Surgically Corrected Congenital Heart Disease: Utility of 64-MDCT
Abstract
OBJECTIVE. The purpose of this article is to review the CT appearance of postoperative morphology and complications after surgical correction of congenital heart anomalies.
CONCLUSION. Echocardiography is typically the initial imaging technique used for congenital heart disease; however, some thoracic regions are beyond the imaging scope of echocardiography, particularly after surgical revision. This article shows, through a series of illustrative cases, the usefulness of 64-MDCT in these patients.
Introduction
Imaging of surgically treated patients with congenital heart disease is complicated by the variable and complex cardiac anatomy in these patients, modification of complicated anatomy by surgical intervention, and interference from the frequently present conduits, baffles, stents, and coils. The clinician can choose from one of four techniques when imaging the patient with congenital heart disease: echocardiography, catheterization with angiography, MRI, and CT. Catheterization is uncommonly per formed for diagnostic reasons alone and is chosen predominantly to intervene thera peutically. Echocardiography is usually the initial choice because of the excellent resolution and the logistic simplicity. However, the post operative patient is often difficult to image because of degradation of acoustic windows. When echocardiographic images are inadequate to address a specific clinical question, the alternatives of cardiac MRI and CT must be weighed. If primarily additional functional information is im portant, MRI is preferred, particularly if longitudinal exami nations are anticipated. MRI has good spatial and temporal resolution, allowing excellent flow and volumetric functional evaluation and tissue characterization. Although MRI can be performed in some patients with selected implantable pacemakers, for many patients and at many centers, a pacemaker remains a contraindication to performing an MRI examination [1]. CT has outstanding spatial resolution; however, it suffers in its temporal resolution, making it an inferior technique for evaluating ventricular function, particularly when the heart rate is elevated.
If the clinical question requires further delineation of cardiac anatomy and morphology, CT and MRI can both be considered. However, in infants and toddlers, administration of sedation medication or even general anesthesia may be needed to complete MRI examinations. Current generation MDCT scanning times enable examinations to be completed in seconds, obviating sedation in most patients. Submillimeter isotropic resolution results in superior spatial resolution compared with MRI. These advantages must be balanced against the additional risk of ionizing radiation. Fortunately, measures to reduce radiation exposure are evolving, including reduction of tube current based on weight and size, modulation of tube current depending on anatomic position or phase of cardiac cycle (ECG-modulated pulsing), or reduction of tube voltage [2].
Although primary surgical repair of many congenital heart disorders is being performed early in life with greater frequency, some patients who have had palliative procedures at a young age will require imaging. Older children and adults with congenital heart disease who have undergone surgical correction constitute a growing population because of the improved survival after treatment. Accordingly, radiologists per forming MDCT will be imaging more of these patients for clinical assessment, and in the setting of suspected complications. This focused review presents a series of cases to show the utility of cardiac 64-MDCT for imaging patients with congenital heart disease who have undergone various forms of palliative or definitive surgical correction. Using a procedure-based approach, emphasis is placed on the anatomic configuration after surgical correction, potential complications that must be excluded, and the utility of MDCT for visualizing anatomic regions not accessible to echocardiography. The cases show the importance of 2D and 3D multiplanar viewing for revealing vascular anatomy, surgical anastomoses, and shunts or conduits that are often not fully visualized in an axial plane.
Scanning Protocols
In select cases, 16-MDCT may be satisfactory for imaging congenital heart disease. However, the temporal resolution of 64-MDCT results in acceptable image quality for a much higher percentage of patients. The faster acquisition time results in shorter breath-hold duration and a reduction in the volume of IV contrast material needed, with intraindividual comparison revealing improved image quality [3]. The narrow temporal window of data acquisition is particularly important for pediatric patients with high heart rates. Proper patient preparation is essential to performing high-quality studies, and one of our most important assets is a dedicated pediatric nurse who is adept with these delicate patients.
Protocol design is tailored to each case. Leschka et al. [2] have written a detailed summary of 64-MDCT protocols specific to each congenital cardiac anomaly, including contrast infusion, when to use ECG gating, anatomic region covered, and reconstruction parameters. In patients with shunts or baffles, the timing is often more difficult because of differential flow. Depending on the case, a single- or dual-phase acquisition may be indicated. This series of cases was performed on a commercially available 64-MDCT scanner (Sensation 64, Siemens Medi cal Solutions) using the following scanning protocol: 90–150 mAs (depending on patient size and weight); 120 kVp; pitch, 0.7; detector thickness, 0.6 mm; slice thickness, 0.75 mm; reconstruction interval, 0.5 mm; scanner rotation time, 0.33 second. Total study time is in the range of 1.5–5 seconds for a child and 10–12 seconds for an adult. Patients were injected with up to 2 mL/kg of Visipaque 320 (iodixanol, GE Healthcare) at an injection rate of 2 mL/s for children and 3–5 mL/s for adults. For complicated congenital heart cases, a saline flush is usually not used, to enable contrast opacification of both right and left sides of the heart. A flush will be used if the coronary arteries are being evaluated. We use a test-bolus technique to delineate the timing, using 5–10 mL of contrast agent with a 10-mL saline flush.



In selected patients, such as those in whom coronary artery or cardiac wall motion evaluation is indicated, a gated study is performed. Retrospective gating as we perform it requires a slower heart rate, ideally less than 65 beats per minute (bpm) for adults with a 64-MDCT scanner. In children, gating can be performed up to a heart rate of 90 bpm with a 64-MDCT scanner. The protocol for adults in these cases was 400 mAs; 120 kVp; pitch, 0.2; detector thickness, 0.6 mm; slice thickness, 0.75 mm; reconstruction interval, 0.5 mm; scanner rotation time, 0.33 second; and total scanning time, less than 5 seconds.
All scan data were sent to an independent workstation (Leonardo, Siemens Medical So lutions) running InSpace software (Siemens). Three-dimensional renderings were developed using a combination of volume rendering and maximum intensity projection (MIP), in addition to 2D multiplanar reconstructions (MPRs). We perform 3D rendering on all patients because it adds additional information. Once the CT data are acquired, our goal is to extract the maximum information from the volume. These patients have complex anatomy that is often best visualized with volume rendering because it conveys 3D relationships not shown on MPRs or MIP. Although 3D rendering requires minimal additional time, it is available on most workstations and does not necessitate any additional radiation exposure. Further more, our referring physicians prefer to see complex anatomy in 3D.
Clinical Applications
Systemic and Venous Shunts to Augment Effective Pulmonary Blood Flow
A systemic shunt redirects blood from the aorta or a branch of the aorta to the pulmonary arteries in order to increase oxygen saturation. These include central shunts (surgical graft to the pulmonary arteries from the ascending aorta) (Fig. 1A, 1B, 1C), Potts shunts (from descending aorta to left pulmonary artery, no longer performed), Waterston shunts (from ascending aorta to right pulmonary artery), and Blalock-Taussig shunts (subclavian artery to ipsilateral pulmonary artery). The contemporary modified Blalock-Taussig shunt is performed by interposing a prosthetic graft from the subclavian or innominate artery to the ipsilateral branch pulmonary art ery via an end-to-end anastomosis [4]. Postoperative complications include distortion of the ipsilateral pulmonary artery, identified in 24–33% after a modified Blalock-Taussig shunt [5, 6], occurring more commonly in those who undergo shunt placement earlier in life [5]. Major (≥ 50%) pulmonary artery stenosis occurs in 14% of cases, and rarely, there is complete occlusion of the pulmonary artery or the shunt [5]. In approximately 50%, shunt narrowing occurs, usually at the anastomosis [6].



Although echocardiography is often the initial imaging technique used to evaluate the branch pulmonary arteries, usually only the central pulmonary arteries are imaged because of acoustic interference by the lungs. Frequently in the postoperative patient, surgical clips or catheter-placed coils create imaging artifacts on MRI, which may obscure visualization of the branch pulmonary arteries. In this case, MDCT is helpful, particularly for the left pulmonary artery. The appearance of the left pulmonary artery and the extent of distortion are important to define preoperatively, before any additional sur gical intervention. If the area of distortion is too posterior to adequately visualize from a midsternotomy incision, the surgeon must approach it from a left thoracotomy.
Venous shunts, although not performed in neonates because of elevated pulmonary resistance, can be used to increase pulmonary blood flow after infants are approximately 3 months old. A classic Glenn shunt is a surgical connection between the vena cava and the transected ipsilateral pulmonary artery. These have been performed in the setting of tricuspid atresia and single ventricle [7], but they are essentially no longer used because of the undesirable creation of discontinuous branch pulmonary arteries. The currently performed bidirectional Glenn shunt connects the cranial segment of the transected superior vena cava to the right pulmonary artery via an end-to-side anastomosis, allowing blood to flow to both lungs [4] (Fig. 2A, 2B, 2C). A hemi-Fontan procedure, a variant of the bidirectional Glenn shunt, involves a second anastomosis of the caudal superior vena cava segment with the inferior right pulmonary artery and may facilitate a subsequent Fontan procedure. Sequelae after a Glenn shunt include decreased arterial diameter and flow in the contralateral pulmonary artery as well as development of decompressing venous collaterals (superior vena cava-to-inferior vena cava circulation), resulting in reduced oxygen saturation [7, 8]. Patients are also at risk for developing pulmonary arteriovenous fistulas (AVFs) (20%); the incidence correlates with the number of years after the procedure [7].
A Kawashima procedure is performed in the setting of heterotaxy syndrome and refers to placement of a bidirectional Glenn shunt in a patient with interruption of the inferior vena cava and azygous extension to the superior vena cava. This therefore redirects all systemic venous blood to the lungs except the hepatic and coronary venous return. A left-sided Kawashima vascular connection is anasto mosed to the left branch pulmonary artery. Postoperatively, patients have mild cyanosis because the desaturated hepatic venous blood is pumped along with the pulmonary venous blood to the body. Surgeons sometimes avoid redirecting hepatic venous drainage in patients with an interrupted inferior vena cava because of the complexity of incorporating this return. Unfortunately, failure to incorporate hepatic venous blood results in a number of patients after Kawashima repair developing a pulmonary arteriovenous malformation (AVM) [9]. When an AVM is present, a conduit may be placed to incorporate hepatic blood flow into the pulmonary circuit to facilitate its resolution [10]. Difficult to visualize with echocardiography, these conduits from the hepatic veins to the innominate vein can be evaluated with MDCT to exclude thrombosis or stenosis (Fig. 3A, 3B, 3C, 3D).




Separation of Systemic and Pulmonary Circulations: Single and Biventricular Circuits
The Fontan procedure was initially conducted for tricuspid atresia and is currently performed as the final palliative procedure in any heart with single-ventricle physiology, such as hypoplastic left heart syndrome. This procedure involves separation of the systemic and pulmonary circulations. Systemic venous blood is routed directly to the lungs, which may be performed initially during a staged hemi-Fontan procedure or a bidirectional Glenn shunt. The single ventricle pumps exclusively (or nearly exclusively) to the body. Historically, the inferior vena cava connection involved an anastomosis between the superior aspect of the right atrium with the right pulmonary artery. More recently, the connection is made using an intraatrial baffle or an external conduit from the inferior vena cava to the pulmonary artery [11, 12] (Fig. 4A, 4B, 4C, 4D).
The Fontan cavopulmonary pathway is at risk for thrombosis or suboptimal flow dynamics [11, 12]. Systemic circulation complications include subaortic obstruction, systemic ventricular dysfunction, and atrioventricular valve regurgitation [12]. Aortocollateral vessels and pulmonary AVFs may also develop, and alterations in venous pressure result in an increased propensity for pleural effusion and protein-losing entero pathy [12]. In the early postoperative period, an unusual complication is a large fluid collection adjacent to the external conduit (Fig. 5A, 5B). MDCT with IV contrast material can delineate the extent of such a collection, determine whether it represents a contained collection or a pseudoaneurysm, and show resolution after treatment.
In truncus arteriosus, a rare congenital anomaly, the pulmonary arteries arise from the aorta, which overrides a ventricular septal defect. Classification systems reflect the origins of the branch pulmonary arteries and whether the aortic arch is interrupted. During surgical repair, the ventricular septal defect is closed, and the pulmonary arteries are anastomosed to the right ventricle using a conduit. Follow-up requires evaluation for conduit or truncal valve stenosis or regurgitation, adequacy of ventricular septal defect closure, stenosis of the pulmonary arteries (Fig. 6), dilatation of the aortic root, and ventricular dysfunction [13]. Cardiac angiography can assess for hemodynamically significant stenosis and residual ventricular septal defects; however, the conduit to pulmonary artery anastomosis may not be fully visualized to enable adequate assessment of the pulmonary artery origins, which are seen well with MDCT (Fig. 6).
A Norwood procedure is the initial palliation for patients with hypoplastic left heart; it establishes a reliable egress of systemic arterial blood from the systemic right ventricle via the main pulmonary artery to the reconstructed aortic arch [11, 14]. Pulmonary blood flow is then provided by either a systemic shunt or a restrictive conduit from the right ventricle to the branch pulmonary arteries (Sano modification) [14]. After a Norwood procedure, CT is performed to visualize the aortic arch and ventricular or aortic pulmonary artery anastomoses (Fig. 7A, 7B, 7C). Distal aortic arch narrowing is an important complication to identify in order to avoid a pressure overload on the systemic right ventricle. Elevation in ventricular end-diastolic pressure related to ventricular dysfunction will raise pulmonary venous pressure (and then pulmonary artery pressure) and increase the risk of subsequent Glenn or Fontan procedures.
Augmentation of Pulmonary Artery Flow for Pulmonic Atresia with Tetralogy of Fallot
In patients with severe pulmonary atresia resistant to interventional dilatation, arterial stents or surgical conduits may be required to augment pulmonary artery flow distal to the region of narrowing (Fig. 8A, 8B).










Revising Transposed Outflow Tracts
Older techniques used to correct transposition of the great vessels include the Senning and Mustard procedures, aimed at revising the venous return. This was accomplished with an atrial baffle composed of autologous tissue (Senning) or synthetic material (Mustard) in conjunction with atrial septal resection [15]. Venous inflow was diverted via the baffle to the contralateral atrio ventricular valve and ventricle [15]. Com plications include arrhythmia, sudden death, right ventricular dysfunction, tricuspid regurgitation, obstruction of the atrial baffle due to narrowing, systemic and pulmonary venous obstruction, and atrial baffle leak [16, 17]. Repair for transposition of the great vessels is currently most commonly performed with an arterial switch procedure; however, not all patients are candidates. The pulmonary artery and aorta superior to the sinotubular junction (i.e., distal to coronary artery origins) are transected and reattached in the alternative location, and the coronary arteries migrate to the neoaortic root [15, 18]. Potential cardiopulmonary complications after an arterial switch operation include anastomotic obstruction (most commonly pulmonary), ventricular dysfunction, central and peripheral pulmonary stenosis (with branch pulmonary artery distortion occurring more commonly on the left side), neoaortic insufficiency, mitral regurgitation, left mainstem bronchus compression resulting from posteriorly displaced aorta, and less frequently, coronary artery stenoses [15, 18]. In the setting of compromised oxygenation postoperatively, MDCT can add considerable information in addition to that provided by echocardiography to determine whether a pulmonary artery anomaly is the underlying cause (Fig. 9A, 9B, 9C, 9D).






Discussion
The cases presented show how MDCT with 2D and 3D renderings can play an important role in the care of the patient with congenital heart disease by providing information not available with echocardiography. For certain anatomic regions, including the cardiac septa and valves, echocardiography has been shown to be superior to CT [19, 20]. However, CT has particular value beyond echocardiography for evaluating branch pulmonary arteries, particularly the left [19, 21, 22]; the aortic arch [23, 24]; complex abnormalities of systemic and pulmonary venous return [20, 25, 26]; the coronary sinus and inferior vena cava [19]; the coronary arteries [19, 20]; aortopulmonary collaterals [19, 21]; surgically placed conduits, baffles, and shunts [21]; and associated airway abnormalities [27, 28].
It is our practice to initially image all patients with echocardiography. If clinically important questions remain, we identify alternative techniques expected to offer adequate spatial and temporal resolution to answer the remaining questions. If more than one technique is suitable, we choose the technique on the basis of the patient characteristics (e.g., length of time able to tolerate scanning, clinical stability, feasibility of transport) and examination requirements (e.g., need for sedation).
In certain ways, we are at a golden moment in cardiac imaging because of the remarkable advances to date and the anticipated advances over the next 5 years. For echocardiography, anticipated changes are the increased application of 3D scanning and regional measures of ventricular function. For MRI, wider application of parallel processing and navigator sequences should decrease the need for patient sedation and should shorten scanning time. Scanning time may further decrease (or resolution increase) with more extensive use of 3-T scanners. For CT, dual-source scanners, which can decrease scanning time by half and reduce dose as well, should allow still further improvements in spatial resolution and shorter scanning times. Perhaps even more important, the improvements in temporal resolution will allow better quanti fication of ventricular function.
As cross-sectional imaging techniques become more widely used to evaluate patients with congenital heart disease, radiologists who interpret these studies will need to thoroughly understand the clinical and surgical history to assist in managing each of these challenging patients, and will need to maintain an understanding of the normal postoperative appearances and potential complications after surgical correction.
Footnotes
Address correspondence to P. J. Spevak ([email protected]).
CME
This article is available for CME credit.
See www.arrs.org for more information.
References
1.
Nazarian S, Roguin A, Zyiman MM, et al. Clinical utility and safety of a protocol for noncardiac and cardiac magnetic resonance imaging of patients with permanent pacemakers and implantable cardioverter defibrillators at 1.5 Tesla. Circulation 2006; 19:114:1277 –1284
2.
Leschka S, Oechslin E, Husmann L, et al. Pre- and postoperative evaluation of congenital heart disease in children and adults with 64-section CT. RadioGraphics 2007; 27:829–846
3.
Dewey M, Hoffmann H, Hamm B. CT coronary angiography using 16 and 64 simultaneous detector rows: intraindividual comparison. Rofo 2007 179:581 –586. Epub 2007 May9
4.
Baum VC. Pediatric cardiac surgery: an historical appreciation. Ped Anesthesia 2006; 16:1213 –1225
5.
Godart F, Qureshi SA, Simha A, et al. Effects of modified and classic Blalock-Taussig shunts on the pulmonary arterial tree. Ann Thorac Surg 1998; 66:512 –518
6.
Gladman G, McCrindle BW, Williams WG, Freedom RM, Benson LN. The modified Blalock-Taussig shunt: clinical impact and morbidity in Fallot's tetralogy in the current era. J Thorac Cardiovasc Surg 1997; 114:25 –30
7.
Kopf GS, Laks H, Stansel HC, Hellenbrand WE, Kleinman CS, Talner NS. Thirty-year follow-up of superior vena cava–pulmonary artery (Glenn) shunts. J Thorac Cardiovasc Surg 1990; 100:662–671
8.
Mendelsohn AM, Bove EL, Lupinetti FM, Crowley DC, Lloyd TR, Beekman RH. Central pulmonary artery growth patterns after the bidirectional Glenn procedure. J Thorac Cardiovasc Surg 1994; 107:1284 –1290
9.
Srivastava D, Preminger T, Lock JE, et al. Hepatic venous blood and the development of pulmonary arteriovenous malformations in congenital heart disease. Circulation 1995; 92:1217 –1222
10.
McElhinney DB, Kreutzer J, Lang P, Mayer JE, del Nido PJ. Lock JE. Incorporation of the hepatic veins into the cavopulmonary circulation in patients with heterotaxy and pulmonary arteriovenous malformations after a Kawashima procedure. Ann Thorac Surg 2005; 80:1597 –1603
11.
van Doorn CA, de Leval MR. The Fontan operation in clinical practice: indications and controversies. Nat Clin Pract Cardiovasc Med 2005; 2:116 –117
12.
de Leval MR. The Fontan circulation: what have we learned? what to expect? Pediatr Cardiol 1998; 19:316–320
13.
Bashore TM. Adult congenital heart disease: right ventricular outflow tract lesions. Circulation 2007; 115:1933 –1947
14.
Alsoufi B, Bennetts J, Verma S, Caldarone CA. New developments in the treatment of hypoplastic left heart syndrome. Pediatrics 2007; 119:109–117
15.
Warnes CA. Transposition of the great vessels. Circulation 2006; 114:2699 –2709
16.
Wells WJ, Blackstone E. Intermediate outcome after Mustard and Senning procedures: a study by the Congenital Heart Surgeons Society. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 2000; 3:186 –197
17.
Helbing WA, Hansen B, Ottenkamp J, et al. Long-term results of atrial correction for transposition of the great arteries: comparison of Mustard and Senning operations. J Thorac Cardiovasc Surg 1994; 108:363 –372
18.
Haas F, Wotke M, Poppert H, Meisner H. Long-term survival and functional follow-up in patients after the arterial switch operation. Ann Thorac Surg 1999; 68:1692 –1697
19.
Lee JJ, Kang D-S. Feasibility of electron beam tomography in diagnosis of congenital heart disease: comparison with echocardiography. Eur J Radiol 2001; 38:185–190
20.
Beier UH, Jeinin V, Jain S, Ruiz CE. Cardiac computed tomography compared to transthoracic echocardiography in the management of congenital heart disease. Catheter Cardiovasc Interv 2006; 68:441 –449
21.
Westra SJ, Hill JA, Alejos JC, Galindo A, Boechat MI, Laks H. Three-dimensional helical CT of pulmonary arteries in infants and children with congenital heart disease. AJR 1999; 173:109–115
22.
Paul JF, Lambert V, Losay J, et al. Three-dimensional multislice CT scanner: value in patients with pulmonary atresia with septal defect [in French]. Arch Mal Coeur Vaiss 2002; 95:427–432
23.
Cinar A, Haliloglu M, Karagoz T, Karcaaltincaba M, Celiker A, Tekinalp G. Interrupted aortic arch in a neonate: multidetector CT diagnosis. Pediatr Radiol 2004; 34:901–903
24.
Sridharan S, Yates R, Taylor AM. Optimizing imaging after coarctation stenting: the clinical utility of multidetector computer tomography. Catheter Cardiovasc Interv 2005; 66:420 –423
25.
Chen SJ, Wang JK, Li YW, et al. Validation of pulmonary venous obstruction by electron beam computed tomography in children with congenital heart disease. Am J Cardiol 2001; 87:589–593
26.
Sridhar PG, Kalyanpur A, Suresh PV, et al. Total anomalous pulmonary venous connection: helical computed tomography as an alternative to angiography. Indian Heart J 2003; 55:624–627
27.
Chen SJ, Lee WJ, Wang JK, et al. Usefulness of three-dimensional electron beam computed tomography for evaluating tracheobronchial anomalies in children with congenital heart disease. Am J Cardiol 2003; 92:483 –486
28.
Chen SJ, Shih TT, Liu KL, et al. Measurement of tracheal size in children with congenital heart disease by computed tomography. Ann Thorac Surg 2004; 77:1216 –1221
Information & Authors
Information
Published In
Copyright
© American Roentgen Ray Society.
History
Submitted: July 19, 2007
Accepted: March 24, 2008
Keywords
Authors
Metrics & Citations
Metrics
Citations
Export Citations
To download the citation to this article, select your reference manager software.