点击显示 收起
1 From the Jean Mayer US Department of Agriculture Human Nutrition Research Center at Tufts University, Boston (JDR-M, GGD, and RMR); the Center for Studies of Sensory Impairment, Aging and Metabolism, Guatemala City (NWS and JB); and the Universidad Nacional Autónoma de Nicaragua, Managua, Nicaragua (YM and CBW)
2 Field operations and infrastructure were supported by contributions from Micronutrient Initiative (Ottawa, Canada), Task Force Sight & Life (Basel, Switzerland), Roche Interamericana (Sao Paolo, Brazil), and UNICEF (New York and Managua, Nicaragua). The deuterium-labeled retinyl acetate was provided by Task Force Sight & Life, and the laboratory analyses of plasma retinol isotopes and carotenoids were supported by the US Department of Agriculture. 3 Reprints not available. Address correspondence to JD Ribaya-Mercado, Jean Mayer US Department of Agriculture Human Nutrition Research Center, Tufts University, 711 Washington Street, Boston, MA 02111. E-mail: judy.ribaya-mercado{at}tufts.edu.
ABSTRACT
Background: Nicaragua initiated a national program of vitamin A fortification of its domestic sugar supply starting with the 1999-2000 sugarcane harvest.
Objective: This study was conducted to document any change in the vitamin A status of a cohort of children during the first year of the program.
Design: The vitamin A status of 21 Nicaraguan schoolchildren (mean age: 6.7 y; range: 5.3–9.3 y) was assessed in March 2000 and in March 2001. Total-body vitamin A stores and liver vitamin A concentrations were estimated with the deuterated-retinol-dilution (DRD) technique at a dose of 5 mg [2H4]retinyl acetate at baseline and 5 mg [2H8]retinyl acetate during the repeat test 1 y later. Plasma retinol and carotenoids were measured by HPLC.
Results: Median total-body vitamin A stores increased from 0.33 to 0.72 mmol (P = 0.0001), liver vitamin A concentrations from 0.52 to 0.78 µmol/g (P = 0.0003), and plasma retinol concentrations increased from 0.97 to 1.17 µmol/L (P = 0.01).
Conclusion: The vitamin A status of Nicaraguan schoolchildren improved during the year after the initial distribution of vitamin A–fortified sugar in Nicaragua.
Key Words: Vitamin A • total-body vitamin A stores • liver vitamin A • deuterated retinol dilution • stable-isotope dilution • retinol • carotenoids • ß-carotene • vitamin A–fortified sugar • Nicaraguan children
INTRODUCTION
Vitamin A deficiency is a serious health problem that can result in anemia, a reduced resistance to infection, xerophthalmia, and ultimately blindness and death (1). The ingestion of a sufficient amount of vitamin A is a persistent challenge for populations subsisting on a largely plant-based diet (2). Hypovitaminosis A is endemic primarily in societies that rely on provitamin A carotenoids as the source of dietary vitamin A (3). The feasibility of obtaining sufficient vitamin A from plant-based diets became even more challenging with the readjustment of the conversion factors for plant carotenoids into vitamin A from 6:1 to 12:1 for ß-carotene and from 12:1 to 24:1 for other provitamin A carotenoids (4).
Hypovitaminosis A was found to be endemic in Central America in the isthmus-wide nutrition survey of the mid-1960s (5). The nations in Central America have responded to their precarious dependence on vitamin A from a largely vegetal diet by instituting mandatory national programs of fortifying table sugar sold in these nations with vitamin A. By early 1999, 3 countries of Central America (Guatemala, El Salvador, and Honduras) had put into practice legislation mandating the fortification of table sugar with vitamin A, following the plan developed by Arroyave (6) in Guatemala. Nicaragua was the fourth of the Central American republics to initiate the vitamin A fortification of table sugar for domestic consumption. The Nicaraguan sugar-fortification program commenced with the sugarcane harvest and milling in November 1999, and fortified sugar became available in markets by early 2000. With the de novo introduction of this public health measure, the opportunity arose to document any effect of the program on vitamin A status.
The goal of our study was to assess the effect, with the use of stable isotopes of vitamin A, of the Nicaraguan national program of sugar fortification with vitamin A on the vitamin A status of a cohort of Nicaraguan schoolchildren. The deuterated-retinol-dilution (DRD) technique (7, 8) is one of the most sophisticated methods currently available for assessing vitamin A status. This procedure has been used to assess the total-body stores of vitamin A in children, adults, or both in the United States (7, 9), Bangladesh (9-11), Guatemala (12, 13), China (14, 15), Philippines (13, 16, 17), and Peru (18). The late Professor James A Olson stated, "Further uses of labeled tracers both in determining endogenous reserves of nutrients as well as in quantitating rate processes in humans are clearly a wave of the future in human nutrition" (19).
We report here our findings from the monitoring of vitamin A status with the use of the DRD technique in Nicaraguan schoolchildren at baseline and 1 y after the start of the availability of vitamin A–fortified sugar in their local markets.
SUBJECTS AND METHODS
Subjects
The study participants were 5–9-y-old children (n = 21) who resided in Sabana Grande, a low socioeconomic suburb of Managua. Only children who did not receive a vitamin A dose during the previous 8 mo were eligible to participate in this study. The Nicaraguan National Health Campaign, which started in 1994, administers vitamin A yearly to children <5 y of age (20). The participants were in general good health, with no major chronic diseases. At the time of the vitamin A status assessments, the subjects had no acute illnesses, no febrile conditions, and no gastrointestinal problems. Anthelmintic treatment was not provided during the study. Some of the study participants may have received anthelmintic treatment 8 mo before the study, because albendazole was provided with vitamin A in the Nicaraguan National Health Campaign. In this study, each child served as his or her own control because of the impossibility of having a separate study group of persons unexposed to fortified sugar since the fortification program is implemented nationwide in Nicaragua as well as in other countries in Central America. Although the children and their caregivers were informed about the purpose of the study, no attempt was made by the study investigators to influence their usual consumption of sugar or other aspects of their diet, although medical consultation, referrals to health units, and basic medicines were available to the children if needed. Written informed consent was obtained from the children's caregivers. Approval to conduct the study was obtained from the Tufts University–New England Medical Center Human Investigation Review Committee and the Ethics Committee of the Universidad Nacional Autónoma de Nicaragua.
Methods
Vitamin A status was assessed in March 2000 just after vitamin A–fortified sugar first became available in the local markets and 1 y later (March 2001) in the same cohort of children. Total body stores of vitamin A, liver vitamin A concentrations, and plasma retinol concentrations were measured, as were plasma carotenoid concentrations. Baseline and 1-y plasma samples from each child were analyzed simultaneously to minimize interassay variability.
Stable isotopes of vitamin A
Tetradeuterated retinyl acetate (ie, all-trans-retinyl-10,19,19,19-[2H4]acetate) and octadeuterated retinyl acetate (ie, all-trans-retinyl-10,14,19,19,19,20,20,20-[2H8]retinyl acetate) were synthesized by Cambridge Isotope Laboratories (Andover, MA). We prepared capsules containing 5.0-mg amounts (15.04 µmol [2H4]retinyl acetate or 14.86 µmol [2H8]retinyl acetate) of these stable isotopes dissolved in corn oil.
Deuterated-retinol-dilution procedure for estimating total-body vitamin A
The study participants ingested a capsule of deuterated retinyl acetate with a fat-containing meal at the study center. [2H4]Retinyl acetate was administered at baseline, and [2H8]retinyl acetate was administered 1 y later to distinguish plasma [2H8]retinol from any residual [2H4]retinol. After 21 d, the participants returned to the study center for a nonfasting venous blood draw. In a study of 5- to 8-y-old children, it was reported that blood samples can be obtained either fasting or within 4 h after breakfast without altering the results for blood concentrations of retinol or carotenoids (21). All blood handlings were done in a manner that would protect the specimens from heat and light. Plasma samples were prepared and pipetted into cryovials, which were stored at –20 °C; the frozen samples were hand-carried in dry ice to the Human Nutrition Research Center at Tufts University in Boston, where they were kept at –70 °C until analyzed. All laboratory procedures in Boston were carried out under dim red light to prevent the photodegradation of retinoids and carotenoids.
The plasma samples were analyzed for deuterated- and nondeuterated retinol isotopes by separating retinol from other constituents of plasma with the use of HPLC, collecting the retinol fraction, converting retinol into trimethylsilyl derivatives, and using gas chromatography–mass spectrometry (GC-MS) to analyze retinol isotopes (22). The ratio of deuterated to nondeuterated retinol (D:H) in plasma was determined and used in the Olson equation (7) to obtain a numerical estimate of total-body stores of vitamin A. This formula, which is a modified version of the formula of Bausch and Rietz (23) for the assessment of vitamin A stores in rats, is as follows:
RESULTS
The combined data (n = 21) from 13 male and 8 female subjects are presented, because no difference was observed by sex. The body weights and heights of the study participants increased significantly (Table 1), as expected, during the study year; there was no change in body mass index. As assessed in relation to the US Centers for Disease Control and Prevention juvenile reference (32), 14.3% of children were underweight at baseline, and 9.5% were underweight at 1 y. Stunting was observed in 9.5% of the subjects at both evaluation time points. Four subjects had a body mass index above the 85th percentile of the Centers for Disease Control and Prevention curve (32), which placed them at risk of overweight; there were 3 such subjects after 1 y. No wasting was observed at anytime during the study, as assessed with the use of the US National Center for Health Statistics percentiles (33).
View this table:
TABLE 1. Characteristics of the study participants at baseline and after 1 y of study1
The distributions and individual changes in total-body vitamin A stores, liver vitamin A concentrations, and plasma retinol concentrations from 2000 to 2001 are illustrated in Figure 1. Total-body vitamin A increased in all participants during the year, with a median increase of 0.373 mmol, or 112.0% (0.337 mmol, or 101.3%, if one outlier at the top of the distribution is excluded from analyses).
FIGURE 1.. Individual changes in the 3 measures of vitamin A status (total-body vitamin A, liver vitamin A, and plasma retinol) between March 2000 and March 2001 in the 21 children in the cohort. Total-body vitamin A stores, as estimated with the deuterated-retinol-dilution technique, rose from a mean (±SD) of 0.389 ± 0.177 (median: 0.333; range: 0.169–0.943) mmol retinol at baseline to 0.929 ± 0.553 (median: 0.718; range: 0.376–2.596) mmol retinol after 1 y [P = 0.0001 for the mean and median, including the value for the outlier (dashed line)]. Liver vitamin A concentrations, estimated on the assumption that liver weight is 3% of body weight in children and that 90% of body vitamin A is in the liver, rose from 0.57 ± 0.27 (median: 0.52; range: 0.22–1.35) to 1.20 ± 0.74 (median: 0.78; range: 0.55–3.24) µmol retinol/g [P = 0.0001 for the mean and P = 0.0003 for the median, including the value for the outlier (dashed line)]. Plasma retinol concentrations rose from 0.99 ± 0.17 (median: 0.97; range: 0.74–1.31) to 1.18 ± 0.27 (median: 1.17; range: 0.77–1.68) µmol/L (P = 0.002 for the mean and P = 0.01 for the median). The significant differences are based on Student's paired t test (mean) or the Mann-Whitney U test (median).
With respect to hepatic vitamin A concentrations, none of the subjects had a concentration below the deficiency cutoff of <0.07 µmol/g at either baseline or 1 y later (Figure 1). Increases in liver vitamin A concentrations were observed in all subjects, except for the child who manifested the smallest increase in total-body vitamin A stores (0.058 mmol); for this subject, the estimated hepatic vitamin A concentration actually decreased from 1.083 to 1.003 µmol/g over the year.
Mean plasma retinol values increased on average by 19.6% over the year of prospective monitoring, with a median increment of 18.6%. Individual values increased in 16 subjects and decreased in 5 (Figure 1), despite a concurrent increase in total-body vitamin A in all 21 subjects. Thus, the changes in total body stores or liver concentrations of vitamin A were poorly related to individual changes in plasma retinol concentrations. None of the study participants had plasma retinol concentration <0.70 µmol/L at baseline or 1 y later.
The mean (±SD) percentage enrichment of plasma retinol with [2H4]retinol 21 d after an oral dose of 5 mg [2H4]retinyl acetate was 1.31 ± 0.49; the residual percentage enrichment 1 y later was 0.59 ± 0.20%. The percentage enrichment of plasma retinol with [2H8]retinol 21 d after an oral dose of 5 mg [2H8]retinyl acetate was 0.69 ± 0.35%.
The distributions and changes for 7 classes of carotenoids for each study participant are shown in Figure 2. Reductions in plasma trans-ß-carotene concentration were observed over the year in 95% of the subjects and to an extent of 54.5% of the median value; decreases in 13-cis-ß-carotene were seen in 100% of the subjects, and the reduction in the median value was 41.7%. Median plasma concentrations rose significantly over the year by 107.1% for ß-cryptoxanthin and by 37.5% for lutein, whereas no significant changes were observed for -carotene, zeaxanthin, and lycopene. Changes in individual plasma carotenoids did not correlate with changes in the 3 vitamin A status measures. For example, the extent of changes in plasma trans-ß-carotene in study participants did not show any proportionality to the extent of changes in their total-body vitamin A stores (r = –0.21, P = 0.37), liver vitamin A concentrations (r = –0.20, P = 0.41), or plasma retinol concentrations (r = 0.26, P = 0.26).
FIGURE 2.. Individual changes in the 7 plasma carotenoids between March 2000 and March 2001 in the 21 children in the cohort. trans-ß-Carotene decreased from a mean (±SD) of 1.04 ± 0.71 (median: 0.77; range: 0.26–2.81) µmol/L at baseline to 0.43 ± 0.27 (median: 0.35; range: 0.11–1.19) µmol/L after 1 y (P = 0. 0.0001 for the mean and P = 0.0004 for the median). 13-cis-ß-Carotene decreased from 0.048 ± 0.036 (median: 0.036; range: 0.007–0.140) to 0.023 ± 0.017 (median: 0.021; range: 0.004–0.076) µmol/L (P = 0.0001 for the mean and P = 0.003 for the median). ß-Cryptoxanthin rose from 0.23 ± 0.27 (median: 0.14; range: 0.04–1.29) to 0.33 ± 0.18 (median: 0.29; range: 0.08–0.76) µmol/L (P = 0.03 for the mean and P = 0.005 for the median). Lutein rose from 0.20 ± 0.10 (median: 0.16; range: 0.08–0.43) to 0.26 ± 0.11 (median: 0.22; range: 0.10–0.44) µmol/L (P = 0.007 for the mean and P = 0.03 for the median). -Carotene changed from 0.14 ± 0.09 (median: 0.12; range: 0.04–0.44) to 0.19 ± 0.13 (median: 0.19; range: 0.04–0.56) µmol/L (P = 0.002 for the mean and P > 0.05 for the median). Zeaxanthin changed from 0.038 ± 0.018 (median: 0.034; range: 0.012–0.075) to 0.032 ± 0.014 (median: 0.027; range: 0.012–0.066) µmol/L (P > 0.05 for the mean and the median). Lycopene changed from 0.10 ± 0.07 (median: 0.09; range: 0.003–0.32) to 0.12 ± 0.11 (median: 0.09; range: 0.02–0.47) µmol/L (P > 0.05 for the mean and the median). The significant differences (P < 0.05) are based on Student's paired t test (mean) or the Mann-Whitney U test (median).
DISCUSSION
The findings of this study suggest a strong effect of the Nicaraguan national program of fortifying domestic sugar with vitamin A on the vitamin A status of school-aged children in Nicaragua. During the inaugural year of the program, total-body and liver vitamin A reserves more than doubled, and plasma retinol concentrations increased 1.2-fold. Because of the magnitude of the change in vitamin A status, the conclusion that this was attributable to the national fortification program seems reasonable, despite the lack of any control group because of the impossibility of finding an unexposed cohort for simultaneous comparison.
The estimated mean (0.39 mmol) or median (0.33 mmol) total-body vitamin A stores in Nicaraguan children at baseline were lower than the value of 1.02 mmol reported for one 6-y-old child in the United States (9) and higher than the mean values reported for 2 groups of 5- to 7-y-old (0.09 and 0.13 mmol, respectively) and one group of 10- to 11-y-old (0.27 mmol) children in China (14, 15). As in the present study, the US and Chinese values were also obtained by using the DRD procedure (7). After 1 y of exposure to vitamin A–fortified sugar, the estimated mean (0.93 mmol) or median (0.72 mmol) total-body vitamin A stores in the Nicaraguan study participants approached the value reported for the US child.
According to Olson (25, 28, 29), a minimally adequate liver vitamin A concentration is 0.07 µmol/g (20 µg/g), because this concentration will meet all physiologic needs for the vitamin and maintain a reserve for 3–4 mo when intakes are low or during stress. None of the study participants had a liver vitamin A concentration <0.07 µmol/g. We compared the liver vitamin A values in Nicaraguan schoolchildren at baseline and after 1 y of exposure to vitamin A–fortified sugar with published reports of liver vitamin A, as estimated with the DRD technique (9, 14, 15), or by direct measurements of liver specimens from children who died from various causes (34-42). These comparisons are illustrated in Table 2 and show that the liver vitamin A stores in Nicaraguan schoolchildren seemed adequate, even at baseline, but that the vitamin A reserves doubled during the year to values similar to those in US and Canadian children.
View this table:
TABLE 2. Liver vitamin A concentrations in school-aged children as estimated with the deuterated-retinol-dilution (DRD) technique and in school-aged children who died from various causes as obtained by direct measurement in liver specimens
Plasma retinol values of 0.70 or 1.05 µmol/L (20 or 30 µg/dL) are regarded as alternative cutoffs below which clinical deficiency can become evident (29). At baseline, none of the study participants had plasma retinol values <0.70 µmol/L, but 14 children (70%) had values <1.05 µmol/L. After 1 y, 6 children (30%) had plasma retinol values <1.05 µmol/L.
The greater improvement in vitamin A body stores compared with the improvement in plasma retinol is not surprising because it is recognized that circulating retinol is homeostatically controlled over the physiologic range of liver vitamin A concentrations (28). Plasma retinol values tend to fall precipitously when liver vitamin A concentrations are <0.07 µmol/g (20 µg/g) and tend to increase steeply when liver concentrations are >1.05 µmol/g (300 µg/g) (28). Within the cited range, however, circulating concentrations respond with a very shallow slope to increments in liver vitamin A across the aforementioned range. In this study, a small but significant rise in plasma retinol was observed after 1 y. However, the plasma retinol response was not consistent, with some subjects showing little change or even a decrease in plasma retinol (Figure 1).
One child had a high baseline liver vitamin A concentration of 1.35 µmol/g (outlier, Figure 1), which rose 2.4-fold to 3.24 µmol/g after 1 y; the corresponding plasma retinol values were unremarkable, ie, 1.20 and 1.68 µmol/L, respectively. According to Olson (28), the normal physiologic range of vitamin A stores is 0.07–1.05 µmol/g liver. After 1 y, 9 of the 21 study participants had liver vitamin A stores that were >1.05 µmol/g. The upper safe limit of hepatic vitamin A stores is not known, but should be carefully assessed in all vulnerable groups, such as children and pregnant and lactating women, to ascertain that continued increased intakes of vitamin A pose no risk of adverse health effects. A clear indicator of vitamin A toxicity is a markedly elevated concentration of circulating vitamin A in the form of retinyl esters (43); none of the participants in this study had detectable plasma retinyl ester concentrations.
The target concentration of vitamin A in sugar was 10–15 µg retinol activity equivalents (RAE)/g [1 µg retinol = 1 RAE (4)]. However, a large variation in vitamin A concentrations was found in batches of the sugar samples analyzed. In one study, the analyses of sugar samples taken from 131–150 households in 5 regions in Nicaragua, conducted 5 times between March 2000 and March 2001, gave mean values of 5.5, 4.1, 4.0, 4.7, and 5.0 µg RAE/g; individual values ranged from 0 to 28.4 µg RAE/g sugar (E Boy, O Dary, unpublished observations, 2003). Another set of assays from another laboratory found a mean value of 16.4 µg RAE/g sugar, with values ranging from 2.5 to 61.1 µg RAE/g (44) in sugar from among the same sites. The apparent discrepancy in the results of these 2 experiences could possibly reflect differences in laboratory methods used in the analysis or true differences associated with quality-control measures during the fortification process.
Precise sugar intake data are not available for Nicaraguan school-aged children, but, on the basis of the mean reported sugar consumption of Guatemalan toddlers (ie, 30 g/d, or 2 tablespoons; 45) and on the basis of the mean vitamin A content in sugar obtained in the 2 studies cited above, we estimated that an average of 140 or 492 µg RAE/d (176 or 618 µmol/y) was consumed by the Nicaraguan schoolchildren from fortified table sugar alone. These estimates may be conservative because mean sugar intakes of 61 and 85 g/d were reported in Guatemalan rural and urban populations, respectively (46). Industrially processed sweetened foods would also contribute substantially to the amounts of vitamin A consumed by the Nicaraguan study participants, but precise intakes of vitamin A from sweetened foods and other dietary sources are not available.
The US Institute of Medicine (4) has defined the Tolerable Upper Intake Level (UL) as the highest level of daily vitamin A intake that is likely to pose no risk of adverse health effects in almost all individuals. For children 4–8 y of age, the UL is 900 µg preformed vitamin A/d; for those between 9 and 13 y of age, the UL is 1700 µg preformed vitamin A/d (4).
In the current study, the Nicaraguan schoolchildren had a median increase in vitamin A body stores of 373 µmol during the year (or 337 µmol with the outlier excluded), an increase that is reasonable and attainable based on the aforementioned conservative intake estimates and considerations, and a partition of >50% for hepatic vitamin A storage (accumulation), and <50% for turnover (and loss).
It is intriguing that plasma concentrations of trans- and cis-ß-carotene decreased almost consistently in all subjects during the study year, whereas other plasma carotenoids either increased or remained unchanged. The Nicaraguan study participants had a baseline trans-ß-carotene concentration (mean: 1.04 µmol/L; median: 0.77 µmol/L) that was higher than the mean (or median) baseline circulating concentrations reported for other school-aged children (values are in µmol/L): 0.43 in China (14), 0.34 (median: 0.28) in the United States (47), 0.31 (median: 0.24) and 0.27 in Guatemala (48, 49), 0.14 in Indonesia (50), and 0.09 (median: 0.08) in malnourished schoolchildren in rural Philippines (16). Despite the reduction in plasma ß-carotene in Nicaraguan children during the repeat measurement 1 y after baseline (mean: 0.43 µmol/L; median: 0.35 µmol/L), the values were still higher than or similar to those mentioned above in other school-aged children.
Nevertheless, it would be prudent to further investigate the significance of this observation. It is possible that the reduction in plasma ß-carotene with the increased body pool size of vitamin A may have been due to an increase in tissue uptake of ß-carotene. However, a decrease in the absorption of dietary ß-carotene is also a possibility. In chicks, decreased carotene absorption and decreased carotene concentrations in serum, liver, and toe-web skin were found after dietary vitamin A was increased (51). The possibility that the observed reduction in plasma ß-carotene might have been due to the reduced consumption of ß-carotene–rich foods, because of the seasonal unavailability, was unlikely because the blood draw for plasma measurements was done during the same calendar month of March, albeit 12 mo apart. The possibility that the reduction in plasma ß-carotene was due to the increased conversion of ß-carotene to vitamin A by the increased activity of intestinal ß-carotene 15,15-dioxygenase was also unlikely. In schoolchildren in rural Philippines, it was found that the bioconversion of plant carotenoids to vitamin A varies inversely with vitamin A status (16). The activity of intestinal ß-carotene 15,15-dioxygenase was reported to be higher in rats fed low amounts than in those fed high amounts of vitamin A (52, 53), although other researchers reported that neither retinol depletion nor excess feeding affected this enzyme's activity in rats (54). In hamsters, intestinal enzyme activity was enhanced by feeding a vitamin A–deficient diet; no relation was noted between liver vitamin A and enzyme activity (55).
In conclusion, a stable-isotope-dilution procedure was used to monitor changes in vitamin A status in a cohort of schoolchildren during the year after the initiation of the Nicaraguan national program of sugar fortification with vitamin A. Median total-body vitamin A increased 2.2-fold (112.0%), and plasma retinol increased 1.2-fold (18.6%), which suggested that the program was successful in achieving its goal of improving the vitamin A status of Nicaraguan children residing in a low-income community. However, because everyone in Nicaragua is exposed to the sugar-fortification program, it would be prudent to monitor the vitamin A status of children and pregnant and lactating women, not only from low-income communities but also from higher-income communities, to assess whether the continued increased intakes of vitamin A pose any risk of adverse health effects.
ACKNOWLEDGMENTS
We thank the Nicaraguan children for their participation in this study. We are grateful for the administrative and technical support from Erick Boy (Micronutrient Initiative, Ottawa).
JDR-M participated in the study design, HPLC and GC-MS analyses of plasma, data analyses, and writing of the manuscript. NWS conceived and initiated the study, was the overall study coordinator, and participated in the study design, data analyses, and writing of the manuscript. YM was field director of the protocol. JB coordinated the protocol application and standardized the sample collection and handling procedures. GGD participated in the GC-MS analyses. RMR coordinated the procedures at Tufts University. CBW was the overall coordinator in Nicaragua. The authors had no conflict of interest with the sponsoring organizations.
REFERENCES