Neonatal lupus erythematosus (NLE) is a syndrome first described in 1954 by McCuistion and Schoch.1 The syndrome, while rare, has received considerable attention in recent years because of the almost universal finding of Ro(SSA) antibody in both mothers and infants with this disorder, suggesting a possible cause-and-effect relationship. The syndrome affects the majority of infants at birth or within the first 2 months of life. The two major components of this disorder are congenital heart block and lupus dermatitis, which may be seen singularly or together in the same infant. Additional systemic manifestations that have been described less frequently include hepatomegaly, splenomegaly, pneumonitis, lymphadenopathy, and hematologic abnormalities (leukopenia, anemia, and thrombocytopenia).2 Curiously, arthritis or nephritis have not been described in this syndrome. Whereas many of the initial reports of NLE were of infants bom to mothers with an overt connective tissue disease, eg, systemic lupus erythematosus, or Sjogren's syndrome, 3 it is now evident that the majority of mothers (70% to 80%) are asymptomatic.
The Ro antibody system is present in both symptomatic and asymptomatic mothers of NLE infants.4,5 This antibody is directed against an RNA protein complex present predominantly within the cell cytoplasm.6 The exact function of the Ro antigen is unknown but is believed to play a vital role in cell function because it is highly conserved among different animal species. The Ro(SSA) antibody system is felt to be highly specific for connective tissue disease. It is seen in 25% of patients with SLE and 40% to 45% of patients with Sjogren's syndrome.7 That the potential risk for asymptomatic mothers of NLE infants to develop connective tissue disease at a later date is probably high is suggested by many documented cases of NLE mothers subsequently developing connective tissue disease, most commonly SLE or Sjögren's syndrome.
We had the opportunity to document in detail the clinical and serologic features of 10 neonatal lupus infants and their mothers.5 All of our patients developed lupus dermatitis, one of whom presented with congenital heart block and later developed the typical skin rash. Approximately 50% of the cases described in the literature have congenital heart block. The characteristic lupus dermatitis may be present at birth and appeared in all but one infant in our series before age 2 months (Figure). The rash presents as discrete erythematous macules, papules, or annular plaques. Scaling and central atrophy are frequent. The face, scalp, and upper trunk, ie, exposed sites, were most frequently involved, although lesions have also been described on unexposed skin. The rash usually resolves by age 6 months with minimal or no scarring; however, changes in pigmentation may persist in some. Definite photosensitivity was documented in 20% (2/10) of our patients. The rash started in one infant after prolonged sun exposure and the other infant had sudden extension of his rash after 1 day of prolonged sun exposure. The differential diagnosis of the lupus dermatitis includes eczema, psoriasis, erythema multiforme, and the genodermatoses that display photosensitivity. It is our impression that the lupus dermatitis associated with neonatal lupus syndrome is under-recognized because the transient nature of the skin lesions in the neonate generally precludes histological investigation (in 50% of our own series the diagnosis of NLE was not considered until the results of biopsy were available). Biopsy of the rash reveals the characteristic changes of lupus erythematosus, ie, epidermal atrophy with basal vacuolization and a mononuclear perivascular infiltrate in the dermis.
Figure. A) Annular plaques noted on the forehead of a 4week-old infant born to an Ro positive asymptomatic mother. B) Higher magnification of annular lesions on forehead of infant in A. C) Facial rash in neonatal lupus infant born to mother with Sjögren's syndrome. Note periorbital distribution of rash. DJ Minimally elevated erythematous and hyperpigmented plaques on the forearm of infant seen in C.
The second major clinical feature of NLE syndrome is complete heart block, which is found in approximately 50% of infants and usually detected at or before birth. Isolated congenital heart block occurs in 1 in 20,000 births and the presence of Ro antibody in over 80% of the mothers who give birth to these infants suggests that most cases of isolated congenital heart block are, in fact, variants of the neonatal lupus syndrome.9,10 The complete heart block occurs as an isolated phenomenon in most, but is associated with other structural abnormalities and/or cardiomyopathy in 25% of infants. The most commonly reported structural abnormalities include ventricular septal defect, patent ductus arteriosus, and transposition of the great vessels. Some of these infants have died shortly after birth with congestive heart failure. In these infants, an unusual incidence of extensive endocardial fibroelastosis has been found at autopsy.3 Cardiac conduction defects (first degree AV block, Wenckebach phenomenon, and bundle branch block) have also been described in NLE infants.
COMPLETE HEART BLOCK
While the lupus dermatitis is transient, complete heart block is permanent. Autopsy examination of these infants reveals diffuse fibrosis and calcification affecting predominantly the AV node, but the intraatrial and intraventricular conduction tissue may also be involved.3,8 This finding, coupled with the presence of a diffuse cardiomyopathy in some infants, suggests the presence of an ongoing inflammatory process. The conduction system may be preferentially affected because of its rich vascular supply. The Ro antigen has been demonstrated in fetal myocardium and antibody deposits have been demonstrated diffusely throughout the myocardium by immunofluorescence techniques in an infant who died of the neonatal lupus syndrome.11 The finding in some infants of a structural disorder, ie, transposition, is not explained by an ongoing inflammatory process but suggests cardiac injury occurring during a critical period of cardiac embryogenesis. The occasional finding of less complete conduction defects, ie, bundle block, in NLE infants may represent a less intensive inflammatory response.
The Ro antibody system was present in 100% of the mothers and affected infants in our series. Thus far, this antibody system has been found in 64 of 73 mothers of infants with lupus dermatitis and/or isolated congenital heart block and in 32 of 34 infants tested.5 While the Ro antibody is almost universal in the neonatal lupus syndrome, other autoantibodies have also been described, notably antinuclear antibody, rheumatoid factor, and the La(SSB) antibody. A number of these mothers may demonstrate ANA positivity; however, they may be ANA-negative on certain substrates, ie, mouse liver, and, thus, the syndrome may be missed if antibodies to the Ro and La system are not examined. The La antibody is another autoantibody to a cytoplasmic RNA macromolecule that is often found in association with the Ro antibody. Thus far, this antibody system has been detected in 18 of 29 NLE infants and 30 of 54 NLE mothers.5 Whether this antibody system is of pathogenetic importance in NLE is not known. The antibodies characteristically disappear from the infant's serum at approximately age 6 months at the time of resolution of the dermatitis. These findings suggest that the Ro antibody is transferred across the placenta to the infants in utero, the majority of antibodies crossing in the third trimester. The disappearance of the Ro antibody from the infant's serum correlates well with the resolution of lupus dermatitis, suggesting a cause-andeffect relationship. How the Ro antibody may cause the dermatitis is unknown at this time. A similar annular dermatitis occurs in adults with a subacute cutaneous form of lupus, and in adults with C2 and C4 deficiency LE syndromes in whom the incidence of Ro antibody is 70% to 75% respectively. The Ro antigen has been demonstrated in both the epidermis of the normal neonate and the myocardium of apparently normal fetuses. In vitro studies show that ultraviolet irradiation of human epidermal cells in culture may lead to expression of the Ro antigen on the cell surface, thus making it available for interaction with the Ro antibody.12 This finding may explain why these infants have their disease limited primarily to sunexposed areas; however, it is obvious that exposure to ultraviolet light is not an absolute prerequisite for the disease, as some infants display cutaneous lesions at birth. Recently, Norris et al suggested that the Ro antibody may be capable of affecting antibody-dependent cellular cytotoxicity which may be important in the genesis of the lupus lesions. 13
The majority of mothers are asymptomatic (75%). Of the remainder, SLE is the predominant underlying disorder, although Sjogren's syndrome, rheumatoid arthritis, and undifferentiated connective tissue disease have been described. It is estimated that some 30% to 60% of mothers of these infants ultimately develop clinical or laboratory evidence of connective tissue disease, primarily SLE. The interval between the birth of an NLE infant and development of connective tissue disease may be as long as 16 years. A strong family history of connective tissue disease has been found in some mothers. These women may also be at increased risk for fetal wastage. Isolated reports suggest this increase in wastage, and we demonstrated in a retrospective study of 50 Ro positive women that fetal wastage was significantly increased in black Ro positive women with SLE when compared with black nuclear RNP positive women with SLE. 15 A prospective study is required to clarify this point. Disease expression may vary in siblings of NLE infants. Siblings born after an NLE infant may be normal or have similar (ie, both with rash) or different (ie, one with rash and one with CHB) manifestations. The exact incidence of sibling involvement is unknown.
Although our level of suspicion is high for a causeand-effect role of the Ro antibody in the etiology of neonatal lupus, it is still possible that this antibody is merely an epiphenomenon. We have documented 12 incidences in which mothers known to be Ro positive during their pregnancy have produced normal asymptomatic babies.5 Comparison of the HLA phenotypes of NLE mothers and Ro positive mothers of normal infants reveals an increased incidence of HLAA1B8DR3 in NLE mothers, and an increased incidence of HLA-DR2 in Ro positive mothers of normal infants.5,14 Ro antibody production is associated with both the HLA-DR2 and DR3 phenotypes. The highest titers of Ro antibody are found in DR3 positive individuals. Thus, one might speculate that the titer of Ro antibody may be important in the expression of this syndrome. Neonatal lupus infants demonstrate no increase in HLA, DR2 or DR3 phenotypes and, thus, the clinical expression of this syndrome does not appear to be HLA dependent.
The recent serologic findings in neonatal lupus raise even more questions. What are the chances of a Ro positive woman delivering an NLE infant? What percentage of pregnant women possess the Ro antibody? How does one manage a pregnant Ro-positive woman? Approximately 1% of pregnant women have been shown to possess the Ro antibody. The risk of an Ro positive woman delivering an infant with neonatal lupus is probably low, less than 10% as shown by our series of 10 Ro positive women followed throughout pregnancy, none of whom had an infant with NLE. Careful follow-up is required during the pregnancy in order to detect the development of connective tissue disease in the mother or the development of congenital heart block in the infant. Careful monitoring of fetal heart rate is required, and the infant may be followed by ultrasound examination and fetal echocardiography if necessary. What happens if a fetus shows signs of heart block or cardiomyopathy? The appropriate treatment is as yet unknown, but may well depend on the trimester of presentation. Suggested treatment modalities have included prednisolone and plasmapheresis to remove antibody from the maternal circulation.
With regard to NLE infants the question remains as to their prognosis. A small number of these infants have developed adult LE, juvenile rheumatoid arthritis, or Sjogren's syndrome. In the absence of complete heart block the immediate prognosis is excellent, with the clinical and serological features disappearing by age 6 months to 1 year. Many infants with isolated complete heart block have led productive lives, but the presence of other structural defects and cardiomyopathy is associated with increased mortality in these infants. Infants with profound bradycardia and poor cardiac output will require pacemaker insertion.
In summary, NLE is an inflammatory disorder of neonates characterized by a transient cutaneous lesion and/or congenital heart block. Photosensitivity is recognized as a component of this syndrome. While the majority of mothers are asymptomatic, a large proportion of these women go on to develop connective tissue disease. The Ro antibody is a serologic marker for the syndrome and is present in NLE mothers and their infants during the first 6 months of life. HLA studies reveal that HLA-A1B8, DR,, MB2 and MT2 positive women who possess the Ro antibody are at significant risk of delivering an NLE infant.
The recent findings in NLE have provided an exciting area for further exploration by dermatologists, rheumatologists, and cardiologists alike as this syndrome is the first association of an autoantibody with the pathogenesis of congenital heart block and the cutaneous lupus lesion.
1. McCuistionCH, Schoch EP: Possible discoid lupus erythematosus in newborn infant. Arch Dermatol SmAIoI 1954: 70:782.
2. Draznin TH, Esterly NB, Rirey NL, et al: Neonatal lupus erythematosus. } Am Acad Dermatol 1979; 1:437.
3. Chameides L, Truex RC, Vetter V, et al: Association of maternal lupus erythematosus with congenital complete heart block. N Engl ) Med 1977; 297:1204.
4. Franco HL, Weston WL, Peebles C, et al: Autoantibodies directed against sicca syndrome antigens in the neonatal lupus syndrome. J Am Acad Dermatol 1981; 4:67.
5. Watson RM, Lane AT. Bamett NK. et al: Neonatal lupus erythematosus. A clinical, serological and immunogenetic study with review of the literature. Medicine 1984; 63:362.
6. Clark G, Retchlin M, Tornasi TB Jr: Characterization of a soluble cytoplasmic antigen reactive with sera (rom patients with systemic lupus erythematosus. ] Immunol 1969; 102:117.
7. Alexander EL, Hirsh TJ, Amen FC, et al: The Ro(SSA) and U(SSB) antibody systems in Sjogren's syndrome. ] Rheumatol 1982; 9:239.
8. Esscher E, Scott JS: Congenital heart block and maternal systemic lupus erythematosus. Br Med J 1979; 1:1235.
9. Michaelsson M, Engle MA: Congenital complete heart block: An international study of the natural history. Cardiovasc CIm 1972; 4:85.
10. Scott JS, Maddison PJ, Taylor PV, et al: Connective tissue disease, antibodies to ribonucleoprotein and congenital heart block. N Engl J Med 1983; 309:209.
11 . Litsey SE, Noonan JA. O'Connor WN, et al: Maternal connective tissue disease and congenital heart block. Demonstration of immunoglobulin in cardiac tissue. N Engl] Med 1985; 312:98.
12. LeFeber WP. Norris DA, Weston WL, et al: Induction of nuclear antigen expression in human keratinocytes by ultraviolet light (UVL), abstracted. CIm Res 1983; 148A.
13. Norris DA, Ryan SR, Fritz KA, et al: The role of RNP. Sm and SSA specific antisera from patients with lupus erythematosus in inducing antibody dependent cellular cytotoxicity (ADCC) of ribonucleoprotein-coated targets, abstracted. Clin Res 1982; 30:264a.
14. Lee LA, Bias WB, AmettFC, et al: lmmunogenetics of the neonatal lupus syndrome. Ann lruem Med 1984; 99:592.
15. Watson RM, Braunstein BL, Watson AJ, et al: Fetal wastage in women with antiRo(SSA) antibody, j Rheumatol 1986; 13:90.