Newborns of mothers positive for anti-Ro/SSA autoantibodies may develop a series of electrocardiographic (EKG) disturbances. Prolongation of the corrected QT (QTc) interval was recently reported in a significant proportion of children with maternally acquired anti-Ro/SSA antibodies, with a concomitant disappearance of EKG abnormalities and acquired maternal autoantibodies during the first year, suggesting a direct, reversible electrophysiologic effect of anti-Ro/SSA antibodies on the ventricular repolarization. On this basis, we investigated whether these antibodies may also affect cardiac repolarization in anti-Ro/SSA–positive adult patients with connective tissue diseases.
Fifty-seven patients with connective tissue diseases were selected: 31 had anti-Ro/SSA antibodies and 26 did not (controls). In all subjects, we analyzed the QTc interval, heart rate variability, and signal-averaged high-resolution EKG recording.
Anti-Ro/SSA–positive patients showed a significant prolongation of the mean QTc interval compared with the controls (mean ± SD 445 ± 21 versus 419 ± 17 msec; P = 0.000005). Eighteen of the 31 anti-Ro/SSA–positive patients (58%) and none of the 26 anti-Ro/SSA–negative patients had QTc values above the upper limit of normal (440 msec). Both groups had a reduction in heart rate variability, with a prevalence for the sympathetic nervous system and a high incidence of ventricular late potentials; these values were not significantly different between the 2 groups.
Adult patients with anti-Ro/SSA–positive connective tissue diseases showed a high prevalence of QTc interval prolongation. This feature, with the concomitant abnormalities in the autonomic tone and ventricular late excitability observed in all patients studied, suggests that anti-Ro/SSA–positive patients may have a particularly high risk of developing life-threatening arrhythmias.
Neonatal lupus is a rare syndrome related to the transplacental passage of autoantibodies from mothers positive for anti-Ro/SSA (and anti-La/SSB) to their newborns. These mothers frequently have autoimmune disorders, especially connective tissue diseases; however, they are sometimes entirely asymptomatic. The main feature of neonatal lupus is congenital heart block (CHB), an irreversible disturbance in cardiac conduction. Many recent studies have investigated the possible electrophysiologic and molecular mechanisms of CHB, and the findings suggest a direct arrhythmogenic activity of anti-Ro/SSA antibodies. In particular, it has been demonstrated that affinity-purified anti-Ro/SSA antibodies from mothers of children with CHB are able to induce a complete atrioventricular block in the human fetal heart as well as in a rat model of the heart, which inhibits L-type calcium currents (1).
CHB is the main, but not the only, electrocardiographic abnormality associated with the transplacental passage of anti-Ro/SSA antibodies. Mazel and colleagues (2) reported sinus bradycardia in a murine model injected with IgG obtained from mothers of children with CHB. These findings were later confirmed by Brucato and colleagues (3), who studied infants born to anti-Ro/SSA–positive mothers. The molecular basis for these findings seems to be related to the fact that these antibodies also block the T-type calcium channels, which play a pivotal role in the pacemaker activity of the heart (1). Moreover, it has been reported that a significant proportion of children with maternally acquired anti-Ro/SSA antibodies have a prolongation of the QT interval (4), but the possible underlying electrophysiologic mechanism is still unknown.
Other investigators have suggested a potential role for these autoantibodies in the cholinergic dysfunction. In fact, there is evidence of an interaction of the autoantibodies with muscarinic receptors (5). Several studies have shown the presence of cardiac autonomic dysfunction in Sjögren's syndrome patients, particularly in those with anti-Ro/SSA autoantibodies (6). Autonomic dysfunction has also been observed in other anti-Ro/SSA–positive connective tissue diseases, such as systemic lupus erythematosus and systemic sclerosis (7, 8). These observations suggest that anti-Ro/SSA autoantibodies may determine a dysfunction in cardiac autonomic activity.
A prolongation of the corrected QT interval (QTc) and a cardiac dysautonomia with an imbalance of the sympathetic nervous system have been identified as important risk factors for sudden death from cardiac arrhythmias (9). Since the incidence of cardiac arrhythmias and sudden death in patients with connective tissue diseases is higher than that in the general population (10), we investigated the possible influence of anti-Ro/SSA antibodies on the QTc interval and on heart rate variability in adult patients with connective tissue diseases. Other risk factors for sudden death of cardiac origin, such as ventricular late potentials (9), which are possibly influenced by anti-Ro/SSA antibodies, were also evaluated.
PATIENTS AND METHODS
The study population consisted of 57 Caucasian patients with the following connective tissue diseases: Sjögren's syndrome (SS), systemic lupus erythematosus (SLE), systemic sclerosis (SSc), diagnosed according to the criteria of the American College of Rheumatology (formerly, the American Rheumatism Association) and the European Community Study Group (11–13), undifferentiated connective tissue disease (UCTD) (14), and mixed connective tissue disease (MCTD) (15). Patients were grouped according to positivity (n = 31) and negativity (n = 26) for anti-Ro/SSA antibodies.
The anti-Ro/SSA–positive group (30 women and 1 man) consisted of 15 patients with SS, 6 with SLE, 4 with SSc, 5 with UCTD, and 1 with MCTD. Their mean ± SD age was 41.4 ± 15.4 years. The anti-Ro/SSA–negative (control) group (22 women and 4 men) consisted of 1 patient with SS, 4 with SLE, 17 with SSc, 3 with UCTD, and 1 with MCTD. Their mean ± SD age was 48.1 ± 13.0 years. None of the patients were taking drugs that had the potential to influence the QTc interval (class I and class III antiarrhythmics, antihistamines, quinolone and macrolide antibacterials, azole antifungals, phenothiazines, tricyclic antidepressants, cisapride), except for hydroxychloroquine (HCQ). None of the patients had abnormal findings on EKG and/or none had a history consistent with coronary artery disease, nor did any of them have diabetes, renal failure, or an electrolyte imbalance (Na+, K+, Ca++, or Mg++) (16, 17). Ongoing treatments are reported in Table 1.
Table 1. Ongoing therapy in the study subjects, by anti-Ro/SSA group
Patients with bundle branch block (standard QRS duration >120 msec) and/or an R-R interval shorter than 521 msec or longer than 1,111 msec were excluded from the study. Informed consent was obtained from patients before entering the study.
Heart rate variability.
A 12-lead EKG was continuously monitored and recorded for up to 10 minutes while the patient was in a supine position and breathing normally. A commercially available imaging system (Cardioline δ 612; Remco, Milan, Italy) was used. Detection of the QRS complex and measurement of the R-R interval were performed automatically, using the R wave peak as a reference point. Premature beats, missed beats, and artifacts were visually identified using an interactive graphic interface, and these were evaluated by the operator. In this way, an R-R tachogram, that is, a discrete series of successive R-R intervals as a function of the number of recognized QRS complexes, was obtained. The algorithm used in the analysis of the tachogram was a spectral method (fast Fourier transformation). Three main spectral components were distinguished in a spectrum calculated from short-term recordings of 5 minutes' duration: 1) a very low-frequency (VLF) component (<0.04 Hz), 2) a low-frequency (LF) component (range 0.04–0.15 Hz), and 3) a high-frequency (HF) component (range 0.15–0.4 Hz). Measurements of the VLF, LF, and HF power components as well as total power were made in absolute values of power (msec2). The ratio of low to high frequency was calculated as an expression of the sympathovagal balance (18).
QTc interval and QTc dispersion.
On the same 12-lead EKG used to measure heart rate variability, we measured the QT interval, according to standard criteria. The QT interval was measured from the onset of the Q wave or the onset of the QRS complex to the end of the T wave, defined as the return to the T-P baseline. When U waves were present, the QT interval was measured to the nadir of the curve between the T and U waves.
The QT interval, determined as the longest hand-measured QT interval in any lead, was corrected for the heart rate by Bazett's formula to yield the QTc value. (The QTc value was calculated by dividing the QT interval by the square root of the R-R interval, excluding intervals shorter than 521 msec and longer than 1,111 msec; Bazett's formula considers values outside this range to be unreliable). The QTc was considered abnormal if it was longer than 440 msec. QTc dispersion was defined as the difference between the minimum and maximum heart rate–adjusted QT interval among the 12 EKG leads. QTc dispersion was considered abnormal if it was longer than 50 msec (19).
Signal-averaged high-resolution EKG.
Bipolar pseudo-orthogonal X, Y, and Z EKG leads were acquired using an ART 1200 EPX high-resolution signal averaging system (Arrhythmia Research Technology, Austin, TX). The sampling rate was 1,000 samples · seconds–1, with a resolution of 12 bits. Signal averaging was performed from a mean ± SD of 155 ± 74 beats (range 60–500 beats). The noise level was determined from a 40–250-Hz filtered-vector magnitude. Conventional time-domain analysis was performed using ART analysis software (version 2.0), according to the method of Breithardt (20). The filtered setting was 40–250 Hz.
Three parameters were considered for detection of late potentials: the filtered QRS duration, the root mean square voltage of the terminal 40 msec of the filtered QRS (RMS40), and the low-amplitude (<40 μV) signal duration (LAS40). This methodology fits the standards proposed by the task force committee of the American Heart Association, the American College of Cardiology, and the European Society of Cardiology (20). Late potentials were identified in the high-resolution EKG according to the following conventional criteria: QRS duration >114 msec, RMS40 <20 μV, and LAS40 >38 msec.
Statistical evaluation was performed with Student's t-test for unpaired data. Data are presented as the mean ± SD. The possible influence of HCQ on the QTc interval was evaluated with Student's t-test for unpaired data and with the chi-square test by comparing anti-Ro/SSA–positive and anti-Ro/SSA–negative patients who were and were not taking HCQ. P values less than 0.05 were considered significant.
Characterization of anti-Ro/SSA antibodies.
Among the 31 subjects in the patient group, 11 were positive for antibodies to 52-kd and 60-kd Ro proteins, 11 were positive for anti–60-kd Ro, and 9 were positive for anti–52-kd Ro. Eight patients were also positive for anti-La/SSB antibodies. None of the control group patients was positive for anti-Ro/SSA and/or anti-La/SSB antibodies.
Findings of QTc interval and QTc dispersion.
Anti-Ro/SSA–positive patients showed a significant prolongation of the mean QTc interval compared with anti-Ro/SSA–negative patients (mean ± SD 445 ± 21 msec versus 419 ± 17 msec; P = 0.000005), with a mean value higher than the upper limit of normal (Table 2 and Figure 1). Fifty-eight percent of the anti-Ro/SSA–positive patients (18 of 31) had a QTc interval that was longer than 440 msec, but none of the control patients had a prolonged QTc interval.
Table 2. Parameters of the corrected QT interval, heart rate variability, and signal-averaged high-resolution electrocardiography in the study subjects, by anti-Ro/SSA group
Anti-Ro/SSA positive (n = 31)
Anti-Ro/SSA negative (n = 26)
* Except where indicated otherwise, values are the mean ± SD. P values were determined by Student's t-test for unpaired data.
Corrected QT interval, msec
445 ± 21
419 ± 17
% with corrected QT interval longer than 440 msec
Corrected QT interval dispersion, msec
37 ± 19
29 ± 9
Heart rate variability, msec2
1,642 ± 1,684
1,249 ± 784
Very low-frequency component, msec2
575.0 ± 559
609.1 ± 449
Low-frequency component, msec2
509.3 ± 628
413.7 ± 316
High-frequency component, msec2
499.5 ± 874
241.0 ± 267
Low-frequency:high-frequency component ratio
2.60 ± 2.8
2.86 ± 2.4
R-R interval, msec
836.5 ± 138
786.9 ± 106
% positive for ventricular late potentials
A slightly higher proportion of patients in the anti-Ro/SSA–positive group (11 of 31 [35%]) than in the control group (5 of 26 [19%]) were taking HCQ. We also evaluated the possible influence of HCQ on the QTc interval. Concomitant treatment with HCQ did not show any significant influence on the QTc interval prolongation in anti-Ro/SSA–positive patients (446 ± 20 msec in patients taking HCQ versus 445 ± 22 msec in those not taking HCQ; P = 0.99) or in the anti-Ro/SSA–negative patients (416 ± 18 msec in patients taking HCQ versus 418 ± 17 msec in those not taking HCQ; P = 0.78). Moreover, an overall influence of HCQ treatment on the QTc interval in all study subjects was reasonably ruled out by the observation that a significant QTc interval prolongation still existed in anti-Ro/SSA–positive patients as compared with anti-Ro/SSA–negative patients when only those who were not taking HCQ in both groups (n = 20 and n = 21, respectively) were considered (χ2 = 18.86, P < 0.001).
The slightly different mean QTc dispersion observed in anti-Ro/SSA–positive and anti-Ro/SSA–negative subjects (mean ± SD 37 ± 19 msec versus 29 ± 9 msec, respectively) did not reach statistical significance (P = 0.06) (Table 2).
Findings of heart rate variability.
Both anti-Ro/SSA–positive and anti-Ro/SSA–negative patients showed a reduction in mean values for heart rate variability as compared with reference values (Table 2). Many of the subjects had values that were lower than the lower limit of normal (21 of 31 [68%] in the patient group, and 22 of 26 [85%] in the control group). Indeed, heart rate variability in the 2 groups was not statistically different (Table 2).
An alteration in the sympathovagal balance was present in the majority of patients (25 of 31 [81%] in the patient group, and 20 of 26 [77%] in control group), and the mean values of the ratio of low-frequency to high-frequency components showed a sympathetic tone in both groups (2.60 ± 2.82 in the patient group, and 2.86 ± 2.41 in the control group). In this case as well, no statistically significant difference was found between the 2 groups (Table 2).
Findings of signal-averaged high-resolution EKG.
Both groups of patients presented a high incidence of ventricular late potentials, with respect to the general population. However, no significant difference was found between anti-Ro/SSA–positive and anti-Ro/SSA–negative patients (32% versus 37%) (Table 2).
In a previous study, a QTc prolongation in asymptomatic anti-Ro/SSA–positive infants without congenital heart block was reported (4). The same investigators demonstrated a concomitant disappearance of EKG abnormalities and acquired maternal autoantibodies in these infants during their first year of life (21). These data, together with the results of other in vivo and in vitro studies, suggested a direct, reversible electrophysiologic effect of anti-Ro/SSA antibodies on ventricular repolarization. On this basis, we investigated whether these antibodies may affect cardiac repolarization in anti-Ro/SSA–positive adult subjects as well.
In the present study, we found that a high proportion of patients with connective tissue diseases who were positive for anti-Ro/SSA antibodies had a prolongation of the QTc interval. In contrast, all of the control patients, who had connective tissue diseases and were anti-Ro/SSA negative, had a QTc interval that was lower than the upper limit of normal (440 msec).
These findings may be very relevant in light of the fact that many studies have shown that a prolongation of the QTc interval predisposes a subject to the development of malignant ventricular arrhythmias, with an increased risk of sudden death (9, 16). It has been demonstrated that in patients with connective tissue diseases, the incidence of life-threatening arrhythmias and sudden death is higher than that in the general population (10). Therefore, at least in a subset of connective tissue disease patients, a direct pathogenetic role of anti-Ro/SSA antibodies in increasing arrhythmogenic risk may be suggested.
One limitation of our study is the fact that the SS and SSc patients were not equally distributed among the anti-Ro/SSA–positive and anti-Ro/SSA–negative groups (as would be expected from epidemiologic studies). This would be considered a potential confounder.
Even though the results of our study and the study reported by Cimaz et al (4) seem to indicate an interference of anti-Ro/SSA antibodies in ventricular repolarization, the possible underlying electrophysiologic mechanism is still unknown. Recent data have shown that anti-Ro/SSA antibodies are able to inhibit calcium currents (L- and T-type) (1), thereby providing evidence for a putative pathogenetic mechanism of complete atrioventricular block and sinus bradycardia in anti-Ro/SSA–positive newborns. However, in ventricular cardiomyocytes, the repolarization period is under the main regulatory control of the potassium channels (16). It has been demonstrated that anti-Ro/SSA antibodies do not affect the potassium currents, the inward rectifier potassium current (IK1) and the transient outward current (Ito), in a rat model of the heart (1). However, no experimental data are available on the possible effects of these antibodies on other important potassium currents, such as the rapid cardiac delayed rectifier potassium current (IKr) and the slow cardiac delayed rectifier potassium current (IKs). Thus, the possibility of a selective blocking activity of anti-Ro/SSA antibodies on these latter currents cannot be ruled out.
A preponderant proportion of subjects in both groups showed a reduction in heart rate variability, with a concomitant alteration in the sympathovagal balance, represented by sympathetic tone. However, the incidence of sympathetic tone was not statistically significantly different in the 2 groups, which suggests that the observed QTc interval prolongation in anti-Ro/SSA–positive patients was not influenced by the increased sympathetic tone, as might possibly be expected (22). Furthermore, in both the anti-Ro/SSA–positive and the anti-Ro/SSA–negative groups, a relevant percentage of patients had ventricular late potentials. These findings suggest that in all the connective tissue disease patients, common mechanisms possibly operate that lead to an increase in the arrhythmogenic risk by producing changes in the autonomic tone and ventricular late excitability, without any relevant role for anti-Ro/SSA antibodies. However, the demonstration that the presence of anti-Ro/SSA antibodies is the determining factor for the occurrence of an important prolongation in the QTc interval leads to the conclusion that these antibodies may further enhance the incidence of sudden death in adult patients with connective tissue diseases. Based on the above data, a prospective study in patients with and without anti-Ro/SSA antibodies is now in progress in order to verify the possible difference in the occurrence of life-threatening ventricular arrhythmias and/or sudden death in the 2 groups.