Risk of Intracranial Hemorrhage in Users of Oral Antithrombotic Drugs: a Nationwide Study

The most serious adverse effect of antithrombotic therapy is bleeding. Combinations of antithrombotic agents are now frequently used (e.g. after use of drug eluting stents or after ischemic stroke), and this may lead to an increased frequency of significant bleeding complications. Among hemorrhagic complications of antithrombotic drugs, intracranial hemorrhage (ICH) may have particularly devastating consequences with high morbidity, disability and even mortality rates. Intracerebral hemorrhage (hemorrhagic stroke) is generally associated with a higher risk for death and incurs greater loss of health over a lifetime than ischemic stroke.

This makes antithrombotic therapy a double-edged sword. Although a certain risk for bleeding may be acceptable in the context of even greater protection against ischemic events, it is important to quantify the magnitude of bleeding risk. So far the efficacy and safety profile of antithrombotic agents are generally assessed in randomized controlled trials (RCT). However, extrapolating the results from randomized clinical trials to the general patient population in this context is challenging. Patients who participate in clinical trials are frequently highly selected and therefore somewhat unrepresentative. In addition, their numbers are limited and the treatment period is often much shorter than in routine management of a chronic disease or condition. Finally, patients in clinical trials are often monitored more closely than in routine practice.

The incidence of intracranial hemorrhage due to antithrombotic therapy could theoretically be monitored by post-marketing surveillance by including spontaneously reported events. Unfortunately, it seems this does not provide more reliable estimates. A recent study from Finland has shown that bleeding complications due to oral anticoagulation with Warfarin are underreported in daily clinical practice. Further, it has been shown that reporting rates of side effects following medical therapy tend to decrease over time indicating that it is more likely that adverse events to a newer drug are reported than to a drug that has been available for many years. This is why population-based large-scale pharmaco-epidemiological studies are needed, in which cohorts of patients exposed to antithrombotic medications are monitored to determine a valid and reliable risk of the treatment.