Multidetector Computed Tomography (MDCT) in Coronary Surgery: First Experiences With a New Tool for Diagnosis of Coronary Artery Disease Hendrik Treede, MD, Christoph Becker, MD, Hermann Reichenspurner, MD, PhD, Andreas Knez, MD, Christian Detter, MD, Maximilian Reiser, MD, and Bruno Reichart, MD Departments of Cardiac Surgery, Clinical Radiology, and Cardiology, Ludwig-Maximilians-University Munich, Germany, and Cardiovascular Surgery, University Hospital Eppendorf, Hamburg, Germany Background. Selective coronary angiography (SCA) is the standard invasive procedure for diagnosis in patients eligible for coronary artery bypass grafting (CABG). A recently developed, highly sensitive multidetector computed tomography (MDCT) scan holds promise to be of almost comparable quality and predictiveness. We examined a blinded series of preoperative patients who were admitted to hospital for conventional and minimally invasive CABG procedures. Patients underwent CT scans in addition to SCA; findings were compared regarding location and degree of coronary artery stenosis. Methods. Twenty patients underwent electrocardiogram-gated helical CT scanning. Images with 250 ms effective exposure time were reconstructed with retrospective electrocardiogram gating. Location and degree of coronary stenoses were described and compared with findings of SCA. The study was limited to patients with a heart rate of less than 70 beats per minute and who had the ability to hold their breath for 20 to 30 seconds. Coronary artery disease (CAD) is the leading cause of morbidity and mortality in western countries. The diagnostic gold standard in determining the site and severity of coronary artery stenoses and occlusions in individuals with CAD has been selective coronary angiography (SCA). Although SCA offers the highest resolution of all available imaging techniques, this expensive, invasive procedure carries multiple risks and reveals only calcifications or lesions that have already led to significant coronary stenoses in patients who are mainly already symptomatic. Selective coronary angiography may fail to determine CAD in early stages in which atheromatous lesions lack significant pathologic changes. Moreover, the procedure does not permit differentiation of the quality of the narrowing, such as calcified, fatty, or fibrous proliferation, and therefore cannot offer information that may affect specific patient risk profiles. Presented at the Eighth Annual Cardiothoracic Techniques and Technologies Meeting 2002, Miami Beach, FL, Jan 23 26, 2002. Address reprint requests to Dr Reichenspurner, Department of Cardiovascular Surgery, University Hospital Eppendorf, Hamburg, Martinistr. 15, 20246 Hamburg, Germany; e-mail: treede@uke.uni-hamburg.de. Results. Coronary arteries were clearly displayed by MDCT. Compared with SCA, sensitivity was 92%, specificity 84%, and negative predicted value 89% for significant stenosis (more than 50%). Early forms of atherosclerotic changes were even clearer on MDCT. In addition, the CT examination allowed differentiation of calcified and fatty or fibrous stenoses. Conclusions. Multidetector CT scanning is an effective noninvasive technique for the diagnosis of coronary artery disease. In selected patients, MDCT scanning might be able to replace SCA as a preoperative test for CABG procedures. The intrathoracic situs can be clearly exposed as it is important for the planning of minimally invasive CABG procedures. (Ann Thorac Surg 2002;74:S1398 402) 2002 by The Society of Thoracic Surgeons Recently, noninvasive cardiac imaging techniques such as transesophageal echocardiography, magnetic resonance imaging (MRI), and electron beam computed tomography (EBCT) have had an increasing role in the diagnosis of CAD and also in the preoperative planning of regular and minimally invasive coronary artery bypass grafting (CABG) procedures. The newly developed multirow detector helical CT (MDCT) has rotation times of 500 ms that allow image acquisitions with 250-ms effective exposure time. Electrocardiogram (ECG)-gated helical reconstruction in the end-diastolic slow motion phase of the cardiac cycle results in high image resolution without artifacts due to cardiac motion, as are seen on conventional helical CT scans. Noncontrast-enhanced MDCT scans allow detection and quantification of coronary calcifications [1] as can be detected on EBCT and have been described by Agatston and colleagues [2]. Results from MDCT scans have been useful in estimating the risk for future cardiac events, especially in asymptomatic patients [3]. Contrastenhanced MDCT angiography in patients with slow heart rates in sinus rhythm offers an excellent imaging quality that promises high sensitivity and specificity in the de- 2002 by The Society of Thoracic Surgeons 0003-4975/02/$22.00 Published by Elsevier Science Inc PII S0003-4975(02)04010-9
Ann Thorac Surg CTT SUPPLEMENT TREEDE ET AL 2002;74:S1398 402 MDCT IN CORONARY SURGERY S1399 tection and grading of CAD that is comparable to that of SCA. In this study we compared coronary MDCT angiography and conventional SCA for the detection and quantification of CAD in patients eligible for CABG procedures. Material and Methods Fig 1. American Heart Association model for segmental analysis of coronary arteries. Segments are marked by gray boxes and are numbered from 1 (proximal RCA) to 15 (posterolateral branch of LCX). (RCA right coronary artery; LCA left coronary artery; LAD left anterior descending artery; LCX left circumflex artery; V. cava sup. superior vena cava; V. cava inf. inferior vena cava.) Twenty patients admitted to our hospital for CABG or minimally invasive direct coronary artery bypass (MIDCAB) procedures consented to participate in this study after being given detailed information about the procedure. Fifteen patients were male (75%), 4 patients had insulin-dependent diabetes, and 17 were treated with antihypertensive medication. All patients fulfilled the inclusion criteria defined as a sinus rhythm below 70 beats per minute and unimpaired renal function. All patients underwent SCA within the last 4 weeks before admittance. Computed tomography angiography was performed using a MDCT scanner (Somatom Volume Zoom, Siemens Medical Systems, Forchheim, Germany) and special software for ECG-gated cardiac reconstruction. A 20-mL test bolus of nonionic contrast medium (Ultravist 300, Schering, Berlin, Germany) was given at a flow rate of 3 ml/s through an 18-gauge needle placed in the cubital vein. The ascending aorta was scanned repeatedly to determine the scan delay, which was defined as the time interval between the start of the injection and peak enhancement of the contrast medium in the aorta plus 3 seconds. Helical MDCT scan was started after the determined delay period; 140 ml of contrast medium was needed for full enhancement of the coronary artery tree. The ECG was simultaneously recorded. Patients had to hold their breath for the examination period of approximately 20 to 30 seconds. After the CT scan, the raw data and ECG traces were used to reconstruct axial images at the end of diastole with a temporal resolution of 250 ms per slice. Slice thickness was only 1.25 mm, with increments of 0.5 mm and a resulting total of about 220 slices covering the entire volume of the heart. A special three-dimensional processing workstation (InSight, NeoImagery Technologies, City of Industry, CA) allowed interactive postprocessing of the raw data in a reasonable time frame of only a few minutes per examination. Interpretation of MDCT scan was performed by a radiologist blinded to the results of SCA. Selective coronary angiography was performed in different cardiology centers using the common Judkins technique. Catheterization of the femoral artery was followed by selective intubation of the right and left coronary orifices for the contrast-enhanced twodimensional imaging of the coronary vessels in three biplane projections. All patients included in the study underwent SCA within 4 weeks before hospital admission for the operation. Interpretations of the angiograms were performed independently by cardiologists and a cardiac surgeon who were fully blinded to the results of the MDCT scan. Interpretation of SCA and MDCT findings were therefore performed by examiners fully blinded to results of the corresponding technique. To compare MDCT and SCA findings, a segmental analysis of the coronary vessels was performed as described in a model from the American Heart Association [4] (Fig 1). Due to technical limitations in spatial resolution, only vessels of more than 1.5 mm in diameter were analyzed. Stenosis grading was classified according to Schmermund and colleagues [5]: Grade 1: Normal vessel Grade 2: Stenosis (1% to 49%) Grade 3: Significant stenosis (50% to 74%) Grade 4: High-grade stenosis (74% to 99%) Grade 5: Vessel occlusion. For statistical analysis, SCA findings were taken as standard values and MDCT findings were compared in terms of sensitivity, specificity, and negative predictive value for significant (more than 50%, grade higher than 2) and high-grade (more than 75%, grade higher than 3) stenoses in all examined segments. In addition, densitometry was performed to quantify coronary calcifications. In noncontrast-enhanced, native MDCT scans, calcifications were defined as lesions in the
S1400 CTT SUPPLEMENT TREEDE ET AL Ann Thorac Surg MDCT IN CORONARY SURGERY 2002;74:S1398 402 Table 1. Stenosis Scores in SCA and MDCT MDCT Scores SCA Scores 1 2 3 4 5 Total Number of Segments 1 102 5 0 0 0 107 2 38 32 3 2 1 76 3 5 4 24 2 0 35 4 2 1 2 12 1 18 5 0 0 0 0 3 3 Total Number of Segments 147 42 29 16 5 240 A total of 240 segments in 20 patients were evaluated. Bolded numbers represent similar scores in SCA and MDCT. Score 1 normal vessel, score 2 luminal narrowing (1% 49%), score 3 significant stenosis (50% 74%), score 4 high grade stenosis (75% 99%), score 5 total occlusion. SCA selective coronary angiography; MDCT multidetector computed tomography. coronary artery wall with densities above 130 HU as already described for EBCT [2, 6]. Results A total of 240 coronary segments were analyzed. Of those, 173 (72.1%) segments showed identical scores in blinded SCA and MDCTA analysis. Forty-five segments (18.8%) appeared normal in SCA (grade 1) but showed signs of atherosclerosis in MDCT (grade 2 or higher), whereas five segments (2.1%) seemed to be free of stenoses in MDCT, but showed wall irregularities in SCA (Table 1). One segment of a vessel less than 1.5 mm in diameter showed a grade 3 stenosis in SCA and was underestimated as grade 2 in MDCT. Sensitivity and specificity for the detection of significant stenoses (more than 50%, grade 3 or higher) in MDCT was 92% and 84%, respectively, with a negative predictive value of 89%. Figure 2 shows MDCT images of a 69-year-old symptomatic patient with multiple grade 2 lesions clearly attributed to the mainstem, left anterior descending artery (LAD), and right coronary artery (RCA) showing calcified nodules, a finding that corresponded with wall irregularities in SCA and patchy calcifications. In 6 patients (33%) MDCT and SCA produced exactly the same results in all examined segments. For high-grade stenoses (more than 75%, grade 4 or higher), sensitivity decreased to 79% and specificity to 74%. Negative predictive value for high-grade stenoses was 75%. Sensitivity and specificity decreased because severe coronary calcifications lead to decreased contrast in CT imaging. The exact degree of high-grade coronary Fig 2. Multidetector computed tomography image of a 69-year-old symptomatic patient with multiple calcified nodules in the mainstem, left anterior descending artery, and right coronary artery corresponding to wall irregularities observed on selective coronary angiography. Fig 3. Multidetector computed tomography images of a 63-year-old patient with generalized severe coronary artery disease. Multiple high-grade calcified stenoses are found in the entire left anterior descending artery (A) and the right coronary artery (B).
Ann Thorac Surg CTT SUPPLEMENT TREEDE ET AL 2002;74:S1398 402 MDCT IN CORONARY SURGERY S1401 Fig 4. Three-dimensional volume-rendered multidetector computed tomography (MDCT) image of a 47-year-old male patient with a coronary abnormality: left and right coronary artery arise from a common orifice. The patient underwent MDCT scanning after the right coronary artery could not be found during selective coronary angiography. Comment Recent progress in radiologic imaging techniques such as MDCT have led to excellent-quality resolution and a variety of postprocessing options, which account for their application in the diagnosis of various diseases. Although SCA is still the gold standard in the diagnosis of CAD and in planning CABG surgery, less invasive techniques such as transesophageal echocardiography, EBCT, MRI, and MDCT are gaining more importance. Recent studies have already proved the quality and reliability of MRI and CT scans in evaluating the patency of arterial and venous bypass grafts after surgery [7, 8]. While the number of minimally invasive approaches to cardiac surgery (MIDCAB, robotically enhanced endoscopic CABG) increases, the need for less invasive diagnostic tools increases simultaneously. The newly developed multidetector helical CT scan holds promise to serve as a diagnostic instrument of SCA, with comparable sensitivity and specificity in se- artery stenosis is difficult to evaluate when contrast medium is almost completely covered by circumferential calcifications (Fig 3). Two patients had previously undergone CABG surgery. The MDCT and SCA showed open bypass grafts in 1 patient (left internal mammary artery [LIMA] to LAD, aortocoronary venous bypass to RCA and marginal branch) and occluded vein grafts in another patient (aortocoronary venous bypass to LAD and RCA). Three patients had a total of five coronary artery stents. One stent appeared to be occluded in SCA and also showed no contrast enhancement in MDCT (Fig 4). Four stents with in-stent contrast enhancement appeared to be patent in MDCT. The SCA results supported these findings, but also verified two in-stent stenoses that were not seen in MDCT. Also, coronary anomalies that are difficult to be exposed in SCA can clearly be shown in MDCT scans (Fig 4). In contrast to SCA, densitometry in MDCT allowed differentiation of calcified and noncalcified lesions. Figure 5A shows an MDCT scan of a patient who was admitted to hospital with an SCA diagnosis of high-grade main stem stenosis (Fig 5B). Whereas SCA findings depicted merely the stenosis, MDCT additionally determined the stenosis was a noncalcified low-density lesion corresponding to fatty atheroma. In all examined patients, MDCT exactly verified the intrathoracic anatomy and topographic relation of intrathoracic organs and the chest wall. Patients eligible for MIDCAB procedures were successfully screened for the topographic relation of the LIMA and LAD and for intramyocardial courses of the LAD. The planning of MIDCAB procedures included measurements of the distance between LIMA and LAD (ideally less than 5 cm at the site of the distal anastomosis). Fig 5. (A) Multidetector computed tomography image of a 71-yearold male patient with high-grade symptomatic mainstem stenosis. The scan shows a single noncalcified low-density (corresponding to fat) plaque right before the left anterior descending artery branch, otherwise normal coronary arteries. (B) Coronary angiogram of the same patient. No statement about the origin of the stenosis can be made.
S1402 CTT SUPPLEMENT TREEDE ET AL Ann Thorac Surg MDCT IN CORONARY SURGERY 2002;74:S1398 402 lected patients. To achieve a good quality image, MDCT is technically limited by an effective exposure time of 250 ms, which currently excludes patients with a heart rate of more than 70 beats per minute. Patients eligible for MDCT scans should additionally be able to hold their breath for approximately 30 seconds and should not have implanted pacemakers or defibrillators, again, in order to achieve a good-quality image that is clear enough to be used as a diagnostic tool during CABG surgery. Coronary artery diameters of less than 1.5 mm are likely responsible for false-positive results and should therefore be excluded from evaluation. The lower limit of the diameter of coronary arteries eligible for bypass grafting is usually 1 mm; therefore, a diagnostic gap of 0.5 mm may complicate the planning of CABG anastomoses in small and peripheral coronary arteries. In the present study we found a high sensitivity (92%) and specificity (84%), especially for significant stenoses (more than 50%). Because the same results for significant stenoses were often found in SCA and MDCT, the use of MDCT alone for strategic planning of CABG procedures might be possible in selected patients fulfilling the criteria mentioned above. Sensitivity and specificity for high-grade stenoses (more than 75%) were not as good as for significant stenoses, because differentiating between calcium and intraluminal contrast medium is difficult in patients with severe coronary calcifications. The exact grade of stenosis may be over- or underestimated. Thus, MDCT seems more suitable for detection of significant stenoses. Unlike SCA, MDCT allows statements about the origin of coronary stenoses because the technique can measure the density of exposed lesions. Pixel density above 130 HU indicates calcifications, whereas measurements of less than 130 HU indicate fatty or fibrous lesions. The origin of the stenoses may affect the patient s prognosis. Although the hypothesis has not yet been proved, patients with inhomogeneous artherosclerotic plaques with calcified and noncalcified components susceptible to rupture and thrombotic occlusion are theorized to be at higher risk for myocardial infarction. Patients in early stages of CAD without coronary stenoses that are detectable in SCA could profit from detection of arising fatty or fibrous lesions in the vessel walls; close medical supervision and prophylactic measures could be undertaken. The high negative predicted value of MDCT scans helps to reliably and noninvasively rule out CAD in patients who are asymptomatic or show unclear symptomatic signs. Additionally, MDCT may be useful before sternotomy for patients undergoing redo CABG by clarifying the spatial relation between the graft and sternum. Four of the patients in the study were eligible for single-vessel bypass performed through a small lateral thoracotomy in beating-heart technique (MIDCAB approach). In these patients the intrathoracic situs and spatial relationship between LIMA and LAD was clearly exposed by MDCT scan as an essential condition for preoperative planning of MIDCAB procedures and exclusion of unsuitable patients, as recently published for EBCT [9]. To summarize, MDCT is a promising technique that is of almost comparable quality and predictiveness as SCA in selected patients and for specific indications. It is not yet able to totally replace conventional angiography for diagnosis of CAD and planning of CABG in all patients, but MDCT already has advantages in screening examinations and in assessing early stages of the disease. Future developments including increased gantry speed and improved software solutions will furthermore result in better spatial and temporal resolution, fewer limitations, and higher accuracy for a quality in cardiac imaging that has the potential to replace diagnostic SCA. References 1. Becker CR, Kleffel T, Crispin A, et al. Coronary artery calcium measurement: agreement of multirow detector and electron beam CT. AJR Am J Roentgenol 2001;176:1295 8. 2. Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M Jr, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol 1990;15:827 32. 3. Arad Y, Spadaro LA, Goodman K, et al. Predictive value of electron beam computed tomography of the coronary arteries. 19-month follow-up of 1173 asymptomatic subjects. Circulation 1996;93:1951 3. 4. Austen WG, Edwards JE, Frye RL, et al. A reporting system on patients evaluated for coronary artery disease. Report of the Ad Hoc Committee for Grading of Coronary Artery Disease, Council on Cardiovascular Surgery, American Heart Association. Circulation 1975;51(Suppl):5 40. 5. Schmermund A, Rensing BJ, Sheedy PF, Bell MR, Rumberger JA. Intravenous electron-beam computed tomographic coronary angiography for segmental analysis of coronary artery stenoses. J Am Coll Cardiol 1998;31:1547 54. 6. Achenbach S, Moshage W, Ropers D, Nossen J, Daniel WG. Value of electron-beam computed tomography for the noninvasive detection of high-grade coronary-artery stenoses and occlusions. N Engl J Med 1998;339:1964 71. 7. Brenner P, Wintersperger B, von Smekal A, et al. Detection of coronary artery bypass graft patency by contrast enhanced magnetic resonance angiography. Eur J Cardiothorac Surg 1999;15:389 93. 8. Engelmann MG, von Smekal A, Knez A, et al. Accuracy of spiral computed tomography for identifying arterial and venous coronary graft patency. Am J Cardiol 1997;80:569 74. 9. Gulbins H, Reichenspurner H, Becker C, et al. Preoperative 3D-reconstructions of ultrafast-ct images for the planning of minimally invasive direct coronary artery bypass operation (MIDCAB). Heart Surg Forum 1998;1:111 5.