Jpn J Radiol (2011) 29:342–347 DOI 10.1007/s11604-011-0565-y
Usefulness of multidetector computed tomography coronary venous angiography examination before cardiac resynchronization therapy Selim Doganay · Adem Karaman · Fuat Gündogdu Cihan Duran · Ahmet Yalcin · Mecit Kantarci
Received: October 4, 2010 / Accepted: January 16, 2011 © Japan Radiological Society 2011
Abstract Purpose. Cardiac resynchronization therapy (CRT) is a treatment option for selected heart failure patients. In this study, the aim was to evaluate the usefulness of noninvasive cardiac vein imaging using multidetector computed tomography (MDCT) angiography before CRT. Materials and methods. The MDCT scans of 34 patients (20 men; age range 47–65 years) with a history of cardiac failure were studied for CRT in two centers prospectively. The anatomy of the cardiac venous system, particularly the target veins [left marginal vein (LMV) and posterior vein of the left ventricle (PVLV)], was evaluated with noninvasive MDCT. Result. The coronary sinus, anterior interventricular vein, and posterior interventricular vein were observed in all patients. The PVLV was present in 30 (88.2%) patients. The PVLV was chosen in 30 (88.2%) patients for CRT. If the PVLV had two or more branches, the S. Doganay Department of Radiology, Erciyes University, School of Medicine, Kayseri, Turkey A. Karaman · A. Yalcin · M. Kantarci (*) Department of Radiology, School of Medicine, Atatürk University, 200 Evler Mah. 14. Sok No 5, Dadaskent, Erzurum, Turkey Tel. +90-442-2361212, ext. 1521; Fax +90-442-2361301 e-mail: [email protected]
F. Gündogdu Department of Cardiology, School of Medicine, Atatürk University, Erzurum, Turkey C. Duran Department of Radiology, School of Medicine, Bilim University, stanbul, Turkey
widest branch was chosen for lead implantation. In four (11.7%) patients, the PVLV was absent and the LMV was chosen instead for lead implantation. In one patient (2.9%), partial thrombosis was detected in the coronary sinus with MDCT angiography. Conclusion. MDCT can be used to guide interventionalists for CRT by providing anatomical details of the cardiac venous system rapidly and noninvasively. Key words Coronary veins·MDCT·Cardiac resynchronization therapy
Introduction The coronary venous system (CVS) is increasingly being used for various electrophysiological purposes, including, for example, radiofrequency catheter ablation, mapping, defibrillation, local drug and gene delivery, and cardiac resynchronization therapy (CRT).1,2 In CRT, left ventricular (LV) pacing is achieved by positioning the LV lead in one of the tributaries of the coronary sinus (CS). Transvenous LV lead placement does not succeed in 5%–12% of patients.3 Failure of LV lead placement has been attributed to the inability to insert catheters into the CS and the lack of suitable side branches.3 Previsualization of the venous system of the heart might help an invasive interventionalist recognize some anatomical aspects before CRT implantation, such as the CS ostium (angle and direction), the number of target cardiac veins, and an evaluation of their characteristic features. This knowledge might aid for selecting optimal placement of the LV lead implantation or might even sometimes indicate that the intravenous implantation should not be performed.4 Thus, there is a need for
Jpn J Radiol (2011) 29:342–347
Table 1. Patient characteristics No. of patients
Age (years) Sex (no.) Male Female Heart rate (beats/min)a Ejection fraction (%) Interval between MDCT and CRT (days) Completion time of CRT (minutes) with evaluating MDCT angiography
59 ± 2.5 (47–65) 20 14 72.0 ± 1.5 (65–75) 32 ± 1 (30–34) 7.2 ± 0.8 (6–8) 94 ± 2 (90–100)
Results are the mean ± SD and range unless otherwise noted MDCT, multidetector computed tomography; CRT, cardiac resynchronization therapy a Without premedication
imaging the CVS. The left marginal vein (LMV) or the posterior vein of the left ventricle (PVLV) is often the target vein for pacemaker lead placement for CRT.5 There are only a few methods for visualizing the cardiac venous system, including angiography, echocardiography, electron-beam computed tomography, and, recently, contrast-enhanced multidetector computed tomography (MDCT).1,4 Several studies have demonstrated the capability of cardiac magnetic resonance (CMR) to depict the anatomy of the venous system of the heart in vivo.6,7 MDCT is noninvasive and can provide high-quality images, making this method potentially useful for assessing the coronary venous anatomy.4 The use of MDCT, with a radiation dose as small as 1 mSv due to technological advances, is becoming increasingly widespread.8 In this study, the aim was to evaluate the usefulness of noninvasive cardiac vein imaging using contrast-enhanced MDCT angiography before CRT.
Materials and methods Study population A total of 34 patients (20 men, 14 women) with heart failure who had indications for CRT were prospectively enrolled and underwent contrast-enhanced MDCT. The inclusion criteria were a normal sinus rhythm on electrocardiography (ECG), a QRS complex duration ≥120 ms, ejection fraction (EF) 1.3 mg/
dl), hyperthyroidism, or an inability to suspend breathing for about 25 s. None of the patients had diagnostic cardiac enzyme elevation or ECG findings diagnostic of acute myocardial infarct. Table 1 summarizes characteristics of the patients. The mean age of the patients in this study group was 59 ± 2.5 years (range 47–65 years). The mean heart rate was 72 ± 1.5 beats/min (range 65–75 beats/min), with 24 (70%) of the patients taking β-blockers for heart failure treatment. The institutional ethics committee approved the study, and a signed informed consent was obtained from each patient. MDCT scan protocol The MDCT was performed at 7.2 ± 0.8 days before CRT (range 6–8 days) using two different MDCT scanners (LightSpeed VCT 64, GE Healthcare, Milwaukee, WI, USA; or Aquillon 16, Toshiba Medical Systems, Tokyo, Japan). No extra premedication such as preoxygenation or administration of β-blockers was performed for MDCT. For the first scanner, the following parameters were applied: retrospective ECG gating, 912-channel detectors along the gantry and 64-channel detectors along the z-axis, tube voltage 120 kV, tube current 550–750 mA (depending on the patient’s size), scan field of view (SFOV) 50 cm, gantry rotation 0.35 s/rotation, matrix 512 × 512, slice width 0.625 mm, and range of helical pitch 0.18–0.24. The pitch was selected based on the patient’s heart rate. With the second scanner, the following parameters were used: 16 × 0.5 mm collimation, 1.0 mm slice thickness, and 1.0-mm reconstruction interval; the tube voltage was 120 kV at 400 mA. Nonionic contrast medium (85 ml) (Ultravist 350/ml, Iopromidum; Schering, Berlin, Germany) was injected intravenously at 4.5 ml/s, followed by 40 ml of saline at 2.5 ml/s for both scanners. MDCT was performed 35 s after the intravenous contrast medium was administered.
Jpn J Radiol (2011) 29:342–347
MDCT data reconstruction and analysis Retrospective ECG-gated reconstructions were generated every 10% during 40%–90% of the R-R interval. The cardiac veins were evaluated from the high-quality images acquired during the appropriate R-R interval. Best images were acquired during diastole phase and 60%–80% of the R-R interval. Reconstructed images were then transferred to a processing workstation (Vitrea 2, Vital Images, Minnetonka, MN, USA; Leonardo, Siemens Medical Solutions, Erlangen, Germany) for further analysis with specialized software. Axial source, multiplanar reconstruction, maximum intensity projection, and volume-rendered images were evaluated by two interventionalists and two radiologists by consensus, and the presence or absence of the anterior interventricular vein (AIV), posterior interventricular vein (PIV), and PVLV was recorded. CRT procedure
Fig. 1. Posterior view. Three-dimensional (3D) volume-rendering image shows the posterolateral vein. When it was patent and had the most distal course, the posterior vein of the left ventricle (PVLV) was suggested in 88% of patients for intravenous lead implantation during the cardiac resynchronization therapy (CRT) procedure. CS, coronary sinus
After accessing the left subclavian vein, a guiding catheter was introduced and placed into the ostium of the coronary sinus. Venography of the coronary veins was then performed to establish their anatomy for occlusion with a Berman catheter. The recommended LV lead position is the lateral or posterolateral wall, midway between the base and the apex.2 We chose the PVLV for the intravenous lead implantation procedure for CRT. If the PVLV had two or more branches, we chose the widest branch for lead implantation site via MDCT. For maximum efficiency, the LV lead was placed as distally as possible. If the PVLV was absent, the LMV was chosen instead. Statistical analysis Statistical analysis was performed using SPSS software (version 16.0; SPSS, Chicago, IL, USA). Continuous variables are presented as the mean value ± standard deviation. Categorical variables are presented as absolute number (percentage). The unpaired Student’s t-test was used to determine the significance of the differences in the completion time of CRT with or without evaluating MDCT venous angiography prior to procedure. P < 0.05 was considered statistically significant.
Results No patients were excluded from the study due to suboptimal study quality. The CS, AIV, and PIV were observed on MDCT images in all patients. The PVLV
Fig. 2. Posterior view. 3D volume-rendering image shows the left marginal vein (LMV). The LMV was suggested in two patients who had no posterior veins of the left ventricle for intravenous lead implantation. CS, coronary sinus
was present in 30 (88.2%) patients (Fig. 1). In 8 (23.5%) patients, the PVLV had three side branches; another 8 (23.5%) patients’ PVLVs had two side branches; and the remaining 14 (41.1%) patients had PVLVs with one side branch. In this study, we chose the PVLV in 30 (88.2 %) patients for the intravenous lead implantation procedure for CRT. In four (11.7%) patients, the PVLV was absent. In these four patients (11.7%), the LMV was chosen instead (Fig. 2). In one patient (2.9%) with a partial thrombosis, which was identified as a luminal
Jpn J Radiol (2011) 29:342–347
Fig. 3. a, b Posterior view. 3D volume-rendering (a) and coronal oblique maximum intensity projection (MIP) (b) images show thrombosis (T, arrows) in the coronary sinus (CS)
filling defect, more than 50% in all reconstructed images in the appropriate phase was detected in the CS with MDCT angiography (Fig. 3). With prior knowledge of the venous anatomy revealed by MDCT, interventionalists dilated the lumen with balloon angioplasty, which was followed by lead implantation to the PVLV. The time required for CRT was 94 ± 2 min (range 90–100 min) in the current study population with MDCT, which was significantly shorter than that for the routine CRT procedure without MDCT in our clinic (120 ± 15 minutes, range 105–145 minutes, n = 42; P = 0.01). Reduced symptoms of systolic heart failure (e.g., shortness of breath, dry cough), as well as improved exercise tolerance and LV ejection fraction, were noted in all patients. No complications associated with the CRT were observed during the follow-up time of 3 months after the CRT procedure in 34 patients. Cardiac resynchronization therapy acutely improved the hemodynamics in all patients with systolic heart failure in the study. It has been shown to reduce symptoms (e.g., shortness of breath, dry cough), increase exercise tolerance, and improve the ejection fraction in all patients.
Discussion Cardiac resynchronization therapy has become an attractive treatment option for highly symptomatic heart failure patients with poor LV systolic function.9 Numerous clinical investigations have demonstrated that, in selected patients, CRT significantly improves the cardiac output, systolic pressure, maximum rate of pressure rise, magnitude of wall contraction, mitral regurgitation, and left atrial pressure.10–12
With CRT, LV pacing is achieved by positioning the LV lead in one of the tributaries of the CS. The main source of potential difficulties during CRT lead implantation using the coronary venous system is the anatomical variability of the CS ostium and target veins, as well as ventricular remodeling that may influence the shape and size of the heart.4 Identifying patients who do not have suitable coronary veins for CRT lead implantation is important. These patients may be best treated by an epicardial pacemaker lead implantation that requires a minimal surgical approach.13,14 Major components of the cardiac venous system include the coronary sinus, the great cardiac vein, and the anterior and posterior interventricular veins. The great cardiac vein courses alongside the circumflex artery and subsequently drains into the coronary sinus.5 The great cardiac vein receives two main branches: the LMV, which courses along the lateral border of the left ventricle; and the posterior vein or posterolateral vein.5 Until recently, the cardiac venous system could only be evaluated invasively using retrograde venography, either by direct manual contrast injection or after occlusion of the CS.9,15,16 In most patients, cardiac vein anatomy can be visualized and venous accessibility determined, but these can only be assessed in a twodimensional projection. Reported disadvantages of this method are incomplete direct vein visualization, complicated localization of the CS ostium, excessive use of contrast material (up to 500 ml in one study), balloonrelated trauma to the coronary sinus, and an inability to visualize the venous anatomy in relation to the ventricular wall.17,18 Several studies have demonstrated that the anatomy of the coronary venous system can be noninvasively
evaluated by CMR as well.6,7 CMR does not require administration of iodinated contrast medium or exposure of ionizing radiation to the patients. However, CMR cannot be used for patients with pacemakers or intracardiac defibrillators or those who had other contraindications for MRI.6,7,13 In recent years, MDCT has become the tool for imaging coronary arteries. It can also disclose the coronary venous system anatomy noninvasively. For this reason, MDCT coronary venous angiography is used to demonstrate the vascular territory. MDCT may also identify patients who do not have adequate venous side branches available or those who have obstruction and ostial narrowing.13 In some patients with systolic heart failure, right ventricular pacing is not effective and biventricular pacing is required.10 For providing optimal LV lead placement for successful CRT, the vein or tributary that corresponds to the site of the latest LV myocardial activation should be chosen.13 The posterolateral LV wall is usually the optimal location for pacing.2,19 LMVs are preferred less than PVLVs because the left phrenic nerve passes at a distance of < 3 mm from the LMV in 43% of cadaveric hearts. Left phrenic nerve stimulation after CRT is a well-recognized complication.5 MDCT studies have shown that the posterior veins of the LV and LMV were absent in 5% and 39% of patients, respectively, and variations in venous drainage patterns largely agree with the results of anatomical studies.13,20 When no PVLV was identified on MDCT, we chose the LMV for intravenous lead implantation. Venous anatomy demonstrated by MDCT seems to be effective for reducing the risk of complication in such patients. MDCT was also useful for detecting a partial thrombosis in the CS. Based on the information obtained by MDCT, interventionalists successfully performed lead implantation in the PVLV after balloon angioplasty. The MDCT was performed using two scanners (16and 64-channel detectors). However, no substantial difference in image quality was noted between them. In comparison to coronary arteries, the diameter of the cardiac vein is much larger. Several prior studies demonstrated the feasibility of evaluating the cardiac venous system prior to CRT using 4-, 8-, or 16-detector MDCT.1,2,17 The most important point in the MDCT screening protocol is the timing of contrast medium administration. Abbara et al. suggested that MDCT acquisition after contrast administration should be delayed for appropriate imaging of the cardiac venous system.17 In their study with 16-dedector MDCT, the imaging sessions started at a mean time of 19.4 s after intravenous contrast medium administration. Tada et al.,1 in a study
Jpn J Radiol (2011) 29:342–347
with 8-detector MDCT, the imaging sessions started at a mean time of 25 s after intravenous contrast medium administration. At heart rate of 65 beats/min, optimal visualization of coronary arteries and veins can be obtained when imaging session starts at 17 s and 20–22 s after contrast injection, respectively, in our hospital. However, in patients with cardiac failure, as the cardiac output decreases maximum enhancement in the coronary veins is observed during the later phase. In our current study population, all subjects exhibited systolic dysfunction with an ejection fraction of