Effect of dietary modification by calorie restriction on cholesterol levels in lipoprotein(a) and other lipoprotein classes

Abstract

Background

Dietary habits are associated with obesity which is a risk factor for coronary heart disease. The objective is to estimate the change of lipoprotein(a) and other lipoprotein classes by calorie restriction with obesity index and Framingham risk score.

Methods

Sixty females (56 ± 9 years) were recruited. Their caloric intakes were reduced during the six-month period, and the calorie from fat was not more than 30%. Lipoprotein profiles were estimated at baseline and after the six-month period of calorie restriction. Cholesterol levels in six lipoprotein classes (HDL, LDL, IDL, VLDL, chylomicron and lipoprotein(a)) were analysed by anion-exchange liquid chromatography. The other tests were analysed by general methods. Additionally, Framingham risk score for predicting 10-year coronary heart disease risk was calculated.

Results

Body mass index, waist circumference, insulin resistance, Framingham risk score, total cholesterol, LDL-cholesterol and IDL-cholesterol were significantly decreased by the calorie restriction, and the protein and cholesterol levels of lipoprotein(a) were significantly increased. The change of body mass index was significantly correlated with those of TC, VLDL-cholesterol and chylomicron-cholesterol, and that of waist circumference was significantly correlated with that of chylomicron-cholesterol. The change of Framingham risk score was significantly correlated with the change of IDL-C.

Conclusion

Obesity indexes and Framingham risk score were reduced by the dietary modification. Lipoprotein profile was improved with the reduction of obesity indexes, but lipoprotein(a) was increased. The changes of obesity indexes and Framingham risk score were related with those of triglyceride-rich lipoproteins, e.g. IDL, VLDL and CM.

Keywords

Lipoprotein dietary modification exchange liquid chromatography

Introduction

Numerous clinical studies indicated that increased LDL cholesterol (LDL-C) and triglyceride (TG) and decreased HDL cholesterol (HDL-C) were risk factors for coronary heart disease (CHD).^1–5 Recently, several clinical studies showed that non-HDL-C has more predictive potential to estimate CHD risk than LDL-C.^6–8 The guideline for National Cholesterol Education Program (NCEP)-Adult Treatment Panel-III indicates that LDL-C, HDL-C and TG are important targets for primary and secondary prevention of CHD, and non-HDL-C is an established secondary therapeutic target for CHD prevention.⁹ The non-HDL contains LDL and TG-rich-lipoproteins, i.e. IDL, VLDL and chylomicron (CM). The VLDL cholesterol (VLDL-C) was indicated to be associated with CHD risk in Framingham Heart Study, independently of LDL-C.¹⁰ The association between IDL-C and the severity of CHD was shown.¹¹

Body weight (BW), body mass index (BMI) and waist circumference (WC) are reduced by the change of lifestyle, diet and exercise.^12–14 The lifestyle change improves lipid and lipoprotein profiles.^12–14 BMI and WC of type 2 diabetic patients with insulin resistance (HOMA) are higher than those with insulin sensitivity, and the insulin-resistant patients have decreased HDL-C and increased VLDL-C.¹⁵ We thought that BMI, WC, HOMA, LDL-C, TG-rich lipoproteins are decreased, and HDL-C is increased with decrease of CHD risk by improvement of dietary composition. Increased lipoprotein(a) (Lp[a]) concentration is associated with the risk of CHD.^16–18 Diabetes was reported to increase the risk of atherosclerotic diseases, e.g. CHD, intermittent claudication.¹⁹ However, it has been reported that Lp(a) concentration was associated inversely with the risk of type 2 diabetes.²⁰ Therefore, the change of Lp(a) concentration with the improvement of dietary composition could not be expected. Furthermore, in most of the studies, the Lp(a) protein concentration is estimated, and there are only a few studies that include the estimation of Lp(a) cholesterol.

HDL, LDL, IDL, VLDL, CM and Lp(a) in serum can be separated by anion-exchange chromatography (AEX-HPLC).^21,22 Lipoprotein profile can be accurately and conveniently estimated by AEX-HPLC. We studied the changes of cholesterol concentrations in Lp(a) and other lipoprotein classes obtained by AEX-HPLC in females with the dietary modification by calorie restriction, and the relationship between the change of BMI, WC, HOMA and Framingham risk score (FRS) and the change of the lipoprotein profiles.

Methods

Subjects and study protocol

Sixty Japanese females (age: 56 ± 9 years, postmenopausal females; n = 46) were recruited for the healthy diet course of the Nutrition Clinic of Kagawa Nutrition University, Japan. In the healthy diet course during the six-month period, patients’ lifestyle was improved by using nutrition education session and individual guidance. The volunteers were divided into obese group (BMI >25 kg/cm², n = 37, postmenopausal females; n = 29) and non-obese group (BMI ≦25 kg/cm², n = 23, postmenopausal females; n = 17). All patients did not have any medication, smoking and heavy drinking. At entry, written informed consent was obtained from all volunteers. The study protocol was approved by the ethical committees of Kagawa Nutrition University (No. 222-G) and Tosoh Corporation (No. 10-02).

All volunteers attended nutrition education session and individual guidance during the six-month period, which were provided once a week for the first one month and once two weeks after that. In the education session, we showed the target of dietary composition as follows: caloric intake from diet is 1600 kcal/day, and the percentages of calories from protein, fat and carbohydrate are 13–20%, 20–30% and 55–65%, respectively. The target of the caloric intake was reduced 16% from the caloric intake 1900 kcal/day of females (50–69 years) reported in 2015 by Japanese Ministry of Health, Labour and Welfare (JMHLF). The percentages of calories were also according to the dietary reference intake of females (50–69 years) reported in 2015 by JMHLF. In the individual guidances, the records of eaten foods in each day by themselves were checked. The check points were reduction of caloric intake from diet and percentages of calories from fat as follows: the caloric intake <1600 kcal/day, <1800 kcal/day, <1900 kcal/day, <2000 kcal/day and ≥2000 kcal/day in the start point of study are reduced by >0%, ≥5%, ≥10%, ≥15% and ≥20%, respectively, and the percentages of calories from fat is not more than 30%. All volunteers could meet the demand of the check points during the study period, and did not drop out.

Nutrient intake from three-day food records and 24-h dietary recalls was assessed using a computerized dietary analysis system (Excell-EIYO Ver.3, Kenpaku Crop.) at two points: the start of study and after the six-month period (Table 1). The reduced ratio of caloric intake and the percentages of calories from fat in the volunteers (n = 18) of caloric intake <1600 kcal/day in the start point of study were −8 ± 9% and 23.6 ± 3.8%, respectively. Those in caloric intake <1800 kcal/day (n = 25), <1900 kcal/day (n = 6), <2000 kcal/day (n = 4) and ≥2000 kcal/day (n = 7) were −17 ± 15% and 24.8 ± 4.9%, −19 ± 8% and 23.6 ± 3.4%, −18 ± 5% and 26.7 ± 2.3%, and −29 ± 11% and 26.7 ± 3.8%, respectively. Blood samples were obtained from antecubital vein after an overnight fast of at least 12 h.

Table 1.

Comparison between baseline diet composition and diet composition after six months of lifestyle intervention in females.

	Total (n = 60)				Obese group (n = 37)				Non-obese group (n = 23)
	Baseline data	After six months	Change value	P	Baseline data	After six months	Change value	P	Baseline data	After six months	Change value	P*
Energy
Total energy (kJ)	1750 ± 387	1444 ± 308	−283 ± 322	<0.0001	1773 ± 430	1452 ± 198	−321 ± 363	<0.0001	1713 ± 310	1431 ± 307	−282 ± 231	<0.01
Carbohydrate (%)	55.5 ± 6.4	57.4 ± 5.7	2.3 ± 6.1	<0.01	54.6 ± 7.0	56.3 ± 5.0	1.8 ± 7.1	NS	56.9 ± 5.0	60.1 ± 4.5	3.2 ± 4.0	<0.01
Fat (%)	27.0 ± 5.1	24.7 ± 4.2	−2.3 ± 5.6	<0.005	27.1 ± 5.4	25.8 ± 3.8	−1.3 ± 6.3	NS	26.8 ± 4.8	22.9 ± 4.2	−3.8 ± 3.9	<0.01
Protein (%)	15.7 ± 2.1	16.6 ± 2.1	0.9 ± 2.3	<0.005	15.7 ± 2.2	16.9 ± 1.8	1.2 ± 2.4	<0.05	15.7 ± 1.9	16.6 ± 2.5	0.9 ± 2.5	<0.05
Fat (g)
Saturated fatty acid	14.7 ± 5.8	10.6 ± 5.0	−3.9 ± 5.6	<0.0001	15.3 ± 6.3	11.3 ± 4.3	−3.9 ± 6.0	<0.001	13.2 ± 4.6	9.4 ± 3.4	5.9 ± 1.7	<0.01
Monounsaturated fatty acid	19.0 ± 3.9	14.4 ± 6.0	−4.5 ± 6.3	<0.0001	19.3 ± 7.7	15.4 ± 4.6	−3.9 ± 7.4	<0.005	18.3 ± 5.5	12.9 ± 4.8	−5.4 ± 3.9	<0.01
Polyunsaturated fatty acid	11.4 ± 4.2	9.8 ± 3.7	−1.7 ± 5.4	<0.005	11.6 ± 4.7	10.5 ± 4.9	−1.1 ± 6.3	NS	11.1 ± 3.2	8.6 ± 2.9	−2.5 ± 3.5	<0.01
n-3 polyunsaturated fatty acid	2.2 ± 1.0	1.7 ± 0.9	−0.5 ± 1.2	<0.005	2.1 ± 1.0	1.8 ± 0.8	−0.3 ± 1.1	<0.0001	2.2 ± 1.0	1.5 ± 0.8	−0.8 ± 1.2	<0.01
n-6 polyunsaturated fatty acid	9.2 ± 3.4	8.1 ± 4.1	−1.2 ± 4.7	<0.05	9.5 ± 3.9	8.7 ± 4.3	−0.8 ± 5.5	<0.0001	8.9 ± 2.5	7.1 ± 2.5	−1.8 ± 2.9	<0.01
Fat (%)
Saturated fatty acid (%)	32.1 ± 6.6	30.3 ± 6.2	−1.8 ± 8.6	NS	33.0 ± 6.5	30.2 ± 6.8	−2.6 ± 8.8	NS	30.7 ± 6.7	30.4 ± 5.2	−0.4 ± 8.4	NS
Monounsaturated fatty acid (%)	42.1 ± 4.7	49.7 ± 20.0	7.6 ± 20.0	<0.05	41.6 ± 4.3	53.0 ± 18.9	11.4 ± 19.2	<0.005	42.9 ± 5.3	44.3 ± 21.2	1.4 ± 20.1	NS
Polyunsaturated fatty acid (%)	25.8 ± 5.8	28.4 ± 7.7	2.6 ± 8.9	<0.05	25.4 ± 6.5	28.4 ± 8.8	3.0 ± 10.3	NS	26.4 ± 4.4	28.2 ± 5.6	1.8 ± 6.3	NS
N-3/n-6 polyunsaturated fatty acid ratio	0.24 ± 0.09	0.22 ± 0.10	−0.02 ± 0.12	NS	0.23 ± 0.08	0.23 ± 0.11	0.00 ± 0.00	NS	0.25 ± 0.09	0.21 ± 0.09	−0.04 ± 0.13	NS
Cholesterol (mg)	311 ± 145	273 ± 125	−38 ± 138	<0.05	323 ± 153	284 ± 117	−39 ± 140	<0.0001	292 ± 134	256 ± 107	−36 ± 138	NS
Sodium chloride (g)	8.7 ± 2.0	7.7 ± 2.1	−1.1 ± 2.5	<0.005	8.8 ± 2.2	7.7 ± 2.2	−1.1 ± 2.8	<0.05	8.5 ± 1.7	7.5 ± 2.1	−1.0 ± 2.0	<0.05
Dietary fiber (g)	15.7 ± 5.4	18.5 ± 7.0	2.9 ± 6.5	<0.005	14.4 ± 4.1	17.5 ± 6.3	3.1 ± 6.8	<0.01	17.6 ± 6.7	20.2 ± 7.9	2.6 ± 6.2	<0.05

Not significant is indicated by NS. *p value is decided on P < 0.05 or P < 0.01 by using Wilcoxon-test table.

Measurement

GA08 (A and T Corporation), HLC-723G9 (Tosoh Corporation), Chemilumi-Insulin-Centaur (Siemens Healthcare Diagnostics) and Cholestest-CHO, Cholestest-NHDL and Cholestest-TG(Sekisui Medical), and Lp(a)-latex and sd-LDL-C (Denka Seiken) were used to measure fasting blood glucose (FBG), glycated haemoglobin A1c (A1c), Insulin, total cholesterol (TC), HDL-C, TG, Lp(a) protein (Lp(a)-P) and small dense LDL cholesterol (sdLDL-C), respectively. LDL-C was calculated by Friedewald equation (LDL-C = TC−[HDL-C]−[TG÷5]). Non-HDL-C was calculated using the equation: (non-HDL-C = TC−[HDL-C]). Insulin resistance (HOMA) was calculated by insulin and glucose concentrations in fasting blood: HOMA = insulin concentration × glucose concentration÷405.²³ FRS for predicting 10-year CHD risk was calculated by using gender, age, LDL-C, HDL-C, blood pressure, diabetes and smoking details.²⁴

Cholesterol concentrations of six lipoprotein classes (HDL, LDL, IDL, VLDL, CM, and Lp[a]) were measured by using AEX-HPLC,²² which consists of a column filled with diethylaminoethyl-ligand non-porous gel, two eluents for step gradient of sodium percorolate, and a reagent containing cholesterol esterase and cholesterol oxidase for postcolumn reaction.

Statistical analysis

The data were presented as mean ± standard deviation (SD). Wilcoxon test was used to compare between baseline data and that after six months. Non-obese group was small (n = 23), and therefore P value was decided on P < 0.05 or P < 0.01 by using the Wilcoxon test t table. Correlations were estimated by Spearman’s rank test.

Results

Data of baseline and after six months of dietary modification by calorie restriction are shown in Table 2. BMI, BW, WC, blood pressures, FGB, A1c, insulin, HOMA, FRS, TC, non-HDL-C, sdLDL-C, LDL-C and IDL-C were significantly decreased by the dietary modification in total and obese groups, and the protein and the cholesterol concentration of Lp(a) were significantly increased. In the non-obese group, BMI, BW, WC, systolic blood pressure, A1c and IDL-C were significantly decreased by the dietary modification, and the cholesterol concentrations of Lp(a) were significantly increased. The concentration of HDL was significantly decreased in the only obese group. TG, VLDL-C and chylomicron cholesterol CM-C did not significantly change in all groups, and CM-C was lower than cholesterol concentrations in other lipoproteins.

Table 2.

Comparison between baseline data and data after six months of lifestyle intervention in females.

	Total (n=60)				Obese group (n=37)				Non-obese group (n=23)
	Baseline data	After 6 months	Change value	P	Baseline data	After 6 months	Change value	P	Baseline data	After 6 months	Change value	P*
Basic data
Age (years)	56 ± 9				57 ± 8				56 ± 10
Body mass index (kg/cm²)	26.0 ± 4.0	24.7 ± 3.9	−1.3 ± 1.0	<0.0001	28.6 ± 2.6	27.0 ± 2.9	−1.5 ± 1.1	<0.0001	22.1 ± 1.6	21.1 ± 1.4	−1.0 ± 0.9	<0.01
Body weight (kg)	64.2 ± 9.8	61.0 ± 9.3	−3.3 ± 2.6	<0.0001	70.3 ± 6.5	66.5 ± 6.7	−3.8 ± 2.8	<0.0001	54.4 ± 5.0	52.0 ± 4.8	−2.3 ± 2.1	<0.01
Waist circumference (cm)	89.9 ± 10.1	86.3 ± 10.4	−3.5 ± 3.4	<0.0001	95.3 ± 7.7	92.3 ± 8.2	−3.0 ± 3.1	<0.0001	81.8 ± 6.3	77.5 ± 5.0	−4.3 ± 3.7	<0.01
Systolic blood pressure (mmHg)	129 ± 18	121 ± 15	−7 ± 13	<0.0001	129 ± 18	123 ± 14	−7 ± 13	<0.01	126 ± 17	118 ± 15	−8 ± 12	<0.01
Diastolic blood pressure (mmHg)	76 ± 13	70 ± 10	−2 ± 10	<0.005	77 ± 13	74 ± 10	−3 ± 10	<0.05	73 ± 12	72 ± 10	−1 ± 10	NS
Fasting blood glucose (mg/dL)	103 ± 40	97 ± 31	−6 ± 24	<0.0001	114 ± 50	107 ± 38	−8 ± 30	<0.0005	89 ± 8	87 ± 7	−2 ± 7	NS
Glycated haemoglobin A1c	5.5 ± 0.9	5.3 ± 0.6	−0.3 ± 0.4	<0.0001	5.9 ± 1.7	5.5 ± 1.5	−0.3 ± 0.5	<0.0005	5.6 ± 0.3	5.4 ± 0.3	−0.2 ± 0.3	<0.01
Insulin (×10⁻³U/L)	7.2 ± 6.3	5.7 ± 5.3	−1.7 ± 3.0	<0.0005	10.0 ± 7.0	7.8 ± 6.1	−2.5 ± 3.4	<0.0005	3.2 ± 2.0	2.7 ± 1.9	−0.5 ± 1.9	NS
Insulin resistance (HOMA)	1.9 ± 1.8	1.5 ± 1.6	−0.5 ± 1.1	<0.0005	2.7 ± 2.0	2.1 ± 1.9	−0.8 ± 1.3	<0.001	0.7 ± 0.5	0.6 ± 0.4	−0.1 ± 0.5	NS
Framingham risk score (%)	6.4 ± 4.6	5.1 ± 3.1	−1.3 ± 3.0	<0.0001	7.5 ± 5.1	5.5 ± 3.2	−2.1 ± 3.3	<0.0001	4.5 ± 3.0	4.4 ± 3.0	−0.1 ± 1.8	NS
Lipid data
Total cholesterol (mg/dL)	221 ± 36	212 ± 32	−9 ± 29	<0.05	218 ± 37	203 ± 29	−15 ± 27	<0.005	228 ± 30	227 ± 32	−1 ± 30	NS
Non-HDL cholesterol (mg/dL)	157 ± 35	148 ± 30	−9 ± 25	<0.01	158 ± 36	144 ± 28	−14 ± 25	<0.005	158 ± 31	157 ± 33	−1 ± 24	NS
Triglyceride (mg/dL)	113 ± 56	102 ± 47	−9 ± 50	NS	127 ± 60	113 ± 43	−14 ± 56	NS	95 ± 47	95 ± 51	1 ± 38	NS
Lp(a) protein (mg/dL)	14 ± 15	17 ± 17	3 ± 8	<0.01	15 ± 17	19 ± 20	4 ± 8	<0.05	13 ± 10	14 ± 11	1 ± 7	NS
Small dense LDL cholesterol (mg/dL)	35.8 ± 14.7	31.3 ± 11.1	−4.5 ± 11.1	<0.005	37.3 ± 16.2	31.2 ± 11.8	−6.1 ± 12.4	<0.01	34.6 ± 11.9	32.5 ± 9.9	−2.0 ± 8.5	NS
Lipoprotein data by anion-exchange liquid chromatography
HDL cholesterol (mg/dL)	63.2 ± 16.9	61.0 ± 16.3	−2.5 ± 10.9	NS	58.7 ± 14.1	54.5 ± 13.1	−4.2 ± 9.5	<0.05	70.1 ± 19.7	70.2 ± 17.6	0.1 ± 12.5	NS
LDL cholesterol (mg/dL)	142.4 ± 36.0	125.7 ± 31.6	−18.5 ± 33.5	<0.0005	142.7 ± 38.6	116.4 ± 26.7	−26.3 ± 33.2	<0.0001	147.0 ± 32.4	140.9 ± 35.1	−6.0 ± 30.6	NS
IDL cholesterol (mg/dL)	9.5 ± 4.7	8.3 ± 3.4	−1.3 ± 3.7	<0.05	9.9 ± 4.5	8.3 ± 3.4	−1.7 ± 3.5	<0.01	9.5 ± 5.2	8.8 ± 3.4	−0.8 ± 3.9	<0.01
VLDL cholesterol (mg/dL)	13.3 ± 8.7	13.8 ± 9.9	1.0 ± 10.2	NS	15.3 ± 9.7	15.3 ± 9.8	0.0 ± 10.8	NS	10.7 ± 6.6	13.3 ± 10.4	2.6 ± 9.4	NS
Chylomicron cholesterol (mg/dL)	0.5 ± 0.1	0.5 ± 0.3	0.0 ± 0.3	NS	0.5 ± 0.2	0.6 ± 0.3	0.0 ± 0.3	NS	0.5 ± 0.1	0.5 ± 0.3	0.1 ± 0.3	NS
Lp(a) cholesterol (mg/dL)	3.2 ± 2.8	3.5 ± 3.2	0.3 ± 0.8	<0.05	3.6 ± 3.2	3.9 ± 3.6	0.3 ± 0.8	<0.05	2.7 ± 2.0	2.9 ± 2.4	0.2 ± 0.8	<0.01

Note: P values were obtained from a paired t-test or Wilcoxon-test.

To convert cholesterol concentrations in mg/dL to mmol/L multiply by 0.0259.

To convert triglyceride concentrations in mg/dL to mmol/L multiply by 0.0113.

P value is decided on P < 0.05 or P < 0.01 by using the Wilcoxon-test t table.

The comparison between the changes of basic data and those of lipids and lipoproteins are shown in Tables 3 and 4, respectively. The changes of BMI and BW were significantly correlated with those of TC, non-HDL-C, sdLDL-C, VLDL-C and CM-C in total and non-obese groups. The change of WC was only significantly correlated with that of CM-C in total and non-obese groups. The changes of systolic and diastolic blood pressures were significantly correlated with those of CM-C and IDL-C, respectively, in total and obese groups. The change of A1c was significantly correlated with those of TG and VLDL-C in total and was significantly correlated with that of TG in the obese group. The changes of insulin and HOMA were significantly correlated with those of TG and VLDL-C in total and non-obese groups. The change of FRS was significantly correlated with IDL-C in total and obese groups and was significantly correlated with those of VLDL-C and TG in non-obese groups. The change of caloric intake was significantly correlated with that of IDL-C in total and non-obese groups and was significantly correlated with that of TG in the obese group.

Table 3.

Comparison between changes in basic data and lipid profile in females.

	Regression coefficient
	Total caloric intake	Total cholesterol	Non-HDL cholesterol	Triglyceride	Lp(a) protein	Small dense LDL cholesterol
Total (n=60)
Body mass index	0.0408	0.3960***	0.4065***	0.2204	−0.1223	0.3376*
Body weight	0.0567	0.3918***	0.4008***	0.2072	−0.0987	0.3185*
Waist circumference	−0.0242	0.2123	0.1776	0.0701	0.0971	0.0739
Systolic blood pressure	0.0607	0.1271	0.0751	0.1004	0.0604	0.0761
Diastolic blood pressure	0.2658*	0.1980	0.2376	0.2210	−0.2021	0.1926
Fasting blood glucose	−0.0300	−0.0513	−0.0418	0.1605	−0.2024	−0.0002
Glycated haemoglobin A1c	0.1012	0.0442	0.0778	0.3270**	−0.1718	0.1407
Insulin	0.0715	0.1221	0.1754	0.2791*	−0.1380	0.1949
Insulin resistance (HOMA)	0.0669	0.1053	0.1518	0.2773*	−0.1200	0.1966
Framingham risk score	−0.1006	0.2190	0.2323	0.2201	−0.1356	0.0609
Total caloric intake		0.2373	0.1437	0.2201	−0.0997	0.2261
Obese group (n=37)
Body mass index	−0.0281	0.2463	0.2847	0.1586	−0.1215	0.1362
Body weight	−0.0114	0.2534	0.2818	0.1328	−0.1044	0.1260
Waist circumference	−0.1329	0.2002	0.1885	−0.0105	0.0200	−0.0343
Systolic blood pressure	0.1104	0.0274	−0.0746	−0.0740	0.0426	−0.1429
Diastolic blood pressure	0.1725	0.3148	0.2885	0.2184	−0.3486*	0.2811
Fasting blood glucose	−0.0527	−0.0908	−0.1310	0.0805	−0.1332	0.0200
Glycated haemoglobin A1c	0.2889	0.2068	0.1771	0.3878*	−0.1509	0.2060
Insulin	0.0300	−0.0078	−0.0453	0.2194	−0.0465	0.1288
Insulin resistance (HOMA)	0.0721	0.0039	−0.0246	0.2913	−0.0418	0.2073
Framingham risk score	−0.1250	0.1096	0.1215	−0.0241	−0.1074	−0.1753
Total caloric intake		0.0826	0.0681	0.3608*	−0.1043	0.2449
Non-obese group (n=23)
Body mass index	0.2730	0.6520***	0.6160**	0.3309	−0.0428	0.6894***
Body weight	0.2909	0.6342**	0.6295**	0.3550	−0.0419	0.6764***
Waist circumference	0.0818	0.3117	0.2392	0.0771	0.1776	0.2636
Systolic blood pressure	0.0424	0.3608	0.3622	0.3662	0.0604	0.5437
Diastolic blood pressure	0.4032	0.2494	0.3606	0.2679	0.0616	0.2212
Fasting blood glucose	0.0382	−0.1234	−0.0810	0.3637	−0.3150	−0.0877
Glycated haemoglobin A1c	−0.1901	−0.3540	−0.2804	0.2595	−0.2077	−0.0822
Insulin	0.4166*	0.1394	0.2637	0.5548**	−0.3473	0.2083
Insulin resistance (HOMA)	0.3637	0.1273	0.2479	0.5364**	−0.3504	0.1704
Framingham risk score	−0.0483	0.2655	0.3230	0.5790**	−0.1723	0.3690
Total caloric intake		0.3039	0.3215	0.0870	−0.1070	0.1887

Note: Regression coefficient and P value were obtained from Spearman's rank test.

P < 0.05, P < 0.01, and P < 0.005 were indicated by *,** and ***, respectively.

Table 4.

Comparison between changes in basic data and lipid profile obtained by AEX-HPLC method in females.

	Regression coefficient
	HDL cholesterol	LDL cholesterol	IDL cholesterol	VLDL cholesterol	Chylomicron cholesterol	Lp(a) cholesterol
Total (n=60)
Body mass index	0.0787	0.1512	0.1197	0.3220*	0.3085*	0.1037
Body weight	0.0556	0.1435	0.1238	0.2990*	0.2756*	0.0686
Waist circumference	−0.0186	−0.0576	0.0166	0.1775	0.3200**	0.1919
Systolic blood pressure	−0.1126	−0.0878	0.0166	0.1775	0.3200**	0.1919
Diastolic blood pressure	−0.1339	0.0067	0.2941*	0.2433	0.1063	0.1154
Fasting blood glucose	0.1484	−0.1274	0.1316	0.1886	0.1248	0.1229
Glycated haemoglobin A1c	−0.2230	−0.2102	0.0096	0.3112*	−0.0607	−0.1578
Insulin	0.1434	0.1277	0.2372	0.3541***	0.1085	−0.0778
Insulin resistance (HOMA)	0.1648	0.0812	0.2475*	0.3446**	0.1037	−0.0391
Framingham risk score	−0.0988	0.1824	0.3135*	0.0970	−0.0657	−0.0162
Total caloric intake	−0.0112	−0.0498	0.2817*	0.2431	0.0499	0.0608
Obese group (n=37)
Body mass index	0.0452	0.3540	−0.0049	0.2172	0.1239	−0.0591
Body weight	0.0512	0.0527	0.0037	0.1807	0.1065	−0.0691
Waist circumference	0.0789	−0.0466	−0.0373	0.0829	0.1425	0.0106
Systolic blood pressure	−0.0249	−0.0260	0.1133	−0.1981	−0.3308*	−0.0737
Diastolic blood pressure	0.0333	0.2506	0.4249**	0.0941	−0.0174	0.1870
Fasting blood glucose	0.3450*	−0.1663	0.1848	0.1312	0.1534	0.3370
Glycated haemoglobin A1c	−0.1232	−0.1820	0.1210	0.3118	−0.1289	−0.1344
Insulin	0.4350**	−0.0457	0.1490	0.3369	0.0415	0.0334
Insulin resistance (HOMA)	0.4164*	−0.0992	0.2290	0.3696	0.0320	0.1031
Framingham risk score	−0.1332	0.2712	0.3655*	−0.1767	−0.2003	0.0329
Total caloric intake	−0.0568	−0.1129	0.2227	0.2479	−0.1301	−0.1081
Non-obese group (n=23)
Body mass index	0.0769	0.2405	0.3520	0.5039*	0.6717***	0.4711*
Body weight	−0.0129	0.1804	0.3752	0.5424**	0.6433***	0.4174*
Waist circumference	0.1185	0.1314	0.1114	0.3228	0.5128*	0.4724*
Systolic blood pressure	−0.0854	0.0413	0.1793	0.2391	0.0616	0.0079
Diastolic blood pressure	−0.2613	−0.0806	0.3168	0.4292*	0.2948	0.1128
Fasting blood glucose	−0.1668	−0.2048	−0.0546	0.2241	0.0433	−0.0886
Glycated haemoglobin A1c	−0.3268	−0.4436*	−0.1114	0.3748	0.0341	−0.0707
Insulin	−0.2383	−0.0054	0.2986	0.4909*	0.1242	−0.0656
Insulin resistance (HOMA)	−0.2230	0.0130	0.2617	0.4515*	0.1222	−0.0552
Framingham risk score	−0.2073	−0.0945	0.1787	0.5282**	0.0910	−0.1062
Total caloric intake	0.0035	0.0613	0.4538*	0.2194	0.3240	0.3284

Note: Regression coefficient and P value were obtained from Spearman's rank test.

P < 0.05, P < 0.01, and P < 0.005 were indicated by *,** and ***, respectively.

In comparison between the baseline data of basic data and those of lipids and lipoproteins, FRS and WC were highly correlated with TG and VLDL-C, and CM-C, respectively (P < 0.0005), in the total group. In comparison between data after six months of dietary modification of basic data and those of lipids and lipoproteins, FRS and insulin were highly correlated with HDL-C and TG, and CM-C, respectively (P < 0.0005), in the total group. The cholesterol concentrations of Lp(a) were correlated with the protein concentrations of Lp(a) (r = 0.901, y = 0.172 × +0.736, x; protein concentration, y; cholesterol concentration) (Figure 1). We found that the ratio of Lp(a) cholesterol (Lp[a]-C) to Lp(a)-P in baseline data (0.301 ± 0.191 cholesterol/mg) was similar to those in data after six months of dietary modification (0.313 ± 0.267 cholesterol/mg).

Figure 1.

Correlation of Lp(a) cholesterol with the protein concentration. Data of baseline and after six months of dietary modification were indicated by closed and open circles, respectively.

Discussion

We have estimated the change of lipid and lipoprotein profiles in Japanese females with six-month period of dietary modification by the calorie restriction and the change of caloric percentage from fat. Those changes during the six-month period were −283±322 kcal/day (−16.1±13.1%) and −2.3±5.6%, respectively, in the total group. Then, both BMI and WC decreased by 5% in the total group (Table 2). LDL-C, non-HDL-C and IDL-C decreased by 10%, 4% and 4%, respectively in the total group (Table 2). Additionally, those changes in the obese group were −321±363 kcal/day (−16.0±13.3%) and −1.3±6.3%, respectively. LDL-C, non-HDL-C and IDL-C decreased by 15%, 7% and 10%, respectively. The several previous papers showed the effect of diet, exercise and diet-plus-exercise on lipoprotein profile.^25–27 Stefanick et al. reported that LDL-C of males and females significantly decreased by 12% and 9% in a one-year programme of diet-plus-exercise (National Cholesterol Education Program [NCEP] Step2 diet and aerobic exercise), respectively, but not significantly decreased in only diet programme.²⁵ They additionally showed that HDL-C did not significantly change by diet-plus-exercise or only diet programme.²⁵ Beard et al. reported that total cholesterol, LDL-C, HDL-C and glucose decreased by 20%, 20%, 17% and 16% in a three-week programme of diet-plus-exercise, respectively.²⁶ The diet programme was that the ratios of caloric intake from fat, protein and carbohydrate were <10%, 10–20% and 70–80%, respectively, and the exercise programme was that the exercise classes were held five days per week, and included 15–20 min of stretching and 45 min of aerobic exercise.²⁶ Varady et al. reported that LDL-C decreased by 10% and 8% in 12-week diet programmes, 75% and 25% energy restriction, respectively and HDL-C increased by 16% with three days per week of aerobic exercise.²⁷ Additionally, Varady et al. indicated that the proportion of small LDL particle size decreased in the programme of 75% energy restriction.²⁷ We indicated that small dense LDL-C decreased by 13% in the six-month period (Table 2). The cause for not increasing HDL-C (Table 2) may be that lipoprotein profile was improved only by dietary modification, not using exercise, in this study.

In our study, TG, VLDL-C and CM-C were not significantly decreased by the intervention of calorie restriction and the change of caloric percentage from fat. Stefanick et al. reported that TG was not significantly decreased in the one-year programme of diet-plus-exercise, and the baseline of TG was 159 mg/dL.²⁵ Beard et al. reported that TG decreased by 17% in a three-week programme of diet-plus-exercise, and the baseline of TG was 221 mg/dL.²⁶ Varady et al. reported that TG significantly decreased in a 12-week diet programme of 75% energy restriction, but not significantly changed in the 12-week diet programme of 25% energy restriction.²⁷ We thought that TG can be significantly decreased by extreme energy restriction, or in the case of high TG concentrations of subjects. In this study, the calorie restriction was 16±13%, and the baseline TG concentration was normal (113±56 mg/dL). Therefore, the change of TG was small (Table 2), and the correlations with BMI, BW, WC and FRS were not significant (Table 3). A recent paper reported that the reference range for VLDL-C was 19.8±23.8 mg/dL,²⁸ whereas in our study this was 13.3±8.7 mg/dL. The CM-C concentration in our study was low, 0.5±0.1 mg/dL, as shown in in Table 2. No decrease of VLDL-C and CM-C in our study might be caused by these low baseline concentrations.

In the Framingham Heart Study cohort, the correlation between lifestyle and the risk factor change of CHD in young adults during an eight-year follow-up period was reported.¹³ The report showed that the change of BMI was significantly correlated with those of total cholesterol, LDL-C and VLDL-C, and significantly inversely correlated with that of HDL-C.¹³ We indicated strong correlations of the change of BMI and BW with those of TC and non-HDL-C (P < 0.005) in the total group, and additionally, weak correlations with those of sdLDL-C, VLDL-C and CM-C (P < 0.05) (Tables 3 and 4). In the non-obese group, the changes of BMI and BW were significantly correlated with those of TC, non-HDL-C, sdLDL-C, VLDL-C and CM-C. Furthermore, the change of FRS and HOMA was significantly correlated with the change of IDL-C, which is associated with the severity of CHD¹¹ in the total and obese groups (Table 4). The change of caloric intake was only significantly correlated with that of IDL-C (P < 0.05) in the total and non-obese groups (Table 4). We thought that lipid and lipoprotein profiles were improved by decreasing obesity indexes, BMI and BW, with calorie restriction.

In our study, LDL-C, non-HDL-C and IDL-C decreased in the six-month period of dietary modification, but the cholesterol and protein concentrations of Lp(a) increased (Table 2). Many studies indicated that increased Lp(a)-P concentration is associated with the risk of CHD.^16–18 It was shown that the risk of clinical atherosclerotic disease, e.g. CHD, intermittent claudication, congestive heart failure, increased by twofold to threefold with diabetes in the Framingham cohort study.¹⁹ Prospective studies in US and Danish, however, showed that an increased Lp(a)-P concentration is associated inversely with a risk of incident type 2 diabetes in subjects without CHD.²⁰ Recently, a paper reported that increased Lp(a) was associated with worse outcomes in type 2 diabetic patients with CHD.²⁹ Seman et al. reported that an increased Lp(a)-C (>10 mg/dL) was an independent CHD risk in males with a relative risk of more than 2, and the Lp(a)-C concentrations were measured by using lectin affinity binding method.³⁰ A previous report indicated that high concentration of insulin suppressed apolipoprotein(a) synthesis in monkey hepatocytes.³¹ In our study, insulin concentration decreased in the six-month period of dietary modification by calorie restriction (Table 2). Therefore, Lp(a)-C and Lp(a)-P in each of subjects might increase with the decreases of insulin and HOMA by the intervention of calorie restriction in our study. We thought that high Lp(a) concentration is the risk factor for CHD, but the Lp(a) concentration increases with the decreases of insulin and HOMA in individuals.

It is known that atherosclerotic lesion of early stage appears as chronic inflammation.³² In the atherosclerotic vascular inflammation, platelet is activated and adhered onto the endothelial cells.³³ The endothelial cells expresse P-selectin glyocoprotein ligand-1 (PSGL-1) under an inflammation condition, and P-selectin which is up-regulated on activated platelets can be bound to the PSGL-1.³³ Apolipoprotein(a) was reported to be strongly bound to platelet activated with strong agonist, i.e. thrombin, arachidonic acid (AA), etc.³⁴ It was also reported that LDL and Lp(a) were found in atherosclerotic lesion, and Lp(a) was increased in advanced lesions.³⁵ In our study, blood pressures, LDL-C, FGB and A1c decreased during the six-month period of dietary modification and FRS decreased (Table 2). Therefore, the activated platelet and the bound Lp(a) might decrease with the improvement of inflammation in atherosclerotic lesions during the six-month period. For this reason, the cholesterol and protein concentrations of Lp(a) in serum might increase in the six-month period of dietary modification (Table 2). Increased Lp(a) concentration might be related to not only CHD risk but also insulin concentration or atherosclerotic vascular inflammation, but this remains to be elucidated. Nauck et al. indicated that the ratio of Lp(a)-C to Lp(a)-P was low in CHD patients with immunoassay for the measurement of Lp(a)-P and electrophoresis with ultracentrifugal separation for Lp(a)-C.³⁶ Konerman et al. indicated that HDL-C in the subjects with high Lp(a)-C and normal Lp(a)-P and high Lp(a)-C and high Lp(a)-P was higher than in the subjects with normal Lp(a)-C and high Lp(a)-P and normal Lp(a)-C and normal Lp(a)-P.³⁷ In our study, the ratios of Lp(a)-C to Lp(a)-P in baseline data and data after six months of dietary modification were similar.

It was reported that the increment from monounsaturated fatty acid (MUFA) and polyunsaturated fatty acid (PUFA) was inversely associated with CHD.³⁸ This study showed that the dietary intakes of saturated fatty acid (SFA), MUFA and PUFA significantly decreased by the calorie restriction and the change of caloric percentage from fat (<30%) in total subjects, and the ratio of MUFA and PUFA significantly increased (Table 1). Additionally, FRS significantly decreased in the intervention period (Table 2), but the change of FRS was not correlated with those of MUFA and PUFA. It was also reported that MUFA was positively correlated with TG and VLDL particle concentration, and n-6 PUFA was inversely correlated with those.³⁹ This study showed that MUFA and n-6 PUFA significantly decreased by the intervention (Table 1), but TG and VLDL-C were not significantly changed (Table 2). The change of MUFA was positively correlated with that of VLDL-C (P < 0.05). Eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) are major n-3 PUFAs, and AA is n-6 PUFA. High EPA/AA ratio was known to be associated with a low incidence of CHD.^40,41 High EPA/AA and DHA/AA ratio were associated with an increase and a decrease in serum HDL-C, respectively, and do not relate to LDL-C.⁴¹ Additionally, high EPA/AA and DHA/AA ratio were associated with a decrease and an increase in serum TG or VLDL, respectively.⁴¹ In this study, n-3/n-6 PUFA ratio was not changed by the intervention (Table 1). In this study, EPA, DHA and AA were not measured, and this remains to be estimated.

The study is limited by the small number of subjects and not including males and should be interpreted with caution. Additionally, the study is a one-armed design and does not contain the group without intervention by calorie restriction. We need a large scaled and randomized study containing the subjects with and without intervention by calorie restriction to further clarify the results obtained in our study.

In conclusion, this study showed that BMI, WC, HOMA and FRS were reduced by dietary modification with calorie restriction during the six-month period, and lipid and lipoprotein profiles were improved, but Lp(a) concentration increased. The high Lp(a) concentration is the risk for CHD, but the Lp(a) concentration may increase with the decreases in insulin and HOMA in individuals, but remains to be elucidated. It was additionally shown that the changes of BMI, WC and FRS were significantly correlated with those of TG-rich-lipoprotein; IDL, VLDL and CM.

Footnotes

Acknowledgements

We would like to thank Ms. Syuko Onodera for the excellent technical assistance.

Declaration of conflicting interests

YH was an employee of TOSOH Corp. (Tokyo, Japan) till 31 March 2015. DM is an employee of TOSOH Corp. (Tokyo, Japan).

Funding

Research funds were provided in part by the Kagawa Nutrition University Research Fund (to KK and AT) and in part by the Tosoh Research Fund (to DM).

Ethical approval

The study protocol was approved by the ethical committees of Kagawa Nutrition University (No. 222-G) and Tosoh Corporation (No. 10-02).

Guarantor

YH.

Contributorship

YH, DM and KH corrected data. YH contributed to the discussion and wrote the manuscript. YH and AT designed the study protocol. KK and AT reviewed and edited the manuscript.

References

Després

Lemieux

Dagenais

et al.

HDL-cholesterol as a marker of coronary heart disease risk: the Québec cardiovascular study. Atherosclerosis 2000; 153: 263–272.

Eberly

Stamler

Neaton

. Multiple Risk Factor Intervention Trial Research Group. Relation of triglyceride levels, fasting and nonfasting, to fatal and nonfatal coronary heart disease. Arch Intern Med 2003; 163: 1077–1083.

Cholesterol Treatment Trialists’ (CTT) Collaboration; Baigent C, Blackwell L, Emberson J, et al. Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials. Lancet 2010; 376: 1670–1681.

Chapman

Ginsberg

Amarenco

et al.

European Atherosclerosis Society Consensus Panel. Triglyceride-rich lipoproteins and high-density lipoprotein cholesterol in patients at high risk of cardiovascular disease: evidence and guidance for management. Eur Heart J 2011; 32: 1345–1361.

Fernce

Yoo

Alesh

et al.

Effect of long-term exposure to lower low-density lipoprotein cholesterol beginning early in life on the risk of coronary heart disease: a Mendelian randomization analysis. J Am Coll Cardiol 2012; 60: 2631–2639.

Packard

Saito

. Non-HDL cholesterol as a measure of atherosclerotic risk. J Atheroscler Thromb 2004; 11: 6–14.

Robinson

Wang

Smith

et al.

Meta-analysis of the relationship between non-high-density lipoprotein cholesterol reduction and coronary heart disease risk. J Am Coll Cardiol 2009; 53: 316–322.

Tanabe

Iso

Okada

et al.

Japan Arteriosclerosis Longitudinal Study Group. Serum total and non-high-density lipoprotein cholesterol and the risk prediction of cardiovascular events – the JALS-ECC. Circ J 2010; 74: 1346–1356.

National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation 2002; 106: 3143–3421.

10.

Liu

Sempos

Donahue

et al.

Non-high-density lipoprotein and very-low-density lipoprotein cholesterol and their risk predictive values in coronary heart disease. Am J Cardiol 2006; 98: 1363–1368.

11.

Tatami

Mabuchi

Ueda

et al.

Intermediate-density lipoprotein and cholesterol-rich very low density lipoprotein in angiographically determined coronary artery disease. Circulation 1981; 64: 1174–1184.

12.

Sopko

Leon

Jacobs

Jr et al.

The effects of exercise and weight loss on plasma lipids in young obese men. Metabolism 1985; 34: 227–236.

13.

Hubert

Eaker

Garrison

et al.

Life-style correlates of risk factor change in young adults: an eight-year study of coronary heart disease risk factors in the Framingham offspring. Am J Epidemiol 1987; 125: 812–831.

14.

Nieman

Brock

Butterworth

et al.

Reducing diet and/or exercise training decreases the lipid and lipoprotein risk factors of moderately obese women. J Am Coll Nutr 2002; 21: 344–350.

15.

Haffner

D'Agostino

Jr Mykkänen

et al.

Insulin sensitivity in subjects with type 2 diabetes. Relationship to cardiovascular risk factors: the Insulin Resistance Atherosclerosis Study. Diabetes Care 1999; 22: 562–568.

16.

Wild

Fortmann

Marcovina

. A prospective case-control study of lipoprotein(a) levels and apo(a) size and risk of coronary heart disease in Stanford Five-City Project participants. Arterioscler Thromb Vasc Biol 1997; 17: 239–245.

17.

Suk Danik

Rifai

Buring

et al.

Lipoprotein(a), measured with an assay independent of apolipoprotein(a) isoform size, and risk of future cardiovascular events among initially healthy women. JAMA 2006; 296: 1363–1370.

18.

Lamon-Fava

Marcovina

Albers

et al.

Lipoprotein(a) levels, apo(a) isoform size, and coronary heart disease risk in the Framingham Offspring Study. J Lipid Res 2011; 52: 1181–1187.

19.

Kannel

McGee

. Diabetes and cardiovascular disease. The Framingham study. JAMA 1979; 241: 2035–2038.

20.

Mora

Kamstrup

Rifai

et al.

Lipoprotein(a) and risk of type 2 diabetes. Clin Chem 2010; 56: 1252–1260.

21.

Hirowatari

Yoshida

Kurosawa

et al.

Measurement of cholesterol of major serum lipoprotein classes by anion-exchange HPLC with perchlorate ion-containing eluent. J Lipid Res 2003; 44: 1404–1412.

22.

Hirowatari

Yoshida

Kurosawa

et al.

Analysis of cholesterol levels in lipoprotein(a) with anion-exchange chromatography. J Lipid Res 2010; 51: 1237–1243.

23.

Matthews

Hosker

Rudenski

et al.

Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28: 412–419.

24.

Wilson

D'Agostino

Levy

et al.

Prediction of coronary heart disease using risk factor categories. Circulation 1998; 97: 1837–1847.

25.

Stefanick

Mackey

Sheehan

et al.

Effects of diet and exercise in men and postmenopausal women with low levels of HDL cholesterol and high levels of LDL cholesterol. N Engl J Med 1998; 339: 12–20.

26.

Beard

Barnard

Robbins

et al.

Effects of diet and exercise on qualitative and quantitative measures of LDL and its susceptibility to oxidation. Arterioscler Thromb Vasc Biol 1996; 16: 201–207.

27.

Varady

Bhutani

Klempel

et al.

Comparison of effects of diet versus exercise weight loss regimens on LDL and HDL particle size in obese adults. Lipids Health Dis 2011; 10: 119–123.

28.

Manita

Hirowatari

Yoshida

. A rapid anion-exchange chromatography for measurement of cholesterol concentrations in five lipoprotein classes and estimation of lipoprotein profiles in male volunteers without overt diseases. Ann Clin Biochem 2015; 52: 638–646.

29.

Kwon

Kim

Sung

et al.

Elevated lipoprotein(a) has incremental prognostic value in type 2 diabetic patients with symptomatic coronary artery disease. J Atheroscler Thromb 2015; 22: 527–534.

30.

Seman

DeLuca

Jenner

et al.

Lipoprotein(a)-cholesterol and coronary heart disease in the Framingham Heart Study. Clin Chem 1999; 45: 1039–1046.

31.

Neele

de Wit

Princen

. Insulin suppresses apolipoprotein(a) synthesis by primary cultures of cynomolgus monkey hepatocytes. Diabetologia 1999; 42: 41–44.

32.

Fan

Watanabe

. Inflammatory reactions in the pathogenesis of atherosclerosis. J Atheroscler Thromb 2003; 10: 63–71.

33.

Projahn

Koenen

. Platelets: key players in vascular inflammation. J Leukoc Biol 2012; 92: 1167–1175.

34.

Martínez

Rivera

Loyau

et al.

Binding of recombinant apolipoprotein(a) to human platelets and effect on platelet aggregation. Thromb Haemost 2001; 85: 686–693.

35.

Smith

Crosbie

Cochran

. The inter-relation of fibrin, lipoprotein(a) and plasminogen in human atherosclerotic lesions. Blood Coagul Fibrinolysis 1990; 1: 495–498.

36.

Nauck

März

Wieland

. Is lipoprotein(a) cholesterol a significant indicator of cardiovascular risk? Clin Chem 2000; 46: 436–437.

37.

Konerman

Kulkarni

Toth

et al.

Lipoprotein(a) particle concentration and lipoprotein(a) cholesterol assays yield discordant classification of patients into four physiologically discrete groups. J Clin Lipidol 2012; 6: 368–373.

38.

Guasch-Ferré

Babio

Martínez-González

et al.

PREDIMED Study Investigators. Dietary fat intake and risk of cardiovascular disease and all-cause mortality in a population at high risk of cardiovascular disease. Am J Clin Nutr 2015; 102: 1563–1573.

39.

Jelenkovic

Bogl

Rose

et al.

Association between serum fatty acids and lipoprotein subclass profile in healthy young adults: exploring common genetic and environmental factors. Atherosclerosis 2014; 233: 394–402.

40.

Domei

Yokoi

Kuramitsu

et al.

Ratio of serum n-3 to n-6 polyunsaturated fatty acids and the incidence of major adverse cardiac events in patients undergoing percutaneous coronary intervention. Circ J 2012; 76: 423–429.

41.

Tani

Takahashi

Nagao

et al.

Polyunsaturated fatty acids and cardiovascular risk with the prevalence of coronary artery disease. Int Heart J 2015; 56: 260–268.