Citation, DOI & article data
Patients tend to be older middle age, typically less than 60 years old 2. Subarachnoid haemorrhage accounts for 3% of stroke and 5% of stroke deaths 2.
Risk factors include 2:
- family history
- heavy alcohol consumption
- abnormal connective tissue
- female gender: ~1.5x baseline risk 10
- African descent: ~2x baseline risk ref
- Japanese or Finnish descent 12
Patients typically present with a thunderclap headache, described as a sudden-onset headache that is the worst headache of their life. It is often associated with photophobia and meningism. Focal neurological deficits often present either at the same time as a headache or soon thereafter 2.
In a substantial number of patients (almost half 2), it is associated with collapse and decreased or loss of consciousness, even in those patients who subsequently regain consciousness and have a good grade.
Patients can be graded into five groups based on their clinical presentation, using the commonly employed Hunt and Hess grading system, which is predictive of outcome.
Three distinct patterns of subarachnoid haemorrhage have been described, each with its own aetiology and treatment/prognostic implications 4:
- suprasellar cistern with diffuse peripheral extension
- perimesencephalic and basal cisterns
- isolated cerebral convexity
Causes include 1:
- ruptured berry aneurysm: 85% 1
- perimesencephalic haemorrhage: 10% 4
- arteriovenous malformation
- cerebral amyloid angiopathy
- ruptured mycotic aneurysm
- reversible cerebral vasoconstriction syndrome
- dural arteriovenous fistula
- spinal arteriovenous malformation
- venous infarction 1
- intradural arterial dissection
- pituitary apoplexy
- sympathomimetic drugs (e.g. cocaine)
- cerebral vasculitis 6
- anticoagulation therapy 9
Although MRI is thought to be more sensitive, non-contrast CT is frequently performed first due to wider availability. As well as being more sensitive to haemorrhage, MRI has greater sensitivity to the wide range of causative lesions.
A description of the radiographic features of each causative underlying lesion is clearly beyond the scope of this article; these are discussed separately (see above).
The sensitivity of non-contrast CT to the presence of subarachnoid blood is strongly influenced by both the amount of blood and the time since the haemorrhage.
The diagnosis is suspected when a hyperdense material is seen filling the subarachnoid space. Most commonly, this is apparent around the circle of Willis, on account of the majority of berry aneurysms occurring in this region (~65%), or in the Sylvian fissure (~30%) ref needed.
Small amounts of blood can sometimes be appreciated pooling in the interpeduncular fossa, appearing as a small hyperdense triangle, or within the occipital horns of the lateral ventricles 5.
Subarachnoid haemorrhages are grouped into four categories according to the amount of blood on unenhanced CT by the Fisher scale. This scale has been updated to the modified Fisher scale, which more accurately correlates the risk of vasospasm.
MRI is sensitive to subarachnoid blood and is able to visualise it well in the first 12 hours, typically as a hyperintensity in the subarachnoid space on FLAIR 3.
Susceptibility-weighted sequences are also exquisitely sensitive to blood products.
MR angiography and MR venography are also able to detect a causative aneurysm or another source of bleeding, although in general, MRI suffers from poor availability (compared to CT), longer scan times, and greater difficulty in transferring and looking after patients who are often unstable and intubated.
In aneurysm-associated subarachnoid haemorrhage, diffusion weighted imaging may demonstrate early ischaemic changes (within 0-3 days) in more than half of all patients 8. Additionally, delayed ischaemia detected on DWI, associated with vasospasm developing 4-21 days after ictus, may develop in about half of all patients 8.
Digital subtraction catheter angiography remains the gold standard for diagnosis and characterisation of vascular abnormalities, and in many centres, even if the causative lesion is identified on MRA or CTA and it is thought to require surgical management, a catheter study is carried out.
The benefit of DSA is two-fold:
- higher spatial resolution: better able to delineate small vessels and characterise vascular morphology (e.g. aneurysm neck and incorporation of adjacent vessels)
- temporal resolution: contrast can be seen to wash into and out of vascular malformations, giving important information in regards to the feeding vessels (e.g. arteriovenous malformations or dural arteriovenous fistulas)
Additionally, depending on the cause, endovascular therapy (e.g. aneurysm coiling) may be appropriate.
Treatment and prognosis
Treatment will vary according to the underlying cause; however, regardless of the source of subarachnoid blood, a number of treatment principles and potential complications are encountered:
- elevated intracranial pressure
cerebral vasospasm causing delayed cerebral ischaemia
- triple H therapy (Haemodilution, Hypertension, Hypervolaemia)
- calcium channel blockers (e.g. nimodipine)
- endovascular intervention (e.g. intra-arterial delivery of vasodilating agents (such as NO) and/or balloon angioplasty)
- coronary spasm
- neurogenic pulmonary oedema
- pulseless electrical activity (PEA) 13
Prognosis varies greatly depending on:
- cause of subarachnoid haemorrhage
- grade of subarachnoid haemorrhage
- presence of other injuries/pathologies/co-morbidities
A small amount of traumatic subarachnoid haemorrhage or small perimesencephalic blood has an excellent prognosis with little if any significant long-term sequelae. A grade V aneurysmal subarachnoid, on the other hand, has a dismal prognosis, despite aggressive treatment.
It is important to realise that apparent hyperdensity in the subarachnoid space is not pathognomonic of subarachnoid haemorrhage. Other diagnostic possibilities include:
- pseudosubarachnoid haemorrhage
- cortical laminar necrosis: affects the cortex, not the cisterns
- post-thrombectomy iodine extravasation vs haemorrhagic transformation of ischaemic stroke extending to cisterns differentiated on DECT iodine map
- bacterial meningitis
- tuberculous meningitis
- granulomatous meningitis
- 1. Oppenheim C, Domigo V, Gauvrit JY et-al. Subarachnoid hemorrhage as the initial presentation of dural sinus thrombosis. AJNR Am J Neuroradiol. 2005;26 (3): 614-7. AJNR Am J Neuroradiol (full text) - Pubmed citation
- 2. Van gijn J, Rinkel GJ. Subarachnoid haemorrhage: diagnosis, causes and management. Brain. 2001;124 (Pt): 249-78. doi:10.1093/brain/124.2.249 - Pubmed citation
- 3. Sohn CH, Baik SK, Lee HJ et-al. MR imaging of hyperacute subarachnoid and intraventricular hemorrhage at 3T: a preliminary report of gradient echo T2*-weighted sequences. AJNR Am J Neuroradiol. 2005;26 (3): 662-5. AJNR Am J Neuroradiol (full text) - Pubmed citation
- 4. Marder CP, Narla V, Fink JR et-al. Subarachnoid hemorrhage: beyond aneurysms. AJR Am J Roentgenol. 2014;202 (1): 25-37. doi:10.2214/AJR.12.9749 - Pubmed citation
- 5. Brant WE, Helms C. Fundamentals of Diagnostic Radiology. LWW. (2012) ISBN:1608319113. Read it at Google Books - Find it at Amazon
- 6. Berlit P. Diagnosis and treatment of cerebral vasculitis. Ther Adv Neurol Disord. 2010;3 (1): 29-42. doi:10.1177/1756285609347123 - Free text at pubmed - Pubmed citation
- 7. Rohit Sharma, Stephanie Dearaugo, Bernard Infeld, Richard O'Sullivan, Richard P Gerraty. Cerebral amyloid angiopathy: Review of clinico‐radiological features and mimics. (2018) Journal of Medical Imaging and Radiation Oncology. 62 (4): 451. doi:10.1111/1754-9485.12726 - Pubmed
- 8. Van der Kleij LA, De Vis JB, Olivot JM, Calviere L, Cognard C, Zuithoff NP, Rinkel GJ, Hendrikse J, Vergouwen MD. Magnetic Resonance Imaging and Cerebral Ischemia After Aneurysmal Subarachnoid Hemorrhage: A Systematic Review and Meta-Analysis. (2017) Stroke. 48 (1): 239-245. doi:10.1161/STROKEAHA.116.011707 - Pubmed
- 9. Dähnert W. Radiology Review Manual. (2011) ISBN: 9781609139438
- 10. Eden SV, Meurer WJ, Sánchez BN, Lisabeth LD, Smith MA, Brown DL, Morgenstern LB. Gender and ethnic differences in subarachnoid hemorrhage. (2008) Neurology. 71 (10): 731-5. doi:10.1212/01.wnl.0000319690.82357.44 - Pubmed
- 11. Levis JT. ECG Diagnosis: Deep T Wave Inversions Associated with Intracranial Hemorrhage. (2017) The Permanente journal. 21: 16-049. doi:10.7812/TPP/16-049 - Pubmed
- 12. F.H.H. Linn, G.J.E. Rinkel, A. Algra, J. van Gijn. Incidence of Subarachnoid Hemorrhage. (1996) Stroke. 131 (24): e535. doi:10.1161/CIR.0000000000000219 - Pubmed
- 13. Beun L, Yersin B, Osterwalder J, Carron PN. Pulseless electrical activity cardiac arrest: time to amend the mnemonic "4H&4T"?. (2015) Swiss medical weekly. 145: w14178. doi:10.4414/smw.2015.14178 - Pubmed
- 14. Imaging of Hemorrhagic Stroke. (2016) Continuum (Minneapolis, Minn.). 22 (5, Neuroimaging): 1424-1450.