Skip Navigation


American Journal of Epidemiology Advance Access originally published online on August 1, 2007
American Journal of Epidemiology 2007 166(8):912-923; doi:10.1093/aje/kwm161
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplemental Material
Right arrow All Versions of this Article:
166/8/912    most recent
kwm161v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (10)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Cust, A. E.
Right arrow Articles by Riboli, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cust, A. E.
Right arrow Articles by Riboli, E.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

American Journal of Epidemiology © The Author 2007. Published by the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org.

ORIGINAL CONTRIBUTIONS

Dietary Carbohydrates, Glycemic Index, Glycemic Load, and Endometrial Cancer Risk within the European Prospective Investigation into Cancer and Nutrition Cohort

Anne E. Cust1,2,3, Nadia Slimani1, Rudolf Kaaks1,4, Marit van Bakel1, Carine Biessy1, Pietro Ferrari1, Martine Laville3,5, Anne Tjønneland6, Anja Olsen6, Kim Overvad7, Martin Lajous8,9, Francoise Clavel-Chapelon8, Marie-Christine Boutron-Ruault8, Jakob Linseisen4, Sabine Rohrmann4, Ute Nöthlings10, Heiner Boeing10, Domenico Palli11, Sabina Sieri12, Salvatore Panico13, Rosario Tumino14, Carlotta Sacerdote15, Guri Skeie16, Dagrun Engeset16, Inger Torhild Gram16, J. Ramón Quirós17, Paula Jakszyn18, María José Sánchez19, Nerea Larrañaga20, Carmen Navarro21, Eva Ardanaz22, Elisabet Wirfält23, Göran Berglund24, Eva Lundin25, Göran Hallmans26, H. Bas Bueno-de-Mesquita27, Huaidong Du27, Petra H. M. Peeters28, Sheila Bingham29,30, Kay-Tee Khaw31, Naomi E. Allen32, Timothy J. Key32, Mazda Jenab1 and Elio Riboli33

1 Nutrition and Hormones Unit, International Agency for Research on Cancer, Lyon, France
2 School of Public Health, University of Sydney, Sydney, Australia
3 Center for Research in Human Nutrition Rhône-Alpes, Université Lyon 1, Lyon, France
4 Division of Cancer Epidemiology, German Cancer Research Center (DKFZ), Heidelberg, Germany
5 UMR Inserm U449/INRA 1235, Lyon, France
6 Institute of Cancer Epidemiology, Danish Cancer Society, Copenhagen, Denmark
7 Department of Clinical Epidemiology, Aarhus University Hospital, Aalborg, Denmark
8 Inserm ERI 20, Paris-Sud University EA4045, and Institut Gustave Roussy, Villejuif, France
9 National Institute of Public Health, Center for Investigation on Population Health, Cuernavaca, Mexico
10 German Institute of Human Nutrition Potsdam-Rehbruecke, Nuthetal, Germany
11 Molecular and Nutritional Epidemiology Unit, CSPO-Scientific Institute of Tuscany, Florence, Italy
12 Nutritional Epidemiology Unit, National Cancer Institute, Milan, Italy
13 Department of Clinical and Experimental Medicine, Federico II University, Naples, Italy
14 Cancer Registry, Azienda Ospedaliera "Civile M. P. Arezzo," Ragusa, Italy
15 CPO-Piemonte, Torino, Italy
16 Faculty of Medicine, Institute of Community Medicine, University of Tromsø, Tromsø, Norway
17 Public Health and Health Planning Directorate, Asturias, Spain
18 Department of Epidemiology and Cancer Registry, Catalan Institute of Oncology, Barcelona, Spain
19 Andalusian School of Public Health, Granada, Spain
20 Public Health Department of Gipuzkoa, Basque Government, San Sebastian, Spain
21 Epidemiology Department, Murcia Health Council, Murcia, Spain
22 Public Health Institute of Navarra, Pamplona, Spain
23 Nutrition Epidemiology, Department of Clinical Sciences in Malmö, Lund University, Malmö, Sweden
24 Department of Medicine, Lund University, Malmö University Hospital, Malmö, Sweden
25 Department of Medical Biosciences, Umeå University, Umeå, Sweden
26 Department of Public Health and Clinical Medicine, Umeå University, Umeå, Sweden
27 Centre for Nutrition and Health, National Institute of Public Health and the Environment, Bilthoven, The Netherlands
28 Julius Center for Health Sciences and Primary Care, University Medical Center, Utrecht, The Netherlands
29 MRC Dunn Human Nutrition Unit, Cambridge, United Kingdom
30 MRC Centre for Nutritional Epidemiology in Cancer Prevention and Survival, Department of Public Health and Primary Care, University of Cambridge, Cambridge, United Kingdom
31 Clinical Gerontology, Department of Public Health and Primary Care, University of Cambridge, Cambridge, United Kingdom
32 Cancer Research UK Epidemiology Unit, University of Oxford, Oxford, United Kingdom
33 Department of Epidemiology and Public Health, Imperial College London, London, United Kingdom

Correspondence to Anne Cust, Level 2, K25 - Medical Foundation Building, School of Public Health, The University of Sydney, Sydney, NSW 2006, Australia (e-mail: annec{at}health.usyd.edu.au).

Received for publication December 20, 2006. Accepted for publication April 27, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
The associations of dietary total carbohydrates, overall glycemic index, total dietary glycemic load, total sugars, total starch, and total fiber with endometrial cancer risk were analyzed among 288,428 women in the European Prospective Investigation into Cancer and Nutrition cohort (1992–2004), including 710 incident cases diagnosed during a mean 6.4 years of follow-up. Cox proportional hazards models were used to estimate relative risks and 95% confidence intervals. There were no statistically significant associations with endometrial cancer risk for increasing quartile intakes of any of the exposure variables. However, in continuous models calibrated by using 24-hour recall values, the multivariable relative risks were 1.61 (95% confidence interval: 1.06, 2.45) per 100 g/day of total carbohydrates, 1.40 (95% confidence interval: 0.99, 1.99) per 50 units/day of total dietary glycemic load, and 1.36 (95% confidence interval: 1.05, 1.76) per 50 g/day of total sugars. These associations were stronger among women who had never used postmenopausal hormone therapy compared with ever users (total carbohydrates pheterogeneity = 0.04). Data suggest no association of overall glycemic index, total starch, and total fiber with risk, and a possible modest positive association of total carbohydrates, total dietary glycemic load, and total sugars with risk, particularly among never users of hormone replacement therapy.

cohort studies; diet; dietary carbohydrates; dietary fiber; endometrial neoplasms; glycemic index; insulin; nutrition assessment

Abbreviations: EPIC, European Prospective Investigation into Cancer and Nutrition; HRT, hormone replacement therapy


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
There is increasing evidence that insulin resistance, chronic hyperinsulinemia, and diabetes are implicated in the etiology of endometrial cancer (1–4). Obesity is a major determinant of hyperinsulinemia (5) and strongly increases endometrial cancer risk by aromatization of androgens to estrogens in adipose tissue (1). However, insulin could also influence risk through direct actions on endometrial tissue as a mitogenic and antiapoptotic growth factor or indirectly by increasing bioavailable estrogen and insulin-like growth factor 1 levels (1–3). Postprandial and average insulin concentrations are directly influenced by the type, amount, and rate of digestion of dietary carbohydrates (6–9). Thus, it has been speculated that the quantity and quality (i.e., type, source, component) of dietary carbohydrates could have a role in endometrial carcinogenesis (10).

Previous studies of the association between dietary carbohydrate intake and endometrial cancer risk (11–23), few of which used a cohort design (11–14), have observed inconsistent but generally non-statistically significant results. A common limitation is the large random error associated with using dietary questionnaires to estimate food and nutrient intakes that tends to attenuate relative risk estimates and make it difficult to detect true associations (24, 25).

Glycemic index and glycemic load reflect the metabolic effects of dietary carbohydrates (6, 26). The glycemic index ranks carbohydrate foods based on their postprandial blood glucose response and hence their effect on blood insulin levels (6, 8, 27). The glycemic load combines the glycemic index value and the quantity of carbohydrate to quantify the overall estimated glycemic effect of a portion of food (26, 27). Four previous studies (10, 11, 13, 28) examined the association between glycemic index, glycemic load, and endometrial cancer risk and observed generally null associations or modest increased risks. There were suggestions that the associations may differ according to menopausal status, obesity, physical activity, use of hormone replacement therapy (HRT), and diabetes status (10, 11, 13, 28).

The European Prospective Investigation into Cancer and Nutrition (EPIC) cohort is well suited to study the influence of diet on cancer risk because of the large variation in dietary patterns across the 10 collaborating western European countries (29, 30). Additionally, 24-hour recall values collected from a sample of EPIC participants are available that enable partial correction for dietary measurement errors. To our knowledge, this is the largest prospective analysis to examine the association of endometrial cancer risk with dietary total carbohydrates, glycemic index, and glycemic load.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Study cohort
The EPIC study design, study population, and baseline data collection methods have been described in detail previously (31, 32), including baseline assessment of habitual diet (31, 33), lifestyle factors (31), physical activity (34), anthropometric measures (35), and menopausal status (34). Briefly, diet and lifestyle data were collected from approximately 370,000 women and 150,000 men aged 20–85 years enrolled between 1992 and 2000 in 23 centers throughout 10 western European countries (Denmark, France, Germany, Greece, Italy, Norway, Spain, Sweden, the Netherlands, and United Kingdom) (31). Participants were mostly recruited from the general population residing in defined geographic areas (31). Approval for this study was obtained from the ethical review boards of the International Agency for Research on Cancer (Lyon, France) and from all local recruiting institutions. All participants provided written informed consent.

A total of 288,428 women were included in the present analysis after a priori exclusion of 19,246 with prevalent cancer (other than nonmelanoma skin cancer), 34,972 who reported a hysterectomy at enrollment, 4,158 for whom follow-up data were incomplete, 33 with in situ or nonepithelial incident endometrial cancers, and 6,104 who were in the top or bottom 1 percent of the distribution of the ratio of reported total energy intake to estimated energy requirement (36). All women from Greece (n = 13,748) were also excluded because of lack of access to data on the carbohydrate content of questionnaire food items, which prevented estimation of glycemic index and glycemic load.

Dietary measurement
Usual diet during the previous 12 months was assessed with country-specific, validated (33) dietary assessment instruments (31). Food frequency questionnaires were used in all centers, and some centers combined questionnaires with food records (United Kingdom) or a 7-day menu book (Malmö, Sweden) (31). Intakes (in grams per day) of total carbohydrate, sugars, starch, and fiber (including resistant starch) were estimated from the dietary instruments by using country-specific food composition tables, which were standardized to a certain extent across countries to allow calibration at the nutrient level. To improve comparability of dietary data across centers and to partially correct for dietary measurement error arising from center-specific bias and random and systematic within-person errors (24, 37), a second dietary measurement was taken from an 8 percent stratified random sample (a total of 36,900 participants) of the cohort by using a standardized, computer-assisted, 24-hour dietary recall method (38). A single 24-hour recall among a sufficiently large sample provides a good estimate of mean intakes of foods and nutrients at a population level (39).

A glycemic index database was compiled whereby published glycemic index values (26, 40, 41) were assigned to carbohydrate-providing food items to reflect their blood glucose response (detailed in the supplementary material posted on the Journal's website (http://aje.oupjournals.org/)). Total dietary glycemic load was calculated by multiplying the digestible carbohydrate content of a given food item (g/100 g) by the quantity of that food item consumed per day and its glycemic index value and then summing the values for all food items reported (26). The overall glycemic index was calculated by dividing the total dietary glycemic load by the daily total dietary carbohydrate intake. The overall glycemic index reflects the average quality of carbohydrates consumed, whereas the total dietary glycemic load reflects both the average quantity and the quality of carbohydrates (26).

Follow-up for cancer incidence and mortality
Methods for follow-up for cancer incidence and vital status in each EPIC country have been described in detail elsewhere (31). Vital status was known for 98.4 percent of EPIC participants at the end of April 2004. Endometrial cancers were classified by using the International Classification of Diseases for Oncology, second edition (code C54). There were 710 eligible incident endometrial cancer cases identified by the end of censoring periods ending between December 1999 and March 2004 in the EPIC centers. Cancer diagnosis was microscopically verified for 89.3 percent of cases and by clinical examination for 8.5 percent; the remaining 2.2 percent were verified by self-report, tomography scan, surgery, autopsy, or death certificate. Morphology was specified for 239 (34 percent) cases, of which 229 cases (96 percent) were classified as type I and 10 cases (4 percent) as type II (42).

Statistical methods
Age- and center-adjusted Pearson's partial correlation coefficients were estimated to assess the correlations between nutrient intakes. Dietary exposure intakes were analyzed as both continuous and categorical variables. We used Cox proportional hazards models to calculate hazard ratios as estimates of relative risks and 95 percent confidence intervals. Age was used as the underlying time variable, with entry and exit time defined as the subject's age at recruitment and age at endometrial cancer diagnosis or censoring date, respectively. Quartile cutpoints were based on study-wide energy-adjusted nutrient intake distributions, computed as the residuals from a linear regression of nutrient intake on total energy intake (24, 43) with additional adjustment for country. Trend tests were estimated on integer scores (1–4) applied to the quartiles and were entered as a continuous term in the regression models.

All Cox models were stratified by study center to control for differences in questionnaire design and follow-up procedures, and by age at recruitment in 1-year categories. All models were also adjusted for total energy intake by using the residual method (24, 43) to control partly for the error in nutrient intake (44–46) and because we were primarily interested in dietary composition rather than absolute intake (43, 44). The multivariable models were additionally adjusted for body mass index (kg/m2, continuous), height (cm, continuous; representing lean body mass (43)), and total physical activity level (inactive, moderately inactive, moderately active, active, unknown), to control for other determinants of energy balance (43, 45, 47), and for cigarette smoking status (never, former, current, unknown) because it was associated with nutrient intake and its inclusion in the models influenced the risk estimates. Using body mass index squared, log(body mass index), or sqrt(body mass index) as potentially nonlinear confounders did not further influence the associations. Other potential confounders examined, but not included in the final models because their inclusion had little influence on the risk estimates, were age at menarche, menopausal status, age at menopause, number of full-term pregnancies, age at birth of last child, ever use of HRT, ever use of oral contraceptives, self-reported presence of hypertension or diabetes, and education.

Nutrient intakes including total energy intake were calibrated by using a multivariable fixed-effects linear model in which 24-hour recall values were regressed on the main dietary questionnaire values for the calibration study participants (refer to the supplementary online material for further details). The calibration model was used to compute individual predicted values for each of the dietary exposures of interest. Cox regression models were run by using the predicted (calibrated) values on a continuous scale, for all main models. The standard error of the deattenuated coefficient was estimated with bootstrap sampling to take into account the uncertainty related to measurement error correction (48).

We used multivariate nutrient density models (24, 49) to evaluate the possible confounding effects of other macronutrients. We estimated the effect of a 5 percent increase in the percentage of energy from carbohydrates relative to an identical decrease in the percentage of energy from another macronutrient, enabling us to compare isocaloric diets. Additionally, we examined macronutrient composition by using the energy decomposition (partition) method (24) (refer to the supplementary online material).

On the basis of previous reports of possible effect modification, we examined heterogeneity of risk estimates according to body mass index (<25, 25–<30, ≥30 kg/m2), waist-hip ratio (<0.78, ≥0.78 cm), menopausal status (premenopausal, perimenopausal, postmenopausal), total physical activity level (inactive + moderately inactive, moderately active + active), use of HRT by postmenopausal women (never/ever used), use of oral contraceptives (never/ever used), self-reported diabetes (yes/no), and country. Chi-square tests were used to calculate the deviations of beta coefficients obtained from continuous uncalibrated models in each of the subgroups relative to the overall beta coefficients. All analyses were performed by using SAS software (version 9.1; SAS Institute, Inc., Cary, North Carolina), and statistical significance was inferred at two-sided p <0.05.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
During a mean 6.4 years and 1,842,995 person-years of follow-up, 710 incident endometrial cancer cases were diagnosed among 288,428 women (table 1). Women who developed endometrial cancer during follow-up were, at baseline, more likely to be older, be obese, be never smokers, have diabetes, and have different patterns of exogenous hormone use compared with women who did not develop cancer, but they had similar dietary intake and physical activity levels (table 2). When the dietary questionnaire variables were categorized into quartiles, the mean values in the fourth quartiles were two- to threefold higher than those in the first quartiles, except for glycemic index, where the difference in means was only 10 glycemic index units (data not shown).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Distribution of cases and intake of dietary carbohydrates, glycemic index, and glycemic load, by subcohort, European Prospective Investigation into Cancer and Nutrition, 1992–2004

 

View this table:
[in this window]
[in a new window]

  		TABLE 2. Baseline demographic and lifestyle characteristics of 288,428 women* according to endometrial cancer status at the end of the follow-up period, European Prospective Investigation into Cancer and Nutrition, 1992–2004

 
Total energy intake was strongly correlated with carbohydrate intake (r = 0.86) and total dietary glycemic load (r = 0.84) but weakly correlated with overall glycemic index (r = 0.11) (data not shown). After adjustment for total energy intake in addition to age and center, the following correlations were observed: total carbohydrate intake with total dietary glycemic load (r = 0.93), overall glycemic index (r = 0.17), and total fats (r = –0.71); and overall glycemic index with total dietary glycemic load (r = 0.51), total starch (r = 0.66), total sugars (r = –0.44), and total fiber (r = 0.00).

We found no statistically significant associations or dose-response trends with endometrial cancer risk for increasing quartile intakes of any of the nutrient exposure variables (table 3). However, when total carbohydrates, total dietary glycemic load, total sugars, and total fiber were analyzed with continuous scales, the associations for these dietary factors were marginally statistically significant, and the point estimates more than doubled when calibrated data were used to partially control for measurement error (table 3). Additional adjustment for total dietary fiber did not appreciably alter the relative risk estimates, but the fiber-risk association was partly attenuated after adjustment for total carbohydrates (data not shown), suggesting no independent association of total fiber with risk. There was no association between overall glycemic index or total starch intake and risk, and neither adjustment for potential confounders nor calibration appreciably affected these relative risk estimates (table 3).


View this table:
[in this window]
[in a new window]

  		TABLE 3. Relative risk estimates and 95% confidence intervals for endometrial cancer, by quartiles* of energy-adjusted total carbohydrates, glycemic index, glycemic load, and other carbohydrate components, European Prospective Investigation into Cancer and Nutrition, 1992–2004

 
In isocaloric multivariable nutrient density models (data not shown in tables), increasing the energy intake from total carbohydrates by 5 percent was associated with a marginally statistically significant increased risk when offset by a 5 percent decrease in the percentage of energy from total fats (relative risk = 1.07, 95 percent confidence interval: 1.00, 1.15), polyunsaturated fats (relative risk = 1.18, 95 percent confidence interval: 0.97, 1.44), and, to a lesser extent, monounsaturated fats (relative risk = 1.08, 95 percent confidence interval: 0.90, 1.29). However, there was no association when increased energy from carbohydrates was offset by an identical decrease in energy intake from saturated fats, protein, or alcohol.

In subgroup analyses (table 4), the associations between total carbohydrates, total sugars, and risk were stronger among postmenopausal women who had never used HRT compared with ever users (pheterogeneity = 0.04 for total carbohydrates, pheterogeneity = 0.14 for total sugars). Among women who had never used HRT, there were statistically significant dose-response trends with increasing quartiles and use of continuous measures. The risk estimates appeared slightly stronger among postmenopausal women than among premenopausal women, but the tests for heterogeneity were not statistically significant. When stratified by body mass index subgroups, the calibrated continuous models suggested a possibly stronger association among normal-weight women, but this finding was not reflected in the quartile risk estimates. Similarly, risk estimates for total carbohydrates and total sugars appeared higher for women with a waist-hip ratio below the median versus above the median, but there was no significant heterogeneity (pheterogeneity = 0.10 and pheterogeneity = 0.31, respectively; data not shown in tables). There was no evidence that use of oral contraceptives, physical activity, or diabetes modified the associations, nor was there any evidence of heterogeneity of risk estimates across countries. The subgroup results for total dietary glycemic load were very similar to those for total carbohydrates. The association between overall glycemic index and risk did not differ by any of the subgroup factors examined (pheterogeneity > 0.10 for all comparisons). The relative risk estimates did not meaningfully change in sensitivity analyses that excluded women with less than 1 year of follow-up, women with self-reported diabetes, or in analyses restricted to cases with known type I tumors.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Multivariable* relative risk estimates and 95% confidence intervals for endometrial cancer, by quartiles{dagger} of energy-adjusted total carbohydrates and total sugars, according to body mass index, menopausal status, and use of exogenous hormones, European Prospective Investigation into Cancer and Nutrition, 1992–2004

 
We also examined the associations between main carbohydrate-providing food groups (29) and endometrial cancer risk (supplementary online table 1). None of the associations was statistically significant, although there was a marginally significant positive trend for potatoes (ptrend = 0.06).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
Overall, no significant association with endometrial cancer risk was observed for quartile intakes of any of the exposure variables. However, when continuous estimates were used, there was a suggestion of a modest increased risk associated with dietary carbohydrates, particularly total sugars, which was further supported by the calibrated estimates. Our results also indicate that HRT may modify the carbohydrate-risk association.

Most previous studies (11, 13, 14, 17–19) observed a null association between dietary carbohydrate intake and endometrial cancer, whereas other studies have suggested positive (12, 15, 16) or inverse (20–23) associations. We found that total sugars, but not total starch, might increase risk. Studies in animals and humans have found that diets high in sugars, particularly fructose and sucrose, increase insulin concentrations to a larger extent than starch-rich diets and are associated with insulin resistance (50, 51). Other epidemiologic studies have found no convincing association of total sugars (13, 52) or total starch (22) with endometrial cancer. Studies that examined dietary fiber intake have generally reported possible inverse (16, 18, 21, 22) or null (12, 19, 53) associations. It is possible that the association may differ according to different fiber fractions such as soluble and insoluble fiber; however, these data were not available in our study.

Although glycemic load, by definition, is a product of carbohydrate quantity and quality, it was correlated much more strongly with total carbohydrates than with overall glycemic index in this study. As such, the risk estimates for total dietary glycemic load were very similar to those for total carbohydrates. Similar correlations were reported in a recent study (54), in which the authors suggested that the narrow range of observed glycemic index values could partly explain the low correlation between glycemic load and glycemic index, and the lack of association between glycemic index and risk. Four previous studies (10, 11, 13, 28) examined the association between glycemic load, glycemic index, and endometrial cancer. Consistent with our results, these studies generally observed a weak positive association between total dietary glycemic load and risk. For glycemic index, two studies (11, 28) observed no association between overall glycemic index and risk, and two studies (10, 13) reported positive associations.

The associations between total carbohydrates, total sugars, total dietary glycemic load, and endometrial cancer risk were modified by HRT use, such that risk was increased for never users but not ever users. The underlying mechanisms are unclear, although it is possible that exogenous estrogens in HRT preparations modify circulating hormone levels to such an extent that dietary factors have little additional effect. Two previous studies (13, 28) did not observe statistically significant heterogeneity between glycemic load and risk according to HRT use. Kasum et al. (55) observed an inverse association between intake of whole grains and risk among never users but not ever users.

Contrary to our results, Silvera et al. (13) reported a possibly stronger association between glycemic load, glycemic index, and risk for women who were premenopausal at baseline; however, there was no significant heterogeneity, and many of those women would have experienced menopause during the mean 16-year study follow-up. Our study also measured only baseline menopausal status, but mean follow-up was shorter (6.4 years) and so would be less affected by changes in menopausal status. Although previous studies have reported possibly stronger associations between overall glycemic index or total dietary glycemic load and endometrial cancer risk for women who are overweight or obese (10, 11, 13), inactive (11, 13), or nondiabetic (28), we did not find clear evidence for effect modification by these factors.

Under isocaloric conditions, increasing the proportion of carbohydrates in the diet increased risk only when carbohydrates replaced energy from total fats, monounsaturated fats, and polyunsaturated fats, but not energy from saturated fats, protein, or alcohol. There is evidence from other studies (7, 8, 40, 56, 57) that the amount and type of dietary fat can modify the glycemic and insulinemic response to a carbohydrate food, and that insulin sensitivity may be improved by monounsaturated fats but worsened by saturated fats (58, 59). Future studies should consider exploring further the associations of dietary patterns with risk.

The main mechanisms by which carbohydrate-rich foods and their glycemic index could influence endometrial cancer risk relate to the development or exacerbation of insulin resistance, chronic hyperinsulinemia, hyperglycemia, obesity, or diabetes (1, 6, 8, 27, 60). The type and amount of carbohydrates directly influence the glycemic and insulinemic response (6–8, 50), and some evidence suggests that slowly absorbed, low glycemic index foods improve insulin sensitivity through their maintenance of relatively low plasma fatty acid levels (6, 8, 27). Furthermore, dietary interventions that modify carbohydrate and fat intake and overall glycemic index have been shown to significantly alter peptide and sex steroid hormone levels (61, 62).

A major strength of this study is its large, prospective cohort design; the wide variation in consumption of carbohydrate-rich foods between EPIC countries (29); and information on many potential confounders. Our study also has several limitations common to observational dietary studies. Estimating food and nutrient intakes by questionnaires is associated with large random error that tends to attenuate relative risk estimates (24, 25). It is also probable that differential systematic error was present because of underreporting of intake within specific population subgroups of women, such as obese women (36). Because total energy and many nutrients have correlated errors, adjustment for total energy is thought to partly remove some errors in estimated nutrient intakes (44, 45).

The calibration method used to correct for dietary measurement errors may not completely account for measurement error because of likely correlated errors between the 24-hour recalls and the dietary questionnaires (25, 46). Thus, the relative risk estimates observed in this study may be underestimates of the true association. On the other hand, some caution is needed when interpreting these calibrated estimates, particularly because of several statistically marginal associations and some associations that were seen with only the calibrated estimates. Measurement error correction will generally always strengthen associations away from the null because of imperfect associations between the dietary questionnaires and the 24-hour recall data. Calibrated estimates are thought to give a more precise point estimate but not a more precise estimate of statistical significance.

Diet and other covariates were measured at baseline only; thus, we were unable to adjust for possible changes in exposures, including diet and exogenous hormones, which may have occurred during follow-up. Some limitations also affect the estimation and interpretation of glycemic index and glycemic load. Reference glycemic index values have been determined primarily by using US and Australian foods; however, botanical variation, processing, and cooking methods, which have a significant impact on glycemic index (9, 26, 40), may differ in European countries. Glycemic index values have been determined for only a limited number of foods, and the methods used to assign glycemic index values to food items can differ between studies. It has also been suggested that glycemic index and glycemic load may not always adequately reflect the glycemic or insulinemic response to food when used in the context of a usual diet (56, 57, 63).

In conclusion, our data suggest no association of overall glycemic index, total starch, and total fiber with risk, and a possible modest positive association of total carbohydrates, total dietary glycemic load, and total sugars with risk, particularly among never users of HRT.


   ACKNOWLEDGMENTS
 
Anne Cust was supported by a PhD scholarship from the University of Sydney and by a Research Scholar Award from the Cancer Institute NSW, Australia. The EPIC study has received financial support from the "Europe Against Cancer Program" of the European Commission (SANCO); Danish Cancer Society; German Cancer Aid; Ligue Nationale contre le Cancer, 3M Company, INSERM; German Cancer Research Center; German Federal Ministry of Education and Research; Dutch Ministry of Public Health, Welfare and Sports; National Cancer Registry and the Regional Cancer Registries Amsterdam, East and Maastricht of the Netherlands; Norwegian Cancer Society; Norwegian Research Council; Health Research Fund (FIS) of the Spanish Ministry of Health; the ISCIII Network RCESP (C03/09) and RETICC C03/10, Spanish Regional Governments of Andalusia, Asturias, Basque Country, Murcia, and Navarra, and the Catalan Institute of Oncology; Greek Ministry of Health; Greek Ministry of Education; Italian Association for Research on Cancer; Swedish Cancer Society; Swedish Scientific Council; Regional Government of Skane, Sweden; Cancer Research UK; Medical Research Council, UK; Stroke Association, UK; British Heart Foundation; Department of Health, UK; Food Standards Agency, UK; Wellcome Trust, UK.

Conflict of interest: none declared.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Kaaks R, Lukanova A, Kurzer MS. Obesity, endogenous hormones, and endometrial cancer risk: a synthetic review. Cancer Epidemiol Biomarkers Prev (2002) 11:1531–43.[Abstract/Free Full Text]
  2. Lukanova A, Zeleniuch-Jacquotte A, Lundin E, et al. Prediagnostic levels of C-peptide, IGF-I, IGFBP -1, -2 and -3 and risk of endometrial cancer. Int J Cancer (2004) 108:262–8.[CrossRef][Web of Science][Medline]
  3. Cust AE, Allen NE, Rinaldi S, et al. Serum levels of C-peptide, IGFBP-1 and IGFBP-2 and endometrial cancer risk; results from the European Prospective Investigation into Cancer and Nutrition. Int J Cancer (2007) 120:2656–64.[CrossRef][Web of Science][Medline]
  4. Cust AE, Kaaks R, Friedenreich C, et al. Plasma adiponectin levels and endometrial cancer risk in pre- and postmenopausal women. J Clin Endocrinol Metab (2007) 92:255–63.[Abstract/Free Full Text]
  5. Bergman RN, Mittelman SD. Central role of the adipocyte in insulin resistance. J Basic Clin Physiol Pharmacol (1998) 9:205–21.[Medline]
  6. Ludwig DS. The glycemic index: physiological mechanisms relating to obesity, diabetes, and cardiovascular disease. JAMA (2002) 287:2414–23.[Abstract/Free Full Text]
  7. Wolever TM. Dietary carbohydrates and insulin action in humans. Br J Nutr (2000) 83(suppl 1):S97–102.[Web of Science][Medline]
  8. Augustin LS, Franceschi S, Jenkins DJ, et al. Glycemic index in chronic disease: a review. Eur J Clin Nutr (2002) 56:1049–71.[CrossRef][Web of Science][Medline]
  9. Bjorck I, Granfeldt Y, Liljeberg H, et al. Food properties affecting the digestion and absorption of carbohydrates. Am J Clin Nutr (1994) 59(suppl):699S–705S.[Medline]
  10. Augustin LS, Gallus S, Bosetti C, et al. Glycemic index and glycemic load in endometrial cancer. Int J Cancer (2003) 105:404–7.[CrossRef][Web of Science][Medline]
  11. Larsson SC, Friberg E, Wolk A. Carbohydrate intake, glycemic index and glycemic load in relation to risk of endometrial cancer: a prospective study of Swedish women. Int J Cancer (2007) 120:1103–7.[CrossRef][Web of Science][Medline]
  12. Jain MG, Rohan TE, Howe GR, et al. A cohort study of nutritional factors and endometrial cancer. Eur J Epidemiol (2000) 16:899–905.[CrossRef][Web of Science][Medline]
  13. Silvera SA, Rohan TE, Jain M, et al. Glycaemic index, glycaemic load and risk of endometrial cancer: a prospective cohort study. Public Health Nutr (2005) 8:912–19.[CrossRef][Web of Science][Medline]
  14. Zheng W, Kushi LH, Potter JD, et al. Dietary intake of energy and animal foods and endometrial cancer incidence. The Iowa Women's Health Study. Am J Epidemiol (1995) 142:388–94.[Abstract/Free Full Text]
  15. Tzonou A, Lipworth L, Kalandidi A, et al. Dietary factors and the risk of endometrial cancer: a case-control study in Greece. Br J Cancer (1996) 73:1284–90.[Web of Science][Medline]
  16. Barbone F, Austin H, Partridge EE. Diet and endometrial cancer: a case-control study. Am J Epidemiol (1993) 137:393–403.[Abstract/Free Full Text]
  17. Shu XO, Zheng W, Potischman N, et al. A population-based case-control study of dietary factors and endometrial cancer in Shanghai, People's Republic of China. Am J Epidemiol (1993) 137:155–65.[Abstract/Free Full Text]
  18. Jain MG, Howe GR, Rohan TE. Nutritional factors and endometrial cancer in Ontario, Canada. Cancer Control (2000) 7:288–96.[Medline]
  19. Salazar-Martinez E, Lazcano-Ponce E, Sanchez-Zamorano LM, et al. Dietary factors and endometrial cancer risk. Results of a case-control study in Mexico. Int J Gynecol Cancer (2005) 15:938–45.[CrossRef][Web of Science][Medline]
  20. Goodman MT, Hankin JH, Wilkens LR, et al. Diet, body size, physical activity, and the risk of endometrial cancer. Cancer Res (1997) 57:5077–85.[Abstract/Free Full Text]
  21. Potischman N, Swanson CA, Brinton LA, et al. Dietary associations in a case-control study of endometrial cancer. Cancer Causes Control (1993) 4:239–50.[Web of Science][Medline]
  22. Goodman MT, Wilkens LR, Hankin JH, et al. Association of soy and fiber consumption with the risk of endometrial cancer. Am J Epidemiol (1997) 146:294–306.[Abstract/Free Full Text]
  23. Littman AJ, Beresford SA, White E. The association of dietary fat and plant foods with endometrial cancer (United States). Cancer Causes Control (2001) 12:691–702.[CrossRef][Web of Science][Medline]
  24. Willett W. Nutritional epidemiology. (1998) 2nd ed. New York, NY: Oxford University Press. (Monographs in epidemiology and biostatistics, vol 30).
  25. Kipnis V, Midthune D, Freedman L, et al. Bias in dietary-report instruments and its implications for nutritional epidemiology. Public Health Nutr (2002) 5:915–23.[CrossRef][Web of Science][Medline]
  26. Foster-Powell K, Holt SH, Brand-Miller JC. International table of glycemic index and glycemic load values: 2002. Am J Clin Nutr (2002) 76:5–56.[Abstract/Free Full Text]
  27. Jenkins DJ, Kendall CW, Augustin LS, et al. Glycemic index: overview of implications in health and disease. Am J Clin Nutr (2002) 76:266S–73S.[Abstract/Free Full Text]
  28. Folsom AR, Demissie Z, Harnack L. Glycemic index, glycemic load, and incidence of endometrial cancer: the Iowa Women's Health Study. Nutr Cancer (2003) 46:119–24.[CrossRef][Web of Science][Medline]
  29. Wirfalt E, McTaggart A, Pala V, et al. Food sources of carbohydrates in a European cohort of adults. Public Health Nutr (2002) 5:1197–215.[CrossRef][Web of Science][Medline]
  30. Slimani N, Fahey M, Welch AA, et al. Diversity of dietary patterns observed in the European Prospective Investigation into Cancer and Nutrition (EPIC) project. Public Health Nutr (2002) 5:1311–28.[CrossRef][Web of Science][Medline]
  31. Riboli E, Hunt KJ, Slimani N, et al. European Prospective Investigation into Cancer and Nutrition (EPIC): study populations and data collection. Public Health Nutr (2002) 5:1113–24.[CrossRef][Web of Science][Medline]
  32. Riboli E, Kaaks R. The EPIC Project: rationale and study design. European Prospective Investigation into Cancer and Nutrition. Int J Epidemiol (1997) 26(suppl 1):S6–14.[Abstract/Free Full Text]
  33. Margetts BM, European Prospective Pietinen P. Investigation into Cancer and Nutrition: validity studies on dietary assessment methods. Int J Epidemiol (1997) 26(suppl 1):S1–5.[Free Full Text]
  34. Friedenreich CM, Cust A, Lahmann PH, et al. Physical activity and risk of endometrial cancer: the European prospective investigation into cancer and nutrition. Int J Cancer (2007) 121:347–55.[CrossRef][Medline]
  35. Haftenberger M, Lahmann PH, Panico S, et al. Overweight, obesity and fat distribution in 50- to 64-year-old participants in the European Prospective Investigation into Cancer and Nutrition (EPIC). Public Health Nutr (2002) 5:1147–62.[CrossRef][Web of Science][Medline]
  36. Ferrari P, Slimani N, Ciampi A, et al. Evaluation of under- and overreporting of energy intake in the 24-hour diet recalls in the European Prospective Investigation into Cancer and Nutrition (EPIC). Public Health Nutr (2002) 5:1329–45.[Web of Science][Medline]
  37. Ferrari P, Kaaks R, Fahey MT, et al. Within- and between-cohort variation in measured macronutrient intakes, taking account of measurement errors, in the European Prospective Investigation into Cancer and Nutrition study. Am J Epidemiol (2004) 160:814–22.[Abstract/Free Full Text]
  38. Slimani N, Kaaks R, Ferrari P, et al. European Prospective Investigation into Cancer and Nutrition (EPIC) calibration study: rationale, design and population characteristics. Public Health Nutr (2002) 5:1125–45.[CrossRef][Web of Science][Medline]
  39. Kaaks R, Riboli E. Validation and calibration of dietary intake measurements in the EPIC project: methodological considerations. European Prospective Investigation into Cancer and Nutrition. Int J Epidemiol (1997) 26(suppl 1):S15–25.[Abstract/Free Full Text]
  40. Henry CJ, Lightowler HJ, Strik CM, et al. Glycaemic index and glycaemic load values of commercially available products in the UK. Br J Nutr (2005) 94:922–30.[CrossRef][Web of Science][Medline]
  41. Human Nutrition Unit, University of Sydney. The official website of the glycemic index and GI database. Sydney, Australia, 2006. (www.glycemicindex.com).
  42. Tavassoli FA, Devilee D, eds. Pathology and genetics: tumours of the breast and female genital organs. In: World Health Organization classification of tumours (2003) Lyon, France: IARC Press.
  43. Willett W, Stampfer MJ. Total energy intake: implications for epidemiologic analyses. Am J Epidemiol (1986) 124:17–27.[Free Full Text]
  44. Willett W. Commentary: dietary diaries versus food frequency questionnaires—a case of undigestible data. Int J Epidemiol (2001) 30:317–19.[Free Full Text]
  45. Spiegelman D. Commentary: correlated errors and energy adjustment—where are the data? Int J Epidemiol (2004) 33:1387–8.[Free Full Text]
  46. Kipnis V, Subar AF, Midthune D, et al. Structure of dietary measurement error: results of the OPEN biomarker study. Am J Epidemiol (2003) 158:14–21. discussion 22–6.[Abstract/Free Full Text]
  47. Jakes RW, Day NE, Luben R, et al. Adjusting for energy intake—what measure to use in nutritional epidemiological studies? Int J Epidemiol (2004) 33:1382–6.[Abstract/Free Full Text]
  48. Rosner B, Gore R. Measurement error correction in nutritional epidemiology based on individual foods, with application to the relation of diet to breast cancer. Am J Epidemiol (2001) 154:827–35.[Abstract/Free Full Text]
  49. Smith-Warner SA, Spiegelman D, Adami HO, et al. Types of dietary fat and breast cancer: a pooled analysis of cohort studies. Int J Cancer (2001) 92:767–74.[CrossRef][Web of Science][Medline]
  50. Daly ME, Vale C, Walker M, et al. Dietary carbohydrates and insulin sensitivity: a review of the evidence and clinical implications. Am J Clin Nutr (1997) 66:1072–85.[Abstract/Free Full Text]
  51. Daly M. Sugars, insulin sensitivity, and the postprandial state. Am J Clin Nutr (2003) 78:865S–72S.[Abstract/Free Full Text]
  52. Petridou E, Kedikoglou S, Koukoulomatis P, et al. Diet in relation to endometrial cancer risk: a case-control study in Greece. Nutr Cancer (2002) 44:16–22.[CrossRef][Web of Science][Medline]
  53. Terry P, Vainio H, Wolk A, et al. Dietary factors in relation to endometrial cancer: a nationwide case-control study in Sweden. Nutr Cancer (2002) 42:25–32.[CrossRef][Web of Science][Medline]
  54. Flood A, Peters U, Jenkins DJ, et al. Carbohydrate, glycemic index, and glycemic load and colorectal adenomas in the Prostate, Lung, Colorectal, and Ovarian Screening Study. Am J Clin Nutr (2006) 84:1184–92.[Abstract/Free Full Text]
  55. Kasum CM, Nicodemus K, Harnack LJ, et al. Whole grain intake and incident endometrial cancer: the Iowa Women's Health Study. Nutr Cancer (2001) 39:180–6.[CrossRef][Web of Science][Medline]
  56. Flint A, Moller BK, Raben A, et al. The use of glycaemic index tables to predict glycaemic index of composite breakfast meals. Br J Nutr (2004) 91:979–89.[CrossRef][Web of Science][Medline]
  57. Holt SH, Miller JC, Petocz P. An insulin index of foods: the insulin demand generated by 1000-kJ portions of common foods. Am J Clin Nutr (1997) 66:1264–76.[Abstract/Free Full Text]
  58. Rivellese AA, De Natale C, Lilli S. Type of dietary fat and insulin resistance. Ann N Y Acad Sci (2002) 967:329–35.[Web of Science][Medline]
  59. Xiao C, Giacca A, Carpentier A, et al. Differential effects of monounsaturated, polyunsaturated and saturated fat ingestion on glucose-stimulated insulin secretion, sensitivity and clearance in overweight and obese, non-diabetic humans. Diabetologia (2006) 49:1371–9.[CrossRef][Web of Science][Medline]
  60. Furberg AS, Thune I. Metabolic abnormalities (hypertension, hyperglycemia and overweight), lifestyle (high energy intake and physical inactivity) and endometrial cancer risk in a Norwegian cohort. Int J Cancer (2003) 104:669–76.[CrossRef][Web of Science][Medline]
  61. Berrino F, Bellati C, Secreto G, et al. Reducing bioavailable sex hormones through a comprehensive change in diet: the diet and androgens (DIANA) randomized trial. Cancer Epidemiol Biomarkers Prev (2001) 10:25–33.[Abstract/Free Full Text]
  62. Kaaks R, Bellati C, Venturelli E, et al. Effects of dietary intervention on IGF-I and IGF-binding proteins, and related alterations in sex steroid metabolism: the Diet and Androgens (DIANA) Randomised Trial. Eur J Clin Nutr (2003) 57:1079–88.[CrossRef][Web of Science][Medline]
  63. Mayer-Davis EJ, Dhawan A, Liese AD, et al. Towards understanding of glycaemic index and glycaemic load in habitual diet: associations with measures of glycaemia in the Insulin Resistance Atherosclerosis Study. Br J Nutr (2006) 95:397–405.[CrossRef][Web of Science][Medline]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Am J EpidemiolHome page
S. M. George, S. T. Mayne, M. F. Leitzmann, Y. Park, A. Schatzkin, A. Flood, A. Hollenbeck, and A. F. Subar
Dietary Glycemic Index, Glycemic Load, and Risk of Cancer: A Prospective Cohort Study
Am. J. Epidemiol., February 15, 2009; 169(4): 462 - 472.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Clin. Nutr.Home page
W. Wen, X. O. Shu, H. Li, G. Yang, B.-T. Ji, H. Cai, Y.-T. Gao, and W. Zheng
Dietary carbohydrates, fiber, and breast cancer risk in Chinese women
Am. J. Clinical Nutrition, January 1, 2009; 89(1): 283 - 289.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Clin. Nutr.Home page
P. Gnagnarella, S. Gandini, C. La Vecchia, and P. Maisonneuve
Glycemic index, glycemic load, and cancer risk: a meta-analysis,
Am. J. Clinical Nutrition, June 1, 2008; 87(6): 1793 - 1801.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Clin. Nutr.Home page
H. Du, D. L van der A, M. M. van Bakel, C. J. van der Kallen, E. E Blaak, M. M. van Greevenbroek, E. H. Jansen, G. Nijpels, C. D. Stehouwer, J. M Dekker, et al.
Glycemic index and glycemic load in relation to food and nutrient intake and metabolic risk factors in a Dutch population
Am. J. Clinical Nutrition, March 1, 2008; 87(3): 655 - 661.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplemental Material
Right arrow All Versions of this Article:
166/8/912    most recent
kwm161v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (10)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Cust, A. E.
Right arrow Articles by Riboli, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cust, A. E.
Right arrow Articles by Riboli, E.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?