Retinal artery occlusion during carotid artery stenting with distal embolic protection device

Abstract

Retinal artery occlusion associated with carotid artery stenosis is well known. Although it can also occur at the time of carotid artery stenting, retinal artery occlusion via the collateral circulation of the external carotid artery is rare. We encountered two cases of retinal artery occlusion that were thought to be caused by an embolus from the external carotid artery during carotid artery stenting with a distal embolic protection device for the internal carotid artery. A 71-year-old man presented with central retinal artery occlusion after carotid artery stenting using the Carotid Guardwire PS and a 77-year-old man presented with branch retinal artery occlusion after carotid artery stenting using the FilterWire EZ. Because additional new cerebral ischaemic lesions were not detected in either case by postoperative diffusion-weighted magnetic resonance imaging, it was highly likely that the debris that caused retinal artery occlusion passed through not the internal carotid artery but collaterals to retinal arteries from the external carotid artery, which was not protected by a distal embolic protection device. It is suggested that a distal protection device for the internal carotid artery alone cannot prevent retinal artery embolisation during carotid artery stenting and protection of the external carotid artery is important to avoid retinal artery occlusion.

Introduction

In recent years, evidence of carotid artery stenting (CAS) has been established with the advances in endovascular treatment, and it is widely performed.^1,2 Compared to carotid endarterectomy, one problem of CAS is the higher frequency of distal embolisation, such as cerebral embolisation during the surgical procedure.³ Therefore, many embolic protection devices (EPDs) have been developed and employed during CAS to prevent distal embolisation.

Retinal artery occlusion (RAO) has also been known to occur in CAS, and various causes have been reported.^4–9 According to previous reports, the prevalence of all retinal artery embolisation during CAS is 5–16.9% and that of symptomatic RAO is 1.4–1.7%.^5,7,8 Thus central RAO and branch RAO are not rare complications of CAS; these are very severe complications because they are difficult to treat and result in a decline in activities of daily living caused by the irreversible visual disturbance.

We encountered a case of central and branch RAO during CAS using a distal EPD for the internal carotid artery (ICA) alone, the Carotid Guardwire PS (Guardwire; Medtronic, Minneapolis, MN, USA) and the FilterWire EZ (FilterWire; Boston Scientific, Marlborough, MA, USA).

Case report

Case 1

A 71-year-old man presented with right hemiparesis, dysarthria, right superior quadrantanopsia and ideational apraxia. He had a medical history of hypertension, and had no history of smoking but consumed alcohol almost every day. Diffusion-weighted magnetic resonance imaging (MRI-DWI) revealed a high-intensity area in the left posterior limb of the internal capsule and the left parietal lobe (Figure 1(a and b)). Magnetic resonance angiography showed severe left ICA stenosis (Figure 2(a)). Left carotid ultrasonography demonstrated 81% stenosis according to the criteria of the European Carotid Surgery Trial.

Figure 1.

Preoperative diffusion-weighted magnetic resonance imaging (MRI-DWI) shows a high-intensity area in the left posterior limb of the internal capsule and the left parietal lobe (a, b). Postoperative MRI-DWI shows no new infarction (c, d).

Figure 2.

Magnetic resonance angiography shows severe stenosis and decreased blood flow in the left internal carotid artery (ICA) (arrowhead) (a). Left carotid angiography (b) shows severe stenosis and ulcer formation (arrowhead). Carotid artery stenting was performed under distal balloon protection (c) and good dilation of the left ICA was achieved (d).

The left CAS was selected for symptomatic ICA stenosis because the plaque was in the distal part of the ICA. A 9 Fr Super Sheath (Medikit, Tokyo, Japan) was introduced into the right common femoral artery while keeping the patient under local anaesthesia. Initially, a 5000 IU heparin bolus was administered intravenously and then additional heparin was administered to maintain an active clotting time greater than 300 seconds. A 9 Fr Optimo guiding catheter (Tokai Medical Products, Nagoya, Japan) was advanced to the left common carotid artery (CCA); common carotid angiography demonstrated severe and long ICA stenosis with ulceration (Figure 2(b)). A guardwire was advanced to the distal ICA. No neurological change was detected at this point. Balloon of the guiding catheter was inflated only while crossing the stenotic lesions to prevent ischaemic attack caused by intolerance. Subsequently, precise function of the distal protection was demonstrated by inflation of the guardwire (Figure 2(c)). After intravenous injection of 0.5 mg atropine sulfate, a 3.0 × 40-mm Sterling angioplasty balloon (Boston Scientific) was used for pre-stent dilation (6 atm × 30 s). Then, a 6.0 × 30-mm Precise stent (Johnson & Johnson, New Brunswick, NJ, USA) and a 9.0 × 40-mm Precise stent (Johnson & Johnson) were placed as they overlapped each other. A 4.0 × 20-mm Aviator Plus angioplasty balloon (Johnson & Johnson) was used for post-stent dilation (10 atm × 30 s). An aspiration catheter was inserted just below the distal balloon on the guardwire, and approximately 80 mL blood was aspirated. After confirmation that there was no debris in the aspirated blood, the guardwire was deflated. Post-stenting common carotid angiography revealed successful stenting and no abnormalities of the arteries, including the left ophthalmic artery (Figure 2(d)).

Soon after the CAS procedure, decreased visual acuity in the left eye with almost no light perception was observed. There were no changes in other neurological signs. Central RAO was diagnosed by an ophthalmologist (Figure 3), and visual acuity did not improve after ocular massage and continuous anticoagulant therapy. Post-stenting MRI-DWI showed no new cerebral ischaemic lesions (Figure 1(c and d)).

Figure 3.

The central part of the retina of the left eye shows edematous change and a ‘cherry red spot’. R: right, L: left.

Case 2

A 77-year-old man presented with hypesthesia in the left corner of the mouth and in the left upper limb. The patient had a history of type 2 diabetes mellitus and had a history of smoking for 40 years. MRI-DWI showed a diffuse high-intensity area in the right parietal lobe (Figure 4(a and b)) and magnetic resonance angiography showed right ICA stenosis (Figure 5(a)). Right carotid ultrasonography demonstrated 76% stenosis according to the criteria of the European Carotid Surgery Trial and ulceration.

Figure 4.

Preoperative diffusion-weighted magnetic resonance imaging (MRI-DWI) shows a high-intensity area in the right parietal lobe (a, b). Postoperative MRI-DWI shows no new infarction (c, d).

Figure 5.

Magnetic resonance angiography shows a stenotic lesion in the right internal carotid artery (ICA) (arrowhead) (a). Right carotid angiography (b) shows severe stenosis of the ICA and external carotid artery. Carotid artery stenting was performed under distal filter protection and no flow was detected (c). No flow resolved after collecting the distal filter device (d).

For symptomatic ICA stenosis, right CAS was performed 34 days after the occurrence of cerebral infarction. An 8 Fr sheath (Terumo, Tokyo, Japan) was introduced into the right common femoral artery while the patient was under local anaesthesia. Initially, a 5000 IU heparin bolus was administered intravenously and then additional heparin was administered to maintain an active clotting time greater than 300 seconds. A 8 Fr Launcher (Medtronic) was advanced to the right CCA, and common carotid angiography revealed irregular ICA and external carotid artery (ECA) stenosis (Figure 5(b)). A FilterWire was advanced to the distal ICA and deployed. No neurological change was detected at this point. After intravenous injection of 0.5 mg atropine sulfate, a 3.0 × 40-mm Jackal RX angioplasty balloon (St Jude Medical, St Paul, MN, USA) was used for pre-stent dilation (8 atm × 30 s). Then a 10.0 × 24-mm Carotid Wallstent (Boston Scientific) was placed. A 4.0 × 30-mm Jackal RX angioplasty balloon (St Jude Medical) was used for post-stent dilation (10 atm × 30 s). Right CCA showed no flow of the right ICA soon after the first stenting (Figure 5(c)). Because it was suspected that the proximal portion of the ICA was linearised by the stent and the distal portion of the ICA became kinked, an additional 8.0 × 21-mm Carotid Wallstent (Boston Scientific) was placed distal to the first stent as it overlapped with the first stent and resolved the kink. The FilterWire was retrieved and a considerable amount of debris was found in the retrieved filter device. Post-stenting right CCA showed improved flow in the right ICA and no abnormalities of arteries, including the right ophthalmic artery (Figure 5(d)).

Soon after the CAS procedure, the patient complained of blurred vision in the right eye and was then diagnosed with branch RAO (Figure 6). Post-stenting MRI-DWI showed no new cerebral ischaemic lesions (Figure 4(c and d)).

Figure 6.

Edematous change is observed outside the retina of the right eye (black arrow). R: right, L: left.

Discussion

RAO during CAS can be caused by debris that is not captured using an EPD, debris that occurs when an EPD passes through the lesion or when a guiding catheter is advanced, and debris passing through the collaterals from the ECA to the retinal arteries. In our cases, it was highly likely that the debris passed through collaterals from the ECA, which had not been protected by an EPD, to retinal arteries and caused RAO because post-stenting MRI-DWI showed no new cerebral ischaemic lesions in either of the cases. However, it cannot be denied that both the cases could have been caused during other procedures. One way to find the temporal nature is to do a visual/neuro examination when placing an EPD. Furthermore, debris causing central RAO is 20 µm or more. Because the mesh size of the filter wire is 110 µm, debris can flow to the distal ICA through the mesh.⁸

Currently, devices such as distal filter, distal balloon (distal EPD for the ICA alone) and proximal balloon (proximal EPD for the CCA alone) are being used as EPDs, and significant statistical differences between them have not been established regarding their merits and demerits.^10,11 A report on the number of new cerebral ischaemic lesions found on MRI-DWI after CAS did not show statistical differences between the distal balloon and distal filter.¹² Several reports showed that, compared to the distal filter, the proximal balloon significantly reduced the number of cerebral embolic lesions after CAS.^13–15

In addition to the main blood supply to the retina through the ophthalmic artery, it is well known that collaterals can supply blood from the middle meningeal arteries as branches from the ECA to retinal arteries. Flushing debris in the distal balloon-occluded ICA towards the ECA was associated with a higher incidence of retinal artery embolisation than when no flushing towards the ECA was performed.⁸ These reports suggest that using a proximal occlusion device attached to an ECA protection balloon, such as the Mo Ma Ultra (Medtronic), and using a combination of proximal balloon protection with a balloon guiding catheter and distal balloon device can reduce the incidence of distal embolisation, retinal embolisation in particular, compared to using a distal protection device alone. However, these systems may not be useful in the presence of contralateral occlusion and a large proximal superior thyroid artery branch.¹⁶

Ulceration is considered to be a feature of the vulnerable plaque,¹⁷ and visible debris was captured during CAS more frequently in the vulnerable than in the non-vulnerable plaque groups.¹⁸ A report indicated that ulceration was a risk factor of retinal artery embolisation,⁵ and ulceration was observed in both of our cases. A past medical history of radiotherapy to the cervical part, which was described as a risk factor for carotid endarterectomy in the Stenting and Angioplasty with Protection in Patients at High Risk for Endarterectomy (SAPPHIRE) study,² was reported to be related to the frequent occurrence of retinal artery embolisation.⁷ Lesions longer than 20 mm are more often associated with strokes than shorter lesions (from CREST, EVA-3S). Therefore, long lesions may also be a risk factor for retinal artery embolisation.^1,19

Regarding the type of stent and distal embolisation, post-stenting new cerebral ischaemic lesions that were detected on MRI-DWI occurred more frequently in the open cell than in the closed cell type groups.³ However, there is no statistical difference between the groups in the incidence of retinal artery embolisation.⁷ In our cases, an open-cell type stent was used in case 1 and a closed-cell type stent was used in case 2, and no significant protrusion and no new cerebral embolisation was observed in either of the cases.

Conclusion

We encountered two cases of RAO after CAS using distal protection alone, a distal balloon device and a distal filter device. Both devices were functioning precisely during the procedure. It is suggested that a distal protection device alone cannot prevent retinal artery embolisation during CAS for carotid artery stenosis with ulceration. RAO causes severe and irreversible visual disturbance. Therefore, we believe that protection of the ECA or the proximal CCA will also be necessary to avoid retinal artery embolisation during CAS.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.