A condition known as stress hyperglycemia, sometimes called “stress diabetes,” can occur temporarily after a heart attack or stroke. Even in non-diabetic patients, blood glucose levels may spike during recovery. Such spikes are problematic: People diagnosed with acute stress hyperglycemia are more likely to die following a stroke or heart attack.

Now, we know they’re also at risk for a secondary stroke within several months after the first. In a study published Thursday in Stroke, a team of Chinese and U.S. researchers found that stroke patients who dealt with serious hyperglycemia were 1.5 times more likely to have a secondary stroke than those with minor or no hyperglycemia.

Knowing this risk of recurrence may eventually help doctors provide appropriate care for at-risk patients, said the study’s corresponding author Yilong Wang, M.D., who is a researcher and clinician with the Capital Medical University in Beijing, China. “But it cannot lead to a change in recommended standard procedures at this time,” Wang said.

The study is particularly interesting, if not yet definitive, largely because of the novel way that Wang and his coauthors defined stress hyperglycemia, said an expert who was not involved in the study.

“The uniqueness of this study is the use of a not tremendously well-founded biomarker, the glucose-to-glycated-albumin ratio,” said endocrinologist Robert Eckel, M.D., endowed chair and professor at the University of Colorado Hospital.

“This is an interesting concept. It’s eyebrow-raising,” he said. “It makes this study worthy of attention … Are we missing something?”

Stress hyperglycemia is an exceedingly complicated response to underlying illness, involving what Wang called a “derangement” of the complex choreography of hormones and immune system chemicals called cytokines.

It’s unclear even how many people suffer such a response following an initial stroke, he said. Prevalence estimates range from 8 percent to 63 percent, according to studies that Wang and his colleagues cited. The range of those figures may reflect the fact that it can be a challenge to distinguish stress hyperglycemia from undiagnosed diabetes or prediabetes.

So the Chinese-led team — which included coauthors from the University of Illinois College of Medicine and Dell Medical School at the University of Texas at Austin — diagnosed stress hyperglycemia severity with a ratio of fasting glucose levels to levels of glycated albumin. The latter reflects a person’s average glucose concentration from the past two to four weeks.

This “relative” measure of current hyperglycemia seemed to be a better predictor of another stroke than an absolute measure. Such a predictor would be valuable to intensive care and other doctors, who can try to anticipate likely risks to patients dealing with hyperglycemia.

Yet previous research shows intensive insulin control isn’t as effective as people assumed at reducing the risk of dying or developing other health problems. So medical societies now recommend looser control that should be modified based on the individual patient’s circumstances.

The new study included more than 3,000 patients who had a minor ischemic stroke or transient ischemic attack. The team looked at stroke recurrence within three months, and found that about 12 percent of patients with the highest ratios of glucose-to-glycated albumin had a secondary stroke compared to only about 9 percent of those with the lowest ratios.

Wang agreed with Eckel that results shouldn’t be over-interpreted yet. His team’s ratio approach, comparing highest to lowest measures, isn’t finely tuned enough — yet.

“This is interesting, but there’s a little equivocation for me,” Eckel said. “If this measure is applied to some larger groups, does it have generalizability? Could it predict things like future onset of diabetes? It deserves some more attention.”

If you have questions or comments about this story, please email