Ultimaster TANSEI Stent in Complex Coronary Lesions: The EPIC08 TANSEI COMPLEX Study

Coronary artery disease (CAD) is one of the major causes of morbidity and mortality worldwide.¹ Percutaneous coronary intervention (PCI) has become the preferred treatment option for most patients with CAD.² However, complex coronary lesions, including long, calcified, bifurcation lesions and small vessels, pose a challenge for interventional cardiologists. The development of new technologies and devices, such as drug-eluting stents (DES), has improved the safety and efficacy of PCI for complex lesions. The Ultimaster TANSEI DES (Terumo Corporation) is designed to improve the safety and efficacy of PCI for complex coronary lesions.³ This cobalt-chromium stent features a thin-strut and unique tri-layered hybrid design, combining a biocompatible and bioabsorbable polymer with a drug coating to provide better drug elution while minimising the risk of stent thrombosis and restenosis. Its uniform scaffolding and an open-cell, 2-link design, which allows easy side-branch access, are essential for effective bifurcation stenting.⁴ These characteristics position the Ultimaster TANSEI stent as a promising option for patients with complex coronary lesions undergoing percutaneous revascularisation.

However, the evidence supporting the safety and efficacy of the Ultimaster TANSEI stent in complex lesions is still limited.^5,6 This study aims to provide real-world data on the performance of the Ultimaster TANSEI stent in an all-comer patient population with complex anatomical features across multiple Spanish centres.

Methods

Study Population

The EPIC 08 TANSEI COMPLEX study was designed as a prospective, single-arm, multicentre, observational registry for the evaluation of the safety and effectiveness of the Ultimaster TANSEI stent in patients with complex coronary lesions. All patients undergoing PCI with Ultimaster TANSEI DES were consecutively included between August 2020 and November 2022 across 17 Spanish centres if they met the following criteria for complex lesions: left main lesions, bifurcations, small vessels (<2.5 mm in visual diameter estimation) and long lesions (>35 mm in length). The exclusion criteria were as follows: patients in cardiogenic shock, those with a life expectancy <1 year, a contraindication to dual antiplatelet therapy (DAPT) or lesions caused by stent restenosis.

Patient demographic and clinical characteristics, as well as lesion characteristics and procedural details, were collected. Clinical follow-up was conducted through scheduled medical visits or telephone calls at 1, 6 and 12 months, with an active assessment of clinical outcomes.

The study design received prior approval from all participating investigators. The study protocol and informed consent forms were reviewed and approved by a designated reference ethics committee and accepted by the local institutional review board of each participating centre in accordance with national regulations and in compliance with the standards of the Declaration of Helsinki. Written informed consent was obtained from all enrolled patients in the study.

Study Device and Procedure

The Ultimaster TANSEI is a cobalt-chromium L605 DES with an ultra-thin strut design (80 μm) and a tri-layered hybrid structure combining a biocompatible and bioabsorbable polymer with a drug coating. It has a thin bioabsorbable polymer layer of poly(lactic-co-glycolic acid) and polycaprolactone, eluting sirolimus 3.9 μg/mm² over approximately 3 months. Its open-cell, 2-link design facilitates side-branch access. The available lengths are 9, 12, 15, 18, 24 mm, 28, 33 and 38 mm, with diameters of 2.25, 2.5, 2.75, 3.0, 3.5 and 4.0 mm.⁴ The device is CE marked.

All interventions were performed in accordance with current clinical practice guidelines for the management of PCI, with procedural strategy decisions left to the operator’s discretion, including the need for pre- or post-dilation, intravascular imaging-guided procedures or other relevant considerations.^2,7 Data on intravascular imaging use were collected only for the left main subgroup.

Preoperatively, a 300 mg loading dose of aspirin plus a second antiplatelet agent (clopidogrel, ticagrelor or prasugrel according to the clinical settings and operator’s preference) were administered to all the consecutive patients included. The duration of DAPT and the management of potential complications were also entrusted to the operator’s clinical judgement, based on current clinical practice guidelines.

Endpoints and Definitions

The primary endpoint was the 1-year device-oriented composite endpoint (DoCE), which included cardiovascular death, target-vessel MI (TV-MI), target vessel revascularisation (TVR), and definite or probable stent thrombosis, using the definitions of the Academic Research Consortium-2 (ARC-2) consensus.⁸

The secondary endpoints included the patient-oriented composite endpoint (PoCE), which comprised all-cause death, any MI and any revascularisation. Additionally, individual events were registered, including all-cause death, cardiovascular death, TV-MI, TVR and stent thrombosis (as per ARC-2 definitions). Other analysed events included major bleeding (Bleeding Academic Research Consortium [BARC] 2–5), defined according to the BARC criteria, as well as stroke, procedural success (defined as final Thrombolysis in Myocardial Infarction 3 flow and residual stenosis <20%) and procedural complications (coronary dissection, perforation, no reflow and distal embolism).⁹ Furthermore, the proportion of patients receiving DAPT at 6 and 12 months were evaluated, stratified by bleeding risk according to the PRECISE-DAPT score. MI was adjudicated using the fourth universal definition of MI.¹⁰ Cardiac death was defined as any death resulting from MI, sudden cardiac death, heart failure mortality or stroke. The classification of the lesions was established according to the American Heart Association/American College of Cardiology criteria (A, B1, B2, C).¹¹

Statistical Analysis

Since this cohort study lacked an intervention and control group, a descriptive analysis was performed. Measures of central tendency and variability were assessed: categorical variables are presented as counts and percentages (%), while continuous variables are reported as either median with interquartile range or mean ± SD, depending on the normality of the distribution. A bar chart was used to represent both short-term complications and follow-up events. First-year incidence rates (IRs) were computed as events per 100 person-years with exact Poisson 95% CIs. For subgroup comparisons we used a ‘subgroup versus others’ approach, reporting RRs with Katz 95% CIs and two-sided Fisher’s exact p-values. Event-free survival curves for DoCE and PoCE were estimated by Kaplan–Meier. A two-tailed p-value of <0.05 was considered statistically significant for all analyses. Statistical analyses were performed using SPSS 27.0 (IBM).

Results

Clinical and Procedural Features

During the study period, a total of 501 patients were enrolled and 591 complex coronary lesions were treated. The cohort was predominantly male with a mean age of 66.7 ± 11.3 years. Cardiovascular risk factors were common, with hypertension and/or dyslipidaemia in nearly two-thirds of patients, diabetes and or active smoking in about one-third; one-fifth had received prior PCI. The index presentation was predominantly acute (ST-elevation MI 40.6%, non-ST-elevation MI 27.8%, unstable angina 14.2%). Baseline characteristics and clinical data are presented in Table 1 and angiographic features are shown in Supplementary Table 1. Patients with complex coronary lesions included 55/501 (11.0%) with left main disease, 220/501 (43.9%) with bifurcation lesions, 204/501 (40.7%) with small-vessel disease and 172/501 (34.3%) with long lesions. Up to 23% of cases exhibited more than one complex feature.

A total of 591 complex lesions were treated (mean 1.18 ± 0.77 complex lesions per patient), with overlap of ≥2 complexity categories in up to 29.4% of lesions. Supplementary Table 2 shows distribution details of overlap among groups and Table 2 shows procedural features. The mean number of angiographically significant lesions per patient was 1.65 ± 0.72, while the mean number of treated vessels per patient was 1.44 ± 0.63. Angiographic analysis revealed that the majority of lesions were located in the left anterior descending artery (52.1%). 30.0% of the lesions were classified as type B2 and 31.6% as type C. Severe calcification was observed in 11% of the lesions, while chronic total occlusions (CTO) were present and treated in 16.6% (Supplementary Table 1 ). Radial access was the most common approach, performed in 87.6% of cases. The treated lesions included a single complex lesion in 85.2% of cases, two lesions in 12.8% and three or more lesions in 2% of the cases. Intravascular ultrasound (IVUS) assessment was performed in 15 of 52 patients (28.8%) in the left main subgroup. Owing to their particular characteristics, the angiographic features of bifurcation lesions are reported in Supplementary Table 3, highlighting that the left anterior descending was the most frequent target vessel (nearly two-thirds) and that around 50% were true bifurcations. In addition, a provisional stenting strategy was used in 81.3% and a two-stent technique in 18.6%.

Outcomes

One-year follow-up was successfully completed in 99.0% of patients. Five individuals were lost to follow-up (two withdrawals and three lost to contact), while 24 patients died during the follow-up period. Figure 1 displays Kaplan–Meier curves for PoCE, DoCE and overall mortality in the full cohort. The IRs and RRs for each endpoint are reported in Table 3 for the entire cohort and for each subgroup. DoCE reached an IR of 3.23 per 100 person-years (95% CI [1.84–5.24]), PoCE 5.04 per 100 person-years (95% CI [3.26–7.44]) and overall mortality 2.82 per 100 person-years (95% CI [1.54–4.74]). Across outcomes, the long-lesion (>35 mm) group consistently exhibited higher 1-year event rates than the other groups: DoCE RR 4.22 (95% CI [1.49–11.95]; p=0.006); PoCE RR 4.93 (95% CI [2.10–11.57]; p<0.001); overall mortality RR 4.79 (95% CI [1.53–15.06]; p=0.007); and TVR RR 17.60 (95% CI [2.22–139.51]; p<0.001), as shown in Table 3 and Figure 2.

Complications at hospital discharge were low, as shown in Supplementary Figure 1, with no differences across subgroups. Eighteen (3.6%) early complications were observed, including five (1%) periprocedural deaths (three of which were of cardiac origin), one (0.2%) of MI and two (0.4%) of urgent cardiac surgery. No cases of acute stent thrombosis were detected.

The use of DAPT was 95.6% at discharge, declining to 84.4% at 6 months and 61.3% at 12 months. Supplementary Table 4 summarises DAPT use across subgroups. For each subgroup, comparisons were made against the pooled remainder of the cohort. Differences were observed only in patients with LM disease, who had higher 12-month DAPT use (80.0 versus 59.1%; p=0.004), while the long-lesion subgroup had lower DAPT use at discharge (92.9 versus 96.9%; p=0.039). Among 464 patients with available PRECISE-DAPT scores (mean 29.9 ± 11.78), 58.4% were classified as high risk (≥25) and 41.6% as low risk (<25). Exposure to DAPT did not differ significantly between strata at discharge, 6 or 12 months (Supplementary Table 5). Moreover, no differences in bleeding events were observed after stratification by PRECISE-DAPT risk categories (Supplementary Table 6).

Chronic Total Occlusion Post Hoc Analysis

A CTO subanalysis included 89 patients with 97 lesions, in which 138 stents were implanted (mean 1.42 per lesion). Lesion locations were left anterior descending 37.1%, right coronary artery 36.1% and left circumflex 26.8%. Major procedural complications occurred in two of 89 (2.2%); there were no intraprocedural deaths, MIs, urgent cardiac surgery or acute stent thrombosis. Most lesions received a single stent (70.1%) and two stents were used in 20.6% of cases. Multivessel disease was present in 40.2% of patients, while treatment was limited to a single vessel in 81.4%. Stent implantation success was 97.9% for the first stent and 100% for additional stents. At 1 year, DoCE was 4.49 per 100 person-years (95% CI [1.23–11.51]) and PoCE 8.99 per 100 person-years (95% CI [3.88–17.71]). All-cause death had an incidence rate of 4.49 per 100 person-years (95% CI [1.23–11.51]), with no cardiovascular deaths. Stent thrombosis occurred at 3.37 per 100 person-years (95% CI [0.70–9.85]) and no bleeding events were recorded. Compared with non-CTO patients, there was a nonsignificant trend toward higher event rates in the CTO group, with PoCE: RR 2.15 (95% CI [0.96–4.83]; p=0.102) and stent thrombosis: RR 4.57 (95% CI [0.94–22.29]; p=0.074).

Discussion

This prospective and multicentre study provides real-world data on the performance of the Ultimaster TANSEI stent in a patient population with a wide range of complex coronary lesions. The main findings of this study are threefold. First, there was a high procedural success rate, close to 99%, in patients with complex coronary lesions. Second, there were relatively low incidence rates in the first year for both the primary endpoint (DoCE) at 3.23 per 100 person-years and the secondary endpoint (PoCE) at 5.04 per 100 person-years, suggesting a favourable safety profile in this challenging patient population. Finally, consistent outcomes across the predefined complex subgroups were observed, with no significant differences in DoCE or PoCE according to lesion type, except for patients with long lesions (>35 mm), who showed higher event rates.

In the CENTURY II study, which demonstrated the non-inferiority of the TANSEI stent compared with the everolimus-eluting stent with a permanent polymer (Xience stent, Abbott Vascular), a procedural success rate of 95.6% was observed, along with a combined rate of cardiac death and MI of 2.9%.⁶ These results are consistent with those observed in our study. Although approximately 80% of the lesions in the CENTURY II trial were classified as B2 and C, the overall lesion complexity in that study was considerably lower than in ours: in CENTURY II, only 14% of lesions involved bifurcations, 1.27% affected the left main artery, the average lesion length was 16.92 ± 9.73 mm and 5% were classified as total occlusions.⁶ These differences underscore the greater anatomical complexity of the lesions treated in our cohort, further reinforcing the favourable outcomes achieved in a more challenging clinical scenario. In line with these findings, the e-Ultimaster registry also underscored the impact of procedural complexity on clinical outcomes. Defined as multivessel PCI, ≥3 stents, ≥3 lesions, bifurcation with ≥2 stents, total stent length >60 mm or CTO, complex PCI was shown to be linked with higher rates of adverse events. Notably, most individual complexity features were associated with an increased risk of composite complications and definite/probable stent thrombosis, with the exception of CTO.⁵

The EBC Left Main trial compared single versus double stent strategies in bifurcated left main lesions, reporting a target lesion revascularisation (TLR) rate of 8% for the provisional strategy and 14% for the two-stent strategy (p=0.02).¹² In our study, this subgroup achieved a 97% procedural success rate, with no cases of MI or the need for new revascularisation procedures and a cardiovascular mortality IR of 1.8 per 100 person-years. Although these findings suggest a potentially favourable scenario for the TANSEI DES, dedicated studies are needed to confirm its role in this subgroup. The EXCEL trial, in contrast, reported a revascularisation rate of 12.6% at 3 years with everolimus-eluting stents.¹³ However, the 1-year follow-up in our study limits the ability to compare outcomes with EXCEL and other left main trials. Another point of interest is that IVUS guidance during PCI for complex lesions is associated with consistently lower rates of adverse cardiac events, mortality and repeat revascularisation compared with angiography alone, with the greatest benefits seen in challenging anatomies.^14,15 These data, together with recent guideline endorsements, support broader adoption in complex interventions.¹⁶ Nevertheless, in our cohort, IVUS was used in less than one-third of left main interventions, which may have influenced the observed outcomes, even though this subgroup showed a favourable event profile and remained in line with real-world registry rates.^17,18

Another complex lesion profile of interest relates to long lesions. Although definitions vary across studies, lesions >35 mm are often categorised as very long. In our cohort, the long-lesion subgroup exhibited higher 1-year event rates than the remainder, with more than a fourfold relative risk for DoCE, PoCE and all-cause mortality. Notably, TVR showed an incidence rate of 5.16 per 100 person-years and a much greater risk compared with other lesion groups supporting a potential link between lesion length and restenosis. This association has also been observed in other studies, such as the GRAN-DES registry, where lesions ≥40 mm were linked to a nearly twofold increase in TLR compared with shorter lesions.¹⁹ Notably, the outcomes observed in our cohort are comparable to those reported with stents specifically designed for long lesions, such as those used in the BILLAR study.²⁰ In fact, our observed TLR rate appears similar to the 5.3–6.5% range reported in other dedicated long-lesion trials, where the use of long or overlapping stents was associated with a higher risk of restenosis and thrombosis.¹⁹ These differences in the long lesion subgroup may reflect a composite mechanism that could include greater total stented length with overlap zones that might predispose to restenosis and thrombosis, a higher prevalence of smaller reference vessel diameters that may suggest a diffuse disease phenotype, and an increased likelihood of underexpansion or malapposition when intracoronary imaging is not used routinely. Taken together, these observations may support the hypothesis that intracoronary imaging-guided PCI could mitigate risk by optimising stent expansion and apposition, with minimal stent area targets around 4.5 to 5.0 mm² where appropriate, which might translate into fewer downstream events and may justify further evaluation to determine whether imaging guidance provides a tangible benefit in this subgroup.²¹

Similarly, previous studies on revascularisation in small vessels have shown an increased risk of restenosis and thrombosis with DES and TVR rates notably exceeding those observed in our study.^22,23 For this reason, drug-coated balloon therapy has emerged as a viable alternative in this scenario, reinforcing its potential role in small vessel interventions, with some studies reporting non-inferiority compared with stents. However, the results obtained in our cohort suggest favourable performance in this lesion profile, further supporting the effectiveness of the TANSEI stent in this context.^23,24

In our cohort, 43.9% of the treated lesions were bifurcations. The TVR IR among bifurcated lesions was 3.14 per 100 person-years. Although not negligible, this figure remains lower than those reported in some studies of non-left main bifurcations, where TVR rates have reached 6.8% with provisional stenting and up to 9.3% with the culotte double-stent technique.²⁵ These findings underscore the intrinsically higher risk profile of bifurcation lesions and highlight the importance of adopting individualised, anatomy-driven revascularisation strategies to optimise clinical outcomes. Nonetheless, caution is needed when interpreting and comparing PCI outcomes for bifurcation lesions across studies. One plausible explanation for the favourable results is the lesion profile and technique selection. Only about half of the bifurcation lesions were true bifurcations, defined by side-branch involvement (Medina classification 1,1,1; 1,0,1; 0,1,1) and the double-stent technique was used in fewer than one in five cases. This may have influenced outcomes, as non-true bifurcations generally yield better long-term results with lower rates of adverse events and mortality.²⁶ Factors such as lesion complexity, bifurcation angle, side-branch diameter, operator experience and technique choice can also shape procedural and long-term outcomes.²⁷ Even when accounting for these variables, bifurcation results in our cohort appear favourable.

Our CTO subgroup showed low in-hospital complication rates and high device success, consistent with contemporary CTO-PCI benchmarks reporting technical success around 80–90% with 1–3% major complications in expert centres.²⁸ At 1 year, TVR among evaluable patients was very close to the 4.8% TLR reported in the multicentre Japanese CTO-PCI Expert Registry. Overall event rates in that registry (1-year major adverse cardiac and cerebrovascular events 14.3%) provide a reasonable upper bound against which our results appear modest, acknowledging differences in definitions and case mix.²⁹ Overall, these findings are broadly consistent with contemporary CTO reports. However, the stent thrombosis rate in our series was slightly higher than previously reported. This is not unexpected in the context of high-risk presentations and complex anatomy, but it warrants cautious interpretation given the subgroup sample size.³⁰

In addition, although this study deliberately evaluated a complex PCI cohort, calcium burden was not an inclusion criterion. Plaque-modification techniques were omitted in up to one-fifth of lesions, consistent with lesion biology: 16.9% were thrombotic and more than half lacked angiographic calcification. Likewise, direct stenting occurred predominantly in acute coronary syndrome (ACS), where softer/thrombotic plaque often obviates predilation and device escalation. Overall, 8.4% of cases required advanced plaque modification (cutting balloon, atherectomy, laser or intravascular lithotripsy), broadly in line with the modest calcification burden observed and comparable to rates reported in large contemporary cohorts.³¹

Notably, more than half of patients were at high bleeding risk by PRECISE-DAPT, yet DAPT exposure remained similar across strata at discharge, 6 and 12 months, with only modest de-escalation by 6 months and DAPT maintained in more than half at 1 year. A plausible explanation is the high proportion presenting with ACS and the underlying anatomical complexity, both of which increase thrombotic risk. There were no clinically meaningful differences in DAPT use by risk subgroup, except that left main disease showed higher 12-month DAPT use than the remainder of the cohort (80.0 versus 59.1%; p=0.004). In addition, a recent meta-analysis indicates that PRECISE-DAPT ≥25 predicts not only greater bleeding but also higher ischaemic risk and mortality (composite ischaemic OR 2.16; major bleeding OR 3.62) and the optimal PRECISE-guided strategy remains undefined, particularly in anatomically complex patients.³² Despite this, bleeding rates were low and comparable between risk groups, with no difference in bleeding-free survival. However, the study was not powered to detect differences in bleeding outcomes.

Limitations

This study has several limitations that must be acknowledged. Its observational design introduces potential biases, limiting the ability to establish causal relationships. The absence of a control group further restricts direct comparisons with other stent types. Moreover, the 1-year follow-up period may have been insufficient to capture late adverse events, and certain complications could have been underreported. Since the study was conducted exclusively in Spanish centres, the generalisability of the findings to other healthcare systems may be limited. Additionally, procedural strategies were left to the operator’s discretion, introducing variability in techniques that could have influenced outcomes. The absence of intracoronary imaging data for the remaining lesion subgroups limits our ability to evaluate imaging-guided effects beyond the left main cohort. In the bifurcation subgroup, a substantial proportion of cases lacked side-branch involvement, and the specific double-stent technique employed was not reported, introducing potential confounding in the interpretation of these results. Finally, the relatively small sample size in some subgroups, combined with the low event rates observed, may have further limited the statistical power to identify subtle yet clinically relevant differences between groups.

Conclusion

The Ultimaster TANSEI stent was associated with a high procedural success rate and favourable clinical outcomes in a contemporary real-world setting, suggesting a potentially positive safety and efficacy profile in patients with complex coronary lesions. However, patients with long lesions (>35 mm) experienced significantly higher events and TVR indicating that lesion length remains a key driver of risk. These findings point to a possible benefit of using this stent in this challenging patient population.

Click here to view Supplementary Material.