Background: Severe coronary calcification and chronic total occlusion (CTO) are among the most difficult scenarios in percutaneous coronary interventions (PCI) and the combination of both is even more challenging. There is little evidence about the long-term outcomes of rotational atherectomy (RA) for the treatment of CTO and data comparing the long-term outcome of RA in CTO and RA in non-CTO lesions are scarce.
The aim of this study is to determine in-hospital and long-term outcomes after RA for CTO PCI compared to RA for non-CTO PCI.

Methods and results: This analysis is from an observational registry, which enrolled all consecutive patients undergoing RA at a single centre (Heart Centre, Segeberger Kliniken GmbH) during the period from 2002 to 2019. Major adverse cardiac events (MACE) was defined as a composite endpoint of all-cause death, myocardial infarction and target-vessel revascularization (TVR). 
During the study period, RA was performed in 813 patients to treat 870 lesions (RA-CTO, n=81; RA non-CTO, n=789). The mean age of the CTO population was 72.4 and 73.1 years in the non-CTO group, p= 0.53. The CTO population had a higher prevalence of previous PCI (50% vs 38%, p= 0.04), other baseline characteristics were not significantly different between both groups. Resistant CTO lesions represented the main indication for RA 61% (33.3% balloon non crossable and 27.7% balloon non dilatable lesions) while in 39% RA was performed electively due to angiographic evidence of heavy calcification. The mean J-CTO score in RA-CTO lesions was 2.39 ± 0.9 and 82% of the lesions were ≥ 2. In RA-CTO group the number of stents and total stent length were significantly higher (2.7 ± 1.3 vs 1.8 ± 0.9 and 67.7 ± 32.6 mm vs 37.3 ± 22.1 mm; p< 0.001 respectively). The most frequently used burr size in RA-CTO was 1.25 mm (63%) whereas the 1.5 mm burr was the most frequently used in non-CTO (46%).Consequently, the mean burr size was smaller in RA-CTO PCI (1.37 ± 0.19 mm vs 1.5 ± 0.2 mm; p< 0.001). The total procedural and radiation time were significantly longer in the RA-CTO group (median 124 min (IQR 98-167) vs. 79 min (IQR 58-113) and 49 min (IQR 35-73) vs 23.5 min (IQR 15-35), p< 0.001 respectively). Angiographic success (defined as <30% residual stenosis with TIMI 3 flow at the end of the procedure) was achieved in 92.7% in RA-CTO vs 96.6% in non-CTO (p= 0.06). In-hospital MACE was not significantly different between the two groups 3% in RA-CTO vs 2% in non-CTO (p=0.36). After a median follow up period of 720 days (IQR 300-1260), we observed a trend towards higher long-term MACE in the RA-CTO population, which was not statistically significant (28% vs 31.5% in RA-CTO vs non-CTO, p=0.08). However, cardiac death (11% vs 7%; p =0.03) and target lesion revascularisation (TLR) (12.5% vs 10%; p=0.045) were higher in the RA-CTO and the high TLR was mainly derived from more in stent restenosis.

Conclusion: RA in CTO was achieved with high procedural success rate and similar in-hospital outcome compared to RA for non-CTO. Nevertheless, cardiac death and TLR were higher after RA for CTO at long-term follow up.