Enteric pathogen infection and consequences for child growth in young Aboriginal Australian children: a cross-sectional study

Infections with enteric pathogens are common in disadvantaged populations with low socio-economic status due to inadequate sanitation and hygiene, overcrowded living conditions, and gaps in health literacy [1]. During the first two years of life, enteric infections have a particularly devastating effect on child health as a result of poor nutritional status, loss of appetite and diarrhea, disturbances in intestinal function that lead to impaired absorption of micronutrients and low grade inflammation [2]. More specifically, infection with Shigella sp.and Giardia lamblia have been associated with a significantly higher risk of stunting [2]; enterotoxigenic Escherichia coli with a higher risk of being underweight [3] Campylobacter jejuni has been shown to be associated with lower growth rates for both length and weight [4]; and hookworm and Trichuris trichiura have been associated with anaemia [5, 6]. As well as impaired growth and cognitive development, children may have anaemia and experience lethargy, which can result in reduced educational achievement and decreased opportunities in later life [7, 8]. Impaired child growth in the first 1000 days of life is also associated with long-lasting adverse consequences for adult health, including increased risk of chronic diseases [8‐10].

A high prevalence of enteric infections has been previously reported in Australian Aboriginal communities, from a small number of surveys in older children and adults [11‐15]. However there is a paucity of studies of enteric infections in very young children in these communities. This is mainly due to difficulties in collecting and transporting samples from isolated locations, the limitations of conventional methodology, [11, 12] and challenges in achieving an accurate diagnosis in a field setting. New technologies such as multiplex molecular assays now allow more rapid and sensitive and simultaneous diagnosis of multiple enteric organisms [16] and more reliable diagnosis in remote settings. Recent studies suggest that “silent” enteric infections, in the absence of overt diarrhea, may still result in significantly adverse clinical outcomes, such as growth faltering [17‐19]. The goal of our study was to determine: i) the prevalence of enteric carriage in all Aboriginal children aged between 0 and 2 years living in a remote Northern Australian community using conventional and molecular techniques, and ii) the association between the presence of enteric pathogens/non-pathogens and child growth outcomes.

We enrolled 70 of the ~ 100 children aged between 0 and 2 years resident in the community. Stool samples were collected from 62 (88.5%) children. In the sample of children who had stool samples collected, more than half of mothers/carers (51.6%) were educated to year 11 level. The prevalence of maternal underweight was 23%, overweight 19.4%, and obesity 8.1%. A smoker resided in 87.1% of households, and the average number of residents per household was six. Traditional medicine was used by 70.9% of families. The majority of children (87.1%) were currently breastfed. Food insecurity was common with 58.1% of households reporting running out of food in the previous two weeks. Almost all women (98.4%) received government benefits. Baseline characteristics are shown in Table 1.

There is a paucity of data on the prevalence of enteric pathogen infections in infants and toddlers living in remote Aboriginal communities. We found only two studies in a remote Aboriginal community in the Northern Territory which documented the presence of intestinal parasites in 0–2 year old children [12]. One study in 1994–1996 at the same location identified the presence of Ancylostoma. duodenale (A.duodenale) (9%), T.trichiura (36%), R.nana (9%) and Entamoeba spp (9%). in < 24 months old children (n = 11), and S. stercoralis (27%), R. nana (17%), Giardia duodenalis (13%), Entamoeba spp. (10%), A. duodenale (13%), and T.trichiura (87%) in children aged 2–4 years 11 months (n = 30) between July 1994 to October 1996 [12]. Another study in the same location between 2010 and 2011 in children aged < 24 months demonstrated a prevalence of S.stercoralis (3% on microscopy (n = 29), 9% on agar plate culture (n = 23), 8% on PCR [11] (personal communication Dr. Deborah Holt Sept 2019), 28% on serology (n = 38) [27] (personal communication Dr. Kate Mounsey Sept 2019), T.trichiura (14% on microscopy) and R.nana (7% on microscopy) (n = 29) [11] These findings are in contrast to those of our study, in which no cases of strongyloidiasis or hookworm were observed and the percentage of T. trichiura and R. nana positive were reduced. This reflects the implementation of public health measures (preventive chemotherapy, test and treat strategies) in this area that were introduced for soil transmitted helminths (albendazole from 1995, and an ivermectin mass drug administration in 2010 and 2011), leading to a reduction in S. stercoralis and hookworm prevalence [11, 15]. Another study examined long-term cryptosporidiosis patterns across Western Australia between 2002 and 2012, using data obtained from the Western Australian Notifiable Infectious Disease Database, and demonstrated a high burden of cryptosporidiosis in Aboriginal children aged 0–4 years of age (33.3%) [28]. Specific practices and environmental factors (e.g. hygiene and socioeconomic factors, presence of livestock and other animals, previous de-worming programs, weather patterns) may account for the different rate in our study, in which there were only five (8.2%) isolates of cryptosporidium.

There is growing evidence of the importance of pathogens in a child’s gut, even when asymptomatic. This “pathobiome” is thought to be causal in the development of ‘environmental enteropathy’, with associated chronic intestinal inflammation, villous blunting and intestinal leakage and malabsorption, leading to a cycle of malnutrition, micronutrient deficiencies, and growth impairment [29]. Childhood diarrhoea has been conceptualized as a syndrome of enteropathogen excess, rather than a single infection, with diarrhoea more likely to occur when a quantitative threshold of pathogen burden is exceeded [30]. Indeed in our study, the majority of children did not present with diarrhoea, and pathogen burden was likely to be low (as most isolates were only identified by PCR and diarrhoea was uncommon); however carriage of two or more enteric pathogens or Blastocystis hominis was associated with impaired linear growth (lower height for age z scores). This confirms that pathogens not causing diarrhoea may still cause significant morbidity in resource- constrained settings, and efforts towards improving environmental conditions including improvements in clean drinking water, housing standards and hygiene and sanitation are critical.

B.hominis is usually considered a non-pathogen but in some instances has been reported as a pathogen [31]. We observed a significant association between carriage of B. hominis and lower height for age z scores. This has been documented previously, with infection with B. hominis shown to be associated with lower anthropometric indexes in young children in Turkey [32]. These findings suggest that organisms characterised as ‘non-pathogenic’ may in fact carry some pathogenic potential, possibly through disturbance of the gut microbiome and/or damage to the intestinal mucosa resulting in increased intestinal permeability and impaired nutrient absorption [28]. Poor nutritional status may also have preceded infection by B.hominis [33].. The finding of B. hominis in stool may also represent exposure to organisms spread by the faecal-oral route [34].

Our study also provides important evidence on the application of molecular diagnostics to identify enteric pathogen burden. Use of PCR has previously been shown to substantially alter previous estimates of pathogen prevalence. Re-analysis of specimens from the MAL-ED cohort study using quantitative PCR demonstrated a predominance of viral causes of diarrhea (including sapovirus, rotavirus and adenovirus) with only two pathogens (rotavirus and enterotoxigenic E. coli) remaining in the top five causes of diarrhea identified by the original microbiological work up [35]. In our study, the majority of pathogens in stool samples from children were detected by PCR (qualitative), allowing increased detection rates and correlation with clinical outcomes.

Strengths of this study are that we worked closely with the local primary health care providers and trained Aboriginal Community Based Researchers to carry out the research, resulting in a high representative sample of young children in this remote Aboriginal community. We were also able to collect comprehensive information on possible nutritional, environmental and socio-demographic influences on clinical outcomes. Limitations are the relatively small sample size (even though it represented 68% of the target population), stool samples were not collected from mothers or carers of the child and hygiene practices being assessed only by questionnaire and not by direct observation (which may have introduced some recall bias); and we only asked about diarrheal infections in the last 24 h.

We found a high prevalence of enteric pathogen and non-pathogen carriage in Australian Aboriginal infants and toddlers and a significant association between infection with multiple pathogens and child growth (impaired height for age z scores). These findings suggest that environmental contamination in remote Northern communities where extreme heat and humidity is common, may contribute to the high rates of childhood growth impairment and stunting reported, [36] with health implications extending into adulthood. The findings highlight the urgent need for sustainable strategies to improve housing, hygiene and sanitation in remote Aboriginal communities. Our findings also have relevance for many other First Nations Communities around the world that face similar challenges with regard to poverty, infections, and malnutrition.