Systemic lupus erythematosus (SLE) predominantly affects women of reproductive age, constituting a significant gestational risk factor. Women with SLE are at increased risk of severe pregnancy complications, such as preeclampsia (PE), preterm delivery, fetal death, and several other adverse pregnancy outcomes (APOs) (1–4). Further, pregnancy can exacerbate disease manifestations leading to SLE flares and irreversible organ damage, contributing to considerable maternal morbidity (5–7).

Thromboembolic manifestations are common during SLE pregnancy, as shown by several studies (8–10). The underlying hemostatic disturbances during SLE pregnancy are poorly studied, despite the known propensity to develop a hypercoagulable state during the gestation period, as also shown in healthy women (11). Recently, in a study on the global hemostatic potential, we demonstrated enhanced coagulation and impaired fibrinolysis in healthy women suffering from PE during the third gestational trimester (12). As patients with SLE have an increased risk for PE development, we hypothesized that similar findings could contribute to adverse pregnancy outcomes in SLE pregnancy.

Antiphospholipid antibodies (aPL) and lupus nephritis (LN) significantly contribute to fetal and maternal complications and are known risk factors for the development of PE (13, 14). In addition, positivity for aPL further enhances the risk for thromboembolic complications during pregnancy (15). In a previous study, we demonstrated an increased risk of PE specifically in women with LN compared to patients with non-renal SLE (26 vs. 3%), regardless of the presence of concurrent antenatal flares (16).

This study aimed to investigate hemostatic variables in pregnant women with SLE in relation to the use of anticoagulant prophylaxis, aPL status, and development of thromboembolic complications and PE.

In total, 34 pregnant patients with SLE followed between 2019 and 2022 at the Rheumatology Clinic, Karolinska University Hospital, were included in the study. All the patients fulfilled the American College of Rheumatology (ACR) criteria and/or the Systemic Lupus International Collaborating Clinics for SLE (17, 18). Patients older than 18 years of age with an SLE diagnosis prior to conception who were followed during pregnancy at the SLE outpatient clinic were included. Clinical and laboratory data were collected and the patients were sampled during the first (weeks 10–12, T1) and the third (weeks 32–34, T3) trimester of gestation for research purposes.

Thus, 22 of the patients had samples taken during T1, of whom 2 miscarried, 28 patients were sampled in T3, and 16 patients had samples taken at both study timepoints.

As a control population, 80 pregnant healthy controls (HCs) who had been sampled during T3 of gestation were included (12). HCs scheduled for routine prenatal care visits during T3 were included in the study. All HCs had uncomplicated pregnancies and had no history of cardiovascular disease (CVD), venous thromboembolism (VTE), or treatment with antithrombotic or anticoagulant drugs.

All the participants had given their oral and written consent to participate, and the study was performed according to the Declaration of Helsinki. The study was approved by the Swedish Ethical Review Authority (2021-02559) and the local Ethics Committee of Gynaecology and Obstetrics Clinic “Narodni Front,” Belgrade, Serbia (24/5-1).

Clinical variables were retrieved from electronic medical records. This included age at conception, disease duration, occurrence of lupus nephritis (ever), ongoing immunosuppressing therapy, and a special focus on the use of anticoagulant therapy. The occurrence of secondary antiphospholipid syndrome (sAPS) defined according to the Sapporo criteria (19) was recorded.

We also retrieved data regarding disease activity 12 months before conception where active disease was defined as having a SLE Disease Activity Index 2000 (SLEDAI-2K) score >4 (20).

The general characteristics of the patients and controls are presented in Table 1.

The median age at inclusion was 33 (range 23–42) years, and the median disease duration was 10 (range 0–26) years in the patients with SLE. The median age in the controls was 31 (26–38) years. In total, 29 patients (85%) were of Caucasian origin. None of the patients were smokers but one patient used smokeless tobacco (snuff). None of the patients developed gestational diabetes.

Furthermore, 20 of the patients were primiparous, 13 had one previous delivery, and 1 patient had two. One patient had a previous episode of hepatic vein thrombosis during a previous pregnancy and diagnosis of SLE/APS was established. Her pregnancy was medically terminated in the second trimester of pregnancy (here regarded as primiparous). Otherwise, there were no cases of PE during the previous pregnancies in the parous patients.

Three of 33 (10%) patients with available data were defined as having active disease the year before the pregnancy, of whom 1 developed PE. There was no association with active disease within 12 months of gestation and development of PE. Finally, 10/34 (29.4%) had a history of LN confirmed by a kidney biopsy in all but one case.

In this study, we investigated hemostatic parameters in pregnant women with SLE in early and late gestation in association with the use of anticoagulant therapy and the development of PE and thrombotic complications. To the best of our knowledge, this is the first study investigating hemostatic variables during SLE pregnancy. In general, hemostatic alterations toward enhanced coagulation were found in SLE pregnancy compared to healthy pregnant women. In contrast to the parameters of the OHP assay, routine analyses of fibrinogen and D-dimer were not useful for monitoring the effect of LMWH in treated patients.

Pregnancy is recognized as a major risk factor for thromboembolism, with pregnant women being up to six times more likely to develop VTE compared with the general population (25). Thromboembolic complications during pregnancy contribute significantly to maternal morbidity and treatment burden in SLE. A large meta-analysis demonstrated a risk ratio of 11 for the occurrence of thromboembolism during SLE pregnancy as compared to pregnant women without SLE (8). None of the patients included in our study developed thrombotic complications, though enhanced coagulation was found already in early pregnancy and increased over time.

A large proportion of the patients had significantly elevated OCP, OHP, and OFP levels in the third trimester compared to the healthy pregnant women. Yet, the women with SLE had comparably lower OHP during early pregnancy than the healthy controls in the third trimester, pointing toward the importance of sample timing for the interpretation of results. The levels of fibrinogen and D-dimer also increased during pregnancy, yet without differences between the patients with SLE and healthy women in the third trimester of pregnancy. Not unexpectedly, a large individual variation was noted in the markers as we have chosen to display the results of all hemostatic parameters from real-life settings during SLE pregnancies without omitting outliers.

Despite a high prophylactic LMWH dose and LDASA in risk individuals, PE occurred in two cases of which one did not achieve a significant reduction of OCP, OHP, and OFP (for detailed information on the LMWH-treated patients, see Table 2). Overall, the response to the prophylactic LMWH regimen was variable and none of the patients treated with low prophylactic doses achieved a significant anticoagulation measured by the parameters of the OHP assay. The routine coagulation parameters fibrinogen and D-dimer were not affected by LMWH prophylaxis.

The representative fibrin clot images of two patients showed denser clots composed of thinner fibers in T3 compared to T1, revealing the formation of prothrombotic clots during pregnancy in patients with SLE. The increased fibrin polymerization rate indicated faster fibrin formation and Max Abs confirmed higher fibrin density in T3 compared to the healthy controls sampled at the same gestation stage. These clots were more difficult to lyse, however, the CLT did not differ between the healthy pregnant women and SLE pregnancy in our study. This might be due to the increased TAFI activation demonstrated by increased dCLT in the patients with SLE compared to the controls. TAFI (carboxy-peptidase B) is an enzyme involved in both coagulation and fibrinolysis (26) and TAFI activation due to enhanced thrombin formation inhibits fibrinolysis, as reflected by the inverse association between dCLT and OFP in the pregnant patients with SLE. However, OFP levels in patients with SLE were elevated compared to the HCs, indicating that other factors may be involved in the overall regulation of the fibrinolytic process during SLE pregnancy, which should be an objective of a larger study.

In one patient treated with a high prophylactic dose of LMWH, SEM images revealed a fibrin clot formed of thicker fibers and larger intrinsic pores. LMWH and LDASA are known to influence fibrin structure and may lead to increased porosity of fibrin clots and enhanced fibrinolysis (27, 28). However, we could not observe a difference in OFP and CLT between patients treated prophylactically with LMWH and the rest of the patients in our study. Further, despite LDASA, fibrin structure was changed toward a prothrombotic pattern during gestation. This may further support our observation that LDASA and prophylactic LMWH are not sufficient to achieve significant anticoagulation in pregnant patients with SLE.

We found a prevalence of PE of 12.5%, thus a similar frequency as previously reported in pregnant women with SLE (4, 6), although higher frequencies (13%–35%) have also been reported (29). A history of lupus nephritis, chronic hypertension, disease activity before and at conception, and sAPS are known risk factors for PE development in SLE (14). In our cohort, the patient with sAPS with significant anticoagulation developed PE despite full dose LMWH, LDASA, and HCQ treatment. Thus, additional treatment modalities are still warranted to prevent PE development.

No previous PE had occurred in the 14 parous patients. In the two patients who developed PE in their second pregnancy, gestational hypertension was present in the late stages of the previous pregnancies, of whom one also had a previous history of LN. Neither of these two cases had been given LMWH as prophylactic therapy, and LDASA treatment was here not found to be protective for PE development. The use of HCQ, given here to most patients, was in a recent publication not confirmed to decrease the risk for PE development (30).

Prevention of PE and close monitoring of patients at risk are central in the care of pregnant patients with SLE. Furthermore, in a long-term follow-up, women with SLE experiencing hypertensive disorders of pregnancy, including PE, are at increased risk of developing CVD later in life (31). Thus, women with SLE who have developed PE should be closely monitored for hypertension and other CVD risk factors and subjected to vigorous prophylactic treatment to protect them from later vascular events (32, 33).

Major bleeding complications were seen in 5/32 (15.6%) in the total patient population, of which three were treated with LMWH, in all cases given in combination with LDASA. However, other obstetric complications could also have contributed to the hemorrhages and despite anticoagulant treatment, we found no clear signal of increased bleeding risks in our study population. A majority of patients were on LDASA which was not found to increase the rate of severe hemorrhage. However, the numbers are small, and firm conclusions cannot be drawn from the present study.

The major limitation of this study is the small sample size, as larger studies in all trimesters may increase our understanding of the effects of anticoagulation on the risk for bleeding complications, PE development, and prevention of thromboembolism. Our study was performed in a single center, where most of the patients were of Caucasian ethnicity, thus hemostatic alterations may differ between different ethnic groups. In addition, patient data were sourced from clinical records retrospectively, which may have affected data consistency and follow-up compliance. Furthermore, we did not analyze whether the altered hemostatic balance had an impact on fetal outcomes, and no comparison between conventional assays and the OHP/OCP results was performed.

Although all the patients were closely monitored by specialists in obstetrics and rheumatologists in collaboration, compliance to treatment must be considered when evaluating the effects of treatment. The method for the coagulation analysis used in the present study is not routinely available and needs to be automatized to be used in clinical practice to fully monitor the effects of anticoagulant treatment. However, by using this method, we clearly demonstrate that a prothrombotic condition is present among pregnant patients with SLE, which was not captured by conventional assays.

In summary, in this exploratory pilot study aiming at investigating hemostatic alteration in a pregnant SLE population, we found a clear procoagulant state which may have an impact on the risk for the development of PE and thromboembolism. Prophylactic treatment with LMWH was safe regarding bleeding complications but did not ameliorate the hemostatic disturbances. Monitoring of hemostatic alterations may improve personalized treatment strategies in patients at risk.