Clinical impact of in-stent late loss after drug-eluting coronary stent implantation.

AIMS Controversy exists about the clinical significance of in-stent late loss (ISLL) after drug-eluting stent (DES) implantation. We sought to clarify whether ISLL after DES implantation is related to a potential clinical impact. METHODS AND RESULTS We included in a meta-regression analysis 21 trials (8641 patients) that randomly compared DES with bare-metal stents (BMS). We evaluated the relationship between angiographic behaviour of DES and the clinical impact of using DES instead of BMS in each trial using meta-regression techniques, weighting by the number of patients included in each trial. Mean ISLL in patients allocated to DES and DeltaISLL (difference in ISLL in patients allocated to BMS and DES) were used as angiographic parameters of efficacy of DES. The number of patients needed to be treated (NNT) to prevent one target lesion revascularization (TLR) was used to quantify the clinical impact of using DES instead of BMS. There was a significant relationship between mean ISLL in patients allocated to DES and the clinical benefit of using DES instead of BMS, as measured with the NNT for TLR: NNT for TLR = 6.2 + 18.4 [ISLL-DES] (R = 0.62; P = 0.007). Therefore, a 0.1 mm increase in mean ISLL-DES was associated with a 1.8 increase in NNT for TLR. There was also a significant association between the degree of inhibition of neointimal hyperplasia of DES in comparison with BMS with the NNT for TLR: NNT for TLR = 17.1-11.8 [DeltaISLL] (R = 0.61; P = 0.008). Therefore, a 0.1 mm reduction in ISLL by using DES instead of BMS was associated with a 1.2 decrease in mean NNT for TLR. CONCLUSION There is a strong and significant association between the degree of inhibition of neointimal formation with the use of DES and the clinical impact of using DES instead of BMS.