Performance of direct estradiol immunoassays with human male serum samples.
Thus, measurement of serum [E.sub.2] is as important for understanding numerous developmental and pathophysiological processes in men as it is in women. Yet although modern commercial [E.sub.2] immunoassays, which have been optimized for unextracted human serum ("direct"), may be adequate for measuring the high blood [E.sub.2] concentrations in women between menarche and menopause, and especially during ovarian stimulation, circulating [E.sub.2] concentrations are 10- to 100-fold lower in children, men, and postmenopausal or aromatase inhibitor--treated women and are challenging to measure. Awareness of problems with validity (analytical specificity, accuracy) of direct immunoassays, which was identified >25 years ago (4, 5), grew over the last decade following the recognition of limitations of T immunoassays, particularly at low circulating concentrations (6-8). Direct T immunoassays display assay-specific bias and analytical nonspecificity, most notably at low blood T concentrations in children, women, and androgen--deficient men (6-10). Therefore, aiming to measure the expected (picomolar) serum [E.sub.2] concentrations that render T immunoassays unreliable makes it unsurprising that direct [E.sub.2] immunoassays lack accuracy and validity for measurements of serum [E.sub.2] in not only men but also children (11) as well as postmenopausal (4, 12-14) or aromatase inhibitor-treated (15) women who have comparably low blood [E.sub.2] concentrations (16). Comparisons of direct [E.sub.2] immunoassays with mass spectrometry--based reference methods in men have previously been confined to single commercial [E.sub.2] immunoassays (17-19), whereas other studies focused on in-house [E.sub.2] immunoassays (15,20) or studied only women (4,12,13). Here we used a reference panel of sera to calibrate and compare the performance of widely used commercial direct [E.sub.2] immunoassays relative to the reference LC-MS method using a reference panel of healthy older men (8).
Materials and Methods
A reference panel of sera was formed from a nested cohort of participants in the Healthy Man Study (21) who were >40 years old and reported excellent or very good health with no health symptoms including sexual dysfunction or gynecomastia. Previously unthawed aliquots of serum (n = 101) collected during the study were distributed frozen to 5 laboratories, all holding accreditation to International Organization for Standardization (ISO) 15189 and participating in the national RCPAQAP (Royal College of Pathologists of Australasia Quality Assurance Programs) quality assurance program. The laboratories, selected to represent different commercial direct [E.sub.2] immunoassays, each ran the samples in duplicate in each assay.
The 5 commercial direct [E.sub.2] immunoassays were (A) Siemens ADVIA Enhanced Estradiol assay run on Centaur analyzer; (B) Siemens IMMULITE 2000 Estradiol assay run on IMMULITE 2000 analyzer; (C) Abbott ARCHITECT System Estradiol assay run on ARCHITECT analyzer; (D) Roche cobas Estradiol II assay run on Roche E170 analyzer; and (E) Beckman Coulter Access Estradiol assay run on DxI800 analyzer (these 5 immunoassays are listed in this article as A-E, respectively). All 5 methods provide reference intervals for [E.sub.2] in males, indicating that measurement of male sera is included in the intended use of the assays, and 4 of the 5 methods claim traceability to GC-MS. The samples were also measured by an LC-MS [E.sub.2] assay (22). To validate the serum [E.sub.2] measurements in the LC-MS assay, 3 certified reference materials (CRMs) for serum [E.sub.2] (CRM 576, CRM 577, and CRM 578) obtained from the European Commission's Institute for Reference Materials and Measurements (23) were run in duplicate on consecutive days in the LC-MS assay, including 1 set run together with the samples in this study. Serum T was measured in the same sample by the same LC-MS method (22) and serum luteinizing hormone (LH), follicle-stimulating hormone (FSH), and sex hormone--binding globulin (SHBG) by immunoassays as reported previously (21).
For each immunoassay, the distribution of values was determined by nonparametric methods, distributional deviation from gaussian by Shapiro--Wilk statistic, and bias (relative to MS) by Passing--Bablok regression and illustrated by deviance (modified Bland--Altman) and mountain plots (24) by use of NCSS, MedCalc, and Sigmaplot software. Samples with undetectable serum [E.sub.2] values were deleted from analysis of that immunoassay. We calculated body mass index (BMI) as weight (in kg) divided by the square of height (in meters) and body surface area (BSA) in [m.sup.2] by the Gehan--George formula (25). We calculated CV for an assay as the mean of the CVs for the 2 replicates for each sample without adjustment for serum [E.sub.2] concentration. Correlation was performed by linear least-squares analysis with a Bonferroni adjustment for multiple comparisons. Results are presented as mean and SE if gaussian, or as median and interquartile range (IQR) (25th, 75th centiles) otherwise.
The 101 participants had a median (IQR) age of 54 (48-62) years and BMI 25.9 (23.9-28.6) kg/[m.sup.2]. The group had a mean (SE) height 176 (0.7) cm, weight 81.2 (1.1) kg, and BSA 2.00 (0.01) [m.sup.2].
In terms of the assigned target serum [E.sup.2] concentrations, the CRM samples had a median accuracy of 100% [101% (6%)] at low concentrations (nominal certified concentration 31 pg/mL), 93% [93% (3%)] at medium concentrations (188 pg/mL), and 89% [89% (10%)] at high concentrations (365 pg/mL) without any significant between-day difference (P > 0.8).
The overall analysis of serum [E.sub.2] measurements (repeated-measures ANOVA) showed highly significant difference between assays ([F.sub.5995] = 8.84, P < [10.sup.-6]) and between men ([F.sub.95995] = 9.10, P < [10.sup.-6]) but not for intraassay replication (F1995 = 0.0, P = 1.0). Three immunoassays detected serum [E.sub.2] in all samples, whereas [E.sub.2] was detected in only 53% of samples in assay B and 72% in assay E. Table 1 lists the nonparametric centiles of distribution and pooled within-assay CVs for each immunoassay, and the immunoassays are compared in Table 2. Fig. 1 shows the overall distribution of serum [E.sub.2] concentrations displayed as notched box plots. For each immunoassay, the serum [E.sub.2] measurements had a nongaussian distribution in the natural scale (Table 1), and these were normalized by a logarithmic transformation for each assay except assay B, which had numerous undetectable samples (data not shown).
The bias parameters for each immunoassay relative to LC-MS are displayed as Passing--Bablok regressions (Fig. 2) and regression estimates (Table 1) as well as mountain plots (see Supplemental Fig. 1, which accompanies the online version of this article at http://www.clinchem.org/content/vol60/issue3) and deviance plots (Fig. 3). The bias of each method was upward (positive) ranging from 12% to 53% (at mean) and 11% to 74% (at 25th centile), 6% to 55% (at median), and 11% to 48% (at 75th centile). Residual SD in the Passing--Bablok regressions were similar for 4 immunoassays and LC-MS but were significantly higher for assay B, which reflected its lower analytical sensitivity, lack of linearity, and greater deviation from regression slope of 1.0. The CVs were lower for 4 assays (A, B, C, and D) relative to LC-MS but higher in 1 (E).
To evaluate biological correlates of each assay, serum [E.sub.2] concentrations from the 5 [E.sub.2] immunoassays and LC-MS [E.sub.2] assay were correlated by linear regression with weight, BMI, and reproductive hormone variables (serum LH, FSH, T, SHBG) (Table 3). Serum [E.sub.2] correlated with serum LH and SHBG in the LC-MS assay but not with either gonadotropin in any immunoassays. Assay B results correlated negatively with weight and BMI, but this was not observed with any other assay. None of the assays generated serum [E.sub.2] results that correlated with age, height, or bone density at hip or spine (in either absolute bone density or T scores; data not shown).
In this study, we used a defined reference panel (8) comprising 101 serum samples from healthy older men, a nested cohort of the Healthy Man Study (21), to calibrate the performance of 5 commercial direct [E.sub.2] immunoassays as has been done previously for serum T immunoassays (8). On the basis of measurements of this reference panel, each immunoassay deviated significantly from the reference LC-MS method throughout the working range, with the magnitude of positive (upward) bias varying between assays from 6% to 74%. As the variability seen here is based on single versions of each assay, a significantly greater variability might be expected when interlaboratory variability is included (26). It is possible that the immunoassays are able to separate concentrations within reference intervals from clearly increased serum [E.sub.2] in men; however, method-dependent reference intervals would be required. This study did not address that question.
Our findings extend previous reports comparing direct [E.sub.2] immunoassays with MS-based methods in men but where only a single commercial immunoassay (17-19) or in-house immunoassays (15, 20) were examined. Our results are also consistent with the findings in postmenopausal or aromatase inhibitor--treated women with comparably low circulating [E.sub.2] concentrations (4,12,13,27). The strong, positive correlation of the LC-MS assay with serum T and SHBG was not matched by any immunoassay. The failure of serum [E.sub.2] concentration to correlate with BMI, proposed as a biological validity criterion for serum [E.sub.2] measurements in women (28), may reflect the biological differences in the range of circulating [E.sub.2] concentrations between sexes. The striking negative correlation of weight or BMI with serum [E.sub.2] concentration by assay B, but no other assay, may reflect the unknown cross-reacting steroids, responsible for the overestimation of serum [E.sub.2] measurements and poor correlation with LCMS, which may correlate inversely with weight or obesity. Because this assay produced fewer detectable values covering a narrower range of results, the weaker correlation with LC-MS is an expected finding.
The reasons for these discrepancies are related to the origins of steroid immunoassays, which were developed in 1969 (29) but delayed for a decade after the invention of peptide immunoassay because additional steps were required to adapt immunoassay methodology to valid measurement of nonimmunogenic small molecules such as steroids. Developing steroid-specific antibodies required conjugating steroids as haptens to larger immunogenic carrier proteins via small multivalent reactive bridge compounds. However, this made steroid antibodies that were epitope "blind" to the conjugation site, allowing for undesirable cross-reactivity with structurally related steroids (e.g., structurally related steroid precursors, metabolites, and conjugates produced by phase II metabolism). The original in-house steroid immunoassays developed in the 1970s used solvent extraction, chromatography, and structurally authentic ([H.sup.3], [C.sup.14]) tracers, a triplet of validity criteria for steroid immunoassay that removed structurally related steroids as well as nonspecific matrix interference in the immunoassay reactions. In the ensuing decades, [E.sub.2] immunoassays were commercialized primarily to monitor ovarian responses to gonadotropin stimulation, where excessive serum [E.sub.2] response was a risk indicator for dangerous overdosage effects such as high-order multiple pregnancies and/or life-threatening ovarian hyperstimulation syndrome. However, making a quantal distinction between dangerously high [E.sub.2] concentrations (>2000 pmol/L) and typical premenopausal concentrations (200-800 pmol/L) did not require quantitative accuracy at physiological [E.sub.2] concentrations. Subsequently, ultrasound monitoring of follicular growth has reduced the importance of this application of serum [E.sub.2] immunoassays. The subsequent growing demand for steroid immunoassays in clinical practice and research resulted in assay simplification to adapt steroids into semi-automated multiplex platforms and 1-plate/tube kit formats, preferred by routine pathology and research labs, respectively. This simplification eliminated preassay purification steps (extraction, chromatography) as well as [beta]-scintillation counting, with the latter forcing the replacement ofstructurally authentic steroid tracers with bulky conjugated steroids, which allowed for more convenient nonradioactive assay readouts. However, eliminating the triplet of validity criteria (preassay extraction and chromatography, authentic tracers) sacrificed the specificity of [E.sub.2] immunoassays and revealed the vulnerabilities of direct steroid immunoassay to artifacts from steroid cross-reactivity and matrix interference. Although the original validated steroid immunoassays are now confined to a few long-established laboratories, there is evidence that some but not all the limitations of direct [E.sub.2] immunoassays can be overcome by preassay solvent extraction (12, 28). However, since our study did not include any classical (indirect) estradiol immunoassays, it is not clear whether meeting the original triplet validity criteria is sufficient to overcome all limitations of direct [E.sub.2] immunoassays, and it remains possible that suboptimal antibody specificity could also contribute to the nonspecificity of steroid immunoassays.
The analytical sensitivity of 3 direct [E.sub.2] immunoassays was adequate to detect serum [E.sub.2] in all serum samples from healthy older men. This reflects generational improvement over less analytically sensitive previous [E.sub.2] immunoassays (13). Replicate imprecision was better with most platform [E.sub.2] immunoassays owing to the use of automated rather than manual pipetting; however, the most accurate direct [E.sub.2] immunoassay had replicate imprecision inferior to that of LC-MS, a long-recognized limitation of [E.sub.2] immunoassays (26). This presumably reflects the worse imprecision of immunoassays at the low end of their working range. On the other hand, the CV for the LC-MS assay in this study based on 2 replicates was higher than when a larger number of QC samples was pooled (22), although this does not exclude effects resulting from differences in internal standard: analyte ratios (30) or other deviations from ideal internal standard performance (31).
Although assay-dependent bias and analytical nonspecificity of direct [E.sub.2] immunoassays is established (11, 12, 14, 15, 27, 32-35), direct [E.sub.2] assays are often used in research involving tissue extracts and nonhuman serum samples, where their validity is doubtful (36, 37).
Overall, these findings indicate that direct [E.sub.2] immunoassays are suboptimal for use with human male serum where either absolute measurement is required, such as in clinical practice or research that hinges on interpreting [E.sub.2] measurements in single serum samples, or even in more self-contained research studies where small differences in measurements may be decisive, such as analytical epidemiology (16, 38, 39). These limitations underestimate the problems with direct [E.sub.2] immunoassays over time, since commercial kit or platform formats are subject to changes in proprietary components, notably (but not only) new antibodies, that have to be replaced. Hence direct [E.sub.2] immunoassays exhibit not only assay-dependent bias but are subject to unpredictable variability over time, all of which detract from assay performance and stability. By contrast, the advent of lower-cost, benchtop MS equipment that retains reference concentration specificity but now features analytical sensitivity matching the best steroid immunoassays renders MS-based [E.sub.2] assays more widely accessible. Although MS-based [E.sub.2] assays provide a durable chemical analysis free from method-dependent bias of immunoassays, like all biochemical measurements, MS-based assays maybe subject to different features such as those involving matrix effects, chromatography, and differences in monitored transitions, all of which require rigorous standardization and ongoing quality control (16, 28, 40), an essential but challenging process that remains to be completed.
Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.
Authors' Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest:
Employment or Leadership: None declared.
Consultant or Advisory Role: R.I. McLachlan, Eli Lilly Advisory Board.
Stock Ownership: None declared.
Honoraria: None declared.
Research Funding: R.I. McLachlan, Australian NHMRC research support.
Expert Testimony: None declared.
Patents: None declared.
Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript.
Acknowledgments: The authors are grateful to Dr. James Doery, Southern Cross Pathology Australia, Victoria; Dr. Chris Farrell, Laverty Pathology, NSW; Dr. Greg Ward, Sullivan and Nicholaides Pathology, Queensland; and Dr. Paul Willams, Sydney South West Pathology, NSW, as well as to Dr. Ken Sikaris, Melbourne Pathology, for encouraging the study.
(1.) Naftolin F. Brain aromatization of androgens. J Reprod Med 1994;39:257-61.
(2.) Vanderschueren D, Vandenput L, Boonen S, Lindberg MK, Bouillon R, Ohlsson C. Androgens and bone. Endocr Rev 2004;25:389-425.
(3.) Santen RJ. Gynecomastia. In: Jameson JL, De Groot LJ, eds. Endocrinology: adult and pediatric, Vol. 2. 6th Ed. St. Louis, MO: Saunders Elsevier, 2010:2544-55.
(4.) Schioler V, Thode J. Six direct radioimmunoassays of estradiol evaluated. Clin Chem 1988;34:949-52.
(5.) Thomas CM, van den Berg RJ, Segers MF. Measurement of serum estradiol: comparison of three "direct" radioimmunoassays and effects of organic solvent extraction. Clin Chem 1987;33: 1946-7.
(6.) Taieb J, Mathian B, Millot F, Patricot MC, Mathieu E, Queyrel N, et al. Testosterone measured by 10 immunoassays and by isotope-dilution gas chromatography-mass spectrometry in sera from 116 men, women, and children. Clin Chem 2003; 49:1381-95.
(7.) Wang C, Catlin DH, Demers LM, Starcevic B, Swerdloff RS. Measurement of total serum testosterone in adult men: comparison of current laboratory methods versus liquid chromatography-tandem mass spectrometry. J Clin Endocrinol Metab 2004;89:534-43.
(8.) Sikaris K, McLachlan RI, Kazlauskas R, de Kretser D, Holden CA, Handelsman DJ. Reproductive hormone reference intervals for healthy fertile young men: evaluation of automated platform assays. J Clin Endocrinol Metab 2005;90:5928-36.
(9.) Herold DA, Fitzgerald RL. Immunoassays for testosterone in women: better than a guess? Clin Chem 2003;49:1250-1.
(10.) Rosner W, Auchus RJ, Azziz R, Sluss PM, Raff H. Utility, limitations, and pitfalls in measuring testosterone: an Endocrine Society position statement. J Clin Endocrinol Metab 2007;92:405-13.
(11.) Ankarberg-Lindgren C, Norjavaara E. A purification step prior to commercial sensitive immunoassay is necessary to achieve clinical usefulness when quantifying serum 17beta-estradiol in prepubertal children. Eur J Endocrinol 2008;158:117-24.
(12.) Lee JS, Ettinger B, Stanczyk FZ, Vittinghoff E, Hanes V, Cauley JA, et al. Comparison of methods to measure low serum estradiol levels in postmenopausal women. J Clin Endocrinol Metab 2006;91:3791-7.
(13.) Yang DT, Owen WE, Ramsay CS, Xie H, Roberts WL. Performance characteristics of eight estradiol immunoassays. Am J Clin Pathol 2004;122:332-7.
(14.) Rothman MS, Carlson NE, Xu M, Wang C, Swerdloff R, Lee P, et al. Reexamination of testosterone, dihydrotestosterone, estradiol and estrone levels across the menstrual cycle and in postmenopausal women measured by liquid chromatographytandem mass spectrometry. Steroids 2011;76:177-82.
(15.) Santen RJ, Demers L, Ohorodnik S, Settlage J, Langecker P, Blanchett D, et al. Superiority of gas chromatography/tandem mass spectrometry assay (GC/MS/MS) for estradiol for monitoring of aromatase inhibitor therapy. Steroids 2007;72: 666-71.
(16.) Rosner W, Hankinson SE, Sluss PM, Vesper HW, Wierman ME. Challenges to the measurement of estradiol: an Endocrine Society position statement. J Clin Endocrinol Metab 2013;98:1376-8.
(17.) Huhtaniemi IT, Tajar A, Lee DM, O'Neill TW, Finn JD, Bartfai G, et al. Comparison of serum testosterone and estradiol measurements in 3174 European men using platform immunoassay and mass spectrometry; relevance for the diagnostics in aging men. Eur J Endocrinol 2012;166:983-91.
(18.) Khosla S, Amin S, Singh RJ, Atkinson EJ, Melton LJ 3rd, Riggs BL. Comparison of sex steroid measurements in men by immunoassay versus mass spectroscopy and relationships with cortical and trabecular volumetric bone mineral density. Osteoporos Int 2008;19:1465-71.
(19.) Sluss PM, Hayes FJ, Adams JM, Barnes W, Williams G, Frost S, et al. Mass spectrometric and physiological validation of a sensitive, automated, direct immunoassay for serum estradiol using the Architect. Clin Chim Acta 2008;388:99-105.
(20.) Hsing AW, Stanczyk FZ, Belanger A, Schroeder P, Chang L, Falk RT, Fears TR. Reproducibility of serum sex steroid assays in men by RIA and mass spectrometry. Cancer Epidemiol Biomarkers Prev 2007;16:1004-8.
(21.) Sartorius G, Spasevska S, Idan A, Turner L, Forbes E, Zamojska A, et al. Serum testosterone, dihydrotestosterone and estradiol concentrations in older men self-reporting very good health: the Healthy Man Study. Clin Endocrinol (Oxf) 2012; 77:755-63.
(22.) Harwood DT, Handelsman DJ. Development and validation of a sensitive liquid chromatographytandem mass spectrometry assay to simultaneously measure androgens and estrogens in serum without derivatization. Clin Chim Acta 2009;409:78-84.
(23.) Tai SS, Welch MJ. Development and evaluation of a reference measurement procedure for the determination of estradiol-17beta in human serum using isotope-dilution liquid chromatographytandem mass spectrometry. Anal Chem 2005;77:6359-63.
(24.) Krouwer JS, Monti KL. A simple, graphical method to evaluate laboratory assays. Eur J Clin Chem Clin Biochem 1995;33:525-7.
(25.) Bailey BJ, Briars GL. Estimating the surface area of the human body. Stat Med 1996;15:1325-32.
(26.) Middle JG, Kane JW. Oestradiol assays: fitness for purpose? Ann Clin Biochem 2009;46:441-56.
(27.) Dowsett M, Folkerd E. Deficits in plasma oestradiol measurement in studies and management of breast cancer. Breast Cancer Res 2005;7:1-4.
(28.) Stanczyk FZ, Lee JS, Santen RJ. Standardization of steroid hormone assays: why, how, and when? Cancer Epidemiol Biomarkers Prev 2007;16:1713-9.
(29.) Abraham GE. Solid-phase radioimmunoassay of estradiol-17 beta. J Clin Endocrinol Metab 1969; 29:866-70.
(30.) Botelho JC, Shacklady C, Cooper HC, Tai SS, Van Uytfanghe K, Thienpont LM, Vesper HW. Isotopedilution liquid chromatography-tandem mass spectrometry candidate reference method for total testosterone in human serum. Clin Chem 2013;59:372-80.
(31.) Stokvis E, Rosing H, Beijnen JH. Stable isotopically labeled internal standards in quantitative bioanalysis using liquid chromatography/mass spectrometry: necessity or not? Rapid Commun Mass Spectrom 2005;19:401-7.
(32.) Stanczyk FZ, Cho MM, Endres DB, Morrison JL, Patel S, Paulson RJ. Limitations of direct estradiol and testosterone immunoassay kits. Steroids 2003;68:1173-8.
(33.) Dorgan JF, Fears TR, McMahon RP, Aronson Friedman L, Patterson BH, Greenhut SF. Measurement of steroid sex hormones in serum: a comparison of radioimmunoassay and mass spectrometry. Steroids 2002;67:151-8.
(34.) Leung YS, Dees K, Cyr R, Schloegel I, Kao PC. Falsely increased serum estradiol results reported in direct estradiol assays. Clin Chem 1997;43: 1250-1.
(35.) Taieb J, Benattar C, Birr AS, Lindenbaum A. Limitations of steroid determination by direct immunoassay. Clin Chem 2002;48:583-5.
(36.) McNamara KM, Harwood DT, Simanainen U, Walters KA, Jimenez M, Handelsman DJ. Measurement of sex steroids in murine blood and reproductive tissues by liquid chromatographytandem mass spectrometry. J Steroid Biochem Mol Biol 2010;121:611-8.
(37.) Haisenleder DJ, Schoenfelder AH, Marcinko ES, Geddis LM, Marshall JC. Estimation of estradiol in mouse serum samples: evaluation of commercial estradiol immunoassays. Endocrinology 2011; 152:4443-7.
(38.) Santen RJ, Lee JS, Wang S, Demers LM, Mauras N, Wang H, Singh R. Potential role of ultra-sensitive estradiol assays in estimating the risk of breast cancer and fractures. Steroids 2008;73: 1318-21.
(39.) Key TJ. Endogenous oestrogens and breast cancer risk in premenopausal and postmenopausal women. Steroids 2011;76:812-5.
(40.) Vesper HW, Botelho JC. Standardization of testosterone measurements in humans. J Steroid Biochem Mol Biol 2010;121:513-9.
David J. Handelsman,  * Julie D. Newman,  Mark Jimenez,  Robert McLachlan,  Gideon Sartorius,  and Graham R.D. Jones 
 ANZAC Research Institute, Concord Hospital, University of Sydney, Sydney, NSW, Australia;  Southern Cross Pathology Australia, Clayton, VIC, Australia;  Prince Henry's Institute of Medical Research, Monash Medical Centre, Clayton, VIC, Australia;  SydPath, Department of Chemical Pathology, St Vincent's Hospital, Darlinghurst, NSW, Australia.
 Nonstandard abbreviations: [E.sub.2], estradiol; T, testosterone; ISO, International Organization for Standardization; RCPAQAP, Royal College of Pathologists of Australasia Quality Assurance Programs; CRM, certified reference material; LH, luteinizing hormone; FSH, follicle-stimulating hormone; SHBG, sex hormone--binding globulin; BMI, body mass index; BSA, body surface area; IQR, interquartile range.
* Address correspondence to this author at: ANZAC Research Institute, Concord Hospital, NSW 2139, Australia. E-mail email@example.com.
Received July 25, 2013; accepted December 3, 2013.
Previously published online at DOI: 10.1373/clinchem.2013.213363
Table 1. Serum [E.sub.2] distribution centiles and Passing-Bablok analysis. (a) LC-MS A (b) B (c) Maximum 208 293 363 97.5th centile 147 195 287 75th centile 98 133 113 50th centile 82 118 87 25th centile 62 103 79 2.5th centile 36 76 73 Minimum 34 74 73 Detectable, % 100 100 53 Gaussiang 0.95 (h) 0.86 (h) 0.56 (h) Mean (SD) 83 (28) 122 (30) 102 (43) CV, %i 13.6 7.5 5.4 Correlation 1.0 0.71 0.55 Passing-Bablok Slope Reference 0.92 0.82 Intercept Reference 45 (k) 17 Residual SD Reference 16 28 Linearity Reference [check] x C (d) D (e) E (f) Maximum 285 344 323 97.5th centile 185 173 191 75th centile 145 120 115 50th centile 127 98 100 25th centile 108 81 83 2.5th centile 72 51 75 Minimum 68 43 75 Detectable, % 100 100 72 Gaussiang 0.92 (h) 0.81 (h) 0.67 (h) Mean (SD) 127 (30) 104 (37) 106 (35) CV, %i 10.8 10.9 23.7 Correlation 0.65 0.74 0.64 Passing-Bablok Slope 1.02 1.07 0.82 Intercept 45 (k) 15 (k) 28 Residual SD 18 17 20 Linearity [check] [check] [check] (a) Serum [E.sub.2] concentrations are in pmol/L. To convert to pg/mL, divide by 3.67. (b) A, Siemens ADVIA. (c) B, Siemens IMMULITE 2000. (d) C, Abbott ARCHITECT. (e) D, Roche cobas. (f) E, Beckman Coulter Access. (g) Shapiro-Wilks W. (h) Significant (P < 0.05) deviation from gaussian distribution. (i) Mean coefficient of variation of duplicate measurements. (j) Correlation with LC-MS. (k) Significant (P < 0.05) difference from slope = 1.0 or intercept = 0.0. Table 2. Passing-Bablok regression and Pearson correlation of serum [E.sub.2] concentrations measured by 5 immunoassays pair-wise. (a) A (b) B (c) C (d) A (b) 1 0.660 0.699 B (c) 0.805, -5.44 1 0.646 C (d) 1.091, -5.75 1.230,13.5 1 D (e) 1.127, -36.0 1.154, -2.01 1.116, -43.6 E (f) 1.145, -43.8 1.442, -32.3 1.045, -41.7 D (e) E (f) A (b) 0.802 0.692 B (c) 0.807 0.780 C (d) 0.728 0.671 D (e) 1 0.801 E (f) 0.927, -4.21 1 (a) Values above the diagonal indicate the Pearson correlation, and below the diagonal the slope and intercept of the Passing--Bablok regression. (b) A, Siemens ADVIA. (c) B, Siemens IMMULITE 2000. (d) C, Abbott ARCHITECT. (e) D, Roche cobas. (f) E, Beckman Coulter Access. Table 3. Correlation between serum [E.sub.2] concentrations measured by 5 immunoassays and LC-MS assay and weight, BMI, and reproductive hormones. MS A (a) B (b) C (c) D (d) E (e) Weight -0.14 -0.17 -0.38 (f) -0.20 -0.21 -0.21 BMI -0.18 -0.09 -0.41 (f) -0.27 -0.18 -0.24 LH 0.08 0.08 0.20 0.13 0.19 -0.07 FSH -0.06 -0.03 0.03 0.02 0.03 -0.15 T 0.45 (f) 0.26 0.07 0.12 0.25 0.06 SHBG 0.34 (f) 0.17 0.24 0.05 0.26 0.03 (a) A, Siemens ADVIA. (b) B, Siemens IMMULITE 2000. (c) C, Abbott ARCHITECT. (d) D, Roche cobas. (e) E, Beckman Coulter Access. (f) P < 0.001 for a nominal P < 0.05 with Bonferroni adjustment for 36 pair-wise comparisons.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Endocrinology and Metabolism|
|Author:||Handelsman, David J.; Newman, Julie D.; Jimenez, Mark; McLachlan, Robert; Sartorius, Gideon; Jones,|
|Date:||Mar 1, 2014|
|Previous Article:||Rapid electrokinetic isolation of cancer-related circulating cell-free DNA directly from blood.|
|Next Article:||Pediatric within-day biological variation and quality specifications for 38 biochemical markers in the CALIPER cohort.|