cai hu cheng qi decoction, a compound of Chinese medicinal herbs, plus an antibiotic at an unspecific dose and for an unspecified treatment period. The control group (18 people) received the same antibiotic at an unspecific dose and for an unspecified treatment period. Detailed results about upper abdominal pain, biliary colic, nausea, vomiting, and icterus were not specified; the authors simply mentioned that there was no significant difference between the intervention and control groups in the disappearance time of upper abdominal pain and in the rate with which fever resolved. Therefore, outcomes related to symptoms, signs, and the underlying gallstone syndrome could not be assessed.

In the trial of Zhao 2004, the intervention group (100 people) was given Rongdanpaishi capsule, a compound of Chinese medicinal herbs, at a dose of 5 # three times per day for six months, and the control group (100 people) received ursodesoxycholic acid at 0.5 g every night for six months. In addition to the significant difference in general clinical curative effects, the authors reported improvement in upper abdominal pain, bitter taste, anorexia, and abdominal fullness in the intervention group. However, the authors did not report specific quantitative data for these outcomes, so the results could not be assessed in detail. Similarly, the other seven trials did not report detailed results on upper abdominal pain, biliary colic, nausea, vomiting, or icterus (Zhu 1997; Zhang 2001b; Deng 2005; Huang 2005; Xing 2009; Shi 2012; Su 2012); the trial authors mentioned only that there was a significant difference between the intervention and control groups in general clinical curative effects. Therefore, symptoms and signs could not be assessed.

Number of participants without reduction, excretion, or dissolution of gallstones

The effects of Chinese medicinal herbs on reduction, excretion, or dissolution of gallstones were assessed by B‐ultrasound at the end of the treatment period (Table 5). 'Excretion' refers to the disappearance of the stone in follow‐up ultrasound, presumably due to migration of the stone from the bile system to the intestine, while 'dissolution' refers to detectable reduction in stone size (e.g., from 2 cm to 1 cm), with the residual stone still detectable by ultrasound. The results show a significant difference in the effect on reduction, excretion, or dissolution of the gallstones between the intervention and control groups in two trials (She 1981; Dai 2000). In the study of She 1981, gallstones were excreted in 10 of 16 participants in the intervention group, while excretion did not occur in any of the 18 participants in the control group (P value = 0.000); this trial did not report the numbers of participants in whom the gallstones partially dissolved or reduced. Of the three trials pooled for meta‐analysis (Dai 2000; Zhang 2001a; Zhang 2001b), only Dai 2000 presented data on the numbers of participants not showing reduction, excretion, or dissolution of gallstones; however, these numbers, determined by B‐ultrasound, did not differ significantly between the intervention group (54/60 people) and the control group (25/30 people) (P value = 0.496).

5. The statistical results of symptoms and signs in two trials (Chi² test or Fischer's exact test).

The statistical results of trials (Chi2(P))	Upper abdominal pain	Nausea and vomiting	Abdominal distension	Anorexia	Belching	Icterus	Murphy's sign	Tenderness in right upper abdomen	Without reductionof gallstones	Without increase or change of gallstones	Without excretion or dissolution of gallstones
She 1981	‐	‐	‐	‐	‐	‐	‐	‐	‐	‐	Chi2=15.9 (P value = 0.000)
Dai 2000	‐	‐	‐	‐	‐	‐	‐	‐	Chi2= 0.26 (P value > 0.10)	Chi2= 0.02 (P value > 0.10)	Chi2= 0.83 (P value = 0.496)

Chi² = value of Chi² tests; P = value of Fisher's exact test.

The other trials did not present specific data on numbers of participants not showing reduction, excretion, or dissolution of gallstones.

Participants with serious adverse events

No serious adverse events were reported in the included trials.

Participants with non‐serious adverse events

Number and type of adverse effects

Adverse effects were mentioned in six of the 11 included trials (54.5%) (Table 6). Three of these six trials compared one Chinese herbal medicine (Gandanxiaoshi tablet) versus another Chinese herbal medicine, which was either Aihuodantong capsule (Dai 2000; Zhang 2001a) or Lidanpaishi capsule (Su 2012). The other three trials compared the Western medicine, ursodesoxycholic acid, versus different Chinese medicinal herbs, which were either Rongpaidanshi capsule (Zhao 2004), Danshu capsule (Huang 2005), or Paishi capsule (Shi 2012). Shi 2012 reported no adverse events in either the intervention or control group, whereas the numbers of participants experiencing adverse effects varied from 2 to 25 in the other five trials (Dai 2000; Zhang 2001a; Zhao 2004; Huang 2005; Su 2012). The following adverse effects were reported: vertigo, nausea, abdominal pain, and diarrhoea. The trial authors did not indicate whether the adverse effects were ascertained by standardised monitoring or voluntary self report, neither did they attempt to isolate the specific Chinese medicinal herbs that might be inducing those adverse effects. Fortunately, all adverse effects were mild and disappeared upon discontinuation of herbal therapy or reduction in herbal dosage.

6. The statistical results of adverse effects in one trial (Chi² test or Fischer's exact test).

The statistical results of trials (Chi2(P))	Diarrhoea	Abdominal pain with vertigo	Abdominal pain	Obstructive cholangitis
Huang 2005	‐	‐	Chi2= 1.03 (P value = 0.548)	‐
Zhao 2004	Chi2= 65.91 (P value < 0.001)	‐	‐	‐

Chi² = value of Chi² tests; P = value of Fisher's exact test.

Diarrhoea

Of the 11 trials, three trials presented specific data on occurrence of diarrhoea as an adverse effect (Table 6). In the trial of Zhao 2004, diarrhoea occurred in 11 of 100 participants in the intervention group (Rongpaidanshi capsule) compared to 67 of 100 participants in the control group (ursodesoxycholic acid) (P value < 0.001). In the trials of Dai 2000 and Zhang 2001a, diarrhoea occurred in 7 of 120 participants in the intervention group (Gandanxiaoshi tablet) and in none of the 60 participants in the control group (Aihuodantong capsule) (RR 4.07; 95% CI 0.52 to 31.78) (Analysis 1.2).

1.2. Analysis.

Comparison 1 Trials of Gandanxiaoshi tablet (herbs) versus Aihuodantong capsule (herbs), Outcome 2 Adverse effects: diarrhoea.

Abdominal pain

In this systematic review, we assessed the occurrence of abdominal pain as a side effect of treatment; this was different from the upper abdominal pain caused by cholelithiasis. Of the 11 trials, only one presented data on abdominal pain as an adverse effect (Huang 2005) (Table 6). That study found no significant difference between the frequency of occurrence in the intervention group (2 of 40 participants), which received Danshu capsule, and the control group (0 of 20 participants), which received ursodesoxycholic acid (P value = 0.548).

The number of participants undergoing biliary surgical interventions

No participant in any of the included trials underwent biliary surgical intervention.

Quality of life

No study reported data on participant quality of life.

Participant's economic condition for medical costs during hospitalisation, medication, and medical examination

No trials reported data on participants' medical costs during hospitalisation, medication, or medical examination.

Subgroup analysis and investigation of heterogeneity

We could not perform subgroup analyses because we could not include a sufficient number of comparable and homogeneous clinical trials in this review.

Discussion

Summary of main results

The results of the systematic review suggest that Gandanxiaoshi tablet, a patent Chinese medicinal herb, has equivalent clinical effects on cholelithiasis as compared to another herbal medicine, Aihuodnatong capsule; however, the absolute efficacy of either Gandanxiaoshi tablet or Aihuodantong capsule is unknown because we did not find any trials comparing these herbal medicines to placebo or no treatment. The result from the meta‐analyses in this review showed that Gandanxiaoshi tablet showed no significant difference from Aihuodantong capsule in the improvement of cholelithiasis symptoms. The other studied Chinese medicinal herbs in the included studies, while traditionally believed to reduce upper abdominal pain, biliary colic, nausea, vomiting, and icterus, did not show any significant difference from control groups for the majority of outcome measures. The therapeutic effects on reduction, excretion, or dissolution of gallstones in the bile duct or the gallbladder, or both were unclear in comparisons of Gandanxiaoshi tablet and Aihuodantong capsule.

Overall completeness and applicability of evidence

The long‐term goal of cholelithiasis treatment is to prevent recurrence of gallstones and onset of severe complications. The investigators of the included trials relied on a general clinical curative index that aggregated changes in certain specific symptoms and signs, as well as reduction, excretion, or dissolution of gallstones, but that did not take into account changes in other symptoms, such as upper abdominal pain, biliary colic, nausea, vomiting, and icterus. However, all the included participants in the trials are asymptomatic or just mild‐to‐moderate symptoms, participants with severe symptoms were excluded from the research because Chinese medicine practitioners believe medicinal herbs to play only an accessory role in treating cholelithiasis, therefore, people with severe cholelithiasis are usually treated with Western approaches. Furthermore, most trials did not report specific data on the reduction, excretion, or dissolution of gallstones, so we could not evaluate these outcomes. None of the trials reported how many participants had undergone biliary operations after treatment, so the proportion of participants who had undergone these operations could not be determined. In fact, most trials did not report other important outcomes, such as mortality, patient quality of life, medical costs, incidence of complications, or follow‐up. As a result, we could not draw any definite conclusions about the outcomes of interest to our review.

No serious adverse effects of the Chinese medicinal herbs for cholelithiasis were reported. However, definitive conclusions about adverse effects could not be drawn from this review as the sample size of the included trials was not large, and the period of treatment and follow‐up seemed insufficient. Though some slight symptoms, such as diarrhoea, were observed both in the Chinese medicinal herbs group (Dai 2000; Zhang 2001a), and Western medicine control group (Zhao 2004), reporting of adverse effects was poor and unsystematic, so we could not draw any conclusion about which medicine had more frequent or serious adverse effects. In fact, most Chinese clinical trials on Chinese medicinal herbs do not report adverse effects of the herbs or simply do not mention adverse effects, consistent with our previous observations that harm reporting is unreliable in studies on Chinese medicinal herbs (Chen 2010).

A plausible explanation for this lack of reporting is that Chinese clinical practitioners traditionally believe that Chinese medicinal herbs have few or no adverse effects since the herbs are natural products that have been used in China for many years. However, a growing number of studies show that Chinese medicinal herbs do cause adverse effects (Singh 2012). Therefore, medical researchers as well as clinical practitioners should pay more attention to such adverse effects and should consider them during clinical trial design and therapy planning. 

Quality of the evidence

All randomised clinical trials included in this review were of poor methodological quality, which led to a relatively high risk of bias. Trial reports provided only a limited description of study design and randomisation, and not all trials used allocation concealment. All trials stated only that random assignment was used, which we confirmed with the trial authors by telephone, but the reported outcomes data were incomplete, preventing us from drawing appropriate conclusions about the curative effects of the Chinese medicinal herbs on cholelithiasis. Most trials provided only general statements about the effects of the herbs; specific outcomes data, such as the numbers of participants experiencing reduction, excretion, or dissolution of gallstones, were not presented. We could not obtain this missing information, even after contacting the trial authors.

As a result of the different medicinal herbs used in the included trials and the high risk of bias, the therapeutic effects of several commonly used Chinese medicinal herbs, such as Herba Lysimachiae, Radix Paeonia Alba, Fructus Aurantii, Rheum Offcinale Baillon, Rhizoma Corydalis Fructus, and Radix Curcumae, remain uncertain. The trials also presented different clinical prescriptions in the diversity of Chinese medicinal herbs applied, as well as in baseline patient parameters such as age, sex ratio, and cholelithiasis aetiology and severity, preventing us from performing meta‐analyses.

The evidence we collected for our analyses is weak and we can neither recommend nor argue against the use of the Chinese medicinal herbs studied in our review for treatment of cholelithiasis.

Potential biases in the review process

We found the design, execution, and reporting of the included clinical trials to be inadequate: important outcomes, such as mortality, biliary operations, quality of life, and medical costs, were not reported. There were no treatment withdrawals in the included trials, but information on adverse events, participant follow‐up, and outcomes was incomplete. Participant characteristics at baseline were not clearly described in some of the trials. All these factors led us to estimate the trials to be at high risk of bias.

Agreements and disagreements with other studies or reviews

The results of one systematic review are in agreement with our review (Chen 2010), that is, that Chinese medicines do affect clinical symptoms of cholelithiasis. However, the long‐term therapeutic effects have rarely been reported. The conclusions of the present review require verification due to the low methodological quality, heterogeneity, and inherent risk of bias in the included trials.

Authors' conclusions

Implications for practice.

According to this systematic review, Chinese medicinal herbs should not be claimed to have strong positive therapeutic effects on the reduction, excretion, or dissolution of gallstones in the bile duct or the gallbladder even for people with asymptomatic or mild‐to‐moderate cholelithiasis. The risk of bias of the included randomised trials is high, and the available clinical data remain scanty. There is no strong evidence for the routine application of Chinese medicinal herbs in treatment of cholelithiasis.

Implications for research.

The methodological quality of future randomised clinical trials for treatment of cholelithiasis with Chinese medicinal herbs should be improved. In particular, the trials should present how allocation sequences were generated and how allocation was concealed in detail. Trials should use blinding and should be placebo‐controlled. Withdrawals and dropouts during the trial should be clearly reported. In fact, trials should report on all clinically important outcomes over a sufficiently long follow‐up period.

Future studies of medicinal herbs to treat cholelithiasis should be well‐designed, multicentre, large‐scaled randomised trials that compare Chinese medicinal herbs versus no treatment/placebo and versus a standard treatment for which we have data (e.g., ursodeoxycholic acid and the dissolution of gallstones). Trials should present in detail how allocation sequences were generated and how allocation was concealed. Specific outcomes should be assessed, including reduction, excretion, or dissolution of gallstones; recurrence of gallstones; quality of life; mortality; adverse effects; symptoms; signs; complications; and cost‐effectiveness, in addition to general clinical curative effects. Adverse effects should be critically assessed by standardised monitoring or by an effective self report system, or both. More attention should be paid to the long‐term adverse effects caused by Chinese medicinal herbs.

Appendices

Appendix 1. Search strategies

Database	Time of search	Search strategy
The Cochrane Hepato‐Biliary Group Controlled Trials Register	January 2013.	(((traditional OR Chinese OR oriental OR alternative OR complementary) AND medicine*) OR herb* OR plant*) AND cholelithiasis
The Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library	Issue 12 of 12, 2012.	#1 MeSH descriptor Phytotherapy explode all trees in MeSH #2 MeSH descriptor Drugs, Chinese Herbal explode all trees in MeSH #3 MeSH descriptor Medicine, Herbal explode all trees in MeSH #4 MeSH descriptor Plants, Medicinal explode all trees in MeSH #5 MeSH descriptor Medicine, Traditional explode all trees #6 ((traditional or Chinese or oriental or alternative or complementary) and medicine*) or herb* or plant* ? #7 (#1 OR #2 OR #3 OR #4 OR #5 OR #6) #8 MeSH descriptor Cholelithiasis explode all trees #9 cholelithiasis #10 (#8 OR #9) #11 (#7 AND #10)
MEDLINE (Ovid SP)	1946 to January 2013.	1. exp Medicine, Traditional/ 2. exp Drugs, Chinese Herbal/ 3. exp Medicine, Herbal/ 4. exp Phytotherapy/ 5. exp Plants, Medicinal/ 6. (((traditional or Chinese or oriental or alternative or complementary) and medicine*) or herb* or plant*).mp. [mp=title, original title, abstract, name of substance word, subject heading word] 7. 1 or 2 or 3 or 4 or 5 or 6 8. exp Cholelithiasis/ 9. cholelithiasis.mp. [mp=title, original title, abstract, name of substance word, subject heading word] 10. 8 or 9 11. 7 and 10 12. (random* or blind* or placebo* or meta‐analysis).mp. [mp=title, original title, abstract, name of substance word, subject heading word] 13. 11 and 12
EMBASE (Ovid SP)	1974 to January 2013.	1. exp Traditional Medicine/ 2. exp Medicinal Plant/ 3. Plant Medicinal Product/ 4. exp Phytotherapy/ 5. (((traditional or Chinese or oriental or alternative or complementary) and medicine*) or herb* or plant*).mp. [mp=title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer name] 6. 1 or 2 or 3 or 4 or 5 7. exp Cholelithiasis/ 8. Cholelithiasis.mp. [mp=title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer name] 9. 8 or 7 10. 6 and 9 11. (random* or blind* or placebo* or meta‐analysis).mp. [mp=title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer name] 12. 11 and 10
Science Citation Index Expanded (apps.webofknowledge.com)	1900 to January 2013.	# 3 #2 AND #1 # 2 TS=cholelithiasis # 1 TS=(((traditional or Chinese or oriental or alternative or complementary) and medicine*) or herb* or plant*)
Chinese Medicine Conference Disc	1979 to January 2013.	(((traditional OR Chinese OR oriental OR alternative OR complementary) AND medicine*) OR herb* OR plant*) AND (cholelithiasis OR cholelith OR stone of bile duct)
Chinese Bio‐Medicine Disc	1974 to January 2013.	(((traditional OR Chinese OR oriental OR alternative OR complementary) AND medicine*) OR herb* OR plant*) AND (cholelithiasis OR cholelith OR stone of bile duct)

Data and analyses

Comparison 1. Trials of Gandanxiaoshi tablet (herbs) versus Aihuodantong capsule (herbs).

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Upper abdominal pain	2	153	Risk Ratio (M‐H, Fixed, 95% CI)	1.21 [0.71, 2.05]
2 Adverse effects: diarrhoea	2	180	Risk Ratio (M‐H, Fixed, 95% CI)	4.07 [0.52, 31.78]

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Dai 2000.

Methods	Randomised clinical trial. Sample size: estimation not reported. Diagnostic criteria: B‐ultrasound. Baseline was comparable among the intervention and control groups.
Participants	Intervention group: 60 participants; M 20, F 40. Control group: 30 participants; M 11, F 19.
Interventions	Intervention group: Gandanxiaoshi tablet 3.0 g t.i.d. for 90 days. Control group 1: Aihuodantong capsule 100 mg t.i.d. for 90 days.
Outcomes	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones were reported. Clinical symptoms, including upper abdominal pain, abdominal distension, bitter taste, constipation, and anorexia were reported. Adverse events were reported.
Notes	We contacted the primary author by telephone regarding the unclear information in the trial in March 2009, but they could not remember the details.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not stated.
Allocation concealment (selection bias)	Unclear risk	Not stated.
Blinding (performance bias and detection bias) All outcomes	High risk	Not used.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Losses to follow‐up were not addressed.
Selective reporting (reporting bias)	High risk	General clinical curative effects, a comprehensive index including clinical symptoms, signs, and the reduction, dissolution, or excretion of gallstones were reported. The clinical symptoms, including upper abdominal pain, abdominal distension, bitter taste, constipation, anorexia were reported. The adverse events were reported. The specific results of symptoms, including biliary colic, nausea, vomiting, icterus, and signs were not reported. Mortality, biliary operation, quality of life, medical costs during hospitalisation, medication and medical examination were not reported.
Other bias	Unclear risk	Not stated.

Deng 2005.

Methods	Randomised clinical trial. Sample size: estimation not reported. Diagnostic criteria: B‐ultrasound. Baseline was comparable among the intervention and control groups.
Participants	Intervention group: 60 participants; M 25, F 35 Control group: 60 participants; M 20, F 40
Interventions	Intervention group: TCMHs 250 mL b.i.d. for 8 weeks. Control group: Danweita 1# t.i.d. for 8 weeks.
Outcomes	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones.
Notes	We could not contact the primary author by letter and telephone regarding the unclear information in the trial.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not stated.
Allocation concealment (selection bias)	Unclear risk	Not stated.
Blinding (performance bias and detection bias) All outcomes	High risk	Not used.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Losses to follow‐up were not addressed.
Selective reporting (reporting bias)	High risk	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones were reported. Specific results of symptoms, including upper abdominal pain, biliary colic, nausea, vomiting, icterus, and signs were not reported. Mortality, biliary operation, quality of life, medical costs during hospitalisation, medication, medical examination, and adverse events were not reported.
Other bias	Unclear risk	Not stated.

Huang 2005.

Methods	Randomised clinical trial. Sample size: estimation not reported. Diagnostic criteria: B‐ultrasound. Baseline was comparable among the intervention and control groups.
Participants	Intervention group: 40 participants; M 22, F 18. Control group: 20 participants; M 12, F 8.
Interventions	Intervention group: Danshu capsule 2# t.i.d. for 3 months. Control group: ursodeoxycholic acid 0.2 g t.i.d. for 3 months.
Outcomes	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones.
Notes	We contacted the primary author regarding the unclear information in the trial in March 2009, but the authors of the publication could not remember the details.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Sequence generation was not offered by the authors.
Allocation concealment (selection bias)	Unclear risk	Allocation concealment was not offered by the authors.
Blinding (performance bias and detection bias) All outcomes	High risk	Not used.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Losses to follow‐up were not addressed.
Selective reporting (reporting bias)	High risk	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones were reported. Adverse events were reported. Specific results on symptoms, including upper abdominal pain, biliary colic, nausea, vomiting, icterus, and signs were not reported. Mortality, biliary operation, quality of life, medical costs during hospitalisation, medication, and medical examination were not reported.
Other bias	Unclear risk	Not stated.

She 1981.

Methods	Randomised clinical trial. Sample size: estimation not reported. Diagnostic criteria: B‐ultrasound. Baseline was comparable among the intervention and control groups.
Participants	Intervention group: 16 participants; M 6, F 10. Control group: 18 participants; M 8, F 10.
Interventions	Intervention group: TCMHs. Control group: Western medicine (antibiotic) (the period of treatment is unclear).
Outcomes	Clinical manifestation: the TCMHs led to excretion of gallstones, from the bile system, time to disappearance of upper abdominal pain, fever.
Notes	We could not contact the primary author by letter and telephone regarding the unclear information in the trial.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not stated.
Allocation concealment (selection bias)	Unclear risk	Not stated.
Blinding (performance bias and detection bias) All outcomes	High risk	Not used.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Losses to follow‐up were not addressed.
Selective reporting (reporting bias)	High risk	Excretion of gallstones from the bile system was reported. Dissolution or excretion of gallstones from the bile system were not reported. Time for disappearance of upper abdominal pain and fever was mentioned without complete data. Specific results of symptoms, including upper abdominal pain, biliary colic, nausea, vomiting, icterus, and signs were not reported. Mortality, biliary operation, quality of life, medical costs during hospitalisation, medication, medical examination, and adverse events were not reported.
Other bias	Unclear risk	Not stated.

Shi 2012.

Methods	Randomised clinical trial. Sample size: estimation not reported. Diagnostic criteria: B‐ultrasound, ERCP or PTC. Baseline was comparable among the intervention and control groups.
Participants	Intervention group: 44 participants; M 15, F 29. Control group: 38 participants; M 13, F 25.
Interventions	Intervention group: Jinjidanshi decoction 1 dose b.i.d. for 6 months. Control group: ursodeoxycholic acid 0.3 g b.i.d. for 6 months.
Outcomes	The general clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not stated.
Allocation concealment (selection bias)	Unclear risk	Not stated.
Blinding (performance bias and detection bias) All outcomes	High risk	Not used.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Losses to follow‐up were not addressed.
Selective reporting (reporting bias)	High risk	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones were reported. Adverse events were reported. Specific results on symptoms, including upper abdominal pain, biliary colic, nausea, vomiting, icterus, and signs were not reported. Mortality, biliary operation, quality of life, medical costs during hospitalisation, medication, and medical examination were not reported.
Other bias	Unclear risk	Not stated.

Su 2012.

Methods	Randomised clinical trial. Sample size: estimation not reported. Diagnostic criteria: B‐ultrasound, or CT. Baseline was comparable among the intervention and control groups.
Participants	Intervention group: 64 participants; M 42, F 22. Control group: 64 participants; M 37, F 27.
Interventions	Intervention group: Paishi capsule 2˜4 # t.i.d. for 20 days. Control group: Lidanpaishi tablet 3 # t.i.d. for 20 days.
Outcomes	The general clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones.
Notes	‐
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not stated.
Allocation concealment (selection bias)	Unclear risk	Not stated.
Blinding (performance bias and detection bias) All outcomes	High risk	Not used.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	The losses to follow‐up were not addressed.
Selective reporting (reporting bias)	High risk	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones were reported. Adverse events were reported. Specific results on symptoms, including upper abdominal pain, biliary colic, nausea, vomiting, icterus, and signs were not reported. Mortality, biliary operation, quality of life, medical costs during hospitalisation, medication, and medical examination were not reported.
Other bias	Unclear risk	Not stated.

Xing 2009.

Methods	Randomised clinical trial. Sample size: estimation not reported. Diagnostic criteria: B‐ultrasound. Baseline was comparable among the intervention and control groups.
Participants	Intervention group: 100 participants; M 25, F 75. Control group: 101 participants; M 25, F 76.
Interventions	Intervention group: Jinzhihuang capsule 6# t.i.d. for 12 weeks. Control group: chenodeoxycholic acid 0.5 g t.i.d. for 12 weeks.
Outcomes	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones.
Notes	We contacted the primary author regarding the unclear information in the trial in October 2009, but the author could not remember the requested information.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not stated.
Allocation concealment (selection bias)	Unclear risk	Not stated.
Blinding (performance bias and detection bias) All outcomes	High risk	Not used.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Same number of participants randomised and analysed.
Selective reporting (reporting bias)	High risk	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones were reported. Specific results of symptoms, including upper abdominal pain, biliary colic, nausea, vomiting, icterus, and signs were not reported. Mortality, biliary operation, quality of life, medical costs during hospitalisation, medication, medical examination, and adverse events were not reported.
Other bias	Unclear risk	Not stated.

Zhang 2001a.

Methods	Randomised clinical trial. Sample size: estimation not reported. Diagnostic criteria: B‐ultrasound. Baseline was comparable among the intervention and control groups.
Participants	Intervention group: 60 participants; M 23, F 37. Control group: 30 participants; M 14, F 16.
Interventions	Intervention group: Gandanxiaoshi tablet 3.0 g t.i.d. for 3 months. Control group: Aihuodantong capsule 0.1 g t.i.d. for 3 months.
Outcomes	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones were reported. Clinical symptoms, including upper abdominal pain, abdominal distension, bitter taste, constipation, and anorexia were reported. Adverse events were reported.
Notes	We contacted the primary author by telephone regarding the unclear information in the trial in March 2009, but the authors of the publication could not remember the trial in detail.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random number table was used.
Allocation concealment (selection bias)	Unclear risk	Not stated.
Blinding (performance bias and detection bias) All outcomes	High risk	Not used.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Same number of participants randomised and analysed.
Selective reporting (reporting bias)	High risk	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones were reported. Clinical symptoms, including upper abdominal pain, abdominal distension, bitter taste, constipation, anorexia were reported. Adverse events were reported. Specific results of symptoms, including biliary colic, nausea, vomiting, icterus, and signs were not reported. Mortality, biliary operation, quality of life, medical costs during hospitalisation, medication, and medical examination were not reported.
Other bias	Unclear risk	Not stated.

Zhang 2001b.

Methods	Randomised clinical trial. Sample size: estimation not reported. Diagnostic criteria: B‐ultrasound. Baseline was comparable among the intervention and control groups.
Participants	Intervention group: 80 participants. Control group: 30 participants; the ratio of M/F is unclear.
Interventions	Intervention group: Gandanxiaoshi tablet 3.0 g t.i.d. for 3 months. Control group 1: Aihuodantong capsule 1# t.i.d. for 3 months.
Outcomes	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones.
Notes	We contacted the primary author by telephone regarding the unclear information in the trial in March 2009, but the authors of the publication could not remember the trial in detail.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not stated.
Allocation concealment (selection bias)	Unclear risk	Not stated.
Blinding (performance bias and detection bias) All outcomes	High risk	Not used.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Losses to follow‐up were not addressed.
Selective reporting (reporting bias)	High risk	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones were reported. Specific results of symptoms, including upper abdominal pain, biliary colic, nausea, vomiting, icterus, and signs were not reported. Mortality, biliary operation, quality of life, medical costs during hospitalisation, medication, medical examination, and adverse events were not reported.
Other bias	Unclear risk	Not stated.

Zhao 2004.

Methods	Randomised clinical trial. Sample size: estimation not reported. Diagnostic criteria: B‐ultrasound. Baseline was comparable among the intervention and control groups.
Participants	Intervention group: 100 participants; M 31, F 69. Control group: 100 participants; M 33, F 67.
Interventions	Intervention group: Rongpaidanshi capsule 5# t.i.d. for 6 months; Control group: UDCA 500 mg q.n. for 6 months.
Outcomes	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones.
Notes	We contacted the primary author by telephone regarding the unclear information in the trial in March 2009, but the authors of the publication could not remember the trial in detail.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not stated.
Allocation concealment (selection bias)	Unclear risk	Not stated.
Blinding (performance bias and detection bias) All outcomes	High risk	Not used.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Same number of participants randomised and analysed.
Selective reporting (reporting bias)	High risk	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones were reported. Time for disappearance of upper abdominal pain and fever were mentioned without complete data. Adverse events were reported. Specific results of symptoms, including upper abdominal pain, biliary colic, nausea, vomiting, icterus, and signs were not reported. Mortality, biliary operation, quality of life, medical costs during hospitalisation, medication, and medical examination were not reported.
Other bias	Unclear risk	Not stated.

Zhu 1997.

Methods	Randomised clinical trial. Sample size: estimation not reported. Diagnostic criteria: B‐ultrasound or X‐ray cholecystography Baseline was comparable among the intervention and control groups.
Participants	Intervention group I: 30 participants. Intervention group II: 30 participants. Control group: 30 participants. Ratio of M/F unclear.
Interventions	Intervention group I: Qingdan capsule 3# t.i.d. for 3 months. Intervention group II: danning tablet 5# t.i.d. for 3 months. Control group: UDCA 10 mg/kg/day q.n. for 3 months.
Outcomes	The general clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones.
Notes	We contacted the primary author by telephone regarding the unclear information in the trial in April 2009, but the authors of the publication could not remember the trial in detail.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not stated.
Allocation concealment (selection bias)	Unclear risk	Quote: ''Concealed envelops''.
Blinding (performance bias and detection bias) All outcomes	High risk	Not used.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	The losses to follow‐up were not addressed.
Selective reporting (reporting bias)	High risk	General clinical curative effects, a comprehensive index including clinical symptoms; signs; and the reduction, dissolution, or excretion of gallstones were reported. Specific results of symptoms, including upper abdominal pain, biliary colic, nausea, vomiting, icterus, and signs were not reported. Mortality, biliary operation, quality of life, medical costs during hospitalisation, medication, medical examination, and adverse events were not reported.
Other bias	Unclear risk	Not stated.

b.i.d.: twice daily; CT: computer tomography; ERCP: endoscopic retrograde cholangiopancreatography; F: female; M: male; PTC: percutaneous transhepatic cholangiography; q.n.: every night; TCMH: traditional Chinese medicinal herb; t.i.d.: three times a day; UDCA: ursodeoxycholic acid.

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Bie 1992	The article was not an RCT; information confirmed over the telephone by the author.
Binghui 2002	The article was not an RCT according to the author's description of random principle.
Bu 2008	The article was not an RCT; information confirmed over the telephone by the author.
Chen 1991	The article was not an RCT; information confirmed over the telephone by the author.
Chen 1993	The article was not an RCT; information confirmed over the telephone by the author.
Chen 1997	The article was not an RCT; information confirmed over the telephone by the author.
Chen 2000	The article was not an RCT; information confirmed over the telephone by the author.
Chen 2001	The article was not an RCT; information confirmed over the telephone by the author.
Chen 2003a	The article was not an RCT; information confirmed over the telephone by the author.
Chen 2003b	The article was not an RCT; information confirmed over the telephone by the author.
Chen 2003c	The article was not an RCT; information confirmed over the telephone by the author.
Chen 2006	The article was not an RCT; information confirmed over the telephone by the author.
Cheng 2005	The article was not an RCT; information confirmed over the telephone by the author.
Cui 1995	The article was not an RCT; information confirmed over the telephone by the author.
Dai 2009	The article was not an RCT; information confirmed over the telephone by the author.
Du 2004	The article was not an RCT; information confirmed over the telephone by the author.
Fan 2009	The article was not an RCT; information confirmed over the telephone by the author.
Gao 2004	The article was not an RCT; information confirmed over the telephone by the author.
Gao 2007	The article was not an RCT; information confirmed over the telephone by the author.
Guan 2001	The article was not an RCT; information confirmed over the telephone by the author.
Guo 2003	The article was not an RCT; information confirmed over the telephone by the author.
He 2004	The aim of this RCT was to appraise the effect of TCMHs for acupoint‐massage in ears.
Hou 1995	The article was not an RCT; information confirmed over the telephone by the author.
Huang 1998	The article was not an RCT; information confirmed over the telephone by the author.
Huang 2002	The article was not an RCT; information confirmed over the telephone by the author.
Huang 2004	The article was not an RCT; information confirmed over the telephone by the author.
Huang 2007	The article was not an RCT; information confirmed over the telephone by the author.
Jia 2003	The article was not an RCT; information confirmed over the telephone by the author.
Jin 2004	The article was not an RCT; information confirmed over the telephone by the author.
Jin 2007	The article was not an RCT; information confirmed over the telephone by the author.
Kang 2004	The article was not an RCT; information confirmed over the telephone by the author.
Kui 1996	The article was not an RCT; information confirmed over the telephone by the author.
Li 1998a	The article was not an RCT; information confirmed over the telephone by the author.
Li 2001	The article was not an RCT; information confirmed over the telephone by the author.
Li 2003	The article was not an RCT; information confirmed over the telephone by the author.
Li 2005a	The article was not an RCT; information confirmed over the telephone by the author.
Li 2005b	The article was not an RCT; information confirmed over the telephone by the author.
Li 2005c	The article was not an RCT; information confirmed over the telephone by the author.
Li 2007	The article was not an RCT; information confirmed over the telephone by the author.
Li 2009a	The article was not an RCT; information confirmed over the telephone by the author.
Li 2012a	The article was not an RCT.
Li 2012b	The article was not an RCT.
Li 2012c	The article was not an RCT.
Lian 2006	The article was not an RCT; information confirmed over the telephone by the author.
Liang 2000	The article was not an RCT; information confirmed over the telephone by the author.
Lin 1989	The article was not an RCT; information confirmed over the telephone by the author.
Lin 1999	The article was not an RCT; information confirmed over the telephone by the author.
Lin 2003	The article was not an RCT; information confirmed over the telephone by the author.
Liu 1997	The article was not an RCT; information confirmed over the telephone by the author.
Liu 1999	The article was not an RCT; information confirmed over the telephone by the author.
Liu 2000	The article was not an RCT; information confirmed over the telephone by the author.
Liu 2002a	The article was not an RCT; information confirmed over the telephone by the author.
Liu 2002b	The article was not an RCT; information confirmed over the telephone by the author.
Liu 2003	The article was not an RCT; information confirmed over the telephone by the author.
Liu 2008	The article was not an RCT; information confirmed over the telephone by the author.
Liu 2012	The article was not an RCT.
Long 1998	The article was not an RCT; information confirmed over the telephone by the author.
Long 2002a	The article was not an RCT; information confirmed over the telephone by the author.
Long 2002b	The article was not an RCT; information confirmed over the telephone by the author.
Lu 2003	The article was not an RCT; information confirmed over the telephone by the author.
Lv 1989	The article was not an RCT; information confirmed over the telephone by the author.
Ma 2002	The article was not an RCT; information confirmed over the telephone by the author.
Ma 2003	The article was not an RCT; information confirmed over the telephone by the author.
Mu 1992	The article was not an RCT; information confirmed over the telephone by the author.
Qing 2000	The article was not an RCT; information confirmed over the telephone by the author.
Qiu 2006	The article was not an RCT; information confirmed over the telephone by the author.
Shang 2006a	The article was not an RCT; information confirmed over the telephone by the author.
Shang 2006b	The article was not an RCT; information confirmed over the telephone by the author.
Shao 2010	The article was not an RCT according to the author's description of random principle.
Shen 1991	The article was not an RCT; information confirmed over the telephone by the author.
Shi 1999	The article was not an RCT; information confirmed over the telephone by the author.
Shi 2002	The article was not an RCT; information confirmed over the telephone by the author.
Shi 2003	The article was not an RCT; information confirmed over the telephone by the author.
Song 1992	The article was not an RCT.
Song 2005	The article was not an RCT; information confirmed over the telephone by the author.
Song 2006	The article was not an RCT; information confirmed over the telephone by the author.
Sun 2005	The article was not an RCT; information confirmed over the telephone by the author.
Tan 2005	The article was not an RCT; information confirmed over the telephone by the author.
Tang 1997	The article was not an RCT; information confirmed over the telephone by the author.
Tang 2002	The article was not an RCT; information confirmed over the telephone by the author.
Tao 2006	The article was not an RCT; information confirmed over the telephone by the author.
Tian 2005	The article was not an RCT; information confirmed over the telephone by the author.
Wan 2006	The article was not an RCT; information confirmed over the telephone by the author.
Wang 1995a	The article was not an RCT; information confirmed over the telephone by the author.
Wang 1995b	The article was not an RCT; information confirmed over the telephone by the author.
Wang 2000	The article was not an RCT; information confirmed over the telephone by the author.
Wang 2003a	The article was not an RCT; information confirmed over the telephone by the author.
Wang 2003b	The article was not an RCT; information confirmed over the telephone by the author.
Wang 2003c	The article was not an RCT; information confirmed over the telephone by the author.
Wang 2005	The article was not an RCT; information confirmed over the telephone by the author.
Wang 2007a	The article was not an RCT; information confirmed over the telephone by the author.
Wang 2009	The article was not an RCT; information confirmed over the telephone by the author.
Wang 2012	The article was not an RCT, but a retrospective study.
Wei 1995	The article was not an RCT; information confirmed over the telephone by the author.
Wu 2011	The article was not an RCT, but a retrospective study
Xie 1992	The article was not an RCT; information confirmed over the telephone by the author.
Xie 1997	The article was not an RCT; information confirmed over the telephone by the author.
Xu 1996	The article was not an RCT; information confirmed over the telephone by the author.
Xu 2006	The article was not an RCT; information confirmed over the telephone by the author.
Yang 1998	The article was not an RCT; information confirmed over the telephone by the author.
Yang 2002	The article was not an RCT; information confirmed over the telephone by the author.
Yang 2012	The article was not an RCT.
Yao 2001	The article was not an RCT; information confirmed over the telephone by the author.
Yu 2005	The article was not an RCT; information confirmed over the telephone by the author.
Yuan 1989	The article was not an RCT; information confirmed over the telephone by the author.
Yuan 1999	The article was not an RCT; information confirmed over the telephone by the author.
Zang 2009	The article was not an RCT; information confirmed over the telephone by the author.
Zeng 1994	The article was not an RCT; information confirmed over the telephone by the author.
Zeng 2004a	The article was not an RCT; information confirmed over the telephone by the author.
Zeng 2004b	The article was not an RCT; information confirmed over the telephone by the author.
Zhang 1999	The article was not an RCT; information confirmed over the telephone by the author.
Zhang 2000a	The article was not an RCT; information confirmed over the telephone by the author.
Zhang 2000b	The article was not an RCT; information confirmed over the telephone by the author.
Zhang 2001	The article was not an RCT; information confirmed over the telephone by the author.
Zhang 2003	The article was not an RCT; information confirmed over the telephone by the author.
Zhang 2005	The article was not an RCT; information confirmed over the telephone by the author.
Zhao 1993	The article was not an RCT; information confirmed over the telephone by the author.
Zheng 2006	The article was not an RCT; information confirmed over the telephone by the author.
Zhou 2004	The article was not an RCT; information confirmed over the telephone by the author.
Zhu 2007	The article was not an RCT; information confirmed over the telephone by the author.
Zhu 2009	The article was not an RCT; information confirmed over the telephone by the author.
Zhuang 2002	The article was not a RCT; information confirmed over the phone by the author.

TCMHs: Traditional Chinese Medicinal Herb; RCT: randomised controlled trial.

Differences between protocol and review

We also included people with gallstones in the gallbladder or the biliary ducts, or both as confirmed by a surgical procedure.

Contributions of authors

Tao Gan designed and revised the review, and controlled the overall quality of the review.
 Tao Gan wrote the first draft of the review.
 Jun Chen and Shuli Jin performed handsearches, retrieved the papers, and extracted the data.
 Yiping Wang conceived the idea for the review and gave some suggestions to the review.

Sources of support

Internal sources

• Chinese Cochrane Center, China.

• Chinese Center of Evidence‐based Medicine, China.

• Huaxi Hospital of Sichuan University, China.

External sources

• China Medical Board of New York, China.

Declarations of interest

None known.

New