Neonatal Outcomes in Pregnant Women with Systemic Lupus Erythematosus: A 13-Year Experience in Southern Thailand

Abstract

Objectives

The purpose of the study was to determine the clinical features of NLE and to compare the neonatal outcomes between newborns born to pregnant women with SLE and healthy pregnant women.

Methods

We conducted a retrospective cohort analysis between 2007 and 2019 in a tertiary referral hospital in Thailand. A total of 118 pregnant women with SLE with 132 neonates compared with 264 randomly selected healthy pregnant women.

Results

The median (interquartile range) gestational age and birth weight of 132 neonates born to women with SLE were 37 (35–38) weeks and 2687 g (2045–3160 g), respectively. The clinical features of NLE infants were hemolytic anemia (8%), thrombocytopenia (2.7%) and hyperbilirubinemia (5.3%). There was no neonate with a congenital complete heart block or skin lesion. Moreover, logistic regression analysis found that neonates born to women with SLE increased the risk of preterm birth [odd ratio (OR) 8.87, 95% confidence interval (95% CI) 4.32–18.21, p < 0.001], low birth weight (OR 10.35, 95% CI 5.08–21.08, p < 0.001), birth asphyxia (OR 2.91, 95% CI 1.26–6.73, p = 0.011) and NICU admission (OR 4.26, 95% CI 2.44–7.42, p < 0.001). SLE disease activity and corticosteroid and azathioprine usage were associated with preterm delivery in pregnant women with SLE.

Conclusion

The major clinical features of NLE patients were hematologic and hepatobiliary abnormalities in our study. Pregnancies with SLE dramatically increased the risk of preterm delivery and neonatal complications.

Lay summary

Neonatal lupus erythematosus (NLE) is the consequence of the transplacental passage of autoantibodies to newborns during pregnancy. The clinical features of NLE infants in our study were hemolytic anemia (8%), thrombocytopenia (2.7%) and hyperbilirubinemia (5.3%). There was no neonate with a congenital complete heart block or skin lesion. We also compared the neonatal outcomes between 118 pregnant women with SLE and 264 randomly selected healthy pregnant women. Our study found that the neonates born to women with SLE increased the risk of preterm birth, low birth weight, birth asphyxia and NICU admission. Moreover, SLE disease activity and corticosteroid and azathioprine usage were associated with preterm delivery in pregnant women with SLE.

INTRODUCTION

Systemic lupus erythematosus (SLE) is the most common autoimmune disease generally seen in women of child-bearing age. Clinical manifestations of SLE were diversified, arising from the presence of multiple autoantibodies. The consequence of the transplacental passage of autoantibodies to newborns during pregnancy was known as neonatal lupus erythematosus (NLE) [1]. The incidence of NLE was 1 in 12 500–20 000 live births among women with or without SLE, but the exact incidence of NLE is still unknown due to underdiagnosis [2]. The criteria for diagnosis of NLE were unclarified, but the recognized manifestations involved were congenital complete heart block (CCHB) and skin lesions. The other organs involved in NLE patients were usually found in hematologic and hepatobiliary manifestations. However, most of those abnormalities can be resolved spontaneously along with the absence of autoantibodies in neonates, with the exception of CCHB, which was permanent and contributed to severe neonatal morbidities [1].

A literature review stated that SLE increased the risk of obstetric complications, such as pregnancy-induced hypertension, pre-eclampsia and premature membrane rupture [3]. In addition, the fetal/neonatal complications of SLE during pregnancy included abortion, preterm birth, intrauterine growth restriction and neonatal death [4, 5]. Previous studies found an association between anti-SSA/Ro and NLE, in particular CCHB, in pregnant women with or without connective tissue disease. However, studies on other autoantibodies or other neonatal complications were inconclusive. Disease activity during pregnancy is an indicator of pregnancy outcomes and high activity associated with poor obstetric outcomes [6]. Given its own SLE disease, treatment of SLE can have an effect on obstetric and neonatal complications. Corticosteroid, the common medication used in SLE patients, might increase risk of pregnancy complications, such as gestational diabetes, pregnancy-induced hypertension and fetal growth restriction, as well as complications from other immunosuppressive drugs [7].

In our study, the research focused on the manifestations of NLE in newborns born to pregnant women with SLE and the comparison of neonatal outcomes between newborns born to pregnant women with SLE and healthy pregnant women without SLE. Furthermore, this study also discusses factors associated with neonatal complications in newborns born to pregnant women with SLE.

MATERIALS AND METHODS

Study population

This retrospective cohort study examined 132 neonates born to 118 pregnant women with SLE between January 2007 and December 2019 at Songklanagarind Hospital, a tertiary referral hospital in Southern Thailand. SLE was diagnosed by rheumatologists according to the criteria for the classification of SLE [8]. Additionally, all pregnant women with SLE were assessed for clinical condition, and medications were modified by rheumatologists.

During the study period, all neonates born to pregnant women with SLE were recruited for statistical analysis. The control group (n = 264) was randomly selected from healthy pregnant women who were admitted to Songklanagarind Hospital during the study period. Control pregnant women were screened and matched with age, parity and time (year) of delivery in a ratio of 2 : 1 (two controls were selected for every pregnant woman with SLE).

Data collection

Demographic data were collected from maternal medical history records, including age at delivery, pregnancy complications, timing of SLE diagnosis (before, during or after pregnancy), SLE disease activity during pregnancy, medications during pregnancy and autoantibodies (especially anti-SSA/Ro and SSB/La). Active SLE disease was described as women with clinical manifestations of SLE activity, including renal and central nervous system involvement, fever, arthritis, vasculitis, pleurisy, pericarditis, rash and hypocomplementemia [3]. Furthermore, neonatal data were reviewed from the neonatal chart, including sex, gestational age, birth weight, Apgar score (at 1 and 5 min), neonatal comorbidities (any causes of respiratory distress, hypotension, patent ductus arteriosus and clinical/culture-proven sepsis) and neonatal intensive care unit (NICU) admission, and were compared with the control group as secondary outcomes.

Preterm birth was defined as delivery <37 weeks, while low birth weight and extremely low birth weight were defined as birth weights of <2500 and 1000 g, respectively. Asphyxia was diagnosed when the Apgar score <7 at 1 min. Effects of NLE in newborns born to pregnant women with SLE were reviewed in each system: cardiovascular manifestations (arrhythmias, especially CCHB), hematologic manifestations (hemolytic anemia, autoimmune thrombocytopenia and leucopenia), hepatobiliary manifestations (hyperbilirubinemia and transaminitis) and common skin manifestations. Laboratory tests, including total and differential blood cell counts, liver function tests, anti-SSA/Ro and anti-SSB/La levels and electrocardiography, were examined. Anemia was defined when hemoglobin level was two standard deviations (SDs) below the normal mean value for gestational age, leucopenia was defined as white blood cell counts of <5000/µl and thrombocytopenia was defined as platelet counts <150 000/µl. Moreover, hepatic transaminitis was defined as alanine aminotransferase and/or aspartate aminotransferase >3 times the upper limit of normal. The standard 12-lead electrocardiographs (ECG) were performed prior to discharge from the hospital.

The study was approved by the Institutional Review Board and the Ethics Committee of the Faculty of Medicine, Prince of Songkla University, Songkhla, Thailand (REC 61-278-1-1).

Statistical analysis

Descriptive data were presented as mean ± SD or median with interquartile range (IQR). Nominal variables are expressed as the number of infants and percentages. The statistical analysis was conducted by using the Student’s t-test and Wilcoxon rank sum test for continuous variables and the chi-square test and Fisher exact test for categorical variables for comparison between two groups. Odd ratio (OR) with corresponding 95% confidence interval (95% CI) was calculated for neonatal outcomes using the conditional logistic regression model. A p-value of <0.05 was found to be statistically significant. All analyses were performed using the Epicalc package in R Software version 3.3.1 (R Foundation for Statistical Computing, Vienna, Austria).

RESULTS

Pregnant women with SLE characteristics

In this study, 118 women with SLE had 132 pregnancies, with a median age of 31 (26–35) years for each delivery. Additionally, 48 (36.4%) pregnant women with SLE had active SLE, the most common active symptoms being proteinuria or lupus nephritis (77%) and autoimmune hemolytic anemia (25%). In addition, during pregnancy, medications, such as corticosteroid, hydroxychloroquine and azathioprine, were used as 66.7%, 46.2% and 14.4%, respectively. SSA/Ro antibodies were determined in 82 patients and positive for anti-SSA/Ro in 49 (60%) patients, while SSB/La antibodies were determined in 89 patients, and positive results were found in 13 (14.6%) patients.

Neonatal baseline characteristics and outcomes

During the study period, 132 neonates were born to 118 pregnant women with SLE. Of the 132 neonates, 75 (56.8%) neonates were classified as preterm infants. Low birth weight and extremely low birth weight infants were found to be 40.9% and 5.3%, respectively. There were 70 male neonates and 62 female neonates. The ratio of male to female newborns was 1.13 : 1. Respiratory distress syndrome (RDS) was the most common cause of respiratory distress in newborns born to pregnant women with SLE, which accounted for 7.6% of all newborns. Other comorbidities, including hypotension and clinical sepsis, were also found to be 8.3% and 18.2% of all newborns, respectively. However, there was only one case reported death from prematurity with intracranial bleeding.

NLE manifestation

Diagnosis of neonatal SLE occurs when the neonate develops CCHB, a common skin rash, or hepatic or hematologic manifestations in the absence of another explanation. In our study, there was no reported case of CCHB or any arrhythmias. However, seven (5.3%) neonates born to pregnant women with SLE had structural heart defects, including four patent ductus arteriosus, two ventricular septal defects and one Tetralogy of Fallot, whereas there were two neonates (0.8%) with patent ductus arteriosus in the control group. There was also no reported case of either typical or atypical skin manifestation of NLE prior to discharge from the hospital. Moreover, 41 (31%), 12 (9.1%) and 6 (4.5%) patients were followed up at 1, 2 and 4 months of age, respectively. There were no abnormal skin rashes in any of the patients who were followed up.

In this study, the majority of manifestations were observed in two organ involvements, such as hematologic and hepatobiliary, summarized in Table 1. Blood tests were conducted in 112 patients (85%), and hematologic diseases, such as hemolytic anemia, autoimmune thrombocytopenia and leucopenia, were observed in 14 patients (12.5%). Neonatal jaundice was screened in all patients, and hyperbilirubinemia from hemolysis was found to be 5.3%. Moreover, liver function tests were analyzed in 44 patients (33.3%), and only 1 (2.3%) patient had hepatic transaminitis and was spontaneously resolved prior to discharge.

In our study, 48 (36.4%) pregnant women with SLE had active SLE and 10 (21%) neonates born to pregnant women with active SLE had NLE manifestations. There were three neonates with hyperbilirubinemia, six neonates with hemolytic anemia and one neonate with thrombocytopenia. Therefore, blood tests screening for hematologic abnormalities should be performed before hospital discharge especially in neonates born to pregnant women with active SLE.

A comparison of neonatal outcomes between newborns born to pregnant women with SLE and healthy pregnant women was shown in Table 2. The median (IQR) birth weight of neonates born to maternal SLE and healthy pregnant women were 2687 g (2045–3160 g) and 3069 g (2845–3324 g), respectively (p < 0.001). The rate of preterm birth in pregnant women with SLE was statistically significantly higher than in healthy pregnant women. Furthermore, low birth weight infants, birth asphyxia and RDS were significantly higher in neonates born to pregnant women with SLE. Other complications, such as hypotension, clinical sepsis, hyperbilirubinemia and bronchopulmonary dysplasia were also found to be significantly higher in neonates born to pregnant women with SLE. Moreover, the NICU admission of newborns born to pregnant women with SLE was also higher than that of newborns in the control group. However, there was no statistically significant difference between the two groups in length of hospital stay and the mortality rate.

Factors associated with neonatal outcomes

Medication data used in SLE patients during pregnancy were analyzed, and we found that corticosteroid use was associated with preterm birth and low birth weight (p = 0.006 and p = 0.015, respectively). Moreover, azathioprine was used in 19 patients and was associated with preterm birth (p= 0.014) the same as corticosteroid. We also analyzed the association between disease activity and neonatal outcomes. Hence, statistically significantly higher incidences of preterm birth, low birth weight and NICU admission were observed in pregnant women with active SLE disease (Table 3).

In our study, there were 75/132 (56.8%) preterm infants born to maternal SLE and 29/75 (38.7%) preterm infants born to maternal SLE with active disease. Of 29 preterm infants, 10 infants born to maternal active SLE received azathioprine, while 19 infants born to maternal active SLE did not receive azathioprine. So, the main factor associated with premature birth could be disease severity/flare up rather than drugs exposure.

In addition, the serological findings as anti-SSA/Ro and anti-SSB/La were reviewed, and there was no statistically significant difference in neonatal outcomes and neonatal lupus manifestations between positive or negative anti-SSA/Ro and anti-SSB/La.

DISCUSSION

NLE is a rare disease that results from maternal autoantibodies against RNA–protein complex, Ro/SSA, or SSB/La, and may be present in newborns born to pregnant women with SLE, other connective tissue diseases, or healthy women with autoantibodies [9]. Previous studies reported NLE was found in ∼1–2% of infants born to maternal anti-Ro or anti-La antibodies [10]. However, the true incidence of NLE is not known.

In our study, 22 (16.6%) neonates born to pregnant women with SLE had NLE manifestations. Clinical manifestations of NLE patients were cytopenia, jaundice and hepatic transaminitis. In our study, we did not find skin lesions or CCHB. This finding is different from what has been widely documented in previous studies of skin rashes 16–78% [2, 11] and CCHB 4–15% [11, 12]. However, there were fewer patients with CCHB (1.6–2%) and skin rashes (3.5%) in neonates born to pregnant women with SLE in some studies [2, 13, 14], similar to our study.

Skin lesions in NLE can occur at birth but typically occur within a few weeks of birth [9]. In our study, 69% of neonates did not attend clinical follow-up at 1 month of age, so the late skin manifestation would be underdetermined. The previous study reported the relationship between CCHB and maternal anti-Ro/SSA antibodies [15]. In comparison, 49 pregnant women with SLE had anti-Ro/SSA in this study, but their neonates did not have CCHB. Similar to our findings, 3–8% and 4% of NLE patients reported anemia and hyperbilirubinemia, respectively, in previous studies [11, 16].

Our study revealed adverse neonatal outcomes in pregnant women with SLE. Almost 60% of neonates born to pregnant women with SLE were preterm birth, which is higher than in the previous study (28–39%) [3, 5, 13]. The recent publication reported that low birth weight and NICU admission were significantly associated with SLE [17], which is close to our findings. Moreover, asphyxia is a major neonatal morbidity, and asphyxia has been found to be significantly higher in neonates born to pregnant women with SLE, consistent with the findings of the recent study [17]. However, our study had high rate of premature birth which is a factor associated with birth asphyxia. There was not substantial difference in the number of infants with asphyxia between pregnant women with SLE and control group in Nili, et al. [12] and Wu, et al. [3] studies.

In our study, we analyzed the other preterm complications between pregnant women with SLE and the control group. Significantly higher incidences of RDS and bronchopulmonary dysplasia were found in neonates born to pregnant women with SLE, similar to the findings in a study by Nili, et al. [12]. In addition, we also observed significantly higher incidences of neonatal sepsis and hypotension in the maternal SLE group.

This study reported that preterm birth, low birth weight infants and NICU admission were significantly higher in neonates born to women with active SLE, compared to previous studies. In a study conducted by Georgiou, et al. [5], a higher incidence of fetal loss was identified in women with active SLE, while Yang, et al. [18] also found that pregnant women with active SLE had a 3-fold greater risk of preterm deliveries. In addition to the disease activity, we found the association between medication usage for SLE during pregnancy and neonatal outcomes. Corticosteroid increased the risk of preterm birth in pregnant women with SLE in our study. Similar to a study by Skorpen, et al. [19], which reported a statistically significant 3-fold increase in preterm birth and low birth weight in women using prednisolone, Clark, et al. [20] also found a significantly higher rate of preterm delivery in women, taking more than 10 mg/day of prednisone. Azathioprine is an immunosuppressive drug, and the previous study indicated that azathioprine is safe and not associated with adverse neonatal outcomes, such as preterm birth, low birth weight and neonatal death [21], whereas there was significantly increased preterm birth and a propensity to increase low birth weight infants (p = 0.06) in SLE patients who received azathioprine during pregnancy in our study.

This study had some notable strengths and limitations. The main strength is that all of neonates who born to maternal SLE in our hospital were enrolled in this study and underwent the routine jaundice screening and performed ECG before discharge. The main limitation is that there was a lack of complete long-term follow-up visits to assess late manifestations of NLE, growth and neurodevelopmental outcomes.

CONCLUSIONS

The most clinical manifestations of NLE are hematologic and hepatobiliary abnormalities. Pregnancies with SLE dramatically increased the risk of preterm delivery, low birth weight and NICU admission. Disease activity of SLE and corticosteroid and azathioprine usage were associated with preterm delivery in pregnant women with SLE. Well-controlled disease activity and close monitoring of pregnant women with SLE and their treatments result in improved fetal outcomes. Moreover, the carefully physical examination with standard routine laboratory investigations in neonates born of maternal SLE to detect lupus symptoms can be helpful in optimizing care and improving neonatal outcomes.

ACKNOWLEDGEMENTS

The authors thank Ms Nannapat Pruphetkaew in the Epidemiology Unit at the Faculty of Medicine, Prince of Songkla University, Thailand for the statistical analysis.

REFERENCES