A Study on the Efficacy of Thrombectomy in Patients with Atherosclerotic and Cardioembolic Basilar Artery Occlusion

Abstract

Background  Studies on basilar artery occlusion are relatively few compared with those of anterior circulation stroke. The aim of the present study was to compare the efficacy of endovascular therapy (EVT) in patients with basilar artery occlusion classified as large artery atherosclerosis (LAA) and cardioembolism (CE), and to analyze the independent risk factors affecting the prognosis of EVT.

Methods  A total of 123 people were assigned to the LAA and CE groups (97 to the LAA and 26 to the CE). The primary outcome was a modified Rankin Scale (mRS) score of 2 or lower at 90 days. The primary safety outcome was mortality at 90 days. Secondary safety endpoints included the rates of symptomatic intracranial hemorrhage and reinfarction. Multiple logistic regression was used to screen out independent risk factors for EVT prognosis of the LAA and CE groups.

Results  In the analysis, the patients with LAA stroke had better collateral circulation (American Society of Interventional and Therapeutic Neuroradiology/Society of Interventional Radiology [SIR] score of 2–4; 61.9 vs. 19.2%, p  = 0.000), and higher angioplasty rate (32.0 vs. 3.8%, p  = 0.002). The proportions of patients with a 90-day mRS score of 0 to 2 and 90-day mortality were not found to be statistically significant between the two groups. Multivariate logistic regression analysis indicated that age, SIR, white blood cell, blood glucose, and modified thrombolysis in cerebral infarction were independent risk factors for the poor prognosis of EVT in the LAA group.

Conclusion  Although there were differences in clinical characteristics and imaging features between LAA and CE, there was no evidence of a significant difference in prognosis after EVT. In addition, the National Institutes of Health Stroke Scale score was not among the independent risk factors affecting the prognosis of the LAA group.

Introduction

As the disease spectrum changes from infectious diseases to chronic noncommunicable diseases, stroke has become one of the global public health problems. ¹ Acute ischemic stroke (AIS) accounts for about 70% of all strokes. More than 2 million new cases of AIS are reported in China every year. ² The treatment of AIS has made great progress in recent years. Especially for the occlusion of large vessels in the anterior circulation, DAWN, DEFUSE-3, and other trials have confirmed the safety and efficacy of mechanical thrombectomy. However, there are few studies on posterior circulation stroke. What is the efficacy of thrombectomy in patients with basilar artery occlusion (BAO)? On December 9, 2019, Professor Liu Xinfeng published the world's first randomized controlled trial on posterior circulation thrombectomy in the Lancet Neurology—BEST study, ³ but no evidence was given on whether a difference exists in the ratio of good prognosis between patients with endovascular therapy (EVT) and patients with standard medication. Until May 6, 2022, the results of the BAOCHE study led by Professor Ji Xunming were released at the European Stroke Conference (ESOC 2022). It adds high-level clinical evidence for EVT in patients with acute BAO within the 6 to 24-hour time window, and has a profound impact on the clinical diagnosis and treatment of BAO. The prognosis of mechanical thrombectomy in patients with AIS is not only related to the duration of onset, clinical symptoms, habits, and customs, but also the etiological classification of stroke also plays an important role in treatment and prognosis. ⁴ The present study analyzed the prognosis and independent risk factors of EVT on patients with BAO who have been diagnosed with the classifications of large artery atherosclerosis (LAA) and cardioembolism (CE).

Materials and Methods

Research Object

Between April 10, 2020 and July 10, 2022, we assessed 198 patients for eligibility, 123 people were eventually included in the study (97 to the LAA and 26 to the CE). The inclusion criteria were as follows: (1) acute stroke of BAO was confirmed by imaging; (2) aged ≥18 years; (3) National Institutes of Health Stroke Scale (NIHSS) score >5 at admission; (4) symptom onset time ≤24 hours; and (5) who underwent transfemoral cerebral angiography for endovascular reperfusion therapy. The exclusion criteria were as follows: (1) the cause of stroke was dissection, Moyamoya disease, or vasculitis; (2) previous modified Rankin Scale (mRS) score >2 points; (3) history of hemorrhagic cerebrovascular disease or bleeding tendency; and (4) presence of important organ dysfunction, failure, or malignancy and with expected life span of <2 years.

Treatment Method

According to the guidelines for endovascular treatment of AIS, ⁵ 123 patients chose EVT according to the time window and the risk of bleeding. The methods of endovascular treatment are stent thrombectomy or catheter aspiration. If reocclusion occurs after thrombectomy, remedial measures will be taken, including balloon angioplasty, arterial thrombolysis, and stent implantation. The LAA endovascular treatment process is shown in Fig. 1 , and CE is shown in Fig. 2 .

Fig. 1.

LAA was defined as significant fixed focal stenosis in the occlusion site (arrow), evidenced in the final angiography or during the endovascular treatment procedure. A patient with acute occlusion of the basilar artery, a focal stenosis was seen during endovascular treatment, and reocclusion repeatedly occurred even after thrombectomy with a stent retriever and angioplasty with a balloon catheter twice for each. Finally, the basilar artery was recanalized after a stent was deployed. LAA, large artery atherosclerosis.

Fig. 2.

CE was defined as no evidence of focal stenosis after recanalization (arrow). A patient with acute occlusion of the basilar artery, after the first trial of forced arterial suction thrombectomy, the artery was fully recanalized. CE, cardioembolism.

Research Methods

The following patient information was collected: (1) general information: age, sex, blood pressure, blood glucose (GLU), blood lipids, white blood cell (WBC) count, neutrophil percentage (GRA%), albumin (ALB), smoking history, drinking history, cardio-cerebrovascular history; (2) auxiliary examination results: head computed tomography, digital subtraction angiography (DSA), collateral circulation, responsible blood vessels, and revascularization; (3) treatment and prognosis: baseline NIHSS score, time from symptom onset to revascularization, successful recanalization rate, incidence of symptomatic intracranial hemorrhage after operation, mRS score, and mortality at 3 months after operation. This study was approved by the hospital ethics committee, and all patients signed an informed consent form for the operation.

The American Society of Interventional and Therapeutic Neuroradiology (ASITN)/Society of Interventional Radiology (SIR) ⁶ classification system using DSA images was used to assess collateral circulation. SIR grades 2 to 4 are defined as collateral circulation. Modified thrombolysis in cerebral infarction (mTICI) ⁶ vascular recanalization grade was used to evaluate the degree of vascular recanalization. mTICI grades 2b–3 are defined as successful recanalization. Three months after surgery, mRS ≤2 indicates a good prognosis.

Statistical Analysis

SPSS 26.0 software was used for statistical analysis. The measurement data that conform to the normal distribution were expressed by‾x ± s, and the comparison between groups was performed by an independent sample t-test. The measurement data that do not conform to the normal distribution were described in median and interquartile range, and the Mann–Whitney U test was used for comparison between groups. Counting data were expressed in terms of cases and percentages, and comparisons between groups were tested by chi-square test or Fisher's exact probability method. Univariate analysis results with p  < 0.1 may be considered related to the prognosis. p  < 0.05 was used in the multivariate logistic regression analysis.

Furthermore, we used a prespecified multivariable regression model to adjust for any potential imbalances in prognostic factors across the two groups, including age, gender, and baseline NIHSS score.

Results

Prognostic Analyses of LAA and CE

Baseline Characteristics of Clinical Data

The basic clinical data of the patients are shown in Table 1 . The patients with LAA stroke were younger (60.76 ± 12.88 vs. 72.54 ± 10.22, p  = 0.001), had higher prevalence in male (78.4.7 vs. 57.7%, p  = 0.033), higher preoperative systolic blood pressure (149.25 ± 21.24 vs. 134.79 ± 22.83, p  = 0.010), lower NIHSS score (17.3 vs. 22.2, p  =  0.040), and lower prevalence of heart-related underlying diseases (i.e., coronary heart disease, atrial fibrillation, and heart valve disease). Gender, history of smoking, and other baseline characteristics were not statistically remarkable between the two groups.

Table 1. Baseline characteristics.

Clinical features	Total ( n  = 123)	LAA ( n  = 97)	CE ( n  = 26)	p- Value
Demographic characteristics
Age (yr x̅ ± s)	62.70 ± 12.26	60.76 ± 12.88	72.54 ± 10.22	0.001 a
Male (%)	91 (74.0)	76 (78.4)	15 (57.7)	0.033 a
Past disease history
Hypertension (%)	71 (57.7)	52 (53.6)	19 (73.1)	0.074
Diabetes (%)	40 (32.5)	33 (34.0)	7 (26.9)	0.493
Hyperlipidemia (%)	60 (48.8)	54 (55.7)	6 (23.1)	0.003 a
Atrial fibrillation (%)	26 (21.1)	7 (7.2)	19 (73.1)	0.000 a
CHD (%)	31 (25.2)	13 (13.4)	18 (69.2)	0.000 a
Valvular disease (%)	7 (5.7)	1 (1.0)	6 (23.1)	0.000 a
Cerebral infarction (%)	21 (17.1)	15 (15.5)	6 (23.1)	0.360
Smoke (%)	57 (46.3)	47 (48.5)	10 (38.5)	0.364
Laboratory index
WBC	10.76 ± 3.65	10.74 ± 3.41	11.12 ± 4.57	0.987
GRA%	82.59 ± 10.31	82.84 ± 9.06	82.65 ± 14.51	0.885
ALB	37.47 ± 4.97	37.78 ± 4.67	36.41 ± 6.04	0.498
GLU	8.82 ± 4.31	8.63 ± 4.23	8.26 ± 4.71	0.907
Clinical features
SBP, mmHg	146.4 ± 22.19	149.25 ± 21.24	134.79 ± 22.83	0.010 a
Baseline NIHSS 0.040 a
≤10 (%)	18 (14.6)	17 (17.5)	1 (3.9)
11–20 (%)	49 (39.8)	42 (43.3)	7 (26.9)
≥20 (%)	56 (45.6)	38 (39.2)	18 (69.2)

Abbreviations: ALB, albumin; CE, cardioembolism; CHD, coronary heart disease; GLU, blood glucose; GRA%, neutrophil percentage; LAA, large artery atherosclerosis; NIHSS, the National Institutes of Health Stroke Scale; SBP, systolic blood pressure; WBC, white blood cell.

^aStatistically significant.

Basic Characteristics of Patient EVT

All the included patients received EVT treatment, and the operation-related information is shown in Table 2 . No statistically significant difference was found between the two groups ( p  = 0.095). The ratio of postoperative revascularization (mTICI: 2b–3) was higher in the LAA group than the CE group but without statistical difference (86.6 vs. 84.6%, p  =  0.756). The LAA group had a higher proportion of collateral circulation (ASITN/SIR: 2–4; 61.9 vs. 19.2%, p  = 0.000) and intraoperative angioplasty (stent implantation and balloon dilation; 32.0 vs. 3.8%, p  = 0.002) compared with the CE group.

Table 2. Related characteristics of intravascular therapy.

Variable	Total ( n  = 123)	LAA ( n  = 97)	CE ( n  = 26)	p- Value
ASITN/SIR (%) 0.000 a
0–1	58 (44.9)	37 (38.1)	21 (80.8)
2–4	65 (55.1)	60 (61.9)	5 (19.2)
mTICI (%) 0.756
0–2a	17 (13.8)	13 (13.4)	4 (15.4)
2b–3	106 (86.2)	84 (86.6)	22 (84.6)
Angioplasty (%)
	32 (26.5)	31 (32.0)	1 (3.8)	0.002 a

Abbreviations: ASITN/SIR, American Society of Interventional and Therapeutic Neuroradiology/Society of Interventional Radiology; CE, cardioembolism; LAA, large artery atherosclerosis; mTICI, modified thrombolytic therapy for cerebral infarction score.

^aStatistically significant.

Clinical Outcome

The LAA and CE groups had no statistical difference in the 90-day prognosis (mRS: 0–2) after EVT (59.8 vs. 42.3%, p  = 0.111). See Table 3 for details.

Table 3. Outcomes of clinical effects.

Variable	Total ( n  = 123)	LAA ( n  = 97)	CE ( n  = 26)	p- Value
90-day mRS 0.111
0–2 (%)	69 (56.1)	58 (59.8)	11 (42.3)
3–6 (%)	54 (43.9)	39 (40.2)	15 (57.7)

Abbreviations: CE, cardioembolism; LAA, large artery atherosclerosis; mRS, modified Rankin Scale.

Note: p  = 0.093, prompt two groups have no statistical significance.

Security Outcome

The LAA and CE groups had no significant difference in mortality during the 90-day follow-up period after EVT (19.6 vs. 34.6%, p  = 0.105). Both groups had no significant difference in postoperative reocclusion (4.1 vs. 3.8%, p  = 1.000) and the conversion of intracranial hemorrhage within 72 hours (2.1 vs. 0%, p  = 1.000). See Table 4 and Fig. 3 for details.

Table 4. Security outcome.

Variable	Total ( n  = 123)	LAA ( n  = 97)	CE ( n  = 26)	p- Value
90-day death toll (%)	28 (22.8)	19 (19.6)	9 (34.6)	0.105
Reocclusion (%)	5 (4.1)	4 (4.1)	1 (3.8)	1.000
Hemorrhage (%)	2 (1.6)	2 (2.1)	0 (0)	1.000

Abbreviations: CE, cardioembolism; LAA, large artery atherosclerosis.

Fig. 3.

Modified Rankin Scale score at 3 months for each group. There was no statistical difference between LAA and CE (19.6 vs. 34.6%, p  = 0.105).

Analysis of Independent Risk Factors Affecting the Poor Prognosis of EVT in LAA and CE

Analysis of Independent Risk Factors in LAA

Single-factor logistic regression analysis showed that age; male; WBC, GRA%, ALB, and GLU within 24 hours; baseline NIHSS; SIR; and mTICI may be risk factors related to patients' poor prognosis ( p  < 0.1). Multivariate logistic regression analysis of these risk factors showed that age, SIR, WBC, GLU, and mTICI are independent risk factors for patients with poor prognosis. See Table 5 for details.

Table 5. Analysis of independent risk factors affecting the poor prognosis of intravascular treatment in LAA.

Factor	B	p- Value	OR	OR (95% CI)
Male	0.559	0.481	1.748	0.370–8.261
Age	0.072	0.024 a	1.074	1.009–1.143
WBC	0.268	0.019 a	1.308	1.045–1.637
GRA%	0.019	0.705	1.019	0.924–1.124
ALB	−0.118	0.108	0.889	0.769–1.026
GLU	0.131	0.032 a	1.279	1.038–1.324
Baseline NIHSS	0.025	0.606	1.025	0.932–1.128
ASITN/SIR	−1.801	0.034 a	0.165	0.031–0.871
mTICI	−2.514	0.045 a	0.081	0.007–0.943

Abbreviations: Logistic, In( P /1 −  P ) =  a  +  B ₁ X ₁  +  B ₂ X ₂  +  B ₃ X ₃  + …… +  B _i X _i ( P : probability of poor prognosis in the LAA group); B , coefficients for factors (male, age, WBC, etc.), called logistic regression coefficient; ALB, albumin; ASITN/SIR, American Society of Interventional and Therapeutic Neuroradiology/Society of Interventional Radiology; CI, confidence interval; GLU, blood glucose; GRA%, neutrophil percentage; mTICI, modified thrombolytic therapy for cerebral infarction score; OR, odds ratio; WBC, white blood cell.

^aStatistically significant.

Analysis of Independent Risk Factors in CE

Single-factor logistic regression analysis revealed that SIR and prolonged time from symptom onset to recanalization may be related to poor prognosis ( p  < 0.1), but the validity of the multivariate logistic regression analysis is difficult to guarantee because the sample size is too small; hence, independent prognosis factors were not analyzed.

Discussion

In the Western society, large artery occlusions are caused by embolism from proximal sources. By contrast, artery occlusions in Asian patients are commonly caused by intracranial atherosclerotic disease (IAD). ⁷ Patients with CE mostly have blocked distal blood vessels because of emboli instead of arteriosclerosis or stenosis. Thus, the focus in patients with CE is the presence of emboli instead of narrowed blood vessels. By contrast, occlusions in patients with LAA are caused by vascular stenosis; thus, the focus is on narrowed blood vessels and not emboli. The efficacy of EVT in patients with AIS in the anterior circulation has been demonstrated, but there are few studies in the posterior circulation. In this study, we collected data on patients with BAO to compare the difference in efficacy between LAA and CE. At the same time, independent risk factors affecting the prognosis of patients with two types of AIS were analyzed.

LAA and CE are the main causes of macrovascular stroke. The present study found differences in the EVT-related characteristics between patients with LAA and CE in the posterior circulation AIS, but no differences in clinical efficacy and safety indicators after EVT were found.

Our study found that LAA and CE had substantial differences in clinical baseline characteristics, such as age, blood lipids, and baseline NIHSS scores. A study ⁸ found that external stimuli, such as drinking and smoking, cause vascular endothelial cell metabolism disorders, which result in a large amount of lipid component deposition and ultimately lead to atherosclerosis, spasm, and even blood flow interruption. In addition, a genetic history of cerebrovascular disease, hyperlipidemia, type 2 diabetes, and hypertension are all risk factors for its onset. CE is not a risk factor for arteriosclerosis; however, the presence of cardiogenic emboli, such as old myocardial infarction, rheumatic heart disease, and other heart valve diseases, such as combined atrial fibrillation and other diseases, that clearly produce cardiogenic emboli, predisposes to CE. ⁹ The clinical baseline characteristics of this study are similar to previous reports, that is, the LAA group has a higher proportion of hyperlipidemia and preoperative systolic blood pressure, whereas the CE group has a higher proportion of atrial fibrillation and valvular disease. In addition, the patient's age and baseline NIHSS score were also higher in the CE group, because the prevalence of atrial fibrillation is increased in the elderly. Sudden embolism prevents the body from establishing collateral circulation in time because of its acute onset, and cerebral blood vessels are prone to spasm that causes more severe cerebral ischemic symptoms; thus, the NIHSS score is higher. These differences in baseline characteristics guide us to focus on the prevention and treatment of LAA stroke by strengthening antiplatelets, regulating lipid, and detecting changes in blood pressure, whereas the prevention and treatment of CE comprise applying effective anticoagulation measures, detecting changes in international normalized ratio, and focusing on the treatment of primary cardiac disease in clinical practice.

This study also found that although the LAA group had better collateral circulation, the in situ stenosis of the infarct site affected the recanalization effect and further increased the rate of angioplasty during surgery. First, atherosclerotic embolism has stronger adhesion to local blood vessel walls than cardiogenic embolism, and often only partial recanalization can be achieved after stent embolization (mTICI 2). Residual stenosis >70% is considered an important risk factor for the recurrence of ischemic stroke. Insufficient local forward blood flow still leads to an increased risk of reocclusion even if residual stenosis is <70%. ¹⁰ Second, the outward radial support of the thrombectomy stent is insufficient, and temporary release could hardly overcome the local stenosis caused by severe arteriosclerosis. Distal and proximal stent grids cannot be tightly embedded in the thrombus. Moreover, multiple thrombectomy operations are likely to cause lesions. Plaque disintegration, vascular intima damage, and local platelets are easily activated and induce secondary thrombosis. ¹¹ Finally, atherosclerotic emboli are different from cardiogenic emboli in pathological constitution: atherosclerotic emboli are interwoven by fibrin and platelets, whereas cardiogenic emboli are composed of wrapped red blood cells. ^{9 12}

The present study showed that there was no difference in the outcome of EVT between LAA and CE patients with BAO (59.8 vs. 42.3%, p  = 0.111). Kim et al pointed out that the incidence of posterior circulation IAD is high and the prognosis is poor. ¹³ Their findings are different from our findings. The difference may be caused by regional differences, that is, they enrolled more patients in the embolization group, and postoperative angioplasty was not included in the analysis of efficacy. For anterior circulation AIS, studies have shown that patients with CE stroke in Western countries who receive EVT are superior to patients with LAA stroke in terms of complications, revascularization rate, and long-term prognosis. ¹⁴ In Asia, patients with LAA stroke have better effects after EVT than patients with CE. ^{15 16} The LAA group in this study had better collateral circulation than the CE group, and its NIHSS score was lower, and the results were different from the anterior circulation. The reason may be that endovascular treatment of BAO is more laborious, has more complications, and has a higher rate of ineffective recanalization. ¹⁷ The posterior cranial fossa is smaller than the anterior cranial fossa. The degree of cerebral edema is more serious and the patient's prognosis becomes worse once a large vessel occlusion occurs. Even if the LAA group showed better collateral circulation, the posterior circulation occlusion not only affected the blood perfusion of the distal cortical vascular distribution area, but also affected the reticular ascending activation system due to the involvement of pontine and thalamus perforators. The study showed that although the proportion of mRS = 6 in LAA is less than that in CE, the final prognosis between the two groups had no difference. This result suggests that the LAA group had more neurological deficits after thrombectomy and higher invalid recanalization rate than the CE group.

Although there was no significant difference in prognosis between LAA and CE in this study, the risk factors affecting their respective prognosis were different. Thus, identifying the risk factors for these causes and treating patients individually can benefit patients more in clinical work.

The results of this study showed that age, SIR, WBC, GLU, and mTICI are the independent risk factors that affect the prognosis of patients in the LAA group. Similar to other studies, ^{18 19} advanced age is an independent risk factor for adverse neurological outcomes after the revascularization of large vessel occlusion. The elasticity of blood vessels decreases and the body's compensatory ability becomes poor with increasing age. The process of atherosclerosis is accelerated under the action of the cumulative effects of various metabolites and inflammatory factors in the body and causes the involvement of single or multiple vascular beds and ultimately leads to poor prognosis. ²⁰ Good collateral circulation can delay the apoptosis of damaged nerve cells in the ischemic penumbra and reduce the volume of cerebral infarction. ^{21 22} Small emboli can also be removed through their own collateral circulation to improve the benefits of recanalization. Therefore, collateral circulation is an important predictor of neurological prognosis in acute cerebral infarction. The results of this study showed that poor collateral circulation is an independent risk factor for poor clinical outcome after EVT in LAA-type cerebral infarction. Huber et al ²³ showed that increased WBC levels are independent risk factors for poor prognosis. Our research confirmed that an increase in WBCs 24 hours after MT is also an independent risk factor. The recanalization of occluded blood vessels as soon as possible is the most effective treatment for AIS as it restores local blood flow and saves endangered brain tissues. ^{24 25} We also confirmed that mTICI is an independent predictor of the prognosis.

Unfortunately, we could not draw the independent predictors because of the insufficient sample size of the CE group. However, univariate analysis suggested that SIR and prolonged time from symptom onset to recanalization may be related to poor prognosis. Thus, doctors need to pay more attention to these indicators in clinical practice.

Finally, the baseline NIHSS score was not an independent risk factor for EVT prognosis in posterior circulation stroke. The limitations of the NIHSS include its focus on limb and speech impairments and its less emphasis on cranial nerve lesions. ²⁶ Therefore, NIHSS is mostly used for the evaluation of the acute occlusion of anterior circulation.

Conclusion

The safety and efficacy of EVT did not differ between the LAA and CE groups. Age, SIR, mTICI, GLU, and WBC are the independent risk factors for the poor prognosis of EVT in patients with LAA. SIR, symptom onset, and prolonged recanalization may be related to the poor prognosis of patients with CE. Thus, screening for relevant risk factors according to the cause may improve treatment prognosis in clinical practice.

Funding Statement

Funding This research received funding from National Natural Science Foundation of China, No. 81960222; The Young and Middle-Aged Leading Talents in Scientific and Technological Innovation of XPCC, No. 2020CB011.