Feasibility and reliability of Doppler flow recordings in the fetal aortic isthmus: a multicenter evaluation

Abstract

Objectives: To evaluate the performance of three different centers with respect to their ability to identify the fetal aortic isthmus (AoI) adequately and place a Doppler sample volume in the AoI correctly, and to address the reproducibility of the isthmic flow index (IFI) calculated from Doppler waveforms recorded in the three centers.

Methods: The three collaborating centers sent several ultrasonographic recordings taken at random over a 6-week period to the Saint-Justine Fetal Cardiology Unit (StJ-FCU). A performance quotient ((number of total readings - number of unsatisfactory results)/number of total readings) was calculated for each center by each of three judges, who were experienced fetal cardiologists, to assess the ability of each center to identify the isthmus and to place the Doppler sample volume (DSV) adequately. Intraclass correlation coefficients (ICC) were computed to quantify the variability of IFI measurements ((systolic + diastolic)/systolic flow velocity integrals).

Results: Fifty-five recordings were available for this study. Concerning isthmus identification, there was 100% agreement between the three judges from StJ-FCU and the performance quotients of Centers A, B and C were: 0.90, 0.95 and 1.00, respectively. For DSV positioning, agreement between the judges varied; for Judge 1 vs. Judge 2, kappa = 0.836 (95% CI, 0.651-1.000); for Judge 1 vs. Judge 3, kappa = 0.773 (95% CI, 0.557-1.000); for Judge 2 vs. Judge 3, kappa = 0.941 (95% CI, 0.805-1.000). The performance quotients of the three centers for DSV positioning were consistently lower than were those for identification of the isthmus, being 0.85, 0.76 and 0.92, respectively. The ICC between the first and second measurements of the IFI by Rater 1 was 0.96 (95% CI, 0.93-0.98, P < 0.001) and that between Raters 1 and 2 was 0.97 (95% CI, 0.95-0.99, P < 0.001).

Conclusion: Adequate imaging of the fetal AoI can be achieved easily by a trained sonographer, while DSV positioning is challenging. The intra- and interrater variability of the IFI are low.