Early CT ‘definitive’ for subarachnoid hemorrhage diagnosis

Aug 6 2012

By Eleanor McDermid

Cerebrospinal fluid (CSF) analysis is only necessary for patients with thunderclap headache and a negative computed tomography (CT) scan if they undergo imaging more than 6 hours after headache onset, research suggests.

Mervyn Vergouwen (University Medical Center Utrecht, the Netherlands) and colleagues found that CT findings were 100% accurate for diagnosing subarachnoid hemorrhage (SAH) provided patients had a typical presentation (thunderclap headache, neurologically normal) and the scan was performed within 6 hours of onset.

"Given this analysis, we believe that practice should change," say Jonathan Edlow and Jonathan Fisher (Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA) in an editorial accompanying the study in Stroke. They believe that lumber puncture is no longer indicated in patients with typical presentation who have a negative CT scan within 6 hours.

They note, however, that the study took place in a referral center with experienced neuroradiologists available to interpret the scans. They reference a previous study in which there were four false-negative scans among 953 performed within 6 hours of headache onset. Three of these were initially read by emergency physicians and one by a trainee radiologist.

"This underscores the importance of having well-trained, experienced physicians interpret the CT scans," say Edlow and Fisher.

In the current study, 69 of 137 patients scanned within 6 hours had negative or inconclusive findings and so underwent lumbar puncture. This was negative in all but one patient, but that patient had an atypical presentation, with acute neck pain, nausea, and neck stiffness, but no headache. They were eventually diagnosed with a bleeding cervical arteriovenous malformation.

A further 76 of 113 patients scanned more than 6 hours after onset had negative or inconclusive findings. Five of these patients had positive lumber puncture results, giving CT a sensitivity of 92.3% when restricted to patients presenting with headache, and a specificity of 100%.

The sensitivity, specificity and positive and negative likelihood ratios were all 100% for patients with headache scanned within 6 hours of onset, making CSF analysis unnecessary.

Edlow and Fisher note that some have advised CT angiography in the event of a negative scan, to directly detect culprit aneurysms. But they say: "We believe that this 'technology creep' only leads to diagnosis of asymptomatic aneurysms (not the hemorrhage) and has many other unintended negative consequences."

Licensed from medwireNews with permission from Springer Healthcare Ltd. ©Springer Healthcare Ltd. All rights reserved. Neither of these parties endorse or recommend any commercial products, services, or equipment.