Evaluation of a comprehensive cardiovascular magnetic resonance protocol in young adults late after the arterial switch operation for d-transposition of the great arteries

Abstract

Background: In adults with prior arterial switch operation (ASO) for d-transposition of the great arteries, the need for routine coronary artery assessment and evaluation for silent myocardial ischemia is not well defined. In this observational study we aimed to determine the value of a comprehensive cardiovascular magnetic resonance (CMR) protocol for the detection of coronary problems in adults with prior ASO for d-transposition of the great arteries.

Methods: Adult ASO patients (≥18 years of age) were recruited consecutively. Patients underwent a comprehensive stress perfusion CMR protocol that included measurement of biventricular systolic function, myocardial scar burden, coronary ostial assessment and myocardial perfusion during vasodilator stress by perfusion CMR. Single photon emission computed tomography (SPECT) was performed on the same day as a confirmatory second imaging modality. Stress studies were visually assessed for perfusion defects (qualitative analysis). Additionally, myocardial blood flow was quantitatively analysed from mid-ventricular perfusion CMR images. In unclear cases, CT coronary angiography or conventional angiography was done.

Results: Twenty-seven adult ASO patients (mean age 23 years, 85% male, 67% with a usual coronary pattern; none with a prior coronary artery complication) were included in the study. CMR stress perfusion was normal in all 27 patients with no evidence of inducible perfusion defects. In 24 cases the coronary ostia could conclusively be demonstrated to be normal. There was disagreement between CMR and SPECT for visually-assessed perfusion defects in 54% of patients with most disagreement due to false positive SPECT.

Conclusions: Adult ASO survivors in this study had no CMR evidence of myocardial ischemia, scar or coronary ostial abnormality. Compared to SPECT, CMR provides additional valuable information about the coronary artery anatomy. The data shows that the asymptomatic and clinically stable adult ASO patient has a low pre-test probability for inducible ischemia. In this situation it is likely that routine evaluation with stress CMR is unnecessary.

Figure 1

Schematic demonstrating the timeline of the integrated CMR/SPECT assessment of myocardial function, perfusion, scar and coronary anatomy.

Figure 2

Real and ‘pseudo-real’ late gadolinium enhancement (LGE) in 3 patients with positive LGE studies. (Ai-iii) Genuine LGE is evident in the mid inferior wall in the 2-chamber view (Ai) with both linear (black arrows) and more focal (white arrow) enhancement. Confirmation of these findings is provided by short axis cross cuts through this region, which also show subendocardial (white arrows) and midwall nodular (dotted arrow) myocardial scar. (Bi-ii) “Pseudo” LGE (white arrow) is present at the site of ventricular septal defect (VSD) repair shown in 4-chamber (i) and short axis (ii) views. (C) “Pseudo” LGE (black arrow) evident in a large surgical patch placed for VSD repair.

Figure 3

Whole heart coronary magnetic resonance angiography (MRA) at 1×1×1 mm resolution with multiplanar maximum intensity projection reformats to demonstrate the coronary arteries. (A) The right coronary artery (RCA) is seen to have a normal origin and course. (B) Excellent image quality is evident from the depiction of a small RCA marginal branch (arrow). (C) A large conal branch (arrows) takes a pre-pulmonic course and anastomoses with the left coronary system at the apex. (D, E) The left anterior descending (D solid arrow) and left circumflex (D dotted arrow) coronary arteries are shown in their proximal portions. Although the bifurcation of these vessels is clearly depicted, note that the left main coronary artery is not visible and that there is a’gap’ between the aortic root and the LAD/Cx bifurcation even on ultrahigh resolution (0.5 × 0.5 × 0.5 mm) MRA (E arrows). This was misinterpreted as an occlusion of the left main segment as the diagnosis of single coronary artery had not been recognized. RV = right ventricle; LV = left ventricle; nAo = neo-aorta; RA = right atrium; LA = left atrium.

Figure 4

Evaluation of a “kinked” right coronary artery. (A) Straight axial image from a whole heart magnetic resonance angiogram demonstrates an apparent “kink” at the right coronary artery origin. (B-D) Subsequent multiplanar and centre line reformats however demonstrate that the ostium is in reality unobstructed. (E) The left coronary origin is also unobstructed. (Fi-iii) Stress perfusion magnetic resonance frames at basal, mid and apical left ventricular level show no evidence of any inducible perfusion defect. Quantitative perfusion measured in the right coronary territory was normal (not shown).

Figure 5

Bilateral pulmonary artery stents obscuring coronary origins at CMR. (A, B) Metallic artifact from bilateral pulmonary artery (PA) stents (arrows) on steady state free precession cine imaging obscures the coronary arteries as they emerge from the neo-aortic root. (C) Multiplanar reformat from low dose cardiac gated computed tomography reveals normal origins of the right coronary artery (solid arrow), left anterior descending (dashed arrow) and circumflex (dotted arrow) coronary arteries. (D) Lack of effect of bilateral PA stents (arrows) on visibility of the coronary arteries at cardiac CT is apparent on this coronal reformat. RV = right ventricle; LV = left ventricle; nAo = neo aorta.

Figure 6

Example of false positive SPECT study with a fixed apical defect. (A) Steady state free precession imaging demonstrates a normal right coronary origin (solid arrow) but reveals a retro-aortic course of the left main coronary artery (dotted arrows). (B) The anomalous left main regains the normal position and bifurcates in to LAD (dashed arrow) and Cx (dotted arrow). (C-D) Conventional coronary angiography confirms undistorted coronary origins (solid arrows) and substantiates the presence of an anomalous left main coronary artery (dotted arrows). (E) Bullseye plot from resting SPECT indicates a fixed apical defect suggestive of infarction. (F) Stress perfusion CMR is normal. (G, H) Late gadolinium enhancement images in 2 and 4 chamber orientations show no evidence of any apical infarction.

Figure 7

Example of a false positive SPECT with moderate ‘inducible ischemia’. (Ai-ii) MIBI bulls eye plots demonstrating a defect at stress in the basal to mid anterolateral wall (i) which appears fully reversible at rest (ii) with a SDS of 5 (iii) suggesting a moderate inducible perfusion defect. (Bi-iii) Basal, mid and apical short axis slices from stress CMR show visually normal perfusion in all territories. Quantitative perfusion (not shown) confirmed normal stress perfusion values at the sites of the defect seen on MIBI. (Ci-iii) Basal, mid and apical short axis slices from LGE stack demonstrate lack of any myocardial scar. (Di-ii) CT coronary angiogram performed in view of the discordance between SPECT and CMR shows that the coronary arteries are entirely normal.