Fusion of optical coherence tomography and coronary angiography — In vivo assessment of shear stress in plaque rupture

Abstract

Advancements in cardiovascular imaging [e.g. computed tomo-graphic coronary angiography (CTCA)] and new developments inimage processing [e.g. fusion of intravascular ultrasound (IVUS) withangiographic data, fusion of IVUS with CTCA] permitted complete andcomprehensive three dimensional (3D) reconstruction of coronaryarteries and allowed us to study in-vivo the role of blood flowdynamics in the atherosclerotic process. Over the last years severalstudies used 3D reconstruction techniques to demonstrate that localhemodynamics are involved in the atherosclerotic evolution andaffect the composition of the plaque [1]. However, indigenouslimitations of the implemented imaging techniques (mainly theincreasednoise and the reduced resolution)have not allowed reliableidentification of the vulnerable/ruptured plaques and thus the effectofthebloodflowonplaquedestabilizationandruptureisstillunclear.In this report we fused optical coherence tomography (OCT) withcoronary angiographic data, in order to identify the location of plaquerupture and so to examine the association between flow hemody-namics and plaque rupture.A 55 year old patient with a medical history of hypertension whosustained an acute inferior myocardial infarction and treatedsuccessfully with thrombolysis was subsequently transferred to ourhospital, within 24 h from the event, to undergo coronary angiogra-phy.Thisdemonstratedanormalleftcoronarysystemandahazynon-flow limiting lesion in the distal right coronary artery (RCA) (Fig. 1A).To study in more detail the morphology of the culprit lesion, OCTexamination was performed with a M3CV OCT system (LightLabImaging Inc., Westford, MA, USA). Before OCT interrogation, thesystemwasconnectedwithaviewermixerthatallowedsimultaneousvisualizationof theelectrocardiogramandthe OCTsequence.TheOCTcatheter was advanced in the RCA distal to the culprit lesion, and cineangiographic images from two different views were obtained duringdiluted contrast agent injection. Then, the OCT catheter waswithdrawn, under saline purge through the guiding catheter, withthe use of an automated pull-back device (pull-back speed: 3 mm/s;15 frames/s). Three consecutive pull-backs were performed in orderto examine a 45-mm segment. From the obtained sequences, theframes that corresponded to the end-diastolic phase of the cardiaccircle (peak of R wave on electrocardiogram) were selected and anexpert observer identified the ruptured plaque (Fig. 1B), andextracted manually the luminal borders. At the site of rupture, twotracings of the luminal borders were performed: one along theresidualintimalflaprepresentingthetruelumenwithplaquerupture,and the other by extrapolation along the plaque cavity representingthe lumen before plaque rupture [2]. Fusion of the OCT with theangiographic data was performed to reconstruct the luminal surface(Fig. 1C) by adapting our previously described methodology devel-oped for the fusion of IVUS and angiography [3]. In brief, anatomicalmarkers seen in both angiographic and OCT images were used toidentify the proximal and the distal end of the pull-back. Then, thecatheter path was extracted from two end-diastolic angiographicimages, the OCT borders were placed perpendicularly onto the 3Dpath and their absolute orientationwas determined. A good matchingwas achieved between the projections of the 3D model with theruptured plaque onto the angiographic images and the luminalsilhouette (Fig. 1D and E).Inthe3Dmodelrepresentingthelumenbeforeplaquerupture,thelocal shear stress distribution was calculated by computational fluiddynamics (Fig. 1F) [4]. Plaque rupture was located on the outercurvature of the vessel where the velocity profile was skewed leadingto locally elevated shear stress (Fig. 1G and H).