Abstract
In a relatively small sample of acute stroke patients treated with intravenous tissue plasminogen activator (IV-tPA), Meurer et al. (1) found no association between admission hyperglycemia and clinical outcomes (death, disability and intracerebral hemorrhage (ICH)). These results are counter to the conclusions of several other studies of similar size (2–4) and one larger prospective cohort study that found a clear association between admission hyperglycemia and an increased risk of death, symptomatic ICH (sICH) and disability outcomes (5). Thus, these results may simply represent a type II error (accepting the null-hypothesis when it should be rejected).
However, other explanations for these results should be considered. Only in-hospital death was reported, while other studies have generally examined mortality at 30 or 90 days. This would presumably lead to fewer outcome events in the current study and potentially underestimate the association between hyperglycemia and death, assuming that hyperglycemia is associated with an increased risk of death beyond the early phase after stroke. Similarly, considering disability at discharge rather than at a later time point may mask an eventual divergence in functional outcome between hyperglycemic and nonhyperglycemic patients over time. The use of the European Cooperative Acute Stroke Study definition of sICH (parenchymal hematoma type 2), resulted in a lower overall sICH rate compared with the use of the National Institute of Neurological Disorders and Stroke IV-tPA trial definition, and this may have diluted the association between glucose and sICH in this study.
The rate of overall ICH (9·9%) was surprisingly low as compared with other studies (2, 5) and may help explain the absence of an association between hyperglycemia and ICH. It is also possible that the population studied was comprised of milder strokes (median baseline National Institutes of Health Stroke Scale (NIHSS) is not provided) or a preponderance of lacunar strokes, a sub-type for which hyperglycemia may be less deleterious – or even beneficial (6). We found no evidence, on examining our own data, of effect modification of hyperglycemia on poor outcomes by baseline NIHSS score or by the Oxfordshire Community Stroke Project stroke type. Akin to the Meurer study, one other group also found no association between isolated admission hyperglycemia and outcome. In that study, among a mixed cohort of thrombolysed and nonthrombolysed patients, only persistent hyperglycemia at 24 h was associated with worse outcome. Unfortunately, 24 h glucose values were not available in the Meurer study.
We believe that the association between admission hyperglycemia and worse outcome in acute stroke patients treated with IV-tPA is well established and real. The more important question is whether aggressively treating hyperglycemia can improve patient outcomes.