BACKGROUND: Despite suggestions that glucose levels rise after stroke before falling within a few hours, the natural history and determinants of this phenomenon remain unclear. We aimed to better characterize the time course of changes in glucose levels after ischemic stroke and to identify factors that affect poststroke glycemia. METHODS: Patients with ischemic stroke without previously diagnosed diabetes had blood glucose measured at least 4-hourly until 48 hours poststroke. The relationship between baseline factors, such as the NIH Stroke Scale, and blood glucose was assessed with mixed-effects models. The behavior of glucose over time was modeled in the whole cohort, and for the cohort partitioned into two around an admission glucose of 6.0 mmol/L. RESULTS: In the cohort of 124 patients the mean glucose was 6.6 mmol/L throughout the period of monitoring, with no change over time. Mixed-effects models identified more severe stroke and glucose-lowering therapy to be associated with higher poststroke glucose levels. When the cohort was partitioned, the mean glucose of those below 6.0 mmol/L at admission increased and the mean glucose of those above 6.0 mmol/L at admission decreased to the overall mean. CONCLUSIONS: Mean glucose levels remain static in patients with ischemic stroke without diabetes until at least 48 hours poststroke. Serial glucose levels are higher in patients with more severe stroke. Initially high or low mean glucose recordings exhibit regression to the mean over time, a change which may merely be a statistical phenomenon without necessarily indicating resolution of abnormal glycemia.
BACKGROUND: Despite suggestions that glucose levels rise after stroke before falling within a few hours, the natural history and determinants of this phenomenon remain unclear. We aimed to better characterize the time course of changes in glucose levels after ischemic stroke and to identify factors that affect poststroke glycemia. METHODS:Patients with ischemic stroke without previously diagnosed diabetes had blood glucose measured at least 4-hourly until 48 hours poststroke. The relationship between baseline factors, such as the NIH Stroke Scale, and blood glucose was assessed with mixed-effects models. The behavior of glucose over time was modeled in the whole cohort, and for the cohort partitioned into two around an admission glucose of 6.0 mmol/L. RESULTS: In the cohort of 124 patients the mean glucose was 6.6 mmol/L throughout the period of monitoring, with no change over time. Mixed-effects models identified more severe stroke and glucose-lowering therapy to be associated with higher poststroke glucose levels. When the cohort was partitioned, the mean glucose of those below 6.0 mmol/L at admission increased and the mean glucose of those above 6.0 mmol/L at admission decreased to the overall mean. CONCLUSIONS: Mean glucose levels remain static in patients with ischemic stroke without diabetes until at least 48 hours poststroke. Serial glucose levels are higher in patients with more severe stroke. Initially high or low mean glucose recordings exhibit regression to the mean over time, a change which may merely be a statistical phenomenon without necessarily indicating resolution of abnormal glycemia.
Authors: N D Kruyt; G J Biessels; T M Vriesendorp; J H Devries; J B L Hoekstra; P W Elbers; L J Kappelle; P Portegies; M Vermeulen; Y B W E M Roos Journal: Neurocrit Care Date: 2009-05-27 Impact factor: 3.210
Authors: Askiel Bruno; Valerie L Durkalski; Christiana E Hall; Rattan Juneja; William G Barsan; Scott Janis; William J Meurer; Amy Fansler; Karen C Johnston Journal: Int J Stroke Date: 2013-03-19 Impact factor: 5.266