Aims: Collateral flow may influence long-term results after percutaneous coronary intervention (PCI) because of haemodynamic forces compete with the antegrade flow through the dilated lesion. The aim of the study was to assess the influence of recruitable collateral blood flow on restenosis in patients undergoing PCI with bare metal stents and using optimal antithrombotic treatment. Methods and results: In 95 patients, 95 de novo lesions were treated with PCI and a bare metal stent. Fractional flow reserve (FFR) at maximum hyperaemia induced by intravenous adenosine was determined. The pressure-derived collateral flow index (CFI) was determined as (Pw-P cvp)/(Pa-Pcvp), where Pw represents coronary wedge pressure, Pcvp central venous pressure, and P a mean aortic blood pressure. Both were measured during transient coronary occlusion by a balloon inflation of 30 s. Pre-interventional FFR (0.65 ± 0.20) correlated inversely with the CFI (0.18 ± 0.11), r =- 0.356, P < 0.001. After 9 months, binary angiographic restenosis (≥50% diameter stenosis) was seen in 29.1%. Compared to patients with poorly developed collaterals (CFI < 0.25), patients with well-developed collaterals (CFI ≥ 0.25) had a lower pre-interventional FFR (0.50 ± 0.14 vs. 0.72 ± 0.18, P < 0.001), a higher CFI (0.33 ± 0.08 vs. 0.13 ± 0.07, P < 0.001), and a higher binary restenosis rate (54.2% vs. 19.4, P = 0.003). CFI*100 was an independent predictor of restenosis after 9 months (odds ratio 1.07, 95% CI 1.02-1.12, P = 0.016). Conclusion: Recruitable collateral blood flow measured during balloon inflation predicts angiographic instent restenosis in PCI patients treated with bare metal stents.