Article Text

Original research
Ultrasound for the evaluation of stenosis after flow diversion
1. Cameron M McDougall1,2,
2. Khurshid Khan3,
3. Maher Saqqur3,
4. Andrew Jack1,
5. Jeremy Rempel4,
6. Carol Derksen3,
7. Yin Xi2,
8. Michael Chow1
1. 1Division of Neurosurgery, University of Alberta, Edmonton, Canada
2. 2Department of Radiology, University of Texas Southwestern, Dallas, Texas, USA
3. 3Department of Neurology, University of Alberta, Edmonton, Canada
1. Correspondence to Dr Cameron M McDougall, Division of Neurosurgery, University of Alberta Hospital, 8440 112th Street NW, Edmonton, Canada; cameron.mcdougall{at}gmail.com

## Abstract

Background and purpose Flow diversion is a relatively new strategy used to treat complex cerebral aneurysms. The optimal method for radiographic follow-up of patients treated with flow diverters has not been established. The rate and clinical implications of in-stent stenosis for these devices is unclear. We evaluate the use of transcranial Doppler ultrasound (TCD) for follow-up of in-stent stenosis.

Materials and methods We analyzed 28 patients treated with the Pipeline embolization device (PED) over the course of 42 months from January 2009 to June 2012. Standard conventional cerebral angiograms were performed in all patients. TCD studies were available in 23 patients.

Results Angiographic and TCD results were compared and found to correlate well.

Conclusions TCD is a potentially useful adjunct for evaluating in-stent stenosis after flow diversion.

• Aneurysm
• Blood Flow
• Flow Diverter
• Stenosis
• Ultrasound

## Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

## Introduction

In general, flow diverters are low porosity endovascular stents used for the treatment of complex cerebral aneurysms. The hemodynamic changes created by the stent lead to stagnation and delayed thrombosis of the aneurysm with preservation of flow in the parent artery as well as significant branching arteries.1 The long-term obliteration rate of aneurysms treated with flow diversion is becoming clear; however, the rate of in-stent stenosis remains uncertain.2–5 The optimal strategy for radiographic follow-up of these patients is unknown given the relatively new technology and corresponding lack of data regarding long-term outcome. CT angiography (CTA) and MR angiography (MRA) offer excellent non-invasive means of following the vascular lesions being treated. However, artifact created by the stents themselves hinders accurate evaluation of in-stent stenosis. Formal angiography is the best means of monitoring for stenosis but the invasive nature of the test, associated risks, and cost make it a less desirable option.

In this study we evaluate the use of transcranial Doppler ultrasound (TCD) as a means of monitoring in-stent stenosis after flow diversion.

## Methods

A retrospective chart review of all 28 patients treated with flow diversion by the senior author (MC) using the Pipeline embolization device (PED; Covidien, Irvine, California, USA) between January 2009 and June 2012 was performed. Two patients died early during the follow-up period and never obtained their 6-month follow-up angiogram. One had a previously unruptured posterior circulation aneurysm that ruptured 20 days after treatment.6 The other patient suffered a devastating brainstem infarction secondary to thrombosis after temporary stoppage of antiplatelet agents with a heparin bridge for a minor procedure and went on to die. One patient refused angiographic follow-up of any kind and another patient died from unrelated medical causes. Of the remaining 24 patients, we obtained TCD studies in 23 patients. The TCD results were then compared with the most contemporaneous angiographic study. In most cases, the TCD was obtained within days of the comparative angiogram. Following flow diversion, patients were scheduled for routine follow-up angiography at 6 months. There was variation in the timing of obtaining TCDs over the course of the study. Those treated early on were not considered for TCD at their initial follow-up and therefore the studies were performed later and compared with the most contemporaneous angiographic study (generally performed within 5 days of each other).

In-stent stenosis on TCD was classified as either present or absent based on evaluation by the reporting technician. The mean flow velocity cut-off used to define stenosis was >120 cm/s. The angiographic data were dichotomized into the presence or absence of stenosis within the stented segment as evaluated by a study neuroradiologist (JR) blinded to the TCD results. Angiographic stenosis was also further subclassified as mild (0–50%), moderate (50–75%), and severe (>75%).

The agreement between TCD angiography was evaluated by Cohen's κ. A κ of 0–0.20 is considered slight, 0.21–0.40 is fair, 0.41–0.60 is moderate, 0.61–0.80 is substantial, and 0.81–1 is considered almost perfect agreement.

## Results

Angiographic and TCD information were available in 23 of 25 surviving patients (table 1); 22 anterior circulation aneurysms and 3 posterior circulation aneurysms were evaluated in 21 women and 4 men. Average patient age was 45 years. Average time from treatment to TCD study was 17.7 months. Angiographic narrowing was reported in 35% (8/23) of patients, most of which were ‘mild’ (63%). The presence of stenosis on TCD was detected in 22% (5/23) of cases. One case of mild angiographic narrowing was not detected on TCD (patient 22). In one case the TCD results were reported as ‘turbulent flow, unable to rule out stenosis’ (patient 21). In two additional cases the angiographic spasm had resolved by the time the TCD study was obtained (patients 5 and 8). The Cohen's κ coefficient was 0.78 (95% CI 0.48 to 1), indicating substantial agreement between ultrasound and angiography. If stenosis <50% is excluded, the correlation becomes perfect in this series.

Table 1

Patient demographics

Using TCD for the detection of in-stent stenosis in the setting of flow diversion was found to have a sensitivity of 71% and specificity of 100% with a negative predictive value of 89% and no false positive results.

### Case example

A middle-aged patient (patient 14) presented with a sudden onset headache. CT showed diffuse subarachnoid hemorrhage with early hydrocephalus. CTA was negative and formal angiography demonstrated a blister type aneurysm on the right supraclinoid internal carotid artery (ICA) (figure 1). The patient underwent flow diversion with the PED on post-hemorrhage day 4 and was discharged home on post-hemorrhage day 11. Follow-up cerebral angiography at 3 months demonstrated 50% narrowing within the stent (figure 2). TCD at 3 months was also compatible with stenosis, with velocities in the right ICA (figure 3A) significantly elevated relative to the left ICA (figure 3B). The patient remained asymptomatic and at 6 months the angiographic appearance had improved with minimal residual stenosis (figure 4), which was also reflected in normalized TCD velocities in the right ICA (figure 5).

Figure 1

Right internal carotid artery (ICA) digital subtraction angiogram lateral projection. A 2 mm ventral blister-type aneurysm is seen on the ICA (black arrow).

Figure 2

Right internal carotid artery digital subtraction angiogram lateral projection 3 months after deployment of the Pipeline embolization device showing moderate in-stent stenosis. The outline of the stent is visible (black arrows).

Figure 3

(A) Right terminal internal carotid artery (ICA) transcranial Doppler ultrasound (TCD) demonstrating elevated flow velocities on the stented side. (B) Left terminal ICA TCD demonstrating normal velocities.

Figure 4

Right internal carotid artery digital subtraction angiogram lateral projection at 6 months showing minimal residual stenosis. The blister aneurysm is no longer seen.

Figure 5

Right terminal internal carotid artery transcranial Doppler ultrasound (TCD) demonstrating resolution of in-stent stenosis.

## Discussion

The number of cases treated with flow diversion continues to increase as experience grows and new devices come on the market.7–12 The best means of following these patients after treatment will continue to evolve with this experience.

MRA and CTA are useful non-invasive tools for following the treated vascular lesions.13 However, they are problematic in their evaluation of in-stent stenosis due to metal artifact created by the device, making an accurate assessment difficult.14 TCD has been used in the past for monitoring intracranial stents. Abou Chebl et al described the use of TCD in the follow-up of drug eluting stents used to treat intracranial atherosclerotic lesions. The TCD results or any relationship with angiographic findings were not discussed.15 Moreira et al used TCD exclusively to follow arterial stents in 26 patients. Only four of the stents were intracranial, but no angiographic follow-up data were provided for comparison.16

The accuracy of TCD in the evaluation of vessel caliber is well established in the cerebral vasospasm population.16 ,17 TCD is also useful for identifying emboli.18 There are limitations to TCD and, even with attempts at standardization, reproducibility can be operator-dependent.19 In this study the correlation between TCD and angiography was near perfect (κ 0.78). Our very high rate of angiographic stenosis (26%) reflects the fact that even mild narrowing was registered as stenosis.

Formal cerebral angiography remains the gold standard for the evaluation of in-stent stenosis but has a well-defined set of associated risks.14 ,15 Obviously these risks would be additive with repeat examinations. TCD may be a useful and cost-effective means of monitoring patients with stenosis requiring prolonged follow-up. Total allowable expenses from the Centers for Medicare and Medicaid Services for TCD study is $204 and$1200 for a single vessel cerebral angiogram. Our standard follow-up protocol consists of routine angiography and baseline TCD at 6 months followed by TCDs yearly thereafter.

In-stent stenosis after flow diversion is beginning to be better understood as a common, mostly benign, process.2–4 Rates of angiographic stenosis in the literature are highly variable and depend upon the definition used and the extent of follow-up. Lubitz et al and Nelson et al both reported no stenosis in their series of 20 and 31 patients, respectively,20 ,21 while Lylyk et al reported a stenosis rate of 10.8% in a series of 53 patients.22 In the most comprehensive study to date, Chalouhi et al reported a stenosis rate of 15.8% in their experience of 139 patients treated with the PED. They identified two independent risk factors for the development of in-stent stenosis (anterior circulation location and no pretreatment with aspirin).

Most authors advise caution regarding the treatment of stenosis as it is often transient. Chalohi et al instead suggest close radiographic follow-up ‘preferably with angiography’ as well as continued dual antiplatelet agent therapy.2 ,4 However, symptomatic cases of in-stent stenosis have been reported, and it remains unclear whether delayed thrombosis after flow diversion is related to underlying stenosis.3 ,23 As more devices begin to come on the market, their safety profile and rate of stenosis will need to be carefully observed.

## Conclusion

TCD could potentially play an important role in the follow-up of in-stent stenosis for patients treated with flow diversion.

## Footnotes

• Contributors All authors contributed to the conception, design, creation and revision of this manuscript.

• Competing interests None declared.

• Ethics approval Ethics approval was obtained from the University of Alberta Institutional Review Board.

• Provenance and peer review Not commissioned; externally peer reviewed.