Efficacy and Safety of Distal Radial Artery Access versus Proximal Radial Artery Access for Cardiac Procedures: A Systematic Review and Meta-Analysis

Percutaneous coronary procedures using proximal radial artery (PRA) access have been found to diminish risk of complications compared to the trans-femoral access. Moreover, American Heart Association guidelines recommend it as the class I option [1]. As a consequence, the use of PRA has become more widespread for treating coronary conditions. Nevertheless, it is essential to recognize the severe consequences of PRA, such as vasoconstriction of the artery, the formation of fistula between the radial artery and the radial vein or its tributaries, radial artery occlusion (RAO), and pseudoaneurysm [2]. One of the most frequent complications of PRA is RAO, which can happen either early or late. The prevalence of RAO varies significantly across studies, with reported rates spanning from 5% to 30% [3]. In the radial fossa or on the dorsum of the hand, the distal radial artery (DRA) access has become an increasingly popular new approach to minimize the risk of RAO caused by the constant perfusion of radial artery through the anastomotic network [4]. Numerous trials have demonstrated that DRAs offer several advantages over PRAs, such as lower RAO rates, more rapid hemostasis, lesser fluoroscopy time which reduces exposure to radiation, and greater patient comfort [5‒14]. Despite these promising results, interventionalists may be reluctant to incorporate it into their daily routine due to its novel nature, ergonomic challenges, and less experience [15, 16]. The DRA approach might be more technically demanding due to the superficial palmar branch of the radial artery situated in the fovea radialis of the wrist: being smaller, more convoluted, and having more angles than the segment of the radial artery punctured as compared to the PRA [16, 17].

Previous meta-analyses comparing DRA with PRA for cardiac procedures yielded inconclusive results with limited outcomes [18‒24]. Updated results of two randomized controlled trials (RCTs) have recently become available [25, 26]. To address the discrepancies and incorporate all pertinent randomized data, we performed an updated meta-analysis. This comprehensive analysis seeks to highlight the variations in outcomes between DRAs and PRAs, thus offering a better understanding of their distinct benefits and hazards in cardiac procedures.

This systematic review and meta-analysis employed procedures detailed in the Cochrane Handbook for Systematic Reviews of Interventions, complied to the reporting guidelines established by the PRISMA [27]. The study protocol was preregistered in PROSPERO before conducting the review (CRD42024565730).

Medline, Excerpta Medica Database, Clinicaltrial.gov, and Cochrane Central Registry were searched from the outset till June 2024 for studies comparing DRA and PRA for cardiac procedure, including percutaneous coronary intervention (PCI) and percutaneous coronary angiography using a comprehensive search strategy. The following terms (“Coronary Angiography” OR “Percutaneous Coronary Intervention”) AND (“Radial Artery”) were combined as either Medical Subject Heading (MeSH) terms or keywords. Furthermore, we systematically examined the bibliographies of incorporated studies and similar systematic reviews to locate more pertinent studies.

The included studies had the following characteristics: (1) study design – RCTs, (2) population – adults (>18 years of age) undergoing PCI or percutaneous coronary angiography, (3) intervention – DRA, (4) comparator – PRA. Non-randomized clinical trials, cross-sectional studies, cohorts, case-control and case series, animal studies, and review articles were excluded.

The studies retrieved by the search strategy were imported on EndNote 20. The duplicates were eliminated. The two authors compared the titles, abstracts, and manuscripts of the selected studies against the eligibility criteria. In case of any discrepancy, it was resolved by a senior author through discussion and consensus between authors.

Data from the included articles were extracted using a reconstructed Microsoft Excel sheet. Dichotomous outcomes are reported as events and total, whereas continuous outcomes are measured with mean, standard deviation (SD), and the total number of patients extracted. Data on the name of the author, the year of publication, the country of the studies, the sample size of the research population, age, gender, hypertension, diabetes, smoking, prior PCI, and median follow-up were extracted.

The primary outcome reported was the RAO at 24 h. The secondary outcomes were puncture attempts, success rate of puncture, the success rate of procedure, bleeding, hematoma, crossover rate, time to hemostasis in minutes, fluoroscopy time in minutes, and radial artery spasm.

To evaluate the risk of bias, the version 2 of the Cochrane risk-of-bias tool for randomized trials (RoB-2) was used [28]. This tool assesses the bias for five domains: (1) the bias stemming from the randomization process itself, (2) the bias due to deviation from the intended intervention, (3) the bias originated from missing outcome data, (4) the bias that is caused by measurement of outcome, and (5) the bias because of the selection of reported results. This was applied to all the studies included, and their bias was rated as high, low, and some concerns. Two authors evaluated bias in studies independently. Differences in the evaluation were resolved via consultation with the third senior author.

The statistical analysis was executed using Review Manager (RevMan) version 5.4.1. The inverse variance method of analysis was utilized for the continuous variables, and mean ± SD was reported along with 95% confidence interval (CI). Studies reporting the median and the ranges of interquartile were converted to mean ± SD using a converter. The dichotomous outcomes were analyzed using the Mantel-Haenszel method and reported in the form of the risk ratio (RR) with 95% CI. The model of random effects was utilized. The pooled estimates were represented as a forest plot. The I² statistics were implemented to gauge the heterogeneity of included studies.

The literature search initially retrieved 7,348 articles. After removing 464 duplicates, 6,884 unique articles remained. The title and the abstract screening excluded 6,626 articles, leaving 258 studies for full-text screening against the eligibility criteria. Eventually, 21 RCTs were included in this meta-analysis [5‒14, 25, 26, 29‒37]. The selection of studies is summarized in Figure 1.

This meta-analysis enrolled 21 RCTs that have met the eligibility criteria. Among these, four studies were carried out in China, three in India, two each in the USA, Russia, Turkey, and Greece, and one each in Palestine, Mexico, Italy, Romania, Egypt, Europe, and Japan. The publication years ranged from 2018 to 2024. The total patients across these studies were 9,539 (4,761 DRA, 4,778 PRA). Table 1 provides the summary of baseline characteristics for included studies.

Twenty-one RCTs were included in the quality assessment, and fifteen were judged to have a low risk of bias. Five RCTs were considered to have some concerns due to the domains of the randomization, deviations from the intended interventions, and selection of reported outcomes. One RCT was evaluated having an increasingly high risk of bias. The randomization process domain contributed significantly to this high bias. The quality evaluation of included studies is given in online supplementary Figure 1 (for all online suppl. material, see https://doi.org/10.1159/000543817).

Fifteen studies were included for the analysis of RAO at 24 h, involving a total of 7,859 patients (3,909 DRA vs. 3,950 PRA). The analysis revealed DRA had a significantly lessened risk of RAO at 24 h compared to PRA (RR 0.30, 95% CI: 0.23–0.40, p < 0.00001, I² = 7%) (Fig. 2).

This outcome was reported by 15 studies. The total patients were 6,571 (3,272 DRA vs. 3,299 PRA). According to the analysis, the success rate of puncture was significantly lower in the DRA group compared to the PRA group (RR 0.96, 95% CI: 0.93–0.99, p = 0.01, I² = 88%) (online suppl. Fig. 2).

Eight studies reported a success rate of the procedure having 4,352 patients (2,170 DRA vs. 2,182 PRA). The analysis showed comparable results (RR 0.98, 95% CI: 0.95–1.00, p = 0.07, I² = 79%) between the two groups (online suppl. Fig. 3).

Eight studies were examined for the analysis of puncture attempts having 3,272 patients (1,626 DRA vs. 1,646 PRA). The analysis showed that there were significantly more puncture attempts with DRA as compared to PRA (mean difference [MD] 0.69, 95% CI: 0.37–1.00, p < 0.0001, I² = 94%) (online suppl. Fig. 4).

This outcome was reported by 13 studies. The total patients were 6,886 (3,436 DRA vs. 3,450 PRA). According to the analysis, PRA was significantly superior to DRA (RR 2.89, 95% CI: 2.02–4.15, p < 0.00001, I² = 40%) (online suppl. Fig. 5).

The analysis for the fluoroscopy time involved 10 studies with a total of 4,273 patients (2,113 DRA vs. 2,160 PRA). There was no significant variation between the two groups for this outcome (MD 0.01, 95% CI: −0.25 to 0.27, p = 0.95. I² = 52%) (online suppl. Fig. 6).

The outcome of bleeding was assessed by 5 studies having 4,061 patients (2,019 DRA vs. 2,042 PRA). According to the analysis, the DRA group was related to a significantly reduced risk of bleeding in contrast to the PRA group (RR 0.37, 95% CI: 0.14–0.98, p = 0.04, I² = 89%) (online suppl. Fig. 7).

Fifteen studies reported the outcome of hematoma. The total patients were 7,345 (3,658 DRA vs. 3,687 PRA). The DRA indicated a significantly decreased rate of hematoma development as compared to the PRA (RR 0.65, 95% CI: 0.44–0.97, p = 0.04, I² = 76%) (online suppl. Fig. 8).

The outcome of radial artery spasm was assessed by 15 studies having a total of 6,881 patients (3,426 DRA vs. 3,455 PRA). The two groups were comparable in terms of the spasm of radial artery (RR 0.79, 95% CI: 0.48–1.30, p = 0.35, I² = 77%) (online suppl. Fig. 9).

Sixteen studies were included in the analysis which assessed the outcome of time to hemostasis in the two groups. The total patients were 7,029 (3,501 DRA vs. 3,528 PRA). The analysis revealed that DRA had a significantly lower time to hemostasis as compared to PRA (MD −44.46, 95% CI: −50.64 to −38.29, p < 0.00001, I² = 100%) (online suppl. Fig. 10).

This meta-analysis, comprised of 21 RCTs having 9,539 participants, revealed reduced occurrences of RAO, bleeding, hematoma formation, and hemostasis time with the DRA approach. However, there were notably higher crossover rates and puncture attempts compared to PRA.

DRA is a new technique that presents technical difficulty owing to the intricate anatomy of the DRA, which has a smaller diameter and is more convoluted than the proximal section. The heightened occurrence of crossover rates noted in the DRA group within our study corresponds with recent studies showing rates reaching up to 30% in DRA group [18, 20‒22, 26, 38].

Our meta-analysis shows lesser RAO in DRA which aligns with results of previous studies; Barbarawi et al. [39] undertook a meta-analysis on eleven trials having 5,700 patients, which showed that PRA was linked with higher rates of RAO compared with DRA (RR 3.05, p < 0.01). Similarly, Rigatelli et al. [20] conducted a meta-analysis involving 7,073 patients, which shows a decreased risk of RAO when DRA is used (RR 0.46, p = 0.002). RAO was also significantly lower with DRA for the meta-analyses conducted by Isath et al. [40] (RR 0.36) and Sattar et al. [41] (OR 0.51, p = 0.02). Injury to endothelium and diminished flow of blood from the sheath and insertion of catheter play a valuable role in clot formation and increase the risk of RAO [42, 43]. Repeated catheterization of the radial artery can lead to hyperplasia and thickening of intima-media causing undesirable alterations of the vessel wall and further increasing susceptibility to obstruction [44‒47]. The radial artery stenosis manifests in 31% of individuals within 2 days following PRA procedures, with a subsequent occurrence of 28% observed at a later stage post-procedure [48].

According to our study findings, the success rate of puncture was higher in the PRA group. Similarly, a study conducted by Hammami et al. [49], including 177 subjects, shows a lower rate of cannulation failure in PRA (2.1% vs. 4.8%, p = 0.15) which demonstrates higher success rate of puncture in PRA. A notable portion of participants in the DRA required frequent puncture attempts, resulting in longer access times, which is in line with previous meta-analysis findings. Feghaly et al. [38] who included 28 studies having 9,151 patients also found that DRA access has extended accessibility time (MD: 0.31, p < 0.00001). Similarly, a meta-analysis done by Isath et al. [40], including 14,071 patients from 23 studies, shows increased risk of failure of vascular accessibility (RR = 2.38). These results indicate that there is a learning path associated with mastering this innovative procedure. However, our analysis shows that compared to PRA, DRA had a higher success rate of puncture, although more puncture attempts were needed compared to PRA. According to the study by Sharma et al. [13], the distal arterial method exhibits a learning curve of a minimum of 50 punctures, even for veteran professionals highly skilled at the conventional method of radial artery cannulation.

The use of ultrasonography allows for a live view of the needle path and may increase the success rate for DRA cannulation. Among the reports included in our study, only the CONDITION, PRESERVE RADIAL, and TENDERA trials used ultrasonography regularly, resulting in a decreased crossover rate of 4%, 3%, and 5% in the DRA group, respectively, without affecting the total duration of the process [25, 26, 31]. An RCT by Elmahdy et al. [50] had 300 patients in two groups: one had a planned access site based on vascular ultrasound results (group A), and the other had the access site decided by the operator (group B). Group A had significantly less arterial entrance time (1.25 ± 0.17 min) compared to group B (4.95 ± 0.87 min) with a p value of 0.02. Use of ultrasonography before cannulation may aid patient selection by assessing diameter and thickness of intima of the vessels, as well as any vascular complications such as signs of vessel injury, including intimal tears, medial dissections, clotting, occlusion, pseudoaneurysms, and arteriovenous fistula [51].

Our study revealed a notable decrease in the occurrence of RAO when using DRA, aligning with previous research findings [24, 31]. The puncture location for DRA is situated distally into the deep and the superficial palmar branches of the radial artery. The specific location guarantees the uninterrupted flow of blood within the artery, thus diminishing the likelihood of obstruction. The reduced incidence of obstruction is particularly advantageous for individuals who have undergone coronary artery bypass surgery necessitating arterial channels or those in need of arteriovenous fistulas for hemodialysis. By puncturing at this specific site, the preservation of palmar collaterals occurs, thereby lowering the chances of ischemic hand complications, even in scenarios of arterial occlusion post-procedure at the access point [51, 52].

Anatomical reasons can explain why the DRA group needed less time to establish hemostasis. The reduced caliber of the superficial palmar branch of the radial artery leads to a decrease in the blood volume, lower pressure of blood, and diminished tension in wall, causing quicker hemostasis [24, 40, 48]. Moreover, the superficial nature of bones (the scaphoid and trapezium) at this location enables effective compression and swifter hemostasis [37]. This abbreviated hemostasis period is likely to be a contributing factor in the lower occurrence of RAO seen in the DRA group. The latest meta-analyses of RCTs have also observed reduced time for hemostasis in the DRA group [18, 38, 40, 48]. However, they did not demonstrate any difference in the incidence of hematoma development [11, 19, 38, 40]. Our findings are different from the previous meta-analysis regarding hematoma formation as our meta-analysis shows less hematoma formation with DRA, probably due to increasing expertise in the distal approach with time [53]. Previous investigations found a higher incidence of hematoma, despite faster hemostasis, which could be attributed to factors such as a lack of operational skill and variability in hemostasis strategies between areas. This is obvious in the repeated puncture attempts, absence of appropriate compressing equipment, and differences observed in sensitivity analysis about hematoma development [24].

Our study observed no difference in the fluoroscopy time of contrast utilized, which is consistent with prior research [20, 22, 38, 40]. There was no mismatch between the two groups for radial artery spasm; these findings are corroborated by a meta-analysis conducted by Mhanna et al. [18], having 1,634 patients and 12 RCTs, which reported no significant difference.

Cardiac catheterization using DRA poses unique challenges. DRA’s performance with thicker guide wire is limited, often necessitating an alternative access strategy [54]. There is a need for equipment innovation specifically customized for the DRA approach to effectively address these challenges. Specialized equipment for the DRA method, such as slimmer catheters, compressing devices optimized for the anatomical snuff box, and the use of ultrasonography before and during the procedure, represents key advances [24]. Furthermore, meticulous technique can significantly enhance success rates. A major hindrance to a successful cardiac catheterization using DRA is operator experience, as unfamiliarity with the procedure can significantly accentuate the difficulty in achieving puncture due to anatomical reasons [55]. Roh et al. [56] evaluated 1,000 patients who had undergone DRA to identify risk factors for failure. Although the overall success rate was 95.2%, risk of failure was greater in females and in patients with a systolic blood pressure less than 120 mm Hg. Success rate for operators increased with procedures undertaken, stabilizing at 94% after 200 procedures had been carried out. Thus, DRA has a steep learning curve and initiation of large-scale training programs is crucial in establishing this route as a viable option for cardiac catheterization.

There are several constraints that need to be considered when reporting these results. First, limited research exists on hand sensory functioning, osteonecrosis, and appropriate DRA techniques, indicating a need for further investigation. Second, the percentage of PCI procedures examined in the studies was minimal; thus, caution is warranted in applying the results to PCI patients. Third, the RCTs predominantly utilized 5F sheaths, which may limit the applicability of the results to PCI patients with varying sheath sizes. Future studies need to evaluate the efficacy of sheath sizes other than 5F. Fourth, the meta-analysis included only RCTs, potentially excluding valuable data from other study designs. Fifth, results for hematoma, spasm, and RAO were not adjusted for puncture attempts, which may be a significant confounder for these outcomes. Lastly, operator experience and use of ultrasound can both significantly influence outcomes following DRA, but could not be adjusted for in our analysis. Future studies should assess the impact of ultrasound guidance to determine the most optimal strategy to perform a DRA.

DRA offers benefits in reducing RAO and hemostasis time but requires more puncture attempts. Further studies are needed to confirm its long-term efficacy.

This study does not require ethical approval.

The authors have no conflicts of interest to declare.

This study was not supported by any sponsor or funder.

Mohammad Ebad Ur Rehman: conceptualization, methodology, and writing – original draft, review, and editing. Hafsa Arshad Azam Raja: conceptualization, formal analysis, and writing – original draft, review, and editing. Muhammad Osama and Muhammad Hassan Waseem: methodology, formal analysis, and writing – original draft. Aisha Kakakhail: conceptualization, methodology, and writing – original draft. Muhammad Mukhlis, Muhammad Abdullah Ali, Zain Ul Abideen, Muhammad Shoaib, and Ammara Tahir: data curation, methodology, and writing – original draft. Zahir ud Din: data curation, methodology, and writing – review and editing. Muhammad Zohaib Ul Hassan and Usman Mazhar: data curation, formal analysis, and writing – original draft. Syed Tehseen Haider, Sajeel Saeed, and Abdulqadir J. Nashwan: formal analysis and writing – review and editing.

All data generated or analyzed during this study are included in this article and its supplementary material files. Further inquiries can be directed to the corresponding author.