Improvement of angiographic and clinical outcomes of percutaneous coronary intervention for chronic total occlusion after implementation of a dedicated team: a single-centre experience

In June 2017, the CTO team was implemented, made up of three interventional cardiologists who were dedicated to perform CTO-PCI. Before this date, all staff members performed this procedure. The dedicated CTO interventional cardiologists received full training and carried out multiple hands-on CTO-PCI procedures in patients under direct supervision of highly experienced CTO expert operators. Furthermore, a complete CTO state-of-the-art equipment set was collected and stored in a mobile storage cabinet. This was present during every procedure in the operating room, thereby enabling a fast exchange of dedicated materials, which is mandatory for a successful hybrid approach.

All catheterisation laboratory nurses were trained in CTO recanalisation techniques, use of the materials and complication management. Prior to the procedure, each CTO case was discussed by at least two interventional cardiologists of the CTO team to determine the Multicentre CTO Registry of Japan (J-CTO) score and the appropriate approach according to the hybrid algorithm. All procedures for cases with higher complexity (J-CTO score > 3) or for previously failed cases were scheduled with two interventional cardiologists (‘double scrub’).

Prior to June 2017, patients were indicated for CTO-PCI if they presented with angina and had proven ischaemia. In June 2017, we changed these criteria to the presence of angina and viability (instead of ischaemia) as shown by normal wall motion on any imaging modality [12]. In case of hypokinetic or akinetic wall motion in the myocardial area of a CTO, additional viability testing with either single-photon emission computed tomography, magnetic resonance imaging or positron emission tomography was required. Patients’ quality of life (QoL) was routinely assessed using the Short Form Health Survey (SF-36) from 2018 onward.

All patients who underwent CTO-PCI between January 2013 and June 2020 were included and divided into two groups: the pre–CTO team group and the post–CTO team group. Patients could be included in both Pre -and Post-CTO team group, if the previous target vessel CTO-PCI failed and was repeated at least 1 year after the first attempt. All patients in whom with CTO-PCI was performed by a non–CTO team member after June 2017, were excluded. Ethical approval was obtained by the local ethics committee.

The primary endpoint was defined as the angiographic success rate of CTO-PCI, either at the first attempt or at subsequent attempts. The first secondary endpoint was the angiographic success rate of CTO-PCI, stratified by complexity according to the J‑CTO score. Additional secondary endpoints comprised the following clinical outcomes: in-hospital and procedural complications (consisting of myocardial infarction (MI), all-cause mortality, urgent pacemaker implantation, urgent cardiac surgery, stroke and in-stent thrombosis), and MI, target vessel revascularisation (TVR), all-cause mortality, QoL and major adverse cardiac events (MACE) at 30-day and 1‑year follow-up.

Baseline and procedural data were collected from electronic patient files. Angiographic endpoints were assessed and adjudicated by two independent experienced reviewers (LP, ML) and a third reviewer (KT) in case of disagreement. All angiographic and clinical endpoint definitions are listed in Table S1 (see Electronic Supplementary Material).

Event-free survival curves for adverse cardiac events were estimated using the Kaplan-Meier method and compared with the log-rank test. A multivariate Cox proportional hazard analysis was performed. The Canadian Cardiovascular Society grading of angina pectoris (CCS) classification progress was compared using the Wilcoxon signed-rank test. For both the group with angiographic successful CTO-PCI and the group with angiographic unsuccessful CTO-PCI, baseline scores on 8 QoL domains were compared with the follow-up scores on these domains. For inclusion, the SF-36 questionnaire had to be completed both at baseline and at 1‑year follow-up and at least 50% of the items of each domain had to be answered.

A two-tailed probability value of < 0.05 was considered statistically significant. All statistical analyses were performed with IBM SPSS Statistics version 24.0 (IBM Corp., Armonk, NY, USA).

During the study period, a total of 662 patients underwent CTO-PCI, of whom 376 (58.3%) with 389 lesions were included in the pre–CTO team group and 269 (41.7%) patients with 279 lesions in the post–CTO team group; 17 patients were excluded because a non–CTO team interventional cardiologist had performed the procedure after June 2017. Seven patients were included in both groups, and one patient was included twice in the post-CTO team group (Fig. 1). Baseline characteristics were comparable, except for hypertension and age (Tab. 1; and Tables S2 and S3 in Electronic Supplementary Material). The mean J‑CTO score was comparable between both groups (2.00 vs 1.97; p = 0.64).

The angiographic success rate of the first attempt was significantly higher in the post–CTO team group than in the pre–CTO team group (81.4% vs 62.7%; p < 0.001). The angiographic success rate after the final attempt increased by 5.3% in the post–CTO team group and by 11.1% in the pre–CTO team group (86.7% vs 73.8%; p = 0.001). The success rate was significantly higher for CTO lesions classified as ‘difficult’ or ‘very difficult’ based on the J‑CTO score after the first and final attempt in the post–CTO team group than in the pre–CTO team group (Fig. 2).

The total in-hospital complication rate was 7.3% with a higher major in-hospital complication rate in the post-CTO team group than in the pre-CTO team group (4.7% vs 1.7%; p = 0.02). However, the incidence of missing data on in-hospital MI was higher in the pre–CTO team group than in the post–CTO team group (45.6% vs 9.6%). Procedural or in-hospital MI occurred more often in the post–CTO team group than in the pre–CTO team group (3.2% vs 0.2%; p < 0.01) (Tab. 2). The incidence of MACE was significantly lower in the post–CTO team group during the first year after the intervention (6.4% vs 16.0%; p < 0.01) (Tab. 2 and Fig. 3). An additional Cox proportional hazard analysis was performed for potential confounders age, hypertension and assignment to post-CTO team group. The model showed that hypertension was not an independent predictor of MACE but age (hazard ratio (HR): 1.03; 95% confidence interval (CI): 1.01–1.06) and assignment to the post–CTO team group (HR: 0.37; 95% CI: 0.20–0.66). The angina pectoris classification (CCS) was significantly improved after 30 days of follow-up (p < 0.001) and was sustained after 1 year (p < 0.001) in both the pre–CTO team and post–CTO team groups compared with the pre-procedural CCS (see Fig. S1 in Electronic Supplementary Material).

In the post–CTO team group, a total of 115 serial baseline and 1‑year follow-up completed SF-36 forms were collected (response rate: 42.6%). In the group with angiographic successful CTO-PCI (n = 96), mean scores on all 8 QoL domains were significantly higher at 1‑year follow-up compared with the baseline scores (see Fig. S2 in Electronic Supplementary Material). In the group with angiographic unsuccessful CTO-PCI (n = 19), the mean score for physical functioning was significantly higher at 1‑year follow-up.

This study was limited by its observational nature with the possibility of introducing potential confounders. Differences were observed in baseline characteristics such as age and hypertension between the groups. An additional analysis showed independent association of age and assignment to the post-CTO team to MACE. This analysis strengthened our overall findings on clinical outcomes by showing a lower MACE risk despite an incremental risk of events related to higher mean age in the post–CTO team group.

Furthermore, there were more missing data on clinical follow-up of MI (9.6% vs 45.6%) in the pre–CTO team group than in the post–CTO team group, which might have underestimated the event rate in the first group.

Another limitation of our study is the lower number of respondents included in the QoL analyses. As QoL questionnaires have been introduced for all PCI patients in 2018, they were only available for the post–CTO team group. Consequently, it was not possible to compare the pre–CTO team group with the post–CTO team group with regard to changes in QoL.