Comparison of PDA diameter measurements using color and 2D echocardiography in newborn infants

Abbreviations

2D

2 dimensional

1 Introduction

Echocardiography is the gold standard for the diagnosis of patent ductus arteriosus (PDA) and is increasingly being used by clinicians in the neonatal intensive care unit in conjunction with the physical examination to assess the cardiovascular system of the newborn infant and make decisions about treating the PDA. Several studies have examined various echocardiographic parameters interrogating the PDA [1, 2] to assist clinicians with optimal management strategies, but PDA diameter remains one of the most frequently measured parameters. Though it is frequently measured among several other echocardiographic parameters in the management of PDA, there is paucity of data on how comparable PDA diameters are when measured using 2D and color echocardiography. The aim of this work was to examine the agreement between PDA diameter measured using 2D and color echocardiography in newborn infants.

2 Methods

This retrospective study was carried out in a tertiary neonatal unit where newborn infants underwent echocardiography either for evaluation of a murmur or as part of cardiovascular assessment in selected infants with hypoxic ischaemic encephalopathy, sepsis or persistent pulmonary hypertension of the newborn. Infants with congenital heart disease and where the borders of the PDA were not clearly delineated on 2D echocardiography were excluded. PDA was visualised using a 4–10 Hz sector probe on Logiq S8 ultrasonography system (GE Healthcare, GE Medical Systems, UK) through the high parasternal ductal view. During the acquisition of these images by a single operator (SP), every care was taken to ensure that the color gain and scale was adjusted to ensure that color flow was contained to within the PDA vessel wall. Using color compare, three consecutive cardiac cycles were used to measure PDA diameter at the narrowest point (reflecting the point of maximal ductal constriction) where the PDA joined the left pulmonary artery in both 2D and color echocardiography (figure 1a) using PACS Vue software by a single operator (SP). The PDA 2D diameter measurements were performed first followed by color diameter measurements after 4 to 6 months, blinded to initial measurements, in order to minimise operator bias. Clinical data such as gestation, birth weight and diagnosis were obtained from electronic patient records. The mean (SD) was calculated for PDA diameter measured in 2D and color. Agreement between the two parameters was assessed using Bland-Altman plot with limits of agreement. Intra rater reliability was assessed using intraclass correlation (ICC), absolute agreement, using two-way mixed ANOVA where raters were fixed with random measurements. The grading and strength of correlation was assessed using previously described methods [3]. All statistical analysis was carried out using SPSS v26 (SPSS, Chicago, Illinois, USA). This study was reviewed and approved by the research ethics committee (22/HRA/1108).

OPEN IN VIEWER

Fig. 1a-b

Images of PDA using 2D and color echocardiography from a preterm infant. Line across the PDA demonstrates the point of measurement of PDA diameter. RPA –right pulmonary artery, LPA –left pulmonary artery, PDA –patent ductus arteriosus, 2D –2 dimensional (1a). Bland-Altman plot illustrating the bias (solid line) between PDA diameter measured using color and 2D along with the 95% limits of agreement (dashed lines) (1b).

A total of 25 infants were studied. Two infants were excluded, as the PDA border was not clearly defined. Forty-six measurements were examined from 23 newborn infants with a mean (SD, range) gestation of 28.7 (6.3, 23.0 to 40.7) weeks and mean (SD, range) birth weight of 1303 (1182, 430 to 4670) grams underwent echocardiography at a mean (range) age of 11 (0 to 54) days. The mean (SD) PDA measurement of color and 2D were 2.7 (0.8) mm and 2.2 (0.7) mm respectively. There was good correlation between PDA color and 2D measurement (r = 0.962, p < 0.0001). The mean (SD, 95% limits of agreement) bias between PDA color and 2D measurement were 0.45 (0.23, –0.005 to 0.91) mm (figure 1b). There was a trend towards smaller bias for PDA diameters below 2 mm. The intra-observer variability for PDA diameter measurement in color was 0.968 (95% CI of 0.936 to 0.985) whereas for 2D PDA diameter measurement was 0.983 (95% CI 0.959 to 0.993).

4 Discussion

This single observer study found that PDA diameter measured using colour echocardiography was, on average, 0.45 mm larger when compared to 2D echocardiography in this cohort of newborn infants. These findings were replicated in animal models [4] and is also supported by recommendations that 2D should be used for measurement of PDA diameter [5]. In older children, PDA measured at the pulmonary end using 2D echocardiography correlated well with PDA diameter measured using angiography [6]. PDA was measured at the pulmonary end as ductal constriction occurs first here than at the aortic end [7]. PDA ampulla length was not examined as this has shown to have poor correlation between measurements [8]. Though it is easier to measure the PDA diameter using color echocardiography and challenging to clearly delineate the margins of the PDA in 2D especially in infants requiring significant respiratory support, clinicians should be aware of this when using color Doppler for measurement of PDA.

This study is not without limitations. Comparison of PDA diameter using 2D and color was only performed by a single operator in a single centre. This study did not explore which of these measurements were more accurate by comparing to intra operative diameters or diameter measured using other imaging modalities. Infants whose PDA borders were not well defined on 2D echocardiography were excluded from this work, an important factor which should be explored in future studies. Intra-observer reliability was good and higher for 2D measurements when compared to color measurements. Due to the single operator nature of this study, inter-observer studies were not performed but an inter-observer variability of 0.84 (95% CI 0.722 to 0.953) has been reported [9]. Babla et al found poorer PDA diameter repeatability values than this study [10]. Potential reasons include babies of a lower gestation, who were older at scanning and the inclusion of all scans where imaging was imperfect. Though this retrospective study only examined a small sample and explored one commonly measured parameter of PDA characteristic, there was greater consistency in measurements as a single operator performed all the measurements.

5 Conclusion

This study concludes that PDA diameter measured using color echocardiography exaggerated PDA diameter measurements when compared to 2D PDA measurements by a single operator. Further studies are required to compare how well PDA diameter measured using echocardiography compares between operators and with other modalities of imaging or intraoperative findings in newborn infants. Though there was a strong correlation and clinically acceptable bias observed in this study, clinicians and researchers need to be aware of this and consider this when performing echocardiographic examination and planning echocardiographic studies involving PDA diameter measurements.

Funding source

None.

Financial disclosure

The author has indicated that he has no financial relationships relevant to this article to disclose.

Conflict of interest

No potential conflicts of interest to disclose.