Introduction

The recent European Society of Cardiology guidelines on myocardial revascularization recommend transradial access (TRA) as the standard approach for any percutaneous coronary intervention (PCI), irrespective of clinical presentation, unless there are overriding procedural considerations (recommendation: Class I, and level A)1. For diagnostic coronary angiography and percutaneous coronary intervention for coronary artery disease and acute coronary syndrome, TRA reduce short-term net adverse clinical events, cardiac death, all-cause mortality, bleeding, and access site complications, when compared to transfemoral access (TFA)2,3. TRA also provides many benefits including lower access-site complications, increased patient comfort, early ambulation, and shorter hospital stay3,4,5.

However, potential disadvantages of TRA include smaller vessel size, vessel spasm, and more techniques for guiding catheter placement2,6,7. The main reason for requiring a large guiding system is to be able to use an anchor balloon to help deliver antegrade dissection re-entry equipment or intravascular ultrasound guide puncture of the proximal cap with a microcatheter in situ, or for the need for a debulking device (large-burr rotablator)8,9. TFA provides strong backup support, enables the use of multiple equipment combinations, and allows unrestricted use of the trapping technique10,11. With the development of Glidesheath and improvement of wires for chronic total occlusion (CTO), the interventionist could try larger sheaths via TRA in recent decades12. Therefore, TRA has already become a popular method for CTO PCI, regardless of whether the antegrade or retrograde method was used.

Due to significant advances in specious materials and techniques along with increased operator experience and hybrid strategies, dramatic increment in success rate of CTO PCI in expert hands and low complication rates were reported13,14,15. Most studies comparing between TRA and TFA for CTO PCI were observational studies and did not involve recent improvements16,17. Therefore, due to the recent improvements in method and devices for CTO PCI, we focused on recent studies18,19,20,21,22,23 and compared the efficacy and safety between TRA and TFA for CTO PCI.

Methods

Search strategies, trial selection, quality assessment, review process, and data extraction

Figure 1 presents the literature search and screening protocol. A systematic literature searches for published articles between January 1, 2015, and December 31, 2020 in the PubMed, Embase, ProQuest, ScienceDirect, Cochrane Library, ClinicalKey, Web of Science, and ClinicalTrials.gov databases were separately performed by two cardiologists. The searched key terms “chronic total occlusion”, “percutaneous coronary intervention”, “transradial access”, and “transfemoral access” were used. We did not set language restrictions to increase the number of eligible articles. Disagreements were resolved by a third reviewer. Only randomized controlled trials and cohort studies that compared the clinical outcomes of the comparison between TRA and TFA for CTO PCI were included in the present meta-analysis. The inclusion criteria were human studies with a parallel design, with comparison of efficacy and safety between patients with TRA or TFA for CTO PCI. The exclusion criteria included conference abstracts, case reports or series, animal studies, and review articles.

Figure 1
figure 1

Flowchart of the selection strategy and inclusion and exclusion criteria for this meta-analysis. CTO chronic total occlusion.

Full size image

Definitions

Technical success of six studies was defined as achievement of final flow of target vessel of TIMI grade 3 with at least < 50% residual stenosis of the target CTO lesion at procedure end and was listed in Table 1. Procedural success of two included studies22,23 was the same as the definition of clinical trial design principles for CTO therapies24 and was defined as technical success plus the absence of an in-hospital major adverse cardiovascular event (MACE) including death, myocardial infarction, or clinically driven target vessel revascularization.

Table 1 Characteristics of the 6 included studies.
Full size table

Statistical analysis

All analyses were performed by using Comprehensive Meta-Analysis software, version 3 (Biostat, Englewood, NJ, USA). The frequency of each evaluated outcome was extracted from each study and were showed as cumulative rates. A random effects model was applied to pool individual odds ratios (ORs). The chi‐square test was used to evaluate heterogeneity across trials, (p ≤ 0.1 was considered significant). I2 statistics (> 50% was considered high heterogeneity) was employed to examine each outcome. Funnel plots and Egger’s test were used to inspect potential publication bias (p ≤ 0.1 was considered significant). p values < 0.05 was defined statistical significance.

Results

Characteristics of included studies

The study selection process is displayed in Fig. 1 and six studies met the inclusion criteria. A total of 11,736 participants (mean age of 64.2 ± 10.6 years; 81.7% male) were included. The study design, definition of TRA and TFA, and participants’ characteristics were described in Table 1. The study period, approach method for CTO PCI, and the definition of technical/procedural success were shown in Table 1.

Patient demographics

Table 2 describes the basic demographics, comorbidities, mean Japan-chronic total occlusion (J-CTO) score, average sheath size, mean procedural time, mean fluoroscopy time, and mean contrast volume of the study patients. The TFA group was older (TRA vs. TFA group, 63.8 ± 10.8 years vs. 64.5 ± 10.5 years, p = 0.001). The TFA group had a higher prevalence of diabetes mellitus, hypertension, dyslipidemia, heart failure, prior myocardial infarction, and prior coronary artery bypass grafting than the TRA group. The TFA group had a higher prevalence of right coronary artery involvement than the TRA group (47.4% vs. 54.8%, p < 0.001). Higher mean J-CTO scores were noted in the TFA group (2.0 ± 1.2 vs. 2.4 ± 1.3, p < 0.001).

Table 2 Patients’ demographics and CTO target vessel.
Full size table

The TFA group had a larger sheath size (6.6 ± 0.5 Fr vs. 7.3 ± 0.7 Fr, p < 0.001) than the TRA group. The TRA group had less procedural time (87.8 ± 36.9 min vs. 106.8 ± 47.5 min, p < 0.001), less fluoroscopy time (34.2 ± 22.1 min vs. 40.5 ± 21.9 min, p < 0.001), and less contrast volume (244.2 ± 128.2 ml vs. 272.4 ± 101.2 ml, p < 0.001) than the TFA group.

Pooled odds ratio of technical/procedural success rate and retrograde success rate of CTO PCI between the TRA and TFA groups

The overall odds ratio (OR) of the technical/procedural success rate of CTO PCI in the TRA group versus the TFA group was 0.846 (95% confidence interval (CI), 0.749–0.956; Fig. 2), with non-significant heterogeneity (Cochran Q, 5.764; df, 5; I2, 13.251%; p = 0.330) and non-significant publication bias according to Egger regression (t, 0.171; df, 4; p = 0.873) on inspection of the funnel plot (Supplemental Fig. 1).

Figure 2
figure 2

Forest plots of the overall odds ratio (OR) of procedural success rate of chronic total occlusion (CTO) percutaneous coronary intervention (PCI) between the transradial access (TRA) and transfemoral access (TFA) groups from 6 studies. CI confidence interval, TRA transradial access, TFA transfemoral access.

Full size image

According to 2 studies, the OR of the retrograde success rate showed that TRA versus TFA for CTO PCI was 0.965 (95% CI 0.382–2.435; Fig. 3), with high heterogeneity (Cochran Q, 7.794; df, 1; I2, 87.169%; p = 0.005).

Figure 3
figure 3

Forest plots of the OR of retrograde success rate of CTO PCI between the TRA and TFA groups from 2 studies.

Full size image

Pooled odds ratios of vascular complication after CTO PCI

The OR of vascular complication of CTO PCI in the TRA group versus the TFA group was 0.323 (95% CI 0.203–0.515; Fig. 4), with non-significant heterogeneity (Cochran Q, 2.997; df, 5; I2, 0%; p = 0.700) and non-significant publication bias according to Egger regression (t, 0.607; df, 4; p = 0.577) on inspection of the funnel plot (Supplemental Fig. 2).

Figure 4
figure 4

Forest plots of the OR of vascular complication of CTO PCI between the TRA and TFA groups from 6 studies.

Full size image

Pooled odds ratios of in-hospital mortality and major adverse cardiovascular event rate after CTO PCI between TRA and TFA groups

According to three studies, the OR of in-hospital mortality rate in the TRA group versus the TFA group after CTO PCI was0.527 (95% CI, 0.187–1.489; Fig. 5), with non-significant heterogeneity (Cochran Q, 0.814; df, 2; I2, 0%; p = 0.666) and non-significant publication bias according to Egger regression (t, 0.277; df, 1; p = 0.878) on inspection on the funnel plot (Supplemental Fig. 3).

Figure 5
figure 5

Forest plots of the OR of in-hospital mortality rate of CTO PCI between the TRA and TFA groups from 3 studies.

Full size image

According to six studies, the OR of MACE rate of the TRA group versus the TFA group after CTO PCI was 0.729 (95% CI, 0.504–1.054; Fig. 6), with non-significant heterogeneity (Cochran Q, 4.229; df, 4; I2, 5.408%; p = 0.376) and non-significant publication bias according to Egger regression (t, 1.969; df, 3; p = 0.144) on inspection of the funnel plot (Supplemental Fig. 4).

Figure 6
figure 6

Forest plots of the OR of major adverse cardiovascular event rate of CTO PCI between the TRA and TFA groups from 5 studies.

Full size image

Discussion

PCI for CTO lesions in the presence of viable myocardium and improvement of life quality is well accepted by most cardiovascular interventionists25. An increasing number of newly developed devices and techniques for CTO intervention and the success rate has increased gradually26. The overall procedure success is highly dependent on successful antegrade or retrograde wiring across the occlusion, as well as the support of guiding catheters, guidewires, and other devices. Although the TFA for CTO is still the first choice for most interventional cardiologists, our previous studies have proven the benefits of the TRA in CTO PCI27. In the Japan CTO registry, the use of TRA CTO PCI significantly increased over time gradually in recent years23. Bedsides, the TFA may be avoided in special situations including abdominal aortic atherosclerotic disease, severe aortoiliac disease, and previous iliofemoral bypass graft placement28.

The important limitation of radial artery is relatively small size. The radial artery is smaller than the femoral artery and may not be suitable for larger sheaths and guiding catheters. On routine diagnostic coronary angiography and PCI for acute coronary syndrome (ACS), TRA is a valuable alternative to TFA associated with a reduction in vascular complications29,30. However, routine diagnostic coronary angiography and ACS PCI is not like CTO PCI. Also, complex techniques including the anchor balloon technique, antegrade dissection re-entry technique (ADR), or intravascular ultrasound (IVUS) guide puncture, may require larger bore catheters. After the development of Glidesheath (Terumo, Japan), interventionists could try larger artery sheaths via radial arteries or even distal radial arteries for complex PCI12. A high success and low complication rate of the hybrid approach to CTO crossing was reported13. The introduction of using enabling strategies including antegrade or retrograde wire escalation, and dissection reentry techniques also brought an increasing success rate of CTO PCI14. Therefore, we need to evaluate the efficacy of TRA CTO PCI in recent decades.

This meta-analysis showed a better procedural success rate in TRA CTO PCI than in TFA CTO PCI. However, more comorbidities and higher mean J-CTO scores were noted in the TFA group, which also influenced the procedural success rate. The retrograde success rate was similar between the TRA and TFA groups. Therefore, TRA provides a better procedural success rate and does not improve the retrograde success rate. In addition, less procedural time (87.8 ± 36.9 min vs. 106.8 ± 47.5 min, p < 0.001), less fluoroscopy time (34.2 ± 22.1 min vs. 40.5 ± 21.9 min, p < 0.001), and less contrast volume (244.2 ± 128.2 ml vs. 272.4 ± 101.2 ml, p < 0.001) were seen in the TRA group. In the enrolled studies, two studies reported the incidence of contrast-induced nephropathy, which did not differ between the TRA and TFA groups21,22. Shorter procedural time, fluoroscopy time, and less contrast exposure may contribute to fewer procedural complications.

In our study, fewer vascular complications (OR: 0.323; 95% CI 0.203–0.515) were observed in TRA CTO PCI. This may bring about a net clinical benefit by decreasing ischemic events due to cessation of antiplatelet and antithrombotic agents if bleeding, and the adverse effects of blood transfusion31. Because TRA reduces vascular access-site complications in patients undergoing PCI for simple or complex procedures, mortality and ischemic events may also be reduced by TRA when compared with TFA. However, a more prevalence of multiple comorbidities and higher J-CTO score also effect the results of vascular complications, associated MACE and mortality. The complexity of baseline characteristics may let interventionists to choose TFA approach for CTO PCI. Therefore, the results of observational studies existed selection bias. Kinnaird et al. reported higher short-term and one-year mortality rates in patients with access-site complications20. Tajti et al. reported no significant difference in in-hospital MACE between the bilateral TRA and bilateral TFA groups23. In our study, for the incidence of in-hospital mortality (OR: 0.527; 95% CI 0.187–1.489) and MACE (OR: 0.729; 95% CI 0.504–1.054), TRA CTO PCI showed a non-significant trend when compared with TFA CTO PCI. In addition, a recent randomized study reported TRA is associated with a significant reduction in clinically relevant access-site bleeding or vascular complications, without affecting procedural success when compared with TFA in patients undergoing PCI of complex coronary lesions (≥ 50% CTO PCI) with large-bore access32. In our study, higher procedural success rates were noted in the TRA CTO PCI may be contributed by better baseline characteristics and lower J-CTO score, but fewer vascular complications may be associated with better vascular condition and the benefit of TRA.

Limitations

This study has several limitations. First, all the studies were observational cohort studies and not all studies provide detailed information about the procedure about CTO PCI. Because of observational studies, selection bias could not be excluded totally, and the difference of baseline characteristics could influence the results. Second, we could not decrease the bias from the difference of baseline characteristics. Third, the operator’s experience and CTO PCI volume are not available in all enrolled studies, but all data was from experienced centers or registry of muti-center. Only one study mentioned the operators’ experience as a minimum of 25 hybrid procedures and certification for the controlled antegrade dissection re-entry technique19. Third, the definition of technical/procedural success was different in the included studies. However, a total of 11,736 participants were enrolled from 6 studies. The present study still provides some important findings on the outcomes of TRA CTO PCI after the improvement of the CTO technique and devices in recent decades.

Conclusion

Due to the improvement of the CTO technique, fewer vascular complications and higher procedural success rates were noted in the TRA CTO PCI population. However, similar retrograde success rates and clinical outcomes were noted between the groups. More complex comorbidities and higher J-CTO scores still influenced outcomes.