Cardiac defects, the most common form of congenital anomaly, are found in 2–8 of every 1000 pregnancies. Increased nuchal translucency (NT) on ultrasound is an established marker for first-trimester detection of congenital heart defects (CHDs), with or without structural anomalies, in chromosomally normal and abnormal fetuses. However, due to modest sensitivity and specificity, many fetuses with increased NT who do not have CHD may undergo detailed cardiac scans. Moreover, NT screening may miss some fetuses with CHD and normal karyotype. Ductus venosus (DV) Doppler waveform has been investigated to improve the diagnostic discrimination of ultrasound among cardiac anomalies in chromosomally abnormal and normal fetuses. DV has shown variable results in several studies in fetuses with normal karyotype; its value for identifying fetuses with CHD in the setting of a normal versus an increased NT is unclear.
The aim of this meta-analysis was to evaluate the diagnostic performance of first-trimester DV for detection of CHD in chromosomally normal fetuses. A search of MEDLINE, ISI, SCOPUS, and EMBASE was performed for studies published between 1999 and 2011. Key words included “ductus venosus,” “DV,” “chromosomal abnormalities,” “congenital heart disease,” and “nuchal translucency.” The diagnostic accuracy of DV was examined in 7 studies involving 50,354 fetuses with normal karyotype regardless of NT status, 9 studies including 2908 fetuses with increased NT, and 7 studies involving 47,610 fetuses with normal NT. Hierarchical summary receiver operating characteristic curves were drawn based on parameters of the fitted models.
Analysis of the hierarchical summary receiver operating characteristic curves showed that the sensitivity and specificity of DV to detect CHD in the normal karyotype population regardless of the NT status were 0.50 and 0.93, respectively. Summary sensitivity and specificity were 83% and 80% among participants with increased NT, and 19% and 96% among those with normal NT, respectively.
The findings of this meta-analysis showing the estimated performance of DV for detecting CHD in chromosomally normal fetuses should be considered in evaluating the potential use of this screening test and its limitations.