ORIGINAL ARTICLE

Maternal sugar consumption and risk of preeclampsia innulliparous Norwegian womenI Borgen1, G Aamodt2, N Harsem1, M Haugen3, HM Meltzer3 and AL Brantsæter3

BACKGROUND/OBJECTIVES: Dietary factors have been hypothesized to influence the risk of preeclampsia. The aim of this studywas to investigate the association between maternal intake of sugar and foods with a high content of added or natural sugars andpreeclampsia.SUBJECTS/METHODS: A prospective study of 32 933 nulliparous women in the Norwegian Mother and Child Cohort Study,conducted by the Norwegian Institute of Public Health. Participants answered a general health questionnaire and a validated foodfrequency questionnaire during pregnancy. Information about preeclampsia was obtained from the Medical Birth Registry ofNorway. The relative risk of preeclampsia was estimated as odds ratios (OR) and 95% confidence intervals (CIs) and adjusted forknown confounders.RESULTS: The intake of added sugar was higher in women who developed preeclampsia than in healthy women in the unadjustedanalysis, but not in the adjusted model. Of food items with a high content of added sugar, sugar-sweetened carbonated and non-carbonated beverages were significantly associated with increased risk of preeclampsia, both independently and combined, withOR for the combined beverages 1.27 (95% CIs: 1.05, 1.54) for high intake (4¼ 125ml/day) compared with no intake. Contrary tothis, intakes of foods high in natural sugars, such as fresh and dried fruits, were associated with decreased risk of preeclampsia.CONCLUSIONS: These results suggest that foods with a high content of added sugar and foods with naturally occurring sugars aredifferently associated with preeclampsia. The findings support the overall dietary advice to include fruits and reduce the intake ofsugar-sweetened beverages during pregnancy.

European Journal of Clinical Nutrition (2012) 66, 920–925; doi:10.1038/ejcn.2012.61; published online 20 June 2012

Keywords: preeclampsia; dietary sugars; sugar-sweetened beverages; food frequency questionnaire; cohort study; pregnancy

INTRODUCTIONPreeclampsia is a syndrome unique for human pregnancy andcharacterized by hypertension and proteinuria in a previouslyhealthy woman. It is a common pregnancy complication and amain cause of morbidity and mortality among both pregnantwomen and their offspring.1 In Norway, the prevalence ofpreeclampsia is about 4%.2 The pathogenesis is still unknown,but disturbed placentation, oxidative stress, inflammatoryresponse to pregnancy and endothelial dysfunction have beenidentified as risk factors.3

The role of maternal diet in the etiology of preeclampsia hasreceived much attention, and many nutritional factors have beensuggested among potential triggers of this syndrome.4 Differentdietary components have been investigated as targets forintervention without convincing results.5,6 However, results fromobservational studies indicate that high intake of sugar or sugarcontaining foods and low intake of dietary fiber may influence therisk of preeclampsia.7–9 Possible mechanisms by which dietarycomponents, including sugar, could influence the development ofpreeclampsia have been suggested to be through modification ofendothelial, vascular and inflammatory responses.4,5 In non-pregnant populations, sugar consumption has been associatedwith development of obesity, metabolic and cardiovascular

diseases.10 We have previously examined the overall dietarypatterns and risk of preeclampsia in a large pregnancy cohort inNorway.11 The results showed lower risk of preeclampsia inwomen adhering to a dietary pattern characterized by vegetables,plant foods and vegetable oils and higher risk in women adheringto a pattern characterized by high consumption of processedmeat products, salty snacks and sugar-sweetened beverages. As afollow-up of that study, we wanted to examine more specificallythe intakes of added sugar, mono- and disaccharides and sugar-rich food items.

Sugar belongs to the carbohydrate family and providesvery rapidly available energy. Simple sugars such as sucrose(commercial ‘sugar’), fructose (fruit sugar) or glucose (presentin fruits and also used as an ingredient) are natural ingredients inmany foods, including fruits, but are also added to manufacturedproducts. In sweet beverages and confectionery, the presenceof sugars is obvious, while the occurrence in foods such asyoghurt, breakfast cereals, bread spread and condiments is lessobvious. The aim of this study was to investigate the associationbetween maternal intake of sugar and food items with a highcontent of added or natural sugars and the development ofpreeclampsia in a large prospective cohort of nulliparous womenin Norway.

1Department of Obstetrics and Gynecology, Women and Children’s Division, Oslo University hospital, Oslo, Norway; 2Division of Epidemiology, Norwegian Institute of PublicHealth, Oslo, Norway and 3Division of Environmental Medicine, Norwegian Institute of Public Health, Oslo, Norway. Correspondence: Dr AL Brantsæter, Division of EnvironmentalMedicine, Norwegian Institute of Public Health, Postbox 4404 Nydalen, 0403 Oslo, Norway.E-mail: Anne.Lise.Brantsaeter@fhi.noContributors: IB, GAa and ALB planned the study and carried out the statistical analyses; MH estimated all dietary intakes, NH and HMM contributed to the interpretation of theresults. IB drafted the manuscript. All authors took part in the preparation of the final manuscript.Received 14 December 2011; revised 9 May 2012; accepted 11 May 2012; published online 20 June 2012

European Journal of Clinical Nutrition (2012) 66, 920–925& 2012 Macmillan Publishers Limited All rights reserved 0954-3007/12

www.nature.com/ejcn

SUBJECTS AND METHODSPopulation and study designThe data set is part of the Norwegian Mother and Child Cohort Study(MoBa) initiated by and maintained at the Norwegian Institute of PublicHealth.12 In brief, MoBa is a nation-wide pregnancy cohort that in the yearsfrom 1999 to 2009 included 108 000 pregnancies. Participants wererecruited from all over Norway from 1999–2008, and 38.5% of invitedwomen consented to participate. The women were asked to providebiological samples and to answer three questionnaires duringpregnancy.12 The data included in this study are collected from the twoquestionnaires answered in gestational weeks 15 (Q1) and 18–22 (Q2),respectively. Q2 is a food frequency questionnaire (FFQ), while Q1 is ageneral questionnaire covering health, exposures, lifestyles andbackground factors. Pregnancy and birth records from the Medical BirthRegistry of Norway (MBRN) are linked to the MoBa database.13 Informedconsent was obtained from each participant before the study. The studywas approved by the Norwegian Data Inspectorate and the RegionalCommittee for Ethics in Medical Research.

This study used the quality-assured data files released for research in2009 (version 4). At the time of this analysis 39 199 nulliparous women hadanswered the first MoBa questionnaire (Q1) and were recorded in MBRNwith singleton births. Of these, 34 160 (87%) had also answered version 2of the FFQ and 33 549 had valid energy intakes (total energy between 1080and 4780 kcal/day). The range of acceptable energy intake in MoBa hasbeen evaluated elsewhere.14 Furthermore, we excluded women withchronic hypertension or diabetes mellitus as registered in MBRN (n¼ 616),resulting in a final study sample of 32 933 (84%).

Dietary informationThe dietary data used in this study were collected from February 2002 toNovember 2008. The MoBa FFQ (downloadable at http://www.fhi.no/dokumenter/011fbd699d.pdf) was completed in weeks 17–22 of gestation.This FFQ is a semi-quantitative questionnaire that asks about the intake of255 food items and is designed to capture dietary habits and intake ofdietary supplements during the first 4–5 months of pregnancy.14

Respondents were asked to fill in the mean intake of the food itemseaten since becoming pregnant. The frequency intervals ranged fromnever to 48 times a day. Portion size was only given for units of fruit,bread (slices) and liquids (cups/glasses). The questionnaires were opticallyread. FoodCalc15 and the Norwegian Food Composition Table16 were usedfor calculation intakes of foods and nutrients. The validity of the MoBa FFQhas been thoroughly evaluated.17–19

The Norwegian Food Composition Table16 lists the concentrations ofadded sugar and mono- and disaccharides for all food items separately,although the values for mono- and disaccharides also include added sugarin addition to the naturally occurring sugars such as fructose, maltose,glucose and lactose. Added sugar comprises refined and industrialprocessed sugars such as glucose, sucrose, fructose and glucose sirup.

The daily intakes (g/d) of the 255 food and beverages were aggregatedinto 100 non-overlapping food items/groups based on structure, nutrientprofile or culinary usage.

In this study, we examined 22 of the 100 food groups based on theircontent of added sugar or mono and disaccharides. We included foodgroups with B10 g or more of simple carbohydrates per 100 g food.

Dietary intakes of nutrients and food groups were treated both ascontinuous variables and by ordered categories. The ordered categorieswere based on quintiles for food items with few or no non-consumers.Food items with a high number of non-consumers were treated either asno/yes variable or divided into four groups, one ‘no-intake group’ andtertiles among consumers. In the adjusted analysis, the lowest intake group(representing women with no or low intake of the food item) was used asthe reference category.

PreeclampsiaThe main outcome was preeclampsia as registered in the MBRN, definingpreeclampsia as de novo hypertension and proteinuria in previouslyhealthy women.20 Information provided to the registry was based on formsroutinely completed by midwives shortly after each delivery. Preeclampsiawas divided into subgroups according to early or late onset and clinicalseverity. Preeclampsia in this study includes any of these. The validity ofthe preeclampsia diagnosis in MBRN is presently being assessed, but theresults are not available yet. A recent study from Denmark examined thevalidity of preeclampsia and related diagnoses recorded in a mandatory

Table 1. Dietary intake of added sugar and of mono- anddisaccharides according to maternal characteristics in 32 933nulliparous women in the Norwegian Mother and Child Cohort Study,2002–2008

N % Addedsugar, g/d

Mono- anddisaccharides, g/da

Mean(s.d.)

Mean(s.d.)

All 32 933 100 62 (39) 152 (62)

Maternal age in yearso 25 6110 18.6 75 (51) 165 (76)25–29 16 882 51.3 61 (36) 151 (59)30–34 7955 24.2 57 (33) 147 (57)35þ 1986 6.0 53 (31) 142 (55)P-valueb o0.001 o0.001

Maternal education, yearso 12 1722 5.2 77 (57) 169 (83)12 7614 23.1 70 (47) 160 (72)13–16 14 073 42.7 60 (35) 149 (58)17þ 8803 26.7 56 (30) 148 (54)P-valueb o0.001 o0.001Other/missingdata

721 2.2 63 (41) 153 (67)

Pre-pregnancy BMI, kg/m2

o18.5 1097 3.3 74 (48) 169 (73)18.5–24.9 21 970 66.7 62 (38) 153 (61)25–29.9 6481 19.7 61 (39) 149 (63)30–34.9 1960 6.0 60 (43) 145 (65)35þ 696 2.1 60 (40) 146 (61)P-valueb o0.001 o0.001Missing data 729 2.2 67 (44) 158 (68)

Maternal height (quartiles) (m)o1.65 8883 27.0 63 (41) 152 (65)1.65–1.68 8301 25.2 62 (39) 151 (62)1.69–1.72 7725 23.5 62 (38) 152 (61)41.73 7723 23.5 62 (38) 154 (61)P-valueb 0.234 0.017Missing data 301 0.9 67.0 (46) 163 (69)

Smoking in pregnancyNo 30 210 91.7 61 (37) 151 (61)Occasional 945 2.9 76 (51) 164 (73)Daily 1542 4.7 84 (58) 170 (79)P-valueb o0.001 o0.001Missing data 236 0.7 68 (49) 162 (74)

Exercise during pregnancyNo exercise 3952 12.0 67 (47) 151 (68)oweekly 5679 17.2 64 (40) 149 (63)1–2 times/week 9878 30.0 62 (37) 151 (60)3þ times/week 11 143 33.8 59 (35) 155 (59)P-valueb o0.001 o0.001Missing data 2281 6.9 64 (44) 156 (71)

Dietary fiber intakeQuartile 1 8233 25.0 54 (36) 117 (47)Quartile 2 8233 25.0 57 (36) 136 (48)Quartile 3 8234 25.0 63 (38) 158 (53)Quartile 4 8233 25.0 74 (44) 197 (68)P-valueb o0.001 o0.001

Total energy intakeQuartile 1 8233 25.0 38 (19) 99 (28)Quartile 2 8233 25.0 50 (23) 130 (31)Quartile 3 8234 25.0 64 (28) 158 (36)Quartile 4 8233 25.0 97 (50) 221 (66)P-valueb o0.001 o0.001

aMono and disaccharides include added sugar. bP-value for trend acrossordered categories using linear regression (category with missing data notincluded).

Dietary sugars and preeclampsiaI Borgen et al

921

& 2012 Macmillan Publishers Limited European Journal of Clinical Nutrition (2012) 920 – 925

Danish national discharge registry, which is comparable to the MBRN.The results showed a high positive predictive value of the preeclampsiadiagnosis, and all registrations of serious preeclampsia subtypes reflectedtrue cases.21 We believe similar results will apply to the MBRN.

Other variablesAlthough the aim of this study was to examine the influence of sugar andsugar-rich food items, the total diet needed to be taken into account.Therefore, we included dietary fiber among the confounding variables.Dietary fiber reflects the intake of vegetables, fruits and unrefined grainsand can be viewed as a marker of a healthy diet. Furthermore, we adjustedfor variables known to be associated with the disease outcome; maternalage, smoking habits, maternal education, body mass index (BMI), maternalheight, total energy intake (kcal) and exercise during pregnancy. Maternalage at time of delivery was obtained from MBRN and was grouped intofour categories (o25, 25–29.9, 30–34.9 and 35þ ). Smoking habits duringpregnancy were divided into three categories (non-smokers, occasionalsmokers and daily-smokers), exercise habits into four categories (noexercise, less than once a week, one to two times weekly and three timesweekly or more). Maternal education was divided into four categories(o12 years, 12 years, 13–16 years and 17þ years) and maternal prepregnancy BMI was divided according to the WHO cutoff points for adults.Maternal height was grouped according to quartiles.

Statistical methodsWe used linear regression to find P-value for trend across orderedcategories of maternal characteristics. Differences in dietary intakesbetween women who developed and those who did not developpreeclampsia were examined using independent samples t-tests. Forskewed intake variables, the non-parametric test (Mann–Whitney) resultedin comparable P-values. We estimated odds ratios (OR) and 95%confidence intervals (95% CI) to model the risk of preeclampsia as a

function of the dietary variables and the known confounders. Theconfounding variables were as follows: maternal age at delivery,educational attainment, prepregnant BMI, height, smoking, exercise andintake of dietary fiber and energy. Dietary variables were examined in thelogistic model if intakes differed between women with and withoutpreeclampsia with a P-value o0.1 in bivariate analyses and dietaryvariables with P-values for trend o0.2 are reported. All analyses wereperformed using PASW 17 (SPSS Inc., Chicago, IL, USA).

RESULTSAmong 32 933 nulliparous women, 1703 women (5.2%) developedpreeclampsia. The intakes of added sugar and mono- anddisaccharides differed significantly by maternal characteristics(Table 1). Intake of added sugar decreased with increasing age,education BMI and exercise, and was higher in smokers than innon-smoker. Intake of mono- and disaccharides followed the samepattern except for exercise, for which the intake of mono-and disaccharides was increased with increasing exercise activity(Table 1).

In the crude analyses, we observed that the women whodeveloped preeclampsia had higher intake of added sugar thanthe healthy women, while the intake of total energy and of mono-and disaccharides did not differ between the groups (Table 2).Women who developed preeclampsia had higher intakes of thesugar-sweetened beverages and lower intakes of fresh and driedfruits, honey as spread, jam and sugar or honey in tea or coffee(Table 2).

The results from the adjusted regression analyses are shown inTable 3. We did not observe an increased risk of preeclampsiain women ranked in the highest categories of added sugar.

Table 2. Mean (s.d.) intake of nutrients and food items according to preeclampsia in 32 933 nulliparous women in the Norwegian Mother and ChildCohort Study, 2002–2008

Preeclampsia 1703(5.2%)

No preeclampsia 31 230(94.8%)

P-valuea

Mean (s.d.) Mean (s.d.)

Nutrient intakesAdded sugar, g/d 64.8 (42.9) 61.9 (38.8) 0.009Mono- and disaccharides, g/d 153 (64) 152 (62) 0.624Total energy intake, kcal/d 2291 (618) 2280 (620) 0.455Energy % from added sugar 11.5 (5.7) 11.1 (5.1) 0.005

Food intakes, g/dSugar-sweetened soft beverages, carbonated 102 (233) 84 (189) 0.001Sugar-sweetened non-carbonated beverages, fruit sirup 73 (170) 61 (144) 0.002Sugar/honey on porridge and in tea and coffee 1.0 (2.3) 1.2 (2.7) 0.002Jam 7.8 (12) 8.5 (13) 0.036Honey as bread spread 0.2 (1.0) 0.3 (1.6) o0.001Traditional fruits; apples, oranges, bananas, grapes, peaches, grapefruits, plums 233 (179) 247 (179) 0.003Exotic fruits: papaya, kiwi, mango, melon 19 (40) 19 (38) 0.577Dried fruits: apricots, raisins, prunes, figs, dates 2.4 (6.5) 3.3 (8.7) o0.001Berries; blueberries, strawberries, raspberries 16 (41) 16 (40) 0.938Brown whey cheese 5.9 (10) 6.4 (10) 0.055Nut spread 3.2 (9) 2.8 (8) 0.094Yoghurt, full fat and low fat 71 (111) 72 (103) 0.674Dairy desserts; ice cream, puddings 21 (20) 20 (20) 0.258Corn flakes and sugar-coated cereals 2.3 (7.5) 2.0 (6.3) 0.070Fruit juices 183 (191) 180 (194) 0.510Buns 5.2 (8.2) 5.3 (7.8) 0.497Waffles 9.8 (8.8) 9.4 (8.8) 0.148Cakes 7.9 (7.1) 8.2 (8.6) 0.186Sweet biscuits 2.6 (5.7) 2.8 (6.0) 0.180Chocolate 18 (19) 18 (18) 0.833Caramel sweets 26 (31) 26 (30) 0.924Marzipan 1.4 (4.0) 1.2 (3.3) 0.085

aIndependent samples t-test.

Dietary sugars and preeclampsiaI Borgen et al

922

European Journal of Clinical Nutrition (2012) 920 – 925 & 2012 Macmillan Publishers Limited

However, high intake of either carbonated or non-carbonatedsugar-sweetened beverages was associated with a significantlyincreased risk of preeclampsia. Combining the carbonated andnon-carbonated drinks strengthened the association (Table 3).Food groups with a high content of natural sugars, such as freshfruit and dried fruit, were inversely associated with the risk ofpreeclampsia (Table 3). Adjusting for dietary fiber tended toattenuate the associations. BMI was the variable with the largestinfluence on the estimates and therefore we also examined theassociation between sugar-sweetened beverages and preeclamp-sia by stratifying women by BMI o25 and BMI 25 or above.High intake of sugar-sweetened beverages was associated withincreased risk in both groups, but was stronger in women in thelowest BMI group (crude OR: 1.32 and 1.28, respectively).

Finally, we examined the OR for more extreme intake categoriesof sugar-sweetened beverages by categorizing into no intake,o500 ml, 500–1000 ml and 1000þ ml daily. The OR forconsuming at least 1000 ml daily of non-carbonated sugar-

sweetened beverages was twice as high as for no intake(OR: 2.04; 95% CI: 1.21, 3.45).

DISCUSSIONOur working hypothesis was that consumption of food items witha high content of added sugar was associated with increased riskof preeclampsia. We did not observe any independent associationfor the total estimated intake of added sugar or of mono- anddisaccharides. The hypothesis was confirmed only for high intakeof sugar-sweetened beverages but not for foods with a highcontent of naturally occurring sugars. Our results indicate adifferential influence of naturally occurring and industrially addedsugars. A possible explanation could be that foods with highcontent of natural sugars also have high content of dietary fiber.This is certainly true for dried fruit.

The results of this study resemble those of other studies thathave used a case–control or cohort design. In a case–control study

Table 3. Associations between added sugar and food intakes and preeclampsia in 32 933 nulliparous women in the Norwegian Mother and ChildCohort Study, 2002–2008

Total n (%) Daily intake,range

Preeclampsian (%)

ORa (95% CI) ORb (95% CI) ORc (95% CI)

Added sugar 8233 (25) o36g 453 (5.5) 1 1 18233 (25) 36–53g 376 (4.6) 0.85 (0.74, 0.98) 0.85 (0.74, 0.98) 0.84 (0.73, 0.97)8233 (25) 53–77g 394 (4.8) 0.91 (0.79, 1.05) 0.91 (0.78, 1.05) 0.88 (0.76, 1.03)8233 (25) 77þ g 480 (5.8) 1.12 (0.98, 1.28) 1.10 (0.93, 1.29) 1.04 (0.87, 1.23)

Sugar-sweetened carbonated 7151 (22) No intake 344 (4.8) 1 1 1beverages 8909 (27) 8–30ml 428 (4.8) 1.05 (0.91, 1.22) 1.04 (0.91, 1.22) 1.05 (0.90, 1.21)

7409 (22) 30–70ml 360 (4.9) 1.04 (0.90, 1.22) 1.04 (0.89, 1.21) 1.03 (0.88, 1.20)9464 (29) 70þml 571 (6.0) 1.23 (1.07, 1.42) 1.22 (1.05, 1.40) 1.18 (1.02, 1.37)

Sugar-sweetened non-carbonated 8747 (27) No intake 425 (4.9) 1 1 1beverages 7755 (24) 3–15ml 383 (4.9) 1.05 (0.91, 1.21) 1.05 (0.91, 1.21) 1.05 (0.91, 1.21)

8736 (26) 15–40ml 455 (5.2) 1.10 (0.96, 1.26) 1.09 (0.95, 1.25) 1.08 (0.94, 1.24)7695 (23) 40þ ml 440 (5.7) 1.21 (1.06, 1.39) 1.19 (1.04, 1.37) 1.17 (1.01, 1.35)

Sugar-sweetened beverages 3172 (10) No intake 147 (4.6) 1 1 1combined 103 70 (31) 3–40ml 506 (4.9) 1.10 (0.91, 1.33) 1.10 (0.91, 1.33) 1.09 (0.90, 1.32)

9427 (29) 40–125ml 446 (4.7) 1.05 (0.87, 1.28) 1.05 (0.86, 1.27) 1.03 (0.85, 1.25)9964 (30) 125þ ml 604 (6.1) 1.33 (1.10, 1.61) 1.31 (1.09, 1.59) 1.27 (1.05, 1.54)

Pancakes and waffles 3665 (11) No intake 176 (4.8) 1 1 19761 (30) 1–7g 496 (5.1) 1.08 (0.91, 1.29) 1.08 (0.90, 1.29) 1.08 (0.90, 1.29)9751 (30) 7–12g 478 (4.9) 1.04 (0.87, 1.24) 1.03 (0.86, 1.23) 1.02 (0.85, 1.22)9756 (30) 12þ g 553 (5.7) 1.21 (1.01, 1.44) 1.19 (1.00, 1.42) 1.18 (0.99, 1.41)

Corn flakes and sugar-coatedcereals

23 766 (72) No intake 1174 (4.9) 1 1 1

9167 (28) 1–100g 529 (5.8) 1.14 (1.02, 1.26) 1.13 (1.01, 1.25) 1.11 (1.00, 1.24)

Honey as bread spread 26 988 (80) No intake 1490 (5.5) 1 1 16561 (20) 0.01–50.0 304 (4.6) 0.89 (0.78, 1.02) 0.90 (0.77, 1.01) 0.90 (0.78, 1.01)

Fruits, traditionald 8426 (25) o120g 515 (6.1) 1 1 18384 (25) 120–205g 443 (5.3) 0.91 (0.79, 1.04) 0.89 (0.78, 1.02) 0.90 (0.78, 1.03)8364 (25) 205–330g 439 (5.2) 0.90 (0.79, 1.04) 0.88 (0.76, 1.01) 0.89 (0.77, 1.03)8375 (25) 330þ g 398 (4.7) 0.81 (0.70, 0.93) 0.76 (0.65, 0.88) 0.79 (0.67, 0.93)

Dried fruitse 13 780 (41) No intake 855 (6.2) 1 1 16089 (18) 0.7–1.5 g 316 (5.2) 0.88 (0.77, 1.01) 0.88 (0.77, 1.04) 0.88 (0.77, 1.02)7092 (21) 1.6–3.9 g 343 (4.8) 0.82 (0.72, 0.94) 0.81 (0.71, 0.93) 0.82 (0.72, 0.94)6588 (20) 4.0þ g 280 (4.3) 0.79 (0.68, 0.91) 0.77 (0.66, 0.89) 0.79 (0.68, 0.92)

Abbreviations: CI, confidence interval; OR, odds ratio. aAdjusted for maternal age at delivery, education, prepregnant BMI, height, smoking and leisure exercisein first pregnancy trimester. bAdjusted as above and additionally adjusted for total energy intake. cAdjusted as above and additionally adjusted for total energyintake and dietary fiber. dFruits, traditional: apples, oranges, bananas, grapes, peaches, grapefruits and plums. eDried fruits: apricots, raisins, prunes, figs anddates.

Dietary sugars and preeclampsiaI Borgen et al

923

& 2012 Macmillan Publishers Limited European Journal of Clinical Nutrition (2012) 920 – 925

with 172 preeclamptic and 339 normotensive controls, the authorsreported that consumption of some food items, including fruits,and intakes of magnesium, potassium and total fiber wereassociated with reduced risk of preeclampsia.9 Similar resultswere reported in a prospective study of maternal dietary fiberintake in early pregnancy and subsequent preeclampsia risk.8

Clausen et al.7 found increased risk of preeclampsia in womenwith high intakes of sugar, total energy and polyunsaturated fattyacids in a previous study from Norway. They also reported thathigh intake of sugar-sweetened soft drinks was associated withincreased risk, while no association was observed for intake offruits. In the earlier study from MoBa where we examined the riskof preeclampsia and overall dietary patterns, fruit was among thefood groups with high positive loadings on the pattern associatedwith lower risk while sugar-sweetened drinks were among foodgroups with high positive loadings on the pattern associated withincreased risk of preeclampsia.11 In the present study, theassociation between sugar-sweetened drinks and preeclampsiawas evident independent of adjustment for dietary fiber. Althoughno increased risk was observed for added sugar, these resultsindicated that the observed association for the sweet drinkscannot be explained by their contribution of added sugar.Contrary to the results of the present study, a prospective studyin the US found that chocolate consumption was associated withreduced risk of preeclampsia and gestational hypertension.22

In spite of the associations between sugar-sweetened beveragesand preeclampsia observed in this study it is not possible toestablish any causal implications. Adjusting for dietary fibertended to attenuate the associations between single food groupsand preeclampsia and corroborate that single nutrients orcomponents need to be evaluated in a context of the total diet.Our results support the current dietary advice to reduce the intakeof sugar-sweetened beverage and to include more fruits,vegetables and plant based foods.23 Preeclampsia includesmany pathophysiological mechanisms similar to atherosclerosisand type II diabetes, including oxidative stress, insulin resistance,endothelial dysfunction and inflammation.24–26 In non-pregnantpopulations, diets characterized by low intake of added sugaror by high intake of plant foods have been shown to bene-ficially influence markers of inflammation and endothelialdysfunction.27–29 The scientific evidence has not been sufficientto establish a causal link between high overall sugar consumptionand the development of obesity, metabolic and cardiovasculardiseases.10 However, several studies have reported associationsbetween sugar-sweetened beverages and disease outcomes.30–32

Cohen et al.,33 recently showed that consumption of sweetenedbeverage, but not sugar contributed by the beverage, wasassociated with an increased risk of incident hypertension inthree large prospective US cohorts.

This study has both strengths and limitations. A major strengthis that the women were recruited from all parts of Norway andrepresent diversity with regard to dietary habits, food intakes, ageand sociodemographic variables. Furthermore, information aboutdietary habits was obtained before disease onset, which reducesthe likelihood of misreporting as a consequence of the disease inpreeclamptic women. However, it is not possible to obtain a ‘true’picture of food intakes using a self-reported method such as aFFQ. Misreporting is a serious error in all dietary assessmentmethods.34 It has been shown that food items perceived as‘unhealthy’ are underreported to a larger degree than foodperceived as ‘healthy’.35 This is especially relevant for this study ofsugar and sugar-rich foods, and may explain why no associationwas seen between added sugar and preeclampsia in the adjustedmodel.

In conclusion, the results of the current study suggest adifferential influence of foods with a high content of added sugarand foods with naturally occurring sugars on the risk ofpreeclampsia. It must be taken into account that there was large

heterogeneity between the sugar-rich food groups examined inthe study, but our findings support the overall dietary advice toinclude fruit and to reduce the intake of sugar-sweetenedbeverages during pregnancy.

CONFLICT OF INTERESTThe authors declare no conflict of interest.

ACKNOWLEDGEMENTSWe acknowledge the financial support provided by the Norwegian University of LifeSciences and Sandviks research grant. The Norwegian MoBa is supported by theNorwegian Ministry of Health and the Ministry of Education and Research, NIH/NIEHS(contract no NO-ES-75558), NIH/NINDS (Grant no. 1 UO1 NS 047537-01) and theNorwegian Research Council/FUGE (Grant no. 151918/S10). We are grateful to allthe participating families in Norway who took part in this ongoing cohort study.

REFERENCES1 Lain KY, Roberts JM. Contemporary concepts of the pathogenesis and manage-

ment of preeclampsia. JAMA 2002; 287: 3183–3186.2 Nilsen RM, Vollset SE, Gjessing HK, Skjaerven R, Melve KK, Schreuder P et al. Self-

selection and bias in a large prospective pregnancy cohort in Norway. PaediatrPerinat Epidemiol 2009; 23: 597–608.

3 Redman CW, Sargent IL. Immunology of pre-eclampsia. Am J Reprod Immunol2010; 63: 534–543.

4 Xu H, Shatenstein B, Luo ZC, Wei S, Fraser W. Role of nutrition in the risk ofpreeclampsia. Nutr Rev 2009; 67: 639–657.

5 Roberts JM, Balk JL, Bodnar LM, Belizan JM, Bergel E, Martinez A. Nutrient invol-vement in preeclampsia. J Nutr 2003; 133: 1684S–1692SS.

6 Rumbold A, Duley L, Crowther CA, Haslam RR. Antioxidants for preventingpre-eclampsia. Cochrane Database Syst Rev 2008; CD004227.

7 Clausen T, Slott M, Solvoll K, Drevon CA, Vollset SE, Henriksen T. High intake ofenergy, sucrose, and polyunsaturated fatty acids is associated with increased riskof preeclampsia. Am J Obstet Gynecol 2001; 185: 451–458.

8 Qiu C, Coughlin KB, Frederick IO, Sorensen TK, Williams MA. Dietary fiber intake inearly pregnancy and risk of subsequent preeclampsia. Am J Hypertens 2008; 21:903–909.

9 Frederick IO, Williams MA, Dashow E, Kestin M, Zhang C, Leisenring WM. Dietaryfiber, potassium, magnesium and calcium in relation to the risk of preeclampsia.J Reprod Med 2005; 50: 332–344.

10 Arola L, Bonet ML, Delzenne N, Duggal MS, Gomez-Candela C, Huyghebaert Aet al. Summary and general conclusions/outcomes on the role and fate of sugarsin human nutrition and health. Obes Rev 2009; 10(Suppl 1): 55–58.

11 Brantsæter AL, Haugen M, Samuelsen SO, Torjusen H, Trogstad L, Alexander Jet al. A dietary pattern characterized by high intake of vegetables, fruits, andvegetable oils is associated with reduced risk of preeclampsia in nulliparouspregnant Norwegian women. J Nutr 2009; 139: 1162–1168.

12 Magnus P, Irgens LM, Haug K, Nystad W, Skjaerven R, Stoltenberg C. Cohortprofile: The Norwegian Mother and Child Cohort Study (MoBa). Int J Epidemiol2006; 35: 1146–1150.

13 Irgens LM. The Medical Birth Registry of Norway. Epidemiological research andsurveillance throughout 30 years. Acta Obstet Gynecol Scand 2000; 79: 435–439.

14 Meltzer HM, Brantsæter AL, Ydersbond TA, Alexander J, Haugen M. Methodolo-gical challenges when monitoring the diet of pregnant women in a large study:experiences from the Norwegian Mother and Child Cohort Study (MoBa). MaternChild Nutr 2008; 4: 14–27.

15 Lauritsen J. FoodCalc (online). Available at: http://www.ibt.ku.dk/jesper/foodcalc(accessed February 2006); 2005.

16 Norwegian Food Safety Authority, Norwegian Directorate of Health, Departmentof Nutrition - University of Oslo. Matvaretabellen (The Norwegian Food Compo-sition Table). Oslo: Available online at http://www.matportalen.no/Matvar-etabellen; 2006.

17 Brantsæter AL, Haugen M, Alexander J, Meltzer HM. Validity of a new foodfrequency questionnaire for pregnant women in the Norwegian Mother and ChildCohort Study (MoBa). Matern Child Nutr 2008; 4: 28–43.

18 Brantsæter AL, Haugen M, Rasmussen SE, Alexander J, Samuelsen SO, Meltzer HM.Urine flavonoids and plasma carotenoids in the validation of fruit, vegetable andtea intake during pregnancy in the Norwegian Mother and Child Cohort Study(MoBa). Public Health Nutr 2007; 10: 838–847.

19 Brantsæter AL, Haugen M, Julshamn K, Alexander J, Meltzer HM. Evaluation ofurinary iodine excretion as a biomarker for intake of milk and dairy products in

Dietary sugars and preeclampsiaI Borgen et al

924

European Journal of Clinical Nutrition (2012) 920 – 925 & 2012 Macmillan Publishers Limited

pregnant women in the Norwegian Mother and Child Cohort Study (MoBa). Eur JClin Nutr 2009; 63: 347–354.

20 The Norwegian Society of Obstetrics and Gynecology. Retningslinjer forsvangerskapsomsorgen 2005 (Clinical Guidelines in Obestetrics). Oslo, Norway:Norwegian Medical Association. Available online at: http://www.legeforeningen.no/id/78144.0 (accessed October 2010); 2006.

21 Klemmensen AK, Olsen SF, Østerdal ML, Tabor A. Validity of preeclampsia-relateddiagnoses recorded in a national hospital registry and in a postpartum interviewof the women. Am J Epidemiol 2007; 166: 117–124.

22 Saftlas AF, Triche EW, Beydoun H, Bracken MB. Does chocolate intake duringpregnancy reduce the risks of preeclampsia and gestational hypertension? AnnEpidemiol 2010; 20: 584–591.

23 WHO. Diet nutrition, and the prevention of chronic diseases: report of a joint WHO/FAO expert consultation GenevaWorld Health Organization, 2003.

24 Dusse LM, Rios DR, Pinheiro MB, Cooper AJ, Lwaleed BA. Pre-eclampsia: relationshipbetween coagulation, fibrinolysis and inflammation. Clin Chim Acta 2010; 412: 17–21.

25 Clausen T. Maternal and placental factors and risk of preeclampsia: A prospectivestudy of dietary, lipid, hormonal and inflammatory factors (Doctoral Thesis). Faculty ofMedicine, University of Oslo, 2002.

26 Borzychowski AM, Sargent IL, Redman CW. Inflammation and pre-eclampsia.Semin Fetal Neonatal Med 2006; 11: 309–316.

27 O’Keefe JH, Gheewala NM, O’Keefe JO. Dietary strategies for improving post-prandial glucose, lipids, inflammation, and cardiovascular health. J Am Coll Cardiol2008; 51: 249–255.

28 Nettleton JA, Steffen LM, Mayer-Davis EJ, Jenny NS, Jiang R, Herrington DM et al.Dietary patterns are associated with biochemical markers of inflammation andendothelial activation in the Multi-Ethnic Study of Atherosclerosis (MESA). Am JClin Nutr 2006; 83: 1369–1379.

29 Sofi F, Cesari F, Abbate R, Gensini GF, Casini A. Adherence to Mediterranean dietand health status: meta-analysis. BMJ 2008; 337: a1344.

30 Hu FB, Malik VS. Sugar-sweetened beverages and risk of obesity and type 2diabetes: epidemiologic evidence. Physiol Behav 2010; 100: 47–54.

31 Malik VS, Popkin BM, Bray GA, Despres JP, Hu FB. Sugar-sweetened beverages,obesity, type 2 diabetes mellitus, and cardiovascular disease risk. Circulation 2010;121: 1356–1364.

32 Aeberli I, Gerber PA, Hochuli M, Kohler S, Haile SR, Gouni-Berthold I et al. Low tomoderate sugar-sweetened beverage consumption impairs glucose and lipidmetabolism and promotes inflammation in healthy young men: a randomizedcontrolled trial. Am J Clin Nutr 2011; 94: 479–485.

33 Cohen L, Curhan G, Forman J. Association of sweetened beverage intakewith incident hypertension. J Gen Intern Med; e-pub ahead of print 27April 2012.

34 Westerterp KR, Goris AH. Validity of the assessment of dietary intake: problems ofmisreporting. Curr Opin Clin Nutr Metab Care 2002; 5: 489–493.

35 Olafsdottir AS, Thorsdottir I, Gunnarsdottir I, Thorgeirsdottir H, Steingrimsdottir L.Comparison of women’s diet assessed by FFQs and 24-hour recalls with andwithout underreporters: associations with biomarkers. Ann Nutr Metab 2006; 50:450–460.

Dietary sugars and preeclampsiaI Borgen et al

925

& 2012 Macmillan Publishers Limited European Journal of Clinical Nutrition (2012) 920 – 925

Copyright of European Journal of Clinical Nutrition is the property of Nature Publishing Group and its content

may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express

written permission. However, users may print, download, or email articles for individual use.