Summary
Background
Multidetector computed tomography (MDCT) provides a non-invasive anatomic description of the coronary veins that may be useful in patients candidates to cardiac resynchronization. Prospective gating reduces radiation exposure but its impact on image quality is unknown is this setting.
Aims
This study compared image quality and reliability of MDCT angiography of the coronary veins between prospective and retrospective gating.
Methods
Seven anaesthetized pigs underwent 64-detector row MDCT with prospective and retrospective ECG-gating. MDCT scans were evaluated for visibility of the veins, estimated radiation dose and vein characteristics. Inter- and intra-observer reproducibility was calculated.
Results
Visibility grades of all veins were significantly decreased in prospective (0.82 ± 0.6) compared to retrospective gating (1.68 ± 0.9; P < 0.001), the lateral vein being missed in two cases when using prospective vs. retrospective gating. The maximal vein length was significantly increased when using retrospective gating ( P = 0.015). Inter-observer but not intra-observer reproducibility was dependent on the gating technique for the maximal length and contrast-to-noise ratio ( P = 0.003 for both). Heart rate was 82 ± 13 bpm and 86 ± 11 bpm during retrospective and prospective ECG-gating ( P = ns) despite full dose of atenolol titration.
Conclusion
Retrospective gating seems to be superior to prospective gating MDCT to describe the coronary venous system but the conclusions of our study should be confined to high heart rate condition.
Résumé
Contexte
Le scanner multidétecteurs permet une description anatomique non invasive des veines coronaires qui peut être utile chez les patients candidats à une resynchronisation cardiaque. L’acquisition en mode prospectif réduit l’irradiation mais son impact sur la qualité des images est inconnue dans ce cas précis.
Objectif
Cette étude compare la qualité des images et la précision de l’imagerie des veines coronaires en scanner multidétecteurs entre le mode d’acquisition prospectif et rétrospectif.
Méthodes
Sept cochons anesthésiés ont eu un scanner 64 détecteurs en mode prospectif et rétrospectif couplé à l’ECG. Les images scannographiques ont été évaluées sur la visibilité des veines, la dose de rayons reçue estimée et les caractéristiques des veines. La reproductibilité intra et inter-observateur a été calculée.
Résultats
Les grades de visibilité de l’ensemble des veines coronaires ont été significativement diminués en mode prospectif (0,82 ± 0,6) en comparaison au mode rétrospectif (1,68 ± 0,9 ; p < 0,001), la veine latérale n’ayant pas été visualisée dans deux cas si nous comparons les acquisitions prospectives et rétrospectives. La longueur maximale des veines était significativement plus élevée lors des acquisitions en mode rétrospectif ( p = 0,015). La reproductibilité inter-observateur mais pas celle intra-observateur était dépendante du mode d’acquisition pour la longueur maximale et le rapport contraste sur bruit ( p = 0,003 pour les deux). La fréquence cardiaque était de 82 ± 13 bpm et 86 ± 11 bpm durant les acquisitions en mode rétrospectif et prospectif ( p = ns) malgré une titration à doses pleines d’aténolol.
Conclusion
L’acquisition en mode rétrospectif du scanner multidétecteur semble être supérieure à celle en mode prospectif pour décrire le système veineux coronaire mais les conclusions de notre étude restent confinées à des fréquences cardiaques élevées.
Background
The coronary venous system is increasingly being used with accession of percutaneous cardiac therapeutics, including percutaneous mitral annuloplasty , transvenous cellular cardiomyoplasty , radiofrequency catheter ablation , mapping and defibrillation . Cardiac resynchronization therapy is now a widely used procedure in patients with end stage heart failure. This technique requires the optimal placement of an electrode in a left ventricular coronary vein close to viable myocardial tissue. Since a considerable morphological variability of the coronary veins exists in humans , prior anatomic assessment of these vessels is a critical issue to ensure that therapeutic procedures are achieved.
Recently, multidetector computed tomography (MDCT) emerged as a non-invasive and accurate imaging technique for anatomic description of coronary veins by specifying their number, trajectory, diameter and possible blockages . Despite the need for a higher radiation dose to the patient, retrospective electrocardiographic (ECG) gating remains widely used for assessing coronary artery atherosclerotic lesions due to a better coronary plaque assessment. However, we hypothesized that prospective gating could be sufficient for the description of cardiac venous anatomy, offering the opportunity to reduce radiation exposure without significant loss of detail in the anatomical description of the venous system.
Thus, the purpose of the present study was to compare radiation dose and image quality obtained from coronary venous MDCT angiography between prospective and retrospective gating in anaesthetized pigs.
Methods
Animals and anaesthesia
All experimental studies were approved by our local and national Ethics Committee and conformed to the European Communities Council Directives guidelines (86/609/EEC) for the care and use of laboratory animals. Experiments were performed in seven male Landrace breed conventional pigs with body weights ranging from 50 to 60 kg. The pigs were delivered at least 24 h prior to the day of study and housed in individual cages. All solid foods were withdrawn during the 12 hours before experiments but the animals had access to water ad libitum. Prior to the MDCT procedure, the pigs were sedated by intramuscular injection of ketamine (15 mg/kg) and xylazine (1 mg/kg). Anaesthesia was induced by 2.5% sevoflurane and intravenous propofol (110 μg/kg per minute) via a 20-G venous access placed in an ear vein. Endotracheal intubation was then performed and anaesthesia was maintained with sevoflurane (0.8%) and 66% nitrous oxide in oxygen plus propofol (50–110 μg/kg per hour) along with a continuous intravenous perfusion of atracurium (0.75 mg/kg per hour) to achieve adequate muscular relaxation. Intravenous bupremorphine (0.3 mg/kg) was used for additional analgesia. The pigs were placed on intermittent positive pressure ventilation (Aestiva 5/MRI 7900, GE Healthcare, France) at 15 breaths/min and the respiratory volume (8–12 mL/kg) adjusted so as to maintain normocapnia. Catheters were placed in the carotid artery and jugular vein to allow for the continuous measurement of arterial pressure and administration of drugs and contrast agent (see below), respectively. Thereafter, the pigs were transferred to the scanner for investigation. Before the imaging protocol, and if tachycardia was present, intravenous atenolol (1 mg/min in titration) was administered in order to maintain heart rate below 70 bpm.
At the end of the experiments, sevoflurane was increased to 4% for 15 minutes and a super-saturated solution of potassium chloride (15 mL, i.v.) was injected rapidly to induce cardiac arrest.
Multidetector computed tomography
All examinations were performed at Cyceron research center using a 64-detector MDCT scanner (Lightspeed VCT RX, GE Healthcare, Milwaukee, WI) during a single end-expiratory breath-hold, in the supine position. Each pig underwent a MDCT angiography of the cardiac venous system with both retrospective and prospective ECG-gating (rotation time: 350 ms; detector collimation: 64 × 0.625 mm; tube voltage: 120 to 140 kV at a current of 500 to 600 mA depending on animal weight; craniocaudal direction). For prospectively gated MDCT scans, a commercially available protocol was used (Snapshot Pulse, GE Healthcare, Milwaukee, WI), centred at 65% of the R-R interval with a padding of 125 ms. In each experiment, tube voltage and current were set at similar values for both prospective and retrospective ECG-gated acquisitions. Retrospective and prospective ECG-gated MDCT were performed in a random order. A dual-head power injector (Medrad, Warrendale, PA) was used to provide a three-phase bolus at a rate of 5 mL per second: first, 70 mL of iomeprol 350 mg/mL (Iomeron 350, Bracco Imaging) was administered. Thereafter, 50 mL of a 50%/50% blend of iomeprol and saline was administered. Finally, 20 mL of saline was administered. The MDCT scan acquisition was triggered 7 to 11 seconds (according to heart rate) after the contrast bolus reached the ascending aorta.
Terminology
The terminology used to describe the various tributary veins is based on previous studies . Anatomical studies of the pig heart showed a coronary venous system close to that of humans with the exception of a prominent left azygous vein draining into the coronary sinus .
Anterior interventricular vein
The anterior interventricular vein (AIV) originates in the lower or middle third of the anterior interventricular groove. It follows this groove then turns posteriorly at the left atrioventricular groove to enter the great cardiac vein (GCV).
Great cardiac vein
The great cardiac vein corresponds to the venous segment that runs in the left atrioventricular groove before entering the coronary sinus at approximately a 180° angle. Its ostium coincides with the valve of Vieussens in the majority of cases.
Middle cardiac vein
The middle cardiac vein (MCV) originates close to the left ventricular apex, and courses within the posterior interventricular groove toward the base. It usually drains into the coronary sinus just before it arrives in the right atrium.
Lateral vein
Also called left marginal vein, the lateral vein (LV) originates from the lateral face of the left ventricle and drains into the great cardiac vein.
Image analysis
Multiphase reconstructed MDCT acquisitions were processed on a dedicated workstation (Advantage Workstation 4.4; GE Healthcare, Milwaukee, WI). A systematic review of reconstructed images from 5 to 95% R-R interval was undertaken to select the stage with fewest motion artefacts. Three-dimensional volume-rendered reconstructions were used to obtain a global approach of the anatomy of the cardiac veins. Then, the course of the veins was evaluated in three orthogonal planes using multiplanar reformatting. As described previously , visibility of the coronary sinus and each vein was evaluated visually using a 4-point scale (grade 0: not visible; grade 1: visible with discontinuity; grade 2: visible with irregular borders; grade 3: visible as a smoothly bordered vascular structure). According to this classification, visibility was classified as good (grade 2–3) or poor (grade 0–1). The length of the centre lumen line of each vein from the ostium to the most proximal discernable point was measured through the use of multiplanar reconstructions. The ostium of the coronary sinus was defined as the site where the coronary sinus makes an angle with the right atrium. The size of the ostia of the coronary sinus was measured in two orthogonal axes (anteroposterior and superoinferior axis) and all the visible veins perpendicularly to the vessel axis using multiplanar reconstructions. The contrast-to-noise ratio (CNR) of the coronary sinus and each vein was calculated using a previously described technique . A 3-dimensional region of interest (ROI) encompassing the vein lumen was placed centrally in the vein close to its ostium and then copied and pasted in adjacent tissues near the ostium of the vein. The difference between mean vessel and mean adjacent tissue density (in CT units) was divided by image noise, determined as the standard deviation of CT attenuation in a 3-dimensional ROI placed within the contrast-enhanced ascending aorta. The CNR could then be calculated as follows:
C N R = C T a t t e n u a t i o n i n l u m e n − C T a t t e n u a t i o n i n a d j a c e n t t i s s u e / i m a g e n o i s e
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
