Aims: The aim of this study was to evaluate treatment of coronary in-stent restenosis (ISR). Methods and results: We investigated interventions for ISR and the occurrence of re-restenosis in the Swedish Angiography and Angioplasty Registry (SCAAR). From January 1st 2005 to March 3rd 2012, 212,166 coronary segments were treated and 7,806 restenoses analysed. During seven years of follow-up 1,079 re-restenoses were registered on clinically driven angiography. For BMS-ISR the adjusted risk of re-restenosis was significantly lower with DES (adjusted hazard ratio [BR] 0.71,95% confidence interval [CI]: 0.61-0.82), tended to be lower with DEB (HR 0.84, 95% CI: 0.62-1.16), but higher with BMS (BR 1.24, 95% CI: 1.0-1.55) as compared to balloon angioplasty. For DES-ISR a new DES was associated with a significantly lower adjusted risk of re-restenosis (HR. 0.80, 95% CI: 0.66-0.99), and a similar but non-significant reduction with DEB (BR 0.86, 95% CI: 0.57-1.30) and BMS (BR 0.81, 95% CI: 0.53-1.24) compared to balloon angioplasty. For DES-ISR a DES with a different drug was not more effective than a DES with the same drug. Conclusions: ISR in BMS should be treated with DES or DEB while the optimal treatment of ISR in DES remains to be proven.