Antimicrobial and micronutrient interventions for the management of infants under 6 months of age identified with severe malnutrition: a literature review

Efficacy

In the management of ‘SAM’ in children over 6 months, a recent MA and comment on two RCTs assessing the efficacy of oral amoxicillin compared to placebo reported a 3% increase in survival with antibiotic treatment with amoxicillin (Million, Lagier & Raoult, 2017), whilst the individual trials noted an increased rate of MUAC gain with amoxicillin (Trehan et al., 2013), improved rate of recovery and decreased risk of transfer to inpatient care (Isanaka et al., 2016). An older RCT assessed the efficacy of two days of intramuscular ceftriaxone in comparison to five days of oral amoxicillin and suggested benefits in terms of growth, recovery rate and case fatality rate in the ceftriaxone treated group, but this was not statistically significant (Dubray et al., 2008).

The only study to include infants U6M (as well as those over 6 months) with ‘SAM’ examined 12 month mortality with or without six months of co-trimoxazole prophylaxis after inpatient admission for ‘SAM’. This found no evidence of mortality or growth benefits (Berkley et al., 2016).

With regard to inpatient IV therapy for ‘complicated SAM’, Williams & Berkley identified one interventional trial from 1996 and argue that current recommendations are based on no supporting evidence for either the type and duration of antibiotics (Williams & Berkley, 2018).

In the pre-term population, one study provides evidence of a role for erythromycin in establishing exclusive breastfeeding (EBF) in pre-term infants (Gokmen et al., 2011), likely secondary to its gastrointestinal prokinetic effects (a potential avenue for a future package of interventions in this group). Indeed, there is increasing interest in the potential roles of macrolides in reducing childhood mortality. A cluster-randomised trial based in Tanzania, Malawi and Niger reported in 2018 an overall reduction of mortality by 13.5% in those communities receiving twice-yearly mass administration of azithromycin compared to those receiving placebo. Of particular note is that the largest reductions in mortality were seen in the infants 1-5 months group (nutritional status not reported) and in the most disadvantaged communities (Keenan et al., 2018).

Safety

No study addressed the safety profile of oral amoxicillin or first-line parental antibiotics in malnourished infants U6M. Co-trimoxazole prophylaxis was associated with an increased risk of neutropenia (grade 4), but age-specific effects were not reported (Berkley et al., 2016).

Williams & Berkley’s systematic review addressed the safety profiles of antibiotics used in the included studies, showing there was not a “significant rate of adverse effects documented in antibiotic intervention group(s)”, but they do raise safety concerns about the currently recommended seven day course of gentamicin in the ‘SAM’ population, given its known renal toxicity, uncertain pharmacokinetics in the ‘SAM’ population and inability to monitor renal function and serum concentrations in many low-income settings (Williams & Berkley, 2018).

Summary

Current evidence, high mortality rates, higher rates of bacteraemia in infants than older children and lower specificity of clinical signs for serious infections than in older age groups make divergence from current guidelines difficult to justify for infants U6M.

Urgent research is required on this topic, especially for those infants who appear clinically stable and for whom risks and costs of routine antibiotic use may outweigh potential benefits. As one SR comments “given that these antibiotics have side-effects, costs, and risks as well as benefits, their routine use needs urgent testing…there is sufficient equipoise for placebo controlled RCTs.” (Alcoba et al., 2013) The potential roles for macrolide antibiotics in vulnerable populations, particularly in the context of global increases in antimicrobial resistance, also require further evaluation. Mass antibiotic distribution using azithromycin may become policy in some low resource settings following recent trial and risks relating to antimicrobial resistance will need careful monitoring.

Efficacy

In children over 6 months, the benefits of mass deworming are debatable, with the included MAs providing evidence that it offers no benefit in terms of mortality, nutritional status or cognition (GRADE Low/Moderate) (Welch et al., 2017; Taylor-Robinson et al., 2015; Clarke et al., 2017; Thayer, Clermont & Walker, 2017), although based on trials carried out in heterogenous populations with differing disease prevalence, age profiles and with nutritional status. Among individual studies, in malnourished pre-school aged children, one cluster-randomised trial in India showed significant improvements in weight gain amongst stunted and wasted infants treated with albendazole (p < 0.0001) (Awasthi et al., 2008), whereas an RCT amongst 222 children under 5 years in the Democratic Republic of Congo showed negative impacts of mebendazole treatment on weight, height and MUAC gain (p-values 0.002, 0.028 and 0.012 respectively) (Donnen et al., 1998a). More recently, in Malawi, a cluster-randomised trial of a package of interventions, that included deworming as a component, at discharge from supplementary feeding programmes for ‘moderate acute malnutrition (MAM)’ showed no significant impact on rates of ‘MAM’ relapse at 12-month follow-up (Stobaugh et al., 2017).

Relevant to malnourished infants U6M, we identified one RCT that looked at infant outcomes with albendazole treatment of breastfeeding mothers, showing a substantial mean difference gain of 0.5 (95% CI [0.2–0.8], p = 0.003) in length-for-age Z score (LAZ) in infants of treated mothers who had stool smears positive for helminth infections (Mofid et al., 2017).

Safety

Deworming treatment was associated with no adverse effects in included studies. WHO recommend deworming from the age of 12 months (World Health Organization, 2019). This is because of the presumed epidemiology of helminth infections and although there is less safety data there are no specific additional safety concerns in younger infants (Montresor, Awasthi & Crompton, 2003).

Summary

There is no evidence to support introduction of routine deworming in infants U6M based on current evidence.

There is some evidence for deworming in breastfeeding mothers of malnourished infants U6M that requires further evaluation.

Efficacy

In the U6M population, neither MA identified any impact of vitamin A on mortality (GRADE - Moderate/High) (Imdad, Ahmed & Bhutta, 2016; Haider, Sharma & Bhutta, 2017). No impact was noted on diarrhoea point prevalence in one of these (GRADE –Moderate) (Imdad, Ahmed & Bhutta, 2016). In the child population over the age of 6 months, one MA reported there to be strong, evidence of vitamin A supplementation being associated with reduction in all cause- and diarrhoea-associated mortality by 12% (GRADE –High) and of diarrhoea incidence by 15% (Grade –Low) (Imdad et al., 2017).

In VLBW infants, a MA shows a similar level of reduction in all-cause mortality but this is not statistically significant (GRADE –Moderate) (Darlow, Graham & Rojas-Reyes, 2016). MA evidence of maternal post-partum vitamin A supplementation reported no benefit in terms of infant morbidity or mortality outcomes (GRADE - Very Low/Low) (Oliveira, Allert & East, 2016).

Among children 0-72 months of age with ‘MAM’, one RCT of single high dose of vitamin A compared to placebo (Donnen et al., 1998b) reported increases in annual weight gain (Mean Difference 0.91 kg, p = 0.029) and in annual MUAC gain (Mean Difference 1.29 cm, p = 0.012) with but there seems to be no benefit of high dose supplementation compared to low dose in other RCTs (Donnen et al., 1998b; Sattar et al., 2012; Donnen et al., 2007).

Safety

Adverse effects were noted in both infant and child populations. Vitamin A supplementation was shown to be associated with a 1.5–3 times increased risk bulging of the anterior fontanelle 48–72 h after first dosing. In all cases this resolved spontaneously and was associated with no neurological sequelae (GRADE –High) (Imdad, Ahmed & Bhutta, 2016; Haider, Sharma & Bhutta, 2017). A two times increased risk of one or more episode of vomiting on commencing supplementation was noted in infants and child between the ages of 6-72 months (GRADE –Moderate) (Imdad et al., 2017), and on subgroup analysis in two RCTs there was some evidence of an elevated risk of diarrhoea in children of 6-60 months age without ‘SAM’ (Fawzi et al., 2000) or with high-dose vitamin A in ‘malnourished pre-school children without oedema’ (Donnen et al., 1998b).

Summary

Low-dose supplementation shows the potential for significant benefit in terms of mortality and diarrhoea incidence in deficient populations and in such settings should be given. However, outside of such situations of specific clinical need, routine use cannot be currently recommended given strong evidence of mild to moderate safety concerns.

More research on which populations/individuals do and which do not need extra vitamin A would be valuable.

Efficacy

The only RCT, conducted in the urban setting in India, to supplement breastfed infants U6M, either directly orally or by supplementing breastfeeding mothers orally, showed no significant difference in weight, length or head circumference between vitamin D and placebo. However, the mean number of days with respiratory or diarrhoeal illness was reduced by 33.5 days by 9 months of life (p < 0.05) in infants supplemented with vitamin D orally (Chandy et al., 2016).

In children over 6 months, one recent RCT carried out in ‘uncomplicated SAM’ in Pakistan was identified. Children received either 2 high doses of vitamin D or placebo with follow-up at 8 weeks. This study showed significant anthropometric effects in those receiving vitamin D, with increases in weight-for height z score (WHZ)/weight-for-length z score (WLZ) (adjusted mean difference: 1.07; 95% CI: 0.49,1.65, P < 0.001), and significant improvements across multiple developmental indices (gross motor, fine motor and language) (Saleem et al., 2018). In MAs of children under 5 years, no impact was noted on growth metrics, incidence of pneumonia (GRADE –Moderate), recovery time from pneumonia (GRADE –Low) and pneumonia-specific (GRADE –Very Low) and all-cause mortality (GRADE –Low) rates (Yakoob et al., 2016; Rooze et al., 2016; Das, Singh & Naik, 2018).

In VLBW/LBW infants, vitamin D supplementation resulted in increases in all growth metrics (height, weight, MUAC) in two RCTs (Kumar et al., 2011; Mathur, Saini & Mishra, 2016) but follow-up of one of these studies at 3–6 years showed no lasting difference between groups (Trilok-Kumar et al., 2015) and one RCT in extremely pre-term infants (23-27 weeks gestation) showed no impact in terms of mortality (Fort et al., 2016).

Two RCTs of maternal vitamin D supplementation showed 13% increased prevalence of reported EBF at 6 months, a mean of 28 fewer days of respiratory and diarrhoeal illness by age 9 months (p < 0.01), but no impact on infant growth (Chandy et al., 2016; Czech-Kowalska et al., 2014).

Safety

No adverse effects noted with vitamin D supplementation in any of the included studies.

Summary

Vitamin D supplementation is safe at doses reviewed within this report with evidence of efficacy in terms of growth in children over 6 months with ‘uncomplicated SAM’ and LBW infants, reduced morbidity in children and through maternal supplementation, potential roles in sustaining EBF. Given the fact that a considerable proportion of infants U6M were born at LBW (Kerac et al., 2019), the current WHO recommendations for LBW of 6 months supplementation can reasonably be followed in nutritional programming for malnourished infants U6M where birth weight is unknown.

Further trials of vitamin D in malnourished infants U6M who are not LBW are warranted. There are also questions about the optimum dose, duration and mode of delivery.

Efficacy

In children, of the two included MAs reporting on mortality, both generated insufficient data on which to estimate mortality (Neuberger et al., 2016; De-Regil et al., 2011). All four MAs in both pre-terms and children showed iron supplementation to be associated with an increase in mean haemoglobin concentration (GRADE High or Moderate where reported) (Mills & Davies, 2012; De-Regil et al., 2011; Thompson, Biggs & Pasricha, 2013; Low et al., 2013). Amongst infants, one RCT in Sweden and Honduras reported no impact on weight gain, but in sub-group analyses reported reduced length gain in Swedish infants 4–9 months of age and in Honduran non-anaemic infants (Dewey et al., 2002). Psychomotor development in infants U6M was found to be significantly improved in the iron supplemented group in one MA (Szajewska, Ruszczynski & Chmielewska, 2010) but a further two RCTs of multiple micronutrients not included in this MA showed opposite effects in mean time to walking unassisted in subgroups analysing iron supplementation (Olney et al., 2006; Katz et al., 2010).

In children, three MAs showed inconsistent effects of iron on growth with one MA (wide age range from infants to >5 years) suggesting small benefits in terms of weight-for-age (WAZ) but reductions in some developed-setting subgroups in HAZ and rate of length gain (Sachdev, Gera & Nestel, 2005), one MA showing no effect on growth in children 2-5 years (GRADE - Very Low) (Thompson, Biggs & Pasricha, 2013), and one showing mild increases in HAZ and no impact on WAZ (5-12 years) (Low et al., 2013). There was evidence of improved cognitive performance in two MAs in those supplemented with iron from ages 2-12 years (GRADE –Very Low, where reported) (Thompson, Biggs & Pasricha, 2013; Low et al., 2013).

In LBW infants, the included MA (Mills & Davies, 2012) identified only one poor-quality RCT, out of 13 included studies reporting on growth, that showed improvements in growth (Mills & Davies, 2012) and a subsequent RCT has shown no benefit in terms of growth (Berglund, Westrup & Domellöf, 2015). This MA did not identify any studies comparing iron supplementation to placebo for neurodevelopmental outcomes and one study was identified that showed no difference in neurodevelopment between high- and low-dose iron but a higher incidence of an abnormal neurological examination at 5 years of age with late-onset supplementation of iron (Mills & Davies, 2012). Child behaviour but not intelligence scores were reported in one RCT as significantly better at seven-year follow-up of LBW infants supplemented with iron from birth compared with unsupplemented infants (Berglund et al., 2018).

Safety

A MA of iron supplementation to children under 18 years in malaria-endemic areas showed no impact on overall malaria incidence and indeed 10% relative risk reduction of severe malaria was noted with iron supplementation (GRADE –High) (Neuberger et al., 2016). Iron was well tolerated in 10 of the 11 studies that documented adverse effects with only one MA noting an 11% increased risk of diarrhoea in iron supplemented groups (p = 0.04) (Gera et al., 2002).

Summary

There is a lack of any strong evidence for benefits of iron supplementation in terms of mortality and morbidity but evidence of increased haemoglobin status and some neurodevelopmental benefits across age groups. Concerns, such as those raised in the WHO guidelines on iron supplementation in children (World Health Organization, 2016), exist about potential negative impacts of iron supplementation on growth, infection risk and malaria risk in malaria-endemic settings where regions where malaria prevention and treatment systems are not in place and/or where children are already iron replete (Lönnerdal, 2017). Routine use for all malnourished infants U6M cannot therefore be recommended, but there can be exceptions for specific individuals and/or populations for treatment of iron deficiency.

Further trials are investigating potential alternatives to simple iron salts and ways to target iron therapy.

Efficacy

Within the ‘SAM’ population, we identified two RCTs in infants U6M and children over 6 months of age showing a reduced total number of total infectious episodes (13 days fewer in the supplemented population over 90-days of follow-up, p < 0.025), reduction in number of diarrhoeal episodes (0.6 less than control group, p = 0.04) at 8 weeks follow-up following admission for diarrhoea, and evidence of slight length gain (4.4 mm greater compared to control group at 8 weeks follow-up, reported as p < 0.05) but not weight gain (Castillo-Duran et al., 1987; Roy et al., 1999). In children over the age of 6 months with ‘SAM’, two further RCTs identified significant increased anthropometric indices associated with zinc supplementation (one in length, two in weight and one in MUAC) (Makonnen, Venter & Joubert, 2003; Umeta et al., 2000). In one of these trials, examining 10mg zinc daily from admission to hospital until 90 days post-discharge, a significant 11% reduction in in-hospital mortality in zinc-supplemented children was also reported (Makonnen, Venter & Joubert, 2003) and in the other, there was reduced number of morbid episodes (cough, diarrhoea, fever, vomiting) in stunted children (Umeta et al., 2000). One further RCT showed no significant anthropometric impacts of high-dose compared to low-dose zinc (Doherty et al., 1998).

In a MA of the infant U6M population as a whole, no impact on all-cause mortality or diarrhoea duration or presence at day 7 was demonstrated (GRADE –Very Low and Low respectively) (Lazzerini & Wanzira, 2016). In combination with the child population a 13% reduction in pneumonia incidence (GRADE –Low) (Lassi, Moin & Bhutta, 2016) and slightly improved WAZ and WLZ (pooled effects +0.06 and +0.05 respectively) but not MUAC were found (Nissensohn et al., 2016). A single RCT showed that zinc when used as an adjunct to treatment for neonatal sepsis resulted in significantly reduced mortality and improved mental development at 12 months (Banupriya et al., 2018).

Among children aged over 6 months, one MA showed an 18% reduction in all-cause mortality in those over 12 months (Brown et al., 2009), but two other MAs demonstrated no effect on mortality (GRADE –High and Very Low) (Lazzerini & Wanzira, 2016; Mayo-Wilson et al., 2014). Furthermore, three MAs demonstrated a reduction in incidence and duration of all-cause diarrhoea (GRADE –Moderate) (Lazzerini & Wanzira, 2016; Brown et al., 2009; Mayo-Wilson et al., 2014), and one MA identified a significant reduction in mortality when used as an adjunct to treatment for severe pneumonia (Wang & Song, 2017). A further four MAs noted improved, albeit across different anthropometric indices, with zinc supplementation (Brown et al., 2009; Ramakrishnan, Nguyen & Martorell, 2008; Imdad et al., 2011; Liu et al., 2018), with more marked improvements when supplemented after two years of age (Liu et al., 2018). No impacts on mental or psychomotor development were demonstrated by one MA (GRADE –Moderate) (Gogia & Sachdev, 2012).

In the LBW population, concerning indices of growth, one SR of three trials identified no impact of zinc supplementation on length or weight (Gulani, Bhatnagar & Sachdev, 2011), but four of five RCTs identified by our search demonstrate significant effects in terms of length and weight gain (El Sayed & Elghorab, 2016; El-Farghali et al., 2015; Terrin et al., 2013; Friel et al., 1993; Díaz-Gómez et al., 2003). The same SR included one trial that showed no benefit in terms of mortality, whereas our search identified two RCTs suggesting a reduction in mortality with zinc supplementation by 58–68% (Terrin et al., 2013; Sazawal et al., 2001). Concerning morbidity outcomes, one SR identified two trials showing no overall impacts on number of diarrhoeal illnesses but fewer days of diarrhoea following cessation of breastfeeding and one trial showing no impact on acute lower respiratory tract infection incidence (Gulani, Bhatnagar & Sachdev, 2011). Furthermore, one RCT showed a reduction in combined neonatal morbidities (late-onset sepsis, necrotising enterocolitis, bronchopulmonary dysplasia etc.) with zinc supplementation (Terrin et al., 2013). Finally, one RCT reporting on neurodevelopmental outcomes showed increased alertness and attention at term corrected gestational age and reduced hyper-excitability at three months follow-up in zinc supplemented infants (Mathur & Agarwal, 2015).

Safety

Adverse effects and safety concerns associated with zinc supplementation were poorly reported with only 10 of 24 included articles explicitly mentioning this. Of articles that documented adverse effects or none, one RCT in Bangladesh, of 141 children 6 months–3 years old with ‘SAM’ identified a 4.5 times increased mortality with high-dose (6.0mg/kg for 15+ days) compared to low-dose (1.5 mg/kg for 15 days) zinc supplementation (p = 0.03) (Doherty et al., 1998) and two MAs in infants and children identified a significant 29–57% increased risk of one or more episodes of vomiting upon commencement of zinc supplements (GRADE –High and Moderate) (Lazzerini & Wanzira, 2016; Mayo-Wilson et al., 2014).

Summary

Consistent evidence across age groups exists of zinc supplementation being associated with reduced morbidity and improved anthropometry, whilst mortality and neurodevelopmental impacts are more unclear. Zinc should not be supplemented as a high-dose, given mortality concerns, and its tolerability should be considered in the context of vomiting risk. Zinc should be supplemented as per diarrhoea guidelines for all severely malnourished infants U6M with diarrhoeal illnesses (World Health Organization, 2005). In cases where severely malnourished infants U6M are not affected by diarrhoea, in the absence of further evidence, we suggest that 2mg/day (1 recommended daily allowance (Institute of Medicine , US)) of zinc be supplemented with other micronutrients in regions where zinc deficiency has been documented. This is because breastmilk zinc concentrations have been shown to be insufficient in zinc deficient mothers (Dumrongwongsiri et al., 2015; Yalcin, Yalcin & Gucus, 2015).

Research is urgently needed to establish zinc requirements for malnourished infants U6M and the fact that there is some evidence of mortality benefit when supplemented over an extended period suggests that randomised trials are warranted in this age group.

Efficacy

Folic acid supplementation was shown in two RCTs in infants and children to have no impact on mortality (Sazawal et al., 2006; Tielsch et al., 2006). In children over 6 months of age, two RCTs showed reduced incidence of acute diarrhoeal disease and lower respiratory tract infection (Tielsch et al., 2006; Taneja et al., 2013) and one RCT showed increased total weight gain and WAZ scores with folic acid supplementation compared to placebo (Medeiros et al., 2015).

Amongst LBW infants, of three RCTs included all showed no impact of folic acid on weight gain, one noticed length gains in a subgroup of infants of birthweight >1750g, and one showed no improvement in infectious disease incidence or haemoglobin status (Foged et al., 1989; Kendall et al., 1974; Ek et al., 1984).

Safety

A safety debate around folic acid supplementation has centred on its usage in malaria-endemic settings as malaria parasites can utilise exogenous folate during co-administration with sulfadoxine/pyrimethamine (SP) (a folate antagonist used in the prevention and treatment of malaria). In a paired set of RCTs studying folic acid supplementation amongst children 1-36 months, the Tanzanian trial was stopped early because of increased adverse events and hospital admissions in the supplemented arm (Sazawal et al., 2006) but this was not replicated in the Nepalese study (Tielsch et al., 2006). Subsequent post-hoc analysis of the Tanzanian study showed adverse events were not associated with SP co-administration. Three further RCTs looking at SP and folic acid co-administration showed some evidence of parasitological but not clinical treatment failure with SP and folic acid co-administration (Carter et al., 2005; Van Hensbroek et al., 1995; Mulenga et al., 2006).

Folic acid was well tolerated in other studies with the exception of one RCT amongst Indian children that showed a two times increased incidence of persistent diarrhoea in the supplemented group (Taneja et al., 2013), something that was not replicated in the two other RCTs reporting on diarrhoea incidence.

Summary

There is limited evidence of benefit in the child population in terms of morbidity and growth and therefore this is not recommended as a routine intervention for malnourished infants U6M. Any benefits that may exist are not consistent across age groups. Safety in a malaria endemic setting remains uncertain but is less concerning given SP is no longer routinely used for the treatment of malaria. Safety of folic acid supplementation should be considered in areas where SP is used for Seasonal Malarial Chemoprophylaxis (SMC) or Intermittent Preventive Treatment in pregnancy (IPTp).

Further studies investigating the role of folic acid in malnourished infants U6M are warranted but the evidence presented does not identify this as a priority area for research.

Efficacy

Maternal post-natal multiple micronutrient supplementation was associated with no quantitative evidence of improvement in infant and child mortality or morbidity outcomes (Abe et al., 2016). One of the three MAs looked at PUFA supplementation compared to placebo to breastfeeding mothers postnatally and demonstrated no benefit in terms of infant length, weight and head circumference (GRADE Moderate) (Delgado-Noguera et al., 2015). These results of no significant growth effects were supported in the two other MAs but the population included supplementation during both gestation and lactation and no postnatal subgroup analysis was presented (Quin et al., 2016; Li et al., 2018). A statistically significant improvement was noted in indices of child attention beyond 24 months of age in the first of these MAs (GRADE Low) (Delgado-Noguera et al., 2015). An RCT of maternal calorie supplementation in conjunction with a breastfeeding support intervention showed no improvement in WAZ or LAZ but increased infant breast milk intake and a two times increase in EBF (likely secondary to the support intervention) was demonstrated (Huynh et al., 2017a).

Safety

No clinically detectable adverse effects were reported in any of the included studies.

Summary

We found insufficient evidence of the benefits of maternal supplementation to infants to justify routine use in current programming for malnourished infants U6M. However, potential benefits to the mother, not included in this review, should be considered and evaluated in more detail in further research to inform decisions in this area.

Efficacy

Both studies reported on time to rehydration, with Alam et al. (2003) showing no difference between ReSoMal and ORS and Kumar et al. (2015) reporting a shorter time to rehydration with ReSoMal (16.1 h vs 19.6 h p = 0.036). Stool frequency and number of patients requiring IV fluids after attempting oral rehydration was shown to be similar between treatment groups in both studies. Only Alam et al. (2003) reported on mortality, reporting no deaths. Neither study reported on anthropometric or neurodevelopmental outcomes.

Safety

The primary outcome of the included SR was the occurrence of hyponatraemia. Alam et al. (2003) ReSoMal was associated with a higher incidence of severe hyponatraemia (Na<120mmol compared to an older WHO ORS formulation (5% versus 2%,) a lower mean serum sodium at 24 and 48 h (p < 0.01 and <0.001 respectively) and with one episode of hyponatraemic seizures (n = 130). Similarly, Kumar et al. (2015), comparing ReSoMal with standard ORS, showed ReSoMal to be associated with increased incidence of hyponatraemia (15.4% vs 1.9%, n = 110). Alam et al. (2003) reported there to be no difference in frequency of fluid overload between groups.

Summary

There are no studies of ReSoMal in infants U6M. Limited evidence from inpatient studies of older children with ‘SAM’ suggests that ReSoMal is of similar efficacy in terms of rehydration to standard ORS but that there are significant safety concerns in terms of risk of hyponatraemia. On the basis of current evidence, and the fact that infants in the first few months of life are at increased risk of water and salt retention due to immature hormonal and renal excretion mechanisms, there is no reason to change current recommendations for use of ReSoMal in malnourished infants U6M. This age group may differ from older children in both risks and responses to treatment and is thus a priority area for clinical trials.