Faculty Bibliography

Introduction: Patients receiving radiofrequency ablation (RFA) for the treatment of atrial fibrillation frequently undergo pre-procedural cardiac CT for evaluation of the left atrium and pulmonary veins. Left atrial outpouchings (LAO), including diverticula and accessory appendages, can be mistaken for an ostium of a pulmonary vein, which are important to identify as there is a potential risk of complications during RFA. The prevalence of these outpouchings has been described to be as high as 27 percent in the population of patients undergoing routine cardiac CT.1 The purpose of this study is to describe the prevalence, morphology, and size of LAO in patients undergoing RFA for treatment of atrial fibrillation. Methods: Fifty consecutive patients referred for RFA were identified from our registry of patients undergoing gated cardiac CT. Data was independently analyzed by two blinded readers for LAO. Images were evaluated using multiplanar reformatted and 3D reconstruction. The presence of LAO was defined as any abnormality that had a discernable ostium stemming from the left atrial wall. The number and size of LAO were recorded. Comparison of prevalence was evaluated using the Fisher's exact test. Results: There were a total of 29 LAO found in 24 of the 50 patients for a calculated prevalence of 48% (95% CI: 33.6 to 62.6). The prevalence in our population was significantly higher than reported in the general cohort of patients undergoing routine cardiac CT (p=0.003). The average size (length, width, and depth) of the LAO were 0.54 +/- 0.28 by 0.39 +/- 0.20 by 0.56 +/- 0.26 cm. Conclusions: Patients undergoing RFA for atrial fibrillation have a high prevalence of L

Introduction: Patients receiving radiofrequency ablation (RFA) for treatment of atrial fibrillation typically undergo pre-procedural cardiac computed tomography angiography (CCTA) to delineate pulmonary venous anatomy and transesophageal echocardiogram (TEE) to exclude left atrial and/or left atrial appendage thrombus (LAT). The addition of a late phase acquisition is theorized to aid CCTA identification and discrimination of LAT from slow left atrial appendage filling. The purpose of this study is to evaluate the diagnostic accuracy of dual-phase, ECG-gated dual-source CCTA (64-slice Definition, Siemens) compared to TEE for identification of thrombus and to assess the added value of a late phase CCTA acquisition. Methods: Fifty-three consecutive patients (37 men; mean age 63) had both dual-phase CCTA and TEE prior to RFA. Mean time between CCTA and TEE was 9 days (range 1-22). Mean early phase and late phase scan acquisition delay times were 29 sec and 30 sec, respectively. Presence of LAT was independently graded on both early phase and combined early and late phase (CP) CCTA acquisitions using a 5-point Likert scale by 2 readers blinded to the TEE results. Diagnostic accuracy for LAT was assessed for early phase and CP CCTA acquisitions using TEE results as truth. Results: CCTA identified LAT in 2 out of 3 patients with thrombi on TEE (67%). Relative to TEE, early phase and CP CCTA acquisitions demonstrated: 47% and 67% sensitivity, 84% and 100% specificity, 54% and 100% PPV, 80% and 98% NPV, respectively. Overall diagnostic accuracy was significantly improved for CP compared to early phase acquisition (98% and 77%, respectively, p<0.001). Conclusions: CCTA has excellent specificity (100%) but only modest sensitivity (66.7%) for identification of LAT in patients undergoing RFA. Addition of a late phase CCTA acquisition significantly improves overall diagnostic accuracy

OBJECTIVE: Published cardiac CT angiography (CTA) reference measurements for the cardiac chambers, aorta, and pulmonary artery (PA) are incomplete and compromised by study population, coronary artery disease (CAD), or its risk factors. The purpose of our study was to establish sex-specific normalized ranges of cardiac chamber size, wall thickness, ejection fraction (EF), and aorta and PA diameter on cardiac CTA in a population without CAD or its risk factors. MATERIALS AND METHODS: Seventy-six patients (38 men and 38 women) without known diabetes; hypertension; smoking history; or evidence of structural heart, vascular, or coronary artery diseases underwent 64-MDCTA. Obtained left atrial (LA) size, left ventricular (LV) volumes, LV wall thickness, thoracic aorta, and PA diameter measurements were normalized to body surface area (BSA). RESULTS: There were statistically significant differences noted between men and women for all measured left-sided heart and great vessel measurements. After normalization to BSA, only chamber dimensions and ascending aorta and left PA sizes remained significantly different. Selected normalized measurements for men versus women, respectively, include LA area, 10.6 +/- 2.1 versus 12.3 +/- 2.1 cm(2)/m(2); LV end-diastolic size, 72.4 +/- 15.1 versus 60.9 +/- 13.3 mL/m(2); EF, 67% +/- 7% versus 72% +/- 8%; aortic sinus, 1.6 +/- 0.2 versus 1.7 +/- 0.2 cm/m(2); ascending aorta, 1.4 +/- 0.2 versus 1.6 +/- 0.2 cm/m(2); descending aorta, 1.1 +/- 0.1 versus 1.2 +/- 0.1 cm/m2; main PA, 1.3 +/- 0.1 versus 1.4 +/- 0.1 cm/m(2); right PA, 1.1 +/- 0.1 versus 1.1 +/- 0.2 cm/m(2); and left PA, 1.0 +/- 0.1 versus 1.1 +/- 0.1 cm/m(2). CONCLUSION: Cardiac CTA measurements of the left cardiac chambers, thoracic aorta, and pulmonary arteries were established for a population without CAD or its risk factors

Accurate interpretation of cardiac computed tomography requires fundamental knowledge of the normal cardiac anatomy and its common variations. This article reviews the normal anatomy of the coronary arteries, cardiac chambers, and cardiac valves

Cardiac computed tomographic angiography (CTA) using multidetector computed tomographic scanners has proven to be a reliable technique to image the coronary vessels. CTA also provides excellent visualization of the mitral and aortic valves, and yields useful information regarding valve anatomy and function. Accordingly, an assessment of the valves should be performed whenever possible during CTA interpretation. In this paper, we highlight the imaging features of common functional and structural left-sided valvular disorders that can be seen on CTA examinations

BACKGROUND: Dual-source computed tomography (DSCT) provides diagnostic quality images of the coronary arteries over a wide range of heart rates (HRs). Current dose reduction techniques, including electrocardiographic (ECG) dose modulation and prospective triggering, are optimized for use in patients with relatively slow (<70 beats/min) HRs by limiting radiation dose to the ideal phases of image acquisition. OBJECTIVE: We evaluated coronary vessel image quality (IQ) at different reconstruction phases in patients with fast HRs (>80 beats/min) to assess potential feasibility of prospective triggering techniques on DSCT. METHODS: Patients (n=101) underwent 64-slice DSCT with retrospective ECG-gating without beta-blocker premedication. Image reconstructions were performed at 10% R-R wave phase intervals (0%-90%). Patients were grouped by mean HR: group A, <60 beats/min (n=22); group B, 60-80 beats/min (n=57); group C, >80 beats/min (n=22). Coronary artery IQ was assessed by 2 readers in consensus on a 5-point scale. RESULTS: Optimal IQ occurred at 70% phase for all arteries in groups A and B. In group C, optimal IQ occurred at 30% and 40% phases. The 70% phase achieved diagnostic IQ in 97% of group A and 86% of group B. A widened reconstruction window (30%-50%) was necessary for diagnostic IQ in a similar high proportion (84%) of group C. CONCLUSION: Optimal IQ occurs during late-systolic phases for patients with fast HRs (>80 beats/min). Late-systolic phase prospective triggering is potentially feasible in these patients; however, given the widened reconstruction windows required, a higher radiation dose may be required compared with patients with slower HRs (<80 beats/min)

OBJECTIVE: Dual-source CT improves temporal resolution, and theoretically improves the diagnostic image quality of coronary artery examinations without requiring preexamination beta-blockade. The purpose of our study was to show the improved diagnostic image quality of dual-source CT compared with single-source CT despite the absence of preexamination beta-blockade in the dual-source CT group. MATERIALS AND METHODS: We performed a retrospective analysis of consecutive patients who underwent coronary artery evaluation with either single-source CT or dual-source CT at our institution between February 2005 and October 2006. Examination reports were analyzed for the presence of image artifacts, and image quality was graded on a 3-point scale (no, mild, or severe artifact). Type of artifact (motion, calcium, quantum mottle) was also noted. RESULTS: Examinations (339 single-source CT and 126 dual-source CT) of 465 patients were analyzed. Artifact was reported in 39.8% of examinations using single-source CT and in 29.4% of examinations using dual-source CT (p < 0.05). The number of examinations with motion artifact was significantly higher with single-source CT than with dual-source CT (15.9% vs 4.8%; p < 0.001) despite significantly higher heart rates in the dual-source CT group (59.4 +/- 8.4 vs 68.6 +/- 14.6 beats per minute; p < 0.001). No patients in the dual-source CT group received preexamination beta-blockade compared with 81% of patients in the single-source CT group. The presence of severe (nondiagnostic) calcium artifact was also significantly reduced in the dual-source CT group (13.0% vs 3.2%; p < 0.001). CONCLUSION: Dual-source CT provides significantly better diagnostic image quality than single-source CT despite higher heart rates in the dual-source CT group. These findings support the use of dual-source CT for coronary artery imaging without the need for preexamination beta-blockade