Short Attention Span Summary
Workup is negative - no stroke, right?
Wrong - 9% of the time. When considering stroke, TIA, and SAH, this article states that even in the era of modern neuroimaging, missed diagnoses occured in 9% of ED patients. And if that wasn't terrifying enough, "Risk of misdiagnosis is much greater when presenting neurologic complaints are mild, nonspecific, or transient (range 24%-60%)." When in doubt, patients with strange neurological symptoms need a consult or admission for more workup. And if the patient has a worrisome history or exam, persist until you get a diagnosis.
Even in the era of modern neuroimaging we may miss up to 9% of cerebrovascular events. However, you may want to read the in depth analysis below. This may not reflect missed events purely due to false negative neuroimaging.
ED misdiagnosis of cerebrovascular events in the era of modern neuroimaging: A meta-analysis.
Neurology. 2017 Apr 11;88(15):1468-1477. doi: 10.1212/WNL.0000000000003814. Epub 2017 Mar 29.
Peer Reviewer Comments
I went through the sources of this meta-analysis. The reported sensitivity/specificity was based on clinical diagnosis and did not reflect the sensitivity of the neuroimaging alone. The title of this article was somewhat misleading. Here are two sample articles used in the meta-analysis (reference 22 and reference 23) as examples of the kinds of input used to create the meta-analysis. Both of these studies looked at a consecutive series of SAH and determined that there was a clinical miss rate of 12 and 25%, respectively. The most common error in diagnosis was failure to consider SAH on the differential, leading to a failure to order CT. The most common misdiagnoses were viral meningitis and migraine headache. Interestingly, in reference 23, 7 of 8 misdiagnosed as "viral meningitis" failed to get CT or LP. How else do you diagnose viral meningitis? And the vast majority of missed cases of SAH were Hunt and Hess grade 1 or 2 on initial presentation--consistent with the meta-analysis conclusion that most missed diagnoses had more subtle presentations.
So it appears that the 9% figure is not the miss rate of neuroimaging. Instead, it seems to be a vaguely defined "clinical miss rate." This may or may not have included imaging. However, many prior studies have found MRI can miss strokes, especially early in presentation - "31% of patients with vertebrobasilar ischemic stroke had a false-negative initial DWI study during the first 24 hours."
Diagnosing neurological complaints is tough. Even neurologists miss about 25% of initial ED presentations. In particular, double check your thought process before discharging a patient home with a diagnosis of headache or vertigo since these were the two most common incorrect diagnoses.