Neonatal lupus: Management and outcomes

INTRODUCTION — Neonatal lupus (NL) is an autoimmune disease that is passively transferred from the mother to the fetus. The major manifestations are cardiac and cutaneous findings. The most serious complication of NL is complete heart block (approximately 20 percent have an associated cardiomyopathy at the initial diagnosis, whereas other infants may develop it later [1-4]). In this topic review, the cardiac manifestations of NL are referred to as cardiac-NL and can include any degree of block (referred to as congenital heart block) that may or may not be accompanied by extranodal disease such as valvular abnormalities, endocardial fibroelastosis, and/or dilated cardiomyopathy. Very rarely, those manifestations may occur in the absence of any conduction defect [5].

This discussion emphasizes issues related to the treatment and potential prevention of NL. Testing for candidate antibodies is important prior to initiating therapy for a presumed case of cardiac-NL because there are cases of heart block not associated with anti-Ro/SSA (Sjögren syndrome type A antigen) or La/SSB (Sjögren syndrome type B antigen), and, thus, the management may be different. This testing is discussed separately. The specific indications for cardiac pacing in infants with congenital complete heart block are also discussed separately. (See "Neonatal lupus: Epidemiology, pathogenesis, clinical manifestations, and diagnosis", section on 'Screening and surveillance' and "Congenital third-degree (complete) atrioventricular block".)

The epidemiology, pathogenesis, clinical manifestations, diagnosis, screening, and surveillance of NL are discussed in greater detail separately. Pregnancy in women with systemic lupus erythematosus (SLE) and diagnosis and management of fetal arrhythmias are also reviewed elsewhere. (See "Neonatal lupus: Epidemiology, pathogenesis, clinical manifestations, and diagnosis" and "Pregnancy in women with systemic lupus erythematosus" and "Fetal arrhythmias".)

IN UTERO MANAGEMENT — Complete heart block, once identified, is ultimately irreversible (rare cases of transient reversal [6]) despite all therapies attempted to date, including glucocorticoids, apheresis, intravenous immune globulin (IVIG), and hydroxychloroquine [2,7-9]. Second-degree heart block may be reversible, but it also may progress to complete heart block despite therapy [8-13]. The clinical relevance of first-degree heart block is unclear since progression from first-degree block (defined as a PR interval longer than 150 msec) to more advanced heart block in untreated fetuses has not been well documented but can occur. Monitoring for heart block is discussed in detail separately. (See "Neonatal lupus: Epidemiology, pathogenesis, clinical manifestations, and diagnosis", section on 'Fetal surveillance for heart block'.)

Efficacy and side effects of fluorinated glucocorticoids — Published data are limited and discordant regarding the efficacy of fluorinated glucocorticoids in reducing mortality in cardiac-NL. Dexamethasone and betamethasone, which are not inactivated by placental 11-beta dehydrogenase, may ameliorate pleuropericardial effusions or hydrops, and there are reports of improved outcomes [8,14-16]. However, there are risks of glucocorticoid therapy to both the mother (eg, infection, hypertension, avascular necrosis, insulin resistance, and gestational diabetes) and the infant (eg, oligohydramnios, growth restriction, and the still undetermined potential effect upon neurocognitive development) [17,18]. No neurodevelopmental abnormalities were noted in 14 offspring of anti-Ro/SSA-positive (Sjögren syndrome type A antigen-positive) mothers exposed to antenatal betamethasone [19].

In a meta-analysis that included over 1000 cases (albeit there may have been overlap within registries, and most series were retrospective), fluorinated glucocorticoids did not improve fetal or neonatal survival [20]. In addition, prolonged regimens of fluorinated glucocorticoids were associated with increased risk of fetal and maternal complications. A marginal but significant downgrading of congenital heart block was seen when comparing fluorinated glucocorticoids with or without other treatments, but this improvement was not evident with the use of fluorinated glucocorticoids alone. Moreover, it seems all degrees of block were included, making interpretation difficult. The use of fluorinated glucocorticoids for each degree of heart block is discussed in the sections below. (See "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'Glucocorticoids' and 'First-degree heart block' below and 'Second-degree heart block' below and 'Third-degree heart block' below.)

First-degree heart block — Treatment of isolated first-degree block (defined in some series as a PR interval >150 msec) with glucocorticoids in utero remains controversial because of the risks of therapy, the evidence that first-degree block can revert to normal sinus rhythm (NSR) without therapy, and the inconsistent evidence that untreated first-degree block in the fetus can progress to more advanced block.

Overall, the author's approach to first-degree block is to repeat the echocardiogram within 24 hours to confirm that the PR interval is indeed prolonged (at least 150 msec). If verified, the mother is started on oral fluorinated glucocorticoid (dexamethasone 4 mg per day or betamethasone 3 mg per day), and fetal monitoring by echocardiography is performed weekly. If there is progression to complete block and no evidence of extranodal disease, then dexamethasone is discontinued. If the block remains at first degree or reverts to NSR, then dexamethasone may be continued to 26 weeks gestation and then discontinued since vulnerability is thought to decrease after that time period and further inflammatory insult may be less likely. In contrast, other experts would not treat fetal first-degree block and take an anticipatory approach. Daily home Doppler monitoring of the fetus by the mother can detect second-degree block [21] and is another potential monitoring strategy for identifying progression of first-degree block (see "Neonatal lupus: Epidemiology, pathogenesis, clinical manifestations, and diagnosis", section on 'Fetal surveillance for heart block'). In sum, the overall absence of substantial evidence regarding management is due to the rarity of disease precluding prospective randomized studies, the fact that treatment is often provided to mothers with fetuses with first-degree block, and the variability in both the technique of detection and cutoff values to define first-degree heart block among the different studies. Particularly challenging is the continued management approach if the block does revert to NSR with glucocorticoid treatment since it is unclear whether improvement supports continuing therapy for the remainder of the pregnancy.

In a Canadian study of 165 fetuses of 142 anti-Ro/La antibody-positive women referred for serial echocardiography, fetal atrioventricular (AV) prolongation or type 1 second-degree block (occurring in 15) was not treated, and none progressed [22]. Spontaneous resolution was reported in other case series [23,24]. In a series of 273 mothers participating in a home monitoring study, first-degree block was noted in four fetuses, one of whom also had endocardial fibroelastosis (EFE) and was treated with dexamethasone. None of the four fetuses developed advanced block [21].

Other small studies and case reports have shown normalization of AV conduction within one to two weeks in fetuses with first-degree heart block whose mothers were treated with dexamethasone [9,25-27]. In a report from China, four anti-Ro/SSA-exposed fetuses with a PR interval of ≥160 msec treated with dexamethasone 4.5 mg and 400 mg hydroxychloroquine showed resolution with normal electrocardiograms (ECGs) from 8 to 53 months postpartum [28].

Progression from first- to third-degree heart block was reported in two cases [23,29].

Progression of first-degree block to second-degree block was observed, despite dexamethasone, in one case in the Preventive Approach to Congenital Heart Block With Hydroxychloroquine (PATCH) study [30].

Second-degree heart block — Most second-degree block detected in utero eventually progresses to complete heart block [10,31,32]. However, it is hypothesized that incomplete block represents a critical and potentially reversible inflammatory state (more advanced than first-degree block and thus more clinically concerning) in which the AV node has not already been damaged and irreversibly fibrosed. Thus, the goal of glucocorticoids and intravenous immune globulin (IVIG) [21] in the case of second-degree block is to decrease inflammation and prevent third-degree block. However, second-degree block can revert to NSR without treatment, and unfortunately not all cases respond to treatment.

Small, uncontrolled case series document that both treated and untreated patients with second-degree block can progress, stabilize, or revert to NSR [8,9,11,22]. In a systematic review and meta-analysis of five observational studies that included 71 fetuses exposed to maternal anti-Ro, the rate of progression in those treated and not treated with fluorinated glucocorticoids (primarily dexamethasone) was 52 (95% CI 23-79) and 73 percent (95% CI 39-94); the rate of regression (including to first degree, intermittent first/second degree, or NSR) was 25 and 23 percent; and the rate of complete regression to NSR was 21 (95% CI, 6-42) and 9 percent (95% CI, 0-41), respectively [12]. Given the small number of included cases and the subsequent absence of statistical power, it was not possible to quantify objectively the strength of association. In sum, antenatal glucocorticoid therapy was not associated with a statistically significant reduced rate of progression or increased rate of regression. The interval between prenatal diagnosis and treatment was not systematically recorded, and thus it is possible (and should be prospectively evaluated by employing home Doppler) that initiation of therapy immediately after diagnosis is critical with regard to efficacy. The authors of the meta-analysis concluded that the use of fluorinated glucocorticoids once second-degree heart block is diagnosed should "not be discouraged" until more robust evidence is available.

In another study evaluating nine studies involving 727 cases that overlap with the above study, the authors concluded that fluorinated glucocorticoids did not affect progression of incomplete block [33].

IVIG is an alternative option [21]. To address whether very early detection of second-degree block by the fetal home Doppler followed by rapid treatment with dexamethasone and IVIG reverses second-degree block, a large, multicenter clinical trial (Surveillance and Treatment to Prevent Fetal Atrioventricular Block Likely to Occur Quickly [STOP BLOQ]) is underway (NCT04474223).

These considerations notwithstanding, prenatal treatment with fluorinated glucocorticoids (eg, oral dexamethasone 4 to 8 mg per day or betamethasone 3 mg per day) with or without IVIG at 1 g/kg is usually prescribed by the author for mothers of fetuses with second-degree heart block, beginning as soon after detection as is feasible. Monitoring with at least weekly echocardiography is continued. The challenge is the decision of when to stop such therapy. If the fetus improves, continuing therapy through the end of pregnancy is reasonable. Gradual tapering of dexamethasone from 8 to 2 mg is probably warranted to limit side effects. However, discontinuation with vigilant observation (ie, weekly monitoring by echocardiography) is also an option since the vulnerability of the fetal heart decreases after 26 weeks gestation. If the fetus progresses to third-degree block or does not respond, discontinuation of therapy is reasonable if there is no other indication (eg, cardiomyopathy, myocarditis, hydrops) for treatment with glucocorticoids. However, this approach is not uniformly agreed upon, as many clinicians still worry about progression. For IVIG, it is also unclear whether to continue every three weeks or discontinue after 26 weeks.

Third-degree heart block — Complete heart block is associated with fetal demise in 5 to 20 percent [1,14,31,34]. Fluorinated glucocorticoids have not been shown to reverse third-degree block. In addition, most studies have not supported the use of fluorinated glucocorticoids to prevent disease progression or death. Thus, treatment of isolated third-degree block is generally not advised by the author, unless there are other factors that might indicate ongoing inflammation and glucocorticoid treatment (eg, cardiomyopathy, EFE). Management of the fetus with complete AV block is primarily expectant. Fetal echocardiography is usually performed weekly to detect extranodal disease, which may warrant treatment. Postnatal management, including indications for pacemaker implantation, and mortality risk factors are discussed in detail separately. (See "Congenital third-degree (complete) atrioventricular block", section on 'Postnatal treatment' and "Congenital third-degree (complete) atrioventricular block", section on 'Prognosis' and 'Cardiomyopathy/endocardial fibroelastosis' below.)

In a multicenter, open-label, nonrandomized study, there was no reversal in 31 fetuses with third-degree heart block regardless of whether they were treated with glucocorticoids (n = 22) or not (n = 9) [9]. In a study of 202 cases of third-degree block reported from France, the use of fluorinated glucocorticoids was not associated with improved survival [11]. In another retrospective study that included 71 fetuses with isolated advanced heart block without extranodal disease in utero exposed to fluorinated glucocorticoids within one week of detection and 85 that were not treated, fluorinated glucocorticoids did not significantly prevent development of disease beyond the AV node, reduce mortality, or forestall/prevent pacemaker implantation [2].

In a subsequent meta-analysis covering nine studies inclusive of 747 patients with varying degrees of congenital heart block, fluorinated glucocorticoids did not influence the overall birth rate, overall survival, progression of incomplete congenital heart block, pacing, or presence of extranodal disease [33]. However, in a subgroup analysis, glucocorticoids did have a significant protective role in survival and pacing in patients with third-degree block with or without extranodal disease. The findings in third-degree block may reflect the higher rate of extranodal disease in untreated patients compared with patients exposed to fluorinated glucocorticoids. This implies that extranodal disease drives mortality, which has long been appreciated [1].

Maternal beta agonist therapy — Fetuses generally tolerate the arrhythmia well when ventricular rates are >55 beats per minute (bpm) in the absence of anomalies [35,36]. Administration of maternal beta agonist therapy when the fetal heart rate is <50 to 55 bpm was shown to increase the heart rate and stroke volume in small case series [15,37]. Most centers use this approach routinely if the fetal heart rate is <50 bpm, although it has not been evaluated by comparative studies.

In a retrospective series of 44 fetuses with third-degree block all exposed to maternal anti-Ro/SSA antibodies, prognosis was evaluated based upon stratification into a high heart rate group (n = 18), for which all prenatal ventricular rate measurements were above the age-specific mean of the total cases, and a low heart rate group (n = 26), defined as having at least one observation below the mean during follow-up [38]. Overall, the average heart rate decreased significantly during gestation, from 65 bpm at 20 weeks to 55 bpm at 38 weeks. Significantly worse perinatal outcomes, defined by in utero or perinatal death, neonatal heart rate <50 bpm, or hemodynamic failure requiring emergency pacing, occurred in the low-rate group compared with the high-rate group (85 versus 6 percent). One-third of the high-rate cases were nonpaced at >6 months.

Indications for early delivery — Early delivery should be avoided unless there is evidence of poor overall health, intrauterine growth restriction, hydrops, or other indications since early delivery in the absence of these factors does not improve outcomes [2]. At the time of labor, it is possible to monitor fetal well-being using the atrial rate as detected by Doppler devices normally used in fetal monitoring. A normal atrial rate with normal heart rate variability is reassuring. It is also possible to perform serial biophysical profile assessments or fetal pH measurements if these resources are available. (See "Intrapartum fetal heart rate monitoring: Overview" and "Intrapartum category I, II, and III fetal heart rate tracings: Management" and "Biophysical profile test for antepartum fetal assessment".)

Cardiomyopathy/endocardial fibroelastosis — Fluorinated glucocorticoids (eg, dexamethasone, betamethasone) and/or intravenous immune globulin (IVIG) are often used if there is extranodal disease (eg, a more global cardiomyopathy). However, the effectiveness of these agents in the treatment of EFE is unclear [14]. In an interinstitutional study from Toronto, a retrospective review of the use of glucocorticoids and IVIG, both in utero and after birth, suggested a potential survival benefit in cases with advanced heart block associated with cardiomyopathy/EFE (rate of demise at a median follow-up of three years was 20 percent compared with a rate of demise or need for cardiac transplant of 85 percent in historical controls) [39]. A study from France reported five cases of anti-Ro-associated EFE without conduction abnormalities [5]. Two mothers were treated with betamethasone, one chose to have a therapeutic abortion, and two chose no treatment. None of the electrocardiograms at birth (n = 4) showed heart block.

POSTNATAL MANAGEMENT — Postnatal management primarily depends upon what degree of heart block, if any, was noted in utero and what the findings are on initial neonatal electrocardiogram (ECG).

Infants and children at risk for complete heart block — An ECG should be performed on all neonates born to mothers with anti-Ro/SSA (Sjögren syndrome type A antigen) and/or anti-La/SSB (Sjögren syndrome type B antigen) antibodies, even in the absence of any cardiac abnormalities detected with in utero monitoring. In addition, consultation with a pediatric cardiologist should be obtained if fetal monitoring has detected any degree of heart block in utero (including PR intervals >3 standard deviations [SD] or 150 msec, or second-degree block that reversed) and/or if the neonatal ECG is abnormal.

●First- or second-degree block identified after birth – Infants with first- or second-degree heart block identified after birth (especially those who had one of these incomplete blocks noted during in utero surveillance) are at risk of postnatal progression to a higher-degree block, including complete heart block [10,31,32]. Thus, careful observation of these infants is necessary in the postnatal period under the oversight of a pediatric cardiologist.

●Transient in utero second-degree block – Infants with second-degree block in utero that reverted to normal sinus rhythm (NSR) and that have NSR at birth should be evaluated within the first three months of life by a pediatric cardiologist with performance of an ECG and echocardiogram as they are still at risk of developing third-degree heart block [10]. The need for further monitoring can be assessed at that time, but injury in utero probably suggests that cardiac monitoring be continued indefinitely after birth.

●Transient first-degree block – Obtaining an echocardiogram and ECG at one year is a conservative approach for those with transient first-degree block, defined as PR prolongation >3 SD, in utero and a normal ECG at birth. In a study of 57 anti-Ro-exposed children, 6 (38 percent) of 16 with first-degree block in utero (using 2 SD, not 3 SD) who had NSR on birth ECG were found to again have first-degree block during preschool years [40]. Conclusions based upon these data are limited in part by the less stringent cutoff of first-degree block to a 95 percent reference range.

●Noncardiac manifestations of NL after birth – Patients who are diagnosed with NL manifest as noncardiac involvement (rash or hematologic/liver abnormalities) after birth and have no evidence of heart block of any degree at birth (by exam and ECG) are unlikely to develop cardiac disease. For those who do, referral to cardiology is indicated.

The rash of NL generally does not cause scarring and disappears within six to eight months. Appearance of NL skin lesions postnatally is independent of breastfeeding [41,42]. Thus, breastfeeding is not contraindicated in mothers with anti-Ro/SSA and/or anti-La/SSB antibodies.

●Long-term outcome of children with cardiac manifestations of NL – Cardiac dysfunction in the first year often normalizes during childhood. New-onset dysfunction, although rare, can occur de novo after the first year.

Using data from the US Research Registry for Neonatal Lupus, patients were assessed for cardiac dysfunction [3]. Echocardiogram reports were evaluated in 239 persons with cardiac-NL (143 from age 0 to 1 year, 176 from age >1 to 17 years, and 64 from age >17 years). The composite outcome for cardiac dysfunction at the time of the last available echocardiogram in each age group was defined as having at least one of the following: qualitative description of left ventricular (LV) dysfunction on echocardiogram report; concurrent cardiac medication use (inclusive of beta blockers, angiotensin-converting enzyme [ACE] inhibitors, and/or digoxin and excluding use solely for hypertension), and/or heart transplant. Cardiac dysfunction was identified in 22.4 percent at age 0 to 1 year, 14.8 percent at age >1 to 17 years, and 28.1 percent at age >17 years. Dysfunction in various age groups was significantly associated with male sex, African American ancestry, lower fetal heart rate, fetal extranodal cardiac disease, and length of time paced. In 106 children with echocardiograms at ages 0 to 1 year and >1 to 17 years, 43.8 percent with dysfunction at age 0 to 1 year were also affected at age >1 to 17 years, while the others reverted to normal. Of children without dysfunction at age 0 to 1 year, 8.9 percent developed new dysfunction between ages >1 and 17 years. Among 34 persons with echocardiograms at ages >1 to 17 years and >17 years, 6.5 percent with normal function at age >1 to 17 years developed dysfunction in adulthood.

In this same study, the authors evaluated aortic root or ascending aorta dilation based upon postnatal echocardiogram reports in 141 persons at age 0 to 1 year, 173 at age >1 to 17 years, and 64 age >17 years. Aortic dilation was present in 19/141 (13.5 percent) of the echocardiograms performed in the first year of life, 26/173 (15 percent) in the age >1 to 17 years group, and in 6/64 (9.4 percent) in those >17 years. Of the 15/105 children with echocardiogram reports for the two earliest age groupings who had aortic dilation at age 0 to 1 year, nine (60 percent) still had dilation at age >1 to 17 years, while the other six reverted to normal. In this cohort, aortic dilation in cardiac-NL did not lead to aortic dissection or progress to the point of requiring aortic surgery.

Based upon data from the Swedish National Patient Register, long-term outcomes were evaluated in 119 persons with congenital heart block, their 128 unaffected siblings, and 1190 matched healthy controls [4]. Persons with congenital heart block had a significantly increased risk of cardiovascular comorbidity, with cardiomyopathy and/or heart failure observed in 20 (16.8 percent) patients versus 3 (0.3 percent) controls, yielding a hazard ratio (HR) of 70.0 (95% CI 20.8-235.4), and with a HR for cerebral infarction of 39.9 (95% CI 4.5-357.3). Patients with congenital heart block also had a higher risk of infections. Pacemaker treatment was associated with a decreased risk of cerebral infarction but increased risks of cardiomyopathy/heart failure and infection [4].

●No in utero or neonatal evidence of NL – Late de novo development of complete heart block is rare in children of women who test positive for Ro/SSA and La/SSB autoantibodies with no in utero conduction abnormalities and a normal ECG at birth. Thus, further cardiac surveillance is probably not justified in these children.

In a study of 57 anti-Ro-exposed children, none of the children with completely normal prenatal echocardiograms developed any conduction abnormality [40]. However, persons with a prenatal history of prolonged atrioventricular (AV) time intervals had longer PR intervals on ECGs that those with normal prenatal findings. The clinical significance of these observations awaits longer follow-up. In another study, several cases of postnatal de novo complete heart block were identified in a retrospective study [43], and rare cases of isolated cardiomyopathy (one fatal) have been reported in the Research Registry for Neonatal Lupus. Although a normal pre- and/or perinatal heart rhythm was confirmed in the patients in this retrospective series, the shortcoming of this observation regarding de novo postnatal cardiac disease is the absence of frequent serial in utero echocardiographic surveillance, which may have revealed a transient or incomplete block that went clinically undetected. Thus, in the author's opinion, unless there is evidence of in utero injury, a normal ECG and echocardiogram at birth supports that continued concern for the development of cardiac disease is not warranted.

Complete heart block — Some infants with complete heart block require insertion of a cardiac pacemaker, especially if the heart rate at delivery is <55 beats per minute (bpm). Infants and young children with complete heart block who are asymptomatic usually end up requiring a pacemaker later in childhood, adolescence, or adulthood. However, exercise limitation and even death are possible in the absence of pacing. The prognosis following pacemaker implantation is excellent for most children, although development of heart failure may occur. The postnatal management and prognosis of complete heart block are discussed in greater detail separately. (See "Congenital third-degree (complete) atrioventricular block", section on 'Postnatal treatment' and "Congenital third-degree (complete) atrioventricular block", section on 'Prognosis'.)

Neonatal lupus rash — The rash associated with NL typically resolves by six to eight months of age without sequelae. Thus, treatment is not required. However, treatment with topical corticosteroids may hasten the resolution of lesions [44,45]. Sun protection is also advised, including avoidance of direct sunlight and fluorescent light. (See "Overview of cutaneous photosensitivity: Photobiology, patient evaluation, and photoprotection".)

Adrenal insufficiency — Prolonged in utero exposure to fluorinated glucocorticoids (eg, betamethasone or dexamethasone) can lead to adrenal hypoplasia and result in neonatal adrenal insufficiency [46]. This is a rare complication that can be anticipated and for which neonates can be tested. Neonatal hypotension that potentially results from adrenal insufficiency should be treated empirically with hydrocortisone in addition to standard supportive care. The diagnosis and treatment of adrenal insufficiency are discussed in greater detail separately. (See "Clinical manifestations and diagnosis of adrenal insufficiency in children" and "Treatment of adrenal insufficiency in children".)

Autoimmune and/or rheumatic disease — Children who have had NL may be at increased risk of developing an autoimmune and/or rheumatic disease, albeit this is rare. In a cohort of 49 children followed to at least the age of eight years, six patients (12 percent) developed a well-defined systemic rheumatic and/or autoimmune disease [47]. The disorders noted in these children included oligoarticular and polyarticular juvenile idiopathic arthritis, psoriasis, thyroid disease, iritis, type 1 diabetes mellitus, and nephrotic syndrome. None of the children with neonatal disease, nor any of their unaffected siblings, developed systemic lupus erythematosus (SLE) during at least eight years of follow-up.

The long-term outcomes of children exposed to maternal anti-Ro/SSA (119 with congenital heart block and their 128 siblings without congenital heart block) compared with matched healthy controls (n = 1190) and their siblings (n = 1071) were reported based upon data retrieved from the Swedish National Patient Register [4]. Both children with congenital heart block and their siblings without congenital heart block experienced a significantly higher frequency of several autoimmune diseases than their respective non-anti-Ro/SSA-exposed controls. Specifically, over a follow-up of approximately 20 years, three exposed children with congenital heart block and two of their siblings developed lupus, with no cases of lupus noted in the control groups.

PREVENTION OF NL IN SUBSEQUENT PREGNANCIES — Hydroxychloroquine is an antimalarial drug that inhibits nucleic acid ligation of endosomal Toll-like receptors (TLRs) by preventing endosomal acidification or direct binding to nucleic acids [48,49]. It is one of the drugs used to treat systemic lupus erythematosus (SLE) and is often continued during pregnancy because it is associated with minimal risk to the fetus and mother. As detailed below, data suggest it may also decrease the risk of the fetus developing cardiac-NL when there is a prior history of cardiac-NL in a sibling. The efficacy in the setting of prior cutaneous-NL is not known. (See "Safety of rheumatic disease medication use during pregnancy and lactation" and "Pregnancy in women with systemic lupus erythematosus", section on 'Selective use allowed during pregnancy'.)

The author suggests preemptive treatment with hydroxychloroquine (400 mg orally once a day) in pregnant patients with anti-Ro/SSA (Sjögren syndrome type A antigen) and/or anti-La/SSB (Sjögren syndrome type B antigen) antibodies who have previously given birth to a child with cardiac-NL, regardless of maternal health status. Hydroxychloroquine is initiated between 6 and 10 weeks gestation in patients who are not already on the medication to optimize effective exposure by 16 weeks of gestation. Preemptive treatment is not used if there is only a history of noncardiac-NL. Likewise, if a mother does not have SLE and therefore is not on hydroxychloroquine for maternal disease, treatment of a primigravid anti-Ro/SSA-positive mother or one who has never given birth to a child with heart block is less well supported. Preventive treatment with glucocorticoids is not recommended, even in mothers with a previously affected fetus, because the risks of these medications in the majority of expectant mothers and their fetuses who would be unnecessarily exposed outweigh the uncertain potential benefits. Intravenous immune globulin (IVIG) is also not recommended prophylaxis.

Several retrospective studies have suggested that hydroxychloroquine decreases the overall risk of cardiac-NL [50-52]. One retrospective study based upon data from NL registries in the United States, France, and the United Kingdom included pregnancies of patients who had previously given birth to a child with cardiac-NL and had anti-Ro/SSA antibodies, regardless of maternal health status (ie, the mother could have been asymptomatic at the time of pregnancy or have had an associated autoimmune disease) [51]. Two-hundred fifty-seven pregnancies (40 exposed and 217 unexposed to hydroxychloroquine) were identified. Cardiac-NL developed in 3 of 40 (7.5 percent) of exposed fetuses and 46 of 217 (21.2 percent) of unexposed fetuses, with 10 fatalities in the unexposed group. Another retrospective, single-center study examined 268 pregnancies, 73 of which had exposure to hydroxychloroquine throughout the pregnancy [52]. Ninety-nine offspring developed cardiac-NL, with a nonsignificant trend toward a protective effect of hydroxychloroquine on cardiac-NL.

The efficacy of hydroxychloroquine in preventing the recurrence of congenital heart block was reported in a prospective open-label, single-arm, two-stage clinical trial designed using Simon's optimal design [30]. Anti-Ro/SSA-positive mothers with a previous pregnancy complicated by congenital heart block were recruited (n = 19 first stage; n = 35 second stage). Patients received 400 mg daily of hydroxychloroquine, starting prior to completion of gestational week 10 and maintained through pregnancy. The primary outcome was second-degree or third-degree congenital heart block any time during pregnancy, and secondary outcomes included isolated endocardial fibroelastosis, first-degree congenital heart block at birth, and/or skin rash. By intention-to-treat (ITT) analysis, 4/54 evaluable pregnancies resulted in a primary outcome (7.4 percent). Since nine mothers took potentially confounding medications (fluorinated glucocorticoids and/or intravenous gamma globulin) after enrollment but prior to a primary outcome, to evaluate hydroxychloroquine alone, nine additional mothers were recruited and followed the identical protocol. In the per-protocol analysis restricted to pregnancies exposed to hydroxychloroquine alone, 4/54 (7.4 percent) fetuses developed a primary outcome as in the ITT. Secondary outcomes included mild endocardial fibroelastosis (n = 1) and cutaneous-NL (n = 4). Overall, 5/63 fetuses developed a primary outcome. These prospective data support that hydroxychloroquine significantly reduces the recurrence of congenital heart block below the historical rate by more than 50 percent, suggesting that this drug be prescribed for secondary prevention of fetal cardiac disease in anti-Ro/SSA-exposed pregnancies.

In two multicenter, prospective, observational studies, IVIG at replacement doses (400 mg/kg) given every three weeks from weeks 12 to 24 was not effective in preventing congenital heart block [53,54]. It is unknown whether higher antiinflammatory (1 to 2 g/kg) doses would be efficacious.

Treatment of the mother with hydroxychloroquine may have a protective effect on development of NL rash. In a multicenter, case-control study that included 122 children with cutaneous-NL and 434 controls born to women with a rheumatologic disease who had documentation of maternal anti-Ro±anti-La antibodies at pregnancy and confirmation of medication use and the child's outcome, exposure to hydroxychloroquine was associated with a reduced risk of cutaneous disease (16 versus 34 percent) [55].

SUMMARY AND RECOMMENDATIONS

●Neonatal lupus (NL) is a passively acquired autoimmune disease that occurs in offspring of mothers with anti-Ro/SSA (Sjögren syndrome type A antigen) and/or anti-La/SSB (Sjögren syndrome type B antigen) antibodies. The most serious complication of NL is complete heart block. (See 'Introduction' above.)

●Prenatal treatment of first-degree block is perhaps the most controversial because of the risks of therapy and absence of consistent evidence regarding benefit. We would first confirm within 24 hours that the PR interval is indeed prolonged (>150 msec). If verified, we suggest starting the mother on a fluorinated glucocorticoid (oral dexamethasone 4 mg per day or betamethasone 3 mg per day) (Grade 2C). The cutoff for consideration is not universally agreed upon. In the Surveillance and Treatment to Prevent Fetal Atrioventricular Block Likely to Occur Quickly (STOP BLOQ) trial, the protocol calls for treatment only if the PR interval is ≥170 msec. The glucocorticoid is discontinued if there is progression to complete block and no evidence of extranodal disease. If the block remains at first degree or reverts to normal sinus rhythm (NSR), then the glucocorticoid is continued to 26 weeks gestation and then stopped since vulnerability decreases after that time period and further inflammatory insult is less likely. Observation rather than treatment is a reasonable alternative. (See 'First-degree heart block' above.)

●We suggest prenatal treatment with fluorinated glucocorticoids (oral dexamethasone 4 to 8 mg per day or betamethasone equivalent) for mothers of fetuses with second-degree heart block, beginning as soon after detection as is feasible (Grade 2C). Therapy is tapered, but dexamethasone 2 mg is usually continued through the end of pregnancy even if the heart block regresses to NSR, although discontinuation after 26 weeks (less vulnerable period) is also reasonable. Glucocorticoids are typically discontinued if the fetus progresses to third-degree heart block or does not improve if there is no other indication for treatment with glucocorticoids (eg, cardiomyopathy, myocarditis), although continuing therapy through the end of pregnancy is also reasonable. Intravenous immune globulin (IVIG) at 1 mg/kg is also an option for second-degree block. (See 'Second-degree heart block' above.)

●Reversal of prenatal third-degree block with glucocorticoid therapy has not been documented, and most studies suggest that treatment with glucocorticoids does not improve survival or prevent the development of extranodal disease. Thus, we suggest against treating in utero third-degree block without any signs of myocarditis or cardiomyopathy with glucocorticoids (Grade 2C). Most of these patients will require cardiac pacing after birth. (See 'Third-degree heart block' above and "Congenital third-degree (complete) atrioventricular block".)

●An electrocardiogram (ECG) should be performed on all neonates born to mothers with anti-Ro/SSA and/or anti-La/SSB antibodies, even in the absence of any cardiac abnormalities detected with in utero monitoring. In addition, a pediatric cardiologist should be consulted for further evaluation including echocardiograms and additional ECGs if fetal monitoring has detected any degree of heart block in utero (including PR intervals >3 standard deviations [SD] or 150 msec, or second-degree block that reversed) and/or if the neonatal ECG is abnormal. (See 'Infants and children at risk for complete heart block' above.)

●There appears to be no risk of later cardiac involvement in patients who have no evidence of heart block of any degree in utero or at birth. Second-degree block detected in utero or at birth can progress to complete heart block. Infants and young children with complete heart block usually require a pacemaker at some point in life. The prognosis following pacemaker implantation is excellent for most children, although unfortunately heart dysfunction may develop despite the complete clearance of maternal autoantibodies from the child's circulation. Thus, it is imperative for all children with advanced block to continue close follow-up with their cardiologists. The rash of NL generally does not cause scarring and disappears within six to eight months. (See 'Postnatal management' above and "Congenital third-degree (complete) atrioventricular block", section on 'Prognosis'.)

●We suggest preemptive treatment with hydroxychloroquine (400 mg orally once a day) in pregnant patients who have previously given birth to a child with cardiac manifestations of NL (cardiac-NL) and who have anti-Ro/SSA and/or anti-La/SSB antibodies, regardless of maternal health status (Grade 2B). Hydroxychloroquine should be initiated between 6 and 10 weeks gestation in patients who are not already on the medication. (See 'Prevention of NL in subsequent pregnancies' above.)