Dosing Errors Did Not Have a Major Impact on Outcome in the NINDS t-PA Stroke Study

Abstract

Intravenous tissue plasminogen activator (IV t-PA) for acute stroke is dosed by weight, which is typically estimated, and requires a multistep infusion, which may lead to administration errors. We evaluated the impact of dosing errors on outcome in the National Institute of Neurological Disorders and Stroke (NINDS) t-PA Stroke Study. Using logistic regression, we investigated the relationship between actual weight-based dose (administered dose divided by actual weight, capped at 100 kg) and intracerebral hemorrhage (ICH) at 36 hours and likelihood of good recovery (modified Rankin score of 0 or 1) at 3 months. Estimated weight, actual weight, and total t-PA dose administered were available for 311 of 312 patients; 3 received only the bolus and were excluded. In multivariable logistic regression, there was no association between the risk of any ICH using actual dose as a continuous variable (P = .65), or when comparing the highest quintile of actual dose (>0.944 mg/kg) to the lower 4 quintiles (9.8% vs 10.9%; odds ratio [OR], 0.60; 95% confidence interval [CI], 0.22-1.61; P = .31), adjusting for age, baseline National Institutes of Health Stroke Scale (NIHSS) score, and changes on baseline computed tomography. Similarly, there was no association between likelihood of good outcome and actual dose as a continuous variable (P=.87), or when comparing the lowest quintile (<0.879 mg/kg) of actual dose to the 4 higher quintiles (44.3% vs 42.5%; OR, 1.08; 95% CI, 0.55-2.09; P = .83), adjusting for age, baseline NIHSS score, and time of onset to treatment. Our findings indicate that it is reasonable to estimate weight when administering IV t-PA for acute stroke, because dosing errors did not have a major impact on the risk of ICH or the likelihood of good outcome in the NINDS t-PA Stroke Study.