Transradial Woven EndoBridge device to treat a ruptured basilar tip aneurysm under conscious sedation

Abstract

We present a patient with a ruptured basilar tip aneurysm treated under conscious sedation via a right radial approach using the Woven EndoBridge device. We discuss the technical aspects of the case as the well as the available literature on the treatment of ruptured aneurysms treated without general anaesthesia. We discuss the potential benefits of the Woven EndoBridge device in such situations.

Introduction

The endovascular approach to the treatment of ruptured aneurysms has largely replaced neurosurgical clipping and this is especially true of basilar tip aneurysms. Typically patients are treated whilst under general anaesthesia (GA) as it is likely to decrease the risk of the procedure, potentially decrease the radiation exposure and improve image quality, However, in certain instances GA may not be suitable and in these situations aneurysm embolisation can be performed with conscious sedation and/or local anaesthesia (LA). As with procedures performed under GA, coil embolisation under conscious sedation and/or LA has been the preferred treatment strategy and is well documented.^1–4 To our knowledge there are no documented cases on the treatment of ruptured aneurysms under conscious sedation with the Woven EndoBridge (WEB) (MicroVention, Tustin, CA, USA).

Here we present the first documented case of a WEB device deployed in a ruptured basilar tip aneurysm under conscious sedation. We discuss the potential advantages of the WEB device for patients with ruptured aneurysms that may not be suitable for GA and review the literature regarding the treatment of ruptured aneurysms without GA.

Case description

A 63-year-old female patient was admitted to our tertiary referral unit via the emergency department with a 72 h history of acute onset headache, photophobia and signs of meningism. She had a background history of severe COPD that resulted in restricted mobility and the requirement for ambulatory home oxygen. On admission she was apyrexial and in sinus rhythm with oxygen saturation on room air of 92%. A subarachnoid haemorrhage was suspected and confirmed on unenhanced CT scan (Figure 1(a)) with a solitary aneurysm arising from the basilar tip (Figure 1(b) and (c)). Although there was no significant haemorrhage located around the aneurysm, it was felt that this was likely due to re-distribution of the blood as a result of the delayed presentation. After discussion with the neurosurgical team, endovascular treatment was decided upon. Discussion with the anaesthetic team raised several concerns regarding suitability for GA. The patient’s exercise tolerance was extremely poor with mobility limited to walking in her house. She required ambulatory oxygen, home nebulisers, and she required three pillows to sleep when well from a respiratory perspective. Her baseline oxygen saturation on room air was 92% and her most recent lung function tests showed a FEV 1 of 0.89 (40% predicted) and FVC of 2.95 (112% of predicted) consistent with obstructive lung disease. The main concern regarding GA was that there would be accompanying lung atelectasis with the likelihood of hypoxia post-procedure given the lack of respiratory reserve. Another worry was causing post-procedure bronchospasm that would also contribute to respiratory failure. The probability that she would require post-procedural respiratory support was high and in the worst-case scenario may require ventilation on the intensive care unit with the potential for significant complications. As with any patient with severe respiratory disease, the risks of GA were significant, performing the case under LA and conscious sedation may provide an overall safer treatment option.

Figure 1.

A plain CT head shows sub-arachnoid haemorrhage over the left frontal lobe (a). This was thought to be due to distribution of the blood within the cerebrospinal fluid secondary to the delayed presentation. A CT angiogram shows a solitary basilar tip aneurysm ((b) and (c)).

In the angio-suite the patient was positioned with a single pillow behind her shoulders and her head in a headrest, and given humidified oxygen at 30 l/min via Optiflow nasal cannulae. A head strap was gently placed across the forehead (Figure 2(a)). The arm was gently restrained by wrapping a folded sheet around the elbow and then tucked under the lower back (Figure 2(b)) with the thumb taped to the armrest to help maintain supination (Figure 2(c)). Conscious sedation was achieved using 2 mg IV bolus of Midazolam and a Remifentanil target controlled infusion via an injectomat TIVA pump (Fresnius Kabi, Homburg, Germany) set on Minto model and targeting a plasma concentration ranging between 3 and 4 ng/ml that was started just prior to the puncture. The patient was alert and responsive to commands with a respiratory rate of 8–12 breaths/min and oxygen saturation of 100%.

Figure 2.

To prove gentle restraint of the head we use a head strap and the dedicated headrest (a). To aid with comfort we place a rolled sheet or gel pad beneath the arm and then tape the thumb to the side of the armrest to ensure comfortable supination of the hand. In order to try and limit movement, we loop a folded sheet around the elbow that is then tucked the lower back of the patient ((b) and (c)). These strategies help to maintain the arm in a comfortable and ‘relaxed’ restraint. Similarly, we avoid tight restraint of the head as this can cause agitation for some patients.

With LA and using ultrasound guidance, a 6Fr Glidesheath (Terumo) was inserted into the right radial artery following which 5000 IU of heparin, 200 mcg of Isoket and 2.5 mg of verapamil were given intra-arterially through the radial sheath. Subsequently, a 6Fr Benchmark (Penumbra Inc., Almeida, CA, USA) guide catheter was tracked into the right vertebral artery. Angiography confirmed the basilar artery aneurysm and after rotational angiography appropriate angles for embolisation were selected (Figure 3(a)). The aneurysm measured approximately 6.5 mm in maximal height and 4.5 mm in maximal orthogonal diameter and based on these measurements we opted to implant a 6 mm × 4 mm WEB SL. After successful catheterisation of the aneurysm with a VIA 17 (MicroVention, Tustin, CA, USA) microcatheter and Synchro 2 microwire (Stryker Neurovascular, Kalamazoo, MA, USA), the WEB was successfully implanted and detached without complication (Figure 3(b)). Delayed angiography showed minimal contrast opacification at the neck of the aneurysm and no evidence of thrombus formation on the surface of the WEB or within the daughter branches (Figure 3(c)). As the detachment zone of the WEB was within the parent artery, it was decided to start the patient on a four-week course of aspirin with an oral loading dose of 500 mg given at the end of the procedure. The patient remained at baseline neurology throughout the procedure and a StatSeal and TR Band were used to achieve patent haemostasis at the puncture site.

Figure 3.

Catheter angiography confirmed the presence of the basilar tip aneurysm (a), which was approached via the right vertebral artery. After deployment of the WEB (b) within the aneurysms, there was virtually complete lack of contrast opacification within the dome of the aneurysm and only minimal contrast opacification seen at the neck (c). There was no evidence of thrombus formation at the base of the WEB and no evidence of thromboembolic complication.

To ensure adequate occlusion of the aneurysm, intravenous digital subtraction angiography was performed at 72 h post-embolisation and this confirmed occlusion of the aneurysm (Raymond Roy class I) (Figure 4).

Figure 4.

An IV flat panel DSA performed 72 h post-operatively did not demonstrate any evidence of opacification within the aneurysm. Both the posterior cerebral arteries remained patent with no displacement of the WEB and no evidence of thrombus formation.

There was no subjective or objective deterioration in respiratory function or any other systemic complications from either the procedure or the sedation. She was discharged home seven days later and is awaiting delayed MR imaging.

Discussion

This case illustrates several developments in the endovascular treatment of intracranial aneurysms: the use of LA with conscious sedation, transradial access and intrasaccular flow disruption with the WEB device. To the authors’ knowledge, this is the first time these advancements have been described concurrently in a single procedure.

The use of LA or conscious sedation has been studied in the setting of both unruptured and ruptured intracranial aneurysm embolisation.^1–4 Qureshi et al.² were the first to document the use of coils to treat intracranial aneurysms under local anaesthesia. In this study, 150 endovascular coiling procedures were reported including 58 procedures for ruptured aneurysms. The aneurysms treated in this series were located in both the anterior and posterior circulation; however, the most common location was the basilar artery (n = 14). The average number of coils used was 4.6 ± 4.3. The procedure was completed without complication in 48 (83%) of the 58 procedures. In five cases, the procedure was completed, however there were complications and in two of these cases, GA was required. In five cases, the procedure was aborted because of morphological characteristics that that made embolisation technically difficult (n = 2), aneurysm rupture (n = 2) and the appearance of new transient neurological deficit (n = 1). In total, of the 10 cases that had complications, intra-operative aneurysm rupture was the single most common complication (n = 6, 60%) with morphological issues comprising the next most frequent cause of complication or abandonment of the LA procedure. The procedure related morbidity was 1.6% and mortality was 3%; however, the authors did not report the individual results of patients and therefore it is difficult to ascertain whether the deaths occurred secondary to rupture or other causes. This initial study demonstrated that endovascular coiling of aneurysms, both ruptured and unruptured, was possible and carried an acceptable safety profile in comparison to other studies.⁵ More recently, Song et al. reported their results on the treatment of ruptured aneurysms with respect to the depth of anaesthesia. In this retrospective study of 183 patients with 183 aneurysms, 80 (43.7%) of which were in the anterior circulation and 68 (39.9%) were <5 mm, were treated successfully. Ninety-seven (53%) patients were Hunt and Hess grade 1 or 2, 52 (28.4%) were grade 3 or 4 and 34 (18.6%) were grade 5. In total, 103 patients underwent treatment with LA or conscious sedation, 78 patients were treated whilst under deep sedation and only 2 patients required GA. The overall complication rate was 8.7% (n = 16) with 14 of these complications being symptomatic. The majority of the complications in this series were thromboembolic (n = 11, 69%) and the remainder (n = 5, 31%) being haemorrhagic. Of the patients with haemorrhagic complications, three occurred in patients under conscious sedation and two were in patients under deep sedation. One of the patients required a craniotomy due to raised intracranial pressure, two patients had a clinical deterioration and two patients died. At last follow-up (average 12.4 months) 21 patients had died (11.5%) and of these 21 patients, two had an intra-procedural rupture and one had a thromboembolic complication. The mortality rate in the patients treated under LA or with conscious sedation was 1% (n = 1). These results provide further evidence that the treatment of patients with ruptured aneurysms can be performed without the need for GA. The lower mortality seen in patients treated under LA and conscious sedation is likely to reflect the fact that 76.3% of this group (n = 74/97) presented with Hunt and Hess grade 1 or 2 haemorrhages. Similar studies by other authors have shown low rates of intra-procedural complications⁶ and low rates of conversion to GA in carefully selected patients.⁷ In our case, we believe that at least five coils would be required in order to achieve satisfactory occlusion of the aneurysm, and given the relatively wide neck of the aneurysm, balloon or stent assistance and possibly permanent stent implantation would be required. Similarly, the early results from the CLARYS study have shown that the WEB can provide robust protection for ruptured aneurysms⁸ although longer-term follow-up data are awaited. For these reasons, it was felt that a WEB may provide a simpler and faster treatment option than coiling. Rupture of the aneurysm during WEB implantation has been noted; however, it does not appear to occur commonly. In the series of Van Rooij et al.⁹ that used exclusively the VIA 17 system, there were no intra-procedural ruptures. In previous reports^10–15 that included 346 aneurysms in 341 patients, rupture occurred in nine patients (2.6%) and of the nine ruptures noted in all series, seven occurred on the anterior communicating or posterior communicating arteries and five of the ruptures were caused by the microcatheter tip before WEB deployment. Given the straightforward access and ability to use a VIA 17 system for the deployment of the WEB in our case, we felt the chances of intra-operative rupture were low and in all likelihood lower than placing several coils. Additionally, had a rupture occurred full deployment of the WEB would hopefully have resulted in rapid cessation of the haemorrhage^11,16 and in such a scenario, we were willing to accept a neck remnant. In preparation for an intra-operative rupture, we did not load the patient with anti-platelet agents pre-operatively. Thromboembolic complications have also been reported in up to 25% cases of ruptured aneurysms treated with the WEB device¹³ and in the event of such complications, anti-platelet agents have been used with success. We chose to administer a short course of aspirin after the deployment of the WEB; in our case, however, there was no evidence of thrombus formation on the surface of the WEB or within any of the arteries. We believe that one of the main advantages of the WEB is that it can be easily recaptured and repositioned and that there is a 10 min time window in which to do this. Similarly, the device can be manipulated once fully deployed, e.g. it can be repositioned slightly even after full deployment. For this reason, we would advocate a ‘deploy and see’ strategy. For example, if the patient were to move during the deployment, we would advocate deploying the WEB fully and then performing angiography to ensure that the position was satisfactory. If there was significant movement then there should be enough time to perform a rotational angiogram and check the position of the device as well as the status of the parent and daughter branches.

It is well documented that induction and maintenance of GA is associated with alterations in cardiopulmonary physiology that may lead to complications including cardiac failure, bronchospasm, myocardial ischaemia, myocardial infarction and respiratory failure.¹⁷ Predictors for severe adverse cardiovascular or respiratory outcomes with GA include ASA status 3 or 4, age > 50 years, a history of cardiac failure or myocardial infarction ≤1 year, COPD, obesity and smoking.¹⁷ The use of LA and/or conscious sedation minimises these risks. Additional benefits may include shorter procedural time, lower procedural cost and shorter hospital stay;^1–3 however, these were not our primary concerns with regard to this case. The risk of respiratory failure and prolonged ventilation on the intensive care unit with the associated complications were the principal reasons for avoiding GA. The main disadvantage of conscious sedation is the potential for decreased image quality due to patient motion. Also in case of an intra-operative complication or patient agitation, rapid conversion to GA can be required, which may be challenging.^1,2 Having the equipment and medication for conversion to GA prepared is essential. Gently fixing the head using straps can reduce the motion artefact. We believe that it is important to titrate the sedation so that the patient is comfortable, but remains calm and can follow instructions if needed. In our opinion, the best way to achieve this is by sedating the patient to the appropriate level before starting the procedure and then to maintain this level of sedation throughout. For this reason, we do not advocate varying the sedation during the procedure as this can have unintended consequences, e.g. increased agitation, etc. Rather optimising the level of sedation once the patient is in the angio-suite and prior to cleaning, draping or gaining access should be the goal. However, during the procedure it is prudent to ask the patient if they remain comfortable and if they can ‘feel’ any intracranial sensations. If the patients can feel manipulation of the microwire and microcatheter prior to catheterisation of the aneurysm then titrating the degree of sedation up would seem prudent. There should then be a short period of observation to ensure the patient remains calm and able to follow instructions before the aneurysm is catheterised and the WEB or other devices are deployed.

Transradial access offers multiple advantages over traditional femoral access, which can largely be ascribed to the anatomy. The location of the puncture site at the wrist enables effective compression against the volar aspect of the distal radius. This reduces the incidence of bleeding complications and allows angiography or intervention to be performed in the setting of anticoagulation, thrombolytic or antiplatelet agents, without the need for intra-vascular closure devices or prolonged bed rest.¹⁸ The close proximity of the ipsilateral vertebral artery permits ready access to the vertebrobasilar system. In addition, the relatively small diameter of the vessels allows stable guide catheter position, as compared with the large diameter aortic arch in femoral procedures.¹⁸ Patients are able to sit immediately post-procedure rather than staying supine as a femoral arterial puncture mandates. This is hugely advantageous to any patients with advanced respiratory pathology as in this case.

Conclusion

We believe this the first case to report the use of the WEB device to treat an acutely ruptured aneurysm in a patient treated whilst under conscious sedation via a right transradial approach. This option should be considered in patients unfit for GA or in whom more complicated coiling procedures would be needed.

ORCID iDs

P Bhogal https://orcid.org/0000-0002-5514-5237

N Chan https://orcid.org/0000-0002-0153-0199