SHINE: Intense IV Glucose Control Does Not Improve Functional Stroke Outcomes

Deborah Brauser

February 07, 2019

HONOLULU — Intense intravenous (IV) glucose control is no better than using standard subcutaneous "sliding scale" insulin for improving functional outcomes after an ischemic stroke — and may even increase risk for hypoglycemia, new research suggests.

Primary results from the randomized, multicenter Stroke Hyperglycemia Insulin Network Effort (SHINE) study were presented at a late-breaking science session here at the International Stroke Conference (ISC) 2019.

Among more than 1100 hyperglycemic patients, those who received (within 12 hours of stroke symptom onset) up to 72 hours of continuous IV insulin infusion targeted to 80 to 130 mg/dL did not have greater functional improvement scores on the modified Rankin Scale (mRS) 90 days post-stroke compared with those who received subcutaneous insulin targeted to 80 to 179 mg/dL — the primary efficacy outcome.

In addition, 2.6% of the group receiving intense IV therapy had severe hypoglycemia during the treatment period vs none of those receiving standard care.

"We designed this trial so that no matter what the results were, we wanted to make sure that we answered the question and filled a gap in the stroke community's knowledge base in terms of how we should manage these patients," lead author Karen C. Johnston, MD, professor of neurology and associate vice president of Clinical and Translational Research at the University of Virginia in Charlottesville, told Medscape Medical News.

Dr Karen C. Johnston

"Now we know how to treat them. Our data very clearly show that we should not be using insulin infusions and treating patients to that tighter glucose control," she added. "This is not a negative trial because it efficiently answered the question of which of these two treatment protocols was the preferred therapy."

Worldwide Debate

Although increased glucose is common in patients who have experienced an acute ischemic stroke, management of this condition has been debated "worldwide," Johnston told meeting attendees.

"Preclinical/clinical data show that hyperglycemia during acute cerebral ischemia is associated with worse outcome, while severe hypoglycemia increases injury to the ischemic brain," she said. However, it's been "unclear if glucose-lowering improves outcomes."

The investigators enrolled 1151 adult patients at 63 sites in the United States from April 2012 to August 2018. All had received a clinical diagnosis of ischemic stroke and had either type 2 diabetes and glucose greater than 110 mg/dL, or had no known diabetes and glucose of 150 mg/dL or greater. They also had a baseline National Institutes of Health Stroke Scale (NIHSS) score that ranged between 3 and 22.

Main exclusion criteria included having type 1 diabetes or pre-existing confounding conditions.

The participants were randomly assigned to intensive IV treatment (n = 581; 55% men; 63% white, 31% black; mean age, 66 years) or to standard subcutaneous treatment (n = 570; 54% men; 65% white, 27% black; mean age, 66 years).

Previous ischemic stroke was reported for about 17% of each treatment group, type 2 diabetes mellitus was reported for about 80% of each group, and hypertension was reported by about 88% of each group. The median eligibility glucose level was 188 and 187 mg/dL, respectively, and the baseline NIHSS score was 7 in each group.

In addition, 93.3% and 91.9% of the groups, respectively, had an ischemic stroke right before treatment initiation, and 1.4% and 2.1% had a transient ischemic attack. The median time to treatment randomization was 7.1 hours for both groups.

"The study design required a maximum of 1400 enrollments and included 4 interim analyses for overwhelming efficacy or futility," the researchers report.

The primary efficacy measure was improvement from baseline on the modified Rankin Scale. The primary safety outcome was severe hypoglycemia, defined as less than 40 mg/dL, throughout the treatment period.

Key secondary measures included the 90-day NIHSS, the Barthel Index, and stroke-specific quality of life (SSQOL)

"Sweet Spot" Analyses

Results showed that the primary efficacy outcome did not differ significantly between the treatment groups (met by 20.5% of those receiving intensive therapy vs 21.6% of those receiving standard therapy). After adjusting for baseline stroke severity and use of thrombolysis, the adjusted relative risk [RR] for this outcome was 0.97 (95% confidence interval [CI], 0.87 – 1.08; P = .55).

Severe hypoglycemia occurred in 15 members of the intensive treatment group and zero members of the standard treatment group (risk difference, 2.58; 95% CI, 1.29 – 3.87).

A favorable NIHSS score at 90 days, defined as 0 or 1, was achieved by 43.7% and 44.7% of the groups, respectively (RR, 0.98); and a favorable Barthel Index score of 95 to 100 was achieved by 55.2% and 54.7% (RR, 1.01). Mean SSQOL scores were 3.8 and 3.7, respectively.

"We have answered the question of best glucose control for hyperglycemic acute ischemic stroke patients," Johnston said at the end of her presentation. "Our data suggest that subcutaneous insulin treatment with a target of less than 180 mg/dL is the preferred treatment."

Session moderator Louise McCullough, MD, PhD, professor and chair of neurology at UT Health and chief of neurology at Memorial Hermann Hospital System, Houston, Texas, agreed that the trial "answered a lot of unanswered questions" about how best to treat these patients.

"It contributes greatly to our knowledge," she told meeting attendees.

Asked by Medscape Medical News whether it could be the use of IV insulin alone that affected the outcomes, or the intensive glucose targeting range chosen, or a combination of both, Johnston said that the trial design "was based on the equipoise in the stroke community."

"Going to 80 to 130 with an insulin infusion was the only intensive therapy that our research community would accept. And going below 180 for the comparative group, based on the American Heart Association/American Stroke Association [AHA/ASA] guidelines, was the closest to a 'higher' glucose concentration," she said.

"That was kind of the 'sweet spot.' It was [clinical] equipoise across the country that helped us determine what the two groups should be," Johnston added. Although the trial provided clear results, "there's no doubt that there needs to be further research. If hyperglycemic ischemic stroke patients are having worse outcome and addressing their glucose acutely is not helping, what else can we do?"

Clear Evidence?

In video comments posted to the AHA/ASA website, Miguel Perez-Pinzon, PhD, chair of the ISC 2019 program committee and professor of neurology/neuroscience at the University of Miami, Florida, noted that there has been "a bit of a debate in the field" on how to control the hyperglycemia that ensues after a stroke.

"I think this study provides clear evidence to guide control of glucose levels in patients," said Perez-Pinzon, who also directs the Cerebral Vascular Disease Research Center at UM. In addition to not improving outcomes at 90 days compared with standard glucose control, and increasing the risk for hypoglycemia, the intense IV glucose therapy "requires more resources, such as increased supervision from nursing staff."

Johnston agrees. "For the stroke community across the country — the physicians, the nurses, and all the caretakers — offering that subcutaneous protocol is dramatically easier and uses far less resources. So unless the intensive treatment was dramatically better and clearly safe, it is not a good use of resources," she said.

Asked to comment, Karen Furie, MD, chair of the Stroke Council and chair of neurology at the Warren Alpert Medical School at Brown University, Providence, Rhode Island, told Medscape Medical News that "this is an important study" about what's been a controversial topic.

Dr Karen Furie

"I think the results will have a profound effect on management now. The strategy of very intensive therapy will fall by the wayside and people will be managing blood sugars a bit more liberally to achieve that 'less than 180 [mg/dL]' that was used in the trial," said Furie, who was not involved with the research.

She reiterated Johnston's comment that this was not a negative trial. "The concept that you have to show a benefit in the active arm is a fallacy. If the results of this study affect patient care and patient outcomes, then that study has tremendous value."

"There's a perception that the more aggressive you are, the better off the patient will be — and that's not always true, or may not be true in the absolute sense. There may be a threshold. So these trials about fundamental questions are really important," Furie said.

The study was funded by the National Institute of Neurological Disorders and Stroke (NINDS). Johnston reported having received modest honoraria from NINDS, the US Food and Drug Administration, the American Neurological Association, and the American Academy of Neurology. Johnston also reported being a consultant and/or serving on advisory boards for Biogen and Diffusion Pharm Inc. Furie has disclosed no relevant financial relationships. McCullough reported receiving a significant research grant from the National Institutes of Health.

International Stroke Conference (ISC) 2019: Abstract LB1.
Presented February 6, 2019.

Follow Deborah Brauser on Twitter: @MedscapeDeb

For more Medscape Neurology news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.