Pre- and Postoperative Evaluation of Partial Anomalous Pulmonary Venous Return by 3-Dimensional Cardiovascular Magnetic Resonance Imaging and Cardiovascular Computed Tomography

Abstract

The pre- and postoperative evaluation of anomalous pulmonary venous return usually requires multiple invasive and noninvasive tests in order to obtain complete anatomic and functional data. Conversely, in a single setting, either cardiovascular magnetic resonance imaging or cardiovascular computed tomography can sufficiently reveal this information in adult patients. Herein, we present the cases of 2 patients with partial anomalous pulmonary venous return who underwent preoperative and postoperative evaluation by either method alone, and we discuss the benefits and limitations of each technique.

The pre- and postoperative evaluation of anomalous pulmonary venous return usually requires multiple invasive and noninvasive tests in order for the clinician to achieve a complete anatomic and functional assessment.¹ Cardiovascular magnetic resonance imaging (CMRI) and cardiovascular computed tomography (CCT) can provide this information noninvasively, in a single setting.^2,3 Here, we present the cases of 2 patients with partial anomalous pulmonary venous return, one of whom underwent evaluation by CCT and the other by CMRI. We discuss the advantages and limitations of each method.

Case Reports

Patient 1

In October 2007, a 49-year-old man presented with symptoms of congestive heart failure. Echocardiography showed enlarged right-sided cardiac chambers and a dilated coronary sinus. The atrial and ventricular septa were intact. Cardiovascular CT confirmed an intact atrial septum, a large coronary sinus, and the drainage of both left pulmonary veins into the coronary sinus (Fig. 1A). Catheterization revealed a substantial left-to-right shunt, with a calculated Qp/Qs ratio of 1.6:1 by oximetry. Surgical correction was accomplished by creating an atrial septal defect, unroofing the coronary sinus into the left atrial wall, and constructing a baffle with autologous pericardium in order to redirect the coronary sinus drainage into the left atrium and close the atrial septal defect. A small, mandatory right-to-left shunt resulted, with no physiologic consequences. Figure 1B shows the postoperative CCT imaging.

Fig. 1 A) Patient 1. Preoperative cardiovascular computed tomography (volume-rendered image) shows the left-sided pulmonary veins draining into a large coronary sinus (CS). B) Postoperative image shows the surgically redirected CS draining into the left atrium. The arrow indicates the wide communication between the 2 structures. Ao = descending aorta; IPV = inferior pulmonary vein; LA = left atrium; LV = left ventricle; PA = pulmonary artery; RV = right ventricle; SPV = superior pulmonary vein

Patient 2

In November 2007, a 75-year-old woman with a previously resected left atrial myxoma presented with symptoms of congestive heart failure. Echocardiography showed enlargement of the superior vena cava, right atrium, and right ventricle. The estimated right ventricular systolic pressure was 52 mmHg. There was no atrial or ventricular septal defect. Cardiovascular MRI revealed that the left pulmonary veins drained into a left vertical vein, which ascended anterior to the left pulmonary artery and drained into the brachiocephalic vein (Fig. 2). The calculated Qp/Qs ratio was 1.5:1. Surgical correction was achieved by separating the left vertical vein from its confluence with the brachiocephalic vein. The vertical vein was then anastomosed to the left atrial appendage (Fig. 3A). Postoperative CMRI showed an intact anastomosis and resolution of the shunt (Fig. 3B).

Fig. 2 Patient 2. Preoperative cardiovascular magnetic resonance imaging shows the left pulmonary veins draining into the left vertical vein (arrow).

Fig. 3 Patient 2. A) Intraoperative photograph shows the left vertical vein (LVV) anastomosed to the left atrial appendage (LAA). B) Post-operative 3-dimensional volume-rendering cardiovascular magnetic resonance image of the heart and great vessels shows the LVV anastomosed to the LAA. The stump of the LVV (asterisk) remains in the brachiocephalic vein.

Discussion

Partial anomalous pulmonary venous return might not be diagnosed until a patient's adulthood.⁴ The decision to perform surgical correction relies upon preoperative knowledge of the pulmonary venous anatomy, associated cardiac anomalies, and the magnitude of the left-to-right shunt. Multiple complementary tests are usually necessary in order to obtain complete anatomic and functional data. The limited acoustic windows in transthoracic echocardiography may preclude adequate inspection of all pulmonary veins, because views are generally limited to the venous drainage into the left atrium, yielding no information on mediastinal or extramediastinal origin, trajectory, or drainage.² Transesophageal echocardiography may provide better evaluation of the pulmonary venous atrial drainage and mediastinal trajectory; however, this technique is invasive and may be poorly tolerated.² Although right-sided cardiac catheterization with chamber oximetry is considered the gold standard in the measurement of shunt fraction, this method is also invasive and might not prove accurate when the pulmonary veins drain into the brachiocephalic vein.² Venous contrast angiography is time-consuming and might not provide complete or detailed information about the venous anatomy.⁵

In a single setting, CMRI or CCT can noninvasively provide volumetric data that are essential for diagnosis and therapeutic planning—notably, detailed anatomic information regarding the number, origin, course, and drainage of all pulmonary veins, and their relationships to cardiac and pulmonary structures. Associated great vessels and intracardiac anomalies can also be evaluated.² In addition, multiplanar 3--dimensional reconstructions enable precise anatomic views and surgical planning. Cardiovascular CT yields imaging data with isotropic voxels, wherein the spatial resolution is the same in all 3 dimensions; this enables volume--rendering and multiplanar reformatting in any plane or volume. Although CMRI is typically not isotropic in resolution, its superior temporal resolution affords dynamic views of blood flow upon the use of a small bolus of gadolinium. Gadolinium is less nephrotoxic than is the iodinated medium that is used in CCT or angiography.⁶ In addition, CMRI can provide 3--dimensional anatomic detail similar to that obtained with CCT, along with cine and shunt-fraction quantification. Both CMRI and CCT perform well in quantifying right and left ventricular volumes and systolic function,⁷ although CMRI may be slightly better because it provides most or all of the necessary preoperative data without the use of ionizing radiation. In comparison, CCT cannot quantify shunt fraction; however, it provides 3-dimensional anatomic detail that is superior to that of angiography.

Cardiovascular MRI and CCT are also useful postoperatively. Both facilitate the evaluation of patency, anastomotic narrowing, and right ventricular systolic function; and CMRI provides precise quantification of any residual shunt.

We have found that CMRI enables complete anatomic and functional evaluation of partial anomalous pulmonary venous return in a single noninvasive setting, with minimal risk to the patient. It has become our method of choice in the preoperative and postoperative evaluation of these patients. In many instances, the information obtained by use of CMRI renders cardiac catheterization unnecessary. However, not all patients are candidates for CMRI. The most common contraindications are the presence of pacemakers or other implants subject to magnetism, and patients' fear of closed spaces. In these circumstances, CCT provides the needed anatomic and functional data. Disadvantages of CCT include its inability to calculate shunt fractions, the patients' exposure to ionizing radiation, and a greater risk of nephrotoxicity than that which occurs after the use of CMRI.