Transesophageal Doppler measurement of renal arterial blood flow velocities and indices in children.

BACKGROUND: Doppler-derived renal blood flow indices have been used to assess renal pathologies. However, transesophageal ultrasonography (TEE) has not been previously used to assess these renal variables in pediatric patients. In this study, we (a) assessed whether TEE allows adequate visualization of the renal parenchyma and renal artery, and (b) evaluated the concordance of TEE Doppler-derived renal blood flow measurements/indices compared with a standard transabdominal renal ultrasound (TAU) in children.

METHODS: This prospective cohort study enrolled 28 healthy children between the ages of 1 and 17 years without known renal dysfunction who were undergoing atrial septal defect device closure in the cardiac catheterization laboratory. TEE was used to obtain Doppler renal artery blood velocities (peak systolic velocity, end-diastolic velocity, mean diastolic velocity, resistive index, and pulsatility index), and these values were compared with measurements obtained by TAU. Concordance correlation coefficient (CCC) was used to determine clinically significant agreement between the 2 methods. The Bland-Altman plots were used to determine whether these 2 methods agree sufficiently to be used interchangeably. Statistical significance was accepted at P ≤ 0.05.

RESULTS: Obtaining 2-dimensional images of kidney parenchyma and Doppler-derived measurements using TEE in children is feasible. There was statistically significant agreement between the 2 methods for all measurements. The CCC between the 2 imaging techniques was 0.91 for the pulsatility index and 0.66 for the resistive index. These coefficients were sensitive to outliers. When the highest and lowest data points were removed from the analysis, the CCC between the 2 imaging techniques was 0.62 for the pulsatility index and 0.50 for the resistive index. The 95% confidence interval (CI) for pulsatility index was 0.35 to 0.98 and for resistive index was 0.21 to 0.89. The Bland-Altman plots indicate good agreement between the 2 methods; for the pulsatility index, the limits of agreement were -0.80 to 0.53. The correlation of the size of the measurement and the mean difference in methods (-0.14; 95% CI = -0.28, 0.01) was not statistically significant (r = 0.31, P = 0.17). For the resistive index, the limits of agreement were -0.22 to 0.12. The correlation of the size of the measurement and the mean difference in methods (-0.05; 95% CI = -0.09, -0.01) was not statistically significant (r = 0.10, P = 0.65).

CONCLUSION: This study confirms the feasibility of obtaining 2-dimensional images of kidney parenchyma and Doppler-derived measurements using TEE in children. Angle-independent TEE Doppler-derived indices show significant concordance with those derived by TAU. Further studies are required to assess whether this correlation holds true in the presence of renal pathology. This technique has the potential to help modulate intraoperative interventions based on their impact on renal variables and may prove helpful in the perioperative period for children at risk of acute kidney injury.