Hyperglycemia is a risk factor for stroke and augurs a poor outcome in the aftermath of stroke; however, treating patients with post-stroke hyperglycemia (PSH) with infusions of glucose, potassium, and insulin (GKI) to achieve euglycemia does not provide a survival benefit.
Hyperglycemia is a risk factor for stroke and augurs a poor outcome in the aftermath of stroke; however, treating patients with post-stroke hyperglycemia (PSH) with infusions of glucose, potassium, and insulin (GKI) to achieve euglycemia does not provide a survival benefit. Furthermore, hyperglycemic stroke patients are at risk for symptomatic intracerebral hemorrhage (sICH). Nevertheless, recombinant tissue plasminogen activator (rtPA) may provide more benefit than risk in these patients, although the bottom line is that despite interventions, hyperglycemic patients can be expected to fare worse than normoglycemic patients. These were the findings of studies presented at the International Stroke Conference (ISC), which was held February 7 to 9 in San Francisco.
Irrespective of hyperglycemic status pre-stroke, hyperglycemia appears to be common immediately post-stroke, with moderately elevated glucose levels occurring in the first 6 hours of the index event and then resolving.1,2 Given that hyperglycemia is strongly associated with poor outcome following acute stroke, should PSH be treated? Final results from the United Kingdom Glucose Insulin Stroke Trial (GIST-UK) suggest that the answer is no. The trial determined that induction of euglycemia via GKI infusion does not provide a survival benefit in acute stroke patients in whom hyperglycemia develops.3 Although GKI was associated with a significant reduction in blood glucose levels (mean 0.57 mmol/L within 24 hours), outcome (incidence of death at 90 days) was similar between treatment and control groups.
Intensive intervention for hyperglycemia in patients following myocardial infarction and other acute critical illnesses has proved beneficial, noted study presenter Christopher S. Gray, MD, professor of clinical geriatrics at the University of Newcastle Upon Tyne, in explaining the study's rationale. "Although hyperglycemia has repeatedly been reported to be associated with poor outcome, the risk and benefits of glucose lowering in stroke is unknown," he said. The GIST-UK is the first randomized, placebo-controlled, multicenter clinical trial to look at the effect of therapeutic blood glucose lowering after acute stroke.
The study population consisted of 933 persons who had experienced cerebral infarction or ICH and had plasma glucose levels of 6.0 to 17.0 mmol/L within 24 hours of symptom onset. Patients were randomly selected to receive either a continuous infusion of GKI (72 to 126 g/dL to maintain capillary glucose levels of 4 to 7 mmol/L and avoid hypoglycemia) or saline for 24 hours. "The 2 groups were extremely well-matched," remarked Gray.
The primary end point was all-cause death at 90 days and the secondary end point was avoidance of severe disability. At 90 days, 30% of the 464 patients in the GKI group had died and 27.3% of the 469 patients in the control group had died. "We saw no difference in mortality or time to death and no difference in avoidance of severe disability between groups," said Gray.
Gray concluded that evidence does not support aggressive intervention for PSH although he asserted that the trial provided important information on the management of PSH. "We found that the majority of patients had only moderate elevation in blood glucose-a mean 8.4 mmol/L. Glucose levels fall spontaneously with saline hydration," noted Gray.
HYPERGLYCEMIA LINKED TO POORER OUTCOME
A study that drew on data from the Virtual International Stroke Trials Archive confirmed the findings of smaller single-center trials that linked PSH (defined in the study as blood glucose levels in excess of 7.0 mmol/L) with higher mortality rates and poorer functional outcomes.4 The study, which was reported by Keith W. Muir, MD, MRCP, professor of neurology at the University of Glasgow, United Kingdom, looked at 2645 patients who received medical attention-including evaluation of blood glucose levels-at a median of 5.5 hours after stroke onset. PSH was present in 1126 (42.6%) patients on hospital admission and in 1421 (53.7%) patients within the first 48 hours after admission. PSH developed in 19.4% of patients who were normoglycemic before stroke occurred.
Hyperglycemic patients who received rtPA had better outcomes than those who did not receive rtPA; however, the mortality rate was higher and functional outcome was poorer at 90 days post-stroke in these patients compared with normoglycemic patients. Furthermore, post-stroke blood glucose levels had a significantly greater impact on outcomes at 90 days than did use or non-use of rtPA. The odds of having a poor outcome increased by 44%, reported Muir.
HYPERGLYCEMIA AND ICH RISK
Another study,5 presented by David M. Kent, MD, associate professor in the Department of Clinical Research at the Sackler School for Graduate Biomedical Studies, Tufts-New England Medical Center, Boston, suggested that hyperglycemia is a strong risk factor for stroke-associated sICH regardless of whether rtPA is administered for treatment of acute stroke. Furthermore, Kent and colleagues found that administration of rtPA does not modify risk of sICH in hyperglycemic patients. Some clinicians have argued that rtPA should be contraindicated in hyperglycemic patients because it might enhance risk of sICH.
To get a definitive picture of glycemic status and outcomes in terms of sICH occurrence in patients with PSH, Kent and colleagues analyzed data from the National Institute of Neurological Disorders and Stroke tPA study trials 1 and 2,6 the Alteplase Thrombolysis for Acute Noninterventional Therapy in Ischemic Stroke A and B study,7 and the European Cooperative Acute Stroke Study 2.8 These randomized clinical trials used the current recommended dose of tPA within a 6-hour post-stroke window.
Hyperglycemia strongly correlated with poor outcome. Indeed, the risk of sICH increased in relation to increased blood glucose levels. Outcome was related to glycemic status regardless of whether the patient received rtPA.
Animal models have shown that brain injury worsens in the presence of hyperglycemia, Kent explained. Clinical studies also have shown that hyperglycemic survivors of acute stroke have early neurological worsening, poorer long-term outcomes, lack of improvement, higher rates of sICH, and lower rates of lesion recanalization. These studies, however, have been based on nonrandomized, observational data, according to Kent. "The aim of our study was to explore the effects of hyperglycemia on outcome, complications, and the treatment effect of tPA using a database of pooled randomized clinical trials," he explained.
The study included 2162 stroke patients; 1090 received rtPA and 1072 did not (controls). Data on patient age, NIH Stroke Scale severity, rtPA treatment, systolic blood pressure, history of diabetes, presence of atrial fibrillation, symptom onset to treatment time, and onset to treatment time multiplied by rtPA treatment were collected. Interactions between the treatment effects and symptom onset to treatment time also were noted.
The results were consistent between hyperglycemic and normoglycemic patients and among those treated with rtPA before and after 3 hours of symptom onset. The proportion of poor outcomes was similar in both groups; the rate of thrombolytic-related sICH was about 8%.
An increase in baseline serum glucose level of 1 mmol/L was associated with a 6% increase in poor outcome. In the group that was treated with thrombolytic therapy, a baseline serum glucose level of 1 mmol/L was associated with a 7% increase in sICH.
"These findings basically were in agreement with the previous results showing that outcomes in these patients are worse. Hyperglycemia seems to have an independent effect on that outcome," Kent said. However, he noted that hyperglycemia was not shown to modify the treatment effects of rtPA. "It really didn't matter what the patient's glucose level was, whether 5 mmol/L to 9.5 mmol/L: everybody benefited from treatment within the 0- to 3-hour window," he said. "And the impact of hyperglycemia on the treatment effect is very modest at best."
Although his team's findings do not contradict earlier studies about the effects of hyperglycemia in stroke, the interpretation of the findings differ, Kent said. "Increasing baseline serum glucose is associated with worse outcomes and more complications, which appears to be independent of stroke severity and other prognostic factors. Changes in serum glucose do not appear to modify the treatment effect of rtPA. Hyperglycemia should not be considered a relative contraindication to therapy. Lastly, caution should be used when interpreting the results of uncontrolled cohort analyses exploring the influence of prognostic factors on the effects of therapy."
Christensen H, Boysen G. Blood glucose increases early after stroke onset: a study on serial measurements of blood glucose in acute stroke.
Eur J Neurol.
Gray CS. The United Kingdom Glucose Insulin in Stroke Trial (GIST-UK) final results. Presented at: the International Stroke Conference 2007; February 7-9, 2007; San Francisco.
Gray CS, Hildreth AJ, Alberti GK, O'Connell JE. Poststroke hyperglycemia: natural history and immediate management.
2004;35:122-126 [Epub 2003 Dec 11].
Muir KW, McCormick M. Hyperglycemia in acute stroke trials: prevalence, predictors, and prognostic value: an analysis of the Virtual International Stroke Trials Archive (VISTA).
Kent DM, Selker HP, Ruthazer R, et al. Glycemic index predicts poor prognosis and risk of thrombolytic-related symptomatic intracerebral hemorrhage but does not appear to modify the effectiveness of rtPA therapy.
Tissue plasminogen activator for acute ischemic stroke. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group.
N Engl J Med.
Clark WM, Wissman S, Albers GW, et al. Recombinant tissue-type plasminogen activator (Alteplase) for ischemic stroke 3 to 5 hours after symptom onset. The ATLANTIS Study: a randomized controlled trial. Alteplase Thrombolysis for Acute Noninterventional Therapy in Ischemic Stroke.
Hacke W, Kaste M, Fieschi C, et al. Randomised double-blind placebo-controlled trial of thrombolytic therapy with intravenous alteplase in acute ischaemic stroke (ECASS II). Second European-Australasian Acute Stroke Study Investigators.