Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Circulation. Author manuscript; available in PMC 2010 August 25.
Published in final edited form as:
PMCID: PMC2833014

Plasma Vitamin B6 and Risk of Myocardial Infarction in Women

John H Page, MBBS, MSc, ScD,1 Jing Ma, MD, PhD,2 Stephanie E Chiuve, ScD,3 Meir J Stampfer, MD, DrPH,1,2,3,4 Jacob Selhub, PhD,5 JoAnn E Manson, MD, DrPH,1,2,4 and Eric B Rimm, ScD1,2,3



Vitamin B6 is widely involved in amino acid metabolism and is a modulator of several reactions important to cardiovascular health. We prospectively evaluated relationships between fasting plasma levels of vitamin B6, as pyridoxal phosphate (PLP), to subsequent myocardial infarction risk in women. We also evaluated the predictors of fasting plasma concentration of pyridoxal phosphate.


Participants were adult nurses who completed questionnaires, and updated exposures every 2 years since 1976. Subjects for this analysis were selected by a nested case control design. Blood samples were collected between 1989 and 1990. We restricted our analysis to those women who had provided fasting blood samples (≥10 hours since last meal). During follow-up through June 1998, 144 were diagnosed with myocardial infarction (fatal and non-fatal). Cases were matched 1:2 by age, cigarette smoking status, and month and fasting status at the time of blood collection. Conditional logistic regression was used to adjust for potential confounders, including anthropometric factors, dietary intake, and selected biomarkers. Linear regression was used to determine which variables predict fasting total PLP concentration among control women.


Median age at blood collection was 63. Among controls, lower estimated creatinine clearance, plasma total homocysteine and body mass index were statistically significant predictors of higher plasma PLP, as were higher dietary vitamin B6, and folate intake (all P <0.05). Plasma levels of pyridoxal phosphate were inversely associated with risk of MI, the multivariable adjusted rate ratio (RR) between extreme quarters was 0.22 (95% CI 0.09,0.55; Ptrend=0.05). The effect of plasma PLP varied by age. Among women who were aged less than 60 at blood sampling, the RR (95%CI) comparing top vs. bottom quarter was 0.03 (0.002,0.48), whereas among older women the corresponding RR (95%CI) was 0.43 (0.15,1.25).


Fasting plasma concentration of pyridoxal phosphate was inversely associated with MI risk. Plasma PLP is positively correlated with dietary vitamin B6, and is inversely correlated with renal function and body mass index. Future studies are needed to better understand both dietary and non-dietary determinants of plasma and tissue vitamin B6 status, and how these can be optimized to prevent MI and other diseases.

Keywords: vitamins, nutrition, women, myocardial infarction, risk factors


Coronary Artery Disease (CAD) remains the leading cause of mortality among postmenopausal women in the USA 1,2. Several epidemiological studies suggest that vitamin B6 may be important in the prevention of CAD3,4. Few studies have adequately assessed the role of plasma levels of vitamin B6 in relation to MI risk. Vitamin B6 is widely involved in one-carbon transfer5,6 and is a cofactor for more than 100 human enzymes7, that are essential in amino acid metabolism (including the homocysteine-cystathionine-cysteine conversion), carbohydrate and lipid metabolism, and neurotransmitter production. In circulation, it mainly exists as pyridoxal 5′ phosphate (PLP). In addition to its role in macronutrient metabolism, vitamin B6 is important in steroid receptor interactions8, immune response9, and mineral transport across cell membranes10. One published study found that plasma concentrations of PLP are inversely associated with C-reactive protein concentration, independent of plasma total homocysteine. Another epidemiological study has demonstrated that dietary vitamin B6 is inversely related to risk of myocardial infarction (MI)3, and others have suggested an inverse relationship between plasma PLP and CAD risk2,11,12.

Apart from dietary intake plasma concentration of PLP likely depends on variability in absorption by the gastrointestinal tract, its uptake in the liver, its activation and de-activation by various metabolic processes, and its excretion by the kidneys. These factors have so far not been systematically studied in epidemiological studies. Plasma concentration of PLP is sensitive to feeding status13 and changes after glucose ingestion14.

In this paper, using prospectively collected samples, selected using a nested case control design within the Nurses’ Health Study cohort, we seek to answer two primary questions: 1) What are the major determinants of the fasting plasma concentration of PLP in predominantly postmenopausal women?; and 2) What is the relationship between plasma levels of PLP, in the fasting state, on subsequent risk of myocardial infarction in predominantly postmenopausal women? An additional aim of this research is to determine the extent to which this relationship is independent of dietary pyridoxine intake. Further, we explored whether the association varied by fasting homocysteine concentration, renal function, body mass index (BMI), or age at blood sampling.


Study Population

The Nurses’ Health Study cohort was established in 1976 when 121,700 female registered nurses, 30–55 years of age, completed and returned a mailed questionnaire in order to study the relationship between diet and lifestyle and subsequent disease. The cohort continues to be followed every 2 years by questionnaire to update exposure status and to identify cases of newly diagnosed disease. Data have been collected on many coronary artery disease risk factors, including height, weight, cigarette smoking, alcohol use, physical activity, age at menopause, post-menopausal hormone use, diagnosis of hypertension and diabetes mellitus, history of aspirin use, and parental family history of myocardial infarction. Body mass index was calculated by dividing the most recent weight prior to blood collection by the square of height reported in 1976.

Dietary information was collected using food frequency questionnaires that had been completed by the participants in 1980, 1984, and 1986. These questionnaires assessed the average consumption of a specific amount of each food during the past year, and it allowed nine frequency responses, ranging from “never” to “six or more times per day.” Nutrient intake per day was calculated by multiplying the frequency response by the nutrient content of the specified portion sizes. Total alcohol intake per day was calculated as the sum of the alcohol content contributed from beer, wine, and liquor, assuming 12.8 g of ethanol for 360 mL (12 oz) of beer, 11.0 g for 120 mL (4 oz) of wine, and 14.0 g for 45 mL (1.5 oz) of liquor. Duration, brand, and type of multivitamin supplement use were updated in the biennial questionnaires or the food frequency questionnaires, and a comprehensive database on the vitamin content of the multivitamin preparations was developed. The validity and reliability of the food frequency questionnaires used in the Nurses’ Health Study have been described previously1517.

From 1989 through 1990, blood samples were collected from 32,826 cohort members (27% of the original cohort) who were then 43–69 years of age. Details regarding the blood collection methods have been published previously18. Briefly, each woman arranged to have her blood drawn and then shipped, via overnight courier with an ice pack, to our laboratory, where it was processed and separated into plasma, red blood cells, and white blood cell components. Within 24–36 hours of being drawn, 97% of the samples were received in our laboratory. Since collection, samples have been archived at −130 °C or colder in continuously monitored liquid nitrogen freezers. As of 1998, follow-up of women who provided blood samples was 99.8% complete.

We included as cases, women who provided a blood sample after fasting 10 hours or more, reported no myocardial infarction before blood collection, and who were diagnosed with myocardial infarction after blood collection but before June 1, 1998. Overall, 144 cases of myocardial infarction (21 fatal) with adequate plasma were identified. For all cases of myocardial infarction, we requested hospital records (we could not obtain records for three cases, but all cases were retained for analysis). Myocardial infarction was classified as confirmed if symptoms met the criteria of the World Health Organization (typical symptoms and either diagnostic electrocardiographic changes or elevated cardiac enzymes).

The median (10th to 90th percentiles) time from blood collection to diagnosis was 48 months (16, 91). Two control subjects were selected at random, matched to the case subject by age (±2 years), cigarette smoking status (current, past, and never smoker), month of blood collection, and fasting status at the time of blood collection (≥10 hours since a meal). 81% percent of control matches were exact; the most relaxed match was within ±3 years of age, and within ±6 months of blood collection. The matched sets were analyzed together. The study complies with the Declaration of Helsinki, and was approved by the Committee on the Use of Human Subjects in Research at the Brigham and Women’s Hospital. All subjects gave their informed consent in order to participate.

Since exposure and covariate data were collected prospectively, they are independent of case control status.

Laboratory Analyses

Plasma levels of pyridoxal phosphate (PLP) were determined by an enzymatic procedure using radioactive tyrosine and the apo-enzyme tyrosine decarboxylase.19,20 During the assay process, we interspersed replicate plasma samples, which were labeled to preclude their identification by the laboratory, to assess laboratory precision. The intra-assay coefficients of variation for plasma pyridoxal phosphate was 8.9%.

Creatinine was measured using a modified Jaffe method. The intra-assay coefficient of variation was 12.8%. We used a modified version of the Cockcroft-Gault formula to estimate creatinine clearance2123. This formula is based on fat-free body mass and has the advantage of attenuating the overestimation of creatinine clearance (CrCl) in obese individuals with the Cockcroft-Gault equation while providing similar results in average-weight women22. The formula for women is (146 − age) ×[(0.287 × weight) + (9.74 × height2)]/(60 × creatinine), where age is measured in years, weight is measured in kilograms, height is measured in meters, and creatinine is measured in mg/dl; the units for CrCl are ml/min. This formula has been validated compared with measured CrCl22.

Plasma homocysteine concentrations were measured with high performance liquid chromatography using fluorescence detection at the Jean Mayer United States Department of Agriculture Human Nutrition Research Center on Aging at Tufts University19,24. The intra-assay coefficients of variation for plasma total homocysteine, was 7.5%. Total cholesterol was measured enzymatically25 with an intra-assay coefficient of variation of 1.7%. High-density lipoprotein cholesterol (HDL) was measured using Hitachi 911 analyzer26, with an intra-assay coefficient of variation of 2.5%. C-reactive protein (CRP) was determined with a high sensitivity immunoturbidimetric assay on a Hitachi 911 analyzer, with an intra-assay coefficient of variation of 1.4%.

All matched case–control blood samples were handled identically and together, handled in the same batch, and assayed in the same analytical run. All assays were conducted by the laboratory personnel without knowledge of the case–control status of the samples.

Data Analysis

Medians and percentiles were calculated for continuous variables in case women and in controls as two separate groups.

Among the samples provided by controls, we conducted linear regression analyses to evaluate the relationship of fasting PLP concentration to quarters of the continuous variables and indicator functions of the categorical variables while controlling for an indicator function of blood analysis batch. The continuous variables were age, plasma total homocysteine, CrCl, plasma CRP, HDL-to-total cholesterol, physical activity, body mass index, and dietary intake of energy-adjusted total protein, total caloric intake, vitamin B6, folate, vitamin B12, vitamin B1, vitamin B2, vitamin C, magnesium, and alcohol. The categorical variables were menopausal status (pre-menopausal, post-menopausal, unknown menopausal status), current post-menopausal hormone use at blood collection, a personal history of being diagnosed with diabetes mellitus, a personal history of being diagnosed with hypertension, parental history of myocardial infarction prior to age 60, and cigarette smoking status (never smoker, past smoker, current smoker <15 cigarettes per day, 15–24 cigarettes per day, and greater than 24 cigarettes per day). A modified backwards stepwise procedure was used to determine which variables should remain in the model as judged by statistical significance (p value<0.05) of the F-test27 (p value<0.05). The initial model included all of the non-biomarker variables listed above and creatinine clearance; and variables were removed individually in order of largest to smallest p-value on the F-test until the only variables remaining were statistical significant predictors. All three indicators for each continuous variable were considered one variable. The previously removed variables were then individually returned to determine statistical significance in the modified model, at which point, all previously removed variables would be returned and backwards elimination repeated. This procedure was repeated iteratively until no added or removed variable resulted in statistically significant changes to the model, the “final model”. The relationship of biomarkers to plasma PLP were determined in the multivariable model by individually adding the respective variable to this “final model”.

Conditional logistic regression was used to estimate odds ratios between quarters of plasma PLP concentration and risk of MI. These odds ratios were taken as direct estimates28 of rate ratios (RRs) and 95% confidence intervals (CIs)29. We additionally controlled for potential confounders that were not part of the matching scheme. To control more appropriately for confounding and other co-variation by continuous variables, natural cubic splines (28, 29) with four degrees of freedom (three when four was not feasible) were used to smooth the relationships with the log-odds of myocardial infarction. Physical activity in metabolic equivalents (MET) – hours per week in 1988 was entered into the regression models as indicator functions of the quarters since natural cubic splines did not determine a good fit for this variable. Indicator functions of the categorical covariates were used. We estimated the following models: i) a model that controls for matching factors only; ii) a model that controls for conventional non-biomarker risk factors for MI (considered the final model); iii) a model that additionally controls for creatinine clearance, the ratio of plasma total to HDL-cholesterol, and plasma high-sensitivity C-reactive protein, iv) a model that additionally adjusts for plasma concentrations of total homocysteine; and v) A model that controls for predictors (including dietary pyridoxine and folate) of plasma PLP. There were very few missing values in the covariates physical activity, CRP, creatinine, and HDL to total cholesterol ratio (all less than 4%), and thus medians were used for imputation. Tests for interaction were done by including product terms between indicator functions of halves of the respective variable and indicator functions of quarters of PLP.

We conducted tests for trend by modeling the hormone level as a linear continuous covariate and calculating a Wald statistic29. All P values were based on two-sided tests. The software used for statistical analysis were SAS release 9.130 and S-plus version 831.


Table 1 shows the distribution of risk factors for myocardial infarction at the time of blood sampling among case women and matched controls. As expected, average body mass index, total cholesterol, CRP, histories of diabetes mellitus, hypertension, and having had a parent with early myocardial infarction were all higher in case women. Similarly, the levels of physical activity, alcohol consumption and plasma HDL cholesterol were lower in those who later developed myocardial infarction, compared with control women.

Table 1
Distribution of covariates in MI case and matched control participants*

Plasma concentrations of PLP were inversely correlated with estimated creatinine clearance and body mass index, and directly correlated with dietary intake of vitamin B6 (table 2,,3).3). Our model that included dietary B6, folate, body mass index, estimated creatinine clearance explained 28.6% of the variation in plasma PLP. After adjustment for estimated creatinine clearance, body mass index, dietary intake of vitamin B6, and folate, plasma PLP was also inversely correlated with plasma homocysteine (table 3). Plasma concentration of PLP was not predicted by age, menopausal status, use of postmenopausal hormones, protein intake, total caloric intake, physical activity, alcohol consumption, plasma total to HDL cholesterol ratio, or by history of hypertension, diabetes mellitus, cigarette smoking, or by parental history of MI before age 60 (table 3).

Table 2
Spearman’s correlation coefficients (P value) between continuous covariates and batch adjusted residuals of loge(Pyridoxal Phosphate) among control study participants who provided blood after fasting for 10 hours or more.
Table 3
Predictors of Loge(Plasma Pyridoxal Phosphate) among control participants

Fasting concentrations of PLP were significantly inversely associated with subsequent risk of MI [RR (95%CI) for top vs. bottom quarter 0.22 (0.09,0.55), p value for linear trend 0.047] (Table 4, Figure 1). The relationship between log rate of MI and plasma PLP is more linear than table 4 suggests. Compared to plasma PLP <20 pmol/mL, concentrations of 20–39,40–59,60–79,80–99,100–119,≥120 pmol/mL were associated with MI with estimated rate ratios of 0.4, 0.5, 0.3, 0.3, 0.1, and 0.1 respectively. The shape and magnitude of the relationships changed little with addition of plasma homocysteine or dietary determinants of plasma PLP (dietary pyridoxine and folate).

Figure 1
Smoothed relationship between fasting plasma levels of Vitamin B6 as Pyridoxal 5′ Phosphate and the log odds of myocardial infarction* in predominantly post-menopausal women in the Nurses’ Health Study
Table 4
Odds ratio (OR) of myocardial infarction and 95% confidence interval (CI) by quarter of fasting plasma Pyridoxal Phosphate, in the Nurses’ Health Study.

The relationship between pyridoxal phosphate and risk of MI appeared to vary by age ( χ3df2=11.2, p=0.011). The effect of plasma PLP was stronger among women who were aged less than 60 at blood sampling [RR (95%CI) for top vs. bottom quarter 0.03 (0.002,0.48)], compared to older women [RR (95%CI) for top vs. bottom quarter 0.43 (0.15,1.25)]. There was no statistically significant effect modification by halves of body mass index, estimated creatinine clearance, or by plasma total homocysteine.


The questions addressed by the present study were whether fasting concentrations of plasma pyridoxal phosphate predict later risk of MI among predominantly post-menopausal women, and to determine the major epidemiological predictors of the plasma concentration. The main finding of the study is that a low plasma concentration of pyridoxal phosphate is associated with increased rate of MI among predominantly post-menopausal women. This was demonstrated by the statistically and clinically significant lower risk of MI associated with higher PLP concentration (table 4, figure 1). Although there was a steep decline in MI risk for PLP concentrations above 20 pmol/mL relative to lower, risk continued to decline with higher levels. The effect was more pronounced in younger relative to older women. We also found that estimated creatinine clearance, dietary vitamin B6 intake, dietary folate intake, and body mass index were statistically significant predictors of fasting plasma levels of PLP. Fasting concentration of PLP was also negatively correlated with total homocysteine.

The finding that plasma concentrations of PLP is inversely related to risk of MI is consistent with the knowledge of the widespread role of vitamin B6 in protein metabolism, including homocysteine metabolism, and its likely effects on mineral transport across cell membranes10. This is also in agreement with the results of other epidemiological studies that have also shown that dietary vitamin B6 and plasma PLP are inversely related to risk of MI3 and to the risk of coronary artery disease2,12. This is also accordantwith the role of pyridoxal phosphate as a cofactor in the conversion of homocysteine to cysteine in the transulfuration pathway. Although it was not statistically significantly related to plasma levels of CRP in this study, PLP was inversely correlated with CRP in one previous study32.

The finding that the plasma concentration of PLP is positively correlated with dietary intake of vitamin B6 is consistent with a number of studies that demonstrated that dietary supplementation of vitamin B6 results in an increase in plasma concentration3336. The positive association between dietary folate and plasma PLP concentration may reflect a vitamin B6 - sparing effect of increased folate, by its reduction of the flux of homocysteine via the transulfuration pathway. The negative correlation between estimated creatinine clearance and plasma levels of PLP is consistent with the fact that vitamin B6 is almost completely excreted via the kidneys37. It is unclear whether the negative correlation between BMI and plasma PLP is due to a causal effect of BMI on vitamin B6. Although animal studies suggest that high protein diets increase requirements for vitamin B6, we did not find a relationship between protein intake and plasma PLP. Chronic alcohol use results in increased alkaline phosphatase38, and thus may affect concentration of plasma PLP39, but we did not find significant relationships in this dataset. Older individuals have been found to have lower PLP concentrations relative to the young in another study40, but this was not the case in this study. This may be related to the fact all the women in this study were aged above 40 and thus there was not enough variation in age. The fact that our linear regression model explained only 28.6% of the variation in plasma PLP reinforces the fact that there are important determinants not accounted for by the measured variables. This also helps to explain our finding that the relationship between plasma PLP and MI risk was little affected by including the measured determinants of PLP concentration such as dietary pyridoxine in the regression model (table 4).

To our knowledge, this is the first prospective study that has looked specifically at the relationship of fasting levels of PLP with MI in predominantly post-menopausal women. Major strengths of this study are its prospective design, and its careful control of confounding factors. In addition, we were able to reduce measurement error in PLP as well as other plasma covariates by focusing on those women who had provided blood after fasting for 10 hours or more. However, given the observational nature of the study, we cannot be sure that all unknown confounders have been adequately controlled. Further, inference from our linear regression models are limited in that the estimation was performed in a non-random selection of women, since the controls were matched to the MI cases.

In summary, our investigation revealed that lower fasting concentration of pyridoxal phosphate is significantly associated with increased risk of myocardial infarction in predominantly post-menopausal women, a relationship that may be causal. We also confirmed that plasma PLP is positively correlated with dietary vitamin B6, and found that it is inversely correlated with a measure of renal function and with body mass index. Future studies are needed to better understand both dietary and non-dietary determinants of plasma and tissue vitamin B6 status, and how these can be optimized to prevent disease.


1. Selhub J. Homocysteine metabolism. Annu Rev Nutr. 1999;19:217–46. [PubMed]
2. Folsom AR, Nieto FJ, McGovern PG, et al. Prospective study of coronary heart disease incidence in relation to fasting total homocysteine, related genetic polymorphisms, and B vitamins: the Atherosclerosis Risk in Communities (ARIC) study. Circulation. 1998;98(3):204–10. [PubMed]
3. Rimm EB, Willett WC, Hu FB, et al. Folate and vitamin B6 from diet and supplements in relation to risk of coronary heart disease among women. Jama. 1998;279(5):359–64. [PubMed]
4. Schnyder G, Roffi M, Flammer Y, Pin R, Hess OM. Effect of homocysteine-lowering therapy with folic acid, vitamin B12, and vitamin B6 on clinical outcome after percutaneous coronary intervention: the Swiss Heart study: a randomized controlled trial. Jama. 2002;288(8):973–9. [PubMed]
5. Selhub J. Folate, vitamin B12 and vitamin B6 and one carbon metabolism. J Nutr Health Aging. 2002;6(1):39–42. [PubMed]
6. Smith CM, Marks AD, Lieberman M, Marks DB, Marks DB. Marks’ basic medical biochemistry: a clinical approach. 2. Philadelphia: Lippincott Williams & Wilkins; 2005.
7. Clayton PT. B6-responsive disorders: a model of vitamin dependency. J Inherit Metab Dis. 2006;29(2–3):317–26. [PubMed]
8. Tully DB, Allgood VE, Cidlowski JA. Modulation of steroid receptor-mediated gene expression by vitamin B6. Faseb J. 1994;8(3):343–9. [PubMed]
9. Salhany JM, Schopfer LM. Pyridoxal 5′-phosphate binds specifically to soluble CD4 protein, the HIV-1 receptor. Implications for AIDS therapy. J Biol Chem. 1993;268(11):7643–5. [PubMed]
10. Abraham GE, Schwartz UD, Lubran MM. Effect of vitamin B-6 on plasma and red blood cell magnesium levels in premenopausal women. Ann Clin Lab Sci. 1981;11(4):333–6. [PubMed]
11. Chasan-Taber L, Selhub J, Rosenberg IH, et al. A prospective study of folate and vitamin B6 and risk of myocardial infarction in US physicians. J Am Coll Nutr. 1996;15(2):136–43. [PubMed]
12. Friso S, Girelli D, Martinelli N, et al. Low plasma vitamin B-6 concentrations and modulation of coronary artery disease risk. Am J Clin Nutr. 2004;79(6):992–8. [PubMed]
13. Vermaak WJ, Barnard HC, Potgieter GM, Marx JD. Plasma pyridoxal-5′-phosphate levels in myocardial infarction. S Afr Med J. 1986;70(4):195–6. [PubMed]
14. Leklem JE, Hollenbeck CB. Acute ingestion of glucose decreases plasma pyridoxal 5′-phosphate and total vitamin B-6 concentration. Am J Clin Nutr. 1990;51(5):832–6. [PubMed]
15. Willett W. Nutritional epidemiology. 2. New York: Oxford University Press; 1998.
16. Willett WC, Sampson L, Browne ML, et al. The use of a self-administered questionnaire to assess diet four years in the past. Am J Epidemiol. 1988;127(1):188–99. [PubMed]
17. Willett WC, Sampson L, Stampfer MJ, et al. Reproducibility and validity of a semiquantitative food frequency questionnaire. Am J Epidemiol. 1985;122(1):51–65. [PubMed]
18. Hankinson SE, Willett WC, Manson JE, et al. Alcohol, height, and adiposity in relation to estrogen and prolactin levels in postmenopausal women. J Natl Cancer Inst. 1995;87(17):1297–302. [PubMed]
19. Wu K, Helzlsouer KJ, Comstock GW, Hoffman SC, Nadeau MR, Selhub J. A prospective study on folate, B12, and pyridoxal 5′-phosphate (B6) and breast cancer. Cancer Epidemiol Biomarkers Prev. 1999;8(3):209–17. [PubMed]
20. Camp VM, Chipponi J, Faraj BA. Radioenzymatic assay for direct measurement of plasma pyridoxal 5′-phosphate. Clin Chem. 1983;29(4):642–4. [PubMed]
21. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976;16(1):31–41. [PubMed]
22. Salazar DE, Corcoran GB. Predicting creatinine clearance and renal drug clearance in obese patients from estimated fat-free body mass. Am J Med. 1988;84(6):1053–60. [PubMed]
23. Knight EL, Rimm EB, Pai JK, et al. Kidney dysfunction, inflammation, and coronary events: a prospective study. J Am Soc Nephrol. 2004;15(7):1897–903. [PubMed]
24. Araki A, Sako Y. Determination of free and total homocysteine in human plasma by high-performance liquid chromatography with fluorescence detection. J Chromatogr. 1987;422:43–52. [PubMed]
25. Allain CC, Poon LS, Chan CS, Richmond W, Fu PC. Enzymatic determination of total serum cholesterol. Clin Chem. 1974;20(4):470–5. [PubMed]
26. Sugiuchi H, Uji Y, Okabe H, et al. Direct measurement of high-density lipoprotein cholesterol in serum with polyethylene glycol-modified enzymes and sulfated alpha-cyclodextrin. Clin Chem. 1995;41(5):717–23. [PubMed]
27. Armitage P, Berry G, Matthews JNS. Statistical methods in medical research. 4. Malden, MA: Blackwell Science; 2001.
28. Rothman KJ, Greenland S. Modern epidemiology. 2. Philadelphia, PA: Lippincott-Raven; 1998.
29. Lachin JM. Biostatistical methods: the assessment of relative risks. New York: Wiley; 2000.
30. SAS/GRAPH Software [program] Cary, NC: SAS Institute; 2004. Version 9.
31. S-PLUS (Version 8) Insightful Corporation; Seattle, WA: 2007.
32. Friso S, Jacques PF, Wilson PW, Rosenberg IH, Selhub J. Low circulating vitamin B(6) is associated with elevation of the inflammation marker C-reactive protein independently of plasma homocysteine levels. Circulation. 2001;103(23):2788–91. [PubMed]
33. Chiang EP, Selhub J, Bagley PJ, Dallal G, Roubenoff R. Pyridoxine supplementation corrects vitamin B6 deficiency but does not improve inflammation in patients with rheumatoid arthritis. Arthritis Res Ther. 2005;7(6):R1404–11. [PMC free article] [PubMed]
34. Bor MV, Refsum H, Bisp MR, et al. Plasma vitamin B6 vitamers before and after oral vitamin B6 treatment: a randomized placebo-controlled study. Clin Chem. 2003;49(1):155–61. [PubMed]
35. Lonn E, Yusuf S, Arnold MJ, et al. Homocysteine lowering with folic acid and B vitamins in vascular disease. N Engl J Med. 2006;354(15):1567–77. [PubMed]
36. Bates CJ, Pentieva KD, Prentice A, Mansoor MA, Finch S. Plasma pyridoxal phosphate and pyridoxic acid and their relationship to plasma homocysteine in a representative sample of British men and women aged 65 years and over. Br J Nutr. 1999;81(3):191–201. [PubMed]
37. Lui A, Lumeng L, Aronoff GR, Li TK. Relationship between body store of vitamin B6 and plasma pyridoxal-P clearance: metabolic balance studies in humans. J Lab Clin Med. 1985;106(5):491–7. [PubMed]
38. Nishmura M, Teschke R. Effect of chronic alcohol consumption on the activities of liver plasma membrane enzymes: gamma-glutamyltransferase, alkaline phosphatase and 5′-nucleotidase. Biochem Pharmacol. 1982;31(3):377–81. [PubMed]
39. Fonda ML, Brown SG, Pendleton MW. Concentration of vitamin B6 and activities of enzymes of B6 metabolism in the blood of alcoholic and nonalcoholic men. Alcohol Clin Exp Res. 1989;13(6):804–9. [PubMed]
40. Bates CJ, Pentieva KD, Prentice A. An appraisal of vitamin B6 status indices and associated confounders, in young people aged 4–18 years and in people aged 65 years and over, in two national British surveys. Public Health Nutr. 1999;2(4):529–35. [PubMed]