Smectite for acute infectious diarrhoea in children

  • Review
  • Intervention

Authors


Abstract

Background

As mortality secondary to acute infectious diarrhoea has decreased worldwide, the focus shifts to adjuvant therapies to lessen the burden of disease. Smectite, a medicinal clay, could offer a complementary intervention to reduce the duration of diarrhoea.

Objectives

To assess the effects of smectite for treating acute infectious diarrhoea in children.

Search methods

We searched the Cochrane Infectious Diseases Group Specialized Register, the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE (Pubmed), Embase (Ovid), LILACS, reference lists from studies and previous reviews, and conference abstracts, up to 27 June 2017.

Selection criteria

Randomized and quasi-randomized trials comparing smectite to a control group in children aged one month to 18 years old with acute infectious diarrhoea.

Data collection and analysis

Two review authors independently screened abstracts and the full texts for inclusion, extracted data, and assessed risk of bias. Our primary outcomes were duration of diarrhoea and clinical resolution at day 3. We summarized continuous outcomes using mean differences (MD) and dichotomous outcomes using risk ratios (RR), with 95% confidence intervals (CI). Where appropriate, we pooled data in meta-analyses and assessed heterogeneity. We explored publication bias using a funnel plot.

Main results

Eighteen trials with 2616 children met our inclusion criteria. Studies were conducted in both ambulatory and in-hospital settings, and in both high-income and low- or middle-income countries. Most studies included children with rotavirus infections, and half included breastfed children.

Smectite may reduce the duration of diarrhoea by approximately a day (MD -24.38 hours, 95% CI -30.91 to -17.85; 14 studies; 2209 children; low-certainty evidence); may increase clinical resolution at day 3 (risk ratio (RR) 2.10, 95% CI 1.30 to 3.39; 5 trials; 312 children; low-certainty evidence); and may reduce stool output (MD -11.37, 95% CI -21.94 to -0.79; 3 studies; 634 children; low-certainty evidence).

We are uncertain whether smectite reduces stool frequency, measured as depositions per day (MD -1.33, 95% CI -2.28 to -0.38; 3 studies; 954 children; very low-certainty evidence). There was no evidence of an effect on need for hospitalization (RR 0.93, 95% CI 0.75 to 1.15; 2 studies; 885 children; low-certainty evidence) and need for intravenous rehydration (RR 0.77, 95% CI 0.54 to 1.11; 1 study; 81 children; moderate-certainty evidence). The most frequently reported side effect was constipation, which did not differ between groups (RR 4.71, 95% CI 0.56 to 39.19; 2 studies; 128 children; low-certainty evidence). No deaths or serious adverse effects were reported.

Authors' conclusions

Based on low-certainty evidence, smectite used as an adjuvant to rehydration therapy may reduce the duration of diarrhoea in children with acute infectious diarrhoea by a day; may increase cure rate by day 3; and may reduce stool output, but has no effect on hospitalization rates or need for intravenous therapy.

Resumen

Esmectita para la diarrea infecciosa aguda en niños

Antecedentes

Debido a que la mortalidad secundaria a la diarrea infecciosa aguda ha disminuido en todo el mundo, el foco se traslada a los tratamientos coadyuvantes para reducir la carga de morbilidad. La esmectita, una arcilla medicinal, podría ofrecer una intervención complementaria para reducir la duración de la diarrea.

Objetivos

Evaluar los efectos de la esmectita para el tratamiento de la diarrea infecciosa aguda en niños.

Métodos de búsqueda

Se hicieron búsquedas en el registro especializado del Grupo Cochrane de Enfermedades Infecciosas (Cochrane Infectious Diseases Group Specialized Register), Registro Cochrane Central de Ensayos Controlados (Cochrane Central Register of Controlled Trials) (CENTRAL), MEDLINE (Pubmed), Embase (Ovid), LILACS, en listas de referencias de estudios, en revisiones anteriores y en resúmenes de congresos, hasta el 27 junio 2017.

Criterios de selección

Ensayos aleatorios y cuasialeatorios que compararan esmectita con un grupo de control en niños de un mes a 18 años de edad con diarrea infecciosa aguda.

Obtención y análisis de los datos

Dos autores de la revisión, de forma independiente, revisaron los resúmenes y los textos completos para su inclusión, extrajeron los datos y evaluaron el riesgo de sesgo. Los resultados primarios fueron la duración de la diarrea y la resolución clínica el día 3. Los resultados continuos se resumieron mediante las diferencias de medias (DM) y los resultados dicotómicos mediante los cocientes de riesgos (CR), con intervalos de confianza (IC) del 95%. Cuando fue apropiado, se agruparon los datos en los metanálisis y se evaluó la heterogeneidad. Se exploró el sesgo de publicación mediante un gráfico en embudo (funnel plot).

Resultados principales

Dieciocho ensayos con 2616 participantes cumplieron los criterios de inclusión. Los estudios se realizaron en contextos tanto ambulatorios como hospitalarios, y en países tanto de ingresos altos como de ingresos bajos o medios. La mayoría de los estudios incluyó a niños con infecciones por rotavirus, y la mitad incluyó a niños que estaban siendo amamantados.

La esmectita puede reducir la duración de la diarrea en aproximadamente un día (DM -24,38 horas, IC del 95%: -30,91 a -17,85; 14 estudios; 2209 niños; evidencia de certeza baja); puede aumentar la resolución clínica al día 3 (cociente de riesgos [CR] 2,10; IC del 95%: 1,30 a 3,39; cinco ensayos; 312 niños; evidencia de certeza baja); y puede reducir la producción de heces (DM -11,37; IC del 95%: -21,94 a -0,79; tres estudios; 634 niños; evidencia de certeza baja).

No existe seguridad en cuanto a si la esmectita reduce la frecuencia de las heces, medida según las deposiciones por día (DM -1,33; IC del 95%: -2,28 a -0,38; tres estudios; 954 niños; evidencia de certeza muy baja). No hubo evidencia de un efecto sobre la necesidad de hospitalización (CR 0,93; IC del 95%: 0,75 a 1,15; dos estudios; 885 niños; evidencia de certeza baja) y la necesidad de rehidratación intravenosa (CR 0,77; IC del 95%: 0,54 a 1,11; un estudio; 81 niños; evidencia de certeza moderada). El efecto secundario informado con mayor frecuencia fue el estreñimiento, que no difirió entre los grupos (CR 4,71; IC del 95%: 0,56 a 39,19; dos estudios; 128 niños; evidencia de certeza baja). No se informaron muertes o efectos adversos graves.

Conclusiones de los autores

Basado en la evidencia de certeza baja, la esmectita utilizada como un coadyuvante al tratamiento de rehidratación puede reducir la duración de la diarrea en los niños con diarrea infecciosa aguda en un día; puede aumentar la tasa de curación al día 3; y puede reducir la producción de heces, aunque no tiene ningún efecto sobre las tasas de hospitalización o la necesidad de tratamiento intravenoso.

Résumé scientifique

La smectite pour la diarrhée infectieuse aiguë chez les enfants

Contexte

Comme la mortalité résultant de la diarrhée infectieuse aiguë a diminué à l’échelle mondiale, l’intérêt se porte sur les traitements adjuvants pour diminuer la morbidité de la maladie. La smectite, une argile médicinale, pourrait servir d’intervention complémentaire pour réduire la durée de la diarrhée.

Objectifs

Évaluer les effets de la smectite pour le traitement de la diarrhée infectieuse aiguë chez les enfants.

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), MEDLINE (Pubmed), Embase (Ovid), LILACS, les listes bibliographiques des études et des précédentes revues et les résumés de conférences, jusqu'au 27 juin 2017.

Critères de sélection

Les essais randomisés et quasi-randomisés comparant la smectite à un groupe témoin chez des enfants âgés d’un mois à 18 ans atteints d’une diarrhée infectieuse aiguë.

Recueil et analyse des données

Deux auteurs de la revue ont indépendamment passé au crible les résumés et les textes complets pour l’inclusion, extrait les données et évalué le risque de biais. Nos critères d’évaluation principaux étaient la durée de la diarrhée et la résolution clinique au 3e jour. Nous avons résumé les résultats continus en utilisant les différences moyennes (DM) et les résultats dichotomiques en utilisant le risques relatif (RR), avec un intervalle de confiance (IC) à 95 %. Lorsque cela était approprié, nous avons regroupé les données dans des méta-analyses et évalué l’hétérogénéité. Nous avons étudié le biais de publication à l’aide d’un graphique en entonnoir.

Résultats principaux

Dix-huit essais portant sur 2616 enfants répondaient à nos critères d'inclusion. Les études étaient menées aussi bien dans le contexte ambulatoire qu’hospitalier, et à la fois dans des pays à haut revenu et des pays à revenu faible ou intermédiaire. La plupart des études incluaient des enfants atteints d’infections à rotavirus, et la moitié incluait des enfants allaités.

La smectite pourrait réduire la durée de la diarrhée d’environ un jour (DM -24,38 heures ; IC à 95 % -30,91 à -17,85 ; 14 études ; 2209 enfants ; preuves de faible niveau de certitude) ; elle pourrait augmenter la résolution clinique au 3e jour (RR 2,10 ; IC à 95 % 1,30 à 3,39 ; 5 essais ; 312 enfants, preuves d’un faible niveau de certitude) ; et elle pourrait réduire la production de selles (DM -11,37 ; IC à 95 % -21,94 à -0,79 ; 3 études ; 634 enfants ; preuves d’un faible niveau de certitude).

Nous ne savons pas si la smectite réduit la fréquence des selles, mesurée en nombre de dépôts par jour (DM -1,33 ; , IC à 95 % -2,28 à -0,38 ; 3 études ; 954 enfants ; preuves d’un très faible niveau de certitude). Il n’y avait aucune preuve d’un effet sur la nécessité de l’hospitalisation (RR 0,93, IC à 95 % 0,75 à 1,15 ; 2 études ; 885 enfants ; preuve faible-certitude) et la nécessité d’une réhydratation par voie intraveineuse (RR 0,77, IC à 95 % 0,54 à 1,11 ; 1 étude ; 81 enfants ; preuves d’un niveau modéré de certitude). L’effet secondaire le plus fréquemment signalé était la constipation, qui ne différait pas entre les groupes (RR 4,71 ; IC à 95 % 0,56 à 39,19 ; 2 études ; 128 enfants ; preuves d’un faible niveau de certitude). Aucun décès ou effet indésirable grave n’a été signalé.

Conclusions des auteurs

Sur la base de preuves d’un faible niveau de certitude, la smectite utilisée comme adjuvant au traitement de réhydratation pourrait réduire d’un jour la durée de la diarrhée chez les enfants atteints de diarrhée infectieuse aiguë ; elle pourrait augmenter le taux de guérison au 3e jour ; et elle pourrait réduire la production de selles, mais elle n’a aucun effet sur le taux d’hospitalisation ou la nécessité d’un traitement par voie intraveineuse.

Plain language summary

Smectite for treating children with acute diarrhoea

What is the aim of this review?

The aim of this Cochrane Review was to find out if smectite (or diosmectite), a medicinal clay commonly prescribed to people who have diarrhoea in order to reduce their stool output, helps children with acute diarrhoea. We collected and analysed all relevant studies to answer this question and found 18 relevant studies.

Key messages

Giving smectite to children with acute diarrhoea may reduce its duration. However, more high-quality studies are still needed, including studies that assess different causes of diarrhoea and the economic effects of this treatment.

What was studied in the review?

Acute diarrhoea is one of the most common diseases in children. It is usually caused by a viral infection. The main aim of treatment is to maintain a good level of hydration. This is achieved with oral rehydration solutions, and few children need to be hospitalized or require intravenous rehydration. Still, even with proper hydration, having loose stools is a burden for both parents and patients.

Smectite may help by reducing inflammation in the gut; by acting as a barrier to reduce the penetration of toxins; or by increasing water absorption.

What are the main results?

We found 18 relevant studies with 2616 children that were conducted in both high-income and low- or middle-income countries. These studies compared children receiving smectite with children receiving routine care or a placebo (a pill or liquid that contains no medicine). Eight studies were funded by the manufacturer.

Smectite may reduce the duration of diarrhoea by one day (low-certainty evidence); may increase the number of children cured by day 3 (low-certainty evidence); and may slightly reduce the quantity of loose stools (low-certainty evidence).

We are uncertain whether smectite has an effect on how many stools children have (very low-certainty evidence). It may not have an effect on how many children need to be hospitalized (low-certainty evidence), and probably does not have an effect on how many children need intravenous rehydration (moderate-certainty evidence).

We found no reports of serious adverse effects. Minor adverse effects included constipation, vomiting, and bad taste, but these did not differ between groups.

How up-to-date is this review?

We searched for studies published up to 27 June 2017.

Resumen en términos sencillos

Esmectita para el tratamiento de niños con diarrea aguda

¿Cuál es el objetivo de esta revisión?

El objetivo de esta revisión Cochrane era determinar si la esmectita (o la dioesmectita), una arcilla medicinal prescrita comúnmente a los pacientes que presentan diarrea para reducir la producción de heces, ayuda a los niños con diarrea aguda. Se recopilaron y analizaron todos los estudios relevantes para responder a esta pregunta y se encontraron 18 estudios pertinentes.

Mensajes clave

La administración de esmectita a los niños con diarrea aguda puede reducir su duración. Sin embargo, todavía se necesitan más estudios de alta calidad, incluidos los estudios que evalúan diferentes causas de la diarrea y los efectos económicos de este tratamiento.

¿Qué se estudió en la revisión?

La diarrea aguda es una de las enfermedades más frecuentes en los niños. Por lo general está causada por una infección viral. El objetivo principal del tratamiento es mantener un buen nivel de hidratación. Lo anterior se logra con soluciones de rehidratación orales, y pocos niños necesitan ser hospitalizados o requieren rehidratación intravenosa. De todos modos, incluso con la hidratación adecuada, la producción de heces blandas es una carga tanto para los padres como para los pacientes.

La esmectita puede ayudar a reducir la inflamación en los intestinos al actuar como una barrera para reducir la penetración de las toxinas, o al aumentar la absorción de agua.

¿Cuáles son los principales resultados?

Se encontraron 18 estudios relevantes con 2616 niños que se realizaron en países de ingresos altos y de ingresos bajos o medios. Estos estudios compararon a los niños que recibieron esmectita con los niños que recibieron atención habitual o un placebo (una pastilla o líquido que no contiene medicación). Ocho estudios fueron financiados por el fabricante.

La esmectita puede reducir la duración de la diarrea en un día (evidencia de certeza baja); puede aumentar el número de niños curados el día 3 (evidencia de certeza baja); y puede reducir levemente la cantidad de heces blandas (evidencia de certeza baja).

No existe seguridad en cuanto a si la esmectita tiene un efecto sobre la cantidad de heces que producen los niños (evidencia de muy baja certeza). Puede no tener un efecto sobre cuántos niños necesitan ser hospitalizados (evidencia de certeza baja) y probablemente no tiene un efecto sobre cuántos niños necesitan rehidratación intravenosa (evidencia de certeza moderada).

No se encontraron informes de efectos adversos graves. Los efectos adversos menores incluyeron estreñimiento, vómitos y sabor desagradable, aunque los mismos no difirieron entre los grupos.

¿Cuál es el grado de actualización de esta revisión?

Se buscaron estudios publicados hasta el 27 junio 2017.

Notas de traducción

La traducción y edición de las revisiones Cochrane han sido realizadas bajo la responsabilidad del Centro Cochrane Iberoamericano, gracias a la suscripción efectuada por el Ministerio de Sanidad, Servicios Sociales e Igualdad del Gobierno español. Si detecta algún problema con la traducción, por favor, contacte con Infoglobal Suport, cochrane@infoglobal-suport.com.

淺顯易懂的口語結論

使用Smectite治療兒童急性腹瀉

本篇文獻回顧之目的為何?

本篇考科藍文獻回顧旨在找出smectite(或 diosmectite)是否有助於患有急性腹瀉之兒童。 smectite為一種常見治療腹瀉之處方藥, 可減少糞便的排出。我們收集及分析所有相關研究來回答這個問題,並找出18個相關研究。

主要訊息

兒童急性腹瀉給予藥物smectite可能可縮短其腹瀉持續時間。然而, 尚需更多高品質研究, 包括評估腹瀉之不同原因和腹瀉治療之經濟影響。

本篇文獻回顧內容為何?

急性腹瀉是兒童最常見的疾病之一。它通常是由病毒感染引起的。治療之主要目的為避免脫水。口服電解質補充液可避免脫水,但少數孩童需要住院治療或靜脈注射補充液。即使有適當的補水, 腹瀉仍是父母和病人的一大負擔。

Smectite可助於減少腸道發炎;作為屏障以減少毒素滲透;或增加水分的吸收。

主要結果為何?

我們找到18個相關研究, 其為在高收入和低或中等收入國家中,對2616位兒童進行研究。這些研究比較smectite藥物治療組與接受常規照護或安慰劑(不含藥物之藥丸或藥水)組之兒童。其中8個研究由藥廠資助。

Smectite可能可縮短腹瀉持續時間至一天 (低確定性證據);可能增加兒童第三天痊癒之人數(低確定性證據);及可能輕微減少兒童腹瀉量 (低確定性證據)。

我們不確定smectite對於兒童糞便量產生的影響 (非常低確定性證據)。它可能不會影響需要住院的兒童數目 (低確定性證據), 也可能不會影響需要靜脈注射補充液之兒童數目 (中度確定性證據)。

我們沒有發現任何嚴重不良副作用之報告。輕微不良副作用包括便秘、嘔吐及糞便氣味不佳, 但這些不良副作用在組別間並無差異。

本文獻回顧資料更新日期為何?

我們搜尋至2017年6月27日前發表之研究。

譯註

翻譯者:江吉文
服務單位:衛生福利部屏東醫院
職稱:藥師

本翻譯計畫由臺北醫學大學考科藍臺灣研究中心(Cochrane Taiwan)、台灣實證醫學學會及東亞考科藍聯盟(EACA)統籌執行
聯絡E-mail:cochranetaiwan@tmu.edu.tw

Streszczenie prostym językiem

Smektyn (Smecta - wyrób medyczny; przyp. tłum.) w leczeniu dzieci z ostrymi biegunkami.

Co jest celem tego przeglądu?

Celem niniejszego przeglądu Cochrane była ocena, czy smektyn (lub diosmektyt) - lecznicza glinka powszechnie stosowana w leczeniu biegunki, pomaga dzieciom z ostrą biegunką. Aby odpowiedzieć na to pytanie, zebraliśmy i przeanalizowaliśmy odpowiadające badania. Znaleźliśmy 18 badań spełniających kryteria.

Główne wnioski

Podawanie smektynu dzieciom z ostrą biegunką może skrócić czas jej trwania. Jednak potrzeba większej liczby badań wysokiej jakości, w tym badań oceniających różne przyczyny biegunki i skutki ekonomiczne jej leczenia.

Co badano w ramach tego przeglądu?

Ostra biegunka jest jedną z najczęstszych chorób u dzieci. Zwykle jest ona spowodowana infekcją wirusową. Głównym celem leczenia jest utrzymanie prawidłowego nawodnienia organizmu. Osiąga się to dzięki doustnym roztworom nawadniającym, niewiele dzieci musi być hospitalizowana lub wymaga nawodnienia dożylnego. Mimo to nawet przy odpowiednim nawodnieniu, oddawanie luźnych stolców jest obciążające dla pacjentów i ich rodziców.

Smektyn może pomóc poprzez zmniejszenie stanu zapalnego w jelitach, działając jak bariera ochronna zmniejszająca przenikanie toksyn lub zwiększając wchłanianie wody.

Jakie są główne wyniki?

Znaleźliśmy 18 badań spełniających kryteria, w którym udział wzięło łącznie 2616 dzieci. Badania przeprowadzono w krajach o wysokich, średnich oraz niskich dochodach. W badaniach tych porównano dzieci przyjmujące smektyn z dziećmi otrzymującymi rutynową opiekę lub placebo (pigułki lub płyny nie zawierające leku). Osiem badań zostało sfinansowanych przez producenta.

Smektyn może skrócić czas trwania biegunki o jeden dzień (dane o małym stopniu wiarygodności), zwiększyć liczbę dzieci wyleczonych w ciągu 3 dni (dane o małym stopniu wiarygodności) i nieznacznie zmniejszyć objętość luźnych stolców (dane o małym stopniu wiarygodności).

Nie ma pewności, czy smektyn ma wpływ na liczbę wypróżnień u dzieci (dane o małym stopniu wiarygodności). Stosowanie smektynu może nie mieć wpływu na liczbę dzieci hospitalizowanych (dane o małym stopniu wiarygodności) i prawdopodobnie nie ma wpływu na to, ile dzieci potrzebuje dożylnego nawodnienia (dane o umiarkowanym stopniu wiarygodności).

Nie znaleziono żadnych doniesień o poważnych działaniach niepożądanych. Niewielkie skutki uboczne obejmowały zaparcia, wymioty i nieprzyjemny posmak, ale nie różniły się one pomiędzy grupami.

Na ile aktualny jest ten przegląd?

Wyszukiwaliśmy badań opublikowanych do 27 czerwca 2017 r.

Uwagi do tłumaczenia

Tłumaczenie: Weronika Tępczyk Redakcja: Piotr Szymczak, Małgorzata Kołcz

Résumé simplifié

La smectite pour le traitement des enfants atteints de diarrhée aiguë

Quel est l'objectif de cette revue ?

L’objectif de cette revue Cochrane était de déterminer si la smectite (ou la diosmectite), une argile médicinale souvent prescrite aux personnes qui ont la diarrhée afin de réduire leur production de selles, aide les enfants atteints de diarrhée aiguë. Nous avons rassemblé et analysé toutes les études pertinentes afin de répondre à cette question et nous avons trouvé 18 études pertinentes.

Messages clés

Donner de la smectite aux enfants atteints de diarrhée aiguë pourrait réduire sa durée. Toutefois, des études supplémentaires de haute qualité restent nécessaires, notamment des études évaluant différentes causes de diarrhée et l’incidence économique de ce traitement.

Quel est le sujet de la revue ?

La diarrhée aiguë est l’une des maladies les plus fréquentes chez l’enfant. Elle est généralement causée par une infection virale. Le principal objectif du traitement est de maintenir un bon niveau d’hydratation. Ceci est accompli à l’aide de solutions de réhydratation orale, et peu d’enfants ont besoin d’être hospitalisés ou nécessitent une réhydratation par voie intraveineuse. Pourtant, même avec une bonne hydratation, les selles molles sont un fardeau à la fois pour les parents et pour les patients.

La smectite pourrait aider en réduisant l’inflammation dans les intestins, en agissant comme une barrière afin de réduire la pénétration des toxines, ou en augmentant l’absorption d’eau.

Quels sont les principaux résultats ?

Nous avons trouvé 18 études pertinentes portant sur 2616 enfants, qui ont été menées à la fois dans des pays à haut revenu et des pays à revenu faible ou intermédiaire. Ces études ont comparé des enfants recevant de la smectite à des enfants recevant des soins de routine ou un placebo (un comprimé ou un liquide qui ne contient pas de médicament). Huit études avaient été financées par le fabricant.

La smectite pourrait réduire la durée de la diarrhée d’un jour (preuves d’un faible niveau de certitude), pourrait augmenter le nombre d’enfants guéris au 3e jour (preuves d’un faible niveau de certitude), et pourrait légèrement réduire la quantité de selles molles (preuves d’un faible niveau de certitude).

Nous ne savons pas si la smectite a un effet sur le nombre de selles des enfants (preuves d’un très faible niveau de certitude). Elle pourrait ne pas avoir d’effet sur le nombre d’enfants qui ont besoin d’être hospitalisés (preuves d’un faible niveau de certitude), et elle n’a probablement aucun effet sur le nombre d’enfants qui nécessitent une réhydratation par voie intraveineuse (preuves d’un niveau de certitude modéré).

Nous n’avons trouvé aucun signalement d’effets indésirables graves. Des effets indésirables mineurs incluaient la constipation, les vomissements et un mauvais goût dans la bouche, mais ceux-ci ne différaient pas entre les groupes.

Cette revue est-elle à jour ?

Nous avons cherché des études publiées jusqu'au 27 juin 2017.

Notes de traduction

Traduction réalisée par Sophie Fleurdépine et révisée par Cochrane France

Резюме на простом языке

Смектит для лечения детей с острой диареей

Какова цель этого обзора?

Целью этого Кокрейновского обзора было выяснить, помогает ли детям с острой диареей смектит (или диосмектит), представляющий собой лекарственную глину, который обычно назначают людям с диареей для уменьшения стула. Мы собрали и проанализировали все соответствующие исследования, чтобы ответить на этот вопрос, и нашли 18 подходящих исследований.

Ключевая информация

Назначение смектита детям с острой диареей может уменьшить продолжительность диареи. Однако, необходимы исследования более высокого качества, включая исследования, в которых оценивают различные причины диареи и экономические последствия проводимого лечения.

Что было изучено в этом обзоре?

Острая диарея является одним из наиболее распространенных заболеваний у детей. Причиной острой диареи обычно является вирусная инфекция. Основная цель лечения заключается в поддержке достаточного уровня гидратации. Это достигается с помощью пероральных регидратационных растворов, но некоторым детям может потребоваться госпитализация и проведение внутривенной регидратации (введение растворов в вену). Однако даже при адекватной регидратации жидкий стул является проблемой для пациентов и их родителей.

Смектит может помочь уменьшить воспаление в кишечнике, действуя в качестве барьера, препятствующего проникновению токсинов, или увеличивая всасывание воды.

Каковы основные результаты?

Мы нашли 18 соответствующих исследований с участием 2616 детей, которые были проведены в странах с высоким, средним и низким уровнем дохода. В этих исследованиях сравнивали детей, получающих смектит, с детьми, получающими стандартную помощь или плацебо (таблетки или жидкость, не содержащие лекарства). Восемь исследований были финансированы производителем.

Смектит может сократить продолжительность диареи на один день (низкая уверенность в доказательствах); может увеличить число детей, вылечившихся на третий день (низкая уверенность в доказательствах); и может немного уменьшить количество жидкого стула (низкая уверенность в доказательствах).

Мы не уверены, влияет ли смектит на частоту стула у детей (очень низкая уверенность в доказательствах). Возможно, он не влияет на частоту госпитализаций детей (низкая уверенность в доказательствах) и вероятно не влияет на потребность детей во внутривенной регидратации (умеренная уверенность в доказательствах).

Мы не нашли сообщений о серьезных неблагоприятных эффектах. Среди небольших неблагоприятных эффектов были запор, рвота, неприятный вкус, но они не отличались между группами.

Насколько актуален этот обзор?

Мы провели поиск исследований, опубликованных до 27 июня 2017 года включительно.

Заметки по переводу

Перевод: Юдина Екатерина Викторовна. Редактирование: Зиганшина Лилия Евгеньевна. Координация проекта по переводу на русский язык: Cochrane Russia - Кокрейн Россия (филиал Северного Кокрейновского Центра на базе Казанского федерального университета). По вопросам, связанным с этим переводом, пожалуйста, обращайтесь к нам по адресу: cochrane.russia.kpfu@gmail.com; cochranerussia@kpfu.ru

Summary of findings(Explanation)

Summary of findings for the main comparison. Smectite compared to control for acute infectious diarrhoea in children
  1. 1Four trials are quasi-randomized and without adequate blinding of participants.
    2High heterogeneity (I2 = 96%) among studies that may be explained by differences in age and definition of resolution, although the effect in all studies points in the same direction.
    3Three studies have high risk of selection bias, including one that is quasi-randomized, and three did not perform adequate blinding of participants.
    4High heterogeneity (I2 = 81%), although the effect in all studies points in the same direction.
    5High heterogeneity (I2 = 97%), although all effects point in the same direction.
    6Two of the three studies are classified as quasi-randomized with inadequate blinding of participants.
    7A wide CI that does not exclude the threshold of appreciable clinical benefit.
    8One quasi-randomized study was not pooled because the authors reported stool output as stool weight in total grams per day with an effect estimate favouring smectite (mean of 255.67 g in the smectite group versus 741.33 g in the control group) at day 3 of treatment.
    9Wide CI that does not exclude an appreciable benefit or harm.

Smectite compared to control for acute infectious diarrhoea in children
Patient or population: acute infectious diarrhoea in children
Setting: hospital and outpatients
Intervention: smectite
Comparison: control
OutcomesAnticipated absolute effects* (95% CI)Relative effect
(95% CI)
Number of participants
(studies)
Certainty of the evidence
(GRADE)
Comments (compared with control)
Risk with controlRisk with smectite
Duration of diarrhoea
assessed with: clinical and parental assessment, measured in total hours
Follow-up: mean 1 week
The mean duration of diarrhoea ranged from 32.6 to 118.92 hoursMD 24.38 hours fewer
(30.91 fewer to 17.85 fewer)
-2209
(14 RCTs)
⊕⊕⊝⊝
LOW1,2
Smectite may reduce the duration of diarrhoea
Clinical resolution at day 3
assessed with: clinical assessment by parents and clinicians
Follow-up: mean 3 days
Study populationRR 2.10
(1.30 to 3.39)
312
(5 RCTs)
⊕⊕⊝⊝
LOW3,4
Smectite may increase the resolution of diarrhoea by the third day
342 per 1000718 per 1000
(445 to 1000)
Stool frequency assessed with: clinical assessment as number of depositions per day
Follow-up: mean 1 week
The mean stool frequency was 0 depositions per dayMD 1.33 depositions per day fewer
(2.28 fewer to 0.38 fewer)
-954
(3 RCTs)
⊕⊝⊝⊝
VERY LOW5,6,7
We are uncertain whether or not smectite reduces stool frequency
Stool output assessed with: grams of stool output per kg of body weight in a 72-hour period
Follow-up: mean 1 week
The mean stool output ranged from 90.7 to 118.8 g/kgMD 11.37 g/kg fewer
(21.94 fewer to 0.79 fewer)
-634
(3 RCTs)
⊕⊕⊝⊝
LOW7,8
Smectite may decrease stool output
Need for hospitalization
Follow-up: mean 1 week
Study populationRR 0.93
(0.75 to 1.15)
885
(2 RCTs)
⊕⊕⊝⊝
LOW6,9
Smectite may make little or no difference in the need for hospitalization
85 per 100079 per 1000
(64 to 98)
Need for intravenous access for rehydration
Follow-up: mean 1 week
Study populationRR 0.77
(0.54 to 1.11)
81
(1 RCT)
⊕⊕⊕⊝
MODERATE9
Smectite probably makes little or no difference in the need for intravenoous access
676 per 1000520 per 1000
(365 to 750)
Adverse events – constipation
Follow-up: mean 1 week
Study populationRR 4.71
(0.56 to 39.19)
128
(2 RCTs)
⊕⊕⊝⊝
LOW3,9
Smectite may make little or no difference in the appeareance of adverse events
0 per 10000 per 1000
(0 to 0)
Death-----There were no deaths in the included studies
Serious adverse events-----There were no serious side effects in the included studies

*The risk in the intervention group (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).

Abbreviations: CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio

GRADE Working Group grades of evidence
High certainty: We are very confident that the true effect lies close to that of the estimate of the effect.
Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.
Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.
Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Background

Description of the condition

Acute diarrhoea is defined as the passage of unusually loose or watery stools, usually at least three times in a 24-hour period, for less than 14 days (King 2003; WHO 2005; WHO/UNICEF 2013). Incidence of acute diarrhoea in children under five years of age is approximately two to three episodes per child per year (Walker 2013). The aetiology is usually infectious, and is usually transmitted by faecal-oral route, or by contaminated water or food. Although most cases of acute diarrhoea are self limited, the most common complication is dehydration where children are at higher risk compared to adults. The objective of treatment in many countries is to relieve symptoms and avoid complications. In low- and middle-income countries there are additional concerns to prevent dehydration and prevent the illness contributing to malnutrition. Therapeutic options for the latter objective include probiotics (Allen 2010), zinc (Lazzerini 2016), lactose-free formula (MacGillivray 2013), antibiotics, and antidiarrhoeal agents such as loperamide, racecadotril, and smectite.

Description of the intervention

Smectite is a medicinal clay commonly prescribed to reduce stool output in people with diarrhoea. A survey conducted in 29 European countries with a response rate of 34% found that 22% of physicians (9% in Western European countries and 41% in Eastern European countries) would give smectite as an adjuvant treatment to children with gastroenteritis (Szajewska 2000). In France, the use of smectite by private paediatricians may be as high as 84% (Uhlen 2004). Another survey, conducted in Prague, Czech Republic, found that 45.7% of children with acute diarrhoea received smectite (Kudlova 2010). A survey carried out in 20 hospitals in two Chinese provinces found that smectite was prescribed to 59.3% of adults with acute infectious diarrhoea (Hou 2013).

How the intervention might work

Dioctahedral smectite, or diosmectite, is a natural clay formed from sheets of aluminium and magnesium silicate. Its proposed mechanism of action differs from other antidiarrhoeal agents such as loperamide, which is an opioid-receptor agonist, and racecadotril, which acts as an enkephalinase inhibitor. Three possible mechanisms of action of smectite against diarrhoea have been proposed: an anti-inflammatory activity, alteration of the gut mucus barrier to reduce penetration of toxins, and adsorptive properties. These mechanisms have been replicated mainly in vitro and in animal models (Dupont 2009). In theory, these mechanisms would reduce stool output in children, thereby providing symptomatic relief and possibly preventing dehydration.

Why it is important to do this review

In many countries, symptomatic relief of diarrhoea is important to the public. Smectite is one such option for providing this relief. Two previous systematic reviews including 13 randomized controlled trials published between 1986 and 2013 provide evidence that smectite reduces the frequency and duration of diarrhoea in children (Das 2015; Szajewska 2006). The only reported adverse event was constipation. Since acute diarrhoea is usually a self limited illness, provided the person is properly hydrated, it is important to assess the efficacy and safety of adjuvant therapies such as smectite. With the publication of recent trials, there was a need to update the evidence on this topic.

Objectives

To assess the effects of smectite for treating acute infectious diarrhoea in children.

Methods

Criteria for considering studies for this review

Types of studies

Randomized and quasi-randomized trials comparing children with acute diarrhoea treated with smectite against a control group.

Types of participants

We included trials evaluating children, aged one month to 18 years old, with clinically defined diarrhoea of less than 14 days duration, presumed to be caused by an infectious agent. We excluded studies with other causes of diarrhoea, such as chronic or antibiotic-associated diarrhoea.

Types of interventions

We included trials assessing oral smectite against a control group, either placebo or no smectite. We did not exclude trials that administered other interventions, such as probiotics or zinc, provided that the intervention and control arms were treated identically.

Types of outcome measures

Primary outcomes
  • Duration of diarrhoea, measured in hours.

  • Clinical resolution at day 3 after starting treatment.

Secondary outcomes
  • Stool frequency, measured as number of depositions per day, on day 3 after starting treatment.

  • Stool output, measured in g or mL/kg per day.

  • Need for hospitalization.

  • Need for intravenous access for rehydration.

  • Death (from any cause or diarrhoea-related).

  • Adverse events:

    • serious adverse events (life-threatening events).

    • other adverse events (for example, constipation, vomiting, among others).

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 and strategy described in Appendix 1: Cochrane Infectious Diseases Group Specialized Register (27 June 2017); Cochrane Central Register of Controlled Trials (CENTRAL) (27 June 2017), published in the Cochrane Library (2017, Issue 5); MEDLINE (Pubmed; 1946 to 27 June 2017); Embase (Ovid; 1974 to 27 June 2017); and LILACS (Latin American and Caribbean Health Sciences Literature) (1982 to 27 June 2017). We also searched the metaRegister of Controlled Trials (mRCT) (27 June 2017) and the World Health Organization International Clinical Trials Registry Platform (WHO ICTRP) (27 June 2017) using ‘smectite' and ‘diosmectite' as search terms (Appendix 1).

Searching other resources

Conference proceedings

We searched the following conference proceedings of the last two years (2014 to 2016) for relevant abstracts.

  • Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC).

  • Infectious Diseases Society of America (IDSA) conferences.

  • International Congress on Infectious Diseases (ICID) from the International Society for Infectious Diseases (ISID).

Researchers and organizations

We contacted researchers, authors of included trials, other experts in the field of infectious diseases, and pharmaceutical companies that manufacture smectite.

Reference lists

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

Data collection and analysis

Selection of studies

Two review authors (GP and CC) independently screened the search results to identify potentially relevant trials and retrieved the full-text articles of these trials. GP and CC independently applied the inclusion criteria using an eligibility form, resolving any differences by discussing them with a third review author (VP or IF). We scrutinized the trial reports to ensure that multiple publications from the same trial were included only once. We listed the excluded studies and the reasons for their exclusion in the ‘Characteristics of excluded studies' section. Finally, when we were unsure whether a trial should be included because further information was needed, we attempted to contact the trial authors for clarification and allocated the trial to the ‘Studies awaiting classification' section. We have presented an adapted PRISMA flowchart showing study selection (Liberati 2009).

Data extraction and management

Two review authors (GP and CC) independently extracted prespecified characteristics of each trial using a standardized, piloted data extraction form. We attempted to contact trial authors in cases of unclear or missing data. We extracted the following data.

  • The numbers of randomized and analysed participants in each treatment group for each outcome.

  • The mean and standard deviation (SD) for each treatment group for continuous outcomes, and the number of participants with the event for each treatment group for dichotomous outcomes. If these values were not explicitly presented, we attempted to transform data where possible from available numbers such as 95% confidence intervals (CIs), standard errors, range or test statistics (that is, t, F, Z scores, P values, etc.). We obtained the SD from 95% CIs in one study according to the guidelines in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We imputed SDs for studies that did not present any measure of data dispersion. We extracted information from figures in three trials that presented the results in this format and did not provide numerical values for measures of dispersion (Dupont 2009a; Dupont 2009b; Pociecha 1998a; Pociecha 1998b), using the Plot Digitizer open source software (Jelicic 2016). Two trials presented the information using median and 95% CI and provided a Kaplan-Meier curve with the data for both intervention and control group (Dupont 2009a; Dupont 2009b). We applied the Hozo and colleagues approach to calculate the best estimation of mean and SD (Hozo 2005).

Assessment of risk of bias in included studies

Two review authors (GP and CC) independently assessed the risk of bias of the included studies, resolving any disagreements by discussion with a third review author (VP or IF). We attempted to contact trial authors regarding unclear or unspecified information. We used the Cochrane ‘Risk of bias' assessment tool, which includes the following domains (Higgins 2011).

  • Sequence generation: describe the method used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.

  • Allocation concealment: describe the method used to conceal the allocation sequence in sufficient detail to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment.

  • Blinding (masking) of participants, personnel, and outcome assessors: describe all measures used, if any, to mask trial participants, personnel, and outcome assessors from knowledge of which intervention a participant received. Provide any information relating to whether the intended masking was effective.

  • Incomplete outcome data: describe the completeness of outcome data for each main outcome, including attrition and exclusions from the analysis. State whether attrition and exclusions were reported, the numbers in each intervention group (compared with total randomized participants), reasons for attrition or exclusions where reported, and any re-inclusions in analyses performed by the review authors.

  • Selective outcome reporting: state how the possibility of selective outcome reporting was examined by the review authors and what was found.

  • Other sources of bias: state any important concerns about bias not addressed in the other domains in the tool.

We assessed the risk of bias for each component using ‘yes', ‘no', or ‘unclear' to indicate a low, high, or unclear risk of bias, respectively. We have presented the ‘Risk of bias' assessment in a ‘Risk of bias' graph and the ‘Risk of bias' tables.

Certainty of the evidence

We have presented the certainty of the evidence according to the GRADE approach. Two review authors (GP and CC) independently rated the certainty of the evidence for each outcome. Since we included randomized controlled trials, which are considered as high certainty, review authors could downgrade the body of evidence depending on five criteria: limitations, inconsistency, indirectness, imprecision, and publication bias. Evidence could be upgraded if a large effect size was found, if there was a dose-response association, or if trial authors considered plausible confounding factors. We have presented a summary of the evidence in a ‘Summary of findings' table, which provides key information about the best estimate of the magnitude of the effect in relative terms and absolute differences for each relevant comparison of alternative management strategies, numbers of participants and studies addressing each important outcome, and the rating of the overall certainty in effect estimates for each outcome. We used GRADEpro GDT to create the ‘Summary of findings' table (GRADEpro GDT).

Measures of treatment effect

For continuous outcomes, we used mean differences (MD) as the measure of effect with 95% CIs. For outcomes with different measurements, for example stool output, which can be measured in grams or mL per kg, we used standardized mean differences (SMD). For dichotomous outcomes, we used risk ratios (RR) as the measure of effect with 95% CIs.

Unit of analysis issues

Given the condition under study and the trial participants, we did not expect to find cluster randomized controlled trials or cross-over trials. When we found trials with repeated measurements, we decided on a single time point (for example, diarrhoea resolution at day 3).

Dealing with missing data

When there were no missing data, we carried out analyses according to the intention-to-treat principle, that is all children were analysed according to the group to which they were initially randomized. If there were missing data, we attempted to contact trial authors to request any missing data. If the trial authors did not respond within four to eight weeks, we conducted the analyses based on only the available information.

Assessment of heterogeneity

We used forest plots to detect overlapping CIs, and applied the Chi2 test with a P value < 0.10 to indicate statistical significance for heterogeneity. We investigated inconsistency with the I2 statistic, considering a value from 0% to 40% as not important.

Assessment of reporting biases

We assessed reporting biases by examining asymmetry of funnel plots.

Data synthesis

One review author (GP) analysed the data using Review Manager 5 (RevMan 2014). When appropriate, we combined data by meta-analysis using a fixed-effect model. When we found inconsistency (I2 statistic > 40%) or heterogeneity (Chi2 test at a significant P value < 0.10), we combined the results using the random-effects model.

Subgroup analysis and investigation of heterogeneity

We expected to perform subgroup analysis based on age groups, given that severity of disease might be different among infants, children, and adolescents. Since the higher burden and mortality of acute diarrhoea is in infants (Walker 2013), we analysed subgroups under and over two years of age.

Sensitivity analysis

We performed sensitivity analyses regarding risk of bias to investigate the robustness of the results, that is restricting the analysis by taking into account trials at low versus high or unclear risk of bias, as specified in the Assessment of risk of bias in included studies section. We explored if the following markers affected the direction of results: randomization, allocation concealment, blinding, follow-up, and missing data. We also performed a sensitivity analysis excluding the trials that required estimations and figure extractions (Dupont 2009a; Dupont 2009b; Pociecha 1998a; Pociecha 1998b).

Results

Description of studies

See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classification.

Results of the search

Our search strategy identified 34 potentially relevant studies, of which 22 studies were screened in full text. Eighteen studies met the inclusion criteria, and four were excluded (Dupont 1991; Dupont 1992; Karas 1996; Madkour 1994). The study flow diagram is shown in Figure 1. One reference included two studies (Dupont 2009a; Dupont 2009b). Another study is presented in the results as two separate studies because data were divided by age group (Pociecha 1998a; Pociecha 1998b).

Figure 1.

Study flow diagram.

Included studies

Study location

Eleven studies were conducted in low- or middle-income countries: Peru, Malaysia, Egypt, Thailand, India, Pakistan, Indonesia, and China (Dupont 2009a; Dupont 2009b; Lachaux 1986; Lexomboon 1994; Madkour 1993; Mujawar 2012; Rehman 2013; Vivatvakin 1992; Wang 1995; Widiasa 2009; Zong 1997). Seven were conducted in high-income countries: France, Italy, Lithuania, and Poland (Gilbert 1991; Guarino 2001; Lachaux 1986; Milocco 1999; Narkeviciute 2002; Pieścik-Lech 2013; Pociecha 1998a; Pociecha 1998b). Most trials were conducted in hospitals, with two studies conducted in both hospital and an ambulatory setting (Madkour 1993; Wang 1995), three exclusively with outpatients (Guarino 2001; Lexomboon 1994; Mujawar 2012), and two that did not specify (Gilbert 1991; Zong 1997).

Participants

Most studies included infants aged one to 24 months. One study did not include infants (Mujawar 2012), and one did not report age (Wang 1995). Nine studies included children aged two to 12 years old. No trials included adolescents. Two trials included only males (Dupont 2009a; Dupont 2009b). One report divided its results into two age groups: less than 12 months and 13 to 36 months (Pociecha 1998a; Pociecha 1998b).

Two studies included exclusively breastfed infants (Dupont 2009a; Dupont 2009b), and seven studies included children who were breastfed (Lexomboon 1994; Osman 1992; Pieścik-Lech 2013; Pociecha 1998a; Pociecha 1998b; Rehman 2013; Vivatvakin 1992; Widiasa 2009). One study excluded breastfed infants (Narkeviciute 2002). Thirteen trials reported rotavirus as the most frequent gastroenteritis aetiology. No studies included dysentery or bloody diarrhoea or children with cholera. One study included children with moderate malnutrition (Widiasa 2009), while the other studies excluded children with any degree of malnutrition.

Most trials defined diarrhoea as three or more loose stools, but the duration varied among studies: four defined it as less than two days (Guarino 2001; Lexomboon 1994; Mujawar 2012; Widiasa 2009); six as less than three days (Dupont 2009a; Dupont 2009b; Narkeviciute 2002; Pociecha 1998a; Pociecha 1998b; Rehman 2013; Vivatvakin 1992); one as less than four days (Lachaux 1986); five as less than five days (Madkour 1993; Milocco 1999; Pieścik-Lech 2013; Wang 1995; Zong 1997); one as less than seven days (Osman 1992); and one referred to it as "recent" (Gilbert 1991).

Interventions

Doses of smectite varied between 1 g and 6 g per dose, and frequency of administration varied from once daily to every six hours. Most trials used 1.5 g per dose in infants less one year and 3 g in older infants or children. Two trials administered 3 g twice a day for three days, and then once a day in infants less than one year, and double the dose in older children (Dupont 2009a; Dupont 2009b). Five trials gave 1.5 g of smectite twice a day to infants less than one year, with double the dose for older children (Gilbert 1991; Guarino 2001; Milocco 1999; Pociecha 1998a; Pociecha 1998b; Wang 1995). Two studies gave a loading dose of 3 g (Lexomboon 1994; Narkeviciute 2002). Two trials administered smectite every eight hours (Mujawar 2012; Rehman 2013), and one study gave it every six hours (Madkour 1993). Two trials gave smectite every eight hours to children weighing less than 10 kg, and every six hours to children above 10 kg (Osman 1992; Vivatvakin 1992). Two studies gave Lactobacillus rhamnosus GG to both the intervention and the control group (Pieścik-Lech 2013; Pociecha 1998a; Pociecha 1998b). Two studies did not report the dose (Widiasa 2009; Zong 1997).

The duration of treatment also differed among studies. Four studies gave smectite until recovery (Dupont 2009a; Dupont 2009b; Narkeviciute 2002; Pieścik-Lech 2013); two administered the treatment for three days (Madkour 1993; Milocco 1999); five for five days (Mujawar 2012; Osman 1992; Rehman 2013; Vivatvakin 1992; Widiasa 2009); and one for six days (Pociecha 1998a; Pociecha 1998b). The remaining studies did not specify the duration of treatment.

Outcomes
Primary outcomes

Fifteen studies reported the duration of diarrhoea (Dupont 2009a; Dupont 2009b; Gilbert 1991; Guarino 2001; Lachaux 1986; Madkour 1993; Milocco 1999; Mujawar 2012; Narkeviciute 2002; Pieścik-Lech 2013; Pociecha 1998a; Pociecha 1998b; Rehman 2013; Vivatvakin 1992; Widiasa 2009; Zong 1997), but the outcome was defined differently. Six trials defined it as time to the last loose stool (Guarino 2001; Madkour 1993; Narkeviciute 2002; Pieścik-Lech 2013; Vivatvakin 1992; Widiasa 2009); three as time to first formed stool (Dupont 2009a; Lachaux 1986; Rehman 2013); one as time to first soft or formed stool (Dupont 2009b); three as time to normalization of stools (Gilbert 1991; Mujawar 2012; Pociecha 1998a; Pociecha 1998b); and two did not provide a clear definition (Milocco 1999; Zong 1997).

Five trials reported clinical resolution of diarrhoea at day 3 (Lachaux 1986; Lexomboon 1994; Madkour 1993; Osman 1992; Vivatvakin 1992).

Secondary outcomes

Four studies reported stool frequency: three as number of depositions per day (Guarino 2001; Madkour 1993; Osman 1992), and one as the total number of stools during follow-up (Milocco 1999). Three trials reported stool output as grams per kilogram of child's weight at 72 hours (Dupont 2009a; Dupont 2009b), and one in grams per day (Osman 1992). Two studies reported need for hospitalization (Guarino 2001; Pieścik-Lech 2013). One study reported need for intravenous access for rehydration (Pieścik-Lech 2013). No studies reported deaths.

Risk of bias in included studies

See: Characteristics of included studies; Figure 2; Figure 3 for the risk of bias in included studies.

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

Seven studies had an adequate description of randomization method (Dupont 2009a; Dupont 2009b; Lachaux 1986; Madkour 1993; Pieścik-Lech 2013; Rehman 2013; Widiasa 2009). In five trials the information about random allocation was unclear (Gilbert 1991; Lexomboon 1994; Pociecha 1998a; Pociecha 1998b; Wang 1995; Zong 1997). Five studies were quasi-randomized trials in which children were allocated alternately, by birthday or serial number (Guarino 2001; Milocco 1999; Mujawar 2012; Narkeviciute 2002; Osman 1992). We suspected selection bias in one study as groups differed in the aetiology of diarrhoea, and the method of randomization was not described (Vivatvakin 1992).

Five studies adequately described allocation concealment (Dupont 2009a; Dupont 2009b; Madkour 1993; Pieścik-Lech 2013; Widiasa 2009). We considered all quasi-randomized trials as having high risk of bias regarding allocation concealment.

Blinding

Eight trials were reported as double-blind and used a placebo as control (Dupont 2009a; Dupont 2009b; Gilbert 1991; Lachaux 1986; Madkour 1993; Pieścik-Lech 2013; Rehman 2013; Widiasa 2009). The remaining trials were not blinded (Guarino 2001; Lexomboon 1994; Milocco 1999; Mujawar 2012; Narkeviciute 2002; Osman 1992; Pociecha 1998a; Pociecha 1998b; Vivatvakin 1992; Wang 1995; Zong 1997).

Incomplete outcome data

Fourteen trials had appropriate follow-up and analysis of more than 90% of participants. Two included less than 90% in the analysis (Dupont 2009a; Osman 1992). In one trial information was insufficient to permit judgement (Guarino 2001).

Selective reporting

Two trials had a registered protocol (Dupont 2009a; Dupont 2009b).

Effects of interventions

See: Summary of findings for the main comparison Smectite compared to control for acute infectious diarrhoea in children

Primary outcomes

1.1 Duration of diarrhoea

Overall, duration of diarrhoea was reduced by approximately 24 hours (mean difference (MD) -24.38, 95% confidence interval (CI) -30.91 to -17.85; 14 trials; 2209 children, Analysis 1.1; low-certainty evidence). There was significant heterogeneity (I2 = 96%). This high inconsistency was due to differences in effect size of the benefit, not because of opposing directions of effects (Figure 4).

Figure 4.

Forest plot of comparison: 1 Diarrhoea primary outcomes, outcome: 1.1 Mean duration of diarrhoea (hours).

A sensitivity analysis exploring the effect of randomization, allocation concealment, blinding, and follow-up did not change the result of the meta-analysis significantly. Sensitivity analysis excluding the trials that required estimations and figure extractions did not significantly change the result of the meta-analysis (MD -22.07, 95% CI -30.38 to -13.76) (Dupont 2009a; Dupont 2009b; Pociecha 1998a; Pociecha 1998b).

On visual inspection, the funnel plot was roughly symmetric, with most studies centred together at the top, probably reflecting spuriously small standard deviations of the continuous outcome that is skewed (Figure 5).

Figure 5.

Funnel plot of comparison: 1 Diarrhoea primary outcomes, outcome: 1.1 Mean duration of diarrhoea (hours).

1.2 Duration of diarrhoea, infants less than two years

Five studies included only infants younger than two years (Gilbert 1991; Lachaux 1986; Madkour 1993; Rehman 2013; Vivatvakin 1992). One study reported results for infants less than 12 months (Pociecha 1998a). Smectite reduced the duration of diarrhoea by 24 hours (MD -24.11, 95% CI -31.35 to -16.87; 441 infants; Analysis 1.2). Other studies included both infants and children, but they did not provide enough information to be able to perform a subgroup analysis according to age (Figure 6).

Figure 6.

Forest plot of comparison: 1 Diarrhoea primary outcomes, outcome: 1.2 Mean duration of diarrhoea, studies including only infants < 2 years.

1.3 Clinical resolution at day 3 after starting treatment

Smectite increased the rate of clinical resolution at day 3 (risk ratio (RR) 2.10, 95% CI 1.30 to 3.39; 5 trials; 312 children; Analysis 1.3; low-certainty evidence) (Figure 7). After performing a sensitivity analysis excluding trials with high risk of bias (Osman 1992; Vivatvakin 1992), the pooled effect was not significant (RR 1.90, 95% CI 0.96 to 3.77; 3 trials; 190 children).

Figure 7.

Forest plot of comparison: 1 Diarrhoea primary outcomes, outcome: 1.3 Clinical resolution at day 3 after starting treatment.

Secondary outcomes

2.1 Stool frequency

Three studies measured stool frequency as number of depositions per day, all of them reporting data on day 3 (Guarino 2001; Madkour 1993; Osman 1992). Smectite reduced stool frequency by one (MD -1.33, 95% CI -2.28 to -0.38; 3 trials; 954 children; Analysis 2.1; very low-certainty evidence) (Figure 8). One study measured stool frequency as total number of depositions during follow-up; the mean number of depositions was 10 in both groups (Milocco 1999).

Figure 8.

Forest plot of comparison: 2 Diarrhoea secondary outcomes, outcome: 2.1 Stool frequency, measured as number of depositions per day, on day 3 after starting treatment.

2.2 Stool output

Four studies evaluated stool output. Three studies reported cumulative stool output at 72 hours (Dupont 2009a; Dupont 2009b; Madkour 1993). Smectite reduced stool output by 11 g/kg (MD -11.37, 95% CI -21.94 to -0.79; 3 trials; 634 children; Analysis 2.2; low-certainty evidence) (Figure 9). Another study was not pooled because the authors reported stool output as stool weight in total grams per day with an effect estimate favouring smectite (mean of 255.67 g in the smectite group versus 741.33 g in the control group) at day 3 of treatment (Osman 1992).

Figure 9.

Forest plot of comparison: 2 Diarrhoea secondary outcomes, outcome: 2.2 Stool output, measured in g/kg at 72 hours.

2.3 Need for hospitalization

Two studies reported data on need for hospitalization. There was no evidence of benefit using smectite (RR 0.93, 95% CI 0.75 to 1.15; 2 trials; 885 children; Analysis 2.3; low-certainty evidence) (Figure 10).

Figure 10.

Forest plot of comparison: 2 Diarrhoea secondary outcomes, outcome: 2.3 Need for hospitalization.

2.4 Need for intravenous access for rehydration

There was no evidence of an effect on need for intravenous rehydration (RR 0.77, 95% CI 0.54 to 1.11; 1 trial; 81 children; Analysis 2.4; moderate-certainty evidence).

2.5 Death (from any cause or diarrhoea-related)

No deaths were reported in any of the included trials.

2.6 Serious adverse events (life-threatening events)

There were no reports of serious adverse events.

2.7 Other adverse events (constipation, vomiting)

The most commonly reported adverse effect was constipation. However, the risk of constipation using smectite was very uncertain due to imprecision, with very few events and wide confidence intervals (RR 4.71, 95% CI 0.56 to 39.19; 2 trials; 128 children; Analysis 2.5; low-certainty evidence) (Figure 11). There were also no differences between groups regarding vomiting or fever. Another minor adverse event mentioned in trials was bad taste, but there were no specific numbers for the intervention and control groups.

Figure 11.

Forest plot of comparison: 2 Diarrhoea secondary outcomes, outcome: 2.5 Constipation.

Discussion

Summary of main results

We identified 18 studies that compared smectite to a control group. Overall, smectite reduced the duration of diarrhoea by approximately a day, increased clinical resolution by day 3, and had a modest benefit on stool frequency and output. This evidence of benefit persisted after a sensitivity analysis accounting for randomization method, even though five trials were quasi-randomized. Eight trials reported the inclusion of breastfed infants.

There was no evidence of an effect on the need for hospitalization or intravenous rehydration, deaths, or serious side effects.

Overall completeness and applicability of evidence

Studies were conducted in diverse settings in both high-income and low- or middle-income countries, and including both ambulatory and hospital patients. Aetiology also varied, with most trials including a large proportion of children with rotavirus. Most studies excluded children with malnutrition. Most of the studies were funded by the industry. Although external funding and commercial interests are well recognized as a potential source of bias in clinical trials, most investigators provided reasonable information that shows that the manufacturers had no, or a very limited, active role in the design and conduct of the studies.

Quality of the evidence

We assessed the certainty of the evidence using the GRADE system, which is displayed in ‘Summary of findings' table 1 (Summary of findings for the main comparison). Overall, the certainty of the body of evidence ranged from very low to moderate. For our primary outcomes, the certainty of evidence was low mainly due to concerns of risk of bias and inconsistency of the results. Regarding risk of bias, we included four quasi-randomized trials, and another four trials did not clearly describe the randomization process. Also, seven trials were not blinded.

The high heterogeneity observed may be due to differences in the definition of the condition, the age of participants, and the different aetiologies. In one study, Pieścik-Lech 2013, another explanation for heterogeneity could be that both the intervention and the control group received a probiotic, but the other study that added a probiotic as a co-intervention did not contribute to such inconsistency (Pociecha 1998a; Pociecha 1998b). The high inconsistency observed was mainly due to differences in effect size of the benefit and not because of opposing directions of effects.

Potential biases in the review process

We made every attempt to limit biases during the review process by ensuring a comprehensive search for potentially eligible studies. We believe that the authors’ independent assessments of eligibility of studies for inclusion and data extraction have minimized the potential for additional bias beyond that detailed in the ‘Risk of bias' tables in the Characteristics of included studies and in the funnel plot.

Agreements and disagreements with other studies or reviews

Our findings agree with those of previous systematic reviews (Das 2015; Szajewska 2006). Due to the differences in time of publication, our review includes more trials than the review by Szajewska 2006, and assesses the certainty of the evidence based on the GRADE approach. The review by Das 2015 included 13 out of the 18 studies that were included in this review. Szajewska 2006 reported a reduction of 22.7 hours in the duration of diarrhoea, while Das 2015 reported 22.39 hours. Szajewska 2006 and Das 2015 also report significant results for cure rate at day 3. While Das 2015 reported clinical resolution at day 5 and 7, we considered day 3 to be more clinically relevant.

Authors' conclusions

Implications for practice

Smectite reduces the duration of symptoms of infectious diarrhoea by a day, and at least 17 hours, and increases clinical resolution at day 3. The effect on stool frequency and output is modest. Although smectite did not have an effect on other relevant outcomes such as the need for intravenous therapy or hospitalization, fewer hours of diarrhoea may be considered clinically significant in different settings and contexts, taking into account that most cases of infectious diarrhoea are self limited and resolve within three to five days with adequate hydration and medical care.

Implications for research

Further research with a focus on adequate randomization and blinding is needed. Future studies may explore the causes of heterogeneity in the effect of smectite, its possible benefit in vulnerable populations such as children under two years of age or with malnutrition, and its effect on certain specific aetiologies such as rotavirus or dysentery producing bacteria. Economic analyses will also provide information to guide practice in different countries or settings.

Acknowledgements

The editorial base for the Cochrane Infectious Diseases Group is funded by UK aid from the UK Government for the benefit of low- and middle-income countries (Grant: 5242).

Data and analyses

Download statistical data

Comparison 1. Diarrhoea primary outcomes
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Mean duration of diarrhoea152209Mean Difference (IV, Random, 95% CI)-24.38 [-30.91, -17.85]
2 Mean duration of diarrhoea, studies including only infants < 2 years6441Mean Difference (IV, Random, 95% CI)-24.11 [-31.35, -16.87]
3 Clinical resolution at day 3 after starting treatment5312Risk Ratio (M-H, Random, 95% CI)2.10 [1.30, 3.39]
Analysis 1.1.

Comparison 1 Diarrhoea primary outcomes, Outcome 1 Mean duration of diarrhoea.

Analysis 1.2.

Comparison 1 Diarrhoea primary outcomes, Outcome 2 Mean duration of diarrhoea, studies including only infants < 2 years.

Analysis 1.3.

Comparison 1 Diarrhoea primary outcomes, Outcome 3 Clinical resolution at day 3 after starting treatment.

Comparison 2. Diarrhoea secondary outcomes
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Stool frequency, measured as number of depositions per day, on day 3 after starting treatment3954Mean Difference (IV, Random, 95% CI)-1.33 [-2.28, -0.38]
2 Stool output, measured in g or mL/kg per day3634Mean Difference (IV, Random, 95% CI)-11.37 [-21.94, -0.79]
3 Need for hospitalization2885Risk Ratio (M-H, Random, 95% CI)0.93 [0.75, 1.15]
4 Need for intravenous access for rehydration181Risk Ratio (M-H, Random, 95% CI)0.77 [0.54, 1.11]
5 Constipation2128Risk Ratio (M-H, Random, 95% CI)4.71 [0.56, 39.19]
Analysis 2.1.

Comparison 2 Diarrhoea secondary outcomes, Outcome 1 Stool frequency, measured as number of depositions per day, on day 3 after starting treatment.

Analysis 2.2.

Comparison 2 Diarrhoea secondary outcomes, Outcome 2 Stool output, measured in g or mL/kg per day.

Analysis 2.3.

Comparison 2 Diarrhoea secondary outcomes, Outcome 3 Need for hospitalization.

Analysis 2.4.

Comparison 2 Diarrhoea secondary outcomes, Outcome 4 Need for intravenous access for rehydration.

Analysis 2.5.

Comparison 2 Diarrhoea secondary outcomes, Outcome 5 Constipation.

Appendices

Appendix 1. Search strategy

Search setCIDG SRaCENTRALMEDLINEbEmbasebLILACSb
1smectiteSmectite [Supplementary concept]Smectite [Supplementary concept]Smectite ti, absmectite
2diosmectitesmectite* ti, absmectite* ti, abDiosmectite ti, abdiosmectite
31 or 2Diosmectite ti, abDiosmectite ti, ab1 or 21 or 2
4"smecta"[Supplementary Concept]"smecta"[Supplementary Concept]Limit 3 to human-
5-1 or 2 or 3 or 41 or 2 or 3 or 4--
6-Limit 5 to humansLimit 5 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 2011); upper case: MeSH or EMTREE heading; lower case: free text term.

Contributions of authors

Giordano Pérez-Gaxiola and Carlos Cuello-García prepared the protocol and manuscript. Ivan D Florez checked the protocol and manuscript, and provided advice. Víctor Pérez-Pico checked the protocol and provided advice as an infectious diseases specialist.

Declarations of interest

Giordano Pérez-Gaxiola, Carlos A Cuello-García, Ivan D Florez, Víctor M Pérez-Pico: we certify that we have no affiliations with or involvement in any organization or entity with a direct financial interest in the subject matter of this Cochrane Review (for example, employment, consultancy, stock ownership, honoraria, or expert testimony).

Sources of support

Internal sources

  • Liverpool School of Tropical Medicine, UK.

External sources

  • Department for International Development (DFID), UK.

    Grant: 5242

Differences between protocol and review

We added Ivan D Florez as an author.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Dupont 2009a

Methods

Randomized controlled trial

Length of follow-up: not stated

Participants

Number: 300 enrolled children

Inclusion criteria: inpatients; well-nourished male infants and children aged 1 to 36 months with watery diarrhoea < 3 days duration, with 3 watery stools per day and at least 1 watery stool in the past 24 hours; mild-to-moderate dehydration

Exclusion criteria: severe dehydration or malnutrition, bloody diarrhoea, fever 39 ºC or higher, previous medications

Breastfeeding: exclusively breastfed infants were excluded

Interventions

Intervention group: diosmectite. Dosage 3 g twice a day for 3 days, then 3 g daily for infants younger than 12 months. Double the dose for older children

Control: placebo

Outcomes

Duration of diarrhoea (until first formed stool)

Stool output in g/kg in the first 72 hrs

Notes

Location: Peru

Setting: urban

Cause of diarrhoea: rotavirus 22%. Other aetiologies not specified.

Source of funding: industry

Registration number: NCT00352716

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskDescribed as randomized in sequential ascending order by a statistician
Allocation concealment (selection bias)Low riskSponsor-assigned biostatistician prepared a list of treatment allocation codes.
Blinding of participants and personnel (performance bias)
All outcomes
Low riskPlacebo was identical to diosmectite in size, weight, colour, smell, taste, and appearance, and was inert. Blinding seems appropriate.
Blinding of outcome assessment (detection bias)
All outcomes
Low riskBlind review of data by outcome assessors
Incomplete outcome data (attrition bias)
All outcomes
High risk40 children (13%) were non-adherent, and the rest analysed as per protocol.
Selective reporting (reporting bias)Low riskNone detected. Registered trial
Other biasLow riskNo other biases detected.

Dupont 2009b

Methods

Randomized controlled trial

Length of follow-up: not stated

Participants

Number: 302 enrolled children

Inclusion criteria: inpatients; well-nourished male infants and children aged 1 to 36 months with watery diarrhoea < 3 days duration, with 3 watery stools per day and at least 1 watery stool in the past 24 hours; mild-to-moderate dehydration

Exclusion criteria: severe dehydration or malnutrition, bloody diarrhoea, fever 39 ºC or higher, previous medications

Breastfeeding: exclusively breastfed infants were excluded

Interventions

Intervention group: diosmectite. Dosage 3 g twice a day for 3 days, then 3 g daily for infants younger than 12 months. Double the dose for older children

Control: placebo

Outcomes

Duration of diarrhoea (until first soft or formed stool)

Stool output in g/kg in the first 72 hrs

Notes

Location: Malaysia

Setting: urban

Cause of diarrhoea: rotavirus 12%. Other aetiologies not specified.

Source of funding: industry

Registration number: NCT00352989

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskDescribed as randomized in sequential ascending order by a statistician
Allocation concealment (selection bias)Low riskSponsor-assigned biostatistician prepared a list of treatment allocation codes.
Blinding of participants and personnel (performance bias)
All outcomes
Low riskPlacebo was identical to diosmectite in size, weight, colour, smell, taste, and appearance, and was inert. Blinding seems appropriate.
Blinding of outcome assessment (detection bias)
All outcomes
Low riskBlind review of data by outcome assessors
Incomplete outcome data (attrition bias)
All outcomes
Low risk16 children (5%) were non-adherent, and the rest analysed as per protocol.
Selective reporting (reporting bias)Low riskNone detected. Registered trial
Other biasLow riskNo other biases detected.

Gilbert 1991

Methods

Randomized controlled trial

Length of follow-up: not stated

Participants

Number: 56 enrolled children

Inclusion criteria: inpatients; children aged 2 to 24 months with moderate-to-severe acute diarrhoea

Exclusion criteria: malnutrition

Breastfeeding: not specified

Interventions

Intervention group: diosmectite. Dosage 1.5 g twice a day for infants younger than 12 months. Double the dose for older children

Control: placebo

Another control group received loperamide 0.11 mg/kg every 8 hours.

Outcomes

Duration of diarrhoea (time to normalization of stools)

Stool frequency on day 5

Notes

Location: France

Setting: urban

Cause of diarrhoea: rotavirus 18%, Staphylococcus aureus 3%, Escherichia coli 3%, Campylobacter spp. 3%, Candida spp. 1%

Source of funding: not stated

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskStated as randomized but no method described.
Allocation concealment (selection bias)Unclear riskNo method described.
Blinding of participants and personnel (performance bias)
All outcomes
Low riskUse of placebo; probably adequate blinding
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskInsufficient information to permit judgement
Incomplete outcome data (attrition bias)
All outcomes
Low riskPer-protocol analysis. 4 children (7%) excluded and not analysed.
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasLow riskNo other biases detected.

Guarino 2001

Methods

Quasi-randomized controlled trial

Length of follow-up: not stated

Participants

Number: 804 enrolled children

Inclusion criteria: outpatients; well-nourished children aged 3 to 60 months with acute diarrhoea of mild-to-moderate severity < 2 days duration, with 3 watery stools per day

Exclusion criteria: malnutrition, chronic diseases, previous medications

Breastfeeding: not specified

Interventions

Intervention group: diosmectite. Dosage 1.5 g twice a day for infants younger than 12 months. Double the dose for older children

Control: no medication

Outcomes

Duration of diarrhoea (from first to the last liquid–loose stool output preceding the return of normal stools)

Diarrhoea at day 7

Vomiting

Fever

Hospitalization rate

Notes

Location: Italy

Setting: urban

Cause of diarrhoea: rotavirus 4%, not specified bacterial aetiology 1%

Source of funding: industry

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High riskNot randomized. Participants selected in sequential one-to-one basis.
Allocation concealment (selection bias)High riskNot concealed
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot blinded
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot blinded
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInsufficient information to permit judgement
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasLow riskNo other biases detected.

Lachaux 1986

Methods

Randomized controlled trial

Length of follow-up: not stated

Participants

Number: 36 enrolled infants

Inclusion criteria: inpatients; infants aged 2 to 24 months with acute watery diarrhoea < 4 days duration, with mild-to-moderate dehydration

Exclusion criteria: previous medications, concomitant illness

Breastfeeding: not stated

Interventions

Intervention group: diosmectite. Dosage 3 g per day to infants < 1 year, 6 g per day to infants > 1 year. Duration not stated.

Control: placebo

Outcomes

Duration of diarrhoea (from first drug administration to last liquid stool before a formed one)

Clinical resolution at day 3 and 5

Adverse events

Notes

Location: France

Setting: urban

Cause of diarrhoea: rotavirus 77%

Source of funding: not specified

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskStated as "drawing lots"
Allocation concealment (selection bias)Unclear riskInsufficient information to permit judgement
Blinding of participants and personnel (performance bias)
All outcomes
Low riskUse of placebo; probably adequate blinding
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskUse of placebo; probably adequate blinding
Incomplete outcome data (attrition bias)
All outcomes
Low riskOnly 1 loss per group
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasLow riskNo other biases detected.

Lexomboon 1994

Methods

Randomized controlled trial

Length of follow-up: 5 days

Participants

Number: 66 enrolled children

Inclusion criteria: outpatients; well-nourished children aged 1 to 24 months with acute diarrhoea < 2 days duration, with 3 watery stools within 24 hours

Exclusion criteria: severe dehydration, dysentery, fever higher than 38.5 ºC, previous medications

Breastfeeding: included

Interventions

Intervention group: diosmectite. Dosage: loading dose of 3 g, then 1.5 g twice a day

Control: no medication

Outcomes

Diarrhoea at day 3 and 5

Tolerability

Notes

Location: Thailand

Setting: urban

Cause of diarrhoea: rotavirus 27%, Campylobacter jejuni 8%, enteropathogenic Escherichia coli 5%, Salmonella spp. 6%, Shigella spp. 3%, Plesiomonas shigelloides 2%

Source of funding: industry

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskStated as randomized. No method described.
Allocation concealment (selection bias)Unclear riskNo method described.
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot blinded
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo losses to follow-up
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasLow riskNo other biases detected.

Madkour 1993

Methods

Randomized controlled trial

Length of follow-up: until recovery from diarrhoea

Participants

Number: 90 enrolled children

Inclusion criteria: inpatients; well-nourished male children aged 3 to 24 months with watery diarrhoea < 5 days duration, with dehydration of any severity

Exclusion criteria: prolonged diarrhoea, malnutrition, major illnesses

Breastfeeding: not specified

Interventions

Intervention group: diosmectite. Dosage 1.5 g, 4 times daily for 3 days

Control group: placebo

Outcomes

Duration of diarrhoea (from enrolment to last liquid stool)

Frequency of diarrhoea

Duration of vomiting

Feeding pattern

Notes

Location: Egypt

Setting: urban

Cause of diarrhoea: rotavirus 16%. Other aetiologies not specified.

Source of funding: WHO and industry

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskBlock randomization by Diarrheal Disease Control Programme of the WHO
Allocation concealment (selection bias)Low riskNumerically coded envelopes
Blinding of participants and personnel (performance bias)
All outcomes
Low riskUse of placebo
Blinding of outcome assessment (detection bias)
All outcomes
Low riskUse of placebo
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo losses to follow-up
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasLow riskNo other biases detected.

Milocco 1999

Methods

Quasi-randomized controlled trial

Length of follow-up: not stated

Participants

Number: 35 enrolled children

Inclusion criteria: inpatients; children aged 0 to 60 months with watery diarrhoea < 5 days duration

Exclusion criteria: not stated

Breastfeeding: not specified

Interventions

Intervention group: diosmectite. Dosage 1.5 g twice a day for infants younger than 12 months. Double the dose for older children

Control: no medication

Outcomes

Number of stools at 48 hrs

Duration of diarrhoea (not clearly defined)

Fever, vomiting, weight loss

Notes

Location: Italy

Setting: urban

Cause of diarrhoea: rotavirus 23%, Salmonella spp. 11%, Cryptosporidium 6%

Source of funding: not stated

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High riskParticipants selected "alternatively"; not truly random.
Allocation concealment (selection bias)High riskNot concealed
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot blinded
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot blinded
Incomplete outcome data (attrition bias)
All outcomes
Low risk2 losses to follow-up in intervention group
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasLow riskNo other biases detected.

Mujawar 2012

Methods

Quasi-randomized controlled trial

Length of follow-up: not stated

Participants

Number: 117 enrolled children

Inclusion criteria: outpatients; well-nourished children aged 24 to 60 months with watery diarrhoea < 2 days duration; mild-to-moderate dehydration

Exclusion criteria: bloody diarrhoea, chronic illness, previous medications

Breastfeeding: not specified

Interventions

Intervention group: diosmectite. Dosage 1.5 g thrice a day for 5 days

Control group: no medication

Outcomes

Duration of diarrhoea (until normal stool consistency)

Complications: severe dehydration, severe dysentery, respiratory infection, and anaemia

Notes

Location: India

Setting: urban

Cause of diarrhoea: not specified

Source of funding: not specified

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High riskNot truly random; participants selected by serial number.
Allocation concealment (selection bias)High riskNot concealed
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot blinded
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskIntention-to-treat analysis. 8 children (7%) were lost to follow-up and were included in the analysis.
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasLow riskNo other biases detected.

Narkeviciute 2002

Methods

Quasi-randomized controlled trial

Length of follow-up: not stated

Participants

Number: 54 enrolled children

Inclusion criteria: inpatients; well-nourished infants and children aged 6 to 48 months with watery diarrhoea < 3 days duration, with 3 watery stools per day; mild-to-moderate dehydration

Exclusion criteria: severe dehydration, malnutrition, chronic or concomitant illness

Breastfeeding: excluded

Interventions

Intervention group: diosmectite. Dosage: loading dose of 3 g; 1.5 g, 3 times a day for children < 10 kg, 4 times a day for > 10 kg

Control: no medication

Outcomes

Duration of diarrhoea (time to last watery/semiliquid stool)

Length of stay

Notes

Location: Lithuania

Setting: urban

Cause of diarrhoea: rotavirus 70%, enteropathogenic Escherichia coli 4%, Campylobacter spp. 8%

Source of funding: industry

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High riskNot truly random; participants selected by birthday.
Allocation concealment (selection bias)High riskNot concealed
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot blinded
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo losses to follow-up
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasLow riskNo other biases detected.

Osman 1992

Methods

Quasi-randomized controlled trial

Length of follow-up: 5 days

Participants

Number: 71 infants and children

Inclusion criteria: inpatients; infants and children (no age limit specified, mean age of 13 months) with acute watery diarrhoea < 7 days duration, with mild-to-moderate dehydration

Exclusion criteria: systemic illness; previous use of antibiotics or antidiarrhoeal agents; malnutrition

Breastfeeding: included

Interventions

Intervention group: diosmectite. Dosage 1.5 g every 8 hours to infants < 10 kg, 1.6 g every 6 hours to infants > 10 kg for a maximum of 5 days

Control: no medication

Outcomes

Clinical resolution (return of stools to previous formed consistency and average number of frequency)

Stool output (g/kg)

Stool frequency

Notes

Location: Egypt

Setting: urban

Cause of diarrhoea: rotavirus 43%, bacterial (not specified) 23%

Source of funding: not specified

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High riskNot truly random; participants selected alternately.
Allocation concealment (selection bias)High riskNot concealed
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot blinded
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot blinded
Incomplete outcome data (attrition bias)
All outcomes
High riskPer-protocol analysis with 4 exlusions in intervention group (12%) and 7 losses in control group (19%)
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasUnclear riskNo other biases detected.

Pieścik-Lech 2013

Methods

Randomized controlled trial

Length of follow-up: 7 days

Participants

Number: 88 enrolled children

Inclusion criteria: inpatients/outpatients; well-nourished infants and children aged 4 to 60 months with watery diarrhoea < 5 days duration, with 3 watery stools per day

Exclusion criteria: recent history of diarrhoea, chronic diseases

Breastfeeding: included

Interventions

Intervention group: diosmectite, dose 3 g once daily until diarrhoea stopped, plus Lactobacillus GG, dose of 6 x 109 colony forming units, once daily for 7 days

Control group: placebo plus Lactobacillus GG

Outcomes

Duration of diarrhoea (time from randomization until the last watery stool, or at least 12 h with no stool)

Stool frequency

Consistency of stools

Need for antibiotic therapy

Diarrhoea recurrence

Need for hospitalization

Need for intravenous rehydration therapy

Notes

Location: Poland

Setting: urban

Cause of diarrhoea: rotavirus 60%, adenovirus 5%, Salmonella spp. 5%, Staphylococcus aureus 3%, enteropathogenic Escherichia coli 1%

Source of funding: Medical University of Warsaw

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated block randomization
Allocation concealment (selection bias)Low riskRandomization prepared by independent investigator.
Blinding of participants and personnel (performance bias)
All outcomes
Low riskUse of placebo
Blinding of outcome assessment (detection bias)
All outcomes
Low riskUse of placebo
Incomplete outcome data (attrition bias)
All outcomes
Low riskPer-protocol analysis with 7 losses in control group (8%)
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasLow riskNo other biases detected.

Pociecha 1998a

Methods

Randomized controlled trial

Length of follow-up: 28 days

Participants

Number: 56 enrolled infants

Inclusion criteria: inpatients; well-nourished infants ≤ 12 months with watery diarrhoea of rotaviral aetiology < 3 days duration, with moderate dehydration

Exclusion criteria: chronic diseases, aetiologies other than rotavirus

Breastfeeding: included

Interventions

Intervention group: diosmectite, dose 1.5 g twice daily for 6 days, plus Lactobacillus GG in "age dependent dose"

Control group: Lactobacillus GG

A third group received polyvinylpolypyrrolidone plus Lactobacillus GG.

Outcomes

Duration of intravenous rehydration

Duration of fever and vomiting

Duration of diarrhoea (time to normalization of consistency of the stool or a day without stool)

Need of hospitalization after discharge

Notes

Location: Poland

Setting: urban

Cause of diarrhoea: rotavirus 100%

Source of funding: not stated

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskStated as randomized. No method described.
Allocation concealment (selection bias)Unclear riskNo method described.
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot blinded
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo losses to follow-up
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasUnclear riskNo other biases detected.

Pociecha 1998b

Methods

Randomized controlled trial

Length of follow-up: 28 days

Participants

Number: 105 enrolled infants

Inclusion criteria: inpatients; well-nourished infants > 12 months with watery diarrhoea of rotaviral aetiology < 3 days duration, with moderate dehydration

Exclusion criteria: chronic diseases, aetiologies other than rotavirus

Breastfeeding: included

Interventions

Intervention group: diosmectite, dose 3 g twice daily for 6 days, plus Lactobacillus GG in "age dependent dose"

Control group: Lactobacillus GG

A third group received polyvinylpolypyrrolidone plus Lactobacillus GG.

Outcomes

Duration of intravenous rehydration

Duration of fever and vomiting

Duration of diarrhoea (time to normalization of consistency of the stool or a day without stool)

Need of hospitalization after discharge

Notes

Location: Poland

Setting: urban

Cause of diarrhoea: rotavirus 100%

Source of funding: not stated

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskStated as randomized. No method described.
Allocation concealment (selection bias)Unclear riskNo method described.
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot blinded
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo losses to follow-up
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasUnclear riskNo other biases detected.

Rehman 2013

Methods

Randomized controlled trial

Length of follow-up: 6 days

Participants

Number: 206 enrolled children

Inclusion criteria: inpatients; well-nourished infants and children aged 6 to 24 months with watery diarrhoea < 3 days duration, with 3 watery stools per day and at least 1 in the past 24 hours; mild-to-severe dehydration

Exclusion criteria: bloody diarrhoea, medications, malnutrition, systemic infection

Breastfeeding: included

Interventions

Intervention group: diosmectite, dose 1 g in infants < 12 months and 1.5 g in older children, every 8 hours, plus zinc (dose not specified) for 5 days

Control group: placebo plus zinc

OutcomesDuration of diarrhoea (until first stool of pre-diarrhoeal consistency)
Notes

Location: Pakistan

Setting: urban

Cause of diarrhoea: not specified

Source of funding: not specified

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandomized by lottery
Allocation concealment (selection bias)Unclear riskInsufficient information to permit judgement. Mentions only "lottery method".
Blinding of participants and personnel (performance bias)
All outcomes
Low riskUse of placebo
Blinding of outcome assessment (detection bias)
All outcomes
Low riskUse of placebo
Incomplete outcome data (attrition bias)
All outcomes
Low riskPer-protocol analysis. 10 losses to follow-up (5%), 4 in intervention group, 6 in control group
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasLow riskNo other biases detected.

Vivatvakin 1992

Methods

Randomized controlled trial

Length of follow-up: 5 days

Participants

Number: 62 enrolled children

Inclusion criteria: inpatients; infants/children aged 1 to 24 months with acute secretory diarrhoea < 3 days duration, with 3 watery stools per day

Exclusion criteria: severe dehydration, third-degree malnutrition, other medications, chronic illnesses

Breastfeeding: included

Interventions

Intervention group: diosmectite. Dosage 1.5 g, every 12 hrs for infants < 3 kg; every 8 hrs for infants 4 to 10 kg; every 6 hrs for children 11 to 15 kg, for a maximum of 5 days

Control: no medication

Outcomes

Duration of diarrhoea (from first intervention dose until last liquid stool)

Number of stools

Change in weight

Oral liquid intake

Notes

Location: Thailand

Setting: urban

Cause of diarrhoea: rotavirus in 3% of children in intervention group, 19% in control group. Stool cultures were reported positive for Salmonella and Aeromonas spp. in 7% and 9% of children in the control and study group, respectively, but numbers of each bacterial aetiology per group were not stated.

Source of funding: industry

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High riskStated as randomized, but no method of randomization described. Selection bias is suspected as groups were different in the aetiology of diarrhoea.
Allocation concealment (selection bias)Unclear riskInsufficient information to permit judgement
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot blinded
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo children were lost to follow-up
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasLow riskNo other biases were detected.

Wang 1995

Methods

Randomized controlled trial

Length of follow-up: not stated

Participants

Number: 55 enrolled children

Inclusion criteria: acute diarrhoea < 5 days duration. No age limit or other inclusion criteria stated.

Exclusion criteria: not stated

Breastfeeding: not stated

Interventions

Intervention group: diosmectite. Dosage 3 g per day in infants < 1 year old, 6 g per day in > 1 year

Control: Bismuth complex. Dosage 5 mL, 3 times per day

OutcomesClinical resolution at day 5
Notes

Location: China

Setting: not clear

Cause of diarrhoea: not reported

Source of funding: not stated

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo method of randomization described.
Allocation concealment (selection bias)Unclear riskNo method described.
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot blinded
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo losses to follow-up
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasLow riskNo other biases detected.

Widiasa 2009

Methods

Randomized controlled trial

Length of follow-up: until recovery

Participants

Number: 68 enrolled infants

Inclusion criteria: inpatients; infants aged 6 to 12 months with watery diarrhoea < 2 days duration, with 3 watery stools per day and mild-to-moderate dehydration

Exclusion criteria: severe malnutrition, bloody diarrhoea, severe disease

Breastfeeding: included

Interventions

Intervention group: diosmectite. Dose not specified.

Control group: placebo

OutcomesDuration of diarrhoea (time until normal consistency and frequency)
Notes

Location: Indonesia

Setting: urban

Cause of diarrhoea: not specified

Source of funding: industry

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskBlock randomization
Allocation concealment (selection bias)Low riskCoded, sealed envelopes
Blinding of participants and personnel (performance bias)
All outcomes
Low riskUse of placebo
Blinding of outcome assessment (detection bias)
All outcomes
Low riskUse of placebo
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo losses to follow-up
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasLow riskNo other biases detected.

Zong 1997

  1. a

    WHO: World Health Organization

Methods

Randomized controlled trial

Length of follow-up: not stated

Participants

Number: 45 enrolled children

Inclusion criteria: infants and children aged 2 to 30 months with watery diarrhoea of < 5 days duration

Exclusion criteria: not specified

Breastfeeding: not reported

Interventions

Intervention group: diosmectite. Dose not specified.

Control: lactein tablet. Dose not specified.

A third group received diosmectite and antibiotics.

OutcomesDuration of diarrhoea (not clearly defined)
Notes

Location: China

Setting: unclear

Cause of diarrhoea: rotavirus 100%

Source of funding: not stated

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo method of randomization described.
Allocation concealment (selection bias)Unclear riskNo method described.
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot blinded
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo losses to follow-up
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement. No protocol registered.
Other biasLow riskNo other biases detected.

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
Dupont 1991Wrong outcome: permeability to mannitol and lactulose
Dupont 1992Wrong outcome: permeability to mannitol and lactulose
Karas 1996Wrong population: neonates
Madkour 1994Duplicate

Ancillary