Abstract
Acute diarrhoeal diseases remain a leading cause of global morbidity and mortality particularly among young children in resource-limited countries. Recent large studies utilizing case–control design, prospective sampling and more sensitive and broad diagnostic techniques have shed light on particular pathogens of importance and highlighted the previously under recognized impact of these infections on post-acute illness mortality and growth. Vaccination, particularly against rotavirus, has emerged as a key effective means of preventing significant morbidity and mortality from childhood diarrhoeal disease. Other candidate vaccines against leading diarrhoeal pathogens, such as enterotoxigenic Escherichia coli and Shigella spp., also hold significant promise in further ameliorating the burden of enteric infections in children. Large studies are also currently underway evaluating novel and potential easy-to-implement water, sanitation and hygiene (WASH) preventive strategies. Given the ongoing global burden of this illness, the paucity of new advances in case management over the last several decades remains a challenge. The increasing recognition of post-acute illness mortality and growth impairment has highlighted the need for interventions that go beyond management of dehydration and electrolyte disturbances. The few trials of novel promising interventions such as probiotics have mainly been conducted in high-income settings. Trials of antimicrobials have also been primarily conducted in high-income settings or in travellers from high-income settings. Bloody diarrhoea has been shown to be a poor marker of potentially treatable bacterial enteritis, and rising antimicrobial resistance has also made empiric antimicrobial therapy more challenging in many settings. Novel effective and sustainable interventions and diagnostic strategies are clearly needed to help improve case management. Diarrhoeal disease and other enteric infections remain an unmet challenge in global child health. Most promising recent developments have been focused around preventive measures, in particular vaccination. Further advances in prevention and case management including the possible use of targeted antimicrobial treatment are also required to fully address this critical burden on child health and human potential.
Keywords
Epidemiology
The burden of diarrhoeal diseases
Diarrhoeal diseases are one of the leading causes of morbidity and mortality globally and account for more deaths in early childhood after the neonatal period than any other aetiology save pneumonia.1–3 Diarrhoeal diseases are associated with an estimated 1.3 million deaths annually, 4 with most occurring in resource-limited countries;4–6 note that up to 25% of deaths in young children living in Africa and south-east Asia are attributable to acute gastroenteritis. 7 The youngest children are most vulnerable with the incidence of severe gastroenteritis being highest in the first 2 years of life.8,9 Morbidity due to diarrhoea is further concentrated in marginalized communities within resource-limited countries. 10 Despite improvements in standard of living, advances in sanitation, water treatment and food safety awareness, diarrhoeal disease still accounts for significant economic and societal losses. 11 Although over the last several decades there has been a decline in the total global mortalities due to diarrhoea, the morbidity from diarrhoea may not have shown a similar decline. 12 These improvements in mortality have been attributed to increased use of oral rehydration therapy (ORT), improved nutrition, increased breastfeeding, better supplemental feeding, female education, measles immunization and improvements in hygiene and sanitation. 13 There are, however, concerns about potential rebound increases in diarrhoea-associated mortality in the future due to increasing urbanization and climate change.14,15
Despite how commonly gastroenteritis occurs, the pathogenesis and actual timing of diarrhoea-associated mortality is not well understood. The landmark recent Global Enteric Multicenter Study (GEMS) enrolled 9980 children with moderate-to-severe or severe gastroenteritis in seven sites across sub-Saharan Africa and South Asia and found that their odds of mortality was 8.5 times higher [95% confidence interval (CI): 3.9–11.5] than that of the 12,390 enrolled controls. 5 However, only 34% of deaths occurred within 7 days of enrolment, with 33% occurring between days 8 and 21 and 33% after day 21; in other words, a majority of deaths occurred long after being discharged from a health facility or from a health practitioner’s care. Several smaller cohort studies conducted in Asia had also suggested that a significant proportion of mortalities from childhood diarrhoeal disease actually occurred after discharge from clinic/hospital. 16 This is at odds with the traditional conception of diarrhoea-related mortality being closely tied to dehydration and hypovolemic shock. In the GEMS study, children who died were younger (predominantly infants and toddlers) and were statistically more likely to have had infection with enteropathogenic Escherichia coli, enterotoxigenic E. coli (ETEC) and Cryptosporidium. Another study describing a cohort of children hospitalized for severe gastroenteritis showed that the majority of children who died were shown (using molecular methods) to have a treatable pathogen present in their stools; the presence of either ETEC, Cryptosporidium or Shigella was associated with over a two-fold increase in the odds of mortality. 17
Morbidity and illness in young children is particularly problematic because early childhood is a critical period in terms of development. 18 Physical growth during this period is faster than during any other time and many essential cognitive pathways are established prior to the age of 18 months. Disruption of these processes by diarrhoeal disease can lead not only to mortality in the short term but also to impaired cognitive development, less schooling, less economic productivity in adulthood, a predilection to develop metabolic disease in adulthood and reduced offspring birth weight. 19 There have been numerous cohort studies that measured cognitive development in children with repeated episodes of diarrhoea in early life; one series of studies conducted in a Brazilian resource-limited peri-urban community followed a cohort of young infants very closely to document all episodes of gastroenteritis and, 4–10 years later, found in these children a correlation between the total burden of diarrhoea and impairment of visual-motor coordination, auditory short-term memory and information processing, in addition to lower scores on the Test of Non-Verbal Intelligence-III and the Wechsler Intelligence Scale for Children.20–22
Some authors have shown that there is no association between diarrhoea burden and cognition when stunting is controlled for; 23 however, the link between diarrhoea and the development of stunting has been well established. A pooled analysis of nine separate cohort studies (spanning a 20-year time period and five countries) demonstrated that the likelihood of stunting was directly proportional to both the cumulative incidence and the longitudinal prevalence of diarrhoea, with each five episodes of gastroenteritis increasing the odds of being stunted by the age of 2 years by 13% (95% CI: 7–19%). 24
Aetiology of diarrhoea in young children
Diarrhoea is caused by a wide range of aetiological agents which include viruses, bacteria and parasites.25,26 In low- and middle-income countries (LMICs), bacterial enteritis has been shown to be very common.5,17,27 A study on causes of diarrhoea among children presenting to health facilities in sub-Saharan Africa and Asia prior to rotavirus vaccine introduction found that across all sites, the majority of diarrhoea cases were attributed to rotavirus, Cryptosporidium, ETEC and Shigella. 28 When stool specimens from the original GEMS study were re-analysed using highly sensitive quantitative polymerase chain reaction (PCR) methods, it was found that the leading pathogens causing diarrhoea in young children were (in descending order), Shigella spp., rotavirus, adenovirus 40/41, ST-positive ETEC, Cryptosporidium spp. and Campylobacter spp. 29
Accurate identification of gastroenteritis pathogens is crucial for surveillance purposes in order to understand which organisms are most prevalent in which areas and to design specific prevention measures, vaccination strategies and empiric treatment regimens.30,31 Prompt pathogen identification is also frequently required in the context of outbreak investigations in order to quickly implement effective control measures. Laboratory identification can also be very helpful to guide the management of individual patients particularly for those with severe or persistent disease or in immunocompromised hosts. However, there is evidence that diagnosis and treatment of diarrhoeal disease in LMICs is hampered by a lack of laboratory capacity. 32 This poor diagnostic capacity has greatly limited knowledge pertaining to the aetiology and epidemiology of enteric illnesses in these settings. 33
One of the main challenges for the laboratory diagnosis of diarrhoeal infections is that there are over 40 pathogens causing gastroenteritis that include bacteria, viruses, parasites and even potentially fungi. Conventional microbiologic detection techniques require multiple modalities (e.g. bacterial culture, antigen detection and specific staining followed by microscopy) along with the need for specifically trained laboratory staff, various programmes for quality assurance and the ability to source multiple reagents. This has meant that – in resource-limited settings – many pathogens, even those that are common, are often only identifiable at specific national reference/academic laboratories, if at all. Costs for broader testing using more conventional methods can climb to greater than US$200/sample, with results usually available only after many days of specimen processing. 34
In remote settings, attempting conventional culture-based testing may be even more challenging due to adverse transport conditions. 35 It was recently determined that quantitative PCR dramatically improved ascertainment of Shigella spp. burden in children with moderate-to-severe diarrhoea when compared with culture detection in a case–control analysis of samples from multiple LMIC sites; it was estimated that culture likely underestimated the burden of disease due to this organism by approximately half. 36 Additionally multiplex quantitative PCR testing in Bangladeshi infants revealed that it is actually the norm in this context to co-detect >4 enteropathogen targets irrespective of whether samples are tested during a diarrhoeal episode or when diarrhoea is not present. 37 This has led to the concept of the ‘pathobiome’ that often forms the intestinal milieu for young children in these settings. 38
For these reasons, in many settings, the diagnosis of gastroenteritis is now moving away from culture towards rapid nonculture-based methods. 39 This includes rapid antigen detection as well as various nucleic acid amplification methods. These are beginning to replace and/or complement traditional microbiological tests due to their improved sensitivity, specificity, reproducibility and improved practicality.25,40 Several commercial multiplex molecular assays have been developed for the detection of gastrointestinal pathogens directly from clinical stool samples, including those that allow for the detection and identification of greater than 20 pathogens in as little as 1 h. 41 The advent of these nucleic acid–based tests has the potential to revolutionize the landscape of gastroenteritis diagnostics. 42 For some enteric pathogens (particularly viruses), detection with nucleic acid amplification tests have become the gold standard. 43 Due to this improved diagnostic performance, molecular detection methods will likely enhance clinical management, aid infection control efforts and reduce overall social and health-care costs through the rapid identification of the aetiology of diarrhoeal cases. 44 One of the main disadvantages of these methods, particularly for bacterial pathogens, is that viable isolates are not recovered for additional typing and susceptibility profiling, although methods for molecular characterization (including for antimicrobial resistance) continue to improve. 45
Several groups have sought to develop and evaluate rapid and low-cost diagnostics for diarrhoeal infections that can be potentially be used for case management or surveillance at point of need in low-resourced settings. These include simple dipstick immunoassays for Shigella 46 or Cryptosporidium, 47 to rapid isothermal molecular assays for Vibrio cholerae 48 and Entamoeba histolytica. 49 The availability of such diagnostics for use by front-line clinicians may allow for much more rational use of antimicrobials. This would be particularly helpful in settings of high mortality (e.g. severe gastroenteritis requiring admission to hospital) where empiric antibiotics are often administered but rarely actually provide adequate coverage for detected enteropathogens.17,50,51
Risk factors for the development of diarrhoea
Factors that are involved in the occurrence of diarrhoea in children are complex, and the relative contribution of each factor varies as a function of interaction between socio-economic, environmental and behavioural variables. 52 The documented risk factors for diarrhoea include younger age, malnutrition, early weaning, seasonal patterns, low maternal education, lack of piped water supply, poor water-storage practices, lack of vigilant hand washing, poor sanitation and not treating water in the home.53,54 In LMICs, gastrointestinal infections in children are often associated with poor hygienic conditions; 55 poor sanitation, lack of access to clean water supply and inadequate personal hygiene are responsible for 90% of diarrhoeal disease occurrence. 56 Inadequate drinking water, poor sanitation and hygiene are important risk factors that drive diarrhoea diseases in low-income settings,57,58 in addition to socio-economic and age-related risk factors. 58 It has been shown that paediatric diarrhoea incidence is inversely related to socio-economic status, with children in poverty much more vulnerable to acute diarrhoeal episodes, as well as to more severe and longer duration illness. 59 In a study conducted in South Africa, children living in poverty were approximately 10 times more likely to die from diarrhoea than their more privileged counterparts. 60 More than 2.5 billion people lack access to an improved sanitation facility in low-income countries, and inadequate hand hygiene practices have been estimated to affect 80% of the population globally. 57
Prevention
Water, sanitation and hygiene interventions
The agents of acute gastroenteritis are transmitted mainly by the faecal–oral route, either through direct person-to-person contact or through contaminated food or water; therefore, sanitation, good hand washing and hygienic measures prevent spread. 61 Prevention strategies also include policies to address chronic poor nutrition, lack of adequate sanitation and access to safe drinking water, 62 as stipulated by UNICEF and World Health Organization (WHO). 63 As some authors have stated, there are two main approaches to primary prevention of enteric infections: (a) improved water and sanitation and (b) vaccination. 64 Therefore, the majority of diarrhoeal diseases should be preventable by implementing water, sanitation and hygiene (WASH) programmes, which all aim at interrupting faecal–oral transmission pathways, commonly referred to as the five ‘F’ (fluids, fields, flies, fingers and food), 65 as well as promotion of breastfeeding.27,33 Although improvements in water/sanitation infrastructure and hygiene can diminish transmission of enteric pathogens, vaccines can hasten the decline of diarrhoeal disease morbidity and mortality. 42
Basic prevention strategies for interrupting faecal–oral transmission routes focus mainly on hand washing, sanitation and access to sufficient safe water,58,66 and these can potentially be improved through health promotion and education 56 specifically through group or individual training on hygiene education, germ-health awareness, use of posters, leaflets, comic books, songs and drama. 67 It has been estimated that these interventions against gastroenteritis may be able to reduce deaths due to diarrhoeal diseases by up to two-thirds.68,69 In terms of water access, sub-Saharan African cities are some of the worst off in the world, with 20% of populations supplied by an untreated water source especially in informal settlement areas. 70 WASH campaigns, however, have shown substantial variation in effectiveness and questions regarding ongoing user compliance and sustainability remain unanswered. There are fortunately several large, multifaceted community-based trials being carried out in Zimbabwe 71 as well as Kenya and Bangladesh 72 that will hopefully provide much better understanding of the effectiveness, sustainability and medium-term impacts of these interventions.
Vaccination
Rotavirus is the most common cause of acute gastroenteritis in children, both in upper income countries and LMICs.61,73 It is generally spread by direct contact, and therefore, improvements in water quality are unlikely to have significant impact on transmission. Rotavirus is estimated to account for ~39% of diarrhoea hospitalizations 74 and an estimated 199,000 (95% CI: 165,000–241,000) deaths each year, mostly in children under the age of 2 years. 4 More than 85% of these deaths occur in low-income countries in Africa and Asia. 75 Nearly every child will have experienced a symptomatic infection before the age of 5 years,76,77 with the peak incidence occurring among children aged 4–23 months.78,79 Rotavirus results in more diarrhoea-related deaths than any other single agent in countries with high childhood mortality, and it is also a common cause of diarrhoea-related hospital admissions in countries with low childhood mortality. 80 This has prompted the international prioritization of rotavirus vaccines as a primary strategy for the reduction of the mortality associated with this infection. 81 The Rotarix™ (GlaxoSmithKline Biologicals Rixensart, Belgium) and RotaTeq® (Merck and Co., Inc. Whitehouse Station, New Jersey, USA) vaccines are effective for reducing the morbidity and mortality of rotavirus infection82–84 by preventing severe and fatal rotavirus disease. 76 More recently, Rotavac®, a rotavirus vaccine produced in India has shown promise in preventing disease in that context. 85 More recently, in a Phase III randomized controlled trial (RCT), an Indian manufactured heat-stable, live, oral bovine rotavirus pentavalent vaccine (BRV-PV, Serum Institute of India) was found to have an efficacy of 66.7% against severe rotavirus gastroenteritis among infants in Niger. 86 A specific benefit of this vaccine is that it is heat stable which would greatly facilitate distribution in resource-limited and remote settings.
The incidence of severe rotavirus gastroenteritis dropped substantially in the United States such that norovirus became the leading cause of medically attended gastroenteritis within 5 years of the inclusion of the two licensed vaccines into the universal immunization programme. 87 Recent studies from the Latin American and sub-Saharan African context have demonstrated the impact of these vaccines on all-cause childhood diarrhoea mortality.88,89 However, although rotavirus vaccines are available and supported through Global Alliance for Vaccines and Immunization (GAVI), their effectiveness appears to be significantly lower in low-income countries,90,91 and programme implementation has not yet reached many countries particularly in Africa and Asia where the burden is highest. This approach obviously holds considerable promise for reducing the burden of childhood diarrhoeal disease, and although rotavirus vaccines are the first vaccines to be widely implemented that specifically target diarrhoea, there is already some consideration of other possible vaccine targets such as norovirus. 92
Among bacterial pathogens, ETEC and Shigella contribute significantly to mortality and morbidity due to diarrhoeal diseases. 93 Currently, there are two candidates being considered for licensure in the near future as a combined ETEC and Shigella vaccine for use on national Expanded Programme on Immunisation (EPI) schedule. 93 The first candidate ETVAX is a formalin-inactivated ETEC vaccine consisting of four E. coli preparations each engineered to hyper-produce the CFA/1, CS3, CS5 and CS6 antigens of ETEC. In addition, the vaccine contains a cholera B subunit modified to be more cross-reactive with the B subunit of ETEC. 93 ETVAX is co-administered with a double mutant of the ETEC heat-labile toxin (dmLT), which serves as a potent mucosal adjuvant. The second candidate is TSWC which includes formalin-killed Shigella flexneri 2a and 3a and Shigella sonnei which is prepared as a trivalent vaccine. Clinical trials to test both TSWC and ETVAX have been initiated in North America and Sweden and are expected to be expanded to other countries such as Finland, Benin and Bangladesh in 2017. 93 Vaccine effectiveness of these antibacterial vaccines is yet to be established.
Cholera remains a major public health problem in many sub-Saharan African countries causing deaths and retarding development, 94 and although knowledge on stopping cholera transmission and deaths is well documented and was used successfully by countries in South America to eliminate cholera during the previous decades, similar successes have not been replicated in Africa. 94
WHO has prequalified three cholera vaccines which are (a) Dukoral (Crucell Sweden AB, Stockholm, Sweden), a monovalent oral killed vaccine based on whole cells of V. cholerae O1 and recombinant cholera toxin B subunit; (b) Shanchol (Shantha Biotechnics Ltd, Telangana, India); and (c) Euvichol (Eubiologics, Chuncheon-si, South Korea), both bivalent oral killed vaccines based on serogroups O1 and O139. 95 The Dukoral and Shanchol vaccines have shown protective efficacy of 66–86% at 4–6 months after vaccination, 45–62% at 1 year and 58–77% at 2 years, whereas Euvichol vaccine efficacy was 65% after 5 years for those older than 5 years. 95 Herd immunity has been demonstrated for unvaccinated individuals if vaccine coverage is sufficiently high, based on modelling work which predicted that 50% coverage could avoid transmission in endemic areas. 96 Another cholera vaccine available is Vaxchora, which is the only Food and Drug Administration (FDA)-approved, single-dose oral vaccine for the prevention of cholera caused by V. cholerae serogroup O1 in adult travellers from the United States going to cholera-affected areas. 97
Treatment
Re-evaluating individual case management
For some time now, the foundation of clinical management of acute diarrhoeal disease in children in resource-limited settings has consisted of the following three interventions: prevention and treatment of dehydration with ORT, providing adequate nutritional support (including early re-feeding) during the acute episode and provision of oral zinc therapy. There is over a half century of clinical experience with ORT, 98 and it alone is estimated to have saved millions of lives since its more widespread promotion in the 1970s. Nutritional support has also for some time been understood to be an important aspect of managing diarrhoeal disease in resource-limited settings where malnutrition and related ‘environmental enteropathy’ are common. 99 Although historically there were those who recommended delaying re-feeding until improvement of symptoms from the acute illness, numerous studies have shown that early (immediate) re-feeding is safe and beneficial. 100 Oral zinc as a therapy for acute diarrhoeal disease has also been evaluated and in several larger trials has been shown to be effective in shortening the duration of diarrhoeal symptoms in children, with stronger evidence of effect in children aged greater than 6 months of age, 101 although the evidence for mortality reduction is less clear. It was incorporated into WHO and UNICEF management recommendations in 2004. 102 Other important aspects of supportive management in certain circumstances include use of intravenous rehydration, correction of electrolyte disturbances and prevention of hypoglycemia. 103
Use of antimicrobials for the treatment of acute diarrhoeal disease
The WHO’s guidelines for the treatment of acute diarrhoeal disease have long advocated against providing antimicrobials to children with gastroenteritis who do not have blood in their stools; 103 this may have been related to the assumption that the vast majority of these episodes were caused by viral pathogens. These same guidelines recommend treatment with antimicrobials active against Shigella spp. if children with gastroenteritis have bloody stool, presumably with the assumption that the presence of blood in the stool is a sensitive or specific marker for shigellosis and/or that the presence of blood in the stool is prognostic of poorer outcomes.
However, much information has recently been collected that questions the validity of these assumptions. Many studies, including GEMS, have demonstrated that a significant proportion of children with moderate-to-severe or severe diarrhoea in resource-limited settings have detectable enteropathogens that are treatable with antimicrobials, of which some (such as enteropathogenic E. coli, ETEC and Campylobacter) may be even more associated with mortality than Shigella, despite the lack of accompanying dysentery.5,17 Antibiotics were also used to treat diarrhoeal diseases in many of MAL-ED sites; 104 however, these investigators are of the view that although antibiotics should be used in cases of serious bacterial infections, narrow spectrum antibiotics could be used to target specific pathogens in order to reduce disruption of the microbiota that may have long-lived effects. A clinical trial examining the effects of mass distribution of azithromycin to prevent trachoma in Ethiopia unexpectedly found a 49% mortality reduction; 105 on re-examining study data, many experts felt that this was likely due in large part to treatment of enteric infections in these children.106,107 An earlier trial of mass drug administration of azithromycin (also for trachoma prevention) in the Gambia showed a 40% reduction in vomiting and diarrhoea episodes in the treated group at follow-up. 108 We note that many travel medicine guidelines recommend presumptive treatment with antimicrobials for healthy individuals who develop diarrhoea in endemic areas regardless of whether they have bloody stools, despite the fact that many of these people will not have bacterial enteritis (https://www.canada.ca/en/public-health/services/travel-health/about-catmat/statement-travellers-diarrhea.html). 109 Additionally, the risk for mortality, stunting or long-term neurocognitive sequelae in healthy travellers who acquire enteric infection is likely to be negligible, in direct contrast to young children living in resource-limited settings. These recommendations were made on the basis of numerous randomized high-quality blinded placebo-controlled trials showing that duration of travellers’ diarrhoea was shortened by the use of antimicrobial therapy.110,111 Given that many of the participants in these trials might have had viral enteritis, the true benefit afforded by prompt antibiotic therapy targeted at bacterial enteritis might be much larger.
Aside from the fact that enteropathogens besides Shigella have been associated with significant mortality and morbidity, it should be highlighted that the standard ‘method’ for identification of shigellosis (bloody diarrhoea) appears to have very poor performance. It has been repeatedly demonstrated that the sensitivity of bloody stools for diagnosis of Shigella is often far below 60% and may have decreased over time.17,112–116 Recent studies have also found that the presence of blood in the stool did not appreciably modify the association between shigellosis and death 17 nor did it increase the risk of death 117 providing yet another reason to cease using bloody stool as a decision point on diarrhoea management algorithms.
For all these reasons, it may be worthwhile to consider providing antimicrobials to selected children with gastroenteritis in resource-limited settings. Ideally, these would be restricted to those with severe disease, at high risk for mortality or neurocognitive sequelae, or those found to harbour-specific enteropathogens, in an effort to minimize cost, avoid antibiotic adverse events and prevent the rapid development of antimicrobial resistance. In many settings, it appears that antimicrobials are frequently being provided to children with acute diarrhoea; however, they are generally being prescribed empirically often providing poor coverage for the actual pathogens identified. 51 Low-cost and field-ready diagnostics for point-of-care testing might be required to ensure that appropriate antimicrobials are used as well as the sustainability of antimicrobials as a resource. The Antibiotics for Children with Severe Diarrhoea (ABCD) trial is currently enrolling children aged 2–23 months of age with acute diarrhoea and either some/severe dehydration, moderate wasting or severe stunting to determine whether the provision of 3 days of azithromycin is associated with lower mortality and/or less stunting than placebo; unfortunately, recruitment is slated to continue until the end of 2020, so results will not be immediately forthcoming (https://clinicaltrials.gov/ct2/show/study/NCT03130114?view=record). A pilot trial utilizing rapid diagnostics to guide targeted treatment for children admitted to hospital with severe gastroenteritis did demonstrate benefits (e.g. reduced risk of diarrhoea recurrence) for children enrolled in the test and treat arm although the small sample size precludes any definitive conclusions as to impact of this approach. 118 Azithromycin is currently active against many enteropathogens causing acute diarrhoea in resource-limited settings, including Shigella, Campylobacter, ETEC and enteroaggregative E. coli, Yersinia and V. cholerae; whether this drug will maintain activity against these bacteria after widespread use is questionable. A parasitic pathogen responsible for significant morbidity and mortality5,119,120 is Cryptosporidium, for which there are few treatment options available. Nitazoxanide has been shown to confer benefit in the treatment of those with intestinal cryptosporidiosis as compared to placebo in randomized trials; however, this benefit is modest in immunocompetent individuals and essentially absent in HIV-infected patients. 120 Interestingly, nitazoxanide may have some anti-viral activity as well; a small placebo-controlled double-blind randomized trial enrolling children with confirmed rotavirus infection showed that nitazoxanide treatment was associated with a shorter duration of diarrhoea as compared to placebo, 121 and nitazoxanide has been shown in various reports to be active against norovirus as well.122,123
Probiotics
The mechanisms underlying the effect of probiotics in the context of acute diarrhoea are unclear but may be related to improved epithelial barrier function, 124 impact on the intestinal microbiome, modulation of the immune response or direct competition with infecting enteric pathogens.125,126 Studies conducted in upper income countries have demonstrated benefit of probiotic therapy as add-on treatment for acute diarrhoea in children, with a decrease in diarrhoea duration by 14%. 126 Supported by these data, the European Society for Paediatric Gastroenterology Hepatology and Nutrition (ESPGHAN) recommends the use of probiotics, such as Lactobacillus rhamnosus GG or Saccharomyces boulardii, as routine treatment for children with acute diarrhoea 127 with some children’s hospitals in North America following similar clinical management guidelines (https://webcache.googleusercontent.com/search?q=cache:UmnYNc7q_nAJ:https://www.cincinnatichildrens.org/WorkArea/DownloadAsset.aspx%3Fid%3D93672+&cd=3&hl=en&ct=clnk&client=safari). However, since the prognosis and aetiology of acute diarrhoeal disease in children in LMICs are different and it affects the microbiome differently in each location,128,129 clinical evidence in these populations is necessary to routinely recommend probiotics for gastroenteritis treatment.
The most affected regions by diarrhoeal disease in children are South Asia and Africa, and most of the studies of probiotics in LMIC settings were conducted in India. A large randomized trial of the probiotic L. rhamnosus GG in hospitalized children in India observed a reduction of 2 days in duration of diarrhoea and 3 days in hospital stay in the experimental arm; 130 however, 2 years prior, the same authors did a similar study with a lower dose of L. rhamnosus GG, and no significant benefit of probiotic was observed as compared to placebo. 131 In the outpatient setting, L. rhamnosus GG was also found to reduce the duration of diarrhoea by 18 h 132 though another study of this probiotic in hospitalized moderately malnourished children in rural India, with nonbacterial diarrhoea, showed no effect. 133 Still in India, the use of S. boulardii in children was associated with a reduction of 12 h in the duration of diarrhoea. 134 Other probiotics such as L. sporogenes or Lactobacillus acidophilus showed no significant changes in diarrhoea frequency or duration.135,136 In an Indonesian outpatient clinic, a combination of L. rhamnosus and L. acidophilus showed no benefit in reducing diarrhoea duration. 137 Although there have been no high-quality RCTs done using probiotics in Africa, one open-label randomized study in Ghana found no benefit of a fermented millet drink containing Lactobacillus spp. in reducing diarrhoea duration in children in outpatient clinics. 138 In other less-affected LMICs, in Brazil a study with S. boulardii 139 and in Bolivia a study with S. boulardii or a compound containing L. acidophilus, L. rhamnosus, Bifidobacterium longum and S. boulardii, 140 showed reduction in diarrhoeal duration in children with mainly viral gastroenteritis.
Clearly, although studies have shown benefit associated with the use of probiotics in acute diarrhoea in children, results have varied widely; this is likely due to the substantial heterogeneity observed in these trials, relating to geographical location (country, region and locale), probiotic formulation (genus, species, dose and duration), study population (age, severity of illness, presence of malnutrition and comorbid conditions) and diarrhoea aetiology (viral, bacterial, mixed and undefined) impairing the generalization of the results. It is, therefore, important to promote more high-quality RCTs of specific probiotics in defined groups of children with acute diarrhoea in settings where the burden of disease is highest.
Other emerging or potential therapies
Other potentially promising therapeutic interventions are also being explored. Synbiotics – a combination of probiotics with prebiotic – have also been studied as potential therapeutic interventions in various contexts. Prebiotic is a nutritional additive that is used to help ensure that the probiotic strain(s) establish colonization. A recent large RCT of a symbiotic in young Indian infants utilizing a locally developed probiotic strain showed significant reductions in various infections, including diarrhoea. 141 Antiemetics have long been used for the management of acute gastroenteritis associated with vomiting but widespread use was mostly limited due to significant side effects associated with first-generation agents. More recently, the antiemetic with a more favourable side effect profile ondansetron has been studied in children in upper income settings and its use has been associated with lower risk of admission to hospital and lowered rates of intravenous rehydration requirement in paediatric acute gastroenteritis. 142 There is, however, essentially still no data regarding its effectiveness in the LMIC context. The antimotility agent loperamide has also been studied in several settings but has shown most promise in reducing duration of traveller’s diarrhoea. 143 Unfortunately, its use in young and more severely ill children has been associated with potentially severe adverse events, and therefore, it is not generally recommended for use in LMICs or for any children with more severe illness. 144 Finally, the anti-secretory agent racecadotril has been promoted as a potential adjunct to ORT particularly for watery diarrhoea, but recent placebo-controlled trials in Kenyan 145 and Indian 146 children did not demonstrate any measurable significant benefit, including for children with confirmed rotavirus infection.
Future perspectives
Conclusion
Our understanding of paediatric gastroenteritis has advanced dramatically in the recent past. Although it has been known for decades that acute diarrhoeal disease is a leading cause of death for young children, we are now much more aware of how gastroenteritis contributes to the development of severe malnutrition, stunting, cognitive dysfunction and decreased adult accomplishment and productivity. Recent advances in nonculture-based diagnostics, particularly molecular methods, have facilitated numerous epidemiologic studies demonstrating which pathogens are associated with diarrhoeal disease of varying severity in a number of different regions and confirming that children living in resource-limited settings commonly experience enteric co-infections. Vaccination has proven to be an extremely effective tool against rotavirus, the pathogen responsible for more diarrhoeal episodes than any other, in countries that have made it part of their universal immunization programmes. WASH interventions are effective to prevent disease caused by the majority of agents that result in paediatric gastroenteritis, and interrupt transmission of infection, though they are often more difficult to implement, especially in low-resource settings where the burden of disease is highest. Prompt treatment of gastroenteritis with oral rehydration solutions and zinc has been the mainstay of therapy for decades and continues to be of primary importance; however, it seems likely that antibiotic treatment of selected bacterial and/or protozoal infections may result in more rapid improvement, preservation of growth and less mortality. Probiotic formulations have been shown in numerous trials to consistently decrease the duration of diarrhoea in acute gastroenteritis; whether they will also facilitate growth and decrease mortality is yet to be determined. Much work remains to be done to further diminish the impact of acute diarrhoeal disease on the growth and development of young children living in resource-limited settings.
Footnotes
Acknowledgements
J. Pernica is supported by a Hamilton Health Sciences Foundation Early Career Award.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: M.M., J.M.P. and D.M.G. have received funding from Grand Challenges Canada which supports this work. Drs Goldfarb and Pernica have received investigator initiated funding for diarrhoeal diagnostics research from bioMérieux, Inc.
Conflict of interest statement
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
