Headaches

Shortly after the bleed, red blood cells appear in the CSF, but rapidly diminish in numbers over time due to CSF flow. In the hours to days that follow, xanthochromia, a yellowish discoloration, appears. This is due to degradation of hemoglobin from lysed red blood cells into bilirubin, oxyhemoglobin, and methemoglobin. Xanthochromia can be measured visually or by spectrophotometry, each of which has advantages and disadvantages.¹

Almost every North American hospital laboratory uses visual inspection.¹⁰ Measuring opening pressure helps to diagnose the occasional case of pseudotumor cerebri, CVST (elevated pressure), or spontaneous intracranial hypotension (low pressure). It can also help to distinguish a traumatic LP (pressure is low) versus SAH (pressure is elevated in approximately 60%).

Is Lumbar Puncture Required After a Negative CT Within 6 Hours of Headache Onset?

It has long been recognized that CT sensitivity for SAH decreases over time and is most sensitive in the first 24 hours. Over the last few years, increasing data and expert opinion suggest that CTs performed within 6 hours of onset of classic thunderclap headache in neurologically normal patients are nearly 100% sensitive, obviating the need for LP.^11-13 Although another study called these data into question, the methodology did not allow distinction between traumatic taps in patients with incidental aneurysms and true SAH and not all of their “negative” CT scans were truly negative.¹⁴

The author’s practice no longer uses LP in patients with classic thunderclap headache who are neurologically normal and whose CT scans are performed within 6 hours of headache onset and are read as negative by an attending-level radiologist. Physicians who follow this method should strictly adhere to each component of the preceding sentence. It is estimated that such an approach will miss as many as one in 600 to 1,000 patients. Because this represents a change in practice that has not yet been incorporated into any published guidelines, if a physician chooses to skip the LP in this circumstance, a discussion with the patient is warranted, the nature of which should focus on the balance of benefits (eg, LP picking up the very rare SAH in this group of patients) versus the harm (eg, side effects of the LP, which the author believes is mostly the increased number of advanced imaging due to traumatic taps and from incidental findings). This introduces the concept of “testing threshold,” an interesting and increasingly important concept for EPs to consider in diagnosing any low-frequency but high-stakes condition.^15,16 

Newer Diagnostic Strategies
Over the last few years, various alternative workups, including LP-first, magnetic resonance imaging (MRI) only and CT/CT angiography (CTA), have been proposed to replace the standard workup. Various authors have analyzed both the clinical and economic advantages and disadvantages of each modality.^17,18

LP-First. The logic for the LP-first approach is because physicians often do not perform the LP in routine practice. A careful neurological examination is critical to ensure that this method is restricted to neurologically normal patients. Some neurologically intact patients with a lesion on CT that does not affect the CSF will be missed with this approach. In addition, if this approach is used, the opening pressure of the CSF must be measured in every case. The advantage is that it spares radiation exposure and forces an LP to be performed.

Primary MRI. The advantage of primary MRI is that current MRI sequences are as sensitive as CT for acute blood, and more sensitive for subacute and chronic blood. There is also no radiation exposure. Furthermore, depending on the sequences, which may include cerebrovascular imaging, one may be able to diagnose other rare causes of thunderclap headache beside SAH or unruptured aneurysm. It is important to note, however, that as with CT, smaller bleeds may appear negative on MRI—thus, spectrum bias exists for both modalities.¹ The obvious disadvantage is cost and more importantly, availability in real time. As MRI technology and penetration into routine practice advance, this would become a very reasonable approach.

CT AND CTA. Finally, some have suggested that CT followed by CTA should be the new paradigm in order to avoid LP. These tests are easy for the EP to do and less painful to the patient, but there are unintended consequences to this strategy, including missing CSF-diagnosable conditions, diagnosing incidental aneurysms and other findings (all of which invariably lead to more imaging), and increased radiation and contrast dye exposure.¹⁹ The CT/CTA strategy does make sense in patients who cannot undergo LP (eg, patient refusal, unfavorable body habitus, anticoagulation use).

Beyond SAH

For most ED patients with thunderclap headache, CT and LP—the first two steps in the diagnostic workup—are sufficient to diagnose SAH. A meta-analysis of seven studies of neurologically normal subjects with thunderclap headache and normal CT and CSF results (813 patients) found no cases of SAH or occurrence of sudden death during at least 3 months of follow-up.²⁰ Using the statistical worst-case scenario (upper bound of the 95% confidence interval) would be that four of 1,000 patients could have an SAH.