Abstract
Background
Obstetric antiphospholipid syndrome (OAPS) is a subtype of antiphospholipid syndrome associated with adverse pregnancy outcomes. Non-criteria antiphospholipid antibodies (aPLs) were focused in recent years, but the diagnostic and prognostic values of them in OAPS were rarely explored.
Methods
A single-center retrospective study enrolled 283 pregnant women with history of pregnancy loss at an university hospital in sourthwest China. All participants were tested for criteria and non-criteria aPLs, including anticardiolipin antibodies (aCL) IgA, anti-β2 glycoprotein I antibodies (aβ2GPI) IgA, anti-β2 glycoprotein I domain 1 (aβ2GPID1) IgG and anti-phosphatidylserine/ prothrombin (aPS/PT) IgG/IgM. The subsequent pregnancy outcomes were followed up. The prevalence of non-criteria aPLs were compared among patients who classified into OAPS, non-criteria OAPS (NOAPS) and control groups. Receiver operating characteristic (ROC) curve, area under the curve (AUC), sensitivity, specificity and odds ratios (OR) with 95% confidence interval (CI) were calculated to estimate diagnostic and prognostic values of these non-criteria aPLs for OAPS and NOAPS patients.
Results
The positive rate of aPS/PT IgM was higher in OAPS (44.4%) compared to NOAPS (18.1%) and controls (3.5%) (P < 0.001). aPS/PT IgM presented moderate diagnostic performance in OAPS (AUC = 0.72) but poor performance in NOAPS (AUC = 0.57). Parallel testing of lupus anticoagulant (LAC) and aPS/PT IgM enhanced sensitivity in diagnosing OAPS, and improved accuracy and sensitivity in diagnosing NOAPS. After adjustment, aPS/PT IgM remained a significant predictor of both overall adverse pregnancy outcomes (OR = 3.85, 95% CI: 1.10–13.40, P = 0.03) and early pregnancy loss before 10 weeks (OR = 4.87, 95% CI: 1.03–23.05, P = 0.04) in OAPS patients. However, due to the extremely low positive rates, we did not observe clinical significance of aCL IgA, aβ2GPI IgA, aß2GPID1 IgG and aPS/PT IgG.
Conclusions
aPS/PT IgM exhibits diagnostic and prognostic value for OAPS patients. For NOAPS patients, parallel testing of aPS/PT IgM and LAC may improve diagnostic performance.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12884-026-08672-7.
Keywords: Obstetric antiphospholipid syndrome, Non-criteria antiphospholipid antibodies, Anti-phosphatidylserine/prothrombin, Pregnancy complications
Introduction
Antiphospholipid syndrome (APS) is an autoimmune disease characterized by continuously positive antiphospholipid antibodies (aPLs) [1]. Based on clinical manifestations, APS is primarily classified into thrombotic APS and obstetric APS (OAPS) [2]. While its true prevalence remains unclear, OAPS is recognized as a key cause for adverse pregnancy outcomes, including recurrent pregnancy loss (RPL), preeclampsia, and placental insufficiency [3].
Currently, the classification of APS is primarily based on the Sydney criteria (2006) and the updated criteria (2023) issued by the American College of Rheumatology/ European League Against Rheumatism (ACR/EULAR) [4, 5]. The Sydney criteria require the presence of clinical manifestations (pregnancy morbidity) combined with laboratory evidence of positive aPLs, including lupus anticoagulant (LAC), anticardiolipin antibodies (aCL) IgG/IgM, and/or anti-β2 glycoprotein I antibodies (aβ2GPI) IgG/IgM [4]. The ACR/EULAR criteria introduce a scoring system, requiring at least 3 points from both clinical and laboratory domains [5]. Moreover, a subset of patients meeting the clinical but not the full laboratory criteria were categorized into non-criteria OAPS (NOAPS) [6]. These patients differ from those with OAPS in lab findings, clinical features, and pregnancy outcomes, but NOAPS is also a significant cause of adverse pregnancy outcomes [7, 8]. Although the concept of NOAPS was generated, the clinical diagnosis of OAPS and NOAPS were still difficult because of the lack of clinical diagnostic criteria at present [9]. It’s important to explore novel biomarkers to improve the diagnosis and prognostic estimation for these patients suspectiable for OAPS/NOAPS.
In addition to IgA isotypes of aCL and aβ2GPI, other non-criteria aPLs, such as anti-β2 glycoprotein I domain 1 (aβ2GPID1) and anti-phosphatidylserine/prothrombin (aPS/PT), have been identified as valuable supplementary markers for APS [10], especially due to their strong predictive value for thrombotic events [11–14]. Compared with thrombotic APS, OAPS exhibits unique clinical features, underlying mechanisms, and serological profiles [2, 15]. Although the association between non-criteria aPLs and OAPS has been confirmed [15–17], their performance in diagnosis and prognostic guidance remains controversial. This study aimed to explore the values of aCL IgA, aβ2GPI IgA, aβ2GPID1 IgG, and aPS/PT IgG/IgM in the diagnosis and prognostic evaluation of OAPS and NOAPS.
Materials and methods
Study design
This study is a retrospective cohort design and was approved by the Ethics Committee of West China Second University Hospital (No. 2024-034). This study complies with the Declaration of Helsinki. Informed consent was obtained from all participants prior to their inclusion in the study.
Study participants
From October 2021 to June 2023, 339 patients with history of pregnancy loss and suspectable for OAPS/NOAPS were enrolled from outpatients and inpatients of West China Second University Hospital. All patients underwent a comprehensive evaluation, including testing for criteria and non-criteira aPLs, parental chromosomes, prethrombotic indicators, endocrine factors and reproductive tract anatomical factors. Autoimmune antibodies were tested on at least two separate occasions, with a gap of more than 12 weeks between tests. Patients were excluded for: (a) Abnormal anatomic reproductive tracts; (b) Chromosomal abnormalities in couples and embryos; (c) Inherited thrombophilia; (d) Endocrine dysregulation; (e) History of arterial and venous thrombosis. Patients lacking criteria and non-criteria aPLs results or those without available follow-up pregnancy outcomes were further excluded.
Finally, a total of 54 patients met the Sydney criteria for OAPS [4]. The NOAPS group included 144 patients with clinical manifestations consistent with OAPS but who did not fully meet the laboratory criteria, such as intermittent aPL positivity (two positive tests performed < 12 weeks apart, thus not fulfilling the ≥ 12-week persistence requirement of the Sydney criteria) or low-titer IgG/IgM aCL and/or aβ2GPI antibodies between the 95th and 99th percentiles (below the locally established diagnostic cut-off, expressed in manufacturer-defined Units), or atypical clinical presentations (such as two consecutive or at least three non-consecutive unexplained pregnancy losses, or preterm birth after 34 weeks due to eclampsia or severe preeclampsia) accompanied by standard laboratory tests [18]. Given that proposed definitions and classification schemes for NOAPS remain controversial, the categorization of NOAPS patients in this study was derived from existing guidelines and published studies and was finalized by consensus among several reproductive immunology specialists at our center. Patients who did not meet either the clinical or laboratory criteria were classified into the control group, which ultimately included 85 individuals with a history of at least two consecutive early pregnancy losses before 10 weeks or at least one fetal loss at or beyond 10 weeks.
Data collection
Patient data were collected through the Hospital Information Management System, including general demographic characteristics (age, parity, history of adverse pregnancy outcomes and thrombosis), criteria and non-criteria aPLs profiles, approaches to subsequent pregnancy, and treatment strategies. In addition, subsequent pregnancy outcomes were continuously followed up via medical visit records and telephone until January 2025. The follow-up pregnancy outcomes primarily include healthy full-term live births, early pregnancy loss before 10 weeks, fetal loss after 10 weeks, preterm birth before 37 weeks and fetal growth restriction. For each reported pregnancy, outcome and gestational age were confirmed, whenever available, by reviewing hospital discharge summaries, ultrasound reports and delivery records provided by the patient or retrieved from our hospital information system. Pregnancy outcomes were defined as follows: healthy full-term live birth (live birth at ≥ 37 + 0 weeks of gestation), early pregnancy loss < 10 weeks (spontaneous loss of an intrauterine pregnancy before 10 weeks), fetal loss ≥ 10 weeks (intrauterine fetal death or stillbirth at or beyond 10 weeks), preterm birth (live birth between 24 and 36 + 6 weeks of gestation), and fetal growth restriction (birth weight < 10th percentile for gestational age). Patients were asked to provide reliable medical records to ensure the accuracy of follow-up outcome classification.
Laboratory studies
Criteria aPLs
LAC was tested by the dilute Russell viper venom time (dRVVT) test and activated partial thromboplastin time [19]. The ratio of dRVVT screening time/dRVVT confirming time > 1.20 was considered positive. aCL IgG/IgM and aβ2GPI IgG/IgM were detected by the chemiluminescent immunoassay (CLIA) (Kangrun, Kaeser, China). The recommended threshold values for these antibodies were set at 20 RU/ml, based on the 99th percentile of measurements in healthy individuals. All criteria aPLs were retested after an interval of at least 12 weeks.
Non-criteria aPLs
The detection of aCL IgA and aβ2GPI IgA were similar with IgG/IgM. The CLIA was used to test aβ2GPID1 IgG, with 20 CU as the cut-off value (Werfen, QUANTA Lite, Germany). aPS/PT IgG/IgM antibodies were detected by enzyme-linked immunosorbent assay (Werfen, QUANTA Lite, Germany), with suggested cut-off values of 30 Units according to the recommendation of the manufacturer.
Statistical analysis
Missing data were handled using listwise deletion, as the proportion of missing values was low and the patterns of missingness did not indicate systematic bias. All statistical analyses were performed using R version 4.3.2. Continuous variables were tested for normality and presented as mean (standard deviation) and compared using Welch’s ANOVA when normally distributed, or as median (interquartile range) and compared using the Kruskal–Wallis test when non-normally distributed. Categorical variables were analyzed using Fisher’s exact test or the χ² test as appropriate. Receiver operating characteristic (ROC) analysis was conducted to assess the diagnostic performance of non-criteria aPL assays, and comparisons of area under the curve (AUC) between different antibodies were performed using the DeLong test. Univariable and multivariable logistic regression models with heteroscedasticity-robust (Huber–White) standard errors were used to estimate odds ratio (OR) for the associations between non-criteria aPLs and adverse pregnancy outcomes. Calibration curves based on logistic regression were generated to evaluate the agreement between predicted and observed risks. Multiple comparisons were adjusted using false discovery rate (FDR) correction. The McNemar test was used to compare diagnostic performance between different testing methods.
To assess the robustness of our findings to group-size imbalance, a sensitivity analysis was conducted by randomly down-sampling the NOAPS and control groups to match the sample size of the OAPS group, followed by repeating the ROC analyses in the balanced dataset. A two-tailed p value < 0.05 was considered statistically significant.
Results
Demographic characteristics
Of the 339 initially enrolled patients, 25 were excluded due to missing criteria and non-criteria aPL test results, 18 due to missing follow-up pregnancy outcomes, and 13 due to missing data on anticoagulant treatment, pregnancy type, or previous miscarriage history. Finally, A total of 283 patients were enrolled in the study analysis, including 54 in the OAPS group, 144 in the NOAPS group, and 85 in the control group (Fig. 1). No differences were found among the three groups in terms of age [31.7 (3.5) vs. 31.3 (3.5) vs. 31.6 (4.1), P = 0.65] and types of pregnancy (χ²=1.0, P = 0.62). Although there were no statistically significant differences (P = 0.06, 0.07), both numbers of early pregnancy loss and fetal loss showed a trend towards higher values in the OAPS group compared to the NOAPS and control groups. Additionally, OAPS patients also received more anticoagulant treatment (χ²=28.9, P < 0.001). Among three groups, patients were more frequently treated with the combination of low-dose aspirin (LDA) and low molecular weight heparin (LWMH) than with either medication alone (Table 1).
Fig. 1.
Diagram of Patient Enrollment and Classification. OAPS: obstetric antiphospholipid syndrome; NOAPS: non-criteria obstetric antiphospholipid syndrome; aPL: antiphospholipid antibody; LAC: lupus anticoagulant; aCL: anti-cardiolipin antibody; aβ2GPI: anti-β2 glycoprotein I antibody; aβ2GPID1: anti-β2 glycoprotein I domain1 antibody; aPS/PT: anti-phosphatidylserine/prothrombin antibody
Table 1.
Clinical characteristics and the prevalence of criteria aPLs among study groups
| Variables | Total (n = 283) | OAPS (n = 54) | NOAPS (n = 144) | Control (n = 85) | Statistic | P |
|---|---|---|---|---|---|---|
| Age, Mean (SD) | 31.5 (3.7) | 31.7 (3.5) | 31.3 (3.5) | 31.6 (4.1) | F = 0.4 | 0.65 |
| Early pregnancy loss (< 10 weeks), M (Q₁, Q₃) | 2.0 (1.0, 2.0) | 2.0 (1.0,3.0) | 2.0 (1.0, 2.0) | 2.0 (1.0, 2.0) | χ²=5.6# | 0.06 |
| Fetal loss (≥ 10 weeks), M (Q₁, Q₃) | 0.0 (0.0, 1.0) | 1.0 (0.0, 1.0) | 0.0 (0.0, 1.0) | 0.0 (0.0, 1.0) | χ²=5.4# | 0.07 |
| Types of pregnancy, n (%) | χ²=1.0 | 0.62 | ||||
| Natural pregnancy | 203 (71.7) | 37 (68.5) | 107 (74.3) | 59 (69.4) | ||
| Assisted pregnancy | 80 (28.3) | 17 (31.5) | 37 (25.7) | 26 (30.6) | ||
| Anticoagulant treatment, n (%) | χ²=28.9 | < 0.001 | ||||
| No | 24 (8.5) | 7 (13.0) | 9 (6.3) | 8 (9.4) | ||
| LDA | 29 (10.3) | 0 (0.00) | 10 (6.9) | 19 (22.3) | ||
| LWMH | 23 (8.1) | 2 (3.7) | 11 (7.6) | 10 (11.8) | ||
| LDA + LWMH | 207 (73.1) | 45 (83.3) | 114 (79.2) | 48 (56.5) | ||
| Criteria aPLs | ||||||
| LAC, n (%) | χ²=125.3 | < 0.001* | ||||
| 0 | 159 (56.2) | 3 (5.6) | 71 (49.3) | 85 (100.0) | ||
| 1 | 124 (43.8) | 51 (94.4) | 73 (50.7) | 0 (0.0) | ||
| aCL IgG, n (%) | - | 1.00* | ||||
| 0 | 282 (99.6) | 54 (100.0) | 143 (99.3) | 85 (100.0) | ||
| 1 | 1 (0.4) | 0 (0.0) | 1 (0.7) | 0 (0.0) | ||
| aCL IgM, n (%) | χ²= 10.6 | 0.002* | ||||
| 0 | 261 (92.2) | 47 (87.0) | 129 (89.6) | 85 (100.0) | ||
| 1 | 22 (7.8) | 7 (13.0) | 15 (10.4) | 0 (0.0) | ||
| aβ2GPI IgG, n (%) | - | 1.00* | ||||
| 0 | 282 (99.6) | 54 (100.0) | 143 (99.3) | 85 (100.0) | ||
| 1 | 1 (0.4) | 0 (0.0) | 1 (0.7) | 0 (0.0) | ||
| aβ2GPI IgM, n (%) | - | 0.41* | ||||
| 0 | 279 (98.6) | 54 (100.0) | 140 (97.2) | 85 (100.0) | ||
| 1 | 4 (1.4) | 0 (0.0) | 4 (2.8) | 0 (0.0) | ||
OAPS obstetric antiphospholipid syndrome, NOAPS non-criteria obstetric antiphospholipid syndrome, aPLs anti-phospholipid antibodies, LAC lupus anticoagulant, aPS/PT anti-phosphatidylserine/prothrombin antibody, aCL anti-cardiolipin antibody, aβ2GPI anti-β2 glycoprotein I antibody, aβ2GPID1 anti-β2 glycoprotein I domain1 antibody, F Welch’s ANOVA; #: Kruskal-wallis test; χ²: Chi-square test; -: Fisher exact; *: False discovery rate correction; SD: standard deviation; M: Median; Q₁: 1st Quartile; Q₃: 3st Quartile; 0: negative; 1: positive
All incorporated patients were tested by APS criteria assays. The detection results of LAC (χ²=125.3, P < 0.001) and aCL IgM (χ²=10.6, P = 0.002) were statistically different among three groups. The positive rates of aCL IgG and aβ2GPI IgG/IgM were insufficient for meaningful statistical analysis.
Levels and prevalence of non-criteria aPLs in patients and controls
As shown in Supp. Table 1, after FDR correction, only the level of aPS/PT IgM remained significantly different among the three groups (P < 0.001). The median level of aPS/PT IgM was highest in the OAPS group [26.5 (13.9, 36.5)], followed by the NOAPS group [18.1 (11.8, 26.4)] and controls [16.4 (10.4, 22.1)]. No significant differences in aβ2GPI IgA, aCL IgA, aβ2GPID1 IgG, or aPS/PT IgG were found after correction. The prevalence of aCL IgA, aβ2GPID1 IgG and aPS/PT IgG/IgM in OAPS, NOAPS, and control groups were shown in Supp. Table 2. The positive rate of aPS/PT IgM differed significantly among the three groups (χ² = 36.4, P < 0.001), with the highest rate observed in the OAPS group (44.4%), followed by the NOAPS (18.1%) and control groups (3.5%). In contrast, the positive rates of aCL IgA, aβ2GPI D1 IgG, and aPS/PT IgG were all low across the groups and showed no statistically significant differences (P > 0.05). Notably, all patients tested negative for aβ2GPI IgA.
Table 2.
Diagnostic properties of LAC, aPS/PT IgM and their combination for OAPS
| Accuracy (95%CI) | Sensitivity (95%CI) | Specificity (95%CI) | PPV (95%CI) | NPV (95%CI) | OR (95%CI) | |
|---|---|---|---|---|---|---|
| LAC positive† | 0.98 (0.94-1.00) | 0.94 (0.88–1.00) | 1.00 (1.00–1.00) | 0.97 (0.93–1.00) | 1.00 (1.00–1.00) | 2516.14 (127.38–49703.15) |
| aPS/PT IgM positive‡ | 0.76 (0.68–0.83) | 0.44 (0.31–0.58) | 0.96 (0.93–1.00) | 0.89 (0.77–1.00) | 0.73 (0.65–0.81) | 18.93 (5.73–62.52) |
| LAC and aPS/PT IgM | 0.77 (0.69–0.84) | 0.41 (0.28–0.54) | 1.00 (1.00–1.00) | 1.00 (1.00–1.00) | 0.73 (0.65–0.81) | 118.38 (6.98–2008.98) |
| LAC or aPS/PT IgM | 0.97 (0.93–0.99) | 0.98 (0.95–1.00) | 0.96 (0.93–1.00) | 0.95 (0.89–1.00) | 0.99 (0.96–1.00) | 840.71 (120.40 -5870.60) |
OAPS obstetric antiphospholipid syndrome, LAC lupus anticoagulant, aPS/PT anti-phosphatidylserine/prothrombin antibody, PPV positive predictive value, NPV negative predictive value, OR odds ratio, CI confidence interval
† Defined as more than 1.20
‡ Defined as more than 30 units
Of all patients in the OAPS group, triple-positivity for LAC, aCL, and aPS/PT was observed in only one case (Supp. Figure 1A). Additionally, three patients were positive for both aCL and aPS/PT, and 22 patients showed dual positivity for aPS/PT and LAC. Among the aPS/PT-positive cases, the IgM isotype was predominant, with only one case testing positive for the IgG isotype (Supp. Figure 1B). In NOAPS group, two patients were positive for LAC, aCL, aβ2GPI and aPS/PT. Dual positivity was observed in 15 patients for aPS/PT and LAC, in four patients for aPS/PT and aβ2GPI, and in three patients for aPS/PT and aCL (Supp. Figure 1C). By comparison, 12 patients were double-positive for aPS/PT IgM and LAC (Supp. Figure 2D).
Diagnostic performances of non-criteria aPLs in OAPS and NOAPS patients
Among non-criteria aPLs, ROC analysis for the detection of OAPS showed that aPS/PT IgM had the highest AUC value (0.72), followed by aβ2GPI IgA (0.61) and aCL IgA (0.60) (Fig. 2A). aβ2GPID1 IgG (AUC = 0.52) and aPS/PT IgG (AUC = 0.57) demonstrated minimal diagnostic utility. The diagnostic performance of aPS/PT IgM was significantly superior to that of aβ2GPID1 IgG and aPS/PT IgG (P < 0.05), but there was no significant difference compared to aCL IgA (P = 0.08) and aβ2GPI IgA (P = 0.13). In contrast, none of the markers demonstrated adequate discriminative power for NOAPS, with all AUC values remaining below 0.60 (Fig. 2B). The cut-off and Youden index of aPS/PT IgM were shown in Fig. 2C and D. Additionally, in both the OAPS and NOAPS groups, aPS/PT IgM shows reasonable calibration, particularly in the moderate to high probability ranges (Supp. Figure 2).
Fig. 2.
Receiver operating characteristic analysis of non-criteria aPLs and LAC in distinguishing OAPS and NOAPS patients. (A) OAPS group, comparison of LAC with non-criteria aPLs (B) NOAPS group, comparison of LAC with non-criteria aPLs (C) ROC curve for aPS/PT IgM in the OAPS group with AUC, 95% CI, and cut-off value indicated (D) ROC curve for aPS/PT IgM in the NOAPS group with AUC, 95% CI, and cut-off value indicated OAPS: obstetric antiphospholipid syndrome; NOAPS: non-criteria obstetric antiphospholipid syndrome; LAC: lupus anticoagulant; aPS/PT: anti-phosphatidylserine/prothrombin antibody; aCL: anti-cardiolipin antibody; aβ2GPI: anti-β2 glycoprotein I antibody; aβ2GPID1: anti-β2 glycoprotein I domain1 antibody; AUC: area under the curve; ROC: receiver operating characteristic; CI: confidence interval
Except for aPS/PT IgM, other non-criteria aPLs have very low positive rates, so the diagnostic analysis primarily focuses on aPS/PT IgM (Tables 2 and 3). In OAPS, aPS/PT IgM positivity demonstrated high specificity (0.96) and positive predictive value (PPV) (0.89), low sensitivity (0.44) and negative predictive value (NPV) (0.73), moderate overall accuracy (0.76), and a strong association with disease (OR = 18.93, 95% CI: 5.73–62.52). Although the parallel test of LA with aPS/PT IgM showed no significant difference in overall diagnostic performance compared to the single LAC test (P = 0.07), its sensitivity was slightly higher than that of the single LAC test (0.98 vs. 0.94). For NOAPS, aPS/PT IgM alone showed high specificity (0.96) and PPV (0.90), but low sensitivity (0.18) and NPV (0.41), with an overall accuracy of 0.47 (OR: 5.27, 95% CI: 1.67–16.64). However, the parallel testing of LAC and aPS/PT IgM improved both accuracy (0.72) and sensitivity (0.60), while also maintaining a relatively high specificity (0.93), showing a significant improvement in diagnostic performance compared with LAC testing alone (P < 0.01).
Table 3.
Diagnostic properties of LAC, aPS/PT IgM and their combination for NOAPS
| Accuracy (95%CI) | Sensitivity (95%CI) | Specificity (95%CI) | PPV (95%CI) | NPV (95%CI) | OR (95%CI) | |
|---|---|---|---|---|---|---|
| LAC positive† | 0.68 (0.61–0.74) | 0.51 (0.43–0.59) | 0.96 (0.93–1.00) | 0.96 (0.92–1.00) | 0.54 (0.46–0.61) | 24.23 (7.92–74.12) |
| aPS/PT IgM positive‡ | 0.47 (0.41–0.54) | 0.18 (0.12–0.24) | 0.96 (0.93–1.00) | 0.90 (0.79–1.00) | 0.41 (0.34–0.48) | 5.27 (1.67–16.64) |
| LAC and aPS/PT IgM | 0.42 (0.36–0.49) | 0.08 (0.04–0.13) | 1.00 (1.00–1.00) | 1.00 (1.00–1.00) | 0.39 (0.33–0.46) | 16.13 (0.94–276.06) |
| LAC or aPS/PT IgM | 0.72 (0.66–0.78) | 0.60 (0.52–0.68) | 0.93 (0.87–0.98) | 0.94 (0.89–0.99) | 0.58 (0.50–0.66) | 18.61 (7.83–44.25) |
NOAPS non-criteria obstetric antiphospholipid syndrome, LAC lupus anticoagulant, aPS/PT anti-phosphatidylserine/prothrombin antibody, PPV positive predictive value, NPV negative predictive value, OR odds ratio, CI confidence interval
† Defined as more than 1.20
‡ Defined as more than 30 units
A sensitivity analysis based on down-sampled, size-balanced groups yielded similar AUCs and diagnostic indices (Supp. Figure 3; Supp. Table 3), supporting the robustness of the findings.
Predictive value of aPS/PT IgM for adverse pregnancy outcomes
A total of 23 patients in the OAPS group and 44 in the NOAPS group experienced adverse pregnancy outcomes. Most of these were early pregnancy loss occurring before 10 weeks of gestation (OAPS: 17, NOAPS: 34) (Supp. Table 4). In the OAPS group, fetal loss after 10 weeks was not observed, whereas preterm birth before 37 weeks occurred in 6 patients (11.1%). In the NOAPS group, 4 patients (2.7%) experienced fetal loss after 10 weeks, and 6 patients (4.1%) had preterm births. Additionally, no cases of fetal growth restriction were observed in either group. Therefore, the overall adverse pregnancy outcomes were defined as the total number of early pregnancy loss, fetal losses, and preterm birth. In OAPS patients, aPS/PT IgM positivity was significantly associated with both overall adverse pregnancy outcomes (OR = 3.27, 95% CI: 1.06–10.07, P = 0.04) and early pregnancy loss before 10 weeks of gestation (OR = 3.85, 95% CI: 1.11–13.41, P = 0.03). After adjustment for age, anticoagulant treatment, types of pregnancy, numbers of early pregnancy loss (< 10 weeks) and fetal loss (≥ 10 weeks), aPS/PT IgM remained a significant predictor for both overall adverse pregnancy outcomes (OR = 3.85, 95% CI: 1.10–13.40, P = 0.03) and early pregnancy loss before 10 weeks (OR = 4.87, 95% CI: 1.03–23.05, P = 0.04) (Table 4). Additionally, the ROC analysis found that aPS/PT IgM has moderate predictive efficacy for both overall adverse pregnancy outcomes (AUC = 0.67) and early pregnancy loss (AUC = 0.71) in OAPS patients (Supp. Figure 4), which is slightly higher than that of LAC (AUC = 0.60, 0.63).
Table 4.
Association of aPS/PT IgM with overall adverse pregnancy outcomes and early pregnancy loss
| Variable | Overall adverse pregnancy outcomes | Early pregnancy loss (< 10 weels) | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P | OR | 95% CI | P | |
| aPS/PT IgM | 3.85 | (1.10 ~ 13.40) | 0.03 | 4.87 | (1.03 ~ 23.05) | 0.04 |
| Age | 1.08 | (0.91 ~ 1.27) | 0.39 | 1.08 | (0.88 ~ 1.32) | 0.46 |
| Anticoagulant treatment | 0.39 | (0.08 ~ 1.91) | 0.25 | 0.22 | (0.04 ~ 1.31) | 0.10 |
| Types of pregnancy | 1.24 | (0.34 ~ 4.54) | 0.75 | 0.86 | (0.19 ~ 3.83) | 0.84 |
| Early pregnancy loss (< 10 weeks) | 1.09 | (0.61 ~ 1.97) | 0.77 | 1.20 | (0.57 ~ 2.51) | 0.64 |
| Fetal loss (≥ 10 weeks) | 1.16 | (0.44 ~ 3.06) | 0.76 | 1.65 | (0.51 ~ 5.40) | 0.41 |
aPS/PT anti-phosphatidylserine/prothrombin antibody, OR odds ratio, CI confidence interval
Nevertheless, aPS/PT IgM was not associated with the occurrence of overall adverse pregnancy outcomes (OR = 1.54, 95% CI: 0.64–3.74, P = 0.34) or early pregnancy loss (OR = 1.89, 95% CI: 0.75–4.79, P = 0.18) in the NOAPS group.
Discussion
OAPS is linked to serious adverse pregnancy outcomes, causing significant physical, emotional, and economic burdens [20]. Due to the heterogenity in clinical manifestations and laboratory indicators, a comprehensive evaluation of both clinical and serological data is recommended [21]. However, current laboratory criteria may fail to identify OAPS and NOAPS in some patients, potentially delaying diagnosis and worsening clinical outcomes [12]. Currently, non-criteria aPLs, which fall outside both current laboratory classification criteria for APS, represent a central area of investigation in translational research. Exploring the novel indicators may facilitate the appropriate management for these patients, thereby improving live birth rates [16, 17].
In our study, aCL IgA, aβ2GPI IgA, aβ2GPID1 IgG, and aPS/PT IgG showed no diagnostic or predictive value in ROC or logistic regression analyses, likely due to their low titers and limited positivity in both OAPS and NOAPS groups. Our results are different to previous studies. In a previous study with 1,404 APS patients, aβ2GPI IgA had the highest positivity rate (56.3%) among the non-criteria aPLs [22]. And in a Chinese cohort study, 30.46% and 6.62% patienst with APS were positive for aCL and aβ2GPI IgA, respectively [23]. To date, most research has concentrated on the link between aCL IgA and aβ2GPI IgA with thrombotic APS, while evidence supporting their predictive utility in APS-related pregnancy complications remains lacking [22–24]. The diagnostic and prognostic relevance of aβ2GPID1 IgG in thrombotic APS were also supported by several studies [25–27]. Retrospective analyses by Chighizola et al. first linked aβ2GPID1 to adverse pregnancy outcomes [28], and emerging evidence consistently supports its predictive value for such outcomes [29]. Studies further suggested that aβ2GPID1 is primarily associated with late-onset preeclampsia, with limited relevance to early pregnancy loss [28, 30]. Nonetheless, the potential role of aPS/PT IgG as a novel predictive marker for adverse pregnancy outcomes is still a matter of debate [17, 29]. The discrepancy between our study and previous studies may be attributed to different patients’ subsets, as well as differences in assay methods and reference value definitions. Compared to previous studies, this study focused on the specific subset of patients with obstetric manifestations, may provide more important clinical guidance for OAPS and NOAPS patients.
Recent developments in APS classification contextualize our findings. The 2023 ACR/EULAR APS classification criteria have de-emphasized IgM isotypes of aCL and aβ2GPI, while highlighting aPS/PT antibodies as part of the research agenda for improving laboratory characterization [5]. In our study, we observed that aPS/PT IgM was more prevalent in patients with OAPS compared to both NOAPS groups and the controls. These observations are broadly in line with the results reported in previous studies [16, 17, 31]. LAC is not commonly performed in many laboratories due to its complex technical requirements. In contrast, aPS/PT antibodies are considered potential parameters for diagnosing OAPS [32]. We found that aPS/PT IgM exhibited moderate diagnostic utility in identifying OAPS, and also a strong relationship between aPS/PT IgM and LAC. This suggests that aPS/PT could be considered an adjunct to the LAC test [16, 33]. Both in OAPS and NOAPS, parallel testing of LAC and aPS/PT IgM demonstrates higher diagnostic efficacy compared to combination testing. Especially for the NOAPS group, parallel testing improved the accuracy and sensitivity of the LA single-marker test.
Additionally, our findings revealed an association between aPS/PT IgM and adverse pregnancy outcomes in OAPS, particularly early pregnancy loss before 10 weeks, consistent with Žigon et al. [34]. ROC analysis further supported its predictive value, showing moderate efficacy for both overall adverse pregnancy outcomes and early pregnancy loss. Nevertheless, the association between aPS/PT and obstetric events is contradictory among previous studies. In a systematic review and meta-analysis, Radin et al. reported a strong association between pregnancy complications and the presence of aPS/PT IgG and/or IgM. However, no significant correlation was found between pregnancy morbidity and aPS/PT IgM alone [35]. This may be related to the considerable heterogeneity in the studies included in the article. Vandevelde et al. found a higher positivity rate of aPS/PT IgM in cases of late pregnancy loss compared to early pregnancy loss [17], whereas Zhang et al. reported that patients positive for aPS/PT IgM were more likely to experience early RPL [16]. Limited by sample size, this study could not determine whether aPS/PT IgM has differential predictive value for pregnancy complications at specific gestational stages, a question that remains controversial in the current literature.
Taken together, our findings found the potential role of aPS/PT IgM as a clinically relevant factor for identifying patients with OAPS. Incorporating aPS/PT IgM testing into serological assessments may aid in earlier recognition and risk stratification of OAPS patients whose LAC may be unavailable [36, 37]. Parallel testing for LAC and aPS/PT IgM may be considered for patients suspectble for NOAPS. Moreover, in OAPS patients, aPS/PT IgM should be considered as a potential risk factor for adverse pregnancy outcomes. For OAPS patients positive for aPS/PT IgM, enhanced prenatal monitoring may be benefit.
A key limitation of this study lies in its limited sample size, single-center setting and retrospective study design, potentially enhancing the risk of bias and affecting the external validity of the results. Secondly, the use of manufacturer-defined reference intervals for non-criteria antibodies, in the absence of harmonized cut-off values, may limit comparisons with data from other assay systems. Thirdly, the limited information on late-term pregnancy complications restricts the comprehensiveness of the analysis on the predictive value of non-criteria antibodies. Finally, the unequal group sizes may reduce statistical precision, though sensitivity analyses confirmed robustness. Future multicenter, prospective studies with standardized testing platforms and longitudinal antibody measurements are therefore needed to validate our findings and further clarify the clinical utility of non-criteria aPLs.
Conclusion
Detecting aPS/PT IgM could provide information on the diagnosis and risk prediction for OAPS patients. Although aPS/PT IgM demonstrated limited clinical utility in diagnosing NOAPS, the parallel testing of LAC and aPS/PT IgM may be a selective method. However, exception of aPS/PT IgM, we cannot recommend routine use of other non-criteria aPL tests in current clinical practice for obstetric patients. Further standardization of testing platforms and reference values may enhance the clinical utility of non-criteria aPLs.
Supplementary Information
Acknowledgements
Not applicable.
Abbreviations
- APS
Antiphospholipid Syndrome
- OAPS
Obstetric Antiphospholipid Syndrome
- NOAPS
Non-criteria Obstetric Antiphospholipid Syndrome
- aPLs
Antiphospholipid Antibodies
- aCL
Anticardiolipin Antibodies
- aβ2GPI
Anti-β2 glycoprotein I antibodies
- aβ2GPID1
Anti-β2 glycoprotein I domain 1 antibodies
- aPS/PT
Anti-phosphatidylserine/prothrombin antibodies
- LAC
Lupus Anticoagulant
- RPL
Recurrent Pregnancy Loss
- dRVVT
Dilute Russell Viper Venom Time
- LDA
Low-dose Aspirin
- LMWH
Low Molecular Weight Heparin
- ROC
Receiver Operating Characteristic
- AUC
Area Under the Curve
- FDR
False Discovery Rate
- PPV
Positive Predictive Value
- NPV
Negative Predictive Value
- OR
Odds Ratio
- CI
Confidence Interval
- ACR/EULAR
American College of Rheumatology / European League Against Rheumatism
Authors’ contributions
Gao Rui: Conceptualization, Writing - Review & Editing. Han Jinbiao: Validation, Investigation, Resources, Writing - Original Draft. Cai Rui: Methodology, Software, Formal analysis, Writing - Original Draft. Chen Qiqi: Investigation, Resources, Methodology. Huang Wanrong: Data curation, Visualization. Zeng Rujun: Funding acquisition. Qin Lang: Supervision.
Funding
Key R&D projects of sichuan provincial science and technology department (2024YFFK0366, 2024YFFK0076).
Data availability
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
Declarations
Ethics approval and consent to participate
This retrospective cohort study was approved by the Ethics Committee/Institutional Review Board of West China Second University Hospital (Approval No. 2024-034). And this study complies with the Declaration of Helsinki. Informed consent was obtained from all participants prior to their inclusion in the study.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Jinbiao Han and Rui Cai contributed equally to this work.
Contributor Information
Rujun Zeng, Email: 603350162@qq.com.
Rui Gao, Email: kaojui@yeah.net.
References
- 1.Tao JJ, Adurty S, D’Angelo D, DeSancho MT. Management and outcomes of women with antiphospholipid syndrome during pregnancy. J Thromb Thrombolysis. 2023;55(4):751–9. [DOI] [PubMed] [Google Scholar]
- 2.D’Ippolito S, Barbaro G, Paciullo C, Tersigni C, Scambia G, Di Simone N. Antiphospholipid syndrome in pregnancy: new and old pathogenetic mechanisms. Int J Mol Sci. 2023;24(4):3195. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Murvai VR, Galiș R, Panaitescu A, Radu CM, Ghitea TC, Trif P, et al. Antiphospholipid syndrome in pregnancy: a comprehensive literature review. BMC Pregnancy Childbirth. 2025;25(1):337. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Miyakis S, Lockshin MD, Atsumi T, Branch DW, Brey RL, Cervera R, et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J Thromb Haemost JTH. 2006;4(2):295–306. [DOI] [PubMed] [Google Scholar]
- 5.Barbhaiya M, Zuily S, Naden R, Hendry A, Manneville F, Amigo MC, et al. The 2023 ACR/EULAR antiphospholipid syndrome classification criteria. Arthritis Rheumatol Hoboken NJ. 2023;75(10):1687–702. [DOI] [PubMed] [Google Scholar]
- 6.Alijotas-Reig J, Esteve-Valverde E, Anunciación-Llunell A, Marques-Soares J, Pardos-Gea J, Miró-Mur F. Pathogenesis, diagnosis and management of obstetric antiphospholipid syndrome: A comprehensive review. J Clin Med. 2022;11(3):675. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Beça S, Borrell M, Cervera R, Figueras F, Nadal A, Espinosa G, et al. Non-Criteria obstetric antiphospholipid syndrome: myth or reality? J Clin Med. 2025;14(4):1299. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Liu H, Han J, Gao R, Hu Z, Tang Y, Qin L. Clinical features between primary obstetric antiphospholipid syndrome and Non-Criteria obstetric antiphospholipid syndrome and risk factors of adverse pregnancy outcomes: A retrospective study of 1006 Cases. Am J reprod Immunol N Y N 1989. 2024 Sept;92(3):e13931. [DOI] [PubMed]
- 9.Devreese KMJ, Bertolaccini ML, Branch DW, de Laat B, Erkan D, Favaloro EJ, et al. An update on laboratory detection and interpretation of antiphospholipid antibodies for diagnosis of antiphospholipid syndrome: guidance from the ISTH-SSC subcommittee on lupus Anticoagulant/Antiphospholipid antibodies. J Thromb Haemost JTH. 2025;23(2):731–44. [DOI] [PubMed] [Google Scholar]
- 10.Funke A, Staub HL, Monticielo OA, Balbi GGM, Danowski A, Santiago MB, et al. Non-criteria antiphospholipid antibodies: a narrative review. Rev Assoc Med Bras 1992. 2020;66(11):1595–601. [DOI] [PubMed] [Google Scholar]
- 11.Truglia S, Mancuso S, Capozzi A, Recalchi S, Riitano G, Longo A, et al. Non-criteria antiphospholipid antibodies’: bridging the gap between seropositive and seronegative antiphospholipid syndrome. Rheumatol Oxf Engl. 2022;61(2):826–33. [DOI] [PubMed] [Google Scholar]
- 12.Liu X, Zhu L, Liu H, Cai Q, Yun Z, Sun F, et al. Non-criteria antiphospholipid antibodies in antiphospholipid syndrome: diagnostic value added. Front Immunol. 2022;13:972012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Pignatelli P, Ettorre E, Menichelli D, Pani A, Violi F, Pastori D. Seronegative antiphospholipid syndrome: refining the value of ‘non-criteria’ antibodies for diagnosis and clinical management. Haematologica. 2020;105(3):562–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.de Laat B, Gehlen R, de Groot PG, Viewpoint. The value of non-criteria antiphospholipid antibodies. Rheumatol Oxf Engl. 2024;63(SI):SI64–71. [DOI] [PubMed] [Google Scholar]
- 15.Anunciación-Llunell A, Muñoz C, Roggenbuck D, Frasca S, Pardos-Gea J, Esteve-Valverde E, et al. Differences in antiphospholipid antibody profile between patients with obstetric and thrombotic antiphospholipid syndrome. Int J Mol Sci. 2022;23(21):12819. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zhang X, Liu Z, Guo F, Wang Q, Bai W, Zhao A. Antiphosphatidylserine/prothrombin antibodies (aPS/PT) and risk of obstetric anti-phospholipid syndrome. Am J Reprod Immunol. 2023 June;89(6):e13621. [DOI] [PubMed]
- 17.Vandevelde A, Gris JC, Moore GW, Musiał J, Zuily S, Wahl D, et al. Added value of antiphosphatidylserine/prothrombin antibodies in the workup of obstetric antiphospholipid syndrome: communication from the ISTH SSC subcommittee on lupus Anticoagulant/Antiphospholipid antibodies. J Thromb Haemost JTH. 2023 July;21(7):1981–94. [DOI] [PubMed]
- 18.Arachchillage DRJ, Machin SJ, Mackie IJ, Cohen H. Diagnosis and management of non-criteria obstetric antiphospholipid syndrome. Thromb Haemost. 2015;113(1):13–9. [DOI] [PubMed] [Google Scholar]
- 19.Favaloro EJ, Pasalic L. An overview of laboratory testing for antiphospholipid antibodies. Methods Mol Biol Clifton NJ. 2023;2663:253–62. [DOI] [PubMed] [Google Scholar]
- 20.Soto-Peleteiro A, Gonzalez-Echavarri C, Ruiz-Irastorza G. Obstetric antiphospholipid syndrome. Med Clin (Barc). 2024;163(Suppl 1):S14–21. [DOI] [PubMed] [Google Scholar]
- 21.Gris JC, Guillotin F, Chéa M, Bourguignon C, Nouvellon É, Bouvier S. Antiphospholipid antibodies in pregnancy: maternal and neonatal implications. Semin Thromb Hemost. 2023 June;49(4):337–47. [DOI] [PubMed]
- 22.Rodríguez-García V, Ioannou Y, Fernández-Nebro A, Isenberg DA, Giles IP. Examining the prevalence of non-criteria anti-phospholipid antibodies in patients with anti-phospholipid syndrome: a systematic review. Rheumatol Oxf Engl. 2015;54(11):2042–50. [DOI] [PubMed] [Google Scholar]
- 23.Hu C, Li S, Xie Z, You H, Jiang H, Shi Y, et al. Evaluation of the diagnostic value of Non-criteria antibodies for antiphospholipid syndrome patients in a Chinese cohort. Front Immunol. 2021;12:741369. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Mattia E, Ruffatti A, Tonello M, Meneghel L, Robecchi B, Pittoni M, et al. IgA anticardiolipin and IgA anti-β2 glycoprotein I antibody positivity determined by fluorescence enzyme immunoassay in primary antiphospholipid syndrome. Clin Chem Lab Med. 2014 Sept;52(9):1329–33. [DOI] [PubMed]
- 25.Li L, Chen J, Feng J, Zhao H, Liu X, Jiang Y. Anti-β2 glycoprotein domain 1 antibody as a diagnostic marker for antiphospholipid syndrome and a predictor of thrombosis: a systematic review and meta-analysis. Front Immunol. 2025;16:1541165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Zhou Y, Hu C, Qi W, Long Y, Huang C, Wang Q, et al. Anti-β2GPI-domain I antibody is associated with extra-criteria manifestations in a large prospective antiphospholipid syndrome cohort in China. Lupus Sci Med. 2023;10(2):e000924. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.De Craemer AS, Musial J, Devreese KMJ. Role of anti-domain 1-β2 glycoprotein I antibodies in the diagnosis and risk stratification of antiphospholipid syndrome. J Thromb Haemost JTH. 2016 Sept;14(9):1779–87. [DOI] [PubMed]
- 28.Chighizola CB, Pregnolato F, Andreoli L, Bodio C, Cesana L, Comerio C, et al. Beyond thrombosis: Anti-β2GPI domain 1 antibodies identify late pregnancy morbidity in anti-phospholipid syndrome. J Autoimmun. 2018 June;90:76–83. [DOI] [PubMed]
- 29.Moyle KA, Branch DW, Peterson LK, Guerra MM, Allshouse AA, Benson AE, et al. Association between novel antiphospholipid antibodies and adverse pregnancy outcomes. Obstet Gynecol. 2025;145(1):55–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Guo H, Zhang Y, Li A, Wang C, Yang S, Zhang Y, et al. Anti-domain 1 of beta2-glycoprotein I aids risk stratification in lupus anticoagulant-positive patients. Clin Exp Med. 2019;19(3):339–45. [DOI] [PubMed] [Google Scholar]
- 31.Zhu R, Cheng CY, Yang Y, Denas G, Pengo V. Prevalence of aPhosphatidylserine/prothrombin antibodies and association with antiphospholipid antibody profiles in patients with antiphospholipid syndrome: A systematic review and meta-analysis. Thromb Res. 2022 June;214:106–14. [DOI] [PubMed]
- 32.Zhu L, Li C, Liu N, Yang X, Jia RL, Mu R, et al. Diagnostic value of antibodies to phosphatidylserine/prothrombin complex for antiphospholipid syndrome in Chinese patients. Clin Rheumatol. 2017;36(2):401–6. [DOI] [PubMed] [Google Scholar]
- 33.Pham M, Orsolini G, Crowson C, Snyder M, Pruthi R, Moder K. Anti-phosphatidylserine prothrombin antibodies as a predictor of the lupus anticoagulant in an all-comer population. J Thromb Haemost JTH. 2022 Sept;20(9):2070–4. [DOI] [PubMed]
- 34.Žigon P, Perdan Pirkmajer K, Tomšič M, Kveder T, Božič B, Sodin Šemrl S, et al. Anti-Phosphatidylserine/Prothrombin antibodies are associated with adverse pregnancy outcomes. J Immunol Res. 2015;2015:975704. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Radin M, Foddai SG, Cecchi I, Rubini E, Schreiber K, Roccatello D, et al. Antiphosphatidylserine/Prothrombin antibodies: an update on their association with clinical manifestations of antiphospholipid syndrome. Thromb Haemost. 2020;120(4):592–8. [DOI] [PubMed] [Google Scholar]
- 36.Tripodi A, Cohen H, Devreese KMJ. Lupus anticoagulant detection in anticoagulated patients. Guidance from the scientific and standardization committee for lupus anticoagulant/antiphospholipid antibodies of the international society on thrombosis and haemostasis. J Thromb Haemost JTH. 2020 July;18(7):1569–75. [DOI] [PubMed]
- 37.Knight JS, Branch DW, Ortel TL. Antiphospholipid syndrome: advances in diagnosis, pathogenesis, and management. BMJ. 2023;380:e069717. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.