| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
University Hospital Freiburg, Department of Clinical Chemistry, D-79106 Freiburg, Germany.
2
Laboratorio Chimica Clinica, Verona 37126,
Italy.
3
Evaluation Department, Roche Diagnostics, Indianapolis,
IN 46250.
4
Academic Hospital Rotterdam, Department of Clinical
Chemistry, Lipid Reference Laboratory, 3015 GD Rotterdam, The
Netherlands.
5
Washington University School of Medicine, Core
Laboratory for Clinical Studies, St. Louis, MO 63110.
6
Hôpital Robert Debré, Laboratoire Central de
Biochimie, Reims 51092, France.
7
Institut für Klinische Chemie und
Hämatologie des Kantons St. Gallen, St. Gallen 9001,
Switzerland.
8
The Johns Hopkins University School of Medicine,
Baltimore, MD 21205.
9
Childrens’s Hospital and Harvard Medical School,
Department of Laboratory Medicine, Boston, MA 02115.
a Address correspondence to this author at: University Medical Center Freiburg, Department of Clinical Chemistry, Hugstetter Strasse 55, D-79106 Freiburg i. Br., Germany. Fax 49-761-270-3444; e-mail manauck{at}med1.ukl.uni-freiburg.de
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
12% established by the National Cholesterol Education
Program. Methods: In a multicenter study, we evaluated the analytical and clinical performance of a homogeneous LDL-C assay (LDL-CRoche; Roche Diagnostics, Indianapolis, IN) in a comparison with a ß-quantification method.
Results: This direct assay correlated highly with a ß-quantification method (r = 0.968; y = 1.037x - 95.8 mg/L; n = 355; 95% confidence intervals, 1.011–1.063 for the slope and -129.5 to 62.0 mg/L for the y-intercept) and met the current total error requirement. The assay was not affected significantly by concentrations of hemoglobin up to 6000 mg/L or bilirubin up to 500 mg/L. However, a negative bias of 10% was seen when triglyceride concentrations exceeded 10 000 mg/L. At the medical decision cut-point range, the LDL-CRoche assay showed positive predictive values of 91–100% and negative predictive values of 80–99%. Furthermore, the clinical utility of the assay seemed unaffected in samples obtained postprandially.
Conclusions: The homogeneous LDL-CRoche assay meets the currently established analytical performance goals and may be useful for the diagnosis and management of hyperlipidemic patients.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The ß-quantification method, which involves an ultracentrifugation
step (LDL-CUC), is a generally accepted method to
determine LDL-C. In addition, the Friedewald calculation
(LDL-CFried), which is the most commonly used
procedure in clinical laboratories for the estimation of LDL-C, is the
routine method recommended by the NCEP Working Group on Lipoprotein
Measurement (6)(7). Although the latter method
correlates highly with the ß-quantification, it has several
shortcomings (8): it is invalid when a specimen is collected
in the nonfasting state or from a patient with type III
hyperlipoproteinemia, or in the presence of increased triglycerides
(TGs >4000 mg/L). Furthermore, because this calculation
requires the determination of three different measurements [TC, TGs,
and HDL-cholesterol (HDL-C)], each with its own analytical CV, it may
not always meet the performance criteria of total error 12%
established by the NCEP. Therefore, the NCEP Working Group on
Lipoprotein Measurement recommended the development of direct methods
for LDL-C measurement (7). A new homogeneous assay for the
determination of LDL-C has been developed recently and is being
introduced by Roche Diagnostics (LDL-CRoche;
Roche Diagnostics, Indianapolis, IN) (9). The goal of this
multicenter study was to evaluate the analytical and clinical
performance of this new assay.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
32 g of fat), were collected from
43 healthy subjects to determine the postprandial effect on the
measurement of LDL-C by this direct homogeneous method. In addition, 29
fresh-frozen human serum samples were sent to seven laboratories in the
United States and Europe, where the homogeneous
LDL-CRoche assay was performed using the
Roche/Hitachi 704, 911, 912, or 917 analyzers. These samples were
characterized by the LDL-CCRMLN method at
Northwest Lipid Research Laboratories in Seattle, WA, a Cholesterol
Reference Method Laboratory Network (CRMLN) Laboratory.
lipid measurements
TC and TGs were determined enzymatically with the CHOD-PAP (cat.
no. 450061; Roche Diagnostics) and GPO-PAP (cat. no. 1488872; Roche
Diagnostics) methods, respectively, according to the manufacturer’s
specifications. The day-to-day imprecision of the two methods,
reflected by the CV when Precinorm® L and
Precipath® HDL/LDL controls were used, was
<3%. HDL-C was measured using a homogeneous assay (cat. no. 1930672;
Roche Diagnostics) (10)(11)(12)(13). HDL-C was measured with a
day-to-day CV of <3%.
ldl-croche assay
At neutral pH (pH 7.0) and in the presence of
MgCl2, sulfated 
-cyclodextrin, and dextran
sulfate, the enzymatic reaction for cholesterol in VLDL and
chylomicrons is markedly reduced (reagent 1, cat. no. 1985604; Roche
Diagnostics). The nonionic detergent in reagent 2, which selectively
solubilizes LDL-C but not HDL-C, enables the measurement of LDL-C by a
conventional enzymatic reaction (9). The assay was
calibrated as recommended with the Calibrator for automated systems
(C.f.a.s.) LDL-C Plus calibrator, which is standardized to the
CRMLN reference method, and performed according to the manufacturer’s
recommendation.
ß-quantification (ldl-cuc)
An accurately measured volume of serum was place into an
ultracentrifugation tube, overlayed with sufficient 0.15 mol/L NaCl to
fill the tube, and centrifuged at 105 000g for 18
h at 10 °C in Baltimore and St. Louis. In Boston,
ultracentrifugation was performed at 250 000g for
3 h at 10 °C. The floating layer containing VLDL and
chylomicrons (if present) was removed, and the infranatant was
reconstituted to known volume and analyzed for cholesterol using the
enzymatic cholesterol assay. LDL-CUC was calculated as the
difference between the cholesterol concentration of the infranatant and
HDL-C, measured by dextran sulfate precipitation in Boston and St.
Louis and by homogeneous assay in Baltimore. All three laboratories
that performed the ß-quantification are certified by the CDC National
Heart, Lung and Blood Institute Lipid Standardization Program
(TC, TG, HDL-C) and two of them participate in the Alert Proficiency
Survey (Pacific Biometrics Research Foundation, Seattle, WA) to
periodically check the accuracy of the ß-quantification procedure.
The LDL-C concentrations of the 29 samples involved in the
interlaboratory survey were determined by the CRMLN Method
(LDL-CCRMLN), which uses a heparin/Mn2+
precipitation with subsequent cholesterol determination by the
Abell-Kendall method.
ldl-cfried
LDL-CFried (6) was calculated
using the following equation:
 |
after excluding samples with TGs >4000 mg/L. [TG]/5 is an estimate of VLDL-C. All concentrations are in mg/L (6).
linearity
Human LDL concentrates (Scantibodies, Inc.) were serially diluted
with 9 g/L NaCl at 11 different concentrations by the eight
participating laboratories. The predicted values were calculated by a
regression line according to Passing and Bablok, using
LDL-CRoche concentrations up to 4000 mg/L
(14).
interferences
Interferences from hemoglobin and bilirubin were determined
according to Glick et al. (15). In addition, isolated VLDL
and chylomicrons were added to different pooled sera at various
concentrations to determine the effect of increased TGs on the
measurement of LDL-CRoche.
method comparison
In three laboratories, the LDL-CRoche assay
and the LDL-CFried were compared with the
LDL-CUC. A total of 355 fresh sera were analyzed
in parallel. In addition, the influences of increasing concentrations
of LDL-CUC and TGs on the
LDL-CRoche assay were examined by bias plots
using the 355 samples mentioned above. To account for the
interlaboratory biases of the lipid determinations in the three
laboratories, the data were adjusted to CDC reference values using the
results of CDC National Heart, Lung and Blood Institute Lipid
Standardization Program Part III measurements made in the three
laboratories during the study period January 1998 through March
1998). In each laboratory, four different samples for the lipid
measurements of TC, TGs, and HDL-C were analyzed 18 times. The biases
for TC, TGs, and HDL-C were calculated for each sample, using the
differences of the results by the CDC and each laboratory; the
mean biases for TC, TGs, and HDL-C were then calculated for
each laboratory. The mean biases were -43 to 17 mg/L for TC, -12 to
24 mg/L for TGs, and 2–11 mg/L for HDL-C, respectively.
The adjustments of the data were then calculated specifically for each
laboratory. The adjustments for LDL-CUC and
LDL-CFried were performed according to the
following equations:
 |
 |
precision study
Two commercial control sera, with low and high LDL-C
concentrations, and two human serum pools were used to assess the
precision of the new homogeneous LDL-C assay in eight laboratories,
according to the NCCLS EP5-T protocol (16).
postprandial study
In this study, the effect of feeding on the determination of LDL-C
concentration by the Roche assays was examined in paired samples from
43 subjects.
storage of samples
Fresh serum samples (n = 125) were assayed using the
LDL-CRoche assay immediately after
collection and after storage at -20 °C for up to 12 months in a
noncycling freezer to determine the effect of storage on the
measurement of LDL-CRoche. In addition, the
short-term storage up to 14 days at 4 °C was also examined using two
serum pools.
statistical methods
Regression analyses were performed using the method of Passing and
Bablok (14). Total error was calculated as the sum of the
systematic error plus random error (17)(18).
Systematic error was calculated from the linear regression equation
yc =
bxc + a, where b is the slope of the
regression line, and a is the y-axis intercept
(LDL-CUC vs LDL-CRoche). At
an LDL-C concentration of xc,
systematic error was the absolute value of
yc -
xc. Random error was 1.96 x CV,
based on the run-to-run precision study. Results were considered
statistically significant at P <0.05. The positive
predictive value (PPV) of an LDL-C assay at each specified cut-point
was calculated as: [true positive/(true positive + false positive)]
x 100, where "true positive" means that the LDL-C results of both
the comparison procedure (LDL-CUC) and the test
method (LDL-CRoche or
LDL-CFried) were greater than or equal to the
cutoff concentration, and "false positive" means that the test
method LDL-C result was greater than the cut-point when the reference
procedure LDL-C value was less than the cut-point. The negative
predictive value (NPV) of an LDL-C assay at each specified cut-point
was calculated as [true negative/(true negative + false negative)] x
100, where "true negative" means that the LDL-C results of both the
reference procedure and the test method were less than the cut-point
concentration, and "false negative" means that the test method
LDL-C result was less than the cut-point when the reference procedure
LDL-C value was greater than or equal to the cut-point concentration.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
700 mg/L (human serum pool 1), 950 mg/L
(Precinorm L), 1300 mg/L (human serum pool 2), and 2100 mg/L (Precipath
L) were between 0.7% and 3.1% (Table 1
) and did not vary significantly with concentration.
|
linearity
The dilution experiment conducted in all participating
laboratories demonstrated that the assay is linear up to 4000 mg/L of
LDL-C [y = 0.996x - 1.29 mg/L;
r = 1.000; n = 52; 95% confidence interval (CI),
0.989–1.002 for the slope, and -7.26 to 1.28 mg/L for the
y-intercept]. At LDL-C concentrations >4000 mg/L, negative
biases from the expected values of approximately -5%, -9%, and
-12% were seen at 5000, 6000, and 7000 mg/L, respectively.
method comparisons
A total of 355 freshly collected sera were analyzed in three
laboratories with the LDL-CUC, the
LDL-CRoche assay, and the
LDL-CFried calculation. Means, standard
deviations (± 1 SD), and ranges for TC, TGs, HDL-C, and
LDL-CUC were 2195 ± 494 (990–3880),
2017 ± 1632 (180–12 290), 470 ± 133 (189–1081), and
1357 ± 436 (240–2758) mg/L, respectively. The
comparison-of-methods plot [LDL-CUC
(x) vs test method (y)] showed regression lines
according to Passing and Bablok of y =
1.037x - 95.8 mg/L (r = 0.968; n
= 355; 95% CI, 1.011–1.063 for the slope and -129.5 to 62.0 mg/L for
the y-intercept) for the LDL-CRoche
assay, and y = 1.025x - 69.0 mg/L
(r = 0.974; n = 313; 95% CI, 1.004–1.047 for the
slope, and -98.3 to 40.6 mg/L for the y-intercept) for the
LDL-CFried calculation (Fig. 1
). For the row data, the equations of the regression lines
according to Passing and Bablok were: y =
1.042x - 124.2 mg/L (r = 0.961; n
= 355; 95% CI, 1.011–1.072 for the slope, and -158.4 to 81.1 mg/L
for the y-intercept) for the
LDL-CRoche assay; and y =
1.024x - 66.5 mg/L (r = 0.973; n
= 313; 95% CI, 1.002–1.046 for the slope, and -6.9 to 38.0 mg/L for
the y-intercept) for the LDL-CFried
assay (data not shown). In these method comparisons, the
LDL-CFried concentrations were included only for
those samples with TG concentrations <4000 mg/L. To better illustrate
the performance of the LDL-CRoche assay in
hypertriglyceridemic samples, specimens with TG concentrations <4000
mg/L and those with TG concentrations 
4000 mg/L were compared
separately with the LDL-CUC procedure. A
regression line of y = 1.035x - 111.0
mg/L (r = 0.957; n = 313; 95% CI, 1.002–1.068
for the slope, and -151.1 to 68.4 mg/L for the y-intercept)
was obtained for samples with TG concentrations <4000 mg/L; and a line
of y = 1.074x - 169.5 mg/L
(r = 0.954; n = 42; 95% CI, 0.957–1.188 for the
slope, and -293.9 to 44.3 mg/L for the y-intercept) was
obtained for samples with TG concentrations 4000 mg/L.
|
concentration difference plots
The concentration differences of the homogeneous
LDL-CRoche assay from
LDL-CUC and LDL-CFried from
LDL-CUC were examined as a function of either
increased LDL-CUC or TGs. In samples with TG
concentrations <4000 mg/L, the LDL-CRoche assay
and the LDL-CFried showed a relatively constant
mean negative bias of -60 and -30 mg/L, respectively, independent of
the LDL-CUC concentration. In samples with TG
concentrations 4000 mg/L, the scattering of
LDL-CRoche was comparable to the results of
samples with TG concentrations <4000 mg/L, whereas the
concentration differences increased tremendously for
LDL-CFried (Fig. 2
, A and B). The negative bias of the
LDL-CRoche assay increased slightly from a mean
of -56 mg/L at a TG concentration of 1000 mg/L to a mean of -82
mg/L at a TG concentration of 10 000 mg/L. The mean bias of -30 mg/L
of LDL-CFried was only marginally influenced by
increasing TG concentrations up to 4000 mg/L. However, this negative
bias increased dramatically in samples with TG concentrations 4000
mg/L: up to more than -500 mg/L at a TG concentration of 10 000 mg/L
(Fig. 2
, C and D).
|
total error
The systematic error of the LDL-CRoche assay
at various LDL-C concentrations, encompassing the clinical decision
cut-points (1000–1900 mg/L), ranged from 0.8% to 6.3% (Table 2
). The random error was always <4%, as recommended by the NCEP
(7). This assay fulfilled the current NCEP total error
requirements for LDL-C at the clinical decision cut-point
range.
|
interlaboratory survey
The LDL-CRoche values of the 29 fresh-frozen
serum samples reported by seven participating laboratories were
compared with those obtained by the LDL-CCRMLN
method (Table 3
). Means, standard deviations (± 1 SD), and ranges for TC, TGs,
HDL-C, and LDL-CCRMLN were 2186 ± 381
(1290–2950), 1313 ± 788 (520–3510), 529 ± 160 (303–973),
and 1427 ± 348 (690–2080) mg/L, respectively. The slopes and
y-intercepts of the regression lines according to Passing
and Bablok were comparable. All of the slopes were >1, whereas the
y-intercepts were negative. The mean biases ranged from -12
to 72 mg/L. When the imprecision from these measurements was
calculated, the interlaboratory CVs of the samples were on average
2.5%, with a range of 1.7–4.4%. These findings confirm the good
agreement among the seven laboratories in the measurement of the
LDL-CRoche assay, regardless of the type of the
Roche/Hitachi analyzer used.
|
interferences
Hemoglobin at a concentration of 6000 mg/L caused a positive bias
in the LDL-CRoche assay by 10% (data not
shown). Unconjugated and conjugated bilirubin added to serum pools up
to concentrations of 500 and 250 mg/L, respectively, showed a tendency
to decrease the LDL-CRoche value by <5% (data
not shown). Isolated VLDL and chylomicrons, which were added to serum
pools at different concentrations, showed a negative bias slightly
>10% when the TG concentration exceeded 10 000 mg/L (data not
shown).
classification of patients according to ncep guidelines
The LDL-CRoche assay and the
LDL-CFried method were compared for their ability
to appropriately classify patients into treatment groups as established
by the NCEP (Fig. 3
) (4). For this purpose, the UC method used in the
three laboratories involved in the method comparison, adjusted to CDC
reference values, was considered the reference method. In those
subjects with LDL-C below 1000 and 1300 mg/L, 96% were classified
correctly using the LDL-CRoche assay. In those
with LDL-C concentrations of 1300–1600 and >1600 mg/L, 68–81% of
subjects were classified correctly by the homogeneous method. The
LDL-CFried correctly classified a slightly higher
percentage of subjects than the homogeneous method. It is
important to note, however, that with the Friedewald method, only those
samples with TG concentrations <4000 mg/L were included in the
analysis. In those with LDL-C >1900 mg/L, both methods classified at
least 89% of subjects correctly.
|
The effects of TGs on the classification of subjects into NCEP
cut-points, using both the LDL-CRoche and
LDL-CFried methods, are presented in Table 4
. The classification in samples with TG concentrations of <2000
mg/L and 2000–4000 mg/L was slightly better when the
LDL-CFried was used (88.6% vs 85.3% and 92.9%
vs 88.9%). In samples with TG concentrations >4000 mg/L, for which
the Friedewald calculation could not be used, the percentage of
correctly classified subjects by the LDL-CRoche
was nearly perfect (97.5% for TG concentrations of 4000–6000 mg/L,
and 100.0% for TG concentrations >6000 mg/L).
|
ppv and npv
The ability of the LDL-CRoche assay to
correctly classify subjects at the medical decision cut-points was
evaluated in this study population (n = 355), using the
LDL-CUC concentrations as the true values. For
the interest of comparison, the PPV and NPV of
LDL-CFried were also examined using only those
subjects with TG concentrations <4000 mg/L (n = 313). The PPV of
LDL-C estimated by either method decreased as LDL-C concentrations
increased (PPV range, 91–100% for LDL-CRoche
assay and 90–99% for LDL-CFried; Fig. 4
A). In contrast, the NPV of LDL-C estimated by either method
increased as LDL-C concentrations increased (NPV range, 80–99% for
LDL-CRoche assay and 81–99% for
LDL-CFried; Fig. 4B
). The PPVs were similar for
both methods, whereas the NPV was slightly better for the
LDL-CFried assay.
|
postprandial study
As expected, TGs increased significantly (1365 vs 2036 mg/L;
P <0.001) after feeding. Statistically significant
decreases were also seen in both TC (2009 vs 1973 mg/L; P
<0.01) and HDL-C (513 vs 490 mg/L; P <0.001) in samples
obtained in the fasting state and postprandially. Although a good
correlation was seen in LDL-CRoche concentrations
among the paired samples (y = 0.95x -
5.4 mg/L; r = 0.98), significant decreases occurred
postprandially [1345 vs 1262 mg/L (-6.2%); P <0.001].
When we used the LDL-CFried, which is not
recommended in the postprandial state, the equation of the regression
line showed a negative intercept (y =
0.97x - 90.9 mg/L; r = 0.98),
producing a significant discrepancy between the paired samples [1294
vs 1146 mg/L (-11.4%)]. However, despite this difference,
LDL-CRoche measured after feeding was equally
effective in classifying these 43 participants into NCEP cut-points. In
fact, 86% of subjects (37 of 43) were correctly classified when
nonfasting samples were used compared with 83% (36 of 43) when fasting
samples were used.
storage
When serum samples were reanalyzed 12 months after storage at
-20 °C, a good agreement (y =
1.012x - 26.41 mg/L; r = 0.984; 95%
CI, 0.991–1.035 for the slope, and -40.7 to 7.3 mg/L for the
y-intercept; baseline mean ± SD, 844 ± 425.0 mg/L; mean ±
SD after storage 834 + 441.7 mg/L; P, not
significant) between these LDL-C values and those obtained at baseline
was seen in 121 of the 124 samples examined. The three discrepant
samples were grossly hypertriglyceridemic (11 590, 16 990, and
16 700 mg/L). After storage, these samples showed a positive bias in
the LDL-CRoche assay of 2054, 1189, and 1389
mg/L, respectively. Furthermore, storage of two serum pools at 4 °C
for up to 14 days did not affect the measurement of LDL-C by the
homogeneous assay (pools 1 and 2, 725 and 1274 mg/L, respectively, at
baseline and 730 and 1277 mg/L, respectively, after 14 days).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The NCEP recommends that the total error for LDL-C determination be
<12%, which can be achieved by an imprecision of 4% and a bias
from the reference method of 4% (7). The imprecision
criterion was met by all eight laboratories that participated in this
study, whereas the bias exceeded 4% at different LDL-C concentrations.
This bias tended to increase with decreasing LDL-C concentrations.
Because the imprecision seen was low, this homogeneous assay met the
NCEP total error requirements in the range of the clinical decision
cut-points.
Good agreement was seen between the LDL-CRoche
and the LDL-CUC procedures. The bias plots
revealed a constant negative bias that was independent of the
LDL-CUC concentration. In samples with TG
concentrations 4000 mg/L, this negative bias was slightly accentuated
when the LDL-CRoche assay was used. However, the
negative bias of LDL-CFried was much more
pronounced in such samples, so that this method is obsolete in samples
with TG concentrations 4000 mg/L, as already published by Friedewald
himself (6). In addition, the results of the method
comparisons further support that the homogeneous
LDL-CRoche assay can be used in the determination
of LDL-C in hypertriglyceridemic samples. Therefore, the
LDL-CRoche assay provides laboratorians the means
to measure LDL-C in samples with increased TGs. The UC method measures
the cholesterol component of the wide-density LDL fraction, which
includes LDL, intermediate density lipoprotein, and lipoprotein(a). The
good agreement between the UC procedure and this homogeneous assay
suggests that the latter is capable of measuring the cholesterol
content of the broad density LDL fraction. However, the constant
negative bias seen between the LDL-CRoche and the
UC may be an indication of a suboptimal standardization of the new
homogeneous LDL-CRoche assay.
The effects of common interferents were investigated by addition experiments and bias plots using the serum samples included in the method comparison study. Addition experiments showed no significant interference caused by free hemoglobin up to concentrations of at least 6000 mg/L and unconjugated or conjugated bilirubin up to 500 and 250 mg/L, respectively. Different addition experiments with isolated TG-rich lipoproteins showed deviations of >10% for LDL-CRoche values when TGs exceeded 10 000 mg/L. A slight but incremental negative bias in LDL-CRoche was also apparent from the bias plot as TG concentrations increased. Although this method enables the determination of LDL-C in samples with TG concentrations >4000 mg/L, the LDL-C results tended to decrease with increasing TG values, whereas an earlier reported homogeneous method tended to increase (20).
According to the NCEP-ATP II guidelines, the management of
hyperlipidemic patients, using either dietary or drug therapy, is based
on four LDL-C cut-points (1000, 1300, 1600, or 1900 mg/L). LDL-C
concentrations determined by either the
LDL-CRoche assay or the adjusted
LDL-CFried correctly classified 86% and 89% of
the subjects, respectively, into the above mentioned cut-points. The
LDL-CRoche assay was able to correctly classify
into NCEP cut-points nearly all subjects with TG concentrations 4000
mg/L, samples in which the Friedewald LDL should not be applied. This
will provide the clinical laboratory the ability to measure LDL-C in
hypertriglyceridemic samples and alleviate the need for the expensive,
time-consuming, cumbersome UC procedure. Furthermore, our data indicate
that the LDL-CRoche assay is effective in
classifying subjects into NCEP cut-points when samples collected
postprandially are used.
Our data show that storage at -20 °C for up to 12 months did not influence the measurement of samples with TG concentrations <10 000 mg/L. However, in turbid samples with TG concentrations >10 000 mg/L, unacceptable overestimation of the LDL-C concentration (>1000 mg/L) was seen. Therefore, this assay is not adequate for use in grossly lipemic samples stored frozen at -20 °C. In addition, our data indicate that samples stored at 4 °C for up to 2 weeks did not experience any change in their LDL-C concentrations when measured by the homogeneous assay. This observation could have a practical importance to the clinical laboratory.
In conclusion, the homogeneous LDL-CRoche assay is precise and acceptably accurate. It represents an improvement in the measurement of LDL-C concentration in samples with increased TGs or samples collected postprandially and may assist in the identification of individuals at increased risk of CHD and the management of patients with hyperlipoproteinemia.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-cyclodextrin sulfate. Clin Chem 1998;44:522-531.-cyclodextrin. Clin Chem 1995;41:717-723.The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  S. Mora, N. Rifai, J. E. Buring, and P. M Ridker Comparison of LDL Cholesterol Concentrations by Friedewald Calculation and Direct Measurement in Relation to Cardiovascular Events in 27 331 Women Clin. Chem., May 1, 2009; 55(5): 888 - 894. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. T. Bairaktari, K. I. Seferiadis, and M. S. Elisaf Evaluation of Methods for the Measurement of Low-Density Lipoprotein Cholesterol Journal of Cardiovascular Pharmacology and Therapeutics, January 1, 2005; 10(1): 45 - 54. [Abstract] [PDF]
|
|
|
|
|
|
S. Usui, H. Kakuuchi, M. Okamoto, Y. Mizukami, and M. Okazaki Differential Reactivity of Two Homogeneous LDL-Cholesterol Methods to LDL and VLDL Subfractions, as Demonstrated by Ultracentrifugation and HPLC Clin. Chem., November 1, 2002; 48(11): 1946 - 1954. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. G. Miller, P. P. Waymack, F. P. Anderson, S. F. Ethridge, and E. C. Jayne Performance of Four Homogeneous Direct Methods for LDL-Cholesterol Clin. Chem., March 1, 2002; 48(3): 489 - 498. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Nauck, G. R. Warnick, and N. Rifai Methods for Measurement of LDL-Cholesterol: A Critical Assessment of Direct Measurement by Homogeneous Assays versus Calculation Clin. Chem., February 1, 2002; 48(2): 236 - 254. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Ordonez-Llanos, A. M. Wagner, R. Bonet-Marques, J. L. Sanchez-Quesada, F. Blanco-Vaca, and F. Gonzalez-Sastre Which Cholesterol Are We Measuring with the Roche Direct, Homogeneous LDL-C Plus Assay? Clin. Chem., January 1, 2001; 47(1): 124 - 126. [Full Text] [PDF]
|
|
|
|
|
|
B. D. Ragland, R. J. Konrad, C. Chaffin, C. A. Robinson, and R. W. Hardy Evaluation of a Homogeneous Direct LDL-Cholesterol Assay in Diabetic Patients: Effect of Glycemic Control Clin. Chem., November 1, 2000; 46(11): 1848 - 1851. [Full Text] [PDF]
|
|
|
|
|
|
D. Furuya, A. Yagihashi, S. Nasu, T. Endoh, T. Nakamura, R. Kaneko, C. Kamagata, D. Kobayashi, and N. Watanabe LDL Particle Size by Gradient-Gel Electrophoresis Cannot Be Estimated by LDL-Cholesterol/Apolipoprotein B Ratios Clin. Chem., August 1, 2000; 46(8): 1202 - 1203. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, samples with TGs <4000 mg/L (n = 313); 
, samples with TGs
