Off-label thrombolysis for acute ischemic stroke: rate, clinical outcome and safety are influenced by the definition of 'minor stroke'

Background: Several contraindications for intravenous thrombolysis are not based on controlled trials. Specialized stroke centers often apply less restrictive criteria. The aim of our study was to analyze how many patients at our institution receive off-label thrombolysis. In addition, clinical outcome and safety data were compared to those from patients treated on-label, and the influence of different definitions of 'minor stroke' were examined.

Methods: Consecutive thrombolysis patients treated between January 2006 and January 2010 were included. Patients treated off-label were compared to patients given on-label therapy according to the European license. Since no specified definition for 'minor neurological deficit' exists in the license, two distinct definitions were considered off-label, i.e. National Institutes of Health Stroke Scale score (NIHSSS) <1 (definition 1) and NIHSSS ≤4 (definition 2).

Results: Of a total of 422 patients, 232 (55%) were treated off-label. The most prevalent off-label criteria (OLCs) were the following: age >80 years (n = 113), minor stroke (definition 1, n = 3; definition 2, n = 84), elevated blood pressure necessitating aggressive treatment (n = 75), time window >3 h (n = 71) and major surgery or trauma within the preceding 3 months (n = 20). In group comparisons, off-label patients had an overall worse outcome using definition 1 for minor stroke, while there was no difference when definition 2 was applied. In multivariate analysis, off-label therapy (definition 1) in general and age >80 years were independent predictors of poor outcome. None of the contraindications were associated with an increased bleeding risk.

Conclusions: Off-label therapy is frequently applied at our center and is not associated with higher complication rates. Overall outcome of off-label treatment largely depends on the definition used for minor stroke. Besides age >80 years, a known poor prognostic factor, no other specific OLC was associated with poor outcome. Our data suggest that the criteria in the European license may be too restrictive.