From the Guidelines
The risk of missing a subarachnoid hemorrhage (SAH) in a 45-year-old female with a rapid onset headache and a normal CT head scan performed 7 hours after symptom onset is approximately 2-5%, and therefore, a lumbar puncture is necessary to rule out SAH. Given her presentation with a thunderclap headache, this patient should still be considered at risk for SAH despite the normal CT scan result. The next appropriate step would be to perform a lumbar puncture to analyze the cerebrospinal fluid for xanthochromia or red blood cells, which can detect SAH even when CT is negative, as recommended by the American Heart Association/American Stroke Association in their 2023 guideline for the management of patients with aneurysmal subarachnoid hemorrhage 1. Modern multidetector CT scanners have improved sensitivity (around 95-98%) for detecting SAH within the first 6-12 hours, but they still miss a small but significant percentage of cases, particularly those with small bleeds. The sensitivity of CT decreases further with time from symptom onset, dropping to about 50% at one week, as blood in the subarachnoid space gets diluted and reabsorbed over time. For this patient, given the potentially devastating consequences of a missed SAH, additional testing is warranted despite the normal CT result. Some guidelines, such as the Canadian Stroke Best Practice Recommendations, suggest that a lumbar puncture may not be required if a third-generation or higher CT scan is performed within six hours of onset of headache and read as normal by a neuroradiologist 1. However, in this case, the CT scan was performed 7 hours after symptom onset, and the patient's clinical presentation warrants further evaluation. Therefore, a lumbar puncture is the most appropriate next step to rule out SAH, as it can detect xanthochromia or red blood cells in the cerebrospinal fluid, which is a sensitive and specific test for SAH, with a sensitivity of 100% and specificity of 95.2% 1.
From the Research
Risk of Missing a Subarachnoid Hemorrhage
The risk of missing a subarachnoid hemorrhage (SAH) in a 45-year-old female with a rapid onset headache and a normal computed tomography (CT) head scan 7 hours after symptom onset can be assessed based on the sensitivity and specificity of CT scans and lumbar puncture (LP) tests.
- The sensitivity of modern multislice CT for subarachnoid haemorrhage at incremental timepoints after headache onset is high, with a sensitivity of 99.6% for aneurysmal SAH and 99.0% for all SAH at 48 hours after headache onset 2.
- A normal CT head scan within 6 hours of headache onset is highly sensitive for the detection of SAH, but the sensitivity may decrease beyond this time frame 3.
- However, a study found that a normal head CT has a very high negative predictive value to exclude SAH when carried out within 24 hours from ictus, provided a 3rd generation CT scanner is utilized, and the scan is reported by a neuroradiologist 4.
- LP is often performed if the CT is negative for blood beyond 6 hours of headache onset, but the yield of LP for diagnosing SAH in adults with nontraumatic headache after a normal head CT is very low, with a number needed to diagnose of 151 5.
- The risk of LP-related complications is higher than the risk of SAH diagnosis, with 5.96% of patients experiencing an LP-related complication, including low-pressure headaches, pain at the LP site, and contaminated CSF cultures 5.
Diagnostic Accuracy of CT Scans and LP Tests
- The diagnostic accuracy of CT scans and LP tests for SAH can be assessed based on their sensitivity, specificity, and positive and negative predictive values.
- A study found that CSF-xanthochromia testing using the revised UK-NEQAS method has a sensitivity of 100%, specificity of 98.1%, positive predictive value of 65.9%, and negative predictive value of 100% for diagnosing SAH 6.
- Another study found that noncontrast cranial CT within 6 hours of headache onset accurately rules in and rules out SAH, with a likelihood ratio of 230 and 0.01, respectively 3.
- The diagnostic accuracy of LP tests, including CSF xanthochromia analysis, is lower than that of CT scans, with a likelihood ratio of 5.7 and 0.21 for a threshold RBC count of 1,000 × 10(6) /L 3.