Rescue Stenting for Failed Mechanical Thrombectomy in Acute Ischemic Stroke

A Multicenter Experience

Yoonkyung Chang, MD; Byung Moon Kim, MD; Oh Young Bang, MD; Jang-Hyun Baek, MD; Ji Hoe Heo, MD; Hyo Suk Nam, MD; Young Dae Kim, MD; Joonsang Yoo, MD; Dong Joon Kim, MD; Pyoung Jeon, MD; Seung Kug Baik, MD; Sang Hyun Suh, MD; Kyung-Yul Lee, MD; Hyo Sung Kwak, MD; Hong Gee Roh, MD; Young-Jun Lee, MD; Sang Heum Kim, MD; Chang-Woo Ryu, MD; Yon-Kwon Ihn, MD; Byungjun Kim, MD; Hong Jun Jeon, MD; Jin Woo Kim, MD; Jun Soo Byun, MD; Sangil Suh, MD; Jeong Jin Park, MD; Woong Jae Lee, MD, PhD; Jieun Roh, MD; Byoung-Soo Shin, MD; Jeong-Min Kim, MD


Stroke. 2018;49(4):958-964. 

In This Article

Materials and Methods

The data that support the findings of this study are available from the corresponding author on reasonable request.

Patient Enrollment

We identified all patients who underwent MT using a stent retriever (Solitaire AB/FR, Covidien/ev3, Irvine, CA; Trevo Proview, Stryker, CA), Penumbra system (Penumbra, Alameda, CA), or both for acute ischemic stroke because of LVO between September 2010 and December 2015 and had the 3-month modified Rankin Scale score (mRS) available. This study was initiated by the Korea Health Technology R&D Project, and the study population was recruited from the cohorts prospectively registered in 16 participating comprehensive stroke centers. Tandem cervical ICA and intracranial large artery occlusions were included. Cervical ICA dissection was included, but intracranial artery dissection was excluded. Bilateral large artery occlusions was also excluded. Enrollment criteria were as follows: (1) age ≥18 years old, (2) initial National Institutes of Health Stroke Scale score (NIHSS) ≥4, (3) onset-to-puncture time ≤600 minutes, (4) mRS of 0 or 1 before qualifying stroke, (5) collateral grading on computed tomographic angiogram (CTA) assessable, (6) recanalization success or failure assessable, and (7) intracranial ICA or MCA-M1 occlusion documented during the endovascular therapy. The institutional review boards of all participating hospitals approved this study and waived the requirement for informed consent for study inclusion based on the retrospective study design.

Data Collection and Assessment

All data including clinical and laboratory findings were obtained from the each participating hospital and then filled-up in the predefined case report form. The procedural details, including complications during endovascular therapy, were obtained from the neurointerventional database of each participating center and were also added to the case report form. The case report forms were anonymized and then sent to the central core laboratory. All imaging data, including pretreatment nonenhanced computed tomography (CT) and CTA, catheter angiograms during the endovascular therapy, and follow-up CT or magnetic resonance (MR) imaging, were anonymized and sent to the central core laboratory as digital imaging and communication in medicine files. Two neuroradiologists independently assessed the images to determine Alberta Stroke Program Early CT Score and collateral status on CTA using a commercialized digital imaging and communication in medicine viewer (OsiriX, Pixmeo, Geneva, Switzerland). The Alberta Stroke Program Early CT Score was evaluated on 5-mm thickness nonenhanced CT images, and the CTA collateral grade was assessed on 20-mm thickness maximum intensity projection images of single-phase CTA. The CTA collateral grade was dichotomized into either good or poor collateral. Poor collateral was defined when there were few or no vessel markings in more than one half of the MCA territory of the affected side. Recanalization success was defined as a modified Tissue Thrombolysis in Cerebral Ischemia grade of 2b or 3 on the final control angiogram. Modified Tissue Thrombolysis in Cerebral Ischemia grades were assessed independently by 2 interventional neurointerventionalists in the core laboratory using the same software. Those reviewers were blind to Alberta Stroke Program Early CT Score, collateral status, and clinical outcome. The κ value of the dichotomized modified Tissue Thrombolysis in Cerebral Ischemia grades (0–2a and 2b–3) was 0.813. Discrepant cases were resolved by consensus between raters who were still blind to both the patient's clinical outcome and the findings on follow-up CT or MR.

Endovascular Treatment

MT was performed under local anesthesia or conscious sedation in all participating centers. Because this study was a retrospective analysis of the data collected from the each participating stroke center, use of a balloon guide catheter and choosing the type of MT device as the first tool were at the operator's discretion in each center. After MT failed, RS was performed at the discretion of the operator. The indications for RS were retrospectively surveyed among the centers where RS was done. In addition, it was surveyed among the 16 participating centers how many LVO might be attributable to intracranial atherosclerotic stenosis (ICAS). It depended on the each participating hospital's protocol whether glycoprotein IIb/IIIa inhibitor was administered, whether balloon angioplasty was performed, and which type of stent was used for RS.

Clinical and Imaging Follow-up

All patients were examined via MR imaging or CT within 3 days after endovascular treatment depending on each participating center's follow-up protocol. Although the MR imaging sequences were different depending on each center's protocol, T2*, Flair, and diffusion-weighted imaging were included in common. When the patient's neurological status abruptly deteriorated, a CT scan was immediately obtained. Intracranial hemorrhage was evaluated on CT or T2* MR images. Intracranial hemorrhage was regarded as symptomatic (sICH) if the patient's NIHSS score increased to ≥4 and there were no other evident causes for the increased NIHSS score. The clinical follow-up schedule followed each center's protocol.

Outcome Measurement

The rate of recanalization success was evaluated in the RSG. Then, we compared RSG with NSG patients in clinical and laboratory findings, the use of an intravenous tissue-type plasminogen activator, MT method, the rate of good outcome, and incidence of sICH and mortality. Good outcome was defined as a mRS of 0 to 2 at 3 months. Finally, we evaluated whether RS remained independently associated with good outcome after adjustment of other factors potentially associated with functional outcome.

Statistical Analysis

All statistical analyses were performed with IBM SPSS Statistics version 23 (IBM Corp, Armonk, NY). All categorical variables are presented as number and frequency (%), and continuous variables are presented as the mean±SD or the median±interquartile range. In the univariate analysis to find the factors associated with good outcome in the patients with MT failure, and to compare RSG with NSG patients, clinical and laboratory findings, the use of intravenous tissue-type plasminogen activator, MT tool, incidence of sICH, and functional outcome (mRS) were evaluated using the χ2 test, Fisher exact test, student t test, and Mann–Whitney U test as appropriate. Finally, to find independent factors associated with good outcome in the patients with MT failure, multivariate binary logistic regression analysis was performed. We included variables with a potential association in the univariate analysis (P<0.20) in the multivariate binary logistic regression analysis. We assessed the model's goodness of fit with the Hosmer and Lemeshow test.

A P<0.05 is considered significant with a 95% confidence interval.