Evaluation of Robotic Coronary Surgery With Intraoperative Graft Angiography and Postoperative Multislice Computed Tomography

Evaluation of Robotic Coronary Surgery With Intraoperative Graft Angiography and Postoperative Multislice Computed Tomography Thomas Schachner, MD,* Gudrun M. Feuchtner, MD,* Johannes Bonatti, FETCS, MD, Nikolaos Bonaros, MD, Armin Oehlinger, MD, Eva Gassner, MD, Otmar Pachinger, MD, FESC, and Guy Friedrich, MD, FESC Departments of Cardiac Surgery, Cardiology, Radiology II, and Radiology I, Innsbruck Medical University, Innsbruck, Austria Background. Robotically assisted totally endoscopic coronary artery bypass graft surgery (TECABG) is an innovative minimally invasive procedure requiring proof of immediate- and short-term patency of grafts to compete with conventional bypass surgery or percutaneous coronary interventions. The purpose of this study was to evaluate the combination of invasive and noninvasive coronary angiography methods in innovative cardiac surgery as an approach to optimal quality control. Methods. In 86 patients after robotic coronary surgery (62 arrested-heart TECABG, 20 through sternotomy with robotically assisted anastomoses, 4 beating-heart TECABG), intraoperative coronary angiography was performed with a mobile C-arm. All patients underwent multislice computed tomography angiography, and invasive coronary angiography was performed in 48 patients within 3 months after surgery. Results. Bypass grafts could be visualized by intraoperative angiography in 84 patients (98%). Spasm of target vessels or bypass grafts, or both (reversible after intraluminal nitroglycerine application), was observed in 47%. In 9 patients, surgical revisions were performed owing to inadequate revascularization results. No angiographyrelated complications occurred. The sensitivity and specificity of multislice computed tomography for the evaluation of graft patency were 100% and 97%, respectively. The visualization of distal anastomoses and distal target vessels was good in 90% but limited in 10% because of artifacts, limited spatial resolution, and high image noise. Conclusions. The combination of intraoperative angiography and postoperative multislice computed tomography allows safe and high-quality evaluation of immediate- and short-term outcome in innovative robotic coronary surgery. Immediate revisions of bypass grafts are possible, to ensure that all patients leave the operating room with patent bypass grafts. (Ann Thorac Surg 2007;83:1361 7) 2007 by The Society of Thoracic Surgeons Totally endoscopic coronary artery bypass graft surgery (TECABG) is a new, minimally invasive coronary surgery technique developed in 1998 [1]. Considering the technically difficult surgical challenge of this operation type, as well as the concurrent possible prolonged learning curve, its implementation should be accompanied by a robust program of quality control [2 6]. Having built up logistics for intraoperative diagnostic coronary angiography [7], we sought to investigate the immediate results of TECABG using the standard diagnostic method and to compare these findings with those obtained through multislice computed tomography (MSCT) coronary angiography. This noninvasive imaging modality has recently been shown to provide a high diagnostic accuracy in the assessment of coronary artery bypass grafts [8 14]. Accepted for publication Oct 16, 2006. *Doctors Schachner and Feuchtner contributed equally to this article. Address correspondence to Dr Schachner, Innsbruck Medical University, Department of Cardiac Surgery, Innsbruck, Austria; e-mail: thomas. schachner@uibk.ac.at. Patients and Methods Study Population Between February 2002 and March 2006, intraoperative graft angiographies were performed in 86 patients who underwent robotically assisted coronary artery bypass grafting (CABG). The Institutional Review Board approved the totally endoscopic CABG operations, and informed consent was obtained from all patients. The operations performed were single-vessel, arrested-heart, totally endoscopic CABG (AHTECABG) using the left internal mammary artery (LIMA) to left anterior descending coronary artery (LAD) in 56 patients; double-vessel AHTECABG using LIMA to obtuse marginal (OM) and right IMA (RIMA) to LAD in 6 patients; beating-heart (off-pump), totally endoscopic CABG (BHTECABG, LIMA to LAD) in 4 patients; and robotically assisted, multivessel CABG through sternotomy in 20 patients. Patient characteristics are given in Table 1. The patient exclusion criteria for totally endoscopic CABG were age greater than 75 years, impaired renal function, moderate or severely impaired lung function with forced expiratory volume in 1 second of less than 70%, and severe peripheral vascular disease. 2007 by The Society of Thoracic Surgeons 0003-4975/07/$32.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2006.10.054

1362 SCHACHNER ET AL Ann Thorac Surg INTROPERATIVE GRAFT ANGIOGRAPHY 2007;83:1361 7 Abbreviations and Acronyms AHTECABG arrested-heart totally endoscopic CABG BHTECABG beating-heart totally endoscopic CABG CABG coronary artery bypass graft surgery LAD left anterior descending artery LIMA left internal mammary artery MSCT multislice computed tomography OM obtuse marginal branch RIMA right internal mammary artery Table 1. Demography of 86 Patients Who Underwent Intraoperative Graft Angiography After Innovative Coronary Artery Bypass Graft Surgery Variable Median (Minimum Maximum) or Number (%) Age (years) 59 (38 76) Male sex 69 (80%) Body mass index (kg/m 2 ) 27 (19 36) History of smoking 33 (38%) Family history of coronary artery 16 (19%) disease Hypertension 62 (72%) Hypercholesterolemia 65 (76%) Diabetes mellitus 5 (6%) Chronic obstructive pulmonary 16 (19%) disease Peripheral vascular disease 2 (2%) History of neurological events (TIA, 3 (4%) stroke) Preoperative NYHA stage 2.5 (1 4) Preoperative LVEF (%) 63 (36 88) Previous myocardial infarction 22 (26%) Preoperative serum creatinine 0.98 (0.58 1.53) (mg/dl) EuroSCORE a 1 (0 5) a The EuroSCORE is a prediction system of early mortality in patients undergoing cardiac surgery and was applied as described by Nashef and coworkers [15]. EuroSCORE European System for Cardiac Operative Risk Evaluation; LVEF left ventricular ejection fraction; TIA transient ischemic attack; NYHA New York Heart Association. Operative Technique In all operations, the davinci telemanipulator (Intuitive Surgical, Sunnyvale, California) was used through a left thoracic approach. Two working arms and one camera arm were placed in the third, fifth, and seventh intercostal spaces, respectively. For AHTECABG, the heart-lung machine was installed through the left femoral vessels using balloon-carrying arterial perfusion catheters for endoocclusion of the ascending aorta. For robotically assisted operations through sternotomy, either the LIMA was prepared endoscopically, or the LIMA-to-LAD anastomosis was sutured using the telemanipulator. Cardiopulmonary bypass time was 116 (57 to 342) minutes, and aortic cross-clamp time (endo-occlusion time) was 75 (31 to 223) minutes. Endoscopic LIMA takedown time was 45 (19 to 71) minutes. If conversions of TECABG operations were necessary, the sternotomy approach was used. The intention was to safely finish the operation through a standard approach. Intraoperative Graft Angiography Intraoperative graft angiography (Fig 1) was performed through a 7F femoral access in the left or right groin using the Judkins technique. Fluoroscopy was carried out with a GE OEC9800 mobile C-arm (General Electric; GE Healthcare, Chalfont St. Giles, UK), which was covered with sterile plastic sheets (frame rate: 25 frames per second). These sequences were transferred to a computer work station and visualized using OEC CRS-PC Software. Two to four projections (anterior posterior, left anterior oblique 20 degrees, right anterior oblique 20 degrees, or left anterior oblique 90 degrees) were used to visualize the grafts. Depending on individual anatomical features, standard Judkins right and left or bypass catheters (Boston Scientific, Natick, MA) were used. The operating table was a Maquet 1150.10D0 (Maquet, Rastatt, Germany). Multislice CT Multislice computed tomography angiography was performed using a 16-slice CT scanner (Sensation 16; Siemens, Erlangen, Germany [detector collimation 16 0.75 mm, gantry rotation time 0.42 s, tube output 120 kv/400 to 500 mas]) in 78 patients, or by using a 64-slice CT unit (Cardiac Sensation 64; Siemens [64 0.6 mm, 0.33 s, 120 kv/700 to 800 mas]) in 8 patients during a single inspiratory breath-hold of 15 s to 30 s. Scan delay was calculated with bolus tracking (ascending aorta, threshold 100 HU) with the 64-slice CT scanner or by applying the test bolus technique (16-slice CT). A bolus of 100 to 120 ml nonionic iodine contrast agent (Visipaque 320; GE Amersham Health, Buckinghamshire, UK) was injected intravenously at a flow rate of 3 to 5 ml/s. A -blocker was given intravenously (5 mg metoprolol) before the examination if the heart rate was greater than 80 beats per minute. Exclusion criteria were atrial fibrillation, renal dysfunction (serum creatinine 1.5 mmol/ L), hyperthyreosis, and known allergy to the iodine contrast agent. CT Image Reconstruction and Analysis A dataset of transaxial slices (effective slice width 1 mm, 16-slice CT; or 0.75 mm, 64-slice CT; 60% overlap; smooth convolution kernel B 20 f) was reconstructed by using retrospective electrocardiographic gating (at mid to late diastole [50% to 70% of RR interval] or mid to late systole [30% to 40%], dependent on heart rate). Postprocessing was performed by applying curved multiplanar reformations, maximum intensity projection, and a volumerendering technique. Graft patency was evaluated as

Ann Thorac Surg SCHACHNER ET AL 2007;83:1361 7 INTROPERATIVE GRAFT ANGIOGRAPHY 1363 After conversion to sternotomy, a saphenous vein graft was placed onto the circumflex system, and the LIMA was used as a free graft coming from the vein. A target vessel spasm was found in 37 patients (43%), and a spasm of the LIMA graft in 3 patients (4%), all of which were relieved after application of nitroglycerine. No angiography-related complications occurred. Clinical Outcome of the Patients The length of stay at the intensive care unit was 20 (11 to 389) hours. Respiratory support was necessary with a ventilation time of 8 (0 to 260) hours. No hospital death was observed in this series. Fig 1. Setting of intraoperative graft angiography in a patient who underwent totally endoscopic coronary artery bypass graft surgery. The left internal mammary artery catheter is advanced through the left femoral artery, which was already exposed for extracorporeal circulation. A mobile C-arm is used for fluoroscopy. The patient is still draped sterilely and fully anesthetized, enabling immediate bypass revisions if necessary. patent or nonpatent. The distal target vessel was graded as 1 well opacified, 2 weakly opacified, or 3 not visible by one experienced radiologist (G.M.F.) masked to the invasive graft angiography. Results Intraoperative Graft Angiography Intraoperative graft angiography was performed successfully in 84 of 86 patients (98%). In 1 case, the subclavian artery could not be intubated because of massive kinking of the iliac vessels. In another patient, the catheter could not be advanced into the left subclavian artery, and the angiography was terminated because of time concerns (an emergency case was scheduled). The mean (range) duration of intraoperative graft angiography was 20 (10 to 110) minutes. Fluoroscopy time was 367 (96 to 2,282) seconds, and the cumulative radiation (dose area product) reached 43,896 (10,261 to 429,787) mgy/cm 2. The amount of contrast agent (Iodixanol) used was 150 (20 to 500) ml. Surgical revision due to inadequate revascularization results detected by intraoperative angiography was performed in 9 of 86 cases (11%): in 5 cases, a proximal or distal target vessel stenosis in the anastomotic area occurred and was treated by revision of the anastomosis. In 1 patient early in the series, anastomotic bleeding was treated with a repair stitch. In another case, intramural hematoma of the LIMA graft was noticed. The distal part of the LIMA was resected, and the LIMA reanastomosed to the LAD. In 1 patient, an erroneous anastomosis of the LIMA to the coronary vein was performed, and was revised. In another patient, no inflow into a LIMA to marginal branch bypass was noticed, and that was due to tension of the vessel that was grafted endoscopically. Postoperative MSCT Angiography All 86 patients underwent MSCT angiography within 3 months after surgery. Overall, 92 grafts were assessed (n 82 LIMA/LAD, see Fig 2; n 5 LIMA/circumflex, n 5 RIMA/LAD; see Fig 3). A total of 90 of 92 grafts were found patent down to the anastomosis (Fig 4). Postoperative MSCT Angiography Versus Invasive Graft Angiography 3-Month Follow-Up A total of 48 patients underwent MSCT and invasive angiography after 3 months. Overall, 50 grafts were assessed (n 46 LIMA/LAD; n 2 LIMA/OM; n 2 RIMA/LAD). A total of 48 of 50 grafts were found patent on both MSCT and invasive angiography. The opacification of the distal target vessel was graded as well in 42 of 50 patients (84%), as weak in 5 of 50 (10%), and as absent in 3 of 50 (6%). Accordingly, in 1 of the latter 3 patients, inadequate image quality was noted because of stair-step motion artifacts explained by an unexpected heart rate increase greater than 80 to 90 beats per minute during the scan. In the remaining 2 of those patients, one proximal 75% LIMA stenosis and one 99% LIMA/OM anastomosis stenosis was found. Of 5 patients with a weak distal target vessel opacification on 16-MSCT, 1 patient had an 80% stenosis distal to the anastomosis (Fig 5) on invasive graft angiography; 1 patient had a previously known small and severely calcified distal LAD due to a total proximal LAD occlusion; and in 3 patients, invasive angiography showed no stenosis, but overall CT image quality was hampered by high image noise (body mass index 30) or motion artifacts leading to a weak display of the distal LAD. The sensitivity and specificity of MSCT for the assessment of graft patency when compared with invasive angiography were 100% and 98%, respectively (positive predictive value 67% and negative predictive value 100%). In summary, the visualization of distal anastomosis and the distal target vessel was good in 90% of patients (43 of 48) but technically limited in 10% (5 of 48). Follow-Up Invasive Coronary Angiography Short-term follow-up angiography was performed in 48 patients at 3 (0 to 15) months postoperatively. Early postoperative angiography (before discharge) was performed in 3 patients: in 1 patient, intermittent postoperative ST elevations occurred, but the angiography

1364 SCHACHNER ET AL Ann Thorac Surg INTROPERATIVE GRAFT ANGIOGRAPHY 2007;83:1361 7 Fig 2. Hybrid revascularization. (a) The 64-slice computed tomography images of a 52-year-old man with a poststenotic coronary aneurysm (C) of the left anterior descending (LAD) artery, which precluded LAD stenting. Therefore, the patient was treated with totally endoscopic single left internal mammary artery (LIMA) grafting to the LAD and the simultaneous implantation of a cypher stent (Cordis; Johnson & Johnson, New Brunswick, NJ [black arrow]) into the circumflex artery (CX) intraoperatively during one session (hybrid revascularization). The coronary aneurysm (C) releases both diagonal branches (DG). The image was reconstructed in three dimensions by applying a volume-rendering technique. Note that four vessels are arising from the aortic arch (1 innominate artery; 2 left common carotid artery; 3 left vertebral artery; and 4 left subclavian artery), which is useful information before performing arrested-heart totally endoscopic coronary artery bypass graft surgery. White spots at the aortic arch indicate calcifying plaque. (b) The corresponding catheterized angiography image shows the LIMA graft patent, with a good runoff distal into the LAD. showed a good operative result. In another patient, the LAD was severely atherosclerotic and showed a narrowing immediately distal from the anastomosis in the intraoperative angiography. The postoperative angiography confirmed an adequate LIMA-to-LAD anastomosis and a borderline stenosis distal from the anastomotic site. The third patient with a perianastomotic hematoma in the MSCT had an adequate LIMA-to-LAD anastomosis. In 2 of 48 patients (4%), an anastomotic stenosis was present: in 1 patient, with an originally LIMA-to-LAD anastomosis revision, 3-month angiography showed an 80% stenosis of the anastomosis. Because the patient presented no clinical symptoms, we opted for conservative therapy. The 3-month MSCT showed a good bypass runoff, but narrowing of the anastomosis could not be detected. In the second patient, a LIMA-to-OM anasto- Fig 3. Double-vessel totally endoscopic coronary artery bypass graft. Image from a 66- year-old man who underwent double-vessel endoscopic revascularization is shown. (a) The right internal mammary artery (RIMA) was sutured to the left anterior descending artery (LAD). (b) The left internal mammary artery (LIMA) was connected to the circumflex artery (CX). The black arrow indicates the anastomosis (A). Both grafts were patent on both 64- slice computed tomography and intraoperative angiography. Note that the distal target vessel was found to be well opacified on multislice computed tomography, suggesting a good distal runoff.

SCHACHNER ET AL INTROPERATIVE GRAFT ANGIOGRAPHY 1365 Ann Thorac Surg 2007;83:1361 7 Fig 4. Single-vessel totally endoscopic coronary artery bypass graft. (a) Image from a 59-year-old man in whom single left internal mammary artery (LIMA) to left anterior descending artery (LAD) grafting was performed. Graft patency was assessed with 16-slice computed tomography by applying a volume-rendering technique. (b) Computed tomography images using a maximum intensity projection technique, which is advantageous for the display of the graft lumen and for determining patency. (c) Invasive angiography confirmed LIMA graft patency. mosis showed a severe narrowing. It has to be pointed out that the target vessel was very thin, and the patient was clinically asymptomatic (RIMA-to-LAD bypass showing an excellent result). In accordance with these results, the 3-month MSCT showed a compromised runoff (LIMA-to-OM) in this patient. Comment Multislice computed tomography [8 14] and intraoperative graft angiography [16 18] are valuable and sensitive tools to evaluate bypass graft patency after CABG. Intraoperative angiography is of special interest because it offers the possibility of performing revisions while the patient is still in the operating room. In our series, we revised one tenth of patients after robotically assisted CABG. Two factors may explain this relatively high number: the learning curve of a technically difficult procedure, and the invasive gold standard method of graft patency control, which is supposed to be superior to conventional approaches. On the other hand, we observed that all patients left the operating room with patent bypass grafts. However, not all patients are suitable candidates for intraoperative angiography because of anatomical features: we experienced severe technical difficulties in 2% of the patients, with subsequent termination of the angiographic approach. Multislice computed tomography angiography showed a high diagnostic accuracy for the evaluation of bypass graft patency with a sensitivity of 100% and a specificity of 97%. The visualization of the distal target vessel was graded as well in 90%, which has been previously reported [10]; however, it was impaired in 10% because of technical limitations such as motion artifacts, high image noise in obese patients, or a small-size target vessel with severe atherosclerosis. The spatial resolution of 16-MSCT of 0.5 0.5 0.6 mm3 [19] is still lower when compared with invasive graft angiography, which limits the visualization of the distal target vessel and may lead to difficulties in detecting stenosis of the distal anastomosis (Fig 4). Most (90.8%) of our patients were scanned with a 16-slice CT scanner. The recently introduced 64-slice CT technique offers improved spatial (0.4 0.4 0.4 mm3) and temporal resolution [20], thus improving visualization of the distal anastomosis and reducing artifacts from residual cardiac motion. Therefore, only the combination of intraoperative graft angiography and early postoperative MSCT angiography allows excellent control of coronary bypass grafts without the disadvantage of ruling out too many patients for technical imaging reasons. Follow-up invasive coronary angiography showed a stenosis in 1 of 46 (2%) of the LIMA-to-LAD anastomoses and in 1 of 2 LIMA-to-OM anastomoses. This outcome most probably reflects the difficulties of endoscopic coronary surgery beyond LIMA-to-LAD grafting; the LIMAto-OM bypass intervention is limited by the distance of the circumflex system to the mammary artery. One possible solution to this problem would be the endo-

1366 SCHACHNER ET AL Ann Thorac Surg INTROPERATIVE GRAFT ANGIOGRAPHY 2007;83:1361 7 Fig 5. Single-vessel totally endoscopic coronary artery bypass graft. Image from a 61-year-old asymptomatic man presenting with an 80% stenosis (S) of the left anterior descending artery (LAD) distal to the anastomosis (A) after left internal mammary artery (LIMA) grafting. (A) Black arrows indicate the stenotic segment (S) after the anastomosis (A) on both (a, b) 16-multislice CT and (c) invasive angiography. (A) White arrows denote an aortocoronary venous graft to the obtuse marginal branch. (c) Note that the retrograde filling of the LAD was absent on invasive angiography. scopic employment of the aortocoronary saphenous vein or radial artery bypass grafts in addition to LIMA-to-LAD grafts. The detection of anastomotic stenoses is still the domain of invasive angiography. In agreement, we observed that a significant stenosis could not be detected by MSCT angiography in a graft with a good distal runoff, whereas the stenosis was identified (indirectly) in another patient with impaired distal runoff. However, a weak runoff was also observed in patients with a very small-size target vessel or in patients in whom technical limitations were noted (eg, artifacts or high image noise). Therefore, impaired distal runoff cannot be used as a reliable criterion for the detection of distal graft stenosis with 16-MSCT. Limitations Although we observed no invasive- or CT-related angiographic complications, our study protocol has some limitations: The patient number investigated was small, but the TECABG technique is very innovative, and selection of patients presenting adequate anatomical features for the use of endoscopic instruments and not suitable for percutaneous interventions is mandatory. Furthermore, the cumulative radiation dose from intraoperative and follow-up MSCT angiography is high. This factor, however, may be outweighed by prevention of postoperative ischemic events and the intended strategy of leaving out invasive angiographic monitoring in these patients based on MSCT findings. In conclusion, the combination of intraoperative graft angiography and postoperative MSCT angiography allows a safe and reliable quality control of immediate and follow-up results after robotically enhanced endoscopic coronary surgery. However, visualization of the distal LAD and distal anastomosis stenosis is technically limited by using 16-MSCT angiography. References 1. Loulmet D, Carpentier A, d Attellis N, et al. Endoscopic coronary artery bypass grafting with the aid of robotic assisted instruments. J Thorac Cardiovasc Surg 1999; 118:4 10. 2. Falk V, Diegeler A, Walther T, et al. Total endoscopic computer enhanced coronary artery bypass grafting. Eur J Cardiothorac Surg 2000;17:38 45. 3. Bonatti J, Schachner T, Bernecker O, et al. Robotic totally endoscopic coronary artery bypass: program development and learning curve issues. J Thorac Cardiovasc Surg 2004; 127:504 10.

Ann Thorac Surg SCHACHNER ET AL 2007;83:1361 7 INTROPERATIVE GRAFT ANGIOGRAPHY 1367 4. Bonatti J, Schachner T, Bonaros N, et al. Ongoing procedure development in robotically assisted totally endoscopic coronary artery bypass grafting (TECAB). Heart Surg Forum 2005;8:E287 91. 5. Schachner T, Bonaros N, Feuchtner G, Muller L, Laufer G, Bonatti J. How to handle remote access perfusion for endoscopic cardiac surgery. Heart Surg Forum 2005;8:E232 5. 6. Bonatti J, Schachner T, Bonaros N, et al. Technical challenges in totally endoscopic robotic coronary artery bypass grafting. J Thorac Cardiovasc Surg 2006;131:146 53. 7. Friedrich GJ, Jonetzko P, Bonaros N, et al. Hybrid coronary artery revascularization: logistics and program development. Heart Surg Forum 2005;8:E258 61. 8. Pache G, Saueressig U, Frydrychowicz A, et al. Initial experience with 64-slice cardiac CT: non-invasive visualization of coronary artery bypass grafts. Eur Heart J 2006;27:976 80. 9. Dewey M, Lembcke A, Enzweitler C, Hamm B, Rogalla P. Isotropic half-millimeter angiography of coronary bypass grafts with 16-slice computed tomography. Ann Thorac Surg 2004;77:800 4. 10. Schlosser T, Konorza T, Hunold P, et al. Noninvasive visualization of coronary artery bypass grafts using 16-detector row computed tomography. J Am Coll Cardiol 2004;44: 1224 9. 11. Martuscelli E, Romagnoli A, D Eliseo A, et al. Evaluation of venous and arterial conduit patency by 16-slice spiral computed tomography. Circulation 2004;16:3234 8. 12. Chiurlia E, Menozzi M, Ratti C, Romagnoli R, Modena MG. Follow-up of coronary artery bypass graft patency by multislice computed tomography. Am J Cardiol 2005;95:1094 7. 13. Stein PD, Beemath A, Skaf E, et al. Usefulness of 4-, 8-, and 16-slice computed tomography for detection of graft occlusion of patency after bypass grafting. Am J Cardiol 2005;96: 1669 73. 14. Anders K, Baum U, Schmid M, et al. Coronary artery bypass graft (CABG) patency: assessment with high-resolution submillimeter 16-slice multidetector-row computed tomography (MDCT) versus coronary angiography. Eur J Radiol 2006;57: 336 44. 15. Nashef SA, Roques F, Michel P, Gauducheau E, Lemeshow S, Salamon R. European System for Cardiac Operative Risk Evaluation (EuroSCORE). Eur J Cardiothorac Surg 1999; 16:9 13. 16. Bonatti J, Danzmayr M, Schachner T, Friedrich G. Intraoperative angiography for quality control in MIDCAB and OPCAB. Eur J Cardiothorac Surg 2003;24:647 9. 17. Mack MJ, Magovern JA, Acuff TA, et al. Results of graft patency by immediate angiography in minimally invasive coronary artery surgery. Ann Thorac Surg 1999;68:383 9. 18. Izzat MB, Khaw KS, Atassi W, Yim AP, Wan S, El-Zufari MH. Routine intraoperative angiography improves the early patency of coronary grafts performed on the beating heart. Chest 1999;115:987 90. 19. Flohr T, Bruder H, Stierstorfer K, et al. New technical developments in multislice CT, part 2. Sub-millimeter 16- slice scanning and increased gantry rotation speed for cardiac imaging. Rofo 2002;174:1023 7. 20. Flohr T, Stierstorfer K, Raupach R. Performance evaluation of a 64-slice CT system with z-flying focal spot. Rofo 2004; 176:1803 10.