Cryptosporidiosis: a Neglected Infection and Its Association with Nutritional Status in Schoolchildren in Northwestern Mexico

Introduction: Undernutrition is a recognized public health problem in Mexican children and cryptosporidiosis is probably a factor contributing to this problem. However, it remains a neglected and poorly attended infection in Mexico. This study aimed to determine the prevalence of Cryptosporidium parvum and to establish its association with the nutritional status in schoolchildren of northwestern Mexico. Methodology: A total of 405 schoolchildren between 6 and 13 years of age were included in this study. Weight-forage (W/A), height-forage (H/A) and body mass index-forage (BMI/A) Z scores were calculated. The Faust technique was used to detect intestinal parasites in stool samples and C. parvum coproantigen was detected by enzyme-linked immunosorbent assay (ELISA). Results: The overall prevalence of C. parvum was 28.4%. Some of the study children were C. parvum free (n = 86) and some were C. parvum infected (n = 77). The prevalences of risk of undernutrition found in both groups of children for H/A, W/A, and BMI/A Z scores were 18%, 21% and 28%, respectively. Weight, ZW/A, and ZH/A were significantly higher in the C. parvum-free group compared to the C. parvum-infected group (p ≤ 0.05). The children with cryptosporidiosis were 2.7 times more likely to be at risk of undernutrition by W/A Z score and 2.9 times more likely to be at risk of undernutrition by ZH/A than were the C. parvum-free children. Conclusions: Cryptosporidiosis may be a contributing factor to childhood undernutrition in northwestern Mexico. The proper authorities must implement control and prevention measurements in Mexico and other developing countries.


Introduction
Cryptosporidiosis is a world public health problem, especially affecting immunocompromised persons, such as children and the elderly.More than 20 species of Cryptosporidium are recognized, but 90% of human cryptosporidiosis is attributed to Cryptosporidium parvum and Cryptosporidiun hominis [1].Its transmission has been associated with contaminated drinking water and food, low socioeconomic status and overcrowding conditions [2].C. parvum has a cosmopolitan distribution and its prevalence can range from 0.1% to 31.5% in developing countries [3].Human cryptosporidiosis can be accompanied by abdominal pain, fever, vomiting, malabsorption and diarrhea [4] that may lead to undernutrition, as found in children in West Africa and South America [5] and in Bangladesh [6].Undernutrition is a well-recognized worldwide health problem, and some years ago, it was found to be responsible of 54% of deaths in children in developing countries [7].In 2011, the prevalence of moderate and severe underweight in preschool children in South Asia, West and Central Africa, sub-Saharan Africa, and Latin America and the Caribbean were estimated to be 33%, 23%, 21% and 3%, respectively.In addition, the prevalences of stunting and wasting, respectively, in the same regions were 39% and 16%, 39% and 12%, 40% and 9%, and 12% and 2%, respectively [8].It was estimated that most of global undernutrition was associated with impaired intestinal absorptive function resulting from multiple and repeated enteric infections [9], which is critical in regions where children are mildly nourished [5].In Mexico, it was estimated that, in 2012, 302, 279 (2.8%) children under five years of age were underweight, 1,467,757 (13.6%) were stunted and 171,982 (1.6%) showed emaciation.The prevalence of low height/age was 9%, 11.4% and 19.2% in north, center, and south Mexican regions, respectively [10].However, human cryptosporidiosis remains neglected in Mexico, and some information is provided only in limited publications.In 2006, a study found cryptosporidiosis in 41% of 100 children under one year of age in south Mexico [11], while in 2010, 16% of 100 patients with diarrhea had cryptosporidiosis in north Mexico [12].Based on this, we investigated the prevalence of C. parvum and its association with the nutritional status of schoolchildren in northwestern Mexico.

Study site and population
This was a cross-sectional study conducted from September 2008 to September 2009 in the state of Sonora (northwestern Mexico).Sonora borders the state of Chihuahua to the east, the state of Sinaloa to the south, the Gulf of California to the west, and the American state of Arizona to the north (Figure 1).Ninety-six percent of Sonora is dry and semi-dry.The average summer temperature is 38°C (June-August), and the temperature ranges from 5°C to 30°C from September to January.Four primary schools in the municipality of Hermosillo were selected based on high rates of gastrointestinal infections in the local population [13] and low socioeconomic status [14] defined as a high percentage of parents with no secondary education, and a high number of households with no drainage, no electricity, no safe drinking water and poor quality of construction materials [15] around the selected schools.A total of 883 children officially enrolled in the selected primary schools were invited to participate.Plastic containers were distributed for the collection of stool samples (three per child).Study protocol was explained to the school authorities and parents.A total of 405 schoolchildren (45.8%) agreed to participate in this study, and they were categorized into either the C. parvum-free group (i.e., free of intestinal parasites or showing only non-pathogenic intestinal parasites such as Entamoeba coli, Endolimax nana, I. butschlii, and Chilomastixmesnili), or the C. parvum-infected group (only with C. parvum alone or plus no pathogen intestinal parasites as already previously indicated).

Ethical consideration
Individual informed consent was obtained from the parents or guardians of the participating children.Of the 883 invited children, 478 were unwilling to participate for various reasons.Both participating and non-participating children lived in the same living conditions around the selected schools.The ethics committee of the Centro de Investigación en Alimentación y Desarrollo approved this study.
Children infected with intestinal parasites received the proper antiparasitic treatment by a qualified physician.

Anthropometric measurements
Standing height was measured using a stadiometer (Holtain Ltd., Dyfed, UK) with 2.05 ± 0.0005 m capacity, and weight was measured to the nearest 50 g using a digital electronic scale (AND FV-150 KA1, A&D Co. Ltd., Toshima-ku, Tokyo, Japan) following the standardized recommendations.Ages were validated from reliable official school records.Weight-for-age (W/A), height-for-age (H/A), and body mass index-for age (BMI/A) Z scores were calculated using Anthro Plus versio 1.0.4[16] and Anthro [17] software.Undernutrition risk was defined from -2 to < -1 Z scores and moderate and severe undernutrition from ˂ -2 Z scores, considering the median reference values [7] of H/A (stunting), W/A, and BMI/A.The valid Z scores for analysis in this study were those considered the most frequently occurring in the study population: -5.0 and +5.0 for H/A, -5.0 to +5.0 for W/A, and between -4 and +4 for BMI/A.Five cases for BMI/A, three for H/A, and two for W/A were outside the above-mentioned ranges.

Fecal sample collection, processing, and analysis
Fecal samples were collected and transported to the parasitology laboratory of the Center of Research of Food and Development and were stored between 5°C and 7°C for 24-72 hours until analysis using the Faust technique [18].Covered slides with fecal material in a drop of iodine solution were observed using the 10x and 40x objectives for identification of protozoan cysts and helminth eggs.In addition, 1 g of feces were homogenized, transferred into cryogenic vials (2 mL), properly labeled and stored at -20°C until analysis.Samples were allowed to thaw at room temperature (24°C), and 5 mL of anti-C.parvum solution was added to each vial.Content was homogenized and 200 uL of a second anti-C.parvum solution was added, forming a "sandwich" with the C. parvum fecal antigen bound by the first antibody.The reaction was visualized by adding a second C. parvum antibody bound to a peroxidase conjugate with a chromogenic tetramethylbenzidine against the second C. parvum antibody.The blue color revealed the presence of C. parvum fecal antigen bound to the anti-C.parvum, and reaction was stopped using phosphoric acid 1M.A yellow color was developed and read using a model 680 microplate reader from Bio-Rad Laboratories (Hercules, USA) in an absorbance range between 450 nm and 650 nm.Positive and a negative standard references for quality control were included for each run.A positive result was considered when the reading was ≥ 0.150, in agreement with the manufacturer's instructions.The DRG ELISA kit used had a sensitivity of 93% and a specificity of 98% for the diagnosis of C. parvum fecal antigens (Cryptosporidium Ag Stool, DRG International, Mountainside, USA).

Statistical analysis
Descriptive statistics were used to express the age and anthropometric indicators of the study children as a mean value.The prevalence of cryptosporidiosis was expressed as the percentage of children with C. parvum in any of the fecal samples provided.The proportions were compared using the Chi-square test with the corresponding odds ratios, 95% confidence intervals, and pvalues (prevalence of infection).The association between the nutritional status and cryptosporidiosis was analyzed using multiple logistic regression models.In all models, the dependent variable was the indicator of mild undernutrition or risk for undernutrition from -2 to < -1 Z scores for H/A, W/A and BMI/A [19], and the C. parvum infection (0 to denote absence and 1 presence) was considered as the hypothesized independent variable.Those variables judged to be possible confounding factors such as community, age and sex (0 = female; 1 = male) were used to construct the stepwise models.Data was analyzed with the statistical software STATA/SE version 12.0 with a significance level of p ≤ 0.05.

Results
A total of 198 girls (49%) and 207 boys (51%) (z = -0.632,p = 0.527) were included in this study.The average age of the participants (n = 405) was 7.7 years ± 1.3.Thirty-four percent (n = 138) and 58% (n = 235) of the children provided two and three fecal samples, respectively.The overall prevalence of C. parvum was 24% (n = 97) in this study.From the 405 schoolchildren, 163 were categorized as 86 C. parvum-free (with no intestinal parasites or with only non-pathogen intestinal parasites) and 77 were categorized as C. parvum-infected (only with C. parvum or C. parvum plus non-pathogen intestinal parasites).The rest did not meet these criteria.Eighty percent of the categorized children provided three samples.Due to the low prevalence of data of severe or moderate undernutrition defined by < -2 Z scores for H/A, W/A and BMI/A (3.4%, 2.1% and 2.8% respectively), they were not used in the multiple logistic regression models.Otherwise, the prevalences of risk of undernutrition found in these groups of children for the ZH/A, ZW/A and ZBMI/A were 18%, 21% and 28%, respectively.No difference in nutritional status was found between boys and girls (p> 0.05; data not shown), Also, the weight, ZW/A, and ZH/A were significantly higher in the C. parvumfree group compared to the C. parvum-infected group (Table 1).No difference was found in the age, height, and BMI/A between the groups (Table 1).Approximately 95% of the fathers and 45% of the mothers had formal jobs at the time of interview.They represented their family's economic support.Secondary education was completed by 60% and 57% of the fathers and mothers, respectively.Chi-square analysis showed no significant difference between the C. parvum-free and C. parvum-infected groups in parental employment and education (preparatory school) (p > 0.05), or in the quality of walls (cement and brick) (p > 0.05) and roof (concrete) (p > 0.05) in their homes.Forty-eight percent and 51% of the C. parvum-free and C. parvum-infected children, respectively, resided in households with no drainage (p > 0.05).Although the percentage of treated drinking water (defined as boiled water or water subjected to cartridge or membrane filtration) used and family income tended to be higher in the C. parvum-free group (p = 0.130 and p = 0.090, respectively) and the overcrowding conditions tended to be higher (> 5 people per room in a household) in the C. parvuminfected group (p = 0.090), both were no different between the groups.Because age, sex, type of community, and socioeconomic conditions may influence nutritional status [20], the nutritional statuscryptosporidiosis relationship was determined after adjusting for those all the significant related variables by univariate and stepwise analysis.Table 2 shows the mentioned association adjusted for those variables with a p value ≤ 0.20 (univariate analysis), and also the nutritional status-cryptosporidiosis association adjusted only for the significant variables (stepwise).The children with cryptosporidiosis were 2.7 times more likely to be at risk of undernutrition by ZW/A and 2.9 times more likely to be at risk of undernutrition by ZH/A than the C. parvum-free children (Table 2).No association was found between cryptosporidiosis and risk of undernutrition by ZBMI/A.

Discussion
This study investigated the prevalence of C. parvum and its association with nutritional status in children, between September 2008 and September 2009 from four public elementary schools belonging to the municipality of Hermosillo in northwestern Mexico.The schoolchildren (n = 405) showed a high prevalence of C. parvum (24%).Otherwise, the prevalence of undernutrition by ZH/A in this study was lower (3.4%)than the national prevalence (13.6%) published by the Mexican survey in 2012, which estimated a prevalence of undernutrition for ZW/A similar to this study (2.1% vs. 2.8%).In addition, the Mexican survey also estimated a prevalence of undernutrition for ZH/A of 8.9% in parvum was found to be a risk of undernutrition in the study children.Finally, like undernutrition, cryptosporidiosis is the result of multiple factors, but outbreaks have been highly associated with unsafe drinking water worldwide [23].It is important to mention that in 1994, the American federal government published the Information Collection Rule, which considered Cryptosporidium to be a primary drinking water contaminant [24].In Mexico, there is a law governing public systems for drinking water operators, whose standard includes determining total fecal coliforms and the presence of thermotolerant organisms such as genera Citrobacter and Enterobacter [25].This study suggests that, in addition to monitoring bacteria in drinking water as dictated by Mexican law, analysis of Cryptosporidium should also be considered in determining the magnitude of the problem of this infection in Mexico.

Conclusions
Complementary studies should be developed to determine the sources of Cryptosporidium in the study area and to promote the dissemination of these results at all levels of government.This will provide support, updates, and expansion of health communication efforts among the proper authorities to prevent and control cryptosporidiosis in Mexico.

Figure 1 .
Figure 1.General location of the study site (municipality of Hermosillo) in northwestern México

Table 1 .
Characteristics of the Cryptosporidium-free and Cryptosporidium-infected groups of schoolchildren in northwestern Mexico a Median (25%-75% inter-quantiles); all other data are presented as mean ± standard deviation; b Kruskal-Wallis rank sum test; c Two samples independent t test; d Significance at p ≤ 0.05.

Table 2 .
[22]justed and adjusted association between the risk of undernutrition and cryptosporidiosis in schoolchildren in northwestern Mexico Significance at p ≤ 0.05 northern Mexico but did not provide information about ZBMI/A.On the other hand, the C. parvum-infected schoolchildren showed lower means of ZH/A and ZW/A than did the C. parvum-free children.In 2006, another cross-sectional study in Haitian children 36 months of age found that 49 C. parvum-infected children showed lower (˂ -2) ZW/A and ZH/A than did 41 C. parvum-free children[21].Some years before Molbak et al.[22]found that Cryptosporidium infection was associated with the highest weight loss for children (both boys and girls) who contracted cryptosporidiosis before reaching one year of age.In addition, the proportion of children at risk for undernutrition was relatively high in this study.Although risk of undernutrition is not an immediate life-threatening condition, frequent reinfections are serious health-deteriorating conditions in regions where children are at risk of undernutrition.Although the sample size in this study was not representative of the general population of northwestern Mexico, C.
OR (SE): odds ratio (standard error); CI: confidence interval; a No occupational activity at the time of interview; b No boiled water or water not subjected to cartridge or membrane filtration; c