- Email: [email protected]
Diet score is associated with plasma homocysteine in a healthy institutionalised elderly population
384
Nutr Metab Cardiovasc Dis (2003) 13:384--390
ORIGINAL
ARTICLE
Diet score is associated with plasma homocysteine in a healthy institutionalised elderly population C. Lasheras 1, J.M. Huerta 1, S. Gonz~ilez I, M. Prada 2, S. Braga 2, S. Fern~indez 1, and A.M. Patterson I 1Departamento de Biologia Funcional, Area de Fisiolog/a,Facultad de Medicina, Universidad de Oviedo, and 2Laboratorio de An~ilisisCl/nicos,Hospital Central de Asturias, Oviedo, Spain
Abstract Background and Aim: Total plasma homocysteine (tHcy) is an independent risk factor for cardiovascular diseases.Among other dietary and non-dietary factors, B vitamins, such as folate, riboflavin and cobalamin, are primary determinants of tHcy in the general population. However, research has concentrated on the relationship of these nutrients with tHcy, and little is known about overall eating patterns and tHcy. Methods and Results: In this study, we analysed whether a diet score based on the consumption of folate-, riboflavinand cobalamin-rich food groups was associated with tHcy in a sample of 140 institutionalised elderly subjects (59 men and 81 women aged 60-80 years) from Northern Spain. The food groups identified as the major contributors to the intake of the three vitamins were vegetables, fruit, fish, meat and milk and dairy products. The mean tHcy level was 13.3+5.1 pmol/L (range: 3.9-30.7 pmol/L). None of the food groups predicted tHcy levels individually, but the overall diet score was inversely associated with tHcy in a multiple linear regression analysis. High tHcy levels (>16 pmol/L) were almost twice as prevalent in the groups scoring less than 7 than in those scoring 7 or more (37.5 vs 19.6%, p=0.021). Conclusions: These data suggest that a dietary pattern characterised by high intakes of B vitamin-rich foods is associat-
Key words: Homocysteine, hyperhomocysteinemia,vitamins,elderly, dietary patterns. Correspondence to: Cristina Lasheras PhD, Departamento de Biologfa Funcional, Area de Fisiologfa,Facultad de Medicina, Universidad de Oviedo, JuMn Claveria s/n, 33006, Oviedo, Spain. E-mail: [email protected] Received: 5 June 2003;accepted: 1 August 2003.
ed with a lower tHcy concentration and a reduced percentage of high tHcy levels in elderly subjects. They also support the use of dietary pattern approaches to evaluate the relationships between diet and health outcomes that go beyond single nutrient analyses. Nutr Metab Cardiovasc Dis (2003) 13:384-390 e2003, Medikal Press
Introduction The level of total plasma homocysteine (tHcy) has attracted much interest over recent years because a high tHcy concentration has been associated with an increased risk of cardiovascular disease (1, 2) and all-cause mortality (3). Vitamin status is one of the major determinants of total plasma homocysteine levels in the general population (4, 5), in which it has been estimated as accounting for two-thirds of the cases of high tHcy concentrations (6, 7). B vitamins such as folate, B 6 and riboflavin are important cofactors for a number of the enzymes involved in homocysteine metabolism (5, 8), and cobalamin (B12) is the final methyl-group donor in the homocysteine to methionine remethylation pathway (8). As a result, a large number of studies have been carried out over the last ten years with the aim of elucidating the relationship between these vitamins and vascular pathologies (9), as well as the potential benefits of supplementation programmes on population tHcy levels (10, 11). Vitamins are, by definition, essential for human life, and most of them can only be provided through diet, As freeliving people do not consume single nutrients but combinations of several foods, a second-stage analysis of dietrelated diseases has to move from nutrient to food intake
385
Diet and homocysteine in the elderly
or dietary patterns in order to understand the nature of such diseases and provide the general population with recommendations for promoting healthy dietary habits. Although there are a number of studies of the importance of B vitamins in tHcy metabolism, very few have analysed the connection between food intake and tHcy levels (5, 12, 13) despite the fact that diet is the most important source of most of these vitamins, specially in countries (such as Spain) in which folate fortification policies have not been implemented. Two previous studies have suggested that dietary patterns may be related to plasma homocysteine levels (12, 13). In this study, we developed a diet score based on the intake of the food groups that make the greatest contribution to folate, cobalamin and riboflavin intake in a sample of elderly people aged 60-80 years living in Institutions in Asturias (Spain). Our hypothesis was that the higher the score, the lower the tHcy concentration should be. This cross-sectional report forms part of an ongoing prospective study of diet and disease.
Me~o~ Study subjects The analysed sample consisted of 140 healthy institutionalised elderly subjects (59 men and 81 women) recruited in seven nursing homes located in Asturias (Northern Spain). Subjects with a history of cancer or cardiovascular disease were not included as they will be end-points in our prospective investigation. Their mean age at the time of enrolment was 73.3_+5.6 years for men and 74.2_+4.8 years for women; all of the subjects had been living at the institutions for at least three years (mean 5.2 years). The medical history of each subject was taken before enrolment. Nine patients taking antiepileptic drugs or thyroid hormones were excluded from the analysis because these drugs can interfere with plasma tHcy levels (14, 15). Although it has been reported that antihypertensive medication can affect tHcy (5), the subjects taking antihypertensive drugs were not excluded as we found no differences in tHcy levels between those who were on this medication and those who were not (data not shown). Two subjects taking vitamin supplements on a regular basis were not included in the analysis. All of the participants were mentally and physically capable of participating in the study, and all of them gave their informed consent. The study design was approved by
the Committee on Ethical Research of the Oviedo University Hospital. After the subjects had given their informed consent, a personal appointment was made for collecting data.
Dietary intake Dietary intake was assessed by means of a food frequency questionnaire (FFQ) specifically designed for each of the seven Institutions after they had kindly provided us with the menus of the previous year. Trained dietitians asked questions about cooking practices, the numbers and amounts of the ingredients used in each recipe, and menu preparation (the type of oil, type of milk, etc.). The FFQ contained all of the individual foods (not food groups) in the menus of each Institution. During an interview, the subjects were asked item-by-item whether they usually ate each food and, if so, how much they ate. For this purpose, pictures of three serving sizes of each cooked food were presented to the participants so that they could choose from up to seven serving portions (from "less than the small one" to "more than the large one"). For some of the foods consumed, the amounts were recorded in household units, by volume or by length measured with a ruler. All of the subjects were asked whether they had access to foods other than those offered by the Institution (through their relatives, for example). Food intake was analysed for energy and macro- and micronutrient content using the nutrient Food Composition Tables developed by the CSIC (16).
Blood measurements Blood samples were drawn by venipuncture after a 12hour fast, and the plasma and serum were collected in separate tubes. The samples were kept on ice and centrifuged within 2-4 hours of collection. Plasma aliquots were kept at -70°C until analysis. Total plasma homocysteine was determined using reversephase HPLC with fluorimetric detection (17) and a Biorad analyser. In order to ensure the stability of the analytical procedure, cysteine and cysteineglycine (two amino thiols) were measured together with homocysteine in the same sample and we checked that their concentrations (absolute and relative to homocysteine) were within the normal range. Serum creatinine was measured in a Hitachi 737 analyser (Boehringer Mannheim Corp, Indianapolis, IN).
Food groupings The foods were grouped on the basis of their characteristics. Vegetables included all types of vegetables except fruits,
386
C. Lasheras, et al
dried fruits (eg raisins, dried figs), nuts and seeds (hazelnuts, almonds, walnuts), which made up the fruit group. The cereals were bread, flour, rice and pasta. Potatoes included boiled potatoes and French fries. Fish and seafood were grouped as fish, and pork, poultry and red meat made up the meat group. The vegetable oils were olive and sunflower oils. Milk and dairy products included all types of milk, yogurt and cheese. The "other" group included bakery products, margarine, sweets, cakes, coffee, beverages and mixed foods (such as pizza, paella, etc.).
TABLE 1
Characteristics and selected total plasma homocysteine (tHcy) determinants in the study sample. Mean value ± SD or percentage Plasma tHcy (iJmol/L) a
13.3±5.1
Serum creatinine (iJmol/L) a
92.1 +19.5
Age (yr)
73.8,+5.2
Female sex (%)
57.9 [81] b
Body mass index (kg/m 2)
28.3±4.9
Mean diet score c
7.5±2.8
Energy (kJ/day)
8.1 ± 1.7
Folate intake (IJg/day)
Diet score The diet score was based on the quartiles of the intakes (grams per day) of the food groups previously identified as being the main contributors to the intake of folate, B n and riboflavin (see Results): ie, meat, fish, milk and dairy products, fruit, and vegetables. Each variable was categorised into quartiles and given a score from 0 to 3 (0 = an intake in the lowest quartile and 3 = an intake in the highest quartile). The total score (range 0-15) was the sum of all of the partial scores.
187.3+81.1
Riboflavin intake (mg/day)
1.8_+0.5
B12 intake (IJg/day)
4.9+1.8
Protein intake (g/day)
76.2±15.2
Coffee intake d (mL/day)
30.1 ±19.5
a Adjusted for age and gender, b Percentage (n in brackets), c Based on intake of vegetables, fruit, meat, fish, milk and dairy products (see text for details). d Only consumers.
Statistical analyses The descriptive data in the tables are expressed as mean values + SD or percentages. The intakes of the food
TABLE 2
Contribution of different food groups to the intake of macronutrients and selected vitamins in the study population. g/day (mean ± SD) a
Energy (%)
Protein (%)
Carbohydrate (%)
Lipid (%)
Folate (%)
Riboflavin (%)
192.5±105.8
3.15
6.57
4.72
1.70
49.55
33.15
0
17.5,+7.3
2.72
4.51
3.7
0.85
5.77
1.73
0
Fruit
275.8+192.2
6.23
1.86
12.87
0.t0
19.94
3.80
0
Cereals
210.0,+89.4
20.52
15.45
36.11
4.44
1.01
4.39
0
Potatoes
41.5+32.1
1.76
1.31
3.15
0.31
2.94
1.21
0
Fish
68.6+34.1
3.60
15.11
0.17
2.95
3.76
4.99
44.41
Meat
108.4+37.8
12.95
26.71
0.53
22.90
4.74
10.96
29.74
Eggs
9.3±5.9
0.72
1.53
0
1.30
1.25
1.69
3.25
300.8±165.0
10.90
16.53
6.07
14.73
7.03
31.47
17.51
Vegetable oils
26.0 ± 6.4
12.03
0
0
32.55
0
0
0
Other foods b
270.8±228.7
25.25
10.55
31.96
18.30
4.17
7.02
4.81
Vegetables Legumes
Milk and dairy products
a
Mean values adjusted for energy intake, b Includes bakery products, margarine, sweets, cakes, coffee, beverages and mixed foods.
Vitamin (%)
B12
Diet and homocysteine in the elderly
387
P5
P25
P~
Prs
P~
Vegetables
69.4
125.9
166.6
225.6
439.8
Fruit
61.0
141.5
229.8
384.2
596.2
Fish
13.1
49.6
64.0
85.4
116.3
Meat
49.8
81.7
105.3
135.6
177.6
Milk and dairy products
118.5
190.5
263.1
377.1
600.8
TABLE 3
Percentile values (g/day) of the food groups used to calculate the diet score.
n=140.
groups used to calculate the diet score were characterised by the the values corresponding to the 5th,, 25 th, 50 th, 75 th, and 95 th, percentiles of the distribution of the different groups. Multiple linear regression analyses were used to study the associations between tHcy and the score components and total score. Each regression model contained tHcy as the dependent variable and was controlled for the effects of age, gender and serum creatinine. When skewed, the variables were transformed by taking natural logarithms or calculating the square root in order to normalise their distribution. All of the statistical analyses were performed using SPSS 11.0 for Windows (SPSS Inc., Chicago, IL). A p value <0.05 was considered statistically significant.
The coefficients for each of the components of the diet score and overall score in the multiple regression models were estimated in order to predict tHcy (Table 4). None of them showed a statistically significant association with tHcy on their own, but the sum of all the partial scores (the total diet score) was negatively associated with tHcy: a 0.28 lamol/L reduction in tHcy for every unit increase in the score. The percentages of subjects in the upper tHcy quartile (tHcy >16 lamol/L) by range of diet scores are shown in Figure 1. The highest percentage (33-38%) was found among the subjects with a score of less than 7, whereas this percentage was _<21% in the groups scoring 7 or more (Chisquare = 5.311,p=0.021). TABLE 4
Results The main characteristics of the sample population are shown in Table 1. Mean tHcy was 13.3 lamol/L, adjusted for age and gender. The total diet score ranged from 1 to 13, without any significant between-gender differences (data not shown). The mean folate intake was far below the recommended 400 lag/day, whereas dietary riboflavin and cobalamin levels were in line with the recommendations (18). The contribution of each food group to energy and macronutrient intake, as well as to the intake of folate, cobalamin and riboflavin, is shown in Table 2. The five groups on which the score is based accounted for scarcely 40% of total energy intake, but provided 85% of folate and riboflavin, and 92% of cobalamin. The "other" group accounted for more than 25% of energy consumption, but less than 5% of folate and cobalamin, and 7% of riboflavin. Table 3 shows the quartile intake values of the food groups used to calculate the diet score, together with the 5th and 95th percentiles of intake distribution.
Coefficients for single components and total diet score in multiple linear regression models to predict total plasma homocysteine (tHcy) concentration. Score component
b coefficient
(95% Cl)
p
Vegetables (g/day) 1
-0.832 (13=-0.102)
(-2.051,0.387)
0.180
Fruit (g/day)2
-0.112 (13=-0.125)
(-0.245, 0.022)
0.1 O0
Fish (g/day)
-0.013 (p=-0.083)
(-0.035, 0.010)
0.276
Meat (g/day)
-0.0005 (p=o.oo4)
(-0.020, 0.021)
0.960
Milk and dairy products (g/day) 1
-0.112 (p=-0.114)
(-2.582, 0.358)
0.137
Total score
-0,280 (-0.545, -0.015) (~=-0.156)
0.039
n=140. All regression models were controlled for the effect of sex, age and serum creatinine. 1Variables transformed by taking the natural logarithm. 2 Variable transformed by calculating the square root.
C. Lasheras, et al
388
FIGURE 1 Percentage of subjects with high tHcy levels by total diet score groups. The diet score is based on the intakes of vegetables, fruit, fish, meat and milk and dairy products (see text). The bars represent the percentage of subjects with total plasma homocysteine levels above 16 pmol/L in each score group. 4O >, O
35
"O
'~ 30 >
-~ 25 .IZ 20 "6 0~ 15 2T to O
10 5 <5
5-6
7-8
9-10
>10
Diet score
Discussion Over recent years, a great deal of attention has been paid to the role of individual foods or food groups in disease promotion and prevention. Given the complexity of human diets, conclusions concerning the effect of the consumption of a single nutrient, food or dietary constituent on a specific health outcome may be misleading. For this reason, we think it useful to examine global indices of food intake that simultaneously express a number of related aspects of dietary intake (19). The main finding of our study was the significant association between the overall diet score and tHcy, and the non-significant association of the individual variables on which the score is based. A diet characterised by a higher intake of folate-, cobalamin- and riboflavin-rich foods at the same time led to significantly lower tHcy levels and a lower percentage of subjects with high tHcy levels, which was not found when each food group was analysed separately. This provides further evidence that a dietary pattern rather than the individual constituents of a diet plays a role in disease prevention, as previously observed by us (20) and others (12). The relationship of tHcy with dietary folate, cobalamin and riboflavin is well established (4-7, 9). Nevertheless, there are fewer studies of food intake and homocysteine, most of which considered the protective effects of pharma-
cological vitamin doses. We therefore thought it would be interesting to discover whether the effects observed with high doses of the individual compounds were related to the same degree or in the same manner as those observed in humans consuming realistic doses as part of a habitual diet. The fact that we did not find any relationship between tHcy and any individual food group alone can be explained in the light of homocysteine metabolism. Homocysteine is remethylated to methionine through a pathway that requires folate, riboflavin and cobalamin, which means that all three vitamins must be simultaneously present in sufficient amounts to prevent an increase in tHcy concentrations. In our sample, vegetables and fruit together accounted for 70% of folate intake; vegetables were also a fair source of riboflavin, but did not provide cobalamin. We found that a higher intake of any of these foods alone was not enough to prevent higher tHcy levels, which remained low only when they were eaten in combination with foods of animal origin (which supply enough Bx2). In line with this, other studies have found that vegetarian diets seem to increase tHcy levels and the risk of hyperhomocysteinemia (21, 22) because of a poor Ba2 status. In addition to their great contribution to cobalamin and riboflavin intake, meat, fish and dairy products are also the main sources of dietary proteins, and it has been shown that protein intake is an independent predictor of tHcy levels even when other dietary factors, such as energy and folate intake, are controlled for (23). The mechanism underlying the protein-induced decrease in tHcy is not known, but it could involve vitamin B12 insofar as all of the good sources of this vitamin are animal products. In our study, as much as 92% of dietary cobalamin intake was provided by these food groups. It is worth noting that this study involved a population with an eating pattern characteristic of the Mediterranean diet which, in terms of the foods selected for our diet score, means a low consumption of meat, milk and dairy products, and a high intake of vegetables and fruit. Therefore, even when meat and milk and dairy products are components of our score (and consequently a larger intake leads to a higher score), it must be remembered that the highest quartile of these foods was still moderate (P75= 135.6 g/day of meat and 377.1 g/day of milk and dairy products). The consumption of vegetables and fruit in our population was considerably higher than that found in most European countries and the United States (24), and the mean intake was above the minimum recommended combined intake of vegetables and fruit of 400 g/day, which is equivalent to five servings per day (25).
Diet and homocysteine in the elderly
To the best of our knowledge, only one other study has analysed the effect of global dietary patterns on plasma tHcy concentrations (12). The authors found that a "prudent dietary pattern" (characterised by higher intakes of fruit, vegetables, poultry, fish, whole grains and legumes) was associated with a significantly lower level of tHcy, whereas a "Western pattern" (with higher intakes of red and processed meat, French fries, eggs, high-fat dairy products, sweets and refined grains) positively correlated with tHcy. However, as the same person can score high in both patterns, the effects of these patterns on tHcy may be contradictory. In another study, Tucker et al examined whether the main dietary sources of folate were associated with plasma folate levels and, subsequently, tHcy levels (13); however, their approach did not reflect the overall dietary pattern of the elderly subjects studied, but the association between tHcy and the individual components of the diet. Most estimate hyperhomocysteinemia on the basis of a cut-off value above the 90th, or 95th, percentile of tHcy distribution in the general population. As no information is available concerning the distribution of tHcy in the general Spanish population, we used the P75 in our group (16 lamol/L) as our cut-off point. This value is near to the 15 lamol/L proposed by Kang et al as defining hyperhomocysteinemia (26). The percentage of subjects with tHcy levels of more than 16 lamol/L was almost twice as high in our subjects with a diet score of less than 7, once again indicating that emphasis must be placed on the diet as a whole rather than its individual components. However, as the aim of this study was to analyse the association between habitual diet and tHcy levels in elderly people, we did not consider serum vitamin levels, and this may represent a limitation of the study. In conclusion, the fact that people eat foods instead of single nutrients means that the overall diet comprises the intake of various factors that may act synergistically or antagonistically in combination. The use of dietary pattern approaches can therefore shed further light on the relationship between diet and disease, and thus complement traditional single-food (or nutrient) analyses.
Acknowledgements This study was supported by grant FISS-03/020141 from the Fondo de Investigaciones Sanitarias (Instituto de Salud Carlos III, Madrid, Spain), and the Grande Covifin grant from the Prince of Asturias Foundation (Oviedo, Spain).
389
References 1. Graham IM, Daly LE, Refsum HM, Robinson K, Brattstrom LE, Ueland PM, Palma-Reis RJ, Boers GH, Sheahan RG, Israelsson B, Uiterwaal CS, Meleady R, McMaster D, Verhoef P, Witteman J, Rubba P, Bellet H, Wautrecht JC, de Valk HW, Sales Luis AC, Parrot-Rouland FM, Tan KS, Higgins I, Garcon D, Andria, G (1997) Plasma homocysteine as a risk factor for vascular disease. The European Concerted Action Project. JAMA 277:1775-1781 2. Boushey CJ, Beresford SA, Omenn GS, Motulsky AG (1995) A quantitative assessment of plasma homocysteine as a risk factor for vascular disease. Probable benefits of increasing folic acid intakes. JAMA 274:1049-1057 3. Vollset SE, Refsum H, Tverdal A, Nygard O, Nordrehaug JE, Tell GS, Ueland PM (2001) Plasma total homocysteine and cardiovascular and noncardiovascular mortality: the Hordaland Homocysteine Study. Am J Clin Nutr 74:130-136 4. Selhub J, Jacques PF, Wilson PW, Rush D, Rosenberg IH (1993) Vitamin status and intake as primary determinants of homocysteinemia in an elderly population. JAMA 270:2693-2698 5. Jacques PF, Bostom AG, Wilson PW, Rich S, Rosenberg IH, Selhub J (2001) Determinants of plasma total homocysteine concentration in the Framingham Offspring cohort. Am J Clin Nutr 73:613-621 6. Swift ME, Schultz TD (1986) Relationship of vitamins B6 and B12 to homocysteine levels: risk for coronary heart disease. Nutr Rep Int. 34:1-14 7. Selhub J, Jacques PF, Rosenberg IH, Rogers G, Bowman BA, Gunter EW, Wright JD, Johnson CL (1999) Serum total homocysteine concentrations in the third National Health and Nutrition Examination Survey (1991-1994): population reference ranges and contribution of vitamin status to high serum concentrations. Ann Intern Med 131:331-339 8. Selhub J (1999) Homocysteine metabolism. Annu Rev Nutr 19: 217-246 9. Robinson K, Arheart K, Refsum H, Brattstrom L, Boers G, Ueland P, Rubba P, Palma-Reis R, Meleady R, Daly L, Witteman J, Graham I (1998) Low circulating folate and vitamin B 6 concentrations: risk factors for stroke, peripheral vascular disease, and coronary artery disease. European COMAC Group. Circulation 97:437-443 10. Venn BJ, Mann JI, Williams SM, Riddell LJ, Chisholm A, Harper MJ, Aitken W, Rossaak JI (2002) Assessment of three levels of folic acid on serum folate and plasma homocysteine: a randomised placebo-controlled double-blind dietary intervention trial. Eur J Clin Nutr 56:748-754 11. Quinlivan EP, McPartlin J, McNulty H, Ward M, Strain JJ, Weir DG, Scott JM (2002) Importance of both folic acid and vitamin B12 in reduction of risk of vascular disease. Lancet 359:227-228 12. Fung TI', Rimm EB, Spiegelman D, Rifai N, Tofler GH, Willett WC, Hu FB (2001) Association between dietary patterns and plasma biomarkers of obesity and cardiovascular disease risk. Am J Clin Nutr 73:61-67
390
13. Tucker KL, Selhub J, Wilson PWF, Rosenberg IH (1996) Dietary intake pattern relates to plasma folate and homocysteine concentrations in the Framingham Heart Study. J Nutr 126: 3025-303l 14. Varela-Moreiras G (2001) Nutritional regulation of homocysteine: effects of drugs. Biomed Pharmacother 55:448-453 15. Catargi B, Parrot-Rouland F, Cochet C (1999) Homocysteine, hypothyroidism, and effect of thyroid hormone replacement. Thyroid 9:1163-1166 16. Instituto de Nutrici6n (1994) Tablas de Composici6n de Alimentos. CSIC, Madrid. 17. Araki A, Sako Y (1987) Determination of free and total homocysteine in human plasma by high-performance liquid chromatography with fluorescence detection. J Chromatogr 422:43-52 18. Institute of Medicine (1998) Dietary Reference Intakes for Thiamin, Riboflavin, Niacin, Vitamin Bt, Folate, Vitamin B12, Pantothenic Acid, Biotin, and Choline. National Academy Press, Washington, DC. 19. Kant AK (1996) Indexes of overall diet quality: a review. J Am Diet Assoc 96:785-791 20. Lasheras C, Fern~indez S, Patterson AM (2000) Mediterranean diet and age with respect to overall survival in institutionalized, nonsmoking elderly people. Am J Clin Nutr 71:987-992
C. Lasheras, et al
21. Herrmann W, Schorr H, Purschwitz K, Rassoul F, Richter V (2001) Total homocysteine, vitamin Ble, and total antioxidant status in vegetarians. Clin Chem 47:1094-1101 22. Bissoli L, Di Francesco V, Ballarin A, Mandragona R, Trespidi R, Brocco G, Caruso B, Bosello O, Zamboni M (2002) Effect of vegetarian diet on homocysteine levels. Ann Nutr Metab 46:73-79 23. Stolzenberg-Solomon RZ, Miller ER II1, Maguire MG, Selhub J, Appel LJ (1999) Association of dietary protein intake and coffee consumption with serum homocysteine concentrations in an older population. Am J Clin Nutr 69:467-475 24. Agudo A, Amiano P, Barcos A, Barricarte A, Beguirist~iin JM, Chirlaque MD, Dorronsoro M, Gonzglez CA, Lasheras C, Martfnez C, Navarro C, Pera G, Quir6s JR, Rodrlguez M, Tormo MJ (1999) Dietary intake of vegetables and fruits among adults in five regions of Spain. EPIC Group of Spain. European Prospective Investigation into Cancer and Nutrition. Eur J Clin Nutr 53:174-180 25. Williams C (1995) Healthy eating: clarifying advice about fruit and vegetables. BMJ 310:1453-1455 26. Kang SS, Wong PW, Malinow MR (1992) Hyperhomocysteinemia as a risk factor for occlusive vascular disease. Annu Rev Nutr 12: 279-298
Nutr Metab Cardiovasc Dis (2003) 13:384--390
ORIGINAL
ARTICLE
Diet score is associated with plasma homocysteine in a healthy institutionalised elderly population C. Lasheras 1, J.M. Huerta 1, S. Gonz~ilez I, M. Prada 2, S. Braga 2, S. Fern~indez 1, and A.M. Patterson I 1Departamento de Biologia Funcional, Area de Fisiolog/a,Facultad de Medicina, Universidad de Oviedo, and 2Laboratorio de An~ilisisCl/nicos,Hospital Central de Asturias, Oviedo, Spain
Abstract Background and Aim: Total plasma homocysteine (tHcy) is an independent risk factor for cardiovascular diseases.Among other dietary and non-dietary factors, B vitamins, such as folate, riboflavin and cobalamin, are primary determinants of tHcy in the general population. However, research has concentrated on the relationship of these nutrients with tHcy, and little is known about overall eating patterns and tHcy. Methods and Results: In this study, we analysed whether a diet score based on the consumption of folate-, riboflavinand cobalamin-rich food groups was associated with tHcy in a sample of 140 institutionalised elderly subjects (59 men and 81 women aged 60-80 years) from Northern Spain. The food groups identified as the major contributors to the intake of the three vitamins were vegetables, fruit, fish, meat and milk and dairy products. The mean tHcy level was 13.3+5.1 pmol/L (range: 3.9-30.7 pmol/L). None of the food groups predicted tHcy levels individually, but the overall diet score was inversely associated with tHcy in a multiple linear regression analysis. High tHcy levels (>16 pmol/L) were almost twice as prevalent in the groups scoring less than 7 than in those scoring 7 or more (37.5 vs 19.6%, p=0.021). Conclusions: These data suggest that a dietary pattern characterised by high intakes of B vitamin-rich foods is associat-
Key words: Homocysteine, hyperhomocysteinemia,vitamins,elderly, dietary patterns. Correspondence to: Cristina Lasheras PhD, Departamento de Biologfa Funcional, Area de Fisiologfa,Facultad de Medicina, Universidad de Oviedo, JuMn Claveria s/n, 33006, Oviedo, Spain. E-mail: [email protected] Received: 5 June 2003;accepted: 1 August 2003.
ed with a lower tHcy concentration and a reduced percentage of high tHcy levels in elderly subjects. They also support the use of dietary pattern approaches to evaluate the relationships between diet and health outcomes that go beyond single nutrient analyses. Nutr Metab Cardiovasc Dis (2003) 13:384-390 e2003, Medikal Press
Introduction The level of total plasma homocysteine (tHcy) has attracted much interest over recent years because a high tHcy concentration has been associated with an increased risk of cardiovascular disease (1, 2) and all-cause mortality (3). Vitamin status is one of the major determinants of total plasma homocysteine levels in the general population (4, 5), in which it has been estimated as accounting for two-thirds of the cases of high tHcy concentrations (6, 7). B vitamins such as folate, B 6 and riboflavin are important cofactors for a number of the enzymes involved in homocysteine metabolism (5, 8), and cobalamin (B12) is the final methyl-group donor in the homocysteine to methionine remethylation pathway (8). As a result, a large number of studies have been carried out over the last ten years with the aim of elucidating the relationship between these vitamins and vascular pathologies (9), as well as the potential benefits of supplementation programmes on population tHcy levels (10, 11). Vitamins are, by definition, essential for human life, and most of them can only be provided through diet, As freeliving people do not consume single nutrients but combinations of several foods, a second-stage analysis of dietrelated diseases has to move from nutrient to food intake
385
Diet and homocysteine in the elderly
or dietary patterns in order to understand the nature of such diseases and provide the general population with recommendations for promoting healthy dietary habits. Although there are a number of studies of the importance of B vitamins in tHcy metabolism, very few have analysed the connection between food intake and tHcy levels (5, 12, 13) despite the fact that diet is the most important source of most of these vitamins, specially in countries (such as Spain) in which folate fortification policies have not been implemented. Two previous studies have suggested that dietary patterns may be related to plasma homocysteine levels (12, 13). In this study, we developed a diet score based on the intake of the food groups that make the greatest contribution to folate, cobalamin and riboflavin intake in a sample of elderly people aged 60-80 years living in Institutions in Asturias (Spain). Our hypothesis was that the higher the score, the lower the tHcy concentration should be. This cross-sectional report forms part of an ongoing prospective study of diet and disease.
Me~o~ Study subjects The analysed sample consisted of 140 healthy institutionalised elderly subjects (59 men and 81 women) recruited in seven nursing homes located in Asturias (Northern Spain). Subjects with a history of cancer or cardiovascular disease were not included as they will be end-points in our prospective investigation. Their mean age at the time of enrolment was 73.3_+5.6 years for men and 74.2_+4.8 years for women; all of the subjects had been living at the institutions for at least three years (mean 5.2 years). The medical history of each subject was taken before enrolment. Nine patients taking antiepileptic drugs or thyroid hormones were excluded from the analysis because these drugs can interfere with plasma tHcy levels (14, 15). Although it has been reported that antihypertensive medication can affect tHcy (5), the subjects taking antihypertensive drugs were not excluded as we found no differences in tHcy levels between those who were on this medication and those who were not (data not shown). Two subjects taking vitamin supplements on a regular basis were not included in the analysis. All of the participants were mentally and physically capable of participating in the study, and all of them gave their informed consent. The study design was approved by
the Committee on Ethical Research of the Oviedo University Hospital. After the subjects had given their informed consent, a personal appointment was made for collecting data.
Dietary intake Dietary intake was assessed by means of a food frequency questionnaire (FFQ) specifically designed for each of the seven Institutions after they had kindly provided us with the menus of the previous year. Trained dietitians asked questions about cooking practices, the numbers and amounts of the ingredients used in each recipe, and menu preparation (the type of oil, type of milk, etc.). The FFQ contained all of the individual foods (not food groups) in the menus of each Institution. During an interview, the subjects were asked item-by-item whether they usually ate each food and, if so, how much they ate. For this purpose, pictures of three serving sizes of each cooked food were presented to the participants so that they could choose from up to seven serving portions (from "less than the small one" to "more than the large one"). For some of the foods consumed, the amounts were recorded in household units, by volume or by length measured with a ruler. All of the subjects were asked whether they had access to foods other than those offered by the Institution (through their relatives, for example). Food intake was analysed for energy and macro- and micronutrient content using the nutrient Food Composition Tables developed by the CSIC (16).
Blood measurements Blood samples were drawn by venipuncture after a 12hour fast, and the plasma and serum were collected in separate tubes. The samples were kept on ice and centrifuged within 2-4 hours of collection. Plasma aliquots were kept at -70°C until analysis. Total plasma homocysteine was determined using reversephase HPLC with fluorimetric detection (17) and a Biorad analyser. In order to ensure the stability of the analytical procedure, cysteine and cysteineglycine (two amino thiols) were measured together with homocysteine in the same sample and we checked that their concentrations (absolute and relative to homocysteine) were within the normal range. Serum creatinine was measured in a Hitachi 737 analyser (Boehringer Mannheim Corp, Indianapolis, IN).
Food groupings The foods were grouped on the basis of their characteristics. Vegetables included all types of vegetables except fruits,
386
C. Lasheras, et al
dried fruits (eg raisins, dried figs), nuts and seeds (hazelnuts, almonds, walnuts), which made up the fruit group. The cereals were bread, flour, rice and pasta. Potatoes included boiled potatoes and French fries. Fish and seafood were grouped as fish, and pork, poultry and red meat made up the meat group. The vegetable oils were olive and sunflower oils. Milk and dairy products included all types of milk, yogurt and cheese. The "other" group included bakery products, margarine, sweets, cakes, coffee, beverages and mixed foods (such as pizza, paella, etc.).
TABLE 1
Characteristics and selected total plasma homocysteine (tHcy) determinants in the study sample. Mean value ± SD or percentage Plasma tHcy (iJmol/L) a
13.3±5.1
Serum creatinine (iJmol/L) a
92.1 +19.5
Age (yr)
73.8,+5.2
Female sex (%)
57.9 [81] b
Body mass index (kg/m 2)
28.3±4.9
Mean diet score c
7.5±2.8
Energy (kJ/day)
8.1 ± 1.7
Folate intake (IJg/day)
Diet score The diet score was based on the quartiles of the intakes (grams per day) of the food groups previously identified as being the main contributors to the intake of folate, B n and riboflavin (see Results): ie, meat, fish, milk and dairy products, fruit, and vegetables. Each variable was categorised into quartiles and given a score from 0 to 3 (0 = an intake in the lowest quartile and 3 = an intake in the highest quartile). The total score (range 0-15) was the sum of all of the partial scores.
187.3+81.1
Riboflavin intake (mg/day)
1.8_+0.5
B12 intake (IJg/day)
4.9+1.8
Protein intake (g/day)
76.2±15.2
Coffee intake d (mL/day)
30.1 ±19.5
a Adjusted for age and gender, b Percentage (n in brackets), c Based on intake of vegetables, fruit, meat, fish, milk and dairy products (see text for details). d Only consumers.
Statistical analyses The descriptive data in the tables are expressed as mean values + SD or percentages. The intakes of the food
TABLE 2
Contribution of different food groups to the intake of macronutrients and selected vitamins in the study population. g/day (mean ± SD) a
Energy (%)
Protein (%)
Carbohydrate (%)
Lipid (%)
Folate (%)
Riboflavin (%)
192.5±105.8
3.15
6.57
4.72
1.70
49.55
33.15
0
17.5,+7.3
2.72
4.51
3.7
0.85
5.77
1.73
0
Fruit
275.8+192.2
6.23
1.86
12.87
0.t0
19.94
3.80
0
Cereals
210.0,+89.4
20.52
15.45
36.11
4.44
1.01
4.39
0
Potatoes
41.5+32.1
1.76
1.31
3.15
0.31
2.94
1.21
0
Fish
68.6+34.1
3.60
15.11
0.17
2.95
3.76
4.99
44.41
Meat
108.4+37.8
12.95
26.71
0.53
22.90
4.74
10.96
29.74
Eggs
9.3±5.9
0.72
1.53
0
1.30
1.25
1.69
3.25
300.8±165.0
10.90
16.53
6.07
14.73
7.03
31.47
17.51
Vegetable oils
26.0 ± 6.4
12.03
0
0
32.55
0
0
0
Other foods b
270.8±228.7
25.25
10.55
31.96
18.30
4.17
7.02
4.81
Vegetables Legumes
Milk and dairy products
a
Mean values adjusted for energy intake, b Includes bakery products, margarine, sweets, cakes, coffee, beverages and mixed foods.
Vitamin (%)
B12
Diet and homocysteine in the elderly
387
P5
P25
P~
Prs
P~
Vegetables
69.4
125.9
166.6
225.6
439.8
Fruit
61.0
141.5
229.8
384.2
596.2
Fish
13.1
49.6
64.0
85.4
116.3
Meat
49.8
81.7
105.3
135.6
177.6
Milk and dairy products
118.5
190.5
263.1
377.1
600.8
TABLE 3
Percentile values (g/day) of the food groups used to calculate the diet score.
n=140.
groups used to calculate the diet score were characterised by the the values corresponding to the 5th,, 25 th, 50 th, 75 th, and 95 th, percentiles of the distribution of the different groups. Multiple linear regression analyses were used to study the associations between tHcy and the score components and total score. Each regression model contained tHcy as the dependent variable and was controlled for the effects of age, gender and serum creatinine. When skewed, the variables were transformed by taking natural logarithms or calculating the square root in order to normalise their distribution. All of the statistical analyses were performed using SPSS 11.0 for Windows (SPSS Inc., Chicago, IL). A p value <0.05 was considered statistically significant.
The coefficients for each of the components of the diet score and overall score in the multiple regression models were estimated in order to predict tHcy (Table 4). None of them showed a statistically significant association with tHcy on their own, but the sum of all the partial scores (the total diet score) was negatively associated with tHcy: a 0.28 lamol/L reduction in tHcy for every unit increase in the score. The percentages of subjects in the upper tHcy quartile (tHcy >16 lamol/L) by range of diet scores are shown in Figure 1. The highest percentage (33-38%) was found among the subjects with a score of less than 7, whereas this percentage was _<21% in the groups scoring 7 or more (Chisquare = 5.311,p=0.021). TABLE 4
Results The main characteristics of the sample population are shown in Table 1. Mean tHcy was 13.3 lamol/L, adjusted for age and gender. The total diet score ranged from 1 to 13, without any significant between-gender differences (data not shown). The mean folate intake was far below the recommended 400 lag/day, whereas dietary riboflavin and cobalamin levels were in line with the recommendations (18). The contribution of each food group to energy and macronutrient intake, as well as to the intake of folate, cobalamin and riboflavin, is shown in Table 2. The five groups on which the score is based accounted for scarcely 40% of total energy intake, but provided 85% of folate and riboflavin, and 92% of cobalamin. The "other" group accounted for more than 25% of energy consumption, but less than 5% of folate and cobalamin, and 7% of riboflavin. Table 3 shows the quartile intake values of the food groups used to calculate the diet score, together with the 5th and 95th percentiles of intake distribution.
Coefficients for single components and total diet score in multiple linear regression models to predict total plasma homocysteine (tHcy) concentration. Score component
b coefficient
(95% Cl)
p
Vegetables (g/day) 1
-0.832 (13=-0.102)
(-2.051,0.387)
0.180
Fruit (g/day)2
-0.112 (13=-0.125)
(-0.245, 0.022)
0.1 O0
Fish (g/day)
-0.013 (p=-0.083)
(-0.035, 0.010)
0.276
Meat (g/day)
-0.0005 (p=o.oo4)
(-0.020, 0.021)
0.960
Milk and dairy products (g/day) 1
-0.112 (p=-0.114)
(-2.582, 0.358)
0.137
Total score
-0,280 (-0.545, -0.015) (~=-0.156)
0.039
n=140. All regression models were controlled for the effect of sex, age and serum creatinine. 1Variables transformed by taking the natural logarithm. 2 Variable transformed by calculating the square root.
C. Lasheras, et al
388
FIGURE 1 Percentage of subjects with high tHcy levels by total diet score groups. The diet score is based on the intakes of vegetables, fruit, fish, meat and milk and dairy products (see text). The bars represent the percentage of subjects with total plasma homocysteine levels above 16 pmol/L in each score group. 4O >, O
35
"O
'~ 30 >
-~ 25 .IZ 20 "6 0~ 15 2T to O
10 5 <5
5-6
7-8
9-10
>10
Diet score
Discussion Over recent years, a great deal of attention has been paid to the role of individual foods or food groups in disease promotion and prevention. Given the complexity of human diets, conclusions concerning the effect of the consumption of a single nutrient, food or dietary constituent on a specific health outcome may be misleading. For this reason, we think it useful to examine global indices of food intake that simultaneously express a number of related aspects of dietary intake (19). The main finding of our study was the significant association between the overall diet score and tHcy, and the non-significant association of the individual variables on which the score is based. A diet characterised by a higher intake of folate-, cobalamin- and riboflavin-rich foods at the same time led to significantly lower tHcy levels and a lower percentage of subjects with high tHcy levels, which was not found when each food group was analysed separately. This provides further evidence that a dietary pattern rather than the individual constituents of a diet plays a role in disease prevention, as previously observed by us (20) and others (12). The relationship of tHcy with dietary folate, cobalamin and riboflavin is well established (4-7, 9). Nevertheless, there are fewer studies of food intake and homocysteine, most of which considered the protective effects of pharma-
cological vitamin doses. We therefore thought it would be interesting to discover whether the effects observed with high doses of the individual compounds were related to the same degree or in the same manner as those observed in humans consuming realistic doses as part of a habitual diet. The fact that we did not find any relationship between tHcy and any individual food group alone can be explained in the light of homocysteine metabolism. Homocysteine is remethylated to methionine through a pathway that requires folate, riboflavin and cobalamin, which means that all three vitamins must be simultaneously present in sufficient amounts to prevent an increase in tHcy concentrations. In our sample, vegetables and fruit together accounted for 70% of folate intake; vegetables were also a fair source of riboflavin, but did not provide cobalamin. We found that a higher intake of any of these foods alone was not enough to prevent higher tHcy levels, which remained low only when they were eaten in combination with foods of animal origin (which supply enough Bx2). In line with this, other studies have found that vegetarian diets seem to increase tHcy levels and the risk of hyperhomocysteinemia (21, 22) because of a poor Ba2 status. In addition to their great contribution to cobalamin and riboflavin intake, meat, fish and dairy products are also the main sources of dietary proteins, and it has been shown that protein intake is an independent predictor of tHcy levels even when other dietary factors, such as energy and folate intake, are controlled for (23). The mechanism underlying the protein-induced decrease in tHcy is not known, but it could involve vitamin B12 insofar as all of the good sources of this vitamin are animal products. In our study, as much as 92% of dietary cobalamin intake was provided by these food groups. It is worth noting that this study involved a population with an eating pattern characteristic of the Mediterranean diet which, in terms of the foods selected for our diet score, means a low consumption of meat, milk and dairy products, and a high intake of vegetables and fruit. Therefore, even when meat and milk and dairy products are components of our score (and consequently a larger intake leads to a higher score), it must be remembered that the highest quartile of these foods was still moderate (P75= 135.6 g/day of meat and 377.1 g/day of milk and dairy products). The consumption of vegetables and fruit in our population was considerably higher than that found in most European countries and the United States (24), and the mean intake was above the minimum recommended combined intake of vegetables and fruit of 400 g/day, which is equivalent to five servings per day (25).
Diet and homocysteine in the elderly
To the best of our knowledge, only one other study has analysed the effect of global dietary patterns on plasma tHcy concentrations (12). The authors found that a "prudent dietary pattern" (characterised by higher intakes of fruit, vegetables, poultry, fish, whole grains and legumes) was associated with a significantly lower level of tHcy, whereas a "Western pattern" (with higher intakes of red and processed meat, French fries, eggs, high-fat dairy products, sweets and refined grains) positively correlated with tHcy. However, as the same person can score high in both patterns, the effects of these patterns on tHcy may be contradictory. In another study, Tucker et al examined whether the main dietary sources of folate were associated with plasma folate levels and, subsequently, tHcy levels (13); however, their approach did not reflect the overall dietary pattern of the elderly subjects studied, but the association between tHcy and the individual components of the diet. Most estimate hyperhomocysteinemia on the basis of a cut-off value above the 90th, or 95th, percentile of tHcy distribution in the general population. As no information is available concerning the distribution of tHcy in the general Spanish population, we used the P75 in our group (16 lamol/L) as our cut-off point. This value is near to the 15 lamol/L proposed by Kang et al as defining hyperhomocysteinemia (26). The percentage of subjects with tHcy levels of more than 16 lamol/L was almost twice as high in our subjects with a diet score of less than 7, once again indicating that emphasis must be placed on the diet as a whole rather than its individual components. However, as the aim of this study was to analyse the association between habitual diet and tHcy levels in elderly people, we did not consider serum vitamin levels, and this may represent a limitation of the study. In conclusion, the fact that people eat foods instead of single nutrients means that the overall diet comprises the intake of various factors that may act synergistically or antagonistically in combination. The use of dietary pattern approaches can therefore shed further light on the relationship between diet and disease, and thus complement traditional single-food (or nutrient) analyses.
Acknowledgements This study was supported by grant FISS-03/020141 from the Fondo de Investigaciones Sanitarias (Instituto de Salud Carlos III, Madrid, Spain), and the Grande Covifin grant from the Prince of Asturias Foundation (Oviedo, Spain).
389
References 1. Graham IM, Daly LE, Refsum HM, Robinson K, Brattstrom LE, Ueland PM, Palma-Reis RJ, Boers GH, Sheahan RG, Israelsson B, Uiterwaal CS, Meleady R, McMaster D, Verhoef P, Witteman J, Rubba P, Bellet H, Wautrecht JC, de Valk HW, Sales Luis AC, Parrot-Rouland FM, Tan KS, Higgins I, Garcon D, Andria, G (1997) Plasma homocysteine as a risk factor for vascular disease. The European Concerted Action Project. JAMA 277:1775-1781 2. Boushey CJ, Beresford SA, Omenn GS, Motulsky AG (1995) A quantitative assessment of plasma homocysteine as a risk factor for vascular disease. Probable benefits of increasing folic acid intakes. JAMA 274:1049-1057 3. Vollset SE, Refsum H, Tverdal A, Nygard O, Nordrehaug JE, Tell GS, Ueland PM (2001) Plasma total homocysteine and cardiovascular and noncardiovascular mortality: the Hordaland Homocysteine Study. Am J Clin Nutr 74:130-136 4. Selhub J, Jacques PF, Wilson PW, Rush D, Rosenberg IH (1993) Vitamin status and intake as primary determinants of homocysteinemia in an elderly population. JAMA 270:2693-2698 5. Jacques PF, Bostom AG, Wilson PW, Rich S, Rosenberg IH, Selhub J (2001) Determinants of plasma total homocysteine concentration in the Framingham Offspring cohort. Am J Clin Nutr 73:613-621 6. Swift ME, Schultz TD (1986) Relationship of vitamins B6 and B12 to homocysteine levels: risk for coronary heart disease. Nutr Rep Int. 34:1-14 7. Selhub J, Jacques PF, Rosenberg IH, Rogers G, Bowman BA, Gunter EW, Wright JD, Johnson CL (1999) Serum total homocysteine concentrations in the third National Health and Nutrition Examination Survey (1991-1994): population reference ranges and contribution of vitamin status to high serum concentrations. Ann Intern Med 131:331-339 8. Selhub J (1999) Homocysteine metabolism. Annu Rev Nutr 19: 217-246 9. Robinson K, Arheart K, Refsum H, Brattstrom L, Boers G, Ueland P, Rubba P, Palma-Reis R, Meleady R, Daly L, Witteman J, Graham I (1998) Low circulating folate and vitamin B 6 concentrations: risk factors for stroke, peripheral vascular disease, and coronary artery disease. European COMAC Group. Circulation 97:437-443 10. Venn BJ, Mann JI, Williams SM, Riddell LJ, Chisholm A, Harper MJ, Aitken W, Rossaak JI (2002) Assessment of three levels of folic acid on serum folate and plasma homocysteine: a randomised placebo-controlled double-blind dietary intervention trial. Eur J Clin Nutr 56:748-754 11. Quinlivan EP, McPartlin J, McNulty H, Ward M, Strain JJ, Weir DG, Scott JM (2002) Importance of both folic acid and vitamin B12 in reduction of risk of vascular disease. Lancet 359:227-228 12. Fung TI', Rimm EB, Spiegelman D, Rifai N, Tofler GH, Willett WC, Hu FB (2001) Association between dietary patterns and plasma biomarkers of obesity and cardiovascular disease risk. Am J Clin Nutr 73:61-67
390
13. Tucker KL, Selhub J, Wilson PWF, Rosenberg IH (1996) Dietary intake pattern relates to plasma folate and homocysteine concentrations in the Framingham Heart Study. J Nutr 126: 3025-303l 14. Varela-Moreiras G (2001) Nutritional regulation of homocysteine: effects of drugs. Biomed Pharmacother 55:448-453 15. Catargi B, Parrot-Rouland F, Cochet C (1999) Homocysteine, hypothyroidism, and effect of thyroid hormone replacement. Thyroid 9:1163-1166 16. Instituto de Nutrici6n (1994) Tablas de Composici6n de Alimentos. CSIC, Madrid. 17. Araki A, Sako Y (1987) Determination of free and total homocysteine in human plasma by high-performance liquid chromatography with fluorescence detection. J Chromatogr 422:43-52 18. Institute of Medicine (1998) Dietary Reference Intakes for Thiamin, Riboflavin, Niacin, Vitamin Bt, Folate, Vitamin B12, Pantothenic Acid, Biotin, and Choline. National Academy Press, Washington, DC. 19. Kant AK (1996) Indexes of overall diet quality: a review. J Am Diet Assoc 96:785-791 20. Lasheras C, Fern~indez S, Patterson AM (2000) Mediterranean diet and age with respect to overall survival in institutionalized, nonsmoking elderly people. Am J Clin Nutr 71:987-992
C. Lasheras, et al
21. Herrmann W, Schorr H, Purschwitz K, Rassoul F, Richter V (2001) Total homocysteine, vitamin Ble, and total antioxidant status in vegetarians. Clin Chem 47:1094-1101 22. Bissoli L, Di Francesco V, Ballarin A, Mandragona R, Trespidi R, Brocco G, Caruso B, Bosello O, Zamboni M (2002) Effect of vegetarian diet on homocysteine levels. Ann Nutr Metab 46:73-79 23. Stolzenberg-Solomon RZ, Miller ER II1, Maguire MG, Selhub J, Appel LJ (1999) Association of dietary protein intake and coffee consumption with serum homocysteine concentrations in an older population. Am J Clin Nutr 69:467-475 24. Agudo A, Amiano P, Barcos A, Barricarte A, Beguirist~iin JM, Chirlaque MD, Dorronsoro M, Gonzglez CA, Lasheras C, Martfnez C, Navarro C, Pera G, Quir6s JR, Rodrlguez M, Tormo MJ (1999) Dietary intake of vegetables and fruits among adults in five regions of Spain. EPIC Group of Spain. European Prospective Investigation into Cancer and Nutrition. Eur J Clin Nutr 53:174-180 25. Williams C (1995) Healthy eating: clarifying advice about fruit and vegetables. BMJ 310:1453-1455 26. Kang SS, Wong PW, Malinow MR (1992) Hyperhomocysteinemia as a risk factor for occlusive vascular disease. Annu Rev Nutr 12: 279-298