Vitamin K Antagonist (Phenprocoumon) and Subarachnoid Hemorrhage: A Single-Center, Matched-Pair Analysis.

Abstract

Demographic changes are leading to an aging society with a growing number of patients relying on anticoagulation, and vitamin K antagonists (VKA) are still widely used. As mortality and functional outcomes are worse in case of VKA-associated hemorrhagic stroke, phenprocoumon treatment seems to be a negative prognostic factor in case of subarachnoid hemorrhage (SAH). The purpose of this study was to analyze whether phenprocoumon treatment does worsen the outcome after non-traumatic SAH. All patients treated for non-traumatic SAH between January 2007 and December 2016 in our institution were retrospectively analyzed. After exclusion of patients with anticoagulant or antiplatelet treatment other than phenprocoumon, we analyzed 1040 patients. Thirty-three patients (3%) of those were treated with continuous phenprocoumon. In total, 132 out of all 1007 patients without anticoagulant treatment of the remaining patients were matched as control group (ratio = 1:4). Patients with phenprocoumon treatment were significantly older (66.5years vs. 53.9years; p < .0001), and admission status was significantly more often poor (66.7% vs. 41.8%, p = .007) compared to all patients without anticoagulant treatment. Further, bleeding pattern and rates of early hydrocephalus did not differ. Matched-pair analysis revealed a significant higher rate of angio-negative SAH in the study group (p = .001). Overall rates of hemorrhagic or thromboembolic complications did not differ (21.4% vs. 18.8%; NS) but were more often fatal, and 30-day mortality rate was significantly higher in the phenprocoumon group than in patients of the matched-pair control group (33% vs. 24%; p < .001). 30% of the phenprocoumon group and 37% of the matched-pair control group reached favorable outcome. However, poor outcome was strong associated with the reason for phenprocoumon treatment. Patients with phenprocoumon treatment at the time of SAH are significantly older, admission status is worse, and 30-day mortality rates are significantly higher compared to patients without anticoagulant treatment. However, outcome at 6months did not differ to the matched-pair control group but seems to be strongly associated with the underlying cardiovascular disease. Treatment of these patients is challenging and should be performed on an interdisciplinary base in each individual case. Careful decision-making regarding discontinuation and bridging of anticoagulation and close observation is mandatory.