DOI: https://doi.org/10.21203/rs.3.rs-1615173/v1
Occlusions of the superior cerebellar artery (SCA) are occasionally observed during thrombectomy in patients with posterior circulation stroke. This study aimed to investigate the incidence and clinical impact of SCA occlusion remaining after endovascular thrombectomy for acute basilar artery occlusion. We retrospectively analyzed data from 116 patients who underwent thrombectomy for basilar artery occlusion. The presence or absence of SCA occlusion was assessed on final angiograms. Clinical and radiologic data were compared between patients with and without SCA occlusion. All patients underwent pretreatment and follow-up diffusion-weighted imaging to detect new infarctions in SCA territory. Follow-up computed tomography angiography was performed to detect spontaneous recanalization of occluded SCAs. Ten patients (8.6%) had SCA occlusions on final angiograms. Of these, two patients had bilateral occlusions. A new infarction in corresponding SCA territory occurred in 5 of 10 patients and new midbrain infarction in one patient. The diameter of new infarctions ranged from 4 to 11mm. No patients with SCA occlusions experienced symptomatic cerebellar hemorrhage or malignant cerebellar infarction. There were no differences in the length of hospital stay, rates of favorable outcome, and mortality between patients with and without SCA occlusions. Nine of 12 (75%) SCA occlusions showed spontaneous recanalization. In conclusion, SCA occlusions remaining after thrombectomy for basilar artery occlusion had a benign clinical course. In addition, most of these lesions had recanalized spontaneously. Our study suggests that attempts to recanalize remnant SCA occlusion may be unnecessary after basilar artery thrombectomy.
Although recent randomized controlled trials failed to show the superiority of endovascular treatment over standard medical treatment in patients with posterior circulation stroke [1, 2], endovascular thrombectomy is now being incorporated into daily clinical practice to treat patients with acute occlusion of the basilar artery [3–6].
When performing thrombectomy for basilar artery occlusion (BAO), little attention has been paid to the patency of cerebellar arteries and the clinical impact of their occlusion. Determining the clinical significance of an occlusion of each cerebellar artery may help prognostication after thrombectomy in posterior circulation stroke. To date, however, there have been no reports regarding the clinical consequences of occlusion of cerebellar arteries that remains after thrombectomy for BAO. Unlike the anterior and posterior inferior cerebellar arteries, the superior cerebellar artery (SCA) is the consistent cerebellar artery, and thus no visualization of the SCA confidently indicates its occlusion [7, 8]. The SCA is one of the major posterior fossa arteries and supplies the whole superior aspect of the cerebellar hemisphere, the superior vermis, the largest part of the cerebellar deep white matter, and the lateral and posterior regions of the midbrain [9, 10]. SCA occlusions are occasionally observed on final angiograms after successful removal of clots from the basilar artery. However, the optimal management of such circumstances remains unknown. Thus, this study aimed to investigate the incidence and clinical impact of remaining SCA occlusion after endovascular thrombectomy for acute stroke due to BAO.
Patients
Between January 2011 and December 2020, a total of 128 consecutive patients with acute BAO received endovascular thrombectomy at a comprehensive stroke center. Of these, 5 patients without pretreatment diffusion-weighted imaging (DWI), 6 patients without posttreatment DWI, and one patient without 90-day modified Rankin Scale (mRS) data were excluded from the study. Clinical and radiologic data of the remaining 116 patients were retrieved from a prospectively collected database and analyzed. The institutional review board approved this study and waived the requirement to obtain informed consent on the basis of the retrospective study design.
Endovascular Therapy
Inclusion criteria for endovascular thrombectomy were as follows: occlusion of the basilar artery confirmed on catheter angiography, presentation within 12 hours of stroke onset or last seen normal, baseline National Institutes of Health Stroke Scale (NIHSS) score ≥ 4, and no intracranial hemorrhage on pretreatment computed tomography (CT). Intravenous thrombolysis with recombinant tissue plasminogen activator was performed in eligible patients before thrombectomy. Endovascular procedures were performed under local anesthesia in most patients. Conscious sedation was used at the discretion of the neurointerventionalists. Endovascular thrombectomy was performed with either a stent retriever or a large-bore aspiration catheter as a first-line device. Balloon angioplasty with or without stent placement was performed to treat underlying severe (≥70%) atherosclerotic stenosis, if needed. We did not perform additional recanalization procedures for SCA occlusion even if it was revealed on final angiograms.
Imaging Analysis
All patients included in the study underwent pretreatment DWI and follow-up DWI within 3 days after the endovascular procedure. DWI examination were performed with a 1.5T system (Signa HDxt; GE Medical Systems, Milwaukee, Wisconsin, USA) or a 3.0T system (Ingenia 3.0T CX, Philips Medical Systems, Best, Netherlands). At DWI, the SCA territory infarction was defined as diffusion-restricted lesions in the upper half of the ipsilateral cerebellar hemisphere or middle cerebellar peduncle or in the lateral or posterior region of the midbrain [10,11]. The presence or absence and types of SCA infarction (cerebellar or midbrain) were recorded in each patient. We also assessed the posterior circulation Acute Stroke Prognosis Early CT Score (pc-ASPECTS) on pretreatment DWI.
The patency or occlusion of the superior cerebellar arteries was assessed on final angiograms. The site of BAO was classified as proximal, middle, or distal on initial angiograms in accord with previous studies [12]. Overall reperfusion status was assessed on final angiograms according to the modified Thrombolysis in Cerebral Infarction (mTICI) score. Successful reperfusion was defined as an mTICI score of 2b or 3. All patients with SCA occlusion underwent follow-up brain CT angiography before discharge. CT angiography source images, thick-slab maximum-intensity projection images in axial, coronal, and oblique coronal plains, and volume-rendering images were reviewed to evaluate the patency of SCAs. At CT angiography, we defined the late spontaneous recanalization as clear visualization of the ostium and whole arterial segments of previously occluded SCA. Post-treatment CT scans were evaluated to identify intracranial hemorrhages according to the Heidelberg bleeding classification.13 Symptomatic cerebellar hemorrhage was defined, according to the Heidelberg classification, as any cerebellar hemorrhage associated with clinical evidence of neurological worsening, with the hemorrhage judged to be the principal cause of neurologic decline [13]. Malignant cerebellar infarction was defined as a cerebellar infarction causing mass effect in the posterior cranial fossa resulting in decompressive craniectomy or in-hospital mortality [14]. All imaging examinations including DWI, angiography, CT, and CT angiography were retrospectively assessed by two neuroradiologists who were blinded to clinical information. Conclusions were made by consensus in case of disagreement between two readers.
Clinical outcomes
Clinical outcomes were assessed by stroke neurologists using a modified Rankin scale (mRS) score during an outpatient visit or by telephone interviews at 90 days after endovascular therapy. A favorable outcome was defined as an mRS score of 0 to 3. We also assessed the length of hospital stay and occurrence of malignant cerebellar infarction and in-hospital mortality. Patients with in-hospital mortality were excluded from the analysis of the length of hospital stay.
Statistical analysis
Continuous variables are presented as the median and interquartile range. Categorical variables are presented as the number and percentage. The χ2 test or Fisher exact test was used for categorical variables, and the Mann-Whitney U test was used to compare continuous variables. We compared baseline characteristics, procedural factors, malignant cerebellar infarction, in-hospital mortality, length of hospital stay, and 90-day mRS between patients with SCA occlusion and without it. Statistical analyses were performed using SPSS software (Version 26.0; IBM SPSS, Chicago, IL). A P value <.05 was considered significant. Logistic regression analysis was performed to identify independent predictors of 90-day favorable outcome. The logistic regression analysis included variables that showed a P value of <.05 in univariate analysis.
This study included 116 patients (63 males; median age, 73 years) who underwent DWI before and after endovascular thrombectomy for BAO. The baseline demographics and procedural characteristics are shown in Table 1. Overall, the median baseline NIHSS score was 12. Intravenous thrombolysis with tissue plasminogen activator was performed in 27 patients (23.3%). Sixty-nine patients (59.5%) had an occlusion in the distal segment of the basilar artery, 25 patients (21.6%) in the proximal segment, and 22 patients (18.9%) in the middle segment. Underlying severe atherosclerotic stenosis of the basilar artery was identified in 28 patients (24.3%). For the first-line approach, thrombectomy with a stent-retriever was used in 101 patients (87.1%), and contact aspiration thrombectomy in 15 patients (12.9%).
Ten patients (8.6%) had an SCA occlusion on final angiograms. Eight patients had unilateral SCA occlusion and 2 patients had bilateral occlusion. The site of basilar artery occlusion was the distal segment in 8 patients and middle segment in 2 patients. There were no differences in baseline and procedural characteristics and stroke etiology between patients with SCA occlusion and those without it (Table 1).
All patients (n = 116) |
With SCA occlusion (n = 10) |
Without SCA occlusion (n = 106) |
P value |
|
---|---|---|---|---|
Age, y |
73 (63.75-80) |
72 (59-76.5) |
73 (64–80) |
0.46 |
Sex, male |
63 (54.3) |
5 (50.0) |
58 (54.7) |
0.775 |
Vascular risk factor |
||||
Hypertension |
80 (69.0) |
7 (70.0) |
73 (68.9) |
0.941 |
Diabetes mellitus |
35 (30.2) |
3 (30.0) |
32 (30.2) |
0.990 |
Dyslipidemia |
28 (24.1) |
3 (30.0) |
25 (23.6) |
0.650 |
Smoking |
26 (22.4) |
2 (20.0) |
24 (22.6) |
0.848 |
Atrial fibrillation |
57 (49.1) |
4 (40.0) |
53 (50.0) |
0.545 |
Coronary artery disease |
11 (9.5) |
1 (10.0) |
10 (9.4) |
0.953 |
History of stroke or TIA |
19 (16.4) |
1 (10.0) |
18 (17.0) |
0.569 |
Baseline NIHSS |
12 (7-18.25) |
16 (13.25-19) |
11 (7–18) |
0.194 |
Admission hyperglycemia1 |
51 (43.9) |
4 (40) |
47 (44.3) |
0.792 |
Intravenous thrombolysis |
27 (23.3) |
2 (20.0) |
25 (23.6) |
0.798 |
pc-ASPECTS on DWI |
7 (6–8) |
6 (5.25-7) |
7 (6–8) |
0.05 |
Occlusion sites |
0.208 |
|||
Proximal |
25 (21.6) |
0 (0) |
25 (23.6) |
|
Middle |
22 (19.0) |
2 (20.0) |
20 (18.9) |
|
Distal |
69 (59.5) |
8 (80.0) |
59 (57.5) |
|
Underlying severe ICAS |
32 (24.3) |
0 |
32 (26.7) |
0.041 |
Tandem vertebral artery steno-occlusion |
12 (10.3) |
1 (10) |
11 (10.4) |
0.970 |
Stroke etiology |
0.269 |
|||
Cardioembolic |
56 (48.3) |
5 (50.0) |
51 (48.1) |
|
Large artery atherosclerosis |
43 (37.1) |
2 (20.0) |
41 (38.7) |
|
Undetermined |
17 (14.7) |
3 (30.0) |
14 (13.2) |
|
First-line thrombectomy |
0.12 |
|||
Stent retriever |
101 (87.1) |
7 (70.0) |
94 (88.7) |
|
Contact aspiration |
15 (12.9) |
3 (30.0) |
12 (11.3) |
|
Time to procedure |
277.5 (193.75–378) |
302 (202.75–477.5) |
277.5 (191.25–363) |
0.5 |
Procedure time |
29.5 (18.75-46) |
27 (14.5-50.75) |
29.5 (19–46) |
0.723 |
Time to revascularization |
306 (223.75–432.5) |
313 (222.25–554) |
306 (225.25–411) |
0.552 |
Values are presented as n (%) or median (IQR). 1Admission hyperglycemia was defined as admission serum glucose of ≥ 140 mg/dL. | ||||
DWI, diffusion weighed imaging; ICAS, intracranial atherosclerotic stenosis; mRS, modified Rankin scale; NIHSS, National Institute of Health Stroke Scale; pc-ASPECTS, posterior circulation Acute Stroke Prognosis Early CT Score; SCA, superior cerebellar artery; TIA, transient ischemic attack. |
Of 10 patients with SCA occlusion, 5 patients developed a new infarction in the corresponding SCA territory on follow-up DWI. (Figure 1).
All of these infarctions were punctate lesions, 1–3 in number, and 4–11 mm in diameter. Three patients with SCA occlusion showed increased extent of pre-existing SCA territorial infarction and mild mass effect on the fourth ventricle. No cerebellar infarction occurred in the remaining 2 patients. A new midbrain infarction occurred in only one patient (10%), in the posterior region of the midbrain; this patient also had preexisting cerebellar infarction.
Treatment outcomes after endovascular therapy in 116 patients with acute BAO are presented in Table 2.
All patients (n = 116) |
With SCA occlusion (n = 10) |
Without SCA occlusion (n = 106) |
P value |
|
---|---|---|---|---|
Symptomatic cerebellar hemorrhage |
0 |
0 |
0 |
– |
Malignant cerebellar infarction |
1 (0.8) |
0 |
1 (0.9) |
0.758 |
Hospital length of stay, days |
12 (9–22) |
17 (11.25–27.75) |
12 (8.5–20.5) |
0.202 |
Favorable outcome (mRS 0–3) |
58 (50) |
4 (40.0) |
54 (50.9) |
0.743 |
Mortality |
12 (10.3) |
1 (10) |
11 (10.4) |
0.970 |
Values are presented as n (%) or median (IQR). mRS, modified Rankin Scale; SCA, superior cerebellar artery. |
Overall, successful reperfusion was achieved in 99.1% (115/116) of patients. The median length of hospital stay was 12 days (9–22). A 90-day favorable outcome (mRS 0–3) occurred in 50% (58/116) of patients and the rate of 90-day mortality was 10.3% (12/116). Malignant cerebellar infarction occurred in only one patient, who did not have SCA occlusion. No patients showed symptomatic cerebellar hemorrhage. There were no differences in the length of hospital stay and the rates of mortality, favorable outcome, malignant cerebellar infarction, and symptomatic cerebellar hemorrhage between patients with SCA occlusion and those without it. At follow-up CT angiography, 9 of 12 (75%) occluded SCAs showed late spontaneous recanalization (Fig. 2).
In binary logistic regression analysis with adjustment for potential confounders (hypertension and pc-ASPECTS), the following variables remained independent predictors of 90-day favorable outcome: age (adjusted odds ratio [OR)] per 1-year increase, 0.930; 95% confidence interval [CI], 0.881–0.983, P = 0.010), female gender (adjusted OR, 5.030; 95% CI, 1.670–15.152; P = 0.004), baseline NIHSS (adjusted OR per 1-point increase, 0.864; 95% CI, 0.787–0.948; P = 0.002), admission hyperglycemia (adjusted OR, 0.253; 95% CI, 0.084–0.764; P = 0.015), and any intracranial hemorrhage (adjusted OR, 0.249; 95% CI, 0.072–0.863; P = 0.028).
Unlike other cerebellar arteries, the SCA is the most consistent artery among the infratentorial arteries, and agenesis of the SCA either one side or both sides has not been reported in the literature [7, 8]. The orifice of the SCA can be compromised in patients with BAO, especially when clots lodge in the distal segment of the basilar artery. However, the incidence of remaining occlusion of the SCA in patients undergoing thrombectomy for acute BAO has not yet been reported. In the present study, SCA occlusion occurred in 8.6% of such patients and 80% of SCA occlusion occurred in patients with BAO in the distal segment.
Treatment strategies for SCA occlusion remaining after successful recanalization of the basilar artery have not yet been reported and need to be elucidated. Attempting endovascular recanalization of the occluded SCA could be one option. However, the risk of vessel injury would be high during the endovascular procedure because of the blind navigation of small arteries and acute angled take-off of the SCAs. The results of our study suggest that attempt to recanalize remaining SCA occlusion may be unnecessary in patients with acute BAO. New cerebellar infarction occurred in only half of the patients with SCA occlusion and new midbrain infarction in only one patient (10%). Moreover, the new infarctions were small, with a diameter ranging from 4–11mm. Accordingly, no patients with SCA occlusions developed symptomatic cerebellar hemorrhage or malignant cerebellar infarction. Patients with and those without SCA occlusions did not differ in their length of hospital stay, rates of good outcome, and in-hospital mortality. Therefore, this study suggests that SCA occlusions remaining after endovascular thrombectomy for BAO may have a benign prognosis.
The benign prognosis of SCA occlusions remaining after thrombectomy for BAO may be attributed to several possible mechanisms. In our study, 75% of SCA occlusions seen on final angiograms had recanalized spontaneously on follow-up CT angiography. Successful recanalization of the basilar artery trunk may increase regional blood pressure on the SCA ostium and facilitate endogenous thrombolysis, both of which lead to late spontaneous recanalization. Millán et al. reported that spontaneous complete recanalization (evaluated with CT or magnetic resonance angiography 24 hours after stroke) occurred in 22.3% of patients with anterior circulation large vessel occlusion who received best medical management only [15]. In their study, patients with spontaneous complete recanalization at 24 hours showed better clinical outcome compared with those with no or partial recanalization (90-day mRS 0–2, 57.1% vs. 23.3%). Another possible mechanism is that that recanalization of the basilar artery enhances collateral flow from the inferior cerebellar arteries and posterior cerebral artery to the SCA territory via leptomeningeal anastomoses, possibly preventing the development of large territorial infarction [16].
The major limitations of this study are the small sample size and the retrospective study design. In addition, the patency of inferior cerebellar arteries and leptomeningeal collateral flow were not assessed in this study. However, the exact evaluation of the patency of inferior cerebellar arteries is impossible in patients with BAO because the prevalence of anatomical variations of inferior cerebellar arteries is high. The incidence of congenital absence of the anterior inferior cerebellar artery and posterior inferior cerebellar artery was 36.1% and 38.1%, respectively, in a CT angiographic study [7]. We could not assess leptomeningeal collateral flow because carotid angiography and contralateral vertebral angiography were not performed in most of the cases.
In this retrospective case series study, remaining SCA occlusions were found in 8.6% of patients undergoing thrombectomy for BAO, but they had a benign clinical course. In addition, most SCA occlusions had recanalized spontaneously on follow-up CT angiography. These results suggest that attempts to recanalize remaining SCA occlusion may be unnecessary in patients with acute BAO.
Ethical approval and consent to participate: Ethical approval and consent to participate was waived by the local Ethics Committee of Chonnam National University Hospital in view of the retrospective nature of the study and all the procedures being performed were part of the routine care.
Human and Animal Ethics: Not applicable.
Consent for publication: Not applicable.
Availability of supporting data: Data are available upon reasonable request.
Competing interests: The authors have no competing interests to declare that are relevant to the content of this article.
Funding: No funds, grants, or other support was received.
Author’s contributions: WY: study concept and design, acquisition of data, interpretation of the data, study supervision, statistical analysis, and critical revision of manuscript for intellectual content. BHB, YYL: acquisition of the data and interpretation of the data, writing of the first draft of the manuscript, statistical analysis. SKK, CP, BCL, HOK: acquisition of the data and interpretation of the data.
Acknowledgements: Not applicable.