HbA1c: How low is too low?

I saw an 82-year-old man last week with an HbA1c of 6.1%. His basal insulin had been continued to be increased because of fasting morning hyperglycemia. His fasting blood glucose was often in the 200 mg/dL to 300 mg/dL range. Because of this, his primary care provider raised his dose of bedtime basal insulin. He was not on scheduled pre-meal insulin but had a correction scale of fast-acting insulin, which he rarely used because whenever he did, he became hypoglycemic.

Scenarios such as this are common. The first question I ask when I see a frail elderly individual with such a low HbA1c while treated with insulin or oral hypoglycemic medications is: Where is the hypoglycemia that we are missing?

Although lower HbA1c has been shown to correlate with lower-risk of microvascular complications in studies such as theUnited Kingdom Prospective Diabetes Study and others, recently the benefits of lowering HbA1c have been questioned. Intensive blood pressure and glycemic control reduced microvascular complications in ADVANCE but not macrovascular events. The ACCORD and VADT trials raised concerns about the safety of intensive glycemic control.

In a retrospective cohort study recently published on line in Lancet, Currie et al found that an HbA1c of 7.5% was associated with the lowest mortality. Both high-and-low HbA1c were associated with increased all-cause mortality and cardiovascular events. This U-shaped curve was true for patients on oral combination vs. insulin based regimens but more prominent for those on insulin. The authors suggested that if results are confirmed, guidelines may need to include a minimum HbA1c recommendation.

In my patient, I strongly suspected nocturnal hypoglycemia with rebound hyperglycemia the following morning. I advised decreasing his basal insulin and requested middle of the night fingerstick blood glucose measurements so we could be sure that the problem had been addressed. However, at first his family was hesitant to reduce his insulin. They had been told that his target was an HbA1c of <6.5%. They were under the impression that with HbA1c, lower is better.

I explained that an HbA1c of <6.5% without hypoglycemia may indeed be the target for many people. However, in an 82-year-old man who lives alone and has other comorbidities, the consequences of hypoglycemia would be more serious than the benefits of preventing microvascular complications decades in the future. In someone such as this patient, an HbA1c of between 7% and 8% without hypoglycemia is reasonable.

Many diabetologists would also have decreased his insulin as I did, even before the recent studies raising questions about intensive glycemic control. HbA1c and other therapeutic targets must always be individualized to the patient. Our question now is what are and how do we determine the ideal goals for our patients across the spectrum of presentations we see in clinical practice? Further research is clearly needed. However, until the results of such research are available, we must not forget to use simple common sense and clinical experience when managing our patients.

Treat the patient, not just the numbers!