Angiography-based hemodynamic features predict recurrent ischemic events after angioplasty and stenting of intracranial vertebrobasilar atherosclerotic stenosis

Intracranial atherosclerotic stenosis (ICAS) is a major cause of ischemic stroke worldwide [1], accounting for almost 30% of ischemic strokes in Asia [2]. Hypoperfusion and artery-to-artery embolism caused by ICAS are important mechanisms leading to ischemic stroke [3]. Patients with symptomatic ICAS have a stroke risk of over 10% per year, despite optimal medical treatment [4, 5]. Although not widely applied to the general ICAS population, percutaneous transluminal angioplasty and stenting (PTAS) and angioplasty alone are still important treatment options for high-risk or relapsed ICAS after ineffective medical treatment. Moreover, PTAS undergoing balloon angioplasty before or after stent implantation could achieve better stenosis relief and perfusion improvement. Earlier pivotal trials indicated an apparent increase in stroke and death in patients with ICAS after stenting [5, 6]. In contrast, recent studies revealed positive results of PTAS for symptomatic ICAS with experienced interventionalists and proper patient selection [7, 8]. Current data show that the 1-year recurrence rate of ICAS patients treated with PTAS is around 8.0–8.5%.

To improve the benefit of PTAS treatment, it is crucial to perform comprehensive assessments and identify the risk factors of long-term recurrent stroke. Compared with luminal stenosis, pilot studies have shown that detailed intracranial atherosclerotic features and hemodynamic status were more relevant to clinical presentations and subsequent ischemic events [9‐12]. Digital subtraction angiography (DSA)–defined parameters, in particular, those before and after stent deployment, observed or measured during the PTAS have not been paid sufficient attention. DSA is the current gold standard for luminal stenosis measurement. It, however, can also provide semi-quantitative parameters as a supplement for classical hemodynamic values. Although this can also be quantified by computed tomography perfusion, magnetic resonance perfusion–weighted imaging, and arterial spin labeling, the artifacts of posterior cranial fossa and anatomical vascular variations may hinder the detailed evaluation of the hemodynamic of ICAS in the posterior circulation. Quantitative DSA (q-DSA) allows a real-time evaluation of hemodynamics with superior temporal and spatial resolution. Its feasibility for evaluating blood flow changes has been demonstrated in various cerebrovascular diseases, such as cerebral vasospasm, arteriovenous malformations, and steno-occlusive arterial disease [13‐17].

Quantitative DSA can detect subtle changes in blood flow in the angiography suite during the intervention [18]. The main aim of this study is, therefore, to identify the clinically relevant pre- and postprocedural hemodynamic parameters derived from q-DSA, and assess their potential prognostic values in patients with posterior ICAS who underwent PTAS therapy.

From January 2012 to May 2020, 307 ICAS patients who underwent endovascular therapy were registered in the Nanjing Stroke Registry Program, and a total of 137 patients were included in this study. The flow chart of patient selection is summarized in Supplementary Fig. 1. Patients were 62.1 ± 9.7 years, and 114 (83.2%) were men. Of these 137 patients, 129 patients (94.2%) had ischemic strokes and 8 TIA. Ninety-eight (71.5%) lesions located in vertebral artery (VA) and 39 (28.5%) in basilar artery (Table 1). One hundred and eight cases were identified as Mori A or B type, and 29 cases were Mori C type. The detailed characteristics of all patients are listed in Table 1.

This is a longitudinal study evaluating the prognostic value of hemodynamic parameters derived from q-DSA in ICAS patients treated with PTAS. Several inspiring findings emerged from this study. Above all, even though the PTAS procedure markedly improved the blood flow of the target vessels, the preprocedural hemodynamic status was still significantly related to the prognosis. The prolonged rTTP of the target vessel and shortened translesional TTP difference derived from DSA imaging before the procedure were independently associated with an increased risk of stroke relapse. Secondly, prior TIA, low levels of HDL, and high NLR were also identified as the risk factors for subsequent ischemic stroke. Thirdly, multivariate analyses showed that hemodynamic parameters provided an incremental prognostic value over traditional risk factors for recurrent stroke prediction.

For the first time, we conducted a longitudinal study to evaluate the prognostic value of hemodynamic features in posterior ICAS patients treated with PTAS, using the q-DSA technique. Q-DSA has been used to evaluate subtle hemodynamic changes to determine the treatment efficacy for several cerebrovascular diseases, e.g., cerebral aneurysms, cerebral vasospasm, arteriovenous malformations, steno-occlusive arterial disease, and subarachnoid hemorrhage [16‐18, 24]. Studies also demonstrated the feasibility of q-DSA for the evaluation of peri-therapeutic hemodynamic changes for carotid stenting [25]. Keita et al demonstrated the predictive value of hemodynamic parameters derived from q-DSA in hyperperfusion phenomenon after carotid artery stenting and angioplasty [26]. Jayme et al confirmed that mean transit time generated from q-DSA could predict hemorrhagic transformation in patients following endovascular thrombectomy [14]. Multiple hemodynamic parameters can be obtained from q-DSA, including arrival time, peak density, AUC, TTP, full width half maximum, etc. Several factors that affect the quantitative analyses of DSA include the location of the catheter, injection protocol, and frame rate [27, 28]. However, the differences in TTP in different sites can reflect relative temporal delays, offsetting differences in catheter location and injection. Therefore, TTP is considered one of the most robust parameters in clinical practice, regardless of the DSA acquisition protocol [25, 29].

This study revealed that prolongation of preprocedural rTTP of the target vessel and shortened trans-stenotic TTP at the preprocedural DSA imaging were associated with recurrent events. Prolonged TTP reflects low blood flow velocity implying a low local blood pressure gradient and therefore decreased perfusion. In addition, lumen loss due to stenosis lesions increases focal flow velocity, which is manifested as a shortening of trans-stenotic TTP difference. The association between reduced blood flow and recurrent ischemic events had been suggested in patients with vertebrobasilar artery stenosis [30]. Insufficient perfusion, together with artery-to-artery embolism, and plaque extension over the orifice of penetrating arteries have been hypothesized as the main mechanism of ischemic events [1, 31].

Although PTAS procedures significantly improved perfusion status, patients with more restricted baseline blood flow still showed higher risks of stroke recurrence. Restricted blood flow often indicates more severe and continuous cerebral hypoperfusion. Chronic cerebrovascular hypoperfusion may injure the vascular endothelium, resulting in local neurovascular unit dysfunction and disruption [32‐34]. The mechanisms result in permanent cerebral damage [33]. Therefore, even if the blood flow status is corrected, the residual risk of patients with worse baseline blood flow status is still higher than that of other patients.

The decreased perfusion could be due to insufficient upstream pressure or high downstream resistance. Apart from the insufficient perfusion pressure that can be improved after PTAS procedures, prolonged TTP may also be the result of increased downstream resistance, which was caused by chronic damage due to high-risk cardiovascular factors [35]. Comorbidities worsen cerebral circulation through several mechanisms, including hypoperfusion, diminished autoregulatory capacity, and increased permeability of the blood–brain barrier [36]. This highlights the importance of secondary prevention to prevent recurrent ischemic stroke. For patients with poor baseline hemodynamic status, even underwent PTAS procedure, more attention and intensive medical treatment were still needed. In other words, high-risk patients need more close monitoring, intensified follow-ups, and individualized secondary prevention strategies, such as longer duration of the dual antiplatelet therapy and intensive statin therapy.

Despite interesting and useful findings, limitations exist. Firstly, due to the high image quality requirements, a considerable number of patients failed to meet the inclusion criteria, resulting in a possible selection bias. Secondly, the semi-quantitative assessment from 2D DSA images can only infer hemodynamics, and there is a lack of more pulsatile flow parameters to reveal the detailed hemodynamic state. Thirdly, the small sample size and low heterogeneity of the population may limit the generalizability of conclusions. Therefore, further prospective studies with large samples are warranted to validate these findings.

In the current study, q-DSA was applied to evaluate the pre- and postprocedural hemodynamic status of ICAS patients receiving PTAS. Although PTAS procedures could significantly improve blood flow status, patients with more restricted blood flow at baseline still showed a higher risk of stroke recurrence. More precisely, posterior ICAS patients with the prolonged TTP of the target vessel and shortened trans-stenotic TTP difference at the preprocedural DSA imaging were at a significant higher risk of stroke recurrence in the long-term follow-up. Therefore, baseline hemodynamic information obtained from q-DSA could be used for risk stratification in patients with ICAS.