Background
Systemic lupus erythematous (SLE) is an autoimmune disease that involves multiple organs and develops mainly in women of reproductive age. In the last decades, owing to improvements in multidisciplinary perinatal management, SLE was no longer viewed as a contraindication for pregnancy [
1]. With the increase of pregnancy in women with SLE, numerous studies have reported that SLE is associated with increased risk of fetal complications including fetal loss, preterm delivery (20.8-28.7%)
2,3,4,5, and intrauterine growth retardation.
Such studies have focused on the pregnancy outcomes of SLE [
6‐
9] while there is only very few observational studies have assessed the outcome in preterm offspring of mothers with SLE [
10]. Furthermore, pregnant women with SLE are at greater risk of thrombosis, infection, preeclampsia and multiple drugs would administrate during period of pregancy [
11,
12], but hardly any study have looked at the influence of SLE on the outcome of preterm infants.
Understanding the outcomes of preterm infants born to SLE populations can inform clinical care and potentially provide mechanistic insight into such obstetric complication. Accordingly, the primary objective of this study is to assess the impact of SLE on premature newborns by comparing demographic data, prenatal and postnatal characteristics, laboratory data, and morbidities in a cohort of premature infants born to mothers with and without SLE. The secondary objective is to assess the influence of active SLE on preterm birth infants among infants with maternal SLE.
Materials and methods
Study design, setting and participants
This retrospective cohort study was carried out from February 2012 to May 2021 in the Neonatology Department of Shanghai Children’s Medical Center in Shanghai, China. Infants with preterm birth were included in this study. Infants with maternal SLE were identified from the database of medical records in our neonatology department born between February 2012 and May 2021. The exclusion criteria include: death during hospitalization, neonatal lupus and major congenital anomalies. The study participants were followed up until discharge.
Data collection
Data of all preterm participants were collected by trained data abstractor in Shanghai Children’s Medical Center. All SLE mothers in this study delivered in Ren Ji Hospital which located next to our hospital. Ren Ji Hospital is the largest prenatal center of both autoimmune disease and high-risk pregnancy in Shanghai. Almost all pregnant women with SLE in this region will give birth in Renji Hospital. Maternal history were collected by trained data abstractor in Ren Ji Hospital. Site investigators were responsible for data quality control in both hospitals. Data were collected using the same definition in both hospitals.
Exposure
SLE was diagnosed according to the revised criteria for the classification of SLE developed by the American College of Rheumatology [
13] The non-SLE group was randomly selected from preterm infants admitted in our center. Non-SLE group were 1:2 matched with SLE group based on gender, birth weight (BW) ± 300 g, gestational age (GA) ± 6 days and date of birth ± 6 months. Infants in non-SLE group with major congenital anomalies, confirmed intrauterine infection, death during hospitalization and born to mother with other autoimmune diseases or malignant disease were excluded. The disease activity of SLE was made an appropriate evaluation based on the SLE Disease Activity Index 2000 (SLEDAI-2 K) [
14].
Outcome
The primary outcome was defined as infants who survived without major morbidities. Major morbidities of premature included intraventricular hemorrhage (IVH) (grade ≥ 3) and/or cystic periventricular leukomalacia (PVL), necrotizing enterocolitis (NEC) (stage ≥ 2), sepsis, Retinopathy of prematurity (ROP) (stage ≥ 3), and bronchopulmonary dysplasia (BPD). IVH was defined as greater than or equal to grade 3 according to the Papile criteria [
15]. Cystic PVL was defined as the presence of periventricular cysts identified on cranial ultrasonography or magnetic resonance imaging. NEC was defined according to Bell criteria [
16,
17]. Sepsis was defined as positive blood or cerebrospinal fluid culture (Exception for specimen contamination) [
18]. Retinopathy of prematurity was diagnosed according to the International Classification of Retinopathy of Prematurity [
19]. Bronchopulmonary dysplasia was defined as treatment with FiO
2 > 0.21 for at least 28 days plus failure of room air challenge test at 36 weeks’ postmenstrual age [
20]. Secondary outcome included complete blood count, total bilirubin (TBil) and alanine aminotransferase (ALT).
Definitions of other covariates
Pregnancy-induced hypertension (PIH) was defined as an increase in blood pressure to ≥ 140/90 mmHg on at least two occasions ≥ 6 h apart that arises de novo after the 20th week of pregnancy. Preeclampsia (PE/E) was defined as pregnancy-induced hypertension with proteinuria > 0.3 g/L/d in the absence of a urinary tract infection or the abrupt onset of hypertension and proteinuria after 20 weeks of gestation. Seizures were required for a diagnosis of eclampsia. Gestational diabetes (GDM) was defined as any degree of glucose intolerance with onset or first recognition during pregnancy. HELLP (hemolysis, elevated liver enzymes and low platelet count) syndrome was defined as presence of hemolysis, high levels of lactate dehydrogenase or total bilirubin > 12 mg/L, elevated alanine aminotransferase levels of greater than twofold the upper normal value, and thrombocytopenia < 100*10
9/L. Small for gestational age (SGA) was defined as infants whose weight was lower than the lower 10% limit of the CI of the Fenton curve. An Apgar score > 7 was defined as normal, while a score ≤ 7 was considered indicative of moderate or severe hypoxia [
21].
Statistical analysis
Normally distributed continuous data were expressed as mean and standard deviation (SD) and differences between two groups were tested by independent t-test. Categorical variables were presented with frequency and percentage and using chi-square test to exam the differences among groups.
In order to determine the association of neonatal outcome with maternal SLE in preterm birth, we performed multiple logistic regression analysis after adjustment of 1 min Apgar score and delivery type. Multiple linear regression was also performed in order to determine the association between maternal SLE and biochemical parameters. Log-transformation was conducted if biochemical parameters were not normally distributed before conducting linear regression.
Bi-variate analysis of baseline and neonatal outcome was conducted between active SLE and non-active SLE during pregnancy. In order to further investigate the effect of active SLE during pregnancy on neonatal outcome among infants born to maternal SLE, Multiple logistic regression analysis was conducted after adjustment of potential confounders. Each variable which has p-value < 0.1 was entered into the multivariable logistic regression model to control possible confounders. All tests were two-tailed, and p < 0.05 was considered statistically significant. Statistical analysis was performed using SPSS 22.0 and R 4.1.2.
Ethics statement
This study protocol was reviewed and approved by the Ethics Board of Shanghai Children’s Medical Center, School of Medicine, Shanghai Jiao Tong University (SCMCIRB-K2021016-1). Written informed consent was obtained from all participants’ legal guardian. Approval to obtain clinical data from the database was received from the office of the medical director of the hospital. All patient information was kept confidential.
Discussion
In this retrospective cohort study, we found that there was no significant difference of major morbidities between SLE group and non-SLE group. However, SLE preterm offspring showed a hematologic profile that is different from that of non-SLE preterm offspring, with lower leukocyte counts, lower neutrophile and lower platlet.
The intact survival rate was 83.7% and the average GA was 33 weeks in SLE group, only 1% premature infants had a GA less than 28 weeks of which had most morbidities. No significant difference was found in each single morbidity including BPD, PVL, ROP, NEC, and Sepsis between the two groups. Mei-Ying et al. reported that SLE preterm offspring had much higher rate of neonatal infection as compared to preterm babies from women without SLE (47.62% vs. 16.81%)11. In their study the mean GA was 33 weeks plus 4 days which was similar as our study subject, the definition of sepsis in our study was positive blood or cerebrospinal fluid culture while their definition of neonatal infection was covering a wider range. Our definition was more precise, the broad definition may lead to bias. Furthermore, the non-SLE group they enrolled was all other premature infants delivered during the same time period, whereas our study matched the individuals’ GA and weight, excluding the confounding facts of varied GA and weight.
In our study hematological abnormalities were rare in both groups, however significant lower leukocyte, neutrophile and platelets were found in SLE preterm offspring than in non-SLE group. Maria Gariup et al. found that SLE Offspring (mean age 14.9 years) showed an immune profile that is different from relevant health control, with lower leukocyte counts and higher levels of mainly proinflammatory cytokines [
23]. Besides their study showed SLE offspring had significantly higher proportions of history of nonallergic autoimmune conditions, and asthma was more frequent in SLE offspring than in health control. A cohort of 719 SLE offspring revealed that instead of rheumatic autoimmune disease, children born to women with SLE had an increased risk of non-rheumatic autoimmune diseases [
24]. All these suggested that SLE offspring may have proinflammatory and autoimmune activation. The alteration in immune profile of SLE offspring seems to last for a long duration after birth while the maternal antibodies including Anti-SSA and anti-SSB in the offspring circulation are progressively reduced after birth and diappeared at about 12-month [
25]. This phenomenon cannot be simply explained by maternal antibodies directed against auto-antigens in the fetal blood stream. Studies indicate that multiple molecular as well as cellular components originating in pregnant women are transferred to the fetus and program the fetal immune system [
26]. SLE pregnancy might modify these signals and may consequently alter immunity in early life and childhood. On the other hand, the previous studies have proven familial risks between rheumatic autoimmune diseases (RAIDs), and SLE was associated with four of five RAIDs [
27]. Further studies have highlighted a genetic component in the onset of SLE and other autoimmune diseases that may lead to several different phenotypes even though the genotype is the same [
28]. Many of the shared loci were related to immune processes, such as interferon signalling and polymorphisms of STAT genes that may have a wide spectrum of phenotype [
28,
29].
The incidence of CHD in this study was much higher than previous reports and with no difference from the non-SLE group. In a large population-based study conducted by Evelyne Yinet et al., reported that in comparison with children from the general population, children born to women with SLE have an increased risk of CHD 5.1% [95% CI, 3.7–7.1] versus 1.9% [95% CI, 1.6–2.2] [
30]. One potential reason for these differences in the incidence of CHD is that the GA was significant different between two studies. The mean GA of SLE offspring in their study was 37.7 weeks which was significant different from control children whose mean GA was 38.8 weeks. Epidemiologic study suggests that premature neonates have a more than 2-fold higher risk of cardiovascular abnormalities [
31]. CHD is more likely to be associated with prematurity instead of SLE.
In our study the maternal age, SLE disease duration, SLE active, and antibodies characteristics were similar as several studies previously reported in Asia [
4,
32‐
34]. The incidence of pregnancy complications in this cohort was significantly higher than other SLE cohorts. The results of our study showed that PIH and preeclampsia are the significant problems in SLE pregnancies with preterm infants. In our study, nearly one-third of SLE mothers were complicated with PIH and preeclampsia, whereas according to previous studies only 3.1–19.2% of the pregnancies in SLE women were affected by such conditions [
4,
35]. Several studies have explored that PIH can significantly increase the risk of preterm birth [
36], while our subjects are all preterm infants. Yen-Ju Chen et al. has reported that preterm risks increased markedly in participants with both preeclampsia/eclampsia and SLE (OR: 17.5, 95% CI: 12.6_24.1, p < 0.01)
32. Therefore, the high incidence of PIH and preeclampsia in our cohort is understandable and we speculate that SLE pregnant women complicated with PIH/PE are more likely to give birth prematurely than those SLE mothers without PIH/PE.
The rate of cesarean section in SLE group was significantly higher than non-SLE group in this study (91.8% vs. 78.5%). Jae-kyoon Hwang et al. reported a slightly lower cesarean section rate by 84.8% in preterm SLE offspring [
37]. In SLE women, the general cesarean rate was previously reported to be 30-40%
2,38,39,40. In our study, high cesarean section rate could be the confounding factor of SLE, premature birth and high rate of PIH/PE.
Further analysis in this study showed that maternal status including SLE active during pregnancy, renal and blood system involvement and pregnancy complications were associated with smaller offspring GA and lower offspring birth weight. Studies have showed that maternal SLE activity in the last 6–12 months or at conception increased the risk of maternal disease activity (subsequent flare during pregnancy and puerperium) and hypertensive complications, fetal morbidity and mortality [
41,
42]. Considering the high risk of active SLE during pregnancy, EULAR (European League Against Rheumatism) recommends that SLE patients plan their pregnancies at least 6 months after achieving remission [
42].
Taking aspirin during pregnancy was associated with larger offspring GA and higher offspring birth weight in this study, besides analysis also indicated that aspirin administration during pregnancy would increase the rate of survival without major morbidities. Because of the significant risk of preeclampsia in all lupus pregnancies, especially those with preexisting renal impairment or active lupus, low-dose aspirin is indicated in all women with SLE starting around 12 weeks gestation [
43]. There is no study reported the long-term effect of aspirin on offspring outcomes. Here according to the mediation model, GA interaction can explain the effect of aspirin on survival without major morbidities 51.7% which means aspirin reduces morbidities of preterm infants by increasing gestational weeks. Besides, univariate logistic regression analysis showed that in very preterm infants of taking aspirin during pregnancy with survive without major morbidities in among SLE group.
To our knowledge, this was the first retrospective cohort study based on nearly 10 years clinical data of preterm infants born to mothers with SLE. Additionally almost every key influence of the SLE pregnancy on preterm infants was included in this study. Nevertheless, there were several limitations to our study. Firstly, the sample size of this retrospective study is not very large and the actual sample size for preterm morbidities comparison was considerably small, however, the sample size of this study is one of the largest sample sizes of all studies at present. Secondly, in order to control confounding factors, the matching criteria in this study are relatively strict, we can only achieve 1: 2 matching. Finally, the end point of our study was discharge. The long-term outcome is very important for preterm infants. In the future, we may carry out a prospective cohort study of preterm offspring of SLE.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.