The evaluation of headache in the emergency room can be challenging.
Headache is a common presenting problem, and only a minority of patients with headache harbor malignant pathology. This is offset, however, by the sometimes subtle features of patients with subarachnoid hemorrhage (SAH), a malignant headache with a greater than 40% mortality rate. The clinical presentation of SAH is typified by the sudden onset of severe headache ("worst headache of life"). This is often associated with brief loss of consciousness or nausea and vomiting. If the bleeding is substantial, resultant mass effect can precipitate seizures or lateralizing weakness. Up to half of patients with headache from subarachnoid hemorrhage have mild or otherwise atypical signs and symptoms. This fact underscores the importance of careful evaluation of patients with headache.
Headaches that raise concern for SAH have aphoristically been categorized as "first, worst, or cursed." A new headache should raise suspicion for SAH. This has special bearing on the migraineur population: a new headache is classified as a migraine only after several similar episodes, so the first episode should be regarded with suspicion. Likewise, patients with migraines are often familiar with their headaches, so a report of a "different" type of headache may signify a different process. Likewise, any "worst headache of my life," insofar as it remains the classic symptom, requires evaluation for SAH. Finally, the "cursed" headache is the one associated with an abnormal neurologic exam and should similarly prompt thorough evaluation. The clinical diagnosis of SAH is further informed by epidemiologic factors and associated risks. Prevalence is highest in the 5th through 7th decade with slightly more women than men affected. Family or personal history of cerebral aneurysm, Ehlers-Danlos syndrome, or adult polycystic kidney disease are risk factors. Hypertension, tobacco, and heavy alcohol use have been associated with aneurysmal subarachnoid hemorrhage.
The diagnosis can be straightforward if pursued. After clinical evaluation, noncontrast head CT should be performed. If performed 12 to 24 hours after onset of symptoms, sensitivity is better than 95% . However, imaging can still miss small SAH, especially from a sentinel bleed. When performed more than 48 hours after symptom onset, the sensitivity drops to around 50%, as subarachnoid blood becomes isodense. If head CT is not diagnostic, lumbar puncture is indicated. Opening pressure is elevated (>250mm H2O), protein is commonly elevated, glucose is normal, and red cells are often found. Because the needle can injure dural venules on its course to the subarachnoid space, red cells must be counted in sequential tubes. If the sequential erythrocyte count drops, blood may be "traumatic," from venules. However, cell counts may occasionally fall in sequential tube analysis in SAH. By contrast, if the erythrocyte count does not fall, the blood is from subarachnoid space and SAH is diagnosed. Sometimes with small sentinel SAH, erythrocytes are not found in the CSF sampled in the lumbar space. However, erythrocytes in the CSF are rapidly lysed and, even in small volumes, can be detected by looking for xanthochromia. This is assessed either by visual comparison with a water-containing tube or, preferably with digital spectrophotometry. These hemoglobin degradation products can be found as early as two hours after bleeding and persist two weeks or longer in CSF. Advanced imaging is sometimes used. MR imaging with T2*sequence can be as sensitive as early noncontrast head CT, but likewise is imperfect. More intriguing is the use of CT angiography (CTA) as a non-invasive way to assess vasculature and specifically look for the suspected aneurysm. This modality is gaining popularity, as resolution is approaching that of formal catheter-based angiography. However, it is less reliable for finding aneurysms smaller than 3 mm and, unlike lumbar puncture, cannot identify non-aneurysmal sources of SAH (e.g., perimesencephalic hemorrhage). Additionally, if SAH is suspected and CTA is nondiagnostic, catheter-based angiography is required, thus exposing the patient to dye load and radiation twice. Catheter-based angiography remains the "gold standard" to diagnose SAH. Its diagnostic role is mainly to identify the presence, location, size, and morphology of the suspected aneurysm, but may be used in occasional instances if head CT and lumbar puncture analysis are nondiagnostic.
key points to remember
■ Thunderclap headache, headache with associated neurological signs should raise suspicion for subarachnoid hemorrhage.
■ Sensitivity of noncontrast head CT in diagnosing SAH can be more than 95% 12-24 hours after a bleed but drops to 50% after two days.
■ Lumbar puncture to assess for subarachnoid blood or xanthachromia is most sensitive from 12 hours to 2 weeks after bleeding.
■ CT angiography and MR imaging can be useful non-invasive adjunctive modalities to diagnose SAH, but catheter-based angiography remains the gold standard for diagnosis.
Adams HP, Jergenson DD, Kassell NF, Sahs AL. (1980). Pitfalls In the recognition of subarachnoid hemorrhage. JAMA 244:794-796.
Bederson JB, Connolly ES, Batjer J, et al. (2009). Guidelines for the management of aneurysmal subarachnoid hemorrhage. Stroke 40:994-1025.
Chappell ET, Moure FC, Good MC. (2003). Comparison of computed tomographic angiography with digital subtraction angiography in the diagnosis of cerebral aneurysms: a meta-analysis. Neurosurg 52:624-631.
Edlow JA, Caplan LR. (2000). Avoiding pitfalls in the diagnosis of subarachnoid hemorrhage. N Engl J Med342:29-36.
Kaibara T, Heros R. (2008). Aneurysms. In: Uncommon Causes of Stroke, 2nd edition, (Caplan LR, ed.), pp. 171-180. Cambridge UK, Cambridge University Press.
Sidman R, Connolly E, Lemke T. (1996). Subarachnoid hemorrhage diagnosis: lumbar puncture is still needed when the computed tomography scan is normal. Acad Emerg Med3:827-831.
Sidman R, Spitalnic S, Demelis M, Durfey N, Jay G. (2005). Xanthochromia? By what method? A comparison of visual and spectrophotometric xanthochromia. Ann Emerg Med46:51-55.
Was this article helpful?