Recently contrast-enhanced cardiac computed tomography (CT) was found to be useful for imaging the left atrium and pulmonary veins (PVs) before radiofrequency catheter ablation in patients with atrial fibrillation. However, the risks of contrast agent in patients with impaired renal function must be considered. We investigated the accuracy of low-dose electrocardiographically synchronized nonenhanced cardiac CT (NECT) for identifying PV anatomy. One hundred eight consecutive patients who underwent cardiac CT before radiofrequency catheter ablation of atrial fibrillation were included. Nonenhanced cardiac computed tomogram was retrospectively evaluated for each patient by 2 radiologists for the following PV anatomy: conventional pattern, conjoined ostium, and accessory PVs with number and location. Sensitivity and specificity for variations in PVs were calculated using contrast-enhanced cardiac computed tomogram as the reference standard. Detection rates for each variation were also calculated. Twenty-one right PV (RPV) variations and 11 left PV (LPV) variations were observed. NECT showed a high diagnostic performance in detecting variations in PVs for the 2 observers. For RPV variations overall sensitivity was 97.6% and specificity was 96.6%. For LPV variations overall sensitivity was 90.9% and specificity was 97.9%. Overall detection rates for variation between the 2 observers were 97.1% for accessory RPV from the right middle lobe, 100% for 4 ostia with accessory RPV from the right middle lobe and accessory RPV from the superior segment of the right lower lobe, 100% for accessory RPV from the superior segment of the right lower lobe, 88.9% for conjoined ostium of the LPV, and 100% for accessory LPV from the left lingular segment. In conclusion, variations in PV anatomy were detected with great accuracy by NECT.
Radiofrequency catheter ablation (RFA) of the posterior left atrium and distal pulmonary vein (PV) is increasingly being used by cardiac interventional electrophysiologists to manage patients who have recurrent or refractory atrial fibrillation. The success of RFA is highly dependent on a preprocedural understanding of the complex 3-dimensional anatomy of the posterior left atrium and distal PVs. Previous studies have reported that 3-dimensional contrast-enhanced magnetic resonance angiography provides a good presentation of left atrial and pulmonary venous anatomy. However, many patients with atrial fibrillation have pacemakers or defibrillators cannot undergo magnetic resonance angiography. Recently a 3-dimensional multidetector computed tomographic technique was found to be useful for imaging the left atrium and PVs before RFA in patients with atrial fibrillation who had pacemakers or defibrillators. In contrast, magnetic resonance angiography might be useful for patients with contraindications to intravenously administered iodinated contrast material. However, for multidetector computed tomography (CT) and magnetic resonance angiography, the risks of contrast agent injection in patients with impaired renal function must be considered. The course of the PVs might be distinct from the course of the pulmonary arteries and bronchi, so we believed we could easily detect PVs by nonenhanced cardiac CT (NECT). However, no reports have assessed imaging the left atrium and distal PVs by NECT. Therefore, the purpose of this study was to evaluate the accuracy of NECT in identifying variations in pulmonary venous anatomy through the coronary calcium scoring scan (low-dose electrocardiographically synchronized NECT).
Methods
Institutional review board approval was obtained, and informed patient consent was waived for this retrospective review. We searched our database for radiologic examinations performed and reported at our institution from January 2009 through December 2009. In total 108 consecutive patients underwent cardiac CT before RFA for atrial fibrillation during this period and all were included (85 men and 23 women, mean age 56.7 ± 11.0 years). No specific exclusion criterion was used in this retrospective study.
Drug therapy was adjusted for maximal rate control in patients with persistent atrial fibrillation before CT. If a patient had paroxysmal atrial fibrillation with normal sinus rhythm or had atrial fibrillation that was stable, electrocardiographically gated CT was performed. Cardiac CT at our institution consisted of NECT for coronary calcium scoring and contrast-enhanced computed tomographic angiography. All computed tomographic scans were obtained using 64-slice multidetector computed tomography (Somatom Sensation 64, Siemens Medical Solution, Forchheim, Germany). For coronary calcium scoring, NECT was performed with prospective electrocardiographically triggered acquisitions in mid-diastole (70% of RR interval) using 100 to 120 kVp with 150 to 200 mA depending on patient size, 240-ms exposure time per rotation, 330-ms gantry rotation time, and 64- × 1.5-mm slice collimation. Nonenhanced computed tomogram was reconstructed at 70% of the RR interval using a slice thickness of 3 mm and an increment of 3 mm. Each scan was performed from the lower neck to the diaphragm and field of view was adjusted according to the thorax. Next, contrast-enhanced computed tomographic angiography was performed. A bolus of iopamidol 60 to 80 ml (Iopamiro 370, Bracco, Italy) was injected into an antecubital vein at a flow rate of 5 ml/s, followed by a 50-ml saline chasing bolus at 5 ml/s. Start delay was defined by bolus tracking in the ascending aorta, and scan start was automatically initiated 5 seconds after reaching a threshold of 140 HU. Scanning used the following parameters: retrospective electrocardiographically gated acquisitions, 100 to 120 kV, 600 to 800 mA, 64- × 0.6-mm slice collimation, 330-ms gantry rotation time, and a table feed speed of 18 mm/s. The scan was performed from the tracheal bifurcation to the diaphragm and field of view was adjusted according to the size of the heart. Contrast-enhanced computed tomographic angiogram was reconstructed using a slice thickness of 0.75 mm and an increment interval of 0.5 mm.
Nonenhanced computed tomogram and computed tomographic angiogram were reviewed on an off-line workstation (AquarisNet Viewer 1.8.0.3, TeraRecon, San Francisco, California). Normal pulmonary venous anatomy was defined as the presence of any combination of nonanomalous PVs including conventional anatomy and conjoined or accessory PVs. Conventional pulmonary venous anatomy was defined as the presence of single right and left superior and inferior PVs that drained into the left atrium with no accessory veins. Variations in pulmonary venous anatomy included conjoined or accessory PVs. Conjoined veins were defined as superior and inferior veins combined proximal to the left atrium resulting in only 1 atriopulmonary venous junction on the involved side. Accessory PVs were defined as extra veins with independent atriopulmonary venous junctions separated from the superior and inferior PVs.
A radiologist with 10 years of experience in cardiac multidetector CT evaluated the posterior left atrium and PVs on computed tomographic angiogram to diagnose normal or anomalous pulmonary venous anatomy. For normal pulmonary venous anatomy, the pattern of the PV was determined for the right PV (RPV) and left PV (LPV) and conventional PVs or variations in the PVs (conjoined PVs or accessory PVs). After 1 month 2 radiologists (with 2 and 4 years of experience in cardiac multidetector CT, respectively) blinded to all computed tomographic angiographic data independently reviewed the posterior left atrium and PVs with NECT. Image datasets were analyzed with using multiplanar reformatted images (coronal and sagittal) and thin-slab maximum intensity projection images in addition to axial images. For accessory PVs the number and location of PVs were recorded. Image quality of nonenhanced cardiac computed tomogram for the posterior left atrium and PVs was classified as good (no or minor artifact, good diagnostic quality), fair (moderate artifacts, acceptable for diagnosis), or poor (severe artifact impairing accurate evaluation). Presence of irregular heart rhythm on electrocardiogram during acquisition of cardiac computed tomogram was evaluated.
Using computed tomographic angiography as a standard reference, detection rates and diagnostic indexes including sensitivity, specificity, and accuracy of NECT for the presence of variations in PVs for the RPV and LPV were calculated.
Results
Demographic data are presented in Table 1 . Normal range of serum creatinine was 0.50 to 1.40 mg/dl and 7 patients (6.5%) showed >1.4 mg/dl of serum creatinine before cardiac CT. Mean value of estimated glomerular filtration rate was 75.4 ± 13.0 ml/min/1.73 m 2 (Modification of Diet in Renal Disease formula). There were 4 patients with estimated glomerular filtration rate <50 ml/min/1.73 m 2 and 6 patients with 50 to 60 ml/min/1.73 m 2 . Mean heart rate during cardiac CT was 64.3 ± 10.9 beats/min and irregular heart rhythm during acquisition of cardiac computed tomogram was present in 48.1% (52 of 108). Image quality of nonenhanced cardiac computed tomogram was good in 61.1% (66 of 108) and fair in 38.9% (42 of 108), with no images with poor quality. Irregular heart rhythm was noted in 1/2 the patients but did not seriously affect the image quality of nonenhanced cardiac computed tomogram.
| Age (years) | 56.7 ± 11.0 |
| Men | 85/108 (78.7%) |
| Body mass index (kg/m 2 ) | 26.1 ± 3.0 |
| Heart rate (beats/min) | 64.3 ± 10.9 |
| Arrhythmia | 52/108 (41.4%) |
| Radiation dose (mSv) | |
| Nonenhanced cardiac computed tomography | 1.29 ± 0.22 |
| Contrast-enhanced computed tomography angiography | 8.45 ± 3.12 |
| Serum creatinine (mg/dl) | 1.1 ± 0.2 |
| Estimated glomerular filtration rate (ml/min/1.73 m 2 ) ⁎ | 75.4 ± 13.0 |
On computed tomographic angiogram, no anomalous pulmonary venous return (partial or total) was observed. All patients showed normal pulmonary venous drainage into the left atrium, 28 patients (25.9%, 28 of 108) had variations in PVs (for RPV and/or LPV) and 21 (19.4%, 21 of 108) presented with variations in the RPV. All variations in the RPV were accessory PVs and the most common was separate drainage of the right middle lobe (81.0%, 17 of 21). Accessory PVs from the superior segment of the right lower lobe were found in 2 patients (9.5%, 2 of 21). Combined accessory PVs, 1 from the right middle lobe and the other from the superior segment of the right lower lobe, were found in 2 patients (9.5%, 2 of 21). No conjoined PVs for the RPV were seen. Variations in the LPV were observed in 11 patients (10.2%, 11 of 108) with the most common being conjoined PV (81.8%, 9 of 11). We observed 2 cases of an accessory PV from the lingular segment of the left upper lobe for the LPV (18.2%, 2 of 11). Variations in the RPV and LPV were noted in 4 patients (3.7%, 4 of 108), all of whom had right middle lobe PV for the RPV and conjoined PV for the LPV.
NECT achieved high diagnostic performance in detecting variations in PVs ( Table 2 ). For the 2 observers, RPV sensitivity was 100% to 95.2%, specificity was 96.6% to 96.6%, and accuracy was 97.2% to 96.3%. For LPV sensitivity was 90.9% to 90.9%, specificity was 97.9% to 97.9%, and accuracy was 97.2% to 97.2%. On a patient basis (with any variation in RPV or LPV belonging to the variation-patient group), we also observed a high degree of diagnostic performance with NECT for detecting variations in pulmonary venous anatomy ( Table 2 ).
| Observers | RPV | LPV | Patient Basis |
|---|---|---|---|
| 1 | |||
| Sensitivity | 100% (21/21) | 90.9% (10/11) | 100% (28/28) |
| Specificity | 96.6% (84/87) | 97.9% (95/97) | 93.8% (75/80) |
| Accuracy | 97.2% (105/108) | 97.2% (105/108) | 95.4% (103/108) |
| 2 | |||
| Sensitivity | 95.2% (20/21) | 90.9% (10/11) | 96.4% (27/28) |
| Specificity | 96.6% (84/87) | 97.9% (95/97) | 93.8% (75/80) |
| Accuracy | 96.3% (104/108) | 97.2% (105/108) | 94.4% (102/108) |
| Overall | |||
| Sensitivity | 97.6% (41/42) | 90.9% (20/22) | 98.2% (55/56) |
| Specificity | 96.6% (168/174) | 97.9% (190/194) | 93.8% (150/160) |
| Accuracy | 96.8% (209/216) | 97.2% (210/216) | 94.9% (205/216) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
