Table 1 Characteristics of included studies
Reference [Ref #] | Country | Population | Intervention | Comparator | Number with follow up data | Pre-specified follow up time | Relevant outcomes measured and poweri | Modified GRADE score |
|---|---|---|---|---|---|---|---|---|
Antimicrobial Intervention | ||||||||
Gilman 1980 [8] | Bangladesh | Inpatient adults and children with blood, pus cells, and mucus in stool, 4 or more stools/day, and culture-confirmed Shigella infection | Low-dose ampicillin (50 mg/kg/day) | High-dose ampicillin (150 mg/kg/day) | 56 children | 21 days | Mortality (power/sample size calculations not reported) | Very lowΩ,ΦΨ |
Amadi 2002 [42] | Zambia | Inpatient children 12–85 mo with diarrhea (at least 3 stools per day that take the shape of a container or can be poured) and Cryptosporidium oocysts | 5 mL of 20 g/L nitazoxanide oral suspension, twice daily for 3 days | Placebo | 96 | 8 days | Mortality (powered for outcome of clinical response) | ModerateΩ |
Dietary Supplements | ||||||||
Alam 2000 [26] | Bangladesh | Inpatient males 4–18 mo with acute non-dysenteric diarrhea | WHO ORS with dietary fiber(20 g/L Benefiber®) | Standard WHO-ORS | 150 | 7 days | Weight gain at day 7 from enrollment (powered for the outcome of stool output) | ModerateΩ |
Rabbani 2001 [15] | Bangladesh | Severely malnourished (< 60% NCHS standard), inpatient male infants 5–12 mo with persistent diarrhea (> 3 loose stools/day for 14 days), treated with ciprofloxacin | Rice-based diet with dietary fiber (250 m/L cooked, green banana, 7 days; or rice-based diet with 1 g/kg to 8 g/kg of pectin supplement) 7 days | Calorically equivalent control diet: rice-based diet only, 7 days | 62 | At least 7 days, or until end of diarrhea | Proportions recovered from diarrhea [formed stool] at days 7, 8, 9, and 10 (powered for outcome of diarrhea recovery duration) | ModerateΩ |
Yalcin 2004 [51] | Turkey | Inpatient children 6–24 mo with acute diarrhea | Glutamine supplement - 0.3 g/kg/day, for 7 days | Placebo | 143 | 3 months | Weight gain at day 30, 60, and 90 from enrollment (powered for the outcome of diarrhea duration) | LowΩ, ϑ |
High Protein Diets | ||||||||
Datta 1990 [9] | India | Inpatient children under 5 y/o with acute dysentery (visible blood and mucus in stools), treated with nalidixic acid | Extra servings of milk (30% of total daily caloric requirements), in addition to standard hospital diet | Standard hospital diet | 96 | 15 days | Weight and MUAC at day 15 (power/sample size calculations not reported) | Very lowΩ, Φ, ϑΨ |
Kabir 1992 [10] | Bangladesh | Inpatient children 2–4 y/o with culture-confirmed Shigella dysentery, treated with nalidixic acid or other microbial | High protein diet (15% of total energy from protein), 21 days | Standard diet (7.5% of total energy from protein), 21 days | 22 | 21 days | Change in weight, height, WAZ, WHZ, HAZ, MUAC, and triceps skinfold thickness at day 21 from admission (power/sample size calculations not reported) | Very lowΩ, Φ, ϑΨ |
Kabir 1993 [11] | Bangladesh | Outpatient children 2–5 y/o with acute diarrhea and culture-confirmed Shigella spp., treated with nalidixic acid or pivmecillinam | High protein diet (15% of total energy from protein), 21 days | Standard diet (7.5% of total energy from protein), 21 days | 69 | 21 days | Change in WHZ, WAZ, and HAZ at day 21 from admission (power/sample size calculations not reported) | Very lowΩ, Φ, ϑΨ |
Mazumder 1997 [12] | Bangladesh | Malnourished (< 80% NCHS median), inpatient children 12–48 mo with bloody or bloody mucoid diarrhea and culture-confirmed Shigella, treated with nalidixic acid | High calorie and high protein diet, 4960 kJ/l for 10 days | Standard hospital diet, 2480 kJ/l for 10 days | 75 | 40 days | Percent change in WAZ and WHZ at 10, and 40 from admission (power/sample size calculations not reported) | LowΩ, Φ, Ψ |
Nurko 1997 [36] | Mexico | Inpatient children 3–36 mo with persistent diarrhea (3 or more loose stools/day for 14 days or longer) and third-degree malnutrition (< 60% NCHS median), treated with TMP-SMX or metronidazole | High protein diet (2 groups: chicken-based diet, or soy-based diet)ii | Standard cultural dietary treatment for diarrhea (elemental diet or “Vivonex”) | 41 | 16 days minimum | Weight at end of intervention and at discharge; proportion with nutritional recoveryiii at end of intervention (powered for outcome of diarrhea duration) | Very low Ω, ϑ, Ψ |
Kabir 1998 [13] | Bangladesh | Inpatient children 2–60 mo with acute bloody mucoid diarrhea, treated with nalidixic acid or pivmecillinam | High protein diet (15% of total energy from protein), 21 days | Standard protein diet (7.5% of total energy from protein), 21 days | 59 | 6 months post- intervention | Change in weight, height, WHZ, WAZ, and HAZ at 6 months compared to post-intervention measurements (power/sample size calculations not reported) | Very low Ω, Φ, ϑ, Ψ |
Mazumder 2000 [14] | Bangladesh | Malnourished (< 80% of NCHS median) inpatient children 12–48 mo with acute bloody or mucoid diarrhea, treated with nalidixic acid of pivmecillinim | High calorie & protein milk-cereal formula (4960 kJ/l), 10 days | Control milk-cereal formula (2480 kJ/l), 10 days | 75 | 10 days | Percent change in WAZ at day 10, compared to admission WAZ (power/sample size calculations not reported) | Very lowΩ, ϑ, Ψ |
Valentiner-Branth 2001 [43] | Guinnea-Bissau | Community-based children under 3 y/o with persistent diarrhea per mother’s report | Counseling on the importance of breastfeeding and of a nutritious diet, and a high protein millet gruel with a multivitamin tablet (including zinc), until the end of a 7 day period without diarrhea | Counseling on the importance of breastfeeding and of a nutritious diet | 101 | 9 months | Difference in knee-heel length, height and weight at end of intervention and day 90, compared to admission measurements (powered for outcome of diarrhea duration) | Very lowΩ, Φ, Ψ |
Rollins 2007 [44] | South Africa | Inpatient HIV+ children 6–36 mo with persistent diarrhea (4 or more loose or watery stools/day, for 5 days or more) | Enhanced nutritional support: standard nutritional support + extra protein to provide 150 kcal/kg/day and 4.0–5.5 g protein/kg/day (as milk or powdered protein, depending on age), until diarrhea resolved | Standard nutritional support: maize porridge + milk formula, until diarrhea resolved | 104 | 26 weeks | Mortality, median change in weight-SDs ivand WAZ at 26 weeks; proportions underweight (WAZ < − 2 SDs) and stunted (LAZ < − 2 SDs) at 26 weeks (powered for outcome of weight change) | Very lowΩ, Φ, ϑ, Ψ |
Lactose Free Diet | ||||||||
Bhan 1988 [16] | India | Outpatient children 3–24 mo | Legume and cereal-based formula (lactose-free), until recovery or a minimum of 7 days | Calorically equivalent milk-based formula, until recovery or a minimum or 7 days | 57 | At least 7 days, or until end of diarrhea | Weight gain at day 7 and at recovery compared to admission weight (power/sample size calculations not reported) | Very lowΩ, Φ, ϑ, Ψ |
Bhutta 1991 [50] | Pakistan | Outpatient males 6 mo - 3 yo with persistant diarrhea (increased frequeny and reduced consistency lasting 2 weeks or more) | Soy milk (lactose-free) for 7 days, followed by khitchri and yogurt for 7 days | Khitchri and yogurt for 14 days | 73 | 14 days | Weight gain at day 7 and 14 compared to admission weight (power/sample size calculations not reported) | Very lowΩ, Φ, ϑ, Ψ |
Lozano 1994 [38] | Colombia | Inpatient children 1–24 mo with diarrhea (4 or more watery stools in a 24 h period) and dehydration | Lactose-free feeding formula, 21 days | Feeding formula with lactose, 21 days | 52 | 6 weeks post-discharge | Weight increment at 6 weeks (powered for outcomes of diarrhea duration) | Very lowΩ, Φ, ϑ, Ψ |
Bhatnagar 1996 [17] | India | Inpatient \children 3–24 mo, with persistent diarrhea (3 or more liquid stools/day for 14 days) | Puffed rice cereal, sugar, oil, and milk protein, 120 h | Puffed rice cereal, sugar, oil, and egg white protein (lactose-free), 120 h | 116 | 4–6 weeks after discharge | Proportion of patients whose weight on day 7 was lower than at rehydration; probability of continuing diarrhea at each day to day 12 (power/sample size calculations not reported) | Very lowΩ, Φ, Ψ |
de Mattos 2009 [37] | Brazil | Inpatient male infants 1–30 mo with persistent diarrhea (3 or more liquid stools per day for 14 days) | Amino-acid based diet or soy-based diet, or hydrolyzed protein-based dietv | Yogurt-based diet | 154 | 7 days post-discharge | Difference in weight gain and change in WHZ at discharge compared to admission measurements (powered for outcomes of stool output and diarrhea duration) | Very lowΩ, Φ, ϑ, Ψ |
Other Dietary Interventions | ||||||||
Eichenberger 1984 [35] | Brazil | Inpatient infants 1–11 mo with acute to subacute gastroenteritis with diarrhea | Semi-elemental diet with low osmolarity and high content of hydrolyzed lactalbumin | Standard hospital diet | 38 | 21 days | Weight at day 21 compared to weight at beginning of therapy (power/sample size calculations not reported) | Very low Ω, Φ, ϑ, Ψ |
van der Kam 2016 [45] | Uganda | Non-malnourished, outpatient children 6–59 mo with diarrhea (3 or more loose stools [bloody or nonbloody] per 24 h by mothers’ report), malaria, or lower respiratory tract infections | Ready-to-use Therapeutic Foods (RUTF), plus instructions to feed the child an extra meal/day for 14 d; or micronutrient powder plus instructions to feed the child an extra meal/day for 14 d | An instruction to feed the child an extra meal/day for 14 d | 941 with diarrhea only | 6 months | Incidence of WHZ < − 2, MUAC < 115 mm, or nutritional oedema during follow up (powered for combined outcome of negative nutritional outcome)vi | Low ϑ, Ψ |
van der Kam 2016 [46] | Nigeria | Non-malnourished or moderately malnourished outpatient children 6–59 mo with diarrhea (3 or more loose stools [bloody or nonbloody] per 24 h by mothers’ report), malaria, or lower respiratory tract infections | Ready-to-use Therapeutic Foods (RUTF), plus instructions to feed the child an extra meal/day for 14 d; or micronutrient powder plus instructions to feed the child an extra meal/day for 14 d | An instruction to feed the child an extra meal/day for 14 d | 1171 with diarrhea only | 6 months | For non-malnourished children at enrollment: Incidence of WHZ < − 2, MUAC < 115 mm, or nutritional oedema during follow up. For malnourished children at enrollment: Incidence of WHZ < − 3, MUAC < 115 mm, nutritional oedema. Or > 10% weight loss during follow up. Powered for combined outcome of negative nutritional outcomevii | Low ϑ, Ψ |
Oral Rehydration Solution Formulations | ||||||||
Santosham 1983 [39] | Panama | Inpatient 3 mo - 2 y/o who were well nourished, with acute diarrhea (more than 3 watery stools per day) | High potassium and chloride ORS, or standard WHO-ORS | Standard diet for diarrhea management (aerated beverages, bananas, cereals, and apple sauce) | 93 | 14 days | Weight at day 14, weight gain at day 14 as percent of enrollment weight (power/sample size calculations not reported) | Very lowΩ, Φ, Ψ |
Ribeiro 1991 [40] | Brazil | Inpatient male infants less than 12 mo, with acute diarrhea and dehydration | Standard WHO-ORS with 30 mmol/L alanine | Standard WHO-ORS | 18 | 7 days | Weight gain at day 7 (power/sample size calculations not reported) | ModerateΩ |
Faruque 1997 [21] | Bangladesh | Inpatient children 3–35 mo with acute non-dysenteric diarrhea | Glucose based ORS | Rice-powder based ORS | 471 | 16 days | Proportion with diarrhea at day 14, weight gain at day 16 (powered for outcomes of stool output, diarrhea duration and weight gain [70 g]) | Low Φ, Ψ |
Alam 2009 [55] | Bangladesh | Severely malnourished (< 70% NCHS standard), inpatient infants 6–60 mo with acute diarrhea and culture-confirmed V. cholerae | Glucose-based ORS, or Glucose-based ORS plus amylase resistant starch | Rice-based ORS | 137 | 6 weeks | Time to attain 80% of median WLZ from enrollment; proportion with diarrhea at or after day 7 (power/sample size calculations not reported) | LowΩ, Ψ |
Probiotics | ||||||||
Boudraa 2001 [47] | Algeria | Inpatient children 3–24 mo with acute watery diarrhea (> 3 loose stools in the previous 24 h) | Standard formula fermented with L. bulgaricus and S. thermophilus (lactose and calorically equivalent) | Standard milk-based formula | 97 | 7 days | Weight gain at day 7 (power/sample size calculations not reported) | Very lowΩ, Φ, ϑ, Ψ |
Villaruel 2007 [41] | Argentina | Outpatient children 3 mo - 2 yo, with acute diarrhea (3 or more liquid or loose stools in the preceding 24 h) | WHO-ORS and S. boulardii, 250 mg per day (patients < 1 yo) or 500 mg per day (patients 1 yo and older) | WHO-ORS and placebo | 72 | 1 month | Proportion of patients with diarrhea at or after day 7 (power/sample size calculations not reported) | LowΩ, ϑ |
Misra 2009 [19] | India | Inpatient infants < 36 mo with diarrhea (more than 3 stools per day that take the shape of their container) | Lactobacillus rhamnosus GG (10^9 live bacteria) | Placebo | 207 | 6 weeks | Change in WHZ at 6 weeks(powered for outcomes of stool output and diarrhea duration) | High |
Sindhu 2014 [20] | India | Children 6 m to 5 years with diarrhea testing positive for either rotavirus or Cryptosporidium infection | Lactobacillus rhamnuosus GG (10^10 organisms) | Placebo | 123 | 4 weeks | Proportions stunted (HAZ < − 2 SD), underweight (WAZ < − 2 SD), and wasted (WHZ < − 2 SD) at 4 weeks, proportion with diarrhea or severe diarrhea during follow-up (powered for outcome of L:M ratio) | ModerateΩ |
Dinleyici 2014 [52] | Turkey | Inpatient children 3–60 mo with acute watery diarrhea | WHO-ORS + lactobacillus reuteri 17,938 (10^8 CFU) for 5 days | WHO-ORS only | 127 | 12 days | Proportion with diarrhea at day 12 (powered for outcome of diarrhea duration) | Very lowΩ, Φ,Ψ |
Therapeutic Micronutrients (Vitamin A and Zinc)viii | ||||||||
Faruque 1999 [34] | Bangladesh | Inpatient children 6 mo - 2 yo with acute diarrhea (3 or more liquid stools in the previous 24 h) | 4500 μg vitamin A, 15 day, 14.2 mg Zinc acetate, 15 days, or bothix | Placebo | 656 | 17 days | Proportion with diarrhea at day 7 and 16 (powered for outcome of diarrhea duration) | High |
Khatun 2001 [22] | Bangladesh | Inpatient children 6 mo - 4 yo with persistent diarrhea (diarrhea for > 14 days duration) | Multivitamin (D, C, B1 B2 B6) syrup and 20 mg elemental zinc (as zinc acetate, 5 ml twice daily for 7 days), multivitamin syrup with Vit A (100,000 IU for children < 1 yo, 200,000 for children > 1 yo), or both | Multivitamin (D, C, B1 B2 B6) syrup only | 93 | 7 days | Weight at day 7, weight gain at day 7 compared to day 1, proportions with diarrhea at day 7 (powered for outcome of clinical recovery) | ModerateΩ |
Therapeutic Micronutrients (Zinc Alone) | ||||||||
Sazawal 1995 [23] | India | Inpatient children 6–35 mo, with acute diarrhea (at least 4 unformed stools in the preceding 24 h) | Multivitamin syrup (A, B2, B6, D3, E) plus zinc gluconate (20 mg of elemental zinc) | Multivitamin syrup (A, B2, B6, D3, E) only | 937 | At least 120 days | Proportion of diarrhea episodes that last longer than 7 days, proportion of diarrhea episodes taken to a physician during follow up (power/sample size calculations not reported) | High |
Roy 1998 [24] | Bangladesh | Inpatient 3–24 mo with persistent diarrhea | Multivitamin syrup (Vit A, B1, B2, B3, B6, D, Ca) with 20 mg elemental zinc per day for 14 days | Multivitamin syrup (Vit A, B1, B2, B3, B6, D, Ca) only for 14 days | 141 | 15 days | Mortality, weight gain at discharge compared to admission weight, proportion with diarrhea after day 15 (powered for outcome of diarrhea duration) | Very lowΩ, Φ, ϑ |
Bhutta 1999 [49] | Pakistan | Inpatient children 6–36 mo with persistent diarrhea | Multivitamin syrup (Vit A, B1, B2, B3, B6, B12, C, D, Ca) with 3 mg elemental zinc per kg per day for 28 days | Multivitamin syrup (Vit A, B1, B2, B3, B6, B12, C, D, Ca) only for 28 days | 77 | 28 days | Weight gain at day 7 and 14; overall weightincrement at day 14; MUAC at day 7 and 14; overall MUAC increment (powered for outcome of day 14 weight gain) | LowΩ, ϑ |
Roy 1999 [25] | Bangladesh | Malnourished (< 76% of NCHS median), inpatient children 3–24 months with acute diarrhea | Multivitamin syrup (vit A, B1, B2, B6, D, and Ca) with 20 mg elemental zinc per day for 14 days | Multivitamin syrup (vit A, B1, B2, B6, D, and Ca) only for 14 days | 29 | 10 weeks | Weight gain at each week of for 8 weeks, length gain at each week for 8 weeks (power/sample size calculations not reported) | LowΩ, ϑ |
Baqui 2002 [27] | Bangladesh | Community-based children 3–59 mo with diarrhea of any duration | ORS with 20 mg zinc per day, 14 days | ORS only | 8070 | 2 yearsx | Incidence of diarrhea, mortality (powered for the outcomes of diarrhea duration, diarrhea incidence, acute lower respiratory infections incidence, admission to hospital for diarrhoea or acute lower respiratory infections, and child mortality) | Low Φ, Ψ |
Walker 2007 [53] | Ethiopia, Pakistan, and India | Infants 1–5 mo with acute diarrhea, identified through home visits by health workers and community based study clinics | ORS with 10 mg zinc sulfate, daily for 14 days | ORS with placebo | 1042 | 8 weeks | Weight at week 4 and 8, length at week 4 and 8, proportion of infants with ≥1 episode of any diarrhea, ≥ 2 episode of any diarrhea, or ≥1 episode of dysentery (any day with blood in the stool); incidence and prevalence of diarrhea; mortality (powered for anthropometry and morbidity outcomes) | High |
Roy 2007 [28] | Bangladesh | Convalescent children 3–24 mo, after recovery from persistent diarrhea | Multivitamin syrup (Unspecified) with 20 mg elemental zinc, 14 days | Multivitamin syrup (unspecified) only, 14 days | 147 | 12 weeks | Mortality, gain in length and weight at 12 weeks, incidence of subsequent diarrhea episodes (power/sample size calculations not reported) | LowΩ, Φ, ϑ |
Roy 2008 [29] | Bangladesh | Moderately malnourished (weight/age 61–75% of NCHS median), inpatient children age 12–59 m with acute bloody-mucoid diarrhea or febrile diarrhea, and lab-confirmed Shigella spp | Multivitamin syrup (A, D, B complex, Ca) with zinc acete (10 mg elemental Zn/5 mL), for 14 days | Multivitamin syrup (Vit A, D, B complex, Ca) only | 30xi | 6 months | Diarrhea incidence and duration of episodes during 6 mo follow up (power/sample size calculations not reported) | LowΩ, ϑ |
Fajolu 2008 [48] | Nigeria | Outpatient children 6–24 mo with acute diarrhea (3 or more loose, liquid or watery stools in a 24 h period) | 20 mg of elemental zinc (zinc sulphate monohydrate) for patients > 1 y/o, 10 mg of elemental zinc, 14 days, for patients < 1 y/o | Placebo | 60 | 3 months | Weight gain at 3 months, number and duration of subsequent diarrhea episodes during follow up (power/sample size calculations not reported) | ModerateΩ |
Larson 2010 [30] | Bangladesh | Community-based children 6–23 mo with acute diarrhea and culture-confirmed ETEC | 10 days of zinc (10 mg/d) + additional 3 months of zinc supplementation (10 mg/d) | 10 days of zinc (10 mg/d) only | 333 | 9 months | Incidence rate of diarrhea illness during follow up (powered for the incidence of acute upper respiratory tract infections) | Moderateϑ |
Alam 2011 [31] | Bangladesh | Community-based children 4–59 mo with diarrhea (3 or more loose or liquid stools in the previous 24 h) | Short course zinc - 20 mg elemental zinc, 5 days | Standard course zinc - 20 mg elemental zinc, 10 days | 1622 | 90 days | Number of diarrheal episodes and days of diarrhea during follow up; proportion with at least 1 subsequent episode of diarrhea, prolonged diarrhea, or persistent diarrhea during follow up; day of onset of first subsequent diarrhea episode during follow up (powered for the outcome of diarrhea incidence) | High |
Patel 2013 [32] | India | Outpatient children 6–59 mo with acute diarrhea (> 3 unformed stools in the previous 24 h per mother’s report) | Zinc (2 mg/kg/day) or zinc + copper (Zn 2 mg/kg/day + Cu 0.2 mg/kg/day), 14 days | Placebo | 724 | 12 weeks | Proportion with at least 1 diarrhea episode, 2 diarrhea episodes, or 1 dysentery episode during follow up; number and duration of subsequent diarrhea episodes; change in WAZ, WHZ, and HAZ from enrollment measurements every 2 weeks for 12 weeks (power/sample size calculations not reported) | High |
Negi 2015 [33] | India | Children 5–12 yrs. presenting to pediatric emergency units with acute watery diarrhea (3 or more episodes of loose stools over 24 h of < 72 h duration), with some or severe dehydration, and having had no treatment | Zinc (20 mg/day) for 14 days | Placebo | 134 | 3 months | Risk of having at least 1 episode of diarrhea during follow up (power/sample size calculations not reported) | LowΩ, ϑ |