Clinical Chemistry 47: 532-539, 2001;
(Clinical Chemistry. 2001;47:532-539.)
© 2001 American Association for Clinical Chemistry, Inc.
Triple Lipid Screening Test: A Homogeneous Sequential Assay for HDL-Cholesterol, Total Cholesterol, and Triglycerides
Maureen L. Sampson1,
Andrea Aubry1,
Gyorgy Csako1 and
Alan T. Remaleya,1
1
Clinical Chemistry Service, Department of Laboratory Medicine, Clinical Center, National Institutes of Health, Bethesda, MD 20892.
a Address correspondence to this author at: Department of Laboratory Medicine, National Institutes of Health, Bldg. 10, Rm. 2C-407, Bethesda, MD 20892-1508. Fax 301-402-1885; e-mail
aremaley{at}nih.gov.
 |
Abstract
|
|---|
Background: The analysis of lipids in serum lipoprotein fractions
is useful in assessing the risk for coronary artery disease, but it
typically involves performing multiple tests. An automated single-tube
assay, referred to as the triple lipid screening (TLS) test, can be
used for measuring HDL-cholesterol (HDL-C), total cholesterol,
and triglycerides (TGs) with no specimen pretreatment.
Methods: The first part of the assay is based on a homogeneous
assay for HDL-C that uses either an anti-apolipoprotein B antibody
(TLS-A test) or a polyanion (TLS-B test) that blocks the enzymatic
measurement of cholesterol on the non-HDL fraction. After the addition
of deoxycholate, which solubilizes the unreacted cholesterol from the
non-HDL fraction, the remaining cholesterol in the sample is
subsequently measured enzymatically. Using the same enzyme detection
system as the cholesterol assay, TGs are measured in the last step,
after the addition of the enzymes for the TG assay.
Results: The TLS assay (y) had acceptable analytic
performance and compared favorably with standard tests
(x) for each analyte: for HDL-C, TLS-A =
0.99x + 0.19 (R = 0.980);
TLS-B = 1.00x - 0.15 (R =
0.974); for total cholesterol, TLS-A = 1.03x + 0.12
(R = 0.997); TLS-B = 1.07x
- 0.30 (R = 0.965); and for TGs,
TLS-A = 1.02x + 0.02 (R =
0.988); TLS-B = 1.04x - 0.28
(R = 0.980).
Conclusions: The TLS test is a single-tube homogeneous assay for
the analysis of all of the major serum lipoprotein fractions and can be
used as a simple screening test for the detection of
hyperlipidemia.
|
Introduction
|
|---|
Cholesterol is transported in the blood on various types of
lipoprotein particles that differ in their density and in their role in
the pathogenesis of atherosclerosis (1). The current
guidelines for the laboratory evaluation of hyperlipidemia recommend
measuring the cholesterol content of the serum lipoprotein fractions
for assessing the risk of coronary artery disease (2).
HDL-cholesterol
(HDL-C),1
a negative risk factor for coronary artery disease
(3)(4)(5), and LDL-cholesterol (LDL-C), a positive risk factor
(1)(2), are the most commonly measured
lipoprotein cholesterol fractions. Serum triglyceride (TG)
concentrations are also often determined in the analysis of serum
lipoproteins because they can be used for estimating LDL-C
(6), and because of the role of TGs as a positive
risk factor for coronary artery disease (7). Most
laboratories analyze serum lipoproteins by performing three separate
tests, namely tests for total cholesterol (TC), HDL-C, and TGs.
Because of the relatively high frequency of hyperlipidemia in the
general population and because of the recommendation that all adults
undergo periodic cholesterol screening (2), any procedure
that would simplify and reduce costs for the analysis of serum
lipoproteins would be advantageous. The recent development of
homogeneous assays for lipoprotein cholesterol fractions that do not
require any specimen preprocessing was a major advance in simplifying
lipoprotein lipid analysis (8)(9). There have
also been several reports of single-tube assays that involve the
sequential enzymatic measurement of TC and TGs (10)(11)(12).
These assays use a common enzyme detection system, but they
sequentially measure TC and TGs by taking advantage of the lipid
substrate specificity of the enzymes used in each part of the assay
(10)(11)(12). We recently described a sequential test for HDL-C
and TC (13), referred to as the dual HDL/TC test. Although
these sequential test designs reduce the number of tests needed, none
of them provides all of the necessary results for performing a complete
analysis of the major lipoprotein cholesterol fractions.
In this report, we describe a new sequential test, the triple-lipid
screening (TLS) test, that can be used for measuring HDL-C, TC, and
TGs. The TLS test begins with a homogeneous HDL-C assay that uses
either an antibody (TLS-A) or a polyanion (TLS-B) to block the
measurement of cholesterol from the non-HDL fraction. After the
addition of a detergent, TC and TGs are sequentially measured. The TLS
test is performed in a single tube with no specimen pretreatment, can
be readily automated, and compares favorably with standard lipoprotein
lipid assays.
|
Materials and Methods
|
|---|
reagents
Direct HDL-C (immunoinhibition based; EZ-HDL), direct LDL-C
(EZ-LDL), and GPO-Trinder TG reagent sets, and deoxycholate,
glycerol standards, ATP, lipase, glycerol kinase, glycerol
phosphate oxidase, ascorbic acid, and a lipid linearity set were
obtained from Sigma. A polyanion-based, homogeneous assay for
HDL-C (Liquid N-geneous HDL) was obtained from Genzyme. TG (glycerol
blanked) and TC reagents for the Hitachi 917 were obtained from Roche.
Reagents for a dextran precipitation-based method for HDL-C were
purchased from Polymedco. Unconjugated bilirubin was obtained from
Pfanstiehl Laboratories. Purified human HDL (d =
1.125–1.21 kg/L) and LDL (d = 1.009–1.063
kg/L) were obtained from EDTA-plasma by density gradient
ultracentrifugation (14). Isolated lipoproteins were >95%
pure, as judged by agarose gel electrophoresis (Helena Laboratories).
procedure for the tls test
The Cobas FARA II analyzer (Roche) was used to perform the TLS,
direct LDL-C, and HDL-C tests by the Polymedco method. A description of
the reagents and assay conditions for the two versions of the
TLS test are shown in Table 1
. The first part of the TLS-A test uses the R1 (antibody
reagent) and R2 (cholesterol reagent) components of the EZ-HDL-C
reagent set, whereas the TLS-B test uses the R1 (polyanion reagent) and
R2 (cholesterol reagent) components from the N-geneous HDL-C reagent
set. Randomly collected (fasting and nonfasting) serum samples were
used to develop and evaluate the TLS tests. For the TLS-A assay, the
serum sample was diluted 10-fold with phosphate-buffered saline by the
analyzer before the first step of the assay.
We calibrated both the HDL-C and TC parts of the assay for cholesterol,
using either the Sigma EZ-HDL calibrator for the TLS-A test or the
Genzyme HDL calibrator for the TLS-B test. TC was calibrated by taking
the factor (i.e., the slope of the calibration curve) generated in step
1B (HDL-C) of the assay and applying it to the absorbance data
generated during step 2 (TC). The TG assay was calibrated with glycerol
standards (Sigma). The enzyme reactions were monitored colorimetrically
with peroxidase (4-aminoantipyrine) and
N-ethyl-N-(2-hydroxy-3-sulfopropyl)-3,5-dimethoxy-4-fluoroanaline,
sodium salt (TLS-A), or
N,N-bis(4-sulfobutyl)-m-toluidine-disodium
(TLS-B) supplied in the homogeneous HDL-C reagent sets. Absorbance
values were converted to concentrations by an endpoint calculation,
with the absorbance at the beginning of each step as the baseline and
the absorbance at 5 min from the start of the step as the final point.
LDL-C was calculated using the Friedewald equation (6).
analytical evaluation of the tls test
The specificity of the TLS test for lipoprotein cholesterol
fractions was assessed by analyzing samples supplemented with either
purified HDL or LDL. Interassay precision was calculated from 20
replicate results obtained during a single analysis of a serum pool.
Interassay precision was calculated based on the analysis of a frozen
serum pool measured in 10 separate analytical runs (2 runs per day for
5 days) using a different calibration for each run. Interference in the
TLS assay by hemolysis was tested after the addition of a lysate from
washed human red blood cells. Interference in the TLS assay by
bilirubin and ascorbic acid was tested after the addition of either
unconjugated bilirubin or a fresh solution of ascorbic acid. The
interfering substances were added to a serum pool in eight different
concentrations and assayed in triplicate; deviations >10% compared
with the unadulterated samples were considered significant. Linearity
for the TLS tests was assessed using a lipid linearity set (Sigma) at
seven different concentrations assayed in triplicate. Deviations >10%
from the linear regression line were considered significant. Standard
tests for HDL-C (Polymedco), TC (Roche), TGs (Roche), and direct LDL-C
(Sigma) were compared with the results of the TLS test by Deming
regression analysis and Bland-Altman plots. A different set of serum
samples was used in the comparison study for the TLS-A and TLS-B tests.
|
Results
|
|---|
principle of the tls assay
The reaction profile of serum for the TLS-A test is shown in Fig. 1
. Three stepwise increases in absorbance were observed, which
correspond to the sequential measurement of HDL-C, TC, and TGs. In step
1A, an anti-apolipoprotein B (apoB) antibody was added, which bonded to
the surface of the apoB-containing lipoproteins (LDL, VLDL, and
chylomicrons). After the addition of cholesterol esterase and
cholesterol oxidase (step 1B), only cholesterol from HDL particles was
detected, because the enzymes were sterically blocked by the antibody
from reacting with cholesterol on the apoB-containing lipoproteins. The
residual or unreacted cholesterol on the apoB-containing lipoproteins
(non-HDL-C) was then measured after the addition of deoxycholate in
step 2. Deoxycholate disrupted the antibody-apoB lipoprotein complex
and allowed the subsequent enzymatic measurement of the unreacted
cholesterol in the sample. The sum of the cholesterol measured in step
1B (HDL-C) and step 2 (non-HDL-C) was equal to TC. In step 3, TGs were
measured after the addition of the enzymes for the GPO-Trinder
reaction. The enzymatic reaction of TGs was monitored with the same
peroxidase-based detection system that was used for cholesterol in
steps 1B and 2. The reaction for each step of the assay was essentially
complete after 5 min at 37 °C, based on the reaction profile of
serum.
View larger version (28K):
[in this window]
[in a new window]
|
Figure 1. Absorbance profile of the TLS test.
Serum was analyzed by the TLS-A test, and the change in absorbance was
monitored during the individual steps of the assay.
|
|
The TLS-B test of serum also produces three stepwise increases in
absorbance (data not shown) and was similar in principle to the TLS-A
test, except for the procedure used to block the non-HDL fraction. In
step 1A, a polyanion instead of an antibody was used to sterically
block the enzymatic measurement of cholesterol from the non-HDL
particles. In addition, the polyanion inhibition method contained a
detergent that, in the presence of the polyanion, selectively
solubilized HDL in step 1B. Like the TLS-A test, the addition of
deoxycholate in step 2 of the TLS-B test disrupted the complex between
the polyanion and the apoB-containing lipoproteins, thus allowing
the measurement of cholesterol on the non-HDL particles. TGs were then
measured enzymatically in the final step of the assay.
lipoprotein specificity of the tls assay
The change in absorbance for the two versions of the TLS test for
purified HDL and LDL are shown in Fig. 2
. Cholesterol was detected only during step 1B of the reaction
when HDL was used as a sample. No additional change in absorbance was
observed after the addition of deoxycholate in step 2, which indicated
that the reaction of HDL-C was complete after step 1B. In contrast,
when purified LDL was analyzed, a change in absorbance was observed
only in step 2, which indicated that the enzymatic measurement of
cholesterol on LDL was prevented in step 1B by either the antibody
(Fig. 2A
; TLS-A) or the polyanion (Fig. 2B
; TLS-B). For both the HDL
and LDL samples, TGs were detected in step 3. These results indicated
that steps 1 and 2 of both versions of the TLS test were specific for
measuring cholesterol from HDL and non-HDL lipoproteins, respectively.
View larger version (21K):
[in this window]
[in a new window]
|
Figure 2. Lipoprotein specificity of the TLS tests.
(A), analysis of purified HDL ( ; final concentration,
1.6 mmol/L) and purified LDL ( ; final concentration, 6.6 mmol/L),
diluted in phosphate-buffered saline, by the TLS-A test.
(B), analysis of purified HDL (; final concentration,
1.1 mmol/L) and purified LDL (; final concentration, 4.1 mmol/L),
diluted in phosphate-buffered saline, by the TLS-B test.
|
|
precision, linearity, and interference of the tls assay
The analytical characteristics of the TLS-A and TLS-B tests are
shown in Table 2
. The intra- and interassay precision of the different steps of
the TLS assays were similar to the precision of the commonly used
individual lipid assays described previously (15). The TC
part of the assay had the best precision, followed by the HDL-C and TG
parts of the assay. Both versions of the TLS test were linear
throughout the usual concentration range of the measured analytes.
Interference by bilirubin, hemolysis, and ascorbic acid with the TLS
tests was similar to the limits described by the manufacturers of the
homogeneous HDL-C used in step 1 of the assays. Both TLS tests
were relatively unaffected by bilirubin; no significant bias was
observed for any step of the assays in samples containing up to 880
µmol/L bilirubin. For hemolysis, no significant interference was
observed for samples containing hemoglobin in concentrations
corresponding to gross hemolysis (10 g/L hemoglobin for TLS-A;
5.0 g/L hemoglobin for TLS-B). No significant interference was observed
for either assay when ascorbic acid concentrations were <1000 mg/L,
which was much higher than the concentration typically observed in
human sera.
comparison of tls assay with standard tests
The performance of the TLS tests was compared with the standard
tests for HDL-C, TC, and TGs using serum specimens that had a wide
range of lipid values. The results of the regression analysis for
specimens with <4.5 mmol/L TGs are shown in Table 3
, and the difference plots are shown in Fig. 3
. The HDL-C measurements from both TLS assays agreed well with
the results of the standard HDL-C assay (Table 3
), although the TLS-A
assay showed a small, fixed positive bias of 0.19 mmol/L (Fig. 3A
)
relative to the standard test, whereas the TLS-B test showed a small,
fixed negative bias of 0.15 mmol/L (Fig. 3B
). Measurement of HDL-C by
the immunoinhibition method has been described previously to cause a
positive bias, compared with precipitation-based methods, because of
the underestimation of cholesterol on apoE-containing HDL lipoproteins
by precipitation-based methods (8)(9). Specimens
with a TG concentration >4.5 mmol/L (Fig. 3
; •) did appear to show a
negative bias for HDL-C in the TLS-B assay, but not in the TLS-A assay.
Overall, the TC measurement by both TLS assays also compared well with
the standard TC assay (Table 3
). There was, however, a small, fixed
positive bias of 0.12 mmol/L for the TLS-A assay for TC (Fig. 3C
),
whereas the TLS-B assay showed increased variation relative to the
standard TC test (Fig. 3D
). Both tests yielded TG results comparable to
the standard TG assay, but showed a poorer correlation for those
specimens containing >4.5 mmol/L TGs (Fig. 3, E and F
). When the TLS
results from serum specimens with <4.5 mmol/L TGs were used to
calculate LDL-C, they closely matched the calculated results based on
the standard tests (Table 3
and Fig. 4, A and B
). The calculated LDL-C results from the TLS assay also
matched reasonably well with LDL-C as measured by a direct LDL-C assay
(Fig. 4, C and D
).
View larger version (33K):
[in this window]
[in a new window]
|
Figure 3. Comparison of TLS test to standard tests.
The TLS-A test (A, C, and
E) and the TLS-B test (B,
D, and F) for HDL-C (A and
B), TC (C and D), and TGs
(E and F) were compared with the standard
tests by Bland-Altman plots. The y-axis shows the
difference between the TLS test and the standard test (TLS test -
standard test), and the x-axis is the mean value of the
two assays. •, serum specimens with a TG concentration >4.5 mmol/L;
, serum specimens with <4.5 mmol/L TGs.
|
|
View larger version (27K):
[in this window]
[in a new window]
|
Figure 4. Comparison of LDL-C measurement by the TLS test vs the
standard test.
The results of the TLS-A test (A and C)
and the TLS-B test (B and D) were used to
calculate LDL-C and compared by Bland-Altman plots to the calculated
results based on the standard tests (A and
B) or to the direct LDL-C measurement (C
and D). The y-axis shows the difference
between the TLS test and the standard test (TLS test - standard
test), and the x-axis is the mean value of the two
assays. Only results for specimens with <4.5 mmol/L TGs are shown.
|
|
|
Discussion
|
|---|
We found that the assay order, substrate specificity, and
the detection system were important considerations in the development
of the test. Although a different assay order would have been possible
if only two of the three analytes were sequentially measured, only the
order shown in Fig. 1
(HDL-C 
TC TGs) is compatible for
sequentially determining all three analytes. The measurement of TGs in
an earlier step would have disrupted the structure of the lipoproteins
by the lipolysis of TGs, and the TG measurement is more effective after
the addition of detergent, which solubilizes TGs from the core of the
lipoprotein particles. Similarly, only HDL-C followed by TC is
compatible for sequentially measuring both analytes. Deoxycholate was
found suitable for disrupting both the antibody-apoB lipoprotein and
the polyanion-apoB lipoprotein complexes and allowed the measurement of
TC to proceed after the HDL-C step. The addition of deoxycholate at the
beginning of the HDL-C step would have prevented the selective
measurement of cholesterol from just HDL. The analyte specificity of
each part of the TLS assay was, therefore, dependent on the assay order
and on the inherent substrate specificity of the lipid-modifying
enzymes used in each part of the assay. A common enzyme assay detection
system was used for each part of the TLS test to simplify the test and
to potentially reduce reagent costs. The enzymes used in the
measurement of both TC and TGs generate
H2O2, which in the presence
of peroxidase oxidizes 4-aminoantipyrine and produces a visible color.
The signal from the assay detection system was relatively stable. No
significant change in absorbance was observed after the completion of
steps 1B and 2 (Fig. 1
) for as long as 30 min after the start of the
reaction (data not shown). The absorbance at the end of each step,
therefore, could serve as the baseline for the endpoint calculation of
the subsequent step.
Several procedures for the simultaneous or sequential measurement of
various analytes, aimed at improving the efficiency of laboratory
testing and reducing costs, have been reported
(10)(11)(12)(16)(17)(18)(19)(20) Lipoprotein lipid analysis is
ideally suited for this type of application because multiple lipid
fractions are routinely measured on a single sample. The most
recent recommendations for screening for hyperlipidemia and for
monitoring cholesterol-lowering therapies emphasize the importance of
measuring both LDL-C and HDL-C (2). There have been several
tests developed for the sequential measurement of TC and TGs
(10)(11) and one test that sequentially measures
TC, TGs, and phospholipids (12). These assays, however, are
not as clinically useful because of the importance of measuring not
only TC and TGs, but also the cholesterol content of HDL and LDL
(2). In contrast, the TLS assay provides results for HDL-C,
along with TC and TGs. The TLS results from nonhypertriglyceridemic
specimens (TGs <4.5 mmol/L) matched well with the standard tests (Fig. 3
), particularly in the range of lipid values typically used for
diagnosing hyperlipidemia (1)(2). In addition,
the calculated LDL-C from the TLS tests closely matched the calculated
LDL-C result from the standard tests, as well as the result from a
direct LDL-C assay (Fig. 4
). In an earlier report (20), we
showed that the homogeneous HDL-C method used for the TLS-A test can
also provide an estimate of the apoB concentration by monitoring the
change in turbidity induced after the anti-apoB antibody is added in
step 1A. The TLS tests, therefore, can be used to measure all of the
commonly used lipid values for the screening of hyperlipidemia.
One other advantage of the TLS test compared with the current multiple
test format of lipoprotein lipid analysis is that it is a single-tube
assay and, therefore, potentially requires fewer disposable items, such
as reaction cuvettes. More importantly, there is an overall reduction
in the use of reagents. The HDL-C and TC parts of the TLS test use the
same enzymes, and the TC step does not require any additional reagents
except for deoxycholate (Fig. 1
), which is relatively inexpensive.
Additional enzyme reagents are added for measuring TGs, but the TG
reaction is monitored with the same enzyme detection system used for
the HDL-C and TC parts of the assay. It is estimated that there would
be a >99% reduction in reagent costs for the TC assay and an 
20%
reduction for the TG assay if both tests are performed as part of the
TLS test, based on the retail price of the individual components of the
TC and TG assays and the cost of deoxycholate. Depending on the
operations of a given laboratory, there may also be some savings in
labor for the technologist performing the TLS assay, particularly when
compared with precipitation-based HDL-C assays, which require a
substantial amount of manual specimen preparation
(8)(9).
The setting in which the TLS test may be the most useful is
point-of-care sites, such as physician offices, in which multiple lipid
fractions frequently are measured for screening of hyperlipidemia. Any
abnormal result from the TLS test could subsequently be confirmed by
individual standard lipid tests. A limitation of the TLS assay is that
although the individual components of the test are commercially
available, the packaging of the reagents as a single reagent
set would make it easier and more practical to use the assay
and would likely improve the analytical performance of the assay
because of the improved standardization of the reagents and the
calibrators. Another current limitation of the TLS assay is that it
requires an analyzer, such as the Cobas FARA, that can add
multiple reagents to a reaction cuvette. However, because the TLS assay
is a single-tube and single-wavelength colorimetric assay, it could be
readily adapted to a simple analyzer dedicated to the point-of-care
testing of hyperlipidemia.
In conclusion, the TLS test offers a convenient way to simplify
lipoprotein lipid analysis. It can be used for directly measuring
HDL-C, TC, and TGs, and for calculating LDL-C; thus, the test provides
results for the four commonly monitored lipoprotein lipid values.
Because of the potential of cost savings, the TLS test may be
particularly useful for the screening of hyperlipidemia.
|
Footnotes
|
|---|
1 Nonstandard abbreviations: HDL-C and LDL-C, HDL- and LDL-cholesterol; TG, triglyceride; TC, total cholesterol; TLS, triple lipid screening; and apo, apolipoprotein. 
|
References
|
|---|
-
Wilson PWF, D’Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation 1998;97:1837-1847.[Abstract/Free Full Text]
-
. The Expert Panel. Summary of the second report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection (Adult Treatment Panel II). JAMA 1993;269:3015-3023.[Abstract/Free Full Text]
-
Gordon T, Castelli WP, Hjortland MC, Kannel WB, Dawber TR. High-density lipoprotein as a protective factor against coronary heart disease: the Framingham study. Am J Med 1977;62:707-714.[Web of Science][Medline]
[Order article via Infotrieve]
-
Badimon JJ, Fuster V, Badimon L. Role of high-density lipoproteins in the regression of atherosclerosis. Circulation 1992;86:86-94.
-
Hoeg JM, Remaley AT. Reverse cholesterol transport. Goldbourt U de Faire U Berg K eds. Genetic factors in heart disease 1994:351-372 Kluwer Academic Publishers New York. .
-
Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972;18:499-502.[Abstract]
-
Miller M. The epidemiology of triglyceride as a coronary artery disease risk factor. Clin Cardiol 1999;22:111-116.
-
Lin M, Hoke C, Ettinger B. Evaluation of homogeneous high-density lipoprotein cholesterol assay on a BM/Hitachi 747-200 analyzer. Clin Chem 1998;44:1050-1052.[Free Full Text]
-
Nauck M, Marz W, Wieland H. New immunoseparation-based homogeneous assay for HDL-cholesterol compared with three homogeneous and two heterogeneous methods for HDL-cholesterol. Clin Chem 1998;44:1443-1451.[Abstract/Free Full Text]
-
Zoppi F. Single-cuvet sequential determination of triglyceride and cholesterol. Clin Chem 1985;31:2036-2039.[Abstract]
-
Sharma A, Artiss JD, Zak B. A method for the sequential colorimetric determination of serum triglycerides and cholesterol. Clin Biochem 1987;20:167-172.[Web of Science][Medline]
[Order article via Infotrieve]
-
Nanjee MN, Miller NE. Sequential microenzymatic assay of cholesterol, triglycerides and phospholipids in a single aliquot. Clin Chem 1996;42:915-926.[Abstract/Free Full Text]
-
Sampson ML, Csako G, Remaley AT. Dual HDL/total cholesterol test: a single-tube, homogeneous assay for sequential measurement of HDL cholesterol and total cholesterol. Ann Clin Biochem 2000;37:479-487.
-
Schumaker VN, Puppione DL. Sequential flotation ultracentrifugation. Methods Enzymol 1986;128:155-169.[Web of Science][Medline]
[Order article via Infotrieve]
-
Rifai N, Cole TG, Iannotti E, Law T, Macke M, Miller R, et al. Assessment of interlaboratory performance in external proficiency testing programs with a direct HDL-cholesterol assay. Clin Chem 1998;44:1452-1458.[Abstract/Free Full Text]
-
Zak B, Ressler N. Simultaneous microdetermination of copper and iron using mixed phenanthrolines. Anal Chem 1956;28:1158-1161.
-
Ferguson JW, Richard JJ, O’Laughlin JW, Banks CV. Simultaneous spectrophotometric determination of calcium and magnesium with chlorophosphonazo III. Anal Chem 1964;36:796-799.
-
Yee HY, Goodwin JF. Simultaneous determination of copper and iron in a single aliquot of serum. Clin Chem 1974;20:188-191.[Abstract]
-
McArthur M, Chong-Kit R. Multiple wavelength readings on the eni gemsaec for keto and ketogenic steroids. Clin Chem 1973;19:659.
-
Sampson ML, Csako G, Remaley AT. Estimation of serum apolipoprotein B by a modified homogeneous assay for HDL-cholesterol. Clin Chem 2000;46:869-871.[Free Full Text]
Top
Abstract
Introduction
