A 45 year old male comes into the ED with a sudden, severe headache. It started while he was at work yesterday and was the worst of his life. It started feeling a little better, but hasn’t totally gone away and his wife made him come to get checked out.

There are enough concerning features in this history that most of us would agree, this patient needs a stat CT head. You get the CT, look at the scan, and—no blood.  It’s been over 24 hours since his headache started and you know that the CT head isn’t sufficient enough to rule out this dangerous etiology. So, now what?

The traditional approach is CT head first, followed by LP if the CT is negative. Studies have shown that the sensitivity of this is 100%, while the specificity is only about 65%. There are also a number of drawbacks to LP. They can be painful, time consuming, and often difficult secondary to patient factors.  The results can be difficult to interpret, especially in a traumatic tap. But LP remains the gold standard and recommended as part of the ACEP Clinical Policy for headache (though the policy hasn’t been updated since 2008).

So what’s the other option? CT angiography of the brain is a reasonable one and studies have showed favorable outcomes–a CTA is 98% sensitive and 100% specific for identifying aneurysms >3mm. It’s also potentially more efficient than an LP. But it’s not a perfect test. You may identify aneurysms that are not the cause of the headache–leading to more testing and possibly even procedures. Additionally, you expose your patient to an extra radiation load and renally-impaired patients may not be able to tolerate the contrast.

Final verdict? The jury is still out. No one seems ready to say that CTA is superior to LP. But many, including our own Dr. Probst (who recently wrote a paper on this very topic), are moving in the direction of saying that CTA is an “acceptable alternative diagnosis strategy” for those who are at risk of SAH. And of course, center your approach on shared decision making with your patient–discuss the risks and benefits of both pathways, and come up with a plan together.

tl;dr–LP remains the gold standard recommendation for patients who are at high risk of SAH, but have a negative CT head. However, CTA is gaining credibility and currently viewed as an acceptable alternative. Maybe we’ll move in that direction with a bit more research?

Thanks to Dr. Khan for inspiring this pearl!


Edlow, JA. Managing Patients With Nontraumatic, Severe, Rapid-Onset Headache. Annals of Emergency Medicine, March 2018, Vol 71, Issue 3, 400-408.  https://doi.org/10.1016/j.annemergmed.2017.04.044

Probst, M & Hoffman, J. Computed Tomography Angiography of the Head Is a Reasonable Next Test After a Negative Noncontrast Head Computed Tomography Result in the Emergency Department Evaluation of Subarachnoid Hemorrhage” Annals of Emergency Medicine. June 2016. Volume 67, Issue 6, Pages 773–774

May 2024