Dietary and socioeconomic risk factors for fumonisin exposure among women of reproductive age in 18 municipalities in Guatemala from 2013 to 2014

Ariel V. Garsow; Olga R. Torres; Jorge A. Matute; Ronald T. Riley; Julie R. Harris; Archana P. Lamichhane; Orion McCotter; Barbara B. Kowalcyk

doi:10.1371/journal.pgph.0000337

Abstract

Fumonisin exposure is common in populations where maize is a dietary staple, such as in Guatemala, and has been associated with negative health outcomes including neural tube defects. The objective of this study was to estimate fumonisin B₁ (FB₁) exposure among Guatemalan reproductive-age women and develop a better understanding of the dietary and sociodemographic risk factors for exposure. A cross-sectional study in 18 municipalities in Guatemala was conducted. Midwives and study nurses enrolled consenting women and collected individual and household demographic and socioeconomic data. A food frequency questionnaire was administered to estimate quantity and types of food products consumed. A urine sample was collected and urinary fumonisin B₁ (uFB₁) concentration was measured. A univariable analysis was conducted to identify predictors of low/high uFB₁. Multivariable logistic regression was used to calculate adjusted odds ratios (ORs) and 95% confidence intervals (CIs). In total, 775 women had analyzable urine samples. Higher uFB₁ levels were associated with speaking Mayan (OR = 2.33, 95% CI:1.44–3.77), less than high school education (OR = 1.61, 95% CI:1.12–2.30), increasing dietary proportion of maize-based foods (OR = 1.02, 95% CI:1.01–1.03), and consumption of tostadas (fried tortillas) (OR = 1.11, 95% CI:1.02–1.22). Lower uFB₁ levels were associated with consumption of highly processed maize-based foods (OR = 0.93, 95% CI:0.87–0.99). Tortillas were the most frequently consumed maize-based food among study participants and significantly associated with high uFB₁ exposure in the univariable but not multivariable analysis. Consumption of >4,750 grams/week of maize-based foods, >5,184 g/week of locally produced maize-based foods, and >110 servings/week of tortillas were also significantly associated with high uFB₁ exposure in univariable analysis. Populations with low socioeconomic status/education levels and high consumption of maize-based foods had higher fumonisin exposure. Interventions aimed at reducing the risk of exposure to mycotoxins through maize in Guatemala, including the increased consumption of non-maize-based foods, should be further explored.

Citation: Garsow AV, Torres OR, Matute JA, Riley RT, Harris JR, Lamichhane AP, et al. (2022) Dietary and socioeconomic risk factors for fumonisin exposure among women of reproductive age in 18 municipalities in Guatemala from 2013 to 2014. PLOS Glob Public Health 2(8): e0000337. https://doi.org/10.1371/journal.pgph.0000337

Editor: Cemil Kurekci, Mustafa Kemal University: Hatay Mustafa Kemal Universitesi, TURKEY

Received: February 7, 2022; Accepted: June 27, 2022; Published: August 9, 2022

Copyright: © 2022 Garsow et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The raw data and data dictionary from this study are available in S6 Table and S7 Table.

Funding: Centers for Disease Control and Prevention (CDC) provided financial support through PO 200-2010-M-36759 (O.R.T., R.T.R.). Guatemalan Secretaria Nacional de Ciencia y Tecnología (SENACYT) provided financial support through project FODECYT 066-2012 (O.R.T.). Centro de Investigación en Nutrición y Salud (CIENSA) (J.A.M.), Diagnostico Molecular (O.T.R), United States Department of Agriculture (USDA) at Athens (R.T.R), The Ohio State University (B.B.K.), and RTI International provided in-kind salary support (B.B.K.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Fumonisins are mycotoxins produced primarily by the fungus Fusarium verticillioides, a common contaminant of maize, and inhibit ceramide synthases [1–3]. Fumonisins are involved in the pathogenesis of numerous animal diseases [1, 4], are considered possibly carcinogenic in humans [1, 5], and have been associated with negative health outcomes including neural tube defects [4, 6–8]. Exposure to fumonisins occurs through consumption of contaminated food, predominantly maize and maize-based foods [1, 4]. People in low- and middle-income countries (LMICs) are disproportionately and continually exposed to higher quantities of fumonisins than persons in higher-income countries [4, 9], especially where diets comprise mostly maize, and practices and regulations to control fumonisins in the food supply are not easily implemented because most food is self-produced [1, 4]. While more than 20 fumonisin analogs have been identified, fumonisin B₁ (FB₁) is the predominant type in maize infected with F. verticillioides, accounting for approximately 70% of all forms of fumonisins found in maize.

Maize is a principal dietary component in Guatemala [10, 11]. Guatemalans whose diet is limited to maize-based foods are typically rural residents with a low socioeconomic level and less formal education [12]. Many Guatemalans have a diet low in variation, with maize as the primary source of calories [2, 11, 13]. For example, the diet of urban young Guatemalan adults, as described from a cohort study conducted in the Department of El Progreso, consists largely of maize tortillas, beans, sugar added to coffee, and highly processed foods like pizza and hamburgers in larger cities [14].

Given their diets and the agroecological conditions in Guatemala, Guatemalans are at risk of chronic exposure to fumonisins [1, 3, 13]. High levels of fumonisins have been found in maize grown at lower elevations where environmental conditions are conducive to both cultivation of maize, growth of fungi, and subsequent production of fumonisins [10, 11, 13, 15–17]. One study estimated FB₁ intake in Guatemalan women (based on the levels of fumonisins found in the maize purchased from local markets) to be between 0.20–23 μg/ kg of body weight (bw) per day. Two additional studies were undertaken to estimate FB₁ intake among 1,533 maize-consuming Guatemalan women from six communities representing historically high- and low-exposure regions of Guatemala [2, 11]. Results showed that urine from women in the three high-exposure communities had significantly higher levels of uFB₁ (range: non-detected to 61.9 ng/ml) than urine from women in the three low-exposure communities (range: non-detected to 9.95 ng/ml; p <0.05). Urinary fumonisin B₂ and fumonisin B₃ were also detected at levels below that expected based on the level of contamination found in the maize used to prepare maize-based foods [2, 11]. In total, 11% of women in the low-exposure communities and 75% of women in the high-exposure communities had estimated total FB₁ intakes that exceeded 2 μg/kg bw/day [2, 11], which is the provisional maximum tolerable daily intake (PMTDI) recommended by the World Health Organization (WHO) for FB₁, FB₂, and FB₃ alone or in combination [1]. Studies in other high maize-consuming regions, including Mexico and parts of Sub-Saharan Africa, have demonstrated an association between consumption of maize-based foods and uFB₁ levels [18–20]. In 2017, an evaluation by the Joint FAO/WHO Expert Committee on Food Additives (JECFA) concluded that uFB₁ has been validated as a biomarker for FB₁ intake in multiple human studies [1].

In low- and medium-income countries like Guatemala, regulations to control exposure to mycotoxins do not exist or are not enforced, and usually the analytical capacity to monitor the level of fumonisins in maize is not available [4]. Poor agricultural practices and a diet based on tortillas as staple food result in the possibility of high individual exposures, particularly among subsistence farmers and those dependent on maize sold for human consumption at the local markets [21].

Repeated consumption of fumonisin-contaminated maize has been identified as a potential risk factor for neural tube defects and stunting [22]. The objective of this study was to estimate FB₁ exposure among Guatemalan reproductive-age women and to develop a better understanding of the dietary and sociodemographic risk factors for FB₁ exposure among two departments in Guatemala with varying economic levels and high rates of neural tube defects [23]. The long-term goal is to use the identified risk factors to design future studies to better understand the possibility of fumonisins as a contributing factor in human disease in Guatemala. It was hypothesized that consuming a greater amount of maize and maize-based foods, mainly tortillas, and lower socioeconomic status are associated with higher uFB₁ concentrations among Guatemalan women of reproductive age.

Methods

Ethics statement

The study had ethical approval number 9–13 of the National IRB of the Ministry of Health of Guatemala, and of INCAP Ethics Institutional Committee (CIE by Spanish Acronym) CIE-REV-030-2013. INCAP’s CIE has Federal Wide Assurance and acted as local IRB for Centers for Disease Control and Prevention (CDC); the field workers were trained to comply with all of both IRB’s recommendations. The researchers met all the requirements of both committees. Formal written informed consent or a fingerprint demonstrating consent was obtained by the midwives from the prospective participants. Study participants received a refreshment (a muffin) and Q50 (approximately $6.67 USD) for transportation expenses. If an individual was under 18 years of age, formal written informed consent or a fingerprint demonstrating consent was obtained from a parent or guardian and formal written assent or a fingerprint demonstrating assent was obtained from the individual. Additionally, since midwives were volunteers in the community, written informed consent was obtained from midwives as requested by the IRB at the Institute of Nutrition of Central America and Panama (INCAP).

The target population for this cross-sectional study was reproductive age women (aged 15–49 years) in two of the 22 departments (first-level geopolitical administrative areas in Guatemala) of the Republic of Guatemala, the department of Guatemala and the department of Alta Verapaz. Both departments comprise 17 municipalities (second-level geopolitical administrative areas in Guatemala). Women were recruited from all 17 municipalities in the department of Guatemala and, Cobán, from the department of Alta Verapaz, for a total of 18 study sites.

Women were recruited from March 2013 to January 2014 through a network of 246 midwives, two study nurses, and five field workers. Midwives are registered with the public health services of the county and provide healthcare services for persons who cannot afford private health services. As volunteers of the community health areas and respected members of the community, the midwives identified potential study participants and were paid Q50 to cover transportation expenses. Demographic data were collected from each study participant by hired data collectors, including municipality of residence and maternal age. Due to prior victimization and potential threats of violence in the areas that the survey was conducted [24], it was decided that asking subjects questions about household income could discourage subjects from participating in the survey. As such, indicators of socioeconomic status were collected from each study participant as a proxy for household income, including number of persons living in the household (1 to 7 or 8 to 18), educational level (above or below high school), and paid employment (housewife or paid worker) [25–27]. Low literacy rates are common among Guatemalan women; hence, face-to-face interviews were conducted by trained nurses/field workers who speak Spanish or Spanish-Q‘eqchi (Mayan dialect of Cobán). When a participant spoke another dialect, the interview was conducted with the help of local translators.

Women were also asked about their consumption of 61 food items representing 19 food groups (S1 Table) in the previous week using a food frequency questionnaire (FFQ) used extensively in similar studies [2, 11, 15, 28]. Data was anonymized at the time of collection.

Food groups were determined using the food composition table published by the INCAP. Units of consumption were reported to estimate the grams ingested (e.g.: one tortilla, one glass, one cup, one tablespoon, one piece of fruit, one tamal, one bag of tortrix, etc.) with each unit having a known average weight (S1 Table) [17]. Maize-based foods were categorized as locally produced (not highly processed), industrially produced highly processed (most likely do not contain fumonisins), or micronutrient-fortified (potentially contains fumonisins) to explore the role of processing conditions on fumonisin exposure. For example, industrially produced foods, such as those based on cornstarch (like atol de maicena), are subjected to extrusion during processing, and therefore have a low probability of containing fumonisins (S1 Table) [29]. The number and total quantity of reported foods consumed, the number and amount of reported maize-based foods consumed, and the percentage of total reported food consumption that was maize-based were estimated [17].

Individual exposure to FB₁ was estimated using uFB₁ (ng/ml) levels. A spot urine sample was collected from each woman and shipped frozen to the Laboratorio Diagnóstico Molecular in Guatemala City, where uFB₁ was extracted and sent to USDA’s Russell Research Center in Athens, Georgia to be analyzed following published protocols [2, 28, 29]. Briefly, urine samples (9 ml) were adjusted to 10% acetonitrile using acetonitrile containing 1% formic acid. A total of 40 ng of U-[¹³C₃₄]-FB₁ (Sigma-Aldrich Corp., St. Louis, MO, USA) was added to every urine sample as an internal standard to allow for quantification of uFB₁ levels in samples after extraction. Fumonisins were isolated on C₁₈ solid phase extraction (SPE) cartridges (Sep-Pak R Classic C18 cartridges, Waters Corporation, Milford MA, USA). The loaded solid phase extraction cartridges were shipped to and eluted at USDA’s Russell Research Center in Athens, Georgia as described in Riley et al. [29] using 2 ml of 70% acetonitrile:30% water made to 0.1% formic acid. Separation of the eluates from the SPE cartridges was accomplished by reverse phase HPLC analysis conducted using a Finnigan Micro AS auto sampler coupled to a Surveyor MS pump (ThermoFisher, Woodstock, GA, USA). Separation was accomplished using an Imtakt Cadenza CW-C18 3μ particle size, 150 mm x 2 mm column (Imtakt USA, Philadelphia, PA, USA). The column effluent was directly coupled to a Finnigan LTQ-XL linear ion trap mass spectrometer (ThermoFisher, Woodstock, GA, USA) operated in the electrospray ionization positive ion mode. A complete set of pure standards (FB₁, U-[¹³C₃₄]-FB₁, FB₂, and FB₃ at 1, 10, and 100 pg/μL) was run daily. Additionally, a 10 pg/μL standard was run at the beginning and end of each set of samples. Quantification of FB₁ was based on internal standardization by the comparison of the areas under the chromatographic peaks for FB₁ to the area of the known amount of U-[¹³C₃₄]-FB₁ added to the samples before processing. Results were expressed as ng FB₁/ml of urine and the limit of detection was 0.07 ng/ml [29]. Urine samples were only analyzed for uFB₁ levels. Sample analysis was conducted between 2013–2014.

Descriptive statistics were calculated for demographic, socioeconomic, dietary variables overall, by municipality, and by FB₁ exposure group. Since distributions were right-skewed, nonparametric tests were used to assess differences between groups. Differences in distributions of categorical variables (municipality, language, number of persons living in the household, occupation, age group) were evaluated using chi-square tests. Differences in means of continuous variables (maternal age, uFB₁ level, total grams and servings consumed, total grams and servings by food type, percent of total food consumed that is maize based) were evaluated using the Kruskal-Wallis test.

Univariable logistic regression was used to assess the relationship between FB₁ exposure group and sociodemographics; total number of reported food types consumed; reported consumption of each food type; the reported number of servings of food type consumed; the reported amount of total food consumed in grams; the reported number of maize-based food items consumed; the reported amount of maize-based food consumed in grams; and the reported percentage of food consumed that was maize based. Odds ratios (ORs) and 95% confidence intervals (CIs) for food consumption variables reflect the change in odds of FB₁ exposure with a one serving increase in food intake. Variables significant at α = 0.20 were then included in the multivariable analysis.

Multivariable logistic regression was used to calculate adjusted ORs and 95% CIs for comparing none/low and high FB₁ exposure groups. The multivariable model was fit using stepwise selection and considered the following variables: sociodemographics; total number of reported food types consumed; reported consumption of each food type; the reported number of servings of food type consumed; the reported amount of total food consumed in grams; the reported number of maize-based food items consumed; the reported amount of maize-based food consumed in grams; and the reported percentage of food consumed that was maize based. All possible interactions between main effects were tested but none were significant and, thus, were excluded from the final model.

Sensitivity analyses were conducted to evaluate the impact of imputing missing food consumption data and the cut-off value for assigning high and low FB₁-exposure groups. Imputing missing food consumption data as zero and the median amount consumed produced similar results (not presented).

To further explore the impact of reported consumption of maize-based foods on exposure, the analysis was re-run for high and low consumers of maize-based food groups and individual foods. For each food group and food item, the cut-off for high- and low-exposure groups was defined to be the mid-point of the range for the specific food group or item. Study participants were categorized as high or low consumers and the univariable analysis was repeated for each food group or item. These methods are similar to those conducted in previous research by members of the research team [15]. In brief, 1,100 randomly selected adults from the 22 departments of Guatemala (50 per department) provided information on the amounts of maize-based foods eaten during the previous 7 days. Those in the 60^th percentile of the distribution or higher were identified as high consumers of maize-based foods and consumed 4,750 grams/week. Those below the 60^th percentile were considered low consumers of maize-based foods [30].

Data were recorded by double entry and validation using Epi-Info 6.04d, and all statistical analyses were performed using SAS v.9.4 for Windows (SAS Institute Inc., USA). Ethical approval was obtained from Comité Nacional de Etica from the Ministry of Health of Guatemala, the Hospital General San Juan de Dios, Hospital Roosevelt, Area de Salud de Alta Verapaz, and the United States Centers for Disease Control and Prevention, through the Institutional Review Board (IRB) of INCAP. A final report of the project that included a description of the JECFA PMTDI levels was shared with local health authorities in the departments of Guatemala and Alta Verapaz. Additionally, several workshops about study findings, fumonisins in general and the risk of chronic exposure were conducted for health authorities, agricultural sector authorities, industry delegates, and the press. Additional information regarding the ethical, cultural, and scientific considerations specific to inclusivity in global research is included in the Supporting Information (S1 Checklist).

Results

A total of 802 reproductive-aged women were recruited to participate in the study. Of these, 27 (3.37%) women had spot urine samples that were unanalyzable and were excluded from the analysis. Most of the participants spoke Spanish; consumed ≤4,750 g of maize-based food/week (n = 670, 86.45%); had uFB₁ levels <0.5 ng/ml (532, 68.65%); were younger than 25 years old; had less than a high school education; had 2–7 persons living in their households; and were housewives (Table 1, S3 Table).

Download:

Table 1. Sociodemographic characteristics of study participants.

https://doi.org/10.1371/journal.pgph.0000337.t001

Overall, study participants consumed a median of 6.65 kg of food in the week prior to study enrollment (Table 2). Locally produced maize-based foods constituted 39.06% of all foods reported being consumed during the previous seven days and tortillas constituted 81.33% of all locally produced maize-based foods consumed (Fig 1).

Download:

Fig 1.

A Distribution of food groups, and B, highly processed and locally produced maize-based foods consumed in the week prior to study enrollment.

https://doi.org/10.1371/journal.pgph.0000337.g001

Download:

Table 2. Total estimated weekly food consumption (kg), mean, median and range of number of servings consumed for selected food groups and maize-based food items per week overall and by uFB₁ level (ng/ml)^¹*.

https://doi.org/10.1371/journal.pgph.0000337.t002

Of the 775 women included in the analysis, 398 (51.35%) had detectable uFB₁ levels and 243 (31.35%) had levels that exceeded 0.5 ng/ml and were classified as having high exposure (Table 2). Since levels of uFB₁ were not expected to be uniform across the population, FB₁ exposure was categorized as a dichotomous variable: low/no exposure (uFB₁ < 0.5 ng/ml) and high exposure (uFB₁ ≥ 0.5 ng/ml). The cut-off point of 0.5 ng/ml for high and low exposure was chosen because, in previous studies [2, 11, 28], a uFB₁ level of 0.5 ng/ml was estimated to be approximately equal to the JECFA PMTDI of 2 μg total FB/kg bw/day [1]. Specifically, an uFB₁ concentration of 0.5 ng/ml is estimated to reflect an FB₁ intake of 1.67 μg/kg bw/day [1, 2], assuming that the average woman weighs 60 kg, has a daily urine output of 1,000ml, and 0.5% of total FB₁ intake is excreted [28]. Assuming FB₁ comprises 70% of total FB (fumonisin) exposure, this suggests 0.5 ng/ml of FB₁ in the urine is likely to indicate a total FB intake >2ug/kg bw/day, which is the current JECFA PMTDI above which there is a higher risk for adverse health outcomes associated with consumption of fumonisins [1].The overall mean uFB₁ level was 0.72 ± 2.03 ng/ml, with the low-exposure group having a mean of 0.08 ± 0.15 ng/ml and the high-exposure group having a mean of 2.13 ± 3.20 ng/ml (Table 2, Fig 2A and 2B). Cobán (64.44%) and Chuarrancho (60.00%) were the two municipalities with the highest percentages of high-exposure women, while Mixco (12.50%) and Santa Catarina Pinula (18.18%) had the lowest percentages (Fig 2, S2 Table).

Download:

Fig 2.

A Geographic distribution of median uFB₁ groups in sampled municipalities. Grey shaded areas indicate research sites with scaled grey representing the relative uFB₁ groups. The darkest area is the municipality of Cobán in the department of Alta Verapaz; all other grey shaded areas are municipalities in the department of Guatemala and white areas were not sampled in this study. p-value calculated using chi-square test. B Percentage of the study participants classified as low (<0.5 ng/ml) or high (≥0.5 ng/ml) uFB₁ groups by municipality.

https://doi.org/10.1371/journal.pgph.0000337.g002

In the univariate analysis, women with high uFB₁ levels were more likely to speak Mayan; be less educated; have more persons living in the household; and be 30–34 years of age compared to reproductive-aged women with no/low uFB₁ levels (Fig 3A and 3B). Women in the high-exposure group consumed significantly more maize-based foods than women in the low-exposure group (Fig 3B, S3–S5 Tables). Additionally, the low-exposure group ate a median of 49 tortillas per week while the high-exposure group ate a median of 63 tortillas per week (Table 2). The percent of total foods consumed that were maize-based and consumption of locally-produced maize-based foods, tortillas, and tayuyos (thick tortillas filled with beans) were significantly associated with high exposure (Fig 4). Reported consumption of highly processed maize-based foods, corn flakes, French bread, lettuce, dairy, powdered milk, the fats and oils food group, and oil as a food item were significantly associated with low exposure. Reported consumption of chard was marginally associated with high exposure while reported consumption of beef, whole milk, atol de maicena (a thick beverage made of starch), beef, and papaya were marginally associated with low exposure. Significant differences were also observed for language spoken (Spanish versus Mayan), reported maize consumption (≤ or > 4,750 g/week), uFB₁ levels (< or ≥ 0.5 g/ml), age group, education (high school and above or less than high school), occupation (housewife or paid worker), and number of persons in household (1 to 7 or 8 to 18) when comparisons were made between municipalities using a Kruskal-Wallis test (S2 Table).

Download:

Fig 3.

A Sociodemographic characteristics by low (<0.5 ng/ml) or high (≥0.5 ng/ml) uFB₁ groups. B Univariable results for sociodemographic characteristics, including odds ratios (OR), upper (UCL) and lower (LCL) 95% confidence levels, and p-values from logistic regression.

https://doi.org/10.1371/journal.pgph.0000337.g003

Download:

Fig 4. Dietary factors associated with uFB₁ exposure group in univariable analysis.

Unit is servings unless otherwise noted. Odds ratios (OR), upper (UCL) and lower (LCL) 95% confidence levels, and p-values from logistic regression.

https://doi.org/10.1371/journal.pgph.0000337.g004

In the multivariable analysis, the percent of total food consumed that was maize-based, language spoken, education level, reported consumption of tostadas, and reported consumption of highly processed maize-based foods were significantly associated with exposure group (Fig 5). For every 1% increase in percentage of total food consumed that was maize-based, the odds of having high uFB₁ levels (>0.5 ng/ml) increased by 1.03 (95% CI: 1.02–1.04). Those who spoke a Mayan language had increased odds of having high uFB₁ levels compared to those who spoke Spanish. Women with less than a high school education had significantly higher odds of having high uFB₁ levels than those with more than a high school education. One serving increases in tostadas, which are fried tortillas, were significantly associated with increased odds of high uFB₁ levels. One serving increases in highly processed maize-based foods were significantly associated with decreased odds of high uFB₁ levels. The number of persons living in a household, reported consumption of tayuyos or chard were marginally associated with high exposure, while reported consumption of powdered milk or tomatoes were marginally associated with low exposure. The final multivariable logistic regression model had a significant fit (p < 0.001) and did not violate the Hosmer-Lemeshow goodness of fit test (p > 0.1008).

Download:

Fig 5. Significant demographic and dietary factors significantly associated with uFB₁ exposure group in multivariable analysis (p<0.1).

Odds ratios (OR), upper (UCL) and lower (LCL) 95% confidence levels, and p-values from logistic regression.

https://doi.org/10.1371/journal.pgph.0000337.g005

In the categorical analysis, high reported consumption of maize-based foods, locally-produced maize-based foods, and tortillas were significantly associated with high exposure to uFB₁ (Fig 6). Reported consumption of more than 110 servings/week of tortillas increased the odds of having high exposure to uFB₁ by 1.84 (95% CI: 1.17–2.84).

Download:

Fig 6.

A Significant maize-based food groups and foods associated with low (<0.5 ng/ml) or high (≥0.5 ng/ml) uFB₁ groups in categorical analysis. (B) Odds ratios (OR), upper (UCL) and lower (LCL) 95% confidence levels, and p-values from logistic regression.

https://doi.org/10.1371/journal.pgph.0000337.g006

The study was designed to estimate the prevalence of FB₁ exposure while maximizing geographic diversity, but no formal sample size calculations were performed; rather, the sample size was driven by budgetary and laboratory constraints. A retrospective power analysis showed that a sample size of 775 provided >99% power to detect an OR of 1.09 at a significance level of 0.05, assuming a response rate of 31.35% for high FB₁ exposure.

Discussion

Nearly a third of all women and two-thirds of women in Cobán had uFB₁ levels that exceeded 0.5 ng/ml. This level of uFB₁ has been estimated to indicate a total FB₁ intake that approximates or exceeds the JECFA PMTDI of 2 μg/kg bw/day. The JECFA PMTDI of 2 μg/kg bw/day was derived using dose-response animal toxicity studies (incorporating an uncertainty factor of 100 for intra- and interspecies variation) [1]. Elevation of sphingolipid metabolites in blood spots from Guatemalan women, indicative of FB₁ inhibition of ceramide synthase, have been shown to first occur in the window of uFB₁ concentration that was greater than 0.5 but less than 1.0 ng/ml [2]. The evidence suggests that fumonisin inhibition of ceramide synthase is the proximate cause of animal diseases associated with exposure to fumonisins [1, 2, 31]. These data (along with previous studies) provide an estimate of exposure of FB₁ in women in Guatemala which can be used to better understand the possible adverse health effects of fumonisin on the population.

Reported consumption of maize-based foods was significantly associated with higher uFB₁ levels. Tortillas comprise more than 80% of the Guatemalan diet and are likely one of the principal sources of fumonisin exposure in Guatemala [10, 11, 15, 19]. Reported consumption of maize-based foods was significantly associated with higher fumonisin exposure based on uFB₁ levels. As suspected, processing seemed to reduce exposure. For example, reported consumption of locally processed maize-based foods, such as tortillas and other tortilla-based dishes (tostadas and tayuyos) was associated with higher FB₁ exposure, while reported consumption of highly processed foods that are unlikely to contain fumonisins, such as corn flakes, was associated with lower exposure levels. Many highly processed maize-based foods are prepared with nixtamalized maize. Nixtamalization has been found to reduce some forms of fumonisins in the original maize [32]. However, if fumonisin levels are high in maize, the reduction of fumonisins that occurs through nixtamalization may still result in high levels of contamination in the nixtamalized maize [32, 33]. Due to the widespread consumption of tortillas, the ability to detect differences in the quality of maize used to produce tortillas may have been confounded. The marginal association of higher uFB₁ exposure in both the univariable and multivariable analysis with reported chard consumption demonstrates this confounding as chard is almost exclusively eaten with tortillas in Guatemala. Additional research is needed to further elucidate the role of tortilla consumption in mycotoxin exposure.

These results support the conclusion that communities with high uFB₁ are also likely to have high contamination in the maize sold for preparing maize-based food for human consumption. Two studies in Guatemala have shown that the calculated fumonisin intake based on either the uFB₁ levels in spot urine samples or the total fumonisin in the maize collected at the same time from local markets were correlated [2, 11]. Communities with high uFB₁ levels (Jutiapa, Chiquimula, and Santa Rosa) had higher contamination in the maize sold for preparing maize-based food for human consumption compared to communities with low uFB₁ levels (Chimaltenango, Escuintla, Sacatepéquez). This finding is also supported by the results of a study of the kinetics of urinary FB₁ excretion in humans consuming maize-based diets where uFB₁ increased while consuming the FB-contaminated diets and quickly became non-detectable when consumption of the FB_-contaminated diets ceased [28].

Results from this study suggest that reproductive-aged women in Guatemala consume substantial amounts of maize-based foods and increased reported consumption of these foods was associated with higher uFB₁ levels. Additionally, consumption of multiple types of foods—such as lettuce, dairy, powdered milk, bread, fats and oils—were associated with lower uFB₁ levels. In a randomized controlled trial conducted in 2019 that examined the variation in diet of 240 Guatemalan pregnant women, results indicated that 50% of the participants had diets that included consumption of staple foods (tortillas, bread, and rice) as well as meat, dairy (cheese), eggs, and pulses [34] Multiple studies have shown low diet variation and high maize-based food consumption in Guatemala [10, 11, 21]. Mendoza et al. [21] reported that nearly half of those interviewed in the department of Huehuetenango consumed more than 600 g of maize-derived foods/day, with some individuals ingesting as much as 3,000 g/day demonstrating the low diet variation of these populations [34].

Factors associated with poverty and indicative of lower socioeconomic status (i.e., education level, language spoken, number of persons in household, occupation) were significantly associated with high exposure to FB₁. Specifically, higher uFB₁ levels were observed in Mayan speakers, women living in crowded households, and women of lower education and lower socioeconomic levels. Lower socioeconomic status has been associated with higher reported maize consumption, a less diverse diet, and a potential for increased mycotoxin exposure [4]. The connection between FB₁ exposure and poverty, lack of access to a variety of foods, and culture of high maize consumption need to be further understood to determine their relation to high levels of adverse health outcomes (such as stunting) on a population level in Guatemala. Additionally, research on socially and culturally appropriate interventions that reduce mycotoxin exposure to individuals by reducing maize contamination and increasing consumption of foods that are less likely to be contaminated are needed.

This study expands the knowledge of mycotoxin exposure in Guatemala and further demonstrates the utility of uFB₁ as a biomarker to determine and monitor fumonisin intake. Previous research has demonstrated exposure to other mycotoxins (such as aflatoxins) through consumption of maize-based products. High levels of aflatoxin-albumin adducts in blood samples have been associated with high consumption of tortillas in Guatemala [35]. Additionally, in a cross-sectional study conducted in 2016, aflatoxin-lysine adducts (AFB₁-Lys) were measured in the blood of 100% of study participants (443 adults ≥ 40 years old) living in five departments of Guatemala where the Institute of Cancer in Guatemala (INCAN) detected high rates of liver cancer with low prevalence of viral hepatitis antibodies [36]. Co-exposure to mycotoxins, like aflatoxin B₁ (AFB₁) and FB₁, is a concern because FB₁ is a proven tumor promoter of AFB₁ liver carcinogenicity in animal models. The incidence of chronic liver disease and stunting are high where co-exposure to both mycotoxins are high [1]. As a result, the concluding recommendation of the 83^rd JECFA included reduction of exposure to these mycotoxins in LMICs [1]. Additional studies that estimate mycotoxin (e.g., fumonisin, aflatoxin) levels in maize-based foods throughout Guatemala can provide better insight into relationships between dietary consumption and mycotoxin exposure. Further studies should also be conducted to explore the association between mycotoxin exposure and adverse birth outcomes (e.g., low birth weight, neural tube defects, growth retardation, liver and renal disease) especially in the populations consuming food products highly contaminated with mycotoxins. The findings of this study can be used to inform the design of these studies and develop interventions that can effectively reduce exposure in countries that consume high amounts of foods that are likely to be contaminated, like Guatemala.

This study has several limitations that should be considered when interpreting the results. First, the FFQ asked about food consumption in the previous week, which may not be representative of the woman’s overall diet and could potentially bias results. FFQs estimate an individual’s usual food intake over a defined period, which is easier and more economical to conduct than dietary 24-hour recalls [2, 11, 37]. However, FFQs rely on reported consumption by the respondent, which introduces recall bias. Although, studies have shown that dietary patterns do not change much over time with a consistently high consumption of maize-based foods [10, 11, 28]. Further, the FFQ used in this study was developed in 2010 which could also introduce bias. However, studies have shown that dietary patterns have not changed since 2010 [12, 38, 39]. Calculating a dietary diversity score and total caloric intake was beyond the scope of this manuscript [40]. Due to the high consumption of maize-based foods in this population [10, 11, 28], the FFQ included the consumption of maize-based food items. Additionally, variables representing total servings consumed of 19 food groups and percent of maize-based foods were evaluated for inclusion in the multivariate model indirectly demonstrating dietary diversity and total energy intake in this population. Second, the study focused on only two regions within Guatemala, which weakens the extrapolation of the results to all reproductive-aged women in Guatemala. Third, household maize samples were not collected in this study, which would have provided supporting evidence of exposure estimates based on uFB₁. Knowledge of both the amounts of maize-based foods consumed and the levels of FB₁ contamination of the food are necessary to predict the levels of FB₁ intake. In the absence of analysis of FB₁ in the foods consumed, uFB₁ serves as a biomarker for recent (24 to 48 h) FB₁ intake [1, 2]. Finally, this study examined only one type of mycotoxins, fumonisins, even though maize can be contaminated with other mycotoxins such as aflatoxins. As noted previously, co-exposure to fumonisins and aflatoxins could significantly impact human health and should be the focus of future research.

This is the first cross-sectional study to explore the association between reported food consumption and FB₁ exposure among Guatemalan women of reproductive age. Reproductive-aged women in Guatemala have uFB₁ levels that indicate an exposure to FB₁ that frequently exceeds the JECFA PMTDI. Results from this study suggest that there is an association between low socioeconomic status and/or education and high levels of reported consumption of maize-based foods (mainly tortillas) with FB₁ exposure. Results of this study contribute to the body of evidence that consumption of locally processed maize products (e.g., home-cooked tortillas) is associated with higher exposure to mycotoxins.

Supporting information

S1 Checklist. Ethical, cultural, and scientific considerations specific to inclusivity in global research.

https://doi.org/10.1371/journal.pgph.0000337.s001

(DOCX)

S1 Table. Food groups and serving size (g) for associated food items.

https://doi.org/10.1371/journal.pgph.0000337.s002

(DOCX)

S2 Table. Maize consumption, uFB₁ levels, and socio-demographic characteristics by municipality.

https://doi.org/10.1371/journal.pgph.0000337.s003

(DOCX)

S3 Table. Mean and median intakes (servings) of food items by FB₁ exposure group.

https://doi.org/10.1371/journal.pgph.0000337.s004

(DOCX)

S4 Table. Consumption of food groups by department.

https://doi.org/10.1371/journal.pgph.0000337.s005

(DOCX)

S5 Table. Mean and median consumption of food groups by municipality.

https://doi.org/10.1371/journal.pgph.0000337.s006

(DOCX)

S6 Table. Data dictionary for raw data for article including variable names, description, and units.

https://doi.org/10.1371/journal.pgph.0000337.s007

(CSV)

S7 Table. Raw data for article.

https://doi.org/10.1371/journal.pgph.0000337.s008

(CSV)

Acknowledgments

The authors thank the women who participated in this study, Dr. Graciela Mannucci, Iris Muñoz (National Spina Bifida Unit at Hospital General San Juan de Dios), and Dr. Monica Roy (epidemiologist at CDC) for collecting information and preparing introductory data, Sharito Barrera, Juana Itzept, Hector Gamero, and the midwives of Alta Vera Paz and the 17 municipalities of Guatemala for the fieldwork, Marta Maria Mendez and Cecilia Diaz de Mayorga for the uFB₁ extraction, Ana Ruth Méndez for the laboratory logistics, Jency Showker (USDA/ARS) for the uFB₁ analyses, and Dr. Arun Balajee for helping with coordinating cross-country collaboration.

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

References

1. World Health Organization & Joint FAO/WHO Expert Committee on Food Additives. Evaluation of certain contaminants in food: eighty-third report of the Joint FAO/WHO Expert Committee on Food Additives. Geneva, Switzerland: World Health Organization; 2017. 182 p.
2. Riley RT, Torres O, Matute J, Gregory SG, Ashley-Koch AE, Showker JL, et al. Evidence for fumonisin inhibition of ceramide synthase in humans consuming maize-based foods and living in high exposure communities in Guatemala. Mol Nutr Food Res. 2015 Nov;59(11):2209–24. pmid:26264677
- View Article
- PubMed/NCBI
- Google Scholar
3. Voss K, Riley RT. Fumoninsin Toxicity and Mechanism of Action: Overview and Current Perspectives. Food Safety. 2013;1(1).
- View Article
- Google Scholar
4. International Agency for Research on Cancer (IARC) Working Group Report. In: Wild CP, Miller JD, Groopman JD, editors. Mycotoxin Control in Low-and Middle-income Countries. Lyon (FR): International Agency for Research on Cancer; 2016. p. 55. (IARC Working Group Report; vol. 9).
5. International Agency for Research on Cancer, editor. IARC Working Group on the Evaluation of Carcinogenic Risk to Humans. Some Traditional Herbal Medicines, Some Mycotoxins, Naphthalene and Styrene. [Internet]. Lyon; 2002. (IARC monographs on the evaluation of carcinogenic risks to humans; vol. 82). Available from: https://www.ncbi.nlm.nih.gov/books/NBK326619/ pmid:12687954
- View Article
- PubMed/NCBI
- Google Scholar
6. Chen C, Mitchell NJ, Gratz J, Houpt ER, Gong Y, Egner PA, et al. Exposure to aflatoxin and fumonisin in children at risk for growth impairment in rural Tanzania. Environment International. 2018 Jun;115:29–37. pmid:29544138
- View Article
- PubMed/NCBI
- Google Scholar
7. Chen C, Riley RT, Wu F. Dietary Fumonisin and Growth Impairment in Children and Animals: A Review. Comprehensive Reviews in Food Science and Food Safety. 2018 Nov;17(6):1448–64. pmid:33350142
- View Article
- PubMed/NCBI
- Google Scholar
8. Missmer SA, Suarez L, Felkner M, Wang E, Merrill AH, Rothman KJ, et al. Exposure to Fumonisins and the Occurrence of Neural Tube Defects along the Texas–Mexico Border. Environmental Health Perspectives. 2006 Feb;114(2):237–41. pmid:16451860
- View Article
- PubMed/NCBI
- Google Scholar
9. Wild CP, Gong YY. Mycotoxins and human disease: a largely ignored global health issue. Carcinogenesis [Internet]. 2010 Jan [cited 2018 Dec 3];31(1):71–82. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2802673/ pmid:19875698
- View Article
- PubMed/NCBI
- Google Scholar
10. Torres OA, Palencia E, de Pratdesaba LL, Grajeda R, Fuentes M, Speer MC, et al. Estimated Fumonisin Exposure in Guatemala Is Greatest in Consumers of Lowland Maize. The Journal of Nutrition. 2007 Dec 1;137(12):2723–9. pmid:18029490
- View Article
- PubMed/NCBI
- Google Scholar
11. Torres O, Matute J, Gelineau-van Waes J, Maddox JR, Gregory SG, Ashley-Koch AE, et al. Urinary fumonisin B ₁ and estimated fumonisin intake in women from high- and low-exposure communities in Guatemala. Mol Nutr Food Res. 2014 May;58(5):973–83. pmid:24375966
- View Article
- PubMed/NCBI
- Google Scholar
12. Mayén AL, Stringhini S, Ford ND, Martorell R, Stein AD, Paccaud F, et al. Socioeconomic predictors of dietary patterns among Guatemalan adults. Int J Public Health. 2016 Dec;61(9):1069–77. pmid:27421467
- View Article
- PubMed/NCBI
- Google Scholar
13. Soto-Méndez MJ, Campos R, Hernández L, Orozco M, Vossenaar M, Solomons NW. Food variety, dietary diversity, and food characteristics among convenience samples of Guatemalan women. Salud pública Méx. 2011 Aug;53(4):288–98. pmid:21986785
- View Article
- PubMed/NCBI
- Google Scholar
14. Ford ND, Jaacks LM, Martorell R, Mehta NK, Perrine CG, Ramirez-Zea M, et al. Dietary patterns and cardio-metabolic risk in a population of Guatemalan young adults. BMC Nutr. 2017 Dec;3(1):68.
- View Article
- Google Scholar
15. Torres O, Matute J, Gelineau-van Waes J, Maddox JR, Gregory SG, Ashley-Koch AE, et al. Human health implications from co-exposure to aflatoxins and fumonisins in maize-based foods in Latin America: Guatemala as a case study. World Mycotoxin Journal. 2015 Jan;8(2):143–59.
- View Article
- Google Scholar
16. Riley RT, Voss KA, Showker JL, Torres O, Matute J, Maddox JR, et al. Development of Biomarkers to Assess Fumonisin Exposure and Birth Defects. In: Binder EM, editor. World Nutrition Forum 2012. Leicestershire, England: Anytime Publishing; 2012. p. 249–56.
17. Instituto de Nutrición de Centro América y Panamá (INCAP). Tabla de Composición de Alimentos de Centroamérica [Internet]. Guatemala: Instituto de Nutrición de Centro América y Panamá (INCAP)/Organización Panamericana de la Salud (OPS); 2012 Feb [cited 2020 Oct 22]. Report No.: 2a. Edición. Available from: http://www.incap.int/mesocaribefoods/dmdocuments/TablaCAlimentos.pdf
18. Gong YY, Torres-Sanchez L, Lopez-Carrillo L, Peng JH, Sutcliffe AE, White KL, et al. Association between Tortilla Consumption and Human Urinary Fumonisin B ₁ Levels in a Mexican Population. Cancer Epidemiology Biomarkers & Prevention. 2008 Mar 4;17(3):688–94.
- View Article
- Google Scholar
19. van der Westhuizen L, Shephard GS, Burger HM, Rheeder JP, Gelderblom WCA, Wild CP, et al. Fumonisin B1 as a Urinary Biomarker of Exposure in a Maize Intervention Study Among South African Subsistence Farmers. Cancer Epidemiology Biomarkers & Prevention. 2011 Mar 1;20(3):483–9.
- View Article
- Google Scholar
20. Turner PC, Snyder JA. Development and Limitations of Exposure Biomarkers to Dietary Contaminants Mycotoxins. Toxins. 2021 Apr 28;13(5):314. pmid:33924868
- View Article
- PubMed/NCBI
- Google Scholar
21. Mendoza JR, Rodas A, Oliva A, Sabillón L, Colmenares A, Clarke J, et al. Safety and Quality Assessment of Smallholder Farmers’ Maize in the Western Highlands of Guatemala. Journal of Food Protection. 2018 May 1;81(5):776–84. pmid:29624105
- View Article
- PubMed/NCBI
- Google Scholar
22. Marasas WFO, Riley RT, Hendricks KA, Stevens VL, Sadler TW, Gelineau-van Waes J, et al. Fumonisins Disrupt Sphingolipid Metabolism, Folate Transport, and Neural Tube Development in Embryo Culture and In Vivo: A Potential Risk Factor for Human Neural Tube Defects among Populations Consuming Fumonisin-Contaminated Maize. The Journal of Nutrition. 2004 Oct 1;134(4):711–6. pmid:15051815
- View Article
- PubMed/NCBI
- Google Scholar
23. García VBJD. Caracterización de pacientes con defectos del tubo neural en el Hospital General San Juan de Dios en el período 2009–2013 [Internet]. [Unidad de Espina Bífida e Hidrocefalia, Hospital General San Juan de Dios]: Universidad Rafael Landívar; 2015 [cited 2021 Aug 13]. Available from: http://recursosbiblio.url.edu.gt/tesiseortiz/2015/09/03/Velveth-Duque.pdf
24. United States Department of State. 2013 Country Reports on Human Rights Practices—Guatemala [Internet]. United States Department of State; 2014 [cited 2021 Aug 13]. Available from: https://www.refworld.org/docid/53284ad514.html
25. Gobierno de la República de Guatemala, Instituto Nacional de Estadística (INE), United Nations Population Fund. Censo Nacional XII de Población y VII de Vivienda 2017–2018 Manual Del Censista. Guatemala: Gobierno de la República de Guatemala; 2018.
26. Blakley T, Hales S, Woodward A. Poverty: assessing the distribution of health risks by socioeconomic position at national and local levels. World Health Organization; 2004. (WHO Environmental Burden of Disease Series). Report No.: 10.
27. Gobierno de la República de Guatemala, Instituto Nacional de Estadística (INE), United Nations Population Fund. Principales Resultados Censo 2018. Instituto Nacional de Estadística (INE) Guatemala; 2019.
28. Riley RT, Torres O, Showker JL, Zitomer NC, Matute J, Voss KA, et al. The kinetics of urinary fumonisin B1 excretion in humans consuming maize-based diets. Mol Nutr Food Res. 2012 Sep;56(9):1445–55. pmid:22815244
- View Article
- PubMed/NCBI
- Google Scholar
29. Riley RT, Torres OA, Palencia E. International shipping of fumonisins from maize extracts on C18 sorbent. Food Addit Contam. 2006 Aug;23(8):826–32. pmid:16807208
- View Article
- PubMed/NCBI
- Google Scholar
30. Matute J. Personal communication. 2021.
- View Article
- Google Scholar
31. Riley RT, Merrill AH. Ceramide synthase inhibition by fumonisins: a perfect storm of perturbed sphingolipid metabolism, signaling, and disease. J Lipid Res. 2019 Jul;60(7):1183–9. pmid:31048407
- View Article
- PubMed/NCBI
- Google Scholar
32. Palencia E, Torres O, Hagler W, Meredith FI, Williams LD, Riley RT. Total Fumonisins Are Reduced in Tortillas Using the Traditional Nixtamalization Method of Mayan Communities. The Journal of Nutrition. 2003 Oct 1;133(10):3200–3. pmid:14519811
- View Article
- PubMed/NCBI
- Google Scholar
33. Monroy-Valle M, Coyoy W, De León J, Flórez ID. Determinantes dietéticos del consumo de zinc en menores de cinco años con retardo de crecimiento en comunidades mayas de Guatemala. Rev Peru Med Exp Salud Publica. 2017 Oct 9;34(3):451. pmid:29267769
- View Article
- PubMed/NCBI
- Google Scholar
34. Lander R, Hambidge K, Westcott J, Tejeda G, Diba T, Mastiholi S, et al. Pregnant Women in Four Low-Middle Income Countries Have a High Prevalence of Inadequate Dietary Intakes That Are Improved by Dietary Diversity. Nutrients. 2019 Jul 10;11(7):1560. pmid:31295916
- View Article
- PubMed/NCBI
- Google Scholar
35. Kroker-Lobos MF, Alvarez CS, Rivera-Andrade A, Smith JW, Egner P, Torres O, et al. Association between aflatoxin-albumin adduct levels and tortilla consumption in Guatemalan adults. Toxicology Reports. 2019;6:465–71. pmid:31193789
- View Article
- PubMed/NCBI
- Google Scholar
36. Smith JW, Kroker-Lobos MF, Lazo M, Rivera-Andrade A, Egner PA, Wedemeyer H, et al. Aflatoxin and viral hepatitis exposures in Guatemala: Molecular biomarkers reveal a unique profile of risk factors in a region of high liver cancer incidence. Chemin I, editor. PLoS ONE. 2017 Dec 13;12(12):e0189255. pmid:29236788
- View Article
- PubMed/NCBI
- Google Scholar
37. Rodríguez M, Méndez H, Torún B, Schroeder D, Stein A. Validation of a semi-quantitative food-frequency questionnaire for use among adults in Guatemala. Public health nutrition. 2002 Nov 1;5:691–9. pmid:12372164
- View Article
- PubMed/NCBI
- Google Scholar
38. Bermudez OI, Tucker KL. Trends in dietary patterns of Latin American populations. Cad Saúde Pública. 2003;19(suppl 1):S87–99. pmid:12886439
- View Article
- PubMed/NCBI
- Google Scholar
39. Bermudez OI, Hernandez L, Mazariegos M, Solomons NW. Secular Trends in Food Patterns of Guatemalan Consumers: New Foods for Old. Food Nutr Bull. 2008 Dec 1;29(4):278–87. pmid:19227052
- View Article
- PubMed/NCBI
- Google Scholar
40. Swindale A, Bilinsky P. Household Dietary Diversity Score (HDDS) for Measurement of Household Food Access: Indicator Guide (v.2). Washington, D.C.: FHI 360/Food and Nutrition Technical Assistance Project (FANTA); 2006. Report No.: 2.

[ref1] 1. World Health Organization & Joint FAO/WHO Expert Committee on Food Additives. Evaluation of certain contaminants in food: eighty-third report of the Joint FAO/WHO Expert Committee on Food Additives. Geneva, Switzerland: World Health Organization; 2017. 182 p.

[ref2] 2. Riley RT, Torres O, Matute J, Gregory SG, Ashley-Koch AE, Showker JL, et al. Evidence for fumonisin inhibition of ceramide synthase in humans consuming maize-based foods and living in high exposure communities in Guatemala. Mol Nutr Food Res. 2015 Nov;59(11):2209–24. pmid:26264677
View Article
PubMed/NCBI
Google Scholar

[3] View Article

[4] PubMed/NCBI

[5] Google Scholar

[ref3] 3. Voss K, Riley RT. Fumoninsin Toxicity and Mechanism of Action: Overview and Current Perspectives. Food Safety. 2013;1(1).
View Article
Google Scholar

[7] View Article

[8] Google Scholar

[ref4] 4. International Agency for Research on Cancer (IARC) Working Group Report. In: Wild CP, Miller JD, Groopman JD, editors. Mycotoxin Control in Low-and Middle-income Countries. Lyon (FR): International Agency for Research on Cancer; 2016. p. 55. (IARC Working Group Report; vol. 9).

[ref5] 5. International Agency for Research on Cancer, editor. IARC Working Group on the Evaluation of Carcinogenic Risk to Humans. Some Traditional Herbal Medicines, Some Mycotoxins, Naphthalene and Styrene. [Internet]. Lyon; 2002. (IARC monographs on the evaluation of carcinogenic risks to humans; vol. 82). Available from: https://www.ncbi.nlm.nih.gov/books/NBK326619/ pmid:12687954
View Article
PubMed/NCBI
Google Scholar

[11] View Article

[12] PubMed/NCBI

[13] Google Scholar

[ref6] 6. Chen C, Mitchell NJ, Gratz J, Houpt ER, Gong Y, Egner PA, et al. Exposure to aflatoxin and fumonisin in children at risk for growth impairment in rural Tanzania. Environment International. 2018 Jun;115:29–37. pmid:29544138
View Article
PubMed/NCBI
Google Scholar

[15] View Article

[16] PubMed/NCBI

[17] Google Scholar

[ref7] 7. Chen C, Riley RT, Wu F. Dietary Fumonisin and Growth Impairment in Children and Animals: A Review. Comprehensive Reviews in Food Science and Food Safety. 2018 Nov;17(6):1448–64. pmid:33350142
View Article
PubMed/NCBI
Google Scholar

[19] View Article

[20] PubMed/NCBI

[21] Google Scholar

[ref8] 8. Missmer SA, Suarez L, Felkner M, Wang E, Merrill AH, Rothman KJ, et al. Exposure to Fumonisins and the Occurrence of Neural Tube Defects along the Texas–Mexico Border. Environmental Health Perspectives. 2006 Feb;114(2):237–41. pmid:16451860
View Article
PubMed/NCBI
Google Scholar

[23] View Article

[24] PubMed/NCBI

[25] Google Scholar

[ref9] 9. Wild CP, Gong YY. Mycotoxins and human disease: a largely ignored global health issue. Carcinogenesis [Internet]. 2010 Jan [cited 2018 Dec 3];31(1):71–82. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2802673/ pmid:19875698
View Article
PubMed/NCBI
Google Scholar

[27] View Article

[28] PubMed/NCBI

[29] Google Scholar

[ref10] 10. Torres OA, Palencia E, de Pratdesaba LL, Grajeda R, Fuentes M, Speer MC, et al. Estimated Fumonisin Exposure in Guatemala Is Greatest in Consumers of Lowland Maize. The Journal of Nutrition. 2007 Dec 1;137(12):2723–9. pmid:18029490
View Article
PubMed/NCBI
Google Scholar

[31] View Article

[32] PubMed/NCBI

[33] Google Scholar

[ref11] 11. Torres O, Matute J, Gelineau-van Waes J, Maddox JR, Gregory SG, Ashley-Koch AE, et al. Urinary fumonisin B ₁ and estimated fumonisin intake in women from high- and low-exposure communities in Guatemala. Mol Nutr Food Res. 2014 May;58(5):973–83. pmid:24375966
View Article
PubMed/NCBI
Google Scholar

[35] View Article

[36] PubMed/NCBI

[37] Google Scholar

[ref12] 12. Mayén AL, Stringhini S, Ford ND, Martorell R, Stein AD, Paccaud F, et al. Socioeconomic predictors of dietary patterns among Guatemalan adults. Int J Public Health. 2016 Dec;61(9):1069–77. pmid:27421467
View Article
PubMed/NCBI
Google Scholar

[39] View Article

[40] PubMed/NCBI

[41] Google Scholar

[ref13] 13. Soto-Méndez MJ, Campos R, Hernández L, Orozco M, Vossenaar M, Solomons NW. Food variety, dietary diversity, and food characteristics among convenience samples of Guatemalan women. Salud pública Méx. 2011 Aug;53(4):288–98. pmid:21986785
View Article
PubMed/NCBI
Google Scholar

[43] View Article

[44] PubMed/NCBI

[45] Google Scholar

[ref14] 14. Ford ND, Jaacks LM, Martorell R, Mehta NK, Perrine CG, Ramirez-Zea M, et al. Dietary patterns and cardio-metabolic risk in a population of Guatemalan young adults. BMC Nutr. 2017 Dec;3(1):68.
View Article
Google Scholar

[47] View Article

[48] Google Scholar

[ref15] 15. Torres O, Matute J, Gelineau-van Waes J, Maddox JR, Gregory SG, Ashley-Koch AE, et al. Human health implications from co-exposure to aflatoxins and fumonisins in maize-based foods in Latin America: Guatemala as a case study. World Mycotoxin Journal. 2015 Jan;8(2):143–59.
View Article
Google Scholar

[50] View Article

[51] Google Scholar

[ref16] 16. Riley RT, Voss KA, Showker JL, Torres O, Matute J, Maddox JR, et al. Development of Biomarkers to Assess Fumonisin Exposure and Birth Defects. In: Binder EM, editor. World Nutrition Forum 2012. Leicestershire, England: Anytime Publishing; 2012. p. 249–56.

[ref17] 17. Instituto de Nutrición de Centro América y Panamá (INCAP). Tabla de Composición de Alimentos de Centroamérica [Internet]. Guatemala: Instituto de Nutrición de Centro América y Panamá (INCAP)/Organización Panamericana de la Salud (OPS); 2012 Feb [cited 2020 Oct 22]. Report No.: 2a. Edición. Available from: http://www.incap.int/mesocaribefoods/dmdocuments/TablaCAlimentos.pdf

[ref18] 18. Gong YY, Torres-Sanchez L, Lopez-Carrillo L, Peng JH, Sutcliffe AE, White KL, et al. Association between Tortilla Consumption and Human Urinary Fumonisin B ₁ Levels in a Mexican Population. Cancer Epidemiology Biomarkers & Prevention. 2008 Mar 4;17(3):688–94.
View Article
Google Scholar

[55] View Article

[56] Google Scholar

[ref19] 19. van der Westhuizen L, Shephard GS, Burger HM, Rheeder JP, Gelderblom WCA, Wild CP, et al. Fumonisin B1 as a Urinary Biomarker of Exposure in a Maize Intervention Study Among South African Subsistence Farmers. Cancer Epidemiology Biomarkers & Prevention. 2011 Mar 1;20(3):483–9.
View Article
Google Scholar

[58] View Article

[59] Google Scholar

[ref20] 20. Turner PC, Snyder JA. Development and Limitations of Exposure Biomarkers to Dietary Contaminants Mycotoxins. Toxins. 2021 Apr 28;13(5):314. pmid:33924868
View Article
PubMed/NCBI
Google Scholar

[61] View Article

[62] PubMed/NCBI

[63] Google Scholar

[ref21] 21. Mendoza JR, Rodas A, Oliva A, Sabillón L, Colmenares A, Clarke J, et al. Safety and Quality Assessment of Smallholder Farmers’ Maize in the Western Highlands of Guatemala. Journal of Food Protection. 2018 May 1;81(5):776–84. pmid:29624105
View Article
PubMed/NCBI
Google Scholar

[65] View Article

[66] PubMed/NCBI

[67] Google Scholar

[ref22] 22. Marasas WFO, Riley RT, Hendricks KA, Stevens VL, Sadler TW, Gelineau-van Waes J, et al. Fumonisins Disrupt Sphingolipid Metabolism, Folate Transport, and Neural Tube Development in Embryo Culture and In Vivo: A Potential Risk Factor for Human Neural Tube Defects among Populations Consuming Fumonisin-Contaminated Maize. The Journal of Nutrition. 2004 Oct 1;134(4):711–6. pmid:15051815
View Article
PubMed/NCBI
Google Scholar

[69] View Article

[70] PubMed/NCBI

[71] Google Scholar

[ref23] 23. García VBJD. Caracterización de pacientes con defectos del tubo neural en el Hospital General San Juan de Dios en el período 2009–2013 [Internet]. [Unidad de Espina Bífida e Hidrocefalia, Hospital General San Juan de Dios]: Universidad Rafael Landívar; 2015 [cited 2021 Aug 13]. Available from: http://recursosbiblio.url.edu.gt/tesiseortiz/2015/09/03/Velveth-Duque.pdf

[ref24] 24. United States Department of State. 2013 Country Reports on Human Rights Practices—Guatemala [Internet]. United States Department of State; 2014 [cited 2021 Aug 13]. Available from: https://www.refworld.org/docid/53284ad514.html

[ref25] 25. Gobierno de la República de Guatemala, Instituto Nacional de Estadística (INE), United Nations Population Fund. Censo Nacional XII de Población y VII de Vivienda 2017–2018 Manual Del Censista. Guatemala: Gobierno de la República de Guatemala; 2018.

[ref26] 26. Blakley T, Hales S, Woodward A. Poverty: assessing the distribution of health risks by socioeconomic position at national and local levels. World Health Organization; 2004. (WHO Environmental Burden of Disease Series). Report No.: 10.

[ref27] 27. Gobierno de la República de Guatemala, Instituto Nacional de Estadística (INE), United Nations Population Fund. Principales Resultados Censo 2018. Instituto Nacional de Estadística (INE) Guatemala; 2019.

[ref28] 28. Riley RT, Torres O, Showker JL, Zitomer NC, Matute J, Voss KA, et al. The kinetics of urinary fumonisin B1 excretion in humans consuming maize-based diets. Mol Nutr Food Res. 2012 Sep;56(9):1445–55. pmid:22815244
View Article
PubMed/NCBI
Google Scholar

[78] View Article

[79] PubMed/NCBI

[80] Google Scholar

[ref29] 29. Riley RT, Torres OA, Palencia E. International shipping of fumonisins from maize extracts on C18 sorbent. Food Addit Contam. 2006 Aug;23(8):826–32. pmid:16807208
View Article
PubMed/NCBI
Google Scholar

[82] View Article

[83] PubMed/NCBI

[84] Google Scholar

[ref30] 30. Matute J. Personal communication. 2021.
View Article
Google Scholar

[86] View Article

[87] Google Scholar

[ref31] 31. Riley RT, Merrill AH. Ceramide synthase inhibition by fumonisins: a perfect storm of perturbed sphingolipid metabolism, signaling, and disease. J Lipid Res. 2019 Jul;60(7):1183–9. pmid:31048407
View Article
PubMed/NCBI
Google Scholar

[89] View Article

[90] PubMed/NCBI

[91] Google Scholar

[ref32] 32. Palencia E, Torres O, Hagler W, Meredith FI, Williams LD, Riley RT. Total Fumonisins Are Reduced in Tortillas Using the Traditional Nixtamalization Method of Mayan Communities. The Journal of Nutrition. 2003 Oct 1;133(10):3200–3. pmid:14519811
View Article
PubMed/NCBI
Google Scholar

[93] View Article

[94] PubMed/NCBI

[95] Google Scholar

[ref33] 33. Monroy-Valle M, Coyoy W, De León J, Flórez ID. Determinantes dietéticos del consumo de zinc en menores de cinco años con retardo de crecimiento en comunidades mayas de Guatemala. Rev Peru Med Exp Salud Publica. 2017 Oct 9;34(3):451. pmid:29267769
View Article
PubMed/NCBI
Google Scholar

[97] View Article

[98] PubMed/NCBI

[99] Google Scholar

[ref34] 34. Lander R, Hambidge K, Westcott J, Tejeda G, Diba T, Mastiholi S, et al. Pregnant Women in Four Low-Middle Income Countries Have a High Prevalence of Inadequate Dietary Intakes That Are Improved by Dietary Diversity. Nutrients. 2019 Jul 10;11(7):1560. pmid:31295916
View Article
PubMed/NCBI
Google Scholar

[101] View Article

[102] PubMed/NCBI

[103] Google Scholar

[ref35] 35. Kroker-Lobos MF, Alvarez CS, Rivera-Andrade A, Smith JW, Egner P, Torres O, et al. Association between aflatoxin-albumin adduct levels and tortilla consumption in Guatemalan adults. Toxicology Reports. 2019;6:465–71. pmid:31193789
View Article
PubMed/NCBI
Google Scholar

[105] View Article

[106] PubMed/NCBI

[107] Google Scholar

[ref36] 36. Smith JW, Kroker-Lobos MF, Lazo M, Rivera-Andrade A, Egner PA, Wedemeyer H, et al. Aflatoxin and viral hepatitis exposures in Guatemala: Molecular biomarkers reveal a unique profile of risk factors in a region of high liver cancer incidence. Chemin I, editor. PLoS ONE. 2017 Dec 13;12(12):e0189255. pmid:29236788
View Article
PubMed/NCBI
Google Scholar

[109] View Article

[110] PubMed/NCBI

[111] Google Scholar

[ref37] 37. Rodríguez M, Méndez H, Torún B, Schroeder D, Stein A. Validation of a semi-quantitative food-frequency questionnaire for use among adults in Guatemala. Public health nutrition. 2002 Nov 1;5:691–9. pmid:12372164
View Article
PubMed/NCBI
Google Scholar

[113] View Article

[114] PubMed/NCBI

[115] Google Scholar

[ref38] 38. Bermudez OI, Tucker KL. Trends in dietary patterns of Latin American populations. Cad Saúde Pública. 2003;19(suppl 1):S87–99. pmid:12886439
View Article
PubMed/NCBI
Google Scholar

[117] View Article

[118] PubMed/NCBI

[119] Google Scholar

[ref39] 39. Bermudez OI, Hernandez L, Mazariegos M, Solomons NW. Secular Trends in Food Patterns of Guatemalan Consumers: New Foods for Old. Food Nutr Bull. 2008 Dec 1;29(4):278–87. pmid:19227052
View Article
PubMed/NCBI
Google Scholar

[121] View Article

[122] PubMed/NCBI

[123] Google Scholar

[ref40] 40. Swindale A, Bilinsky P. Household Dietary Diversity Score (HDDS) for Measurement of Household Food Access: Indicator Guide (v.2). Washington, D.C.: FHI 360/Food and Nutrition Technical Assistance Project (FANTA); 2006. Report No.: 2.

Figures

Abstract

Introduction

Methods

Ethics statement

Results

Discussion

Supporting information

S1 Checklist. Ethical, cultural, and scientific considerations specific to inclusivity in global research.

S1 Table. Food groups and serving size (g) for associated food items.

S2 Table. Maize consumption, uFB1 levels, and socio-demographic characteristics by municipality.

S3 Table. Mean and median intakes (servings) of food items by FB1 exposure group.

S4 Table. Consumption of food groups by department.

S5 Table. Mean and median consumption of food groups by municipality.

S6 Table. Data dictionary for raw data for article including variable names, description, and units.

S7 Table. Raw data for article.

Acknowledgments

References

S2 Table. Maize consumption, uFB₁ levels, and socio-demographic characteristics by municipality.

S3 Table. Mean and median intakes (servings) of food items by FB₁ exposure group.