Comparing Outcomes of Patients With Idiopathic Subarachnoid Hemorrhage by Stratifying Perimesencephalic Bleeding Patterns

• Published in: Journal of stroke and cerebrovascular diseases 2019-09, Vol.28 (9), p.2407-2413
• Main Authors: Atchie, Benjamin, McGraw, Constance, McCarthy, Kathryn, van Vliet, Rebecca, Frei, Donald, Bennett, Alicia, Bartt, Russell, Orlando, Alessandro, Wagner, Jeffrey, Bar-Or, David
• Format: Article
• Language: English
• Subjects: computed tomography angiography; hydrocephalus; Nonaneurysmal subarachnoid hemorrhage; perimesencephalic subarachnoid hemorrhage; vasospasm
• ISSN: 1052-3057; 1532-8511
• DOI: 10.1016/j.jstrokecerebrovasdis.2019.06.032

Background: To determine the clinical outcomes of perimesencephalic subarachnoid hemorrhages based on the computed tomography (CT) bleeding patterns. Methods: This retrospective cohort study included: (1) patients (≥18 years) admitted to a comprehensive stroke center (January 2015-May 2018), (2) with angiography-negative, nontraumatic subarachnoid hemorrhage in a perimesencephalic or diffuse bleeding pattern, and (3) had CT imaging performed in ≤ 72 hours of symptom onset. Patients were stratified by location of bleeding on CT: Peri-1: focal prepontine hemorrhage; Peri-2: prepontine with suprasellar cistern +/− intraventricular extension; and diffuse. Results: Of the 39 patients included, 13 were Peri-1, 11 were Peri-2, and 15 were diffuse. The majority were male (n = 26), with a mean (standard deviation) age of 55.3 (11.3) years, who often presented with headache (n = 37) and nausea (n = 28). Overall, patients in Peri-1 were significantly less likely to have hydrocephalus compared to Peri-2 and dSAH (P= .003), and 4 patients required an external ventricular drain. Five patients developed symptomatic vasospasm. Patients in Peri-1, compared to Peri-2 and diffuse, had a significantly shorter median neuro critical care unit length of stay (LOS) and hospital LOS. Most patients (n = 35) had a discharge modified Rankin Score between 0 and 2 with no significant differences found between groups. Conclusion: These data suggest that patients with the best clinical course were those in Peri-1, followed by Peri-2, and then diffuse. Because these patients often present with similar clinical signs, stratifying by hemorrhage pattern may help clinicians predict which patients with perimesencephalic subarachnoid hemorrhage develop complications.