Management of Subarachnoid Haemorrhage: A Decision Analysis

Abstract

The management of subarachnoid haemorrhage is studied by decision analysis. Optimal treatment strategies for individual patients are identified, by combining early surgery (on day 2), delayed surgery (on day 10), late elective surgery after 3 months, antifibrinolytics and nimodipine. The risks of rebleeding, infarction and mortality from the initial haemorrhage were estimated by proportional hazards regression in a reanalysis of a randomized clinical trial of antifibrinolytics against placebo. These combined with data from the literature are used in a decision tree to compute the chances of good recovery, moderate and severe disability and death after 3 months, and the (discounted quality-adjusted) life expectancy for each treatment strategy. Treatment with nimodipine is recommended for all patients. For patients in good clinical condition at admission, antifibrinolytics and delayed surgery form the optimal treatment strategy. However, the benefits of early or delayed surgery are small compared to elective late surgery in this group, and therefore the surgical complication rate should be low. Patients with a heavy load of subarachnoid blood on CT scan should be treated with (delayed) surgery, but not with antifibrinolytics, because of the increased risk of infarction. Late elective surgery is recommended for patients with good outcome after 3 months who have not yet been operated.