Assessment of vitamin D status and treatment of vitamin D deficiency are issues of public health interest [1–3]. This is mainly based on anti-fracture effects of vitamin D and its role in the prevention and treatment of rickets and osteomalacia [1–3]. Stimulated by data suggesting a possible role of vitamin D in extraskeletal diseases, there has been an exponentially increasing demand for measurements of serum 25-hydroxyvitamin D (25[OH]D) levels over the past few years [4–8]. Serum concentrations of 25(OH)D are considered to best reflect whole body vitamin D status and are thus used for its classification [1–3]. Although there is a debate ongoing on cut-off values for vitamin D deficiency as well as vitamin D intoxication, a wide consensus exists that 25(OH)D levels below 20 ng/mL (multiply by 2.496 to convert ng/mL to nmol/L) should be avoided by preventive or therapeutic approaches [1, 2]. As a consequence, clinical decisions regarding vitamin D treatment are increasingly based on laboratory results of 25(OH)D measurements, warranting a critical appraisal of 25(OH)D assay performance. Data on comparisons of automated immunoassays for 25(OH)D are relatively sparse but have already highlighted that some relevant assay differences exist [9–17]. Therefore, the aim of this study was to compare four frequently used automated immunoassays for 25(OH)D.
Materials and methods
The study population consists of hypertensive patients derived from the Graz Endocrine Causes of Hypertension (GECOH) study. Study protocol and characteristics of the GECOH study have been published elsewhere [18, 19]. In brief, we included patients who were referred for screening for endocrine hypertension to a tertiary care center at the Medical University of Graz, Austria. All study participants gave written informed consent and the study was approved by the Local Ethics Committee.
Measurements of 25(OH)D were done in fasting blood samples drawn between February 2009 and August 2010. Automated assays were performed according to manufacturers’ instructions on the IDS iSYS (25-hydroxyvitamin D; Immunodiagnostic Systems Ltd., Boldon, UK), the Liaison (25-OH-vitamin D Total, DiaSorin Inc., Stillwater, MN, USA), the Architect (ARCHITECT 25-OH Vitamin D assay, Abbott Park, IL, USA), and the Cobas 8000 system (Elecsys vitamin D total; Roche Diagnostics, Mannheim, Germany). Serum samples were immediately analyzed by the IDS iSYS on the day of blood sampling. EDTA plasma samples, stored at –20°C to –80°C, were used to measure 25(OH)D by the Liaison in August 2010 and by the Architect and Cobas 8000 system in January 2012. All analyses were performed by experienced technicians at the laboratory of either the Department of Endocrinology and Metabolism (IDS iSYS) or the Clinical Institute of Medical and Chemical Laboratory Diagnostics (all other assays) at the Medical University or Graz, Austria. This work complies with the World Medical Association Declaration of Helsinki regarding ethical conduct of research involving human subjects.
Data are presented as means±standard deviation (SD). Comparisons between assay results were done by Pearson correlation analyses, by Deming regression, and by Bland-Altman plots. Statistical analyses were performed by SPSS (Version 20.0; SPSS Inc., Chicago, IL, USA). A p-value below 0.05 was considered statistically significant.
Data on 25(OH)D concentrations measured by all four automated immunoassays were available in 106 patients (53±14 years; 59% females). There were significant differences in 25(OH)D levels between the different assays (see Table 1). To graphically illustrate the impact of different laboratory methods on vitamin D status classification, we showed the percentages of patients below a certain 25(OH)D threshold in Figure 1. Using a value of ≤20 ng/mL as the cut-off for vitamin D deficiency, the percentages of vitamin D deficient patients were significantly different depending on the method used: 79.2% (Abbott ARCHITECT), 50.0% (DiaSorin Liaison), 28.3% (IDS iSYS), and 23.6% (Roche Cobas 8000). Results of correlation analyses ordered from the highest to the lowest Pearson correlation coefficients were 0.85 (Abbott ARCHITECT vs. Roche Cobas 8000), 0.77 (DiaSorin Liaison vs. IDS iSYS), 0.69 (DiaSorin Liaison vs. Abbott ARCHITECT), 0.67 (IDS iSYS vs. Abbott ARCHITECT), 0.61 (Roche Cobas 8000 vs. IDS iSYS), and 0.57 (Roche Cobas 8000 vs. DiaSorin Liaison) (p<0.001 for all comparisons). Comparisons between these methods are further presented by using scatter plots including Deming regression lines (see Figure 2) and by showing Bland-Altman plots (see Figure 3). Our results were materially unchanged when males and females were analyzed separately.
Levels of 25-hydroxyvitamin D (in ng/mL) measured by different assays in a cohort of 106 hypertensive patients.
SD, standard deviation; IDS, immunodiagnostic systems.
In this study, we compared 25(OH)D levels of four different automated assays in samples derived from 106 hypertensive patients. We found significant differences between the laboratory methods with the percentages of vitamin D deficient patients ranging from 23.6% to 79.2% depending on the assay used. In addition, we observed only moderate correlations between the assays.
Our results fit well to previous literature showing differences in 25(OH)D levels measured by different laboratory methods. In 2004, Binkley et al. showed that when samples were sent to laboratory A 90% were below an arbitrary threshold of 32 ng/mL, whereas only 17% were below this cut-off value in laboratory B . These results were similar to our observed variability across different methods; however, our work has been done after several years in which new assays were introduced, some assays were withdrawn from the market, and some companies tried to further improve their laboratory methods. Indeed, according to data from the Vitamin D External Quality Assessment Scheme (DEQAS), which gives participating laboratories the opportunity to assess their assay reliability in comparison to other methods, the interlaboratory imprecision has improved from >30% in 1995 to 15% in 2011 [14–16]. Nevertheless, some recent studies comparing 25(OH)D assays with each other as well as with the gold standard mass spectrometry method also showed, in line with our findings, remarkable differences and only relatively moderate correlations between different assays [6–17]. In this context, Farrell et al. compared six routine assays (including all of our assays) and a mass spectrometry method . It was concluded that immunoassays demonstrated variable performance with best performance characteristics of the DiaSorin Liaison . Although this work was performed under highly standardized conditions there was a significant bias of all immunoassay methods with an up to 28% deviation from the mass spectrometry method. Hence, the variability of the 25(OH)D assay results in that study and a few other studies was lower compared to our work, but these studies were all performed in well-established reference laboratories taking care of almost all aspects with regard to standardization of laboratory procedures [9–12]. Although such excellent studies are invaluable for validation and further improvements of laboratory methods, they may not adequately reflect daily life of clinicians and researchers who are not frequently working with 25(OH)D measurements performed under such highly standardized conditions in reference laboratories. Thus, our work significantly adds to the existing literature by showing comparisons of 25(OH)D levels obtained under “daily life” conditions by using four of the most frequently used routine assays. It is really of concern that the diagnosis of vitamin D deficiency is so significantly dependent on the assay method used and even the two assays that revealed the most similar absolute values (Roche Cobas and IDS iSYS) showed a relatively poor comparability as evidenced by correlation analyses and Bland-Altman plots. Awareness of these assay problems is crucial for clinicians and researchers. In line with this notion, Schöttker et al. showed that in a large survey the prevalence of women with 25(OH)D levels <12 ng/mL was 48.3% by using an older version of the DiaSorin Liaison and decreased to 15.7% after standardization of this assay with a mass spectrometry method . Translating such assay comparisons into clinical routine is difficult when considering that the direction of systematic bias between different assays is not consistent throughout the literature, that is, whereas Schöttker et al. and some other studies including ours showed higher 25(OH)D levels by the IDS iSYS compared to the DiaSorin Liaison, the opposite was observed by Heijboer et al. [10–12, 20]. Similar inconsistent data with regard to systematic bias have been reported for other assays [6–17, 20, 21]. Several reasons for these assay differences have been discussed and evaluated. Although 25(OH)D is stable after long-term storage even at –20°C, there are several factors that may impact on 25(OH)D determinations. A major problem is the strong binding of 25(OH)D to its carrier vitamin D binding protein (DBP). It has already been demonstrated that DBP concentrations impact on measured 25(OH)D levels, questioning whether separation steps of routine assays that aim to free 25(OH)D from its carrier proteins are working properly . Cross-reactivity of assay antibodies may also account for assay differences and different vitamin D metabolites are indeed an analytical problem . Although the current opinion is that 25(OH)D levels should be reported as the sum of 25(OH)D3 and 25(OH)D2, various other vitamin D metabolites exist, such as the 3-epi-25(OH)D or the 24,25-dihydroxvitamin D with largely unclear biological significance [8, 14, 16, 22]. Although most of these metabolites circulate at relatively low concentrations, they may be relevant for certain populations, for example, relatively high 3-epi-25(OH)D levels are found in children.
Owing to our findings and other reports on significant differences between 25(OH)D assays, it is crucial to standardize 25(OH)D measurements. This is currently the aim of the Vitamin D Standardization Program (VDSP) . This VDSP is currently performed by the NIH Office of Dietary Supplements (ODS) in collaboration with the CDC National Center for Environmental Health (NCEH), the National Institute of Standards and Technology (NIST), and the Ghent University . Remeasurements of 25(OH)D in samples of large surveys and implementation of a laboratory certification program will be one of the main aims of the VDSP. In this context, it should be noted that a certified reference material for 25(OH)D has already been introduced . It might, however, be challenging how to deal with such newly standardized 25(OH)D levels in clinical routine when considering that current cut-off levels for vitamin D status classification have been mainly based on data by older immunoassays, in particular the DiaSorin Radio-Immunoassay (RIA). Depending on how significantly the VDSP will change the results of routine immunoassays, we might have to reconsider our cut-off levels for vitamin D status classification in the future. This might be of great clinical relevance because vitamin D deficiency has been identified as an independent risk factor for mortality in various different populations and vitamin D supplementation reduced mortality significantly in meta-analyses of randomized controlled trials [23–30].
Although our results are clearly limited by the fact that our measurements were performed by two different laboratories and in different samples (serum and plasma) with varying storage procedures, we believe that this is also a strength of our work because this may well reflect the “real life” situation. Missing data on DBP and on the gold standard mass spectrometry method including measurements of different vitamin D metabolites including the 3-epi-25(OH)D are another drawback of our work. This, however, allowed us to present our data without discrediting any of the four automated assays of our work by its comparison with mass spectrometry results.
In conclusion, our work documents assay differences for 25(OH)D measurements that significantly impact on vitamin D status classification and thus likewise also on clinical decisions regarding vitamin D treatment. Clinicians, researchers, and public health authorities must be aware of these limitations when comparing 25(OH)D values derived from different assay methods. Standardization of 25(OH)D measurements is therefore urgently needed and is already planned to be widely introduced within the next few years. New challenges will, however, come up when standardized 25(OH)D measurements will be introduced in the future, because then, past results of vitamin D research might appear in a new light and we might have to reconsider the cut-off values for vitamin D status classification.
We are grateful for all the support by the laboratory staff at the Department of Endocrinology and Metabolism and at the Clinical Institute of Medical and Chemical Laboratory Diagnostics at the Medical University of Graz. We are grateful for Abbott Diagnostics, Roche Diagnostics and DiaSorin for providing 25(OH)D reagents free of charge.
Conflict of interest statement
Authors’ conflict of interest disclosure: The authors stated that there are no conflicts of interest regarding the publication of this article. Research support played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.
Research funding: The GECOH study is supported by the province of Styria. Katharina Kienreich is supported by funding from the Austrian National Bank (Jubilaeumsfond: project numbers: 13878 and 13905). Nicolas Verheyen is supported by funding from the Austrian National Bank (Jubilaeumsfond: project number: 14621). Stefan Pilz is supported by the EFSD Albert Renold Travel Fellowship grant.
Employment or leadership: None declared.
Honorarium: None declared.
Ross AC, Manson JE, Abrams SA, Aloia JF, Brannon PM, Clinton SK, et al. The 2011 report on dietary reference intakes for calcium and vitamin D from the Institute of Medicine: what clinicians need to know. J Clin Endocrinol Metab 2011;96:53–8.
Holick MF, Binkley NC, Bischoff-Ferrari HA, Gordon CM, Hanley DA, Heaney RP, et al. Evaluation, treatment, and prevention of vitamin D deficiency: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2011;96:1911–30.
Souberbielle JC, Body JJ, Lappe JM, Plebani M, Shoenfeld Y, Wang TJ, et al. Vitamin D and musculoskeletal health, cardiovascular disease, autoimmunity and cancer: Recommendations for clinical practice. Autoimmun Rev 2010;9:709–15.
Muscogiuri G, Sorice GP, Ajjan R, Mezza T, Pilz S, Prioletta A, et al. Can vitamin D deficiency cause diabetes and cardiovascular diseases? Present evidence and future perspectives. Nutr Metab Cardiovasc Dis 2012;22:81–7.
Pilz S, Tomaschitz A, Drechsler C, Zittermann A, Dekker JM, März W. Vitamin D supplementation: a promising approach for the prevention and treatment of strokes. Curr Drug Targets 2011;12:88–96.
Herrmann M. The measurement of 25-hydroxy vitamin D – an analytical challenge. Clin Chem Lab Med 2012;50:1873–5.
Pilz S, Maerz W. Vitamin D and cardiovascular diseases: where do we stand in 2012? J Lab Med 2012;36:281–9.
Pilz S, Tomaschitz A, Pieber T, Lafer I, Drechsler C, Meinitzer A, et al. Vitamin D: clinical implications beyond musculoskeletal diseases. J Lab Med 2011;35:211–6.
Farrell C, Soldo J, Williams P, Herrmann M. 25-Hydroxyvitamin D testing: challenging the performance of current automated immunoassays. Clin Chem Lab Med 2012;50:1953–63.
Farrell CJ, Martin S, McWhinney B, Straub I, Williams P, Herrmann M. State-of-the-art vitamin D assays: a comparison of automated immunoassays with liquid chromatography-tandem mass spectrometry methods. Clin Chem 2012;58:531–42.
Janssen MJ, Wielders JP, Bekker CC, Boesten LS, Buijs MM, Heijboer AC, et al. Multicenter comparison study of current methods to measure 25-hydroxyvitamin D in serum. Steroids 2012;77:1366–72.
Heijboer AC, Blankenstein MA, Kema IP, Buijs MM. Accuracy of 6 routine 25-hydroxyvitamin D assays: influence of vitamin D binding protein concentration. Clin Chem 2012;58:543–8.
Lips P, Chapuy MC, Dawson-Hughes B, Pols HA, Holick MF. An international comparison of serum 25-hydroxyvitamin D measurements. Osteoporos Int 1999;9:394–7.
Carter GD. 25-Hydroxyvitamin D: a difficult analyte. Clin Chem 2012;58:486–8.
Carter GD, Berry JL, Gunter E, Jones G, Jones JC, Makin HL, et al. Proficiency testing of 25-hydroxyvitamin D (25-OHD) assays. J Steroid Biochem Mol Biol 2010;121:176–9.
Carter GD. Accuracy of 25-hydroxyvitamin D assays: confronting the issues. Curr Drug Targets 2011;12:19–28.
Binkley N, Krueger D, Cowgill CS, Plum L, Lake E, Hansen KE, et al. Assay variation confounds the diagnosis of hypovitaminosis D: a call for standardization. J Clin Endocrinol Metab 2004;89:3152–7.
Pilz S, Tomaschitz A, Stepan V, Obermayer-Pietsch B, Fahrleitner-Pammer A, Schweighofer N, et al. Graz Endocrine Causes of Hypertension (GECOH) study: a diagnostic accuracy study of aldosterone to active renin ratio in screening for primary aldosteronism. BMC Endocr Disord 2009;9:11.
Pilz S, Kienreich K, Stückler D, Meinitzer A, Tomaschitz A. Associations of sun exposure with 25-hydroxyvitamin D and parathyroid hormone levels in a cohort of hypertensive patients: the Graz Endocrine Causes of Hypertension (GECOH) study. Int J Endocrinol 2012;2012:732636.
Schöttker B, Jansen EH, Haug U, Schomburg L, Köhrle J, Brenner H. Standardization of misleading immunoassay based 25-hydroxyvitamin D levels with liquid chromatography tandem-mass spectrometry in a large cohort study. PLoS ONE 2012;7:e48774.
Cavalier E, Wallace AM, Carlisi A, Chapelle JP, Delanaye P, Souberbielle JC. Cross-reactivity of 25-hydroxy vitamin D2 from different commercial immunoassays for 25-hydroxy vitamin D: an evaluation without spiked samples. Clin Chem Lab Med 2011;49:555–8.
Sempos CT, Vesper HW, Phinney KW, Thienpont LM, Coates PM; Vitamin D Standardization Program (VDSP). Vitamin D status as an international issue: national surveys and the problem of standardization. Scand J Clin Lab Invest Suppl 2012;243: 32–40.
Zittermann A, Iodice S, Pilz S, Grant WB, Bagnardi V, Gandini S. Vitamin D deficiency and mortality risk in the general population: a meta-analysis of prospective cohort studies. Am J Clin Nutr 2012;95:91–100.
Pilz S, Iodice S, Zittermann A, Grant WB, Gandini S. Vitamin D status and mortality risk in CKD: a meta-analysis of prospective studies. Am J Kidney Dis 2011;58:374–82.
Pilz S, Tomaschitz A, Friedl C, Amrein K, Drechsler C, Ritz E, et al. Vitamin D status and mortality in chronic kidney disease. Nephrol Dial Transplant 2011;26:3603–9.
Thomas GN, O Hartaigh B, Bosch JA, Pilz S, Loerbroks A, Kleber ME, et al. Vitamin D levels predict all-cause and cardiovascular disease mortality in subjects with the metabolic syndrome: the Ludwigshafen Risk and Cardiovascular Health (LURIC) study. Diabetes Care 2012;35: 1158–64.
Putz-Bankuti C, Pilz S, Stojakovic T, Scharnagl H, Pieber TR, Trauner M, et al. Association of 25-hydroxyvitamin D levels with liver dysfunction and mortality in chronic liver disease. Liver Int 2012;32:845–51.
Pilz S, Dobnig H, Tomaschitz A, Kienreich K, Meinitzer A, Friedl C, et al. Low 25-hydroxyvitamin D is associated with increased mortality in female nursing home residents. J Clin Endocrinol Metab 2012;97:E653–7.
Bjelakovic G, Gluud LL, Nikolova D, Whitfield K, Wetterslev J, Simonetti RG, et al. Vitamin D supplementation for prevention of mortality in adults. Cochrane Database Syst Rev 2011;7:CD007470.
Rejnmark L, Avenell A, Masud T, Anderson F, Meyer HE, Sanders KM, et al. Vitamin D with calcium reduces mortality: patient level pooled analysis of 70,528 patients from eight major vitamin D trials. J Clin Endocrinol Metab 2012;97:2670–81.