AIMS:

Intra-vascular ultrasound (IVUS) has been utilized as a adjunct to percutaneous coronary intervention (PCI) to guide stent implantation and pre-stent assessment of lesion length. The co-registration feature allows synchronization of angiographic and tomographic IVUS frames. Some IVUS models utilize an manual pullback by the interventionist. It allows us to select, deploy and expand stents optimally. We aim to evaluate the accuracy of manual pullback co-registration by comparing post-PCI stent lengths measured on IVUS to actual stent length.

METHODS AND RESULTS:

38 patients who underwent IVUS-guided PCI were analysed retrospectively in a single centre with five interventionists. We included stents deployed in any major epicardial vessel; overlapping stent were excluded (as actual combined stent length would no longer be accurate given the overlap). A pre-IVUS angiogram of the vessel of interest was first performed to obtain an angiographic map. After stenting, a Volcano Eagle Eye Platinum catheter (Philips, USA) was used to perform IVUS assessment with manual pullback and co-registration was generated with its accompanying software. The length of the stent was then measured from the first to the last frame with the any stent strut visible. This was then compared to the actual length of the stent used and their correlation was analyzed. 38 paired IVUS and angiographic runs were performed. Stent deployment was well distributed between different coronary arteries and multiple performing interventionists. The differences between the reference length from the manufacturer and the length measured via IVUS co-registration ranged from -4.2mm to 3.9mm. Assessment of correlation between actual and measured stent length using Pearson’s correlation coefficient was 0.9447.  

CONCLUSION:

IVUS with co-registration generated by manual pullback appears to be highly accurate and likely independent of vessel assessed or inter-user variability. The use of co-registration via coronary angiography in combination with IVUS allows us to predict stent length reliably and thus increases the ability to select and deploy stents optimally.