Secondary Logo

Journal Logo

Original Article

Dietary Nitrites and Nitrates, Nitrosatable Drugs, and Neural Tube Defects

Brender, Jean D.*; Olive, Janus M.*; Felkner, Marilyn; Suarez, Lucina; Marckwardt, Wendy*; Hendricks, Katherine A.

Author Information
doi: 10.1097/01.ede.0000121381.79831.7b
  • Free

Abstract

Amine- and amide-containing (nitrosatable) drugs could react with nitrite to form N-nitroso compounds or nitrosamines.1–5 Several N-nitroso compounds have been found in animal models to induce congenital malformations, including central nervous system abnormalities.6–9 Relatively few studies have been published concerning the relation between nitrosatable drug exposure during the periconceptional period and congenital malformations in humans. Studies of these drug exposures have suggested increased risks of craniosynostosis,10,11 meningomyelocele/meningocele, hydrocephaly, and abnormalities of the eye and gastrointestinal system.10

Nitrite is a necessary substrate in the formation of N-nitroso compounds from nitrosatable drugs, and its concentration in the body depends on nitrate and nitrite intake in food and water.12 Approximately 5% of ingested nitrate is converted to nitrite in the saliva.13 Higher dietary intake of nitrate and nitrite would most likely result in greater amounts of N-nitroso compounds being formed if exposure to nitrosatable compounds also occurred. None of the aforementioned studies on congenital malformations, however, examined nitrosatable drug exposure in relation to dietary intake of nitrates and nitrites.

In this study, we examined nitrosatable drug exposure during the periconceptional period in relation to dietary intake of nitrates and nitrites and the risk of neural tube defect (NTD)-affected pregnancies among Mexican Americans living along the Texas–Mexico border. We also examined nitrosatable drug exposure in relation to nitrates in drinking water for a subset of the population for whom the usual source of drinking water during the periconceptional period could be tested for nitrates.

METHODS

Data for this analysis of nitrosatable drug exposure and NTDs came from the Texas Neural Tube Defects Project, which included multisource active surveillance, a folic acid intervention, and a case-control study among women who resided in the 14 Texas counties along the Texas–Mexico border. Methods for case finding are described in detail elsewhere.14 Briefly, surveillance involved active case ascertainment from genetic clinics and ultrasound centers for prenatally diagnosed fetuses, as well as from hospitals, birthing centers, abortion centers, prenatal clinics, and lay and certified midwives. Eligible women were defined as all Mexican Americans who delivered a live- or stillborn infant or who had a spontaneous or induced abortion with an NTD during the period June 1995 through May 2000. Of the NTD cases identified among these women, 53% were liveborn; 13%, stillborn; 31%, electively aborted; and 2%, spontaneously aborted. Included in the case series were 83 with anencephaly (International Classification of Diseases, 9th Revision, Clinical Modification [ICD-9-CM] code 740), 84 with spina bifida (741), and 17 with encephalocele (742.0). Control women were identified from Mexican American residents in the study area who delivered normal live births during the same time period as case women; they were randomly selected annually in proportion to the number of live births that occurred 2 years earlier in a given facility, which included hospitals and midwife-attended birthing centers. The Texas Department of Health Institutional Review Board for the protection of human subjects approved the protocol, consent forms, and questionnaires.

Case and control women were interviewed in person, approximately 5 to 6 weeks postpartum in either English or Spanish after obtaining informed consent. The questionnaire was modeled after the Mother Questionnaire, Centers for Disease Control and Prevention, Birth Defect Risk Factor Surveillance (unpublished document, 1992). The instrument contained questions about maternal health and reproductive history, medical history, use of medications and nutritional supplements, and other potential risk factors for NTDs. Before each interview, the study obstetrician–gynecologist and interviewer established the date of conception for the index pregnancy using all gestational age estimates from the medical record. The interviewer used a personalized calendar to focus the respondent's attention on exposures and events during the 6-month interval beginning 3 months before and ending 3 months after the estimated date of conception.

A 98-item food frequency questionnaire (FFQ), specifically designed for this Mexican American border population,15 was administered to all study participants. The FFQ was developed from 100 24-hour dietary recalls elicited from new enrollees into the federal Women, Infants, and Children (WIC) program in the 4 most populous border counties in the study area. A comparable approach in another county in the study area produced a similar FFQ that showed high agreement with 3-day food records for estimates of energy, fats, and cholesterol among low-income Mexican Americans.16 For each food item, women estimated their usual frequency of intake (average number of times per day, week, or month) during the periconceptional period. Periconceptional dietary nitrate and nitrite intakes were calculated from these responses. After identifying the average amounts of nitrates and nitrites of these foods from the published literature,13,17–22 we multiplied these averages by the frequency that the food was reportedly eaten to obtain a daily intake of these compounds for each food. Daily intakes of nitrates and nitrites were then summed across all foods to obtain total daily intakes of nitrate and nitrite for each study participant. We calculated the total nitrite exposure for each participant by using the formula suggested by Choi13 (total nitrite = dietary nitrite intake + 0.05 [dietary nitrate intake]). On average, dietary nitrate and nitrite contributed 52% and 48%, respectively, to total nitrite intake in this population.

All medications reported by women taken during the periconceptional period were reviewed and classified as potentially nitrosatable based on the published literature.1–5,12,23 We identified the active ingredients of over-the-counter preparations with the use of the MICROMEDEX DRUGDEX System,24 the Physicians’ Desk Reference,25 and product information available on the Internet. Assignment of nitrosatable drug exposure and the associated daily intake of nitrates and nitrites was completed without knowledge of case or control status.

During the face-to-face interviews, interviewers determined the usual periconceptional drinking water source (bottled water or tap water), and water samples from that source were collected and measured for nitrates. Water samples were not collected if the periconceptional residence was not accessible or if consent could not be obtained for sampling. Water nitrate was tested by mixing a water sample with NitraVer 5 Reagent Powder Pillow (USA Hach Co., Loveland, CO), followed by 3 consecutive measurements of nitrate with a pocket colorimeter. Measurements were in milligrams per liter nitrate as nitrogen (NO3-N). For the purposes of this study, these values were converted to milligrams per liter total nitrate (NO3) by multiplying them by 4.4. An average of the 3 values was calculated. We were not able to calculate daily intake of water nitrate for each participant because the questionnaire did not contain questions on the frequency and amount of water consumption.

We used logistic regression26 to obtain odds ratios (OR) and 95% confidence intervals (CI) for risk of neural tube defects associated with nitrosatable drug exposure by levels of dietary nitrite and total nitrite. Daily dietary nitrite and total nitrite were divided into tertiles based on the distribution among control women. We considered the following factors as covariates: maternal age (<20, 20–24, 25–29, 30+ years); education (<7, 7–11, 12+ years); household income (categorized into 6 levels that ranged from <$10,000 through >$40,000); body mass index (<30 kg/m2, ≥30 kg/m2), folate intake (diet and supplements combined: < 400, 400+ μg); supplement use (yes, no); energy intake (<2388, 2388–2975, 2976–3908, >3908 kcal); dietary vitamin C intake (< 205, 205–293, 294–412, >412 mg); smoking (yes, no); and dietary intake of nitrosamines (< 0.544, 0.544–0.829, >0.829 μg). Other variables that have been associated with NTDs in this study population were also considered as covariates, including: serum B12 (in quintiles),27 hyperinsulinemia (<11, 11+ uIU/mL),28 maternal solvent exposure (yes, no),29 and stressful life events (yes, no).30 In the final logistic regression analyses, only household income, body mass index, and folate intake changed the risk estimates by more than 10%; therefore, none of the other covariates were included in the final models. Because febrile illnesses during the periconceptional period have been associated with NTD risk,31–33 the analyses for nitrosatable drug exposure were further stratified and adjusted for reported fevers.

Logistic regression was also used to obtain the ORs and 95% CIs for NTDs associated with nitrosatable drug exposure by water nitrate. Drinking water nitrates were divided into 2 groups based on the 50th percentile (3.5 mg/L) of the control-women's distribution; the numbers of exposed cases and controls were insufficient to permit finer stratification by water nitrate levels. Maternal education changed the risk estimates by more than 10% and was therefore included in the final model.

RESULTS

Among the 225 case women and 378 control women identified, 184 (82%) case women and 225 (60%) control women completed the interviews. Table 1 shows the characteristics of these women by age, education, household income, body mass index, folate intake, and reported fevers. Compared with control women, case women were somewhat poorer and less educated; more likely to have a body mass index of greater than or equal to 30 kg/m2; and more likely to report a fever during the periconceptional period.

TABLE 1
TABLE 1:
Characteristics of Mexican American Case Women With NTD-Affected Pregnancies and Control Women in 14 Texas–Mexico Border Counties, 1995–2000

Women who took nitrosatable drugs during the periconceptional period were 2.5 times more likely to have an NTD-affected pregnancy than women without this exposure (95% CI = 1.3–4.8); the odds ratios for anencephaly (OR = 2.5; 95% CI = 1.1–5.5) and spina bifida (2.7; 1.2–5.8) were similar although the OR for encephalocele was somewhat lower (1.8; 0.3–9.3). With adjustment for household income, body mass index, and folate intake, women who took any drugs classified as nitrosatable during the periconceptional period were 2.7 times more likely to have an NTD-affected pregnancy than women who did not take these drugs (CI = 1.4–5.3) (Table 2).

TABLE 2
TABLE 2:
Odds Ratios and 95% Confidence Intervals for Neural Tube Defects and Nitrosatable Drug Exposure During the Periconceptional Period by Dietary Intake of Nitrites and Estimated Total Nitrites

Although information on drug intake was available for all study participants, approximately 89% (n = 365) of participants had complete dietary information for calculation of dietary nitrite intake and 88% (n = 359) had complete dietary nitrate and nitrite information for calculation of total nitrite. Intake of dietary nitrite and total nitrite showed minimal association with NTD risk in this study population. Relative to the lowest tertiles of intake, the odds ratios for the upper tertiles of nitrite intake were 0.8 and 0.9 and for total nitrite, 0.9 and 0.8. When nitrosatable drug exposure was stratified by tertiles of dietary nitrite and total nitrite, the elevated ORs were restricted to the 2 upper tertiles of intake for both indices. The strongest association between nitrosatable drug exposure and NTDs was found among women with an estimated daily total nitrite of greater then 10.5 mg (OR = 7.5; CI = 1.8–45).

Adjustment for reported fevers during the periconceptional period slightly reduced the risk estimates associated with nitrosatable drug exposure (OR = 2.4; CI = 1.3–4.5). The odds ratio for NTDs associated with nitrosatable drug exposure among women with fever was considerably lower (1.4; 0.3–5.9) than that for women with this drug exposure but who did not report a fever (3.0; 1.4–7.0). In the 2 upper tertiles of dietary nitrite (≥3.2 mg/day) and total nitrite (≥7.5 mg/day), however, nitrosatable drug exposure was positively associated with NTD-affected pregnancies (OR = ∞; CI = 0.9–∞ and OR = 12.1; CI = 0.9–895, respectively) as well as for women without reported fevers.

We also examined nitrosatable drug exposure by periconceptional month, but the small numbers of exposed cases and controls limited the analyses to crude ORs. The highest OR for NTDs associated with nitrosatable drug exposure was for those drugs taken during the month before the estimated date of conception (3.8; 0.7–39).

Four drugs or active ingredients accounted for almost three fourths (72%) of nitrosatable drugs taken. These were drugs containing phenylpropanolamine (a sympathomimetic used in cold remedies until recently; n = 16), drugs containing chlorpheniramine (an antihistamine; n = 14 women), ampicillin (n = 12), and various types of penicillin preparations (n = 8). The remainder included other antibiotics, antihistamines, and beta-adrenergic-blocking agents that have been classified as nitrosatable in the literature.

Of the 409 women interviewed, water samples for nitrate measurement were collected for 110 (27%), including 43 case women and 67 control women. Participants in the water nitrate substudy were similar to nonparticipants by age, household income, and body mass index. However, participating case women were more educated (61% with 12+ years of education) than nonparticipating case women (46% with 12+ years of education). Drinking water nitrate levels ranged from 0 to 28 mg/L (median of 5.4 mg/L for case women and 3.5 mg/L for control women). Women whose drinking water nitrates measured 3.5 mg/L or greater were 1.9 times more likely (CI = 0.8–4.6) to have an NTD-affected pregnancy than women with lower levels of nitrate in their water. Table 3 shows the ORs for NTDs associated with nitrosatable drug exposure in relation to drinking water nitrates. With adjustment for maternal education, women who took nitrosatable drugs during the periconceptional period and whose drinking water source had measured nitrate levels of 3.5 mg/L or higher were 14 times (CI = 1.7–660) more likely to have an NTD-affected pregnancy. The odds ratio associated with nitrosatable drug exposure was considerably lower (OR = 2.3) among women whose drinking water source had nitrate levels less than 3.5 mg/L. Although adjustment for reported fevers reduced these risk estimates, nitrosatable drug exposure remained positively associated with NTDs among women whose drinking water nitrates were in the higher range. Further stratification by dietary nitrites could not be accomplished because of the small numbers of exposed case and control women.

TABLE 3
TABLE 3:
Odds Ratios and 95% Confidence Intervals for Neural Tube Defects and Nitrosatable Drug Exposure During the Periconceptional Period by Nitrates in Drinking Water

DISCUSSION

Maternal exposure to nitrosatable drugs was associated with NTD risk. Moreover, this relation was observed only among women who had high dietary intake of nitrite and total nitrite. Women with high intakes of dietary nitrites and total nitrites were more likely to have an NTD-affected pregnancy if they took nitrosatable drugs, regardless of whether they reported fevers during the periconceptional period. Several studies have examined the separate associations of NTD risk with nitrosatable drugs,10,34 dietary nitrates and nitrites,34 and drinking water nitrates.34–39

Using prospective data from the National Collaborative Perinatal Project, Olshan and Faustman10 found that women who took nitrosatable drugs during the first 4 months of pregnancy were 4.6 times more likely to give birth to a child with a meningomyelocele or meningocele (CI = 0.7–31.6); no positive association was seen between these drug exposures and anencephaly. In a study of maternal exposure to nitrate from drinking water and diet and risk for NTDs, Croen et al.34 also examined the relation between nitrosatable drug exposure and NTDs. The proportion of California control women with nitrosatable drug exposure (8.0%) was similar to that (8.7%) of control women in the Texas border county study area. Essentially no association (OR = 1.2; CI = 0.8–1.8) between nitrosatable drug exposure and NTDs were found in the California population. One of the reasons for this finding might be the considerably lower estimated daily dietary intake of nitrite in this population. The highest dietary nitrite intake was 3.9 mg/day compared with 12.6 mg/day for the Texas–Mexico border population in our study. Restricting our analyses of nitrosatable drug exposure to women whose dietary nitrite intake was 3.9 mg/day or less resulted in a similar risk estimate (OR = 1.3; CI = 0.5–3.4) to the one found in the California study.

Based on what is known about the endogenous formation of N-nitroso compounds, it seems important to take into account dietary nitrite and total nitrite when assessing the teratogenic effects of nitrosatable drugs. In acidic aqueous media such as the stomach, N-nitroso compounds can be formed from the reaction of nitrous acid (formed from nitrite) with drugs containing potentially nitrosatable groups such as amines or amides.4,12 Choi13 noted that the formation of nitrosamines will be greater if the nitrite concentration is higher. If N-nitroso compounds are a risk factor for neural tube defects in offspring, it is plausible that exposure to nitrosatable drugs in the presence of higher levels of total nitrite would result in higher risk for these defects.

In light of the low levels of drinking water nitrate found in this study population and the small percentage (5%) of nitrate that is converted into available nitrite, the strong association found between nitrosatable drug exposure and NTD-affected pregnancies among women with higher levels of water nitrate is difficult to explain. Women in this study were not questioned about the frequency and amounts of tap water and bottled water they consumed during the periconceptional period; therefore, we could not directly estimate the amount of nitrate they ingested from water. In a study of tap water and total water ingested by pregnant women, Ershaw et al.40 found that total water intake for this group was approximately 2 L/day. Using 2 L/day as an estimate of water consumption, water nitrate would have contributed an average of 6% of total nitrite intake for the subset of women for whom nitrates could be measured in their usual source of drinking water during the periconceptional period. Therefore, it is unlikely that the amount of nitrate in drinking water directly contributed to the increased risk estimate found among women with nitrosatable drug exposure and higher nitrate levels in their drinking water.

Low participation rates in the water nitrate study, as well as measurement of drinking water nitrates after delivery rather than during the periconceptional period, could have introduced some bias. Participants in this portion of the study were similar in demographic characteristics to all study participants with the exception of a higher education level among case-women participants in the water nitrate study compared with case women who did not participate. Although maternal education was a covariate in the risk estimates, other unmeasured factors related to participation might have biased the odds ratios for NTDs associated with nitrosatable drug exposure in relation to drinking water nitrate.

Several other potential limitations should be discussed. Participation was lower among eligible control women than case women in the questionnaire portion of the study (60% vs. 82%, respectively). However, participating control women had education levels similar to all Texas–Mexico border Hispanic women giving birth during the study years (51% and 50%, respectively, with 12 years or more of education). These women also had a similar household income distribution as that found in a 1997 survey of the population living in 6 of the most populous border counties in the study area.41

Case women in the study might have been more likely to recall various medications taken for health problems compared with control women who delivered normal infants. It is unlikely, however, that this potential recall bias would have operated selectively among women who had higher intakes of dietary nitrite and total nitrite. In the lowest tertiles of both dietary nitrite and total nitrite, the odds ratios for nitrosatable drug exposure were 0.8 and 0.9, respectively.

Some misclassification of nitrosatable drug exposure could have occurred. Potential nitrosatable drugs were identified from published studies; however, hundreds of drugs contain nitrosatable groups.3 To explore the impact of this potential source of misclassification, we examined the molecular structure of other medications containing secondary and tertiary amine groups, which are 2 of the more common nitrosatable groups identified in the literature. This further classification yielded an additional 36 mothers (cases and controls) who took potentially nitrosatable medications during the periconceptional period. The overall risk estimate for NTDs with exposure to this expanded group of nitrosatable drugs was slightly less (OR = 2.4; CI = 1.4–4.1) than the estimate obtained when only medications identified from the published literature were used for classification (OR, 2.7).

Misclassification of the types and frequencies of foods eaten during the periconceptional period might also have occurred. Women were asked about various foods and their frequency of consumption during the entire periconceptional period. It is conceivable that types and amounts of food eaten might have changed during the second or third month of pregnancy. Because both case and control women were pregnant, it is likely that any misclassification of foods eaten would have been nonselective and would have minimal effects on risk estimates for nitrosatable drug exposure in each stratum of dietary nitrite and total nitrite intake.

Finally, it is important to note that this study was conducted exclusively among Mexican Americans, who tend to have higher NTD rates than non-Hispanic whites.14,42,43 Intakes of dietary nitrates and nitrites vary among populations. The study population was estimated to have median daily dietary nitrate and nitrite intakes of 87 mg and 4.1 mg, respectively. Estimates of daily dietary intake of nitrates in other populations have ranged from 31 mg in Norway to 245 mg in Italy.22 Estimates of daily dietary nitrite intake range from 0.8 mg in the United States17 to 8.7 mg in Poland.22 In reviewing the relative significance of dietary sources of nitrate and nitrite, White21 estimated the average U.S. resident ingested 4.1 mg nitrite daily. In the United States, vegetables contribute most of the dietary nitrate intake, whereas cured meats and baked goods and cereals contribute most of dietary nitrite intake.17

In conclusion, it could be important to consider the levels of dietary nitrate and nitrite in investigating the effects of nitrosatable drugs on risks for NTDs and other congenital malformations. This is especially important, because levels of nitrate and nitrite from diet vary widely among populations. Findings of this study suggest that if there is risk for NTD-affected pregnancies associated with nitrosatable drug exposure, it could depend on dietary nitrite and total nitrite.

ACKNOWLEDGMENTS

We thank the following NTD Project team members for their crucial role in interviewing the case and control women: El Paso—Hilda Chavarria, Maria Torres, Carmen Ramos, Donna Brom, and Patricia Velazquez; Harlingen—Oralia Villafranca, San Juana Thompson, Graciela Rubio, Manuela Flores, Rene Rodriguez, Sara Mungia, and Jorge Trevino; Laredo—Ricardo Trevino, Miguel Madrigal, Olivia Macias Gutierrez, Cynthia Medina de Llano, Jackie Bassini, and Armandina Ortiz. We also acknowledge Rich Ann Baetz, Kelly Johnson, Hermia Brooks, Billie Woullard, Jennifer Tisch, John Dunn, and Jackie Stroupe for their careful work that ensured the accuracy of data; Scott Simpson for his early contributions as a coinvestigator and staff obstetrician; Russell Larsen for his invaluable role as project director; and Zunera Gilani for her timely and cheerful assistance in data retrieval.

REFERENCES

1.Lijinsky W, Keefer L, Conrad E, Van de Bogart R. Nitrosation of tertiary amines and some biologic implications. J Natl Cancer Inst. 1972;49:1239–1249.
2.Lijinsky W, Conrad E, Van de Bogart R. Carcinogenic nitrosamines formed by drug–nitrite interactions. Nature. 1972;239:165–167.
3.Brambilla G, Cajelli E, Finollo R, Maura A, Pino A, Robbiano L. Formation of DNA-damaging nitroso compounds by interaction of drugs with nitrite. A preliminary screening for detecting hazardous drugs. J Toxicol Environ Health. 1985;15:1–24.
4.Gillatt PN, Palmer RC, Smith PLR, Walters CL. Susceptibilities of drugs to nitrosation under simulated gastric conditions. Food Chem Toxicol. 1985;23:849–855.
5.Gillatt PN, Hart RJ, Walters CL. Susceptibilities of drugs to nitrosation under standardized chemical conditions. Food Chem Toxicol. 1984;22:269–274.
6.Givelber HM, DiPaolo JA. Teratogenic effect of N-ethyl-N-nitrosourea in the Syrian hamster. Cancer Res. 1969;29:1151–1155.
7.Inouye M, Murakami U. Teratogenic effect of N′-methyl-N′-nitro-N-nitrosoguanidine in mice. Teratology. 1978;18:263–268.
8.Koyama T, Handa J, Handa H, Matsumoto S. Methynitrosourea-induced malformations of the brain in SD-JCL rat. Arch Neurol. 1970;22:342–347.
9.Pfaffenroth MJ, Das GD, McAllister JP. Teratologic effects of ethylnitrosourea on brain development in rats. Teratology. 1974;9:305–315.
10.Olshan AF, Faustman EM. Nitrosatable drug exposure during pregnancy and adverse pregnancy outcome. Int J Epidemiol. 1989;18:891–899.
11.Gardner JS, Guyard-Boileau B, Alderman B, Fernbach SK, Greene C, Mangione EJ. Maternal exposure to prescription and non-prescription pharmaceuticals or drugs of abuse and risk of craniosynostosis. Int J Epidemiol. 1998;27:64–67.
12.International Agency for Research on Cancer. General considerations on N-nitrosatable drugs. IARC Monogr Eval Carcinog Risk Hum. 1980;24:297–314.
13.Choi BCK. N-Nitroso compounds and human cancer. Am J Epidemiol. 1985;121:737–743.
14.Hendricks KA, Simpson JS, Larsen RD. Neural tube defects along the Texas–Mexico border, 1993–1995. Am J Epidemiol. 1999;149:1119–1127.
15.Suarez L, Hendricks KA, Cooper SP, Sweeney AM, Hardy RJ, Larsen RD. Neural tube defects among Mexican Americans living on the US–Mexico border: effects of folic acid and dietary folate. Am J Epidemiol. 2000;152:1017–1023.
16.McPherson RS, Kohl HW III, Garcia G, Nichaman MZ, Han CL. Food-frequency questionnaire validation among Mexican-Americans: Starr County, Texas. Ann Epidemiol. 1995;5:378–385.
17.Committee on Nitrite and Alternative Curing Agents in Food. The Health Effects of Nitrate, Nitrite, and N-Nitroso Compounds. Washington, DC: National Academy Press; 1981.
18.Walters CL. Nitrate and nitrite in foods. In: Hill M, ed. Nitrates and Nitrites in Food and Water. New York: Ellis Howard; 1991.
19.Knight TM, Forman D, Al-Dabbagh SA, Doll R. Estimation of dietary intake of nitrate and nitrite in Great Britain. Food Chem Toxicol. 1987;25:277–285.
20.Siciliano J, Krulick S, Heisler EG, Schwartz JH, White JW. Nitrate and nitrite content of some fresh and processed market vegetables. J Agric Food Chem. 1975;23:461–464.
21.White JW. Relative significance of dietary sources of nitrate and nitrite. J Agric Food Chem. 1975;23:886–891.
22.Gangoli SD, van den Brandt PA, Feron VJ, et al. Nitrate, nitrite, and N-nitroso compounds. Eur J Pharmacol. 1994;292:1–38.
23.Dawson BA, Lawrence RC. Analysis of selected drug formulations for volatile nitrosamines. J Assoc Off Anal Chem. 1987;70:554–556.
24.Hutchinson TA, Shahin DR, eds. DRUGDEX System. Greenwood Village, CO: MICROMEDEX; 2003.
25.Murray L, Kelly GL, eds. Physicians’ Desk Reference. Montvale, NJ: Medical Economics Co; 2001.
26.Mehta C, Patel N. LogXact 5. Cambridge, MA: Cytel Software Corp; 2002.
27.Suarez L, Hendricks K, Felkner M, Gunter E. Maternal B12 levels and risk for neural tube defects in a Texas–Mexico border population. Ann Epidemiol. 2003;13:81–88.
28.Hendricks KA, Nuno OM, Suarez L, Larsen R. Effects of hyperinsulinemia and obesity on risk of neural tube defects among Mexican Americans. Epidemiology. 2001;12:630–635.
29.Brender J, Suarez L, Hendricks K, Baetz RA, Larsen R. Parental occupation and neural tube defect-affected pregnancies among Mexican Americans. J Occup Environ Med. 2002;44:650–656.
30.Suarez L, Cardarelli K, Hendricks K. Maternal stress, social support, and risk of neural tube defects among Mexican Americans. Epidemiology. 2003;14:612–616.
31.Graham JM, Jr., Edwards MJ, Edwards MJ. Teratogen update: gestational effects of maternal hyperthermia due to febrile illnesses and resultant patterns of defects in humans. Teratology. 1998;58:209–221.
32.Botto LD, Erickson JD, Mulinare J, Lynberg MC, Liu Y. Maternal fever, multivitamin use, and selected birth defects: evidence of an interaction? Epidemiology. 2002;13:485–488.
33.Shaw GM, Nelson V, Carmichael SL, Lammer EJ, Finnel RH, Rosenquist TH. Maternal periconceptional vitamins: interactions with selected factors and congenital anomalies? Epidemiology. 2002;13:625–630.
34.Croen LA, Todoroff K, Shaw GM. Maternal exposure to nitrate from drinking water and risk for neural tube defects. Am J Epidemiol. 2001;153:325–331.
35.Scragg RK, Dorsch MM, McMichael AJ, Baghurst PA. Birth defects and household water supply. Epidemiological studies in the Mount Gambier region of South Australia. Med J Aust. 1982;2:577–579.
36.Dorsch MM, Scragg RK, McMichael AJ, Baghurst PA, Dyer KF. Congenital malformations and maternal drinking water supply in rural South Australia: a case-control study. Am J Epidemiol. 1984;119:473–486.
37.Ericson A, Kallen B, Lofkist E. Environmental factors in the etiology of neural tube defects: a negative study. Environ Res. 1988;45:38–47.
38.Arbuckle TE, Sherman GJ, Corey PN, Walters D, Lo B. Water nitrates and CNS birth defects: a population-based case-control study. Arch Environ Health. 1988;43:162–167.
39.Klotz JB, Pyrch LA. Neural tube defects and drinking water disinfection by-products. Epidemiology. 1999;10:383–390.
40.Ershow AG, Brown LM, Cantor KP. Intake of tapwater and total water by pregnant and lactating women. Am J Public Health. 1991;81:328–334.
41.Dutton RJ, Weldon M, Shannon J, et al. Survey of Health and Environmental Conditions in Texas Border Counties and Colonias. Austin, TX: Texas Department of Health; June 2000.
42.Brender JD, Carmichael L, Preece MJ, Larimer GC, Suarez L. Epidemiology of anencephaly in Texas, 1981–1986. Tex Med. 1989;85:33–35.
43.Canfield MA, Annegers JF, Brender JD, Cooper SP, Greenberg F. Hispanic origin and neural tube defects in Houston/Harris County, Texas. 1. Descriptive epidemiology. Am J Epidemiol. 1996;143:1–11.
© 2004 Lippincott Williams & Wilkins, Inc.