Symptomatic and Asymptomatic Campylobacter Infections and Child Growth in South Asia: Analyzing Data from the Global Enteric Multicenter Study

ABSTRACT. Campylobacter is a major cause of food-borne gastrointestinal illnesses worldwide, predominantly affecting children under 5 years of age. This study examined potential associations of symptomatic (with diarrhea) and asymptomatic (without diarrhea) Campylobacter infections with child growth among children under 5 years of age in South Asia. The Global Enteric Multicenter Study was conducted from 2007 to 2011 with a case-control design. Children were followed for 60 days after enrollment. Stool culture was performed to isolate Campylobacter spp. Among the 22,567 enrolled children, 9,439 were symptomatic, with 786 (8.28%) testing positive for Campylobacter. Conversely, 13,128 asymptomatic healthy controls were included, with 1,057 (8.05%) testing positive for Campylobacter. Growth faltering was observed in the symptomatic group, particularly among children aged 0–11 months (−0.19 height-for-age z score [HAZ]; 95% CI: −0.36, −0.03; P = 0.018) and 24–59 months (−0.16 HAZ; 95% CI: −0.28, −0.04; P = 0.010). However, in the asymptomatic group, growth faltering was observed only in the 24- to 59-month age group, in terms of HAZ (−0.15 HAZ; 95% CI: −0.24, −0.05; P = 0.002) and weight-for-height z score (−0.16; 95% CI: −0.26, −0.06; P = 0.001). These findings underscore the importance of immediate and enhanced introduction of preventive modalities to reduce the burden of Campylobacter infections and reduce their long-term sequelae.


INTRODUCTION
2][3] This enteric pathogen is significantly associated with moderate-to-severe diarrhea (MSD) among children aged 0-11 months in Bangladesh and Pakistan; and 24-to 59-month-old children in India and Pakistan. 2Campylobacter enteritis is an acute, self-limiting infection, and clinical manifestation ranges from watery diarrhea to dysenteric illnesses. 4,5This enteric infection damages gut mucosa, disrupts widespread gut commensal flora, and may even cause prolongation of the diarrheal episode. 6Long-term health sequelae include near-fatal Guillain-Barr e syndrome, reactive arthritis, or Reiter's syndrome. 7Postinfectious irritable bowel syndrome and inflammatory bowel disease are also linked with Campylobacter infection. 80][11] A study conducted in Peru reported that younger age, a recent history of diarrheal illness, and lower maternal education were linked to both asymptomatic and symptomatic Campylobacter infections.][14] The isolation of this pathogen in nondiarrheal children could be explained by poor sanitation and early contact with animals. 13Moreover, excretion of Campylobacter by diarrheal infants has been reported in 75% of cases less than 7 days after diarrhea onset.However, such excretion may be prolonged, as high as 15 days and more among infants with both symptomatic infections (18%) and comparable asymptomatic colonization (11%) after diarrhea onset. 15Recent reports from the Etiology, Risk Factors, and Interactions of Enteric Infections and Malnutrition and the Consequences for Child Health and Development Project (MAL-ED) study have shown that the burden of asymptomatic infection by Campylobacter is associated with increased enteric inflammation. 16Previously, Chen et al. hypothesized that infections by certain species of the Campylobacter genus cause interruption of the gut mucosal barrier by targeting tight junctions and inducing a proinflammatory response in colonic epithelial cells, thereby leading to gastroenteritis. 17Recent studies have indicated that Campylobacter can use sufficient nutrients from the gut luminal environment for survival and growth. 18Asymptomatic malnourished children are more likely to carry Campylobacter in developing countries. 12This has led to the hypothesis that, it is an opportunistic infection in this setting, maybe linked to immunosuppression caused by malnutrition. 12,19A study of Peruvian Amazonian children observed reduced weight gain and marginally compromised linear growth over a 3-and 9-month period, respectively, in cases of both symptomatic and asymptomatic Campylobacter infections.Such observations were more pronounced in the case of more severe Campylobacter infections. 12Another study conducted in eight LMICs found Campylobacter jejuni/coli infections were associated with poor growth performance. 16owever, information regarding the impact of Campylobacter infections on childhood growth in LMICs is lacking, especially in South Asia, where Campylobacter infection is a leading cause of MSD (moderate-to-severe diarrhea) episodes. 1,2In this study, our objectives were to evaluate potential associations of symptomatic and asymptomatic Campylobacter infections with child growth in three distinct countries in South Asia.  .Before enrollment in the study, mothers/caregivers of eligible under-5 children were verbally informed about study objectives as well as the protocol; only those who gave written consent voluntarily were enrolled after stool specimens were provided.Subsequently, the mothers/caregivers were interviewed.

MATERIALS AND METHODS
Study sites.The Global Enteric Multicenter Study (GEMS) was a 3-year prospective, age-stratified, matched casecontrol study of MSD (moderate-to-severe diarrhea) among children aged 0-59 months who sought care at sentinel hospitals and health centers in seven study sites in sub-Saharan Africa and South Asia. 21Children from the South Asian countries of the GEMS-Mirzapur, a rural community in Bangladesh; Kolkata, an urban site in India; and Karachi (Bin Qasim Town), a periurban site in Pakistan-were included in our analysis.
Study design and participants.The study was conducted from December 2007 to February 2011, and a case-control design was followed. 20Under-5 children from the demographic surveillance system catchment area presenting with MSD to the sentinel health center within 7 days of an acute illness onset were considered cases.Age-, sex-, and community-matched healthy controls were randomly selected from the community.Children with MSD and the healthy controls were enrolled and followed up after 60 days. 20Follow-up visits took place at the household of the participants.Nutritional assessments were performed at the time of enrollment (after hydration) as well as during follow-up visits.
Specimen collection and laboratory procedure.A fresh whole-stool sample (minimum 3 g) was collected from each enrolled child (both cases and controls).3][24] Bacterial pathogens (Salmonella, Shigella, Campylobacter spp., Aeromonas spp., Vibrio cholerae, and Escherichia coli [enterotoxigenic (ETEC), enteropathogenic, and enteroaggregative (EAEC)]), viruses (rotavirus, norovirus, sapovirus, astrovirus, and adenovirus), and protozoa (Entamoeba histolytica, Giardia intestinalis, and Cryptosporidium spp.) were detected following standard laboratory methods as described elsewhere. 25n the GEMS, stool samples were delivered to the laboratory in cooler boxes after collection.A separate fecal aliquot was placed in two tubes, one containing Cary-Blair media 26 and the other containing buffered glycerol saline (BGS), 27 either at the time of collection or after the samples were received in the laboratory.When a fecal specimen could not be collected, a rectal swab was obtained during enrollment and promptly placed into the tubes containing Cary-Blair media and BGS.The study staff critically monitored the time between sample collection and transport to the laboratory, whereby the interval between stool collection and transport medium inoculation did not exceed 6 hours and the duration between placing the specimen in transport media and accession did not exceed 18 hours.For the following examinations, separate aliquots of the stool samples were made and stored in a 280 C freezer before further analysis.
Fecal microbiology and Campylobacter spp.isolation.The GEMS protocol incorporated traditional bacterial culture, largely to allow central laboratories to independently validate the growth of the involved pathogens and characterize them further for virulence, serologic, and antibiotic resistance features as described elsewhere. 25Swabs were plated onto Campy blood agar plates (Campy-BAPs) from the Cary-Blair tube.After observing the Campy-BAPs for growth on day 3 at 42 C, plates were further incubated.Suspicious colonies were chosen after incubation and subjected to a series of simple biochemical tests that could be easily carried out in resource-limited settings. 25ata collection.The passing of three abnormally loose or watery stools per 24 hours was defined as diarrhea. 20Many of the factors, such as vomiting (at least three times per day), fever (at least 38 C), and the presence of obvious blood in the stools, were examined. 20Both exclusively and partially breastfed children were referred to as "breastfed."The information about the enrolled child's family (defined as a group of people sharing a common cooking fire), included the mother as the major caregiver, maternal education (illiterate or literate), household size (including the number of children under the age of 5), building materials of the household (most common floor material: earth, sand, dung, and other), handwashing practices (before nursing or preparing baby food; after handling animals and cleaning a child), access to the main source of drinking water (tube well water and nontube well water), use of handwashing materials (water with soap or without soap), water treatment (water treatment method of drinking water available or not), improved toilet facilities (toilet facility for disposal of human fecal waste available or not), and presence of domestic animals or pets on the premises (sheep, goat, chicken, cow, dog, and cat), all of which were categorized as the explanatory variables in this analysis. 21To analyze possible factors linked with infection status (the presence of Campylobacter spp.), households were classified into socioeconomic status quintiles based on wealth index quintiles (poorest, lower middle, middle, upper middle, and richest). 20Global Enteric Multicenter Study field staff visited each enrolled child's household approximately 60 days after enrollment, and detailed information on the morbidity of the participant was recorded as published elsewhere. 20nthropometry.For each participant, height, weight, and mid-upper arm circumference (MUAC) were measured at enrollment and the 60-day household follow-up visit; details of the measuring methods have been described elsewhere. 20Using a digital scale calibrated every day (model 314, Tanita Corporation of America, Arlington Heights, IL), weight (to the nearest 0.1 kg) was measured following the standard guidelines for measurement.In the recumbent position, the length of children aged 0-23 months or those who were older but unable to stand unassisted were measured using a board with a fixed head and sliding foot piece (to the nearest 0.1 cm) (Shorr Productions, Olney, MD).For children aged 2 years and older, the same apparatus was used to measure standing height.To calculate the MUAC to the nearest 0.1 cm, a 25-cm paper single-slotted insertion tape was used (Shorr Productions). 28Length/ height and MUAC were measured three times each and the average was estimated. 29Using WHO Child Growth Standards as the reference population, the height/length-for-age, weight-for-age, and weight-for-height/length z scores (HAZ, WAZ, and WHZ) were measured using a WHO SAS macro.After assessing the nutritional status of the children, z scores were calculated and categorized as underweight (WAZ , 22), stunted (HAZ , 22), or wasted (WHZ , 22) following the WHO anthropometry guidelines. 30,31ata analysis.We reported the child-and householdlevel characteristics by using mean and SD for continuous variables and frequency as a percentage for categorical variables to summarize the data.A paired t test was used to test the statistical significance of an observed difference between baseline and endline (follow-up measurement) z scores among these study children.To assess the association between the presence of Campylobacter spp. in stool among the symptomatic and asymptomatic children at baseline and the difference in the child's HAZ, WAZ, and WHZ in the subsequent 60 days, we used a generalized linear model where the explanatory variable was the presence of Campylobacter in stool and the outcome variables were HAZ, WAZ, and WHZ.The repeated measure was used as the time variable in STATA (Stata Corporation, College Station, TX).We analyzed the difference in the z score from baseline to endline (60 days) by comparing the two measures over time.The explanatory variable (the presence of Campylobacter spp.) was used individually in the generalized linear model (simple linear regression) at first to investigate its unadjusted impact on the outcome variable (WAZ, HAZ, and WHZ).All the factors, namely age, sex, diarrhea at enrollment, breastfeeding status, mother's education, number of under-5 children in the household, handwashing before nursing a child and after cleaning a child, handwashing material, main source of drinking water, wealth index, available toilet facility, co-pathogens (ETEC, Shigella, Aeromonas, and rotavirus), comorbidity (malaria, typhoid, pneumonia, diarrhea, and dysentery), time (because it was repeated measured data, the anthropometry was taken in two time points: on enrollment [0] and at day 60 follow-up [1]), and study site (country), suggesting the associations with the outcome as indicated in the literature were chosen for multivariable modeling (multiple logistic regression).The variance inflation factor (VIF) was calculated to detect multicollinearity, and no variable with a VIF value greater than 5 was identified in the final model.We estimated the coefficient and its 95% CI to describe the precision of the point estimate.During the analysis, a P value of , 0.05 was considered statistically significant.there were 230, 246, and 310 cases, respectively, and in the control group, there were 241, 427, and 411 individuals, respectively.There were 1,840 baseline data for HAZ, 1,842 for WAZ, and 1,839 for WHZ that could be used for analysis.However, we had 1,685, 1,686, and 1,681 data at 60 days of follow-up for HAZ, WAZ, and WHZ, respectively (Figure 1), because the height and weight of all the study participants could not be measured during baseline (the child's parents were unwilling to provide the data) and endline during follow-up (dropout).Figure 2 shows the baseline and endline difference of HAZ, WAZ, and WHZ.
Symptomatic Campylobacter (1) children.Those children who were Campylobacter positive with diarrhea were enrolled as cases in the GEMS upon fulfilling the selection criteria as follows: they were between the ages of 0 and 59 months; belonged to the population covered by the demographic surveillance system at the study site; were not currently enrolled as a case; and met the case definition of diarrhea ($ 3 abnormally loose stools in the previous 24 hours).The diarrhea episode also had to be acute (with onset within 7 days of study enrollment) and referred to as a new episode (onset after at least 7 diarrhea-free days); in addition, the child's diarrhea had to be moderate to severe, as determined by at least one of the following criteria: sunken eyes as confirmed by the parent/primary caretaker as more than normal, loss of skin turgor as determined by an abdominal skin pinch (with slow return [# 2 seconds] or very slow return [. 2 seconds]), intravenous rehydration administered or prescribed, dysentery (visible blood in a loose stool), or hospitalization with diarrhea or dysentery.
Asymptomatic Campylobacter (1) children.Those children who were Campylobacter positive without diarrhea were enrolled as controls in the GEMS upon fulfillment of the following criteria: they resided in the demographic surveillance system area and were matched to the index case on the basis of age, sex, and time of enrollment within 14 days of presentation of the case.The age-matching criteria were 62 months for cases aged 0-11 months and 64 months for cases aged 12-59 months, with controls not exceeding the stratum boundaries of the case.In addition, controls must not have had diarrhea in the previous 7 days.
Table 1 illustrates that approximately 48% (N 5 375/786) of symptomatic Campylobacter infection occurred in children aged 0-11 months, followed by 34% (N 5 270/786) in children aged 12-23 months.In the asymptomatic group, 43% (N 5 456/1,057) were in children aged 12-23 months followed by 34% (N 5 361/1,067) in the 0-to 11-month age group.The symptomatic group had a higher percentage of underweight children (41%, N 5 786), whereas the asymptomatic group had a higher percentage of children who were stunted (36%, N 5 1,057).In both groups, the percentage of female sex and wealth index were similar.
The outcomes of the multiple linear regression model are presented in Table 3.After adjusting for sex, breastfeeding status, maternal education, handwashing before nursing a child and after cleaning the child, handwashing material, main source of drinking water, available toilet facility, wealth index, co-pathogens (ETEC, Shigella, Aeromonas, and rotavirus), study site, and comorbidity (malaria, typhoid, pneumonia, diarrhea, and dysentery), significant growth faltering was observed in the symptomatic group (MSD children, who were enrolled as cases) among infants and young children aged 0-11 months (20.19 HAZ difference in 60-day follow-up; 95% CI: 20.36, 20.03; P 5 0.018) and 24-59 months (20.16HAZ difference; 95% CI: 20.28, 20.04; P 5 0.010).However, in the asymptomatic group (healthy children without having diarrhea, enrolled as controls from the community), growth faltering was observed only in the 24-to 59-month age group (20.15 HAZ  3).

DISCUSSION
The study findings supported our hypothesis that both symptomatic and asymptomatic infections by Campylobacter were associated with growth faltering in under-5 children in South Asia.Important findings concerning these children with symptomatic and asymptomatic Campylobacter infections include 1) Campylobacter infections were equally distributed among study children with MSD and children without diarrhea, 2) growth faltering was observed during a follow-up period of approximately 60 days, and 3) co-pathogens that were commonly detected with both symptomatic and asymptomatic Campylobacter infections were ETEC, Aeromonas, EAEC, Cryptosporidium, G. lamblia, and Shigella.We observed reduced growth with both symptomatic and asymptomatic Campylobacter infections in this cohort from the Indian subcontinent.However, after controlling for seasonality and potential confounders such as socioeconomic and demographic factors, one prospective multisite birth cohort study found an association between Campylobacter infection and poor linear growth in children under the age of 2 years, suggesting that Campylobacter may play a role in childhood malnutrition. 16ampylobacter is a bacterial cause of diarrhea that has been reported in infants and young children in both industrialized and developing countries. 32,33In this study, control children were diarrhea free for the previous 7 days at the time of enrollment in the study.Cases and controls were equally infected with Campylobacter spp.All these observations could be explained in several ways; these controls may have experienced symptomatic Campylobacter infections in the recent past.Despite recovery from the diarrheal episode, these control children continued to excrete Campylobacter asymptomatically for an extended period (more than 7 days after onset).However, we do not have any data because the study did not take any proper history in support of our statement.Like other enteric pathogens in circulation, Campylobacter may have heterogeneous circulation, and some study children may have had an infection with those strains that were nonpathogenic or less pathogenic, and despite similar infective doses they were not able to develop manifestations of clinical diarrhea-like pathogenic strains.
Experimental challenge studies in healthy adult volunteers have indicated that the presence of any less pathogenic or nonpathogenic microorganism as a single infection may cause the manifestation of a relatively milder diarrheal episode or no episode at all. 34One study reported that both the susceptibility of the particular hosts and bacterial variables have an impact on the development of campylobacteriosis. 35Numerous virulence genes involved in invasion, colonization of the intestinal mucosa, motility, adhesion, and toxin synthesis may be found in the Campylobacter genome.It was depicted that despite the cytopathic effects on epithelial cells, the capacity of bacteria to enter host cells is frequently not the primary mechanism leading to infection. 35Another study reported that there are numerous species of Campylobacter spp. 36However, C. jejuni/coli infection is predominantly linked to Campylobacter enteritis in humans.The pathogenicity of non-jejuni/coli Campylobacter-like Campylobacter upsaliensis and Campylobacter helveticus (isolated from dogs and cats) is unknown.In our study, some children may have had an infection with Campylobacter as the sole pathogen, and in the absence of pathogenicity or because of being less pathogenic any overt clinical illness was not manifested.Moreover, due to the failure to use molecules present on the surface of human intestinal cells as specific receptors, there were no disease manifestations other than asymptomatic infections.4][45][46][47][48][49][50] Campylobacter-associated episodes are acute but self-limiting and more common among malnourished children. 43,51,52It has also been reported that asymptomatic Campylobacter infection was more common among malnourished children. 19Both symptomatic and asymptomatic infections were associated with the excretion of Campylobacter for a longer duration. 15,53The adverse effects on gut mucosa due to Campylobacter infections may even extend beyond the period of acute infection.Chen et al. reported that disrupted barrier function and increased cytokine secretion by intestinal epithelial cells attract neutrophils, macrophages, and lymphocytes to the gut inflammation site, causing intensification of host responses that may result in uncontrolled postinfectious intestinal complications including irritable bowel syndrome. 17alnourished asymptomatic children were more often infected with Campylobacter than symptomatic children, and this is potentially caused by the presence of their host susceptibility factors and the absence of nonspecific host factors that can confront diarrhea pathogens. 50,54Well-nourished symptomatic children were infected with Campylobacter more frequently than asymptomatic well-nourished children, as Campylobacter has been revealed to be able to obtain sufficient nutrients from the host's gut environment for survival and rapid growth in the intestinal lumen. 18A prospective cohort study of 442 children aged 0-72 months conducted in a semirural community in the Peruvian Amazon reported the nutritional consequences of Campylobacter infections.
The study observed reduced weight gain due to asymptomatic and symptomatic Campylobacter infections over 3 months.However, symptomatic Campylobacter infections were marginally associated with reduced linear growth over 9 months.The study further revealed that relatively severe episodes were associated with reduced linear growth. 12he present study has several strengths.Data were collected from a relatively large number of children presenting with MSD and their concurrent controls and the study gathered a variety of information.The study focuses on health institutions from three distinct South Asian sites.The strengths of our study also include unbiased random sampling and high-quality laboratory performance.A notable feature of this study was the single follow-up household visit, roughly 60 days after enrollment, and consequently the observed association may be due to an acute effect.Separate analyses for C. jejuni/coli, a pathogenic variant of Campylobacter spp., were not possible; this was a limitation of our study.

CONCLUSION
Both symptomatic and asymptomatic Campylobacter infections were associated with growth faltering in under-5 children in South Asia.High preventive public health priorities from different sectors of health policy are imperative to reduce the burden of childhood symptomatic and asymptomatic Campylobacter spp.infections, and thereby their nutritional consequences.* Adjusted for sex, breastfeeding status, mother's education, handwashing before nursing a child and after cleaning a child, handwashing material, main source of drinking water, available toilet facility, wealth index, copathogens (enterotoxigenic Escherichia coli, Shigella, Aeromonas, and rotavirus), sites (country), and comorbidity (malaria, typhoid, pneumonia, diarrhea, and dysentery).Campylobacter was detected from the stool samples during enrollment.Anthropometric measurements were taken during enrollment and 60 days after enrollment (during the follow-up visit).
the United Kingdom.We thank our core donors for their support and commitment to icddr,b's research efforts.Data availability: A publicly available Global Enteric Multicenter Study dataset was analyzed in this study.These data can be obtained from ClinEpiDB (https://clinepidb.org/ce/app/record/dataset/DS_841a9f5259).

Received May 22 ,
2022.Accepted for publication March 14, 2023.Published online May 1, 2023.Acknowledgments: We are grateful to Global Enteric Multicenter Study staff, parents, and children for their contributions.This research was funded by the Bill & Melinda Gates Foundation and by the core donors who provide unrestricted support to the International Center for Diarrheal Disease Research, Bangladesh (icddr,b) for its operations and research.Current donors providing unrestricted support include the governments of Bangladesh, Canada, Sweden, and Funding support: This work was supported, in whole or in part, by the Bill & Melinda Gates Foundation [INV-002050].Under the grant conditions of the Foundation, a Creative Commons Attribution 4.0 Generic License has already been assigned to the Author Accepted Manuscript version that might arise from this submission.
Ethical considerations.Study protocol approval was conferred by the institutional review board of the University of Maryland as well as by the research review committee and ethical review committee of the respective governmental or collaborating local institutions from each study site (Centro de Investigac¸ao em Saude da Manhic¸a, Manhic¸a, Mozam-

TABLE 1
Sociodemographic characteristics and nutritional status of the study children EAEC 5 enteroaggregative Escherichia coli; ETEC 5 enterotoxigenic Escherichia coli.Results are from x 2 tests.