DISCUSSION

In patients with ARVC/D, it has been demonstrated that the RVOT is the most commonly dilated anatomical structure.[4] It has therefore been proposed, when associated with regional RV akinesia, dyskinesia, or aneurysm, as a major diagnostic criterion in the 2010 revised TFC of ARVC/D.[1] Since 2D TTE is the most readily available imaging modality to evaluate cardiac anatomy, RVOT dilatation is defined in the revised TFC as an echocardiographic diameter ≥ 21 mm/m2 in the PSAX view or ≥ 19 mm/m2in the PLAX view. However, accurate measurements of the RVOT diameter by 2D TTE may be difficult to obtain in patients with poor echocardiographic windows and there may be significant variation between operators.[5] CMR may overcome these echocardiographic limitations and provide very accurate and reliable measurements.[6] Since static bSSFP images in the three orthogonal planes are routinely obtained with all CMR protocols in our institution, we sought to provide reference ranges for the RVOT diameter in children and adolescents in the strict transverse and sagittal views and compare the measurements with the echocardiographic PSAX and PLAX views, respectively.
We obtained high SDs and consequently very wide reference ranges for the RVOT diameter by CMR. This could be explained by possible imprecision of the measurements due to lower spatial and temporal resolution of CMR compared to 2D TTE. However, we observed even higher SDs for the RVOT diameter by 2D TTE, with a large proportion of the measurements being greater than the cut-off values for RVOT dilatation according to the 2010 revised TFC. In fact, 22% of the measurements in the PSAX view were ≥ 21 mm/m2 and 30% of the measurements in the PLAX view were ≥ 19 mm/m2. These wide variations may be attributed to the fact that 2D TTE was performed by many different operators, including less experienced ones, and the RVOT may have been imaged in many cases in an oblique view. Alternatively, pediatric data on normal values of echocardiographic RVOT diameter measurements is rare, and the RVOT diameter may normally exhibit wider variations in children compared to adults.[7,8] Although to a lesser degree than in our study, Koestenberger et al. also found wide reference ranges for the RVOT diameter in older children and adolescents.[9] Finally, the echocardiographic measurements of the RVOT diameter as recommended in the 2010 revised TFC may not be optimal for ARVC/D diagnosis, as they have been shown to have low sensitivity.[10]
Despite their wide reference ranges, the RVOT diameter measurements by CMR showed a strong correlation with the echocardiographic measurements, suggesting that the static bSSFP images triggered to end-diastole in the strict transverse and sagittal views may provide accurate measurements of the RVOT diameter. However, there was poor agreement between the CMR and 2D TTE measurements, with CMR on average significantly underestimating the RVOT diameter compared to 2D TTE. This may not be surprising as the strict transverse and sagittal views by CMR are oriented more perpendicular to the RVOT than the PSAX and PLAX views by 2D TTE, which image the RVOT obliquely. Gotschy et al. demonstrated much better agreement between RVOT diameter measurements by CMR and 2D TTE when equivalent views were compared [10]. This was not possible in our study as the CMR views equivalent to the 2D TTE PSAX and PLAX views are not routinely obtained with all CMR protocols in our institution, and therefore too few measurements would have been compared.

Limitations

Limitations of our study include a sample size that is too small to provide normal values for the RVOT diameter by CMR, and the fact that we did not assess intraobserver and interobserver reliability of the measurements.