AJCN EB Program 2010 Early Registration
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sun, Q.
Right arrow Articles by Hu, F. B
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sun, Q.
Right arrow Articles by Hu, F. B
Agricola
Right arrow Articles by Sun, Q.
Right arrow Articles by Hu, F. B
American Journal of Clinical Nutrition, Vol. 86, No. 4, 929-937, October 2007
© 2007 American Society for Nutrition

ORIGINAL RESEARCH COMMUNICATION

Plasma and erythrocyte biomarkers of dairy fat intake and risk of ischemic heart disease1,2,3

Qi Sun, Jing Ma, Hannia Campos and Frank B Hu

1 From the Departments of Nutrition (QS, HC, and FBH) and of Epidemiology (QS and FBH), Harvard School of Public Health, Boston, MA, and the Channing Laboratory, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, MA (JM and FBH)

2 Supported by research grants CA49449, CA42182, HL24074, HL34594, and CA87969 from the National Institutes of Health and by an American Heart Association Established Investigator Award (to FBH).

3 Reprints not available. Address correspondence to FB Hu, Department of Nutrition, Harvard School of Public Health, 665 Huntington Avenue, Boston, MA 02115. E-mail: frank.hu{at}channing.harvard.edu.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: The relation between dairy product intake and the risk of ischemic heart disease (IHD) remains controversial.

Objective: We aimed to explore biomarkers of dairy fat intake in plasma and erythrocytes and to assess the hypothesis that higher concentrations of these biomarkers are associated with a greater risk of IHD in US women.

Design: Among 32 826 participants in the Nurses' Health Study who provided blood samples in 1989–1990, 166 incident cases of IHD were ascertained between baseline and 1996. These cases were matched with 327 controls for age, smoking, fasting status, and date of blood drawing.

Results: Among controls, correlation coefficients between average dairy fat intake in 1986–1990 and 15:0 and trans 16:1n–7 content were 0.36 and 0.30 for plasma and 0.30 and 0.32 for erythrocytes, respectively. In multivariate analyses, with control for age, smoking, and other risk factors of IHD, women with higher plasma concentrations of 15:0 had a significantly higher risk of IHD. The multivariate-adjusted relative risks (95% CI) from the lowest to highest tertile of 15:0 concentrations in plasma were 1.0 (reference), 2.18 (1.20, 3.98), and 2.36 (1.16, 4.78) (P for trend = 0.03). Associations for other biomarkers were not significant.

Conclusions: Plasma and erythrocyte contents of 15:0 and trans 16:1n–7 can be used as biomarkers of dairy fat intake. These data suggest that a high intake of dairy fat is associated with a greater risk of IHD.

Key Words: Dairy fat • plasma • blood cells • coronary disease • women


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Consumption of whole milk and other high-fat dairy products, such as cheese and butter, has been postulated to be associated with the risk of ischemic heart disease (IHD) because of the high saturated fat content (1). Although strong positive associations between the intake of dairy products with high fat content and IHD mortality have been indicated in ecologic studies (2–4), prospective cohort studies showed mixed results for these associations (5). Dairy products are complex with respect to their nutrient composition. Besides fat, dairy products also contain many nutrients, such as calcium, that may play different roles in the development of IHD (1). Dairy fat intake could be a more specific risk factor for IHD, because it is the fat content of dairy products that is suggested to be detrimental to coronary health. Few epidemiologic studies of the association between dairy fat intake and the risk of IHD have been published (6, 7). Measuring dairy fat intake can be challenging to epidemiologic studies that use food-frequency questionnaires (FFQs) and similar methods to estimate nutrient intake, because dairy fat exists in many foods, and fat content varies within these foods (8).

Biological markers are believed to be more objective than FFQs, because biomarkers do not rely on the precision of food composition databases and self-reports or the appropriateness of FFQ items (9, 10). So far, only a few studies have been conducted to determine valid biomarkers of dairy fat intake. The preliminary results showed that fatty acids (FAs) with odd-number carbon chains—ie, 15:0 and 17:0—in human adipose tissue may be used as biomarkers of dairy fat intake (11, 12). The possibility that other FAs exclusively existing in ruminant fats, such as certain trans isomers, could be used as valid biomarkers has not been assessed. Even fewer studies have been conducted to explore the associations between these biomarkers and the risk of IHD. In the current study, we evaluated the validity of odd-numbered FAs and explored other FAs in plasma and erythrocytes as biomarkers of dairy fat intake. We further examined the associations between these biomarkers and the risk of IHD in a prospective case-control study nested in the Nurses' Health Study cohort.


   SUBJECTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Study population and blood collection
In 1976, >120 000 female registered nurses aged 30–55 y were enrolled in the Nurses' Health Study. In 1989–1990, blood samples were provided by 32 826 participants. Ninety-seven percent of the samples arrived within 24 h after blood drawing. Immediately on arrival, the samples were centrifuged (1200 x g, 15 min, room temperature) and divided into aliquots of plasma, erythrocytes, and buffy coat fractions, which were then stored in liquid nitrogen freezers at –130 °C or colder until analysis.

Participants who provided a blood sample and who were free of diagnosed cancer and cardiovascular disease at the time of blood drawing were eligible for the current study. Among these participants, 167 cases of nonfatal myocardial infarction (MI) or IHD death were newly diagnosed between the time of blood drawing and June 1996. For each IHD case, 2 control subjects were randomly selected from the participants who were free of IHD at the time of the case's IHD diagnosis. Cases and controls were matched for age (±1 y), smoking status (never, past, or current), fasting status at blood drawing (fasting for 10 h or not fasting), and the date of the blood drawing. One case and 7 controls had missing data on FA concentrations were missing for 1 case and 7 controls because of contamination, dilution, or loss of blood samples. After exclusion of these participants, 166 cases and 327 controls were available for analysis.

The study protocol was approved by the Institutional Review Board of the Brigham and Women's Hospital and the Human Subjects Committee Review Board of Harvard School of Public Health.

Assessment of ischemic heart disease
Nurses who reported a nonfatal MI in the biennial follow-up questionnaires were asked to provide their medical records, which were then reviewed by study physicians who were blinded as to the exposure status of participants. Telephone interviews were conducted if medical records were not provided by the nurses. Nonfatal MI was confirmed if the World Health Organization criteria, which require typical symptoms plus either diagnostic electrocardiographic findings or elevated cardiac enzyme concentrations, were met (13). We identified deaths among the Nurses' Health Study participants through reports from the next of kin or postal authorities or by searching the National Death Index. At least 98% of deaths were identified (14). IHD deaths were identified if IHD was listed as the cause of death in hospital records, in autopsy reports, or on death certificates. IHD deaths were then confirmed by a previous report of IHD and by a determination that another, more apparent or plausible cause of death was not stated. We included only confirmed IHD cases in the current analysis. Total IHD was defined as nonfatal MI plus fatal IHD.

Assessment of diet
Food and nutrient consumption assessed by semiquantitative FFQs in 1986 and 1990 was primarily used in the present study to represent the usual intake. A detailed description of the validity of the FFQs used in the Nurses' Health Study and a discussion of the FFQs' validity were published elsewhere (10, 15). Briefly, these FFQs inquired about the consumption of major dairy foods and other foods in the previous year. For each food item, the frequency of the consumption of that food in a standard portion size was asked. There were 9 possible coding responses, ranging from "never or <1 time/mo" to "≥6 times/d." The major dairy foods inquired about in these FFQs included skim milk, whole milk, cream, sour cream, sherbet, yogurt, ice cream, cheese, and butter. Dairy fat intake was calculated by multiplying the frequency of consumption of dairy foods by the fat content in the specified amount. Then, the contributions across all foods were summed to represent total dairy fat intake, taking into account the dairy fat intake from other foods in which milk was an ingredient, such as milk chocolate, white bread, home-made bakery products, and other foods. The food composition database was primarily based on US Department of Agriculture publications (16), which were supplemented by data from other publications and individual laboratories (17–19). The intakes of other FAs were also derived by using the same method. Fourteen participants had missing dietary intake. Therefore, when we evaluated biomarkers in relation to dietary intake, we excluded these participants.

Plasma and erythrocyte fatty acid analysis
FA concentrations in plasma and erythrocytes were analyzed by gas-liquid chromatography in 2000 and 2002. A detailed description of the laboratory process has been published elsewhere (12). The concentration of each FA was expressed as a percentage of the total FA content in plasma or erythrocytes. Technicians and laboratory personnel were blinded as to the disease status of the participants. Blood samples from each case-control triplet was shipped in the same batch and analyzed in the same run. Each triplet's samples were assayed by the same technicians in a random sequence and under identical conditions. Laboratory control samples were tested along with the case-control samples.

We quantified 37 FAs in plasma and erythrocytes that had meaningful (ie, >0.01% of total) concentrations. In the current study, we reported FAs that are derived exclusively from ruminant sources and that cannot be synthesized by human body (ie, 15:0, 17:0, and trans 16:1n–7) (20, 21) and other FAs that exist in substantial amounts in dairy fat (ie, 14:0, 16:0, and 18:0) (6). The major trans isomer existing in dairy fat is trans 18:1n–7 (vaccenic acid), and it therefore was also included in the analysis. The intraassay CV was 11.1% (plasma) or 19.3% (erythrocytes) for 14:0 (myristic acid); 4.3% (plasma) or 8.8% (erythrocytes) for 15:0; 1.4% (plasma) or 3.3% (erythrocytes) for 16:0 (palmitic acid); 2.3% (plasma) or 7.2% (erythrocytes) for 17:0; 2.1% (plasma) or 3.4% (erythrocytes) for 18:0 (stearic acid); 3.3% (plasma) or 6.8% (erythrocytes) for trans 16:1n–7; and 10.5% (plasma) or 7.0% (erythrocytes) for trans 18:1n–7. Although 12:0 (lauric acid) also mainly comes from dairy foods (6), we did not report this FA because of the high incidence of laboratory measurement errors in relation to it.

Stability of fatty acid contents in plasma and erythrocytes
One study has shown that FAs in serum stored at a low temperature (–80 °C) for 7–12 y showed minimal degradation over time (22). Among the controls of the current study, the proportions of monounsaturated and polyunsaturated FAs were 18.63% (plasma) and 13.25% (erythrocytes) for oleic acid, 30.52% (plasma) and 13.64% (erythrocytes) for linoleic acid, 0.50% (plasma) and 0.18% (erythrocytes) for {alpha}-linolenic acid, 7.80% (plasma) and 14.61% (erythrocytes) for arachidonic acid, 0.49% (plasma) and 1.15% for eicosapentaenoic acid, and 1.57% (plasma) and 3.72% (erythrocytes) for docosahexaenoic acid. The contents of these FAs were similar to those observed in previous studies with shorter durations of storage (23, 24), which indicates that the FA contents were stable after long-term storage at a very low temperature (–130 °C).

Assessment of covariates
Information on the occurrence of disease, medical history, and major lifestyle IHD risk factors of the participants was collected at baseline and updated biennially thereafter with the use of follow-up questionnaires. We primarily used covariates assessed by the 1990 follow-up questionnaire in the current study. To minimize missing data, we carried forward the values obtained in previous questionnaires for values missing in the 1990 questionnaires.

Plasma total and HDL-cholesterol concentrations were measured enzymatically. LDL cholesterol was measured by using a homogenous direct method. A detailed description of the laboratory methods and procedures was published elsewhere (25).

Statistical analysis
For continuous variables with normal distribution, means (±SDs) were calculated and Student's t tests were used to evaluate the significance of difference between cases and controls. Medians and interquartile ranges and Wilcoxon's rank-sum tests were used for variables with nonnormal distribution. For categorical variables, proportions were calculated, and chi-square tests of significance were used.

Dairy fat intake measured by the FFQs was expressed as a percentage of total fat intake. We determined the biomarkers of dairy fat intake by using data on the controls only, because controls were representative of the source population from which the cases were drawn. Spearman partial correlation coefficients (r) were calculated to determine relations between FA concentrations and dairy fat intake after adjustment for age at blood collection (y), smoking status (never smoker, past smoker, currently smoke 1–14 cigarettes/d, currently smoke 15–24 cigarettes/d, or currently smoke ≥25 cigarettes/d), body mass index (BMI; in kg/m2), current weight (kg), fasting status (yes or no), postmenopausal status (yes or no), postmenopausal hormone use (never, past, or current), and the periods during which the blood samples were assayed. We used both dairy fat intake measured by the 1990 FFQ and the average intake estimated by averaging the 1986 and 1990 FFQ measurements to represent long-term dairy fat intake.

We used multivariate linear regression to detect linear trends of the biomarker concentrations across deciles of average dairy fat intake of controls in 1986–1990. Least-squares means of FA concentrations were calculated for each decile of dairy fat intake after adjustment for total energy intake, age, smoking status, BMI, and other covariates. To allow for deviance from the assumption of normal distribution of dependent variables, we calculated robust estimators of variance for these means (26). Median values of each decile of intake were entered into the models as a continuous variable to calculate P values for linear trend.

We used conditional logistic regressions to estimate the relative risks (RRs) of IHD associated with the biomarkers of dairy fat intake. In nested case-control studies, odds ratios derived from conditional logistic regressions are unbiased estimates of hazard ratios or RRs (27). The distribution among control participants was used to determine the cutoffs for tertiles of these biomarkers. Multivariate RRs were adjusted for established risk factors for IHD: ie, BMI (<25, 25–29, or ≥30), physical activity (≤8.1, 8.2–18.5, or ≥18.6 metabolic equivalent tasks), alcohol intake (0, 1–4, 5–14, or ≥15 g/d), use of aspirin (tablets/wk), parental history of MI before age 65 y (yes or no), history of hypertension (presence or absence), history of hypercholesterolemia (presence or absence), history of diabetes (presence or absence), postmenopausal status (yes or no), and postmenopausal hormone use (never, past, or current). Matching factors were taken into account as well. We included 18:2n–6 (linoleic acid) and total trans FAs in the multivariate models because their inclusion in plasma changed the RR of ≥1 category of 15:0 tertiles in plasma by >15%.

P values for linear trend were calculated by entering a continuous score based on the median value in each tertile of FA into the models. All P values were 2-sided; 95% CIs were calculated for RRs and least-squares means. Data were analyzed with the use of SAS software (version 9.1; SAS Institute Inc, Cary, NC).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
We evaluated the validity of plasma and erythrocyte FAs as biomarkers of dairy fat intake in controls only. The comparison between controls and the overall Nurses' Health Study cohort with respect to baseline characteristic distribution is shown in Table 1Go. As expected, the controls in the present study did not differ significantly from the overall cohort except in age, which was used to match controls with IHD cases. The FA concentrations in plasma and erythrocytes are described in Table 2Go.


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

  		TABLE 1. Comparison between controls and the entire cohort, the Nurses’ Health Study, 1990

 

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

  		TABLE 2. Fatty acids composition of plasma and erythrocytes and dietary intake of fatty acids measured by a food-frequency questionnaire among controls, the Nurses' Health Study, 19901

 
The Spearman partial correlation coefficients between FA concentrations in plasma and erythrocytes and dairy fat and product intake are shown in Table 3Go. Uniformly, each FA in both blood fractions had a stronger correlation with average dairy fat intake in 1986–1990 than with 1990 intake alone, probably because of the reduction of random measurement errors by averaging multiple assessments. In plasma, 15:0 had the strongest correlation with dairy fat intake; the Spearman correlation coefficient between this FA and average dairy fat intake was 0.36. The second-strongest marker of dairy fat intake in plasma was trans 16:1n–7. In erythrocytes, correlations with dairy fat intake did not differ significantly for 15:0 and trans 16:1n–7; both correlation coefficients were {approx}0.30. Other FAs in plasma or erythrocytes were weakly associated with dairy fat intake. Because we observed reasonable correlations for 15:0 and trans 16:1n–7 in both plasma and erythrocytes, we focused on these FAs thereafter. In previous studies (11, 28–31), 17:0 in blood or human tissues was also suggested as a potential biomarker of dairy fat intake. To be comparable to these studies, we also included 17:0 in further analyses. Plasma 15:0 concentrations were correlated with total dairy product intake less than with dairy fat intake; the Spearman correlation coefficient was 0.29 between this FA in plasma and average dairy product intake in 1986–1990. Results are similar for other FAs. Correlations of these FAs with skim milk intake were close to 0 (data not shown). These FAs were not correlated with beef or lamb intake (data not shown). The dose-response relations between these FAs and dairy fat intake are shown in Figure 1Go. As was shown in Table 3Go, we observed clear dose-response curves for 15:0 and trans 16:1n–7, although these curves were flattened or even dropped at the 2 highest deciles of dairy fat intake.


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

  		TABLE 3. Spearman partial correlation coefficients between plasma and erythrocyte fatty acid composition and dairy fat and dairy product intake measured by food-frequency questionnaires during various periods among controls, the Nurses' Health Study, 19901

 

Figure 1
View larger version (11K):
[in this window]
[in a new window]

  		FIGURE 1.. Relation between cumulative average intake of dairy fat in 1986–1990 and biomarker concentrations in plasma and erythrocytes, the Nurses' Health Study, 1990. t, trans configuration. n = 306. Least-squares means of fatty acids in plasma ({diamond}) and erythrocytes ({blacksquare}) are shown as plotted against median values by decile of intake (% of total fat) after adjustment for total energy intake, age, smoking, BMI, current weight, postmenopausal hormone use, fasting status, and period of blood assay. Bars are 95% CIs (x ± SEM: 1.96 x SEM). The range on the y axis in the middle panel differs from that in the top and bottom panels. Median values of each decile of intake were entered into the models as a continuous variable to estimate the P values for linear trend. All P values for trend are based on t tests, and P < 0.001.

 
The baseline characteristics of cases and controls are shown in Table 4Go. Women diagnosed with IHD events had a significantly (P < 0.01) higher BMI, drank significantly (P = 0.01) less alcohol, were significantly (P < 0.001) more likely to have a parental history of MI, and had significantly (P < 0.001) less favorable plasma lipoprotein values than did controls. For the FAs of interest in both blood fractions, the concentrations among IHD cases and controls did not differ significantly except that 17:0 content in erythrocytes was significantly lower in cases than in controls (P = 0.01).


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

  		TABLE 4. Baseline characteristics of study participants, the Nurses' Health Study, 19901

 
RRs of IHD associated with 15:0, 17:0, and trans 16:1n–7 in plasma and erythrocytes are shown in Table 5Go. In crude models (model 1), trans 16:1n–7 in plasma and 17:0 in erythrocytes were associated with lower risks of IHD. These inverse associations were attenuated toward the null after further adjustment of established dietary and lifestyle risk factors for IHD and of other FAs in plasma or erythrocytes, whereas the positive associations for 15:0 in plasma were strengthened after such adjustments. These results indicated the presence of negative confounding on these associations caused by the covariates. The RR of IHD between the highest tertile and the lowest tertile of 15:0 in plasma was 2.36 (95% CI: 1.16, 4.78; P for trend = 0.03). Adjustment for other nutrients in which dairy foods are rich—ie, calcium and vitamin D—did not significantly change the associations. Further adjustment for fiber, folate, and fruit and vegetables did not alter these associations. To examine whether these findings may be due to changes in diet subsequent to the diagnosis of diabetes, we excluded participants who were diagnosed with diabetes at baseline. The results did not differ significantly (data not shown).


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

  		TABLE 5. Relative risks (and 95% CIs) of ischemic heart disease by tertile (T) of fatty acid content of plasma and erythrocytes, the Nurses' Health Study, 19901

 
In a secondary analysis, we found positive but nonsignificant correlations between the biomarker concentrations and the ratio of LDL to HDL cholesterol (LDL:HDL) (data not shown). We observed a correlation coefficient of 0.09 (P = 0.13) between LDL:HDL and 15:0 in plasma. After further adjustment of LDL:HDL in the multivariate models, the associations for 15:0 in plasma were partly attenuated. The RRs across each tertile of 15:0 in plasma were 1.0 (reference), 1.93 (95% CI: 1.0, 3.72), and 2.25 (95% CI: 0.99, 5.15) (P for trend = 0.07). Associations for other biomarkers were attenuated to a similar extent after such adjustment (data not shown).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
In this prospective, nested, case-control study, we explored potential biomarkers of dairy fat intake in plasma and erythrocytesand examined their associations with the risk of IHD. We found that 15:0 and trans 16:1n–7 in plasma and erythrocytes could be used as biomarkers of dairy fat intake. We observed significant associations between plasma 15:0 concentrations and the risks of IHD after adjustment for established dietary and lifestyle IHD risk factors and other FAs in plasma. In comparison with women in the lowest tertile of 15:0 in plasma, women in the highest tertile had IHD risks more than twice as great.

Consumption of whole milk and other high-fat dairy products, such as butter, has been postulated to be associated with the risk of IHD because of the high saturated fat content of those products (1). Although data from ecological studies suggested a strong, positive correlation between dairy food consumption and IHD mortality (2–4), most prospective cohort studies showed mixed results regarding dairy product consumption and the risk of IHD (7, 32–34). These studies varied according to population characteristics, dietary assessment methods, categorization of intake, and assessment and control for confounders, which may explain the inconsistency. Moreover, in addition to the fat content, milk and dairy products contain many nutrients that may play different roles in the development of IHD. For example, intake of calcium, for which dairy products are the main food source, has been associated in some epidemiologic studies with lower risks of IHD (7, 35).

Dairy fat, from which the health concerns arose, has been only indirectly evaluated in a few epidemiologic studies (6, 7). Measuring dairy fat intake can be challenging in epidemiologic studies that use FFQs to quantify it. Apart from the inaccuracies of self-reports or food composition databases (9, 10), measurement errors may also result from the facts that dairy fat exists in many food items and that the content of dairy fat may vary in the same foods, depending on whether whole milk or low-fat milk is used to produce dairy products (8). Biomarkers of dairy fat intake are not subject to these measurement errors and thus are believed to be more precise than self-reports. Adipose tissue or plasma contents of 15:0 and 17:0 have been found to be valid biomarkers of dairy fat intake (11, 12, 28–31). These 2 odd-numbered FAs are exclusively produced by the bacterial flora in the rumen of ruminants (20). In a Swedish population, Wolk et al (11) documented a correlation coefficient of 0.36 between adipose tissue 15:0 content and FFQ dairy fat measurement, which was similar to our results. In 3 other Swedish studies, the validity of serum 15:0 content as a biomarker of dairy fat intake was examined in relation to diet record measurements, and the authors found somewhat higher correlation coefficients (r > 0.40), which may be due to the fact that fewer measurement errors were associated with the diet records (28, 29, 31). In these studies, 17:0 in human tissues or serum fractions was a weaker marker than 15:0 (11, 28, 29, 31). In a study conducted in Norway, a negative association was documented between dairy fat intake and 17:0 in serum or adipose tissue (30). These findings may be due to different metabolic procedures of these 2 FAs in humans.

To our knowledge, this is the first report that trans 16:1n–7 could be used as a biomarker for dairy fat intake. This FA exists exclusively in ruminant fats, especially dairy fat (21). Because trans FAs cannot be synthesized by humans, this FA possesses the characteristics of a potential biomarker. More studies are needed to confirm this novel finding. Although trans 18:1n–7 is the major trans isomer in dairy fat, the fact that this trans isomer also exists in partially hydrogenated oils may explain its weak correlation with dairy fat intake (21).

In the current study, we found that these FAs in plasma or erythrocytes were correlated with dairy product intake less than with dairy fat intake. This finding is consistent with previous observations (28). As expected, these biomarkers were not correlated with low-fat milk intake in our study. Approximately 80% of the participants in the present study reported drinking low-fat milk rather than whole milk. Although 15:0, 17:0, and trans 16:1n–7 also exist in ruminant meats (eg, beef or lamb), they were not valid biomarkers of ruminant meat intake because these foods are not the major food sources of these FAs (21, 36).

Few epidemiologic studies have examined the coronary effects of dairy fat intake, and the results indicated that high dairy fat intake may increase the risk of heart disease (6, 7). It is interesting that, in 2 small case-control studies that used biomarkers to represent dairy fat intake, adipose tissue or serum 15:0 and 17:0 concentrations were associated with nonsignificantly lower risks of CHD after adjustment for covariates (36, 37). Residual confounding cannot be excluded, because only a few lifestyle risk factors were controlled for in these 2 studies. In the current study, after we adjusted for established IHD risk factors and other FAs in plasma, 15:0 in plasma, the strongest biomarker identified in our study, was associated with a higher risk of IHD, whereas associations for other weaker biomarkers were not significant.

Clinical feeding trials have provided solid evidence regarding the effects of butter fat intake in increasing atherogenic plasma LDL-cholesterol concentrations. At the same time, butter fat intake modestly increases HDL-cholesterol concentrations (38–40). As a result, butter fat intake causes a small increase in ratio of total to HDL cholesterol (41), which is a stronger risk factor than either LDL or HDL cholesterol alone (25). In the present study, the 15:0 in plasma was correlated with a slightly increased LDL:HDL. However, adjustment of this lipid variable did not entirely abolish the association between 15:0 concentration in plasma and IHD risk. These results suggested that other pathways linking dairy fat intake with a higher risk of IHD also may exist. Dairy fat consists of {approx}60% saturated FAs, 30% monounsaturated FAs, and 5% trans FAs (8, 21, 42). Saturated fat intake has effects on promoting platelet aggregation (43), activating factor VII (44), and increasing lipoprotein(a) production (45). trans FAs may increase systemic inflammation (46), cause endothelial dysfunction, and impair insulin sensitivity (47).

The present study has two noteworthy strengths. First, we measured a full spectrum of FAs in plasma and erythrocytes, and we thus had the opportunity to identify a greater number of potential biomarkers than did the previous studies. Second, because of the study's prospective design, the occurrence of disease could not influence biomarker concentrations in erythrocytes and plasma.

The present study was also subject to several limitations. First, because we did not have complete data on 15:0 and 17:0 content in foods, we cannot assess whether these FAs reflect their own intakes more than dairy fat intake. Second, single baseline measurements of plasma and erythrocyte FA concentrations will not perfectly reflect long-term dietary intake, which is the variable of interest. Third, we observed higher laboratory measurement errors for erythrocyte FAs than for plasma FAs. This may partly explain why plasma 15:0 content was a better marker and a better predictor than the erythrocyte one. Fourth, in observational studies, residual confounding cannot be entirely ruled out.

In conclusion, 15:0 and trans 16:1n–7 content in plasma and erythrocyte can be used as biomarkers of dairy fat intake as assessed by FFQs. These biomarker data suggest that a high intake of dairy fat may be associated with a moderately greater risk of IHD.


   ACKNOWLEDGMENTS
 
The authors' responsibilities were as follows—QS: analyzed the data and drafted the manuscript; JM: designed the study and directed the blood sample assays; HC: designed the study, assayed the biomarkers, and prepared the data; FBH: designed the study analytic strategy and supervised the data analysis; and all authors: contributed to the revision of the manuscript. None of the authors had any personal or financial conflict of interest.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Tholstrup T. Dairy products and cardiovascular disease. Curr Opin Lipidol 2006;17:1–10.[Medline]
  2. Menotti A, Kromhout D, Blackburn H, Fidanza F, Buzina R, Nissinen A. Food intake patterns and 25-year mortality from coronary heart disease: cross-cultural correlations in the Seven Countries Study. The Seven Countries Study Research Group. Eur J Epidemiol 1999;15:507–15.
  3. Artaud-Wild SM, Connor SL, Sexton G, Connor WE. Differences in coronary mortality can be explained by differences in cholesterol and saturated fat intakes in 40 countries but not in France and Finland. A paradox. Circulation 1993;88:2771–9.
  4. Keys A. Seven countries: a multivariate analysis of death and coronary heart disease. Cambridge, MA: Harvard University Press, 1980.
  5. Elwood PC, Pickering JE, Hughes J, Fehily AM, Ness AR. Milk drinking, ischaemic heart disease and ischaemic stroke II. Evidence from cohort studies. Eur J Clin Nutr 2004;58:718–24.[Medline]
  6. Hu FB, Stampfer MJ, Manson JE, et al. Dietary saturated fats and their food sources in relation to the risk of coronary heart disease in women. Am J Clin Nutr 1999;70:1001–8.[Abstract/Free Full Text]
  7. Bostick RM, Kushi LH, Wu Y, Meyer KA, Sellers TA, Folsom AR. Relation of calcium, vitamin D, and dairy food intake to ischemic heart disease mortality among postmenopausal women. Am J Epidemiol 1999;149:151–61.[Abstract/Free Full Text]
  8. Gebhardt SE, Thomas RG. Nutritive value of foods. Washington, DC: US Department of Agriculture, 2002. (Home and Garden Bulletin 72.)
  9. Cantwell MM. Assessment of individual fatty acid intake. Proc Nutr Soc 2000;59:187–91.[Medline]
  10. Willett WC. Nutritional epidemiology. 2nd ed. New York: Oxford University Press, 1998.
  11. Wolk A, Vessby B, Ljung H, Barrefors P. Evaluation of a biological marker of dairy fat intake. Am J Clin Nutr 1998;68:291–5.[Abstract]
  12. Baylin A, Kabagambe EK, Siles X, Campos H. Adipose tissue biomarkers of fatty acid intake. Am J Clin Nutr 2002;76:750–7.[Abstract/Free Full Text]
  13. Rose GBH. Cardiovascular survey method. 2nd ed. Geneva, Switzerland: World Health Organization, 1982.
  14. Stampfer MJ, Willett WC, Speizer FE, et al. Test of the National Death Index. Am J Epidemiol 1984;119:837–9.[Free Full Text]
  15. Garland M, Sacks FM, Colditz GA, et al. The relation between dietary intake and adipose tissue composition of selected fatty acids in US women. Am J Clin Nutr 1998;67:25–30.[Abstract]
  16. Institute CaFE. Composition of foods: raw, processed, prepared. Washington, DC: US Department of Agriculture, 1989. (Agriculture handbook no. 8.)
  17. Hepburn FN, Exler J, Weihrauch JL. Provisional tables on the content of omega-3 fatty acids and other fat components of selected foods. J Am Diet Assoc 1986;86:788–93.[Medline]
  18. Slover HT, Thompson RH, Davis CA, Merola GB. Lipids in margarine and margarine-like foods. J Am Oil Chem Soc 1985;62:775–86.
  19. Enig MG, Pallansch LA, Sampugna J, Keeney M. Fatty acid composition of the fat in selected food items with emphasis on trans components. J Am Oil Chem Soc 1983;60:1788–95.
  20. Wu Z, Palmquist DL. Synthesis and biohydrogenation of fatty acids by ruminal microorganisms in vitro. J Dairy Sci 1991;74:3035–46.[Abstract/Free Full Text]
  21. Kris-Etherton PM, ed. Trans fatty acids and coronary heart disease risk. Report of the Expert Panel on Trans Fatty Acids and Coronary Heart Disease. Am J Clin Nutr 1995;62(suppl):655S–708S.[Medline]
  22. Zeleniuch-Jacquotte A, Chajes V, Van Kappel AL, Riboli E, Toniolo P. Reliability of fatty acid composition in human serum phospholipids. Eur J Clin Nutr 2000;54:367–72.[Medline]
  23. Baylin A, Kim MK, Donovan-Palmer A, et al. Fasting whole blood as a biomarker of essential fatty acid intake in epidemiologic studies: comparison with adipose tissue and plasma. Am J Epidemiol 2005;162:373–81.[Abstract/Free Full Text]
  24. Lemaitre RN, King IB, Raghunathan TE, et al. Cell membrane trans-fatty acids and the risk of primary cardiac arrest. Circulation 2002;105:697–701.[Abstract/Free Full Text]
  25. Shai I, Rimm EB, Hankinson SE, et al. Multivariate assessment of lipid parameters as predictors of coronary heart disease among postmenopausal women: potential implications for clinical guidelines. Circulation 2004;110:2824–30.[Abstract/Free Full Text]
  26. Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986;42:121–30.[Medline]
  27. Rothman KJ, Greenland S. Modern epidemiology. 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins, 1998.
  28. Smedman AE, Gustafsson IB, Berglund LG, Vessby BO. Pentadecanoic acid in serum as a marker for intake of milk fat: relations between intake of milk fat and metabolic risk factors. Am J Clin Nutr 1999;69:22–9.[Abstract/Free Full Text]
  29. Wolk A, Furuheim M, Vessby B. Fatty acid composition of adipose tissue and serum lipids are valid biological markers of dairy fat intake in men. J Nutr 2001;131:828–33.[Abstract/Free Full Text]
  30. Brevik A, Veierod MB, Drevon CA, Andersen LF. Evaluation of the odd fatty acids 15:0 and 17:0 in serum and adipose tissue as markers of intake of milk and dairy fat. Eur J Clin Nutr 2005;59:1417–22.[Medline]
  31. Rosell M, Johansson G, Berglund L, Vessby B, de Faire U, Hellenius ML. The relation between alcohol intake and physical activity and the fatty acids 14:0, 15:0 and 17:0 in serum phospholipids and adipose tissue used as markers for dairy fat intake. Br J Nutr 2005;93:115–21.[Medline]
  32. Elwood PC, Pickering JE, Fehily AM, Hughes J, Ness AR. Milk drinking, ischaemic heart disease and ischaemic stroke I. Evidence from the Caerphilly cohort. Eur J Clin Nutr 2004;58:711–7.[Medline]
  33. Ness AR, Smith GD, Hart C. Milk, coronary heart disease and mortality. J Epidemiol Community Health 2001;55:379–82.[Abstract/Free Full Text]
  34. Snowdon DA, Phillips RL, Fraser GE. Meat consumption and fatal ischemic heart disease. Prev Med 1984;13:490–500.[Medline]
  35. Van der Vijver LP, van der Waal MA, Weterings KG, Dekker JM, Schouten EG, Kok FJ. Calcium intake and 28-year cardiovascular and coronary heart disease mortality in Dutch civil servants. Int J Epidemiol 1992;21:36–9.[Abstract/Free Full Text]
  36. Biong AS, Veierod MB, Ringstad J, Thelle DS, Pedersen JI. Intake of milk fat, reflected in adipose tissue fatty acids and risk of myocardial infarction: a case-control study. Eur J Clin Nutr 2006;60:236–44.[Medline]
  37. Warensjo E, Jansson JH, Berglund L, et al. Estimated intake of milk fat is negatively associated with cardiovascular risk factors and does not increase the risk of a first acute myocardial infarction. A prospective case-control study. Br J Nutr 2004;91:635–42.
  38. Wardlaw GM, Snook JT. Effect of diets high in butter, corn oil, or high-oleic acid sunflower oil on serum lipids and apolipoproteins in men. Am J Clin Nutr 1990;51:815–21.[Abstract/Free Full Text]
  39. Wood R, Kubena K, O'Brien B, Tseng S, Martin G. Effect of butter, mono- and polyunsaturated fatty acid-enriched butter, trans fatty acid margarine, and zero trans fatty acid margarine on serum lipids and lipoproteins in healthy men. J Lipid Res 1993;34:1–11.[Medline]
  40. Tholstrup T, Hoy CE, Andersen LN, Christensen RD, Sandstrom B. Does fat in milk, butter and cheese affect blood lipids and cholesterol differently? J Am Coll Nutr 2004;23:169–76.[Abstract/Free Full Text]
  41. Mensink RP, Zock PL, Kester AD, Katan MB. Effects of dietary fatty acids and carbohydrates on the ratio of serum total to HDL cholesterol and on serum lipids and apolipoproteins: a meta-analysis of 60 controlled trials. Am J Clin Nutr 2003;77:1146–55.[Abstract/Free Full Text]
  42. Agriculture Research Service, US Department of Agriculture. USDA National nutrient database for standard reference—release 18 (2005). Washington, DC: US Department of Agriculture, 2006.
  43. Kwon JS, Snook JT, Wardlaw GM, Hwang DH. Effects of diets high in saturated fatty acids, canola oil, or safflower oil on platelet function, thromboxane B2 formation, and fatty acid composition of platelet phospholipids. Am J Clin Nutr 1991;54:351–8.[Abstract/Free Full Text]
  44. Mitropoulos KA, Miller GJ, Martin JC, Reeves BE, Cooper J. Dietary fat induces changes in factor VII coagulant activity through effects on plasma free stearic acid concentration. Arterioscler Thromb 1994;14:214–22.[Abstract/Free Full Text]
  45. Tholstrup T, Marckmann P, Vessby B, Sandstrom B. Effect of fats high in individual saturated fatty acids on plasma lipoprotein[a] levels in young healthy men. J Lipid Res 1995;36:1447–52.[Abstract]
  46. Mozaffarian D. Trans fatty acids—effects on systemic inflammation and endothelial function. Atheroscler Suppl 2006;7:29–32.[Medline]
  47. Mozaffarian D, Katan MB, Ascherio A, Stampfer MJ, Willett WC. Trans fatty acids and cardiovascular disease. N Engl J Med 2006;354:1601–13.[Free Full Text]
Received for publication April 26, 2007. Accepted for publication June 8, 2007.




This article has been cited by other articles:


Home page
 J. Am. Coll. Nutr.Home page
P. C. Elwood, D. I. Givens, A. D. Beswick, A. M. Fehily, J. E. Pickering, and J. Gallacher
The Survival Advantage of Milk and Dairy Consumption: an Overview of Evidence from Cohort Studies of Vascular Diseases, Diabetes and Cancer
J. Am. Coll. Nutr., December 1, 2008; 27(6): 723S - 734S.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Clin. Nutr.Home page
P. Elwood, A. Fehily, J. Pickering, and I. Givens
Pentadecanoic acid (15:0), milk, and ischemic heart disease
Am. J. Clinical Nutrition, May 1, 2008; 87(5): 1540 - 1541.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sun, Q.
Right arrow Articles by Hu, F. B
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sun, Q.
Right arrow Articles by Hu, F. B
Agricola
Right arrow Articles by Sun, Q.
Right arrow Articles by Hu, F. B

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS