Vitamin D Deficiency: I Know It When I See It (Or Do I?)

Some things you just know.  Right from wrong, for instance, coming to a complete stop at a red light.  Well, hopefully!  Others are relative.  Driving 65mph on the German autobahn is slow but that same speed in a school zone is ludicrous.  But this latter requires precise measurement.  How do you know that you're driving 65mph if your speedometer doesn't work.  How does the police officer know how fast you were going if his/her radar detector isn't calibrated properly?

Likewise, while we've heard lots about vitamin D (deficiency) lately, how sure are we when making our diagnosis?  Shouldn't our lab results be 100% correct (accurate), or at least demonstrate repeatable results (precise)?  And therein lies the rub.

In a study to be published in print next month in the Journal of Clinical Endocrinology & Metabolism, the authors noted a rather statistically significant difference in results between two common but different assay methods for vitamin D when analyzing the same sample.  Not that the names are important, but the Immunodiagnostic Systems RIA (IDS-RIA) assay delivered higher numerical results than did the DiaSorin Liaison assay when both tested the same blood sample.

Put another way, you had a better than 1 in 3 chance of being declared normal with regards to your vitamin D if your blood sample was tested on an IDS-RIA machine but only a 1 in 5 chance if tested on DiaSorin machine.  Of course, you & I are at the mercy of the laboratory our insurance company sides with, and it's not likely that we can ask for special dispensation to get our blood redrawn and retested using the competitor's equipment.  But this study certainly doesn't help give me more confidence.  Sure, if my patient has a high level on his/her test, it'll probably be high on the other machine.  Likewise, if it's really low.  But most of my patients are just barely normal, which means they might be deficient if tested on the other machine (and vice versa).

This doesn't even take into account what's considered normal, much less optimal.  Clearly we need to come to grips with testing such that results are replicable across town, much less across the country, especially if we're to advance knowledge and promulgate guidelines.  Go ask your insurer since when close enough became good enough.  In the mean time, I'd like to believe that someday "I [will] know it when I see it", to paraphrase US Supreme Court Justice Potter Stewart's 1964 argument, rather than wonder which assay was used.

Follow @alvinblin
Tweet