LAUSR

In this issue of Acta Neurologica Scandinavica, Bergh et al present a study in which they account for reasons why thrombolysis is not given to patients with ischemic stroke.1 They wisely conclude that increasing the use of thrombolysis within the time window of 4.5 h must be weighed against possible harms. In Sweden, the Swedish Stroke Register started in 1995. In 2011, the register decided to present target levels for some variables. One of these variables was the use thrombolysis. The target level for a “moderate” performance was 10%– 14%, while the target level for a “high” performance was 15% or more. (Today, thrombolysis and thrombectomy are merged.) The National Board of Health and Welfare chose to publish these variables in their annual publication “Open comparisons,” which at least during its first year received some media attention. Reporting thrombolysis in this manner gives a clear message to the healthcare: more is better. Is there any reason be that proactive? For those of us who were stroke physicians in the early 2000s, when thrombolysis was introduced, remember that it was only after several years of hesitation that the European Agency for the Evaluation of Medicinal Products (EMEA, now EMA) approved the treatment. This long hesitation was related to safety issues, more specifically the increased mortality that is associated with thrombolysis. Death is the most serious outcome of any medical treatment, and in the case of thrombolysis, the increased mortality is still unsolved. Later studies, such as The Third International Stroke Trial (IST3),2 have confirmed this harmful effect. However, a metaanalysis published after IST3 gave seemingly reassuring news: “Numbers of deaths within 7 days were increased ... but by final followup the excess was no longer significant....”3 This statement seems to have created a belief among stroke physicians that early deaths caused by thrombolysis are unimportant, as thrombolysistreated patients have better longterm prognosis. There are at least three objections to this way of reasoning.