Abstract
Background: The associations of serum omega-6 (n−6) fatty acids with lipoprotein subclasses at the population level are uncertain.
Objective: We aimed to examine associations between major n−6 fatty acids [ie, linoleic acid (LA, 18:2n−6) and arachidonic acid (AA, 20:4n−6)] and the lipoprotein subclasses VLDL, LDL, and HDL.
Design: We conducted a cross-sectional study in 1098 participants using population-based data from US white, Japanese American, Japanese, and Korean men aged 40–49 y. Serum fatty acids were analyzed by capillary gas-liquid chromatography. Lipoprotein subclasses were measured by nuclear magnetic resonance spectroscopy. Multiple linear regression models as a function of each fatty acid were used after adjustment for age, population, body mass index, pack-years of smoking, alcohol consumption, diabetes, hypertension, and omega-3 (n−3) and trans fatty acids.
Results: Serum LA was inversely associated with large VLDL (β = −0.62, P < 0.001), total LDL (β = −22.08, P < 0.001), and small LDL (β = −31.89, P < 0.001) particle concentrations and VLDL size (β = −0.72, P < 0.001). Serum LA was positively associated with large HDL particle concentration (β = 0.21, P < 0.001) and HDL size (β = 0.03, P < 0.001). The patterns of association of AA with large VLDL and large HDL particle concentrations were comparable with those of LA.
Conclusions: At the population level, higher serum concentrations of LA were significantly associated with lower concentrations of total LDL particles. Higher serum concentrations of LA and AA were significantly associated with a lower concentration of large VLDL particles and a higher concentration of large HDL particles. These associations were consistent across the population groups. This trial was registered at clinicaltrials.gov as NCT00069797.
INTRODUCTION
Several epidemiologic studies have shown an association between dietary omega-6 (n−6) polyunsaturated fatty acid (PUFA) intake, more specifically linoleic acid (LA, 18:2n−6) intake and a reduced risk of coronary heart disease (CHD) or stroke (1–3). LA is an essential fatty acid that represents the primary source of n−6 fatty acids. The Nurses’ Health Study has reported that dietary LA was inversely associated with CHD risk (4). Dietary intake of LA is a strong and significant determinant of serum LA (5). Recently, the Kuopio Ischemic Heart Disease Risk Factor (KIHD) study has reported that serum LA is inversely associated with cardiovascular mortality (5). LA has a potent benefit in lowering LDL cholesterol and is an established CHD risk factor when substituted for saturated fatty acids (SFAs) (6–8). LA may also have cardioprotective benefits through its effect on thrombosis, arrhythmia, insulin resistance, and blood pressure (9–10).
Particle numbers of lipoprotein subclasses or their average size can provide additional information for risk prediction of CHD compared with standard measures of serum lipids. Nuclear magnetic resonance (NMR) spectroscopy is one method for quantifying the numbers of and measuring average sizes of lipoprotein subclasses, ie, VLDL, LDL, and HDL. Recently, LDL particle concentration, measured by NMR spectroscopy, has been recognized to be a predictor of CHD risk in healthy men and women in prospective and case-control studies (11–14). Besides LDL particles, small LDL and large VLDL particle concentrations may also be related to the prevalence of subclinical atherosclerosis or cardiovascular disease risk (11, 15, 16). In addition, a large HDL particle concentration may be protective against cardiovascular disease risk, rather than total, medium, and small HDL particle concentrations (11).
The associations of serum n−6 fatty acids with lipoprotein particle concentrations or sizes are uncertain. Although a few clinical trials examined the effect of n−6 PUFA–enriched diets on lipoprotein particle concentrations or sizes, they are likely to have methodologic limitations, such as differences in background diets and a very small sample size (17, 18). No previous study has examined the associations between serum n−6 fatty acids and lipoprotein subclasses in a population consuming their habitual diet.
We previously showed associations between serum n−6 fatty acids and lipids (ie, LDL and HDL cholesterols and triglycerides) in 3 different population groups (19). In the present study, we aimed to examine the associations between serum n−6 fatty acids [ie, total n−6 PUFAs, LA, and arachidonic acid (AA, 20:4n−6)] and the lipoprotein concentrations or sizes of VLDL, LDL, and HDL in 1098 US white, Japanese American, Japanese, and Korean men aged 40–49 y who participated in an electron-beam tomography and risk assessment in the post-World War II birth cohort (ERA-JUMP)—a population-based cross-sectional study.
SUBJECTS AND METHODS
Subjects
During 2002–2006, a population-based sample of 1228 men aged 40–49 y was obtained from 4 centers: 310 whites from Allegheny County, PA; 303 Japanese Americans from Honolulu, HI; 313 Japanese from Kusatsu, Shiga, Japan; and 302 Koreans from Ansan, Gyeonggi-do, South Korea (20). None of the multicenter participants had clinical cardiovascular disease, type I diabetes, or other severe diseases (20, 21). Of the original sample, we excluded those taking lipid-lowering medications (n = 123) or with missing data (n = 7). The final sample consisted of 1098 men (269 whites, 229 Japanese Americans, 303 Japanese, and 298 Koreans).
Written informed consent was obtained from all participants. The study was approved by the Institutional Review Boards of the following institutions: the University of Pittsburgh, PA; the Kuakini Medical Center, Honolulu, HI; Shiga University of Medical Science, Otsu, Japan; and Korea University, Seoul, South Korea.
All participants underwent a physical examination, and completed a lifestyle questionnaire (eg, smoking and alcohol consumption), and a laboratory assessment as described previously (20–22). Venipuncture was performed early in the clinic visit after a 12-h fast, and samples were stored at −80°C and shipped on dry ice to the University of Pittsburgh. Data collection was standardized across all 4 centers.
Nuclear magnetic resonance lipoprotein measurements
NMR spectroscopy (LipoScience Inc, Raleigh, NC) was performed to quantify the particle numbers of VLDL, LDL, and HDL by using each of the serum samples obtained from study participants (23, 24). Particle concentrations were determined for 3 VLDL subclasses (large, >60 nm; medium, 35–60 nm; and small, 27–35 nm), 3 LDL subclasses [intermediate-density lipoprotein (IDL), 23–27 nm; large, 21.3–23 nm; small, 18.3–21.2 nm], and 3 HDL subclasses (large, 8.8–13 nm; medium, 8.2–8.8 nm; and small, 7.3–8.2 nm) (25). We calculated weighted average particle sizes of VLDL, LDL, and HDL.
Measurement of serum fatty acids
Serum fatty acids were analyzed by capillary gas-liquid chromatography (PerkinElmer Clarus 500; PerkinElmer, Waltham, MA) (26, 27). The main sources of the fatty acids are cholesterol esters, phospholipids, and triglycerides. CVs between runs for major n−6 fatty acids (LA and AA) were 1.6% and 2.8%, respectively. The coefficients for marine omega-3 (n−3) PUFAs—eicosapentaenoic acid (EPA, 20:5n−3), docosapentaenoic acid (DPA, 22:5n−3), and docosahexaenoic acid (DHA, 22:6n−3)—were 2.5%, 2.5%, and 7.0%, respectively. The coefficients for other major fatty acids—palmitic acid (16:0), stearic acid (18:0), oleic acid (18:1n−9), α-linolenic acid (ALA, 18:3n−3), and total fatty acids—were 1.2%, 4.0%, 2.3%, 7.9%, and 5.7%, respectively. The coefficients for trans fatty acids—palmitelaidic (16:1), trans-9-octadecanoic (18:1), and linolelaidic acid (18:2)—were 4.1%, 3.3%, and 9.8%, respectively. Fatty acids were expressed as a percentage of total serum fatty acids.
Statistical analysis
To examine differences in clinical, lipid, lipoprotein particle concentrations or sizes, and serum proportions of fatty acids between population groups, we performed an analysis of variance and a post hoc Bonferroni's test. To examine the association between each fatty acid (a predictor variable) and each lipoprotein (an outcome variable), we performed a multiple linear regression analysis by using pooled data from 4 population groups (n = 1098). We have several rationales for performing the analysis of the pooled data. First, the analysis may be a robust statistical method rather than a population-specific data analysis with regard to sample size. Second, the patterns of associations of n−6 fatty acids with lipoprotein subclasses were consistent across the 4 population groups.
In the multiple linear regression analysis, we used adjusted models 1, 2, and 3. In model 1, age and population were adjusted for. In model 2, body mass index, pack-years of smoking, alcohol consumption (g/d), diabetes, and hypertension were further adjusted for. In model 3, total n−3 and trans fatty acids were further adjusted for. When modeling for AA, model 4 was further adjusted for LA, because a few clinical trials reported that dietary AA decreased the serum concentrations of LA (28, 29). Hypertension was defined as a systolic blood pressure ≥140 mm Hg, a diastolic blood pressure ≥90 mm Hg, or hypertensive medication. Diabetes was defined as a fasting glucose concentration ≥126 mg/dL or use of diabetes medication. Total n−3 fatty acids indicate the sum of marine-derived n−3 fatty acids (DHA, DPA, and EPA) and ALA. Statistical significance was considered to be P < 0.05. All statistical analyses were performed by using STATA 10.0 for Windows (StataCorp LP, College Station, TX).
RESULTS
The mean age of the study population (n = 1098) was 45 y (Table 1). Of the study population, 19% and 6% had hypertension and type 2 diabetes mellitus, respectively. Population-specific clinical and lipid variables were reported previously, consisting of data from US white, Japanese American, and Japanese men (19). Concerning lipoprotein subclasses across 4 population groups, US white and Japanese American men had similar concentrations of large VLDL particles; the 2 population groups had significantly higher concentrations of large VLDL particles than did Japanese and Korean men (Table 1). US white, Japanese American, and Japanese men had similar concentrations of total LDL particles; the 3 population groups had significantly higher concentrations of total LDL particles than did Korean men. Japanese men had a significantly higher concentration of large HDL particles than did the US white, Japanese American, and Korean men.
TABLE 1.
Characteristics of the study participants in 2002–2006: clinical, lipid, and lipoprotein subclass variables (n = 1098)1
| All (n = 1098) | W (n = 269) | JA (n = 229) | JJ (n = 302) | Kr (n = 298) | |
| Age (y) | 45.1 ± 2.92 | 44.9 ± 2.83 | 46.0 ± 2.945 | 45.1 ± 2.8 | 44.7 ± 2.8 |
| BMI (kg/m2) | 25.7 ± 3.9 | 27.7 ± 4.245 | 27.4 ± 4.245 | 23.7 ± 3.15 | 24.7 ± 2.7 |
| Systolic BP (mm Hg) | 123.8 ± 13.7 | 122.8 ± 11.4 | 126.5 ± 12.15 | 124.6 ± 15.8 | 121.7 ± 14.1 |
| Diastolic BP (mm Hg) | 75.6 ± 10.4 | 73.2 ± 8.83–5 | 76.8 ± 8.8 | 76.3 ± 11.9 | 76.2 ± 11.1 |
| Hypertension [n (%)] | 213 (19.4) | 36 (13.4) | 55 (24.0) | 77 (25.5) | 45 (15.2) |
| Type 2 diabetes [n (%)] | 70 (6.4) | 8 (3.0) | 17 (7.4) | 17 (5.6) | 28 (9.4) |
| Total cholesterol (mg/dL) | 208.6 ± 36.7 | 214.8 ± 37.05 | 212.1 ± 35.945 | 216.3 ± 35.45 | 192.3 ± 33.5 |
| LDL cholesterol (mg/dL) | 127.8 ± 34.1 | 137.1 ± 33.05 | 127.9 ± 32.25 | 131.6 ± 35.95 | 115.5 ± 31.3 |
| HDL cholesterol (mg/dL) | 49.6 ± 13.1 | 48.0 ± 13.04 | 50.7 ± 12.15 | 54.1 ± 13.75 | 45.7 ± 11.5 |
| Triglycerides (mg/dL) | 133.0 (96.0–191.0)6 | 126.0 (92.0–186.0) | 140.0 (93.0–218.0) | 135.0 (102.0–180.0) | 132.5 (96.0–202.0) |
| Fasting glucose (mg/dL) | 104.6 ± 17.2 | 100.8 ± 13.43,3,4 | 109.1 ± 16.95 | 106.6 ± 18.6 | 102.7 ± 17.9 |
| Fasting insulin (μIU/mL) | 10.7 (8.2–14.1) | 12.5 (10.2–17.2)45 | 13.3 (9.7–16.6)45 | 9.4 (7.4–12.4) | 9.3 (5.5–11.8) |
| Smoking (pack-years) | 11.0 ± 14.7 | 3.8 ± 8.845 | 4.3 ± 9.245 | 19.5 ± 16.95 | 14.1 ± 14.3 |
| Alcohol consumption (g/d) | 19.0 ± 27.5 | 10.3 ± 14.445 | 16.7 ± 28.74 | 26.2 ± 28.1 | 21.3 ± 32.3 |
| Lipoprotein subclass | |||||
| VLDL particles | |||||
| Total (nmol/L) | 91.3 ± 47.9 | 93.1 ± 44.235 | 108.9 ± 54.745 | 91.2 ± 47.05 | 76.5 ± 41.2 |
| Large (nmol/L) | 3.8 ± 6.6 | 4.6 ± 6.74 | 5.4 ± 7.945 | 2.6 ± 5.5 | 3.2 ± 6.1 |
| Medium (nmol/L) | 42.9 ± 36.5 | 41.3 ± 32.33 | 53.1 ± 41.45 | 45.4 ± 37.25 | 33.9 ± 33.1 |
| Small (nmol/L) | 44.6 ± 22.5 | 47.2 ± 21.45 | 50.4 ± 24.245 | 43.1 ± 24.0 | 39.4 ± 19.1 |
| Average size (nm) | 46.9 ± 8.4 | 49.9 ± 7.845 | 48.8 ± 7.345 | 44.1 ± 7.6 | 45.7 ± 9.3 |
| LDL particles | |||||
| Total (nmol/L) | 1351.6 ± 442.3 | 1479.6 ± 400.95 | 1407.2 ± 512.45 | 1382.1 ± 441.05 | 1162.7 ± 353.8 |
| IDL (nmol/L) | 43.1 ± 48.2 | 52.2 ± 49.245 | 66.5 ± 55.445 | 33.6 ± 42.2 | 26.7 ± 37.1 |
| Large (nmol/L) | 461.3 ± 251.3 | 525.1 ± 280.83–5 | 334.0 ± 247.645 | 507.3 ± 227.7 | 454.5 ± 210.3 |
| Small (nmol/L) | 847.2 ± 506.7 | 902.3 ± 513.35 | 1006.7 ± 538.045 | 841.3 ± 508.65 | 681.5 ± 420.2 |
| Average size (nm) | 21.0 ± 0.9 | 20.9 ± 0.93 | 20.6 ± 0.945 | 21.1 ± 0.9 | 21.1 ± 0.9 |
| HDL particles | |||||
| Total (μmol/L) | 32.5 ± 6.6 | 31.1 ± 5.73–5 | 35.8 ± 5.85 | 35.3 ± 6.45 | 28.4 ± 5.6 |
| Large (μmol/L) | 6.2 ± 3.7 | 5.0 ± 3.24 | 6.0 ± 3.34 | 8.6 ± 4.05 | 4.9 ± 2.8 |
| Medium (μmol/L) | 1.9 ± 3.3 | 1.1 ± 2.234 | 2.9 ± 3.65 | 2.8 ± 4.45 | 0.8 ± 1.9 |
| Small (μmol/L) | 24.4 ± 5.2 | 25.0 ± 4.535 | 26.8 ± 4.845 | 23.9 ± 5.5 | 22.7 ± 5.1 |
| Average size (nm) | 8.8 ± 0.5 | 8.6 ± 0.545 | 8.7 ± 0.44 | 9.1 ± 0.55 | 8.8 ± 0.5 |
BP, blood pressure; IDL, intermediate-density lipoprotein; W, US white men; JA, Japanese American men; JJ, Japanese men in Japan; Kr, Korean men. ANOVA and post hoc tests with Bonferroni correction were performed to test between-group differences.
Mean ± SD (all such values).
Significantly different from JA, P < 0.05.
Significantly different from JJ, P < 0.05.
Significantly different from Kr, P < 0.05.
Median; interquartile range in parentheses (all such values).
The pooled data for the 4 population groups showed that PUFAs, SFAs, monounsaturated fatty acids (MUFAs), and trans fatty acids were 45.1%, 32.1%, 20.3%, and 0.9% of total fatty acids, respectively (Table 2). Total n−6 and total n−3 fatty acids were 37.2% and 7.3%, respectively. LA and AA were 27.7% and 7.5%, respectively. Regarding population-specific data, total fatty acids did not differ by population groups (Table 2). Specifically, US white and Japanese American men had similar concentrations of serum LA and AA; the 2 population groups had significantly higher concentrations of serum LA and AA than did Japanese men and Korean men.
TABLE 2.
Serum concentrations of all individual fatty acids in 2002–2006 (n = 1098)1
| All (n = 1098) | W (n = 269) | JA (n = 229) | JJ (n = 302) | Kr (n = 298) | |
| Total fatty acids (mg/L) | 241.0 ± 63.5 | 238.7 ± 51.9 | 236.0 ± 81.3 | 244.5 ± 52.0 | 243.6 ± 67.8 |
| Fatty acid proportion (%) | |||||
| Polyunsaturated fatty acids | 45.1 ± 4.7 | 46.1 ± 4.62–4 | 47.4 ± 4.334 | 44.7 ± 3.84 | 42.6 ± 4.7 |
| Total n−6 fatty acids5 | 37.2 ± 5.9 | 41.4 ± 4.234 | 41.5 ± 4.134 | 34.7 ± 4.24 | 32.6 ± 5.0 |
| Linoleic acid | 27.7 ± 5.0 | 30.0 ± 4.134 | 30.5 ± 4.234 | 26.5 ± 4.14 | 24.7 ± 4.8 |
| Arachidonic acid | 7.5 ± 2.1 | 8.9 ± 1.934 | 8.8 ± 2.234 | 6.6 ± 1.34 | 6.0 ± 1.3 |
| Total n−3 fatty acids6 | 7.3 ± 3.4 | 4.2 ± 1.82–4 | 5.5 ± 2.434 | 9.6 ± 3.0 | 9.0 ± 2.6 |
| Marine-derived n−3 fatty acids7 | 6.6 ± 3.3 | 3.8 ± 1.82–4 | 5.0 ± 2.334 | 9.3 ± 13.04 | 7.6 ± 2.5 |
| Docosahexaenoic acid | 4.2 ± 2.0 | 2.4 ± 1.22–4 | 3.2 ± 1.434 | 5.9 ± 1.74 | 4.8 ± 1.4 |
| Eicosapentaenoic acid | 1.6 ± 1.3 | 0.8 ± 0.62–4 | 1.0 ± 1.034 | 2.5 ± 1.44 | 1.9 ± 1.0 |
| α-Linolenic acid | 0.5 ± 0.6 | 0.3 ± 0.34 | 0.4 ± 0.434 | 0.2 ± 0.24 | 1.0 ± 0.8 |
| Saturated fatty acids8 | 32.1 ± 2.8 | 31.0 ± 2.534 | 30.9 ± 2.234 | 31.7 ± 2.14 | 34.3 ± 2.8 |
| Monounsaturated fatty acids9 | 20.3 ± 3.5 | 20.2 ± 3.223 | 19.1 ± 3.334 | 21.2 ± 3.14 | 20.3 ± 3.9 |
| trans Fatty acids10 | 0.9 ± 0.9 | 1.0 ± 0.53 | 0.9 ± 0.434 | 0.6 ± 0.24 | 1.2 ± 1.6 |
All values are means ± SDs. W, US white men; JA, Japanese American men; JJ, Japanese men in Japan; Kr, Korean men. ANOVA and post hoc tests with Bonferroni correction were performed to test between-group differences.
Significantly different from JA, P < 0.05.
Significantly different from JJ, P < 0.05.
Significantly different from Kr, P < 0.05.
The sum of linoleic acid (18:2n−6), γ-linolenic acid (18:3n−6), dihomo-γ-linolenic acid (20:3n−6), and arachidonic acid (20:4n−6).
The sum of marine-derived n−3 fatty acids, eicosatetraenoic acid (20:4n−3), and α-linolenic acid (18:3n−3).
The sum of docosahexaenoic acid (22:6n−3), docosapentaenoic acid (22:5n−3), and eicosapentaenoic acid (20:5n−3).
The sum of tetradecanoic acid (14:0), palmitic acid (16:0) and stearic acid (18:0).
The sum of palmitoleic acid (16:1n−7), vaccenic acid (18:1n−7), and oleic acid (18:1n−9).
The sum of palmitelaidic acid (16:1tn−7), trans octadecanoic acid (18:1), and linolelaidic acid (18:2ttn−6).
The pooled data for the associations between n−6 fatty acids and lipoprotein subclasses are presented in Tables 3, 4, and 5, with β coefficients, as the estimated average increase in each lipoprotein (mmol/L, μmol/L, or nm) per one-unit (% of total fatty acids) increase in each fatty acid in models 1, 2, 3, and 4. Total n−6 PUFAs and LA were significantly and inversely associated with total, large, and medium VLDL particle concentrations and VLDL size and positively associated with small VLDL particle concentration in models 1, 2, and 3 (Table 3). The associations of total n−6 PUFAs and LA with large and medium VLDL particle concentrations and VLDL size were consistent across the population groups (see Supplemental Table 1 under “Supplemental data” in the online issue).
TABLE 3.
Associations between n−6 fatty acids and VLDL subclasses (n = 1098)1
| Predictor variable |
|||
| Outcome variable | Total n−62 | Linoleic acid | Arachidonic acid |
| % | |||
| Total VLDL (nmol/L) | |||
| Model 1 | −1.84 (0.304)3 | −1.19 (0.323)3 | −5.31 (0.814)3 |
| Model 2 | −1.74 (0.313)3 | −1.08 (0.330)4 | −4.61 (0.822)3 |
| Model 3 | −2.88 (0.349)3 | −2.01 (0.368)3 | −5.22 (0.824)3 |
| Model IV | — | — | −5.59 (0.813)3 |
| Large VLDL (nmol/L) | |||
| Model 1 | −0.52 (0.040)3 | −0.42 (0.043)3 | −1.11 (0.110)3 |
| Model 2 | −0.48 (0.040)3 | −0.38 (0.043)3 | −0.99 (0.107)3 |
| Model 3 | −0.76 (0.041)3 | −0.62 (0.045)3 | −1.03 (0.106)3 |
| Model 4 | — | — | −1.14 (0.096)3 |
| Medium VLDL (nmol/L) | |||
| Model 1 | −1.97 (0.230)3 | −1.37 (0.247)3 | −4.99 (0.620)3 |
| Model 2 | −1.93 (0.239)3 | −1.30 (0.254)3 | −4.57 (0.630)3 |
| Model 3 | −3.16 (0.260)3 | −2.33 (0.279)3 | −5.15 (0.628)3 |
| Model 4 | — | — | −5.58 (0.606)3 |
| Small VLDL (nmol/L) | |||
| Model 1 | 0.64 (0.144)3 | 0.59 (0.152)3 | 0.79 (0.390)5 |
| Model 2 | 0.66 (0.151)3 | 0.60 (0.158)3 | 0.95 (0.399)5 |
| Model 3 | 1.04 (0.170)3 | 0.94 (0.177)3 | 0.97 (0.402)5 |
| Model 4 | — | — | 1.13 (0.398)5 |
| Average VLDL size (nm) | |||
| Model 1 | −0.44 (0.052)3 | −0.37 (0.055)3 | −1.00 (0.140)3 |
| Model 2 | −0.39 (0.052)3 | −0.32 (0.055)3 | −0.88 (0.139)3 |
| Model 3 | −0.83 (0.051)3 | −0.72 (0.054)3 | −0.95 (0.129)3 |
| Model 4 | — | — | −1.08 (0.119)3 |
All values are βs; SEs in parentheses. A multiple regression analysis was performed by using models 1, 2, 3, and 4. For models, the outcome variable is lipoprotein number or size, and the primary predictor variable is each fatty acid. Model 1 was adjusted for age and population; model 2 was further adjusted for BMI, pack-years of smoking, alcohol consumption, diabetes, and hypertension; model 3 was further adjusted for total n−3 and trans fatty acids; and model 4 was further adjusted for linoleic acid.
The sum of linoleic acid (18:2n−6), γ-linolenic acid (18:3n−6), dihomo-γ-linolenic acid (20:3n−6), and arachidonic acid (20:4n−6).
P < 0.001.
P < 0.01.
P < 0.05.
Total n−6 PUFAs and LA were significantly and inversely associated with total and small LDL particle concentrations and significantly and positively associated with large LDL particle concentration and LDL size in models 1, 2, and 3 (Table 4). The associations of LA with total, large, and small LDL particle concentrations and LDL size were consistent across the population groups (see Supplemental Table 2 under “Supplemental data” in the online issue). Regarding HDL particles, total n−6 PUFAs and LA were significantly and positively associated with large HDL particle concentration and HDL size and inversely associated with medium and small HDL particle concentrations in models 1, 2, and 3 (Table 5). The associations of LA with large, medium, and small HDL particle concentrations and HDL size were consistent across the population groups (see Supplemental Table 3 under “Supplemental data” in the online issue).
TABLE 4.
Associations between n−6 fatty acids and LDL subclasses (n = 1098)1
| Predictor variable |
|||
| Outcome variable | Total n−62 | Linoleic acid | Arachidonic acid |
| % | |||
| Total LDL (nmol/L) | |||
| Model 1 | −19.53 (2.733)3 | −21.06 (2.871)3 | −12.61 (7.500) |
| Model 2 | −16.50 (2.762)3 | −18.52 (2.881)3 | −1.78 (7.371) |
| Model 3 | −19.94 (3.109)3 | −22.08 (3.214)3 | −4.71 (7.388) |
| Model 4 | — | — | −8.48 (7.252) |
| IDL (nmol/L) | |||
| Model 1 | −0.45 (0.304) | −1.06 (0.318)4 | 2.25 (0.814)4 |
| Model 2 | −0.42 (0.313) | −1.08 (0.325)4 | 2.84 (0.817)4 |
| Model 3 | −0.93 (0.353)4 | −1.74 (0.364)3 | 2.86 (0.822)4 |
| Model 4 | — | — | 2.58 (0.817)4 |
| Large LDL (nmol/L) | |||
| Model 1 | 7.89 (1.625)3 | 5.99 (1.719)4 | 20.29 (4.370)3 |
| Model 2 | 7.26 (1.662)3 | 5.17 (1.746)4 | 17.68 (4.370)3 |
| Model 3 | 14.49 (1.829)3 | 11.54 (1.918)3 | 19.73 (4.348)3 |
| Model 4 | — | — | 21.79 (4.281)3 |
| Small LDL (nmol/L) | |||
| Model 1 | −26.97 (3.150)3 | −25.99 (3.331)3 | −35.15 (8.674)3 |
| Model 2 | −23.35 (3.123)3 | −22.61 (3.277)3 | −22.31 (8.381)4 |
| Model 3 | −33.51 (3.487)3 | −31.89 (3.639)3 | −27.29 (8.434)4 |
| Model 4 | — | — | −32.85 (8.155)3 |
| Average LDL size (nm) | |||
| Model 1 | 0.05 (0.006)3 | 0.04 (0.006)3 | 0.08 (0.016)3 |
| Model 2 | 0.04 (0.006)3 | 0.04 (0.006)3 | 0.05 (0.015)3 |
| Model 3 | 0.07 (0.006)3 | 0.06 (0.007)3 | 0.06 (0.015)3 |
| Model 4 | — | — | 0.08 (0.015)3 |
All values are βs; SEs in parentheses. IDL, intermediate-density lipoprotein. A multiple regression analysis was performed by using models 1, 2, 3, and 4. For models, the outcome variable is lipoprotein number or size, and the primary predictor variable is each fatty acid. Model 1 was adjusted for age and population; model 2 was further adjusted for BMI, pack-years of smoking, alcohol consumption, diabetes, and hypertension; model 3 was further adjusted for total n−3 and trans fatty acids; and model 4 was further adjusted for linoleic acid.
The sum of linoleic acid (18:2n−6), γ-linolenic acid (18:3n−6), dihomo-γ-linolenic acid (20:3n−6), and arachidonic acid (20:4n−6).
P < 0.001.
P < 0.01.
TABLE 5.
Associations between n−6 fatty acids and HDL subclasses (n = 1098)1
| Predictor variable |
|||
| Outcome variable | Total n−62 | Linoleic acid | Arachidonic acid |
| % | |||
| Total HDL (μmol/L) | |||
| Model 1 | −0.10 (0.043)3 | −0.16 (0.045)4 | 0.50 (0.114)4 |
| Model 2 | 0.00 (0.041) | −0.05 (0.043) | 0.41 (0.108)4 |
| Model 3 | −0.09 (0.046)3 | −0.17 (0.047)4 | 0.31 (0.107)5 |
| Model 4 | — | — | 0.29 (0.107)5 |
| Large HDL (μmol/L) | |||
| Model 1 | 0.14 (0.024)4 | 0.12 (0.025)4 | 0.38 (0.064)4 |
| Model 2 | 0.15 (0.022)4 | 0.14 (0.023)4 | 0.27 (0.059)4 |
| Model 3 | 0.23 (0.024)4 | 0.21 (0.025)4 | 0.25 (0.058)4 |
| Model 4 | — | — | 0.29 (0.056)4 |
| Medium HDL (μmol/L) | |||
| Model 1 | −0.13 (0.021)4 | −0.15 (0.023)4 | 0.03 (0.059) |
| Model 2 | −0.08 (0.021)4 | −0.09 (0.022)4 | 0.04 (0.057) |
| Model 3 | −0.13 (0.024)4 | −0.15 (0.025)4 | 0.02 (0.057) |
| Model 4 | — | — | 0.00 (0.057) |
| Small HDL (μmol/L) | |||
| Model 1 | −0.11 (0.033)5 | −0.14 (0.035)4 | 0.09 (0.090) |
| Model 2 | −0.07 (0.034)3 | −0.10 (0.036)5 | 0.10 (0.091) |
| Model 3 | −0.19 (0.038)4 | −0.23 (0.040)4 | 0.04 (0.091) |
| Model 4 | — | — | 0.00 (0.090) |
| Average HDL size (nm) | |||
| Model 1 | 0.02 (0.003)4 | 0.02 (0.003)4 | 0.03 (0.008)4 |
| Model 2 | 0.02 (0.003)4 | 0.02 (0.003)4 | 0.02 (0.008)5 |
| Model 3 | 0.03 (0.003)4 | 0.03 (0.003)4 | 0.02 (0.008)3 |
| Model 4 | — | — | 0.02 (0.007)5 |
All values are βs; SEs in parentheses. A multiple regression analysis was performed by using models 1, 2, 3, and 4. For models, the outcome variable is lipoprotein number or size, and the primary predictor variable is each fatty acid. Model 1 was adjusted for age and population; model 2 was further adjusted for BMI, pack-years of smoking, alcohol consumption, diabetes, and hypertension; model 3 was further adjusted for total n−3 and trans fatty acids; and model 4 was further adjusted for linoleic acid.
The sum of linoleic acid (18:2n−6), γ-linolenic acid (18:3n−6), dihomo-γ-linolenic acid (20:3n−6), and arachidonic acid (20:4n−6).
P < 0.05.
P < 0.001.
P < 0.01.
As for AA, we added the results from model 4, in which LA was further adjusted for. AA was significantly and inversely associated with total, large, and medium VLDL particle concentrations and VLDL size in models 1, 2, 3, and 4 (Table 5). These associations were consistent across the population groups (see Supplemental Table 1 under “Supplemental data” in the online issue). AA was not significantly associated with total LDL particle concentration. AA was significantly and positively associated with large LDL particle concentrations and LDL size, and inversely associated with small LDL particle concentration in models 1, 2, 3, and 4. Such associations did not, however, remain consistent across the population groups (see Supplemental Table 2 under “Supplemental data” in the online issue). AA was significantly and positively associated with total and large HDL particle concentrations and HDL sizes in models 1, 2, 3, and 4. The association of AA with total HDL particle concentrations did not, however, remain significant across the population groups (see Supplemental Table 3 under “Supplemental data” in the online issue).
DISCUSSION
In the population-based, cross-sectional study of 1098 men aged 40–49 y, we found that serum LA was significantly and inversely associated with large VLDL and total and small LDL particle concentrations. Serum LA was associated, significantly and positively, with large HDL particle concentration. Furthermore, serum AA was significantly and inversely associated with large VLDL particle concentration and significantly and positively associated with large HDL particle concentration.
To the best of our knowledge, this was the first study to report an association between serum LA and total LDL particle concentration in a population consuming their habitual diet. Some case-control studies have shown that a lowered proportion of serum or plasma LA is significantly related to nonfatal CHD events (30). Mechanisms for these observations may include the role of LA in lowering LDL cholesterol and blood pressure and improving insulin resistance (8). Our finding may add one more mechanism to the association between LA and CHD risk. The particle concentration of LDL is reported to be a risk factor for CHD in case-control studies and prospective cohort studies and may be a more predictive measure of the risk than is LDL cholesterol (11, 12, 31, 32). Therefore, the inverse association of serum LA with total LDL particle concentration may be more robust than would be expected from the association with LDL cholesterol alone for suggesting a beneficial role of LA in the reduction of CHD risk. Although the mechanism responsible for the association is unclear, it may be that LA increases the number of hepatic LDL receptors and hence enhances the clearance of LDL (33).
We found that serum LA was associated, significantly and positively, with large HDL particle concentration and average HDL size. Large HDL particle concentration may be associated with a reduced risk of carotid atherosclerosis, coronary progression (ie, lumen diameter change), and cardiovascular events (11, 15, 34). Thus, the association of LA with large HDL may be another potential mechanism for explaining a protective role of LA against CHD risk. Until now, only a few clinical trials have examined the effect of LA on large HDL particles (17, 18). In the clinical trials, nonsignificantly higher concentration of large HDL particles were observed after 5 or 12 wk with an LA-enriched diet than with a baseline diet consisting of a relatively high SFA content or with a diet high in stearic acid or in oleic acid (17, 18). Compared with our positive finding, those nonsignificant findings may have been attributable to a very small sample size, significantly unreached LA concentrations after the end of the treatment, and different background diets. Furthermore, the mechanism linking LA and HDL particle redistribution is unknown.
The present study noted that serum LA was associated, significantly and inversely, with large VLDL, small LDL, and small HDL particle concentrations. Higher large VLDL and small LDL particles may be related to a higher prevalence of coronary calcification (16). There was little evidence of an association between LA and VLDL. One crossover study reported benefits of an LA-enriched diet (ie, sunflower-oil diet) in lowering concentrations of triglycerides and VLDL triglycerides compared with a diet high in SFAs (a baseline diet) and compared with a MUFA-enriched diet (ie, rapeseed-oil diet) (35). The potential mechanisms for the association of LA with total and large VLDL, small LDL, and small HDL particles are unclear. However, an increase in serum LA may be related to the regulation of VLDL overproduction induced by increases in plasma free fatty acids, which leads to an improvement in an insulin-resistant phenotype of elevations in large VLDL, small LDL, and small HDL particles (36–39). LA has been known to improve insulin resistance (40), possibly by directing excess fatty acids toward triacylglycerol storage (41).
In our study, serum AA was significantly and inversely associated with large VLDL particle concentration and positively associated with large HDL particle concentration. These associations were consistent across all population groups. In the human body, the serum concentration of AA can be increased not only with an AA-enriched diet (ie, yolk, lean meat, or fish) but also with endogenous synthesis through desaturation and elongation of LA (28, 29). AA is an important component of cellular membranes. The free AA released from cellular membranes serves as a substrate for the production of eicosanoids, including prostaglandins (PGs). The eicosanoids have well-established roles in many pathologic processes, including thrombosis and inflammation (8, 42). However, lipid metabolism associated with AA was unknown. Recently, in an animal study, Oikawa et al (2009) reported that coadministration of AA with conjugated linoleic acid (CLA) in mice suppressed CLA-induced accumulation of triacylglycerol contents in liver. The authors suggested that AA or its eicosanoid derivatives, PGE2, prevents the gene expression of Spot 14 (a lipogenesis-related nuclear protein) and fatty acid synthase in hepatocytes that induces lipid production in liver (43, 44). However, in human studies, the favorable role of an AA-enriched diet in lipid metabolism has not been supported (28, 29). The discrepancy may possibly be attributed to a decreasing effect of an AA-enriched diet on serum concentrations of LA in the human body (28, 29). The decreased concentrations of serum LA in the clinical studies may have balanced the beneficial effect of AA on lipoproteins because LA has a potent effect on lipid metabolism.
Additionally, we examined population-specific associations of n−3 fatty acids with lipoprotein subclasses (see Supplemental Tables 4ndash6 under “Supplemental data” in the online issue). However, the data for n−3 fatty acids was not able to be pooled, because the distribution of marine-derived n−3 fatty acids considerably differed by population (see Supplemental Figure 1 under “Supplemental data” in the online issue). For example, concentrations of DHA barely overlapped between the Japanese men in Japan and the US white men. The 25th percentile of DHA (4.77%) in the Japanese men in Japan was much higher than the 75th percentile of DHA (2.99%) in US white men.
We also examined population-specific associations of MUFA and SFA with lipoprotein subclasses (see Supplemental Tables 7ndash9 under “Supplemental data” in the online issue). In previous clinical trials, an MUFA-rich diet has been noted to have lipid-lowering effects comparable with those of a PUFA-rich diet (35, 45). In our cross-sectional study, the associations of MUFA with lipoprotein subclasses contrasted somewhat with our expectations: an increase in serum MUFAs had significant associations with a decrease in large HDL particle concentration and an increase in large VLDL particle concentration across all population groups. One possible explanation of the results may be that serum concentrations of MUFA and SFA are unlikely to serve as biomarkers of their dietary intakes, which are endogenously produced in the human body (46). For example, stearic acid (a type of SFA) can be converted to oleic acid (a type of MUFA) (46, 47). The other possible explanation may be that an increased proportion of MUFAs may result in a decreased proportion of SFAs and n−3 fatty acids after the adjustment for n−6 PUFAs and trans fatty acids in the analyses of our study as a percentage of total fatty acids. Previous clinical studies have shown that, given constant SFAs, an MUFA-rich diet may not have any advantage with respect to HDL particle concentration (48), whereas, given the substitution of SFAs, the MUFA-rich diet may decrease large HDL particle concentration (35). In this context, the negative associations of serum MUFAs with large HDL particle concentrations in our data may be expected. In addition, the positive associations between MUFAs and large VLDL particle concentration were no longer significant after further adjustment for n−3 fatty acids (data not shown).
The strengths of this study were that we investigated lipoprotein subclass associations in a large population by using serum concentrations of n−6 fatty acids regarded as biomarkers of dietary intakes, which can show robust findings. However, the study also had several limitations. We did not collect dietary data (eg, total energy intake); thus, some confounding factors may remain even after adjustment. Because of a lack of data, we were unable to observe correlations of serum n−6 PUFAs with dietary n−6 PUFA intake. However, epidemiologic studies have reported that dietary PUFAs (75% of LA) are correlated with serum PUFAs (r = 0.50, P < 0.001 in the KIHD study) (5). In this regard, our data may be of significance to elucidate the role of LA in natural settings (ie, habitual diet) and not observed in clinical settings. Another limitation includes the cross-sectional nature of the study design. The method of measuring the serum fatty acids used in this design cannot determine the temporal relations for lipoproteins, eg, triglyceride-rich lipoproteins (ie, large VLDLs). When expressed either as a percentage of total fatty acids or as the weight of the fatty acids (mg/dL), LA and AA may be lower in triglycerides than in cholesterol esters and phospholipids (27). Thus, the reduced concentration of large VLDL particles could raise serum LA and AA concentrations. Finally, our study population of men aged 40–49 y limited our ability to generalize the results to women or other age groups.
In conclusion, we found that, in a population-based sample of men aged 40–49 y, higher serum LA and AA concentrations were significantly associated with a lower concentration of large VLDL particles and with a higher concentration of large HDL particles. More importantly, a higher serum LA concentration was significantly associated with a lower concentration of total LDL particles—a strong atherogenic lipoprotein.
Supplementary Material
Acknowledgments
The authors’ responsibilities were as follows—JC and AS: developed the hypothesis of this study and prepared the manuscript draft; JC, AS, HU, JDC, CS, RWE, AE-S, TK, and TO: involved in data collection; JC, AS, LHK, KS-T, DE, and RDA: led the data analyses; and AS, HU, RWE, KN, TO, KM, AE-S, RDA, KS-T, DE, and LHK: provided expert consultation on data interpretation. All authors were involved in the review and revision of the manuscript and gave final approval of the version to be published. None of the authors had a conflict of interest.
REFERENCES
- 1.Harris WS. Omega-3 fatty acids and cardiovascular disease: a case for omega-3 index as a new risk factor. Pharmacol Res 2007;55:217–23 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Hu FB, Stampfer MJ, Manson JE, et al. Dietary fat intake and the risk of coronary heart disease in women. N Engl J Med 1997;337:1491–9 [DOI] [PubMed] [Google Scholar]
- 3.Iso H, Sato S, Umemura U, et al. Linoleic acid, other fatty acids, and the risk of stroke. Stroke 2002;33:2086–93 [DOI] [PubMed] [Google Scholar]
- 4.Oh K, Hu FB, Manson JE, Stampfer MJ, Willett WC. Dietary fat intake and risk of coronary heart disease in women: 20 years of follow-up of the nurses’ health study. Am J Epidemiol 2005;161:672–9 [DOI] [PubMed] [Google Scholar]
- 5.Laaksonen DE, Nyyssonen K, Niskanen L, Rissanen TH, Salonen JT. Prediction of cardiovascular mortality in middle-aged men by dietary and serum linoleic and polyunsaturated fatty acids. Arch Intern Med 2005;165:193–9 [DOI] [PubMed] [Google Scholar]
- 6.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 [DOI] [PubMed] [Google Scholar]
- 7.Gardner CD, Kraemer HC. Monounsaturated versus polyunsaturated dietary fat and serum lipids. A meta-analysis. Arterioscler Thromb Vasc Biol 1995;15:1917–27 [DOI] [PubMed] [Google Scholar]
- 8.Harris WS, Mozaffarian D, Rimm E, et al. Omega-6 fatty acids and risk for cardiovascular disease: a science advisory from the American Heart Association Nutrition Subcommittee of the Council on Nutrition, Physical Activity, and Metabolism; Council on Cardiovascular Nursing; and Council on Epidemiology and Prevention. Circulation 2009;119:902–7 [DOI] [PubMed] [Google Scholar]
- 9.Willett WC. The role of dietary n−6 fatty acids in the prevention of cardiovascular disease. J Cardiovasc Med (Hagerstown) 2007;8(suppl 1):S42–5 [DOI] [PubMed] [Google Scholar]
- 10.Miura K, Stamler J, Nakagawa H, et al. Relationship of dietary linoleic acid to blood pressure. The International Study of Macro-Micronutrients and Blood Pressure Study. Hypertension 2008;52:408–14 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Mora S, Otvos JD, Rifai N, Rosenson RS, Buring JE, Ridker PM. Lipoprotein particle profiles by nuclear magnetic resonance compared with standard lipids and apolipoproteins in predicting incident cardiovascular disease in women. Circulation 2009;119:931–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kuller L, Arnold A, Tracy R, et al. Nuclear magnetic resonance spectroscopy of lipoproteins and risk of coronary heart disease in the cardiovascular health study. Arterioscler Thromb Vasc Biol 2002;22:1175–80 [DOI] [PubMed] [Google Scholar]
- 13.Blake GJ, Otvos JD, Rifai N, Ridker PM. Low-density lipoprotein particle concentration and size as determined by nuclear magnetic resonance spectroscopy as predictors of cardiovascular disease in women. Circulation 2002;106:1930–7 [DOI] [PubMed] [Google Scholar]
- 14.Cromwell WC, Otvos JD, Keyes MJ, et al. LDL particle number and risk of future cardiovascular disease in the Framingham Offspring Study—implications for LDL management. J Clin Lipidol 2007;1:583–92 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Mora S, Szklo M, Otvos JD, et al. LDL particle subclasses, LDL particle size, and carotid atherosclerosis in the Multi-Ethnic Study of Atherosclerosis (MESA). Atherosclerosis 2007;192:211–7 [DOI] [PubMed] [Google Scholar]
- 16.Mackey RH, Kuller LH, Sutton-Tyrrell K, Evans RW, Holubkov R, Matthews KA. Lipoprotein subclasses and coronary artery calcium in postmenopausal women from the healthy women study. Am J Cardiol 2002;90:71i–6i [DOI] [PubMed] [Google Scholar]
- 17.Wilkinson P, Leach C, Ah-Sing EE, et al. Influence of alpha-linolenic acid and fish-oil on markers of cardiovascular risk in subjects with an atherogenic lipoprotein phenotype. Atherosclerosis 2005;181:115–24 [DOI] [PubMed] [Google Scholar]
- 18.Thijssen MA, Mensink RP. Small differences in the effects of stearic acid, oleic acid, and linoleic acid on the serum lipoprotein profile of humans. Am J Clin Nutr 2005;82:510–6 [DOI] [PubMed] [Google Scholar]
- 19.Motoyama KR, Curb JD, Kadowaki T, et al. Association of serum n−6 and n−3 polyunsaturated fatty acids with lipids in 3 populations of middle-aged men. Am J Clin Nutr 2009;90:49–55 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Sekikawa A, Curb JD, Ueshima H, et al. Marine-derived n−3 fatty acids and atherosclerosis in Japanese, Japanese-American, and white men: a cross-sectional study. J Am Coll Cardiol 2008;52:417–24 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sekikawa A, Ueshima H, Kadowaki T, et al. Less subclinical atherosclerosis in Japanese men in Japan than in white men in the United States in the post-World War II birth cohort. Am J Epidemiol 2007;165:617–24 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Abbott RD, Ueshima H, Rodriguez BL, et al. Coronary artery calcification in Japanese men in Japan and Hawaii. Am J Epidemiol 2007;166:1280–7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Otvos JD. Measurement of lipoprotein subclass profiles by nuclear magnetic resonance spectroscopy. Clin Lab 2002;48:171–80 [PubMed] [Google Scholar]
- 24.Sekikawa A, Ueshima H, Sutton-Tyrrell K, et al. Intima-media thickness of the carotid artery and the distribution of lipoprotein subclasses in men aged 40 to 49 years between whites in the United States and the Japanese in Japan for the ERA JUMP study. Metabolism 2008;57:177–82 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Freedman DS, Otvos JD, Jeyarajah EJ, et al. Sex and age differences in lipoprotein subclasses measured by nuclear magnetic resonance spectroscopy: the Framingham Study. Clin Chem 2004;50:1189–200 [DOI] [PubMed] [Google Scholar]
- 26.Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Physiol 1959;37:911–7 [DOI] [PubMed] [Google Scholar]
- 27.Yeh LLL, Bunker CH, Evans RW, Kuller LH. Serum fatty acid distributions in postmenopausal women. Nutr Res 1994;14:675–91 [Google Scholar]
- 28.Nelson GJ, Schmidt PC, Bartolini G, et al. The effect of dietary arachidonic acid on plasma lipoprotein distributions, apoproteins, blood lipid levels, and tissue fatty acid composition in humans. Lipids 1997;32:427–33 [DOI] [PubMed] [Google Scholar]
- 29.Kusumoto A, Ishikura Y, Kawashima H, Kiso Y, Takai S, Miyazaki M. Effects of arachidonate-enriched triacylglycerol supplementation on serum fatty acids and platelet aggregation in healthy male subjects with a fish diet. Br J Nutr 2007;98:626–35 [DOI] [PubMed] [Google Scholar]
- 30.Harris WS, Poston WC, Haddock CK. Tissue n−3 and n−6 fatty acids and risk for coronary heart disease events. Atherosclerosis 2007;193:1–10 [DOI] [PubMed] [Google Scholar]
- 31.Barter PJ, Ballantyne CM, Carmena R, et al. Apo B versus cholesterol in estimating cardiovascular risk and in guiding therapy: report of the thirty-person/ten-country panel. J Intern Med 2006;259:247–58 [DOI] [PubMed] [Google Scholar]
- 32.Otvos JD, Collins D, Freedman DS, et al. Low-density lipoprotein and high-density lipoprotein particle subclasses predict coronary events and are favorably changed by gemfibrozil therapy in the Veterans Affairs High-Density Lipoprotein Intervention Trial. Circulation 2006;113:1556–63 [DOI] [PubMed] [Google Scholar]
- 33.Mustad VA, Ellsworth JL, Cooper AD, Kris-Etherton PM, Etherton TD. Dietary linoleic acid increases and palmitic acid decreases hepatic LDL receptor protein and mRNA abundance in young pigs. J Lipid Res 1996;37:2310–23 [PubMed] [Google Scholar]
- 34.Rosenson RS, Otvos JD, Freedman DS. Relations of lipoprotein subclass levels and low-density lipoprotein size to progression of coronary artery disease in the Pravastatin Limitation of Atherosclerosis in the Coronary Arteries (PLAC-I) trial. Am J Cardiol 2002;90:89–94 [DOI] [PubMed] [Google Scholar]
- 35.Valsta LM, Jauhiainen M, Aro A, Katan MB, Mutanen M. Effects of a monounsaturated rapeseed oil and a polyunsaturated sunflower oil diet on lipoprotein levels in humans. Arterioscler Thromb 1992;12:50–7 [DOI] [PubMed] [Google Scholar]
- 36.Ginsberg HN, Huang LS. The insulin resistance syndrome: impact on lipoprotein metabolism and atherothrombosis. J Cardiovasc Risk 2000;7:325–31 [DOI] [PubMed] [Google Scholar]
- 37.Despres JP. Is visceral obesity the cause of the metabolic syndrome? Ann Med 2006;38:52–63 [DOI] [PubMed] [Google Scholar]
- 38.Rosenson RS. Clinical role of LDL and HDL subclasses and apolipoprotein measurement. ACC Curr J Rev 2004;13:33–7 [Google Scholar]
- 39.Garvey WT, Kwon S, Zheng D, et al. Effects of insulin resistance and type 2 diabetes on lipoprotein subclass particle size and concentration determined by nuclear magnetic resonance. Diabetes 2003;52:453–62 [DOI] [PubMed] [Google Scholar]
- 40.Summers LK, Fielding BA, Bradshaw HA, et al. Substituting dietary saturated fat with polyunsaturated fat changes abdominal fat distribution and improves insulin sensitivity. Diabetologia 2002;45:369–77 [DOI] [PubMed] [Google Scholar]
- 41.Lee JS, Pinnamaneni SK, Eo SJ, et al. Saturated, but not n−6 polyunsaturated, fatty acids induce insulin resistance: role of intramuscular accumulation of lipid metabolites. J Appl Physiol 2006;100:1467–74 [DOI] [PubMed] [Google Scholar]
- 42.Seyberth HW, Oelz O, Kennedy T, et al. Increased arachidonate in lipids after administration to man: effects on prostaglandin biosynthesis. Clin Pharmacol Ther 1975;18:521–9 [DOI] [PubMed] [Google Scholar]
- 43.Oikawa D, Tsuyama S, Akimoto Y, Mizobe Y, Furuse M. Arachidonic acid prevents fatty liver induced by conjugated linoleic acid in mice. Br J Nutr 2009;101:1558–63 [DOI] [PubMed] [Google Scholar]
- 44.Kinlaw WB, Quinn JL, Wells WA, Roser-Jones C, Moncur JT. Spot 14: a marker of aggressive breast cancer and a potential therapeutic target. Endocrinology 2006;147:4048–55 [DOI] [PubMed] [Google Scholar]
- 45.Mensink RP, Katan MB. Effect of a diet enriched with monounsaturated or polyunsaturated fatty acids on levels of low-density and high-density lipoprotein cholesterol in healthy women and men. N Engl J Med 1989;321:436–41 [DOI] [PubMed] [Google Scholar]
- 46.Arab L. Biomarkers of fat and fatty acid intake. J Nutr 2003;133(suppl 3):925S–32S [DOI] [PubMed] [Google Scholar]
- 47.Kobayashi M, Sasaki S, Kawabata T, Hasegawa K, Akabane M, Tsugane S. Single measurement of serum phospholipid fatty acid as a biomarker of specific fatty acid intake in middle-aged Japanese men. Eur J Clin Nutr 2001;55:643–50 [DOI] [PubMed] [Google Scholar]
- 48.Dreon DM, Vranizan KM, Krauss RM, Austin MA, Wood PD. The effects of polyunsaturated fat vs monounsaturated fat on plasma lipoproteins. JAMA 1990;263:2462–6 [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.