This is Javier Preventia, and I am a professor of neurology and the director of neurocritical care at the University of Virginia. And I wrote this question. This is a challenging question, and there have been some things brought up by some of the learners that I think deserve more explanation. In summary, this is a question about a 44-year-old man who presents with what looks like a subarachnoid hemorrhage. Interestingly, the patient's already had some of their workup, which is what makes this question so challenging. They've already had a CT angiogram and a digital subtraction angiogram, which is a conventional angiogram we think of when we do workups for cerebral aneurysms. The most important piece of data that one needs to know when somebody has a subarachnoid hemorrhage is whether there's an aneurysm or not. If there's an aneurysm, the risk of rebleeding is quite high, and obliterating that aneurysm is the first order of business. So in this case, the most important piece is to figure out whether there's an aneurysm, but the digital subtraction angiogram and CT angiogram are pretty good for that. Interestingly, the lumbar puncture response is not appropriate because we already know the patient has a subarachnoid hemorrhage, and lumbar puncture is really used to diagnose subarachnoid hemorrhage, not to find the aneurysm. So you're left with three other options, an MRI of the brain, an MRI of the spine, and a digital subtraction angiogram in six weeks. So let me dispel with the first one. A digital angiogram in six weeks is actually a good idea, but it's not the next appropriate step. There need to be some intervening steps before you get there. The MRI of the brain is the one that many people want to select, and the reason for that is that it gives you a better picture of the brain, particularly of things like brain tumors. In this case, the differential diagnosis of this entity is either an aneurysm that has obstructed itself, that has clotted off so you don't see it through any vascular imaging, an AVM, a tumor which leads to subarachnoid hemorrhage, or other odd causes. There is one benign syndrome called a benign perimucinophallic hemorrhage, which is a diagnosis of exclusion, so the rest of the work needed to be completed anyway. So in this case, the utility of an MRI of the brain is less, because on the differential diagnosis, an AVM would have been diagnosed by either the CT angiogram or the digital subtraction angiogram. A tumor also will likely be seen in either of those situations, and a tumor so small as not to be seen in that situation is unlikely to cause a subarachnoid hemorrhage. A more dangerous entity, although uncommon, that needs to be diagnosed is whether somebody has a spinal AVM, which is not typically seen in the digital subtraction angiogram or on a CT angiogram of the brain, but that can lead to a subarachnoid hemorrhage. In that situation, an intervention would need to be done relatively quickly to prevent re-bleeding. So the only test on this list that would help you diagnose a spinal AVM is an MRI of the spine. That doesn't mean to say that the patient wouldn't need another digital subtraction angiogram later, because that is true that a second angiogram can sometimes find a ruptured aneurysm that is clotted off and recanalized, but this would be done after the spinal MRI. I hope that this description gives you a better idea of why the correct answer was the correct answer. The most important teaching point I thought when I wrote this question was to understand that spinal AVMs are actually a cause of subarachnoid hemorrhage, although uncommon, but they can lead to significant morbidity if not identified early in the hospital course.

subarachnoid hemorrhage

aneurysm

diagnosis

MRI of the spine

spinal arteriovenous malformation (AVM)