Systemic Sclerosis and Abortion: Determining Factors and Clinical Significance in a Multicenter Study

1. INTRODUCTION

Systemic Sclerosis (SSc) is a multifactorial autoimmune disease; its pathological hallmark is characterized by progressive fibrosis and vasculopathy. The overall global incidence is 1:10,000. The incidence in women is 4- 9 times higher than in men [1]. This female predominance is even higher during the reproductive period (15–50 years), where females have a 15-fold increase in the incidence of the disease compared to males [2]. Vascular abnormalities are a key driver of disease progression [3]. Whereas fertility may not be reduced in women with SSc [4], pregnancy increases the risk of organ involvement, system dysfunction, and carries risks to the fetus [5].

Diffuse subtype, anti-topoisomerase 1 or anti-RNA polymerase 3 antibodies carry a greater obstetric risk [6]. Complications include hypertension, renal failure, pulmonary hypertension (PAH), interstitial lung disease (ILD), and heart failure, which are usually seen starting from the second trimester of pregnancy [7]. Women with preclinical SSc have higher frequencies of adverse pregnancy outcomes preceding the diagnosis [8, 9]. Similar patterns have been shown in women with other rheumatic diseases (RDs) [10].

The risk of abortion in SSc is controversial, while previous reports found a higher risk of abortion in SSc patients compared to the general population [8, 11, 12]; others found the risk comparable to the general population. In a meta-analysis published in 2020, pregnancies in women with SSc had a higher risk of abortions compared with healthy controls (OR 1.6, 95% CI 1.22–2.22) [13]. Contrarily, other studies suggest that the risk of abortion in SSc women can be controlled when closely monitored [14-16], except for patients with severe diffuse SSc who may experience a higher risk for miscarriage [15]. Severe internal organ involvement (PAH, active Scleroderma Renal Crisis (SRC), severe ILD) represents a near-fatal risk and is the most determining factor for poor outcome. Patients developing SRC necessitated an elective abortion at 19 weeks; similarly, PAH represented a high risk to the mother and the fetus, leading in most cases to elective medical termination [17].

Placental vasculopathy has been observed in SSc, yet the significance of this pathology in inducing abortion in SSc is still unknown [4]. Histopathological analysis of placental biopsies revealed marked decidual vasculopathy, which, when correlated with fetal outcomes, was significantly associated with fetal demise between gestational weeks 16 and 30 [18]. Additional findings included chorioamnionitis and accelerated placental maturation; these histological patterns parallel those observed in pregnancies complicated by pregnancy-induced hypertension. Notably, placental abnormalities may manifest in SSc pregnancies even in the absence of clinical perinatal complications. However, the severity of placental vascular pathology is directly proportional to the risk of spontaneous abortion, Intrauterine Growth Restriction (IUGR), and preterm delivery [19].

A well-timed pregnancy with careful obstetric monitoring will increase the probability of a successful outcome [15]. It has been advised that pregnancy should be delayed in patients with early disease (<4 years), as patients in early disease showed higher rates of miscarriages [20].

Although previous studies have described pregnancy outcomes in women with systemic sclerosis, data evaluating the clinical, serological, and metabolic factors associated with a lifetime history of spontaneous abortion and recurrent spontaneous abortion in large multicenter cohorts remain limited. Therefore, this study aimed to identify the demographic, clinical, laboratory, serological, and treatment-related factors associated with spontaneous abortion in women with systemic sclerosis and to explore potential predictors of recurrent pregnancy loss.

2. MATERIALS AND METHODS

3. RESULTS

3.3. Main Results

Patients with a history of spontaneous abortion had a significantly longer disease duration than those without spontaneous abortion (6.4±4.7 vs. 4.6±4.2 years, p=0.007). They also had a significantly higher frequency of comorbidities (29.7% vs. 15.4%, p=0.03), arthralgia (54.9% vs. 34.1%, p=0.007), and telangiectasia (28.6% vs. 12.1%, p=0.009). Conversely, Raynaud's phenomenon was significantly less frequent among patients with spontaneous abortion than among controls (68.1% vs. 84.6%, p=0.014). (Table 1).

Regarding laboratory investigations, patients with spontaneous abortion had significantly higher serum creatinine levels than those without spontaneous abortion (0.94±0.52 vs. 0.75±0.27 mg/dL, p=0.004). (Table 2).

With respect to treatment, mycophenolate mofetil was prescribed significantly more frequently in patients with spontaneous abortion than in controls (29.7% vs. 15.4%, p=0.03), whereas methotrexate and calcium channel blockers were used more frequently in patients without spontaneous abortion (47.3% vs. 29.7%, p=0.02 and 29.7% vs. 23.1%, p=0.008, respectively (Table 2).

Among patients with a history of spontaneous abortion, those with recurrent spontaneous abortion had significantly higher frequencies of Raynaud's phenomenon (93.3% vs. 55.7%, p<0.0001), hand puffiness (43.3% vs. 21.3%, p=0.047), hypo-/hyperpigmentation (46.7% vs. 23.0%, p=0.03), cardiovascular involvement (40.0% vs. 19.7%, p=0.046), anti-Scl-70 positivity (53.3% vs. 26.2%, p=0.01), and double antibody positivity (33.3% vs. 9.8%, p=0.017). They also had significantly higher serum triglyceride levels (188.7±86.4 vs. 101.6±45.6 mg/dL, p=0.006) and were more frequently receiving calcium channel blockers (66.7% vs. 31.1%, p=0.003), sildenafil (33.3% vs. 13.1%, p=0.048), azathioprine (43.3% vs. 21.3%, p=0.047), leflunomide (33.3% vs. 6.6%, p=0.002), and cyclophosphamide (36.7% vs. 9.8%, p=0.004) (Table 3).

Table 3.

Parameters mean±SD, n (%)	Female SSc patients with abortion (n=91)
	with (n=91)	Once (n=61)	Recurrent (n=30)	p
Clinical manifestations
Raynauds	62 (68.1)	34 (55.7)	28 (93.3)	<0.0001
Hand puffiness	26 (28.6)	13 (21.3)	13 (43.3)	0.047
Hypo/hyperpigmentation	28 (30.8)	14 (23)	14 (46.7)	0.03
Cardiovascular	24 (26.4)	12 (19.7)	12 (40)	0.046
Investigations
Triglycerides	131.4±74.3	101.6±45.6	188.7±86.4	0.006
Anti-Scl-70	32 (35.2)	16 (26.2)	16 (53.3)	0.01
Double-positivity	16 (17.6)	6 (9.8)	10 (33.3)	0.017
Medications
CCB	21 (23.1)	19 (31.1)	20 (66.7)	0.003
Sildinafil	18 (19.8)	8 (13.1)	10 (33.3)	0.048
AZA	26 (28.6)	13 (21.3)	13 (43.3)	0.047
LFN	14 (15.4)	4 (6.6)	10 (33.3)	0.002
CYC	17 (18.7)	6 (9.8)	11 (36.7)	0.004

Abbreviations: SSc: systemic sclerosis, anti-Scl-70: anti-scleroderma-70, CCB: calcium channel blocker, AZA: azathioprine, LFN: leflunomide, CYC: cyclophosphamide. Bold values are significant at p<0.05.

Receiver Operating Characteristic (ROC) curve analysis demonstrated that a disease duration greater than 3.5 years predicted spontaneous abortion with an Area Under the Curve (AUC) of 0.662 (95% CI, 0.58–0.75), a sensitivity of 72.8%, and a specificity of 54.5% (p<0.0001). Similarly, a serum creatinine level greater than 0.8 mg/dL predicted spontaneous abortion with an AUC of 0.63 (95% CI, 0.54–0.72), a sensitivity of 60.5%, and a specificity of 67.5% (p=0.005) (Figure 1).

For recurrent spontaneous abortion, a serum triglyceride level greater than 80 mg/dL predicted recurrence with an AUC of 0.784 (95% CI, 0.61–0.96), a sensitivity of 83.3%, and a specificity of 34.8% (p=0.006) (Figure 2).

4. DISCUSSION

Like many autoimmune diseases, SSc occurs more frequently in women and is relatively common during childbearing age. Women with SSc may have successful pregnancies, although there is an increased occurrence of adverse pregnancy outcomes including abortion, preterm birth, small for gestational age, and low birth weight [10]. The need for modern studies to address whether the risk of adverse outcomes has decreased over time is warranted given improved rheumatologic and obstetric management [23]. To date, literature data on pregnancy outcomes in women with SSc are lacking [24].

The cause of spontaneous abortion in SSc is not fully understood. Pathogenesis is thought to involve autoimmunity, fibrosis, collagen production, and vasculopathy. SSc involves various internal organs, and the placenta may be affected with decidual vasculopathy that may induce ischemia associated with a poor pregnancy outcome [25]. The rate of abortion is higher when pregnancy occurs after the onset of the disease compared to those prior to onset of SSc (15% vs 11%) [20]. Moreover, the overall rate of miscarriage in SSc (12-15% of pregnancies) is higher than in the general population (11% of pregnancies) [26]. In a study from Spain, abortion was higher in SLE patients (22%), followed by SSc (18%), primary Sjögrens disease (15%) and undifferentiated connective tissue disease (UCTD) (15%) [27]. In Portugal, 16.7% of SSc patients had miscarriage [28]. Out of 111 SSc women from the USA (248 pregnancies), 19.8% of patients had 12.9% spontaneous abortions [29].

In the current work, the disease duration was significantly longer in SSc patients with a history of abortion. In agreement, the risk of abortion was reported to increase with disease duration [15].

We found that co-morbidities including hypertension, diabetes and thyroid dysfunction were significantly more frequent in patients with abortion. Women with SSc exhibit a heightened susceptibility to pregnancy complications, such as gestational hypertension, mandating a high-risk approach to obstetrical care [15]. In a large-scale US study, the risk of fetal mortality among women with SSc demonstrated a marked improvement over the last two decades. Nevertheless, SSc deliveries continued to present a significantly higher risk for adverse pregnancy outcomes, including hypertensive disorders, compared to controls; this disparity warrants the implementation of specific clinical strategies to improve outcomes [30]. Similarly, a nationwide Swedish registry reported that while SSc pregnancies were associated with an increased risk of preeclampsia, this did not translate to higher rates of maternal death, neonatal death, or stillbirth [10].

The subtype of SSc had no influence on the frequency of abortion in our study; however, the rate of limited cutaneous systemic sclerosis (lcSSc) was significantly higher in those with abortion. However, an early study revealed that diffuse cutaneous systemic sclerosis (dcSSc) is associated with a higher risk of abortion [15]. A healthy successful pregnancy and outcome are possible for SSc women. Patients with early disease (less than 4 years' duration) and progressive diffuse skin thickening should avoid pregnancy as they are at higher risk of developing renal crisis [31].

Although the frequency of adverse pregnancy outcomes was generally higher in those with dcSSc compared to those with lcSSc (100% vs 53%), miscarriage was reported only in those with lcSSc [31]. Others reported that the subtype had no significant effects on the pregnancy outcomes after the disease onset [32]. In a nationwide study from the USA, out of 111 SSc women with complete pregnancy history, 19.8% of patients had spontaneous abortion; 45% had dSSc, while lcSSc was present in 55% [29].

A recent prospective cohort reported that although over 90% of pregnancies in women with systemic sclerosis resulted in live births, these pregnancies remained associated with increased rates of pre-eclampsia, preterm delivery, fetal growth restriction, low birthweight, and severe postpartum hemorrhage compared with the general population. Moreover, nearly 40% of women experienced disease progression, predominantly during the postpartum period. The major risk factors were dcSSc and previous cutaneous vascular involvement [33].

The frequency of Raynaud’s was significantly lower in the current patients with a history of abortion, perhaps shedding light on the importance of conducting a prospective study to monitor the increase in frequency over the following months and years. In SLE cases, abortions were not increased in patients with Raynaud’s phenomenon regardless of the presence of a positive anticardiolipin antibody [34]. The occurrence of miscarriage, stillbirth, and small gestational age infants was similarly found in patients with primary Raynaud’s phenomenon and controls [35].

In the present study, telangiectasia was significantly increased in those with a history of abortion. Telangiectasia may theoretically be a clue to complications during pregnancy such as miscarriages and abortions. However, this hypothesis was not studied in patients with SSc.

The serum creatinine level in patients with abortion was significantly higher than in those without. Scleroderma Renal Crisis (SRC) is a major vascular life-threatening complication of SSc and is associated with high morbidity and mortality, defined by an increased serum creatinine level [36]. It is stated that more than half of SSc patients have asymptomatic renal involvement such as proteinuria, elevated creatinine level, and hypertension [37].

The patients in this study without abortion were receiving a significantly higher frequency of Calcium Channel Blockers (CCBs) and methotrexate (MTX), signifying better control of the disease status, while receiving biologic therapy tended to be higher in those with abortion. Similarly, women with rheumatoid arthritis (RA) who were taking MTX had a lower rate of induced abortions than unexposed women, while the contrary was found in those who took anti-TNF as compared to those unexposed; the patients who took anti-TNF agents had an increased rate of induced abortions [38].

In the present study, the frequency of patients receiving MMF was significantly higher in SSc patients with a history of abortion. MMF is an established therapy in SSc with a potential efficacy against skin and lung fibrosis. The contribution of medications to obstetric risk could not be adequately evaluated because of the retrospective study design and the absence of detailed pregnancy-specific treatment data. Consequently, the observed associations between medication use and spontaneous abortion may reflect underlying disease severity rather than a direct effect of the medications themselves. Thus, these results may be interpreted as suggesting that these SSc patients had major organ involvement requiring MMF therapy and thus increasing the risks for abortion.

A previous study demonstrated that pregnancy outcomes change markedly according to systemic sclerosis disease activity. Women with active or early disease experienced frequent disease exacerbations during or shortly after pregnancy, higher rates of hypertensive disorders, fetal loss, and intrauterine growth restriction/low birthweight. In contrast, women with stable disease had planned pregnancies with close monitoring, remained clinically stable throughout pregnancy, and had substantially better maternal and fetal outcomes, although low birthweight remained relatively common [39].

Patients with recurrent abortion had a significantly higher frequency of Raynaud’s, hand puffiness, hypo/hyperpigmentation, cardiovascular disease, anti-Scl-70 and double positivity, as well as of receiving CCB, sildenafil, azathioprine, leflunomide and cyclophosphamide compared to those with a single abortion. The serum level of triglycerides was significantly higher than that of those with recurrent abortion. Although the ROC analysis showed low specificity (34.8%), it sheds light on the altered lipid profile in these patients and the need for early detection and management of hypertriglyceridemia. The frequency of adverse pregnancy outcomes was significantly higher in SSc women (60%) compared to SLE (37.5%), anti-phospholipid syndrome (36.6%), or to the normal subjects (10%), with a higher rate of miscarriage [24]. Accordingly, pregnant women with SSc face an elevated risk of adverse obstetrical outcomes. Specifically, rates of spontaneous abortion in this cohort exceed those observed in both the general population and patients with SLE, necessitating a high-risk, multidisciplinary management approach [24]. Conversely, evidence suggests that experiencing multiple pregnancies following an SSc diagnosis does not significantly compromise long-term renal, respiratory, or global functional outcomes. While these findings offer reassurance for patients considering pregnancy, clinicians must remain vigilant regarding potential postpartum complications to optimize the clinical care provided for these women [40].

Preconception counseling, risk stratification, and multidisciplinary management are of utmost importance in these groups of patients. Personalized approaches and shared decisions minimize overall maternal and fetal risks [41, 42].

5. LIMITATIONS

The Retrospective Design, as an observational study extracting data from an electronic database, carries an inherent risk of recall bias or missing documentation regarding past obstetric history, which may lead to an underestimation of early spontaneous abortions. In addition, the contribution of medications to obstetric risk could not be adequately evaluated because of the lack of detailed pregnancy-specific treatment data. Therefore, the observed associations between medication use and spontaneous abortion should be interpreted with caution, as they may reflect underlying disease severity rather than a direct effect of the medications.

CONCLUSION

In conclusion, abortion in SSc patients is intimately related to the presence of telangiectasia, longer disease duration, higher serum creatinine levels, the presence of co-morbidities, and the use of MMF. Many factors determine the recurrence of abortion, with a higher level of triglycerides having a strong link.

REFERENCES