Hydroxychloroquine (HCQ) is an old antimalarial drug, that has reached wide employment for treating systemic autoimmune rheumatic diseases. HCQ blocks the interaction of memory B-cells but not naive B-cells. A major mechanism of HCQ is preventing complement system activation. Moreover, HCQ inhibits various endolysosomal functions, including autophagy, endosomal Toll-like receptor activation, and calcium signaling. These effects impact several immune system processes, collectively reducing the production and release of pro-inflammatory cytokines (1).

HCQ, in APS, shows pleiotropic immunomodulant and protective effects through multiple pathways; HCQ can reduce the binding of the antiphospholipid antibodies (aPL) to the syncytiotrophoblast, restoring the expression of annexin A5 which has an anticoagulant effect (2) and antagonising aPL by the inhibition of trophoblast migration, invasion, and differentiation, even if partially (3). A reduction of antibody production was also demonstrated by a study showing a gradual decrease in the levels of circulating aPL in antiphospholipid syndrome (APS) during pregnancy (4, 5).

Key mechanisms include endothelial cell activation, where aPL bind to endothelial cells and trigger a pro-thrombotic state. This state is reinforced by the activation of monocytes, platelets, and neutrophils, often involving complement activation (6). Cellular activation in APS involves the binding of aPLs to specific receptors within lipid rafts, specialized microdomains in the cell membrane that trigger signal transduction (7). Gender differences have also been linked to APS outcomes, with research indicating a correlation between gender-specific immune responses and clinical manifestations in APS patients (8).

Moreover, its safety in pregnancy has been addressed in numerous studies (9–23). Studies by Parke et al., who examined SLE patients who received HCQ, revealed no association between its use and congenital abnormalities (11). In addition, they highlighted the beneficial effect of HCQ during pregnancy in alleviating skin lesions (9, 11). The 2020 American College of Rheumatology (ACR) recommends HCQ administration to all pregnant women with SLE (Systemic Erythematosus Lupus) if possible (15). Furthermore, several studies have shown that maternal use of HCQ is associated with decreased disease activity and lupus flares and reduced cardiac and cutaneous manifestations of neonatal lupus (13, 14). Other studies have presented the benefits of HCQ in lowering the risk of adverse pregnancy outcomes in patients with SLE (16–20). In recent years, larger literature surveys have also concluded that antimalarial HCQ has to be preferred over chloroquine (21). HCQ is also preferred to Azathioprine (AZA) to control disease activity during pregnancy in patients with SLE because of the association between AZA and the increased risk of childhood infections (22). HCQ is safe during breastfeeding and is present in the breast milk of women undergoing treatment since the level ingested by breastfed infants is less than 0.2 mg/kg/day, which is a small fraction of the adult therapeutic dose of 5 mg/kg/day, as the American Academy of Ophthalmology recommends (23), it unlikely causes toxic effects (24).

Given the increasingly widespread use of HCQ during pregnancy, our study aimed to observe the pregnancy outcome and the postnatal health in infants born to mothers with systemic autoimmune rheumatic diseases, treated or not with HCQ.

This retrospective observational cohort study was conducted at IRCSS-Policlinico “A. Gemelli” in Rome, from 2000 to 2020.

Women included in the study suffered from different systemic autoimmune rheumatic diseases: Systemic Erythematosus Lupus (SLE), Antiphospholipid Syndrome APS, Rheumatoid Arthritis (RA), Primary Sjogren's Syndrome (pSjS), Systemic Sclerosis (SSc), Undifferentiated Connective Tissue Disease (UCTD), Antiphospholipid Antibody positivity with non-criteria manifestations (aPL). Pregnancy and delivery data were collected from delivery room records; a telephone follow-up questionnaire was subsequently performed to describe the long-term pediatric outcome.

A total of 312 pregnancies were considered, 309 singleton and 4 twin pregnancies, which were excluded from statistical analysis in the birth weight and birth weight percentile indicators. Among all pregnancies, 16 spontaneous abortion and 3 intrauterine deaths occurred, and 2 cases were lost to follow-up. Complete data for follow-up were obtained in 182 cases. Patients were divided into two groups: women who took HCQ (400 mg/day) during pregnancy (n = 135 pregnancies) (HCQ group) and women (n = 177 pregnancies) who did not take HCQ during pregnancy (non-HCQ group).

The distribution of the different Rheumatic Diseases among enrolled patients is presented in Figures 1a,b according to the group.

All women in the study underwent clinical evaluation during pregnancy, and anamnestic, laboratory, and therapy data were collected retrospectively.

The anamnestic data included: patients' age at delivery, HCQ therapy taken during pregnancy and pre-conception, and any associated therapies [low-dose Aspirin, Low Molecular Weight Heparin (LWMH), Corticosteroids].

Delivery data were collected through the delivery room records at Policlinico “A. Gemelli” and, if patients had delivered elsewhere, by telephone questionnaire. Data included: mode of delivery (vaginal delivery or cesarean section); gestational age at the time of delivery; neonatal weight and weight percentile related to gestational age calculated according to Italian population curves (25); Apgar score at 1 and 5 min and admission to the neonatal intensive care unit.

Follow-up data included children's age (in months); weight and height with their percentiles calculated according to CDC (26) growth curves; exclusive maternal breastfeeding and its duration; regularity of vaccinations; and presence of diseases such as recurrent upper respiratory tract infections (≥4 episodes/year), gastrointestinal diseases, allergic diseases, atopic dermatitis, anaemia, cardiac diseases, visual impairment, hearing impairment, neurobehavioral developmental disorders, particularly language, and presence of congenital malformations.

Among the women in HCQ group, n = 79 cases completed the telephone questionnaire, while in non-HCQ group n = 103 cases completed the questionnaire; the remaining cases were lost due to: non-traceability by phone, language barrier, and refusal. In the following flow chart, the distribution of the excluded cases is shown.

Women in the non-HCQ group were affected by a single autoimmune disease with a higher prevalence than women in the HCQ group (70% vs. 54%, p-value of 0.0045) (Figures 2a,b; Table 1). Accordingly, women in the HCQ group had a higher prevalence of multiple associated autoimmune diseases (39% vs. 27% in the non-HCQ group, with a p-value of 0.0015 (Figures 2a,b; Table 1).

We analysed retrospectively a large sample of pregnant women complicated by the presence of systemic autoimmune rheumatic diseases.

HCQ has been gradually implemented in our Center over the years, due to improved knowledge about the safety and benefits of its use in pregnancy and lactation (10, 18, 20). As expected, therefore, in our results we confirm that the births throughout the most recent years occurred in pregnancies with wider use of HCQ.

In addition, we observed that women with multiple comorbidities had a higher prevalence of HCQ use compared to women suffering from a single autoimmune disease.

Regarding pregnancy outcome, we found a high rate of live births, with a low occurrence of fetal losses. In part, this can be explained by the fact that women with early-stage miscarriages did not come under our observation and, therefore, do not appear in this case history; another possibility is that the management of systemic autoimmune diseases in pregnancy has been improved over the years. Most obstetric outcome indicators were very similar in the two groups, such as the week of delivery, the birth weight percentile, the rate of low birth weight, and the number of preterm births.

In our opinion, the similar pregnancy outcome in the two groups supports the beneficial effect of HCQ during gestation. In fact, literature data suggest that patients with multiple medical conditions have less favourable pregnancy outcomes (29). On this topic, a recent review by D'Ippolito et al. offers a thorough exploration of APS in pregnancy, delving into both well-established and novel mechanisms by which APS leads to obstetric complications. At the core of APS pathogenicity are aPL, which disrupt normal placental function by targeting trophoblast cells. This leads to placental thrombosis, inflammation, and impaired angiogenesis, causing adverse outcomes like recurrent miscarriages, fetal growth restriction, and preeclampsia (30). Moreover, further mechanisms are emerging for the identification of additional specificities and laboratory tests in those patients defined as seronegative (31).

Considering postnatal health, data on infants' follow-up was obtained through a telephone questionnaire. At follow-up, the children suffered from a substantial number of medical conditions: allergic diseases, atopic dermatitis, and visual impairment were the most represented. Our main finding was emphasized in the follow-up analysis, showing that infants in the HCQ group were disease-free at a significantly higher rate than those in the non-HCQ group. This observation suggests a potential role for HCQ in improving pregnancy outcomes and promoting infant health. Notably, this finding aligns with existing literature, which highlights HCQ benefits in improving live birth rates, reducing preterm births, and lowering the risk of neonatal complications such as cutaneous lupus and congenital heart block (32). These effects are due to the ability of HCQ to reduce inflammation, inhibit cytokine release, and modulate the immune environment in ways that may benefit fetal development and neonatal outcomes (32). The absence of an increased incidence of childhood infections in the HCQ group is particularly intriguing. This addresses an important safety concern regarding prenatal drug exposure and supports HCQ profile as a safe therapeutic option during pregnancy. By contributing to the growing body of evidence on HCQ exposure in pregnancy, our findings provide reassurance to clinicians and patients about its use in managing conditions such as SLE, Sjogren Syndrome and APS in pregnant women.

The present study has some limitations, as the absence of a control group from a general obstetric population limits our ability to identify the underlying causes of the high rate of observed medical conditions in infants. Moreover, incomplete data on NICU admissions prevented its inclusion in the statistical analysis, despite its potential value as a key indicator of neonatal health for future exploration. Additionally, the monocentric and retrospective design, along with the substantial number of cases lost to follow-up, further constrain the study's scope. Future research with a broader, multicentric and prospective approach is essential to generate more robust, long-term data. Moreover, further studies are needed to elucidate the mechanisms by which HCQ might influence fetal immune development and protect against disease manifestations in childhood.