There are doubts about relying solely on the HbA1c tests in medical circles. These links are just a few of many on the subject.
Many thanks!
Summary: In non-diabetic people, red blood cells live longer than in diabetics (as long as 146 days). This can drive up the A1C to mistakenly high levels in non-diabetic people. The converse is true for diabetics, whose red blood cells live for as few as 81 days. This leads to falsely low A1C levels in diabetics. Post-meal home glucose tests are a much better indicator.
My reaction: This seems to raise more questions than it answers. If the readings are "falsely low" in diabetics, and if this is a general problem, surely the cutoff level of 6.5% takes that into account? The A1C test has been standardized since around 2009 (the article dates from early 2011).
Summary: Situations where the average red blood cell lifespan is significantly less than 120 days will usually give rise to low HbA1c results because 50% of glycation occurs in days 90-120.
My reaction: Same as previous article. I honestly do not understand this point at all.
Summary: The test is unreliable in patients whose BG oscillates between very high and very low. In these cases the A1C should be supplemented by home BG tests.
My reaction: Now we are getting somewhere! Mind you, the question of whether very high and very low spikes are bad in themselves (even with a good A1C) is not tackled directly. But it makes sense to me that
very large daily spikes could be dangerous.
Summary: Similar point to previous two articles. A1C can be artifically low because diabetics' red blood cells have shorter life. I still do not understand this point at all! Additional (already well known) points: the test can be unreliable in the presence of anemia, or for certain ethnic groups, or in the presence of certain genetic factors.
Summary: A controlled trial, comparing the A1C method with the CGM (continuous glucose monitoring) method. It was found that, "not infrequently," A1C and CGM gave different results. "Thus, a patient’s CGM glucose profile has considerable value for optimizing his or her diabetes management."
Reaction: This is much better evidence, since it is based on a controlled trial. I would love to know what "not infrequently" means. The trial consisted of 387 people and the article is recent (published this year). I have only consulted the free summary. Unfortunately, to see the actual data costs $35 for a one-day peek. Has anyone seen the full data and can tell us what is meant by "not infrequently"?
Summary: A1C will vary between individuals because of various factors. This makes it unrealistic to apply a "one size fits all" diagnostic level for diabetes, for all individuals. Factors such as anemia make a difference. Plus, in any case, a given A1C test should be considered to be accurate to no better than about plus or minus 0.5 percentage points (which is quite a range!).
Reaction: Fair enough, but already widely known. To me what matters is the personal
trend. I know where
my A1C started, at diagnosis. If I can lop off "X" percentage points with diet/exercise, that is what matters to me, without obsessing too much about the absolute level. Concerning the second point (the margin of error on each A1C test). Personally, I don't obsess about a 0.5-percentage-point shift up or down. On the day of my diagnosis, the first thing I looked up was the margin of error: with my reading of 8.3%, I was obviously diabetic even if the reading had been erroneously high by 0.5 percent.
______
From reading the above articles, I emerge somewhat more convinced of the usefulness of self-testing on a daily basis, after meals, and so forth. However none of them really answers the question, "are daily (or occasional) spikes dangerous, as long as my A1C is under control?"