This report demonstrates the production and use of 3-D reconstruction of a coronary sinus from a single-injection rotational angiogram. The detailed virtual model enabled easy magnetic navigation of a wire for device placement in cardiac resynchronization therapy. The use of the magnetic navigation system (MNS, Stereotaxis, St Louis, Missouri) in electrophysiology is now well established. This system has been described in detail previously.1,2 Using computer-controlled movements of the magnets, a 15–20 cm uniform magnetic field volume can be directed in 360° in all planes to deflect a magnetic wire tip precisely. This control is coupled with an ability to reconstruct a patient-specific vessel and provide image-guided navigation for each and every section of the vessel. A 3-dimensional (3-D) model can be produced that enables magnetic field direction control from the computer-generated image. Published reports have shown the clinical utility of the MNS in routine electrophysiology catheter ablation procedures, including specific reports describing its use in atrial fibrillation,3,4 atrioventricular nodal reentrant tachycardia,5 atrioventricular reentrant tachycardia,6 right ventricular outflow tract tachycardia,7 parahisian accessory pathways,8 and more complex arrhythmia ablation procedures such as ventricular tachycardia.9 Recent studies indicate its usefulness in cardiac resynchronization therapy (CRT) to facilitate the placement of left ventricular leads in cardiac resynchronization therapy device implantation procedures.10,11 This report demonstrates the feasibility of producing an on-table, real-time coronary sinus reconstruction to facilitate lead placement in a cardiac vein. This virtual coronary sinus was produced with a single contrast injection and the use of rotational angiography. Case Report. A 76-year-old female with poor left ventricular function due to ischemic heart disease was admitted for cardiac resynchronization therapy and implantable cardioverter-defibrillator (ICD) deployment. She had previously undergone coronary artery bypass grafting (CABG) and had had a DDD pacemaker implanted for Mobitz II heart block. An old atrial active fixation lead was successfully extracted while the old ventricular tined lead was left in situ. An Endotak Reliance SG active fixation shock lead (Guidant) was placed in the RV apex and a CapSureFix active fixation lead (Medtronic) in the RA. The coronary sinus (CS) was cannulated using a CS guiding catheter. The table was put in the isocenter position, the CS was occluded using a balloon catheter, and 20 ml of diluted contrast used to perform rotational venography. A 3-D virtual coronary sinus was reconstructed using 3-DCA software (Philips Medical Systems) (Figure 1). This model was imported into the MNS (Figure 3). A Titan™ 3 mm angled Soft Support wire passed on the first attempt using navigation based on the model into a side branch of the great cardiac vein. An Easytrak 3 lead (Guidant) passed easily over this. A Contak Renewal 4 RF (VVED-CRT; Guidant) was implanted. All checks were satisfactory at implantation and at follow up. Discussion. CRT is an important technique in reducing morbidity and improving patient well-being. CRT improves quality of life and prolongs survival in appropriate patients with QRS prolongation or left ventricular (LV) dysynchrony.12–15 However, a number of difficulties may hamper lead placement in CRT procedures such as difficulty in obtaining CS access, efficiently selecting CS branch vessels, maintaining lead stability and avoiding both phrenic nerve stimulation and lead dislodgement.16–20 Feasibility studies suggest that the MNS without full 3-D reconstruction appears to be at least as good as manual technique.10,11 The rotational venogram provides a number of possible benefits that may facilitate lead placement in CRT. The production of a patient-specific 3-D model supplies the vectors required for image-guided steering of the wire along the length of the vessel (Figure 3). In addition, this 3-D model supplies a white line overlay (of the center-line of the model) that automatically adjusts to the view on the fluoroscopy screen. This acts as a guide to wire positioning. This technique has two clear advantages for contrast use and procedure time for two reasons. First, a patient-specific 3-D model of the coronary sinus and its branches is produced from a single injection of contrast, thus reducing contrast and time. Second, the use of the patient-specific vectors should facilitate wire placement with facilitation of wire placement and re-positioning potentially leading to reduced contrast use and procedure times. Conclusion. The combination of improved navigation and a reduced burden to the patient and operator may be facilitated by such easy on-table 3-D reconstruction as it allows computer generated image guidance of magnetic field direction control and better wire steering. This may lead to faster wire placement by facilitating wire delivery to a target sidebranch and also easier repositioning should unsatisfactory lead stability or phrenic pacing occur. This will lead ultimately to limitation of contrast load, radiation time, and procedure times. Acknowledgement. The authors are grateful to Joep Maeijer for help with preparation of the images.