What if you have an unexpected lab value?   (June 10, 2009)

As I do every time I'm at a big diabetes meeting, I head for the exhibit hall to see what's new - and to get my A1C checked. This past weekend, I was at the American Diabetes Association meeting in New Orleans, and the exhibit hall had not one, not two, but three different vendors offering free A1C testing. The vendors were using lab devices which are designed for physician office and/or patient home use, so they're not the huge ugly machines that hospital labs use, but petite machines that use a blood sample from a fingerstick, and which display an A1C result a few minutes later.

The last time I'd checked, a few months ago, my A1C had been 5.7. Things have not changed much recently, so I completely expected my results at the ADA meeting to be about the same.

But the A1C result was 6.6 - up almost a full percentage point (which is roughly equivalent to a change in blood glucose of about 35 mg/dl). Boy, was that frustrating - no matter how hard I tried, I couldn't think of a reason why it would have shot up that much.

So, off to the other vendors. The second result was 5.7. And the third result was also 5.7. Very interesting (and very reassuring to me about my diabetes control!) but it certainly did make me wonder about the 6.6 value - so I went back to that vendor, and got another A1C test (that's four in two days - certainly a record for me, and one that is likely to stand for a very long time).

6.3 this time. Totally frustrating. Next stop: to find the most knowledgeable person around at this exhibit booth and talk to him/her. I did. First, I pointed out the two results from his A1C device were "off" by 0.3 points from each other - which he indicated (and I agreed) could be within the range of variability of these devices.

Then I showed him my printouts of the other two results (both 5.7) - and asked him what he thought now. I presumed he'd probably bad-mouth the other companies' technology, but he totally surprised me by doing the reverse - he admitted that his company's readings were running high, and he didn't know why. He mentioned it might have been something to do with shipping, and that the company was saving some of the kits for future reassessment.

That opens a whole new can of worms. Should a vendor continue to test people at a diabetes meeting, knowing full well that his results are "off" from where he expected? Of course not. Especially as not all people will go to the effort to get free double-checks from other vendors as I did.

Yes, this may be a one-of-a-kind story, but my point to you, gentle reader, is simple: don't trust a single value of any lab test, anytime, anywhere, if it's different from what you expected: instead, get it rechecked. That's true for A1C; and it's also true for blood glucose, cholesterol, and indeed for any lab test.