White Matter Hyperintensity Volume and Outcome of Mechanical Thrombectomy with Stentriever in Acute Ischemic Stroke

Abstract

Background and Purpose

Finding of white matter hyperintensity (WMH) has been associated with an increased risk of parenchymal hematoma (PH) and poor clinical outcomes after mechanical thrombectomy using old-generation endovascular devices. Currently, no data exist with regard to the risk of mechanical thrombectomy using stentriever devices in patients with significant WMH. We hypothesized that WMH volume (WMHv) will not affect the hemorrhagic and clinical outcome in acute ischemic stroke (AIS) patients undergoing thrombectomy using new-generation devices.

Methods

A retrospective cohort of consecutive AIS patients >18 year-old receiving mechanical thrombectomy with stentriever devices at a single academic center was examined. WMHv was assessed by a semi-automated volumetric analysis on T2 FLAIR MRI. Outcomes included the rate of any ICH, 90-day modified Rankin score (mRS), the rate of good outcome (discharge mRS ≤2), and the rate of successful reperfusion (TICI score 2b-3).

Results

Between June 2012 and December 2015, 56 AIS patients met the study criteria. Median WMHv was 6.76 cm³ (4.84–16.09 cm³). Increasing WMHv did not significantly affect the odds of good outcome (OR 0.811, 95%CI 0.456–1.442), ICH (OR 1.055, 95%CI 0.595–1.871), PH (OR 0.353, 95%CI 0.061–2.057), successful recanalization (OR 1.295, 95%CI 0.704–2.383), or death (OR 1.583, 95% CI 0.84–2.98).

Conclusions

Mechanical thrombectomy using stentrievers appears to be safe in selected AIS patients with LVO, non-withstanding the severity of WMH burden in this population. Larger prospective studies are warranted to validate these findings.

Introduction

White matter hyperintensity (WMH), a radiographic indicator of cerebral small vessel disease¹, has been linked to poor clinical outcomes after acute ischemic stroke (AIS)². An association has been reported of moderate or severe WMH with increased risk of parenchymal hematoma (PH) after mechanical thrombectomy with MERCI devices³. However, newer stentrievers show better revascularization rates and clinical outcomes, with lower rates of symptomatic intracerebral hemorrhage (ICH) compared to the MERCI retriever system⁴. To date, no data exist regarding the risks of mechanical thrombectomy using current generation devices in patients with advanced WMH. We hypothesized that contemporary thrombectomy approach in AIS patients with significant WMH will demonstrate a better safety profile.

Methods

This was a retrospective, single-center cohort study of AIS patients treated with mechanical thrombectomy using new-generation devices from the prospectively collected institutional stroke database. This study received approval from our institutional review board. We included all consecutive AIS patients with anterior circulation stroke due to large vessel occlusion (LVO), who underwent treatment with stentrievers and who had T2 fluid attenuated inversion recovery (FLAIR) MRI obtained prior to mechanical thrombectomy and available for volumetric analysis of WMH burden.

WMH volume (WMHv) was assessed using a previously published semi-automated volumetric analysis method, by adding WMHv from both hemispheres, with exception of cases where acute infarct area in the affected hemisphere largely overlaps with areas of WMH⁵. One investigator (K.A.), who was trained in the method (ICC >0.92) and blinded to all clinical data, independently analyzed the images. Supplementary Figure I shows WMHv analysis steps used in this study.

Safety outcome included the rate of any ICH and PH classified by European Cooperative Acute Stroke Study (ECASS) criteria⁶. Primary efficacy outcome was 90 day modified Rankin score (mRS). Secondary outcomes included the rate of good outcome (mRS on discharge ≤2), and rate of successful reperfusion (modified Thrombolysis in Cerebral Ischemia (TICI) score 2b or 3). Non-contrast head CT at 24 hours after mechanical thrombectomy was used to assess for hemorrhagic complications.

Statistical Analysis

Statistical analysis was performed using the SPSS (version 24, IBM Corporation, Armonk, NY, USA). Spearman’s rank correlation was used for correlation analyses. Logistic regression analyses were performed to evaluate the primary outcome and safety end points. Variables with p-value <0.1 in univariable analyses were included in the multivariable analyses. Statistical significant p-value was <0.05.

Results

From June 2012 through December 2015, we identified 136 AIS patients receiving stentreiver thrombectomy. Among these, 56 consecutive cases with MRI FLAIR imaging completed on admission were included. Baseline characteristics, procedural details, and outcomes (supplementary Table I) were similar to other reported AIS stentriever thrombectomy patient cohort⁷. Any ICH and PH rates were 32.1% and 5.4%, respectively. Median WMHv was 6.76 cm³ (IQR4.84–16.09 cm³).

Supplementary Table II shows outcomes per WMHv quartile. There was no significant difference in rate of good outcome, death, ICH, PH, and successful recanalization between WMHv quartiles. WMHv was significantly correlated with age (r=0.53, p<0.001), and history of atrial fibrillation (r=0.35, p=0.008), hypertension (r=0.28, p=0.034), or coronary artery disease (r=0.39, p=0.003). There was no significant correlation between WMHv and evidence of ICH, PH, good outcome, or death.

Univariable and multivariable logistic regression analysis of thrombectomy outcome predictors is shown in Supplementary Table III. Successful recanalization was an independent predictor of good outcome (OR 209.99, 95%CI 2.6–16711.6). Increased duration from last known well (LKW) time to recanalization was an independent predictor of ICH (OR 1.005, 95%CI 1–1.01). Congestive heart failure/cardiomyopathy increased the odds of mortality (OR 22.594, 95%CI 1.143–446.488), while successful recanalization significantly decreased the odds of death (OR 0.081, 95%CI 0.007–0.972). Increasing WMHv did not significantly affect the odds of good outcome (OR 0.811, 95%CI 0.456–1.442), ICH (OR 1.055, 95%CI 0.595–1.871), PH (OR 0.353, 95%CI 0.061–2.057), successful recanalization (OR 1.295, 95%CI 0.704–2.383), or death (OR 1.583, 95% CI 0.84–2.98).

Discussion

Our study finds no significant association between burden of WMH and AIS outcome after mechanical thrombectomy using stentrievers. WMH severity did not significantly increase the odds of ICH, PH, or death in AIS patients; therefore, suggesting that thrombectomy using new-generation devices may be safe in this population, as compared to historical cohorts. To our knowledge, this is the first study of WMH and stentriever thrombectomy outcomes.

The proposed pathophysiology of WMH is diseased small cerebral vasculature associated with intimal hyperplasia, atherosclerosis, and lipohyalinosis⁸. While exact mechanisms remain unclear, the possible association of WMHv and hemorrhagic conversion of acute infarct after intravenous thrombolysis or mechanical thrombectomy may be due to severity of small vessel disease causing endothelial dysfunction and blood brain barrier disruption, making extravasation of blood from immediate reperfusion and hemorrhagic complications more likely.

Shi et al reported 65.4% ICH rate and 42.3% PH rate in AIS patients with moderate or severe WMH treated with MERCI thrombectomy³. They utilized a visual rating scale to measure WMH. Using more precise, volumetric WMH measurement in our study, we found that the AIS patients with WMHv in the highest quartile (>16.09 cm³) stentriever treatment resulted in lower ICH (32.1%) or PH (5.2%) rates, compared with the MERCI thrombectomy cohort³ and not significantly increased from the lower WMHv quartile. These findings underscore the safety of contemporary thrombectomy device usage in patients with high WMH burden.

WMH has been linked to decreased odds of good functional outcomes after AIS², but in our cohort of patients treated with contemporary devices, WMHv did not significantly impact the odds of good outcome (OR 0.811, 95%CI 0.456–1.442). However, this finding may be related to the small sample size of our patient cohort, and larger prospective studies are needed to validate these findings.

Studies have suggested that time from symptom onset to recanalization may be associated with ICH/sICH development following stentriever thrombectomy as seen with our patient cohort^{9, 10}. In current American Heart Association/American Stroke Association Guidelines from 2015, mechanical thrombectomy is recommended if groin puncture occurs up to 6 hours after stroke onset¹¹. Ongoing randomized trials aim to extend the therapeutic time window even further^{12, 13}. The majority of patients in our cohort received thrombectomy within 6 hours of symptom onset (median LKW to recanalization time 5.28 hrs). In the extended window for thrombectomy, a higher risk of hemorrhagic outcome in AIS patients with high WMH burden might be expected in association with immediate reperfusion post-thrombectomy. Future studies are warranted to investigate the interaction between WMH burden, time to recanalization, and outcomes in AIS patients.

Our study is limited by its retrospective design, the single center experience, the small number of patients with T2 FLAIR MRI sequences specifically available for volumetric WMH analysis, relatively low range of WMHv burden, and lack of data on symptomatic ICH. Another possible limitation is the use of our previously published institutional protocol of MRI-based election for endovascular therapy¹⁴, which specifies that only patients with small DWI core be recommended for mechanical thrombectomy. This creates a potential bias toward reduced risk of hemorrhagic complications in this highly selected population, possibly negating the detrimental effects of WMH. The strengths of our study include the (a) utilization of a validated, semi-automated WMH volumetric analysis method instead of the simpler visual grading scale, allowing for a more accurate estimate of individual WMH disease burden, (b) uniform exposure of the treated patients given the single center nature of the study, and (c) comprehensive ascertainment of clinical and neuroimaging data in this analysis.

Conclusion

Mechanical thrombectomy using stentrievers appears to be safe in selected AIS patients with LVO, non-withstanding the severity of WMH burden in this population. Larger prospective studies are warranted to validate these findings.