Lempersz C, van Laar JO, Clur SB, Verdurmen KM, Warmerdam GJ, van der Post J, Blom NA, Delhaas T, Oei SG, Vullings R.
PLoS One. 2020 Apr 30;15(4):e0232606. doi: 10.1371/journal.pone.0232606. eCollection 2020.
PMID: 32353083 Free Article
Select item 32347120
Introduction: The examination of the fetal heart in mid-pregnancy is by ultrasound examination. The quality of the examination is highly dependent on the skill of the sonographer, fetal position and maternal body mass index. An additional tool that is less dependent on human experience and interpretation is desirable. The fetal electrocardiogram (ECG) could fulfill this purpose. We aimed to show the feasibility of recording a standardized fetal ECG in mid-pregnancy and explored its possibility to detect congenital heart disease (CHD).
Materials and methods: Women older than 18 years of age with an uneventful pregnancy, carrying a healthy singleton fetus with a gestational age between 18 and 24 weeks were included. A fetal ECG was performed via electrodes on the maternal abdomen. After removal of interferences, a vectorcardiogram was constructed. Based on the ultrasound assessment of the fetal orientation, the vectorcardiogram was rotated to standardize for fetal orientation and converted into a 12-lead ECG. Median ECG waveforms for each lead were calculated.
Results: 328 fetal ECGs were recorded. 281 were available for analysis. The calculated median ECG waveform showed the electrical heart axis oriented to the right and inferiorly i.e. a negative QRS deflection in lead I and a positive deflection in lead aVF. The two CHD cases show ECG abnormalities when compared to the mean ECG of the healthy cohort.
Discussion: We have presented a method for estimating a standardized 12-lead fetal ECG. In mid-pregnancy, the median electrical heart axis is right inferiorly oriented in healthy fetuses. Future research should focus on fetuses with congenital heart disease.
Conflict of interest statement
R. Vullings is a shareholder in Nemo Healthcare BV, the Netherlands. S.G. Oei initiated the scientific research from which Nemo Healthcare originated, there is no financial relationship between Nemo Healthcare and S.G. Oei. All other authors have declared that no competing interests exist.
Fig 1. Measurement setup. Measurement setup with eight electrodes on the maternal abdomen (six recording electrodes, one common reference (ref), one ground (gnd)) and the prototype fetal monitoring system. The bipolar channels are indicated by the arrows and formed by the electrodes 1–6 with respect to the common reference (eg 1 –ref, 2 –ref). The positions of the electrodes and lead vectors of the recorded channels are defined within the xyz-axis system depicted on the bottom right.
Fig 2. Schematic illustration of signal processing to obtain standardized fetal ECG. Schematic illustration of signal processing steps followed to obtain the standardized fetal ECG. From the top left, the consecutive steps are depicted in a clockwise direction. In the first step the raw data was filtered to suppress the maternal ECG. In the next step, the fetal ECG was further enhanced by averaging the ECG over 30 heartbeats. Multi-channel ECG complexes were subsequently combined to calculate the vectorcardiogram (VCG). This VCG is described within a coordinate system (xyz) that is defined with respect to the maternal body. Based on the fetal orientation assessed by ultrasound examination, a mathematical rotation R of the VCG was performed to convert the VCG to a coordinate system (x’y’z’) that is defined with respect to the fetal body. Finally, the rotated VCG was transformed with the Dower matrix to yield an estimate for the standardized, 12-lead fetal ECG.
Fig 3. Rotation of the VCG between heartbeats due to fetal movement. In black the VCG at time t = 0, in red the VCG at time t = 1. Due to fetal movement there was a rotation of the VCG.
Fig 4. Normal ECG waveform. Normal ECG waveform with in black the median over 281 healthy fetuses and in grey the interquartile range, shown for lead I (A) and lead aVF (B).
Fig 5. 12-lead normal fetal ECG in mid-pregnancy. 12-lead ECG of the normal fetal heart at 20 weeks of gestation, calculated as the median over 281 healthy fetuses. Note the rightward QRS axis and the right ventricular dominance (positive R wave in V1 with deep S in V6 and failure of R wave progression precordially). The marker on the bottom left indicates the scale at which an amplitude of 1 μV is depicted.
Fig 6. 12-lead fetal ECG of left isomerism and atrioventricular septal defect. 12-lead ECG of case 1 with the atrio-ventricular septal defect: Note the abnormal QRS axis (-45 degrees) and prominent left ventricle (prominent R in V5 and V6 and deep S in V1 and V2 as well as the negative aVR). The marker on the bottom left indicates the scale at which an amplitude of 1 μV is depicted.
Fig 7. 12-lead fetal ECG of hypoplastic right heart. 12-lead ECG of case 2 with a hypoplastic right heart (hypoplastic right ventricle, tricuspid stenosis and dysplastic pulmonary valve): Note the abnormal QRS axis (+60 degrees) and left ventricular dominance (prominent R waves in V5 and V6). The marker on the bottom left indicates the scale at which an amplitude of 1 μV is depicted.
Fig 8. ECG of cases compared to normal ECG. Leads I (A) and aVF (B) for the two CHD cases, plotted together with the IQR of the normal ECG. The gray area represents the IQR of the normal ECG, the solid line represents the fetus with atrio-ventricular septal defect (Fig 6) and the dashed line represents the fetus with hypoplastic right heart (Fig 7).