Development of a Z-score equation for atrioventricular interval measurement by two-dimensional pulsed Doppler echocardiography in normal fetuses between 16 and 33+6 weeks of gestation

Article information

Ultrasonography. 2025;44(3):212-219

Publication date (electronic) : 2025 April 30

doi : https://doi.org/10.14366/usg.24142

Luciane Alves da Rocha Amorim¹, Nathalie Jeanne Bravo-Valenzuela², Sophia Livas de Morais Almeida¹, Alberto Borges Peixoto³, Luiz Henrique Soares Nicoloso⁴, Márcia Moura Schmidt⁴, Paulo Zielinsky⁴, Edward Araujo Júnior^,5,6

¹Postgraduate Program in Health Sciences, Medical School, Federal University of Amazonas (UFAM), Manaus, AM, Brazil

²Department of Pediatrics, Pediatric Cardiology, School of Medicine, Federal University of Rio de Janeiro (UFRJ), Rio de Janeiro, RJ, Brazil

³Department of Obstetrics and Gynecology, Federal University of Triângulo Mineiro (UFTM), Uberaba, MG, Brazil

⁴Fetal Cardiology Unit, Institute of Cardiology of Rio Grande do Sul, Porto Alegre, RS, Brazil

⁵Department of Obstetrics, Paulista School of Medicine, Federal University of São Paulo (EPMUNIFESP), São Paulo, SP, Brazil

⁶Discipline of Woman Health, Municipal University of São Caetano do Sul (USCS), São Caetano do Sul, SP, Brazil

Correspondence to: Edward Araujo Júnior, PhD, Department of Obstetrics, Paulista School of Medicine - Federal University of São Paulo (EPM-UNIFESP), Rua Belchior de Azevedo, 156 apto. 111 Torre Vitoria Vila Leopoldina, São Paulo, SP, CEP 05089-030, Brazil Tel, Fax. +55+011-99368-5430 E-mail: araujojred@terra.com.br

Received 2024 July 25; Revised 2025 January 30; Accepted 2025 March 11.

Abstract

Graphic Abstract

Introduction

Fetal echocardiography has transformed the diagnosis and management of congenital heart disease (CHD) and fetal arrhythmias since its introduction in the late 20th century [1]. This tool has led to a deeper understanding of fetal cardiac physiology and revolutionized prenatal care by enabling the early detection of cardiac anomalies [1]. Given that postnatal CHD occurs in approximately 1% of cases, early surveillance and diagnosis are essential for timely interventions that can significantly affect pregnancy outcomes and long-term neonatal health [2-4].

Advancements in technology have also propelled fetal cardiology forward. Measurement of the atrioventricular (AV) interval via fetal echocardiography has emerged as a promising indicator of fetal cardiac health. This measurement reflects electrical conduction between the atria and ventricles and may serve as an early marker for detecting cardiac conduction defects [5]. Although existing data support the utility of this parameter in predicting complications such as congenital AV block [5-11], establishing precise reference values remains crucial for its clinical application.

Congenital AV block is a rare but serious arrhythmia affecting approximately 1 in 15-20,000 births [5-11]. It is often caused by the transfer of maternal autoantibodies that lead to inflammation and fibrosis in the fetal myocardium and AV node [5-11]. Risk factors include the use of certain medications, viral infections, structural heart disease, and maternal autoimmune disease [9-11]. Early and accurate identification is crucial, highlighting the importance of AV interval measurement [11].

This article proposes the development of a reference range for the AV interval using a Z-score equation. The goal is to standardize these values and facilitate their practical application, ultimately improving fetal health planning.

Materials and Methods

Interventions

Two-dimensional fetal echocardiography using pulsed Doppler was performed. The AV interval was measured as the time between the onset of the mitral valve A wave and the onset of the aortic valve systolic flow, with both measurements obtained from the same tracing (Fig. 1).

Fig. 1.

Measurement of the atrioventricular interval using pulsed Doppler on fetal echocardiography.

Results

During the data collection period, 1,309 fetal echocardiograms were performed. After applying the eligibility criteria, 1,183 pregnant women were included in the final analysis. The mean age of the participants was 29.7±6 years, with 21% aged 35 years or older. The mean BMI was 27.6 kg/m², and the mean gestational age at the time of fetal echocardiography was 26.5 weeks. Approximately 80% of the pregnant women underwent routine screening (Table 1).

Table 1.

Baseline characteristics of the pregnant women (n=1,183)

A very weak but statistically significant correlation was observed between AV interval measurement and gestational age (r=0.16, P<0.001) (Fig. 2). Similarly, the relationship between AV interval measurement and fetal heart rate was weak (r=-0.21, P<0.001) (Fig. 3).

Fig. 2.

Pearson correlation coefficient assessing gestational age and atrioventricular (AV) interval – very weak correlation (r=0.16), but statistically significant (P<0.001).

Fig. 3.

Pearson correlation coefficient assessing fetal heart rate and atrioventricular interval – weak correlation (r=-0.21), but statistically significant (P<0.001).

Based on the fetal echocardiographic measurements, the 10th, 50th, 90th, and 95th percentiles were determined for each gestational age (Table 2).

Table 2.

The 10th, 50th, 90th, and 95th percentiles of atrioventricular interval measurements (ms) in normal fetuses between 16 and 33+6 weeks of gestation

The Z-score equation for AV interval measurement in milliseconds was derived using the following formula: Z-score=(AV interval measurement-111.3)/8.6.

This equation is based on the mean (111.3) and standard deviation (8.6) of the AV interval measurements obtained during the study. The normal distribution of these measurements, as evidenced by skewness and kurtosis values of 0.07 and 2.7 respectively (and confirmed by the Q-Q plot in Fig. 4), supports the use of this equation without additional transformations. The AV Z-score intervals are provided in Table 3.

Fig. 4.

Quantile-quantile (Q-Q) plot evaluating normality of atrioventricular (AV) interval measurement.

Table 3.

AV intervals Z-scores for gestational age

Discussion

This study presents a Z-score equation and percentile reference values for fetal AV interval measurement in a large sample covering gestational ages from 16 to 33+6 weeks. These results will aid in monitoring fetuses of pregnant women with anti-SSA and anti-SSB antibodies, thereby facilitating the early diagnosis of AV conduction delay and enabling timely therapy [6-9].

Several studies have sought to define the optimal AV interval cutoff value and to evaluate the benefits of corticosteroids in cases of suspected or confirmed AV block [1,15-18]. Donofrio et al. [1] compared fetuses from pregnancies positive for anti-Ro/SSA antibodies with a control group by measuring the AV interval (using Doppler of the left ventricular inflow and outflow tracts) and the AV interval (using magnetocardiography). Their study found no significant differences, with AV interval cutoff values for diagnosing first-degree block ranging from 151 to 167 ms between 16.5 and 38.5 weeks—corresponding to Z-scores >+3. Notably, three of the four cases with AV interval and PR Z-scores above +3 exhibited postnatal intra-atrial and intraventricular block despite dexamethasone treatment. Bergman et al. [17] enrolled fetuses from healthy pregnant women with and without positive anti-SSA antibodies between 18 and 24 weeks and, using simultaneous Doppler recordings of left ventricular inflow and outflow as well as superior vena cava a-wave to aortic flow (SVC-Ao), identified abnormal AV interval cutoff values of 134-138 ms and 132-138 ms, respectively. Friedman et al. [19] in the PRIDE study employed a 150 ms cutoff for prolonged AV intervals in fetuses of women with positive anti-SSA antibodies and noted that complete AV block could develop within one week of a normal echocardiogram; they also suggested that tricuspid regurgitation and increased atrial echodensity might serve as markers for complete AV block. In contrast, Eliasson et al. [8] reported that early corticosteroid administration (6-7 days after the onset of block) might reverse AV block in fetuses of anti-SSA–positive mothers.

Discrepancies in the literature regarding the cutoff for diagnosing AV conduction delay and the benefit of corticosteroids in pregnancies with positive anti-SSA and/or anti-SSB antibodies have been noted [6,8,9,18,20-23]. These controversies likely stem from the rarity of the condition, which complicates a comprehensive understanding of its progression from early detection. Establishing a clearer understanding of normal AV interval values is crucial for a more accurate early diagnosis. This study addresses this need by thoroughly evaluating AV interval measurements in a robust sample spanning a wide range of gestational ages [7,16,24].

The reference values developed in this study make it possible to more confidently identify when a fetus is progressing toward AV conduction delay and to direct therapy at the optimal time. Improved early diagnosis allows for prompt treatment and a more reliable evaluation of therapeutic outcomes. The development of Z-score curves based on normal AV interval values facilitates the design of studies investigating corticosteroid use in the prevention of congenital AV block, reducing uncertainty regarding the normality of measurements obtained from fetuses of treated mothers.

Based on the literature, it is proposed that AV interval values can be classified as normal when the Z-score is ≤2.0 and abnormal when the Z-score exceeds 2. Abnormal values may be further stratified into three levels of AV conduction delay: Z-score between 2.0 and 3.0, between 3.0 and 4.0, and greater than 4.0 [8].

A previous study by a Chinese group provided important reference values for the AV interval expressed as Z-scores for gestational age, using three different measurement techniques—Tissue Doppler, SVC-Ao Doppler, and pulmonary vein–pulmonary artery (PV-PA) Doppler. Although their analysis was based on a sample of 227 fetuses, their work significantly contributed to the understanding of fetal AV conduction [25]. This study focused on a standardized measurement technique—pulsed Doppler with simultaneous acquisition of left ventricular inflow and outflow—which ensured methodological consistency. By evaluating a much larger cohort of 1,183 fetuses, this study robustly demonstrated that AV interval measurements follow a normal distribution. This finding strengthens the reliability of the Z-score equation described in this study and enhances its clinical applicability. In addition to providing Z-scores, the present study established percentile-based reference values, offering a comprehensive tool for fetal echocardiographic assessment.

Although fetal AV block can occur suddenly, guidelines recommend weekly or bi-weekly monitoring of the AV interval between 16 and 28 weeks [1,26]. Corticosteroid therapy is generally considered when the AV interval exceeds 150 ms in fetuses of pregnant women with positive autoantibodies [19,20,26]. In this study, the maximum AV interval values observed were 118 ms at 16 weeks and 127 ms at 33+6 weeks of gestation. Consequently, to minimize errors in the early detection of AV block, it is preferable to express AV interval values as percentiles and Z-scores.

The present study also noted differences in the Z-score equation compared to those reported by Donofrio et al. [1] and Pan et al. [25]. Donofrio et al. [1] employed a polynomial regression model incorporating gestational age squared (GA²) to account for non-linear variations, whereas Pan et al. [25] used multiple regression models that included both gestational age and fetal heart rate—resulting in more complex equations. In contrast, this study utilized a linear Z-score equation based on the normal distribution of AV interval measurements in a significantly larger cohort. While Donofrio et al. [1] focused on a high-risk population and Pan et al. [25] applied different Doppler-based methods, the aim of this study was to provide a clinically practical and statistically robust reference for routine fetal echocardiography. These methodological differences reflect variations in study design, sample characteristics, and the specific objectives in applying Z-score equations within fetal cardiology.

This study also observed that fetal AV interval measurements tend to exhibit a negative relationship with fetal heart rate and a positive relationship with gestational age, which is consistent with previous findings [15,16,22,27-30]. Given the ease and reproducibility of measuring the AV interval using the left ventricular inflow and outflow technique with pulsed Doppler, the authors advocate for its routine inclusion in all fetal echocardiograms. Employing this normality curve will provide fetal cardiologists with greater confidence in diagnosing AV conduction delay, thereby positively impacting management. Moreover, multidisciplinary collaboration among fetal cardiologists, pediatricians, obstetricians, and other specialists will be essential for early therapeutic intervention and improved outcomes.

A limitation of this study is that the AV interval was measured in a single plane (left ventricular inflow and outflow) without employing additional planes to obtain the spectral curve [6-8,16,24]. However, this approach is recognized as the simplest and most reliable, enabling the measurement of the AV interval in virtually all fetuses [29-31]. In cases where proper identification of the left ventricular inflow E and A waves is challenging—such as in early gestation—alternative methods for AV interval assessment may be required [17,27,28,30]. Examples include simultaneous pulsed Doppler recordings from the SVC-Ao, PV-PA, or antegrade telediastolic arterial flow (Fig. 5) [15,27,28,30]; additionally, tissue Doppler imaging may be used, particularly when fetal positioning is unfavorable [15,22,27]. Another limitation is that not all fetal echocardiograms included more than one measurement per spectral curve; however, the large sample size minimizes this potential bias [15,16,22,27,28,30].

Fig. 5.

Fetal echocardiogram showing images of how to measure the atrioventricular (AV) interval using different techniques.

A. Doppler of superior vena cava (SVC)–aorta (Ao) from the onset of the retrograde A-wave of SVC Doppler (A) to the onset of the forward S-wave of aortic flow (V). B. Pulmonary artery (PA)–pulmonary vein (PV) Doppler: note the red line showing the measurement of the AV interval from the beginning of the pulmonary venous A wave to the beginning of the pulmonary arterial ejection flow (S wave). C. Antegrade Late Diastolic Arterial blood Flow (ALDAF): from the beginning of the retrograde A wave (A) to the beginning of the forward S wave in the aortic or pulmonary arterial flow (V). D. This image shows how the Doppler sample should be positioned (red arrow) in the ventricular outflow tract to obtain the ALDAF. A, atrial contraction; V, ventricular contraction; RV, right ventricle.

The main strengths of this study are the robust number of participants and the wide range of gestational ages examined. The use of the left ventricular inflow and outflow plane for AV interval measurement is straightforward and reproducible. Advances in technology have yielded devices with excellent image quality and sensitivity for spectral curve acquisition, and this study encountered no technical difficulties in obtaining the spectral curve or measuring the AV interval regardless of gestational age. Given the simplicity and reproducibility of this method—and its positive correlation with gestational age—we strongly support the use of reference curves expressed as percentiles or, ideally, as Z-scores for monitoring AV conduction in fetuses.

A reference range and Z-score equation for AV interval measurement were established. These findings may facilitate more accurate monitoring of fetuses at risk for developing AV block, particularly in pregnant women with anti-SSA and anti-SSB antibodies, thereby enabling earlier diagnosis and treatment.

Article information Continued

Notes