Prenatal diagnosis of a bundle branch block based on the fetal ECG

Abstract

A non-invasive fetal ECG was performed on a 36-year-old pregnant woman at 24+6 weeks of gestation as part of ongoing clinical research. A paediatric cardiologist suspected an incomplete bundle branch block based on the averaged ECGs from the recording. The characteristic terminal R’ wave was present in multiple leads of the fetal ECGs. A fetal anomaly scan had been performed at 20 weeks of gestation and showed no abnormalities. An incomplete right bundle branch block was confirmed on an ECG recorded at the age of 2 years. This case shows the possibility of novel non-invasive fetal ECG technology as an adjunct to the diagnosis of fetal cardiac anomalies in the future.

Background

Congenital heart disease (CHD) is the most common major congenital anomaly worldwide, with an estimated prevalence of 8 per 1000 live births.^1–3 About 20%–30% of all CHDs are severe, in that they require early surgery or catheter intervention.^4–6 Despite the decrease in mortality rates over the last decade due to improved diagnostics and treatment techniques, CHD remains the leading cause of infant mortality in developed countries.^{7 8} Timely detection of CHD has important advantages: it allows for close monitoring during pregnancy, planning the delivery in a centre with the required treatment facilities and it keeps the option of pregnancy termination open for the parents if the diagnosis is made before 24 weeks gestation. Prenatal diagnosis of CHD has shown to increase survival prior to planned cardiac surgery and decreases preoperative morbidity, especially in the case of ductal-dependent lesions.^{9 10}

Screening for CHD is performed through ultrasound examination by means of the second-trimester anomaly scan around 20 weeks of gestation. Since the introduction of national screening programmes around Europe, the detection rate for CHD in the low-risk population has increased up to 50%–60%.^{4 11–13} However, the detection rate is strongly correlated with the severity of the CHD and highly dependent on the sonographer’s experience.¹⁴ In some specialised tertiary care centres, the general detection rate has risen to 89%.¹⁵ However, only 10% of infants born with CHD are born to mothers with known risk factors, and therefore end up in tertiary care.¹⁶ Most CHDs occur in the low-risk population, where at least 4 out of 10 cases of severe CHD are still missed.^{4 11–13}

Non-invasive fetal ECG (NI-fECG) in addition to ultrasound screening might raise the detection rate for CHD in low-risk populations, by providing a multilead fetal ECG. To date, the diagnosis of significant ECG abnormalities has been difficult prenatally. We report a case where a bundle branch block was diagnosed on an NI-fECG measurement performed at 24 weeks of gestation, as part of ongoing clinical research.¹⁷

Case presentation

A 36-year-old gravida 2 para 1 had a fetal anomaly scan as part of prenatal screening at 20 weeks of gestation. No abnormalities were seen at this time. As part of ongoing clinical research, she received a single fetal ECG measurement of 17 min at 24+6 weeks of gestation. There was no family history of (congenital) cardiac pathology.

Investigations

Non-invasive fetal ECG

The recording was performed with a prototype fetal ECG system (Nemo Healthcare BV, the Netherlands). An electrode patch consisting of eight adhesive electrodes was placed on the maternal abdomen, including one ground and one reference electrode (figure 1). An averaged fetal ECG was calculated for each of the six recording electrodes through a series of signal processing techniques. First, the maternal signal was suppressed using a dynamic template subtraction technique.¹⁸ The signal to noise ratio was then enhanced by spatially combining the remaining signals to filter out electrophysiological interferences from, that is, muscle activity.¹⁹ In this enhanced signal, fetal QRS complexes were detected using a low-complexity R-peak detection method.²⁰ Because fetal QRS complexes occur in each of the six recorded channels at the same time, the QRS complexes, which were detected in the spatially combined signal, were used for each of the six individual channels to segment the recording in individual ECG complexes. These ECG complexes were subsequently averaged across multiple heartbeats, where the number of heartbeats was dynamically varied by an adaptive Kalman filter, depending on the quality and stationarity of the ECG signal, to produce a further enhanced fetal ECG (figure 2).²¹

Figure 1.

The non-invasive fetal ECG. The left picture shows the electrode patch attached to the maternal abdomen, which is connected through a single wire to an amplifier and the base station. The right picture shows the electrode patch with the six recording electrodes (numbered), the GND and the REF. GND, ground electrode; REF, reference electrode.

Figure 2.

Averaged fetal ECG for each electrode. Averaged fetal ECG for each of the six recording electrodes. Channel numbers match those of the corresponding electrodes as displayed in figure 1. Terminal R wave is visible in channel 4–6. Ch, channel.

Differential diagnosis

The six averaged ECGs were presented to a paediatric cardiologist (figure 2). Based on the morphology of the QRS complex, incomplete bundle branch block was suspected. In the absence of information on the fetal orientation, the location of the bundle branch block could not be determined.

Outcome and follow-up

At 40+2 weeks of gestation, a spontaneous vaginal delivery took place. A male neonate was born weighing 3510 g and with an Apgar score of 8 and 9 at 1 and 5 min, respectively. He showed no clinical signs of cardiac pathology. A cardiac ultrasound and ECG were performed at the age of 2 years after a cardiac murmur was heard. The presence of an incomplete right bundle branch block (RBBB) was confirmed. The ECG showed the characteristic late R’ wave in the right precordial lead V1 (figure 3). Duration of the QRS complex (78 ms) lies within the normal ranges described in the literature for this age group.^{22 23} The child is alive and well and has no cardiac symptoms.

Figure 3.

Right precordial lead V1 from the ECG recorded at 2 years of age. The QRS complex shows the characteristic pattern of an incomplete RBBB with an additional terminal R-peak, reflecting the delayed activation of the right ventricle. The QRS complex is not widened (78 ms). RBBB, right bundle branch block.

Discussion

A complete RBBB is an abnormality of cardiac ventricular conduction. Due to a conduction delay in the peripheral conduction system on the right side, activation and subsequent contraction of both ventricles are no longer synchronised. This causes a characteristic pattern of the QRS complex on the ECG which reflects the delayed depolarisation of the right ventricle as an additional R’ wave in the right precordial leads.²⁴ In an incomplete RBBB, the QRS complex is widened, but no more than 120 ms. It is a common finding in paediatric ECGs,^25–30 and in the absence of an underlying cardiac condition, is usually a benign finding without clinical consequences.³⁰

To our knowledge, we report the first case where an incomplete RBBB was detected prenatally on a fetal ECG. NI-fECG uses electrodes placed on the maternal abdomen (figure 1). This non-invasive technique has first been described as early as 1906 by Cremer, but development has lagged behind due to technical challenges.³¹ The maternal electrophysiological signal dominates that of the fetus, making it extremely challenging to extract the fetal ECG with sufficient quality to enable use in diagnostics.³² Other factors, such as the amniotic fluid, the vernix caseosa and fetal movements, also affect the quality of the fetal signal.^{33 34} With the improvement of technology, it is now possible to effectively suppress the maternal signal. Since this is relatively new technology, little is known about the normal ECG morphology in the fetus. The fetal circulatory system has a unique shunting system, bypassing the lungs, resulting in a higher cardiac output of the right ventricle than left ventricle, compared with postnatal life. Also the right ventricle pumps against higher resistance prenatally resulting in a larger mass compared with the left ventricle. These haemodynamic differences can influence fetal ECG morphology. Previous research has shown that the electrical heart axis of the fetus points toward the right.³⁵ Until now, only a few studies have been published studying fetal ECG waveform characteristics.^36–43 Differences in study design (antepartum vs intrapartum), NI-fECG systems used to acquire the signals, and signal processing techniques have led to the large heterogeneity in the studies which complicates combined interpretation of the results. A large variation in gestational age at time of fetal ECG registration is seen between the studies. As the fetal cardiac mass grows with increasing gestational age, cardiac time intervals can be expected to change accordingly.^{37 44}Standardising groups based on gestational age is indispensable when trying to define normal ranges for the population. Knowledge of normal ECG morphology in the fetus is essential before this technology can be clinically implemented for the detection of abnormal ECG variations.

Typically, converting the fetal ECG to a 12-lead ECG aids the interpretation for trained clinicians, since this is the standard format used for postnatal ECGs. Such conversion to a 12-lead fetal ECG has been described in Vullings 2010, but is not yet available for clinical practice.⁴⁵ Moreover, this conversion requires information on the fetal orientation, which was absent in this case. Due to the missing fetal orientation, it was unknown which vector with respect to the heart each of the electrodes comprised and thus the location of the conduction delay (ie, right or left) could not be determined.

Although the incomplete bundle branch block was benign in our patient, this is the first case describing its detection on a fetal ECG and serves as a proof of principle that ECG abnormalities can be detected prenatally using NI-fECG recordings. The NI-fECG can be used as an additional screening tool for CHD. In line with the current prenatal screening process, an NI-fECG measurement could be carried out along with the fetal anomaly scan around 20 weeks of gestation. When combining the NI-fECG technology with the fetal anomaly scan, fetal orientation is available and a 12-lead fetal ECG can become available for interpretation. Abnormalities seen on the NI-fECG should encourage additional awareness for the cardiac anatomy during the anomaly scan. Only in the case that the ultrasound confirms the abnormality or, at the least, cannot refute the abnormality, should the patient be referred for advanced ultrasound examination by an experienced gynaecologist and/or paediatric cardiologist. As this follow-up ultrasound usually takes place within a few days after referral, the expecting parents are left in uncertainty only for a limited amount of time. Since the detection of CHD in advanced ultrasound examination is high (ie, 89%),¹⁵ NI-fECG may play a role in increasing prenatal detection of various cardiac anomalies. Moreover, cardiac conduction disorders without overt anatomic anomalies which are easily missed with ultrasound may become visible on the fetal ECG.

Patient’s perspective.

When I was asked to participate in a study towards the possible use of fetal electrocardiography in monitoring the health of unborn babies in general, I was more than happy to do so. We already had the conventional anomaly scan and had heard that all was fine. Therefore, I did not feel any insecurities towards the outcome of the fetal ECG measurements and assumed I participated as part of the healthy control group. It was quite a shock when I heard that the fetal ECG of our son was abnormal. Luckily, we were put at relative ease of mind when it was explained to us that, although the ECG was abnormal, all signs pointed to an innocent abnormality. When our son was born and showed no signs of disease, I managed to almost forget about the whole thing. When he was two years old, we went to the GP because of a suspected ear infection and as part of his normal routine, the GP listened to the heart and heard some murmur. All memories of heart disease came rushing back and because of this history, the GP referred us to a pediatrician, who in turn referred us to a pediatric cardiologist. When this cardiologist confirmed the incomplete bundle branch block and assured us that she expected our son’s life to not be further affected by this abnormality, I think we could finally get some real closure.

Although in our case the ECG abnormality turned out to be relatively harmless, I still wonder what could have happened if it would have been a more serious abnormality and because of a missed diagnosis, our son would not have had the appropriate treatment. If such a simple ECG measurement can help to prevent missing some cardiac abnormalities, I hope that all expectant mothers soon get this extra check and that maybe some real damage can be prevented.

Learning points.

• Most congenital heart diseases (CHDs) occur in low-risk populations, where 4 out of 10 CHD are still missed.

• The non-invasive fetal ECG (NI-fECG) uses electrodes placed on the maternal abdomen to obtain electrophysiological information on the fetal heart.

• Until recently, it was not possible to obtain a high-quality fetal signal due to electrophysiological interference from, among others, large maternal signals.

• NI-fECG is a promising technology which can aid in diagnosing cardiac anomalies when combined with current screening methods.