Secondary Complications of Subarachnoid Haemorrhage

A thirty eight year old female smoker was admitted via A+E following sudden onset occipital headache with visual disturbance and collapse with loss of consciousness lasting approximately five minutes. She had complained of unusual headaches a week prior to this event, but these were short lived and not associated with any neurology. On arrival in resus she had recovered to a Glasgow coma score (GCS) of 14/15. She demonstrated neck stiffness and photophobia, as well as general irritability. Plain computerised tomography
scan (CT) performed showed a subarachnoid haemorrage in the region of the middle cerebral artery, with the presence of blood in the sylvian fissure.

She was transferred to the ITU for monitoring and blood pressure control with invasive arterial and central venous pressure monitoring. She was treated with nimodipine to prevent vasospasm. Contrast CT performed showed an aneurysm at the bifurcation of the middle cerebral artery, and this was felt to be the origin of the bleed. She underwent uneventful endovascular coiling of this aneurysm the following day under general anesthesia, and was discharged to the neurosurgical team for ongoing care afterwards.

What are the secondary complications of subarachnoid haemorrhage and how are they managed?

David Hepburn

Subarachnoid haemorrage (SAH) can be either traumatic or spontaneous as in this case. Spontaneous SAH are commonly caused by arteriovenous malformations or cerebral artery aneurysms. Uncommon causes of
spontaneous SAH include neoplasms, infections and venous angiomas. The vast majority (>70%) of spontaneous bleeds are due to aneurysms. These are present in approximately 1% of the population and are more common in females than males. 10 to 15% of patients presenting with aneurysmal SAH will have multiple aneurysms. As in this case, patients present with a warning headache or “sentinel bleed” in approximately 50% of SAHs. These are usually followed by a larger more catastrophic haemorrage, classically described as sudden onset “thunderclap headache” which is accompanied by collapse and loss of consciousness. This carries a mortality of around 50%. Patients who do not die may have permanent neurological deficit although effective treatment reduces this chance.

First line investigation is plain CT scan. This has 90% sensitivity at 24 hrs and can also detect intracerebral bleeding, mass effect and hydrocephalus. False-negatives can occur when bleeds are of small volume. If CT scan is negative but clinical suspicion persists, a lumbar puncture can be diagnostic. Classically a positive Xanthochromia index occurs in SAH but this is a later sign. definitive diagnosis can be gained via contrast CT angiography, which will locate any aneurysm as well as source of the haemorrage. If multiple
aneurysmal areas are found, treatment is usually targeted towards the lesion most adjacent to the bleed.

Treatment is SAH is directed at two problems – eliminating the causative lesion such as an aneurysm and prevention of secondary complications. Major complications of SAH include rebleeding, vasospasm and
hydrocephalus.

Rebleeding is usually prevented by treatment of the causative lesion. This can be performed surgically, clipping the aneurysm or by interventional radiology packing the aneurysm with endovascular coils which lead to thrombosis and remove the aneurysm from circulation. The major trial contributing to the evidence base for these treatments was the International subarachnoid aneurysm trial (ISAT) published in 2002 (1). They reviewed the cases of 2143 patients with ruptured intracranial aneurysms, who were randomized to receive either endovascular coiling (n= 1070) or neurosurgical clipping (n=1073). They demonstrated that 23.7% patients allocated endovascular treatment were dependent or dead at 1 year compared with 30.6% allocated neurosurgical treatment (p=0.0019). They concluded that the relative and absolute risk reductions in dependency or death after allocation to coiling versus clipping were 22.6% (95% CI 8.9-34.2) and 6.9% (2.5-11.3), respectively. They found that the rates of rebleeding were slightly less in the clipping group.

Post SAH hydrocephalus can be difficult to prevent. Approximately 20% of SAH patients will experience hydrocephalus. This can either be due to communicating or obstructive hydrocephalus. In communicating
hydrocephalus there is no obvious obstruction to CSF flow, and it is postulated that increased CSF occurs as a result of either higher turnover and production or diminished absorption triggered by the action of blood on arachnoid villi. In obstructive hydrocephalus thrombus blocks the flow of CSF usually at the level of the third or fourth ventricle or the aqueduct of Sylvius. It can be treated with extraventricular drainage, VP shunt or lumbar drain depending on its aetiology.

Vasospasm is a major cause of post-SAH morbidity and mortality. The vascular structures of the brain are easily irritated by extrinsic presence of blood and this can prompt marked arterial vasospasm leading to ischaemia and infarction. Estimates suggest up to 30% SAH patients experience some vasospasm. Peak occurrence is between day 3 and day 10, and it is more likely with increasing age (partly due to diminished tolerance to brain ischaemia) and clot size. It can be prevented by a number of means. Traditionally “3H” therapy has been advocated. This stands for haemodilution (to haematocrit of 30-35%), hypertension (maintaining CPP with vasopressors if required) and hypervolaemia (using crystalloid or colloid to achieve CVP or PCWP of >12 cmH2O). Close monitoring is required to recognize and treat vasospasm as soon as it occurs.

Calcium antagonists such as nimodipine and nicardipine are routinely used to prevent vasospasm by relaxing arterial smooth muscle. Nimodipine in particular has a relatively cerebro-selective effect. (2). A British trial (3) randomizing patients to oral nimodipine or placebo demonstrated in 540 patients that the rate of ischaemic secondary events was reduced from 30% to 22% in the nimodipine group. Meta-analysis performed by the Cochrane collaboration in 2006 (4) collected data from randomized trials of Calcium antagonists including nimodipine in patients following acute SAH. 12 trials totalling 2844 patients with SAH (1396 in the treatment group and 1448 in the control group) were included. They found that overall, calcium antagonists reduced the risk of “poor outcome” (they defined this as death or severe dependency): the relative risk (RR) was 0.82 and the absolute risk reduction was 5.1%, so the corresponding number of patients needed to treat to prevent a single “poor outcome event” was 20.

Vasospasm occurring despite treatment with prophylactic calcium antagonists must be treated aggressively (5). Angioplasty and local intravenous papaverine have both been used effectively. (6) There is a suggestion that the vasospastic state may be associated with the high catecholamine release typically seen in patients with brain injury (which is known to precipitate neurogenic pulmonary oedema and myocardial ischaemia) (7) and thus it has been postulated that there may be a role for beta blockade in secondary prevention after SAH. This has yet to be established experimentally, but warrants further analysis.

Lessons Learnt

Current treatment for aneurysmal SAH relies on close monitoring and preempting secondary complications. The strongest evidence exists for use of oral nimodipine and the use of endovascular coiling and neurosurgical clipping. Therapy has changed little over the last ten years since these therapies became established practice. There remain a number of unexplored areas in which gains could potentially be made in outcomes for what is sudden and often devastating neurological catastrophe.

References