Is it time to take bare metal stents off the catheter laboratory shelf?

Implantation of metallic stents revolutionized percutaneous coronary intervention (PCI) in the early 1990s. Coronary stents restrain dissection flaps and create a regular, usually round vessel lumen which reduces the chances of acute vessel occlusion. Stents also optimize acute lumen gain, prevent early vessel recoil, and limit the constrictive effect of late adverse vessel remodelling. However, the inevitable late increase in neo-intima formation caused by the presence of the stent required the development of drug-eluting stents (DES)-combining polymer technology with anti-proliferative drugs to improve the medium and long-term predictability of stent implantation.¹

Drug-eluting stents implantation necessarily attenuates the vessels healing processes which results in delayed and sometimes incomplete endothelialization. Incomplete vascular healing appears to contribute to stent thrombosis (ST) which is catastrophic and can be fatal in up to 50% of cases.¹ At the European Cardiac Society in 2006, presentation of pooled data following implantation of first-generation DES and preliminary data from the Swedish SCAAR registry raised the spectre that ST appeared to be more prevalent after DES than after bare metal stent (BMS) implantation.^2,3 Consequently, rates of DES implantation plummeted throughout the world and a perception that BMS were safer than DES was initiated. Subsequent review of longer term SCARR data subsequently showed that these initial concerns were unfounded and that long-term outcomes after DES were actually superior to those with BMS.⁴ Concomitant sequential improvements in the efficacy and safety of DES have been evident⁵ (Table 1), but despite this, a perceived ‘niche’ role for BMS has persisted. This misconception about the lower rates of ST with BMS rather than DES has been potentiated by existing Guidelines.

Table 1.

Clinical outcomes in coronary stent trials with primary endpoint assessment at 9–12 months

	No. of contributing patients/trials	Clinical outcomes at 9–12 months median (IQR 25–75%) per 100 person-years
All-cause death (%)
BMS	7011/21	2.29 (1.64–3.79)
DES	63 535/75	1.67 (0.99–2.59)
EARLY DES	31 937/63	1.64 (0.94–2.76)
NEW DES	31 598/37	1.92 (1.05–2.54)
Cardiac death (%)
BMS	5891/15	1.57 (0.88–2.81)
DES	59 334/59	1.00 (0.53–1.69)
EARLY DES	29 149/48	0.98 (0.50–1.83)
NEW DES	30 185/32	1.00 (0.65–1.63)
Myocardial infarction (%)
BMS	6315/19	3.29 (1.97–4.31)
DES	62 347/71	2.88 (1.41–4.57)
EARLY DES	30 976/59	2.88 (1.39–4.59)
NEW DES	31 371/36	2.89 (1.45–4.21)
Target lesion revascularization (%)
BMS	5557/17	12.32 (7.44–13.79)
DES	57 595/67	4.00 (2.05–6.40)
EARLY DES	26 729/56	4.34 (2.40–7.11)
NEW DES	30 866/35	2.91 (1.67–5.94)
Definite stent thrombosis (%)
BMS	6399/19	1.08 (0.57–1.94)
DES	54 393/58	0.61 (0.37–0.99)
EARLY DES	24 221/46	0.74 (0.45–1.19)
NEW DES	30 172/31	0.47 (0.28–0.58)

Results of a systematic review⁵ regarding clinical outcomes at 9–12 months for BMS, early [the Cypher sirolimus-eluting stent, the Taxus paclitaxel-eluting stent (Taxus and Taxus Element), and the Endeavor zotarolimus-eluting stent] and new DES [the Xience, Promus and Promus Element everolimus-eluting stents, the Resolute zotarolimus-eluting stent, the BioMatrix and Nobori biolimus-eluting stents, and the Yukon Choice PC and Yukon Choice PF sirolimus-eluting stents]—median rates and IQR per 100 person-years.

A recent multicentre prospective registry documented the following indications for using BMS (744 consecutive PCIs): large vessel diameter, 241 (32.4%); ST-segment elevation myocardial infarction (STEMI), 132 (17.7%); reimbursement reasons, 70 (9,4%); advanced age, 92 (12.4%); planned non-cardiac surgery within the next year, 41 (5.5%); concomitant treatment with oral anticoagulants (OAC), 84 (11.3%); increased bleeding risk, cancer, or anaemia 71 (9.5%) and anticipated poor dual antiplatelet therapy (DAPT) compliance, 13 (1.7%)⁶ (Figure 1). In our opinion, objective review of the current literature suggests that all of these indications for BMS are mistaken and that each of these patients would probably have been better treated with DES.

Figure 1.

Clinical indications for potential bare metal stents use. BMS, bare metal stents; DES, drug-eluting stents; DAPT, dual antiplatelet therapy.

Large vessel diameter

The Basel Stent Kosten Effektivitäts Trial-Prospective Validation Examination (BASKET-PROVE) trial randomized DES and BMS in patients with large vessels (>3 mm). Patients with DES had a lower risk of cardiac death or MI (pooled DES vs. BMS: RR 0.60; 95% CI 0.39–0.93; P = 0.02) at 2 years of follow-up.⁷ In the BASKET-PROVE II trial, the second-generation biolimus-A9-eluting biodegradable-polymer (BP) DES (Nobori, Terumo) was compared with durable-polymer (DP) DES (Xience Prime, Abbott Vascular) and a thin-strut BMS coated with a biocompatible silicone-carbide layer (ProKinetik, Biotronik) in large vessels (≥3 mm). Biodegradable-polymer-DES (7.6%) was similar to DP-DES (6.8%) but superior to BMS (12.7%) in terms of cumulative incidence of the primary endpoint of cardiac death, MI, or clinically driven target-vessel revascularization (TVR) within 2 years.⁸ Hsieh et al. followed 1096 patients for 4.5 years who underwent either DES or BMS implantation in large coronaries divided into four subgroups by the reference vessel diameter (Q1: 3.0–3.25 mm, Q2: 3.26–3.50 mm, Q3: 3.51–3.75 mm, and Q4: 3.76–4.50 mm). The adjusted risk of major adverse cardiac events, defined as a composite of cardiac death, MI, TVR, and ST in DES recipients was reduced by 82% in Q1 (HR 0.18, 95% CI 0.09–0.38), 49% in Q2 (HR 0.51, 95% CI 0.26–0.98), and 67% in Q3 (HR 0.33, 95% CI 0.15–0.73).⁹

ST-segment elevation myocardial infarction

Two recent trials directly compared DES with BMS among STEMI patients undergoing primary PCI. In the everolimus-eluting stent (EES) vs. BMS in ST-segment elevation myocardial infarction (EXAMINATION) trial, 1498 STEMI patients were randomized to receive EES or BMS. The combined patient-oriented outcome of all-cause death, any MI, or any revascularization was assessed at 5 years, which occurred in 159 (21%) patients in the EES group vs. 192 (26%) in the BMS group (HR 0.80, 95% CI 0.65–0.98; P = 0.033). This difference was mainly driven by a reduced rate of all-cause mortality with DES (65 [9%] vs. 88 [12%]; 0.72, 0.52–0.10; P = 0.047).¹⁰

The Comparison of Biolimus Eluted From an Erodible Stent Coating With Bare Metal Stents in Acute ST-Elevation Myocardial Infarction (COMFORTABLE AMI) compared BMS and biolimus BP-DES. Patients treated with BP-DES had a lower risk of the composite primary endpoint of cardiac death, target-vessel MI, and TLR (4.3 vs. 8.7%; HR 0.49; 95% CI 0.30–0.80; P = 0.004) and a lower risk of target-vessel MI (0.5 vs. 2.7%; HR 0.20; 95% CI 0.06–0.69; P = 0.01).¹¹ A subsequent pooled analysis demonstrated a lower risk of ST and re-infarction with DES than with BMS.¹² In patients with STEMI, very favourable safety and efficacy profiles have also been demonstrated with CoCr-EES¹³ and DES should now be considered to be more effective and probably safer than BMS during primary PCI in STEMI.

Reimbursement

Schafer et al. evaluated clinical outcomes and costs of care over 3 years in 1147 undergoing BMS before the availability of DES and 1247 DES patients at Wake Forest University Baptist Medical Center from 2002 to 2005. Index stenting costs were $1846 higher per patient for DES vs. BMS. At 3 years, absolute TVR rates were 15.2 per 100 DES patients and 24.1 per 100 BMS patients, and as a result, cumulative TVR-related costs were $2065 less per patient for DES vs. BMS.¹⁴ Cost-effectiveness is evident especially among patients with moderate and high risk of TVR.¹⁵

Advanced age

In the Xience or Vision Stents for the Management of Angina in the Elderly XIMA Trial, 800 patients (83.5 ± 3.2 years of age) were randomized to BMS or DES. The primary endpoint (1-year composite of death, MI, cerebrovascular accident, TVR, or major haemorrhage) occurred in 18.7% of patients in the BMS group vs. 14.3% of patients in the DES group (P = 0.09). There was no difference in death, major haemorrhage, or cerebrovascular accident but MI (8.7 vs. 4.3%; P = 0.01) and TVR (7.0 vs. 2.0%; P = 0.001) occurred more often in patients treated with the BMS.¹⁶

Dual antiplatelet therapy duration/elective non-cardiac surgery

Current ESC guidelines recommend that DAPT should be administered for at least 6 months after implantation of new generation DES, and for up to 1 year in patients after ACS, irrespective of the revascularization strategy.¹⁷ Data from the DAPT trial indicate no differentiation in outcome after prolonged DAPT administration irrespective of stent type (BMS vs. DES).¹⁸ Premature discontinuation of DAPT is identified as a potent risk factor for ST.¹ For patients being considered for non-cardiac surgery this has been considered to be problematic when using DES.¹⁹ Using stents which recruit endothelial progenitor cells has been suggested as one approach. Notably, analyses of the risk of adverse events among patients with DAPT cessation and patients undergoing non-cardiac surgery indicates no differences between BMS and DES²⁰ and this was confirmed in a recent single large centre study where no benefit was seen after BMS compared with DES.²¹ Some observational data with zotarolimus stents (ZES) suggest that 1-month duration of DAPT can be sufficient,²² and a randomized study showed a similar outcome in patients treated with 3 and 12 months of DAPT after PCI.²³

Anticoagulation and high bleeding risk

This group of patients may encompass at least 15% of patients undergoing PCI. Current ESC guidelines recommend new generation DES rather than BMS in patients requiring oral anticoagulants at low bleeding risk (HAS-BLED ≤ 2). There is less certainty in the current recommendation for patients at high bleeding risk (HAS-BLED ≥ 3).²⁴ Recent guidance is available, in the Zotarolimus-eluting Endeavor Sprint Stent in Uncertain DES Candidates trial, 1606 patients at either high bleeding risk (52%), high thrombotic risk (17%), or low restenosis risk (31%) were randomized to ZES or BMS. Reductions in TVR and ST [2.0 vs. 4.1%; HR 0.48 (95% CI 0.27, 0.88), P = 0.019] were observed and the benefit of ZES over the BMS remained consistent across all pre-specified subgroups.²⁵ The LEADERS FREE trial randomized 2466 patients at high risk for bleeding who underwent PCI. A polymer-free, Biolimus-coated stent was superior to a BMS with respect to the primary safety and efficacy end points when used with an only 1 month DAPT.²⁶ Consequently, in this group of particularly high-risk patients, data strongly favours use of these particular DES.

Our opinion of these data is that there is a clear benefit from DES implantation compared with BMS and that this benefit is independent of the clinical indication. Any existing differential in price between these stents is marginal and consequently in 2016 implantation of BMS is no longer justified. We suggest that BMS should be honourably retired and added to the history books that document the evolution of interventional therapies.

Authors' contributions

G.K. and A.B. drafted the manuscript, and made critical revision of the manuscript for key intellectual content.

Funding

A.B. is partially funded by the NIHR Oxford Biomedical Research Centre.

Conflict of interest: A.B. has received an unrestricted research funding from Boston Scientific.