Cryptosporidium Prevalence and Risk Factors among Mothers and Infants 0 to 6 Months in Rural and Semi-Rural Northwest Tanzania: A Prospective Cohort Study

Background: Cryptosporidium epidemiology is poorly understood, but infection is suspected of contributing to childhood malnutrition and diarrhea-related mortality worldwide. Methods/Findings: A prospective cohort of 108 women and their infants in rural/semi-rural Tanzania were followed from delivery through six months. Cryptosporidium infection was determined in feces using modified Ziehl-Neelsen staining. Breastfeeding/infant feeding practices were queried and anthropometry measured. Maternal Cryptosporidium infection remained high throughout the study (monthly proportion = 44 to 63%). Infection did not differ during lactation or by HIV-serostatus, except that a greater proportion of HIV-positive mothers were infected at Month 1. Infant Cryptosporidium infection remained undetected until Month 2 and uncommon through Month 3 however, by Month 6, 33% of infants were infected. There were no differences in infant infection by HIV-exposure. Overall, exclusive breastfeeding (EBF) was limited, but as the proportion of infants exclusively breastfed declined from 32% at Month 1 to 4% at Month 6, infant infection increased from 0% at Month 1 to 33% at Month 6. Maternal Cryptosporidium infection was associated with increased odds of infant infection (unadjusted OR = 3.18, 95% CI 1.01 to 9.99), while maternal hand washing prior to infant feeding was counterintuitively also associated with increased odds of infant infection (adjusted OR = 5.02, 95% CI = 1.11 to 22.78). Conclusions: Both mothers and infants living in this setting suffer a high burden of Cryptosporidium infection, and the timing of first infant infection coincides with changes in breastfeeding practices. It is unknown whether this is due to breastfeeding practices reducing pathogen exposure through avoidance of contaminated food/water consumption; and/or breast milk providing important protective immune factors. Without a Cryptosporidium vaccine, and facing considerable diagnostic challenges and ineffective treatment in young infants, minimizing the overall environmental burden (e.g. contaminated water) and particularly, maternal Cryptosporidium infection burden as a means to protect against early infant infection needs prioritization. This study was a prospective birth cohort enrolling newborns and their HIV-seropositive or –negative mothers living in the rural and semi-rural areas of Kisesa Ward (population 30,000) [9] in northwestern Tanzania. Pregnant women receiving antenatal care at Kisesa Health Centre (KHC), a Tanzanian government-administered, publically accessible primary care facility were recruited from March through December, 2012, a period that included both the dry and rainy seasons. Women gave birth between April, 2012 and January, 2013; the study follow-up appointments for mothers and infants were conducted between May, 2012 and July, 2013. Eligibility criteria were gestation <37 weeks at consent, singleton birth, known maternal HIV serostatus (screening with Determine HIV-1/2 [Inverness Medical], confirmation with Uni-Gold HIV-1/2 [Trinity Biotech]), maternal ability to speak and understand the local language of Kiswahili, and stated intention to reside within the clinic catchment at delivery and through six months post-partum. The study was advertised through health workers at KHC as well as rural government-run health dispensaries in the region. All HIV-positive women were receiving anti-retroviral treatment (ART) for their own care or for prevention of mother-to-child transmission by the time of delivery. Infants born to HIV-positive women were given nevirapine daily for six weeks and tested for HIV-infection by dried blood spot DNA-PCR at the regional hospital laboratory at the Month 3 follow-up visit. The study protocol was approved by the ethics review committees of the Tanzania National Health Research Ethics Review Committee and Cornell University. Written informed consent was obtained from mothers for themselves and on behalf of their infants at enrolment with verbal assent re-confirmed at follow-up. All women were encouraged to deliver at KHC unless otherwise medically advised. As many women in this region do not deliver at health clinics, and preliminary research revealed that transportation expenses were the primary barriers to accessing healthcare [10], the study provided transportation compensation and other clinical expenses typically borne by mothers for delivery and follow-up visits. For women who delivered elsewhere, including home births, mothers and infants were requested to attend a follow-up clinic visit within three days of delivery. The study flow chart is summarized in Figure 1. If a mother-infant pair did not return for a regularly scheduled follow-up visit, a field worker traveled to their last known address to invite them to return to the clinic for a follow-up appointment. At each follow-up, the research nurse, under supervision of the study coordinator, administered the Infant Feeding and Health Questionnaire to mothers. This questionnaire was designed to obtain data on a range of feeding, health, and environmental risk factors. Exclusive breastfeeding (EBF-WHO) was defined according to the WHO definition where “the infant receives breast milk (including expressed breast milk or breast milk from a wet nurse) and allows the infant to receive oral rehydration solution (ORS), drops, syrups (vitamins, minerals, medicines), but nothing else” [11]. Duration of EBF-WHO was defined as the time from birth until an infant first received food or liquids other than breast milk or medicines. Diarrhea was defined as loose or watery stools ≥ three times per day that represented a pattern atypical for that individual [2]. The questionnaire included: 1) infant nutrition: breastfeeding and complementary feeding practices; 2) mother-reported infant morbidity: cough, difficulty breathing, fever, convulsions, vomiting, skin rash, anorexia, unscheduled clinic/hospital visits, and episodes of diarrhea; and 3) environment: food security, using an index composed of questions relating to the mother's food consumption pattern, and sanitation and hygiene practices, such as hand-washing behavior, access to safe water, and toilet facilities. Infants exhibiting symptoms of illness were referred to the clinical officer at KHC for follow-up. Anthropometric assessments were collected at each follow-up visit. Maternal height and weight were measured using a standard stadiometer (Health O Meter, Inc., Bridgeview, IL) to the nearest 0.2 kg and nearest 0.1 cm, respectively. Maternal mid-upper arm circumference (MUAC) and triceps skinfold thickness (TSF) were measured to the nearest 0.1 cm and 0.5 mm, respectively. Infant weight and length were measured using a calibrated digital infant scale (Seca 334 Digital Baby Scale) to the nearest 0.01 kg and a standard infant length board to the nearest 0.1 cm, respectively. Infant MUAC, TSF, and head circumference were measured to the nearest 0.1 cm, 0.5 mm, and 0.1 cm, respectively. Active case detection was of interest so maternal and infant fecal samples were collected irrespective of self-reported intestinal symptoms at each follow-up visit. Cryptosporidium infection was detected using fresh stool samples that were stored in a cooler with ice packs for ≤5 hours before being transferred and stored at 4°C in the parasitology laboratory of the Tanzanian National Institute for Medical Research (NIMR), Mwanza Research Centre. Within 24 hours of collection, approximately 5 g of stool was mixed with 5 mL 10% v/v formalin and stored at 4°C until analysis. Presence of Cryptosporidium was confirmed using a modified Ziehl-Neelsen staining procedure [12], which is estimated to have a sensitivity ranging from 32 to 79% and a specificity ranging from 89 to 100% [13]–[15]. After staining, slides were examined by a single technician, without knowledge of participant clinical status, using a light microscope (Olympus model CX41RF) to detect Cryptosporidium oocysts and estimate oocyst burden. Cryptosporidium infection was defined as ≥1 oocyst detected in stained fecal smears. A second technician re-examined a sample (10%) of the slides and inter-observer agreement was 96%. Data were analyzed in STATA10 (STATA Corporation, Texas, USA). Means of normally distributed continuous variables were compared using Student's t-test and proportions of categorical variables were compared using the χ2 test and Fisher's Exact test. Results were considered statistically significant at α = 0.05, two-sided. Univariate and multivariate logistic regression models were used to estimate the odds ratio (OR) and 95% confidence interval (95% CI) of a priori considered potential risk factors for infant Cryptosporidium infection (HIV-exposure, exclusive breastfeeding, maternal Cryptosporidium infection, and household factors, such as animal ownership, sanitation, wealth, and maternal education). This study is registered with ClinicalTrials.gov, number {"type":"clinical-trial","attrs":{"text":"NCT01699841","term_id":"NCT01699841"}}NCT01699841. The sponsors (Cornell University and the National Science Foundation) were not involved in the design or oversight of the study. Members of the writing team had full access to the study data. The authors had final responsibility for the decision to submit for publication.