CoChat
Sign Up
Cited by 95Clinical Laboratory Practices and Quality ControlAdrenal Hormones and DisordersScientific Measurement and Uncertainty Evaluation

Serum cortisol: an up-to-date assessment of routine assay performance

Open access
Hawley, James M, Owen, Laura J, Lockhart, Stephen J, Monaghan, Phillip J, Armston, Annie, Chadwick, Carrie A, Wilshaw, Heather, Freire, Maisa +2 more·2016·Clinical Chemistry
DOI 10.1373/clinchem.2016.255034
Open this paper in CoChat — free accountOpen PDF

Abstract

BACKGROUND: Accurate serum cortisol quantification is required for the correct diagnosis and management of adrenal pathologies. Presently, most laboratories use immunoassay to measure serum cortisol with proficiency schemes demonstrating a wide dispersion of results. Here, we investigate the effects of sex, matrix, and antibody specificity on serum cortisol quantification in 6 routine assays. METHODS: Surplus serum was obtained before disposal and the following cohorts were created: males, nonpregnant females, pregnant patients, and patients prescribed either metyrapone or prednisolone. Samples were anonymized and distributed to collaborating laboratories for cortisol analysis by 6 routine assays. Cortisol was also measured in all samples using an LC-MS/MS candidate reference measurement procedure (cRMP); cortisol-binding globulin (CBG) was measured in the nonpregnant and pregnant female cohorts. RESULTS: Considerable inter- and intraassay variation was observed across the male and nonpregnant female cohorts relative to the cRMP. Four immunoassays underrecovered cortisol in the pregnancy cohort, and CBG was found to be significantly higher in this cohort than in the nonpregnant females. In the metyrapone and prednisolone cohorts, all immunoassays overestimated cortisol. The first generation Roche E170 and Siemens Centaur XP were particularly prone to overestimation. In all cohorts the routine LC-MS/MS assay aligned extremely well with the cRMP. CONCLUSIONS: Despite the clinical importance of serum cortisol, the performance of routine immunoassays remains highly variable. Accurate quantification is compromised by both matrix effects and antibody specificity. Underpinning this study with a cRMP has highlighted the deficiencies in standardization across routine cortisol immunoassays.

AI snapshot

Written from the paper’s full text.

TL;DR

Routine serum cortisol immunoassays show substantial sex-, matrix-, and antibody-specific bias relative to an LC-MS/MS candidate reference measurement procedure, while a routine LC-MS/MS assay aligns well with the cRMP across cohorts. The study highlights compromised standardization and assay-dependent over- or underrecovery in pregnancy, metyrapone, and prednisolone cohorts.

Key findings

  • Considerable inter- and intraassay variation was observed across male and nonpregnant female cohorts relative to the cRMP; first-generation Roche E170 and Siemens Centaur XP showed positive bias (male cohort mean bias 17.0% and 18.5%, respectively; nonpregnant female cohort 11.8% and 14.6%).
  • In pregnancy, four immunoassays underrecovered cortisol, and CBG was significantly higher in pregnant than nonpregnant females (mean 52.02 mg/L vs 27.23 mg/L, P < 0.0001); e.g., Siemens Centaur XP, Beckman Access, and Abbott Architect had mean biases of −5.5%, −23.3%, and −25.0%.
  • In the metyrapone cohort, all immunoassays overestimated cortisol (mean bias 16.4% to 95.0%); in the prednisolone cohort, all immunoassays overestimated cortisol with mean biases from 690.9% to 8169.4% (56.0 to 658.6 nmol/L).
  • The routine LC-MS/MS assay aligned extremely well with the cRMP in all cohorts; in the prednisolone cohort its mean bias was 15.5% but equated to only 3.1 nmol/L, and only 24/42 samples were quantifiable.

Methods

  • Surplus serum cohorts were created: 51 males, 45 nonpregnant females, 72 pregnant females (10–20 weeks), 42 patients prescribed prednisolone, and 27 patients prescribed metyrapone.
  • Cortisol was measured by an LC-MS/MS cRMP, 5 automated commercial immunoassays (Abbott Architect, Beckman Access, Roche E170 Generation I, Roche E170 Generation II, Siemens Centaur XP), and a routine LC-MS/MS assay; CBG was measured by commercial ELISA in nonpregnant and pregnant female cohorts.
  • Analysis used Passing-Bablok regression, Bland-Altman bias plots, Pearson correlation, Shapiro-Wilk test, and dependent-samples t-test or Wilcoxon test as appropriate.

Limitations

  • The paper states that the male and nonpregnant female cohorts lacked particularly high cortisol concentrations, which are difficult to obtain without sample adulteration.
  • The paper notes that information on how commercial immunoassays liberate cortisol from binding proteins remains proprietary, and that further work with larger sample sizes is needed to substantiate the relationship between CBG concentration and cortisol quantification.
  • Evident from the text: samples came from a single center (University Hospital South Manchester), some cohorts were small (metyrapone n = 27; prednisolone n = 42), and in the prednisolone cohort only 24/42 samples were quantifiable by the routine LC-MS/MS assay.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1373/clinchem.2016.255034

Cites

19 of 19 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 30 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    In the male cohort, the routine LC-MS/MS, Abbott Architect, Beckman Access, and second generation Roche E170 assays had mean biases between -1.5% and -1.2% relative to the candidate reference measurement procedure.

    “The routine LC-MS/MS, Abbott Architect, Beckman Access, and second generation Roche E170 all generated a mean bias between Ϫ1.5% and Ϫ1.2%.”
    Quote in the paperSubstantial claim
  • Findingreasoning

    In the male cohort, the first generation Roche E170 assay had a negative constant bias of -19.45 nmol/L.

    “This relationship likely contributes to the negative constant bias of Ϫ19.45 nmol/L.”
    Quote in the paperNot a substantial claim
  • Findingown data

    In the male cohort, for the routine LC-MS/MS, Abbott Architect, Beckman Access, and second generation Roche E170 assays, 95% confidence intervals encompassing 0% supported the absence of bias.

    “In each case, 95% CIs encompassing 0% supported the absence of bias.”
    Quote in the paperSubstantial claim
  • Findingown data

    In the male cohort, the first generation Roche E170 and Siemens Centaur XP assays displayed positive bias relative to target concentrations.

    “In contrast, the first generation Roche E170 and Siemens Centaur XP displayed a positive bias relative to the target concentrations.”
    Quote in the paperSubstantial claim
  • Findingown data

    In the nonpregnant female cohort, all immunoassays demonstrated significant differences from the candidate reference measurement procedure.

    “Although the Beckman Access and second generation Roche E170 exhibited good agreement with the cRMP in the male cohort, all immunoassays demonstrated significant differences with the cRMP in the nonpregnant female cohort.”
    Quote in the paperSubstantial claim
  • May combine two claims. 32
  • The quote only partly states this claim. 6
  • Classified as probably not a substantial claim. 12

  • May combine two claims.

    Findingown data

    In the male cohort, the candidate reference measurement procedure produced normally distributed cortisol results ranging from 24.8 to 399.8 nmol/L.

    “Analysis of this cohort by the cRMP generated normally distributed results ranging from 24.8 to 399.8 nmol/L.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In the male cohort, the first generation Roche E170 assay had an overall mean bias of 17.0%, a proportional bias of 24.0%, and bias that increased with concentration.

    “With an overall mean bias of 17.0%, yet a proportional bias of 24.0%, it is apparent from Fig. 1D that bias increases with concentration for the first generation Roche E170 assay.”
    Quote in the paperSubstantial claim

  • The quote only partly states this claim.

    Findingown data

    In the prednisolone cohort, immunoassays had mean biases ranging from 690.9% to 8169.4%, corresponding to 56.0 to 658.6 nmol/L.

    “In contrast, the immunoassays exhibited mean biases ranging from 690.9% to 8169.4%, corresponding to 56.0 to 658.6 nmol/L.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the nonpregnant female cohort, only the routine LC-MS/MS assay had 95% confidence intervals encompassing 0%; the other assays' intervals fell below 0%, indicating slight negative bias.

    “However, of these, only the routine LC-MS/MS assay possessed 95% CIs encompassing 0%.The 95% CIs of the other assays all fell below 0% indicating a slight negative bias.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the pregnancy cohort, the first generation Roche E170 assay had a mean bias of 10.2% and performed similarly to its performance in the nonpregnant female cohort.

    “With a mean bias of 10.2%, the first generation Roche E170 performed similarly both in the pregnant female and nonpregnant female cohorts.”
    Quote in the paperNot a substantial claim

  • The quote only partly states this claim.

    Findingown data

    In the nonpregnant female cohort, the routine LC-MS/MS, Abbott Architect, Beckman Access, and second generation Roche E170 assays had mean biases ranging from -4.9% to -0.4%.

    “The routine LC-MS/MS, Abbott Architect, Beckman Access, and second generation Roche E170 all generated results with a mean bias ranging from Ϫ4.9% Beckman Access Roche E170 Generation II Siemens Centaur XP Fig. 1.Bland-Altman bias plots depicting the performance of the assays investigated relative to the cRMP in the male cohort (n = 51).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based).

    Findingnone given

    In the pregnancy cohort, the performance of the Siemens Centaur XP, Beckman Access, and Abbott Architect assays was distinctly different from their performance in the nonpregnant female cohort.

    “The performance of these assays in the pregnant female cohort was dis- Dashed lines represent the 5% (95% CI) measurement uncertainty of the cRMP.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In the male cohort, the Siemens Centaur XP assay had an overall mean bias of 18.5%, resulting from a 13.0% proportional bias and a 15.59 nmol/L constant bias.

    “Conversely, in the case of the Siemens Centaur XP, the overall mean bias of 18.5% is the cumulative effect of a 13.0% proportional and a 15.59-nmol/L constant bias.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the prednisolone cohort, LC-MS/MS confirmed prednisolone in all samples, with a mean prednisolone concentration of 245 g/L and a range of 41 to 590 g/L.

    “LC-MS/MS quantification confirmed the presence of prednisolone in all samples.The mean (range) prednisolone concentration was 245 g/L (range 41 to 590 g/L).”
    Quote in the paperNot a substantial claim

  • May combine two claims. The quote only partly states this claim.

    Findingown data

    In the prednisolone cohort, routine LC-MS/MS had a mean bias of 15.5%, equating to 3.1 nmol/L, and all Passing-Bablok regression parameters had 95% confidence intervals including their targets.

    “Although the mean bias for LC-MS/MS was 15.5% this equated to only 3.1 nmol/L and all Passing-Bablok regression parameters produced 95% CIs that included their respective targets.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the pregnancy cohort, the second generation Roche assay compared well with the candidate reference measurement procedure, with a slight negative mean bias of -2.6% and Passing-Bablok slope and intercept 95% confidence intervals encompassing the target parameters.

    “Despite a slight negative mean bias of Ϫ2.6%, Passing-Bablok regression analysis produced a slope and intercept with 95% CIs encompassing the cRMP target parameters.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the prednisolone cohort, Passing-Bablok slope characteristics were positive relative to the line of identity for each assay except the Siemens Centaur XP, which failed the CUSUM linearity test (P < 0.01), invalidating regression analysis.

    “In addition, Passing-Bablok slope characteristics were positive relative to the line of identity for each assay except the Siemens Centaur XP, which failed the CUSUM linearity test (P Ͻ 0.01) thereby, invalidating regression analysis.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In the pregnancy cohort, the first generation Roche E170 assay had Passing-Bablok characteristics markedly different from the nonpregnant cohort: the 95% confidence intervals for the two slopes shared no overlap, and the intercept shifted from negative to positive.

    “Despite this, the Passing-Bablok characteristics were markedly different since the 95% CIs for the 2 slopes shared no overlap, and the intercept shifted from negative in the nonpregnant cohort to positive in the pregnant cohort.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the metyrapone cohort, immunoassay mean bias ranged from 16.4% to 95.0%, with Passing-Bablok slopes from 0.72 to 1.37 and intercepts from 34.02 to 127.82 nmol/L.

    “Of the immunoassays, mean bias ranged from 16.4% to 95.0% with Passing-Bablok regression param-eters ranging from 0.72 to 1.37 (slope) and 34.02 to 127.82 nmol/L (intercept).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the nonpregnant female cohort, the presumed premenopausal subgroup had a mean cortisol-binding globulin concentration of 29.37 mg/L (95% CI 27.65-31.09) and the presumed postmenopausal subgroup had 25.92 mg/L (95% CI 23.64-27.08).

    “Subdivision of the nonpregnant female cohort into their presumed pre-and postmenopausal cohorts provided mean (95% CIs) CBG concentrations of 29.37 (27.65-31.09)and 25.92 mg/L (23.64 -27.08), respectively.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In the pregnancy cohort, the Siemens Centaur XP, Beckman Access, and Abbott Architect assays had mean biases of -5.5%, -23.3%, and -25.0%, equating to -36.7, -143.6, and -156.6 nmol/L relative to the candidate reference measurement procedure.

    “Conversely, with respective mean biases of Ϫ5.5%, Ϫ23.3%, and Ϫ25.0% equating to Ϫ36.7, Ϫ143.6, and Ϫ156.6 nmol/L, the Siemens Centaur XP, Beckman Access, and Abbott Architect displayed markedly negative bias relative to the cRMP.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the nonpregnant female cohort, the candidate reference measurement procedure produced normally distributed cortisol results ranging from 56.5 to 602.7 nmol/L.

    “Analysis of this cohort by the cRMP yielded normally distributed results ranging from 56.5 to 602.7 nmol/L.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based). May combine two claims. The quote only partly states this claim.

    Findingnone given

    In the prednisolone cohort, linearity was detected between the first generation Roche E170 and the candidate reference measurement procedure, but the software could not provide 95% confidence intervals for the slope and intercept.

    “Although linearity was detected between the first generation Roche E170 and Dashed lines represent the 5% (95% CI) measurement uncertainty of the cRMP.”
    Quote in the paperNot a substantial claim

  • May combine two claims. The quote only partly states this claim.

    Findingown data

    In the prednisolone cohort, only 24 of 42 samples were quantifiable by routine LC-MS/MS; the remaining samples were correctly identified as below the assay's limit of quantification.

    “Furthermore, only 24/42 samples were quantifiable by the LC-MS/MS assay; the remaining samples were correctly identified as having a cortisol concentration below the limit of quantification of the assay.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The Abbott Architect consistently underestimated cortisol, while the first generation Roche E170 and Siemens Centaur XP consistently overestimated it.

    “Whereas the Abbott Architect consistently underestimated cortisol, the first generation Roche E170 and Siemens Centaur XP consistently overestimated it.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the nonpregnant female cohort, the first generation Roche E170 and Siemens Centaur XP assays had mean biases of 11.8% and 14.6%, with 95% confidence intervals above 0% indicating positive bias.

    “The first generation Roche E170 and Siemens Centaur XP displayed respective mean biases of 11.8 and 14.6%.Like the male cohort, both assays generated 95% CIs that were above 0% identifying a positive bias.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The first generation Roche E170 and Siemens Centaur XP assays are markedly affected by cross-reactivity, while the Abbott Architect, Beckman Access, and second generation Roche E170 assays are also subject to considerable overrecovery.

    “While our data suggest that the first generation Roche E170 and Siemens Centaur XP are markedly affected by cross-reactivity, the Abbott Architect, Beckman Access and second generation Roche E170 are also subject to considerable overrecovery.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingreasoning

    Overrecovery in the male, nonpregnant, and pregnant female cohorts improved with the second generation Roche E170 assay, with better alignment to the candidate reference measurement procedure.

    “Certainly, the overrecovery exhibited in the male, nonpregnant and pregnant female cohorts has improved, with better alignment to the cRMP.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    The Endocrine Society has set minimum performance criteria for steroid assays, acknowledging that metrological traceability is not yet available for all assays but remains an important goal.

    “The Endocrine Society has now established minimum performance criteria for steroid assays, which recognize that metrological traceability is not currently available for all assays, yet remains an important goal (12)(13)(14).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [12] Handelsman, 2013, Requirement for mass spectrometry sex steroid assays in the journal of clinical endocrinology and metabolism, J Clin Endocrinol Metab source

      Not checked in the free run.

    • cites [13] Wierman, 2014, The new instructions to authors for the reporting of steroid hormone measurements [Editorial], Mol Endocrinol source

      Not checked in the free run.

    • cites [14] Monaghan, 2014, Case for the wider adoption of mass spectrometry-based adrenal steroid testing, and beyond, J Clin Endocrinol Metab source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Mean cortisol concentrations at baseline and 30 minutes after synacthen are assay-specific and show a consistent negative bias in women taking the oral contraceptive pill.

    “Furthermore, mean cortisol concentrations at baseline and 30 min post-synacthen were found to be assay specific and exhibit a consistently negative bias in women taking the oral contraceptive pill (OCP) (9 ).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [9] El-Farhan, 2013, Method-specific serum cortisol responses to the adrenocorticotrophin test: comparison of gas chromatography-mass spectrometry and five automated immunoassays, Clin Endocrinol (Oxf) source

      Not checked in the free run.

  • May combine two claims.

    Premiseown data

    External quality assessment schemes show wide dispersion of serum cortisol results across methods, with a sex-specific bimodal distribution.

    “Although variation for individual methods generally remains within narrow limits, the variation across all methods is considerably wider.This frequently produces a wide dispersion of results with a sexspecific bimodal distribution (see the Data File 1 in the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol62/ issue9).”
    • Also stated in: § Abstract · boosted
      “Presently, most laboratories use immunoassay to measure serum cortisol with proficiency schemes demonstrating a wide dispersion of results.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Most laboratories use automated commercial immunoassays for serum cortisol quantification.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The Partnership for Accurate Testing of Hormones was established to address standardization in selected assays.

    “The Partnership for Accurate Testing of Hormones has been established to address this issue in select assays (15 ).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [15] PATH. Partnership for the accurate testing of hormones. http://www.hormoneassays.org (Accessed January 2016).

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation

    Previous work speculated that cortisol underrecovery may be attributed to increased cortisol-binding globulin concentrations secondary to pregnancy or oral contraceptive pill medication.

    “Previous work has speculated that cortisol underrecovery may be attributed to increased CBG concentra-tions secondary to pregnancy (8 ) or OCP medication (9 ).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [8] Dodd, 2014, The effect of serum matrix and gender on cortisol measurement by commonly used immunoassays, Ann Clin Biochem source

      Not checked in the free run.

    • cites [9] El-Farhan, 2013, Method-specific serum cortisol responses to the adrenocorticotrophin test: comparison of gas chromatography-mass spectrometry and five automated immunoassays, Clin Endocrinol (Oxf) source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Metyrapone attenuates negative feedback on the hypothalamic-pituitary-adrenal axis, resulting in accumulation of cortisol precursors.

    “This attenuates negative feedback on the hypothalamic-pituitaryadrenal axis, resulting in the accumulation of cortisol precursors (6 ).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [6] Nieman, 2015, Treatment of Cushing's syndrome: an endocrine society clinical practice guideline, J Clin Endocrinol Metab source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Target cortisol concentrations for patients prescribed metyrapone should fall between 150 and 300 nmol/L.

    “In the treatment of Cushing syndrome, target cortisol concentrations for patients prescribed metyrapone should fall between 150 and 300 nmol/L (6 ).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [6] Nieman, 2015, Treatment of Cushing's syndrome: an endocrine society clinical practice guideline, J Clin Endocrinol Metab source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductreasoning

    Using a candidate reference measurement procedure ensured each method was compared with a higher-order measurement procedure, certifying metrological traceability to SI units and providing a platform for meaningful statistical analysis.

    “Underpinning this study with a cRMP has ensured that each method was compared to a higher-order measurement procedure.This certifies metrological traceability to SI units thereby, providing a platform for meaningful statistical analysis.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [16] Hawley, 2016, Candidate reference measurement procedure for the quantification of total serum cortisol with LC-MS/MS, Clin Chem source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Manufacturers may displace cortisol from cortisol-binding globulin by adjusting temperature, lowering pH, or using agents such as salicylate or 8-anilino-1-naphthalene sulfonic acid.

    “Nevertheless, it has been reported that manufacturers may displace cortisol from CBG by adjusting temperature, lowering pH or alternatively using agents such as salicylate or 8-anilino-1-napthalene sulfonic acid (ANS) (19 ).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [19] Bertholf, 2012, The adrenal cortex source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S1 (asserted), S4 (based).

    Interpretationnone given

    Over-recovery in the first generation Roche E170 and Siemens Centaur XP assays is secondary to nonspecific cross-reactivity.

    “Whereas underrecovery is attributed to matrix effects such as increased CBG concentrations, we hypothesize that the over-recovery in the first generation Roche E170 and Siemens Centaur XP is secondary to nonspecific cross-reactivity.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Prednisolone and prednisone are the most commonly prescribed exogenous glucocorticoids.

    “Prednisolone and prednisone are the most commonly prescribed exogenous glucocorticoids (20 ).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [20] Forss, 2012, Current practice of glucocorticoid replacement therapy and patient-perceived health outcomes in adrenal insufficiency – a worldwide patient survey, BMC Endocr Disord source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    Cortisol-binding globulin concentrations decline after menopause, possibly secondary to an attenuated estrogenic stimulus.

    “This infers that CBG concentrations decline postmenopause, possibly secondary to the attenuated estrogenic stimulus.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Conditions with a high estrogenic state stimulate cortisol-binding globulin synthesis.

  • Probably not a substantial claim: may not pass S2 (checkable).

    Premisecitation

    For patients not on lifelong treatment, adrenal reserve should be ascertained before altering dosing regimens.

    “For patients not on lifelong treatment, their adrenal reserve should be ascertained before altering dosing regimens (4 ).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [4] Bancos, 2015, Diagnosis and management of adrenal insufficiency, Lancet Diabetes Endocrinol source

      Not checked in the free run.

  • The quote only partly states this claim.

    Interpretationown data

    In the prednisolone cohort, the routine LC-MS/MS assay exhibited superior performance compared with the immunoassays.

    “Despite this, the routine LC-MS/MS assay exhibited superior performance compared to the immunoassays.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductreasoning

    The study is representative of the male and nonpregnant female populations, although it is limited by the absence of particularly high cortisol concentrations.

    “Although limited by an absence of particularly high cortisol concentrations in the male and nonpregnant female cohorts, this study is representative of these populations, with high concentrations difficult to obtain without sample adulteration.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: High cortisol concentrations are difficult to obtain without sample adulteration.

  • May combine two claims.

    Interpretationreasoning

    Underestimation in pregnancy was exacerbated with increasing cortisol concentrations, a relationship seemingly related to increased cortisol-binding globulin concentrations.

    “This underestimation was exacerbated with increasing cortisol concentrations, a relationship that was seemingly related to increased CBG concentrations.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisereasoning

    Cortisol precursors share structural homology with cortisol, necessitating highly specific antibodies for immunoassay selectivity.

    “These precursors share structural homology with cortisol, necessitating that immunoassays require highly specific antibodies to ensure selectivity.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    Sex- and assay-specific cutoffs may be helpful in the short term, but reagent lot changes, recalibration, and assay reformulation may increase result uncertainty and thus not represent a long-term solution.

    “Although these may be helpful in the short-term, reagent lot changes, recalibration and assay reformulation may increase result uncertainty and thus not represent a longterm solution.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    For patients prescribed metyrapone, immunoassay quantification may lead to an inappropriate dosing regimen and may ultimately mask adrenal insufficiency.

    “Similarly, for patients prescribed metyrapone, immunoassay quantification may lead to an inappropriate dosing regimen and may ultimately mask adrenal insufficiency.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    Assays must differentiate cortisol from exogenous glucocorticoids and their metabolites to guarantee the integrity of the short synacthen test.

    “This provides a requirement for assays to differentiate cortisol from exogenous glucocorticoids and their metabolites, to guarantee the integrity of the short synacthen test.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    Results are not comparable across assay, location, or time.

    “Consequently, results are not comparable across assay, location or time.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The case for sex- and assay-specific cutoffs has been made.

    “The case for sex-and assay-specific cutoffs has been made (9 ).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [9] El-Farhan, 2013, Method-specific serum cortisol responses to the adrenocorticotrophin test: comparison of gas chromatography-mass spectrometry and five automated immunoassays, Clin Endocrinol (Oxf) source

      Not checked in the free run.

45 of 45

  • The paper relies on this without support.

    Most laboratories use automated commercial immunoassays for serum cortisol quantification.

    “Currently, the majority of laboratories use automated commercial immunoassays for cortisol quantification.”

  • The paper relies on this without support.

    The poor performance of routine cortisol assays has not yet been considered.

    “However, despite its clinical importance, the poor performance of routine cortisol assays has yet to be considered.”

  • The paper relies on this without support.

    Conditions with a high estrogenic state stimulate cortisol-binding globulin synthesis.

    “These conditions are associated with a high estrogenic state that serves to stimulate CBG synthesis.”

  • The paper relies on this without support.

    Accurate cortisol measurement relies on complete recovery of cortisol from its protein-bound fractions.

    “Accurate cortisol measurement relies on the complete recovery of cortisol from its protein-bound fractions.”

  • The paper relies on this without support.

    Consistent sample preparation and storage minimize variation due to noncommutability and heterogeneity.

    “In all cases, variation due to noncommutability and heterogeneity has been minimized through consistent sample preparation and storage before analysis.”

  • The paper relies on this without support.

    High cortisol concentrations are difficult to obtain without sample adulteration.

    “Although limited by an absence of particularly high cortisol concentrations in the male and nonpregnant female cohorts, this study is representative of these populations, with high concentrations difficult to obtain without sample adulteration.”
6 of 6

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • An increased amount of corticosteroid-binding globulin entails a risk for misleadingly normal total plasma cortisol levels.

    “However, an increased amount of corticosteroid-binding globulin (CBG) entails a risk for misleadingly normal total plasma cortisol levels as more CBG-bound cortisol will be measured, even though cortisol release and free, biologically active cortisol levels may be low (5-9).”

    From Salivary Cortisol and Cortisone Can Circumvent Confounding Effects of Oral Contraceptives in the Short Synacthen Test (2024) · cites this paper as [8]

    The Discussion (page unavailable) links increased CBG in pregnant patients to immunoassay underrecovery of cortisol: the assays may fail to release and measure all CBG-bound cortisol. That is contrary to the attributed explanation that increased CBG causes more bound cortisol to be measured, producing misleadingly normal total levels. The paper does not assess whether free cortisol is low in that situation. (LLM, conf=0.85)

    Checked by a reasoning model — may be wrong.

    Experimental — accuracy not yet measured

  • Immunoassays have been shown to give variable results depending on the specificity and sensitivity of the antibody used.

    “Immunoassays have seen shown to give variable results depending on the specificity and sensitivity of the antibody used”

    From Bioavailability of Oral Hydrocortisone Corrected for Binding Proteins and Measured by LC-MS/MS Using Serum Cortisol and Salivary Cortisone (2018) · cites this paper as [4]

    The Discussion (page unavailable) reports variable cortisol immunoassay results and links overestimation to antibody cross-reactivity; it also suggests that a more specific antibody improved one assay. The paper does not establish that variation depends on antibody sensitivity, and it identifies matrix effects as another cause. (LLM, conf=0.86)

    Experimental — accuracy not yet measured

  • LC-MS/MS is the gold standard method for measuring steroids.

    “LC-MS/MS is now the gold standard method for measuring steroids [4].”

    From Bioavailability of Oral Hydrocortisone Corrected for Binding Proteins and Measured by LC-MS/MS Using Serum Cortisol and Salivary Cortisone (2018) · cites this paper as [4]

    The ASSAYS section uses an LC-MS/MS candidate reference measurement procedure for serum cortisol, and the results report that routine LC-MS/MS performed better than the immunoassays tested. The paper does not establish LC-MS/MS as the gold standard for measuring steroids generally; its comparison is specific to cortisol. (LLM, conf=0.86)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • First-generation Roche assay described as both overestimating and underestimating cortisol in pregnancy

    Major

    The pregnancy results report a positive 10.2% mean bias for the first-generation Roche E170, but the discussion says this assay showed considerable underestimation relative to the cRMP in pregnant patients.

  • Postmenopausal CBG confidence interval is not centered on its reported mean

    Minor

    The reported 95% confidence interval has a midpoint of 25.36 mg/L, not the stated mean of 25.92 mg/L. Unlike the other mean confidence intervals in this paragraph, its bounds are not symmetric about the mean.

    Reported 25.92 · recomputed 25.36

2 of 2

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Hawley, J. M., Owen, L. J., Lockhart, S. J., Monaghan, P. J., Armston, A., Chadwick, C. A., Wilshaw, H., Freire, M., Perry, L., & Keevil, B. G.
Affiliations
  • University Hospital South Manchester, Manchester, UK
  • Manchester Academic Health Science Centre (GB)
  • The Christie Hospital (GB)
  • Southampton Hospital (US)
  • Aintree University Hospital (GB)
  • St. Georges' University Hospital, London, UK
  • Croydon University Hospital (GB)
Corresponding author
Hawley, James M
ORCID
Venue
Clinical Chemistry
Publisher
Oxford University Press (OUP)
Published
2016-09-01
Volume
62
Issue
9
Pages
1220-1229
ISSN
  • 0009-9147 (ISSN-L)
  • 1530-8561
DOI
10.1373/clinchem.2016.255034
OpenAlex
W2501751012
Cited by
95
References
19
Citations by year
2026: 8 · 2025: 10 · 2024: 7 · 2023: 11 · 2022: 9 · 2021: 15 · 2020: 11 · 2019: 9 · 2018: 7 · 2017: 7 · 2016: 1
FWCI
6.44
Type
journal article
Language
en
Open access
Licence
academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model (vor)
Locations
  • Clinical Chemistry (publishedVersion)
  • PubMed (publishedVersion)
  • Research Explorer (The University of Manchester) (publishedVersion, other-oa)
Funders
Abbott Laboratories
Topics
  • Health Sciences › Medicine › Physiology › Clinical Laboratory Practices and Quality Control
  • Health Sciences › Medicine › Endocrinology, Diabetes and Metabolism › Adrenal Hormones and Disorders
  • Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Scientific Measurement and Uncertainty Evaluation
Keywords
serum cortisol, immunoassay performance, matrix effects, antibody specificity, corticosteroid-binding globulin, pregnancy, metyrapone, prednisolone, assay standardization, adrenal function testing
MeSH terms
Adult, Aged, Aged, 80 and over, Female, Humans, Hydrocortisone, Male, Middle Aged, Pregnancy, Cohort Studies

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-09.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-09.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review