Original Article
Ann Clin Biochem 1990; 27: 489-493
Postprandial effect of a high fat meal on plasma lipid, lipoprotein cholesterol and apolipoprotein measurements Nader Rifai*, Jeffrey R Merrill] and Robert G Holly] From the "Department of Laboratory Medicine, Children's National Medical Center, Departments of Pathology and Pediatrics, The George Washington University Medical Center, 111 Michigan Avenue, NW, Washington, DC 20010, t Department of Family Practice, The Moses H. Cone Memorial Hospital, Greensboro, NC 27401, t Department of Physical Education, University of California, Davis, CA 956, USA
We investigated the effect of a fatty meal on plasma concentrations of lipids, apolipoproteins, and the cholesterol component of lipoproteins. Sixteen nonobese, healthy, asymptomatic males, 22-34 years of age, served as subjects for this study. None smoked, consumed more than two alcoholic drinks per day, or took any medication known to alter plasma lipids. After a 12 h fast, baseline plasma samples were obtained just before subjects consumed a high fat meal. The meal, standardized to a 70 kg individual, contained approximately 70 g fat, 580 mg cholesterol, and 1100cal, with 56% of the calories coming from fat. During the 8 h following consumption of the meal, subjects rested quietly and consumed no food or beverages except water. Blood specimens were obtained hourly. There was a significant increase in plasma triglyceride (150% from baseline at 3 h, P < 0·0005). Very low density lipoprotein cholesterol (VLDL-C) concentrations increased 150% at 3 h (P < 0,0005) while low density lipoprotein cholesterol (LDL-C) concentration decreased 37% at 3 h (P < 0'005) when estimated by Friedewald's formula. No statistically significant differences were observed between fasting total cholesterol, high density lipoprotein cholesterol (HDL-C), HDL2-C, and HDU-C, apolipoprotein AI (apo AI, All), and B-IOO concentrations and non-fasting samples. We conclude that plasma triglyceride concentration is significantly affected in the post-prandial state. As a result, VLDL-C and LDL-C when assessed by the Friedewald formula are also altered. A minimum of 8 h fasting is required to assess these concentrations accurately in this population. The concentrations of total cholesterol, HDL-C, HDL2-C, HDU-C, apo AI, apo All, and apo B-lOO can be determined adequately using a non-fasting specimen.
SUMMARY.
Additional key phrases: HDL; LDL; VLDL; Friedewald's formula; hyperlipidaemia Increased concentrations of total cholesterol, low density lipoprotein cholesterol (LDL-C) and apolipoprotein B-lOO (apo B-lOO), and decreased concentrations of high density lipoprotein cholesterol (HDL-C), HDL2-C and apo AI, correlate with increased risk of premature coronary heart disease.l ? In addition to total cholesterol concentration, the US National Cholesterol Education Program (NCEP) Adult Treatment Panel recommended the use of LDL concentration as the other primary parameter for the diagnosis and treatment of hypercholesterolaemia." Although it has repeatedly been shown that the apo AI and B-100 concentrations are better predictors of coronary heart disease risk than the Correspondence: Dr N Rifai.
cholesterol component of the corresponding lipoproteins.r" no specific guidelines have been issued concerning the use of these parameters in the routine clinical setting. Problems of standardization and the lack of well-defined reference ranges are the two main reasons for restricting the measurement of apolipoproteins to clinical trials. Twelve hour fasting samples are recommended by the US Lipid Research Clinics (LRC) for accurate determination of lipid and lipoprotein cholesterol concentrations." The LRC methodology for the isolation of the LDL fractions involves an ultracentrifugation step,' which is time consuming and cumbersome. Most clinical laboratories use Friedewald's equation to calculate an estimate of LDL-C concentration.I 489
490
Rifai, Merrill and Holly
Only one group has looked at the effect of postprandial lipaemia on the determination of LDLC by Friedewald's equation (9) and this Iipaemic effect on the apolipoprotein concentrations remains controversial. HI-I 2 In the present study, we investigated the postprandial effect of a high fat meal on plasma lipid, lipoprotein cholesterol, and apolipoprotein concentrations and determined the minimum fasting period required for lipid testing. MATERIALS AND METHODS Sixteen non-obese, healthy, asymptomatic white males, 22-34 years-of-age, served as subjects for this study. None of them smoked, consumed more than two alcoholic or caffeinated drinks per day, or was taking any medication known to alter plasma lipids. After a 12 h fast, subjects reported to the laboratory, rested quietly for 10 min after which a venous blood sample was obtained to establish baseline values. Immediately thereafter, subjects consumed a high fat meal. The meal, which consisted of approximately two sausages with egg, hash brown potatoes and a glass of iced water, was standardized to a 70 kg individual. This was done in an attempt to allow for the differences in the rates of metabolism that could occur between subjects due to differences in body size alone. The meal contained approximately 70 g fat, 580 mg cholesterol, and 1100 cal, with 56% of the calories coming from fat. During the 8 h following consumption of the meal, subjects rested quietly, consumed no food and drank only water. Venous blood specimens were obtained hourly for lipid testing. Portions of these data have been presented previously.l':" Blood specimens were collected in tubes containing EDTA anticoagulant and centrifuged for 15min to separate the plasma. Plasma HDL-C concentrations were determined by assay of the supernatants remaining after precipitation of plasma LDL and very low density lipoprotein (VLDL) with heparin plus manganese chloride (92 mrnol/L of MnCI 2 ) . 15 HDL2 and HDU were isolated from the HDL fractions by flotation density ultracentrifugation using a Beckman Airfuge (Beckman Instruments Inc., Brea, USA).16 Cholesterol was then measured in both fractions. Total cholesterol, as well as HDL-C, HDL2-C and HDL3-C, were determined enzymatically!' using reagents from BoehringerMannheim Diagnostics (BMD, Indianapolis, USA). Triglyceride was measured enzymatically with correction for endogenous glycerol" using
reagents from the same supplier. Day-to-day imprecision for triglyceride and total cholesterol measurement was less than 3% while for HDL-C it was 5·5%. VLDL-C and LDL-C concentrations were calculated using Friedewald's equation: 8VLDL-C = triglyceride/5, LDL-C = total cholesterol - (HDL-C + VLDL-C). Apolipoproteins were determined by immunoturbidimetric assays" using antisera for apo AI and All from BMD and reference materials for apo AI and All from Washington Research Foundation (Seattle, USA) and antisera and calibrators for apo B-100 from Behring Diagnostics (La Jolla, USA). Day-to-day imprecision for the apolipoprotein assays was < 5%. The measurements of cholesterol, triglyceride and apolipoprotein concentrations were performed on a Cobas-BIO centrifugal analyser (Roche Analytical Instruments, Nutley, USA). The paired r-test was used to compare the significance of mean differences with P < 0·05 being chosen as statistically significant.
RESULTS AND DISCUSSION It has been reported that diurnal variations in fasting plasma triglyceride concentration exist with the highest value obtained in the rnoming.:" In addition, it has been shown that the triglyceride increment is much larger after breakfast than after similar meals later in the day.:" Therefore. subjects in the present study reported to the laboratory early in the morning and received their meals shortly thereafter. Plasma triglyceride concentrations were significantly increased in the postprandial state (Table I). As previously reported," 13 the maximum increase was observed after 3 h (Fig. I). By 8 h, mean triglyceride concentrations did not statistically differ from fasting values. The mean triglyceride concentration was increased by 150% of baseline at 3 h (range 58%232%). Physiological variations in fasting triglyceride concentrations, as determined using two fasting specimens collected I month apart. was expressed as mean difference ± SD (0'14 ± 0·25 mrnol/L). The individual differences between the 8 h triglyceride concentration and its fasting value exceeded I SD of this calculated physiological variation in two subjects and 2 SD in a third. The parallel increase in VLDL-C and triglyceride concentrations observed in this study (Table I, Fig. I) is inevitable since VLDL-C
Postprandial effect of a high fat meal TABLE
I.
491
Post-prandial effect on plasma lipid, lipoprotein cholesterol and apolipoprotein concentrations
Time after consuming the meal (h)
Cholesterol Triglyceride VLDL-C LDL-C HDL-C HDL2-C HDL3-C Apo AI Apo All Apo 8-100
0
2
4
6
8
4·169 ± 0·777 0·79 ± 0·32 0·363 ± 0·155 2·564 ± 0·725 1·243 ± 0·244 0·402 ± 0·117 0·813 ± 0·098 1·34 ± 0·18 0·32 ± 0·03 1·00 ± 0·19
3·963 ± 0·751 1·71 ± 0·880·777 + 0·4142·046 ±0·803' 1·217 ± 0·233 0·391 ± 0·117 0·787 ± 0·137 1·32 ± 0·17 0·33 ± 0·035 1·00 ± 0·21
4·144 ± 0·803 1·72 ± 0·920·777 ± 0-4142·150 ± 0·673t 1·191 ± O'207 0·368 ± 0·109 0·766 ± 0·122 1·35 ± 0·18 0·34 ± 0·048 1·02 ± 0·22
4·196 ± 0·803 1·22 ± 0·57-0·570 ± 0·259-2·383 ± 0·751 1·243 ± 0·212 0·399 ± 0·122 0·785 ± 0·096 1·35 ± 0·16 0·33 ± 0·045 1·02 ± 0·21
4·248 ± 0·829 0·86 ± 0·35 0·414 ± 0·155 2·538 ± 0·751 1·321 ± 0·218 0-459 ± 0·132 0·834 ± 0·109 1·36 ± 0·18 0·33 ± 0·040 1·00 ± 0·22
Values represent mean ± SO in mrnol/L for cholesterol, triglyceride, VLDL-C, LDL-C, HDL-C, HDL2-C and HDL3-C and in giL for apo AI, apo All and apo 8-100. Difference from 0 h values by paired t-test: - P < 0,0005, _. P < 0'005, t P < 0,025, t P < 0·05.
concentrations were estimated by dividing triglyceride concentrations by five, and not measured directly. The postprandial increase in measured plasma VLDL-C concentrations, by the LRC methodology,' was shown to be' significantly less pronounced than that of estimated VLDL-C. 9 Therefore, most of the increase in VLDL-C concentration reported in this study resulted from the methodology used. The increased secretion of VLDL from liver after fat-feeding" may also have contributed to this increase. The LDL-C concentrations, calculated from Friedewald's formula, were significantly decreased in all subjects (Table I). A maximum decrease of 37% was observed 3 h post-feeding (Fig. I). By 8 h, LDL-C values returned to fasting concentrations. As with VLDL-C concentration,
:a--'
1-70
E E
s'"
1'13
;::
E c
g0'57 o
u
0+---,---,---,-_-,-_-,-_-,-_-,-_--,
o
3
4
5
6
7
B
Postprandial period (h)
FIGURE 1. Postprandial effect of a high fat meal on plasma concentrations of triglycerides, VLDL-C and LDL-C. _, triglyceride; 0, LDL-C; ., VLDL-C.
the postprandial effect on the estimated LDL-C was much greater than the measured LDL-C by the LRC methodology." Such a bias is of concern because many clinical laboratories calculate LDL-C by Friedewald's formula and physicians then assess patient risk of coronary heart disease using LDL-C values generated by the LRC methodology. If the patient is not fasting, the LDL-C concentration will be significantly underestimated and the patient's management may be altered since the NCEP recommended the use of LDL-C as a primary parameter in the diagnosis and treatment of hypercholesterolaemia. Therefore, to diminish the discrepancy between the estimated and the measured LDL-C concentration only fasting specimens should be used. When it becomes available, a reliable and simple direct method for the determination of LDL-C concentration might alleviate this problem. In this study, the mean plasma total cholesterol concentration was not significantly altered in the postprandial phase (Table 1). This finding is in agreement with previous reports.IO·12.22 Nonfasting HDL-C and its two subfractions, HDL2C and HDL3-C, concentrations did not differ from fasting values either (Table I). It has been reported previously that the cholesterol component of HDL isolated by density gradient!" or rate-zonal ultracentrifugation" was increased after fat-feeding. This observed increase in HDLC was attributed to an increase in HDL2-C and not in HDL3-C. 33 In contrast, the cholesterol component of HDL isolated by precipitation methods, using heparin-MnCl 211 or dextran sulphate-MgCI/ reagents, was shown to be decreased by 5-10%. Cohn et a/.9 suggested possible explanations for the underestimated HDL-C.
492
Rifai, Merrill and Holly
First, after fat feeding the increase in size and lipid: protein ratio of lipoproteins, including HDL, may cause HDL to co-precipitate with LDL and VLDL, Secondly, the postprandial shift in apo C proteins and apo E from high density to apo B-100-containing lower density lipoproteins may change the polarity of these complexes and increase precipitation, resulting in less cholesterol being measured in the supernatant. The discrepancy between our findings and those previously reported may be attributed to inter-laboratory variations in precipitation methodologies. As with HDL-C, the concentrations of plasma apo AI and All, the major proteins in HDL, were not affected in this study by fat-feeding (Table 1). A slight but statistically significant increase in the concentrations of apo AI and All post-feeding has been reported." Others, however, have reported that apo AI and/or apo All concentrations remained the same after fat-feeding.!':'? As previously reported," no postprandial changes in the concentrations of apo B-100 were observed (Table I). Therefore, the slight decrease in the cholesterol component of LDL isolated by the LRC methodolgy, the recognized reference method, after fat-feeding" suggests a shift toward the formation ofa higher density LDL, the more atherogenic form of LDL. 24 In conclusion, the measurement of plasma triglyceride concentration and the calculation of VLDL-C and LDL-C concentrations using Friedewald's formula are significantly altered in the post-prandial state and a minimum of 8 h fasting is required to assess their concentrations accurately. However, one should be aware that only white, young, normo-triglyceridaemic males were used in this trial and an 8 h fast may not be adequate to determine accurately triglyceride and LDL-C concentrations in older or hypertriglyceridaemic populations. The concentrations of total cholesterol, HDL-C, HDL2-C, HDL3-C, apo AI, apo All, and apo B-100 can be determined adequately using a non-fasting sample. REFERENCES I Stamler J, Wentworth D, Neaton JD. Is relationship between serum cholesterol and risk of premature death from coronary heart disease continuous and graded? JAMA 1986; 256: 2823-8 2 Castelli WP, Doyle JT, Gordon T, et al. HDL cholesterol and other lipids in coronary heart disease - The cooperative Lipoproptein Phenotyping Study. Circulation 1977; 55: 767-77 3 Kreisberg RA. Lipids, Lipoproteins, apolipo
proteins and atherosclerosis. Ann Intern Med 1983; 99: 713-5 4 National Cholesterol Education Program. Report of the expert panel on detection, evaluation, and treatment of high blood cholesterol in adults. Arch Intern Med 1988; 145: 36-62 5 Kukita H, Hiwade K, Kokubu T. Serum apolipoprotein AI, All, and B levels and their discriminative values in relatives with patients with coronary artery disease. Atherosclerosis 1984: 51: 261-7 6 Maciejko 11, Holmes DRF, Kottke BA, et al. Apolipoprotein AI as a marker of angiogrpahically assessed coronary artery disease. N Eng J Med 1983; 309: 385-9 7 Lipid Research Clinics Program: Lipid and Lipoprotein Analysis. In Manual of Laboratory Operations, Vol. I. Department of Health, Education and Welfare Publication (NIH) 75-628. Bethesda, MD: National Institutes of Health, 1874 8 Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chern 1972; 18: 499-502 9 Cohn JS, McNamara JR, Schaefer EJ. Lipoprotein cholesterol concentrations in the plasma of human subjects as measured in the fed and fasted states. Clin Chern 1988; 34: 2456-9 10 Tall AR, Blum CB, Forester GP, Nelson CA. Changes in the distribution and composition of plasma high density lipoproteins after ingestion of fat. J Bioi Chern 1982; 257: 198-207 II Kay RM, Rao S, Arnott C, Miller NE, Lewis B. Acute effects of the pattern of fat ingestion on plasma high density lipoprotein components in man. Atherosclerosis 1980; 36: 567-73 12 Groot PHE, Scheek LM. Effects of fat ingestion on high density lipoprotein profiles in human sera.·J Lip Res 1984; 25: 684-92 13 Merrill JR, Holly RG, Anderson RL, Rifai N, King ME, DeMeersman R. Hyperlipidemic response of young trained and untrained men after a high fat meal. Arteriosclerosis 1989; 9: 217-23 14 Rifai N, King ME, DeMeersman R, Merrill JR, Holy RG. Apolipoprotein and lipoprotein cholesterol concentrations in trained and sedentary males. Clin Chim Acta 1987; 163: 113-7 15 Warnick GR, Albers JJ. A comprehensive evaluation of the heparin-manganese precipitation procedure for estimating high density lipoprotein cholesterol. J Lip Res 1978; 19: 65-76 16 Eyre J, Hammett F, Miller NE. A micro-method for the rapid centrifugal separation of human plasma high density lipoprotein subfractions. HDL2 and HDL3. Clin Chim Acta 1981; 114: 22531 17 Allain C, Po on LS, Chan CSG, Richmond W, Fu Pc. Enzymatic determination of total serum cholesterol. Clin Chern 1974; 20: 470-5 18 Stinshoff K, Weisshaar D, Stachler F, Hess D, Gruber W, Steier E. Relation beteen concentration offree glycerol and triglycerides in human sera. Clin Chern 1977; 23: 1029-32 19 Rifai N, King ME. Immunoturbidimetric assays of apolipoprotein A, AI, All, and B in serum. Clin
Postprandial effect of a high fat meal Chern 1986; 32: 957-61 20 Terpstra J, Hessel LW, Seepers J, Van Gent CM. The influence of meal frequency on diurnal lipid, glucose and cortisol levels in normal subjects. Eur J Clin Invest 1978; 8: 61-6 21 Carey MC, Small DM, Bliss CM. Lipid digestion and adsorption. Annu Rev Phy.l'io11983; 45: 651-77 22 Redgrave TG, Carlson LA. Changes in plasma very low density and low density lipoprotein content, composition, and size after a fatty meal in normoand hypertriglyceridemic man. J Lipid Res 1979; 20:
493
217-29 23 Havel RJ, Kane JP. Kashyap ML. Interchange of apolipoproteins between chylomicrons and high density lipoproteins during alimentary lipemia in man. J Clin Invest 1973; 52 32-8 24 Krauss RL. Relationship of intermediate and lowdensity lipoprotein subspecies to risk of coronary artery disease. Am Heart J 1987; 113: 578-82
Accepted for publication 17 August 1989
Ann Clin Biochem 1990; 27: 489-493
Postprandial effect of a high fat meal on plasma lipid, lipoprotein cholesterol and apolipoprotein measurements Nader Rifai*, Jeffrey R Merrill] and Robert G Holly] From the "Department of Laboratory Medicine, Children's National Medical Center, Departments of Pathology and Pediatrics, The George Washington University Medical Center, 111 Michigan Avenue, NW, Washington, DC 20010, t Department of Family Practice, The Moses H. Cone Memorial Hospital, Greensboro, NC 27401, t Department of Physical Education, University of California, Davis, CA 956, USA
We investigated the effect of a fatty meal on plasma concentrations of lipids, apolipoproteins, and the cholesterol component of lipoproteins. Sixteen nonobese, healthy, asymptomatic males, 22-34 years of age, served as subjects for this study. None smoked, consumed more than two alcoholic drinks per day, or took any medication known to alter plasma lipids. After a 12 h fast, baseline plasma samples were obtained just before subjects consumed a high fat meal. The meal, standardized to a 70 kg individual, contained approximately 70 g fat, 580 mg cholesterol, and 1100cal, with 56% of the calories coming from fat. During the 8 h following consumption of the meal, subjects rested quietly and consumed no food or beverages except water. Blood specimens were obtained hourly. There was a significant increase in plasma triglyceride (150% from baseline at 3 h, P < 0·0005). Very low density lipoprotein cholesterol (VLDL-C) concentrations increased 150% at 3 h (P < 0,0005) while low density lipoprotein cholesterol (LDL-C) concentration decreased 37% at 3 h (P < 0'005) when estimated by Friedewald's formula. No statistically significant differences were observed between fasting total cholesterol, high density lipoprotein cholesterol (HDL-C), HDL2-C, and HDU-C, apolipoprotein AI (apo AI, All), and B-IOO concentrations and non-fasting samples. We conclude that plasma triglyceride concentration is significantly affected in the post-prandial state. As a result, VLDL-C and LDL-C when assessed by the Friedewald formula are also altered. A minimum of 8 h fasting is required to assess these concentrations accurately in this population. The concentrations of total cholesterol, HDL-C, HDL2-C, HDU-C, apo AI, apo All, and apo B-lOO can be determined adequately using a non-fasting specimen.
SUMMARY.
Additional key phrases: HDL; LDL; VLDL; Friedewald's formula; hyperlipidaemia Increased concentrations of total cholesterol, low density lipoprotein cholesterol (LDL-C) and apolipoprotein B-lOO (apo B-lOO), and decreased concentrations of high density lipoprotein cholesterol (HDL-C), HDL2-C and apo AI, correlate with increased risk of premature coronary heart disease.l ? In addition to total cholesterol concentration, the US National Cholesterol Education Program (NCEP) Adult Treatment Panel recommended the use of LDL concentration as the other primary parameter for the diagnosis and treatment of hypercholesterolaemia." Although it has repeatedly been shown that the apo AI and B-100 concentrations are better predictors of coronary heart disease risk than the Correspondence: Dr N Rifai.
cholesterol component of the corresponding lipoproteins.r" no specific guidelines have been issued concerning the use of these parameters in the routine clinical setting. Problems of standardization and the lack of well-defined reference ranges are the two main reasons for restricting the measurement of apolipoproteins to clinical trials. Twelve hour fasting samples are recommended by the US Lipid Research Clinics (LRC) for accurate determination of lipid and lipoprotein cholesterol concentrations." The LRC methodology for the isolation of the LDL fractions involves an ultracentrifugation step,' which is time consuming and cumbersome. Most clinical laboratories use Friedewald's equation to calculate an estimate of LDL-C concentration.I 489
490
Rifai, Merrill and Holly
Only one group has looked at the effect of postprandial lipaemia on the determination of LDLC by Friedewald's equation (9) and this Iipaemic effect on the apolipoprotein concentrations remains controversial. HI-I 2 In the present study, we investigated the postprandial effect of a high fat meal on plasma lipid, lipoprotein cholesterol, and apolipoprotein concentrations and determined the minimum fasting period required for lipid testing. MATERIALS AND METHODS Sixteen non-obese, healthy, asymptomatic white males, 22-34 years-of-age, served as subjects for this study. None of them smoked, consumed more than two alcoholic or caffeinated drinks per day, or was taking any medication known to alter plasma lipids. After a 12 h fast, subjects reported to the laboratory, rested quietly for 10 min after which a venous blood sample was obtained to establish baseline values. Immediately thereafter, subjects consumed a high fat meal. The meal, which consisted of approximately two sausages with egg, hash brown potatoes and a glass of iced water, was standardized to a 70 kg individual. This was done in an attempt to allow for the differences in the rates of metabolism that could occur between subjects due to differences in body size alone. The meal contained approximately 70 g fat, 580 mg cholesterol, and 1100 cal, with 56% of the calories coming from fat. During the 8 h following consumption of the meal, subjects rested quietly, consumed no food and drank only water. Venous blood specimens were obtained hourly for lipid testing. Portions of these data have been presented previously.l':" Blood specimens were collected in tubes containing EDTA anticoagulant and centrifuged for 15min to separate the plasma. Plasma HDL-C concentrations were determined by assay of the supernatants remaining after precipitation of plasma LDL and very low density lipoprotein (VLDL) with heparin plus manganese chloride (92 mrnol/L of MnCI 2 ) . 15 HDL2 and HDU were isolated from the HDL fractions by flotation density ultracentrifugation using a Beckman Airfuge (Beckman Instruments Inc., Brea, USA).16 Cholesterol was then measured in both fractions. Total cholesterol, as well as HDL-C, HDL2-C and HDL3-C, were determined enzymatically!' using reagents from BoehringerMannheim Diagnostics (BMD, Indianapolis, USA). Triglyceride was measured enzymatically with correction for endogenous glycerol" using
reagents from the same supplier. Day-to-day imprecision for triglyceride and total cholesterol measurement was less than 3% while for HDL-C it was 5·5%. VLDL-C and LDL-C concentrations were calculated using Friedewald's equation: 8VLDL-C = triglyceride/5, LDL-C = total cholesterol - (HDL-C + VLDL-C). Apolipoproteins were determined by immunoturbidimetric assays" using antisera for apo AI and All from BMD and reference materials for apo AI and All from Washington Research Foundation (Seattle, USA) and antisera and calibrators for apo B-100 from Behring Diagnostics (La Jolla, USA). Day-to-day imprecision for the apolipoprotein assays was < 5%. The measurements of cholesterol, triglyceride and apolipoprotein concentrations were performed on a Cobas-BIO centrifugal analyser (Roche Analytical Instruments, Nutley, USA). The paired r-test was used to compare the significance of mean differences with P < 0·05 being chosen as statistically significant.
RESULTS AND DISCUSSION It has been reported that diurnal variations in fasting plasma triglyceride concentration exist with the highest value obtained in the rnoming.:" In addition, it has been shown that the triglyceride increment is much larger after breakfast than after similar meals later in the day.:" Therefore. subjects in the present study reported to the laboratory early in the morning and received their meals shortly thereafter. Plasma triglyceride concentrations were significantly increased in the postprandial state (Table I). As previously reported," 13 the maximum increase was observed after 3 h (Fig. I). By 8 h, mean triglyceride concentrations did not statistically differ from fasting values. The mean triglyceride concentration was increased by 150% of baseline at 3 h (range 58%232%). Physiological variations in fasting triglyceride concentrations, as determined using two fasting specimens collected I month apart. was expressed as mean difference ± SD (0'14 ± 0·25 mrnol/L). The individual differences between the 8 h triglyceride concentration and its fasting value exceeded I SD of this calculated physiological variation in two subjects and 2 SD in a third. The parallel increase in VLDL-C and triglyceride concentrations observed in this study (Table I, Fig. I) is inevitable since VLDL-C
Postprandial effect of a high fat meal TABLE
I.
491
Post-prandial effect on plasma lipid, lipoprotein cholesterol and apolipoprotein concentrations
Time after consuming the meal (h)
Cholesterol Triglyceride VLDL-C LDL-C HDL-C HDL2-C HDL3-C Apo AI Apo All Apo 8-100
0
2
4
6
8
4·169 ± 0·777 0·79 ± 0·32 0·363 ± 0·155 2·564 ± 0·725 1·243 ± 0·244 0·402 ± 0·117 0·813 ± 0·098 1·34 ± 0·18 0·32 ± 0·03 1·00 ± 0·19
3·963 ± 0·751 1·71 ± 0·880·777 + 0·4142·046 ±0·803' 1·217 ± 0·233 0·391 ± 0·117 0·787 ± 0·137 1·32 ± 0·17 0·33 ± 0·035 1·00 ± 0·21
4·144 ± 0·803 1·72 ± 0·920·777 ± 0-4142·150 ± 0·673t 1·191 ± O'207 0·368 ± 0·109 0·766 ± 0·122 1·35 ± 0·18 0·34 ± 0·048 1·02 ± 0·22
4·196 ± 0·803 1·22 ± 0·57-0·570 ± 0·259-2·383 ± 0·751 1·243 ± 0·212 0·399 ± 0·122 0·785 ± 0·096 1·35 ± 0·16 0·33 ± 0·045 1·02 ± 0·21
4·248 ± 0·829 0·86 ± 0·35 0·414 ± 0·155 2·538 ± 0·751 1·321 ± 0·218 0-459 ± 0·132 0·834 ± 0·109 1·36 ± 0·18 0·33 ± 0·040 1·00 ± 0·22
Values represent mean ± SO in mrnol/L for cholesterol, triglyceride, VLDL-C, LDL-C, HDL-C, HDL2-C and HDL3-C and in giL for apo AI, apo All and apo 8-100. Difference from 0 h values by paired t-test: - P < 0,0005, _. P < 0'005, t P < 0,025, t P < 0·05.
concentrations were estimated by dividing triglyceride concentrations by five, and not measured directly. The postprandial increase in measured plasma VLDL-C concentrations, by the LRC methodology,' was shown to be' significantly less pronounced than that of estimated VLDL-C. 9 Therefore, most of the increase in VLDL-C concentration reported in this study resulted from the methodology used. The increased secretion of VLDL from liver after fat-feeding" may also have contributed to this increase. The LDL-C concentrations, calculated from Friedewald's formula, were significantly decreased in all subjects (Table I). A maximum decrease of 37% was observed 3 h post-feeding (Fig. I). By 8 h, LDL-C values returned to fasting concentrations. As with VLDL-C concentration,
:a--'
1-70
E E
s'"
1'13
;::
E c
g0'57 o
u
0+---,---,---,-_-,-_-,-_-,-_-,-_--,
o
3
4
5
6
7
B
Postprandial period (h)
FIGURE 1. Postprandial effect of a high fat meal on plasma concentrations of triglycerides, VLDL-C and LDL-C. _, triglyceride; 0, LDL-C; ., VLDL-C.
the postprandial effect on the estimated LDL-C was much greater than the measured LDL-C by the LRC methodology." Such a bias is of concern because many clinical laboratories calculate LDL-C by Friedewald's formula and physicians then assess patient risk of coronary heart disease using LDL-C values generated by the LRC methodology. If the patient is not fasting, the LDL-C concentration will be significantly underestimated and the patient's management may be altered since the NCEP recommended the use of LDL-C as a primary parameter in the diagnosis and treatment of hypercholesterolaemia. Therefore, to diminish the discrepancy between the estimated and the measured LDL-C concentration only fasting specimens should be used. When it becomes available, a reliable and simple direct method for the determination of LDL-C concentration might alleviate this problem. In this study, the mean plasma total cholesterol concentration was not significantly altered in the postprandial phase (Table 1). This finding is in agreement with previous reports.IO·12.22 Nonfasting HDL-C and its two subfractions, HDL2C and HDL3-C, concentrations did not differ from fasting values either (Table I). It has been reported previously that the cholesterol component of HDL isolated by density gradient!" or rate-zonal ultracentrifugation" was increased after fat-feeding. This observed increase in HDLC was attributed to an increase in HDL2-C and not in HDL3-C. 33 In contrast, the cholesterol component of HDL isolated by precipitation methods, using heparin-MnCl 211 or dextran sulphate-MgCI/ reagents, was shown to be decreased by 5-10%. Cohn et a/.9 suggested possible explanations for the underestimated HDL-C.
492
Rifai, Merrill and Holly
First, after fat feeding the increase in size and lipid: protein ratio of lipoproteins, including HDL, may cause HDL to co-precipitate with LDL and VLDL, Secondly, the postprandial shift in apo C proteins and apo E from high density to apo B-100-containing lower density lipoproteins may change the polarity of these complexes and increase precipitation, resulting in less cholesterol being measured in the supernatant. The discrepancy between our findings and those previously reported may be attributed to inter-laboratory variations in precipitation methodologies. As with HDL-C, the concentrations of plasma apo AI and All, the major proteins in HDL, were not affected in this study by fat-feeding (Table 1). A slight but statistically significant increase in the concentrations of apo AI and All post-feeding has been reported." Others, however, have reported that apo AI and/or apo All concentrations remained the same after fat-feeding.!':'? As previously reported," no postprandial changes in the concentrations of apo B-100 were observed (Table I). Therefore, the slight decrease in the cholesterol component of LDL isolated by the LRC methodolgy, the recognized reference method, after fat-feeding" suggests a shift toward the formation ofa higher density LDL, the more atherogenic form of LDL. 24 In conclusion, the measurement of plasma triglyceride concentration and the calculation of VLDL-C and LDL-C concentrations using Friedewald's formula are significantly altered in the post-prandial state and a minimum of 8 h fasting is required to assess their concentrations accurately. However, one should be aware that only white, young, normo-triglyceridaemic males were used in this trial and an 8 h fast may not be adequate to determine accurately triglyceride and LDL-C concentrations in older or hypertriglyceridaemic populations. The concentrations of total cholesterol, HDL-C, HDL2-C, HDL3-C, apo AI, apo All, and apo B-100 can be determined adequately using a non-fasting sample. REFERENCES I Stamler J, Wentworth D, Neaton JD. Is relationship between serum cholesterol and risk of premature death from coronary heart disease continuous and graded? JAMA 1986; 256: 2823-8 2 Castelli WP, Doyle JT, Gordon T, et al. HDL cholesterol and other lipids in coronary heart disease - The cooperative Lipoproptein Phenotyping Study. Circulation 1977; 55: 767-77 3 Kreisberg RA. Lipids, Lipoproteins, apolipo
proteins and atherosclerosis. Ann Intern Med 1983; 99: 713-5 4 National Cholesterol Education Program. Report of the expert panel on detection, evaluation, and treatment of high blood cholesterol in adults. Arch Intern Med 1988; 145: 36-62 5 Kukita H, Hiwade K, Kokubu T. Serum apolipoprotein AI, All, and B levels and their discriminative values in relatives with patients with coronary artery disease. Atherosclerosis 1984: 51: 261-7 6 Maciejko 11, Holmes DRF, Kottke BA, et al. Apolipoprotein AI as a marker of angiogrpahically assessed coronary artery disease. N Eng J Med 1983; 309: 385-9 7 Lipid Research Clinics Program: Lipid and Lipoprotein Analysis. In Manual of Laboratory Operations, Vol. I. Department of Health, Education and Welfare Publication (NIH) 75-628. Bethesda, MD: National Institutes of Health, 1874 8 Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chern 1972; 18: 499-502 9 Cohn JS, McNamara JR, Schaefer EJ. Lipoprotein cholesterol concentrations in the plasma of human subjects as measured in the fed and fasted states. Clin Chern 1988; 34: 2456-9 10 Tall AR, Blum CB, Forester GP, Nelson CA. Changes in the distribution and composition of plasma high density lipoproteins after ingestion of fat. J Bioi Chern 1982; 257: 198-207 II Kay RM, Rao S, Arnott C, Miller NE, Lewis B. Acute effects of the pattern of fat ingestion on plasma high density lipoprotein components in man. Atherosclerosis 1980; 36: 567-73 12 Groot PHE, Scheek LM. Effects of fat ingestion on high density lipoprotein profiles in human sera.·J Lip Res 1984; 25: 684-92 13 Merrill JR, Holly RG, Anderson RL, Rifai N, King ME, DeMeersman R. Hyperlipidemic response of young trained and untrained men after a high fat meal. Arteriosclerosis 1989; 9: 217-23 14 Rifai N, King ME, DeMeersman R, Merrill JR, Holy RG. Apolipoprotein and lipoprotein cholesterol concentrations in trained and sedentary males. Clin Chim Acta 1987; 163: 113-7 15 Warnick GR, Albers JJ. A comprehensive evaluation of the heparin-manganese precipitation procedure for estimating high density lipoprotein cholesterol. J Lip Res 1978; 19: 65-76 16 Eyre J, Hammett F, Miller NE. A micro-method for the rapid centrifugal separation of human plasma high density lipoprotein subfractions. HDL2 and HDL3. Clin Chim Acta 1981; 114: 22531 17 Allain C, Po on LS, Chan CSG, Richmond W, Fu Pc. Enzymatic determination of total serum cholesterol. Clin Chern 1974; 20: 470-5 18 Stinshoff K, Weisshaar D, Stachler F, Hess D, Gruber W, Steier E. Relation beteen concentration offree glycerol and triglycerides in human sera. Clin Chern 1977; 23: 1029-32 19 Rifai N, King ME. Immunoturbidimetric assays of apolipoprotein A, AI, All, and B in serum. Clin
Postprandial effect of a high fat meal Chern 1986; 32: 957-61 20 Terpstra J, Hessel LW, Seepers J, Van Gent CM. The influence of meal frequency on diurnal lipid, glucose and cortisol levels in normal subjects. Eur J Clin Invest 1978; 8: 61-6 21 Carey MC, Small DM, Bliss CM. Lipid digestion and adsorption. Annu Rev Phy.l'io11983; 45: 651-77 22 Redgrave TG, Carlson LA. Changes in plasma very low density and low density lipoprotein content, composition, and size after a fatty meal in normoand hypertriglyceridemic man. J Lipid Res 1979; 20:
493
217-29 23 Havel RJ, Kane JP. Kashyap ML. Interchange of apolipoproteins between chylomicrons and high density lipoproteins during alimentary lipemia in man. J Clin Invest 1973; 52 32-8 24 Krauss RL. Relationship of intermediate and lowdensity lipoprotein subspecies to risk of coronary artery disease. Am Heart J 1987; 113: 578-82
Accepted for publication 17 August 1989
