Drugs for treating giardiasis

  • Review
  • Intervention

Authors


Abstract

Background

Giardiasis infection may be asymptomatic, or can cause diarrhoea (sometimes severe), weight loss, malabsorption, and, in children, failure to thrive. It is usually treated with metronidazole given three times daily for five to 10 days.

Objectives

To evaluate the relative effectiveness of alternative antibiotic regimens for treating adults or children with symptomatic giardiasis.

Search methods

We searched the Cochrane Infectious Disease Group Specialized Register, the Cochrane Central Register of Controlled Trials (CENTRAL) (Issue 6 2012); MEDLINE, EMBASE, LILACS and the International Clinical Trials Registry Platform Search Portal (3 July 2012).

Selection criteria

We included randomized controlled trials (RCT) comparing metronidazole administered for five to 10 days with any of the following drugs: metronidazole (single dose), tinidazole, albendazole, mebendazole, and nitazoxanide. The primary outcomes were parasitological and clinical cure.

Data collection and analysis

Two authors independently assessed studies for inclusion, performed the risk of bias assessment, and extracted data. We summarized data using risk ratios and mean differences and we presented the results in forest plots and performed meta-analyses where possible. We assessed heterogeneity using the Chi2 test, I2 statistic and visual inspection; and we explored this by using subgroup analyses.We assessed the quality of evidence by using the GRADE approach.

Main results

We included 19 trials, involving 1817 participants, of which 1441 were children. Studies were generally small, with poor methods reporting. . Most reported parasitological outcomes rather than clinical improvement.

Ten trials, from India, Mexico, Peru, Iran, Cuba, and Turkey, compared albendazole (400 mg once daily for five to 10 days) with metronidazole (250 mg to 500 mg three times daily for five to 10 days). This once-daily regimen of albendazole is probably equivalent to metronidazole at achieving parasitological cure (RR 0.99, 95% CI 0.95 to 1.03; 932 participants, 10 trials; moderate quality evidence), and improving symptoms (RR 0.98, 95% confidence interval (CI) 0.93 to 1.04; 483 participants, five trials; moderate quality evidence), but the duration of follow-up was short (two to three weeks). Albendazole probably has fewer side effects than metronidazole (gastrointestinal side effects: RR 0.29, 95% CI 0.13 to 0.63; 717 participants, eight trials; moderate quality evidence; neurological side effects: RR 0.34, 95% CI 0.18 to 0.64; 453 participants, five trials; low quality evidence).

Five trials from Turkey, Spain and the UK compared mebendazole (200 mg three times daily for five to 10 days) with metronidazole (5 mg/kg (or 250 mg) three times daily for five to 10 days). These trials were small in size, and at high risk of bias. Consequently, reliable conclusions on the relative effectiveness cannot be made (very low quality evidence).

Five further trials, from Iran, Spain and Peru, have evaluated shortened regimens of tinidazole (single dose; 179 participants, three trials), metronidazole (single dose; 55 participants, one trial), and nitazoxanide (three days; 55 participants, one trial). Again, these trials were at high risk of bias and too small to reliably detect or exclude important differences (very low quality evidence).

Authors' conclusions

Albendazole may be of similar effectiveness to metronidazole, may have fewer side effects, and has the advantage of a simplified regimen. Large, high quality trials, assessing clinical outcomes (such as diarrhoea) will help assess further alternatives.

Résumé scientifique

Médicaments pour le traitement de la giardiase

Contexte

La giardiase peut être asymptomatique ou peut provoquer des diarrhées (parfois graves), une perte de poids, une mauvaise assimilation et, chez l'enfant, un retard de croissance staturo-pondérale. L'infection est généralement traitée au métronidazole administré trois fois par jour pendant 5 à 10 jours.

Objectifs

Evaluer l'efficacité relative d'autres schémas thérapeutiques antibiotiques pour le traitement de l'adulte ou de l'enfant atteint de giardiase symptomatique.

Stratégie de recherche documentaire

Nous avons effectué des recherches dans le registre spécialisé du groupe Cochrane sur les maladies infectieuses, le registre Cochrane des essais contrôlés (CENTRAL) (Numéro 6 2012), MEDLINE, EMBASE, LILACS et le portail de recherche du système d'enregistrement international des essais cliniques (3 juillet 2012).

Critères de sélection

Nous avons inclus les essais contrôlés randomisés (ECR) comparant le métronidazole administré pendant cinq à 10 jours à l'un des médicaments suivants : le métronidazole (dose unique), le tinidazole, l'albendazole, le mébendazole et le nitazoxanide. Les critères de jugement principaux étaient la guérison parasitologique et clinique.

Recueil et analyse des données

Deux auteurs ont évalué les études identifiées pour l’inclusion, procédé à l'évaluation du risque de biais et extrait les données de manière indépendante. Nous avons résumé les données avec les rapports de risques et les différences moyennes, et avons présenté les résultats sous la forme de Forest plots et procédé à des méta-analyses lorsque cela était possible. Nous avons évalué l'hétérogénéité au moyen du test du Chi2, de la statistique I2 et d'un examen visuel, et l'avons étudiée par des analyses en sous-groupes. Nous avons évalué la qualité des preuves en utilisant l'approche GRADE.

Résultats principaux

Nous avons inclus 19 essais, portant sur 1 817 participants, dont 1 441 étaient des enfants. Les études étaient généralement de petite taille et les méthodes utilisées étaient mal rapportées. . La plupart indiquaient les résultats parasitologiques plutôt que l'amélioration clinique.

Dix essais, réalisés en Inde, au Mexique, au Pérou, en Iran, à Cuba et en Turquie, comparaient l'albendazole (400 mg une fois par jour pendant cinq à 10 jours) au métronidazole (250 mg à 500 mg trois fois par jour pendant cinq à 10 jours). Cette posologie d'une dose par jour d'albendazole est vraisemblablement équivalente au métronidazole pour parvenir à une guérison parasitologique (RR 0,99, IC à 95 % 0,95 à 1,03 ; 932 participants, 10 essais ; preuves de qualité modérée) et améliorer les symptômes (RR 0,98, intervalle de confiance (IC) à 95 % 0,93 à 1,04 ; 483 participants, cinq trials ; preuves de qualité modérée), mais la durée du suivi était courte (deux à trois semaines). L'albendazole a vraisemblablement moins d'effets secondaires que le métronidazole (effets secondaires gastro-intestinaux : RR 0,29, IC à 95 % 0,13 à 0,63 ; 717 participants, huit essais ; preuves de qualité modérée ; effets secondaires neurologiques : RR 0,34, IC à 95 % 0,18 à 0,64 ; 453 participants, cinq essais ; preuves de qualité médiocre).

Cinq essais, réalisés en Turquie, en Espagne et au Royaume-Uni, comparaient le mébendazole (200 mg trois fois par jour pendant cinq à 10 jours) au métronidazole (5 mg/kg (ou 250 mg) trois fois par jour pendant cinq à 10 jours). Ces essais étaient de petite taille et présentaient un important risque de biais. En conséquence, il est impossible d'établir des conclusions fiables concernant l'efficacité relative (preuves de très médiocre qualité).

Cinq autres essais, réalisés en Iran, en Espagne et au Pérou, ont évalué des schémas posologiques raccourcis de tinidazole (dose unique ; 179 participants, trois essais), de métronidazole (dose unique ; 55 participants, un essai) et de nitazoxanide (trois jours ; 55 participants, un essai). Cette fois encore, ces essais présentaient un important risque de biais et étaient de trop petite taille pour détecter ou exclure de manière fiable des différences importantes (preuves de très médiocre qualité).

Conclusions des auteurs

L'albendazole pourrait avoir une efficacité semblable à celle du métronidazole, il pourrait avoir moins d'effets secondaires et présente l'avantage d'une posologie simplifiée. Des essais à grande échelle, de grande qualité, évaluant des critères d'évaluation cliniques (tels que la diarrhée), permettront d'évaluer d'autres alternatives.

Plain language summary

Drugs for treating giardiasis

Giardiasis is an infection of the small intestine caused by a microscopic organism called Giardia lamblia. The infection is passed from person to person by ingesting faecally contaminated water or food. Symptoms frequently include diarrhoea, abdominal pain, flatulence, bloating, vomiting, and weight loss. In this review, we assess alternatives to the most commonly used treatment for giardiasis; metronidazole given orally for five or more days.

We identified 19 trials involving 1817 participants, of which 1441 were children. Most trials had a small number of participants and were at high risk of bias. Albendazole is probably of similar effectiveness to metronidazole, probably has fewer side effects, and has the advantage of a simplified regimen. Large, high quality trials, assessing clinical outcomes are required to assess further alternatives.

Résumé simplifié

Médicaments pour le traitement de la giardiase

La giardiase est une infection de l'intestin grêle provoquée par un organisme microscopique appelé Giardia lamblia. L'infection est transmise de personne à personne par l'ingestion d'eau ou d'aliments contaminés par des excréments. Les symptômes comprennent fréquemment des diarrhées, des douleurs abdominales, des flatulences, des ballonnements, des vomissements et une perte de poids. Dans cette revue, nous évaluons des alternatives au traitement le plus couramment utilisé contre la giardiase, le métronidazole administré par voie orale pendant cinq jours ou plus.

Nous avons identifié 19 essais, portant sur 1 817 participants, dont 1 441 étaient des enfants. La plupart des essais avaient un petit nombre de participants et présentaient un important risque de biais. L'albendazole a une efficacité vraisemblablement semblable à celle du métronidazole ; il a vraisemblablement moins d'effets secondaires et présente l'avantage d'une posologie simplifiée. Des essais à grande échelle, de grande qualité, évaluant des critères d'évaluation cliniques, doivent être réalisés pour évaluer d'autres alternatives.

Notes de traduction

Traduit par: French Cochrane Centre 10th January, 2013
Traduction financée par: Minist�re Fran�ais des Affaires sociales et de la Sant�

Summary of findings(Explanation)

Summary of findings for the main comparison. 
  1. 1 These trials were conducted in adults and children in India, Mexico, Peru, Cuba, and Turkey. The dose of metronidazole was 250 mg to 500 mg three times daily for 5 to 10 days. The dose of albendazole was 400 mg once daily for five to 10 days.
    2 Downgraded by 1 for risk of bias as none of these trials adequately described allocation concealment or blinding.
    3 The finding of no difference between treatments is consistent across trials, and the 95% CI excludes clinically important differences.
    4 Downgraded by 1 for imprecision as the number of neurological side effects in these trials was very low.

Albendazole (5+ days) compared to metronidazole (5+ days) for treating giardiasis
Patient or population: People with giardiasis infection
Settings: Endemic settings
Intervention: Albendazole once daily for five or more days
Comparison: Metronidazole three times daily for five or more days
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
Metronidazole
(three times daily for 5 to 10 days)
Albendazole
(once daily for 5 to 10 days)
Parasitological cure (Follow-up: 1 to 2 weeks)91 per 10090 per 100
(86 to 94)
RR 0.99
(0.95 to 1.03)
932
(10 studies)
⊕⊕⊕⊝
moderate 1,2,3
 
Clinical improvement (Follow-up: 2 to 3 weeks)91 per 10089 per 100
(85 to 95)
RR 0.98
(0.93 to 1.04)
483
(5 studies)
⊕⊕⊕⊝
moderate 1,2,3
 
Gastrointestinal side effects (Follow-up: 1 to 3 weeks)29 per 1008 per 100
(4 to 18)
RR 0.29
(0.13 to 0.63)
717
(8 studies)
⊕⊕⊕⊝
moderate 1,2
 
Neurological side effects (Follow-up: 1 to 3 weeks)15 per 1005 per 100
(3 to 10)
RR 0.34
(0.18 to 0.64)
453
(5 studies)
⊕⊕⊝⊝
low 1,2,4
 
*The basis for the assumed risk (eg the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk Ratio
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Summary of findings 2

Summary of findings 2. 
  1. 1These trials were conducted in adults and children from Spain, Turkey, and the UK. The usual dose of mebendazole was 200 mg three times daily for five to 10 days, and metronidazole 5 mg/kg (or 250 mg) three times daily for five to 10 days.
    2 Downgraded by 1 for risk of bias. None of these trials adequately described allocation concealment or blinding.
    3 Downgraded by 2 for imprecision. The trials were very small and underpowered to detect differences. The 95% CI is wide and includes no difference between treatments.

Mebendazole (5+ days) compared to metronidazole (5+ days) for treating giardiasis
Patient or population: People with giardiasis infection
Settings: Endemic settings
Intervention: Mebendazole once daily for five or more days
Comparison: Metronidazole three times daily for five or more days
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
Metronidazole
(three times daily for 5 to 10 days)
Mebendazole
(three times daily for 5 to 10 days)

Parasitological cure

(Follow-up: 1 to 2 weeks)

90 per 10052 per 100
(29 to 95)
RR 0.58
(0.32 to 1.06)
142
(5 studies)
⊕⊝⊝⊝
very low 1,2,3
 
Clinical improvement---(0 studies)-Not reported
Side effects: nausea/vomiting8 per 1003 per 100
(0 to 15)
RR 0.32
(0.05 to 1.9)
100
(3 studies)
⊕⊝⊝⊝
very low 1,2,3
 
Side effects: headache8 per 1002 per 100
(0 to 17)
RR 0.24
(0.03 to 2.06)
74
(2 studies)
⊕⊝⊝⊝
very low 1,2,3
 
Serious adverse events---142
(5 studies)
-Not reported
*The basis for the assumed risk (eg the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk Ratio
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Summary of findings 3

Summary of findings 3. 
  1. 1These trials were conducted in Iran, Spain and Peru. The usual dose of tinidazole was 50mg/kg as a single dose, and metronidazole 5 mg/kg (or 250 mg) three times daily for five to 10 days.
    2 Downgraded by 1 for risk of bias. None of these trials adequately described allocation concealment or blinding.
    3 Downgraded by 2 for imprecision. The trials were very small and underpowered to detect differences. The 95% CI is wide and includes no difference between treatments.

Tinidazole (single dose) compared to metronidazole (5+ days) for treating giardiasis
Patient or population: People with giardiasis infection
Settings: Endemic settings
Intervention: Tinidazole single dose
Comparison: Metronidazole three times daily for five or more days
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
Metronidazole
(three times daily for 5 to 10 days)
Tinidazole
(single dose)
Parasitological cure
(Follow-up: 3 to 4 weeks)
73 per 10089 per 100
(63 to 100)
RR 1.23
(0.87 to 1.73)
179
(3 studies)
⊕⊝⊝⊝
very low 1,2,3
 
Clinical improvement
(Follow-up: 3 to 4 weeks)
76 per 10087 per 100
(76 to 100)
RR 1.15
(1 to 1.32)
179
(3 studies)
⊕⊝⊝⊝
very low 1,2,3
 
Gastrointestinal side effects
(Follow-up: 3 to 4 weeks)
8 per 1005 per 100
(2 to 13)
RR 0.58
(0.21 to 1.62)
179
(3 studies)
⊕⊝⊝⊝
very low 1,2,3
 
Neurological side effects
(Follow-up: 3 to 4 weeks)
4 per 1005 per 100
(1 to 22)
RR 1.3
(0.29 to 5.92)
138
(2 studies)
⊕⊝⊝⊝
very low 1,2,3
 
Serious adverse events---179
(3 studies)
-Not reported
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Background

Description of the condition

Giardia lamblia (also known as Lamblia intestinalis and Giardia duodenalis) is an intestinal protozoan parasite that infects a wide range of host species, including humans and domestic mammals.The prevalence of G. lamblia infection varies in different parts of the world, but it is generally higher in developing countries where reported prevalences range from 3% to 38% (Cifuentes 2000; Cifuentes 2004; Park 2004; Celiksöz 2005; Al-Saeed 2006; Barnawi 2007; de Souza 2007). Seroprevalence studies suggest that in some settings between 23% and 40% of children are infected before six months of age (Miotti 1986; Abdel 1991). In developed countries, G. lamblia infects about 2% of adults and between 6% to 8% of children, and is responsible for frequent waterborne outbreaks of diarrhoea (Garner 2001; Laupland 2005; Yoder 2007). Hörman 2004 estimated the prevalence of G. lamblia infection, determined by stool examination, in the Nordic countries (Denmark, Finland, Norway, and Sweden) by combining data from 13 studies; the estimated prevalence in asymptomatic populations was 2.97% (95% confidence interval (CI) 2.64 to 3.31) and 5.81% (95% CI 5.34 to 6.30) in symptomatic populations.

G. lamblia is transmitted by the ingestion of cysts excreted in the faeces of infected persons or animals. Risk factors for infection include consuming untreated or inadequately treated water (Chute 1987; Pinheiro 2011), travelling to disease-endemic areas (Jelinek 2000; Ekdahl 2005; Freedman 2006), age between one and nine years (particularly in childcare settings), close or intimate contact with infected people (Yoder 2007), poor hygienic conditions (eg living in a household without a latrine or with a mud floor in the sleeping rooms) (Mahmud 1995), contact with infected animals (Yoder 2007), and oral-anal sexual contact (Yoder 2007). Cysts can survive for several months in cold water, but once they are ingested and reach the small intestine, they develop into the disease-causing form (trophozoite). A prospective study suggested that the time between infection and the appearance of G. lamblia cysts in the stool is 12 to 19 days (Jokipii 1985). However, symptoms may appear between one and 75 days after infection and usually at six to 15 days from the date of infection. Symptoms are usually gastrointestinal disturbances (Rendtorff 1954; Brodsky 1974; Fricker 2007).

Giardia trophozoites (disease-causing forms) damage the epithelial absorptive surface and upper intestine. This results in the malabsorption of glucose, sodium, and water, and reduces disaccharidase activity, which can lead to diarrhoea and malnutrition. The trophozoites do not invade the surrounding tissues or enter the blood stream, but the immune response results in an increased inflammation of the intestine (Garner 2001; Buret 2007). Generally, giardiasis is a self-limiting clinical illness. The spectrum of giardiasis disease varies from asymptomatic carriage to severe diarrhoea, weight loss, malabsorption, and failure of children to thrive. Acute giardiasis is typically characterized by the sudden onset of acute, watery diarrhoea usually containing neither blood nor pus. Other symptoms may include abdominal cramps, malaise, nausea, vomiting, epigastric pain, bloating, weight loss, and weakness. Acute giardiasis generally resolves in one to four weeks, but may become chronic and persist for months, leading to malabsorption and malnutrition particularly in children (Yoder 2007). Chronic giardiasis may or may not have been preceded by an acute episode, and is typically characterized by intermittent or periodic episodes of diarrhoea, increased flatulence, epigastric pain, and weight loss (AAP 2006; Yoder 2007). The infection can be prolonged in people who are immunocompromised (Libanore 1991; Newman 2001). Malnutrition and intestinal parasitic infections are common among children in developing countries and have been associated with poor physical development and impaired resistance to infections (Goek 2003; Mukhopadhyay 2007).

Giardiasis is diagnosed by the identification of cysts or trophozoites in stool specimens, duodenal fluid, or small bowel biopsy. Repeated samplings may be necessary to find the parasite in symptomatic people. Different staining techniques of three separate stool specimens are frequently used to identify cysts; trophozoites are rarely identified in stools and are usually detected in duodenal biopsies (Mank 2001). Alternate methods used for detection of parasites include antigen detection tests by enzyme-linked immunosorbent assay (ELISA) and immunofluorescence. These tests have a sensitivity of 88% to 98% and a specificity of 87% to 100% (Schunk 2001; Garcia 2006; Weitzel 2006). About 85% to 90% of cases are detected when three separate stool samples are examined (Hiatt 1995). Stool microscopy is relatively inexpensive and commonly used. However if a single stool exam is performed, G. lamblia cysts may be missed resulting in under-diagnosis of cases. Duodenal aspirate biopsy (needle aspiration) is more invasive and requires an endoscopy. In direct comparison studies to stool microscopy, this method may have a lower diagnostic yield (Havenik 2007). Notably, the copro-antigen assays are less time-consuming, easier to perform, and more sensitive and specific (over 90%) than stool specimen examination (Schunk 2001; Garcia 2006; Weitzel 2006).

Description of the intervention

Many different treatments for giardiasis have been described, especially members of the 5-nitroimidazole family (metronidazole, secnidazole, ornidazole, and tinidazole) and the benzimidazole family (including albendazole and mebendazole)

An earlier Cochrane Review found that treatment options included nitroimidazoles derivatives, particularly a single dose of tinidazole and metronidazole treatment longer than three days (Zaat 1998); however, the included trials were of poor methodological quality. A recent non-Cochrane review reported that albendazole, when given as a single dose of 400 mg/day for five days, was comparable to that of metronidazole, with fewer side effects (Solaymani-Mohammadi 2010). Other reports of decreased susceptibility to metronidazole and poor compliance to metronidazole treatment have raised concerns and have led to the search for other treatments that are in use, such as tinidazole, nitazoxanide, albendazole, and mebendazole (Argüello-García 2004; Escobedo 2007; Solaymani-Mohammadi 2010).

How the intervention might work

The mode of action of the 5-nitroimidazoles is mediated by free nitro radicals, which are formed during the metabolic reduction of the drug and cause cytotoxicity (Argüello-García 2004). Treatment with metronidazole usually ranges from five to seven days (Solaymani-Mohammadi 2010). The benzimidazoles are broad-spectrum drugs that have been used for many years in the treatment of helminthiasis in animals and humans. Their primary mode of action against helminths is via inhibition of tubulin, which causes trophozoite detachment and distortion of morphology and general structure (Morgan 1993).

Why it is important to do this review

This Cochrane Review supersedes the original Cochrane Review on drugs for giardiasis (Zaat 1998). This update focuses on addressing the effects of treatments for symptomatic giardiasis in adults and children.

Objectives

To evaluate the relative effectiveness of alternative antibiotics for treating adults and children with symptomatic giardiasis.

Methods

Criteria for considering studies for this review

Types of studies

Randomized controlled trials (quasi-randomized controlled trials were excluded).

Types of participants

Adults and children with clinically diagnosed symptomatic giardiasis (gastrointestinal symptoms plus positive laboratory results).

Trials were eligible if they included participants with G. lamblia and other intestinal parasitic infections; or if they enrolled symptomatic and asymptomatic participants, and information regarding symptomatic participants could be extracted, or if at least 75% of participants were symptomatic.

Diagnosis of symptomatic giardiasis may have involved the examination of stool specimens or detection of Giardia copro-antigen in faeces, using direct fluorescent antibody (DFA), ELISA, or other methods.

Types of interventions

Intervention

Any of the following drugs used for treating giardiasis: metronidazole (single dose), tinidazole, albendazole, mebendazole, or nitazoxanide.

Control

Metronidazole for treating giardiasis, usually for five to 10 days.

Types of outcome measures

Primary outcomes
  • Parasitological cure (as defined by trialists).

  • Parasitological and clinical cure (as defined by trialists).

Parasitological cure refers to no G. lamblia trophozoites or cysts found in post-treatment in faecal specimens, or no detection of G. lamblia antigen in stool specimens using diverse methods.

Clinical cure refers to clinical improvements of symptoms such as abdominal pain and cramps, diarrhoea, weight loss, and malnutrition as diagnosed by the trialist.

Secondary (measures of clinical improvement)
  • Number of participants with cessation of abdominal pain, vomiting, or diarrhoea at specific times of follow up.

  • Reduction in symptoms of diarrhoea.

  • Relapse (as defined by trial authors).

  • Weight gain.

  • Dehydration.

  • Any other functional indicators such as quality of life, time off work.

  • Time to recovery (to clinical improvement).

Adverse events

  • Serious adverse events (those leading to hospitalizations and/or death).

  • Other adverse events (gastrointestinal, allergic/cutaneous, fever, neurological, hematological, hepatic, among others).

  • Non-compliance to treatment.

Search methods for identification of studies

We attempted to identify all relevant trials regardless of language or publication status (published, unpublished, in press, and in progress).

Electronic searches

We searched the following databases using the search terms detailed in Appendix 1: Cochrane Infectious Disease Group Specialized Register; Cochrane Central Register of Controlled Trials (CENTRAL), published in The Cochrane Library (Issue 6 2012); MEDLINE (1966 to July 2012); EMBASE (1980 to July 2012); and LILACS (1982 to July 2012). We also searched the International Clinical Trials Registry Platform Search Portal (ICTRP) using 'Giardia*' as the search term (July 2012).

Searching other resources

To help identify unpublished and ongoing trials, we contacted individual researchers working in the field and organizations, including the World Health Organization.

We also checked the reference lists of all studies identified by the above methods.

Data collection and analysis

Selection of studies

Two authors (CG, LR) assessed the article titles and abstracts identified from the literature searches for eligibility. If eligibility was unclear, we obtained the full text for assessment. We (CG, LR) decided individually which trials met the inclusion criteria and then we compared decisions to reach a consensus. In cases of disagreement, a third author (LU) was invited to comment. We recorded excluded studies and the reasons for exclusion. In certain cases, we contacted the authors of studies to obtain more information about the trials.

Data extraction and management

One author (CG or LR) extracted data using a pre-designed data extraction form, and a second author (LU or CC) independently cross-checked the data. We extracted data for all outcomes for all relevant drugs, paying particular attention to the dosage and periodicity of treatment. We extracted the number of participants randomized and the number of participants for which outcomes were measured for all outcomes in all treatment arms. For dichotomous data, we extracted the number of events and the number of participants in each treatment arm. For continuous data, we extracted the arithmetic means and standard deviations for each treatment group together with the number of participants in each group. We resolved disagreements by consensus.

Assessment of risk of bias in included studies

Two review authors (CG, LR, LU, or CC) independently assessed the risk of bias of each trial using The Cochrane Collaboration's risk of bias tool (Higgins 2008) and a pre-designed assessment form. We followed the guidance to make judgements on the risk of bias in six domains, focusing on the 'parasitological cure' outcome measure: sequence generation; allocation concealment; blinding of participants, personnel, and outcome assessors; incomplete outcome data; selective outcome reporting; and other sources of bias (eg trial stopped early, no sample size calculation). We categorized these judgements as 'yes' (low risk of bias), 'no' (high risk of bias), or 'unclear'. Where our judgement was unclear we attempted to contact the trial authors for clarification. We summarized the information into a risk of bias summary figure and risk of bias graph.

Measures of treatment effect

We expressed the results for dichotomous data using risk ratios and 95% CIs.

We expressed results for continuous outcomes using the mean difference and 95% CIs when the data were summarized by arithmetic means and standard deviations.

Dealing with missing data

We performed a complete case analysis, which included only patients with reported outcomes.

Assessment of heterogeneity

We inspected visually the forest plots to detect overlapping CIs. We tested for heterogeneity using a standard Chi2 test with significance being set at P < 0.1. We used the I2 test to estimate the total variation across trials due to heterogeneity rather than chance (Deeks 2008).

Assessment of reporting biases

We planned to assess publication bias with a visual inspection of a funnel plot if nine or more trials were meta-analysed.

Data synthesis

We analysed data using Review Manager 5 and stratified the analyses according to treatments and comparators.

A fixed-effect or random-effects model was used according to the heterogeneity assessment. If statistical heterogeneity was detected (Chi2 P < 0.1), we used the random-effects model for the pooled analysis.

Assessment of evidence quality

We assessed the quality of evidence using the GRADE methodology (Balshem 2011). The GRADE system considers quality to be a judgment of the extent to which we can be confident that the estimates of effect are correct. The level of quality is judged on a four-point scale. Evidence from randomized controlled studies is initially graded as high and downgraded by one, two or three levels after full consideration of any limitations in the design of the studies, the directness (or applicability) of the evidence, the consistency and precision of the results, and the possibility of publication bias.

The estimates of effect, and the GRADE assessments of our confidence in these estimates are displayed in 'summary of findings tables' for the main comparisons. Where we have downgraded the evidence our reasons for doing so are displayed in the footnotes.

When making conclusions about the relative effects of the interventions, we have used language which reflects the GRADE assessments and our confidence in the estimates, ie if the evidence was high quality we would say "albendazole is equivalent to...."; moderate quality "albendazole is probably equivalent..."; low quality ("albendazole may be equivalent to...."); and where the evidence is of very low quality we have not drawn conclusions.

Subgroup analysis and investigation of heterogeneity

Where appropriate, we attempted to investigate the following sources of heterogeneity using subgroup analyses: disease severity groups (ie hospitalized patients); participant age (children (aged less 18 years) versus adults (aged 18 or more years); pregnancy (yes or no); co-infection (yes or no); HIV/AIDS (positive/negative); cancer (yes or no); type of diagnostic methods (stool microscopy, copro-antigen detection, duodenal fluid, or small bowel biopsy); country classification (low income, lower middle income, upper middle income, and high income (World Bank 2009)).

Sensitivity analysis

If we identified significant heterogeneity in the trials' risk of bias, we performed a sensitivity analysis against risk of bias (low, unclear, or high risk of bias) when possible.

Results

Description of studies

Results of the search

We identified 1213 references using our search strategy (see Figure 1). After we removed duplicates, we had 1072 references remaining. We screened all 1072 references by abstract and identified 40 potential studies for inclusion. We assessed the full text articles of these 40 studies and we identified 19 studies which met the inclusion criteria, one of which had two publications (Misra 1995 IND).

Figure 1.

Study flow diagram.

Included studies

We identified 19 studies that met the inclusion criteria and included 1817 participants, of which 1441 were children. One study had two publications (Misra 1995 IND). The oldest trial was conducted in 1972 (Bhandari 1972 IND) and the most recent in 2012 (Cañete 2012 CUB). However, most trials were conducted during the 1990s.

Trial setting

Four studies were conducted in Turkey (Kalayci 1995 TUR; Bulut 1996 TUR; Karabay 2004 TUR; Yereli 2004 TUR), four in India (Bhandari 1972 IND; Hall 1993 IND; Dutta 1994 IND; Misra 1995 IND). three in Spain (Gascon 1989 ESP; Perez-Choliz 1989 ESP; Gascón 1990 ESP) and two in Peru (Chan Del Pino 1999 PER; Ortiz 2001 PER), Mexico (Romero-Cabello 1995 MEX; Rodríguez 1996 MEX) andIran (Alizadeh 2006 IRN; Fallah 2007 IRN); one study was conducted in the United Kingdom (Al-Waili 1992 GBR) and Cuba (Cañete 2012 CUB).

Participants

Fourteen studies were conducted exclusively in children (aged above three years); three were exclusively conducted in adults (Gascon 1989 ESP; Karabay 2004 TUR; Cañete 2012 CUB), two in both children and adults (Gascón 1990 ESP; Alizadeh 2006 IRN), and one did not report the proportion of included children (Alizadeh 2006 IRN).

The inclusion criteria of all studies included the presence of cysts or trophozoites, or both, in the faeces on stool examination. All studies included symptomatic participants and eight trials reported improvement of symptoms as an outcome (Perez-Choliz 1989 ESP; Dutta 1994 IND; Romero-Cabello 1995 MEX; Chan Del Pino 1999 PER; Ortiz 2001 PER; Yereli 2004 TUR; Fallah 2007 IRN; Cañete 2012 CUB).

Interventions

Five groups of RCTs were described comparing alternative drug regimens with metronidazole given three times daily for five to 10 days:

A full description of these studies is available in the "Characteristics of included studies" section.

Excluded studies

Twenty studies were excluded. The reasons for exclusion are reported in the "Characteristics of excluded studies" table.

Risk of bias in included studies

The risk of bias assessments are summarised in Figure 2 and Figure 3 and detailed in the "Characteristics of included studies" table.

Figure 2.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 3.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

Only four trials reported the method of random sequence generation, and were judged to be at low risk for this domain (Hall 1993 IND; Chan Del Pino 1999 PER; Ortiz 2001 PER; Cañete 2012 CUB). No information was presented for the remaining trials, although all were described as 'randomized'.

Only one trial adequately described allocation concealment to be at considered at low risk of bias (Alizadeh 2006 IRN).

Blinding

Only one trial adequately blinded participants and personnel to be considered at low risk of bias (Al-Waili 1992 GBR)

In seven trials (Al-Waili 1992 GBR; Hall 1993 IND; Kalayci 1995 TUR; Bulut 1996 TUR; Ortiz 2001 PER; Alizadeh 2006 IRN; Cañete 2012 CUB) the people who performed the stool examinations were blinded. However, the evaluators of other outcomes were only blinded in one (Al-Waili 1992 GBR).

Incomplete outcome data

Fifteen trials were judged to be at low risk for this domain, one was deemed to be high risk (Perez-Choliz 1989 ESP) and three had unclear risk of bias (Gascon 1989 ESP; Gascón 1990 ESP; Bulut 1996 TUR).

Selective reporting

Four trials showed some evidence of selective outcome reporting (Kalayci 1995 TUR; Romero-Cabello 1995 MEX; Chan Del Pino 1999 PER; Ortiz 2001 PER).

Other potential sources of bias

Seven studies were judged as having unclear risk of bias due to the lack of baseline data to asses the risk of confounding (Bhandari 1972 IND; Gascon 1989 ESP; Perez-Choliz 1989 ESP; Gascón 1990 ESP; Hall 1993 IND; Bulut 1996 TUR; Fallah 2007 IRN) while three other trials studies were considered as having high risk of bias due to baseline differences between groups (Misra 1995 IND; Rodríguez 1996 MEX; Chan Del Pino 1999 PER).

Effects of interventions

See: Summary of findings for the main comparison; Summary of findings 2; Summary of findings 3

Albendazole (once daily) versus metronidazole (three times daily)

Dosing: Albendazole was given as 400 mg once daily for five days. For metronidazole, the usual dosage was 5 mg/kg (250 mg or 500 mg in adults) three times daily for five to 10 days.

Four studies were from India, two from Turkey, two from Mexico, one from Peru, and one from Cuba. In addition, six of the included trials were conducted exclusively in children (see Characteristics of included studies).

No significant difference was found between treatments when assessing the rate of parasitological cure after one to three weeks of treatment (932 participants, 10 trials, Analysis 1.1). Although moderate heterogeneity was detected (I2 = 30%), this relates to one small study which found a statistically significant benefit with metronidazole (Chan Del Pino 1999 PER). One trial (Misra 1995 IND) found less time was needed to achieve parasitological cure in children treated with albendazole (Mean difference: -19.00, 95% CI -34.72 to -3.28; 57 participants, one trial, Analysis 1.2).

No significant differences were found in the rate of clinical improvement between treatments (483 participants, five trials, Analysis 1.3) and in the time needed for symptomatic improvement (97 participants, two trials, Analysis 1.4).

There were significantly fewer gastrointestinal mild adverse events in the albendazole group (RR 0.29, 95% CI 0.13 to 0.63; 717 participants, eight trials, Analysis 1.5). There were also significantly lower rates of neurological adverse events (RR 0.34, 95% CI 0.18 to 0.64; 453 participants, five trials, Analysis 1.6) and metallic taste (RR 0.03, 95% CI 0.01 to 0.13; 327 participants, three trials, Analysis 1.7). Other adverse events were uncommon and no significant differences were observed between groups (Analysis 1.8; Analysis 1.9; Analysis 1.10; Analysis 1.11).

One RCT assessed non-compliance to treatment (Karabay 2004 TUR); no significant difference was reported between groups (57 participants, one trial, Analysis 1.12).No serious adverse events were reported in either group.

Mebendazole (three times daily) versus metronidazole (three times daily)

Dosing: The usual dosage of mebendazole was 200 mg three times daily for five to 10 days. For metronidazole, the usual dosage was 5 mg/kg (or 250 mg) three times daily for five to 10 days.

We found two trials from Spain, two from Turkey and one from the United Kingdom.

When pooling all studies, no significant difference was found between mebendazole and metronidazole in the rate of parasitological cure at one or two weeks (142 participants, five trials, Analysis 2.1). Furthermore, no significant difference was found in the subgroup of children (100 participants, three trials, Analysis 2.1). Two studies recruiting mainly adults (Gascon 1989 ESP; Gascón 1990 ESP) found a significantly higher parasitological cure rate with metronidazole (RR 0.15, 95% CI 0.05 to 0.44; 42 participants, two trials, Analysis 2.1). However these studies were small and underpowered to reliably detect or exclude an effect. None of the studies reported clinical recovery.

No significant differences in the rate of mild adverse events was found between groups (Analysis 2.2; Analysis 2.3; Analysis 2.4). No severe adverse events were reported.

Tinidazole (single dose) versus metronidazole (three times daily)

Dosing: Tinidazole was given at 50 mg/kg as a single dose, (maximum dose 1.5 to 2 g). The dose of metronidazole was more variable; 15 to 25 mg/kg/day three times daily for seven to 10 days, and Perez-Choliz 1989 ESP repeated the course three times one week apart.

Two small studies from Iran and Spain found that tinidazole significantly increased parasitological cure compared to metronidazole, while one study from Peru found no difference with a 100% cure rate in both arms. Overall, there was no significant difference detected between groups (179 participants, three trials, Analysis 3.1).

No significant difference was found in the rate of clinical improvement (179 participants, three trials, Analysis 3.2).

The number of adverse events reported was not significantly different between the two treatment arms (Analysis 3.3; Analysis 3.4; Analysis 3.5; Analysis 3.6), and there were no serious adverse events reported from any of the three trials.

Metronidazole (single dose) versus metronidazole (three times daily)

Dosing: The dosage of metronidazole was 35 mg/kg single dose, repeated at two weeks, versus metronidazole 20 mg/kg/day three times daily for 5 days, repeated at two weeks.

In this one small trial in Indian children, both treatment regimens achieved a 100% parasitological cure rate when assessed at two weeks (54 participants, one trial, Analysis 4.1).

Side-effects were rare during the two weeks follow-up with no differences between the groups (54 participants, one trial, Analysis 4.2). There were no serious adverse events reported.

Nitazoxanide (twice daily) versus metronidazole (twice daily)

Dosing: Nitazoxanide was given for three days (100 to 200 mg twice daily) and metronidazole for 5 days (125 to 250 mg twice daily).

In one trial conducted in Peru, no significant differences were found in the rates of parasitological cure (110 participants, one trial, Analysis 5.1), improvement of diarrhoea (110 participants, one trial, Analysis 5.2) or mild adverse events (Analysis 5.3; Analysis 5.4; Analysis 5.5; Analysis 5.6).

Discussion

Summary of main results

This review of interventions for treating giardiasis found 19 RCTs incorporating 1817 participants, of which 1441 were children.

Once-daily albendazole is probably equivalent to metronidazole given three times daily at achieving parasitological cure (moderate quality evidence), and improving symptoms (moderate quality evidence), but the duration of follow-up was short (two to three weeks). Albendazole probably also has fewer gastrointestinal and neurological side effects (moderate quality evidence).

Mebendazole has been evaluated in five studies but these are poorly conducted and too small to make reliable conclusions on the relative effectiveness (very low quality evidence).

Five further trials have evaluated shortened regimens of tinidazole, metronidazole, and nitazoxanide, but again these were at high risk of bias and too small to reliably detect or exclude important differences (very low quality evidence).

Overall completeness and applicability of evidence

The studies included in this review are from a variety of clinical settings across many different countries. Although the major findings can probably be generalized to elsewhere, several important limitations of these data should be noted; only one trial included HIV positive participants (Gascon 1989 ESP), only one included malnourished children (Chan Del Pino 1999 PER), and none of the trials included pregnant women. Further trials in these groups may be helpful.

Quality of the evidence

The quality of evidence was assessed using the GRADE methodology and is displayed in the three summary of findings tables:Summary of findings for the main comparison; Summary of findings 2; Summary of findings 3.

The evidence that albendazole is equivalent to metronidazole for treating giardiasis is considered to be of moderate quality. This implies that we can have reasonable confidence in the estimates but that further research may change the result. This evidence was downgraded due to concerns about the high risk of bias of all the trials comparing these two drugs. However, despite this risk, there was a consistent finding of no statistical or clinically important difference between the drugs across eight trials.

The evidence comparing all other regimens with metronidazole is considered to be of very low quality due to both the high risk of bias of the included trials, and the trials being too small, and significantly underpowered, to reliably detect or exclude a difference. Reliable conclusions on the relative efficacy of tinidazole, nitazoxanide, or single dose metronidazole can therefore not be made.

Potential biases in the review process

We have attempted to avoid publication bias by searching for both published and unpublished studies. However, we did not identify any and we are not aware of any that exist.

Most of the trials included in this review are at unclear or high risk for selection bias or reporting bias as a result of inadequate randomization or blinding. Inclusion of these trials in meta-analyses can produce biased or misleading conclusions. However, we have used the GRADE approach to assessing quality of evidence to allow us to express our decreased confidence in the results of our meta-analysis as a consequence of these faults. Many of the trials are also at unclear or high risk for selective outcome reporting as important outcomes, such as clinical improvement and adverse events, were frequently not reported.

Agreements and disagreements with other studies or reviews

Solaymani-Mohammadi and colleagues published a similar systematic review of albendazole versus metronidazole for treating giardiasis in 2010, and concluded that 'the effectiveness of albendazole..,was comparable to that of metronidazole', and 'patients treated with albendazole tended to have fewer side effects compared with those who took metronidazole's (Solaymani-Mohammadi 2010). Our review includes one additional study for this comparison, and confirms this conclusion.

Authors' conclusions

Implications for practice

Albendazole given once daily is probably as effective as metronidazole given three times daily for treating giardiasis in both children and adults, but with the advantages of a simplified regimen and fewer side effects.

Implications for research

This systematic review has identified the need for well-designed, adequately powered RCTs to assess the benefits and harms of treatments for symptomatic giardiasis. Many questions remain open and some important considerations for future research are as follows.

  1. Clinical outcomes and compliance need to be better addressed and considered for study sample size calculations.

  2. Studies are needed to determine the effects in specific populations such as pregnant women, people infected with human immunodeficiency virus, and malnourished children.

  3. Studies should follow the CONSORT guidelines for randomized trials (Schulz 2010): http://www.consort-statement.org/).

Acknowledgements

We are grateful to the editorial base for the Cochrane Infectious Diseases Group, which is funded by the UK Department for International Development (DFID) for the benefit of developing countries. Dr David Sinclair was the academic editor for this review.

We want to express our gratitude to Luis Gabriel Cuervo who was one of the authors of the protocol. We are also grateful to the Iberoamerican Cochrane Centre for their support.

Data and analyses

Download statistical data

Comparison 1. Albendazole (once daily) versus metronidazole (three times daily)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Parasitological cure (at 1 to 3 weeks)10932Risk Ratio (M-H, Fixed, 95% CI)0.99 [0.95, 1.03]
2 Time to achieve parasitological cure (hours)157Mean Difference (IV, Fixed, 95% CI)-19.0 [-34.72, -3.28]
3 Clinical improvement (at 2 to 3 weeks)5483Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.93, 1.04]
4 Time of symptomatic improvement (hours)297Mean Difference (IV, Fixed, 95% CI)-5.69 [-22.10, 10.73]
5 Other adverse events (gastrointestinal at 1 to 3 weeks)8717Risk Ratio (M-H, Random, 95% CI)0.29 [0.13, 0.63]
6 Other adverse events (neurological at 1 to 3 weeks)5453Risk Ratio (M-H, Fixed, 95% CI)0.34 [0.18, 0.64]
7 Other adverse events (metallic taste at 3 weeks)3327Risk Ratio (M-H, Fixed, 95% CI)0.03 [0.01, 0.13]
8 Other adverse events (allergic/cutaneous at 3 weeks)2154Risk Ratio (M-H, Fixed, 95% CI)0.2 [0.01, 4.08]
9 Other adverse events (fever at 3 weeks)134Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.23, 4.27]
10 Other adverse events (transient elevation of liver enzymes at 3 weeks)1150Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.06, 15.69]
11 Other adverse (transient neutropenia at 3 weeks)1150Risk Ratio (M-H, Fixed, 95% CI)5.0 [0.24, 102.42]
12 Non-compliance to treatment (at 2 weeks)157Risk Ratio (M-H, Fixed, 95% CI)0.74 [0.27, 2.06]
Analysis 1.1.

Comparison 1 Albendazole (once daily) versus metronidazole (three times daily), Outcome 1 Parasitological cure (at 1 to 3 weeks).

Analysis 1.2.

Comparison 1 Albendazole (once daily) versus metronidazole (three times daily), Outcome 2 Time to achieve parasitological cure (hours).

Analysis 1.3.

Comparison 1 Albendazole (once daily) versus metronidazole (three times daily), Outcome 3 Clinical improvement (at 2 to 3 weeks).

Analysis 1.4.

Comparison 1 Albendazole (once daily) versus metronidazole (three times daily), Outcome 4 Time of symptomatic improvement (hours).

Analysis 1.5.

Comparison 1 Albendazole (once daily) versus metronidazole (three times daily), Outcome 5 Other adverse events (gastrointestinal at 1 to 3 weeks).

Analysis 1.6.

Comparison 1 Albendazole (once daily) versus metronidazole (three times daily), Outcome 6 Other adverse events (neurological at 1 to 3 weeks).

Analysis 1.7.

Comparison 1 Albendazole (once daily) versus metronidazole (three times daily), Outcome 7 Other adverse events (metallic taste at 3 weeks).

Analysis 1.8.

Comparison 1 Albendazole (once daily) versus metronidazole (three times daily), Outcome 8 Other adverse events (allergic/cutaneous at 3 weeks).

Analysis 1.9.

Comparison 1 Albendazole (once daily) versus metronidazole (three times daily), Outcome 9 Other adverse events (fever at 3 weeks).

Analysis 1.10.

Comparison 1 Albendazole (once daily) versus metronidazole (three times daily), Outcome 10 Other adverse events (transient elevation of liver enzymes at 3 weeks).

Analysis 1.11.

Comparison 1 Albendazole (once daily) versus metronidazole (three times daily), Outcome 11 Other adverse (transient neutropenia at 3 weeks).

Analysis 1.12.

Comparison 1 Albendazole (once daily) versus metronidazole (three times daily), Outcome 12 Non-compliance to treatment (at 2 weeks).

Comparison 2. Mebendazole (three times daily) versus metronidazole (three times daily)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Parasitological cure (at 1 to 2 weeks)5142Risk Ratio (M-H, Random, 95% CI)0.58 [0.32, 1.06]
1.1 Children3100Risk Ratio (M-H, Random, 95% CI)0.85 [0.62, 1.17]
1.2 Children and adults242Risk Ratio (M-H, Random, 95% CI)0.15 [0.05, 0.44]
2 Other adverse events (nausea/vomiting at 1 to 2 weeks)3100Risk Ratio (M-H, Fixed, 95% CI)0.32 [0.05, 1.90]
3 Other adverse events (headache at 1 to 2 weeks)274Risk Ratio (M-H, Fixed, 95% CI)0.24 [0.03, 2.06]
4 Other adverse events (urticaria at 1 to 2 weeks)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
Analysis 2.1.

Comparison 2 Mebendazole (three times daily) versus metronidazole (three times daily), Outcome 1 Parasitological cure (at 1 to 2 weeks).

Analysis 2.2.

Comparison 2 Mebendazole (three times daily) versus metronidazole (three times daily), Outcome 2 Other adverse events (nausea/vomiting at 1 to 2 weeks).

Analysis 2.3.

Comparison 2 Mebendazole (three times daily) versus metronidazole (three times daily), Outcome 3 Other adverse events (headache at 1 to 2 weeks).

Analysis 2.4.

Comparison 2 Mebendazole (three times daily) versus metronidazole (three times daily), Outcome 4 Other adverse events (urticaria at 1 to 2 weeks).

Comparison 3. Tinidazole (single dose) versus metronidazole (three times daily)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Parasitological cure (at 3 weeks to 1 month)3179Risk Ratio (M-H, Random, 95% CI)1.23 [0.87, 1.73]
2 Clinical improvement (at 3 weeks to 1 month)3179Risk Ratio (M-H, Fixed, 95% CI)1.20 [1.06, 1.36]
3 Other adverse events (gastrointestinal at 3 weeks to 1 month)3179Risk Ratio (M-H, Fixed, 95% CI)0.58 [0.21, 1.62]
4 Other adverse events (neurological at 3 weeks to 1 month)2138Risk Ratio (M-H, Fixed, 95% CI)1.30 [0.29, 5.92]
5 Other adverse events (allergic/cutaneous at 3 weeks to 1 month)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
6 Other adverse events (fever at 3 weeks to 1 month)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
7 Serious adverse events (at 3 weeks to 1 month)3179Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
Analysis 3.1.

Comparison 3 Tinidazole (single dose) versus metronidazole (three times daily), Outcome 1 Parasitological cure (at 3 weeks to 1 month).

Analysis 3.2.

Comparison 3 Tinidazole (single dose) versus metronidazole (three times daily), Outcome 2 Clinical improvement (at 3 weeks to 1 month).

Analysis 3.3.

Comparison 3 Tinidazole (single dose) versus metronidazole (three times daily), Outcome 3 Other adverse events (gastrointestinal at 3 weeks to 1 month).

Analysis 3.4.

Comparison 3 Tinidazole (single dose) versus metronidazole (three times daily), Outcome 4 Other adverse events (neurological at 3 weeks to 1 month).

Analysis 3.5.

Comparison 3 Tinidazole (single dose) versus metronidazole (three times daily), Outcome 5 Other adverse events (allergic/cutaneous at 3 weeks to 1 month).

Analysis 3.6.

Comparison 3 Tinidazole (single dose) versus metronidazole (three times daily), Outcome 6 Other adverse events (fever at 3 weeks to 1 month).

Analysis 3.7.

Comparison 3 Tinidazole (single dose) versus metronidazole (three times daily), Outcome 7 Serious adverse events (at 3 weeks to 1 month).

Comparison 4. Metronidazole (single dose) versus metronidazole (three times daily)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Parasitological cure (at 2 weeks)154Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.93, 1.08]
2 Other adverse events (gastrointestinal at 2 weeks)154Risk Ratio (M-H, Fixed, 95% CI)2.70 [0.26, 27.96]
3 Serious adverse events (at 2 weeks)154Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
Analysis 4.1.

Comparison 4 Metronidazole (single dose) versus metronidazole (three times daily), Outcome 1 Parasitological cure (at 2 weeks).

Analysis 4.2.

Comparison 4 Metronidazole (single dose) versus metronidazole (three times daily), Outcome 2 Other adverse events (gastrointestinal at 2 weeks).

Analysis 4.3.

Comparison 4 Metronidazole (single dose) versus metronidazole (three times daily), Outcome 3 Serious adverse events (at 2 weeks).

Comparison 5. Nitazoxanide (twice daily) versus metronidazole (twice daily)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Parasitological cure (at 1 week)1110Risk Ratio (M-H, Random, 95% CI)0.95 [0.76, 1.20]
2 Reduction in symptoms of diarrhea (at 1 week)1110Risk Ratio (M-H, Random, 95% CI)1.07 [0.90, 1.27]
3 Other adverse events (at 1 week)1110Risk Ratio (M-H, Fixed, 95% CI)1.1 [0.51, 2.38]
4 Other adverse events (abdominal pain at 1 week)1110Risk Ratio (M-H, Fixed, 95% CI)1.67 [0.65, 4.27]
5 Other adverse events (nausea/vomiting at 1 week)1110Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.01, 8.01]
6 Other adverse events (headache at 1 week)1110Risk Ratio (M-H, Fixed, 95% CI)0.14 [0.01, 2.70]
Analysis 5.1.

Comparison 5 Nitazoxanide (twice daily) versus metronidazole (twice daily), Outcome 1 Parasitological cure (at 1 week).

Analysis 5.2.

Comparison 5 Nitazoxanide (twice daily) versus metronidazole (twice daily), Outcome 2 Reduction in symptoms of diarrhea (at 1 week).

Analysis 5.3.

Comparison 5 Nitazoxanide (twice daily) versus metronidazole (twice daily), Outcome 3 Other adverse events (at 1 week).

Analysis 5.4.

Comparison 5 Nitazoxanide (twice daily) versus metronidazole (twice daily), Outcome 4 Other adverse events (abdominal pain at 1 week).

Analysis 5.5.

Comparison 5 Nitazoxanide (twice daily) versus metronidazole (twice daily), Outcome 5 Other adverse events (nausea/vomiting at 1 week).

Analysis 5.6.

Comparison 5 Nitazoxanide (twice daily) versus metronidazole (twice daily), Outcome 6 Other adverse events (headache at 1 week).

Appendices

Appendix 1. Search methods: detailed search strategies

Search setCIDG SRaCENTRALMEDLINEbEMBASEbLILACSb
1Giardia lambliaGIARDIA LAMBLIAGIARDIA LAMBLIAGIARDIA-LAMBLIAGIARDIA LAMBLIA
2giardiasisGIARDIASISGIARDIASISGIARDIASISGIARDIASIS
31 or 21 OR 21 OR 21 OR 21 OR 2
4CARBAMATESCARBAMIC-ACID DERIVATIVECARBAMATES
5BENZIMIDAZOLESBENZIMIDAZOLE-DERIVATIVEBENZIMIDAZOLES
6NITROIMIDAZOLESNITROIMIDAZOLE-DERIVATIVENITROIMIDAZOLES
7IMIDAZOLESIMIDAZOLE-DERIVATIVEIMIDAZOLES
8ANTITRICHOMONAL AGENTSANTITRICHOMONAL AGENTANTITRICHOMONAL AGENTS
9albendazolealbendazolealbendazole
10metronidazolemetronidazolemetronidazole
11mebendazolemebendazolemebendazole
12tinidazoletinidazoletinidazole
13nitazoxanidenitazoxanidenitazoxanide
144-13/OR4-13/OR4-13/OR
153 AND 143 AND 143 AND 14
16Limit 15 to humansLimit 15 to humans

aCochrane Infectious Diseases Group Specialized Register.
bSearch terms used in combination with the search strategy for retrieving trials developed by The Cochrane Collaboration (Lefebvre 2008); upper case: MeSH or EMTREE heading; lower case: free text term.

Contributions of authors

Carlos Granados conceived the review question, developed and coordinated the protocol and the review, completed the first draft of the review, assessed the studies, extracted and analysed data, approved the final version prior to submission, and is guarantor for the review.

Ludovic Reveiz developed the protocol and the review, completed the first draft of the review, assessed the studies, extracted and analysed data, and approved the final version prior to submission. He performed part of the writing/editing of the review. He contributed to this systematic review in a personal capacity and during his spare time. Most of the work was performed before joining the Pan American Health Organization. The Pan American Health Organization does not assume responsibility for the statements contained therein.

Claudia Criollo developed the protocol and the review, completed the first draft of the review, assessed the studies, extracted data, and approved the final version prior to submission.

Luis Uribe developed the protocol and the review, completed the first draft of the review, assessed the studies, extracted data, and approved the final version prior to submission.

Declarations of interest

None known.

Differences between protocol and review

We have modified the 'comparison' as a result of suggestions from the editorial base of the Cochrane Infectious Diseases Group. We only included studies assessing the standard treatment (metronidazole for five or more days) against a number of frequently used drugs.

Luis Gabriel Cuervo was an author of the protocol.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Al-Waili 1992 GBR

MethodsUnited Kingdom. No power calculation made. Intention to treat analysis. Duration: 21 days.
Participants

44 children (28 males); age: three to 13 years.

Symptoms: anorexia, altered bowel habits, abdominal cramps, flatulence.

Asymptomatic: 0%.

Microscopic stools examination: cysts in all cases and trophozoites in only two cases. Technique: not mentioned.

Exclusion criteria: not mentioned.

Interventions

Rx 1. Mebendazole: 200 mg three times daily, 5 days (oral).

Rx 2: Metronidazole: 200 mg three times daily, 5 days (oral).

Outcomes

Parasitological cure: fecal specimens were examined on days 7 and 14 after treatment. Technique: not mentioned.

Any side effects.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"… were randomly to receive either mebendazole or metronidazole..".
Allocation concealment (selection bias)Unclear riskNot mentioned.
Blinding of outcome assessment (detection bias)
Stools examination
Low risk"The examinations were done blindly.."
Blinding of outcome assessment (detection bias)
Other outcomes
Low risk"…. and the investigators were unaware of the nature of the drugs used..".
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo withdrawals or loss to follow up.
Selective reporting (reporting bias)Unclear riskNo mention of a systematic strategy for collecting side effects.
Other biasLow riskNo comments.

Alizadeh 2006 IRN

MethodsIran. Intention to treat analysis. Duration: 10 days.
Participants

120 participants (61 males); age: two to 53 years, mean 22.3 ± 11 years.

Symptoms: were not specified. Asymptomatic: 0%.

Inclusion criteria: acute symptoms of giardiasis and a positive stool examination report for trophozoites of the pathogen.

Microscopic stools examination: cysts in all cases and trophozoites in only two cases. Technique: not mentioned.

Exclusion criteria: not mentioned.

Interventions

Rx 1. Albendazole: 400 mg single dose for 5 days (oral).

Rx 2. Metronidazole: 250 mg three times daily for 5 days (oral).

Outcomes

"A week after starting the treatment, stool examination to detect trophozoites of G. lamblia was performed three times in three consecutive days for all the patients". Technique: not mentioned.

Parasitological cure: negative stool examination for trophozoites and cysts.

Any side effects.

NotesSample size calculation: "(alpha level of 0.05 and beta level of 0.2), the sample size needed was calculated to be 120 documented cases of the disease"
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"before starting the study one number was assigned for each case. All the numbers were assigned in two groups at random..."
Allocation concealment (selection bias)Low risk"and a physician was assigned to give the needed numbers to the clinics".
Blinding of outcome assessment (detection bias)
Stools examination
Low risk"one parasitologist who was blinded to the treatment received was responsible for detecting the trophozoites in the samples before and after treatment".
Blinding of outcome assessment (detection bias)
Other outcomes
Unclear riskNot mentioned.
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll 120 paediatric patients were included in the statistical analysis.
Selective reporting (reporting bias)Unclear riskOne of the expected outcomes is symptomatic improvement, but it was not reported.
Other biasLow riskNo comments.

Bhandari 1972 IND

MethodsIndia. No power calculation made. Intention to treat analysis. Duration: 21 weeks.
Participants

54 participants (41 males), from 18 months to 12 years old; There is not any specification about symptoms. However, improvement on clinical symptoms was reported as outcome.

Inclusion criteria: positive stools reports on two successive days.

Exclusion criteria: not mentioned.

Interventions

Rx 1. Metronidazole: 35 mg/K single dose, repeated at two weeks (oral).

Rx 2. Metronidazole: 20 mg/K/d three times daily for 5 days, repeated at two weeks (oral).

Metronidazole was administrated mixed with honey, after meals.

Outcomes

Clinical assessment and stools were repeated on days four, eight,12 and 16 post-treatment. Any side effect volunteered by the patient was noted.

Parasitological cure: stool negative for G. lamblia. The study does not specify the technique of stool examination.

Cure: stool negative for G. lamblia, and complete relief of symptoms or marked clinical improvement.

NotesAny patient with treatment failure was given a third regimen of therapy, which was not specified. None of the patients received treatment before nine weeks.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"Treatment schedules in a random manner".
Allocation concealment (selection bias)Unclear riskNot mentioned.
Blinding of outcome assessment (detection bias)
Stools examination
Unclear riskNot mentioned.
Blinding of outcome assessment (detection bias)
Other outcomes
High riskNot mentioned. Probably not, this study compares therapy with metronidazole, 5 days versus 1 day. There is no mention about placebo plus metronidazole in the second group.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing outcome data.
Selective reporting (reporting bias)Unclear riskNo mention of a systematic strategy for collecting side effects.
Other biasUnclear riskCharacteristics of included patients were not described in a comparative manner.

Bulut 1996 TUR

MethodsTurkey. No power calculation made. Per protocol analysis. Duration: 14 days.
Participants

Information only on 48 participants (30 males); age: six to 12 years.

Symptoms: two children had diarrhoea. "Nonspecific gastrointestinal symptoms, such as abdominal pain, nausea, and viscid stool, were found in the remaining 46".

Microscopic stools examination: "standard methods" in fresh stool samples. Technique: not mentioned.

Exclusion criteria: not mentioned.

Interventions

Rx 1. Mebendazole 100 mg three times daily for one day.

Rx 2: Mebendazole 100 mg three times daily for seven days.

Rx 3: Metronidazole 15mg/kg single dose for seven days.

Rx 4: Ornidazole 40 mg/day single dose.

Outcomes

Parasitological cure: absence of cystic or trophozoite forms of G. lamblia in the stools (three tests) at two weeks. Technique: not mentioned.

Adverse effects.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskRandomization without more information.
Allocation concealment (selection bias)Unclear riskNot reported.
Blinding of outcome assessment (detection bias)
Stools examination
Low risk"fresh stool samples were examined microscopically for parasites by an examiner blinded".
Blinding of outcome assessment (detection bias)
Other outcomes
High risk"open label".
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskStool test were not done in 12 children. However authors reported that symptoms disappeared. No specific information was provided.
Selective reporting (reporting bias)Unclear riskSymptoms outcomes are mentioned but not reported.
Other biasUnclear riskNo information concerning baseline characteristics for each groups.

Cañete 2012 CUB

MethodsCuba. Duration: seven days.
Participants

150 participants (73 male); age: 18 to 38 years, mean 30 years.

Symptoms: were not specified. Asymptomatic: 0%.

Inclusion criteria: adults (older than 18 years) with acute symptoms of giardiasis (with or without diarrhoea) and a positive stool examination report. No pregnant women were included.

Microscopic stools examination: cysts or trophozoites.

Technique: "microscopic examination of fecal samples, as wet mounts and/or after Ritchie concentration".

Exclusion criteria: (1) history of sensitivity to benzimidazole compounds or, (2) antiparasitic drugs in the preceding four weeks or, (3) diseases other than giardiasis or, (4) attending other clinical trial.

Interventions

Rx 1. Albendazole: 400 mg single dose for five days (oral).

Rx 2. Metronidazole: 250 mg three times daily for 5 five days (oral).

Outcomes

Stool samples from all cases were examined on the 3rd, 5th and 7th days following completion of treatment.

Adverse events: any reported.

NotesSample size calculation: "statistical power 80%, confidence level 95%, cure rate of alternative drug of 70% and 90% of efficacy for the drug applied to the control group ... "
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"A random-number table was used to allocate each of these 150 adults ..."
Allocation concealment (selection bias)Unclear risk" a random table was developed to (blindly) identify with a number the drug to be administered to each participant, only disclosed at the end of the study".
Blinding of outcome assessment (detection bias)
Stools examination
Low risk"Neither investigators nor patients had information on the drug used throughout the study".
Blinding of outcome assessment (detection bias)
Other outcomes
Unclear risk"Neither investigators nor patients had information on the drug used throughout the study". The frequency of administration of both drugs was different.
Incomplete outcome data (attrition bias)
All outcomes
Low risk"All patients who began treatment completed the study period and were included in the statistical analysis".
Selective reporting (reporting bias)Low riskApparently, all relevant outcomes were reported. Protocol was not available.
Other biasLow riskNo comments.

Chan Del Pino 1999 PER

MethodsPeru. No power calculation was done. Per protocol analysis. Duration: 3 weeks.
Participants

79 participants. The mean age was 7.83 ± 2.66 years. The mean weight was 24.06 ± 7.48kg. 38% were preschool children. 67% were malnourished first degree.  

Clinical manifestations were abdominal pain (89.87%), diarrhoea (81.01%), hyporexia (70.88%), and abdominal distension (59,49%). 

Inclusion criteria:

  • Children infected with G. lamblia only. Confirmed by parasitological stool examinations.

  • Children, both sexes, aged three to 15 years of age.

  • Consent of parents or guardians of children.

Exclusion criteria:

  • Children with malnutrition grade II or III.

  • Children who have received antigiardial drugs in the past two months.

  • Children with acute febrile illness.

Interventions

Rx 1. Tinidazole: 50 mg/kg/day, single dose (oral).

Rx 2. Metronidazole: 15 mg/kg/day, three times daily for 10 days (oral).

Rx 3. Albendazole: 400 mg/day for five days (oral).

Rx 4. Furazolidone: 5mg/kg/day, every six hours, for 10 days.

Rx 5. Secnidazole: 30 mg/kg/day, single dose (oral).

Outcomes

Clinical signs and symptoms (cured, improvement, not cured, relapse).

Parasitological (disappearance of cysts or trophozoites, reinfection, infection by other parasites). Coproparasitology examination: direct microscopy, using the technique of Willis. The samples were collected during the first seven days, and days 14 and 21.

Global efficacy (excellent: parasitological cured at 3 days; Good: parasitological cured between three and seven days; moderate: parasitological cured after seven days; bad: parasite presence ).

Tolerance (excellent: no adverse events; good: one mild adverse event with no discontinuation; etc).

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandom number table.
Allocation concealment (selection bias)High riskFirst, children were randomized to 5 groups and drug subsequently randomized to these groups. 16 children were subsequently excluded after being randomized.
Blinding of outcome assessment (detection bias)
Stools examination
High risk"We performed a prospective, longitudinal, comparative, open and randomised study ..."
Blinding of outcome assessment (detection bias)
Other outcomes
High risk"We performed a prospective, longitudinal, comparative, open and randomised study ..."
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo significant differences in loss to follow up.
Selective reporting (reporting bias)Low riskThe study report include expected outcomes, including those that were pre-specified.
Other biasHigh riskRaw water intake was higher in the groups of children assigned to albendazole and secnidazole.

Dutta 1994 IND

MethodsIndia (multicentre study). No power calculation made. Intention to treat analysis. Duration: 21 days.
Participants

150 hospitalised participants (79 males), from two to 10 years old. Diarrhea, abdominal pain and anorexia were the main presenting symptoms.

Inclusion criteria: children (two to 10 years) with proven giardia infection (trophozoites and/or cyst of G lamblia in stool specimens).

Exclusion criteria: "... acute febrile conditions, chronic diarrhoea, severe malnutrition or receiving long term therapy". Treatment of Giardia or intestinal helminthic infection in the previous seven days.

Interventions

Rx 1. Albendazole: 400 mg/d, single dose, for five days (oral suspension).

Rx 2. Metronidazole: 7.5mg/kg three times daily (22.5 mg/kg/day) for five days (oral suspension).

Outcomes

Clinical assessment and stools were repeated on days one to seven, 14 and 21 post-treatment. Stool samples were examined microscopically (direct smear and formal ether concentration tests).

Hematological and biochemical parameters on the 7th and 14th day after treatment.

Reported side-effects.

NotesThis study has a similar methodology to the Misra 1995 IND study, but different patients (different hospitals).
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot stated.
Allocation concealment (selection bias)Unclear risk"eligible children were randomised according to pre-determined lists at each site".
Blinding of outcome assessment (detection bias)
Stools examination
Unclear riskNot mentioned.
Blinding of outcome assessment (detection bias)
Other outcomes
High risk"Patient's response to therapy and severity of side-effects were recorded by a co-investigator who was unaware of the treatment given to the patient". However, patients were probably aware about treatment.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing outcome data.
Selective reporting (reporting bias)Unclear riskNo mention of a systematic strategy for collecting side effects.
Other biasLow riskNo comments.

Fallah 2007 IRN

MethodsIran. No power calculation made. Intention to treat analysis. Duration: three weeks.
Participants

106 participants (69 males). Children, 75% of them were less than 12 years old.

"... from public and private health centres ...". There is no any specification about symptoms

Inclusion criteria: Giardia intestinalis-positive children.

Exclusion criteria: not mentioned.

Interventions

Rx 1. Tinidazole: 50 mg/kg, maximum 2 g, once a day, single dose (tablets).

Rx 2. Metronidazole: 15 mg/kg/day, three times daily for seven days (suspension).

Outcomes

Clinical and parasitological follow-up was carried out before, and at 7, 14, 21 days after treatment.

Parasitological cure was documented when there were three consecutive negative stool examination for G. intestinalis at 1-3 weeks after therapy termination. Stool examination was done by formalin ether concentration technique.

Duration of symptoms of patients.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"were randomly allocated to two groups".
Allocation concealment (selection bias)High risk"a randomised, open-label, clinical trial".
Blinding of outcome assessment (detection bias)
Stools examination
High risk"a randomised, open-label, clinical trial".
Blinding of outcome assessment (detection bias)
Other outcomes
High risk"a randomised, open-label, clinical trial".
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing outcome data.
Selective reporting (reporting bias)Unclear risk"Clinical and parasitological follow-up was carried out ...". However, in the outcomes there is no mention about any clinical improvement.
Other biasUnclear riskNo baseline comparison.

Gascon 1989 ESP

MethodsSpain. No power calculation made. Intention to treat analysis. Duration: 30 days.
Participants

23 adults; 17 male; Age was not specified. All symptomatic (information provided by author).

Microscopic stools examination: "harbouring G. lamblia". Technique: not mentioned.

Exclusion criteria: not mentioned.

Interventions

Rx 1. Mebendazole 200 mg three times daily for one day.

Rx 2: Metronidazole 250 mg every 8 h for seven days.

OutcomesParasitological cure: faecal specimens were examined on days three and seven after treatment, and finally one month later. Technique: not mentioned.
Notes

11 patients included from a travellers' clinic. The infections were acquired in Africa (4), Asia (3), Latin America (4).

4 patients were from an acquired immune deficiency syndrome (AIDS) clinic.

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"They were all given a random number".
Allocation concealment (selection bias)Unclear riskNot mentioned.
Blinding of outcome assessment (detection bias)
Stools examination
Unclear riskNot mentioned.
Blinding of outcome assessment (detection bias)
Other outcomes
Unclear riskNot mentioned.
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskThe authors do not specifically mention that there were no losses to follow-up.
Selective reporting (reporting bias)High riskSymptoms outcomes are mentioned but not reported.
Other biasUnclear riskNo information concerning baseline characteristics for each groups.

Gascón 1990 ESP

MethodsSpain. No power calculation made. Intention to treat analysis. Duration: 30 days.
Participants

19 participants (10 M); mean age: 29 ± 12 years; 16 adults. All symptomatic (information provided by author).

Microscopic stools examination: "excreting cystics of G. lamblia in their stools". Technique: not mentioned.

Exclusion criteria: not mentioned.

Interventions

Rx 1. Mebendazole 200 mg three times daily for five days.

Rx 2: Metronidazole 250 mg every 8 h for seven days.

Outcomes

Parasitological cure: faecal specimens were examined on days three, seven, and 30 after treatment. Technique: not mentioned.

No mention of adverse events.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"They, were all given a random number..."
Allocation concealment (selection bias)Unclear riskNot mentioned.
Blinding of outcome assessment (detection bias)
Stools examination
Unclear riskNot mentioned.
Blinding of outcome assessment (detection bias)
Other outcomes
Unclear riskNot mentioned.
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskThe authors do not specifically mention that there were no losses to follow-up.
Selective reporting (reporting bias)High riskNo mention of symptoms or adverse events.
Other biasUnclear riskNo information concerning baseline characteristics for each groups.

Hall 1993 IND

MethodsIndia. No power calculation made. Per protocol analysis. Duration: 15 days.
Participants

Children between 5 and 10 years old. Confirmed presence of Giardia cysts or trophozoites in faeces. 426 children (222 in the first phase, and 204 in the second phase)

There is not any specification about symptoms.

Exclusion criteria: not mentioned.

Interventions

Phase 1:

Rx 1. Albendazole: 600 mg single dose (chewable tablets).

Rx 2. Albendazole: 400 mg single dose, for three days (chewable tablets).

Rx 3. Metronidazole: 125 mg three times daily (375 mg/d) for five days (oral).

Phase 2:

Rx 4. Albendazole: 800 mg single dose (chewable tablets).

Rx 5. Albendazole: 400 mg single dose, for five days (chewable tablets).

Rx 6. Metronidazole: 125 mg three times daily (375 mg/d) for five days (oral).

Outcomes

Direct microscopy of faeces in saline. Three stools samples collected from each child over a period of 10 days after treatment.

Cure: Stool negative for G. lamblia.

Adverse events.

Notes

This study took into account as the unit of analysis "infections" (one child could participate more than once). However, also considered as the unit of analysis "children in their first (and / or unique) infection". For this systematic review, we consider data from the first Infection of children.

Similarly, for this systematic review, we consider treatments Rx 5 and Rx 6. Since this scheme albendazole is used more frequently, and it was the best response rate in this study.

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"A list of the numbers 1, 2 and 3 was randomly generated by a computer program".
Allocation concealment (selection bias)Low risk"… sealed in serially numbered envelopes by someone not concerned with treatment. The envelopes were opened in the field by the project physician, case by case, …".
Blinding of outcome assessment (detection bias)
Stools examination
Low risk"… each of the 3 stool samples collected after treatment was identified only by a unique serial number when examined microscopically, …".
Blinding of outcome assessment (detection bias)
Other outcomes
High risk"Because of the very different drug regimens being used in the investigation, a double-blind study was considered impracticable.".
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing outcome data.
Selective reporting (reporting bias)High riskThe outcomes of adverse events were incompletely reported. Adverse events were not specified for each treatment group.
Other biasUnclear riskNo baseline comparison.

Kalayci 1995 TUR

MethodsTurkey. No power calculation made. Intention to treat analysis. Duration: 10 days.
Participants

45 children; ages 34 months to 16 years. All symptomatic.

Symptoms: abdominal cramps (82%), vomiting (27%), greasy stools (24%), flatulence (18%), weight loss (18%), and anorexia (67%). The children had these complaints for 45 to 120 days (mean 86 days).

Microscopic stools examination: ".. within one hour after being passed for both cysts and trophozoites of G. lamblia". G. lamblia cysts were seen in all cases and trophozoites in 14 cases. Technique: not mentioned.

Exclusion criteria: children with a history of any treatment for giardiasis.

Interventions

Rx 1. Mebendazole 200 three times daily for 10 days (oral).

Rx 2. Metronidazole 5mg/kg three times daily for 10 days.

Rx 3. Furazolidone 2mg/kg four times daily for 10 days.

Outcomes

Clinical monitoring and three fecal specimens were re-examined (technique not mentioned.) for 7 to 10 days after treatment.

Side-effects.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk".. and chosen randomly to receive.."
Allocation concealment (selection bias)Unclear riskNot mentioned.
Blinding of outcome assessment (detection bias)
Stools examination
Low risk".. by investigators who were unaware of the drugs used or the results of the treatment".
Blinding of outcome assessment (detection bias)
Other outcomes
Unclear riskNot mentioned.
Incomplete outcome data (attrition bias)
All outcomes
Low riskThere were no losses to follow-up.
Selective reporting (reporting bias)Low riskAll outcomes were included in the analyses.
Other biasLow riskNo comments.

Karabay 2004 TUR

MethodsTurkey. No power calculation made. Per protocol analysis. Duration: 15 days.
Participants

Adults with diarrhoea at the outpatient clinic of the Department of Infectious Diseases of Social Security Hospital in Turkey: 67 patients (29 male).

Inclusion criteria: adults with diarrhoea (more than four times of unformed stools per day) and G intestinalis cyst or trophozoites (direct faecal microscopic examination).

Exclusion criteria: antiparasitic drugs during the 10 days prior to beginning the study; fever, pregnant women, mothers who were breast feeding, known hypersensitivity to either treatments, patients for whom any of the treatments used in the study were contraindicated.

Interventions

Rx 1. Albendazole: 400 mg/day single dose, for five days (oral).

Rx 2. Metronidazole: 500 mg three times daily (1500 mg/d), for five days (oral).

Outcomes

Clinical response and parasitological response with 2 stool samples between days 7 and 15 after initiation of treatment (direct faecal microscopic examination).

Noncompliance: failure to attend the controls, or inappropriate use of medications (dose, duration).

After starting treatment, healing of symptoms in hours.

Side effects.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskRandomization without more information.
Allocation concealment (selection bias)High risk"An open randomised clinical study".
Blinding of outcome assessment (detection bias)
Stools examination
High risk"An open randomised clinical study".
Blinding of outcome assessment (detection bias)
Other outcomes
High risk"An open randomised clinical study".
Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data balanced in numbers across intervention groups (with similar reasons).
Selective reporting (reporting bias)Unclear riskIn the methods section, the authors do not describe all outcomes that subsequently reported in results section.
Other biasLow riskNo comments.

Misra 1995 IND

MethodsIndia. No power calculation made. Per protocol analysis. Duration: 21 days.
Participants

64 hospitalised participants (38 males), from twp to 10 years old; Diarrhea, abdominal pain, anorexia, nausea, and vomiting were the more frequent symptoms.

Inclusion criteria: Children (twp to 12 years) with proven Giardia infection (trophozoites and/or cyst of G lamblia in stool specimens).

Exclusion criteria: "... acute febrile conditions, chronic diarrhoea, severe malnutrition or receiving long term therapy". Treatment of Giardia or intestinal helminthic infection in the previous seven days.

Interventions

Rx 1. Albendazole: 400 mg/d, single dose, for five days (oral suspension).

Rx 2. Metronidazole: 7.5mg/kg three times daily (22.5 mg/kg/day), for five days (oral suspension).

Outcomes

Clinical assessment and stools were repeated on days one to seven, 14 and 21 post-treatment. Stool samples were examined microscopically (direct smear and formal ether concentration tests).

Hematological and biochemical parameters on the 7th and 14th day after treatment.

Reported side-effects.

Notes

More severe presenting symptoms (abdominal pain, vomiting and watery stools) in children in the group treated with metronidazole.

This study has a similar methodology to the Dutta 1994 IND study, but different patients (different hospitals).

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"An open randomised parallel..."
Allocation concealment (selection bias)High risk"An open ..."
Blinding of outcome assessment (detection bias)
Stools examination
High risk"An open ..."
Blinding of outcome assessment (detection bias)
Other outcomes
High risk"An open ..."
Incomplete outcome data (attrition bias)
All outcomes
Low riskThe losses were distributed similarly in both groups. "All these children had responded to the therapy and parents did not want the children to be hospitalised any further"
Selective reporting (reporting bias)Unclear riskNo results were reported hematological and biochemical variables. The report mentions that there were no significant differences in these variables.
Other biasHigh riskMore severe presenting symptoms (abdominal pain, vomiting and watery stools) in children in the group treated with metronidazole.

Ortiz 2001 PER

MethodsPeru. Intention to treat and per protocol analysis. Duration: 14 days.
Participants

110 (54 males); ages two to 11 years, mean of age 5.67 (±2.56). All symptomatic.

Symptoms: duration of diarrhoea: 2 to 154 days. Pain, distension, and/or tenderness in 32.7%. Diarrhea was defined as more than three unformed stools per day or unformed stools without increased stool frequency for more than four weeks.

Enzyme immunoassay (Triage Parasite Panel, Biosite Diagnostic, San Diego, California) for G. lamblia, Entamoeba histolytica or Cryptosporidium parvum. Positive results for G. lamblia were confirmed using a direct examination and a Ritchie concentration technique.

Exclusion criteria: children with positive enzyme immunoassay of faecal samples for E. histolytica or C. parvum; children using any drug with antiprotozoal activity within two weeks of enrolment; children known to have or suspected of having AIDS were excluded from the study; patients with positive coprocultures for bacterial causes of diarrhoea were excluded from the analysis of clinical response.

Interventions

Rx 1. Nitazoxanide: 5 mL (ages 2 to 3 years) or 10 mL (ages four to 11 years) of nitazoxanide (100 mg per 5 mL) strawberry-flavoured oral suspension, BID for three days.

Rx 2: Metronidazole: 5 mL of 125 mg per 5 mL (ages two to five years) or 250 mg per 5 mL (ages six to 11 years) of metronidazole suspension BID for five days.

Outcomes

Clinical response (at the day 7 follow-up): "well" (no symptoms, no watery stools and no more than two soft stools, and has no hematochezia within the past 24 h or no symptoms and no unformed stools within the past 48 h); or "continuing illness" (any number of watery stools, more than two soft stools per 24 h, or hematochezia or enteric symptoms plus the passage of any number of soft or watery stools during the past 48 h); or "clinical treatment failure" (clinical deterioration or worsening of symptoms after at least 24 h of treatment resulting in the patients being removed from the study).

Parasitological response: "eradicated" (no cysts observed in either post-treatment stool sample); or "persistence" (cysts observed in either or both post-treatment stool samples). Two stool samples were collected on consecutive days between day 7 and 10, following the initiation of treatment for evaluation of parasitological response . Stool samples collected at follow-up were examined for G. lamblia using a direct examination and a Ritchie concentration technique.

NotesSample size calculation: For the two treatment groups (n 55) (80% of power; 20% in risk differences).
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"Computer-generated randomisation list".
Allocation concealment (selection bias)Unclear riskNot mentioned.
Blinding of outcome assessment (detection bias)
Stools examination
Low risk"parasitologist conducting the stool examinations was blinded to the patients treatment group assignment".
Blinding of outcome assessment (detection bias)
Other outcomes
Unclear riskNot mentioned.
Incomplete outcome data (attrition bias)
All outcomes
Low riskLosses to follow up were reported.
Selective reporting (reporting bias)Low riskAll outcomes were included in the analyses.
Other biasLow riskNo comments.

Perez-Choliz 1989 ESP

MethodsSpain. No power calculation made. Per protocol analysis. Duration: 1 month.
Participants

52 children (30 male) with parasitological diagnosis of giardiasis (paediatric outpatient). The mean age was 7.62 ± 3.23 (1.4 to 14.1 years).

There is not any specification about baseline symptoms, they were presumably 100% symptomatics.

Exclusion criteria: Not mentioned.

Interventions

Rx 1. Tinidazole: 50 mg/kg, maximum 1.5g, single dose (oral). During or after a main meal.

Rx 2. Metronidazole: 20-25 mg/kg/d, max 0.5 g per day (oral). Three cycles of eight days with seven days between them.

Outcomes

Examination of four stool samples, one month after treatment ends. The study does not specify the technique of stool examination.

Parasitological cure: stool negative for G. lamblia.

Symptoms collected in unsystematically through medical charts.

Side effects: questioning the patient and family, especially: headache, allergic reactions or any gastrointestinal symptoms.

NotesThe study provides no explanation for the loss to follow up.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"The allocation was set so random and alternating with tinidazole and metronidazole".
Allocation concealment (selection bias)Unclear riskNot mentioned.
Blinding of outcome assessment (detection bias)
Stools examination
Unclear risk"Randomized, double blind trial". Not specified. Both treatments are different in terms of duration of therapy.
Blinding of outcome assessment (detection bias)
Other outcomes
Unclear risk"Randomized, double blind trial". Not specified. Both treatments are different in terms of duration of therapy.
Incomplete outcome data (attrition bias)
All outcomes
High riskTwo losses to follow up in tinidazole group, and nine in the metronidazole group.
Selective reporting (reporting bias)Unclear riskSymptoms collected in unsystematically through medical charts.
Other biasUnclear riskNo baseline comparison.

Rodríguez 1996 MEX

MethodsMexico. No power calculation made. Intention to treat analysis. Duration: three weeks.
Participants

49 participants (25 males). Ages 3 to 12 years.   

The main clinical manifestations were abdominal pain (65%), diarrhoea (35%), and hyporexia (49%). 8% of children were asymptomatic. 

Inclusion criteria:

  • Children between three and 12 years old

  • Children infected with G. lamblia. Confirmed by parasitological stool examinations (Faust's technique).

  • Consent of parents or guardians of children.

Exclusion criteria:

  • Children who have received antigiardial drugs in the past three months.

Interventions

Rx 1. Albendazole: 200 mg three times daily for 5 days (oral).

Rx 2. Metronidazole: 30 mg/kg/day, three times daily, for 5 days (oral).

Outcomes

Parasitological (disappearance of cysts or trophozoites). Coproparasitology examination: direct microscopy, using the technique of Faust. The samples were collected on days 14 and 21, and he/she was considered cured if both samples were negative.

Reported side-effects.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"Two groups were formed randomly".
Allocation concealment (selection bias)Unclear riskNot mentioned.
Blinding of outcome assessment (detection bias)
Stools examination
Unclear riskNot mentioned.
Blinding of outcome assessment (detection bias)
Other outcomes
Unclear riskNot mentioned.
Incomplete outcome data (attrition bias)
All outcomes
Low riskThere were no losses to follow-up.
Selective reporting (reporting bias)Unclear riskNo mention of a systematic strategy for collecting side effects.
Other biasHigh riskThere are differences in baseline characteristics of both groups. For example, preschool children were 51% in the albendazole group and 40% for metronidazole arm. Similarly, 74% of the children complained of abdominal pain in the albendazole group and 54% did so in the metronidazole group.

Romero-Cabello 1995 MEX

MethodsMexico. No power calculation made. Intention to treat analysis. Duration: 21 days.
Participants

100 participants (49 males), from four to 11 years old (median: eight years old); symptoms (93%): 42% abdominal pain, 39% hyporexia, 8% diarrhoea, 2% nausea, and 2% debility.

Inclusion criteria: children infected with G. lamblia. Stool samples were examined microscopically.

Exclusion criteria: any drug with antiprotozoal activity in the last two months; or another class of treatment in children with denutrition II or III grade.

Interventions

Rx 1. Albendazole: 400 mg/day single dose, for five days (oral).

Rx 2. Metronidazole: 7.5 mg/kg three times daily, for five days (oral).

Outcomes

Parasitological cure: absence of the cystic or trophozoite form of G. lamblia in the stools. Stools were repeated on days one to seven, 14, and 21 after beginning the treatment. Direct microscopy (three successive days).

Clinical response at one to seven days after beginning of treatment. Time to recovery.

Side-effects: on days one to seven, 14, and 21.

Cell blood count and blood chemistry after treatment.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot stated.
Allocation concealment (selection bias)Unclear risk"using an aleatory code A and B".
Blinding of outcome assessment (detection bias)
Stools examination
High risk"Open label".
Blinding of outcome assessment (detection bias)
Other outcomes
High risk"Open label".
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo reported losses to follow up.
Selective reporting (reporting bias)Low riskApparently, all outcomes were reported.
Other biasLow riskNo comments.

Yereli 2004 TUR

MethodsTurkey. No power calculation made. Intention to treat analysis. Duration: 21 days (14 days after completing treatment).
Participants

107 children, from three to 15 years old; Symptoms (gastrointestinal complaints, diarrhoea, weight loss, anorexia, and fatigue) and G. lamblia cysts and or trophozoites in stools.

Stool samples were assessed for giardiasis by the saline–Lugol, formalin ethyl acetate concentration, and trichrome staining methods.

Exclusion criteria: not mentioned.

Interventions

Rx 1. Albendazole: 10 mg/kg/day, single dose, for five days (tablets).

Rx 2. Metronidazole: 20 mg/kg/day, three times daily, for seven days (tablet or liquid).

Outcomes

Stool samples from all cases were examined on the 7th, 10th and 14th days following completion of treatment.

Clinical symptoms: the method was not specified.

Side-effects: the method was not specified.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"was divided into two random groups".
Allocation concealment (selection bias)Unclear riskNot mentioned.
Blinding of outcome assessment (detection bias)
Stools examination
Unclear riskNot mentioned.
Blinding of outcome assessment (detection bias)
Other outcomes
High riskNo masking of drugs.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo withdrawals or loss to follow up.
Selective reporting (reporting bias)High riskSymptoms were not adequately reported.
Other biasLow riskNo comments.

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
Bakshi 1978It is not a randomized clinical trial. Meta-analysis of clinical trials in India.
Baqai 2001No randomization.
Bassily 1987It is not a randomized clinical trial.
Besirbellioglu 2006This study included two groups treated with metronidazole plus Saccharomyces boulardii.
Botero 1974The study was presented in a congress. No data available.
Gazder 1978This study compared single dose of metronidazole against single dose of tinidazole.
Grant 2001This study included two groups treated with metronidazole plus wheat germ, or placebo.
Guerrero 2000Authors were contacted. The study was presented in a congress, data were not provided.
Jokipii 1978It is not a randomized clinical trial.
Jokipii 1979It is not a randomized clinical trial.
Krishnamurthy 1978This study compared single dose of metronidazole against single dose of tinidazole.
Kyrönseppä 1981This study compared treatment with single dose tinidazole, against treatment with metronidazole for two days.
Leary 1974This study compared treatment with metronidazole for one day, against treatment with metronidazole for two days.
Levi 1977It is not a randomized clinical trial.
Morrone 2004It is not a randomized clinical trial.
Nigam 1991This study compared single dose of metronidazole against single dose of tinidazole.
Sabchareon 1980It is not a randomized clinical trial.
Sadjjadi 2001Asymptomatic children.
Speelman 1985This study compared single dose tinidazole against single dose metronidazole, and another arm with metronidazole for three days.
Suntornpoch 1981It is not a randomized clinical trial.

Ancillary