Secondary Logo

Neurology Today At the Meetings: International Stroke Conference

Access daily, concise peer-reviewed reports from the International Stroke Conference selected by the Neurology Today editors.

Tuesday, February 12, 2019

Intravenous Insulin Fails to Improve Recovery of Diabetic Stroke Patients


HONOLULU—Lowering glucose levels in diabetic patients who had acute ischemic strokes by delivering insulin intravenously rather than by standard subcutaneous injections failed to provide a benefit in functional outcomes at 90 days, researchers reported here at the 2019 International Stroke Conference, sponsored by the American Stroke Association.

Researchers from 70 sites in the United States participated in the prospective, randomized, blinded trial, the Stroke Hyperglycemia Insulin Network Effort (SHINE): 581 patients were randomized to intensive therapy and 570 to standard treatment. Patients in the intensive group with hyperglycemia were randomized to receive intensive glucose lowering using an insulin, intravenous drip until glucose reached a target level of  80-130 mg/dL. In the standard treatment group, the goal of therapy was to lower glucose below 180 mg/dL.

In the intensive treatment group who were given intravenous insulin, 20.5 percent of the 581 patients achieved a score of 0–2 on the modified Rankin Scale based on a sliding scale dependent upon baseline scores when they were admitted to treatment compared with 21.6 percent of patients who underwent standard treatment (p=0.55), reported Karen C. Johnston, MD, MSc, professor of neurology at the University of Virginia in Charlottesville.

In her oral late-breaker presentation, Dr. Johnston explained that patients with mild strokes—defined as by a National Institute of Health Stroke Scale (NIHSS) of 3–7—were counted as having achieved the primary endpoint if their 90-day modified Rankin Scale score (mRS) was 0. Those with moderate stroke with an NIHSS score of 8–14 achieved the positive primary endpoint if their 90-day mRS was 0–1; in severe stroke patients, those with NIHSS scores of 15–22, success was measured if they achieved an mRS score of 0–2.

Dr. Johnston also noted that among the intravenous insulin patients there was a 2.6 percent rate of severe hypoglycemia, defined as blood glucose less than 40 mg/dL compared with no cases of hypoglycemia among the patients treated in the standard care arm of the trial [RR 2.58 (1.29-3.870], a significant difference.

The treatment was successful in lowering glucose, Dr. Johnston said, with the intensive therapy group lowering glucose to a mean of 118 mg/dL, while the standard care group lowered their glucose to a mean of 179 mg/dL. But success in lowering glucose did not translate to better outcomes.

"Intensive glucose control does not improve 90-day functional outcomes and increases the risk of hypoglycemia," Dr. Johnston said. She suggested that the study did offer insights on the glycemic goal that should be set for diabetic patients with a stroke who were out of glycemic control or other patients who had hyperglycemia at baseline. "Subcutaneous insulin with a target of less than 180 mg/dL is preferred," she said.

Commenting on the presentation, Philip B. Gorelick, MD, MPH, FAAN, chief medical officer at Thorek Memorial Hospital in Chicago, told Neurology Today At the Meetings: "The purpose of SHINE was to see if lowering the glucose was going to be beneficial in patients with acute ischemic stroke by administering intravenous insulin compared with patients for whom standard care involved subcutaneous insulin on a sliding scale. Unfortunately they found that the intravenous insulin study—or intensive glucose therapy as it was called—did not improve the outcomes at 90 days, which is our standard outcome measure for ischemic stroke patients."

Dr. Gorelick noted that previous studies in other conditions, often performed in the intensive care setting, either had neutral or negative results. "We have to realize that people with diabetes or hyperglycemia tend to have more damage to the brain when there is stroke, and their outcomes tend to be worse," he said. "So it might make sense to try and reverse this damage by having more intensive glucose therapy and compare it with a standard subcutaneous glucose treatment. But it is not likely that in a three- to four-day period of an acute stroke that intensive glucose control will reverse conditions that occur in diabetics that include a higher coagulability status of the blood," he said.

In addition, Dr. Gorelick said, "There is an area surrounding the core infarct, the penumbra, which may need glucose. So if the penumbra is not getting enough glucose, the patient could worsen. If you are using intravenous insulin there is also a risk you may be making these patients hypoglycemic, which can be equally deleterious to the brain."

He suggested that the glucose hypothesis in stroke might spell the end for this area of inquiry. "Given the past work in the intensive care unit, and this work, it is going to be highly unlikely that we are going to be going forward with any tight glucose control regimens in the acute stroke area."

The National Institute of Neurological Disorders and Stroke (NINDS) funded the study. Dr. Johnston has received honoraria from the NINDS and has served as a consultant for and on the data safety and monitoring board for Biogen and Diffusion Pharmaceuticals Inc.


ISC Abstract LB1: Johnston KC, Bruno A, Barrett KM, et al. Stroke Hyperglycemia Insulin Network Effort (SHINE) trial primary results.