Article Text

Comorbidity and long-term outcome in patients with congenital heart block and their siblings exposed to Ro/SSA autoantibodies in utero
  1. Johannes Mofors1,
  2. Håkan Eliasson2,
  3. Aurelie Ambrosi1,
  4. Stina Salomonsson1,
  5. Amanda Skog1,
  6. Michael Fored1,
  7. Anders Ekbom1,
  8. Gunnar Bergman2,
  9. Sven-Erik Sonesson2,
  10. Marie Wahren-Herlenius1
  1. 1 Department of Medicine, Karolinska Institutet, Stockholm, Sweden
  2. 2 Department of Women’s and Children’s Health, Karolinska Universitetssjukhuset, Stockholm, Sweden
  1. Correspondence to Professor Marie Wahren-Herlenius, Department of Medicine, Karolinska Institutet, Stockholm 17176, Sweden; marie.wahren{at}ki.se

Abstract

Objective Congenital heart block (CHB) may develop in fetuses of Ro/SSA autoantibody-positive women. Given the rarity of CHB, information on comorbidity and complications later in life is difficult to systematically collect for large groups of patients. We therefore used nation-wide healthcare registers to investigate comorbidity and outcomes in patients with CHB and their siblings.

Methods Data from patients with CHB (n= 119) and their siblings (n= 128), all born to anti-Ro/SSA-positive mothers, and from matched healthy controls (n= 1,190) and their siblings (n= 1,071), were retrieved from the Swedish National Patient Register. Analyses were performed by Cox proportional hazard modelling.

Results Individuals with CHB had a significantly increased risk of cardiovascular comorbidity, with cardiomyopathy and/or heart failure observed in 20 (16.8%) patients versus 3 (0.3%) controls, yielding a HR of 70.0 (95% CI 20.8 to 235.4), and with a HR for cerebral infarction of 39.9 (95% CI 4.5 to 357.3). Patients with CHB also had a higher risk of infections. Pacemaker treatment was associated with a decreased risk of cerebral infarction but increased risks of cardiomyopathy/heart failure and infection. The risk of systemic connective tissue disorder was also increased in patients with CHB (HR 11.8, 95% CI 4.0 to 11.8), and both patients with CHB and their siblings had an increased risk to develop any of 15 common autoimmune conditions (HR 5.7, 95% CI 2.83 to 11.69 and 3.6, 95% CI 1.7 to 8.0, respectively).

Conclusions The data indicate an increased risk of several cardiovascular, infectious and autoimmune diseases in patients with CHB, with the latter risk shared by their siblings.

  • congenital heart block
  • pacemaker
  • sle
  • sjögren’s syndrome

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Key messages

What is already known about this subject?

  • No population-based health register studies have been performed to investigate comorbidity and long-term outcome in congenital heart block (CHB).

What does this study add?

  • Patients with CHB have a significantly increased risk of cardiovascular disease, including heart failure, cardiomyopathy and cerebral infarction.

  • Autoimmune diseases are significantly more frequent in individuals with CHB, as well as in siblings of patients with CHB.

How might this impact on clinical practice or future development?

  • Our data support a close follow-up of cardiac function in patients with CHB, and that autoimmune conditions should be considered in both patients and their siblings.

Introduction

Complete congenital heart block (CHB) without associated cardiac malformation is a rare condition, affecting 1 in 23,000 births in the general population.1 The association between CHB and maternal anti-Ro/SSA and anti-La/SSB autoantibodies is well established,2–4 with CHB occurring in 1%–2% of anti-Ro/SSA-exposed fetuses in several studies,4–6 although lesser figures have been reported in studies confined to women with systemic lupus erythematosus (SLE).7 Women carrying the autoantibodies are often diagnosed with SLE or Sjögren’s syndrome (SS) but can also be asymptomatic.8–10 During pregnancy, the autoantibodies are transported across the placenta and may induce neonatal lupus, including a complete third-degree atrioventricular (AV) block.11–13 The majority of children with CHB require a pacemaker at an early age to improve cardiac function.14 However, pacemaker treatment may potentially carry negative effects, and right ventricular pacing has been suggested to associate with subsequent development of dilated cardiomyopathy.15–17

Given the rarity of CHB, information on comorbidity and complications later in life is difficult to systematically collect for large groups of patients, and the literature on long-term follow-up and comorbidity in children with CHB is limited. We and others have previously observed that patients with CHB are growth-restricted during the first years of life,18 19 and that there is an increased prevalence of impaired neurodevelopment19 and neuro-psychiatric abnormalities20 in this group of individuals. Recent studies on CHB, including several case reports21–26 and a questionnaire-based study,27 have also indicated that CHB might be a risk factor for the development of rheumatic and autoimmune disease. Whether there may also be a risk for the siblings without CHB is not clear,27 although familial aggregation of autoimmune diagnoses28 29 suggests that also the siblings could have an increased risk to develop autoimmune disease.

To systematically assess morbidity and long-term outcome in individuals with or without CHB born to anti-Ro/SSA autoantibody-positive mothers, we established a cohort of CHB individuals and their siblings based on 115 families in which anti-Ro/SSA antibody-positive mothers had given birth to at least one child with CHB. Ten control families were identified for each index family using Swedish population registers, and healthcare data were subsequently obtained through national healthcare registers. The incidence of International Classification of Diseases 10th revision (ICD10) diagnoses in exposure and matched control groups was then analysed to assess the risk of comorbidity and future disease development in children with CHB and their siblings.

Materials and methods

Study population and data sources used to detect outcomes during follow-up

Individuals in the present study were previously identified through a population-based strategy and included in a cohort of Swedish patients with CHB.30 From this cohort, all individuals with CHB (n=119) and their siblings without CHB (n=128) born to anti-Ro/SSA antibody-positive mothers (n=115) between year 1948 and 2010 were included for analysis (table 1). Siblings were defined as individuals sharing both parents. For each patient with CHB, 10 controls (n=1,190) from the general population matched for sex, year and month of birth, as well as region of birth were randomly selected from the Swedish Total Population Register at Statistics Sweden (www.scb.se). Siblings (n=1,071) of the controls were identified through the Swedish national multi-generation register at Statistics Sweden, and served as controls to the siblings of the patients with CHB. Diagnoses received by patients with CHB, siblings and controls during the period of observation were obtained from the Swedish National Patient Register (inpatient care 1987–2010 and non-primary outpatient care 2001–2010; www.socialstyrelsen.se). For these periods, the register is nationwide, with a coverage of 99% for hospitalisations and 80% for outpatient care. The latter lower coverage is mainly due to lower reporting rates from private care.31 All diagnoses are coded according to the ICD, and the data analysed in the current study are based on the 9th and 10th ICD editions (http://www.who.int/classifications/icd/en/). Diagnoses based on the 9th ICD edition were converted to their corresponding ICD-10 versions, thus enabling an aggregate analysis of comorbidity throughout the observational period. The cohorts were followed from birth or 1st of January 1987 (whichever came last) until death or 31st December 2010, whichever came first.

Table 1

Characteristics of investigated cohorts

The study was approved by the Regional Ethics Committee in Stockholm, and written informed consent was obtained from all participating individuals from the CHB families or their parents if <18 years old.

Statistical analysis

Statistical analysis was performed using STATA MP V. 13.0 (StataCorp LP, College Station, TX, USA). Statistical significance was defined by an alpha level of 0.05. Q-values were calculated to account for false discovery rates, only observations with q<0.2 are reported.

Cox regression was used to estimate HRs of disease during follow-up time. Significance parameters were defined when the sum of events in the exposure and control groups was ≥5 in CHB and ≥10 in siblings and their respective controls. In comparisons between unmatched samples, such as siblings of individuals with CHB versus siblings of controls, HRs adjusted for differences in age. HRs above 100 or below 0.01 are reported as >100 and<0.01, respectively. If no event was present in the exposure or control group, no CI is reported. A Nelson-Aalen estimator was used to calculate cumulative hazard rates of disease.

Results

Demographics of the investigated cohort

In the present study, we included all patients with CHB (n=119) and their siblings without CHB (n=128, referred to hereafter as ‘CHB siblings’) enrolled in a population-based CHB cohort and born to Ro/SSA antibody-positive mothers,30 and assigned them to 1,190 and 1,071 controls, respectively (table 1). Data were extracted from the Swedish National Patient Register for all individuals. The mean age at inclusion in the study was 7.5 years (median 0, range 0–38.6 years) for individuals with CHB, and 11.7 years (median 3.3, range 0–45.7 years) for CHB siblings (table 1,online supplementary table 1). The mean follow-up time was 17.1±8.0 years in the CHB group and 18.4±7.3 years in the CHB sibling group, with a total exposure of 2036 and 2352 patient-years for patients with CHB and their siblings, respectively, and 20 078 and 19 534 comparator-years for the respective control groups.

Comorbidity and long-term outcome in CHB in relation to organ system and aetiology

To investigate comorbidity and long-term health status in individuals affected by CHB, we first analysed the occurrence of diagnoses included in the ICD blocks of chapters I–XIV in CHB individuals and matched controls (table 2). A higher proportion of patients with CHB than controls received diagnoses within the circulatory system ICD chapter during the observation period. This chapter indeed includes the diagnosis of AV block within the ICD block ‘Other forms of heart disease’ (I30-I52), and although this diagnosis was present in all CHB individuals as part of the cohort inclusion criteria, it was also re-assigned to 97% of CHB individuals during the observation period. The AV-block diagnosis is therefore retained in the tables and accounts, at least in part, for a HR of >100 for the ICD block ‘Other forms of heart disease’. Nevertheless, HRs>1 were also observed for two other ICD blocks within the circulatory system ICD chapter, and these blocks were related to disorders not manifested in the heart itself but in the vascular tissue, with HRs ranging from 3.9 to 10.1 (table 2). We further found that patients with CHB had an elevated risk of systemic connective tissue disorders compared with controls (HR 11.8, 95% CI 4.0 to 35.1). In addition, individuals with CHB appeared at higher risk of infectious diseases, including general illnesses and infections of the respiratory system or the skin. Finally, we observed that patients with CHB were at higher risk of developing psychological disorders, metabolic disorders and diseases of the digestive system.

Table 2

ICD blocks associated with significant HRs for patients with CHB

To assess the risk of these diagnoses in the CHB siblings, we next compared their occurrence between CHB siblings and siblings of the controls (table 3). No significant hazard ratios were observed for any of these ICD blocks. We however observed that the incidence of systemic connective tissue disorders was considerably higher among CHB siblings compared with controls (3.1% vs 0.4%, n=4 in each group, HR 12.7). Similarly, the number of events of ‘Other forms of heart disease’ (I30-I52) observed in CHB siblings (3.9%, n=5) was higher than in the control group (1.0%, n=11) (HR 3.0, 95% CI 1.0 to 8.9, corresponding to a p value of 0.067).

Table 3

HR and incidence of disease for siblings of patients with CHB in ICD blocks associated with significant HRs for patients with CHB

Defining comorbidity and long-term outcome at the three-character ICD code level

To more precisely define the observed comorbidities, we next assessed risk at the level of the discrete three-character ICD codes included in the blocks for which patients with CHB had presented significant hazard ratios. We found that patients with CHB had an increased risk of disease across multiple cardiovascular diagnoses (table 4). Overall, 20 (16.8%) patients with CHB were diagnosed with cardiomyopathy and/or heart failure during the observation period (HR 70.0, 95% CI 20.8 to 235.4). More specifically, there were 14 events of cardiomyopathy in the CHB group compared with none in the control group (HR >100), and 10 events of heart failure among patients with CHB versus three among controls (HR 34.4, 95% CI 9.5 to 125.2). In addition, a diagnosis of ‘other arrhythmias’ (I49) was present in 55 (46.2%) patients with CHB versus 5 (0.4%) controls (HR >100, 95% CI 66.0 to 415.0). This included sick sinus syndrome (SSS, I49.5) present in 5 (4.2%) patients with CHB and 1 (0.1%) control, while the majority of cases (50 CHB individuals and four controls) were classified as ‘other specified cardiac arrhythmias’ (I49.8), and could not be further defined. Nine (7.6%) patients with CHB developed atrial fibrillation and flutter (HR 46.7, 95% CI 10.1 to 216.1), of which five either had a previous or subsequent diagnosis of heart failure or cardiomyopathy.

Table 4

Three-character ICD codes within blocks in table 2 associated with significant hazard ratios for patients with CHB

Notably, 4 (3.4%) patients with CHB experienced a cerebral infarction, compared with one individual (0.08%) in the control group (HR 39.9, 95% CI 4.5 to 357.3) (table 4). Two of the patients had previously been diagnosed with cardiomyopathy, and one individual had records of pacemaker treatment before the cerebral infarction. No records of previous or subsequent atrial fibrillation or flutter were found in patients with CHB with cerebral infarction.

We further observed that patients with CHB had a significantly increased risk of the ICD diagnosis code ‘other systemic involvement of connective tissue’ (M35) (HR 7.5, 95% CI 1.7 to 33.4), with 3 (2.5%) patients with CHB versus 3 (0.3%) control individuals diagnosed with such conditions. Other individual 3-character ICD codes within the ‘systemic connective tissue disorder’ block did not fulfil criteria for analysis due to few events, or the estimated HR was not significant.

The analysis at the three-character ICD code level also confirmed the increased risk of multiple infectious diseases in patients with CHB previously noted at the ICD block level, including bacterial infections, sepsis, tonsillitis, upper respiratory tract infections, bronchitis and pneumonia, as well as infections of the skin and subcutaneous tissue (table 4). The proportion of infections occurring before the age of 1 year did not significantly differ between patients with CHB and controls (p=0.41, online supplementary table 2. Skin involvement was also apparent from the observation of an increased risk of atrophic skin disorders (L90). Of the disorders of psychological development, pervasive developmental disordered was more common in patients with CHB than controls. Notably, the siblings of patients with CHB did not present significant hazard ratios for any of the discrete diagnoses cited above (online supplementary table 3).

Effects of pacemaker implantation on cardiovascular morbidity and infections

Considering that pacemaker implants have been shown to associate with various morbidities,17 32–36 we assessed the effect of pacemaker treatment on cardiovascular morbidity and infections in individuals with CHB, performing Cox proportional hazard modelling with pacemaker surgery as time-varying covariate (figure 1, online supplementary table 4). n=107 (90%) individuals with CHB had records indicating pacing treatment during the observational period. A protective effect of pacemaker treatment was observed regarding the risk of developing cerebral infarction (HR 0.1; 95% CI 0.01 to 0.8) and other cardiac arrhythmias (HR 0.4; 95% CI 0.1 to 0.9). Pacemaker treatment was however associated with an increased risk of developing cardiomyopathy and/or heart failure after pacemaker implantation (HR 3.8, 95% CI 1.1 to 12.6). In addition, pacemaker implants were associated with an increased risk of infections (A00-B99 and L00-08) (HR 5.5; 95% CI, 2.7 to 11.3) (figure 1).

Figure 1

Influence of pacemaker treatment on the risk of comorbidities. Hazard ratios and 95% CI for indicated diagnoses with pacemaker surgery as time-varying covariate.

Cumulative risk of autoimmune disease

We observed a significantly increased risk of several conditions at the three character ICD code level related to autoimmunity and a trend for increased risks of others in patients with CHB (tables 2 and 4, and data not shown). To assess the aggregated risk of autoimmune disease, we created a composite outcome variable of common autoimmune diagnoses (thyroid disease, multiple sclerosis, psoriasis, arthritis and systemic rheumatic disease). The ICD codes included in the measure are specified in the legend of figure 2. This variable was then analysed using a Nelson-Aalen estimator to investigate the age-wise accumulation of autoimmune disease, and hazard ratios were defined by Cox proportional regression. Both patients with CHB and their siblings presented a significantly higher frequency of autoimmune diseases as defined by the composite variable than their respective controls, with hazard ratios for autoimmune disease of 5.7 for patients with CHB (p<0.01), and 3.6 (p<0.01) for their siblings (figure 2, online supplementary table 5). Changing the composite outcome variable to only include the systemic rheumatic diseases, or the thyroid diseases, or the arthritic diseases, all demonstrated significantly increased hazard ratios for patients with CHB, while only the thyroid disease composite variable remained significant for siblings of patients with CHB (online supplementary table 6). During the follow-up time, 13 (11%) of patients with CHB developed an autoimmune disease, compared with 24 (2%) of the controls. Six (5%) siblings of patients with CHB versus 33 (3%) of sibling controls received a diagnosis of autoimmune disease. Although the proportion of CHB siblings developing an autoimmune disease was smaller than that seen in the group of CHB individuals, the difference was not statistically significant.

Figure 2

Risk of developing autoimmune disease for patients with CHB and their siblings. Cumulative hazard rates of autoimmune disease as assessed by a composite score and plotted for patients with cHb (blue line) and controls (green line), as well as CHB siblings (red line) and controls’ siblings (yellow line). the autoimmune disease composite score was defined as any of: thyroid diagnoses (E03, E04, E06), multiple sclerosis (G35), psoriasis (L40), systemic rheumatic diseases (M32, L93, M33, M34, M35.0), and arthritis diagnoses (M02, M03, M05, M08). Brackets indicate inter-group comparisons using Cox regression with corresponding hazard ratio (95% CI).

Discussion

Information on comorbidity and complications later in life for individuals affected by CHB is scarce due to the rareness of the condition. We therefore conducted a study based on data available in nationwide Swedish healthcare registers to systematically survey long-term outcome in patients with CHB that had been identified in a population-based manner.30 In our cohort, patients with CHB displayed an increased risk of cardiovascular comorbidities. Specifically, we observed an increased risk of cardiomyopathy and/or heart failure in CHB individuals, confirming findings from previous studies.14 17 35–37 Although we also found an association between pacemaker implantation and an increased risk of developing cardiomyopathy/heart failure, as previously observed,36 it should be noted that our study does not allow investigation of a possible causal relationship, and that it has also been reported that the vast majority of patients with CHB already have an impaired heart function before receiving pacemaker treatment.37 We also observed that patients with CHB had elevated risks of atrial fibrillation and flutter, as well as of SSS. Importantly, the increased risk of cardiovascular morbidity was not just related to conditions affecting the heart itself, but also included an increased risk of other cardiovascular diseases, with patients with CHB displaying a substantially increased risk to develop cerebral infarction. Our analyses revealed that pacing therapy may provide a degree of protection, with three of the four individuals who experienced cerebral infarction not having received pacemaker treatment. It is also worth noting that siblings of patients with CHB had a higher incidence of heart disease than their controls. Our data indicate conduction disorders and tachyarrhythmias, but a statistical relationship could not be established due to the low number of events in this limited cohort. It is important to keep in mind that the higher frequency of minor cardiac abnormalities may also relate to reporting bias, as siblings of patient with CHB may more often be subject to investigation of cardiac function than the general population.

Our study further demonstrates that patients with CHB are more likely of being diagnosed with infections. Several reasons may underlie this increased risk. One explanation, although perhaps less probable, is that individuals with CHB represent a group with an inherently increased susceptibility to infectious disease. The occurrence of infections secondary to pacemaker implantation surgery is well known,33 38 thus constituting a likely explanation, and our analysis of complications secondary to pacemaker implantation indeed confirmed this association. Moreover, the increased prevalence of prematurity reported in individuals with CHB,1 39 which itself is associated with an increased risk of infections,40 may also explain some of the risk. Although we were not able to account for the impact of preterm births, the age-wise distribution of infections did not suggest a significant role of prematurity. In addition, it may be assumed that patients with an increased healthcare exposure and more frequent healthcare visits, such as patients with CHB, are more likely to receive diagnoses for infections compared with the general population.

Our findings also implicate that patients with CHB have an increased frequency of psychological developmental disorders, consistent with previous studies reporting an increased prevalence of neuro-psychiatric dysfunction in this group.19 20 Moreover, we also observed an increased incidence of diagnoses related to metabolic disorders and diseases of the digestive system. However, reporting bias may influence the estimated risk of morbidity in patients with CHB, and false discoveries are likely more probable in disorders with few observed cases such as the above.

The patients with CHB included in this study were all born to anti-Ro/SSA antibody-positive mothers, the majority of whom had a diagnosis of autoimmune disease either at the time of pregnancy or later in life.10 30 41–43 In line with previous reports of familial aggregation of autoimmune diseases,28 29 44 we observed that both patients with CHB and their siblings had an increased risk of developing systemic connective tissue disorders and/or autoimmune diseases. Although the frequency of autoimmune diseases in patients with CHB was not statistically different from that observed in their siblings, the proportion of patients with CHB developing autoimmune disease was nevertheless greater, with the cumulative incidence about 2-fold higher among patients with CHB than their siblings. A relatively low recurrence rate of CHB despite the persisting presence of maternal anti-Ro/SSA antibodies45–47 indicates that fetal genetic susceptibility may modulate pathogenetic mechanisms and influence CHB development.48–50 Similarly, genetic differences could underlie a higher frequency of autoimmune disease in patients with CHB compared with their siblings, as well as the overall increased risk for these two groups in relation to their general population comparators.

Limitations of this study include potential errors related to the registers used with regard to validity and exhaustiveness. This constraint is however somewhat mitigated by the setup, as the information collected for both exposure and control groups is subject to the same limitations. Another drawback is the fact that we only included CHB individuals alive at the time of entry into the cohort, as its original establishment was designed to include biological sample collection, leading to a survival bias. Further, as discussed above, we were not able to assess or control for the impact of prematurity, which may contribute to the risk of some morbidities.18 51 Finally, despite the national coverage and long-term follow-up at a mean of almost two decades, the overall limited number of events makes statistical estimates less stable. We nonetheless think that our dataset, allowing long-term follow-up of a relatively large number of patients born with CHB, is unique, and carries substantial value in understanding the health challenges for this group of patients.

In all, our findings suggest a non-negligible burden of comorbidity for patients with CHB, which is most apparent within cardiovascular, infectious and chronic inflammatory or autoimmune disorders, with the latter risk also shared by their siblings.

Acknowledgments

We thank Ulf Hammar, Karolinska Institutet, for invaluable statistical support.

References

View Abstract

Supplementary materials

  • Lay summary

    Disclaimer : This is a summary of a scientific article written by a medical professional (“the Original Article”). The Summary is written to assist non medically trained readers to understand general points of the Original Article. It is supplied “as is” without any warranty. You should note that the Original Article (and Summary) may not be fully relevant nor accurate as medical science is constantly changing and errors can occur. It is therefore very important that readers not rely on the content in the Summary and consult their medical professionals for all aspects of their health care and only rely on the Summary if directed to do so by their medical professional. Please view our full Website Terms and Conditions.
    Copyright © 2019 BMJ Publishing Group Ltd & European League Against Rheumatism. Medical professionals may print copies for their and their patients and students non commercial use. Other individuals may print a single copy for their personal, non commercial use. For other uses please contact our Rights and Licensing Team.

Footnotes

  • Handling editor Josef S Smolen

  • Contributors JM, SS and MWH conceived the study, with support from MF and AE for the study design. HE, GB, SES recruited patients and recorded clinical data. ASA and MWH extracted register data. JM analyzed the data with input from HE, GB, SES and MWH. JM wrote the first version of the manuscript with input from HE, AA, GB, SES and MWH, and all authors participated in the editing until its final version.

  • Funding The study was supported by grants from the Swedish Research Council, the Swedish Rheumatism association, the King Gustaf the V:th 80-year foundation, the Heart-Lung Foundation, the Freemason’s in Stockholm Foundation for Children’s Welfare, the Stockholm County Council, the Karolinska Institute, and the Torsten and Ragnar Söderberg Foundation.

  • Competing interests None declared.

  • Patient consent Not required.

  • Ethics approval The study was approved by the Regional Ethics Committee in Stockholm, and patients or guardians gave written informed consent.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement The ethical permit granted for this study does not allow for sharing to a third party.