Intra- and Inter-Observer Variability in PDA Size Measured on Transthoracic Echocardiography (TTE) in Extremely Preterm Babies.

Background: The internal diameter of patent ductus arteriosus (PDA) measured on transthoracic echocardiography (TTE) is commonly used in determining the need for treatment in clinical practice and enrolment in research trials. PDA diameter measured on TTE has not been well validated.

Objective: To assess the intra- and inter-observer variability in PDA internal diameter measured on TTE on the same examination in babies born at ≤30 weeks’ gestation.

Methods: We included 56 babies born ≤30 weeks gestation that had clinically indicated TTEs. TTEs were performed by a neonatologist (AK) certified in echocardiography. PDAs were visualised on high parasternal views, and cineloops were stored without measurement annotations.

PDAs were measured at the narrowest diameter on 2-Dimensional (2D) and Colour Doppler echocardiography. A second certified neonatologist (DD) measured the internal diameter of PDA on the same cineloops. They were blinded to each other’s measurements. AK repeated measurements at a later date, blinded to his original measurements.

Results: Table 1 shows results for inter-observer comparisons. Fig 1 and 2 show intra-observer measurement comparisons. Comparison of all measurements showed colour measurements a mean of 0.29 mm higher than 2D.

Table 1. Bland-Altman Analysis of inter-observer measurement comparisons.

Fig 1. Intra-observer measurement comparisons (2D).

Fig 2. Intra-observer measurement comparisons (colour).

Conclusion: The poor repeatability coefficients and repeatability indices lead us to conclude that there is significant inter- and intra-observer variability in 2D and Colour echocardiography measurements of PDA diameter, even on the same cineloops.Thresholds and timing of treatment for PDA remain controversial. We advise caution when using diameter as a surrogate marker for degree of shunting.