Herbal medicinal products for non-ulcer dyspepsia

Authors


Correspondence to: Dr J. Thompson Coon, Department of Complementary Medicine, School of Sport and Health Sciences, University of Exeter, 25 Victoria Park Road, Exeter EX2 4NT, UK. E-mail: J.Thompson-Coon@exeter.ac.uk

Summary

Background : Non-ulcer dyspepsia is predominantly a self-managed condition, although it accounts for a significant number of general practitioner consultations and hospital referrals. Herbal medicinal products are often used for the relief of dyspeptic symptoms.

Aims : To critically assess the evidence for and against herbal medicinal products for the treatment of non-ulcer dyspepsia.

Methods : Systematic searches were performed in six electronic databases and the reference lists located were checked for further relevant publications. No language restrictions were imposed. Experts in the field and manufacturers of identified herbal extracts were also contacted. All randomized clinical trials of herbal medicinal products administered as supplements to human subjects were included.

Results : Seventeen randomized clinical trials were identified, nine of which involved peppermint and caraway as constituents of combination preparations. Symptoms were reduced by all treatments (60–95% of patients reported improvements in symptoms). The mechanism of any anti-dyspeptic action is difficult to define, as the causes of non-ulcer dyspepsia are unclear. There appear to be few adverse effects associated with these remedies, although, in many cases, comprehensive safety data were not available.

Conclusions : There are several herbal medicinal products with anti-dyspeptic activity and encouraging safety profiles. Further research is warranted to establish their therapeutic value in the treatment of non-ulcer dyspepsia.

Introduction

Dyspepsia is a common complaint; in a recent UK survey, 40% of adult responders reported one or more dyspeptic symptoms in the past year. Of all new patients consulting UK general practitioners each year, 3.3% suffer from dyspeptic symptoms; 60% of them will have no clinically significant abnormalities on investigation and will be diagnosed with non-ulcer dyspepsia.1 The majority of patients experience dyspeptic symptoms of short duration and mild severity, and are therefore self-managed,2 with less than half of dyspepsia sufferers seeking medical care for their complaints in the USA and Europe.3, 4

Herbal medicine is becoming increasingly popular, particularly for benign, chronic conditions that are self-managed, such as non-ulcer dyspepsia. Recent surveys suggest that the prevalence of use of herbal medicine in the USA is between 12% and 17%,5 with total retail sales of herbal dietary supplements of $4000 million in 2000.6 The European Scientific Co-operative on Phytotherapy definition of a herbal medicinal product is ‘any medicinal product containing as active ingredients only plants, parts of plants or plant materials or combinations thereof, whether in the crude or processed state’. Standard reference texts suggest that several herbal medicinal products are promoted for non-ulcer dyspepsia7–9(Table 1). It is therefore vital to establish both the efficacy of these supplements in relieving dyspeptic symptoms and their relative safety. This review is an attempt to systematically evaluate the evidence from randomized clinical trials for the efficacy of herbal medicinal products for non-ulcer dyspepsia.

Table 1.  Herbal medicinal products commonly recommended for the treatment of dyspepsia
  1. *  Herbal medicinal products discussed in this review.

Angelica*Angelica archangelicaGinger*Zingiber officinale
AnisePimpinella anisumHorsetailEquisetum arvense
Artichoke*Cynara scolymusHarongaHarungana madagascariensis
Bitter orange peelCitrus aurantiumHorehoundMarrubium vulgare
Blessed thistleCnicus benedictusJuniperJuniper communis
BogbeanMenyanthes trifoliataLemon balm*Melissa officinalis
Boldo*Peumus boldusMeadowsweetFilipendula ulmaria
Caraway*Carum carviMilk thistle*Silybum marianum
CardamomElettaria cardamomumMistletoeViscum album
CentuaryCentaurium minusOregon grapeMahonia aquifolium
ChicoryCichorium intybusPeachPrunus persica
CinchonaCinchona pubescensPeppermint*Mentha piperita
CinnamonCinnamomum verumRadishRaphanus sativus
ClovesSyzygium aromaticumRosemaryRosemarinus officinalis
CorianderCoriandrum sativumSageSalvia officinalis
DandelionTaraxacum officinaleSandy everlastingHelichrysum arenarium
Devil's clawHarpagophytum procumbensSt John's wortHypericum perforatum
DillAnethum graveolensStar aniseIllicium verum
ElecampaneInula heleniumThymeThymus vulgaris
FenugreekTrigonella foenum-graecumTurmeric*Curcuma longa
GalangalAlpinia officinariumWormwoodArtemisia absinthium
Gentian*Gentiana luteaYarrowAchillea millefolium

Materials and methods

Identification of clinical trials

In order to identify clinical trials involving herbal extracts used in the treatment of non-ulcer dyspepsia, systematic literature searches were conducted in the following electronic databases: Medline (Via Pubmed), Embase, CINAHL, Amed (Alternative and Allied Medicine Database, British Library Medical Information Centre), The Cochrane Library (Issue 2, 2001) and CISCOM (Research Council for Complementary Medicine, London, UK) (all from their inception to September 2001). The search terms used were: non-ulcer dyspepsia, functional dyspepsia, dyspepsia, reizmagen, idiopathic dyspepsia, phytomedicine, herbal medicine, botanical medicine, complementary medicine, phytotherapy, pfefferminzöl, kummelöl,celandine,Chelidonium majus, fennel, Foeniculum vulgare, banana, Musapep, capsaicin, Iberogast, Enteroplant, Lomatol and all terms in Table 1. Further relevant papers were located by hand-searching the reference lists of all papers. In addition, experts in the field and manufacturers of relevant herbal products were contacted to provide published and unpublished material. Finally, our own extensive files were hand-searched for further relevant publications.

Inclusion/exclusion criteria

Only randomized clinical trials of herbal remedies administered as supplements to patients with a diagnosis of non-ulcer dyspepsia were included. These could be placebo controlled or equivalent trials. All retrieved data, including uncontrolled trials, case reports, pre-clinical and observational studies, were reviewed for safety information and possible mechanisms of action. No language restrictions were imposed.

Data extraction and quality assessment

All articles were read in full. Data relating to sample size, diagnosis of patients, intervention and control, treatment duration, primary outcome measures and results were extracted by the first author and validated by the second author. The methodological quality of each randomized clinical trial was assessed using thescoring system of Jadad et al.10 This scale ranges from 0 (poorest) to 5 (highest) and assesses methods of randomization and blinding and the description of withdrawals and dropouts.

Results

Seventeen randomized controlled trials of herbal medicinal products for non-ulcer dyspepsia were identified. All trials stated that included patients had been diagnosed with non-ulcer or functional dyspepsia. However, details of the diagnostic tests performed (e.g. ultrasonography, oesophago-gastroduodenoscopy) were provided in only 11 of the trials.11–21 Four trials were of monopreparations (Table 2) and involved turmeric (Curcuma longa), greater celandine (Chelidonium majus), banana and Emblica officinalis. The remaining 13 trials were of combination products (Table 3), including nine in which peppermint (Mentha piperitae) and caraway (Carum carvi) were constituent ingredients. For each herb, possible active constituents and mechanisms of action, where known, are shown in Table 4.

Table 2.  Randomized clinical trials of monopreparations of herbal remedies used in the treatment of non-ulcer dyspepsia
ReferencenDiagnosisTreatmentTreatment
duration
(weeks)
Improvement
in symptoms
(%)
Jadad
score
Ritter et al.2260Functional epigastricStd celandine extract6Celandine 603
 complaints (6 tablets/day) or placebo  Placebo 26.7 
Thamlikitkul et al.23116Acid dyspepsia,Turmeric (2 g/day),1Turmeric 873
 flatulent dyspepsia flatulence (8 capsules/day)  Flatulence 83 
 or atonic dyspepsia or placebo  Placebo 53 
Arora and Sharma1146Non-ulcer dyspepsiaBanana powder
 (8 capsules/day)
 or no treatment
8Banana
 powder 75
 No treatment 20
1
Chawla et al.1238Peptic ulcer (10) andAmalaki (9 g/day) or gel4Both treatments1
 non-ulcer dyspepsia (28) antacids (30 mL every 3 h)  produced a
 significant
 reduction in
 symptom scores
 
Table 3.  Randomized clinical trials of combination preparations of herbal remedies used in the treatment of non-ulcer dyspepsia
ReferencenDiagnosisTreatmentTreatment
duration (weeks)
Improvement in
symptoms (%)
Jadad
score
  • * TJ-43 contains spray-dried aqueous extracts of Atractylodis lanceae rhizoma , Ginseng radix , Pinelliae tuber , Hoelen , Zizyphi fructus , Aurantii nobilis pericarpium , Glycyrrhizae radix and Zingiberis rhizoma.

  • Iberogast contains peppermint leaves, caraway fruit, bitter candy tuft fruit, licorice root, lemon balm leaves, angelica root, celandine herbs, milk thistle fruit and chamomile flowers.

  • Percentage of patients symptom-free.

Tatsuta and Iishi13 42Chronic idiopathic  dyspepsiaTJ-43* (7.5 g/day)  or placebo1Significant reductions in
 symptom scores compared
 with placebo and baseline
1
Chen24100Non-ulcer dyspepsiaShenxiahewining or an
 un-named control drug
 (15 capsules/day)
3Shenxiahewining 92
 Control 20
1
May et al.14 96Functional dyspepsiaPeppermint & caraway oils
 (180 & 100 mg/day) or placebo
4Peppermint/caraway 67
 Placebo 21
5
May et al.15 45Non-ulcer dyspepsiaPeppermint & caraway oils
 (270 & 150 mg/day) or placebo
4Peppermint/caraway 94.7
 Placebo 55
3
Freise and Kohler16223Non-ulcer dyspepsiaEnteric-coated peppermint &
 caraway oils (270 & 150 mg/day)
 or enteric soluble peppermint &
 caraway oils (108 & 60 mg/day)
4Enteric-coated peppermint/
 caraway 90 Enteric soluble
 peppermint/caraway 90
3
Madisch et al.17120Functional dyspepsiaPeppermint & caraway oils
 (180 & 100 mg/day) or cisapride  (30 mg/day)
4Peppermint/caraway 78.6
Cisapride 70.9
2
Madisch et al.18 60Functional dyspepsiaIberogast, Iberogast
 without bitter candy tuft
 (60 drops/day) or placebo
4Significant improvements in
 symptom scores after both herbal  preparations compared with placebo
4
Nicolay19 94Functional
 gastroenteropathy
Iberogast or metaclopramide
  (60 drops/day)
2Iberogast 52.6
Metaclopramide 56.4
2
Holtmann et al.25120Functional dyspepsiaIberogast (no details of dose
 provided) or placebo
8Iberogast 68.4
 Placebo 35.1
2
Rosch26183Functional dyspepsiaIberogast (60 drops/day) or
 cisapride (30 mg/day) or placebo
4Improvements in symptom scores
 after both Iberogast and cisapride
 compared with baseline
2
Westphal et al.20 60Functional dyspepsiaPeppermint, caraway, wormwood
 & fennel (75 drops/day) or
 metaclopramide
2Significant improvements in
 gastrospasmspain, nausea,
 heartburn and compared with
 metaclopramide
3
Stadelmann et al.21 46Functional dyspepsiaPeppermint oil and ginger
 extract (180 & 25 mg/day)
 or placebo
4Peppermint/ginger 74
 Placebo 30
3
Borgia et al.27359Mild to moderate
 functional disorders of
 the gastrointestinal tract
Boldo & cascara, rhubarb & gentian,
 boldo, cascara, rhubarb & gentian  or placebo
2Improvements in loss of appetite,
 dyspepsia and constipation compared
 with baseline and placebo
2
Table 4.  Possible active constituents and proposed mechanisms of action
HerbPossible active constituentsProposed mechanisms of action
Greater celandineApproximately 20 alkaloids with biological activityAnimal experiments suggest a direct anti-spasmolytic
 including chelidonine, sanginarine and chelerythrine, action on smooth muscle and a stimulation of bile
 flavonoids and phenolic acids28 flow28
TurmericVolatile oil composed mainly of turmerone andIncreased bile formation and secretion, promotion of
 zingiberene and phenolic compounds (curcuminoids) gall-bladder contraction, the volatile oil acts as a
  of which curcumin is the most abundant29 carminative improving digestion, some components
 may also be anti-spasmodic23, 29
BananaA flavonoid with anti-ulcerogenic activity has beenNot known
 isolated from unripe banana pulp30 
Emblica officinalisNot knownNot known
Liu-Jun-Zi-TangNot knownIncreased gastric emptying,13, 31 increased plasma
 levels of somatostatin and gastrin,32 promotion of
 gastric adaptive relaxation33
ShenxiahewiningNot knownNot known
PeppermintMenthol and menthone34Inhibition of smooth muscle contractions due to direct
 interaction between peppermint oil and smooth muscle
 calcium channels35, 36
CarawayNot knownInhibition of smooth muscle contraction37
BoldoAlkaloids, e.g. boldine, essential oils, volatile oils,Boldine acts as a smooth muscle relaxant in vitro by
 flavonol glycosides, resin and tannins directly interfering with the cholinergic mechanism
 involved in contraction38 and may cause gall-bladder
 contraction increasing bile secretion39
GentianBitter substances, e.g. gentiopicrosideStimulates salivary and gastric secretions via
 a reflex activity7

Clinical trials of monopreparations

Greater celandine (Chelidonium majus).  Ritter et al. conducted a 6-week, placebo-controlled, double-blind, randomized trial using a standardized celandine extract in 60 patients with functional epigastric complaints and cramp-like pains in the region of the biliary and gastrointestinal tracts.22 After 6 weeks of treatment, patients treated with the celandine extract reported fewer gastrointestinal symptoms compared with the placebo group (P=0.003). The symptoms that showed most improvement were stomach pains, bilious complaints, flatulence, nausea and sensation of fullness. Eight patients reported adverse events during the study, five whilst receiving placebo (insomnia, urticaria and sleepiness) and three whilst receiving celandine extract (dry mouth and insomnia).

In an open clinical trial of greater celandine treatment, six patients reported adverse events (diarrhoea, 3; nausea, 2; tiredness, 1).40 A case series of 10 cases of acute hepatitis following treatment with greater celandine extracts was identified.41 Where available, liver biopsy specimens were consistent with drug-induced damage, and other possible causes of liver disease were excluded by laboratory tests and imaging procedures. Discontinuation of greater celandine resulted in rapid recovery, with liver enzyme levels returning to normal within 2–6 months. One patient was unintentionally re-challenged, resulting in a second episode of hepaticinflammation. A further case series describes twopatients investigated for abdominal pain, who were prescribed celandine extract (Panchelidon, two capsules t.d.s. and one capsule b.d.), and presented 2 and 4 months later with scleral jaundice and elevated transaminase levels. On cessation of therapy, the liver enzymes returned to normal and the patients recovered without sequelae. The patients were not re-challenged.42 A single case report of contact dermatitis in a 64-year-old woman who used greater celandine juice to treat warts was also identified;43 whether this is relevant to the oral administration of celandine tablets is not clear.

Turmeric (Curcuma longa).  One randomized, double-blind, multicentre clinical trial conducted in Thailand was located, in which 116 patients (diagnosed with acid dyspepsia, flatulent dyspepsia or atonic dyspepsia) received capsules of turmeric (2 g/day), placebo or a combination treatment known as flatulence (cascara dry extract, nux vomica extract, asofoetida tincture, capsicum powder, ginger powder and diastase) for 7 days.23 Significantly more patients (P=0.003) reported an overall favourable outcome (resolution or improvement of symptoms) following treatment with turmeric (33 patients) than placebo (20 patients), although there was no significant difference in outcome between those patients treated with turmeric and those treated with flatulence. Twenty-eight patients experienced adverse events: nine in the turmeric group (nausea, diarrhoea, headache, tiredness and sleepiness), nine in the flatulence group (nausea, vomiting, diarrhoea, throat discomfort, headache and flatulence) and 10 in the placebo group (nausea, diarrhoea, pruritis, headache, constipation, anorexia and tiredness). All the side-effects were described as mild and self-limited.

Banana (Musa sapientum).  A prospective, randomized, open study of 46 consecutive patients diagnosed with non-ulcer dyspepsia was conducted in India.11 Twenty-two patients received banana powder (Musapep) capsules for 8 weeks; 24 patients acted as controls and were not given any additional treatment. Gastrointestinal symptoms were partly or completely relieved in three-quarters of patients treated with banana powder and in one-fifth of untreated control patients (P < 0.05). One patient suffered generalized itching during treatment with banana powder.

Emblica officinalis.  One randomized clinical trial was identified in which 38 patients (10 with peptic ulcer and 28 with non-ulcer dyspepsia) were treated with Amalaki (the pericarp of the dried fruits of Emblica officinalis) or gel antacids for 4 weeks.12 All patients were advised to follow a bland diet and to avoid smoking, alcohol, coffee and all other drugs for the duration of the study. Both treatments produced a significant reduction in symptom scores (P < 0.01), accompanied by a decrease in acid output (P < 0.001), compared with baseline, although comparisons were not made between treatments. Adverse events were reported by four patients treated with antacids (weakness and pain in the lower limbs) and three patients treated with Amalaki (loose motions and vomiting); none were severe enough to necessitate withdrawal of treatment.

Clinical trials of combination preparations

Liu-Jun-Zi-Tang.  Liu-Jun-Zi-Tang is a Chinese herbal medicine, also known as Rikkunshi-to and TJ-43, containing spray-dried aqueous extracts of Atractylodis laneae rhizoma, Ginseng radix, Pinelliae tuber, Hoelen, Zizyphi fructus, Aurantii nobilis pericarpium, Glycyrrhizae radix and Zingiberis rhizoma. Forty-two patients with chronic idiopathic dyspepsia were treated with either Liu-Jun-Zi-Tang (2.5 g three times daily) or placebo for 7 days.13 Treatment with Liu-Jun-Zi-Tang resulted in a significantly greater reduction in gastrointestinal symptom scores for epigastric fullness, heartburn, belching and nausea compared to baseline and compared to patients treated with placebo (P < 0.05), although there was no effect on epigastric pain. Compared to both baseline and placebo treatment, gastric emptying, measured by the acetaminophen absorption method, was significantly increased following Liu-Jun-Zi-Tang treatment (P < 0.01). Details of adverse events were not provided; however, a case report of drug-induced pneumonitis due to Liu-Jun-Zi-Tang was identified, in which a 79-year-old woman was admitted to hospital with interstitial pneumonia. Following the cessation of all drugs and the administration of prednisolone, her symptoms and laboratory data improved. Lymphocyte stimulation and migration inhibition tests were positive for Liu-Jun-Zi-Tang.44

Shenxiahewining.  Shenxiahewining is a Chinese herbal medicine which contains Ginseng radix, Pinelliae tuber, Coptidis rhizoma, Zingiberis rhizoma exsiccatum and Glycyrrhizae radix in the ratio 3 : 9 : 3 : 3 : 3. In a randomized clinical trial performed in China, 100 subjects with non-ulcer dyspepsia were treated with either shenxiahewining (15 capsules/day; 0.42 g/capsule) or an un-named control drug in similar capsules (15 capsules/day; 0.42 g/capsule) for 20 days. Improvements in symptoms were reported in 92% of patients treated with the herbal mixture compared with 20% treated with the control medication. No adverse events were reported for either treatment group.24

Peppermint and caraway (Mentha piperita and Carum carvi).  Four randomized clinical trials of a fixed combination of peppermint and caraway oils were identified, five trials of a combination of peppermint leaves, caraway fruit, bitter candy tuft (Iberis amara), licorice root (Glycyrrhiza glabra), lemon balm leaves (Melissa officinalis), angelica root (Angelicae radix), celandine herbs (Chelidonium majus), milk thistle fruit (Silybum marianus) and chamomile flowers (Matricaria recutita) (Iberogast, Steigerwald Arzneimittelwerk GmbH, Darmstadt, Germany), one trial of a combination of extracts of peppermint leaves, caraway fruit, fennel fruit (Foeniculum vulgare) and wormwood herbs (Artemisia absinthium) (Lomatol, Lomapharm, Emmerthal, Germany), and one trial of a combination of peppermint oil and ginger extract.

May et al. performed a double-blind, placebo-controlled trial in 96 out-patients with a diagnosis of functional dyspepsia, defined as diffuse, unspecific, variable, moderately intense epigastric pain with at least one other dyspeptic symptom and an absence of organ pathology.14 Patients received either enteric-coated capsules of a fixed combination of peppermint and caraway oils (180 mg/day peppermint oil and 100 mg/day caraway oil; Enteroplant, Dr Willmar Schwabe GmbH, Karlsruhe, Germany) or identical placebo capsules for 4 weeks. There were statistically significant differences in the reduction of pain intensity (40% vs. 22% of the initial mean value; P < 0.001) and sensation of pressure, heaviness and fullness (43.5% vs. 22.3% of the initial mean value; P < 0.001) between the peppermint/caraway- and placebo-treated patients, with an overall median rating of ‘much improved’ for the peppermint/caraway group and ‘minimally improved’ for the placebo group (P < 0.001).

A similarly designed study from the same group of investigators involved 45 patients with non-ulcer dyspepsia, defined as moderate epigastric pain with atleast one other dyspeptic symptom for the past 14 days.15 A fixed combination of peppermint and caraway oils (270 mg/day peppermint oil and 150 mg/day caraway oil; Enteroplant, Dr Willmar Schwabe GmbH, Karlsruhe, Germany) or placebo was administered for 4 weeks. Improvement in pain intensity after 4 weeks of treatment was significantly greater in the peppermint/caraway-treated patients than in those treated with placebo (17 vs. 9; P=0.015). The physician also estimated the prognosis and severity of the disorder on the basis of a Clinical Global Impressions Scale before and after treatment, and found that 94.7% of patients in the peppermint/caraway group showed improvement compared with only 55% of patients in the placebo group (P=0.008).

Two hundred and twenty-three patients with non-ulcer dyspepsia (defined as dysmotility type or essential/idiopathic dyspepsia with irritable bowel syndrome) were included in a prospective, randomized, double-blind multicentre trial in which two fixed combinations of peppermint and caraway oils were compared.16 Patients received either enteric-coated capsules (270 mg/day peppermint oil and 150 mg/day caraway oil; Enteroplant, Dr Willmar Schwabe, Karlsruhe, Germany) or an enteric soluble formulation (108 mg/day peppermint oil and 60 mg/day caraway oil; Hersteller, Dr Willmar Schwabe, Karlsruhe, Germany) for 4 weeks. The mean baseline measurements of epigastric pain intensity were 6.1 in the former group and 5.9 in the latter group [on a visual analogue scale ranging from 0 (no pain) to 10 (extremely strong pain)]. Statistically significant reductions in pain intensity were seen in both groups (−3.6 vs. −3.3; P=0.001).

A double-blind comparative study of a fixed combination of peppermint and caraway oil and the prokinetic agent cisapride was performed in 120 out-patients with functional dyspepsia.17 Inclusion criteria included the presence of moderate epigastric pain and one other dyspeptic symptom for at least the past 14 days. Patients were treated with cisapride (30 mg/day) or a peppermint–caraway combination (180 mg/day peppermint oil and 100 mg/day caraway oil; Enteroplant, Dr Willmar Schwabe, Karlsruhe, Germany) for 4 weeks. Both groups showed improvements in epigastric pain intensity, pain frequency, dyspeptic discomfort (a combined measure of other dyspeptic intestinal and extra-intestinal symptoms) and Clinical Global Impression, with no clinically relevant differences between the effects of the peppermint–caraway combination and cisapride.

A placebo-controlled, multicentre trial of Iberogast (Steigerwald Arzneimittelwerk GmbH, Darmstadt, Germany) was conducted in 60 patients with functional dyspepsia, in which patients were treated with Iberogast, a similar preparation without extract of bitter candy tuft or placebo (60 drops/day) for 4 weeks.18 Compared with placebo, both herbal preparations produced a statistically significant improvement in the gastrointestinal symptom score after 2 and 4 weeks of treatment (P < 0.001). There was no difference between the two herbal preparations, except that Iberogast appeared to have a more rapid effect on symptom improvement (P=0.023), suggesting that bitter candy tuft is a necessary component of the mixture.

A single-blind, randomized, comparative study was conducted involving 94 patients with functional gastroenteropathy treated with either Iberogast or metaclopramide for 14 days.19 The occurrence of four symptoms was recorded: feeling of pressure and pain, sickness and vomiting, eructation and heartburn. After 14 days, there was no significant difference between the two groups of patients, with approximately 50% of patients in each group reporting no further symptoms. On average, symptoms resolved faster in the Iberogast-treated patients. Seventeen patients withdrew from the study as a result of treatment failure (metaclopramide, 3; Iberogast, 1), non-compliance with the protocol (12) and sickness and vomiting (metaclopramide, 1).

Two studies were found which have only been presented in abstract form to date. The first was a four-way parallel group study, in which each patient received treatment for two consecutive 4-week periods; 120 patients with functional dyspepsia were assigned to one of the following treatment groups: placebo–placebo; placebo–Iberogast; Iberogast–placebo; and Iberogast–Iberogast.25 Details of the doses were not provided in the abstract. After 8 weeks of treatment, 43.3% of patients treated with Iberogast were symptom-free compared to only 3.3% of the placebo group (P < 0.001). The second study had a randomized, double-blind, placebo-controlled, double dummy design and enrolled 124 patients with functional dyspepsia. After 4 weeks of treatment with Iberogast (60 drops/day), cisapride (30 mg/day) or placebo, gastrointestinal symptom scores had decreased in the active treatment groups (Iberogast, 14.3 to 3.3; cisapride, 14.0 to 3.9). The results for placebo were not provided. There were no significant differences between Iberogast and cisapride; however, both showed significant differences from baseline (P < 0.001).26 Neither study provided details of the occurrence of adverse events.

Westphal et al. conducted a comparative, double-blind study of Lomatol (Lomapharm, Emmerthal, Germany) in 60 patients with at least one of the following symptoms: pressure and pain in the upper stomach, sub- or retrosternal gastrospasms, sensation of repletion, belching, nausea, retching, heartburn and loss of appetite.20 Patients were treated with the herbal combination or metaclopramide, a prokinetic used in the treatment of digestive tract complaints, (25 drops in lukewarm water three times daily prior to meals) for a maximum of 14 days. Five symptoms were studied: pain, nausea, heartburn, retching and gastrospasms. Compared with metaclopramide, significantly fewer patients experienced pain, nausea and heartburn after 7 days of treatment with the herbal combination and, after 14 days, the results were also statistically significant for gastrospasms, and two of the secondary outcome measures: belching and sensation of pressure.

Stadelmann et al. reported a randomized, double-blind, placebo-controlled study in 46 patients with functional dyspepsia, who were treated with peppermint oil and ginger extract (180 mg/day and 25 mg/day) or identical placebo for 4 weeks. At the end of the treatment period, 74% of patients treated with the herbal combination reported an improvement in gastrointestinal symptom scores compared with 30% in the placebo group (P=0.012).21

In total, adverse events were reported by 84 of 1046 patients: 26 of these were during peppermint and caraway treatment; one during treatment with peppermint, caraway, fennel and wormwood; 13 during treatment with peppermint and ginger; 14 during cisapride treatment; 18 during metaclopramide treatment; two during treatment with Iberogast; and 10 during placebo treatment. The adverse events included heartburn or eructation with a peppermint taste, diarrhoea, nausea, sickness and vomiting, substernal burning sensation with severe eructation and nausea, flatulence and increasing acid taste, hyperventilation and suspected grand mal with syncope, upperabdominal pain and dizziness, spasm such as abdominal pains with increased gastrointestinal noise and feeling of pressure in the upper abdomen, increased stool evacuation, pasty stools and burning in the rectum, pain in the neck and shoulder, extreme tiredness, haemorrhoids, bronchitis and influenza-like symptoms. Cessation of treatment was necessary in six patients due to the severity of symptoms. A causal relationship with the herbal combination was considered possible in eight patients. No serious adverse events were reported and no abnormality was detected inlaboratory values during or after treatment.Apost-marketing surveillance study of Iberogast was identified, involving 2267 patients, in which only one adverse event, without a causal relationship to Iberogast therapy, was recorded; 95% of doctors reported the tolerability to be ‘good’ or ‘very good’.45 Peppermint oil and menthol have been reported as causes of allergic contact dermatitis in fragrances and toothpastes,46–48 contact urticaria has been reported from menthol in mentholated cigarettes, cough drops, aerosol room spray and topical medicaments,49 and other rare cases of skin irritation have also been reported.50, 51 Exacerbation of asthma has beendescribed with the use of a peppermint-containing toothpaste.52 In a further case report, two patients who were addicted to ‘peppermints’ experienced idiopathic atrial fibrillation, which resolved with the cessation of peppermint ingestion.53 No further information on the safety of caraway was located.

Other herbs.  One randomized clinical trial was identified in which herbal combinations containing boldo (Peumus boldus), cascara (Rhamnus purshianus), gentian (Gentiana lutea) and rhubarb (Rheum sp.) were evaluated. All four herbs or placebo were administered to 359 patients with slight or moderate functional disorders of the gastrointestinal tract for 28 days.27 Compared with baseline and placebo, the combination of all four herbs and rhubarb and gentian alone produced statistically significant improvements in the loss of appetite, dyspepsia and constipation (P < 0.001), whereas boldo and cascara alone only improved symptoms of constipation (P < 0.001). No adverse events were reported during the trial.

Discussion

Several herbal medicinal products have been identified for use in the relief of symptoms of non-ulcer dyspepsia; in all trials, subjects treated with the herbal medicinal product showed improvements in symptom scores (results ranging from 60% to 95%) compared with baseline and/or placebo or the comparator drug. Peppermint and caraway have been most extensively studied, although they have sometimes been administered in the presence of other herbal extracts.

There was a large response to placebo (30–55% of patients showing an improvement in symptoms compared to baseline) reported in five of the 10 placebo-controlled studies (data were not provided for the remaining five studies). A large placebo response has also been noted in clinical trials of conventional therapies for non-ulcer dyspepsia (13–73%).55 In addition, an open, non-randomized study designed to assess the effect of placebo supplementation in patients with non-ulcer dyspepsia found that 80% of patients reported an improvement in symptoms, with the improvement being marked in 47%.54

Due to the episodic nature of non-ulcer dyspepsia, studies with no placebo arm, in which the herbal preparation is compared with a comparator drug, e.g. metaclopramide or cisapride, are difficult to interpret; although they show similar improvements in symptom scores for both treatment options, they provide no measure of the natural progression of symptoms.

A number of weaknesses with the original trials exist. Of the 17 identified randomized clinical trials of herbal medicinal products for non-ulcer dyspepsia, only eight scored three or more points on the scale of Jadad et al.10 The most frequent methodological flaws were single blind or open studies and incomplete reporting of methods of randomization, blinding and subject withdrawals.

Several different methods of scoring symptoms were used and not all were validated techniques. The description of patients entered into each trial was variable; it was not clear whether patients with, for example, predominantly reflux symptoms were always excluded or what the Helicobacter pylori status of each patient was. It is therefore difficult to compare studies or to extrapolate the data to other patient populations whose perception of symptom severity may differ.

However, these results compare favourably with those using conventional treatment approaches for non-ulcer dyspepsia, which are far from ideal. A meta-analysis of 53 studies published in 1996 concluded that, due to methodological flaws, there are no proven effective therapies for the treatment of non-ulcer dyspepsia.55 Antacids appear to provide little relief;56 acid-suppressing agents and prokinetics provide moderate symptomatic improvement, with average differences from placebo of 25%55 and 40%57 in the proportions of patients showing improvement; the eradication of H.pylori, although controversial, appears to offer little benefit.58

Of the 17 trials identified, 13 relate to combination products containing two or more herbal extracts. It is unclear which of the herbs contained in these products is effective or, indeed, if they all play a role in alleviating symptoms. Combination preparations are popular in herbal medicine as many traditional herbalists believe that synergistic interactions can occur between the different components, eliciting a larger effect than would be expected from the individual constituents alone. There is a limited amount of evidence for such interactions with some herbs,59 and it may offer some insight into the mechanism whereby low doses of some extracts are able to produce beneficial therapeutic effects. However, there is no such evidence for the herbs included in this review; much detailed research is needed to elucidate the role of each individual component alone and within combinations.

Smooth muscle relaxation and effects on the gall-bladder are common to several of the herbs, and may be important in eliciting the overall effects on symptoms; however, it is difficult to postulate on and study the mechanisms by which herbal extracts are able to alleviate symptoms when the pathophysiology of the condition is so ill-defined.

The safety profile of these herbs looks encouraging, with the exception of greater celandine, for which there have been reports of hepatotoxicity. Many have been used extensively in traditional medicine and culinary practices around the world. This supports their relative safety, but does not demonstrate it beyond reasonable doubt.60 In particular, the long-term safety for use as herbal medicines, either alone or in conjunction with conventional therapies, has not been established. Comparative and placebo-controlled trials suggest that patients experience no more adverse events with these herbs than with placebo or comparative medications, although short-term clinical trials are not designed to detect rare or delayed adverse events. No evidence for herb–drug interactions with peppermint or caraway was uncovered; however, there is indirect evidence for herb–drug interactions with several of the other herbs contained in the combination preparations.61–63 Further research is necessary to elucidate the importance of these.

To conclude, some of the herbal medicinal products identified, in particular peppermint and caraway, with effects of similar or greater magnitude to conventional therapies and encouraging safety profiles, undoubtedly warrant further investigation, especially in the light of the lack of efficacy of conventional therapies for non-ulcer dyspepsia. A greater understanding of the pathophysiology of non-ulcer dyspepsia would facilitate the improvement of the therapeutic options available to treat this widespread and debilitating condition.

Acknowledgements

The authors wish to thank Jongbae Park, Barbara Wider, Katja Schmidt and Francesca Borrelli, Department of Complementary Medicine, University of Exeter, UK, for the translation of papers from Chinese, Italian, German and French, and Esther Prati, Pharmaton SA,Lugano, Switzerland, for assistance with locating relevant articles.

JTC is supported by a research fellowship provided by Pharmaton SA, Lugano, Switzerland.

Ancillary