Drug eluting stent (DES) has dramatically reduced restenosis risks compared with bare metal stent (BMS) and conventional balloon angioplasty 123. Angiographic analysis found that the majority of DES restenosis were focal, localized, and bordered by segments with no angiographic evidence of neointima, while BMS restenosis tended to be diffuse or proliferative 45. Thereby, DES has revolutionized revascularization therapy and is rapidly becoming the standard for percutaneous coronary intervention (PCI).

Although the incidence of late stent thrombosis is very low, DES may increase the risk for late events, especially associated with discontinuation of dual anti-platelet therapy 67. Considering that the patients are difficult in anti-platelet compliance and more drug cost, we have to ask whether all lesions need DES and what specific lesion types are independent of DES therapy. Factors known to increase the risks of in-stent restenosis include smaller vessel diameter, prior restenosis, length of stented vessel, and diabetes mellitus 89. However, few of study reported that a simple de novo lesion, for example, lesion type A/B11011 in the middle and large vessel was treated with single BMS vs single DES at short- and long-term follow-up.

Thus we investigated the efficacy and safety of single BMS vs. single DES in nondiabetic patients with a simple de novo lesion in the middle and large vessel at 6-month, 1-year and 3-year follow-up in real world.

Baseline clinical characteristics were shown in table 1. Compared to DES group, the patients in BMS group had lower hypercholesteremia rate (22.0% vs 38.8%, P = 0.006), but age, gender, other risk factors for coronary artery disease and left ventricular function were similar in the two groups (all P > 0.05).

During procedure and in-hospital, BMS group had higher proportion of TIMI grade 0 than DES group (12.0% vs 1.2%, P < 0.05), but other variable including calcified lesion (%), lesion length (mm), stent diameter (mm), percentage of lumen stenosis (%), balloon predilatation (%), stent length (mm), post-dilatation (%), vessel dissection (%), postprocedural residual stenosis (%) and in-hospital outcomes did not significantly differ (all P > 0.05) in table 2. Despite that BMS had higher acute thrombosis rate than DES (3.3% vs 0%, P = 0.162), these patients recovered reperfusion after thrombolysis and intra-aortic balloon pump therapy, there were not in-hospital TLR and death in BMS group.

Repeat coronary angiography at 6-month follow-up showed similar acute and subacute thrombosis (%), late thrombosis (%), in-segment restenosis (%), TLR (%) and composite of cardiac death or Re-MI (%) in the two groups (all P > 0.05) in table 3.

The rates of lost to follow up at 3-year follow-up were 6.7% and 1.2% between BMS and DES group. The both groups had not significant differences in primary events including TLR (1.3% vs 1.2%, P = 1.00) and recurrent myocardial infarction (Re-MI) (0% vs 1.2%, P = 0.36) or cardiac death (0.7% vs 1.2%, P = 1.00) between 1- and 3-year, So were Re-MI (%), cardiac death (%), TLR (%) and MACE (%) at 1- and 3-year follow-up (all P > 0.05) in table 4.

The cumulative survival free of cardiac death in BMS group vs DES group was 100% vs 100% (Log rank P = 1.000) at 1-year and 98.67% vs 97.65% (Log rank P = 0.559) at 3-year follow-up (Fig. 1). Similarly, TLR-free cumulative survival between BMS group and DES group was 95.33% vs 95.29% (Log rank P = 0.978) at 1-year and 94.00% vs 94.12% (Log rank P = 0.984) at 3-year follow-up (Fig. 2). Noticeably, there was a trend towards a decrease of Re-MI-free cumulative survival in the DES group compared with the BMS group at 1-year (98.82% vs 100%, Log rank P = 0.183) and 3-year follow-up (97.65% vs 100%, Log rank P = 0.059) (Fig. 3).

This is the first study to investigate efficacy and safety of single BMS vs. single DES in nondiabetic patients with a simple de novo lesion in the middle and large vessel at 3-year follow-up in real world. The present study found that both DES group and BMS group had similar acute and subacute thrombosis (%), late thrombosis (%), in-segment restenosis (%), TLR (%), composite of cardiac death or Re-MI (%) at 6-month follow-up, so were Re-MI (%), cardiac death (%), TLR (%) and MACE (%) at 1- and 3-year follow-up in the two groups, furthermore, nonsignificant difference in the cardiac death-free and TLR-free cumulative survival rates except that there was a trend towards a decrease of Re-MI-free cumulative survival rate in DES group compared with BMS group at 1- and 3-year follow-up (all P > 0.05).

The previous study found that vessel diameter was an established predictor of angiographic outcome after catheter-based intervention, with a higher restenosis rate in smaller vessels 14. Thereby, at the time of these pilot studies, sirolimus-eluting stents were only available in a 3.0 mm or 3.5 mm diameter, limiting treatment to relatively large vessels, these sirolimus-eluting stents showed 0% restenosis at 4-month 15, 6-month 16, and 12-month 17. Later, a study demonstrated that the classic inverse relationship between vessel diameter and restenosis rate was seen in the BMS group but not in the sirolimus-eluting stent group 18, and vessel sizes of 2.5 – 3.5 mm were allowed in the subsequent randomized study with the sirolimus-coated Bx velocity balloon-expandable stent in the treatment of patients with de novo native coronary artery lesions (RAVEL) trial, yet lesions still had to be covered with one stent 19.

Currently, based on a lot of studies, people began to believe that the restenosis at the site of stent implantation seen in 15–60% of patients was dependent on various confounding factors, such as the presence or absence of diabetes mellitus, the size of the targeted coronary artery, the length of the coronary lesion, and the degree of vessel patency achieved by the intervention 202122232425. DES has been shown to reduce the risk of restenosis compared with BMS 1192526. Despite that treatment of specific lesions types, especially in stent restenosis and distal stenosis of left main coronary, as well as diabetic patients, remains suboptimal with DES, whereas considering that DES practice including complex interventions is safe and associated with significant reductions in clinical driven repeat revascularization rates 27. Moreover, DES also can effectively treat in-stent restenosis and saphenous vein graft restenosis 282930, thus it appears to be the advent of transition from BMS to DES in routine PCI practice.

However, we do not disregard an important problem of DES, that is, thrombosis. Especially subacute in-stent thrombosis could occur more frequently with DES than with BMS and a prolonged anti-platelet regimen is mandatory 31. In spite of the use of anti-platelet agents, stent thrombosis occurs in approximately 1% of patients, with an increased likelihood of occurrence in high-risk patients or complex lesion subset of patients 3233. According to the previous report, triple anti-platelet therapy (aspirin + clopidogrel + cilostazol) seemed to be more effective in preventing thrombotic complications after stenting than dual anti-platelet agent 34, but latterly a case report showed a patient with subacute stent thrombosis involving two different arteries simultaneously under the use of triple anti-platelet regimen 31. Therefore, the promises of this potential panacea – DES, have been recently attenuated by the specter of late and very late stent thrombosis because of anti-platelet discontinuation 35363738. However, the large-scale clinical trials and pool analysis demonstrated that the beneficial effect of DES on reducing the need for new revascularization compared with BMS extends to 4 years without evidence of a worse safety profile including thrombosis 39404142.

In our study, the specific lesion was choiced in the nondiabetic patients, and both BMS group and DES group had similar post-procedural outcomes including balloon predilatation (%), stent length (mm), stent diameter (mm), post-dilatation (%), vessel dissection (%) and postprocedural residual stenosis (%), finally we found that the both groups had similar acute and subacute thrombosis (%), late thrombosis (%), in-segment restenosis (%), TLR(%), composite of cardiac death or Re-MI (%) at 6-month follow-up and Re-MI (%), cardiac death (%), TLR (%) as well as MACE (%) at 1- and 3-year follow-up, so were the cardiac death-free and TLR-free cumulative survival rate, however, there was a trend towards a decrease of Re-MI-free cumulative survival rate in the DES group compared with the BMS group at 1- and 3-year follow-up (all P > 0.05), we presumed that the Re-MI might be associated with very late thrombosis. In view of less cost, short-term anti-platelet regimen, less thrombosis incidence, similar restenosis rate and TLR rate in BMS compared with DES, suggesting that BMS may has similar efficacy and superior safety compared with DES at 3-year follow up, thus the nondiabetic patients with a simple de novo lesion in the middle and large vessel seem to have other benefit from BMS instead of DES in real world.

The single BMS has similar efficacy and safety to single DES in non-diabetic patients with a simple de novo lesion in the middle and large vessel at short- and long-term follow-up.

JLC, SBQ, XWQ, MY, JC, YJW, HBL, JiQY, SJY, JJL, JD and RLG carried out the relevant PCI procedures, BX participated in the design of the study, collected and managed the relevant data, YJY conceived of the study, participated in its design and coordination, and revised this manuscript. All authors read and approved the final manuscript.