Does anticoagulation promote mural hematoma growth or delayed occlusion in spontaneous cervical artery dissections?

Background: In case of spontaneous cervical artery dissection (CAD), a medical treatment with anticoagulant or antiplatelet (AP) drugs would avoid the occurrence of an ischemic stroke. Although immediate anticoagulation (AC) is advocated, evidence from randomized trials is lacking. Since CAD is characterized by a mural accumulation of blood, the dissecting hematoma may enlarge under AC, with subsequent lumen narrowing. Although direct evidence of mural hematoma enlargement is lacking in the literature, such a complication may not only be theoretical. Magnetic resonance imaging (MRI) of the mural hematoma on transverse sections through the neck is the current diagnostic gold standard. Our aim was to compare the evolution of the mural hematoma in CAD during the first week after treatment initiation (AP agent: groupAP, AC: groupAC), using dedicated cervical MRI of the arterial wall.

Methods: The study was -approved by the Ethics Committee of Ile de France III. Informed consent was waived. The manuscript was prepared in accordance with the STROBE statement. Fast spin-echo T1-weighted fat-suppressed axial sequences were performed at admission (MRI1) and during the first week after initiation of the treatment (MRI2). Two readers measured volumes, craniocaudal length of the mural hematoma and lumen patency, and searched for early recurrent CAD. They also searched for extension or recurrence of ischemic brain lesions and for hemorrhagic transformation on diffusion-weighted imaging (DWI) and gradient echo T2 (T2*) sequences, respectively.

Results: The population included 44 patients (31 in groupAC, 13 in groupAP) with 49 CAD (35 carotid, 14 vertebral). Recurrent CAD and reduction of the lumen did not occur in either group. We did not observe recurrent DWI lesions or occurrence of hemorrhagic transformation. Interobserver agreement [intraclass correlation coefficient (95% CI)] was excellent for volume measurement [0.98 (0.97-0.99) and 0.99 (0.98-1.0) for volume1 and volume2, respectively]. While mean volumes and length of the mural hematoma decreased after treatment in both groups (volume: groupAC -13 ± 22%, groupAP -12 ± 24%, p = 0.33; length: groupAC -10 ± 27%, groupAP -10 ± 20%, p = 0.18), approximately one third of patients in each group had some growth of the mural hematoma as well as an increase in length.

Conclusion: Limited growth of the mural hematoma was seen with both treatments in approximately one third of patients during the first week after treatment initiation. However, neither AC nor AP agents promote reduction of the lumen or recurrent dissection.