Signs of first-degree heart block occur in one-third of fetuses of pregnant women with anti–SSA/Ro 52-kd antibodies

Authors


Abstract

Objective

To prospectively investigate the development of fetal heart block in anti–SSA/Ro 52-kd–positive women, and to evaluate the usefulness of serial Doppler echocardiography in detecting early signs of congenital heart block.

Methods

Twenty-four women with anti–SSA/Ro 52-kd antibodies and consequently increased risk for fetal heart block were followed up weekly, between 18 and 24 weeks of gestation, with two Doppler echocardiographic methods designed to estimate the time delay between hemodynamic events caused by atrial and ventricular depolarizations. Two hundred eighty-four women with normal pregnancies served as controls. Anti–Ro 52-kd, anti–Ro 60-kd, and anti-La antibodies were investigated by immunoblotting and enzyme-linked immunosorbent assay using recombinant proteins.

Results

In anti–Ro 52-kd–positive women, fetal atrioventricular (AV) time intervals were longer and heart rates were slightly lower compared with those in controls. Eight of 24 fetuses had signs of first-degree block. One of these fetuses had progression to complete block, and another showed recovery from second-degree block to first-degree block with betamethasone treatment. In the remaining 6 fetuses, spontaneous normalization occurred before or shortly after birth. Fetuses with normal AV time intervals at 18–24 weeks had normal electrocardiographic results at birth.

Conclusion

Anti–Ro 52-kd–positive pregnant women frequently carry fetuses with Doppler echocardiographic signs of first-degree AV block. These blocks revert spontaneously in the majority of fetuses, but progression to a more severe degree of block may occur in some. Serial Doppler echocardiographic measurement of AV time intervals is suggested as a useful method for surveillance of these high-risk pregnancies.

Several rheumatic conditions are associated with increased risk of pregnancy complications and fetal loss. Congenital heart block (CHB) without associated cardiac malformation is a rare disease in the general population, with an incidence in newborn babies of 1/15,000–1/20,000 (1). There is, however, a well-known association between CHB and placental transfer of the maternal anti-SSA/Ro and anti-SSB/La autoantibodies (2, 3). Women with these antibodies are commonly diagnosed as having Sjögren's syndrome (SS), systemic lupus erythematosus (SLE), or rheumatoid arthritis (RA), but they may also be asymptomatic. With circulating anti-Ro and/or anti-La antibodies in maternal sera, there is a 2–5% risk of giving birth to a child with CHB (4, 5), while the risk of giving birth to a second child with CHB is 12–25% (6–9). More recent studies also suggest that the risk is higher in women in whom the anti-Ro activity is targeted to the 52-kd component of the antigen rather than to the 60-kd component (9–11). CHB is most frequently diagnosed at 18–24 weeks of gestation, when it results in fetal bradycardia (6, 12). CHB has been associated with substantial perinatal mortality (20–30%), and the majority of children born alive require a pacemaker at an early age (6, 7, 13–15).

Early treatment with fluorinated glucocorticoids has been demonstrated to improve atrioventricular (AV) conduction in fetuses with second-degree AV block and to improve cardiac function in complete AV block (12, 16–20). A complete heart block is, however, commonly considered permanent, but it still remains to be determined whether a third-degree block might be reversible if therapy is initiated immediately upon occurrence.

Assuming that CHB is a gradually progressing disease, a method to diagnose a first-degree AV block should have the potential for early detection of CHB, before it becomes complete. Using standard fetal echocardiographic techniques, atrial and ventricular depolarizations can be identified indirectly by their mechanical (M-mode) or hemodynamic (Doppler) consequences. Recent experimental (21) and clinical (22) studies have demonstrated the superiority of the Doppler technique compared with the M-mode approach for measuring fetal AV time intervals and thus for diagnosing first-degree AV block. Reference values obtained by using different Doppler techniques have been established (23, 24).

Based on these observations and ideas, we devised a protocol for early surveillance of pregnancies at risk for fetal CHB. In addition to the goal of providing early therapy to fetuses developing CHB, a second goal was to investigate whether and to what extent these fetuses demonstrated Doppler echocardiographic abnormalities suggesting first-degree AV block. Such observations would indicate that measuring AV time intervals could be a useful and reliable method for early detection of fetal CHB.

PATIENTS AND METHODS

Patients.

From December 1999 through March 2003, 24 women (Table 1) with a mean ± SD age of 31.2 ± 4.7 years and singleton pregnancies were recruited to undergo weekly fetal echocardiographic examinations between 18 and 24 weeks of gestation. Inclusion criteria were a positive SSA finding on routine serology with a verified positive anti–Ro 52-kd finding by enzyme-linked immunosorbent assay (ELISA). Eight women had SS as defined by the revised European criteria (25), 11 had SLE (2 with secondary SS [25]) according to the 1982 revised criteria of the American College of Rheumatology (ACR; formerly, the American Rheumatism Association) (26), 2 had RA according to the 1987 revised criteria of the ACR (27), 1 had an undifferentiated autoimmune syndrome, and 2 were asymptomatic. Gestational age had been determined by ultrasound biometry before 18 weeks of gestation. All women gave informed consent to participate, and the study was approved by the Ethics Committee at Karolinska Hospital.

Table 1. Pregnant anti-SSA/Ro–positive women included in the study*
Diagnosis of mother (n)Most severe degree of AV block in fetusPositive for autoantibodies
IIIIIIAnti–Ro 52-kdAnti–Ro 60-kdAnti-La
  • *

    AV = atrioventricular; SS = Sjögren's syndrome; SLE = systemic lupus erythematosus; RA = rheumatoid arthritis.

  • Measured by Doppler echocardiography. Most severe degree of block recorded in each fetus is shown.

  • By enzyme-linked immunosorbent assay with recombinant antigen.

SS (8)2118 of 85 of 86 of 8
SLE (11)311 of 115 of 117 of 11
RA (2)2 of 21 of 20 of 2
Undifferentiated autoimmune syndrome (1)1 of 10 of 10 of 1
Asymptomatic (2)12 of 21 of 21 of 2

Serologic analyses.

Production and purification of recombinant Ro 52-kd, Ro 60-kd, and La proteins were performed as described previously (28). The proteins were used in ELISA and Western blotting for detection of antibodies as described previously (11).

Echocardiographic studies.

All fetal echocardiographic recordings and measurements were performed by the same examiner (S-ES) using a Sequoia ultrasound system with a 6C2 transducer (Acuson Computed Sonography, Mountain View, CA). At the first examination, a complete fetal echocardiographic study was performed to identify fetuses with cardiac malformations. At this and at the following examination, AV time intervals were measured by using two different Doppler techniques that have previously been described in detail (24).

Briefly, Doppler recordings were made in a 5-chamber view from a position recording velocities in both the mitral valve and the left ventricular/aortic outflow. The AV time interval was measured from the intersection of the mitral E and A waves to the beginning of the ventricular ejection wave (MV-Ao). To obtain a real-time picture suitable for simultaneous recordings of the superior vena cava and aorta Doppler velocities (SVC-Ao recordings), a 4-chamber view in a vertical position was first obtained, and thereafter 90° rotation allowed a longitudinal view of both the ascending aorta and the SVC in close proximity to each other. The gate of the pulsed Doppler was then opened enough to encompass both vessels. On these simultaneous SVC-Ao recordings, the AV time interval was measured from the beginning of the retrograde venous a wave to the beginning of the aortic ejection wave. Measurements of corresponding complete cardiac cycles were used to calculate heart rate from both the MV-Ao and the SVC-Ao recordings. Three consecutive beats were measured for all intervals and averaged.

Two hundred eighty-four women with normal pregnancies served as controls. The majority of these controls have already been described (24). During the time period of investigating the pregnancies in the present study, an additional 20 normal pregnancies with gestational ages of 17–22 weeks were studied. All AV time interval measurements made on these fetuses fitted well with previously obtained data and were included in the normal reference values.

Statistical analysis.

Statistical analysis was performed using the Mann-Whitney U test (Statistica 6.0; StatSoft, Tulsa, OK). Repeated observations made on the anti–Ro 52-kd–positive pregnancies were divided into 3 gestational age periods. Within each of these periods, an average value for each individual pregnancy was calculated and compared with the single observations made on control pregnancies of the same gestational age. P values less than 0.05 were considered significant.

RESULTS

Longitudinal Doppler echocardiographic observations of fetuses of anti–Ro 52-kd–positive mothers.

Twenty-four anti–Ro 52-kd–positive, singleton-pregnant women (Table 1) were followed between gestational weeks 18 and 24 with fetal echocardiography, and in some cases also at earlier and later weeks. All 24 fetuses had structurally normal hearts and normal hemodynamic findings at the first examination. Longitudinal observations made on these fetuses are illustrated in Figures 1 and 2. For both Doppler methods used, there was a clear tendency toward prolonged AV time intervals in these fetuses compared with normal control fetuses. Actually, as many as 8 of the 24 fetuses under study had values outside the 95% confidence interval for normal fetuses, suggesting an intrauterine first-degree AV block in one-third of these fetuses. One fetus progressed from an abnormal AV time interval to complete AV block in 6 days. Despite betamethasone treatment until delivery, the block was still complete at birth. Another fetus progressed from a normal AV time interval to second-degree AV block in 3 weeks. After a few days of betamethasone treatment, there was recuperation to a first-degree block. Disregarding these 2 fetuses with second-degree or complete AV block, the remaining 22 fetuses had heart rates within the normal range, but with a tendency toward lower values (Figure 2). Individual AV time estimates did not show any correlation with heart rate (Figure 3).

Figure 1.

Longitudinal observations of atrioventricular (AV) time intervals in 24 fetuses with anti–Ro 52-kd–positive mothers. Open squares represent the fetus with progression to complete block and open circles represent the fetus with second-degree block that reversed to first-degree block (see Results). Solid circles represent the other 22 fetuses. Straight lines denote the linear regression and 95% confidence limits for individual observations on normal fetuses. A, Results obtained measuring the AV time interval from the beginning of the retrograde venous a wave to the beginning of the aortic ejection wave. B, Results obtained measuring the AV time interval from the intersection of the mitral E and A waves to the beginning of the ventricular ejection wave. ms = milliseconds; GA = gestational age.

Figure 2.

Longitudinal observations of heart rate (HR) in 24 fetuses with anti–Ro 52-kd–positive mothers. Open square represents the fetus with progression to complete block and open circles represent the fetus with second-degree block that reversed to first-degree block (see Results). Solid circles represent the other 22 fetuses. Straight lines denote the linear regression and 95% confidence limits for individual observations on normal fetuses. bpm = beats per minute; GA = gestational age.

Figure 3.

Observations of AV time intervals in 24 fetuses with anti–Ro 52-kd–positive mothers, plotted against HR. A, Results obtained measuring the AV time interval from the beginning of the retrograde venous a wave to the beginning of the aortic ejection wave. B, Results obtained measuring the AV time interval from the intersection of the mitral E and A waves to the beginning of the ventricular ejection wave. Straight lines denote the linear regression and 95% confidence limits for individual observations on normal fetuses. See Figures 1 and 2 for definitions.

Transient prolonged AV conduction time is common in fetuses of anti–Ro 52-kd–positive mothers.

In the women with anti–Ro 52-kd antibodies, fetal AV time intervals measured with the MV-Ao approach were longer than those in normal fetuses during all 3 gestational age periods (Table 2). With the SVC-Ao approach, which included somewhat fewer observations during the first period, this finding could be demonstrated only for the second and third periods. A slightly lower heart rate could also be found in fetuses with anti–Ro 52-kd–positive mothers during the two later gestational age periods (Table 2). The usability of the suggested fetal Doppler echocardiographic approaches to distinguish between different degrees of AV block is demonstrated in Figures 4 and 5, which illustrate recordings from fetuses included in the study.

Table 2. Heart rates (HRs) and atrioventricular (AV) time intervals obtained with two different Doppler techniques in patients and controls at 3 gestational age (GA) periods*
 GA period, weeks
17–1920–2223–25
  • *

    Values are the mean ± SD (no. of patients or controls). bpm = beats per minute; MV-Ao = measurement from the intersection of the mitral E and A waves to the beginning of the ventricular ejection wave; SVC-Ao = measurement from the beginning of the retrograde venous a wave to the beginning of the aortic ejection wave.

  • P < 0.05 versus controls.

  • P < 0.01 versus controls.

  • §

    P < 0.001 versus controls.

HR, bpm   
 Patients147 ± 6.8 (21)143 ± 4.7 (21)142 ± 5.0 (22)
 Controls148 ± 6.5 (57)147 ± 6.5 (58)146 ± 7.0 (32)
MV-Ao AV time interval, msec   
 Patients123 ± 10.8 (21)§125 ± 11.7 (21)§129 ± 10.9 (22)§
 Controls114 ± 8.5 (55)111 ± 8.3 (56)115 ± 9.7 (31)
SVC-Ao AV-time interval, msec   
 Patients118 ± 11.7 (14)123 ± 11.3 (19)§126 ± 11.6 (22)
 Controls112 ± 7.9 (15)110 ± 7.8 (38)116 ± 9.9 (22)
Figure 4.

Doppler recordings obtained by measuring the atrioventricular (AV) time interval from the intersection of the mitral E and A waves (E and A, respectively) to the beginning of the ventricular ejection wave. Mitral inflow is shown downward. Aortic outflow is shown upward. A, A case with abnormally long time delay (boxed area) between the mitral A wave and the aortic outflow wave, indicating first-degree block. B, Second-degree block in which every second A wave is superimposed on the mitral E wave (E+A). C, Complete block with asynchrony between mitral A waves and aortic outflow waves. m/s = meters per second (sweep speed 100 mm/second in A and B; 50 mm/second in C).

Figure 5.

Doppler recordings obtained by measuring the atrioventricular (AV) time interval from the beginning of the retrograde venous a wave (a) to the beginning of the aortic ejection wave. Flow through the superior vena cava toward the heart is shown upward. Aortic outflow is shown downward. Note the small retrograde venous a wave corresponding to atrial contraction. Shown are the same cases illustrated in Figure 4. A, A case with abnormally long time delay (boxed area) between the venous a wave and the aortic outflow wave, indicating first-degree block. B, Second-degree block in which every second a wave is followed by an aortic outflow wave. C, Complete block with asynchrony between venous a waves and aortic outflow waves. m/s = meters per second (sweep speed 100 mm/second in A and B; 50 mm/second in C).

All women included in the study have delivered their babies, 21 at term and 3 at gestational ages of 33–35 weeks. Results of electrocardiograms (EKGs), usually performed during the first days after birth, were normal in 19 babies and showed a first-degree AV block in 4. Three of these babies had had Doppler echocardiographic signs of first-degree block as fetuses, and the fourth was the fetus with second-degree block that had converted to first-degree block during betamethasone treatment. Within a few weeks, these 4 babies also had normal EKG results. The fetus diagnosed as having complete block was delivered by cesarean section at term. EKG performed after birth confirmed the diagnosis. All of the babies are alive and well.

DISCUSSION

Isolated complete congenital heart block is associated with transfer of maternal anti-Ro/SSA autoantibodies to the fetus. The final stage—a complete third-degree AV block—of the presumed gradual development from normal conduction activity is usually discovered before 24 weeks of gestation. However, little is yet known about the natural progression of this conduction failure, nor is it known whether an observed first-degree block will unconditionally lead to more advanced disease.

An objective of the present study was to determine whether fetuses at risk for developing complete AV block had signs of first-degree block, as documented by fetal Doppler echocardiography. With each of two different techniques, this could be demonstrated in 8 of 24 fetuses. A higher degree of block also occurred in two of these fetuses. In one fetus, there was a progression from first-degree block to complete block, while in the other, recovery from second-degree block to first-degree block was observed, with subsequent complete normalization. Interestingly, 6 fetuses had indirect signs of first-degree block without progressing to a higher degree of AV conduction abnormality. Three of these fetuses had normal EKG results at birth, and 3 still had first-degree AV block. A few weeks after birth, the latter 3 babies also had normal EKG results. No later occurring conduction abnormalities were found in the 16 fetuses with normal AV time intervals during early midtrimester, and they all had normal EKG results at birth.

To our knowledge, these observations are the first reported prospective data to support the idea that CHB is a gradually developing disease starting with a first-degree block. Our findings also indicate that first-degree AV block can be present in the fetus without progressing to complete AV block, and that a first-degree block is spontaneously reversible. To our knowledge, Doppler echocardiographic signs of first-degree AV block have previously only been described in a single fetus initially noted to have a myocardial dysfunction that resolved during dexamethasone therapy (20); in addition, this fetus was retrospectively shown to have had prolonged mechanical time intervals (29).

Treatment with fluorinated steroids in a few fetuses diagnosed as having second-degree AV block has been demonstrated to improve AV conduction, and our observations of one fetus are in accordance with these previous findings. A complete AV block is commonly considered irreversible. This conclusion is based on cases in which bradycardia was the indication for fetal echocardiography and the time delay from the occurrence of the block to treatment was unknown. We know that another fetus in our study had signs of a first-degree AV block, but remained in a one-to-one conduction, only 6 days before a complete block was diagnosed and treatment with betamethasone was started. Still, the complete block remained, suggesting that complete block is irreversible even when treatment is started within a few days of occurrence.

The use of fluorinated steroid as prophylactic or symptomatic treatment of CHB carries potential risks for both the mother and her fetus (16, 30). Thus, observing that first-degree AV block seems to be reversible in the majority of cases, we currently consider this degree of block, without signs of endocardial fibroelastosis, to be an indication for closer surveillance without starting treatment with fluorinated steroids.

Excluding the two fetuses with second-degree or complete block, a slightly lower heart rate was still demonstrated in fetuses with anti–Ro 52-kd–positive mothers compared with that in control fetuses. This observation is interesting, since it might be an indication that the sinus node is also affected by maternal antibodies. This assumption is supported by experimental data showing that anti-Ro antibodies have effects on calcium currents in cardiocytes (31) and by autopsy data showing that in 2 of 7 babies who died of CHB, the sinuatrial nodes were hypoplastic and, in the case of another of these babies, surrounded by extensive fibrosis (32). Furthermore, sinus bradycardia has been described in infants of mothers with anti-Ro antibodies (5), but, to our knowledge, not previously in the fetus.

Fetal heart rate has been inversely correlated with AV time intervals (24). Hence, AV time intervals were plotted against heart rate to investigate whether the slight bradycardia found in the fetuses in our study could explain the prolongation in AV time intervals. No relationship was demonstrated, which confirmed that the abnormally long AV time intervals found in the fetuses of anti–Ro 52-kd–positive mothers were not an effect of a decrease in heart rate.

Our data may be interpreted as supporting the idea that the development of complete congenital heart block is a 2-stage process. In the first step, the maternal autoantibodies are transferred through the placenta and bind structures in the developing fetal heart, leading to a first-degree AV block. Our data indicate that occurrence of first-degree block is much more common than previously appreciated; they also show that this state is spontaneously reversible with further maturation of the fetus. In some cases, however, a second phase is entered and progression to complete heart block commences. Factors involved in this second phase may be mainly fetal and related to the immunogenetic profile of the individual. A predisposition to an intense inflammatory reaction to the antibody deposition might lead to the subsequent mononuclear cell infiltration and fibrosis, with a permanent injury as the outcome.

Our study demonstrates the usability of fetal Doppler echocardiography in the surveillance of fetuses at risk for CHB. By measuring AV time intervals, indirect signs of first-degree block can be documented, and different levels of block can also be distinguished by using these described techniques (Figures 4 and 5). Reference values for both techniques used in the present study have recently been reported (24). To minimize the risk of unexpected methodologic changes with time, new control pregnancies were tested against the original data from that study, and, after analysis, the resulting data were included in the reference data used in the present study.

Although the two Doppler methods used in the present study do not show any systematic differences when compared directly, each has advantages and disadvantages. On the one hand, the advantage of the MV-Ao approach is that these recordings are easier to obtain. On the other hand, in fetuses with long time intervals and/or higher heart rates, when the mitral E and A waves are difficult to separate, measuring the AV time interval may be very difficult. Recordings with the SVC-Ao approach require greater technical skill, but measurements are easier to perform. With this technique, AV time intervals are also less dependent on heart rate (24).

In summary, our data demonstrate that anti–Ro 52-kd–positive pregnant women frequently carry fetuses with Doppler echocardiographic signs of first-degree AV block. In the majority of these fetuses, a spontaneous normalization occurred before or shortly after birth. Still, the long-term prognosis for these infants remains to be established. Two of 8 fetuses in whom signs of first-degree block were documented also had higher degrees of block. Fetuses with normal AV time intervals at 18–24 weeks seemed to maintain normal AV conduction throughout pregnancy. Serial Doppler echocardiographic measurement of AV time intervals is suggested as a most useful instrument for surveillance of pregnancies at risk for fetal CHB.

Acknowledgements

The authors want to thank Professor J. C. Fouron, University of Montreal, Montreal, Quebec, Canada, for use of the reference data.

Ancillary