Approximately 20% of ischemic stroke subtypes occur in the posterior circulation, and vertebral artery (VA) stenosis is a significant but often underestimated cause of posterior ischemic strokes (1, 2). The V1 segment of the vertebral artery originates from the ostium of the fifth or sixth cervical vertebrae. Due to the curved anatomy and slow blood flow velocity of the V1 segment,it is the most frequent site of severe occlusive lesions in the vertebral artery (3) and is associated with an increased risk of poor functional outcomes and mortality compared to carotid circulation infarctions (4). However, the optimal treatment for vertebral artery stenosis remains controversial. On one hand, open surgery for V1 stenosis is constrained by the incidence of perioperative complications (5); on the other hand, endovascular treatment with stenting is an attractive treatment option because of its safety and higher rates of technical success (6, 7). However, the efficacy of stenting remains uncertain due to the complex vascular anatomy and the risk of in-stent restenosis (ISR) (8, 9).

The transradial approach (TRA) has been gaining attention among interventional physicians in recent years as an alternative artery access for endovascular treatments in recent years (10). Neurointerventional physicians have explored the feasibility of TRA in various procedures, including diagnostic cerebral angiography, aneurysm clotting, and endovascular stenting (11–14). Previous studies have reported several advantages of TRA over the transfemoral approach (TFA) in terms of direct access to the vertebral artery, easy navigation to the complex arch anatomy, and fewer perioperative complications (15). However, prospective and randomized evidence remains lacking, and few studies have investigated the feasibility of endovascular treatment via TRA in V1 segment stenosis. Hence, we conducted a retrospective and non-randomized study, reporting our multicenter experience in comparing the efficacy and safety of TRA vs. TFA for stenting in patients with V1 segment stenosis.

During the study period, a total of 109 patients treated with endovascular treatment for the vertebral artery V1 segment stenosis were enrolled. The mean age was 66.2 ± 7 years, and 80.7% were men. A total of 45 patients were treated via TRA, and 64 patients were treated via TFA. Four cases were missing for total cholesterol, four for triglycerides, six for high-density lipoprotein, and seven for low-density lipoprotein. Little’s Missing Completely at Random (MCAR) test indicated missing at random (p = 0.844). The baseline characteristics of the treatment groups are listed in Table 1. They did not differ in demographics, comorbidities, admission features, surgical experience, stent types, stenting techniques, stent location, and lesion characteristics except for the level of triglycerides [TRA: 1.3 (1.0, 1.6), TFA: 1.2 (0.9, 1.5), p = 0.045].

As listed in Table 2, the rate of technical success (TRA: 100%, TFA: 100%) and the degree of residual stenosis [TRA: 2.7% (0.9, 7.5), TFA: 3.8% (0.0, 7.2), p = 0.509] were similar between both treatment groups. The TRA group [43 (39, 45) min] had a significantly shorter procedure time compared to the TFA group [50 (38, 62) min, p = 0.005, Supplementary Figure S2]. The post-procedure complications were slightly higher in the TFA group (10.9%) than in the TRA group (2.2%), although this difference did not reach statistical significance due to the limited sample size. The rate of minor complications was 2.2% (1 radial occlusion) for the TRA group and 7.8% (5 puncture site hematoma) for the TFA group. The rate of major complications was 0.0% for the TRA group and 3.1% (1 thromboembolic event and 1 vessel dissection) for the TFA group. Post-procedure mediations were similar in both groups. The length of stay in the hospital was shorter in the TRA [9 (7, 12) days] group compared to the TFA group [10 (9, 14) days, p = 0.048]. Long-term outcomes were also similar between TRA [90 days mRS: 0 (0, 1) points, ischemic stroke and death: 2.2%] and TFA groups [90 days mRS: 0 (0, 1) points, ischemic stroke and death: 1.6%]. The Kaplan–Meier curve revealed a similar risk of ISR in the TRA and TFA groups (28.9% vs. 34.4%, p = 0.355, Figure 2).

After matching, the TRA group had a shorter procedure time [43 (39, 45) vs. 50 (38, 59) min, SMD = 0.735] and a shorter length of stay [9 (7, 11) vs. 10 (7, 11.2) days, SMD = 0.342]. The association between treatment groups and clinical outcomes was not significant both in the matched cohorts and IPTW analysis (post-procedure complications: OR: 0.231; 95% CI: 0.026–2.070; p = 0.188; ISR: OR: 0.924; 95% CI: 0.356–2.401; p = 0.870; ischemic stroke and death: OR: 1.909; 95% CI: 0.109–33.305; p = 0.655; Table 3 and Supplementary Table S1). Falsification endpoint analysis suggested that unmeasured confounders were well-balanced between treatment groups (OR: 2.079; 95% CI: 0.296–14.586; p = 0.458). The E-value analysis revealed that unmeasured confounders required a risk ratio of 3.528 for post-procedure complications, 1.245 for ISR, and 2.108 for ischemic stroke and death events to explain the association. Additionally, we observed that surgeons with greater surgical experience tend to favor TRA, resulting in shorter procedure times and a slightly lower incidence of surgical complications compared to those with less experience (Supplementary Table S2). Left vertebral artery stenting was associated with an increased risk of procedural complications (total: 10% vs. 5.5%; minor: 6% vs. 5.5%; major: 4% vs. 0%) compared to right vertebral artery stenting. Despite similar success rates between the groups, the mean procedure duration was marginally longer in the left stenting (45.5 min vs. 44.0 min). In the long term, the risks of ischemic stroke (4% vs. 0%) and in-stent stenosis (42% vs. 25.5%) were elevated in the left stenting. In patients treated via TRA, left vertebral artery stenting had a higher prevalence of residual stenosis, accompanied by slightly longer procedural times and a higher incidence of complications compared to right stenting (Supplementary Table S3).

In the present study, we compared the technical feasibility and safety between TRA and TFA for vertebral artery stenting of V1 segment stenosis. We observed a similar rate of technical success and overall complications between the two approaches, as well as short procedure times and lengths of stay of TRA for stenting. To the best of our knowledge, this was the first study comparing TRA with TFA for stenting of V1 segment stenosis.

TRA is an alternative approach for endovascular procedures and has gained popularity among neurointerventionalists over the past years (10). A recent randomized controlled trial reported that TRA was associated with shorter time from sheath insertion to stent insertion and higher patient acceptance and satisfaction (21). Schartz et al. conducted a meta-analysis to investigate the difference in complication rates between TRA and TFA for neurovascular procedures. After analyzing 17 studies, they reported that the rates of access site complications were 1.8 and 3.2% for TRA and TFA, respectively (22). Catapano et al. performed a retrospective analysis among 877 patients receiving neuroangiographic procedures and reported that the overall complication rate with TFA procedures [60/844 (7%)] was significantly higher than TRA procedures [4/206 (2%)] (16). In our study, we reported that TFA procedures had a slightly higher complication rate compared to TRA procedures, although this difference did not reach statistical significance due to the small sample size. Our results also indicated that the length of stay was significantly shorter in TRA procedures. Literature in cardiology suggested that TRA was associated with a reduction in the post-procedural length of stay for acute myocardial infarction patients undergoing rescue angioplasty (TRA: 7.0, TFA: 7.9 days, p < 0.05) (23). Ge and Wei (24) reviewed 1,048 cerebral angiograms and found that the duration of hospital stay was shorter in the TRA group (123.8 h) compared to the TFA group (168.7 h, p < 0.05).

Endovascular treatment for V1 segment stenosis of the vertebral artery is generally preferred over aggressive medical management due to the specific location of the segment, which is challenging to access surgically (25, 26). Recently, the Vertebral Artery Ischemic Stenting Trial (VIST) revealed that stenting for the extracranial vertebral artery was safe with low complications and might have potential efficacy for preventing stroke recurrence (27). However, the main problem with stent therapy is the high incidence of ISR, which varies from 15 to 60% according to different study designs (8, 28). Our results found that TRA did not increase the incidence of ISR after stenting for the V1 segment (28.9% vs. 34.4%). Che et al. (2) investigated the short- and long-term outcomes after stenting for V1 segment stenosis and revealed that the treatment approach was not included in the final multivariable analysis. Guo et al. (29) conducted a retrospective analysis of a multicenter registry among patients with symptomatic intracranial vertebrobasilar artery stenosis treated with TRA or TFA, reporting an incidence of symptomatic ISR of 6% for the TRA group and 13.1% for the TFA group (p = 0.692). These results supported the potential safety and efficacy of TRA for the treatment of vertebrobasilar artery stenosis. Notably, anatomic factors may influence the outcome of vertebral artery stenting. A previous review reported that placement of stents on the left vertebral artery via TRA is affected by the angle and distance between the innominate and left subclavian arteries (30). In our study, left-sided procedures were associated with a higher rate of procedural complications despite similar technical success rates for both left and right approaches. The mean procedure duration was slightly longer for left-sided stenting. In addition, the left vertebral artery is associated with increased risks of ischemic stroke and in-stent stenosis, and patients treated via TRA showed a higher prevalence of residual stenosis.

Despite the technical safety and patient preference for TRA, TFA remains the preferred choice among interventional neurologists (31). Factors limiting the promotion of TRA may include the steep learning curve and the lack of evidence from large randomized controlled trials. Wilkinson et al. retrospectively examined 500 cerebral angiograms and recorded the fluoroscopy time at different stages of surgery proficiency. They suggested 30–50 cases of angiograms would be needed to pass the steepest stage of the learning curve (11). Zussman et al. (32) evaluated the safety and feasibility of TRA in 50 diagnostic cerebral arteriograms and found that neurointerventionalists tend to achieve higher success rates after performing 50 cases. Almallouhi et al. (33) found that the number of crossovers from TRA to TFA rapidly declined after 6 weeks of training. In the present study, procedures were performed by seasoned neurointerventionalists who had navigated beyond the inflection point of the learning curve. The procedure time was notably shorter for the TRA group than for the TFA group, with more experienced neurointerventionalists demonstrating reduced procedure times.

This study had several limitations. First, this was a retrospective study with a small sample size, and the selection of TRA or TFA was based on experience and preference, which may have generated selection biases. Second, this study lacks standard protocols for screening puncture site complications and may underestimate the incidence of radial occlusion, which requires Doppler ultrasound examination. Third, we were unable to evaluate the learning curve of neurointerventionalists participating in this study because of the heterogeneity of each center. Fourth, due to the retrospective nature of the study, we were unable to provide information about patient-reported outcomes or satisfaction scores, which might reflect the additional advantages of TRA. Finally, our study’s sample size was relatively small, particularly when assessing outcomes that were rare but clinically significant, which may have led to limited statistical power and a Type II error.

In conclusion, our study found that TRA was a safe and feasible approach for vertebral artery stenting of V1 segment stenosis. TRA had an advantage over TFA in terms of reduced procedure time and length of stay. Further large-scale and randomized controlled studies were warranted to verify the findings, given the retrospective and non-randomized nature of this study.