METHODS: Thirty-six children with FLSV were prospectively enrolled. The LV volume analysis was performed on QLAB 8.1 (semiautomated border tracking) and TomTec 4D LV 3.0 (manual dominant border tracking) and compared with MRI as the reference standard.
RESULTS: 3DE volume quantification was achieved for 32 children with QLAB and 34 children with TomTec. Analysis time was much shorter for QLAB than TomTec (4.8 ± 1.2 vs. 6.3 ± 1.8 minutes, P < 0.05). Ejection fraction (EF) by either 3DE modality was significantly lower than the published normal values (P < 0.01 for each). End-diastolic volume (EDV), end-systolic volume (ESV), stroke volume, and EF calculated by both 3DE modalities underestimated MRI values. Compared to QLAB, TomTec showed better correlation and smaller intertechnique differences with MRI (the 95% limits of agreement, EDV: -20.84 to 5.18 mL in QLAB, -10.66 to 1.84 mL in TomTec; ESV: -8.94 to 3.07 mL in QLAB, -2.45 to 0.98 mL in TomTec; SV: -13.31 to 3.45 mL in QLAB, -9.34 to 2.0 mL in TomTec; EF: -12.07 to 7.76% in QLAB, -9.64 to 1.52% in TomTec), TomTec was more reproducible with better intraclass correlation coefficients and variation coefficients.
CONCLUSIONS: Both 3DE modalities tend to underestimate LV volumes, but the correlation of LV volumes and EF between 3DE and MRI still holds well. Despite a longer operating time, TomTec analysis is more accurate and reproducible.