Prognostic factors for reocclusion after mechanical thrombectomy plus rescue treatment in acute athereosclerotic steno-occlusion with successful recanalization
Article information
Abstract
Objective
Treatment failure usually occurs within 24 hours after mechanical thrombectomy (MT) for acute intracranial atherosclerotic steno-occlusion (ICAS) and is an unexpected event that adversely influences the clinical outcome. We retrospectively evaluated the factors influencing reocclusion after MT plus rescue treatment in acute ICAS patients with successful recanalization.
Methods
From January 2013 to December 2020, 60 patients with ICAS who underwent MT plus rescue treatment with successful recanalization were included in our study. We classified the patients into a patency group (n=47, 78.3%) and a reocclusion group (n=13, 21.7%) based on CT angiography data obtained the day after MT plus rescue treatment.
Results
Intravenous tissue plasminogen activator (IV t-PA) in the patency group (n=18/47 (38.3%)) significantly differed from that in the reocclusion group (n=1/13 (7.7%)) (p=0.045). The number of thrombectomy attempts in the reocclusion group was significantly greater than that in the patency group (median [interquartile range], 2 [1–3] vs. 1 [0–1.5], p=0.004). According to the univariate logistic regression analysis, the only prognostic factor for reocclusion was the number of thrombectomy attempts (odds ratio (OR), 1.655 [1.046–2.619], p=0.032).
Conclusions
In our study of ICAS patients who achieved successful recanalization after MT plus rescue treatment, the number of thrombectomy attempts was an independent prognostic factor for reocclusion after MT. Accordingly, for highly suspicious ICAS lesions, additional attempts at the MT should be carefully performed to prevent reocclusion.
INTRODUCTION
Mechanical thrombectomy (MT) for acute ischemic stroke caused by large vessel occlusion has been a standard treatment regardless of stroke etiology [10,14,15,17]. However, reocclusion after MT and subsequent rescue treatment, such as balloon angioplasty or intracranial stenting, are unexpected adverse prognostic factors of clinical outcome, especially in acute intracranial atherosclerotic steno-occlusion (ICAS) [11,12,16].
Intra-arterial or intravenous (IV) tirofiban infusion, including preoperative oral antiplatelet medication, has been used during the periprocedural period based on the theoretical background that there are platelet aggregation and activation, which are initiated by endothelial injury or atheromatous plaque rupture during MT plus rescue treatment. Despite maximal medical treatment, reocclusion may be one of the major adverse events in ICAS lesions, with successful recanalization after MT plus rescue treatment of ICAS. Accordingly, determining the factors influencing reocclusion of ICAS with successful recanalization via MT and rescue treatment is crucial. We retrospectively investigated demographic and prognostic factors for reocclusion in ICAS patients with successful recanalization after MT and rescue treatments.
MATERIALS AND METHODS
Study cohort
From March 2010 to December 2020, 580 patients with acute large artery occlusion were treated with MT. Among the 580 patients, 73 patients (12.6%) with ICAS were treated with MT plus rescue treatment. Thirteen patients were excluded for the following reasons: 1) 2 patients were lost to follow-up 3 months after MT and rescue treatment, and 2) 11 patients had unsuccessful recanalization (thrombolysis in cerebral infarction (TICI) I or IIa) after MT and rescue treatment. Therefore, 60 consecutive patients with ICAS with initial successful recanalization after MT plus rescue treatment were included in our study (Fig. 1). This study was approved by the Institutional Review Board (IRB No. 2022-11-030), and informed consent was waived due to the retrospective design. The authors followed the guidelines for an observational case-control study.
Flowchart of our study. MT, mechanical thrombectomy; IAT, intra-arterial recanalization therapy; ICAS, intracranial atherosclerotic occlusion
MT indications and MT strategies
We considered MT for patients with large artery occlusion who met the following criteria: 1) initial National Institutes of Health Stroke Scale (NIHSS) score ≥ 4 points; 2) absence of intracranial hemorrhage on computed tomography (CT); 3) at least one-half mismatch between cerebral blood flow and the cerebral blood volume map on CT perfusion or between diffusion-weighted magnetic resonance image and the blood volume map on CT perfusion; and 4) the sites of large artery occlusions included seven segments: the internal cerebral artery (ICA) occlusion, ICA bifurcation occlusion, M1 segment occlusion, M2 segment occlusion, vertebral artery occlusion, basilar artery occlusion and basilar top occlusion. In our study, the middle cerebral artery was classified into M1 (horizontal), M2 (vertical), M3 (opercular) and M4 (cortical) segments [6,10].
Most procedures are performed under local anesthesia via femoral access. For anterior circulation lesions, a balloon guiding catheter (BGC) (9F Optimo; Tokai Medical Products, Aichi, Japan) was introduced into the proximal cervical internal carotid artery (ICA), as it was possibly safe for preventing circulating blood flow; however, a BGC has not been used routinely in posterior circulation lesions. If the patient showed acute anterior cerebral artery occlusion or posterior cerebral artery occlusion, we initially performed stent retrieval with a partially deployed technique, and serial control angiography revealed reocclusion or subtle changes in the lesion treated by MT. We decided to deploy a stent without balloon angioplasty in the lesion with intra-arterial (IA) or IV tirofiban infusion (Fig. 2).
A 73-year-old female patient whose neurological status showed decreased right ACA territory on time map of CT perfusion (A). Control angiogram showed right distal A1 occlusion (B). We performed MT via stent retrieval via the partially deployed method and revealed reocclusion on repeated cerebral angiograms (C, D). Therefore, because this lesion was steno-occluded intractable to the MT, we decided to perform stenting via a neuroform atlas covering steno-occlusion (E). We injected tirofiban (300 mcg) and actilyse (10 g) intra-arterially into the lesion treated with the deployed stent. A control angiogram revealed complete recanalization (F). The maximal intensity projection image of the CTA 1 day after MT plus stenting showed maintenance of patency in the occluded lesion (G). ACA, anterior cerebral artery; CT, computed tomography; MT, mechanical thrombectomy; CTA , computed tomography angiography
After conventional angiography confirmed the accurate location of the occlusive lesions, the thrombotic burden was removed via stent retrieval (Trevo; Stryker, Kalamazoo, Michigan, USA/Solitare stent; Medtronic, Minneapolis, Minnesota, USA), catheter aspiration (Sofia 6F; Microvention, Orange County, CA, USA) or hybrid mechanical thrombectomy involving the simultaneous use of a Trevo stent and an intermediate catheter [5,6,13]. When ICAS patients present with progressive blood flow stagnation, subtle changes, or reocclusion after MT, we perform balloon angioplasty (Gateway; Stryker, Kalamazoo, Michigan, USA) via the exchange technique. When repeated conventional angiography revealed recoil or reocclusion of the lesion after balloon angioplasty, we subsequently deployed an intracranial stent to cover the entire ICAS, which presented with large artery occlusion. Tirofiban (300 μg) was used intra-arterially during rescue treatment with or without IV t-PA. Intracranial procedural hemorrhage, including contrast leakage, was surveilled by intermittent Dyna CT (Syngo VX91C, Siemens AG) during MT and plain brain CT immediately after MT. If the patients’ neurological statuses were stable or improved after MT and rescue treatment, all patients underwent CT angiography on the first day after MT and rescue treatment to evaluate the patency of the recanalized artery and postoperative hemorrhage. However, CT angiography plus perfusion CT should be routinely evaluated immediately if patients treated with MT and rescue treatment present with neurological deterioration.
IV tirofiban infusion after balloon angioplasty or intracranial stenting
If no definite intracranial hemorrhage was observed on a plain brain CT after MT, intravenous tirofiban was infused automatically according to the protocol dose for 12 hours or 24 hours after rescue treatment to prevent platelet aggregation related to reocclusion. One hundred milligrams of aspirin and 75 mg of clopidogrel were given after switching from an IV tirofiban dose (0.075–0.15 mcg/kg/min) for maintenance medical treatment 24 hours after IAT plus rescue treatment.
Data collection
We retrospectively reviewed the patients’ medical records for their clinical characteristics, treatment details, and clinical and radiological outcomes. The clinical characteristics included sex, age, past history of diabetes mellitus, hypertension, dyslipidemia, atrial fibrillation, coronary artery occlusive disease, smoking history, and occlusion site. Past medical history was provided by the patient or the patient’s family. Smoking history was based on the current smoking state.
The NIHSS score was assessed on admission by a neurologist or a neurointerventionist. Patients presenting within 4.5 hours of acute ischemic stroke symptom onset were considered suitable candidates for intravenous tissue plasminogen activator (IV t-PA) infusion. The time stages were classified into 3 groups: 1) symptom onset to hospital arrival time, 2) door-to-puncture time, and 3) procedure time. The methods of thrombectomy were classified into 3 groups: 1) stent retrieval, 2) catheter aspiration, and 3) hybrid mechanical thrombectomy with simultaneous use of both stent retrieval and catheter aspiration. The number of thrombectomy attempts was the total number of thrombectomy attempts via stent retrieval, catheter aspiration or the combined method. The methods of rescue treatment were classified into 3 groups: 1) balloon angioplasty, 2) balloon angioplasty and intracranial stenting, and 3) intracranial stenting alone. Postprocedure IV tirofiban was defined as continuous intravenous infusion for 12 hours or 24 hours after endovascular treatment.
Postoperative hemorrhage was defined as intracranial hemorrhage on postoperative brain CT or CT angiography after 24 hours of intra-arterial recanalization therapy (IAT) and rescue treatment.
The clinical outcomes were assessed via 3-month modified Rankin scale (mRS) scores. A good clinical outcome was defined as a 3-month mRS score of 2 or less, and an excellent clinical outcome was defined as a 3-month mRS score of 0 or 1. The mRS score was estimated by a neurointerventionist or a stroke neurologist during a routinely scheduled clinical visit. The recanalization rate was assessed via the TICI system. Successful recanalization was defined as a TICI grade of 2b or 3.
Statistical analysis
We described the mean (±standard deviation) or median (interquartile range) value for continuous variables and described the frequency (%) for categorical variables. Univariate analysis was performed via the independent t test or Mann-Whitney U test for continuous variables and the Pearson χ2 test or Fisher’s exact test for categorical variables. Furthermore, through a univariate logistic regression model, we investigated the variables that were significantly associated with reocclusion (p<0.05). We did not perform multivariate logistic regression analysis because the number of thrombectomy attempts was the only significant variable for reocclusion. Statistical analyses and figures were created via the R package version 4.3.0. All reported p values less than 0.05 were considered statistically significant.
RESULTS
Among the 60 patients who achieved successful recanalization after MT and rescue treatment, 37 (61.7%) were male. The mean patient age was 66.9 years (standard deviation (SD), 11.9). The occlusion locations included the internal carotid artery (n=13), middle cerebral artery (n=28), anterior cerebral artery (n=2), basilar artery (n=4), vertebral artery (n=12), and posterior cerebral artery (n=1). The angiographic outcomes were TICI 2b (n=35) and TICI 3 (n=25). Twenty-eight patients (46.7%) had good clinical outcomes (mRS ≤2) 3 months after MT plus rescue treatment). Six patients died after MT plus rescue treatment (10%, n=6), and these patients had pneumonia and sepsis (n=2), advanced gallbladder cancer (n=1), and progression of cerebral infarction (n=3).
The patients were classified into the patency group (n=47, 78.3%) and the reocclusion group (n=13, 21.7%) based on CT angiography or conventional angiography within 24 hours after MT plus rescue treatment.
Table 1 shows the comparison of the demographic data between the patency and reocclusion groups, and the demographic data included sex, age, diabetes mellitus, hypertension, dyslipidemia, smoking history, and National Institutes of Health Stroke Scale (NIHSS) score at arrival. The demographic characteristics of the patients in the patency and occlusion groups were not significantly different.
Comparisons of baseline characteristics, treated lesions, and time of stage between the patency group and occlusion group
Table 2 shows the comparison of the use of IV t-PA as a bridge therapy, influencing factors, the use of IV tirofiban after MT and rescue treatment, procedural hemorrhage, and clinical outcomes between the patency and reocclusion groups. The patency group (n=18, 38.3%) was significantly more frequently treated with IV t-PA as bridge therapy than was the occlusion group (n=1, 7.7%) (p=0.045). The number of thrombectomy trials in the reocclusion group was significantly more than that in the patency group (median [interquartile range, IQR], 2 [1-3] vs. 1 [0-1.5], p=0.004). The use of post-procedural intravenous tirofiban was not significantly different between the patency group (n=18, 38.3%) and the reocclusion group (n=8, 61.5%) (p=0.134).
Comparisons of intravenous thrombolysis, thrombectomy methods, number of thrombectomy attempts, methods of rescue treatment, postprocedure intravenous tirofiban infusion, intracranial hemorrhage and clinical outcomes between the patency group and occlusion group
Procedural hemorrhage after endovascular treatment was not significantly different between the patency group (n=6, 12.8%) and the reocclusion group (n=2, 15.4%) (p=1.000). Good and excellent clinical outcomes were not significantly different between the patency group and the reocclusion group (n=23 (48.9%) vs. n=5 (38.5%), p=0.503, n=10 (21.3%) vs. n=1 (7.7%), p=0.427, respectively).
The univariate logistic regression analysis that included demographic and prognostic factors revealed that reocclusion was independently associated with the number of thrombectomy trials (odds ratio (OR), 1.655 [95% confidence interval (CI), 1.046–2.619], p=0.032) (Table 3).
Univariate logistic regression analysis of reocclusion
DISCUSSION
Highly successful recanalization without reocclusion in large artery occlusion caused by a cardioembolism or an artery-to-artery embolism was reasonable in MT alone. Thrombus removal of the cardioembolism or artery-to-artery embolism is the causative treatment for large artery occlusion. However, because large artery occlusion caused by intracranial stenotic progression or vulnerable plaque rupture is a pathological condition of the artery itself, ICAS may not be recanalized successfully by the MT alone. Therefore, additional rescue treatments, such as balloon angioplasty or intracranial stenting, are frequently needed to maintain the patency of the occluded lesion after MT.
Rescue treatment itself, which includes mechanical expansion of ballooning into the vessel lumen or stent deployment of foreign material into the vessel lumen, will cause additional vascular damage to the active stage of vulnerable plaque in ICAS. Accordingly, antiplatelet preparation and maintenance therapy, such as IV or IA tirofiban, provide a theoretical background to prevent platelet aggregation in the early stage of endothelial injury, preventing reocclusion after ICAS is treated with IAT and rescue treatment [3,4,6,8,11,18]. However, in our study, the number of patients who received IV tirofiban infusion was not significantly different between the patency (n=18, 38.3%) and reocclusion groups (n=8, 61.5%) (p=0.238). Our study did not prove that post-procedural intravenous tirofiban infusion was not significantly different between the patent (n=18 (38.3%)) and reocclusion groups (n=8 (61.5%)) after MT plus rescue treatment (p=0.134). Our results suggest that antiplatelet medication, including intra-arterial or IV tirofiban, is needed to maintain the recanalized state of ICAS treated with MT plus rescue treatment but is not sufficient to prevent reocclusion in the hyperacute stage of ICAS treated with MT plus rescue treatment.
Advanced atheromatous plaques can grow sufficiently large to block blood flow, and the most important clinical complication is acute occlusion due to the formation of a thrombus or blood clot, resulting in acute cerebral infarction [1,2]. Thrombosis is typically associated with rupture or erosion of atheromatous plaques [7,9,13]. MT plus rescue treatment without stabilizing vulnerable plaques is insufficient to prevent fresh thrombus formation in ICAS caused by rupture or erosion of atheromatous plaques. Our finding that IV t-PA significantly differed between the patent (n=18 (38.3%) and reocclusion groups (n=1 (7.7%), p=0.045)) may be due to the thrombolytic effect of t-PA, which may stabilize fresh thrombus formation caused by the active stage of vulnerable plaques. However, our study revealed that the use of IV t-PA was not an independent factor for reocclusion in the univariate logistic regression analysis (OR, 0.135 [CI, 0.016–1.122], p=0.064). In conclusion, the use of t-PA tends to prevent reocclusion in ICAS patients treated with MT plus rescue treatment, but the difference was not statistically significant.
Our study revealed that the number of thrombectomy attempts was a positive independent factor for reocclusion (OR, 2.22 [95% CI, 1.11–4.46], p=0.025). These findings suggest that early detection of ICAS without repeated thrombectomy may be related to a reduction in additional injury to vulnerable atheromatous plaques or activated plaque rupture. We suggest that this theoretical background of reducing MT attempts is a rationale for preventing reocclusion in ICAS patients treated with MT plus rescue treatment. To avoid additional MT attempts, we should perform repeated control angiograms to detect subtle changes or occlusions in the lesion treated by MT in patients with noncardioembolic or nonartery-to-artery embolic etiologies.
There are several limitations to our study. First, our study had a retrospective design and therefore had an underlying selection bias. Second, the study sample size was small, and its statistical power may therefore be low. Further large-scale prospective studies are needed to confirm the procedure factors related to reocclusion after ICAS is treated with the IAT and rescue treatment. Nevertheless, we believe that the findings of this study represent several factors related to reocclusion after ICAS is treated with IAT and rescue treatment.
CONCLUSIONS
Our study revealed that the number of thrombectomy attempts is the only independent modifiable prognostic factor of reocclusion in the ICA with successful recanalization after MT plus rescue treatment. Accordingly, additional attempts at MT should be used with caution in ICAS lesions.
Notes
ACKNOWLEDGEMENTS
The authors would like to acknowledge Dr. Sun Gyu Choi from the Biomedical Research Institute of Haeundae Paik Hospital for helping with the statistical analysis.
Disclosure
The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.