3.1 Clinical characteristics of pregnant APS women with or without adverse outcomes.
A total of 143 pregnant women were included in our study and were divided into the adverse outcomes group (n= 49) and the non-adverse outcomes group (n= 94). Compared with the non-adverse outcomes group, the adverse outcomes group had a lower weight change during pregnancy (11.35±7.01 vs. 15.62±11.00, p=0.039), and the percent of pregnant women with low platelet count (≤50×109/L) was higher in the adverse outcomes group (20.4% vs. 8.4%, p =0.042). The adverse outcomes group had a lower rate of APS diagnosis before pregnancy (61.2% vs. 84.0%, p =0.004), this may be related with the early initiation of treatment (Table 1).
Regarding pregnancy complications and outcomes (Table 2), the data showed that the prevalence of hypertension during pregnancy was significantly higher in the adverse outcomes group (32.7% vs. 4.3%, p =0.001). Nevertheless, there was no significant difference in thrombosis, gestational diabetes, oligohydramnios, premature rupture of membranes, postpartum haemorrhage, and foetal loss between the two groups.
Table 1 Basic and gestational features in the adverse outcomes group and the non-adverse outcomes group.
|
Adverse outcomes
(n=49)
|
Non-adverse outcomes
(n=94)
|
P
|
Age(years)
|
|
|
0.953
|
≥35
|
18(36.7%)
|
35(37.2%)
|
|
<35
|
31(63.3%)
|
59(62.8%)
|
|
BMI (kg/m2)
|
23.39 ±5.59
|
21.53±6.15
|
0.228
|
Weight change during pregnancy(kg)
|
11.35±7.01
|
15.62±11.00
|
0.039*
|
Fertilization way, n (%)
|
|
|
0.406
|
natural conception
|
36(73.5%)
|
63(67.0%)
|
|
test-tube baby
|
13(26.5%)
|
28(29.8%)
|
|
ovulation induction treatment
|
0(0.0%)
|
3(3.2%)
|
|
Pregnancy history, n (%)
|
|
|
0.313
|
Primiparous
|
34(69.4%)
|
73(77.7%)
|
|
Multiparous
|
15(30.6%)
|
21(22.3%)
|
|
Time of diagnosis, n (%)
|
|
|
0.004*
|
Before pregnancy
|
30(61.2%)
|
79(84.0%)
|
|
During pregnancy
|
19(38.8%)
|
15(16.0%)
|
|
Platelet count, n (%)
|
|
|
0.042*
|
Platelet count ≤50 × 109/L
|
10(20.4%)
|
8(8.4%)
|
|
Platelet count >50 × 109/L
|
39(79.6%)
|
86(91.5%)
|
|
Prior spontaneous miscarriage times,
n (%)
|
|
|
0.793
|
Prior spontaneous miscarriage ≥3 times
|
8(16.3%)
|
17(18.1%)
|
|
Prior spontaneous miscarriage <3 times
|
41(83.7%)
|
77(81.9%)
|
|
Number of fetuses produced, n (%)
|
|
|
0.335
|
singleton pregnancy
|
45(91.8%)
|
90(95.7%)
|
|
gemellary pregnancy
|
4(8.2%)
|
4(4.3%)
|
|
*P < 0.05.
BMI: Body mass index. BMI during the first trimester and Platelet count≤50 × 109/L was happened once at least during the pregnancy.
Table 2 Pregnancy complications and outcomes in the adverse outcomes group and the non-adverse outcomes group.
|
Adverse outcomes
(n=49)
|
Non-adverse outcomes
(n=94)
|
P
|
|
Hypertension during pregnancy, n (%)
|
16(32.7%)
|
4(4.3%)
|
0.001*
|
|
Thrombosis, n (%)
|
2(4.1%)
|
4(4.3%)
|
0.961
|
|
Gestational diabetes, n (%)
|
14(28.6%)
|
17(34.7%)
|
0.149
|
|
Oligohydramnios, n (%)
|
6(12.2%)
|
6(6.4%)
|
0.230
|
|
Premature rupture of membranes, n (%)
|
10(20.4%)
|
18(19.1%)
|
0.761
|
|
Postpartum hemorrhage, n (%)
|
12(24.5%)
|
23(24.5%)
|
0.998
|
|
Fetal loss, n (%)
|
3(6.1%)
|
8(8.5%)
|
0.611
|
|
HELLP syndrome, n (%)
|
1(2.0%)
|
|
|
|
Preeclampsia, n (%)
|
16(32.7%)
|
|
|
|
Premature birth, n (%)
|
38(77.6%)
|
|
|
|
Delivery at ≤34 weeks , n (%)
|
13(26.5%)
|
|
|
|
Delivery at 34+1–36+6 weeks , n (%)
|
25(51.0%)
|
|
|
|
Fetal growth restriction, n (%)
|
12(24.5%)
|
|
|
|
Stillbirth, n (%)
|
2(4.1%)
|
|
|
*P < 0.05.
HELLP: hemolysis, elevated liver enzymes and low platelet count.
APOs progression was tightly related with aPLs status, further, the distribution of aPLs status was calculated (Figure 1 and Supplementary Table 1). The results indicated that the percent of pregnant women with single-positivity of aPLs was higher in the adverse outcomes group (75.5% vs. 47.0%, p=0.017), and the prevalence of anti-β2GPI positivity was as high as 36.7% in the adverse outcomes group. However, double-positivity and triple-positivity in the adverse outcomes group were lower than in the non-adverse outcomes group (10.2% vs. 22.3%; 8.2% vs. 9.6%, p =0.017). Of note, treatment with HCQ + LDA/LMWH, HCQ + Pre, HCQ + LDA/LMWH+Pred or other strategies did not effectively improve adverse outcomes during pregnancy between the two groups (p>0.05) (Figure 2 and Supplementary Table 2).
3.2 APOs-related risk factors in APS patients
Univariate logistic regression analysis showed that the low platelet count (≤50×109/L) was associated with the higher odds of APOs (OR = 2.756, 95% CI: 1.010-7.521, p = 0.048), and the time of diagnosis (diagnosis after pregnancy) and hypertension during pregnancy were the evident risk factors for APOs (OR = 3.336, 95% CI: 1.504-7.400, p = 0.003; OR = 10.909, 95% CI: 3.399-35.009, p = 0.001). Single-positivity of aPLs was associated with the higher odds of adverse outcomes (OR = 5.303, 95% CI: 1.457-19.306, p = 0.011), whereas the differences of double- and triple-positivity of aPLs were not significant between the two groups (Table 3).
Next, multivariate analysis was performed after adjusting age, the number of prior spontaneous miscarriages, weight change during pregnancy, gestational diabetes, thrombosis, oligohydramnios, and aPLs status (model Ⅰ). The data suggested that the low platelet count (≤50×109/L), hypertension during pregnancy, and single-positivity of aPLs were the risk factors of adverse outcomes, which is consistent with univariate logistic regression analysis. After adding the adjustments of the time of diagnosis and treatment during pregnancy (model Ⅱ), hypertension during pregnancy and single-positivity of aPLs remained the risk factors of adverse outcomes and the low platelet count (≤50×109/L) was no longer a risk factor. These indicated that treatment could ameliorate adverse outcomes in pregnant APS patients with thrombocytopenia.
Table 3 Univariate and multivariate logistic regression analyses of risk factors associated with APOs in APS patients.
factor
|
Univariate analysis
|
Multivariate analysis(modelⅠ)
|
Multivariate analysis(modelⅡ)
|
|
OR
|
95%CI
|
P
|
OR
|
95%CI
|
P
|
OR
|
95%CI
|
P
|
age(≥35 years)
|
1.025
|
0.500-2.099
|
0.947
|
0.848
|
0.354-2.032
|
0.712
|
1.144
|
0.427-3.065
|
0.789
|
Prior spontaneous miscarriage times
|
0.884
|
0.352-2.222
|
0.793
|
0.953
|
0.311-2.920
|
0.932
|
1.267
|
0.374-4.292
|
0.703
|
Weight change during pregnancy(kg)
|
0.987
|
1.015-1.158
|
0.016*
|
0.917
|
0.849-0.990
|
0.026*
|
1.438
|
0.298-6.934
|
0.651
|
Platelet count ≤50 × 109/L
|
2.756
|
1.010-7.521
|
0.048*
|
3.556
|
1.028-12.303
|
0.045*
|
1.827
|
0.344-9.703
|
0.479
|
Time of diagnosis
|
3.336
|
1.504-7.400
|
0.003*
|
|
|
|
0.382
|
0.126-1.161
|
0.090
|
Hypertension during pregnancy
|
10.909
|
3.399-35.009
|
0.001*
|
9.123
|
2.552-32.616
|
0.001*
|
6.508
|
1.747-24.239
|
0.005*
|
Gestational diabetes
|
1.812
|
0.804-4.082
|
0.152
|
1.419
|
0.498-4.040
|
0.513
|
1.214
|
0.394-3.742
|
0.735
|
Thrombosis
|
0.957
|
0.169-5.420
|
0.961
|
1.131
|
0.164-7.806
|
0.900
|
0.865
|
0.106-7.044
|
0.892
|
Oligohydramnios
|
2.047
|
0.623-0.489
|
0.238
|
1.612
|
0.396-6.558
|
0.505
|
1.438
|
0.298-6.934
|
0.651
|
aPLs status
|
|
|
|
|
|
|
|
|
|
Negative
|
1
|
|
0.019
|
|
|
0.055
|
|
|
0.142
|
Single-positivity
|
5.303
|
1.457-19.306
|
0.011*
|
5.496
|
1.247-24.225
|
0.024*
|
5.882
|
0.997-34.724
|
0.050*
|
Double-positivity
|
1.587
|
0.335-7.530
|
0.561
|
1.684
|
0.284-9.981
|
0.566
|
2.093
|
0.272-16.095
|
0.478
|
Triple-positivity
|
4.444
|
0.903-21.874
|
0.067
|
3.938
|
0.592-26.185
|
0.156
|
4.144
|
0.458-37.497
|
0.206
|
Treatment during pregnancy
|
|
|
|
|
|
|
|
|
|
Anticoagulation alone
|
|
|
0.555
|
|
|
|
|
|
0.824
|
HCQ + LDA/LMWH
|
1.533
|
0.547-4.30
|
0.417
|
|
|
|
0.539
|
0.291-5.006
|
0.795
|
HCQ + Pred
|
0.767
|
0.129-4.556
|
0.770
|
|
|
|
0.755
|
0.043-6.806
|
0.633
|
HCQ + LDA/LMWH+Pred
|
0.776
|
0.316-1.902
|
0.579
|
|
|
|
0.462
|
0.257-2.220
|
0.610
|
others
|
0.567
|
0.135-2.633
|
0.567
|
|
|
|
0.465
|
0.347-3.254
|
0.586
|
*P < 0.05.
aPLs: antiphospholipid antibodies; Pred, prednisone; HCQ, hydroxychloroquine; LDA: low-dose aspirin; LMWH: low molecular weight heparin.
model I: adjust for: age, prior spontaneous miscarriage times, weight change during pregnancy, platelet count ≤50 × 109/L, hypertension during pregnancy,
gestational diabetes, thrombosis, oligohydramnios, aPLs status.
modelⅡ: adjust for: age, prior spontaneous miscarriage times, weight change during pregnancy, platelet count ≤50 × 109/L, hypertension during pregnancy,
gestational diabetes, thrombosis, oligohydramnios, aPLs status, time of diagnosis, treatment during pregnancy.
3.3 Changes of platelet count in pregnant APS patients with thrombocytopenia at different trimesters
As well, we analyzed the changes of platelet count in 18 pregnant APS patients with thrombocytopenia at different trimesters, including the first trimester, the second trimester, the third trimester, and 42 days postpartum (Figure 3). The results showed that platelet count increased after delivery under the treatment (87.53±49.81 vs. 65.06±41.3 vs. 52.75±20.70 vs. 110.19±55.30; p<0.05). Next, our study indicated that the percent of premature birth was higher in pregnant APS patients with thrombocytopenia than the control group (50.0% vs. 24.0%, p =0.021) (Supplementary Table 3). Among 18 pregnant APS patients with thrombocytopenia, 10 patients suffered the APOs and some of them didn’t follow the prescribed treatment. Because of the crucial role of platelet in pregnancy, APS patients with thrombocytopenia need to follow the doctor's advice for regular treatment and timely monitoring of platelet changes.