Endovascular treatment of subarachnoid haemorrhage due to ruptured berry aneurysm
Learning outcome
1 Current anaesthetic management of subarachnoid haemorrhage
2 Anaesthesia for interventional neuroradiology
3 Management of acute rupture of aneurysm.
CPD matrix match
2A06, 2A07, 2F03
Case history
A 46-year-old right-handed man presented to his local district general hospital with a history of sudden-onset headache whist exercising in the gym. A CT brain showed subarachnoid blood. His past medical history includes type II diabetes mellitus, treated with metformin, and a smoking habit of 50 cigarettes per day. Despite this, his usual exercise tolerance was more than a mile. He was referred to the regional neurosurgical unit for operative intervention with a GCS of 15. BP ranged from 127/72 to 147/75; pulse was 82 bpm and regular.
A subsequent diagnostic angiogram under local anaesthetic demonstrated that the likely source of the subarachnoid blood was from a ruptured berry aneurysm of the anterior communicating artery.
What is the aetiology and incidence of subarachnoid haemorrhage (SAH)?
The background incidence of spontaneous SAH in the UK is approximately 1 per 10 000 population per annum. Nearly twice as many females as males are affected, and there can be a familial association in some cases. The peak age for aneurysmal SAH is 45–55 years.
Rupture of berry aneurysms accounts for about 80% of cases, with arteriovenous malformations accounting for a further 5%. The remaining 10% is made up of other vascular malformations, non-aneurysmal perimesencephalic bleeds, and vasculitides.
What are the independent risk factors for an aneurysmal subarachnoid haemorrhage?
These include: hypertension, smoking, binge drinking, drugs (cocaine/amphetamines), substantial positive family history (two or more first-degree relatives affected), autosomal dominant polycystic kidney disease, and certain rare inherited collagen vascular disorders.
Independent predictors of poor outcome after an aneurysmal subarachnoid haemorrhage
◆ Reduced GCS on admission (reflected in the World Federation of Neurosurgeons grade)
◆ Increasing age
◆ High-volume blood load on CT scan.
Non-independent predictors
◆ Larger aneurysm size and posterior circulation aneurysms.
What is the usual current intervention for this disease?
Since the 2002 publication of the International Subarachnoid Aneurysm Trial (ISAT), endovascular coiling has become the mainstay of intervention, replacing clipping in the majority of cases where it is technically feasible. ISAT recruited 2143 patients (mean age 52 years) with a diagnosis of SAH secondary to rupture of a berry aneurysm, between 1994 and 2002, and randomized patients to either endovascular coiling or standard clipping. Measured outcomes were the risk of death or dependency at 1 year and were reported as coiling (24%) vs clipping (31%). This represented an absolute risk reduction of 7.4% and a relative risk reduction of 24%.
The follow-up, reported in 2009, recorded 24 rebleeds after 1 year:
◆ Thirteen from target aneurysm (10/13 in the coil arm), p 0.06.
The 5-year mortality was 11% for the coiled group and 14% for the clipped group (p 0.03). The proportion of independent survivors was the same. If the study were reviewed again, as the technology is improving constantly one could argue that follow-up results now would be even better if more advanced coils had been available in 2002.
In the light of ISAT, are there still indications for the clipping of aneurysms? If so, what are they?
There will still be a requirement for the clipping of aneurysms. The major indications for clipping include:
1 No endovascular services available in the region (and transfer to another centre not feasible)
2 The anatomy of the aneurysm precludes coiling (e.g. branch vessels arising off the aneurysm sac; the neck of the aneurysm is too wide, so a coil will not stay in situ; or impossible to position the coil)
3 A simultaneous urgent surgical procedure also required (such as the evacuation of haematoma).
What are the problems associated with a subarachnoid haemorrhage, and what may be done to lessen them?
Hydrocephalus
Ten to 30% of patients will have significantly impaired consciousness, due to an obstructive hydrocephalus. This may require CSF drainage.
Haematoma
After a bleed from an aneurysm, a haematoma may form around the bleeding point. There may be subsequent swelling, causing significant ICP rises. This may necessitate surgical evacuation. There is a theoretical risk of aneurysmal rupture on opening the dura and evacuating the clot. This could reduce the ICP and therefore remove the tamponade effect of the haematoma, thus encouraging a rebleed. Two broad approaches may be considered:
1 The surgeon evacuating the haematoma must be prepared to clip the aneurysm if further bleeding occurs
2 If it can be arranged in time, the aneurysm is protected by coiling, prior to the haematoma evacuation. Obviously, the ICP pressure problems may preclude the time taken to do this.
Vasospasm
Vasospasm (also referred to as delayed ischaemic neurological deficit, DIND) can be clinical or radiological. Clinical vasospasm is the development of neurological deficits, deterioration in the conscious level, or worsening headache. Radiologically, one may see narrowed arteries on angiography. The two entities do not necessarily coincide, and spasm may be acting at a microcirculation level.
Breakdown products from the haemorrhage may cause vasospasm. Classically, the highest incidence for this was between the 4th and 12th day after the bleed, though there was a large degree of interpatient variation. The traditional management was the ‘triple H’ therapy (hydration, haemodilution, hypertension). The rationale was that an adequate fluid load and driving pressure were required to drive blood through the vasospastic vessels. A haematocrit at 0.3 was said to be the optimum compromise between decreased viscosity and good oxygen-carrying capacity.
However, there are caveats with this. Patients may have neurogenic pulmonary oedema or cardiac impairment, precluding an aggressive fluid management. The aneurysm may not be fully protected, or there may be other aneurysms which do not justify treatment but should not be subjected to hypertension.
In practice, many units strive towards the aims of the triple H therapy, but without slavishly adhering to the original protocols, principally by adequate hydration. Good hydration may maintain an adequate perfusion pressure and dilute the haematocrit, but without using an inotrope or respectively requiring venesection. If a patient with delayed cerebral ischaemia secondary to vasospasm fails to respond to the triple H treatment, then balloon angioplasty under general anaesthesia may be considered. Although anecdotal reports indicate a 15–20% absolute benefit from this technique, robust data on its efficacy are lacking (no RCTs), and there is a 2–5% morbidity/mortality rate associated with the procedure.
What are the anaesthetic issues regarding anaesthetizing for endovascular coiling of a subarachnoid haemorrhage?
The procedure itself is not particularly painful or stimulating to the patient. The major indication for general anaesthesia is to guarantee that the patient will not move during the procedure and to enable control of the HR and BP, as required.
The procedure takes place in an IR suite. An example of such a suite is shown in Figure 7.6. There is restricted access to the patient. There are critical points in the procedure where movement would be not just embarrassing to the professional pride of the anaesthetists, but also life-threatening to the patient. These times are:
1 When the microcatheter is being inserted into the aneurysm under digital fluoroscopic control (‘roadmap’)
2 When coils are being deployed in the aneurysm.
Fig. 7.6 Photograph of a typical interventional radiology suite.
The roadmap involves taking an X-ray at one particular point, subtracting all extraneous details from it, and superimposing the real-time image of the guide catheter upon this. It is done to reduce the dose of radiation administered to the patient. However, if the patient moves after the roadmap image has been taken, the catheter will not actually be where it appears on screen and may be advanced inadvertently through the wall of the vessel or aneurysm. That the patient should not move whist the coil is being deployed is intuitively obvious.
The anaesthetist must not move the table, so care should be taken to avoid resting on it. A free-standing drip pole should be used near, though not directly attached to, the table so that fluid bags or infusions can be changed with minimal disturbance. Similarly, a means of administering bolus drugs without touching the table should be made (‘remote injector’).
Either TIVA, based on propofol and remifentanil, or a gaseous technique, using oxygen/air/sevoflurane, are common. The use of infusions of muscle relaxants is controversial; the author’s preference is to use them.
A depth of anaesthetic monitor has proved useful, especially the recent development of the dual hemisphere BIS. In addition to allowing the measurement of the depth of anaesthesia, this development may allow the activity between the hemispheres to be compared.
Case update
After siting a 20 g left radial arterial line under local anaesthetic and after the administration of 2 mg IV midazolam, the patient was anaesthetized with propofol, remifentanil (bolus and infusion), and atracurium (bolus and infusion). In addition to standard monitoring, a bihemisphere BIS monitor was sited, and a urine catheter was placed after induction. BP was stable at 110/60; pulse was 55, and BIS at 40–45.
About 1 hour into the procedure, at the siting of the second coil, the BP rose suddenly to 200/120, pulse to 80, and BIS fell to 6. The anaesthetic chart is shown in Figure 7.7.
Fig. 7.7 The anaesthetic record of the interventional neuroradiology case that is described in the text.
What do you think has happened, and how should this patient be managed?
The most likely event is an acute rupture of the aneurysm. Several actions must be taken quickly. There is usually a hypertensive response to the rupture as in this case. This must be obtunded. Optimize oxygenation. Reverse any administered heparin (in our unit, with protamine). This event, particularly if surgery is required, is the only time protamine use would currently be considered in our unit. The interventional neuroradiologist may wish to rapidly continue packing the aneurysm with as many coils as practicable to occlude the re-ruptured aneurysm. Our usual practice is to obtain an urgent CT head scan to confirm the diagnosis and to assess the extent of the bleed and damage and also if there are any treatable complications such as an obstructive hydrocephalus which may require drainage.
If an external ventricular drain (EVD) is indicated, then the next step is to take the patient to theatre urgently.
Which problems may present around the time of the procedure? How should these problems be managed?
◆ Rupture of the aneurysm on the table: as discussed previously
◆ Delayed return of consciousness: exclude a pharmacological reason
◆ Consider a thromboembolic event: consider Reopro® or other potent antiplatelet agents
◆ Consider vasospasm: consider glyceryl trinitrate (GTN) or nimodipine intra-arterially (not a licensed use) ± angioplasty
◆ A further management point: at the preoperative assessment stage, it is worthwhile stressing to the patient that their headache will not instantaneously disappear on wakening from their general anaesthetic.
Case update
The patient had an urgent CT scan of head whist under the same general anaesthetic. To facilitate transfer to the CT scanner, then to theatre, and ultimately to ICU, a propofol infusion was substituted for sevoflurane, although the remifentanil and atracurium infusions were continued. The patient was taken to theatre, and an intracranial bolt for pressure monitoring reasons was inserted. The initial ICP was 75 mmHg. A ventricular access device was inserted, and the ICP fell to 6 mmHg almost immediately. The patient was then taken to intensive care for sedation and ventilation.
Sedation was lightened 24 hours later, and he was subsequently extubated, although reintubated 72 hours later for a further 5 days. Four days after the final extubation, he was discharged from the ICU to the neuro HDU, with a GCS of 13 (E4 V3 M6).
After a further 2 weeks on the regional neurosurgical ward, he was discharged to the referring district general hospital for continuing rehabilitation. GCS on discharge was 14, due to confusion. He was scheduled for follow-up in the neurovascular clinic at 6 months.
Summary
The management of subarachnoid treatment has been revolutionized in recent years since the widespread introduction of endovascular treatment. This has taken the management out of the operating theatre and into the radiology suite. Although the procedure is less invasive, the anaesthetic technique should be no less vigilant, as the patient must not move. The anaesthetic technique should allow the administration of drugs and fluids without causing movement of the X-ray table or patient. This may involve mounting infusion pumps and fluids on a free-standing pole and having a remote injection set for bolus drugs.
Although the large majority of subarachnoid haemorrhages (SAHs) are now managed by an endovascular technique, there will still be a requirement for clipping of aneurysms.
Dorairaj IL and Hancock SM (2008). Anaesthesia for interventional neuroradiology. Continuing Education in Anaesthesia, Critical Care & Pain, 8, 86–9.
Molyneux A,Kerr RSC, Birks J, et al. (2009). Risk of recurrent subarachnoid haemorrhage, death or dependence and standardised mortality ratios after clipping or coiling of an intracranial aneurysm in the International Subarachnoid Aneurysm Trial (ISAT): long-term follow-up. Lancet Neurology, 8, 427–33.
Molyneux A,Kerr R,Stratton I, et al. (2002). International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trial. Lancet, 360, 1267–74.