European Journal of Clinical Nutrition (2015), 1–6 © 2015 Macmillan Publishers Limited All rights reserved 0954-3007/15 www.nature.com/ejcn
ORIGINAL ARTICLE
Dietary patterns and bone mineral density in Brazilian postmenopausal women with osteoporosis: a cross-sectional study NAG de França1, MBR Camargo2, M Lazaretti-Castro2, BSE Peters1 and LA Martini1 BACKGROUND/OBJECTIVES: The aim of this study was to investigate the association between dietary patterns and bone mineral density (BMD) in postmenopausal women with osteoporosis. SUBJECTS/METHODS: This cross-sectional study included 156 postmenopausal and osteoporotic Brazilian women aged over 45 years. BMD of lumbar spine, total femur (TF), femoral neck and of total body (TB), as well as body composition (fat and lean mass), was assessed by dual-energy X-ray absorptiometry. Body mass index and lifestyle information were also obtained. Dietary intake was assessed by using a 3-day food diary. Dietary patterns were obtained by principal component factor analysis. Adjusted multiple linear regression analysis was applied in order to evaluate the predictive effect of dietary patterns on BMD. Significance was set at Po 0.05. RESULTS: Five patterns were retained: ‘healthy’, ‘red meat and refined cereals’, ‘low-fat dairy’, ‘sweet foods, coffee and tea’ and ‘Western’. The ‘sweet foods, coffee and tea’ pattern was inversely associated with TF BMD (β = − 0.178; 95% CI: − 0.039 to − 0.000) and with TB BMD (β = − 0.320; 95% CI: − 0.059 to − 0.017) even after adjusting for energy and calcium intake, lean mass, age and postmenopausal time. CONCLUSIONS: A concomitant excessive consumption of sweet foods and caffeinated beverages appears to exert a negative effect on BMD even when the skeleton already presents some demineralization. Food and beverage intake is a modifiable factor that should not be neglected in the treatment of individuals with osteoporosis. European Journal of Clinical Nutrition advance online publication, 25 March 2015; doi:10.1038/ejcn.2015.27
INTRODUCTION Estimates for the year 2000 revealed approximately 9 million new osteoporotic fractures worldwide, an increase of 25% since 1990,1 with osteoporosis affecting an estimated 200 million women worldwide.2 In Brazil, the epidemiological Brazilian Osteoporosis Study—BRAZOS (2010)3—revealed a self-reported prevalence of 15.1 and 12.8% of low-impact fractures among women and men, respectively, placing a financial burden on the public health system, as expenses with hospital admissions for osteoporotic fractures account for 2% of total hospitalization expenditure among the elderly.4 As osteoporosis is considered a public health problem with an increasing prevalence, the development of strategies to improve the treatment, prevent disease progression and ensure a better quality of life for the patients is fundamental. Diet is recognized as one of the modifiable factors that influence bone maintenance, but only the role of specific micronutrients, most notably calcium and vitamin D, is well elucidated.5 However, people consume meals containing a combination of foods and nutrients, and thus an examination of dietary patterns would more closely reflect the real world by considering the overall eating pattern, and the collinearity of nutrients and foods, which could better express the true eating behavior.6 Dietary patterns reflect a blend of social, cultural, environmental and economic aspects, and they are able to provide more
accessible recommendations.7 Previous studies have investigated the association between dietary patterns and bone mineral density (BMD) among different populations and age groups of both sexes. Dietary patterns characterized by high intakes of refined cereals, processed meats, fried foods (mainly French fries) and sweet foods have been negatively associated with bone.8–13 These patterns consist of food related to increased levels of inflammation (increase in inflammatory markers), which seems to have a negative impact on the bone quality.14 On the other hand, patterns with high amounts of fruit and vegetables have shown a positive relationship with skeletal health.8,12,13,15–18 Although the mechanisms involved in this relationship are still not clearly elucidated, it is suggested that the possible positive effect on bone is related to the alkalizing property and the presence of bioactive compounds (such as antioxidants) in the composition of these foods.19 These results are only pertinent to the risk of developing osteoporosis; however, there is a lack of evidence to confirm whether these results could be extended to osteoporotic individuals, or even if dietary pattern can exert an effect on bone when osteoporosis is already established. The use of a dietary pattern approach to evaluate the association between diet and bone health warrants further studies, considering the wide variability in diets among different countries. Furthermore, the study of this association may help bridge the gap in knowledge of the influence of overall diet on
1 Department of Nutrition, School of Public Health, Sao Paulo University, Food and Nutrition Research Center-NAPAN, Sao Paulo, Brazil and 2Division of Endocrinology, School of Medicine, Federal University of Sao Paulo, Sao Paulo, Brazil. Correspondence: Professor LA Martini, Department of Nutrition, School of Public Health, Sao Paulo University, Food and Nutrition Research Center-NAPAN, Sao Paulo 01246-904, Brazil. E-mail: [email protected] Received 30 September 2014; revised 15 December 2014; accepted 26 January 2015
Dietary patterns in women with osteoporosis NAG de França et al
2 osteoporosis treatment, providing data that may contribute to improvements in osteoporotic patient care. Therefore, the aim of the current study was to determine the dietary patterns of postmenopausal women with osteoporosis using factor analysis, and to investigate the association between these patterns and BMD. SUBJECTS AND METHODS Study design and population From among 363 individuals assessed at a specialized outpatient clinic of the Federal University of Sao Paulo in Sao Paulo city, Brazil, for the treatment of osteoporosis, only postmenopausal women aged over 45 years with at least 2 years of absence of menstrual cycles (menopause) and undergoing osteoporosis follow-up for at least 3 months were selected. Subjects with a history of renal failure, absorption disorders, current use of glucocorticoids, current hyperparathyroidism and incomplete data were excluded. Although it is known that corticosteroids and some illnesses, as hyperparathyroidism, are a major cause of secondary osteoporosis, women who were on steroid therapy in the past or those who already have been treated for the disease were not excluded from the study because the authors believe that the study’s purpose was not to investigate the cause of osteoporosis but their current relationship with eating habits; thus, a final total of 156 free-living women were eligible to take part in the study, and they agreed to participate by signing an informed consent form. All subjects formerly received or were receiving treatment for osteoporosis with an approved medication, and/or calcium and/or vitamin D supplementation. The study protocol was approved by the Federal University of Sao Paulo Research Ethics Committee (0839/08), and enrollment occurred between late autumn and late spring, from 2009 to 2012.
Bone mineral density measurement Measurements for the determination of BMD (g/cm2) of the lumbar spine (L1–L4), total femur (TF), femoral neck (FN) and total body (TB) were conducted by using dual-energy X-ray absorptiometry (Discovery A, QDR for Windows XP, Hologic, Inc., Bedford, MA, USA). The analyses were performed by a highly trained technologist using the same technique for all measurements to avoid disparities in size or position of regions of interest, and the scan was determined on the basis of the height and weight of the participant. BMD was assessed and analyzed according to the recommendations of the Brazilian Society for Clinical Densitometry.20 Osteoporosis was defined as a BMD T-score for the TF and/or FN and/or lumbar spine of ⩽ 2.5 SDs.21 The coefficients of variation (CVs) of lumbar spine BMD and TF BMD was 0.8% and of FN BMD was 1.2%.
Table 1.
Dietary assessment and food grouping Dietary intake was assessed by using a 3-day food diary in which the participants were clearly instructed to record information on nonconsecutive days (2 weekdays and 1 weekend day) on their food and beverage intake using household measures. Participants provided details on brand names of food products, food preparation methods and any recipes used. This method reduces the memory bias, provides detailed information about the current food intake and allows estimates of usual intake.22 Food and beverage quantities obtained from the 3-day food diary were converted into grams and milliliters and computed using the Nutrition Data System for Research software (NDS-R, 2007 version, University of Minnesota, MN, USA). To reduce the complexity of the data, the 270 individual food items identified from the diary (in grams/day) were manually allocated into 13 food groups constructed according to the principles of similarity of nutrient profiles or culinary usage of the foods (Table 1), and they were subsequently used to identify the dietary patterns.
Confounding factor measurements Factors that appear to be related to BMD and dietary patterns were assessed, which included body weight, height, body fat, physical activity, smoking habit, alcohol intake, history of bone fracture, supplement use, postmenopausal time and current treatment received. Body weight was measured to the nearest 0.1 kg by using a manual scale (Filizola, Sao Paulo, Brazil), with the participants wearing light clothing and no footwear, and height was measured to the nearest 0.1 cm with a wall-mounted stadiometer (Tonelli, Criciuma, Brazil), after removal of footwear. Body mass index was calculated by dividing the TB weight (kg) by height squared (m2) and classified according to the World Health Organization classification23 for participants aged under 60 years, and according to the Pan American Health Organization24 for subjects aged 60 years or over. Body fat (g) and lean mass (g) were obtained from the TB dual-energy X-ray absorptiometry scan. Age (in years, continuous), practice of physical activity, considering only activities for recreation, sport, exercise or leisure performed for at least 10 continuous minutes (yes or no), current smoking and alcohol intake for the last 3 months (yes or no for both), bone fracture since the age of 45 years (yes or no), postmenopausal time (in years, continuous), calcium or vitamin D supplementation in the last 3 months (yes or no) and the use of bone-related medication (within the last 3 months) were obtained from a two-page questionnaire.
Statistical analyses The derivation of the dietary patterns was conducted by applying the principal component factor analysis method. Principal component factor
The 13 food groupings used in the dietary pattern analysis
Foods group
Foods in the group
Fruit (fresh or dried) and juices
Apple, papaya, banana, tangerine, lemon, peach, grape, strawberry, lemonade, melon, pineapple, watermelon, mango, pear, orange juice, star fruit, kiwi, acerola juice, avocado, coconut, coconut water, cherimoya, melon juice, passion fruit juice, raisins, plum, apricot, guava, fruit salad, persimmon. Fruit-flavored drinks, cola and noncola soft drinks, light and diet soft drinks. Light cream cheese, skimmed milk, semi-skimmed milk, white cheese, low-fat yogurt, cottage cheese, ricotta cheese, light coffee cream. High-fat yogurt, mozzarella cheese, cream cheese, parmesan cheese, whole milk, goat milk. Whitefish, fresh sardines, canned sardines, canned tuna, salmon, snapper, codfish, mullet, shrimp. Beef, hamburger, liver, jerked beef, pork, pork ribs. Cakes, refined sugar, brown sugar, honey, jelly, chocolate, puddings, condensed milk, cappuccino, cookies, mousse, dulce de leche, sugar cane beverage, ice cream, pies, shredded sweetened coconut, chocolate drinks, jam, fruit desserts, chocolate syrup, panettone, canned fruit, baked candy. Carrot, onion, pumpkin, endive, lettuce, peppers, zucchini, eggplant, okra, chard, chayote, cabbage, tomatoes, spinach, garlic, olives, broccoli, beets, cauliflower, cabbage, fennel, kelp, chicory, celery, pumpkin, parsley, cucumber, watercress, turnip, radish. Potato, cassava, yams, sweet potato. Rice, crackers, pasta, wheat bread, wheat flour, cheese bread, corn bread, potato bread, pancakes, cornmeal, gnocchi, croutons, breakfast cereal, canned corn, couscous. Pizza, chips, hotdog, French fries, pies, popcorn, baked and fried snacks. Coffee, decaffeinated coffee, espresso, herbal tea, yerba mate. Margarine, butter, mayonnaise.
Soft drinks and fruit drinks Low-fat dairy High-fat dairy Fish and sea food Beef and pork meat Sweet foods Vegetables Tubers and tuberous roots Refined cereals Snacks, pizzas and pies Coffee and tea Fats
European Journal of Clinical Nutrition (2015) 1 – 6
© 2015 Macmillan Publishers Limited
Dietary patterns in women with osteoporosis NAG de França et al
3 analysis method is a multivariate technique that evaluates the intercorrelations between the initial food variables and reduces food groups into a smaller number of factors (patterns) that can explain variations in the dietary intake. The number of factors retained was based on a combination of food group components with an eigenvalue 41.0, the percentage of variance criteria (sufficient factors to meet at least 60% of variance explained)25 and examination of the breakpoint in the scree plot, resulting in five factors retained for further analyses. The factors were rotated by orthogonal transformation (Varimax option) to achieve a simpler structure with a greater interpretability. Food groups with a factor loading ⩾ 0.45 on a component were considered informative for interpretation of the dietary patterns.25 The factor scores for each pattern and individual were determined by summing the intakes of each food group weighted by the factor loading. The groups that had a negative factor loading were also retained in order to determine the complexity of eating habits. Normality of continuous variables was assessed by the Kolmogorov– Smirnov test. To investigate the predictive effect of dietary patterns on BMD (dependent), Pearson’s correlation was conducted followed by multiple linear regression analysis. The linear regression models included the BMD at each evaluated site (lumbar spine, FN, TF and TB) as a dependent variable and the dietary patterns (in scores) as independent variables. The model was adjusted for energy intake (kcal), calcium intake (mg), lean mass (g), height (cm) and postmenopausal time (years). The principal component factor analysis was performed using SAS Enterprise Guide, version 4.1 (SAS Institute, Inc., Cary, NC, USA), whereas the correlation and regression analyses were performed using the SPSS software, version 20.0 (SPSS, Chicago, IL, USA. The level of significance was set at Po0.05.
RESULTS General characteristics of the participants are presented in Table 2. The participants had a mean age of 68.4 (9.0) years and body mass index value of 25.9 (3.9) kg/m2 (according to body mass index classification, 19% of the sample was underweight, 50% were normal weight and 31% were overweight—table not shown), with body fat altering only slightly. The average intake of calcium was 832.5 (442.7) mg, a very close value to the acceptable by the estimated average requirements for women aged over 50 years (1000 mg).26 Vitamin D ingested by diet was 4.4 (3.0) μg, bellow the requirement of 10 μg.26 Most of the participants had less than 5 years of education (low educational level; 79%), did not exercise (54%), did not smoke (91%) and were taking vitamin D or calcium supplementation (85 and 53.2%, respectively). The main treatment for osteoporosis was through the use of bisphosphonates (75%). The factor-loading matrices for the five patterns retained are shown in Table 3. The high positive loadings indicate strong associations between food groups and patterns, whereas high negative loadings indicate a strong negative relation. Patterns were labeled according to the food groups with positive loadings. Dietary pattern 1 showed heavy loadings on vegetables (average intake of the entire sample: 87.4 ± 94.7 g/day; ranging from 0–725 g/day), fruit (fresh and dried) and fresh juices (251.6 ± 221.2 g/day; 0–1,166 g/day) and tubers and tuberous roots (25.6 ± 54.7 g/day; 0–366.7 g/day) and was labeled ‘Healthy’. Dietary pattern 2, with high loadings for refined cereals and baked products with refined cereals (230.9 ± 129.9 g; 0–786 g/day) and beef and pork meat (59.2 ± 70.3 g/day; 0–450 g/day), was labeled ‘Red meat and refined cereals’. Dietary pattern 3 had heavy loadings on low-fat dairy products and was labeled ‘Low-fat dairy’ (125.9 ± 201.2 g/day; 0–1,019.5 g/day). Dietary pattern 4, with high loadings for sugar, sugary products (42.8 ± 74.9 g/day; 0–502.5 g/ day) and coffee and tea (138.5 ± 149.3 g/day; 0–1,008.9 g/day), was labeled ‘Sweet foods, coffee and tea’. Dietary pattern 5 showed heavy loadings on fats (9.0 ± 19.2 g; 0–207.1 g/day), snacks, pizzas, pies (30.0 ± 97.1 g/day; 0–823.3 g/day) and soft drinks and fruit drinks (59.3 ± 136.6 g/day; 0–1,164.9 g/day), and it was labeled ‘Western’. Overall, the five dietary patterns accounted for 60.4% of the variance in food intakes. © 2015 Macmillan Publishers Limited
Table 2. General characteristics of the osteoporotic postmenopausal women studied (n = 156) Postmenopausal women with osteoporosis Age (years) Weight (kg) Body mass index (kg/m2) Body fat (%) Postmenopausal time (years) Energy intake (kcal) Carbohydrate intake (% of total calories) Fat intake (% of total calories) Protein intake (% of total calories) Calcium intake (mg) Vitamin D intake (μg) Lumbar spine BMD (g/cm2) Lumbar spine BMD (T-score) Femoral neck BMD (g/cm2) Femoral neck BMD (T-score) Total femur BMD (g/cm2) Total femur BMD (T-score) Total BMD (g/cm2) Total BMD (T-score) Fracture (yes)a (%) Educational level ⩽ 5 years (%) 5–8 years (%) 48 years (%) Physical activity (yes) (%) Smoking (yes) (%) Alcohol intake (yes) (%) Vitamin D supplementation (yes) (%) Calcium supplementation (yes) (%) Bisphosphonates (%)
68.4 (9.0) 60.4 (10.1) 25.9 (3.9) 36.0 (5.5) 22.2 (10.1) 1627 (580) 52.2 (9.1) 30.5 (7.1) 18.4 (5.1) 832.5 (442.7) 4.4 (3.0) 0.745 (0.118) − 2.8 (1.1) 0.648 (0.092) − 1.8 (0.8) 0.742 (0.109) − 1.6 (0.9) 0.945 (0.115) − 1.8 (1.3) 42.9 78.8 15.4 5.8 46.2 9.0 0.0 85.3 53.2 75.0
Abbreviation: BMD, bone mass density. Continuous variables are expressed as mean (standard deviation) and categorical variables as percentages. a After age 45.
Table 4 shows the results for the linear regression conducted between the dietary patterns and BMD at each evaluated site. ‘Sweet foods, coffee and tea’ was the only pattern that showed a significant correlation with BMD. As it was inversely associated with TF BMD (β = − 0.206; 95% CI: − 0.042– − 0.004) and with TB BMD (β = − 0.321; 95% CI: − 0.059– − 0.017), this dietary pattern was considered as a negative independent predictor of BMD. DISCUSSION Although there are numerous studies exploring the relationships between dietary patterns and BMD, the present study is unique in that it conducted such an evaluation among postmenopausal women who already presented with osteoporosis. The main finding showed that the ‘sweet foods, coffee and tea’ pattern was inversely correlated with BMD (in femur and TB) for the overall sample. Comparisons of our results with previous findings are hampered by differences in protocols, food habits of the studied populations, food groups formed and in the dietary patterns identified.9,12,13,15,27 Furthermore, previous studies have evaluated the risk factors for the incidence of osteoporosis, in contrast with the present study, which involved osteoporotic women. However, some similarities can be observed with the dietary patterns derived in other studies and their relationships with BMD. The inverse association between BMD and dietary patterns with high loadings for sweet foods was previously shown by Tucker et al.13 in elderly adults from the Framingham Osteoporosis Study. The mean BMD in the ‘candy group’ was significantly lower for the Ward’s area, FN and radius (P o 0.001) among men, and for radius European Journal of Clinical Nutrition (2015), 1 – 6
Dietary patterns in women with osteoporosis NAG de França et al
4 Table 3.
Factor-loading matrix for the five dietary patterns identified among the 156 osteoporotic postmenopausal women
Vegetables Fruit and fresh juices Tubers and tuberous roots Fats Fish and sea food High-fat dairy Low-fat dairy Sweet foods Coffee and tea Snacks, pizzas and pies Refined cereals Beef and pork meat Soft drinks and fruit drinks Percentage of variance (%)
‘Healthy’
‘Red meat and refined cereals’
‘Low-fat dairy’
‘Sweet foods, coffee and tea’
‘Western’
0.768 0.608 0.568 — — — — — — — — — — 19.2
— — — — − 0.472 — — — — — 0.726 0.666 — 13.9
— — — — — − 0.762 0.829 — — — — — — 10.8
— — — — — — — 0.736 0.816 — — — — 8.7
— — — 0.547 — — — — — 0.746 — — 0.554 7.8
Table 4.
Results of adjusted linear regression analysis (β-coefficient), and 95% confidence interval of the dietary patterns (score values) and body mineral density (g/cm2) Dietary patterns ‘Healthy’ ‘Red meat and refined cereals’ ‘Low-fat dairy’ ‘Sweet foods, coffee and tea’ ‘Western’
Lumbar spine BMD β (95% CI) − 0.002 − 0.094 0.141 − 0.066 0.069
(−0.022 (−0.031 (−0.003 (−0.030 (−0.012
to to to to to
0.021) 0.010) 0.035) 0.014) 0.028)
Femoral neck BMD β (95% CI) 0.164 − 0.005 − 0.069 − 0.155 0.099
(−0.002 (−0.016 (−0.021 (−0.031 (−0.007
to to to to to
0.032) 0.015) 0.009) 0.002) 0.025)
Total femur BMD β (95% CI) 0.184 0.038 − 0.067 − 0.206* 0.052
Total body BMD β (95% CI)
(0.001 to 0.038) (−0.014 to 0.022) (−0.024 to 0.010) (−0.042 to − 0.004) (−0.012 to 0.023)
0.087 − 0.019 0.113 − 0.321* 0.039
(−0.011 (−0.023 (−0.006 (−0.059 (−0.016
to to to to to
0.032) 0.018) 0.032) −0.017) 0.025)
Abbreviations: BMD, bone mineral density; CI, confidence interval. *Po0.05. Adjusted for energy intake (kcal), calcium intake (mg), lean mass (g), height (cm) and postmenopausal time (years).
among women (P = 0.004). Similar results were obtained by Hardcastle et al.12 and Park et al.,27 who also found negative correlations between patterns rich in sweet foods and bone health. Both of the studies indicated a negative predictive impact of sugary food intake on the risk of developing osteoporosis. A possible explanation for the negative effect of glucose intake could be related to increased glycemia, which appears to be associated with changes in bone mineral homeostasis.28 A highsucrose diet produces hyperinsulinemia, which in turn could induce hypercalciuria by inhibiting renal tubular resorption of calcium, affecting the quality of bone mineralization.29 There has been little discussion concerning the effects of hyperglycemia on skeletal integrity. Chronically elevated glucose even for short periods during the lifetime could be deleterious for the skeleton, particularly during growth and peak bone mass formation.30 Once an adequate bone formation is essential to attenuate the deleterious effects associated with aging, any change in peak bone mass could lead to the development of osteoporosis and/or to an increased risk of fractures in adulthood.31,32 However, it is emphasized that the present study did not investigate lifetime dietary habits of the women studied, which does not allow making any inference about the impact of glycemic peaks on late bone demineralization. According to the World Health Organization (2002),33 sugar consumption should not exceed 10% of total energy per day; however, the new draft guideline proposes a reduction to below 5%,34 as the consumption of free sugars may result in reduced intake of foods containing more nutritionally adequate calories, leading to an unhealthy diet and increased risk of noncommunicable diseases.33 Among our sample, the average intake of sugar and sugary foods group was 42.8 g/day (74.9 g/day), accounting for 28.6 g of added sugar, which implies 7% of the total average energy intake. This value is in agreement with the current recommendation, but it is above the new draft. Furthermore, the added sugar reached 165 g, which is an excessive intake and European Journal of Clinical Nutrition (2015) 1 – 6
could lead to the negative effects on bone. The results from our sample were similar to the findings reported by Bueno et al.35 showing an average intake of 8.5% (33.4 g) in a sample of 622 elderly individuals, with significant differences between the sexes (7.8% for men vs 8.7% for women; P o 0.05). Taken together, these results hypothesized that consumption of added sugar is a common food habit among elderly individuals, especially women. The ‘sweet foods, coffee and tea’ pattern also contains caffeinated beverages. Results of investigations into the negative impact of caffeine on bone are conflicting. Current evidence suggests that its detrimental effect on BMD might only occur in susceptible individuals (that is, inadequate calcium intake, osteoporosis predisposition, old age), where it has been suggested that caffeine consumption could decrease calcium absorption and/or increase excretion of calcium in this group.36 Rapuri et al. 37 showed that intakes of more than 300 mg of caffeine (≈514 g of brewed coffee) accelerate bone loss at the spine in elderly postmenopausal women, with greater risk among those with the tt genetic variant of the vitamin D receptor. Although the average intake of caffeine among the sample as a whole was 48.2 mg, daily caffeine intake ranged from 0 to 332 mg. It is also suggested that caffeine could have a deleterious effect on osteoblasts by inducing apoptosis via a mitochondria-dependent pathway.38 These apoptotic biochemical changes could be a result of an intracellular oxidative stress stimulated by caffeine, leading to a loss of BMD.38 Furthermore, there are several observations indicating an adverse effect of caffeine on glucose metabolism.39–41 Petrie et al.40 showed a significantly attenuation of whole-body insulin sensitivity in obese, nondiabetic persons, after acute caffeine ingestion, with no evidence over beta cell secretion. The inconsistent results about caffeine indicate that also other coffee constituents may have unfavorable skeletal effects. Recently, in a study with female rats, Folwarczna et al.42 observed a negative impact on bone associated to the presence of © 2015 Macmillan Publishers Limited
Dietary patterns in women with osteoporosis NAG de França et al
trigonelline (a bitter alkaloid related to the production of aroma compounds), according to the estrogen status. The trigonelline administration worsened the mineralization of the vertebra and the strength of the tibial metaphysis, and decreased the width of femoral trabeculae in ovariectomized (estrogen-deficient) Wistar rats, suggesting that the possible detrimental effect of this alkaloid may be stronger in postmenopausal women.42 Thereby, the consumption of caffeinated beverage combined with sugar could exacerbate the detrimental effect on bone through a synergic relationship. BMD is an important predictor of fractures.43 According to a classic meta-analysis conducted by Marshall et al, the risk of hip and total fractures increases 2.6-fold and 1.6-fold, respectively, for each standard deviation decrease in BMD.44 Hip, which includes the proximal femur, accounts for the main site for fractures, causing acute pain, loss of function and usually leading to hospitalization with a complicate recovery.45 It is estimated that the strength of bone decreases about 2–12% per decade in postmenopausal women, with an average age for hip fractures about 80 years.45 Thus, as the ‘sweet foods, coffee and tea’ pattern showed an inverse relationship with TF BMD (β: -0.206; 95% CI: − 0.042 – − 0.004), it is suggested that this dietary habit could increase the risk for latest fracture in hip. Data from the National Health and Nutrition Examination Survey (NHANES 2005–2008)46 determine the TF BMD cutoff for women between 60 and 69 years as 0.855 ± 0.126 g/cm2, classifying our sample between the 15th and 25th percentile (0.742 ± 0.109 g/cm2). Thereby, a decrease of -0.206 g/cm2 associated with the ‘Sweet foods, coffee and tea’ pattern would imply a more prominent demineralization in this site, significantly accelerating the risk of late fractures. Most dietary pattern studies have shown a positive association between a diet rich in fruit and/or vegetables and bone health.8,12,13,16–18 This same association was not observed in our sample, may be because the influence of these foods is too low to detect in a small sample. However, the authors consider that the intake of fruit and vegetables is supported by the beneficial effect of their antioxidant content47 and alkalizing property.48 Furthermore, no correlation was detected between the ‘low-fat dairy’ pattern and BMD in our study. This nonsignificant association may be because the sample comprised osteoporotic women who had already received some orientation to consume higher amounts of dairy, leading to less remarkable results. Average intake from the dairy group was 5 portions/day (data not shown), considering 120 kcal/dairy portion,49 representing a level of consumption above the recommendation of 3 dairy portions/day proposed by the Food Guide for the Brazilian Population49 and an adequate calcium intake (832.5 mg). It should also be emphasized that, although the participants had received or were receiving treatment for osteoporosis, they were not instructed to make changes in their overall diet. This suggests that the specific orientations to increase calcium intake, mainly by increasing dairy consumption, may be insufficient to ensure a protective effect on bone. Besides sample size, the present study has several other limitations. First, this study used a cross-sectional approach, precluding the inference of a causal relationship. Second, as the participants were recruited from a specialized outpatient clinic in Sao Paulo city, our sample was not representative of postmenopausal women with osteoporosis in general. Finally, the minimal nutritional orientation received may have exerted some influence on the results. The dietary pattern approach a posteriori is based on current data, enabling examination of the multidimensionality of the total diet50 and may have major implications for public health, as it is more readily interpreted and translated to the daily diet of the population. In addition, the approach provides underlying knowledge of human feeding practices,6 elucidated in the present study in the form of an inverse association between the ‘Sweet foods, coffee and tea’ pattern and BMD (in femur and TB). © 2015 Macmillan Publishers Limited
CONCLUSIONS A concurrent excessive consumption of sweet foods and caffeinated beverages showed a negative effect on BMD even when the skeleton already presented demineralization. Food and beverage intake is a modifiable factor that should not be neglected in osteoporosis treatment. The findings of the present study suggest that dietary orientations for osteoporotic individuals should not be limited to calcium and vitamin D intake, but extend to encompass the overall diet. CONFLICT OF INTEREST The authors declare no conflict of interest.
ACKNOWLEDGEMENTS We thank Wesley Rodrigues dos Santos and Anatoli Yambartsev, PhD, from the Institute of Mathematics and Statistics of Sao Paulo University, for the dietary pattern analysis, and the funding organization São Paulo Research Foundation (Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)).
REFERENCES 1 Johnell O, Kanis JA. An estimate of the worldwide prevalence and disability associated with osteoporotic fractures. Osteoporos Int 2006; 17: 1726–1733. 2 Kanis JA. Assessment of Osteoporosis at the Primary Health-Care Level Technical Report University of Sheffield. World Health Organization: UK, 2007. 3 Pinheiro MM, Ciconelli RM, Jacques Nde O, Genaro PS, Martini LA, Ferraz MB. The burden of osteoporosis in Brazil: regional data from fractures in adult men and women--the Brazilian Osteoporosis Study (BRAZOS). Rev Bras Reumatol 2010; 50: 113–127. 4 Bortolon PC, Andrade CL, Andrade CA. Characteristics of hospital admissions in the Unified National Health System for osteoporotic hip fracture in elderly people in Brazil, 2006-2008. Cad Saude Publica 2011; 27: 733–742. 5 Bonjour JP, Gueguen L, Palacios C, Shearer MJ, Weaver CM. Minerals and vitamins in bone health: the potential value of dietary enhancement. Br J Nutr 2009; 101: 1581–1596. 6 Hu FB. Dietary pattern analysis: a new direction in nutritional epidemiology. Curr Opin Lipidol 2002; 13: 3–9. 7 National Research Council—Committee on Diet and Health. Diet and Health: Implications for Reducing Chronic Disease Risk. National Academy Press: Washington, DC, 1989. 8 McNaughton SA, Wattanapenpaiboon N, Wark JD, Nowson CA. An energy-dense, nutrient-poor dietary pattern is inversely associated with bone health in women. J Nutr 2011; 141: 1516–1523. 9 Whittle CR, Woodside JV, Cardwell CR, McCourt HJ, Young IS, Murray LJ et al. Dietary patterns and bone mineral status in young adults: the Northern Ireland Young Hearts Project. Br J Nutr 2012; 108: 1494–1504. 10 Fairweather-Tait SJ, Skinner J, Guile GR, Cassidy A, Spector TD, MacGregor AJ. Diet and bone mineral density study in postmenopausal women from the TwinsUK registry shows a negative association with a traditional English dietary pattern and a positive association with wine. Am J Clin Nutr 2011; 94: 1371–1375. 11 Karamati M, Jessri M, Shariati-Bafghi SE, Rashidkhani B. Dietary patterns in relation to bone mineral density among menopausal Iranian women. Calcif Tissue Int 2012; 91: 40–49. 12 Hardcastle AC, Aucott L, Fraser WD, Reid DM, Macdonald HM. Dietary patterns, bone resorption and bone mineral density in early post-menopausal Scottish women. Eur J Clin Nutr 2011; 65: 378–385. 13 Tucker KL, Chen H, Hannan MT, Cupples LA, Wilson PW, Felson D et al. Bone mineral density and dietary patterns in older adults: the Framingham Osteoporosis Study. Am J Clin Nutr 2002; 76: 245–252. 14 Cauley JA, Danielson ME, Boudreau RM, Forrest KY, Zmuda JM, Pahor M et al. Inflammatory markers and incident fracture risk in older men and women: the Health Aging and Body Composition Study. J Bone Miner Res 2007; 22: 1088–1095. 15 Shin S, Joung H. A dairy and fruit dietary pattern is associated with a reduced likelihood of osteoporosis in Korean postmenopausal women. Br J Nutr 2013; 110: 1926–1933. 16 Langsetmo L, Hanley DA, Prior JC, Barr SI, Anastassiades T, Towheed T et al. Dietary patterns and incident low-trauma fractures in postmenopausal women and men aged 4/ = 50 y: a population-based cohort study. Am J Clin Nutr 2011; 93: 192–199. 17 Okubo H, Sasaki S, Horiguchi H, Oguma E, Miyamoto K, Hosoi Y et al. Dietary patterns associated with bone mineral density in premenopausal Japanese farmwomen. Am J Clin Nutr 2006; 83: 1185–1192.
European Journal of Clinical Nutrition (2015), 1 – 6
5
Dietary patterns in women with osteoporosis NAG de França et al
6 18 Wosje KS, Khoury PR, Claytor RP, Copeland KA, Hornung RW, Daniels SR et al. Dietary patterns associated with fat and bone mass in young children. Am J Clin Nutr 2010; 92: 294–303. 19 Macdonald HM, Hardcastle AC. Dietary patterns and bone health. In: Burckhardt P, Dawson-Hughes B, Weaver CM. Nutritional Influences on Bone Health. Springer, pp 135–143, 2010. 20 Brandao CM, Camargos BM, Zerbini CA, Plapler PG, Mendonca LM, Albergaria BH et al. 2008 official positions of the Brazilian Society for Clinical Densitometry-SBDens. Arq Bras Endocrinol Metabol 2009; 53: 107–112. 21 ISCD—The International Society For Clinical Densitometry The Official Positions 2013. 22 Fisberg RM, Slater B, Marchioni DML, Martini LA. Inquéritos Alimentares: Métodos e Bases Científicos. Manole: Barueri, 2005. 23 World Health Oraganization: Physical status: the use and interpretation of anthropometry. Report of a WHO Expert Committee. WHO: Geneva, 1995. 24 Organización Panamericana de la Salud. División de Promoción y Protección de la Salud (HPP) Encuesta multicentrica salud bienestar y envejecimiento (SABE) en América Latina: informe preliminar Multicenter survey aging, health and wellbeing in Latin América and the Caribbean (SABE): preliminary report. Reunión del Comité Asesor de Investigaciónes en Salud: Kingston, Washington, DC. p 19, 2001. 25 Hair Jr J, Black W, Babin B, Anderson R, Tatham R. Multivariate Data Analysis, 6th ed. Prentice Hall: New York, NY, 2006. 26 Food and Nutrition Board, Institute of Medicine. Phosphorus Dietary Reference Intakes: Calcium, Phosphorus, Magnesium, Vitamin D, and Fluoride. National Academy Press: Washington, DC. pp 146–189, 1997. 27 Park SJ, Joo SE, Min H, Park JK, Kim Y, Kim SS et al. Dietary patterns and osteoporosis risk in postmenopausal korean women. Osong Public Health Res Perspect 2012; 3: 199–205. 28 Bass EF, Baile CA, Lewis RD, Giraudo SQ. Bone quality and strength are greater in growing male rats fed fructose compared with glucose. Nutr Res 2013; 33: 1063–1071. 29 Holl MG, Allen LH. Sucrose ingestion, insulin response and mineral metabolism in humans. J Nutr 1987; 117: 1229–1233. 30 Rosen CJ. Sugar and bone: a not-so sweet story. J Bone Miner Res 2008; 23: 1881–1883. 31 Baxter-Jones AD, Faulkner RA, Forwood MR, Mirwald RL, Bailey DA. Bone mineral accrual from 8 to 30 years of age: an estimation of peak bone mass. J Bone Miner Res 2011; 26: 1729–1739. 32 Boot AM, de Ridder MA, van der Sluis IM, van Slobbe I, Krenning EP, KeizerSchrama SM. Peak bone mineral density, lean body mass and fractures. Bone 2010; 46: 336–341. 33 World Health Organization. The World Health Report: Reducing Risks, Promoting Healthy Life. WHO: Geneva, 2002. 34 World Health Organization WHO opens public consultation on draft sugars guideline (acessed on May 2nd, 2014); Available from: http://www.who.int/med iacentre/news/notes/2014/consultation-sugar-guideline/en/ 2014.
European Journal of Clinical Nutrition (2015) 1 – 6
35 Bueno MB, Marchioni DM, Cesar CL, Fisberg RM. Added sugars: consumption and associated factors among adults and the elderly. Sao Paulo, Brazil. Rev Bras Epidemiol 2012; 15: 256–264. 36 Dew TP, Day AJ, Morgan MR. Bone mineral density, polyphenols and caffeine: a reassessment. Nutr Res Rev 2007; 20: 89–105. 37 Rapuri PB, Gallagher JC, Kinyamu HK, Ryschon KL. Caffeine intake increases the rate of bone loss in elderly women and interacts with vitamin D receptor genotypes. Am J Clin Nutr 2001; 74: 694–700. 38 Lu PZ, Lai CY, Chan WH. Caffeine induces cell death via activation of apoptotic signal and inactivation of survival signal in human osteoblasts. Int J Mol Sci 2008; 9: 698–718. 39 Lane JD, Feinglos MN, Surwit RS. Caffeine increases ambulatory glucose and postprandial responses in coffee drinkers with type 2 diabetes. Diabetes Care 2008; 31: 221–222. 40 Petrie HJ, Chown SE, Belfie LM, Duncan AM, McLaren DH, Conquer JA et al. Caffeine ingestion increases the insulin response to an oral-glucose-tolerance test in obese men before and after weight loss. Am J Clin Nutr 2004; 80: 22–28. 41 Keijzers GB, De Galan BE, Tack CJ, Smits P. Caffeine can decrease insulin sensitivity in humans. Diabetes Care 2002; 25: 364–369. 42 Folwarczna J, Zych M, Nowinska B, Pytlik M, Janas A. Unfavorable effect of trigonelline, an alkaloid present in coffee and fenugreek, on bone mechanical properties in estrogen-deficient rats. Mol Nutr Food Res 2014; 58: 1457–1464. 43 Cummings SR, Black DM, Nevitt MC, Browner W, Cauley J, Ensrud K et al. Bone density at various sites for prediction of hip fractures. The Study of Osteoporotic Fractures Research Group. Lancet 1993; 341: 72–75. 44 Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ 1996; 312: 1254–1259. 45 Kanis JA. On behalf of the World Health Organization Scientific Group. Assessment of Osteoporosis at the Primary Health-care Level Technical Report World Health Organization Collaborating Centre for Metabolic Bone Diseases. University of Sheffield: UK, 2007. 46 Looker AC, Borrud LG, Hughes JP, Fan B, Shepherd JA, Melton 3rd LJ . Lumbar spine and proximal femur bone mineral density, bone mineral content, and bone area: United States, 2005–2008. Vital Health Stat 11 2012; (251): 1–132. 47 Nieves JW. Skeletal effects of nutrients and nutraceuticals, beyond calcium and vitamin D. Osteoporos Int 2013; 24: 771–786. 48 Lanham-New SA. The balance of bone health: tipping the scales in favor of potassium-rich, bicarbonate-rich foods. J Nutr 2008; 138: 172S–177S. 49 Ministério da Saúde. Secretaria de Atenção à Saúde. Coordenação-Geral da Política de Alimentação e Nutrição Guia Alimentar Para a População Brasileira: Promovendo a Alimentação Saudável Série A. Normas e Manuais Técnicos: Brasilia, 2006. 50 Newby PK, Tucker KL. Empirically derived eating patterns using factor or cluster analysis: a review. Nutr Rev 2004; 62: 177–203.
© 2015 Macmillan Publishers Limited
ORIGINAL ARTICLE
Dietary patterns and bone mineral density in Brazilian postmenopausal women with osteoporosis: a cross-sectional study NAG de França1, MBR Camargo2, M Lazaretti-Castro2, BSE Peters1 and LA Martini1 BACKGROUND/OBJECTIVES: The aim of this study was to investigate the association between dietary patterns and bone mineral density (BMD) in postmenopausal women with osteoporosis. SUBJECTS/METHODS: This cross-sectional study included 156 postmenopausal and osteoporotic Brazilian women aged over 45 years. BMD of lumbar spine, total femur (TF), femoral neck and of total body (TB), as well as body composition (fat and lean mass), was assessed by dual-energy X-ray absorptiometry. Body mass index and lifestyle information were also obtained. Dietary intake was assessed by using a 3-day food diary. Dietary patterns were obtained by principal component factor analysis. Adjusted multiple linear regression analysis was applied in order to evaluate the predictive effect of dietary patterns on BMD. Significance was set at Po 0.05. RESULTS: Five patterns were retained: ‘healthy’, ‘red meat and refined cereals’, ‘low-fat dairy’, ‘sweet foods, coffee and tea’ and ‘Western’. The ‘sweet foods, coffee and tea’ pattern was inversely associated with TF BMD (β = − 0.178; 95% CI: − 0.039 to − 0.000) and with TB BMD (β = − 0.320; 95% CI: − 0.059 to − 0.017) even after adjusting for energy and calcium intake, lean mass, age and postmenopausal time. CONCLUSIONS: A concomitant excessive consumption of sweet foods and caffeinated beverages appears to exert a negative effect on BMD even when the skeleton already presents some demineralization. Food and beverage intake is a modifiable factor that should not be neglected in the treatment of individuals with osteoporosis. European Journal of Clinical Nutrition advance online publication, 25 March 2015; doi:10.1038/ejcn.2015.27
INTRODUCTION Estimates for the year 2000 revealed approximately 9 million new osteoporotic fractures worldwide, an increase of 25% since 1990,1 with osteoporosis affecting an estimated 200 million women worldwide.2 In Brazil, the epidemiological Brazilian Osteoporosis Study—BRAZOS (2010)3—revealed a self-reported prevalence of 15.1 and 12.8% of low-impact fractures among women and men, respectively, placing a financial burden on the public health system, as expenses with hospital admissions for osteoporotic fractures account for 2% of total hospitalization expenditure among the elderly.4 As osteoporosis is considered a public health problem with an increasing prevalence, the development of strategies to improve the treatment, prevent disease progression and ensure a better quality of life for the patients is fundamental. Diet is recognized as one of the modifiable factors that influence bone maintenance, but only the role of specific micronutrients, most notably calcium and vitamin D, is well elucidated.5 However, people consume meals containing a combination of foods and nutrients, and thus an examination of dietary patterns would more closely reflect the real world by considering the overall eating pattern, and the collinearity of nutrients and foods, which could better express the true eating behavior.6 Dietary patterns reflect a blend of social, cultural, environmental and economic aspects, and they are able to provide more
accessible recommendations.7 Previous studies have investigated the association between dietary patterns and bone mineral density (BMD) among different populations and age groups of both sexes. Dietary patterns characterized by high intakes of refined cereals, processed meats, fried foods (mainly French fries) and sweet foods have been negatively associated with bone.8–13 These patterns consist of food related to increased levels of inflammation (increase in inflammatory markers), which seems to have a negative impact on the bone quality.14 On the other hand, patterns with high amounts of fruit and vegetables have shown a positive relationship with skeletal health.8,12,13,15–18 Although the mechanisms involved in this relationship are still not clearly elucidated, it is suggested that the possible positive effect on bone is related to the alkalizing property and the presence of bioactive compounds (such as antioxidants) in the composition of these foods.19 These results are only pertinent to the risk of developing osteoporosis; however, there is a lack of evidence to confirm whether these results could be extended to osteoporotic individuals, or even if dietary pattern can exert an effect on bone when osteoporosis is already established. The use of a dietary pattern approach to evaluate the association between diet and bone health warrants further studies, considering the wide variability in diets among different countries. Furthermore, the study of this association may help bridge the gap in knowledge of the influence of overall diet on
1 Department of Nutrition, School of Public Health, Sao Paulo University, Food and Nutrition Research Center-NAPAN, Sao Paulo, Brazil and 2Division of Endocrinology, School of Medicine, Federal University of Sao Paulo, Sao Paulo, Brazil. Correspondence: Professor LA Martini, Department of Nutrition, School of Public Health, Sao Paulo University, Food and Nutrition Research Center-NAPAN, Sao Paulo 01246-904, Brazil. E-mail: [email protected] Received 30 September 2014; revised 15 December 2014; accepted 26 January 2015
Dietary patterns in women with osteoporosis NAG de França et al
2 osteoporosis treatment, providing data that may contribute to improvements in osteoporotic patient care. Therefore, the aim of the current study was to determine the dietary patterns of postmenopausal women with osteoporosis using factor analysis, and to investigate the association between these patterns and BMD. SUBJECTS AND METHODS Study design and population From among 363 individuals assessed at a specialized outpatient clinic of the Federal University of Sao Paulo in Sao Paulo city, Brazil, for the treatment of osteoporosis, only postmenopausal women aged over 45 years with at least 2 years of absence of menstrual cycles (menopause) and undergoing osteoporosis follow-up for at least 3 months were selected. Subjects with a history of renal failure, absorption disorders, current use of glucocorticoids, current hyperparathyroidism and incomplete data were excluded. Although it is known that corticosteroids and some illnesses, as hyperparathyroidism, are a major cause of secondary osteoporosis, women who were on steroid therapy in the past or those who already have been treated for the disease were not excluded from the study because the authors believe that the study’s purpose was not to investigate the cause of osteoporosis but their current relationship with eating habits; thus, a final total of 156 free-living women were eligible to take part in the study, and they agreed to participate by signing an informed consent form. All subjects formerly received or were receiving treatment for osteoporosis with an approved medication, and/or calcium and/or vitamin D supplementation. The study protocol was approved by the Federal University of Sao Paulo Research Ethics Committee (0839/08), and enrollment occurred between late autumn and late spring, from 2009 to 2012.
Bone mineral density measurement Measurements for the determination of BMD (g/cm2) of the lumbar spine (L1–L4), total femur (TF), femoral neck (FN) and total body (TB) were conducted by using dual-energy X-ray absorptiometry (Discovery A, QDR for Windows XP, Hologic, Inc., Bedford, MA, USA). The analyses were performed by a highly trained technologist using the same technique for all measurements to avoid disparities in size or position of regions of interest, and the scan was determined on the basis of the height and weight of the participant. BMD was assessed and analyzed according to the recommendations of the Brazilian Society for Clinical Densitometry.20 Osteoporosis was defined as a BMD T-score for the TF and/or FN and/or lumbar spine of ⩽ 2.5 SDs.21 The coefficients of variation (CVs) of lumbar spine BMD and TF BMD was 0.8% and of FN BMD was 1.2%.
Table 1.
Dietary assessment and food grouping Dietary intake was assessed by using a 3-day food diary in which the participants were clearly instructed to record information on nonconsecutive days (2 weekdays and 1 weekend day) on their food and beverage intake using household measures. Participants provided details on brand names of food products, food preparation methods and any recipes used. This method reduces the memory bias, provides detailed information about the current food intake and allows estimates of usual intake.22 Food and beverage quantities obtained from the 3-day food diary were converted into grams and milliliters and computed using the Nutrition Data System for Research software (NDS-R, 2007 version, University of Minnesota, MN, USA). To reduce the complexity of the data, the 270 individual food items identified from the diary (in grams/day) were manually allocated into 13 food groups constructed according to the principles of similarity of nutrient profiles or culinary usage of the foods (Table 1), and they were subsequently used to identify the dietary patterns.
Confounding factor measurements Factors that appear to be related to BMD and dietary patterns were assessed, which included body weight, height, body fat, physical activity, smoking habit, alcohol intake, history of bone fracture, supplement use, postmenopausal time and current treatment received. Body weight was measured to the nearest 0.1 kg by using a manual scale (Filizola, Sao Paulo, Brazil), with the participants wearing light clothing and no footwear, and height was measured to the nearest 0.1 cm with a wall-mounted stadiometer (Tonelli, Criciuma, Brazil), after removal of footwear. Body mass index was calculated by dividing the TB weight (kg) by height squared (m2) and classified according to the World Health Organization classification23 for participants aged under 60 years, and according to the Pan American Health Organization24 for subjects aged 60 years or over. Body fat (g) and lean mass (g) were obtained from the TB dual-energy X-ray absorptiometry scan. Age (in years, continuous), practice of physical activity, considering only activities for recreation, sport, exercise or leisure performed for at least 10 continuous minutes (yes or no), current smoking and alcohol intake for the last 3 months (yes or no for both), bone fracture since the age of 45 years (yes or no), postmenopausal time (in years, continuous), calcium or vitamin D supplementation in the last 3 months (yes or no) and the use of bone-related medication (within the last 3 months) were obtained from a two-page questionnaire.
Statistical analyses The derivation of the dietary patterns was conducted by applying the principal component factor analysis method. Principal component factor
The 13 food groupings used in the dietary pattern analysis
Foods group
Foods in the group
Fruit (fresh or dried) and juices
Apple, papaya, banana, tangerine, lemon, peach, grape, strawberry, lemonade, melon, pineapple, watermelon, mango, pear, orange juice, star fruit, kiwi, acerola juice, avocado, coconut, coconut water, cherimoya, melon juice, passion fruit juice, raisins, plum, apricot, guava, fruit salad, persimmon. Fruit-flavored drinks, cola and noncola soft drinks, light and diet soft drinks. Light cream cheese, skimmed milk, semi-skimmed milk, white cheese, low-fat yogurt, cottage cheese, ricotta cheese, light coffee cream. High-fat yogurt, mozzarella cheese, cream cheese, parmesan cheese, whole milk, goat milk. Whitefish, fresh sardines, canned sardines, canned tuna, salmon, snapper, codfish, mullet, shrimp. Beef, hamburger, liver, jerked beef, pork, pork ribs. Cakes, refined sugar, brown sugar, honey, jelly, chocolate, puddings, condensed milk, cappuccino, cookies, mousse, dulce de leche, sugar cane beverage, ice cream, pies, shredded sweetened coconut, chocolate drinks, jam, fruit desserts, chocolate syrup, panettone, canned fruit, baked candy. Carrot, onion, pumpkin, endive, lettuce, peppers, zucchini, eggplant, okra, chard, chayote, cabbage, tomatoes, spinach, garlic, olives, broccoli, beets, cauliflower, cabbage, fennel, kelp, chicory, celery, pumpkin, parsley, cucumber, watercress, turnip, radish. Potato, cassava, yams, sweet potato. Rice, crackers, pasta, wheat bread, wheat flour, cheese bread, corn bread, potato bread, pancakes, cornmeal, gnocchi, croutons, breakfast cereal, canned corn, couscous. Pizza, chips, hotdog, French fries, pies, popcorn, baked and fried snacks. Coffee, decaffeinated coffee, espresso, herbal tea, yerba mate. Margarine, butter, mayonnaise.
Soft drinks and fruit drinks Low-fat dairy High-fat dairy Fish and sea food Beef and pork meat Sweet foods Vegetables Tubers and tuberous roots Refined cereals Snacks, pizzas and pies Coffee and tea Fats
European Journal of Clinical Nutrition (2015) 1 – 6
© 2015 Macmillan Publishers Limited
Dietary patterns in women with osteoporosis NAG de França et al
3 analysis method is a multivariate technique that evaluates the intercorrelations between the initial food variables and reduces food groups into a smaller number of factors (patterns) that can explain variations in the dietary intake. The number of factors retained was based on a combination of food group components with an eigenvalue 41.0, the percentage of variance criteria (sufficient factors to meet at least 60% of variance explained)25 and examination of the breakpoint in the scree plot, resulting in five factors retained for further analyses. The factors were rotated by orthogonal transformation (Varimax option) to achieve a simpler structure with a greater interpretability. Food groups with a factor loading ⩾ 0.45 on a component were considered informative for interpretation of the dietary patterns.25 The factor scores for each pattern and individual were determined by summing the intakes of each food group weighted by the factor loading. The groups that had a negative factor loading were also retained in order to determine the complexity of eating habits. Normality of continuous variables was assessed by the Kolmogorov– Smirnov test. To investigate the predictive effect of dietary patterns on BMD (dependent), Pearson’s correlation was conducted followed by multiple linear regression analysis. The linear regression models included the BMD at each evaluated site (lumbar spine, FN, TF and TB) as a dependent variable and the dietary patterns (in scores) as independent variables. The model was adjusted for energy intake (kcal), calcium intake (mg), lean mass (g), height (cm) and postmenopausal time (years). The principal component factor analysis was performed using SAS Enterprise Guide, version 4.1 (SAS Institute, Inc., Cary, NC, USA), whereas the correlation and regression analyses were performed using the SPSS software, version 20.0 (SPSS, Chicago, IL, USA. The level of significance was set at Po0.05.
RESULTS General characteristics of the participants are presented in Table 2. The participants had a mean age of 68.4 (9.0) years and body mass index value of 25.9 (3.9) kg/m2 (according to body mass index classification, 19% of the sample was underweight, 50% were normal weight and 31% were overweight—table not shown), with body fat altering only slightly. The average intake of calcium was 832.5 (442.7) mg, a very close value to the acceptable by the estimated average requirements for women aged over 50 years (1000 mg).26 Vitamin D ingested by diet was 4.4 (3.0) μg, bellow the requirement of 10 μg.26 Most of the participants had less than 5 years of education (low educational level; 79%), did not exercise (54%), did not smoke (91%) and were taking vitamin D or calcium supplementation (85 and 53.2%, respectively). The main treatment for osteoporosis was through the use of bisphosphonates (75%). The factor-loading matrices for the five patterns retained are shown in Table 3. The high positive loadings indicate strong associations between food groups and patterns, whereas high negative loadings indicate a strong negative relation. Patterns were labeled according to the food groups with positive loadings. Dietary pattern 1 showed heavy loadings on vegetables (average intake of the entire sample: 87.4 ± 94.7 g/day; ranging from 0–725 g/day), fruit (fresh and dried) and fresh juices (251.6 ± 221.2 g/day; 0–1,166 g/day) and tubers and tuberous roots (25.6 ± 54.7 g/day; 0–366.7 g/day) and was labeled ‘Healthy’. Dietary pattern 2, with high loadings for refined cereals and baked products with refined cereals (230.9 ± 129.9 g; 0–786 g/day) and beef and pork meat (59.2 ± 70.3 g/day; 0–450 g/day), was labeled ‘Red meat and refined cereals’. Dietary pattern 3 had heavy loadings on low-fat dairy products and was labeled ‘Low-fat dairy’ (125.9 ± 201.2 g/day; 0–1,019.5 g/day). Dietary pattern 4, with high loadings for sugar, sugary products (42.8 ± 74.9 g/day; 0–502.5 g/ day) and coffee and tea (138.5 ± 149.3 g/day; 0–1,008.9 g/day), was labeled ‘Sweet foods, coffee and tea’. Dietary pattern 5 showed heavy loadings on fats (9.0 ± 19.2 g; 0–207.1 g/day), snacks, pizzas, pies (30.0 ± 97.1 g/day; 0–823.3 g/day) and soft drinks and fruit drinks (59.3 ± 136.6 g/day; 0–1,164.9 g/day), and it was labeled ‘Western’. Overall, the five dietary patterns accounted for 60.4% of the variance in food intakes. © 2015 Macmillan Publishers Limited
Table 2. General characteristics of the osteoporotic postmenopausal women studied (n = 156) Postmenopausal women with osteoporosis Age (years) Weight (kg) Body mass index (kg/m2) Body fat (%) Postmenopausal time (years) Energy intake (kcal) Carbohydrate intake (% of total calories) Fat intake (% of total calories) Protein intake (% of total calories) Calcium intake (mg) Vitamin D intake (μg) Lumbar spine BMD (g/cm2) Lumbar spine BMD (T-score) Femoral neck BMD (g/cm2) Femoral neck BMD (T-score) Total femur BMD (g/cm2) Total femur BMD (T-score) Total BMD (g/cm2) Total BMD (T-score) Fracture (yes)a (%) Educational level ⩽ 5 years (%) 5–8 years (%) 48 years (%) Physical activity (yes) (%) Smoking (yes) (%) Alcohol intake (yes) (%) Vitamin D supplementation (yes) (%) Calcium supplementation (yes) (%) Bisphosphonates (%)
68.4 (9.0) 60.4 (10.1) 25.9 (3.9) 36.0 (5.5) 22.2 (10.1) 1627 (580) 52.2 (9.1) 30.5 (7.1) 18.4 (5.1) 832.5 (442.7) 4.4 (3.0) 0.745 (0.118) − 2.8 (1.1) 0.648 (0.092) − 1.8 (0.8) 0.742 (0.109) − 1.6 (0.9) 0.945 (0.115) − 1.8 (1.3) 42.9 78.8 15.4 5.8 46.2 9.0 0.0 85.3 53.2 75.0
Abbreviation: BMD, bone mass density. Continuous variables are expressed as mean (standard deviation) and categorical variables as percentages. a After age 45.
Table 4 shows the results for the linear regression conducted between the dietary patterns and BMD at each evaluated site. ‘Sweet foods, coffee and tea’ was the only pattern that showed a significant correlation with BMD. As it was inversely associated with TF BMD (β = − 0.206; 95% CI: − 0.042– − 0.004) and with TB BMD (β = − 0.321; 95% CI: − 0.059– − 0.017), this dietary pattern was considered as a negative independent predictor of BMD. DISCUSSION Although there are numerous studies exploring the relationships between dietary patterns and BMD, the present study is unique in that it conducted such an evaluation among postmenopausal women who already presented with osteoporosis. The main finding showed that the ‘sweet foods, coffee and tea’ pattern was inversely correlated with BMD (in femur and TB) for the overall sample. Comparisons of our results with previous findings are hampered by differences in protocols, food habits of the studied populations, food groups formed and in the dietary patterns identified.9,12,13,15,27 Furthermore, previous studies have evaluated the risk factors for the incidence of osteoporosis, in contrast with the present study, which involved osteoporotic women. However, some similarities can be observed with the dietary patterns derived in other studies and their relationships with BMD. The inverse association between BMD and dietary patterns with high loadings for sweet foods was previously shown by Tucker et al.13 in elderly adults from the Framingham Osteoporosis Study. The mean BMD in the ‘candy group’ was significantly lower for the Ward’s area, FN and radius (P o 0.001) among men, and for radius European Journal of Clinical Nutrition (2015), 1 – 6
Dietary patterns in women with osteoporosis NAG de França et al
4 Table 3.
Factor-loading matrix for the five dietary patterns identified among the 156 osteoporotic postmenopausal women
Vegetables Fruit and fresh juices Tubers and tuberous roots Fats Fish and sea food High-fat dairy Low-fat dairy Sweet foods Coffee and tea Snacks, pizzas and pies Refined cereals Beef and pork meat Soft drinks and fruit drinks Percentage of variance (%)
‘Healthy’
‘Red meat and refined cereals’
‘Low-fat dairy’
‘Sweet foods, coffee and tea’
‘Western’
0.768 0.608 0.568 — — — — — — — — — — 19.2
— — — — − 0.472 — — — — — 0.726 0.666 — 13.9
— — — — — − 0.762 0.829 — — — — — — 10.8
— — — — — — — 0.736 0.816 — — — — 8.7
— — — 0.547 — — — — — 0.746 — — 0.554 7.8
Table 4.
Results of adjusted linear regression analysis (β-coefficient), and 95% confidence interval of the dietary patterns (score values) and body mineral density (g/cm2) Dietary patterns ‘Healthy’ ‘Red meat and refined cereals’ ‘Low-fat dairy’ ‘Sweet foods, coffee and tea’ ‘Western’
Lumbar spine BMD β (95% CI) − 0.002 − 0.094 0.141 − 0.066 0.069
(−0.022 (−0.031 (−0.003 (−0.030 (−0.012
to to to to to
0.021) 0.010) 0.035) 0.014) 0.028)
Femoral neck BMD β (95% CI) 0.164 − 0.005 − 0.069 − 0.155 0.099
(−0.002 (−0.016 (−0.021 (−0.031 (−0.007
to to to to to
0.032) 0.015) 0.009) 0.002) 0.025)
Total femur BMD β (95% CI) 0.184 0.038 − 0.067 − 0.206* 0.052
Total body BMD β (95% CI)
(0.001 to 0.038) (−0.014 to 0.022) (−0.024 to 0.010) (−0.042 to − 0.004) (−0.012 to 0.023)
0.087 − 0.019 0.113 − 0.321* 0.039
(−0.011 (−0.023 (−0.006 (−0.059 (−0.016
to to to to to
0.032) 0.018) 0.032) −0.017) 0.025)
Abbreviations: BMD, bone mineral density; CI, confidence interval. *Po0.05. Adjusted for energy intake (kcal), calcium intake (mg), lean mass (g), height (cm) and postmenopausal time (years).
among women (P = 0.004). Similar results were obtained by Hardcastle et al.12 and Park et al.,27 who also found negative correlations between patterns rich in sweet foods and bone health. Both of the studies indicated a negative predictive impact of sugary food intake on the risk of developing osteoporosis. A possible explanation for the negative effect of glucose intake could be related to increased glycemia, which appears to be associated with changes in bone mineral homeostasis.28 A highsucrose diet produces hyperinsulinemia, which in turn could induce hypercalciuria by inhibiting renal tubular resorption of calcium, affecting the quality of bone mineralization.29 There has been little discussion concerning the effects of hyperglycemia on skeletal integrity. Chronically elevated glucose even for short periods during the lifetime could be deleterious for the skeleton, particularly during growth and peak bone mass formation.30 Once an adequate bone formation is essential to attenuate the deleterious effects associated with aging, any change in peak bone mass could lead to the development of osteoporosis and/or to an increased risk of fractures in adulthood.31,32 However, it is emphasized that the present study did not investigate lifetime dietary habits of the women studied, which does not allow making any inference about the impact of glycemic peaks on late bone demineralization. According to the World Health Organization (2002),33 sugar consumption should not exceed 10% of total energy per day; however, the new draft guideline proposes a reduction to below 5%,34 as the consumption of free sugars may result in reduced intake of foods containing more nutritionally adequate calories, leading to an unhealthy diet and increased risk of noncommunicable diseases.33 Among our sample, the average intake of sugar and sugary foods group was 42.8 g/day (74.9 g/day), accounting for 28.6 g of added sugar, which implies 7% of the total average energy intake. This value is in agreement with the current recommendation, but it is above the new draft. Furthermore, the added sugar reached 165 g, which is an excessive intake and European Journal of Clinical Nutrition (2015) 1 – 6
could lead to the negative effects on bone. The results from our sample were similar to the findings reported by Bueno et al.35 showing an average intake of 8.5% (33.4 g) in a sample of 622 elderly individuals, with significant differences between the sexes (7.8% for men vs 8.7% for women; P o 0.05). Taken together, these results hypothesized that consumption of added sugar is a common food habit among elderly individuals, especially women. The ‘sweet foods, coffee and tea’ pattern also contains caffeinated beverages. Results of investigations into the negative impact of caffeine on bone are conflicting. Current evidence suggests that its detrimental effect on BMD might only occur in susceptible individuals (that is, inadequate calcium intake, osteoporosis predisposition, old age), where it has been suggested that caffeine consumption could decrease calcium absorption and/or increase excretion of calcium in this group.36 Rapuri et al. 37 showed that intakes of more than 300 mg of caffeine (≈514 g of brewed coffee) accelerate bone loss at the spine in elderly postmenopausal women, with greater risk among those with the tt genetic variant of the vitamin D receptor. Although the average intake of caffeine among the sample as a whole was 48.2 mg, daily caffeine intake ranged from 0 to 332 mg. It is also suggested that caffeine could have a deleterious effect on osteoblasts by inducing apoptosis via a mitochondria-dependent pathway.38 These apoptotic biochemical changes could be a result of an intracellular oxidative stress stimulated by caffeine, leading to a loss of BMD.38 Furthermore, there are several observations indicating an adverse effect of caffeine on glucose metabolism.39–41 Petrie et al.40 showed a significantly attenuation of whole-body insulin sensitivity in obese, nondiabetic persons, after acute caffeine ingestion, with no evidence over beta cell secretion. The inconsistent results about caffeine indicate that also other coffee constituents may have unfavorable skeletal effects. Recently, in a study with female rats, Folwarczna et al.42 observed a negative impact on bone associated to the presence of © 2015 Macmillan Publishers Limited
Dietary patterns in women with osteoporosis NAG de França et al
trigonelline (a bitter alkaloid related to the production of aroma compounds), according to the estrogen status. The trigonelline administration worsened the mineralization of the vertebra and the strength of the tibial metaphysis, and decreased the width of femoral trabeculae in ovariectomized (estrogen-deficient) Wistar rats, suggesting that the possible detrimental effect of this alkaloid may be stronger in postmenopausal women.42 Thereby, the consumption of caffeinated beverage combined with sugar could exacerbate the detrimental effect on bone through a synergic relationship. BMD is an important predictor of fractures.43 According to a classic meta-analysis conducted by Marshall et al, the risk of hip and total fractures increases 2.6-fold and 1.6-fold, respectively, for each standard deviation decrease in BMD.44 Hip, which includes the proximal femur, accounts for the main site for fractures, causing acute pain, loss of function and usually leading to hospitalization with a complicate recovery.45 It is estimated that the strength of bone decreases about 2–12% per decade in postmenopausal women, with an average age for hip fractures about 80 years.45 Thus, as the ‘sweet foods, coffee and tea’ pattern showed an inverse relationship with TF BMD (β: -0.206; 95% CI: − 0.042 – − 0.004), it is suggested that this dietary habit could increase the risk for latest fracture in hip. Data from the National Health and Nutrition Examination Survey (NHANES 2005–2008)46 determine the TF BMD cutoff for women between 60 and 69 years as 0.855 ± 0.126 g/cm2, classifying our sample between the 15th and 25th percentile (0.742 ± 0.109 g/cm2). Thereby, a decrease of -0.206 g/cm2 associated with the ‘Sweet foods, coffee and tea’ pattern would imply a more prominent demineralization in this site, significantly accelerating the risk of late fractures. Most dietary pattern studies have shown a positive association between a diet rich in fruit and/or vegetables and bone health.8,12,13,16–18 This same association was not observed in our sample, may be because the influence of these foods is too low to detect in a small sample. However, the authors consider that the intake of fruit and vegetables is supported by the beneficial effect of their antioxidant content47 and alkalizing property.48 Furthermore, no correlation was detected between the ‘low-fat dairy’ pattern and BMD in our study. This nonsignificant association may be because the sample comprised osteoporotic women who had already received some orientation to consume higher amounts of dairy, leading to less remarkable results. Average intake from the dairy group was 5 portions/day (data not shown), considering 120 kcal/dairy portion,49 representing a level of consumption above the recommendation of 3 dairy portions/day proposed by the Food Guide for the Brazilian Population49 and an adequate calcium intake (832.5 mg). It should also be emphasized that, although the participants had received or were receiving treatment for osteoporosis, they were not instructed to make changes in their overall diet. This suggests that the specific orientations to increase calcium intake, mainly by increasing dairy consumption, may be insufficient to ensure a protective effect on bone. Besides sample size, the present study has several other limitations. First, this study used a cross-sectional approach, precluding the inference of a causal relationship. Second, as the participants were recruited from a specialized outpatient clinic in Sao Paulo city, our sample was not representative of postmenopausal women with osteoporosis in general. Finally, the minimal nutritional orientation received may have exerted some influence on the results. The dietary pattern approach a posteriori is based on current data, enabling examination of the multidimensionality of the total diet50 and may have major implications for public health, as it is more readily interpreted and translated to the daily diet of the population. In addition, the approach provides underlying knowledge of human feeding practices,6 elucidated in the present study in the form of an inverse association between the ‘Sweet foods, coffee and tea’ pattern and BMD (in femur and TB). © 2015 Macmillan Publishers Limited
CONCLUSIONS A concurrent excessive consumption of sweet foods and caffeinated beverages showed a negative effect on BMD even when the skeleton already presented demineralization. Food and beverage intake is a modifiable factor that should not be neglected in osteoporosis treatment. The findings of the present study suggest that dietary orientations for osteoporotic individuals should not be limited to calcium and vitamin D intake, but extend to encompass the overall diet. CONFLICT OF INTEREST The authors declare no conflict of interest.
ACKNOWLEDGEMENTS We thank Wesley Rodrigues dos Santos and Anatoli Yambartsev, PhD, from the Institute of Mathematics and Statistics of Sao Paulo University, for the dietary pattern analysis, and the funding organization São Paulo Research Foundation (Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)).
REFERENCES 1 Johnell O, Kanis JA. An estimate of the worldwide prevalence and disability associated with osteoporotic fractures. Osteoporos Int 2006; 17: 1726–1733. 2 Kanis JA. Assessment of Osteoporosis at the Primary Health-Care Level Technical Report University of Sheffield. World Health Organization: UK, 2007. 3 Pinheiro MM, Ciconelli RM, Jacques Nde O, Genaro PS, Martini LA, Ferraz MB. The burden of osteoporosis in Brazil: regional data from fractures in adult men and women--the Brazilian Osteoporosis Study (BRAZOS). Rev Bras Reumatol 2010; 50: 113–127. 4 Bortolon PC, Andrade CL, Andrade CA. Characteristics of hospital admissions in the Unified National Health System for osteoporotic hip fracture in elderly people in Brazil, 2006-2008. Cad Saude Publica 2011; 27: 733–742. 5 Bonjour JP, Gueguen L, Palacios C, Shearer MJ, Weaver CM. Minerals and vitamins in bone health: the potential value of dietary enhancement. Br J Nutr 2009; 101: 1581–1596. 6 Hu FB. Dietary pattern analysis: a new direction in nutritional epidemiology. Curr Opin Lipidol 2002; 13: 3–9. 7 National Research Council—Committee on Diet and Health. Diet and Health: Implications for Reducing Chronic Disease Risk. National Academy Press: Washington, DC, 1989. 8 McNaughton SA, Wattanapenpaiboon N, Wark JD, Nowson CA. An energy-dense, nutrient-poor dietary pattern is inversely associated with bone health in women. J Nutr 2011; 141: 1516–1523. 9 Whittle CR, Woodside JV, Cardwell CR, McCourt HJ, Young IS, Murray LJ et al. Dietary patterns and bone mineral status in young adults: the Northern Ireland Young Hearts Project. Br J Nutr 2012; 108: 1494–1504. 10 Fairweather-Tait SJ, Skinner J, Guile GR, Cassidy A, Spector TD, MacGregor AJ. Diet and bone mineral density study in postmenopausal women from the TwinsUK registry shows a negative association with a traditional English dietary pattern and a positive association with wine. Am J Clin Nutr 2011; 94: 1371–1375. 11 Karamati M, Jessri M, Shariati-Bafghi SE, Rashidkhani B. Dietary patterns in relation to bone mineral density among menopausal Iranian women. Calcif Tissue Int 2012; 91: 40–49. 12 Hardcastle AC, Aucott L, Fraser WD, Reid DM, Macdonald HM. Dietary patterns, bone resorption and bone mineral density in early post-menopausal Scottish women. Eur J Clin Nutr 2011; 65: 378–385. 13 Tucker KL, Chen H, Hannan MT, Cupples LA, Wilson PW, Felson D et al. Bone mineral density and dietary patterns in older adults: the Framingham Osteoporosis Study. Am J Clin Nutr 2002; 76: 245–252. 14 Cauley JA, Danielson ME, Boudreau RM, Forrest KY, Zmuda JM, Pahor M et al. Inflammatory markers and incident fracture risk in older men and women: the Health Aging and Body Composition Study. J Bone Miner Res 2007; 22: 1088–1095. 15 Shin S, Joung H. A dairy and fruit dietary pattern is associated with a reduced likelihood of osteoporosis in Korean postmenopausal women. Br J Nutr 2013; 110: 1926–1933. 16 Langsetmo L, Hanley DA, Prior JC, Barr SI, Anastassiades T, Towheed T et al. Dietary patterns and incident low-trauma fractures in postmenopausal women and men aged 4/ = 50 y: a population-based cohort study. Am J Clin Nutr 2011; 93: 192–199. 17 Okubo H, Sasaki S, Horiguchi H, Oguma E, Miyamoto K, Hosoi Y et al. Dietary patterns associated with bone mineral density in premenopausal Japanese farmwomen. Am J Clin Nutr 2006; 83: 1185–1192.
European Journal of Clinical Nutrition (2015), 1 – 6
5
Dietary patterns in women with osteoporosis NAG de França et al
6 18 Wosje KS, Khoury PR, Claytor RP, Copeland KA, Hornung RW, Daniels SR et al. Dietary patterns associated with fat and bone mass in young children. Am J Clin Nutr 2010; 92: 294–303. 19 Macdonald HM, Hardcastle AC. Dietary patterns and bone health. In: Burckhardt P, Dawson-Hughes B, Weaver CM. Nutritional Influences on Bone Health. Springer, pp 135–143, 2010. 20 Brandao CM, Camargos BM, Zerbini CA, Plapler PG, Mendonca LM, Albergaria BH et al. 2008 official positions of the Brazilian Society for Clinical Densitometry-SBDens. Arq Bras Endocrinol Metabol 2009; 53: 107–112. 21 ISCD—The International Society For Clinical Densitometry The Official Positions 2013. 22 Fisberg RM, Slater B, Marchioni DML, Martini LA. Inquéritos Alimentares: Métodos e Bases Científicos. Manole: Barueri, 2005. 23 World Health Oraganization: Physical status: the use and interpretation of anthropometry. Report of a WHO Expert Committee. WHO: Geneva, 1995. 24 Organización Panamericana de la Salud. División de Promoción y Protección de la Salud (HPP) Encuesta multicentrica salud bienestar y envejecimiento (SABE) en América Latina: informe preliminar Multicenter survey aging, health and wellbeing in Latin América and the Caribbean (SABE): preliminary report. Reunión del Comité Asesor de Investigaciónes en Salud: Kingston, Washington, DC. p 19, 2001. 25 Hair Jr J, Black W, Babin B, Anderson R, Tatham R. Multivariate Data Analysis, 6th ed. Prentice Hall: New York, NY, 2006. 26 Food and Nutrition Board, Institute of Medicine. Phosphorus Dietary Reference Intakes: Calcium, Phosphorus, Magnesium, Vitamin D, and Fluoride. National Academy Press: Washington, DC. pp 146–189, 1997. 27 Park SJ, Joo SE, Min H, Park JK, Kim Y, Kim SS et al. Dietary patterns and osteoporosis risk in postmenopausal korean women. Osong Public Health Res Perspect 2012; 3: 199–205. 28 Bass EF, Baile CA, Lewis RD, Giraudo SQ. Bone quality and strength are greater in growing male rats fed fructose compared with glucose. Nutr Res 2013; 33: 1063–1071. 29 Holl MG, Allen LH. Sucrose ingestion, insulin response and mineral metabolism in humans. J Nutr 1987; 117: 1229–1233. 30 Rosen CJ. Sugar and bone: a not-so sweet story. J Bone Miner Res 2008; 23: 1881–1883. 31 Baxter-Jones AD, Faulkner RA, Forwood MR, Mirwald RL, Bailey DA. Bone mineral accrual from 8 to 30 years of age: an estimation of peak bone mass. J Bone Miner Res 2011; 26: 1729–1739. 32 Boot AM, de Ridder MA, van der Sluis IM, van Slobbe I, Krenning EP, KeizerSchrama SM. Peak bone mineral density, lean body mass and fractures. Bone 2010; 46: 336–341. 33 World Health Organization. The World Health Report: Reducing Risks, Promoting Healthy Life. WHO: Geneva, 2002. 34 World Health Organization WHO opens public consultation on draft sugars guideline (acessed on May 2nd, 2014); Available from: http://www.who.int/med iacentre/news/notes/2014/consultation-sugar-guideline/en/ 2014.
European Journal of Clinical Nutrition (2015) 1 – 6
35 Bueno MB, Marchioni DM, Cesar CL, Fisberg RM. Added sugars: consumption and associated factors among adults and the elderly. Sao Paulo, Brazil. Rev Bras Epidemiol 2012; 15: 256–264. 36 Dew TP, Day AJ, Morgan MR. Bone mineral density, polyphenols and caffeine: a reassessment. Nutr Res Rev 2007; 20: 89–105. 37 Rapuri PB, Gallagher JC, Kinyamu HK, Ryschon KL. Caffeine intake increases the rate of bone loss in elderly women and interacts with vitamin D receptor genotypes. Am J Clin Nutr 2001; 74: 694–700. 38 Lu PZ, Lai CY, Chan WH. Caffeine induces cell death via activation of apoptotic signal and inactivation of survival signal in human osteoblasts. Int J Mol Sci 2008; 9: 698–718. 39 Lane JD, Feinglos MN, Surwit RS. Caffeine increases ambulatory glucose and postprandial responses in coffee drinkers with type 2 diabetes. Diabetes Care 2008; 31: 221–222. 40 Petrie HJ, Chown SE, Belfie LM, Duncan AM, McLaren DH, Conquer JA et al. Caffeine ingestion increases the insulin response to an oral-glucose-tolerance test in obese men before and after weight loss. Am J Clin Nutr 2004; 80: 22–28. 41 Keijzers GB, De Galan BE, Tack CJ, Smits P. Caffeine can decrease insulin sensitivity in humans. Diabetes Care 2002; 25: 364–369. 42 Folwarczna J, Zych M, Nowinska B, Pytlik M, Janas A. Unfavorable effect of trigonelline, an alkaloid present in coffee and fenugreek, on bone mechanical properties in estrogen-deficient rats. Mol Nutr Food Res 2014; 58: 1457–1464. 43 Cummings SR, Black DM, Nevitt MC, Browner W, Cauley J, Ensrud K et al. Bone density at various sites for prediction of hip fractures. The Study of Osteoporotic Fractures Research Group. Lancet 1993; 341: 72–75. 44 Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ 1996; 312: 1254–1259. 45 Kanis JA. On behalf of the World Health Organization Scientific Group. Assessment of Osteoporosis at the Primary Health-care Level Technical Report World Health Organization Collaborating Centre for Metabolic Bone Diseases. University of Sheffield: UK, 2007. 46 Looker AC, Borrud LG, Hughes JP, Fan B, Shepherd JA, Melton 3rd LJ . Lumbar spine and proximal femur bone mineral density, bone mineral content, and bone area: United States, 2005–2008. Vital Health Stat 11 2012; (251): 1–132. 47 Nieves JW. Skeletal effects of nutrients and nutraceuticals, beyond calcium and vitamin D. Osteoporos Int 2013; 24: 771–786. 48 Lanham-New SA. The balance of bone health: tipping the scales in favor of potassium-rich, bicarbonate-rich foods. J Nutr 2008; 138: 172S–177S. 49 Ministério da Saúde. Secretaria de Atenção à Saúde. Coordenação-Geral da Política de Alimentação e Nutrição Guia Alimentar Para a População Brasileira: Promovendo a Alimentação Saudável Série A. Normas e Manuais Técnicos: Brasilia, 2006. 50 Newby PK, Tucker KL. Empirically derived eating patterns using factor or cluster analysis: a review. Nutr Rev 2004; 62: 177–203.
© 2015 Macmillan Publishers Limited
