Why Elevated Cortisol Doesn’t Always Raise Insulin

During a live Q&A session this last week, one of the questions was from Catherine V., a practicing rheumatologist. I thought it was quite fascinating and I’d like to share it.

Catherine’s question centered on her observation of patients with normal fasting insulin levels (around 7 μIU/mL) yet elevated HbA1c levels, particularly in the prediabetic range (mid-to-high 5s), coupled with high cortisol levels. She noted that these cases seemed more hormonally driven than diet-driven, as elevated cortisol didn’t appear to cause a significant rise in insulin. She asked, “Does this observation make sense?”

First, I applaud Catherine for her astute observation! If more endocrinologists approached their work like rheumatologists often do, I think my job would be a little easier! Her observation about the interplay between cortisol, HbA1c, and insulin levels deserves attention.

Let’s break this down. HbA1c is indeed a marker of glucose levels, but it’s also influenced by the lifespan of red blood cells. Longer-lived red blood cells have more time to become glycated, potentially resulting in an elevated HbA1c level even when glucose levels are normal. This can sometimes create a false positive for elevated glucose.

Another aspect to consider is timing. HbA1c reflects an average glucose level over approximately three months, while fasting insulin and cortisol levels can vary significantly throughout the day. Many of these blood tests are done in the morning, a time when cortisol levels naturally peak, and the body tends to be more insulin-resistant. This could partly explain why cortisol is elevated without triggering a corresponding rise in insulin.

High cortisol levels activate gluconeogenesis—the body’s production of glucose—which may elevate glucose but not necessarily to a level requiring significant insulin secretion. However, if cortisol levels persistently rise, it’s reasonable to expect insulin would eventually follow suit. For some individuals, though, even a fasting insulin level of 7 μIU/mL, which is slightly above my ideal cutoff of 6, may not be optimal. These individuals might be more insulin-sensitive at levels closer to 4 or even 3.

To assess insulin resistance further, I recommend looking at additional metrics, such as the triglyceride-to-HDL ratio. If this ratio is above 1.5 (or even in the 2s), it provides more evidence of underlying insulin resistance. Exploring these complementary markers can offer a fuller picture of the patient’s metabolic health.

I’m grateful to Catherine for her insightful question. It’s discussions like these that remind me how much we can learn from each other as we strive to better understand the complexities of human metabolism.

Warm regards,

Ben