Association between pre-pregnancy consumption of meat, iron intake, and the risk of gestational diabetes: the SUN project

Abstract

Purpose

We assessed the association of total meat, processed, and unprocessed red meat and iron intake with the risk of developing gestational diabetes mellitus (GDM) in pregnant women.

Methods

We conducted a prospective study among 3298 disease-free Spanish women participants of the SUN cohort who reported at least one pregnancy between December 1999 and March 2012. Meat consumption and iron intake were assessed at baseline through a validated, 136-item semi-quantitative, food frequency questionnaire. We categorized total, red, and processed meat consumption and iron intake into quartiles. Logistic regression models were used to adjust for potential confounders.

Results

We identified 172 incident cases of GDM. In the fully adjusted analysis, total meat consumption was significantly associated with a higher risk of GDM [OR = 1.67 (95% CI 1.06–2.63, p-trend 0.010)] for the highest versus the lowest quartile of consumption. The observed associations were particularly strong for red meat consumption [OR = 2.37 (95% CI 1.49–3.78, p-trend < 0.001)] and processed meat consumption [OR = 2.01 (95% CI 1.26–3.21, p-trend 0.003)]. Heme iron intake was also directly associated with GDM [OR = 2.21 (95% CI 1.37–3.58, p-trend 0.003)], although the association was attenuated and lost its statistical significance when we adjusted for red meat consumption [OR = 1.57 (95% CI 0.91–2.70, p-trend 0.213)]. No association was observed for non-heme and total iron intake, including supplements.

Conclusions

Our overall findings suggest that higher pre-pregnancy consumption of total meat, especially red and processed meat, and heme iron intake, are significantly associated with an increased GDM risk in a Mediterranean cohort of university graduates.

Introduction

Gestational diabetes mellitus (GDM) is defined as carbohydrate intolerance resulting in hyperglycemia, with first onset during pregnancy [1]. GDM is known to be associated with maternal and perinatal morbidity, including increased risk of pre-eclampsia, caesarean section or premature delivery, macrosomia, or neonatal metabolic abnormalities. Besides, there are long-term complications for the mother, such as type 2 diabetes mellitus [2, 3], but also their offspring [2, 3] has a higher risk of insulin resistance, obesity [4], and type 2 diabetes [5]. Recent evidence suggests that the early detection and treatment of GDM may improve outcomes for both mother and child [6, 7].

The prevalence of GDM is increasing worldwide, due to the presence of overweight and obesity among women of reproductive age [8]. Every year, more than 200,000 cases of GDM are diagnosed in the United States, representing approximately 7% of all pregnancies. In the European Union, the number of cases diagnosed is very similar. The estimated prevalence is on average 2–6% of all pregnancies [9], which means that each year between 104,000 and 312,000 cases of GDM is identified. Hence, it is important to recognize the modifiable factors of GDM, such as, overweight, physical inactivity, or dietary factors, to curb this growing trend.

Pregravid dietary patterns, particularly a diet characterized by high consumption of red meat, processed meats, eggs, refined grain products, sweets, desserts, and pizza were associated with an increased risk of developing GDM [10, 11]. There are few epidemiological studies that have examined the specific role of meat consumption and subtypes on the incidence of GDM and most of them have been conducted in American populations [1012] and recently also in an Australian population [13]. A previous study from our cohort reported a positive association between fast food consumption and GDM [14]. However, the associations of total meat, red meat, or processed meat consumption with GDM in a Mediterranean population are unknown. The objective of this study was to assess the effect of total meat, red meat, processed meat consumption, and iron intake on the incidence of GDM in pregnant women of the SUN Project.

Design

Study population

The SUN project is a multipurpose, prospective, dynamic cohort study of Spanish university graduates. The recruitment started in 1999 and follow-up questionnaires are mailed to participants every 2 years. The objectives, design, and methods of the SUN study have been published in detail elsewhere [15, 16]. The Institutional Review Board of the University of Navarra approved the study protocol. We considered a response to the initial questionnaire to be the appropriate informed consent to participate in the study.

Up to December 2014, 22,175 participants had completed the baseline questionnaire. Participants were excluded if they were male (n = 8623) or women who did not respond the baseline questionnaire before 1st of March 2012 (n = 602) to include only those female participants who had sufficient time to answer at least the first follow-up questionnaire. Participants without any pregnancy during follow-up (n = 9556), 14 women with diagnoses of GDM at recruitment, women who reported caloric intake below the 1st percentile or above the 99th (n = 66), as well as, women with the previous diagnosis of diabetes (n = 16) were also excluded. Participants were considered to have diabetes at baseline if they reported a medical diagnosis of diabetes and/or they were being treated with insulin and/or oral anti-diabetics. In sum, 3298 pregnant women were included in our analyses (Fig. 1).

Fig. 1
figure1

Flowchart of the total sample of participants included in the analysis. *with one or more pregnancies

Dietary assessment

Dietary habits were assessed with a previously validated, semi-quantitative food frequency questionnaire (FFQ) with 136 items [17]. After the first validation of the questionnaire (with 118 items), it was slightly expanded to better adapt to the objectives of the SUN project. A second validation study for the expanded questionnaire was conducted in 2010 [18, 19]. In addition to the 136 foods, the questionnaire now includes a specific section on consumption patterns characteristic of the Mediterranean diet, (e.g., the frequency of consumption of fruit as the usual dessert, further specific details on the type of oil used for cooking), and also attitudes in relation to health practices, such as following a special diet and other questions about taking supplements of vitamins and minerals and other non-specific foods.

The questionnaire included a typical portion size for each food and had nine options for frequency of consumption ranging from “rarely or never” to “more than six times per day”.

Adherence to the Mediterranean Dietary Pattern (MDP) was evaluated combining eight items (fruits and nuts, vegetables, fish, legumes, cereals, dairy products, alcohol intake, and the ratio monounsaturated fatty acids/saturated fatty acids) according to the score proposed by Trichopoulou et al. [20] but excluding meat and meat products.

A group of “total meat consumption” was defined as the sum of all types of meat. Another group, “red meat” included the consumption of unprocessed beef, pork, lamb, liver, entrails, and “processed meat” included sausages, “sobrasada” (a raw cured sausage, typical from the Balearic Islands, Spain), bacon, pate, ham, chorizo, and salami. The food group “poultry” included chicken, turkey, and rabbit.

For the analyses of heme iron, non-heme iron, total iron intake, and iron supplements, the milligrams of iron contents were calculated in each food of the FFQ. “Total iron intake” was defined as the sum of heme iron and non-heme iron from dietary intake, while total iron with supplements was defined as the sum of heme and non-heme iron from dietary intake and the amount (mg) of iron consumed from supplements. Nutrient intakes were calculated by trained dietitians with a computer program based on Spanish food composition tables [21, 22].

Ascertainment of incidence of gestational diabetes mellitus

Participants reported the medical diagnosis of GDM at baseline and on each follow-up questionnaire. We considered a probable case of new-onset GDM if a pregnant women reported a medical diagnosis of GDM in the follow-up questionnaire and did not have a previous history of GDM at baseline. When we observed a probable diagnosis of new-onset GDM, we sent an additional specific questionnaire requesting further information about the confirmation of the diagnosis, the date of diagnosis, whether or not a previous diagnosis of diabetes was received, if diabetes had been diagnosed during the pregnancy period, fasting glucose levels and the values of the first glycated hemoglobin (Hb A1c) during pregnancy, blood glucose after an oral glucose tolerance test, and the use of insulin during pregnancy. We also asked participants to send us their medical records where the diagnosis and other clinical information were reported. An endocrinologist, blinded to dietary exposures, used the information provided by the specific questionnaires and the medical records to adjudicate each case of new-onset GDM. There is no universal diagnostic criterion for GDM [9]. In Spain, there are two level diagnosis: the formal National Diabetes Data Group criteria [23] or Carpenter and Coustan [24] cut-off points after a 100-g oral glucose tolerance test were used by most physicians for the diagnosis of GDM after a positive 50-g glucose challenge test conducted between 24 and 28 weeks in all Spanish pregnant women as the standard medical care.

Other covariates

The baseline questionnaire also collected information, including sociodemographic variables (age), anthropometric measurements [weight, height, and body mass index (BMI)], health-related habits (snacking, alcohol intake, smoking status, and passive exposure to smoke), and clinical variables (use of medication, self-reported pregnancy, cardiovascular disease/hypertension, and parity). Age was calculated from date of birth to the date questionnaire that was returned. The values on height and weight were self-reported. These values have previously been validated in a study of a subsample of the SUN cohort, in which the height and weight of individuals were measured objectively in a period not exceeding 3 months after the baseline questionnaire response [25]. We assessed physical activity at baseline using a previously validated questionnaire with a Spearman correlation coefficient of 0.51 (p < 0.001) between questionnaire information and objectively obtained measurements [26]. To express the degree of physical activity, metabolic equivalents (METs/week) were assigned to each activity in hours per week multiplied by its typical energy expenditure [27].

Statistical analysis

Descriptive statistics are presented as percentages (categorical variables) or means and standard deviations (continuous variables). Participants were classified into quartiles of consumption for meat or meat subtypes. Meat, meat subtypes, and iron intake were adjusted for total energy intake using the residuals method.

Non-conditional logistic regression models were fit to assess the association between meat consumption, meat subtypes, and iron with the risk of GDM. Odds ratios (OR) and their 95% confidence intervals (95% CI) were calculated using the lowest consumption category as the reference. The median value for each category was used as a continuous variable to test for linear trend. The first model was crude. The second model was adjusted for age. The third model additionally included the following covariates: BMI, family history of diabetes, parity, multiple pregnancies, baseline hypertension, smoking, alcohol intake, fiber intake, sugar-sweetened soft drinks, adherence to Mediterranean diet, total caloric energy intake, and physical activity. In the assessment of heme iron intake, the models were additionally adjusted for red meat consumption.

We also analyzed potential interactions between total meat consumption and sugar-sweetened soft drink consumption, physical activity, and family history of diabetes on the risk of GDM, through likelihood ratio tests for each of the product-terms introduced (each at one time) in the fully adjusted models.

Finally, we conducted further sensitivity analyses by additionally adjusting for polycystic ovary disease, by excluding women out of predefined caloric energy limits (<500 and >3500 kcal or outside percentiles 5th–95th), excluding multiparous women and those with prevalent cardiovascular disease.

All p values presented are two-tailed; p < 0.05 was considered statistically significant. Analyses were performed using the STATA program (version 12).

Results

Baseline characteristics of the pregnant women of the SUN Project, across quartiles of total meat consumption, are shown in Table 1. Participants who reported higher meat consumption were on average younger (mean age 28 years), more likely to be active smokers, multiparous, less physically active, and less adherent to the Mediterranean dietary pattern compared to those in the lowest quartile of consumption. In addition, these participants presented a higher percentage of calories from protein and total fat but lower fiber intake in comparison with the lowest quartile of total meat consumption.

Table 1 Final characteristics of 3298 pregnant women in the SUN cohort according to quartiles of total meat consumption

A total of 172 newly diagnosed cases of GDM (corresponding to 5.2% of pregnant participants) were identified among 3298 pregnant women of the SUN Project. Logistic regression models were used to evaluate the association of quartiles of total meat consumption (or meat subtypes) with the incidence of GDM (Table 2). In the crude and age-adjusted analyses total meat consumption, red meat and processed meat were directly associated with GDM. These associations were stronger after adjustment for other risk factors of GDM. Multivariable-adjusted ORs of GDM for comparisons between the highest and the lowest quartiles were 1.69 (95% CI 1.07–2.65) for overall meat consumption, 2.37 (95% CI 1.49–3.78) for red meat and 2.01 (95% CI 1.26–3.21) for processed meat (Table 2). When we additionally adjusted red meat for heme iron intake, the OR was slightly attenuated (adjusted OR = 2.12; 95% CI 1.27–3.54; p for trend = 0.002).

Table 2 OR and 95% confidence interval of incident gestational diabetes mellitus according to quartiles of consumption of meat, red meat, processed meat, poultry, and ham

There was no significant association between poultry consumption [OR = 0.91 (95% CI 0.58–1.41; p for trend 0.725)] or ham consumption [OR = 1.01 (95% CI 0.65–1.57; p for trend 0.990)] and incident GDM (Table 2).

When we analyzed heme iron intake, we observed a statistically significant increased risk of GDM for the highest versus the lowest quartile (adjusted OR = 2.21; 95% CI 1.37–3.58; p for trend = 0.003). When we additionally adjusted for red meat, the OR became non-significant (adjusted OR = 1.57; 95% CI 0.91–2.70; p for trend = 0.213). However, non-heme iron or total iron intake without and with supplements was not associated with GDM (Table 3).

Table 3 OR and 95% confidence interval of incident gestational diabetes mellitus according to quartiles of heme iron, non-heme iron, total iron, and total iron with supplements intake

When total and red meat consumption were analyzed as a continuous variable in multivariable models, the ORs for each increment of 1 serving/day were 1.26 (95% CI 1.07–1.48; p-trend < 0.001) for total meat (serving = 100 g), and 2.11 (95% CI 1.26–2.85; p-trend < 0.001) for red meat (serving = 125 g).

When we additionally adjusted for saturated fat intake the analysis of red meat, the association was slightly attenuated, but it remained statistically significant [adjusted OR of the highest versus the lowest quartile: 2.26 (95% CI 1.41–3.62); p for trend 0.007].

No significant interaction was observed between total meat consumption and sugar-sweetened soft drink consumption (p for interaction 0.29) or between total meat consumption and physical activity (p for interaction 0.80) or between total meat consumption and family history of diabetes (p for interaction 0.79).

Several sensitivity analyses were carried out to evaluate the robustness of our findings: (a) adjusting for the diagnosis of polycystic ovaries, (b) excluding subjects out of predefined caloric energy limits <500 and >3500 kcal and (c) or, alternatively, below percentile 5 and above percentile 95 of total energy intake (d) excluding multiparous women and (e) excluding women with cardiovascular disease. All the results were consistent with our main findings (Table 4).

Table 4 Sensitivity analyses: adjusted OR and 95% confidence interval of incident gestational diabetes mellitus according to quartiles of total, red, and processed meat consumption

Discussion

To our knowledge, this is the first prospective study performed to date in a Mediterranean population, with the objective of assessing the association between total meat consumption, its subtypes, and iron intake on the risk of GDM. We found an overall direct association between consumption of total meat and risk of GDM. Women in the highest quartile of total meat consumption had higher risk of developing GDM compared to women in the lowest quartile. Depending on the specific types of meat, the risk of developing GDM among participants in the highest quartile of processed and red meat consumption were twofold higher than in those in the lowest quartile after adjustment for potential confounders. Besides, a high intake of heme–iron was a risk factor for the development of GDM, although when we adjusted for red meat consumption, the magnitude of the association was attenuated and lost the statistical significance.

Several prospective studies and four meta-analyses have confirmed the association between higher total meat and increased risk of type 2 diabetes [28, 29]. However, there are fewer studies examining the effect of meat consumption on the risk of GDM and they have been mostly conducted in US populations. Recently, Schoenaker et al. [13] have conducted a study in an Australian population, but our study was the first conducted in a European population.

Our study confirms the findings of the previous studies from the Nurses’ Health Study II. In 2006, Zhang et al. [10] reported an increased risk of GDM with increased adherence to the Western pattern, as compared with a Prudent pattern. While the Western dietary pattern was characterized by an increase in red and processed meat, refined grain products, sweets, desserts, and pizza, the Prudent pattern was characterized for being rich in fruit, leafy green vegetables, poultry, and fish as animal protein source. They reported that consumption of red meat and processed meat was both significantly associated with a higher risk of GDM [RR = 1.74 (95% CI 1.35–2.26), p < 0.0001 and RR = 1.68 (95% CI 1.30–2.16), p < 0.00003, respectively]. For an increase in 1 serving/day, they observed an increased risk for GDM with RR = 1.61 (95% CI 1.25–2.07) for red meat and 1.64 (95% CI 1.13–2.38) for processed meat. They documented 758 incident cases of GDM and the risk was measured as RRs across increasing quintiles of consumption of red and processed meat. Meat intake was measured in servings per day across quintiles instead of g/day as in our study. Later, in 2013, this group of authors confirmed these results in the same cohort [12]. They reported a risk for GDM of 2.05 [(95% CI 1.55–2.73), p < 0.001] for total red meat, RR = 1.60 [(95% CI 1.21–2.12), p < 001] for non-processed meat, and RR = 1.36 [(95% CI 1.03–1.80), p = 0.062] for processed meat, comparing the highest versus the lowest quintiles. In this study, they documented 870 incident GDM pregnancies during 10 years of follow-up and they also measured meat intake and types in servings per day. The years of follow-up were similar in the three studies. In our study, no significant association was observed for poultry or ham. Moreover, this direct association of meat on GDM risk was also found in an Australian population [13]. They documented 292 new cases of GDM during 9 years of follow-up and the association was assessed across tertiles of meat intake.

Our data are consistent with those observed in the Nurses’ Health Study II for poultry consumption [12].

It is very interesting to notice that we found the same association in our cohort, since our study population was slightly younger (28.4 versus 31.5 years old) and, what is even more interesting, the average BMI of our participants was 21 kg/m2, while in the NHS cohort or the Australian Longitudinal Study on Women‘s Health, it was slightly higher (BMI = 23.45 kg/m2 in NHS, and BMI = 25.8 kg/m2 in women who developed GDM). Moreover, women of the SUN project with the highest consumption of meat reported lower fiber intake and were less adherent to the Mediterranean diet—a healthy dietary pattern—both fiber and the Mediterranean diet are known factors associated with lower risk of GDM [10, 30]. In fact, a Mediterranean-style pattern was recently associated with lower risk of developing GDM [13].

This study complements the results of a recently published article conducted in our cohort which suggested that higher consumption of fast foods (sausages, hamburgers, and pizza) was an independent risk factor for GDM [14].

Regarding iron intake, our data are consistent with previously reported findings, suggesting that heme–iron intake is a risk factor for the development of GDM. Bowers et al. [31] reported an increased risk of GDM of 58% [RR 1.58 (95% CI 1.21–2.08)] in participants with high heme–iron intake levels compared to those with lower intake. Moreover, when they additionally adjusted for red meat, the RR was attenuated and lost its significance as we observed in our results. Qiu et al. also observed this direct association [32]. Moreover, a high intake of heme iron from meat consumption was associated with an increased risk of developing type 2 diabetes in a cohort study, conducted within the framework of the the PREDIMED study in elderly Mediterranean subjects at high cardiovascular risk [33]. However, the role of non-heme iron is not so well defined. In our results, non-heme iron intake was not associated with increased risk for GDM, as reported in previously cited studies. In regard to iron supplementation, contradictory results have been described. On one hand, Chan et al. [34] did not detect an increased risk for GDM with iron supplementation, whilst Bo et al. [35] observed, with iron supplementation, an altered glucose metabolism, with increased basal glycaemia, and an increase in HOMA-IR was associated with a higher GDM prevalence in women receiving supplementation. As stated above, in our study, neither total iron intake with or without supplements presented a significant association with GDM.

There are different mechanisms that could help to explain a harmful effect of the intake of total meat, red meat, and processed meat on the incidence of GDM. The meat content of fat subtypes (including cholesterol and SFA) can increase the risk of GDM, due to known harmful effects on insulin sensitivity [36]. Other factors that may also contribute are nitrosamines, derived from nitrites present in meat, directly affecting beta-cell function [36]. Besides, advanced glycation end products, derived from heating and processing of meat, are thought to be involved in the development of GDM [36]. And heme–iron itself, especially present in red and processed meat, has been proposed as a cause for an increased risk of GDM, since it can increase insulin resistance when there is an abundance of tissue iron storage. Furthermore, it has been observed that in pregnant women with appropriate iron levels, increased iron deposit levels from supplement intakes are associated with a higher risk of GDM [37]. It seems that the pro-oxidant effect of iron and its ability to produce hydroxyl radicals leads to an increased oxidative stress and contributes to insulin resistance or even to pancreatic failure in the long term [38]. However, our results, as the results previously published by Bowers et al., suggest that heme–iron intake might be a marker of red meat consumption, and in addition to potential metabolic pathways related to heme–iron and GDM, the possible physiopathological explanation for the association between meat, especially red meat, and GDM, might be due to synergies between several components of meat, such as saturated fat, nitrites/nitrates, nitrosamines, advanced glycation end products, and the production of trimethylamine N-oxide [39].

Our study has some potential limitations. Measurement errors related to the process of evaluating nutritional exposures are usually the main problem in nutritional epidemiology. However, during the last two decades, interesting advances have taken place to develop and improve valid methods for measuring food intake, which are economically affordable for studies with a large sample size and can be precise enough to assess the hypotheses proposed [40].

The SUN cohort does not represent a random Spanish sample. However, the high educational level and health concern of participants confer to this cohort large advantages to obtain valid information of high quality. In fact, the restriction to university graduates that we applied in the design of our cohort may also have actually reinforced the internal validity of our findings. The reason is that the high educational level and the socioeconomic homogeneity of our cohort probably substantially reduced the potential for confounding derived from socioeconomic status and other social variables.

The self-declaration of events could be considered as another important disadvantage. However, the previous evidence suggests that the self-reported diagnoses of highly educated participants can be very informative [41].

It is important to emphasize that this project is not intended to estimate the incidence of diseases of interest in the Spanish general population, but rather to evaluate the impact of different lifestyles on the onset of GDM. For this purpose, a sufficiently large cohort is needed, with a relatively high retention rate, wide variability in exposure, and sufficient valid information, features that have been fulfilled in the SUN cohort across different studies. The assessment of associations in analytical epidemiology does not usually requires a “representative” sample in the sense of a probability sample from a large population [42]. According to methodological recommendations, sound selection of subjects for analytical epidemiological studies (as our cohort) should be guided by the need to make a valid comparison, and that criterion usually restricts the eligibility only to participants who meet some criteria to ensure a homogeneous study population. The final cohort has been selected, as in most classical cohort studies only within a narrow range of baseline characteristics. Generalizability (i.e., external validity) in epidemiology is usually based on knowledge of the pathophysiology of the disease and other sources of information rather than in “representativeness” (in the statistical sense of this term) of the cohorts.

This study has a number of strengths, including a prospective design that increases the value of the scientific evidence provided here, and avoids possible recall biases and reverse causality biases that often occur in cross-sectional or case–control studies. Importantly, this cohort has a large sample size with a high retention rate, a lengthy follow-up, and uses a well-validated FFQ. We were able to obtain a high degree of control for confounding, such as for demographic and lifestyle confounders. In addition, the study covers practically the entire Spanish territory, increasing geographic diversity. The high educational level of participants and the proportion of health professionals (with a higher quality of their self-reported information) favors or strengthens the internal validity of our study.

During pregnancy, a healthy diet is recommended including moderate consumption of lean meat [43] and (as for the general population) to limit the consumption of red and processed meat. This observational study supports to limit the consumption of total, and especially red and processed meat before and during pregnancy, to reduce the risk of developing GDM.

However, in the case of iron, according to the results of our study, we can advise pregnant women to consume iron-rich foods from vegetables, fruits or beans, and sometimes even using supplements as recommended by dietary guidelines [44]. However, we advise to reduce the consumption of heme iron from animal sources, including total meat and especially red meat.

In conclusion, our results suggest that total meat, processed meat, and red meat consumption were associated with GDM in a Mediterranean population. Besides, heme–iron intake was also associated with GDM, although when we adjusted for red meat, the association was attenuated and it lost its statistical significance.

References

  1. 1.

    Metzger BE, Coustan DR, the Organizing Committee (1998) Summary and recommendations of the Fourth International Workshop-Conference on gestational diabetes mellitus. Diabetes Care 21(Suppl. 2):161–167

    Google Scholar 

  2. 2.

    Kim C, Newton KM, Knopp RH (2002) Gestational diabetes and the incidence of type 2 diabetes: a systematic review. Diabetes Care 25:1862–1868

    Article  Google Scholar 

  3. 3.

    Bellamy L, Casas JP, Hingorani AD, Williams D (2009) Type 2 diabetes mellitus after gestational diabetes: a systematic review and meta-analysis. The Lancet 373:1773–1779

    CAS  Article  Google Scholar 

  4. 4.

    Boney CM, Verma A, Tucker R, Vohr BR (2005) Metabolic syndrome in childhood: association with birth weight, maternal obesity, and gestational diabetes mellitus. Pediatrics 115:e290–e296

    Article  Google Scholar 

  5. 5.

    Silverman BL, Metzger BE, Cho NH, Loeb CA (1995) Impaired glucose tolerance in adolescent offspring of diabetic mothers. Relationship to fetal hyperinsulinism. Diabetes Care 18:611–617

    CAS  Article  Google Scholar 

  6. 6.

    Landon MB, Spong CY, Thom E, Carpenter MW, Ramin SM, Casey B, Wapner RJ, Varner MW, Rouse DJ, Thorp JM et al (2009) A multicenter, randomized trial of treatment for mild gestational diabetes. N Engl J Med 361:1339–1348

    CAS  Article  Google Scholar 

  7. 7.

    Reece EA, Leguizamón G, Wiznitzer A (2009) Gestational diabetes: the need for a common ground. Lancet 373:1789–1797

    Article  Google Scholar 

  8. 8.

    American Diabetes Association (2009) Standards of medical care in diabetes. Diabetes Care 32(Suppl 1):S13–S61

    Article  Google Scholar 

  9. 9.

    Buckley BS, Harreiter J, Damm P, Corcoy R, Chico A, Simmons D, Vellinga A, Dunne F; DALI Core Investigator Group (2012) Gestational diabetes mellitus in Europe: prevalence, current screening practice and barriers to screening. A review. Diabet Med 29:844–854

    CAS  Article  Google Scholar 

  10. 10.

    Zhang C, Schulze MB, Solomon CG, Hu FB (2006) A prospective study of dietary patterns, meat intake and the risk of gestational diabetes mellitus. Diabetologia 49:2604–2613

    CAS  Article  Google Scholar 

  11. 11.

    Schoenaker DA, Mishra GD, Callway LK, Soedamah-Muthu SS (2016) The role of energy, nutrients, foods and dietary patterns in the development of gestational diabetes mellitus: a systematic review of observational studies. Diabetes Care 39:16–23

    Article  Google Scholar 

  12. 12.

    Bao W, Bowers K, Tobias DK, Hu FB, Zhang C (2013) Prepregnancy dietary protein intake, major dietary protein sources, and the risk of gestational diabetes mellitus: a prospective cohort study. Diabetes Care 36:2001–2008

    CAS  Article  Google Scholar 

  13. 13.

    Schoenaker DA, Soedamah-Muthu SS, Callway LK, Mishra GD (2015) Pre-pregnancy dietary patterns and risk of gestational diabetes mellitus: results from an Australian population-based prospective cohort study. Diabetologia 58:2726–2735

    CAS  Article  Google Scholar 

  14. 14.

    Dominguez LJ, Martínez-González MA, Basterra-Gortari FJ, Gea A, Barbagallo M, Bes-Rastrollo M (2014) Fast food consumption and gestational diabetes incidence in the SUN project. PLoS One 9:e106627

    Article  Google Scholar 

  15. 15.

    Martínez-González MA, Sánchez-Villegas A, De Irala J, Marti A, Martínez JA (2002) Mediterranean diet and stroke: objectives and design of the SUN project. Seguimiento Universidad de Navarra. Nutr Neurosci 5:65–73

    Article  Google Scholar 

  16. 16.

    Seguí-Gómez M, de la Fuente C, Vázquez Z, de Irala J, Martínez-González MA (2006) Cohort profile: the ‘Seguimiento Universidad de Navarra’ (SUN) study. Int J Epidemiol 35:1417–1422

    Article  Google Scholar 

  17. 17.

    Martin-Moreno JM, Boyle P, Gorgojo L, Maisonneuve P, Fernández-Rodríguez JC, Salvini S, Willett WC (1993) Development and validation of a food frequency questionnaire in Spain. Int J Epidemiol 22:512–519

    CAS  Article  Google Scholar 

  18. 18.

    De la Fuente-Arrillaga C, Ruiz ZV, Bes-Rastrollo M, Sampson L, Martinez-Gonzalez MA (2010) Reproducibility of an FFQ validated in Spain. Public Health Nutr 13:1364–1372

    Article  Google Scholar 

  19. 19.

    Fernandez-Ballart JD, Pinol JL, Zazpe I, Corella D, Carrasco P, Toledo E, Pérez-Bauer M, Martínez-González MA, Salas-Salvadó J, Martín-Moreno JM (2010) Relative validity of a semi-quantitative food-frequency questionnaire in an elderly mediterranean population of Spain. Br J Nutr 103:1808–1816

    CAS  Article  Google Scholar 

  20. 20.

    Trichopoulou A, Costacou T, Bamia C, Trichopoulos D (2003) Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med 348:2599–2608

    Article  Google Scholar 

  21. 21.

    Moreiras O (2009) Tablas de composición de alimentos (Food Composition Tables), 16th edn. Ediciones Pirámide, Madrid

  22. 22.

    Mataix J (2003) Tablas de composición de alimentos (Food Composition Tables), 5th edn. Granada

  23. 23.

    National Diabetes Data Group (1979) Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 28:1039–1057

    Article  Google Scholar 

  24. 24.

    American Diabetes Association (2010) Diagnosis and classification of diabetes mellitus. Diabetes Care 33(Suppl 1):S62–S69

    Article  Google Scholar 

  25. 25.

    Bes-Rastrollo M, Pérez-Valdivieso J, Sánchez-Villegas A, Alonso A, Martínez-González MA (2005) Validación del peso e índices de masa corporal auto-declarados de los participantes de una cohorte de graduados universitarios (Validation weight and self-reported body mass index of participants in a cohort of university graduates). Rev Esp Obes 3:183–189

    Google Scholar 

  26. 26.

    Martínez-González MA, López-Fontana C, Varo JJ, Sánchez-Villegas A, Martínez JA (2005) Validation of the Spanish version of the physical activity questionnaire used in the Nurses’ Health Study and the Health Professionals’ Follow-up Study. Public Health Nutr 8:920–927

    Article  Google Scholar 

  27. 27.

    Ainsworth BE, Haskell WL, Whitt MC, Irwin ML, Swartz AM, Strath SJ, O’Brien WL, Bassett DR Jr, Schmitz KH, Emplaincourt PO et al (2000) Compendium of physical activities: an update of activity codes and MET intensities. Med Sci Sports Exerc 32(Suppl 9):S498–S504

    CAS  Article  Google Scholar 

  28. 28.

    van Dam RM, Willet WC, Rimm EB, Stampfer MJ, Hu FB (2002) Dietary fat and meat intake in relation to risk of type 2 diabetes in men. Diabetes Care 25:417–424

    Article  Google Scholar 

  29. 29.

    Pan A, Sun Q, Bernstein AM, Schulze MB, Manson JE, Willett WC, Hu FB (2011) Red meat consumption and risk of type 2 diabetes: 3 cohort of US adults and an updated metaanalysis. Am J Clin Nutr 94:1088–1096

    CAS  Article  Google Scholar 

  30. 30.

    Tobias DK, Zhang C, Chavarro J, Bowers K, Rich-Edwards J, Rosner B, Mozaffarian D, Hu FB (2002) Prepregnancy adherence to dietary patterns and lowers risk of gestational diabetes mellitus. Am J Clin Nutr 96:289–295

    Article  Google Scholar 

  31. 31.

    Bowers K, Yeung E, Williams MA, Qi L, Tobias DK, Hu FB, Zhang C (2011) A prospective study of prepregnancy dietary iron intake and risk for gestational diabetes mellitus. Diabetes Care 34:1557–1563

    Article  Google Scholar 

  32. 32.

    Qiu C, Zhang C, Gelaye B, Enquobahrie DA, Frederick IO, Williams MA (2011) Gestational diabetes mellitus in relation to maternal dietary heme iron and nonheme iron intake. Diabetes Care 34:1564–1569

    CAS  Article  Google Scholar 

  33. 33.

    Fernandez-Cao JC, Arija V, Aranda N, Bullo M, Basora J, Martínez-González MA, Díez-Espino J, Salas-Salvadó J (2013) Heme iron intake and risk of new-onset diabetes in a Mediterranean population at high risk of cardiovascular disease: an observational cohort analysis. BMC Public Health 13:1042

    Article  Google Scholar 

  34. 34.

    Chan KK, Chan BC, Lam KF, Tam S, Lao TT (2009) Iron supplement in pregnancy and development of gestational diabetes—a randomised placebo-controlled trial. BJOG 116:789–798

    CAS  Article  Google Scholar 

  35. 35.

    Bo S, Menato G, Villois P, Gambino R, Cassader M, Cotrino I, Cavallo-Perin P (2009) Iron supplementation and gestational diabetes in midpregnancy. Am J Obstet Gynecol 201(158):e1–e158.e6

    Google Scholar 

  36. 36.

    Zhang C (2010) Risk factors for gestational diabetes-from an epidemiological standpoint. In: Kim C, Ferrara A (eds) Gestational diabetes during and after pregnancy. Springer-Verlag London Limites, London, pp 71–81

    Google Scholar 

  37. 37.

    Lao TT, Ho LF (2004) Impact of iron deficiency anemia on prevalence of gestational diabetes mellitus. Diabetes Care 27:650–656

    Article  Google Scholar 

  38. 38.

    Swaminathan S, Fonseca VA, Alam MG, Shah SV (2007) The role of iron in diabetes and its complications. Diabetes Care 30:1926–1933

    CAS  Article  Google Scholar 

  39. 39.

    Kim Y, Keogh J, Clifton P (2015) A review of potential metabolic etiologies of the observed association between red meat consumption and development of type 2 diabetes mellitus. Metabolism 64:768–779

    CAS  Article  Google Scholar 

  40. 40.

    Willett WC, Colditz GA (1998) Approaches for conducting large cohort studies. Epidemiol Rev 20:91–99

    CAS  Article  Google Scholar 

  41. 41.

    Alonso A, Seguí-Gómez M, de Irala J, Sánchez-Villegas A, Beunza JJ, Martínez-González MA (2006) Predictors of follow-up and assessment of selection bias from dropouts using inverse probability weighting in a cohort of university graduates. Eur J Epidemiol 21:351–358

    Article  Google Scholar 

  42. 42.

    Rothman KJ, Greenland S, Lash TL (2008) Validity in epidemiologic studies. In Modern Epidemiology. 3 th ed. Rothman KJ, Greenland S, Lash TL. Philadelphia, Lippincott Williams & Wilkins, p146–147

    Google Scholar 

  43. 43.

    Food Standards Australia New Zealand (2011) Pregnancy and healthy eating. http://www.foodstandards.gov.au/consumer/generalissues/pregnancy/pages/default.aspx

  44. 44.

    Williamson CS (2006) Nutrition in pregnancy. Br Nutr Found Nutr Bull 31:28–59

    Article  Google Scholar 

Download references

Acknowledgements

We are thankful to Mark Sullivan for the revision of English spelling, grammar, and writing. The SUN Project has received funding from the Spanish Government-Instituto de Salud Carlos III, and the European Regional Development Fund (FEDER) (RD 06/0045, CIBER-OBN, Grants PI10/02658, PI10/02293, PI13/00615, PI14/01668, PI14/01798, PI14/01764, and G03/140), the Navarra Regional Government (45/2011, 122/2014), and the University of Navarra.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Maira Bes-Rastrollo.

Ethics declarations

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Marí-Sanchis, A., Díaz-Jurado, G., Basterra-Gortari, F.J. et al. Association between pre-pregnancy consumption of meat, iron intake, and the risk of gestational diabetes: the SUN project. Eur J Nutr 57, 939–949 (2018). https://doi.org/10.1007/s00394-017-1377-3

Download citation

Keywords

  • Gestational diabetes mellitus
  • Total meat
  • Red and processed meat
  • Heme iron intake
  • Mediterranean population