Vitamin D Deficiency

The ever increasing awareness of vitamin D deficiency and its clinical implications has led to a truly unique phenomenon within the reference and clinical laboratory: Testing for any single serum/plasma biomarker such as what is being experienced with 25-hydroxyvitamin D, 25(OH)D is without precedent.

Reported test volumes for 25(OH)D from America's three larg est reference laboratories range from 300K/month (LabCorp) to 500K/month (Quest). ARUP reports a doubling in volume over 2007.¹Every indication from the field corroborates the fact that fold growth is anticipated yet again for 2009.

Such growth, however, does not come without the concomitant need to examine whether the factors contributing to said growth are sustainable, and which methods, if any, are ably positioned to reliably address the challenge for timely and responsive delivery of accurate results to referring physicians.

Historical Notes

For perspective regarding sustainable growth, a few historical comments regarding vitamin D's supplement usage and clinical utility are warranted. Vitamin D deficiency has long been recognized as the

underlying cause of weak, deformed bones, atrophied muscles and the general malaise associated with rickets in children and osteomalacia in adults. The cumulative effect of vitamin D fortification programs, begun back in the 1930s, is now such that up to 98 percent of the U.S. milk supply is supplemented with vitamin D levels typically not exceeding 500 IU^2-4 er quart. A daily regimen including as little as 100 IU of vitamin D has been demonstrated to be effective in preventing rickets.⁵

A couple of unfortunate industrial instances of errant titration have led to isolated instances of apparent vitamin D toxicity: Significant excesses of cholecalciferol, vitamin D3 in milk⁴and in food supplements⁶have been documented. Vitamin D3 supplements are available in potencies ranging from 400 IU to 5,000 IU per capsule.⁷

Clinically, equally rare iatrogenic occurrences of vitamin D intoxication in adults are reported and have involved ergocalciferol, vitamin D2.⁸At present, only vitamin D2 (Drisdol) is available in high potency form as a medical prescription. Since its use predated the Food and Drug Administration, Drisdol as grandfathered as a pharmaceutical drug.⁹Of concern to the medical community is the fact that there are no firm guidelines for the proper dosing of vitamin D2. Warning statements like "The range between therapeutic and toxic doses is narrow" are prominently displayed on most web sites and product inserts. Yet, in apparent contradiction to these admonitions, commonly prescribed dosages for hypoparathyroidism and refractory rickets range anywhere from 50,000 IU to 200,000 IU and from 12,000 IU to 500,000 IU,

respectively.

A responsible, yet seemingly disproportionate sensitivity to the rare instances of toxicity discussed above, along with the minimal levels of supplementation needed to offset rickets, has led in large measure to the preservation of and adherence to potentially antiquated and overly conservative guidelines.¹⁰Meanwhile, and consequent to the past 15 to 20 years of burgeoning basic and more recent observational clinical research, it has become broadly appreciated that vitamin D deficiency may be linked with the development of numerous diseases involving heart failure, stroke, hypertension, multiple sclerosis, autism, cancer, Type 2 diabetes, systemic lupus erythmatosus, innate immunity and autoimmunity.¹³ Reflective of this nascent awareness, Heaney was one of the first to articulate a conceptual framework¹⁴to deal with what is now recognized as an expanding, global epidemic of vitamin D deficiency: Short-latency disease, like rickets, manifests severe deficiency whereas long-term latencies, like cancer, heart failure, etc. manifest chronic subclinical deficiencies.

Coming of Age

Clinically, juxtaposed against the breadth and depth of peer reviewed literature, failure to diagnose and treat vitamin D deficiency has been questioned as ethically irresponsible.^15-17Since 2003, annual calls for raising the sanctioned levels of daily supplementation to 1,000 IU (infants), 2,000 IU (children) and 4,000-5,000 IU (adults) have appeared.^18-22Safe supplementation levels as high as 10,000 IU have also been published.^23,24

Diagnostically, growth in reference laboratory testing for vitamin D directly reflects the active basic research and growing clinical interest. The DEQAS (Vitamin D External Quality Assessment Scheme) reported more than 470 participants from 33 countries in the third quarter of 2008 reflecting annual growth in the number of laboratories utilizing prevalent semi-automated and automated measurement methods, relative to third quarter 2007 results (Table 1).

As the surveillance of vitamin D intensifies, the need for proper normalization and standardization will increase as well. Whereas improper normalization fostered by antiquated regulatory guidelines has propagated circular epidemiology²⁵and slowed the advance of prospective clinical outcome-based research, non-uniform standardization from the reference laboratory sector translates and contributes directly to increased ambiguity in clinical practice.

Reference Ranges

Presently, normal ranges (or reference ranges) in many reference laboratories are improperly set. The conventional practice of generating Gausian distributions from diverse, asymptomatic populations of sufficient N is not applicable for vitamin D. Nevertheless, for the approximately 40 years prior to 2005 nonrachitic, non-osteomalaic populations, although vulnerable to broad variances attributable to race and degree of sun exposure (as modulated by latitude, sun screen usage and white or blue collar work habit), have been utilized to set "normal" ranges from 10 to 80 ng/ml to direct clinicians.²⁶"For the majority of labs, the bottom of the reference range is set too low due to the previous under appreciation of the clinical benefits of and physiologic requirement for higher vitamin D levels, and the top of the range is too low due to previous misinterpretation of the research resulting in an overestimation of vitamin D toxicity."²⁷i>

With the wealth of observational data published over the last five years, sufficiency and "normality" are now coming to be more properly formulated in terms of levelsthat have been associated clinically with a lower risk of disease development. Prevention incidence rates by 25(OH)D levels across multiple diseases are aligned in a chart presented at the 2008 Vitamin D Symposium.²⁸Based on studies inclusive of those mentioned therein, it is now recommended that 32 to 100 ng/ml be taken as normal and 50 to 80 ng/ml considered optimal.^17,26,27Clinical laboratories, however, have yet to adopt a common practice of reporting optimal levels; therefore, practitioners are being misled unless they stay abreast of current research.

Clinical Versus Analytical Accuracy

While a proper understanding of optimal versus normal range is important for sound clinical practice, an even more fundamental tenet is a proper grasp of clinical versus analytical accuracy. The most important consideration for any laboratory test is its ability to enable physicians to make confident, informed decisions regarding diagnosis and treatment.  Cost and accessibility may factor into the decision for selecting a method for a given laboratory, but pragmatics should never be allowed to override the seminal clinical knowledge based on over 1,000 peer-reviewed research publications. 

At stake is not whether liquid chromatography/mass spectrometry (LC/MS) or immunoassay-based methods are more accurate analytically, but how the results are applied to clinical practice. It is not appropriate to utilize LC/MS results with clinical reference ranges established on the latter. In recognition of this, some LC/MS based 25(OH)D measurement services have RIA-adjusted their LC/MS methodology.³⁰

The issue of whether LC/MS or immunoassays are more accurate analytically awaits further analysis. Plausible mechanisms as to whether immunoassays under-recover relative to LC-MS/MS or whether LC-MS/MS platform methods over-recover relative to immunoassay remain to be fully tested and substantiated. LC-MS/MS methods are presented in the literature as potentially more accurate due to the use of internal deuterated standards to normalize recovery. While deployment of said internal standards enable an exact accounting for absolute recovery from individual, specific platforms,³²this neither addresses, rectifies nor harmonizes inter LC-MS/MS platform differences originating from choice of calibrator matrix, selection/characterization of molar extinction coefficient(s) and/or serum interferents.

Validation Required

Until newer technologies are validated clinically (e.g., how well does the serum 25-OHD result predict serum parathyroid hormone), physicians should expect 25-OHD laboratory results to align with the methods and values prevalent in the evidence-based record of the clinical literature. Elements contributing to responsible practice are twofold. While it is incumbent upon IVD concerns to provide vitamin D testing methods that are clinically useful, reference laboratories should vigilantly assure that their testing procedures are aligned against published clinical guidelines. Ultimately, both utility and veracity are established and validated in clinical practice.

Dr. Blocki is scientific advisor, Bone & Mineral, DiaSorin Inc.