Back to Ask the Diabetes Team Ask the Diabetes Team

From Findlay, Ohio, USA:

On December 12, 2009, my five-year-old son was diagnosed with type 1 diabetes after urinating frequently and drinking more for a few days. His blood sugar was 780 mg/dl [43.3 mmol/L] and his A1c was 10%. The endocrinologist explained that the A1c meant his blood sugar average for the last 90 days was 275 mg/dl [15.3 mmol/L]. My son's A1c on January 14 was 7.2 after 30 days on insulin, a three point drop in the A1c! Is it possible that the A1c could drop that much in such a short time?

As I was concerned about an H1N1 Vaccination on November 10, 2009, I was wondering if the majority of this high daily reading could have happened in the last 30 days. So, I put together a 90 day spreadsheet prior to the December 12 diagnosis and tried to put in normal theoretical daily blood sugar value of 100 or 110 mg/dl [5.6 or 6.1 mmol/L] for the first 60 days and then tried to ramp up the theoretical values for the last 30 days, (after the H1N1 vaccination), such that I could average a daily blood sugar value of 275 mg/dl [15.3 mmol/L].

While it is possible, theoretically, to have values that would make this possible, there is no way to adjust this 90 day sliding window, or any 90 day sliding window, with the new A1c test result on January 14 of 7.2% (which equates to an average daily blood sugar reading of 160 mg/dl [8.9 mmol/L]). No matter how you adjust the theoretical daily numbers, there just does not seem to be any way, one could go from a A1c of 10% (daily average of 275 mg/dl [15.3 mmol/L]) to an A1c of 7.2% (daily average of 160 mg/dl [8.9 mmol/L]) one month later if this A1c test is looking at blood glucose based on red blood cells living 90 days. I even tried redoing the spreadsheet for the original December 12 diagnosis and assuming 275 mg/dl [15.3 mmol/L] every day for 90 days and this sliding 90 day window, once on insulin and for the sake of argument assumed excellent control, such that the last 30 days averaged 110 mg/dl [6.1 mmol/L] from December 12 to January 14, and there is no way to average 160 mg/dl [8.9 mmol/L] per day.

The theoretical daily numbers do not work even if one assumes the red blood cells are living for only 60 days. I did try a six-week scenario, which did provide some overlap between the 10% A1c reading and the 7.2% A1c reading one month later and even this is difficult to rectify with theoretical daily numbers.

If Hemoglobin A1c tests really do look back to provide an average daily blood glucose level for the last 90 days (or even only 60 days), how is it theoretically possible to drop from 10% to 7.2% in 30 days?


It is extremely common for HbA1c values to decrease by 2 or more points within four to six weeks after the INITIAL diagnosis. While the HbA1c is commonly said to reflect the "average" glucose over the prior 90 days, it may really more accurately reflect glucose readings over 30 to 60 or so days. Furthermore, "newer glucose values" tend to "out-weigh" older values.

It may help to review that the HbA1c is essentially measuring the percentage of hemoglobin that has glucose attached to it. Hence, the initial value was 10%. Hemoglobin is the protein found within EVERY red blood cell (the cells that carry oxygen). Whereas oxygen attaches to hemoglobin REVERSIBLY, glucose attaches IRREVERSIBLY; once attached, it remains attached until the red blood cell dies and the hemoglobin is degraded and/or "recycled." The "average" life span for a red blood cell is about 120 days. AT ANY GIVEN MINUTE, assuming your bone marrow and red blood cells and hemoglobin are normal, you are making new red blood cells, maintaining red blood cells, or degrading 120 day old red blood cells. Just as you read this you made new cells and degraded older cells. And, you just did it again, and again and again just now. Get it? New cells with new hemoglobin haven't had a chance to be "frosted" yet by glucose and 120 day old cells are now degraded and unrecognized by the test. So, essentially, at any moment "one-third" of your red blood cells (the maintaining ones) are able to be analyzed by the A1c test. (I think of it as one-third of 120 days equals 40 days.) So, people with normal glucose metabolism have "some" glucose attached to red blood cells -- about 4 to 6%. (The highest I have seen in a teen with very poorly controlled diabetes was 25 or 27%, I think.)

But, your inferences are correct: other factors, besides the "glucose load" will determine and/or impact the A1c value, including variabilities in red blood cell composition, hemoglobin molecules (hemoglobin A1 being the most common normal hemoglobin, but there is also hemoglobin A2, hemoglobin S [sickle cell hemoglobin], hemoglobin C, persistent fetal hemoglobin [hemoglobin F] and MANY others, most of which are minor players in the world of hemoglobin but can impact the life-span of the red blood cell, the degree of "stickiness" to glucose, and even the ability to be measured in some HbA1c assay techniques. In fact, a person can have combinations of hemoglobin and be "carriers" for hemoglobins A1 and S, for example; this is called "sickle cell trait." DO NOT confuse hemoglobin variant to "blood type." They are different and distinct.

The rapid drop in your child's HbA1c value likely reflects the significant impact insulin and meal-planning has had on his glucose readings - and you should have seen this in his glucose diary that you keep. Over time, if the glucose logbook readings do not correlate with the HbA1c value, your pediatric endocrinologist should look to investigate why and that might mean an analysis of hemoglobins in your child. Certainly, if there is a known family history of irregular hemoglobins (such as sickle cell hemoglobin) let your doctors know now. Also, in my own clinical practice, it was common for the initial HbA1c value to be measured by a different laboratory technique (with slight but distinct different specifications and reference range values) than the device I would use in my office. It stressed that HbA1c and the daily glucose readings, while interconnected, are distinct and give you and your physician distinct pieces of information. DO NOT FOREGO glucose checking (and logbook diary) simply to rely on the HbA1c value!

While it might be academically satisfying for you to learn all this, at this point in your learning curve, you probably will do yourself a favor (and avoid an ulcer) by focusing on the things that you must do and accept a little "on faith" for now. There is plenty of time to get more knowledge. But, different people have different learning styles. Keep a dialogue open with your diabetes team.


[Editor's comment: You did not indicate how often you are checking your son's blood sugar. You need to remember that the numbers you are seeing are only those at a given point in time. So, when you find a 130 mg/dl [7.3 mmol/L] at 12:30 p.m. and you do not check again until 3 p.m. (if you check before a snack, which you may or may not provide) or dinner time, you don't what happened to your son's blood sugar in that time you were not checking. This is not to suggest that you should check more often. Your diabetes team should have instructed you when to check your son's blood sugar. Furthermore, your son is probably in his honeymoon phase, which means his pancreas is still producing and releasing insulin. How much and when, no one knows. In the meantime, based on the lower A1c, it looks as if you are doing a good job caring for your son.


Original posting 1 Feb 2010
Posted to A1c, Glycohemoglobin, HgbA1c


  Home Return to Top

Last Updated: Tuesday April 06, 2010 15:10:20
This Internet site provides information of a general nature and is designed for educational purposes only. If you have any concerns about your own health or the health of your child, you should always consult with a physician or other health care professional.

This site is published by T-1 Today, Inc. (d/b/a Children with Diabetes), a 501c3 not-for-profit organization, which is responsible for its contents.
By using this site, you agree to our Terms of Use, Legal Notice, and Privacy Policy.
© Children with Diabetes, Inc. 1995-2015. Comments and Feedback.