Author + information
- Received June 25, 2008
- Revision received October 21, 2008
- Accepted October 26, 2008
- Published online March 3, 2009.
- Francesca Pugliese, MD
- Steve Ramcharitar, BMBCh, DPhil*,
- Carl Schultz, MBChB, DPhil*,†,
- Jurgen Ligthart, BSc*,
- Pim de Feyter, MD, PhD*,†,
- Huling Li, MD†,
- Nico Mollet, MD, PhD†,
- Martin van de Ent, MD, PhD*,
- Patrick W. Serruys, MD, PhD* and
- Robert Jan van Geuns, MD, PhD*,*,† ()
- ↵*Reprint requests and correspondence:
Dr. Robert Jan van Geuns, Thoraxcenter, Ba585, Dr Molewaterplein 40, 3015-GD Rotterdam, the Netherlands
Objectives We hypothesized that percutaneous coronary intervention (PCI) without additional contrast agents can be performed by directly integrating multislice computed tomography coronary angiography (CTCA) within the magnetic navigation system (MNS).
Background Increasingly, CTCA is being used in the diagnostic work-up of patients with coronary disease. Its inherent 3-dimensional information should be exploited, as it potentially offers advantages over 2-dimensional radiography in guiding invasive diagnostic and therapeutic interventions.
Methods CTCA-derived centerlines from 15 patients were coregistered and overlaid on real-time fluoroscopic images employing the MNS. Vessels were manually wired with a magnetically enabled guidewire assisted by variable local magnetic fields. Fractional flow reserve (FFR) determined the lesion severity, and the dimensions were quantified by intravascular ultrasound (IVUS). Locations of the IVUS catheter probe along the lesion were incorporated in software to facilitate stenting without contrast agents.
Results The average crossing and fluoroscopic times were 105.3 ± 35.5 s and 83.4 ± 38.6 s, respectively, with no contrast agents used in 11 of 15 patients (73.3%). Contrast agents were used in only 1 of 10 patients (10%) in whom an IVUS was performed. In 4 patients, apart from a “blinded” safety check angiogram, the entire PCI (lesion crossing, stent sizing, positioning, and deployment) was performed without additional contrast agents following the coregistration of the IVUS probe position in the MNS.
Conclusions The integration of pre-procedural CTCA within the MNS can facilitate PCI without additional contrast agents.
Although over 2 million percutaneous coronary intervention (PCI) procedures are performed annually worldwide, the general methodology has remained unchanged since its conception over one-quarter of a century ago, relying primarily on the manual dexterity of the operator and 2-dimensional (2D) radiographic imaging through the selective injection of contrast agents into the coronary arteries (1). Recently, however, the noninvasive contrast-enhanced multislice computed tomography coronary angiography (CTCA) has been achieving sufficient quality, warranting an increased use for diagnostic workup of patients with chest pain syndrome (2). We hypothesized that the CTCA 3-dimensional (3D) information can provide a tentative roadmap to directly guide a PCI procedure once integrated with the magnetic navigation system (MNS) to eliminate the need for a contrast-filled lumenogram at the beginning of a procedure (3). Furthermore, decisions on treatment indication, stent sizing, and stent location can be based on functionally characterized lesions and intravascular ultrasound (IVUS), respectively, thus enabling the completion of the entire PCI procedure without contrast media (4) (Fig. 1).
Five patients with single-vessel and 10 patients with complex multivessel disease on CTCA were integrated in the MNS (Niobe II, Stereotaxis, St. Louis, Missouri), and the target vessel was crossed using Titan 2- or 3-mm long magnetic tip guidewires (Stereotaxis). Patients were excluded if they were hemodynamically unstable or had acute infarctions, acute coronary syndromes, contraindications for the MNS (e.g., pacemakers or intracardiac defibrillators), left main stem or ostial right coronary artery disease, or a significant amount of calcium on CTCA. Our institutional review board approved the study, with all patients giving written consent to undergo CTCA prior to conventional coronary angiography and enrollment in the study.
The CTCA (Somatom Sensation 64 Cardiac Configuration, Siemens, Erlangen, Germany) were reconstructed using a monosegmental, electrocardiographically gated reconstruction algorithm requiring an 80- to 100-ml intravenous contrast agent injection (5). One experienced observer (F.P.) determined significant stenoses in vessels >2 mm in diameter from multiplanar and curved multiplanar reconstructions.
Magnetic guidewire navigation using CTCA cointegration
The CTCA Digital Imaging and Communications in Medicine images of the extracted coronary arteries (Leonardo Circulation, Siemens Medical Solutions, Forchheim, Germany) (Fig. 2A)were directly uploaded in Navigant (Stereotaxis), the MNS guidance software needed to create a navigational pathway (Fig. 2B). Coregistration with the live fluoroscopic image was achieved by first recording 2 orthogonal non–contrast agent-filled images of the engaged guiding catheter tip in end diastole. This point was aligned to the similar location in the extracted CTCA coronary arteries (Figs. 2C and 2D). Subsequently, a pathway could be displayed on the live fluoroscopic image in any desired orientation. Navigational vectors created along the centerline gave information that allowed 2 large external permanent magnets to rotate, tilt, and translate about their axes to redirect a 0.08-T external magnetic field to precisely guide the magnetically enabled wire (Figs. 2E and 2F). These vectors were updated as the operator manually advanced the guidewire under radiographic X-ray control, but without requiring contrast agents.
Fractional flow reserve (FFR) and IVUS analysis
FFR assessments were performed without contrast agents by exchanging the magnetic-enabled wire with the FFR guidewire (PressureWire, Radi Medical Systems, Uppsala, Sweden) using a 2-lumen multifunctional probing catheter (Boston Scientific Corporation, Santa Clara, California). A FFR ≥0.75 was considered functionally nonsignificant. Intravascular ultrasound was performed using a CVIS Atlantis SR Pro 2.5-F 40-MHz catheter (Boston Scientific Corporation).
Performing an entire PCI procedure without contrast agents
Following appropriate stent sizing, the IVUS catheter was readvanced to the distal landing zone. The position of the piezoelectric crystal, recorded on 2 different angiographic projections (Figs. 3Aand 3B) and identified in the 3D space of the Navigant (Figs. 3C and 3D), provided an additional overlay point on the live fluoroscopic navigational centerline for stent positioning and deployment without contrast agent (Figs. 3E and 3F). Because our aim was to demonstrate the concept of 3D integration to perform a PCI on pre-procedural CTCA, we felt that at this stage, it was necessary to incorporate a “blinded” safety-check angiogram performed by a second operator to check the accuracy of positioning prior to stent deployment. This safety-check film was also used for American College of Cardiology/American Heart Association lesion characterization.
Assessment of lesion lengths and reference diameters by IVUS and CTCA
Off-line IVUS and CTCA measurements were done independently by an IVUS trained technician and a radiologist experienced in interpreting CTCA. Lesion length, minimum luminal diameter (MLD), proximal diameter stenosis (DS), and percent DS were analysed. The absolute differences were calculated by the numerical summation of the overestimated and underestimated values relative to both techniques. Stent apposition was visually assessed, with incomplete apposition defined as at least 1 stent strut not apposed to the vessel wall and clear blood flow behind the stent.
Continuous variables are presented as mean ± SD. Paired Student ttests were used to determine significant differences between the parameters assessed. Analyses were performed using SPSS 11.5 for Windows (SPSS Inc., Chicago, Illinois). A p value <0.05 was considered statistically significant.
FFR and IVUS assessments
The CTCA-derived overlay was successfully incorporated in the MNS to achieve average crossing and fluoroscopic times of 105.3 ± 35.5 s and 83.4 ± 38.6 s, respectively. Overall, contrast media usage was limited to an average of 0.5 ml, and in 11 of the 15 patients (73.3%), no contrast agent was needed for crossing (Table 1).The largest volume of contrast agent used was 3 ml needed to re-enter the main vessel following advancement of the magnetic wire into a small side branch almost parallel to the main branch. In 3 other cases, a small amount of contrast agent (1 to 2 ml) was used to visualize the distal wire position, but not for any readjustments. Consequently, taken as a group to include magnetic wire crossings and lesion evaluation, no contrast agent was used in 63% of patients (5 of 8) having FFR measurements and in 90% of patients (9 of 10) in whom IVUS measurements were performed. In 1 case in which a patient had a negative FFR (0.87), no further investigation/treatment was deemed necessary, and the patient was discharged. In the 4 other patients with single-vessel disease, conventional PCI techniques were employed to treat the lesions. Multivessel FFR successfully identified the culprit lesion/vessel in Patients #6 to #8 and in a further 3 multivessel cases (Patients #9 to #11) in which IVUS was employed to accurately determine the stent size (average diameter, 3.3 mm) and length (average length, 17.6 mm) without contrast agents.
Complete PCI without additional contrast agents
In Patients #12 to #15, the entire PCI procedure from lesion crossing, stent sizing, and positioning to deployment was performed without additional contrast agents, relying exclusively on CTCA and IVUS data incorporated in the Navigant. In only 1 case following the safety-check angiogram was an adjustment (2 mm) required. Repeated IVUS post-stent deployment demonstrated satisfactory stent apposition not requiring post-dilation. In keeping with the hospital's angiographic recording policy, the final results were documented using 2 different angiographic views.
Analysis of IVUS and CTCA measurements
Off-line IVUS and CTCA analyses revealed that longer lesion lengths were determined by CTCA measurements (19.2 ± 13.3 mm vs. 14.5 ± 5.6 mm; p = 0.19) (Table 2).No significant differences were observed in measuring the MLD and %DS. But a near significant difference was observed in proximal reference diameter measurements (3.6 ± 0.6 mm vs. 4.2 ± 0.8 mm; p = 0.05), which were overall smaller with CTCA. In terms of the absolute difference, CTCA measurements would have resulted in 1 stent size smaller (absolute difference: 0.8 mm) and longer (absolute difference: 6.6 mm) than that determined through IVUS analysis.
We present a new concept of performing PCI by directly integrating pre-procedural diagnostic CTCA (3D) data at a later date, thus avoiding an otherwise superfluous contrast agent-filled 2D lumenogram. The precise nature of the CTCA dataset allows the accurate mapping of the native vessel to be exploited in localizing and coregistering an IVUS catheter probe within the 3D space to facilitate stenting without additional contrast agents. Despite its cost and higher radiation exposure, CTCA is noninvasive, and judging by its rapidly gaining popularity, it may be seen in the future as an attractive alternative to conventional angiography (6). Its limited spatial resolution, however, does suggest that in this study, if stent selection had been based primarily on CTCA alone, then 1 stent length longer with a smaller diameter would have been incorrectly chosen, emphasizing the importance of IVUS to quantify the lesion dimensions. In addition, lesion severity can be overestimated due to the blooming artifact of microcalcification seen on the CTCA (7). Similarly, as with conventional angiography, anatomic evaluation on CTCA does not always correspond with functional analysis, and so provides the rationale for using functional assessment (FFR) to fully characterize the lesion (8). FFR-guided therapy in multivessel disease is cost effective and reduces unnecessary stenting (9).
Several key developments using faster systems capable of increased spatial resolution and utilizing single heart beat imaging to limit the radiation exposure are being investigated (10). Nevertheless, with the satisfactory image quality obtained with the modern 64-slice scanners, additional coronary luminology with the use of more radiation and contrast agents seems unnecessary, especially when such use provides less information on the lesion plaque eccentricity and composition, as well as on the 3D relationship of the coronary arteries (11). The 1 case in which an adjustment of 2 mm was required prior to stent deployment, however, identifies a significant limitation in using a single, static CTCA roadmap derived from only 1 phase of the cardiac cycle (diastole) for positioning. This suggests that a dynamic 4-dimensional CTCA roadmap involving more than 1 phase, or ideally involving the whole cardiac cycle, cointegrated with the MNS is desirable to improve accuracy if pre-procedural CTCA is to be exploited in the future to guide PCI without the reliance on conventional 2D imaging and its notable limitations (12).
This was a small feasibility study, and patients were selected based on the availability of the 64-slice CT scanner. Direct comparison of the precision and accuracy of CTCA and IVUS was not our intention, as this has been validated by other groups. CTCA integration in the MNS was time consuming (up to 30 min). Apart from the safety-check angiogram prior to stent deployment, a final film was required for documentation. Therefore, to be truly independent of contrast agents, a validation study based solely on 3D imaging techniques (such as 3D-IVUS) may be required.
Percutaneous coronary intervention without the use of additional contrast agents with optimal results is feasible by integrating CTCA and IVUS with the MNS.
Mr. Ligthart has given lectures about IVUS on Boston Scientific facilitated IVUS courses. He has a contract for this with Boston Scientific Europe.
- Abbreviations and Acronyms
- computed tomography coronary angiography
- diameter stenosis
- fractional flow reserve
- intravascular ultrasound
- magnetic navigation system
- percutaneous coronary intervention
- Received June 25, 2008.
- Revision received October 21, 2008.
- Accepted October 26, 2008.
- American College of Cardiology Foundation
- Meijboom W.B.,
- van Mieghem C.A.,
- Mollet N.R.,
- et al.
- Fearon W.F.,
- Luna J.,
- Samady H.,
- et al.
- Goldstein J.A.,
- Gallagher M.J.,
- O'Neill W.W.,
- Ross M.A.,
- O'Neil B.J.,
- Raff G.L.
- Schroeder S.,
- Achenbach S.,
- Bengel F.,
- et al.
- Meijboom W.B.,
- Van Mieghem C.A.,
- van Pelt N.,
- et al.
- Ragosta M.,
- Bishop A.H.,
- Lipson L.C.,
- et al.
- Husmann L.,
- Valenta I.,
- Gaemperli O.,
- et al.
- Fishman E.K.,
- Horton K.M.
- Topol E.J.,
- Nissen S.E.