determination of plasma free fatty acids, free cholesterol, cholesteryl esters, and triacylglycerols...

ANALYTICALBIOCHEMISTRY 186,243-250 (1990)

Determination of Plasma Free Fatty Acids, Free Cholesterol, Cholesteryl Esters, and Triacylglycerols Directly from Total Lipid Extract by Capillary Gas Chromatography

Alfred Lohninger,” Peter Preis,t Leopold Linhart,* Stephan V. Sommoggy,$ Michael Landau,* and Erich Kaiser* *Department of Medical Chemistry and t2nd Department of Internal Medicine, University of Vienna, Austria, and *Department of Vascular Surgery, Technical University of Munich, Federal Republic of Germany

Received July 13,1989

An accurate capillary gas chromatographic method using different internal standards for determining free fatty acids, cholesterol, cholesteryl esters, and triacylglycerols in plasma and other biological sources is described. It is designed to give information about species composition and, consequently, more detailed information about changes in lipid metabolism of patients suffering from metabolic disorders. After plasma extraction the lipids, except phospholipids, are directly examined without any further derivatization. For free fatty acid determination the programmed temperature vaporizer (PTV) injector was heated from 40°C (sample introduction) to 190°C. In a second gas chromatographic run the PTV-injector system was heated from 60°C (sample introduction) to 4OO”C, enabling the determination of free cholesterol, cholesteryl esters, and triacylglycerol species, differing in the number of carbon atoms. Evaluation of the values obtained resulted in coefficients of variation (%) of 1.0-2.8, 2.0, 1.29- 2.24, and 2.8, for free fatty acid standards, plasma free fatty acids, cholesterol and cholesteryl ester standards, and plasma total cholesterol, respectively. Free fatty acids, cholesterol, and cholesteryl esters were not in- fluenced by storage of plasma at -24OC up to 4 days prior to extraction. The results of the gas chromatographic method and the enzymatic methods correlated well. Determination by gas chromatography yielded higher total cholesterol and lower triacylglycerol values than those values obtained by enzymatic methods. 0 1990 Academic Press, Inc.

Adipose tissue is the main source of blood free fatty acids, which represent the most actively metabolized lipid class. The fatty acid pattern of adipose tissue is

0003-2697/90 $3.00 Copyright 0 1990 by Academic Press, Inc.

. ~

closely related to the fatty acid composition of dietary lipids. But changes in the content of long-chain free fatty acids in human blood are found in many patho- physiological states. Several studies related to the acute phase of myocardial infarction have indicated changes in plasma lipids and lipoproteins (l), and determination of plasma free fatty acids is thought to afford additional information on cardiovascular disease (2). In most retro- spective studies a univariate relation has been reported between plasma triacylglycerol levels and coronary heart disease (3). Also studies dealing with nutritional problems frequently need detailed information about species composition of the main lipid classes. Further changes in plasma lipids related to the presence of a number of tumors have been described: increases of free fatty acids and the ratio of free cholesterol:esterified cholesterol and a decrease of phospholipids were ob- served, both in animal experiments and in the plasma of tumor patients (4,5).

Rapid profiling of intact plasma lipids is preferably done by high temperature gas chromatography with short capillary columns and nonpolar phases (6-9). High temperature gas-liquid chromatography is a sensitive and rapid technique for separation of neutral plasma lipids into individual lipid classes.

It was aim of the present study to adapt a recently described gas chromatographic method (6) for direct determination of plasma free fatty acids, free cholesterol, cholesteryl esters, and triacylglycerols from the total lipid extract without any further sample manipulation.

MATERIALS AND METHODS

Reagents and Standards

Chloroform and methanol were obtained from E. Merck (Darmstadt, FRG), and trihexadecanoylglycerol

243

244 LOHNING~R ET AL.

and trioctadecanoylglycerol from Serva (Heidelberg, FRG). Tridecanoylglycerol; trinonadecanoylglycerol; cholesterol; cholesteryl pentadecanoate; cholesteryl octadecanoate; cholesteryl hexadecanoate; cholesteryl octadecadienoate; tetradecanoic, pentadecanoic, hexadecanoic, heptadecanoic, octadecanoic, octadecenoic, and octadecadienoic acids; phosphatidylcholine from egg yolk; and phosphatidylethanolamine from bovine brain were obtained from Sigma Chemical Co. (St. Louis, MO). Enzymatic triacylglycerol uv test and Celichrom cholesterol enzymatic color test were obtained from Chemie Linz Diagnostica (Linz, Austria), and CHOD- PAP test for determination of total cholesterol was obtained from Boehringer-Mannheim (Mannheim, FRG).

Purity of the synthetic fatty acids, cholesteryl esters, and triacylglyceroh was confirmed by thin-layer chromatography (high-performance thin-layer chromatography plates, 10 X 20 cm, silica gel 60, from E. Merck, Darmstadt, FRG) and by gas chromato~aphic fatty acid analyses as corresponding methyl esters.

Apparatus

The analyses were carried out on a Dani Model 6500 and a Dani Model 8521 gas chromatograph (Dani S.p.A., Monza, Italy) each equipped with a programmed temperature vaporizer (PTV)i injector. A capillary glass liner was inserted into the vaporizer block of the PTV- injection systems. The low thermal mass allowed the temperature to rise within 15-20 s from 40 to 190°C and 60 to 4OO”C, respectively.

Six-meter (0.25 mm i.d.)-fused silica capillary columns with chemically bonded DB-5 (0.1 pm coating thickness) (J&W Scientific, Folsom, CA) were used for all analyses. The approximate column life was 400-500 h. Hydrogen was used as carrier gas at 0.4 bar (8-10 ml/ min flow rate) and nitrogen as the make-up gas. The FID-detector temperature was kept at 340 and 38O”C, respectively, and air and hydrogen flows were adjusted to give maximum detector response. Before use the columns were conditioned at 350°C. Subsequently, the gas chromatographic systems were calibrated with an appropriate mixture of standards. For determination of free fatty acids the oven temperature was programmed from 126 to 156”C, at a rate of 38”C/min, and from 156 to 300°C at a rate of lO”C/min. For determination of cholesterol, cholesteryl esters, and triacylglycerols, the oven temperature was programmed from 200 to 296”C, at a rate of 7.4”C/min, and from 296 to 346°C at a rate of 48”C/min.

The chromatograms were recorded on a Lingo PC-88 XT computer (Taiwan) running under MS-DOS equipped with a 40-Mbyte hard disk (for data storage)

’ Abbreviations used: PTV, programmed temperature vaporizer; CV, coefficient of variation; TMS, trimethylsilyl; FID, flame ioniza- tion detector; t-BDMS, fern-butyldimethylsilyl.

and a 12-bit analog-to-digital converter device and a Lingo AT (Taiwan) computer running under MS-DOS equipped with a 70- and a loo-Mbyte hard disk. The dig- itized FID signal was transferred from the XT to the AT computer (connected by network) and digitally filtered to eliminate the baseline noise. The peaks were marked manually by means of a keyboard-controlled cursor on the computer screen. Subsequently, the peak areas were calculated by standard algorithms and directly pro- cessed by data base software to avoid transcription er- rors.

The correction factors were calculated at four different sample: standard ratios (0.5:1, l:l, 1.5:1,2:1, respectively), multiplying the determined peak area ratios by the reciprocal value of the ratio of the weight of the test component and the internal standard.

The enzymatic total triacylglycerol and total cholesterol determinations were carried out on a Cobas bio Zentri~gan~yzer (Ho~mann-LaRoche, Basel, Switzer- land) and a Parallel Analyzer (American Monitor Corp., IN) in two clinical routine laboratories.

Procedures

Preparation of plasma samples. Blood was taken from 26 healthy male volunteers, all younger than 30 years, and from five patients suffering from multiple myelome. After an overnight fast the blood was collected into tubes containing EDTA and centrifuged. One milli- liter of plasma was mixed with 20 vol of chloroform- methanol (2:l). The lipid extracts were washed as described by Folch et al. (10) and diluted with chloroform to a defined volume. The chloroform solutions were used directly for subsequent analyses. Losses of these lipids during Folch extraction and washing were negligible.

Day-to-day variation was determined by dividing a se- rum pool into three equal portions. One portion was extracted immediately as described above, and two portions were stored at -20°C and extracted on Days 1 and 4, respectively (11).

Separation of plasma phospholipids. Plasma phospholipids were separated from neutral lipids by thin- layer chromatography using the solvent system hep- tane/diethyl ether (6/4, v/v) and recovered by elution with chloroform/methanol (l/l, v/v).

Preparation of standards. Stock solutions (1 mg/ml) of standards were prepared in chloroform. The analyti- cal samples were prepared by mixing appropriate volumes.

Qzumtitatiue determinations. The absolute amounts of plasma neutral lipids were quantitated by means of internal standards, pentadecanoic acid (free fatty acids), tridecanoylglycerol (cholesterol), cholesteryl pentadecanoate (cholesteryl esters), and trinonadecanoylglycerol (triacylglycerols), using appropriate response factors (as described above). The results were expressed in milli- grams per liter.

CAPILLARY GAS CHROMATOGRAPHY OF PLASMA LIPIDS 245

0 5 10 15 20

tirnF(n7in.l FIG. 1. Pyrolysis products of phospholipids at different temperatures of the programmed temperature vaporizer injector, (A) 19O”C, (B) ZlO”C, and (Cf 400°C.

RESULTS

Figure 1 shows the appearance of pyrolysis and dehydration products of ceramide and diacylglyeerol moieties of plasma sphingomyelins and glycerophospholipids at different temperatures of the PTV injector. No pyrolysis products were detected after injecting 4 pg of a mixture of commercial bovine brain phosphatidylethanolamine with egg phosphatidylcholine at 40°C and heating the PTV injector up to 190°C (Fig. lA), simulating the procedure of plasma free fatty acid determination. The same results were obtained for plasma phospholipids (data not shown). Heating the PTV injector from 40 up to 21O”C, small peaks occurred with retention times sim- ilar to those obtained for fatty acids (Fig. lB), and heating up to 400°C (as done for the determination of free cholesterol, cholesteryl esters, and triacylglycerols), sig- nificant amounts of pyrolysis products were found (Fig. 1C). These peaks could be resolved distinctly from those of free cholesterol and tridecanoylglycerol and of cholesteryl pentadecanoate used as internal standard (Figs. 2B and 3B). At 4OO”C, no peaks in the triacylglycerol region were found due to pyrolysis of glycerophospholipids (Fig. 2A), as described by others (12,13). In conclusion, using the PTV injector, gas chromatographic analysis of intact plasma lipids (14,15) could be extended to the quantitative determination of plasma free fatty acids, injecting the total lipid extract without any further sample manipulation. Plots of two represen- tative chromatograms of plasma lipids are shown in Fig.

3 (A shows the separation of free fatty acids; B shows the separation of cholesterol, cholesteryl esters, and triacylglycerols). For glycerolipids the carbon number was introduced as the total number of carbon atoms in the acyl groups, neglecting the three atoms in the glycerol residue.

Determination of Free Fatty Acids

Linearity. The linearity of the gas chromatographic method was determined by means of different concentrations of tetradecanoic acid, hexadecanoic acid, heptadecanoic acid, and octadecenoic acid. The internal standard concentration (pentadecanoic acid) was kept constant, and the concentrations of the other acids were varied from 0.5:1 to 2:l. There was a linear increase of the area response with increasing concentrations normalized on internal standard (data not shown).

Precision. A measure of precision was obtained cal- culating standard deviation and CV of 10 repeated analyses determining tetradecanoic acid, hexadecanoic acid, heptadecanoic acid, octadecadienoic acid, and a human plasma sample (Table 1).

Determination of Cholesterol and Cholesteryl Esters

Linearity. Linearity of the gas chromatographic method was determined by means of different concentrations of cholesterol, cholesteryl hexadecanoate, and cholesteryl octadecadienoate. The internal standard

246 LOHNINGER ET AL.

0 5 10 15 20 time2;mid FIG. 2. Pyrolysis products of a mixture of commercial bovine brain phosphatidylethanolamine and egg phosphatidylcholine (A) and the separation from internal standards used (B). Peaks: 1, cholesterol; 2, tridecanoylglycerol (internal standard); 3, pyrolysis products; 4, cholesteryl pentadecanoate (internal standard); 5, cholesteryl hexadecanoate; 6, cholesteryl octadecadienoate; 7, trihexadecanoylglycerol; 8, trioctadece- noylglyceroi; 9, trinonadecanoylglycerol.

A B

!,

30

52

57

50 I

b 5 lb

~i~~(~~ 0 5 10 15 20

t ime*;mid FIG. 3. Typical gas chromatograms of human plasma: (A) free fatty acids with 14, 15 (internal standard), 16, 18, and 20 acyl carbons (u, unsaturated); (B) 27, cholesterol; 30, tridecanoylglycerol (internal standard); 32-40, pyrolysis and dehydration products of sphingomyelins and glycerophospholipids; 42, cholesteryl pentadecanoate (internal standard); 43-47, cholesteryl esters of fatty acids with a totai of 16-20 acyl carbons; 48-56, triacylglycerols with a total number of 48-56 acyl carbons; 57, tridecanoylglycerol (internal standard).

CAPILLARY GAS CHROMATOGRAPHY OF PLASMA LIPIDS

TABLE 1

Precision of the GC Method: Fatty Acid Standards

Accuracy. Only gas chromatographic runs were taken into consideration exhibiting a sample:standard ratio of 0.5:1 to 1.5:1. Otherwise a new mixture of sample and internal standard was prepared. The amount of a single component injected was between 0.03 and 2 pg de- pending on the split ratio. Figure 5 shows the gas chromatographic total cholesterol values versus the values obtained by the enzymatic method of plasma samples obtained from volunteers. A correlation coefficient of 0.936 was obtained. Total cholesterol values determined by the gas chromatographic method were higher than the values obtained by the Celichrom cholesterol enzymatic test. A correlation coefficient of 0.984 was obtained comparing the gas chromatographic triacylglycerol values and the values of the enzymatic test (data not shown). Determination of plasma triacylglycerols by the present gas chromatographic method was described in detail previously (6). The values obtained from patients suffering from multiple myelome differed in a range of O-9.3% for free fatty acids, of O-3.9% for total cholesterol, and of 0.2-6.5% for total triacylglycerols (parallel determinations, each a single gas chromatographic analysis).

Tetradecanoic acid n Theoretical value Mean Standard deviation cv (W)

Hexadecanoic acid n Theoretical value Mean Standard deviation

cv (%)

Heptadecanoic acid n Theoretical value Mean Standard deviation cv (W)

Octadecadienoic acid n Theoretical value Mean Standard deviation cv (%)

Human plasma (total free fatty acids) n Mean Standard deviation cv (%)

10 917 mg/liter

1066 mg/liter 29.6

2.8

10

1070 mg/liter 1052 mg/liter

21.3 1.9

10 972 mg/liter 914 mg/liter

49.6 2.7

10 1160 mg/liter

950 mg/liter 17.3

1.0

10 79.05 mg/liter

1.60 2.0

concentrations (tridecanoylglycerol, cholesteryl pentadecanoate, respectively) were kept constant. The concentrations of cholesterol and the other cholesteryl esters varied from 0.5:1 to 2:1, analogous to the determination of free fatty acids. Regarding cholesterol, there was a linear increase in the area response with increasing concentrations (data not shown), while cholesteryl esters showed a declining area response with increasing concentrations when normalized on internal standards (Fig. 4). This effect was more pronounced for cholesteryl octadecadienoate than for cholesteryl hexadecanoate.

Precision. Precision was determined analogous to the way described for fatty acids. The highest CV values for a single component were found for cholesteryl octadecadienoate (Table 2). Due to summarizing the integra- tion data of four peaks the CV value of plasma total cholesterol was slightly higher than that of a single standard.

Table 3 shows the results of the evaluated day-to-day variation of free fatty acids, free cholesterol, and total cholesterol. Storage of plasma portions at -20°C for 1 or 4 days had no measurable influence on the data obtained.

247

DISCUSSION

It has been shown that short capillary columns (5-8 m), successfully employed for triacylglycerol separa-

TABLE 2

Precision of the GC Method: Cholesterol Standards

Free cholesterol n Theoretical value Mean Standard deviation cv (%)

Cholesteryl hexadecanoate n Theoretical value Mean Standard deviation cv (%)

Cholesteryl octadecanoate n

Theoretical value Mean Standard deviation cv (W)

Cholesteryl octadecadienoate n Theoretical value Mean Standard deviation cv (%)

Human plasma (total cholesterol) n Mean Standard deviation cv (%)


13.2 1.29


16.4 1.64

10 1009 mg/liter

972 mg/liter

16.8 1.73


26.7 2.24

10 1948 mg/liter

54.5 2.80


o .-

z

0.6

u 0.5

8

2

0.4 0.3

0.2

"0

Geight Ratio (j\nalyte: Int. Siij

2

+ Cholesterylhexadecanoate 0 Cholesteryloctadecadienoate

FIG. 4. Comparison of the evaluated values of cholesteryl hexadecanoate/cholesteryl pentadecanoate (0) and cholesteryl octadecadienoate/ cholesteryl pentadecanoate (+) in ratios 0.5:1, l:l, 1.5:1, and 2:1, respectively.

tions, are capable of resolving much more complex lipid mixtures (9,15-17).

One of the most important steps in high resolution gas chromatography is the method of sample introduction.

TABLE 3

Day-to-Day Variation of GC Method (mg/liter) (Human Plasma)

Free fatty acids (mg/liter) Within day

Immediately after extraction (n = 6)

Day 1 (n = 6) Day 4 (n = 6)

Between days (n = 18)

Free cholesterol (mg/liter) Within day


Day 1 (n = 6) Day 4 (n = 6)


Total cholesterol (mg/liter) Within day


Day 1 (n = 6) Day4(n=6)


Mean SD cv

139.3 2.49 1.8%

139.5 5.01 3.6%

134.0 2.52 1.9%

137.6 6.14 4.5%

Mean SD cv

262 9.4 3.6%

266 10.5 3.9%

256 8.3 3.2%

261 16.3 6.3%

Mean SD cv

1662 19.6 1.3% 1675 32.3 1.9% 1673 23.9 1.4%

1670 44.7 2.7%

High precision and accuracy can be achieved with both the cold sample injection using the PTV-injection system (split or splitless mode) and the cold on-column injection technique, if applied to mixtures ranging widely in volatility (6,18-22). For determination of plasma lipids injecting the Folch extract directly, the advantages of the PTV-injection system, compared to the cold on- column injection made, are (a) the capillary column can be of any diameter, (b) large sample volumes can be injected, and (c) nonvolatile residues are retained by the quartz wool in the glass liner. During cold on-column injection, all sample components enter the column and either unfavorable impurities can irreversibly contami- nate the stationary phase or the top of the column must be removed quite frequently. The most important disad- vantage of the PTV injector is that any small particle in the system results in peak tailing and must be removed.

To determine plasma free fatty acid concentration a number of methods have been proposed. Both titration of the acid group of the fatty acid by alkali and colori- metric determination converting the fatty acids to copper soaps, determining the copper, result in free fatty acid values higher than those of other methods (23-25). Determination of free fatty acids by high-performance liquid chromatography needs an additional derivatization step for fluorescence detection (2,26-28), and the values obtained were in the same range as those of the present method. Gas chromatographic methods either need the recovery of free fatty acids from glyceryl esters, phospholipid esters, and cholesteryl esters or require the

CAPILLARY GAS CHROMATOGRAPHY OF PLASMA LIPIDS 249

H 2000

% E

0 ‘L

fi

P 1000

0 1000 2000 3(

GC - method FIG. 5. Comparison of results (mg/liter) for total plasma cholesterol as obtained by gas chromatographic (x) and enzymatic (y) methods. n = 22, r = 0.9358, x = 1944 mg/liter, i= 1640 mg/liter.

selective extraction after derivatization to either TMS- or t-BDMS-derivatives at moderate temperatures or methylation by diazomethane (14,24,28-31). Diazo- methane is highly toxic and its preparation is hazardous. Consequently, so far, determination of plasma free fatty acids has not become a standard procedure in laboratory medicine. The advantage of the present gas chromatographic method is a rapid and accurate quanti~tion of free fatty acids without the necessity of prefractionation or derivatization after lipid extraction.

Most laboratories use enzymatic methods to measure plasma cholesterol. However, evaluation by the enzymatic methods may not coincide with the evaluations by reference methods (32,33), and any bias of the enzymatic method may cause patients to be misclassified and as- signed to the wrong risk group, according to new guide- lines for inte~retation of cholesterol concentrations (34). Thus, the method of determination must be consid- ered when interpreting blood cholesterol levels. Total plasma cholesterol values determined by the gas chromatographic method were higher than the values obtained by the enzymatic methods. This difference was much more pronounced with regard to the Celichrom cholesterol enzymatic test than to the CHOD-PAP test and might be partly due to incomplete cholesteryl ester hydrolysis (32,35,36).

The present method is suitable for rapid and accurate quantitation of plasma lipids with the advantage of giv- ing information about species composition and, consequently, more detailed information about changes in lipid metabolism of patients suffering from metabolic disorders.

ACKNOWLEDGMENTS

This work was supported by grants from the Austrian Army and the Allianz Insurance Co. (FRG). The authors thank Maria Wohlers, Barbara Lohninger, and Martina Schoderbeck for technical assis- tance.

REFERENCES

1. Mainard, F., Madec, Y., and Robinet, N. (1988) C&z. Chem. 34, 139-I 40.

2. Hatsumi, M., Kimata, S. I., and Hirosawa, K. (1986) J. Chroma- to@-. 380,247-55.

3. Cambien, F., Jacqueson, A., Richard, J. L., Warnet, J. M., Duci- metiere, P., and Claude, J. R. (1986) Amer. J. Epidemiol. 124, 624-632.

4. Barton, T. P., Cruse, J. P., and Lewin, M. R. (1987) &it. J. Cancer 56,451-454.

5. Sailer, E. L., Nordstrand, E., Juves, M. W., and Hartsock, R. J. (1979) Amer. J. Clin. Pathol. 71,83-87.

6. Lohninger, A., Linhart, L., Landau, M., Glogar, D., Kratochwil, C., and Kaiser, E. (1988) Anal. Biochem. 171,366-372.

7. Mares, P., and Husek, P. (1985) J. Chromatogr. 350,87-103.

8. Kuksis, A. (1981) J. Chromatogr. 224, l-23.

9. Myher, J. J., and Kuksis, A. (1984) J. Biochem. Biophys. Methods 10,13-23.

10. Folch, J., Lees, M., and Stanley, G. H. (1957) J. Biol. Chem. 225, 497.-509.

11. Bookbinder, M. J., and Panosian, K. J. (1986) Clin. Chem. 32, 1734-1737.

12. Kuksis, A., Stachnyk, O., and Holub, B. J. (1969) J. Lipid Res. 10, 660-667.

13. Perkins, E. G., and Johnston, P. V. (1969) Lip& 4,301-303.

14. Kuksis, A., Marai, L., and Gornall, D. A. (1967) J. Lipid Res. 8, 352-358.


15. Skorepa, J., Kahudova, V., Kotrlikova, E., Mares, P., and Todoro- vicova, II. (1983) J. C~romo~ogr. 273,180-186.

16. Kuksis, A., and Myher, J. J. (1986) J. Chromatogr. 379,57-90.

17. Geeraert, E., and Sandra, P. (1984) J. High Resold Chromatogr. Chromatogr. 7,431-432.

18. Reglero, G., Herraiz, M., and Cabezudo, M. D. (1986) Chromu~o- gr~ph~a Z&333-336.

19. Schomburg, G., Husman, II., Behlau, H., and Schulz, F. (1983) J. Chromatogr. 279,251-258.

20. Schomburg, G., Husman, H., Schulz, F., Teller, G., and Bender, M. (1983) J. Chromatogr. 279,259-267.

21. Hinshaw, J. V., Jr., and Seferovic, W. (1986) J. High Resolut. Chromatogr. Chromatogr. 9,69-72.

22. Geeraert, E., Sandra, P., and De Schepper, D. (1983) J. Chroma- togr. 279,287-295.

23. Rogiers, V. (197’7) Clin. Chim. Acta 78,227-233.

24. Mulder, C., Sehouten, J. A., and Popp-Snijders, C. (1983) J. Clin. Chem. Clin. Biochem. 2 1,823-827.

25. Regouw, B. J. M., Cornelissen, P. J. H. C., Helder, R. A. P., Spij- kers, J. B. F., and Weeber, Y. M. M. (1971) Clin. Chim. Acta 31, 187-195.

26. Shimomura, Y., Taniguchi, K., Sugie, T., Murakami, M., Sugi- yama, S., and Ozawa, T. (1984) Clin. Claim. Aeta 143,361-366.

27. Yamaguchi, M., Matsunage, R., Fukuda, K., Nakamura, M., and Ohkura, Y. (1986) Anal. Biochem. 155,256-261.

28. Jiingling, E., and Kammermeier, H. (1988) Anal. Biochem. 171, 150-157.

29. Elphick, M. C., and Lawlor, J. P. (1977) J. Ch~rn~t~gr. 130,139- 143.

30. Ko, H., and Royer, M. E. (1974) J. Chromatogr. 88,253-263.

31. Kishiro, K., and Yasuda, H. (1988) Anal. Biochem. 175,516~520.

32. Kroll, M. H., Lindsey, H., Greene, J., Silva, C., Hainline, A., Jr.,

and Elin, R. J. (1988) C&n. Chem. 34,131-135.

33. Kinter, M., Herold, D. A., Hundley, J., Wills, M. R., and Savory, J. (1988) Clin. Chem. 34,531-534.

34. Consensus Conference (1985) J. Amer. Med. Assoc. 253, 2080- 2086.

35. Wiebe, D. A., and Bernert, d. T., Jr. (1984) C&n. Chem. 30,352- 356.

36. Tel, R. M., and Berends, G. T. (1980) J. Clin. Chem. Clin. Biochem. l&595-601.

determination of plasma free fatty acids, free cholesterol, cholesteryl esters, and triacylglycerols...

Documents