Evidence for an Interlude

Hyperglycemia (HG), a common phenomenon in all types of acute strokes, is increasingly considered as a potential therapeutic target in ischemic stroke because there is now strong evidence that high glucose levels are independent predictors of larger infarct size, poor clinical outcome, and higher risk of mortality.1 In the past few years this has led many acute stroke centers to initiate intensive insulin therapy (IIT) policies often modeled on intensive care unit practices. Unfortunately, the initial enthusiasm for IIT in the intensive care unit has disappeared because currently available evidence-based data fail to identify any clinical benefit at the time of continuing to outline the high risk of hypoglycemia.2 Furthermore, in the neurocritical care setting, the increased frequency of hypoglycemia may result in higher mortality.3 Finally, the UK Glucose Insulin in Stroke Trial (GIST-UK) failed to demonstrate any benefit from IIT in 933 patients with stroke.4 Five other small randomized trials were not powered to demonstrate a clinical benefit, but all showed that IIT induced a high risk of hypoglycemia in patients with acute stroke.5–10 It may be time for an interlude. As highlighted by others, there is a need for safer methods of improving glucose control before launching large randomized trials.1,11 We will also argue here that the proper design of large trials may require further experimental work and proof-of-concept human studies, which in turn may benefit from some results obtained in the past decade. It should be stressed that HG is a complex phenomenon in acute stroke and may result from known diabetes or undiagnosed diabetes, metabolic syndrome, acquired insulin resistance, stress response, and lesion size or its location. Furthermore, despite many theories, the mechanism of HG toxicity in acute stroke remains to be clearly elucidated.