Comparative imaging of differential pulmonary blood flow in patients with congenital heart disease: magnetic resonance imaging versus lung perfusion scintigraphy.

摘要

BACKGROUND: Lung perfusion scintigraphy is considered the gold standard to assess differential pulmonary blood flow while magnetic resonance (MR) has been shown to be an accurate alternative in some studies. OBJECTIVE: The purpose of the study was to assess the accuracy of phase contrast magnetic resonance (PC-MR) in measuring pulmonary blood flow ratio compared with lung perfusion scintigraphy in patients with complex pulmonary artery anatomy or pulmonary hypertension and to document reasons for discrepant results. MATERIALS AND METHODS: We identified 25 cases of congenital heart disease between January 2000 and 2003, in whom both techniques of assessing pulmonary blood flow were performed within a 6-month period without an interim surgical or transcatheter intervention. The study group included cases with branch pulmonary artery stenosis, intracardiac shunts, single ventricle circulation, pulmonary venous anomalies and conotruncal defects. The mean age at study was 5.7 years (range 0.33-12) with a mean weight of 20.3 kg (range 6.5-53.6). The two methods were compared using a Bland-Altman analysis, and the Pearson correlation coefficient was calculated using the lung scan as the gold standard. Discrepant results were examined by reviewing the source images to elucidate reasons for error by MR. RESULTS: Bland-Altman analysis comparing right pulmonary artery (RPA) blood flow percentage, as measured by each modality, showed a mean difference of 1.43+/-9.8 (95 limits of agreement: -17.8, 20.6) with a correlation coefficient of r=0.84, P10) was found with a mean difference between techniques of 17.9. The reasons for discrepant results included MR artifacts, dephasing owing to turbulent flow, site of data acquisition and lobar lung collapse. CONCLUSION: When using PC-MR to assess pulmonary blood flow ratio, important technical errors occur in a significant proportion of patients who have abnormal pulmonary artery anatomy or pulmonary hypertension. If these technical errors are avoided, PC-MR is able to supply both anatomic and quantitative functional information in this patient population.