Baseline characteristics and history of pregnancies in the autoantibody subgroups
A total of 124 patients (126 pregnancies) with UCTD were included in this study. These patients exhibited different clinical symptoms, with joint pain in 18 (14.3%) patients, rash in 12 (9.5%) patients, thrombocytopenia in 7 (5.6%) patients, and oral ulcers in 9 (7.1%) patients, 6 (4.8%) patients had alopecia, and 4 (3.2%) patients had other symptoms such as dry eyes at enrollment. Most patients had a history of adverse pregnancy outcome (93/126, 73.8%).
More patients in the non-aPL group had no previous pregnancy (30.2% vs. 13.7%, p=0.024), while more patients in the aPL group have undergone failure pregnancies in the past than those in the non-aPL group (63.0% vs. 45.3%, p=0.048), especially the incidence of failure pregnancies <3 times between the two groups was statistically different (67.1% vs. 48.1%, p=0.041). Patients in the aPL group also experienced more unexplained spontaneous abortion within 10 weeks than the non-aPL group (64.4% vs. 45.3%, p=0.033). (Table 1)
Due to the difference in autoantibodies at enrollment, the positive ANA and anti-SSA/SSB antibodies in the non-aPL group were significantly higher than those in the aPL group (98.1% vs. 49.3%, p<0.001; 32.1% vs. 16.4%, p= 0.040, respectively).
Some patients in the study have been treated with medicine before pregnancy, mainly including prednisone (Pred), hydroxychloroquine (HCQ) and calcineurin inhibitors (CNIs) and low-dose (50-100mg/d) aspirin (LDA). The usage of HCQ in the aPL group was 38.4%, which was higher than that of HCQ in the non-aPL group (20.1%) (p=0.035). The usage of LDA in the aPL group was more than that used in the non-aPL group in the baseline (24.7% vs. 9.4%, p=0.029).
In the aPL group, the percentage of positive anti-β2GPI antibodies (IgG/IgM) was the highest (31/73, 42.5%). To be specific, the proportion of the positive aβ2GPI-IgM (with or without other positive aPLs) was 35.6%. The positive rate of aβ2GPI-IgA reached to 21.9%. In this group, 19 patients agreed to the detection of other NC-aPLs. The percentage of aVim/CL was 52.6% (10/19), and the positive rate of aβ2GPI-D1 and Anxa2-Ab were both 42.1%. (Table 2)
Maternal and fetal DUAEs during pregnancy in the autoantibody subgroups
We observed maternal AEs and fetal DUAEs in the first, second, and third trimesters of pregnancy patients with UCTD and analyzed them in this study. As mentioned above, fetal DUAEs were primarily revealed by ultrasonography during pregnancy, and maternal AEs included thrombocytopenia, abnormal liver and kidney function, low complement C3/C4, and positive urine protein, and patients with one or more AEs (total number of maternal AEs) were also recorded. (Table 3)
A total of 11 (20.8%) patients in the non-aPL group had one or more of their maternal AEs in the first trimester, and a total of 16 (21.9%) patients in the aPL group had a maternal AE. Maternal AEs occurred in 13 patients (26.5%) in the non-aPL group in the third trimester, which was significantly higher than that in the aPL group (10.6%) (p=0.026).
In order to study the impact of fetal adverse events on pregnancy outcomes in patients with UCTD, they were divided patients into with /wo fetal DUAEs in the non-aPL group and aPL group. More patients in the aPL group who had fetal DUAEs experienced failed pregnancies in the past than that of patients in the non-aPL group (73.7% vs. 40.9%, p=0.035). In the non-aPL group, the proportion of thrombocytopenia in patients with fetal DUAEs (22.7%) was higher than that of patients with non-fetal DUAEs (3.2%), but there was no significant difference. (Table 4)
Treatment
The treatment in this study mainly includes Pred, HCQ (200-400mg/d, oral, twice in the morning and evening), and anticoagulant therapy, including LDA (50-100mg/d to 32 weeks) and low molecular weight heparin (LMWH) (4000-6000 IU/d). Patients were treated with immunosuppressants if necessary, such as tacrolimus (TAC), cyclosporin-A (CsA) and azathioprine (AZA). In critical situations during pregnancy, such as massive placental hematoma, sudden drop in amniotic fluid, and severe IUGR, we also recommend intravenous immunoglobulin (IVIG) as appropriate.
In this study, the usage of single anticoagulant therapy (Pred+HCQ+LDA/LMWH) in the non-aPL group was significantly higher than that of the aPL group in the first and third trimester (34.0% vs.11.0%, p=0.002; 26.5% vs. 9.1%, p=0.013, respectively). In the same, the usage of combined anticoagulation therapy (Pred+HCQ+LDA +LMWH) in the aPL group was significantly higher than that of the non-aPL group in the first, second, and third trimester (p=0.001, p=0.006, p=0.001, respectively). (Table 3)
In the non-aPL group, 2(10.5%) patients received IVIG in the third trimester due to acute thrombocytopenia. One delivered prematurely at 32 weeks gestation with a neonatal weight of 1500g, the other patient delivered at 39 weeks gestation and the newborn weighed 3100g. Two other patients (6.5%) were treated with IVIG in the second trimester (one patient with FGR and maternal thrombocytopenia, and the other due to FGR with placental hematoma). The former eventually aborted at 28 weeks gestation, the latter was premature at 35 weeks gestation and the weight of newborn was 2100 g.
In the aPL group, 4 (6.7%) patients were treated with IVIG due to poor doubling of β-HCG in the first trimester, 1 (1.8%) patient was treated with IVIG because of sharp decrease in amniotic fluid and placental hematoma, and 3 (5.5%) patients were used in the third trimester (all of them experienced FGR). A total of 8 patients were treated with IVIG during pregnancy, 2 patients had miscarriage at 8 weeks gestation and 35 weeks gestation (the patient was also complicated with pulmonary arterial hypertension), 1 patient was premature at 30 weeks, and the remaining 5 patients’ gestational week were all >37 weeks, the weight of the newborns was >2500g, and the mean of which was 3186.0±256.87g. There were no differences in the usage of CNIs and IVIG between the two groups according to the types of autoantibodies. (Table 3)
With the idea of using LDA to prevent DUAE and APO during pregnancy, more patients with non-fetal DUAEs in the non-aPL group used LDA than patients with DUAEs (p=0.008). On the other hand, more patients with DUAEs in the aPL group used LDA than those with DUAEs in the non-aPL group, due to the strategy for prevention APO (p=0.043).
There was the same tendency in the usage of LMWH, which was significantly higher in the patients with fetal DUAEs of aPL group than those of non-aPL group (100% vs.63.6%, p=0.011). And the usage of LMWH in the patients without DUAEs in the aPL group was significantly higher than that in the non-aPL group, too (92.6% vs.71.0%, p=0.019). The average dosage of Pred in patients with DUAEs in the aPL group was 12.28±5.32mg/d, which was significantly higher than that in patients with fetal AEs in the non-aPL group (8.40±3.93mg/d) and patients with no DUAEs in the aPL group (7.61±3.92mg/d) (p=0.005, p<0.001, respectively). The usage of IVIG in the patients with fetal DUAEs in the aPL group (26.3%) was significantly higher than that of patients with non-DUAEs (p=0.039). (Table 4)
Pregnancy outcomes
Among all the patients in this study, the rate of live birth in the non-aPL group was 90.6% (48/53), and the rate in the aPL group was 87.7% (64/73). The percentage of APO in the former group was 22.6% (12/53), and it was 27.4% (20/73) in the latter. Three (2.4%) patients were diagnosed with gestational hypertensive disorder, 1(0.8%) of them was diagnosed with gestational hypertension, and 2 (1.6%) with severe PE. Two of them were complicated by APO. 5 (4.0%) patients were diagnosed with gestational diabetes, and these patients ended up with normal pregnancies. Anti-β2GPI-IgG in 4 patients, aβ2GPI-IgM in 4 patients and aβ2GPI-IgA in 5 patients changed from negative to positive during pregnancy. Ten patients were definitively diagnosed with APS[8] at the end of the research investigation.
The percentage of APO in patients with fetal DUAEs (40.9%) was significantly higher than that of patients with non-DUAEs (9.7%) in the non-aPL group (p=0.019). (Table 4) We divided the 126 pregnancies in the study into two groups according to different pregnancy outcomes. They were the APO group (n=32, 25.4%) and NAPO group (n=94, 74.6%). In the APO group, the early fetal loss occurred in 5 (15.6%) patients and late fetal loss in 9 (28.1%) patients (5 of which were >20 weeks). Other adverse pregnancy outcomes included 1 (3.1%) fetal growth restriction, 8 (25.0%) premature rupture of membranes, 1 (3.1%) placental abruption, and 15 (46.9%) preterm birth (<37 weeks) (3 of them were less than 34 weeks) and 11 (34.4%) low birth weight infants (<2500 g).
Compared with the NAPO group, the percentages of previous late fetal loss (≥10 weeks) and positive aβ2GPI-IgG (34.4% vs. 12.8% p=0.006; 31.3% vs. 13.8%, p=0.028, respectively) in APO group were significantly higher. The incidence of thrombocytopenia in the APO group was also significantly higher than that in the NAPO group (p=0.025). The mean dosage of Pred in the APO group during pregnancy was 10.37 ± 5.08 mg/d, which was higher than in the NAPO group (7.35±4.45mg/d) (p=0.006). The cumulative dosage of Pred was not significantly different from that in the NAPO group. Due to adverse events in the adverse pregnancy group, some patients did not complete the entire pregnancy process, so there was a significant difference between the gestational age of APO group (29.42 ± 11.16 weeks) and NAPO group (38.58 ± 0.97 weeks) (p<0.001). (Table 5)
A binary logistic regression model was established to compare the differences between the APO group and the NAPO group by multivariate analysis. The results showed that late fetal loss in the previous pregnancy (≥10 weeks) (OR=4.368, 95%CI (1.509, 12.644), p=0.007), positive aβ2GPI-IgG (OR=3.088, 95%CI (1.039, 9.173), p=0.042) and thrombocytopenia (OR=1.112, 95%CI (1.002, 1.235), p=0.046) were risk factors for patients with APO (Table 6).
In our study, one patient (1/126, 0.8%) experienced late fetal loss in the previous pregnancy and complicated with positive aβ2GPI-IgG and thrombocytopenia simultaneously. This patient was premature at gestational age of 30 weeks (because of IUGR), and the weight of newborn was only 950g. One patient (0.8%) experienced late fetal loss in the previous pregnancy and was complicated with positive aβ2GPI-IgG during this pregnancy. This patient was premature at 36 weeks. One patient (0.8%) experienced late fetal loss in the previous pregnancy and was complicated with thrombocytopenia during this pregnancy. This patient’s pregnancy outcome was late fetal loss (gestational age was 28.29 weeks) because of severe PE. Two patients were complicated with positive aβ2GPI-IgG and thrombocytopenia simultaneously. One of them was premature at 36 weeks, and the other was normal pregnancy (38 weeks).
In all 126 pregnancies, 14 were failed pregnancies, and the other 112 pregnancies were successful. In the patients with successful pregnancies, 15 (13.4%) had premature births. The percentage of late fetal loss in the previous pregnancy, positive aCL(IgG/IgM) with aβ2GPI (IgG/IgM) and positive aβ2GPI-IgG (any combination) in the patients with premature birth were significantly higher than that in patients without premature birth (40.0% vs. 13.4%, p=0.029; 20.0% vs. 2.1%, p=0.017; 40% vs. 14.4%, p=0.041, respectively). The average dosage of Pred in patients in the premature birth group was 10.37 ± 3.71mg/d, and the dosage was 7.33 ± 4.40 mg/d (p=0.003).
At the same time,11 of 112 newborns was low-weight (<2500g) births. The proportion of late fetal loss history in the patients with LBW was 45.5%, which was significantly higher than patients without LBW (13.9%) (p=0.026). The average dosage of Pred of the former was 9.99 ± 4.58 mg/d, significantly higher than the latter (7.49 ± 4.44 mg/d) (p=0.030). (Supplementary material 1)
Adding the outcome of this pregnancy, autoantibody level and maternal AEs in three trimesters to the previous history, we re-divided our patients into three groups. They were changed to UCTD group (n=47), NC-APS group (n=69) and APS group (n=10). Based on failed pregnancies history, the aPL antibodies positive ratio were much higher in APS and NC-APS group, near half of the patient in these two groups have used LDA, HCQ and Pred before pregnancy. More than 75% patient in NC-APS and UCTD group used LDA, LWMH, HCQ and Pred, and 100% APS patient prescribed these four medicines during pregnancy. While the live birth rate was still much lower in APS and NC-APS group than UCTD, that were,60%, 87% vs.97.9%, respectively. (Supplementary material 2)