Subarachnoid Hemorrhage: A Sharp Needle in a Haystack

This systematic review by Carpenter and colleagues1 assesses clinical findings and investigations for emergency department (ED) patients with acute headache. Half of all subarachnoid hemorrhage (SAH) patients, those with global or focal neurologic deficits, are immediately apparent. It is the patient with a new rapidly peaking headache and no other neurologic findings who is difficult to identify and likely to be helped by early management. It is not surprising that in this review, there was no one clinical feature that identifies all cases of SAH. The etiology of SAH is variable. Even among patients with the most common etiology, a ruptured aneurysm, the location, shape, and size vary from patient to patient. It is interesting to note that 1) the presence or absence of the worst headache of the patient's life was not reliable in distinguishing SAH from other headaches and 2) that headaches with an onset of greater than 1 hour were rarely SAH (LR = –0.06). The clinical decision rules created in Ottawa are able to reduce the risk of misdiagnosing a SAH. Patients with none of the high risk features of "Rule 1" had a negative LR of –0.06 and 95% CI of 0.01 to 0.22.2-4 However, "Rule 1" missed two cases of SAH. One was a nonaneurysmal SAH with a good prognosis, and the other an aneurysmal SAH that required intervention in a young woman. Given the severity of a missed SAH, the Ottawa Investigators reanalyzed their data and derived The Ottawa SAH Rule, which identified all cases of SAH. While not decreasing investigations at their study sites, it did not increase them either. It provides evidence to make the decision on who to investigate and who not to. In countries where healthcare resources are less available, using Rule 1 would be a reasonable approach that would still be better than current clinical practice according to a population-based study that identified that up to one in 20 patients presenting to an ED with SAH are initially missed on their first presentation.5 This number is likely to be considerably higher when including patients with a SAH who present to health care facilities other than EDs.6 In this review, the authors suggest that The Ottawa SAH Rule could slightly increase testing rates if strictly applied; however, in the unpublished validation study, investigation rates continue to increase and the rule actually would have modestly decreased testing if it had been strictly applied in the validation study sites. Hence, while The Ottawa SAH Rule will not drastically alter the number of tests, it provides evidence to ensure that the correct patients are being tested. This systematic review demonstrates that new generation computed tomographic (CT) scanning, using thin slices and interpreted by trained radiologists identifies almost all cases of aneurysmal SAH if done soon enough after the onset of headache. Of missed SAH, most are nonaneurysmal SAH (e.g., perimesencephalic bleeds from low-pressure blood vessels). In many centers these patients undergo angiographic evaluation, most frequently CT angiography (CTA), and are discharged home by neurosurgery if the CTA does not detect a vascular lesion predisposing to SAH. Hence, most such patients have a very good prognosis if their initial clinical status, CT imaging, and CTA imaging are benign. These patients do well without intervention other than pain control for their headache. Given the accessibility and relatively small dose of radiation associated with a plain head CT scan, this is generally accepted to be the first step to investigate a suspected SAH. If it has been less than 6 hours from the time of onset, CT alone is a reasonable approach.7 While this approach will not be perfect, it is near perfect. The 95% CI is between 97 and 100%; therefore, unless your pretest probability is extremely high (e.g., a patient with a known aneurysm who presents with an explosive onset, very severe headache), it is reasonable to stop the diagnostic evaluation if the CT scan is negative. This, of course, depends on a quick read by a highly trained radiologist. If further testing is needed, a lumbar puncture is one of the options. Here the systematic review focuses on a red blood cell RBC count of <1,000 × 106/L. It is unclear why they chose this cut point, given that the optimal cut point identified previously by the ROC curve was 2,450 × 106/L in which the authors rounded down to 2,000 × 106/L RBCs.8 However, using a RBC count of <1,000 × 106/L will give a more conservative (i.e., safe) estimate without sacrificing much specificity. While lumbar punctures may be painful, time-consuming, and frightening, they rarely cause serious sequelae and they are still the next step in diagnosing a SAH. A clearly negative lumbar puncture effectively rules out SAH. However, traumatic taps can complicate the picture. CTA, in contrast, is quick and easy for emergency department physicians to perform and has 98% sensitivity for identifying aneurysms.7 The problem is that about one in every 100 patients will have an incidental aneurysm that can be identified on CTA, many of which would have never caused the patient any problems. Once identified, an aneurysm will likely be treated with endovascular coiling or open surgical clipping as an aneurysm of any size that is deemed to be symptomatic will be considered a high-risk lesion. Most neurosurgeons would consider treating the identified aneurysm less risky for the patient (and easier to defend in court) than to assume that an intracranial aneurysm is not the cause of the presenting symptoms that brought the patient to the ED. Some of these patients will have the small risks of the CTA and the more significant risks of endovascular or open surgery for incidental aneurysms. Further, any untreated patients will have to live with the idea that they harbor an unruptured aneurysm that may reduce their quality of life, require surveillance imaging, and affect their insurability. In addition, sudden aneurysm enlargement, even without SAH, can produce a sudden-onset severe headache that mimics the symptoms of an aneurysm warning leak. A sudden-onset headache and a new oculomotor nerve palsy are known to be associated with posterior communicating artery aneurysms ipsilateral to the dilated pupil.8 These headaches are due to the enlargement of the aneurysm and will not have evidence of SAH on imaging or an abnormal lumbar puncture. These aneurysms are at high risk of rupture and require diagnosis with angiography as both head CT scan and lumbar puncture will be normal. This systematic review supports 1) investigations based on the Ottawa SAH Rule, or "Rule 1"; 2) using a noncontrast head CT scan as the initial study; and 3) performing a lumbar puncture for patients with a high pretest probability of SAH and a negative CT scan, especially if the CT scan is performed more than 6 hours after the onset of headache. Consider reserving CTA for patients with a likely SAH based on cerebrospinal fluid results or a positive CT scan or patients with a compelling history and a negative CT scan for whom a lumbar puncture is not feasible (e.g., on anticoagulation, difficult anatomy). A CTA is also indicated if there is a sudden headache with oculomotor nerve palsy. While the focus of this systematic review is on medical care and the decisions to make as a care provider, there is no mention of involving patients in our decisions. Given the extremely low risk of SAH following a normal CT, especially an early CT, it is reasonable to discuss treatment options with our patients. Usually by the time a CT scan is completed and interpreted, we have had time to manage the headache and patients will be in less distress and able to discuss investigation options. In our experience, most patients will decline to have any further testing done if the CT scan is negative and they are reassured by their physician that this is an acceptable option. We document this conversation, manage any remaining symptoms, provide appropriate discharge instructions, and make ourselves available for further questions or concerns in the future. For the minority of patients who wish to have additional tests performed, a lumbar puncture can be done. Subarachnoid hemorrhage is one of the most feared causes of acute headache. While subarachnoid hemorrhage is relatively rare, it can be catastrophic if missed, and a missed subarachnoid hemorrhage often results in litigation against emergency physicians. However, we must not be overly swayed by the fear of missing a subarachnoid hemorrhage nor the fear of legal action initiated against us. Using evidence-based practice is both reasonable and encouraged. This systematic review demonstrates that The Ottawa SAH Rule is an evidence-based tool to identify high-risk and low-risk patients. Investigating only high-risk patients can be justified. When investigating patients, using an unenhanced computed tomography scan alone may be justified if the pretest probability is less than 20% and the testing is done within 6 hours of the headache onset. Following a negative computed tomography scan, a conversation with the patient is a good way to assess if further testing is warranted. Patients are pragmatic and do not want painful procedures done to them if the risks are low. Having this conversation is a good way to determine what the appropriate level of "rule-out subarachnoid hemorrhage" is for acute headache patients. Documenting this conversation, treating symptoms, considering other serious etiologies, and providing sound follow-up instructions will complete the management for these patients.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.