| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
Department of Pathology, University of Utah Health Sciences Center, Salt Lake City, UT 84132.
2
ARUP Institute for Experimental and Clinical Pathology,
Salt Lake City, UT 84108.
3
Departments of Laboratory Medicine and Pathology,
Children’s Hospital and Harvard Medical School, Boston, MA 02115.
a Address correspondence to this author at: c/o ARUP Laboratories, 500 Chipeta Way, Salt Lake City, UT 84108. Fax 801-584-5207; e-mail william.roberts{at}arup-lab.com
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: The Dade Behring BN II, the Abbott IMx, the Diagnostic Products Corporation IMMULITE, and the Beckman Coulter IMMAGE are four automated analyzers with high-sensitivity CRP (hs-CRP) methods. We evaluated these assays for precision, linearity, and comparability with samples from 322 apparently healthy blood donors.
Results: The imprecision (CV) of the BN II, IMx, IMMULITE, and
IMMAGE methods was 
7.6%, 12%, 9.8%, and 9.7% at 3.5
mg/L, respectively. The BN II, IMx, IMMULITE, and IMMAGE methods were
linear down to 0.30, 0.32, 0.85, and 2.26 mg/L, respectively. CRP
concentrations demarcating each quartile in a healthy population were
method dependent. The IMx method gave results comparable to the BN II
method for values in the reference interval. The IMMULITE method had a
positive intercept compared with the BN II method. The IMMAGE method
demonstrated more scatter and a positive intercept compared with the BN
II method, which may reflect the fact that it is a less sensitive
assay.
Conclusions: The four hs-CRP methods exhibited differences in results for a healthy population. Additional standardization efforts are required to ensure that hs-CRP results can be related to large-scale epidemiologic studies.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Chronic inflammation is an important component in the development of atherosclerosis [for recent reviews, see Refs. (1)(2)]. CRP concentrations have been shown to correlate with markers of endothelial dysfunction (3)(4). Numerous recent studies have demonstrated that CRP can be used to help predict the risk of acute events in patients with atherosclerosis. Included in this list are several studies conducted in large populations of apparently healthy men and women who subsequently developed coronary artery disease, cerebrovascular disease, or peripheral arterial disease (5)(6)(7)(8)(9)(10). CRP has also been shown to predict risk of future events in patients with acute coronary syndromes and in patients with stable angina and coronary artery stents (11)(12)(13)(14)(15)(16)(17).
A variety of different methods for quantifying the CRP concentration in serum have been used. When studies have been conducted with apparently healthy individuals, high-sensitivity CRP (hs-CRP) methods that can accurately divide results within the reference intervals into quartiles or quintiles have been required. These high-sensitivity methods usually have used ELISA methodology, and a single in-house ELISA assay was used for several epidemiologic studies (5)(6). This methodology is primarily for research use and generally is not routinely available in clinical laboratories. Standard CRP methods in the clinical laboratory have limits of detection of 3–5 mg/L and are unsuitable for these applications. Recently, several automated hs-CRP assays have been developed that are suitable for routine use in the clinical laboratory. They possess greater precision at low concentrations of CRP. We evaluated the performance characteristics of four of these, including method comparability, using samples from 322 apparently healthy adult blood donors.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
apparatus
The BN II nephelometer was from Dade/Behring, the IMx analyzer was
from Abbott Diagnostics, the IMMULITE analyzer was from Diagnostics
Products Corporation, and the IMMAGE nephelometer was from
Beckman/Coulter.
assay procedures
All four methods used the manufacturers’ reagents as directed.
The BN II N High Sensitivity CRP assay utilizes a monoclonal
antibody coated to polystyrene particles and fixed-time kinetic
nephelometric measurements (18). The BN II nephelometer
makes a 1:400 dilution to measure CRP concentrations between 3.5 and
210 mg/L and a 1:20 dilution below 3.5 mg/L. The IMx assay uses an
automated microparticle capture enzyme immunoassay with two mouse
anti-CRP monoclonal antibodies (19). A 1:50 manual dilution
provides a measurable range of 0.05–30 mg/L. The IMx hs-CRP method is
not and has never been commercially available. The IMMULITE assay is a
two-site chemiluminescent enzyme immunometric assay with one monoclonal
and one polyclonal anti-CRP antibody. A 1:100 manual dilution provides
a measurable range of 0.1–500 mg/L. The IMMAGE assay uses a polyclonal
anti-CRP antibody coated to latex particles and rate nephelometric
measurements. The IMMAGE nephelometer makes a 1:36 dilution for values
up to 80 mg/L and a 1:216 dilution for higher concentrations. The
limits of detection were 0.01 mg/L [zero calibrator + 3 SD
(18)], 0.05 mg/L [zero calibrator + 3 SD (2)],
0.1 mg/L (zero calibrator + 2 SD; manufacturer’s claim), and 1.0 mg/L
(linearity study; manufacturer’s claim) for the BN II, IMx,
IMMULITE, and IMMAGE methods, respectively. Results below the detection
limits of the assays were reported as the detection limits of the
assays. All methods except the IMx have been approved by the Food and
Drug Administration (FDA) for clinical use in the United States. Only
the N High Sensitivity CRP assay has been approved by the
FDA for use in assessing the risk of cardiovascular and peripheral
vascular disease.
Samples for linearity and precision studies were prepared from two
serum pools. The low pool was prepared by combining samples from blood
donors with hs-CRP concentrations in the lowest quartile. The high pool
was prepared by combining patient samples with hs-CRP concentrations of
10 mg/L. The high pool was diluted with the low pool to the
following final percentage of high pool: 100%, 75%, 50%, 30%, 20%,
10%, 5%, and 0%. Samples were assayed in duplicate on 10 different
days.
data analysis
EP Evaluator release 3 software (David G. Rhoads Associates) was
used for Deming regression analysis, calculation of r, and
Sy|x. hs-CRP concentrations were skewed rightward
in samples from blood donors; therefore, percentile values were
estimated and hs-CRP concentrations were log transformed for method
comparison plots.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). The BN II method demonstrated the best precision across the
range of concentrations. The IMx and IMMULITE methods showed comparable
precision, and the IMMAGE showed the poorest precision of the methods
evaluated, with a CV >20% at a CRP concentration of 2.26 mg/L. The BN
II, IMx, and IMMULITE methods were linear down to the lowest
concentration tested, which was a pool of the samples from the lowest
quartile of hs-CRP concentrations from blood donors (Fig. 1
). The IMMAGE method was linear down to a concentration
corresponding to the median value of the group of blood donors. The BN
II, IMMULITE, and IMMAGE methods gave comparable results of 9.79, 9.71,
and 10.13 mg/L, respectively, for the high pool, whereas the IMx method
gave a value of 13.73 mg/L. The BN II and IMx methods gave comparable
results of 0.30 and 0.32 mg/L, respectively, for the low pool, whereas
the IMMULITE method gave a value of 0.85 mg/L. Regression analysis of
the data yielded intercepts that were comparable to the measured values
for the low pool (Fig. 1
, legend). Reanalysis of the data from the BN
II, IMx, and IMMULITE methods by regression of only the four highest
concentrations, which account for 70% of the CRP range, yielded
intercepts of 1.01, 1.38, and 0.80 mg/L, respectively.
|
|
The hs-CRP concentrations of 322 serum samples collected from
apparently healthy adult blood donors were measured by the BN II, IMx,
IMMULITE, and IMMAGE methods simultaneously. Inspection of the data
revealed a highly skewed population. Therefore, values for the 25th,
50th, 75th, 90th, 95th, and 97.5th percentiles were determined for each
method (Table 2
). The BN II and IMx methods yielded comparable results for the
25th through the 95th percentiles. The IMMULITE gave slightly higher
values than the BN II method for the 25th and 50th percentiles, and
comparable values for the 75th through 95th percentiles. The IMMAGE
method gave substantially higher values than the other methods for the
50th and 75th percentiles, whereas the 25th percentile value for this
group of samples was below the detection limit of the assay. The BN II,
IMMULITE, and IMMAGE methods gave comparable results for the 97.5th
percentile, whereas the IMx value was higher.
|
When percentiles were determined by gender, values for the 25th and 50th percentiles were comparable, whereas values for the 75th, 90th, and 95th percentiles were nearly twice as high for women as for men. Because of the limited sample size, the 97.5th percentile was not determined by gender. When the relationship between age and hs-CRP was examined using the BN II method, the Pearson correlation coefficient (r) was 0.238 (P <0.001).
The agreement of the four methods with samples from blood donors was
assessed graphically (Fig. 2
). The BN II method previously had been compared with an
in-house ELISA method that was used in several hs-CRP epidemiologic
studies and has been validated clinically
(18)(20). Furthermore, it is the only method
approved by the FDA for cardiovascular and peripheral vascular risk
assessment. Therefore, it was chosen as the comparative method when
evaluating the other three methods. Deming regression analysis was
performed on all data before log transformation. A comparison of the BN
II and IMx methods gave a slope of 0.99 ± 0.01, an intercept of
0.09 ± 0.03, and a Sy|x of 0.55
(r = 0.991). A comparison of the BN II and IMMULITE
methods gave a slope of 0.93 ± 0.01, an intercept of 0.39 ±
0.01, and a Sy|x of 0.32 (r =
0.997). A comparison of the BN II and IMMAGE methods gave a slope of
0.97 ± 0.01, an intercept of 1.04 ± 0.04, and
aSy|x of 0.59 (r = 0.989).
Statistical analysis of the agreement between the BN II comparative
method and the other three methods for each quartile was performed
(21). The results (Table 3
) indicate that the mean differences and SDs of the differences
for the lowest three quartile were lowest between IMx and the
comparative method, intermediate between the IMMULITE and the
comparative method, and highest for the IMMAGE. For the highest
quartile, these values were lowest for the IMMULITE method, followed by
the IMx and then the IMMAGE method. All methods showed statistically
significant differences from the comparative method (P
<0.05) except for the IMMULITE method with the highest quartile.
|
|
The four methods were compared using samples with increased
concentrations of CRP (Fig. 3
). All methods showed excellent agreement at CRP concentrations
up to 50 mg/L. Above this concentration, there appeared to be
considerably more scatter in the data. Both the IMx and IMMULITE
methods showed differences with the BN II method >2 SD at
concentrations >150 mg/L that could not be explained by proportional
bias alone.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Several previous studies that examined serum CRP concentrations in apparently healthy populations, using highly sensitive ELISA, nephelometric, and turbidimetric methods, have found median values ranging from 0.58 to 1.13 mg/L (5)(18)(19)(22)(23)(24)(25)(26). The median values determined by the BN II, IMx, and IMMULITE methods were 0.8, 0.8, and 1.1 mg/L, respectively, consistent with these earlier studies. Four of these studies found 75th percentile values ranging from 1.44 to 2.10 mg/L, whereas we found values of 1.9, 2.0, and 2.2 mg/L for the BN II, IMx, and IMMULITE methods, respectively, again consistent with the earlier studies (5)(23)(25)(26). Two previous studies found a 90th percentile hs-CRP concentration of 3 mg/L, whereas we found values of 4.1–5.3 mg/L for the four methods we investigated (19)(23). The basis for this difference in the 90th percentiles observed could be attributable to either differences in the populations studied or the methodologies used. The median age of our population was 32 years, whereas the median ages of the two previous studies reporting values for the 90th percentile were not given. A weak positive correlation between age and hs-CRP concentration has been demonstrated (24), and we found a similar correlation. This indicates that younger ages are associated with lower hs-CRP values; however, it is unlikely that the previous studies had a median age substantially younger than 32 years. The good agreement between our data and those generated in previous studies for the median and 75th percentiles indicates similar assay calibration and suggests that population differences may explain the higher values we observed for the 90th percentile.
When our results were examined after segregation by gender, differences became noticeable at the 75th, 90th, and 95th percentiles. Previous studies that looked at the effect of gender failed to find a significant difference (23)(24). However, a recent study found that estrogen replacement therapy caused a large sustained increase in the hs-CRP concentrations in postmenopausal women (27)(28). The basis for the gender differences observed, particularly as it affects results primarily in higher hs-CRP percentiles in our population, is uncertain, but it may reflect hormonal differences.
All of the methods we investigated have been standardized against the WHO International Reference Standard for CRP Immunoassay 85/506 (29). Agreement between all methods for concentrations above the upper limit of the reference interval was acceptable. The BN II and IMx methods showed excellent agreement across the whole range of hs-CRP concentrations encountered in adult blood donors. The IMMULITE showed acceptable agreement with the BN II method for hs-CRP concentrations at or above the 75th percentile, whereas the IMMAGE method showed good agreement only with hs-CRP concentrations above the 90th percentile. Harmonization of hs-CRP results at concentrations below these values will require further work. For clinical risk stratification and epidemiologic studies, there needs to be agreement between methods across the reference interval down to hs-CRP concentrations of 0.2 mg/L. The BN II and IMx methods currently meet this goal, whereas the IMMAGE and IMMULITE methods do not.
The recoveries observed for the IMx method for serum pools prepared
from healthy donors were 30% higher than the other three methods
(Table 1
and Fig. 1
). However, the percentile values for the donor
population determined using the IMx were very similar to the other
methods with the exception that the IMx gave the highest values of the
methods tested for the 95th and 97.5th percentiles, which were 10–30%
higher than the other methods. The IMx method also showed a slope of
1.10 compared with the BN II method for samples from patients with
increased CRP concentrations. The intercept observed for the IMMULITE
method when measured recoveries were plotted vs dilution for donor
serum pools (Fig. 1
) was higher (0.4 mg/L) than those for the
BN II and IMx methods. This was also reflected in higher values for the
25th, 50th, and 75th percentiles. The modest proportional and constant
biases observed for the IMx and IMMULITE methods, respectively, likely
result from differences in calibration. A previous comparison between
the BN II and Hemagen ELISA methods showed a slope of 0.75 and an
intercept of -0.25 mg/L (18). Although these two hs-CRP
methods have been standardized with the same WHO CRP reference
material, they exhibit significant differences in the values obtained
(24). These authors concluded that although the exact source
of differences was uncertain, it was probably attributable to
inadequate standardization. We have confirmed this observation with a
different group of hs-CRP assays. Further effort is required for the
standardization of hs-CRP assays at concentrations comparable to those
seen in healthy subjects.
Another issue that merits discussion is the precision of the methods evaluated. The BN II and IMMAGE methods have automated dilutions, whereas the IMx and IMMULITE methods both require a manual dilution prepared off-line. The BN II method was the most precise, whereas the IMx and IMMULITE methods showed comparable precision with values in the reference interval. The IMMAGE method showed the worst precision at low concentrations of CRP, although acceptable precision with CVs <10% was observed at a CRP concentrations above the 75th percentile of a healthy population. Given the relatively large within-subject variability (CVI = 42%), analytical CVs of 10% or less should be adequate for both clinical and epidemiologic studies (24).
In conclusion, the four hs-CRP methods we evaluated exhibited some differences in results for a healthy population. Additional standardization efforts are required to ensure that results obtained by automated hs-CRP methods used in the clinical laboratory can be related to large-scale epidemiologic studies. Once standardization has been achieved, hs-CRP assays can provide useful data for atherosclerotic risk stratification in apparently healthy individuals.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  K. H. C. Wu, A. G. Tan, E. Rochtchina, E. J. Favaloro, A. Williams, P. Mitchell, and J. J. Wang Circulating Inflammatory Markers and Hemostatic Factors in Age-Related Maculopathy: A Population-Based Case-Control Study Invest. Ophthalmol. Vis. Sci., May 1, 2007; 48(5): 1983 - 1988. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Rifai, C. M. Ballantyne, M. Cushman, D. Levy, and G. L. Myers Point: High-Sensitivity C-Reactive Protein and Cardiac C-Reactive Protein Assays: Is There a Need to Differentiate? Clin. Chem., July 1, 2006; 52(7): 1254 - 1256. [Full Text] [PDF]
|
|
|
|
|
|
J. V. Callaghan and S. I. Gutman Counterpoint: food and drug administration guidance for C-reactive protein assays: matching claims with performance data. Clin. Chem., July 1, 2006; 52(7): 1256 - 1257. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Takemura, H. Matsumoto, A. Niimi, T. Ueda, H. Matsuoka, M. Yamaguchi, M. Jinnai, S. Muro, T. Hirai, Y. Ito, et al. High sensitivity C-reactive protein in asthma. Eur. Respir. J., May 1, 2006; 27(5): 908 - 912. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. E. Laaksonen, L. Niskanen, K. Nyyssonen, K. Punnonen, T.-P. Tuomainen, and J. T. Salonen C-reactive protein in the prediction of cardiovascular and overall mortality in middle-aged men: a population-based cohort study Eur. Heart J., September 1, 2005; 26(17): 1783 - 1789. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. L. Myers, N. Rifai, R. P. Tracy, W. L. Roberts, R. W. Alexander, L. M. Biasucci, J. D. Catravas, T. G. Cole, G. R. Cooper, B. V. Khan, et al. CDC/AHA Workshop on Markers of Inflammation and Cardiovascular Disease: Application to Clinical and Public Health Practice: Report From the Laboratory Science Discussion Group Circulation, December 21, 2004; 110(25): e545 - e549. [Full Text] [PDF]
|
|
|
|
|
|
W. L. Roberts CDC/AHA Workshop on Markers of Inflammation and Cardiovascular Disease: Application to Clinical and Public Health Practice: Laboratory Tests Available to Assess Inflammation--Performance and Standardization: A Background Paper Circulation, December 21, 2004; 110(25): e572 - e576. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. S. Levinson, J. J. Miller, and R. J. Elin Poor Predictive Value of High-Sensitivity C-Reactive Protein Indicates Need for Reassessment Clin. Chem., October 1, 2004; 50(10): 1733 - 1735. [Full Text] [PDF]
|
|
|
|
|
|
T. W. McDade, J. Burhop, and J. Dohnal High-Sensitivity Enzyme Immunoassay for C-Reactive Protein in Dried Blood Spots Clin. Chem., March 1, 2004; 50(3): 652 - 654. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Artieda, A. Cenarro, A. Ganan, I. Jerico, C. Gonzalvo, J. M. Casado, I. Vitoria, J. Puzo, M. Pocovi, and F. Civeira Serum Chitotriosidase Activity Is Increased in Subjects With Atherosclerosis Disease Arterioscler. Thromb. Vasc. Biol., September 1, 2003; 23(9): 1645 - 1652. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. B. Ledue and N. Rifai Preanalytic and Analytic Sources of Variations in C-reactive Protein Measurement: Implications for Cardiovascular Disease Risk Assessment Clin. Chem., August 1, 2003; 49(8): 1258 - 1271. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. S. Ford, W. H. Giles, G. L. Myers, N. Rifai, P. M. Ridker, and D. M. Mannino C-reactive Protein Concentration Distribution among US Children and Young Adults: Findings from the National Health and Nutrition Examination Survey, 1999-2000 Clin. Chem., August 1, 2003; 49(8): 1353 - 1357. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. Kimberly, H. W. Vesper, S. P. Caudill, G. R. Cooper, N. Rifai, F. Dati, and G. L. Myers Standardization of Immunoassays for Measurement of High-Sensitivity C-reactive Protein. Phase I: Evaluation of Secondary Reference Materials Clin. Chem., April 1, 2003; 49(4): 611 - 616. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. C. Vukovich, S. Mustafa, H. Rumpold, and O. Wagner Evaluation of a Turbidimetric Denka Seiken C-Reactive Protein Assay for Cardiovascular Risk Estimation and Conventional Inflammation Diagnosis Clin. Chem., March 1, 2003; 49(3): 511 - 512. [Full Text] [PDF]
|
|
|
|
|
|
T. McLaughlin, F. Abbasi, C. Lamendola, L. Liang, G. Reaven, P. Schaaf, and P. Reaven Differentiation Between Obesity and Insulin Resistance in the Association With C-Reactive Protein Circulation, December 3, 2002; 106(23): 2908 - 2912. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Middleton Effect of Analytical Error on the Assessment of Cardiac Risk by the High-Sensitivity C-Reactive Protein and Lipid Screening Model Clin. Chem., November 1, 2002; 48(11): 1955 - 1962. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Ishibashi, Y. Takemura, H. Ishida, K. Watanabe, and T. Kawai C-Reactive Protein Kinetics in Newborns: Application of a High-Sensitivity Analytic Method in Its Determination Clin. Chem., July 1, 2002; 48(7): 1103 - 1106. [Full Text] [PDF]
|
|
|
|
|
|
M. K.C. Ng, P. Y. Liu, A. J. Williams, S. Nakhla, L. P. Ly, D. J. Handelsman, and D. S. Celermajer Prospective Study of Effect of Androgens on Serum Inflammatory Markers in Men Arterioscler. Thromb. Vasc. Biol., July 1, 2002; 22(7): 1136 - 1141. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Liu, J. E Manson, J. E Buring, M. J Stampfer, W. C Willett, and P. M Ridker Relation between a diet with a high glycemic load and plasma concentrations of high-sensitivity C-reactive protein in middle-aged women Am. J. Clinical Nutrition, March 1, 2002; 75(3): 492 - 498. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. S. Levinson and R. J. Elin What Is C-Reactive Protein Telling Us About Coronary Artery Disease? Arch Intern Med, February 25, 2002; 162(4): 389 - 392. [Full Text] [PDF]
|
|
|
|
|
|
S. Rothkrantz-Kos, M. P.J. Schmitz, O. Bekers, P. P.C.A. Menheere, and M. P. van Dieijen-Visser High-Sensitivity C-Reactive Protein Methods Examined Clin. Chem., February 1, 2002; 48(2): 359 - 362. [Full Text] [PDF]
|
|
|
|
|
|
A. Hamwi, T. Vukovich, O. Wagner, H. Rumpold, R. Spies, M. Stich, and C. Langecker Evaluation of Turbidimetric High-Sensitivity C-Reactive Protein Assays for Cardiovascular Risk Estimation Clin. Chem., November 1, 2001; 47(11): 2044 - 2046. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Feldman, I. Jialal, S. Devaraj, and B. Cryer Effects of low-dose aspirin on serum C-reactive protein and thromboxane B2 concentrations: a placebo-controlled study using a highly sensitive C-reactive protein assay J. Am. Coll. Cardiol., June 15, 2001; 37(8): 2036 - 2041. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. M. Ridker High-Sensitivity C-Reactive Protein : Potential Adjunct for Global Risk Assessment in the Primary Prevention of Cardiovascular Disease Circulation, April 3, 2001; 103(13): 1813 - 1818. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Rifai and P. M. Ridker High-Sensitivity C-Reactive Protein: A Novel and Promising Marker of Coronary Heart Disease Clin. Chem., March 1, 2001; 47(3): 403 - 411. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. L. Roberts, L. Moulton, T. C. Law, G. Farrow, M. Cooper-Anderson, J. Savory, and N. Rifai Evaluation of Nine Automated High-Sensitivity C-Reactive Protein Methods: Implications for Clinical and Epidemiological Applications. Part 2 Clin. Chem., March 1, 2001; 47(3): 418 - 425. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Borque, L. Bellod, A. Rus, M. L. Seco, and F. Galisteo-Gonzalez Development and Validation of an Automated and Ultrasensitive Immunoturbidimetric Assay for C-Reactive Protein Clin. Chem., November 1, 2000; 46(11): 1839 - 1842. [Full Text] [PDF]
|
|
|
|
|
|
N. Gochman High and Higher Sensitivity CRP Clin. Chem., July 1, 2000; 46(7): 1022 - 1022. [Full Text] [PDF]
|
|
Read all eLetters
) and were excluded from the Bland-Altman plots in
B and D. (B,
D, and F), Bland–Altman analyses. The
solid line indicates the mean difference between
methods, and the 95% confidence intervals for the difference are
indicated by dashed lines. (A),
slope = 1.10 (95% confidence interval, 1.06–1.14);
intercept = -4.1 (95% confidence interval, -7.6 to -0.5);
Sy|x = 8.4; r = 0.991.
(B), mean bias and SD, 2.3 and 9.7 mg/L, respectively.
(C), slope = 1.06 (95% confidence interval,
1.02–1.10); intercept = -4.4 (95% confidence interval, -7.7 to
-1.1); Sy|x = 7.7; r = 0.991.
(D), mean bias and SD, -0.6 and 8.0 mg/L, respectively.
(E), slope = 0.92 (95% confidence interval,
0.89–0.95); intercept = -0.9 (95% confidence interval, -3.6 to
1.8); Sy|x = 6.4; r = 0.992.
(F), mean bias and SD, 6.0 and 7.9 mg/L, respectively.