association between organic dietary choice during pregnancy and hypospadias in offspring: a study of...

6
Association Between Organic Dietary Choice During Pregnancy and Hypospadias in Offspring: A Study of Mothers of 306 Boys Operated on for Hypospadias Jeppe Schultz Christensen, Camilla Asklund, Niels E. Skakkebæk, Niels Jørgensen, Helle Raun Andersen, Troels Munch Jørgensen, Lars Henning Olsen, Anette Pernille Høyer, Jan Moesgaard, Jørgen Thorup and Tina Kold Jensen From the Department of Growth and Reproduction (JSC, CA, NES, NJ, TKJ), Department of Plastic Surgery, Breast Surgery and Burns Treatment (CA), and Department of Pediatric Surgery (JM, JT), Rigshospitalet, Copenhagen, Department of Environmental Medicine, University of Southern Denmark, Odense (HRA, TKJ), Department of Urology, Aarhus University Hospital–Skejby, Aarhus (TMJ, LHO) and Department of Plastic Surgery, Copenhagen University Hospital–Herlev, Herlev (APH), Denmark Purpose: The etiology of hypospadias is poorly understood. Exposure to pesti- cides has been considered a risk factor, although findings are inconsistent. Diet constitutes a significant exposure route for pesticides, and pesticide residues are more frequently reported in conventional than organic food products. We exam- ined the association between organic dietary choice during pregnancy and pres- ence of hypospadias in the offspring. Materials and Methods: Mothers of 306 boys operated on for hypospadias were frequency matched for geography and child birth year to 306 mothers of healthy boys in a case-control study. Telephone interviews were conducted regarding demographic and lifestyle factors, including intake and organic choice of selected food items (milk, dairy products, egg, fruit, vegetables and meat). Logistic regres- sion models were constructed for dietary variables, and odds ratios were calcu- lated controlling for maternal age, body mass index and alcohol consumption. Results: Overall organic choice of food items during pregnancy was not associ- ated with hypospadias in the offspring. However, frequent current consumption of high fat dairy products (milk, butter) while rarely or never choosing the organic alternative to these products during pregnancy was associated with increased odds of hypospadias (adjusted OR 2.18, 95% CI 1.09 – 4.36). Conclusions: This large case-control study of boys operated on for hypospadias suggests an association between hypospadias in the offspring and the mother not choosing the organic alternative, and having a high current intake of nonorganic butter and cheese. This finding could be due to chemical contamination of high fat dairy products. However, general lifestyle and health behavior related to choos- ing organic alternatives may also explain the finding. Key Words: diet; food, organic; hypospadias; maternal exposure; pesticides Abbreviations and Acronyms BMI body mass index EDC endocrine disrupting chemical SES socioeconomic status Accepted for publication September 24, 2012. Supported by the Danish Council for Strategic Research, Program Commission on Health, Food and Welfare project No. 2101-08-0058, European Union DEER Grant agreement No. 212844, and Kirsten and Freddy Johansens Foundation Grant 95-103-72087. See Editorial on page 798. HYPOSPADIAS is among the most com- mon congenital anomalies, with a prevalence of up to 4.6% at age 3 years in prospective studies. 1 Various risk factors have been associated with hypospadias, such as low birth weight, low gestational age, and maternal age and education. 2–4 Furthermore, fa- milial clustering implies a strong ge- netic component in the etiology of hy- pospadias. 5–7 A scientific focus has been on expo- sure to EDCs during early fetal life. EDCs mimic, block or alter the action 0022-5347/13/1893-1077/0 http://dx.doi.org/10.1016/j.juro.2012.09.116 THE JOURNAL OF UROLOGY ® Vol. 189, 1077-1082, March 2013 © 2013 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH,INC. Printed in U.S.A. www.jurology.com 1077

Upload: tina-kold

Post on 23-Dec-2016

212 views

Category:

Documents


0 download

TRANSCRIPT

Association Between Organic Dietary Choice During Pregnancy

and Hypospadias in Offspring: A Study of Mothers of 306 Boys

Operated on for Hypospadias

Jeppe Schultz Christensen, Camilla Asklund, Niels E. Skakkebæk, Niels Jørgensen,Helle Raun Andersen, Troels Munch Jørgensen, Lars Henning Olsen,Anette Pernille Høyer, Jan Moesgaard, Jørgen Thorup and Tina Kold JensenFrom the Department of Growth and Reproduction (JSC, CA, NES, NJ, TKJ), Department of Plastic Surgery, Breast Surgery and BurnsTreatment (CA), and Department of Pediatric Surgery (JM, JT), Rigshospitalet, Copenhagen, Department of Environmental Medicine,University of Southern Denmark, Odense (HRA, TKJ), Department of Urology, Aarhus University Hospital–Skejby, Aarhus (TMJ, LHO) andDepartment of Plastic Surgery, Copenhagen University Hospital–Herlev, Herlev (APH), Denmark

Purpose: The etiology of hypospadias is poorly understood. Exposure to pesti-cides has been considered a risk factor, although findings are inconsistent. Dietconstitutes a significant exposure route for pesticides, and pesticide residues aremore frequently reported in conventional than organic food products. We exam-ined the association between organic dietary choice during pregnancy and pres-ence of hypospadias in the offspring.Materials and Methods: Mothers of 306 boys operated on for hypospadias werefrequency matched for geography and child birth year to 306 mothers of healthyboys in a case-control study. Telephone interviews were conducted regardingdemographic and lifestyle factors, including intake and organic choice of selectedfood items (milk, dairy products, egg, fruit, vegetables and meat). Logistic regres-sion models were constructed for dietary variables, and odds ratios were calcu-lated controlling for maternal age, body mass index and alcohol consumption.Results: Overall organic choice of food items during pregnancy was not associ-ated with hypospadias in the offspring. However, frequent current consumptionof high fat dairy products (milk, butter) while rarely or never choosing the organicalternative to these products during pregnancy was associated with increasedodds of hypospadias (adjusted OR 2.18, 95% CI 1.09–4.36).Conclusions: This large case-control study of boys operated on for hypospadiassuggests an association between hypospadias in the offspring and the mother notchoosing the organic alternative, and having a high current intake of nonorganicbutter and cheese. This finding could be due to chemical contamination of high fatdairy products. However, general lifestyle and health behavior related to choos-ing organic alternatives may also explain the finding.

Abbreviations

and Acronyms

BMI � body mass index

EDC � endocrine disruptingchemical

SES � socioeconomic status

Accepted for publication September 24, 2012.Supported by the Danish Council for Strategic

Research, Program Commission on Health, Foodand Welfare project No. 2101-08-0058, EuropeanUnion DEER Grant agreement No. 212844, andKirsten and Freddy Johansens Foundation Grant95-103-72087.

See Editorial on page 798.

Key Words: diet; food, organic; hypospadias; maternal exposure; pesticides

HYPOSPADIAS is among the most com-mon congenital anomalies, with aprevalence of up to 4.6% at age 3years in prospective studies.1 Variousrisk factors have been associated withhypospadias, such as low birth weight,

low gestational age, and maternal age

0022-5347/13/1893-1077/0THE JOURNAL OF UROLOGY®

© 2013 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RES

and education.2–4 Furthermore, fa-milial clustering implies a strong ge-netic component in the etiology of hy-pospadias.5–7

A scientific focus has been on expo-sure to EDCs during early fetal life.

EDCs mimic, block or alter the action

http://dx.doi.org/10.1016/j.juro.2012.09.116Vol. 189, 1077-1082, March 2013

EARCH, INC. Printed in U.S.A.www.jurology.com 1077

ORGANIC DIET DURING PREGNANCY AND HYPOSPADIAS IN OFFSPRING1078

of endocrine hormones and, therefore, can disturburogenital development. Numerous pesticides areknown to have endocrine disrupting abilities, andsome studies have demonstrated associations be-tween parental pesticide exposure and hypospadiasin the offspring,3,8–12 while others have not.13 Diet isconsidered to constitute the main exposure route forpesticides.14 However, few studies of hypospadiashave examined diet as a potential source of exposureto EDCs, and the results are conflicting. Pesticideresidues are more frequently reported in conven-tional foods than in organic foods.15

Organic alternatives to conventional food prod-ucts are widely available to most consumers in Den-mark, where this study was conducted, and mostincome groups have access to the most basic organicproducts. The controls with organic farmers and pro-ducers are strict. The Danish monitoring program,which monitors foreign as well as domestic products,rarely detects pesticide residues in organic products(and low levels when they do) compared to conven-tional products.15 To examine whether maternalchoice of organic food items is associated with hy-pospadias in the male offspring, we conducted acase-control study among 306 boys operated on forhypospadias and 306 healthy controls.

MATERIALS AND METHODS

Data CollectionMothers of all boys (total 437) with isolated hypospadiasreferred for surgery at 3 university hospitals in Denmark(Skejby, Herlev and Rigshospitalet) between 2003 and2005 were offered participation in the study. Cases re-ferred before 2004 (139) and their controls were includedretrospectively, and information on hypospadias severitywas collected through medical records. Information onhypospadias severity for cases included prospectively from2004 to 2005 was collected by the examining surgeon.

For each case a control was randomly selected from theCivil Registration System and frequency matched for yearof birth and postal code to the case until 1:1 matching wasachieved (485 mothers were invited). Mothers who did notspeak Danish were not offered participation in the study.

All participating mothers were telephone interviewedin 2005 or 2006 by trained interviewers. Maternal BMIwas calculated using self-reported information on weightbefore pregnancy and current height. Information on par-ity, maternal age at birth and birth weight was obtained.Maternal alcohol consumption during the first 3 months ofpregnancy was calculated as a cumulated measure ofweekly number of units of wine, beer and spirit. Maternalsmoking was determined as smoking during pregnancy.Maternal educational level was categorized as no educa-tion (only compulsory school), short education (trainee,unskilled worker, skilled worker, other type of worker andshort education) or medium to long education (3 years ormore). Mothers also provided information on whether the

biological father had ever been diagnosed with a reproduc-

tive disorder (testicular cancer, cryptorchidism or hypo-spadias). Use of assisted reproductive technology was de-fined as undergoing insemination, in vitro fertilization orintracytoplasmic sperm injection to conceive.

Organic Choice and Dietary HabitsThe mothers were asked about choice of organic food itemsin the first trimester for milk, other dairy products, eggs,meat, fruit and vegetables. Responses consisted of often,sometimes, rarely and never. Variables were recoded tobinary variables (often/sometimes or rarely/never). A cu-mulated measure of organic choice was computed, addingup how many products the mothers chose often or some-times (ranging from 0 to 6).

Since we did not believe the mothers would recall de-tailed dietary habits during pregnancy several years ear-lier, we asked about current dietary habits, assumingthese to be a proxy for pregnancy habits for cheese, eggs,butter, chicken or turkey, beef or veal, lamb or pork,vegetables and fruit. Answer categories were never, 1 to 3times monthly, once weekly, 2 to 3 times weekly, 4 to 6times weekly, once daily and more than once daily.

Data AnalysisAll analyses were performed using SPSS®, version 18.0.The distributions of all relevant variables were examinedusing frequency tables, and the associations with hypo-spadias were initially examined using cross tables. Fur-ther analysis of the associations between maternal dietaryhabits during pregnancy and hypospadias was performedusing binary logistic regression, calculating OR and cor-responding 95% CI.

Potential confounders were chosen based on publishedliterature and confirmed in our data using chi-square testsand binary logistic regression. All final models were ad-justed for the same characteristics, ie maternal age, alco-hol consumption during the first trimester and BMI. Ma-ternal education was associated with hypospadias andorganic choice. Nevertheless, we did not adjust for educa-tion since this may be an indicator of a healthy lifestyleand a proxy for dietary habits, but we repeated the anal-yses with adjustment.

The first multivariate model included variables on or-ganic choice during pregnancy. Since mothers could reportchoosing the organic alternative often, while rarely con-suming the actual product, we constructed a second modelcontaining variables on choice of organic products duringpregnancy combined with current intake of the product.For example choice of organic nonmilk dairy productsduring pregnancy was combined with current consump-tion of high fat dairy products (cheese and butter) andcategorized as low consumption (butter and cheese lessthan once daily) and medium (daily) to high consumption(more than once daily). Theses categories were dividedbased on the frequency of organic choice of nonmilk dairyproducts of the mother during pregnancy (often/some-times or rarely/never), yielding 6 categories in all. Similarvariables were constructed for organic milk, eggs, meat,fruit and vegetables combined with the current intake ofthe products. Finally, the analyses were repeated afterexclusion of boys whose fathers had been diagnosed with a

reproductive disorder to exclude genetic contribution to

ORGANIC DIET DURING PREGNANCY AND HYPOSPADIAS IN OFFSPRING 1079

the association, and subanalyses for the different subtypesof hypospadias were performed.

RESULTS

The participation rates for cases (70%) were slightlyhigher than for controls (63%). Information on se-verity of hypospadias was available for 269 casesand revealed glanular location in 14.5%, coronal in41.6%, corporeal in 34.9% and penoscrotal, scrotal orperineal in 8.9%.

Table 1 outlines the characteristics of cases and

Table 1. Associations between maternal pregnancyexposures, maternal and birth characteristics, and paternalreproductive health and hypospadias

No. Cases(%)

No. Controls(%)

Unadjusted OR(95% CI)

Birth weight (gm):Less than 2,500 41 (13.4) 10 (3.3) 4.47 (2.19–9.11)2,500–4,500 254 (83.3) 277 (90.8) ReferenceGreater than 4,500 11 (3.6) 18 (5.9) 0.67 (0.31–1.44)

Gestational age (wks):Less than 37 45 (14.8) 16 (5.2) 3.13 (1.73–5.67)37 or More 260 (85.2) 289 (94.8) Reference

Parity:Nullipara 157 (51.3) 131 (42.8) 1.41 (1.02–1.94)Multipara 149 (48.7) 175 (57.2) Reference

Maternal age (yrs):*Younger than 25 24 (7.8) 19 (6.2) Reference25–29 115 (37.6) 102 (33.3) 0.89 (0.46–1.72)30–34 126 (41.2) 128 (41.8) 0.78 (0.41–1.49)35 or Older 41 (13.4) 57 (18.6) 0.57 (0.28–1.17)

Maternal BMI (kg/m2):Less than 19 29 (9.6) 17 (5.8) 1.63 (0.82–3.25)19–25 206 (68.2) 214 (72.5) Reference25 or Greater 67 (22.2) 64 (21.7) 0.92 (0.61–1.36)

Maternal education:None 34 (11.1) 29 (9.5) 1.48 (0.86–2.55)Short 140 (45.9) 112 (36.6) 1.57 (1.12–2.21)Medium/long 131 (43.0) 165 (53.9) Reference

Maternal alcoholconsumption during firsttrimester (units/wk):

Less than 1 220 (71.9) 212 (69.5) Reference1–2 56 (18.3) 69 (22.6) 0.78 (0.52–1.17)3 or More 30 (9.8) 24 (7.9) 1.21 (0.68–2.13)

Maternal smoking duringpregnancy:

No 245 (80.3) 233 (76.1) ReferenceYes 60 (19.7) 73 (23.9) 0.78 (0.53–1.15)

Paternal reproductivedisorders:†

No 267 (87.3) 293 (95.8) ReferenceYes 39 (12.7) 13 (4.2) 3.29 (1.72–6.30)

Use of assistedreproductive technology:

No 276 (90.2) 291 (95.1) ReferenceYes 30 (9.8) 15 (4.9) 2.11 (1.11–4.00)

* Maternal age was statistically significantly associated with hypospadias as acontinuous variable, tending toward decreased risk at greater age.† Testicular cancer, cryptorchidism or hypospadias.

controls in relation to maternal exposures during

pregnancy, birth characteristics and paternal repro-ductive disorders, and the unadjusted odds ratios forconceiving a boy with hypospadias. The presence ofa reproductive disorder in the father was stronglyassociated with hypospadias in the offspring. Birthweight, gestational age, parity and use of assistedreproductive technology were all statistically signif-icantly associated with hypospadias. Maternal agewas also associated with hypospadias in the off-spring when analyzed as a continuous variable,tending toward lower risk at higher ages. Motherswho did not choose the organic alternative duringpregnancy were slightly older, had a higher BMI,were more often of lower education and drank lessalcohol than mothers choosing the organic alterna-tive (data not shown).

Table 2 outlines the odds ratio of giving birth to aboy with hypospadias in relation to maternal or-ganic choice during pregnancy. Rare choice of or-ganic nonmilk dairy products was associated withhigher unadjusted odds of conceiving a boy withhypospadias. However, the association was not sta-tistically significant after adjustment (OR 1.36, 95%CI 0.95–1.94). A similar association was observed formothers rarely or never choosing organic eggs (OR

Table 2. Association between maternal dietary habits duringfirst trimester and hypospadias in offspring

Maternal Choice ofOrganic Alternative

No. Controls(%)

No. Cases(%)

Unadjusted OR(95% CI)

Adjusted OR(95% CI)*

Fruits:Often/sometimes 121 (39.8) 111 (36.5) Reference ReferenceRarely/never 183 (60.2) 193 (63.5) 1.15 (0.83–1.60) 1.08 (0.77–1.53)

Vegetables:Often/sometimes 151 (49.5) 139 (45.7) Reference ReferenceRarely/never 154 (50.5) 165 (54.3) 1.16 (0.85–1.60) 1.10 (0.79–1.53)

Milk:Often/sometimes 155 (50.7) 139 (46.0) Reference ReferenceRarely/never 151 (49.3) 163 (54.0) 1.20 (0.88–1.66) 1.10 (0.79–1.54)

Dairy products:†Often/sometimes 115 (37.6) 88 (29.1) Reference ReferenceRarely/never 191 (62.4) 214 (70.9) 1.46 (1.04–2.06) 1.36 (0.95–1.94)

Eggs:Often/sometimes 162 (53.5) 139 (46.0) Reference ReferenceRarely/never 141 (46.5) 163 (54.0) 1.35 (0.98–1.86) 1.28 (0.92–1.79)

Meat:Often/sometimes 100 (32.9) 85 (28.1) Reference ReferenceRarely/never 204 (67.1) 218 (71.9) 1.26 (0.89–1.78) 1.14 (0.80–1.64)

Cumulated choiceof organicproducts:‡

6 52 (17.1) 44 (14.5) Reference Reference5 26 (8.6) 29 (9.6) 1.32 (0.68–2.56) 1.24 (0.63–2.44)4 39 (12.8) 20 (6.6) 0.61 (0.31–1.19) 0.58 (0.29–1.17)3 38 (12.5) 42 (13.9) 1.31 (0.72–2.37) 1.20 (0.65–2.22)2 27 (8.9) 28 (9.2) 1.23 (0.63–2.38) 1.12 (0.57–2.23)1 40 (13.2) 37 (12.2) 1.09 (0.60–1.99) 0.98 (0.52–1.85)0 82 (27.0) 103 (34.0) 1.48 (0.91–2.44) 1.31 (0.78–2.21)

* Adjusted for maternal age, alcohol consumption during first trimester and BMI.† Nonmilk dairy products.‡ Cumulated measure of organic choices, indicating how many products mother

often or sometimes chose organic alternative to.

ORGANIC DIET DURING PREGNANCY AND HYPOSPADIAS IN OFFSPRING1080

1.28, 95% CI 0.92–1.79). The cumulated choice oforganic products exhibited no statistically signifi-cant association or pattern. However, mothers whonever or rarely chose any organic products had non-significant increased odds of giving birth to a boywith hypospadias (adjusted OR 1.31, 95% CI 0.78–2.21).

The choice of the organic alternative is irrelevantif the frequency of intake of the food item is low and,therefore, we combined choice and frequency of con-sumption. Mothers who currently consumed butteror cheese more than once daily and who less oftenchose the organic alternative during pregnancy hadsignificantly increased odds of conceiving a boy withhypospadias (adjusted OR 1.87, 95% CI 1.01–3.48)compared to the reference group. An even higherodds ratio was seen for mothers who rarely or neverchose the organic alternative during pregnancy andcurrently consumed butter and cheese more thanonce daily (OR 2.18, 95% CI 1.09–4.36, table 3). Noconsistent patterns were observed for organic choiceof eggs, meat, milk, fruit and vegetables duringpregnancy and current consumption of these fooditems (data not shown). However, mothers whorarely or never chose the organic alternative to eggsand consumed eggs more than 3 times weekly hadnonsignificantly increased odds of conceiving a boywith hypospadias (adjusted OR 1.34, 95% CI 0.82–2.19).

Similar associations were found after exclusion ofboys whose fathers had been diagnosed with a re-productive disorder (eg mothers rarely or neverchoosing organic nonmilk dairy products adjustedOR 1.44, 95% CI 0.98–2.07). Additional analyses ofthe different subtypes of hypospadias yielded sim-ilar results. In addition, adjusting for maternaleducation did not change the direction of the find-ings.

DISCUSSION

In this case-control study of 306 boys with hypospa-dias and 306 controls organic choice of meat, egg,

Table 3. Combined choice of organic alternative to nonmilk dabutter and cheese

Choice of Organic Alternative toNonmilk Dairy Products During

First TrimesterCurrent Daily Consumption of

Butter � Cheese No

Often/Sometimes Both less than 1Butter or cheese 1 or moreBoth 1 or more

Rarely/Never Both less than 1Butter or cheese 1 or more

Both 1 or more 52 (17.2

milk, fruit and vegetables during pregnancy was notassociated with hypospadias. However, rarely ornever choosing the organic alternative to nonmilkdairy product or eggs was associated with an in-creased odds ratio of giving birth to a boy withhypospadias (difference not significant after adjust-ing for confounders). Frequent current butter andcheese consumption combined with rarely or neverchoosing the organic alternative during pregnancywas associated with higher odds of giving birth to aboy with hypospadias. However, rare choice of or-ganic dairy products during pregnancy and infre-quent current consumption of butter and cheesewere also associated with hypospadias (differencenot significant), suggesting that general lifestyleand health behavior related to choosing organicsmay be associated with hypospadias.

To our knowledge no other study has examinedthe effect of maternal organic diet during pregnancyon the risk of hypospadias in the offspring. A recentlarge case-control study found that a nonhealth con-scious maternal diet pattern, including infrequentconsumption of organic products, at interview wasassociated with an increased risk of hypospadias inthe offspring.16 One prospective study that included51 cases attempted to investigate the effect of organ-ically grown vegetables but was inconclusive.17 Buy-ing fruit at the market during pregnancy comparedto growing it at home has been associated with hy-pospadias in the offspring.18 We found no associa-tion for choice of organic fruit.

Nonorganic food items may contain traces of pes-ticides,19,20 and maternal exposure to several agri-cultural pesticides is known to be associated withhypospadias in the male offspring of laboratory an-imals.21 In epidemiological studies parental expo-sure to pesticides through work and residence hasbeen associated with hypospadias.3,8–10,12 Further-more, an organic diet has been demonstrated to beassociated with lower concentrations of pesticidemetabolites in the urine.22 Also, substituting con-ventional fruits and vegetables with organic de-creases the amount of detectable nonpersistent pes-

ducts during pregnancy and current maternal consumption of

(%) No. Controls (%) Unadjusted OR (95% CI) Adjusted OR (95% CI)

) 37 (12.1) Reference Reference) 40 (13.1) 1.80 (0.90–3.63) 1.84 (0.90–3.77)) 38 (12.4) 1.41 (0.68–2.92) 1.35 (0.64–2.84)

) 53 (17.3) 1.82 (0.93–3.53) 1.68 (0.85–3.34)) 97 (31.7) 2.10 (1.14–3.86) 1.87 (1.01–3.48)

iry pro

. Cases

20 (6.639 (12.929 (9.6

52 (17.2110 (36.4

) 41 (13.4) 2.35 (1.19–4.64) 2.18 (1.09–4.36)

ORGANIC DIET DURING PREGNANCY AND HYPOSPADIAS IN OFFSPRING 1081

ticides in the urine of Californian children.14 Thesestudies suggest that conventional diets are a sourceof exposure to pesticides. Pesticide residues in feed-stuff supplied to nonorganic dairy cattle can betransferred to the milk,23,24 and the more lipophilic,the greater the risk that these pesticides will beconcentrated in dairy products with high fat con-tent, such as butter and cheese. Several of thesesubstances have endocrine disrupting properties,25

and the presence of these compounds in high fatmilk products could help explain the observed asso-ciations.

The major strengths of this study were its sizeand the extensive interviewer based questionnaire,securing high validity of exposure classification.Only boys referred for surgery were included, secur-ing high validity of disease classification, althoughmild glanular cases were excluded, and if thesecases are related to maternal organic choice, it wasnot detected in this study. Participation rates were70% for cases and 63% for controls, making self-selection a possible source of bias.

The choice of the organic alternative sometimesor rarely is not particularly accurate and may bemisclassified but will most likely induce nondiffer-ential misclassification and cause an underestima-tion of the true associations. In addition, recall biasis possible. However, while the mothers might beaware of the general benefits of an organic diet, theywere unaware of any associations between intake oforganic products and hypospadias, thereby minimiz-ing recall bias. The mothers were asked about theircurrent intake of some food items, including cheeseand butter consumption, as we do not believe thatthey would accurately recall their intake duringpregnancy years earlier. However, we believe thatchoice of organic food items during pregnancy maybe more accurately recalled, as this is not a quanti-tative measure, but a more general lifestyle choice.Dietary habits at interview may not accurately re-

flect habits during pregnancy. Women may become

REFERENCES

Agricultural pesticide use and hypospadias in 1986; 23: 333.

more aware of their diet during pregnancy andchange it. Furthermore, dietary habits may changewith time, becoming more likely as recall time in-creases. We stratified the analyses on recall time,which did not change the direction of the estimates,implying that long recall time was not a majorsource of information bias.

Finally, residual confounding is possible. Womenmay be exposed to pesticides from agriculture neartheir residence, and we were not able to adjust forarea of residence during pregnancy. We also did notadjust for SES, as this factor is a proxy for environ-mental exposures and health behaviors, includingdietary habits.26 By controlling for SES the effect ofdietary habits may vanish, as it may be an interme-diate factor between SES and hypospadias. The ex-clusion of a socioeconomic confounder contrasts withprior studies,18,27 and residual confounding by SESis possible. However, adjusting for maternal educa-tion as a proxy for socioeconomic status did notchange the observed associations.

CONCLUSIONS

Maternal choice of the organic alternative to fooditems during pregnancy was not associated withhypospadias in the offspring. However, rare choiceof organic nonmilk dairy products or eggs was asso-ciated with hypospadias, although the differencewas not statistically significant. Frequent currentconsumption of high fat dairy products and rarely ornever choosing the organic alternative to these prod-ucts during pregnancy were associated with in-creased odds of giving birth to a boy with hypospa-dias. We speculate that conventional dairy productsmay be contaminated with chemicals, possibly lipo-philic pesticides with endocrine disrupting abilities.However, general lifestyle and health behavior re-lated to choosing organic alternatives may contrib-ute to the findings. Future studies should includemore detailed information on dietary habits or mea-

surement of biomarkers for exposures.

1. Boisen KA, Chellakooty M, Schmidt IM et al:Hypospadias in a cohort of 1072 Danish newbornboys: prevalence and relationship to placentalweight, anthropometrical measurements at birth,and reproductive hormone levels at three monthsof age. J Clin Endocrinol Metab 2005; 90: 4041.

2. Pierik FH, Burdorf A, Deddens JA et al: Maternaland paternal risk factors for cryptorchidism andhypospadias: a case-control study in newbornboys. Environ Health Perspect 2004; 112: 1570.

3. Meyer KJ, Reif JS, Veeramachaneni DN et al:

eastern Arkansas. Environ Health Perspect 2006;114: 1589.

4. Fisch H, Golden RJ, Libersen GL et al: Maternalage as a risk factor for hypospadias. J Urol 2001;165: 934.

5. Monteleone NR, Castilla EE and Paz JE: Hypo-spadias: an epidemiological study in Latin Amer-ica. Am J Med Genet 1981; 10: 5.

6. Calzolari E, Contiero MR, Roncarati E et al: Aeti-ological factors in hypospadias. J Med Genet

7. Asklund C, Jørgensen N, Skakkebaek NE et al:Increased frequency of reproductive health prob-lems among fathers of boys with hypospadias.Hum Reprod 2007; 22: 2639.

8. Kristensen P, Irgens LM, Andersen A et al: Birthdefects among offspring of Norwegian farmers,1967–1991. Epidemiology 1997; 8: 537.

9. Bianca S, Li VG, Caruso-Nicoletti M et al: Ele-vated incidence of hypospadias in two Siciliantowns where exposure to industrial and agricul-tural pollutants is high. Reprod Toxicol 2003; 17:

539.

ORGANIC DIET DURING PREGNANCY AND HYPOSPADIAS IN OFFSPRING1082

10. Carbone P, Giordano F, Nori F et al: Cryptorchid-ism and hypospadias in the Sicilian district ofRagusa and the use of pesticides. Reprod Toxicol2006; 22: 8.

11. Dugas J, Nieuwenhuijsen MJ, Martinez D et al:Use of biocides and insect repellents and risk ofhypospadias. Occup Environ Med 2010; 67: 196.

12. Rocheleau CM, Romitti PA and Dennis LK: Pesti-cides and hypospadias: a meta-analysis. J PediatrUrol 2009; 5: 17.

13. Rocheleau CM, Romitti PA, Sanderson WT et al:Maternal occupational pesticide exposure andrisk of hypospadias in the National Birth DefectsPrevention Study. Birth Defects Res A Clin MolTeratol 2011; 91: 927.

14. Lu C, Barr DB, Pearson MA et al: Dietary intakeand its contribution to longitudinal organophos-phorus pesticide exposure in urban/suburbanchildren. Environ Health Perspect 2008; 116: 537.

15. Danish Institute for Food and Veterinary Re-search: Chemical Contaminants: Food Monitoring,1998 –2003. Part 2: Pesticides. Soborg, Denmark:

Danish Veterinary and Food Administration 2005.

16. de Kort CA, Nieuwenhuijsen MJ and MendezMA: Relationship between maternal dietary pat-terns and hypospadias. Paediatr Perinat Epide-miol 2011; 25: 255.

17. North K and Golding J: A maternal vegetariandiet in pregnancy is associated with hypospadias.The ALSPAC Study Team. Avon LongitudinalStudy of Pregnancy and Childhood. BJU Int 2000;85: 107.

18. Giordano F, Carbone P, Nori F et al: Maternal dietand the risk of hypospadias and cryptorchidism inthe offspring. Paediatr Perinat Epidemiol 2008;22: 249.

19. Lairon D: Nutritional quality and safety of organicfood. A review. Agron Sustain Dev 2010; 30: 33.

20. Baker BP, Benbrook CM, Groth E III et al: Pesti-cide residues in conventional, integrated pestmanagement (IPM)-grown and organic foods: in-sights from three US data sets. Food Addit Con-tam 2002; 19: 427.

21. Baskin LS, Himes K and Colborn T: Hypospadiasand endocrine disruption: is there a connection?

Environ Health Perspect 2001; 109: 1175.

22. Curl CL, Fenske RA and Elgethun K: Organophos-phorus pesticide exposure of urban and suburbanpreschool children with organic and conventionaldiets. Environ Health Perspect 2003; 111: 377.

23. Fernandez-Alvarez M, Lamas JP, Garcia-Chao Met al: Monitoring of pesticide residues in dairycattle farms from NW Spain. J Environ Monit2010; 12: 1864.

24. Melgar MJ, Santaeufemia M and Garcia MA:Organophosphorus pesticide residues in raw milkand infant formulas from Spanish northwest. JEnviron Sci Health B 2010; 45: 595.

25. Norgil Damgaard I, Main KM, Toppari J et al:Impact of exposure to endocrine disrupters inutero and in childhood on adult reproduction.Best Pract Res Clin Endocrinol Metab 2002; 16:289.

26. Mackenbach JP: Health Inequalities: Europe inProfile. Rotterdam, The Netherlands: UniversityMedical Center Rotterdam 2006.

27. Akre O, Boyd HA, Ahlgren M et al: Maternal andgestational risk factors for hypospadias. Environ

Health Perspect 2008; 116: 1071.