When a woman with systemic lupus erythematosus learns she is expecting, one of her first questions is often whether she should stop her medicines. For hydroxychloroquine, the most widely used lupus drug, the evidence now gives a fairly clear answer. Across four decades of research, continuing it through pregnancy is associated with fewer lupus flares, and it has not shown a meaningful pattern of harm to the developing baby. It also appears to lower the risk of a rare but serious complication, congenital heart block, in babies of mothers who carry certain antibodies. This article sets out what the studies actually show, where the evidence is strong, where it is still developing, and what questions remain open. Two lives are at stake in every pregnancy, and good decisions start with an honest look at the evidence.
What hydroxychloroquine is and why it matters in lupus
Hydroxychloroquine belongs to a family of medicines with a long history. Its ancestor is quinine, taken from the bark of the South American cinchona tree and used against malaria for centuries. Chemists later developed chloroquine and then hydroxychloroquine, a close relative with fewer side effects. It is a fitting example of how remedies found in the natural world, studied with patience, have become tools for healing.
Hydroxychloroquine is approved in the United States for treating malaria, systemic lupus erythematosus and discoid lupus, and rheumatoid arthritis. It is not an immunosuppressant in the way chemotherapy-type drugs are. It works more as an immune modulator. Laboratory studies show it accumulates in the acidic compartments inside immune cells and interferes with signalling through receptors called toll-like receptors 7 and 9. These receptors help drive the overactive immune response in lupus. Outside pregnancy, long-term observational studies link hydroxychloroquine use with fewer flares, less organ damage and longer survival in lupus. That is why most rheumatologists regard it as a cornerstone treatment.
Flare prevention: the case for continuing
Pregnancy places real demands on a woman with lupus. Active disease in pregnancy is associated with higher rates of miscarriage, preeclampsia, preterm birth and poor fetal growth. Keeping the mother's disease quiet is one of the most effective ways to protect the baby.
The most-cited human evidence comes from two sources:
- A small randomised trial from Brazil. Levy and colleagues, publishing in the journal Lupus in 2001, randomised 20 pregnant women with lupus or discoid lupus to hydroxychloroquine or placebo. The placebo group had higher disease activity scores and more flares by the end of pregnancy. No congenital abnormalities were seen in the babies of either group. The trial was very small, but it is one of the few placebo-controlled studies in this area.
- The Johns Hopkins Lupus Pregnancy Cohort. In a 2006 study in Arthritis & Rheumatism, Clowse and colleagues compared women who continued hydroxychloroquine in pregnancy, women who stopped it in the three months before or during the first trimester, and women who had never used it. Women who stopped had higher lupus activity and were more likely to need higher prednisone doses during pregnancy than those who continued. Those who continued had no more adverse pregnancy outcomes.
This is observational and modest-sized evidence, not a large definitive trial. But it is consistent and biologically plausible, and it fits with the broader non-pregnant data. One practical point is easy to overlook. Hydroxychloroquine has a very long half-life, so stopping it does not cause an immediate change. Its protective effect fades over weeks to months, which can happen to coincide with the middle of pregnancy.
Some studies have also suggested lower rates of preterm birth or preeclampsia among users. Those findings are inconsistent across cohorts, so they should be regarded as possible rather than established benefits.
Fetal safety: what the birth defect data show
Hydroxychloroquine crosses the placenta, so the question of fetal safety is reasonable. The reassuring evidence comes from several lines of research.
- French cohort data. Costedoat-Chalumeau and colleagues, writing in Arthritis & Rheumatism in 2003, compared pregnancies exposed to hydroxychloroquine with unexposed pregnancies in women with similar diseases. They found no increase in malformations and no evidence of visual, hearing or developmental problems in exposed children on follow-up.
- A meta-analysis from Toronto. Osadchy and colleagues at the Motherisk Program, publishing in the Journal of Rheumatology in 2011, pooled studies covering several hundred exposed pregnancies. They found no significant increase in congenital malformations, miscarriage or stillbirth.
The most important recent study adds nuance. In 2021, Huybrechts and colleagues at Brigham and Women's Hospital and Harvard published an analysis in The BMJ of US Medicaid records. It included about 2,000 pregnancies exposed to hydroxychloroquine in the first trimester, compared with more than three million unexposed pregnancies. After adjusting for the underlying autoimmune disease and other factors, they found a small increase in major malformations. The adjusted relative risk was about 1.26. The signal appeared confined to daily doses of 400 mg or more, with a relative risk of about 1.33. Doses below 400 mg showed no increase. The study found no consistent pattern pointing to one particular type of birth defect.
The study's authors described the increase as small. In absolute terms it amounts to a few additional malformations per hundred exposed pregnancies at higher doses, and it is not certain the drug itself caused them. Other large datasets have not found a clear signal, and the finding has not yet been consistently replicated. The fair reading is this. Hydroxychloroquine is not proven to cause birth defects, a modest dose-related signal deserves attention, and that possible risk must be weighed against the documented risks of uncontrolled lupus. It also supports a careful conversation about dosing. Current practice generally keeps the dose at or below 5 mg per kilogram of actual body weight per day.
Concern about eye toxicity in exposed children, a theoretical worry because hydroxychloroquine can affect the adult retina with long-term use, has not been borne out in follow-up studies to date.
Congenital heart block: an emerging protective role
About a third to a half of women with lupus carry antibodies called anti-Ro/SSA, and some also carry anti-La/SSB. These antibodies can cross the placenta. In a small proportion of babies they damage the developing heart's electrical conduction system, usually between about 18 and 24 weeks. The result, complete congenital heart block, is permanent and serious. Many affected children need a pacemaker, and the condition carries meaningful rates of fetal and infant death.
For an anti-Ro-positive mother with no previously affected child, the risk is roughly 2 percent. After one affected child, the recurrence risk rises to somewhere in the range of 13 to 18 percent.
The evidence for hydroxychloroquine here has built step by step:
- Laboratory work. Researchers at NYU Langone Health, led by Jill Buyon and Robert Clancy, showed in cell studies that immune complexes containing anti-Ro antibodies can stimulate toll-like receptor 7 on fetal heart macrophages, driving inflammation and scarring. Hydroxychloroquine blocks this pathway in the laboratory. This provides a plausible mechanism, but cell findings alone do not prove a benefit in people.
- Observational human data. Izmirly and colleagues, publishing in Circulation in 2012, analysed an international registry of anti-Ro-positive mothers who had already had a baby with cardiac neonatal lupus. The recurrence rate was about 7.5 percent among those taking hydroxychloroquine, compared with about 21 percent among those not taking it.
- A prospective trial. The PATCH study (Preventive Approach to Congenital Heart Block with Hydroxychloroquine), reported by Izmirly and colleagues in the Journal of the American College of Cardiology in 2020, enrolled anti-Ro-positive women with a previously affected child. Participants started hydroxychloroquine 400 mg daily by ten weeks of pregnancy. Among 54 evaluable pregnancies, heart block recurred in 4, or 7.4 percent. That is roughly half the expected historical rate of about 18 percent.
PATCH was open-label and single-arm, with no placebo group, and it was small. Its results are encouraging, and they match the earlier registry data. Still, they rest on comparison with historical rates rather than a randomised control group. Not every observational study has found a protective effect. Using hydroxychloroquine specifically to prevent congenital heart block is an off-label use that the FDA has not approved, even though the drug itself is approved for lupus. For many women with lupus the point is somewhat moot, because they would continue the drug for their own disease anyway.
What guidelines say, and the place of the informed patient
The American College of Rheumatology's 2020 Reproductive Health Guideline recommends continuing hydroxychloroquine during pregnancy in women with lupus. Where it is not already in use, the guideline also conditionally recommends starting it for all pregnant women with lupus. It suggests hydroxychloroquine for women with anti-Ro/SSA or anti-La/SSB antibodies, including those without lupus, because of its possible benefit against heart block. The European League Against Rheumatism (EULAR) has issued similar recommendations. Hydroxychloroquine is generally regarded as compatible with breastfeeding, since only small amounts pass into breast milk.
Guidelines summarise evidence, but the decision belongs to a woman and the physicians who know her. Several questions are worth bringing to that conversation:
- Is my current dose appropriate for my body weight, given the dose-related signal in the 2021 study?
- Have I been tested for anti-Ro and anti-La antibodies? If I am positive, what fetal heart monitoring is planned, and when?
- How active has my lupus been in the past six months, and what is my plan if it flares?
- Are my other medicines compatible with pregnancy? Some lupus drugs, such as mycophenolate and methotrexate, are not, and need to be changed before conception.
Ideally, planning starts before conception. Lupus that has been quiet for at least six months before pregnancy is associated with better outcomes. Taking the medicine consistently also matters, because a drug cannot help if it is not being taken. Preparing in advance is a practical way for a mother to care for her child before the child is even born.
Key takeaway: The best available evidence indicates that continuing hydroxychloroquine through a lupus pregnancy helps keep the mother's disease controlled, has not shown a consistent pattern of harm to the baby at standard doses, and may lower the risk of congenital heart block in anti-Ro-positive mothers, so it is a decision worth making deliberately with one's own physician rather than stopping out of caution alone.
