Subscribe to eTOC

A Debate Question Answered: Intensive Glucose Control Does Not Improve Outcomes in Stroke

Article In Brief

Intensive glucose control did not improve 90-day Functional outcomes in patients with acute ischemic stroke and hyperglycemia.

A much-anticipated study may finally settle the debate over whether tight glucose control during the acute phase of ischemic stroke improves longer-term outcomes for patients.

The multicenter trial randomized ischemic stroke patients to either intensive or standard glucose control and found no differences between the two groups in terms of favorable functional outcome at 90 days. The study also found that stroke patients on intensive glucose control were more likely to experience hypoglycemia, which can be dangerous.

The study, published in the July 23/30 issue of the Journal of the American Medical Association, is bound to change the practice of some stroke doctors who believe it makes good sense to keep glucose levels tightly under control to help limit brain damage, experts said.

“When we started this study, we found there was great variability across the country in glucose control during stroke,” said the study's main author, Karen C. Johnston, MD, MSc, the Harrison Distinguished professor of neurology at the University of Virginia.

“Some doctors were so sure that tight, intensive control was the best thing, they were hesitant to participate in the study,” she said. “Other stroke doctors felt equally strongly that more relaxed glucose management was best,” said Dr. Johnston. “For decades, there has been this uncertainty and ultimately providers across the country agreed there was equipoise, so they participated.”

Dr. Johnston said the new evidence that there is no advantage to intensive glucose control during acute ischemic stroke, and that there is potential for harm, will likely bring change to current stroke treatment guidelines issued by the American Heart Association/American Stroke Association, which note there is not adequate data to make a strong recommendation on glucose control.

“When it is time to update, they will be able to say we have clear evidence now that while we should be managing glucose levels, aggressive glucose control is not the preferred therapy,” said Dr. Johnston, who was the lead investigator for the new trial known as SHINE (Stroke Hyperglycemia Insulin Network Effort).

James C. Grotta, MD, FAAN, who was not involved with the study, agreed that the findings will influence clinical practice. “It was a very important study, it was well done, and it came up with a very conclusive answer,” said Dr. Grotta, director of stroke research and the mobile stroke unit at Memorial Hermann Hospital in Houston.

The Trial's Purpose

“Hyperglycemia is present in about 40 percent of stroke patients and is associated with worse clinical outcomes, including greater infarct growth and hemorrhagic stroke conversion,” the study authors noted. And “data suggest that hyperglycemia during acute brain ischemia augments the ischemic injury by multiple potential mechanisms, such endothelial dysfunction, increased oxidative stress, and impaired fibrinolysis,” they wrote.

It seems to intuitively make sense that strictly controlling glucose during the acute phase of ischemic stroke would be advantageous, but clinical research on the issue produced mixed and inconclusive results.


“The beauty of the study was that it was designed to answer a pretty controversial question that has plagued our (stroke) community for years.”—DR. KOTO ISHIDA


“Some doctors were so sure that tight, intensive control was the best thing, they were hesitant to participate” in the study. Other stroke doctors felt equally strongly that more relaxed glucose management was best.”—DR. KAREN C. JOHNSTON

The 2018 ASA/AHA ischemic stroke treatment guidelines suggest treating hyperglycemia to achieve a blood glucose range of 140 to 180 mg/dL and close monitoring to prevent hypoglycemia, the study authors noted. But they said the guideline characterized the strength of evidence behind the recommendation as being supported by limited data.

The Study Design, Findings

The SHINE trial, funded by the National Institutes of Health, included 1,151 patients treated between April 2012 and August 2018 at 63 sites around the US. One group was randomized to receive continuous intravenous insulin using a computerized decision support tool that kept glucose levels at 80 to 130 mg/dL for up to 72 hours. The other group received standard glucose control, which involved subcutaneous insulin administered on a sliding scale to keep insulin in the range of 80-179 mg/dL.

To be eligible, patients had to have hyperglycemia (defined as greater than 110 mg/dL for those with type 2 diabetes and greater than 150 mg/dL for those without diabetes), an acute ischemic stroke, a National Institutes of Health Stroke Score (NIHSS) of 3 to 22 (on a scale of 0 to 42, with higher scores indicative of greater neurological deficits), and present within 12 hours of stroke onset. Patients were ineligible to enroll if they had type 1 diabetes, required kidney dialysis, had a clinical indication for insulin infusion, or had a condition that would confound assessment of the stroke clinical outcome.

The mean age of study participants was 66.6 years. Of the total cohort, 80 percent had a diagnosis of diabetes.

The primary efficacy outcome was the proportion of patients with a favorable outcome at 90 days after randomization. Favorable outcome was defined as a modified Rankin Scale (mRS) score of 0 in patients with a baseline NIHSS score of 3 to 7; a score of 0 to 1 in patients who had a baseline NIHSS score of 8 to 14; a score of 0 to 2 in patients who had a baseline NIHSS score of 15 to 22. (The mRS scores range from 0, for no symptoms, to 6, death.)

A favorable outcome occurred in 119 of 581 patients (20.5 percent) in the intensive treatment group and in 123 of 570 patients (21.6 percent) in the standard treatment group. The difference was not statistically different.

During treatment the mean blood glucose level was 118 mg/dL in the intensive treatment group and 179 mg/dL in the standard treatment group.

Treatment was stopped early for hypoglycemia or other adverse events in 65 patients (11.2 percent) in the intensive treatment group and in 18 patients (3.2 percent) in the standard treatment group. Severe hypoglycemia occurred in 15 patients (2.6 percent) in the intensive group and not at all in the standard group.


NIHSS indicates National Institutes of Health Stroke Scale. The modified Rankin Scale is a global stroke disability scale with scores ranging from 0 (no symptoms or completely recovered) to 6 (death). Each cell corresponds to a score on the modified Rankin Scale; the width of the cell indicates the proportion of patients with equivalent scores, and the percentage of patients is shown within the cell. The diagonal line between the two study groups indicates the dichotomization of favorable outcome in each severity stratum.

“Among patients with acute ischemic stroke and hyperglycemia, treatment with intensive glucose control for up to 72 hours did not result in a significant difference in favorable functional outcome at 90 days,” the researchers concluded. “The findings do not support using intensive glucose control in this setting.”

The researchers noted that the study had limitations, including the fact that 42 percent of the patients were enrolled at six sites, which meant there could have been an influence on the overall outcome numbers because of site-specific practices.


“It was a very important study, it was well done, and it came up with a very conclusive answer.”—DR. JAMES C. GROTTA

Also, treatment with intravenous tissue plasma activator therapy (tPA) in 63 percent of patients suggests a selection bias for trial enrollment toward patients who received a higher level of stroke care.

The trial also did not capture recanalization data. Previous data suggest that hyperglycemia may impede recanalization, the study noted. Another caveat: The intensive treatment group had more frequent glucose checks than the standard group, which could have resulted in more reported cases of hypoglycemia.

Dr. Johnston said she expected that it would be welcome news for stroke doctors that “they don't need to be chasing those high glucose levels in the acute setting.”

She said taking a step back from aggressive glucose control should also save on patient care resources, since it takes time and staffing to operate the equipment used to monitor and administer intravenous insulin.

“That is good news for the health care system because it (standard glucose control) is likely to be safer for the patient and saves on costs for the health care system,” she said, though the study did not measure costs.

Expert Commentary

Dr. Grotta said the new study is a good example of why it is important to rigorously test medical assumptions.

“We know that having a high blood sugar is bad, but we didn't know whether treating that high blood sugar would (in the acute stroke phase) lead to better outcomes,” he said.

He said that doing aggressive glucose control becomes “another thing to take care of,” for nurses and other staff, and likely requires admission to a stroke unit (though he does favor stroke units).

Dr. Grotta said neurologists still need to be attentive to high blood sugar in their patients since it a risk for cardiovascular disease and stroke.

“Many patients who have stroke are found to be diabetic and that was not appreciated beforehand,” he said.

Koto Ishida, MD, associate professor of neurology and medical director of the stroke program at NYU Langone Health, said, “The beauty of the study was that it was designed to answer a pretty controversial question that has plagued our (stroke) community for years.”

The “study helps us differentiate between these two discrete treatment regimens for the first time,” she said.

She thought the results were convincing, but she's not sure the debate over tight versus looser glucose control during stroke treatment is truly over. She noted that more research might show, for instance, that a more aggressive approach is advantageous in subsets of these high-risk patients or that the timing of treatment may be a factor. She also said there is much more to be learned about the connection between blood glucose and stroke at the physiologic level.

Dr. Ishida said it is just as important to pay attention to negative study findings, that something isn't beneficial, as it is for positive ones.

“More aggressive treatment is not always better,” she said, something that has been noted in other areas such as blood control.


Dr. Johnston reported serving on the National Institute of Neurological Disorders and Stroke advisory council and the US Food and Drug Administration neurological devices panel. Dr. Grotta serves on the scientific advisory board of Haemonetics Corp., has received funding for travel or speaker honoraria from Stryker Corp, and consults with Frazer, Ltd and Stryker Corporation; he has received research support from Genentech and Behring. Dr. Ishida had no disclosures.

Link Up for More Information

• Johnston KC, Askiel B, Paul Q, et al Intensive vs. standard treatment of hyperglycemia and functional outcome in patients with acute ischemic stroke: The SHINE randomized clinical trial JAMA 2019;322(4):326–335.