Introduction Rotational atherectomy (RA) during primary PCI (PPCI) for STEMI is relatively contraindicated because of the perceived increased risk of no-reflow. However, RA PPCI may sometimes be required to restore flow in heavily calcified coronary arteries. Previously only very limited observational data has described the use of RA in PPCI.

Aim We report the clinical and procedural characteristics, and in hospital outcomes, of 21 patients who underwent RA PPCI at our centre between 2006 and 2016,

Methods A retrospective review of the PCI database and medical records.

Results 21 patients (age 78(10) years (mean (SD)), 12 men) underwent RA during PPCI (0.4% of all PPCI). 3 patients had cardiogenic shock at presentation and 2 had out of hospital cardiac arrest. Hypertension (n=19), smoking history (n=18), hypercholesterolemia (n=16), diabetes (n=6) and chronic kidney disease (n=6 with eGFR<60) were frequently present. Aspirin was given to 20/21 patients and clopidogrel, ticagrelor and prasugrel to 10, 9 and 3 patients respectively. Heparin was used in all patients, glycoprotein (GP) IIBIIIA inhibitor in 10 and bivalirudin in 2.

Radial access was used in 14 and femoral in 7. Initial TIMI flow grade was 0, 1, 2 and 3 in 11, 2, 1 and 7 patients respectively. The target vessel was the RCA in 14, Cx in 4, LMS in 2 and LAD in 1. All were severely calcified with visible thrombus in 13/21. The lesion length was 36(19) mm (range 12–72 mm). The vessel diameter was 2.5–2.99 mm in 5, 3–3.49 mm in 10 and 3.5–4.0 mm in 6. RA was used because of anticipated difficulty with conventional PCI in 7/21 and uncrossable/unexpandable lesions in 14/21. The number of burrs used was 1.33 (0.48) and the final burr was 1.25, 1.5, 1.75 and 2 mm in 4, 11, 4 and 1 patient respectively. The burr-to-artery ratio was <0.4 in 5 and 0.4–0.6 in 16. The final procedure was DES in 17, BMS in 3 and POBA followed by CABG in 1. A temporary pacemaker was used in 3 cases and intra-aortic balloon pump in 3 cases. There were no perforations, no tamponade and no vascular complications.

The median(range) pain-to-PCI time was 180 min (114–544); door-to-PCI time 49 min (21–186, 14 patients <60 min) and procedure time was 119 min (66-175). Procedural difficulty caused door-to-PCI>60 min in 5/7 patients while medical/logistic reasons contributed in the other 2. 1 patient underwent CT scanning prior to PCI; 1 underwent an urgent MDT discussion before PCI. Final flow was TIMI 3 in 19/21 and TIMI 2 in 2/21 cases. The procedure was complicated by 1 event of distal embolization. There was 1 event of peri-procedural cerebral infarction. 20/21 patients survived to hospital discharge.

Conclusions Our study suggests that RA PPCI can be performed safely in a small group of predominantly elderly, complex patients in whom conventional techniques are inadequate or unsuccessful. It is associated with a relatively long procedure time but we did not experience a marked increase in no-reflow.