Percutaneous transluminal angioplasty and stenting for severe stenosis of the intracranial carotid artery and its branches: Comparison of the Wingspan stent vs the Credo stent

Abstract

Background:

Despite the widespread use of the Wingspan stent system for treating severe medically refractory intracranial artery stenosis (SMR-ICAS), a new Credo stent system was approved because it could integrate stent delivery within the balloon catheter. However, the therapeutic outcomes of these two systems have not been compared. This preliminary study aimed to compare the results of percutaneous angioplasty and stenting (PTAS) in SMR-ICAS patients treated with either Wingspan or Credo stents within the anterior cerebral circulation.

Methods:

SMR-ICAS patients with >70% stenosis in the anterior circulation who underwent PTAS using either the Wingspan or Credo stent system were retrospectively analyzed. We evaluated the technical success, safety, and outcomes of the two-stent systems.

Results:

A total of 29 patients were analyzed, including 17 patients treated with Wingspan stents and 12 with Credo stents. The outcomes of the Wingspan stent vs Credo stent were as follows: technical success (16/17 [94%] vs 11/12 [92%], p = 1.00); periprocedural intracranial hemorrhage (2/17 [12%] vs 0/12 [0%], p = 0.50); silent embolic ischemic lesions on periprocedural magnetic resonance imaging (MRI) (13/17 [76%] vs 7/12 [58%], p = 0.42); and significant (>50%) in-stent restenosis in 1 year (4/17 [24%] vs 2/12 [17%], p = 1.00). No recurrent stroke or mortality was noted within 30 days after the procedures or during the 1-year follow-up period.

Conclusion:

The technical success, safety, and outcomes of the Credo stent system were comparable to those of the Wingspan stent system in the management of SMR-ICAS patients. Further large-scale studies are warranted to substantiate these findings.

1. INTRODUCTION

Intracranial artery stenosis (ICAS) is an important risk factor for recurrent ischemic stroke, the incidence of which varies from 1% to 50% across different populations.^1,2 Medical treatment is the main strategy for treating ICAS. When this strategy fails, percutaneous angioplasty and stenting (PTAS) is one of the best alternative treatments for revascularization.^2–4 The SAMMPRIS trial showed that aggressive medical treatment was significantly superior to endovascular treatment with a Wingspan stent in terms of technical safety and clinical outcomes for patients with severe ICAS.⁵ However, after selecting patients who met the on-label criteria, which included ischemic events despite receiving aggressive medical treatment and delayed stenting for at least 8 days, a subgroup reanalysis of the SAMMPRIS cohort demonstrated comparable rates of technical complications between the medical and PTAS groups.³ Furthermore, the WEAVE⁶ and WOVEN⁷ trials, which adopted strict on-label criteria, also reported low periprocedural complication rates and favorable 1-year outcomes. In addition to the strict patient selection of these studies, improvements in the PTAS technique merit attention. The Wingspan stent (Boston Scientific, Marlborough, MA, USA) was approved by the Food and Drug Administration in 2004 for severe medically refractory ICAS (SMR-ICAS). The procedure involves crossing stenotic lesions using a long exchange wire and a balloon catheter for angioplasty and further exchanging the balloon for stent deployment. The risks during the procedure include vascular injury during the manipulation of the long exchanging wire or atherosclerotic plaque dislodgement due to multiple device passages. The incidence of periprocedural subarachnoid hemorrhage (SAH), which is suspected to be related to guidewire injury, reached 1.8% in the SAMMPRIS trial.⁵

A new percutaneous transluminal angioplasty (PTA) balloon catheter was designed by advancing the stent inside the catheter lumen, which causes technical differences.^8,9 The Credo stent system consists of a NeuroSpeed PTA balloon catheter (Acandis GmbH, Pforzheim, Germany) that can be passed through the stenotic lesion for both angioplasty and stenting delivery inside the lumen and a laser-cut, closed-cell, self-expanding Credo stent (Acandis GmbH) with a retrievable design for adjustable placement. This new stenting within the balloon catheter eliminates the need for a long exchange wire and reduces the contact of device passage to the stenotic lesion. The NeuroSpeed PTA balloon catheter has been utilized with the Acclino Flex (Acandis GmbH) or Credo stent in ICAS⁸ and failed mechanical thrombectomy.^9,10 Although this promising technique was CE approved in 2014, its technical safety and outcomes have not been compared with those of the Food and Drug Administration-approved Wingspan stent system in the management of SMR-ICAS. Additionally, prior large-scale studies, such as the SAMMPRIS and WEAVE/WOVEN trials, used decreased National Institutes of Health Stroke Scale scores as indicators of ischemic events.^5–7 However, silent infarction is approximately 10 times more prevalent than neurological symptoms,¹¹ suggesting that the real incidence of embolic events during stenting may be significantly underestimated. This preliminary study was designed to compare the technical success, safety, and outcomes of the Credo and Wingspan stent systems in PTAS for patients with SMR-ICAS. We also applied periprocedural magnetic resonance imaging (MRI) to detect silent embolic ischemic lesions (SEILs) on diffusion-weighted imaging (DWI) to highlight embolic burdens associated with the two-stent systems.

2. METHODS

2.1. Study population

Data in a prospectively collected database from patients who were diagnosed with SMR-ICAS over the anterior circulation and who received PTAS between 2018 and 2022 were analyzed. Endovascular treatment plans were discussed after consensus meetings between vascular neurologists and interventional neuroradiologists. The indications for PTAS included stenosis exceeding 70% over the cavernous segment of the internal carotid artery (ICA) to the middle cerebral artery (MCA) M1 segment, resistance to medical treatment with persistent or recurrent cerebral ischemic symptoms, absence of evidence of perforator stroke, and a clinically modified Rankin scale score of <3 points. The measurement of percent stenosis was based on the WASID criteria.¹² The patients were separated into two groups based on the type of stent used: Wingspan or Credo. Patients without periprocedural MRI or without imaging follow-up within 1 year after the procedure were excluded from this study. This study was approved by our institutional review board (No.: 2018-07-031BCF).

2.2. Percutaneous angioplasty and stenting

Dual-antiplatelet premedication agents (aspirin 300 mg and Plavix 75 mg daily) were administered for a minimum of 3 days. Heparinization ranging from 3000 to 5000 IU was prescribed based on the activated coagulation time. All patients underwent complete diagnostic neuroangiography, including of the carotid and vertebral arteries, utilizing biplane angiography systems. A three-dimensional angiogram of the target artery was conducted to clarify the stenotic lesion. The procedures were performed under general anesthesia. Following diagnostic angiography, a 6-French guiding sheath (Flexor Shuttle; Cook Medical, Bloomington, IN, USA) was placed in the ipsilateral common carotid artery or cervical ICA of the lesion, followed by the placement of a 6-French intermediate catheter (Neuron 070; Penumbra, Inc., Alameda, CA, USA; or Navien 072; MicroTherapeutics, Inc., DBA ev3 Neurovascular, Irvine, CA, USA) in the petrous or cavernous segment of the ICA.

In the Wingspan group, a microcatheter with a 0.014-inch access guidewire was navigated across the stenosis. The wire was then substituted with a 300-cm exchange guidewire. After removing the microcatheter, a noncompliant balloon catheter was introduced over the guidewire and toward the target lesion for predilation. The chosen diameter of the balloon ranged from approximately 80% to 90% of the adjacent normal target vessel based on diagnostic angiography. PTA was then performed with slow and gradual balloon inflation for 2 to 3 minutes (Fig. 1B), followed by retrieval of the balloon and deployment of a Wingspan stent to the target lesion along the exchange guidewire (Fig. 1C).

Fig. 1.

Two patients with SMR-ICAS treated with different stents: Wingspan for the left cavernous to petrous ICA (A-D) vs Credo for the right MCA M1 segment (E-H). The presenting angiographies showed significant (>70%) stenosis (A, E). In the Wingspan stent system, following guidewire exchange to a 300-cm wire, angiography was performed with a Gateway PTA balloon catheter (B). Subsequently, the balloon was removed and replaced with the Wingspan stent, which was then deployed along the wire (C; the dotted arrows show the stent markers; the solid arrows show the stent catheter tip). For Credo stenting, angioplasty was performed with a NeuroSpeed PTA balloon catheter (F), followed by guidewire retrieval and stent advancement within the balloon catheter without direct contact with the atherosclerotic plaque. Stenting was performed with the unsheathed technique (G; the dotted arrows show the stent markers, and the solid arrows show the balloon catheter markers). Follow-up CT angiography by curved multiplanar reconstruction showing the patent vessels and stents (D, H; white arrows indicate the distal and proximal markers). CT = computed tomography; PTA = percutaneous transluminal angioplasty; SMR-ICAS = severe medically refractory intracranial artery stenosis.

In the Credo group, a NeuroSpeed PTA balloon catheter coupled with a 0.014-inch access guidewire was advanced through the stenotic site. The balloon diameter was selected to be 80% to 90% of the adjacent normal target vessel diameter. The PTA procedure was conducted slowly and gradually for 2 to 3 minutes (Fig. 1F). The guidewire was withdrawn, and the Credo stent was then advanced within the NeuroSpeed balloon catheter and subsequently deployed at the target lesion site (Fig. 1G).

The target pressure of the PTA for all patients was <7 atm.^13,14 After stenting, angiography was performed to evaluate the outcome of revascularization and residual stenosis. Noncontrast on-site cone beam Computed Tomography (CT) of the brain was also conducted to screen for any intracranial hemorrhage.

2.3. Technical success and safety

Technical success was defined as successful deployment of the stent with a residual stenosis <30%.¹³ The technical safety was evaluated by assessing periprocedural neurological complications within 30 days, which included intracranial hemorrhage, SEIL, ischemic stroke, and mortality. Patients underwent periprocedural MRI scans both within 3 days before and within 24 hours after PTAS to detect any newly developed SEILs. The number of recent embolic lesions in the treated vascular territory on DWI was recorded for analysis.

2.4. Outcome evaluation

The outcome of PTAS for SMR-ICAS patients was evaluated based on significant (>50%) in-stent restenosis (ISR) during the 1-year follow-up period¹³ and the presence of clinically recurrent cerebral ischemic stroke in the treated vascular territory. The degree of stenosis was evaluated using CT angiography with a curved MPR (Fig. 1D, H) and based on the WASID method.¹²

2.5. Statistics

Qualitative data are presented as percentages and were evaluated using Fisher exact test due to the limited sample size. Quantitative data are presented as the means ± SDs and were analyzed using the Mann-Whitney U test. A two-tailed p-value <0.05 was considered to indicate statistical significance. Statistical analyses were performed using the Statistical Product and Service Solutions (SPSS; IBM Corporation, Armonk, NY) statistical software package, version 29.0.

3. RESULTS

3.1. Patient characteristics

The demographic data are presented in Table 1. A total of 29 patients—17 who received a Wingspan stent and 12 who received a Credo stent—were included in this study. The patients in both groups were predominantly male. Patients with diabetes mellitus tended to predominate in the Wingspan stent group (p = 0.05). No substantial differences existed in the other baseline characteristics.

Table 1.

Baseline demographic characteristics of the study patients

	Wingspan stent group (n = 17)	Credo stent group (n = 12)	p
Baseline demographic characteristics
Patient no., n (%)	17 (58.6%)	12 (41.4%)
Age, y (mean ± SD)	63.53 ± 16.33 (21-87)	61.42 ± 12.40 (40-79)	0.549
Sex, n (%)			0.665
Male	12 (70.6%)	10 (83.3%)
Female	5 (29.4%)	2 (16.7%)
Underlying disease, n (%)
Diabetes mellitus	9 (52.9%)	2 (16.7%)	0.053
Hypertension	16 (94.1%)	9 (75.0%)	0.279
Dyslipidemia	12 (70.6%)	7 (58.3%)	0.694
History of smoking, n (%)			0.710
No	9 (52.9%)	5 (41.7%)
Yes	8 (47.1%)	7 (58.3%)

The results are shown as the number (%) or mean ± SD (range).

3.2. Percutaneous angioplasty and stenting

The stenosis lesion and stent characteristics are summarized in Table 2. We found no significant differences between the two groups in terms of target vessel laterality (p = 0.72), degree of stenosis (p = 0.74), stent diameter (p = 0.71), or stent length (p = 0.10).

Table 2.

Comparison of stenosis lesions, stent characteristics, technical success rates, periprocedural complications, and short-term outcomes between the Wingspan stent group and the Credo stent group

	Wingspan stent group (n = 17)	Credo stent group (n = 12)	p
Stenosis lesion
Target vessel, n (%)			0.274
MCA	6 (35.3%)	7 (58.3%)
Intracranial ICA	11 (64.7%)	5 (41.7%)
Lesion side, n (%)			0.716
Left	7 (41.2%)	6 (50.0%)
Right	10 (58.9%)	6 (50.0%)
Stenosis (mean ± SD)	82.6% ± 9.6% (70.0%-99.0%)	81.3% ± 8.3% (70.0%-90.0%)	0.735
Stent characteristics
Stent diameter, mm (mean ± SD)	3.74 ± 0.66 (2.50-4.50)	3.71 ± 0.84 (3.00-5.50)	0.713
Stent length, mm (mean ± SD)	19.41 ± 1.66 (15-20)	20.83 ± 2.89 (20-30)	0.102
Technical success, n (%)	16 (94.1%)	11 (91.7%)	1.000
Safety
Periprocedural intracranial hemorrhage, n (%)	2 (11.8%)	0 (0.0%)	0.498
Silent embolic ischemic lesion of ipsilateral MCA territory on DWI, n (%)			0.422
No	4 (23.5%)	5 (41.7%)
Yes	13 (76.5%)	7 (58.3%)
Ischemic stroke or death within 30 d after stenting	0 (0.0%)	0 (0.0%)
Outcomes
Significant in-stent restenosis (>50%) within 1 y	4 (23.5%)	2 (16.7%)	1.000
Recurrent stroke within 1 y	0 (0.0%)	0 (0.0%)

The results are shown as the number (%) or mean ± SD (range).

MCA = middle cerebral artery.

DWI = diffusion-weighted imaging

3.3. Technical success, safety, and outcomes

The technical success, safety, and outcomes are summarized in Table 2. In the Wingspan group, technical success was achieved in 94% (16/17) of the patients. The only failure that occurred was due to the early termination of the procedure without stenting. This patient suddenly developed hypertension exceeding 200 mm Hg after passing the microwire through the stenosis. Subsequent cone beam CT revealed minimal SAH along the sylvian fissure (Fig. 2A). No progression of the hemorrhage was observed on follow-up CT 1 day later, and the patient was discharged 3 days later with no new neurological deficits. In the Credo group, technical success was achieved in 92% (11/12) of the patients. A single failure occurred when a hard stenotic lesion had residual stenosis exceeding 50% after PTAS, which surpasses the definition of technical success in this study (<30% residual stenosis).

Fig. 2.

Periprocedural intracranial hemorrhage during Wingspan stent deployment. A, One patient’s blood pressure suddenly increased to more than 200 mm Hg after the microwire passed through the stenosis; cone beam CT showed SAH in the right sylvian fissure, and the procedure was terminated without stenting. B, One patient who underwent PTAS with a Wingspan stent achieved <30% residual stenosis. A routine postprocedural cone beam CT scan revealed SAH in the left sylvian fissure. CT = computed tomography; PTAS = percutaneous angioplasty and stenting; SAH = subarachnoid hemorrhage.

The cases of two patients in the Wingspan cohort were complicated by intracranial hemorrhage. One patient experienced technical failure without stenting. The other patient underwent successful PTAS, with the residual stenosis decreasing from 95% to <30%. However, a small amount of SAH in the sylvian fissure was noted during the poststenting evaluation (Fig. 2B). The patient presented with postprocedural headache but had no new neurological deficits and was discharged after conservative treatment. No patients who received the Credo stent experienced periprocedural ICH.

A total of 76% of patients in the Wingspan stent group and 58% of patients in the Credo stent group had embolic evidence on periprocedural DWI. No major ischemic stroke or mortality occurred within 30 days following PTAS in either group.

The ISR rate within 1 year was 24% in the Wingspan group and 17% in the Credo group, with no statistically significant difference observed (p = 1.00). The target vessels of most ISR patients were in the ICA, including 3 of the 4 ISR patients in the Wingspan group and both ISR patients in the Credo group. During the postprocedure 1-year follow-up period, no patient experienced recurrent stroke in the treated vascular territory.

4. DISCUSSION

This study preliminarily revealed that the technical success, safety, or outcome of PTAS did not significantly differ between the Credo or Wingspan stent system for patients with SMR-ICAS over the anterior circulation. Notably, no complications or intracranial hemorrhages occurred in the Credo stent group. This study is the first to compare intracranial PTAS using stent delivery using the PTA balloon catheter design (Credo) with the traditional approach of performing the first PTA and then exchanging it with a stent design (Wingspan). This work is also the first clinical study using periprocedural MRI to compare the silent embolic burden between two different intracranial stent systems.

For patients with SMR-ICAS, medical management, including aspirin, blood pressure control, statins, and lifestyle changes, remains the first-line treatment¹⁵ because a previous large-scale randomized trial revealed higher 30-day stroke and mortality rates for ICAD patients receiving intracranial stenting than for those receiving aggressive medical treatment.⁵ One possible reason for this difference may be related to the timing of PTAS: early intervention after symptom onset could cause unstable plaque dislodgement^5,16 and insufficient antiplatelet medication, which increase the risk of thromboembolism^6,16 and reperfusion hemorrhage in the acute stage of stroke.⁶ By delaying the PTAS procedure to more than 7 days after the stroke episode, using a stricter patent selection with the requirements of two episodes of stroke and more than 70% vascular stenosis, having PTAS performed by experienced interventionalists, and using underdilatation during angioplasty, the major periprocedural complication rate for PTAS with a Wingspan stent decreased from 14.7% in the SAMMPRIS trial to 2.6% in the WEAVE trial.^5,6 Notably, this rate was lower than that of the medical management group (5.8%) in the SAMMPRIS trial.⁵ These findings suggest that PTAS may be useful for ICAD patients who fail aggressive medical therapy.

Because vulnerable plaques may have been a key factor in the SAMMPRIS trial, minimizing contact between endovascular devices and atherosclerotic plaques during PTAS could improve technical safety. The NeuroSpeed PTA balloon catheter with the Credo stent combines PTA and stent delivery within the same catheter, eliminating the need for a long exchange wire and decreasing the risk of repeated device passage through vulnerable plaques and wire migration.⁸

Deploying the Wingspan stent requires navigating a microwire within a microcatheter through the stenosis to the target branch artery, followed by exchanging it for a 300-cm long wire. Angioplasty is performed by an over the wire Gateway balloon catheter. After successful dilation, the balloon catheter is withdrawn over the long wire. Finally, the Wingspan stent is delivered to the stenosis site. This procedure involves four device passes: wire, microcatheter, balloon catheter, and stent. In contrast, Credo stent placement involves only two passes through the stenosis. A regular 205-cm wire and a NeuroSpeed PTA balloon catheter are navigated through the stenosis. Following angioplasty, the Credo stent is delivered within the NeuroSpeed catheter to the stenosis site. Although the difference was statistically nonsignificant, the reduced number of device passages in the Credo group might be associated with the numerically lower embolic burden on DWI in this study. The procedure differences are highlighted in Fig. 3.

Fig. 3.

Illustrative comparison of stent deployment for the Wingspan (A-E) and Credo (A, F-H) systems. In the Wingspan stent, we first advanced a microwire through the stenosis (A), followed by a microcatheter (B), and then the microwire was exchanged for a 300-cm long wire. We then exchanged the microcatheter with a Gateway PTA balloon catheter and performed angioplasty (C). The PTA balloon catheter was then changed for a Wingspan stent (D), and the unsheathed technique was used to deploy at the stenosis site (E). A total of four device passages through the atherosclerotic plaque were needed. In the Credo system, the microwire was advanced through the stenosis (A), followed by a Neurospeed PTA balloon catheter, and angioplasty was performed (F). The microwire was then exchanged for a Credo stent inside the original PTA balloon catheter (G), and deployment was performed via the unsheathing technique (H). Two device passages occurred during Credo stent deployment. The number in the black circle indicates the number of devices that passed through the plaque. PTA, percutaneous transluminal angioplasty.

Another difference between the two stents was the wire length. The Wingspan stent requires another 300 cm of exchange wire, while the Credo stent uses 205 cm of wire. In addition to the operator, an experienced assistant is required to exchange and stabilize the long 300-cm wire when the balloon catheter and stent are advanced. The regular-length wire used in the NeuroSpeed balloon catheter is easier to handle and may reduce the risk of accidental injury to distal branches of the intracranial artery. This potential benefit might explain why our study revealed two cases of periprocedural intracranial hemorrhage in the Wingspan group, whereas none occurred in the Credo group.

The other key difference lies in the stent design itself. The Credo stent has a closed-cell design, allowing for retrieval and adjustment before full deployment. In contrast, the Wingspan stent has an open-cell design and cannot be repositioned once deployment begins. In addition, the use of open-cell and closed-cell designs is related to differences in ISR, although the conclusions of different carotid artery studies are inconsistent. A study showed that open-cell stents with a greater cell-free area were related to a higher rate of ISR, but significant restenosis (>70%) was too rare to compare.¹⁷ Another study suggested that the presence of closed-cell stents, which have rigid surfaces and may irritate the intima, was related to a higher rate of moderate or greater restenosis (>50%), but the difference in severe restenosis (70%) was not significant.¹⁸ However, the histological composition of intracranial arteries differs from that of extracranial arteries. These patients have a thinner media and adventitia, thicker internal elastic lamina in the intima, and later development of the vasa vasorum after aging or atherosclerosis changes.¹⁹ Therefore, the applicability of restenosis resulting from extracranial carotid arteries to intracranial arteries is questionable. In this study, the incidence of significant ISR (>50%) was not significantly different between the Wingspan stent group (24%) and the Credo stent group (17%).

The stent design may also influence the risk of distal emboli. Studies have shown that open-cell carotid artery stents are associated with increased rates of subclinical emboli, as detected by MRI.^20,21 In our study, the open-cell design of the Wingspan stent potentially contributed to the observed increase in the SEIL on MRI.

Another difference between closed-cell and open-cell designs is the ability to conform to the vessel wall. The intracranial arteries, particularly at the carotid siphon, are more tortuous than the extracranial carotid arteries. Poor vessel wall apposition is related to flow disturbance and delayed intimal growth in coronary arteries.²² In our study, we observed slightly poor vessel wall apposition with the Credo stent at the carotid siphon. This observation may be due to the closed-cell design and rigid contexture. Incomplete wall apposition potentially increases the risk of late thrombus formation.²³ Furthermore, performing additional procedures, such as poststent angioplasty, is also difficult in our experience. Our study revealed no statistically significant difference in restenosis rates between the two-stent groups within 1 year. However, further large studies with longer follow-up periods are needed to assess these differences.

This preliminary study also highlighted the embolic burden during PTAS. Our findings revealed a significantly greater incidence of SEIL on DWI in the periprocedural MRI of both groups: 76% in the Wingspan group and 58% in the Credo group. This rate is considerably greater than the rate of symptomatic embolic strokes reported in the SAMMPRIS trial (3 out of 129 patients in the anterior circulation).²⁴ Our results align with a study using periprocedural MRI to detect asymptomatic ischemic brain lesions.²⁵ This finding suggested a high incidence of silent embolic events during intracranial PTAS. Silent brain infarction is much more common than symptomatic stroke,¹¹ and increased embolic load during vascular procedures is a major contributor to this condition²⁶ and related to impaired cognitive function following procedures.²⁷ The SAMMPRIS trial included only symptomatic strokes and potentially missed a significant number of asymptomatic embolic events.

Although not encountered in this study, a potential limitation of the NeuroSpeed PTA balloon catheter is the retrieval of the guidewire before stenting. Without the guidewire in place, navigating to the occluded or dissected vessel lumen distal to the deployed Credo stent could be challenging. Therefore, after PTA, the operator should perform a control angiogram via the intermediate catheter to confirm vessel patency before removing the guidewire and remain alert for signs of vasospasm or arterial dissection. The selection of balloon diameter and inflation pressure should also be conservative to minimize the risk.

This study has some limitations. The sample size was limited. Our results did not reach statistical significance, and we could only conclude that the Credo stent was comparable to the Wingspan stent in this preliminary study. Second, our team has experience with the Wingspan stent, which may have influenced our procedural success and patient outcomes. Conversely, the Credo stent is a newly introduced device, and more experience is needed with this device to become familiar with its use. This potential bias could have impacted the comparison between the two systems.

Despite these limitations, this preliminary study demonstrated that the Credo stent system was comparable to the Wingspan stent system and provided insights into the potential of the Credo stent system for treating SMR-ICAS. To solidify these findings and draw more definitive conclusions, larger-scale randomized controlled trials are necessary.

ACKNOWLEDGMENTS

This study was supported by grants from Taipei Veterans General Hospital, Taiwan [V112B-032 (to T.-M.L.); V111B-032, V112B-007 (to C.-H.W.); V110C-037, V111C-028, V112C-059, V113C-182, V112D67-002-MY3-2 (to F.-C.C.)], Veterans General Hospitals and University System of Taiwan Joint Research Program [VGHUST 109V1-5-2 and VGHUST 110-G1-5-2 (to F.-C.C.)], National Science and Technology Council, Taiwan [MOST 110-2314-B-075-005 and 111-2314-B-075-025-MY3 (to C.-H.W.), 109-2314-B-075-036, 110-2314-B-075-032, 112-2314-B-075-066, and 113-2314-B-075-037 (to F.-C.C.)], Yen Tjing Ling Medical Foundation, Taiwan [CI-109-3, CI-111-2, CI-112-2 (to C.-H.W.)], Professor Tsuen CHANG’s Scholarship Program from Medical Scholarship Foundation In Memory Of Professor Albert Ly-Young Shen (to T.-M.L. and C.-H.W.), and Vivian W. Yen Neurological Foundation (to C.-H.W. and F.-C.C.).