7/29/2019 10638_ftp

1/7

ARTHRITIS & RHEUMATISM

Vol. 46, No. 9, September 2002, pp 23772383

DOI 10.1002/art.10638

2002, American College of Rheumatology

Long-Term Followup of Children With Neonatal Lupus and

Their Unaffected Siblings

Victor Martin,1 Lela A. Lee,2 Anca D. Askanase,1 Margaret Katholi,1 and Jill P. Buyon1

Objective. To determine in a longitudinal cohort

study whether children with varied manifestations of

neonatal lupus or their unaffected siblings later develop

autoantibodies and/or rheumatic diseases.

Methods. To obtain information on the health of

children ages >8 years who had manifestations of

neonatal lupus (affected group) and their unaffectedsiblings (unaffected group), questionnaires were sent to

mothers (with anti-SSA/Ro and/or anti-SSB/La anti-

bodies) who were enrolled in the National Institute of

Arthritis and Musculoskeletal and Skin Diseases/

Hospital for Joint Diseases Research Registry for Neo-

natal Lupus. Children of healthy mothers referred by

the Registry enrollees comprised the control group.

Further data were provided by review of medical

records.

Results. Fifty-five mothers enrolled in the Regis-

try returned questionnaires on 49 children with neona-

tal lupus and their 45 unaffected siblings. Six children

with definite rheumatic/autoimmune diseases wereidentified: 2 with juvenile rheumatoid arthritis, 1 with

Hashimoto thyroiditis, 1 with psoriasis and iritis, 1 with

diabetes mellitus and psoriasis, and 1 with congenital

hypothyroidism and nephrotic syndrome. All had neo-

natal lupus, and their mothers had manifestations of

autoimmune diseases (Sjogrens syndrome in 4, sys-

temic lupus erythematosus/Sjogrens syndrome in 1,

and undifferentiated autoimmune disease in 1). Anti-

nuclear antibodies were present in 4 of 55 sera tested (2

of 33 affected children and 2 of 22 unaffected children).

No serum contained antibodies reactive with SSA/Ro or

SSB/La antigens.

Conclusion. These data suggest that children with

neonatal lupus require continued followup, especially

prior to adolescence and if the mother herself has anautoimmune disease. While there was no apparent

increased risk of systemic lupus erythematosus, the

development of some form of autoimmune disease (sys-

temic or organ-specific) in early childhood may be of

concern. During adolescence and young adulthood, in-

dividuals with neonatal lupus and their unaffected

siblings do not appear to have an increased risk of

developing systemic rheumatic diseases.

Neonatal lupus is a model of passively acquiredautoimmunity in which a motherwho may have sys-temic lupus erythematosus (SLE) or Sjogrens syndrome(SS) or may be entirely asymptomaticsynthesizes an-tibodies to SSA/Ro and/or SSB/La ribonucleoproteinsthat enter the fetal circulation via trophoblast FcRnreceptors and presumably cause tissue injury (1,2). Themost serious manifestation of neonatal lupus is cardiacdisease, which is usually third-degree irreversible atrio-

ventricular (AV) block. Other neonatal abnormalitiesaffecting the skin, liver, and blood elements are alsoassociated with maternal anti-SSA/Ro and/or anti-SSB/La antibodies and are grouped under the headingof neonatal lupus syndromes. The noncardiac manifes-tations are transient and disappear in parallel with the

clearance of maternal antibodies from the neonatalcirculation.

Given the rarity of the disease, little informationabout the health outcome of children with neonatallupus and their unaffected siblings is available. This is ofinterest from several perspectives. There is a geneticsusceptibility for the development of SLE (37), andrelatives of SLE patients can have autoantibodies in the

Supported in part by a contract from the National Institute ofArthritis and Musculoskeletal and Skin Diseases (N01-AR-4-2220) forthe Research Registry for Neonatal Lupus.

1Victor Martin, MD, Anca D. Askanase, MD, MargaretKatholi, AA, Jill P. Buyon, MD: Hospital for Joint Diseases, New YorkUniversity School of Medicine, New York, New York; Lela A. Lee,MD: University of Colorado Health Sciences Center, and DenverHealth Medical Center, Denver, Colorado.

Address correspondence and reprint requests to Jill P. Buyon,MD, Department of Rheumatology, Room 1608, Hospital for JointDiseases, 301 East 17th Street, New York, NY 10003. E-mail:

jill.buyon@med.nyu.edu.Submitted for publication December 31, 2001; accepted in

revised form May 9, 2002.

2377

7/29/2019 10638_ftp

2/7

absence of clinical disease (8). Perhaps in addition toautoantibodies, the expression of SLE in the motherincreases the risk of autoimmune disease in her off-spring, independently of whether the child has neonatallupus or not. This generates the hypothesis that children

with neonatal lupus, as well as their unaffected siblings,whose mothers have SLE might be at greater risk for thedevelopment of subsequent disease than children whosemothers are asymptomatic. It remains unknown whysome fetuses and neonates are susceptible to cardiac orcutaneous injury while others exposed to the identicalmaternal antibodies are healthy. Perhaps vulnerability toneonatal lupus is a marker for susceptibility to, orprotection from, the development of actively acquiredautoimmunity. Accordingly, this study was initiated todetermine whether children with varied manifestationsof neonatal lupus or their unaffected siblings later

develop autoantibodies and/or rheumatic diseases, andwhether there is any correlation with the maternaldiagnosis.

PATIENTS AND METHODS

Study group and controls. The National ResearchRegistry for Neonatal Lupus, which was established in Sep-tember 1994 by the National Institute of Arthritis and Muscu-loskeletal and Skin Diseases at the Hospital for Joint Diseases,served as the source of patients for the current study. Details ofthe Registry have been described elsewhere (9,10). Briefly, amother is enrolled in the Registry if her child has anymanifestation of neonatal lupus (AV block, characteristic skin

rash, or hematologic or hepatic abnormalities) and if themother herself has anti-SSA/Ro and/or anti-SSB/La antibod-ies, irrespective of her clinical status. The cardiac manifesta-tions of neonatal lupus are documented by in utero echocar-diogram or postnatal EKG. The cutaneous manifestations aredocumented by a photograph demonstrating the characteristicannular lesions, a clear description of the rash in the childsmedical records, and/or findings on skin biopsy.

The current study included children with manifesta-tions of neonatal lupus (affected group) and their asymptom-atic siblings (unaffected group) who were at least 8 years ofage. Exclusion of children younger than 8 was based on severalconcerns: the invasive nature of blood-drawing in youngerchildren, the greater reliability of answers to the questionnairefor older children, and the presumption that followup for fewerthan 8 years would be too limited to study the development ofrheumatic/autoimmune disease. Mothers of all children hadanti-SSA/Ro and/or anti-SSB/La antibodies documented ei-ther in our laboratory (by JPB) or commercially.

The control group comprised families referred bymothers in the Registry and included 53 friends of children inthe study group whose mothers were willing to participate.Each control subject chosen for study was within 2 years of ageof the index case and of the same ethnic background.

Maternal diagnosis was ascertained and assigned as

previously described (9). Mothers were classified as havingSLE if they met at least 4 of the American College ofRheumatology criteria (11). Mothers were categorized ashaving SS if they had 1) both dry eyes and dry mouth, or 2)either dry eyes or dry mouth along with objective documenta-tion of salivary or lacrimal gland hypofunction or lymphocyticinfiltration of these glands. An undifferentiated autoimmunesyndrome (UAS) was diagnosed in patients with features of arheumatic disease who did not have prominent sicca signs andsymptoms and did not meet the criteria for SLE. Asymptom-atic mothers were those who had no clinical evidence of anyrheumatic illness.

Questionnaire. In order to evaluate the health out-comes of the children with neonatal lupus and their siblings,mothers with antibodies to SSA/Ro and/or SSB/La whosechildren met eligibility criteria were sent a letter inviting themto participate. The consenting mothers were then sent ques-tionnaires consisting of 41 items that specifically focused onsymptoms of rheumatic and autoimmune diseases in theirchildren.

The first group of questions addressed symptoms char-

acteristic of autoimmune diseases, mainly SS and SLE: dryeyes, dry mouth, hair loss, skin rash, photosensitivity, arthral-gias, arthritis, morning stiffness, oral ulcers, nodules,Raynauds phenomenon, pleuritis, pericarditis, renal disease,or central nervous system involvement (seizure, stroke). Thenext group of questions focused on laboratory data: abnormalfindings on urinalysis, low white blood cell count, low plateletcount. The final group of questions specifically inquired aboutany previous diagnosis of autoimmune diseases: SLE, SS,rheumatoid arthritis, ankylosing spondylitis, scleroderma, poly-myositis, dermatomyositis, inflammatory bowel disease, diabe-tes, primary biliary cirrhosis, or thyroid disease. Informationon past and present medications was also obtained. All positiveanswers were followed up in a telephone interview with themother. Medical records from the childs pediatrician, pediat-

ric rheumatologist, dermatologist, or cardiologist were thenrequested and reviewed.Determination of antibodies. All consenting mothers

were asked to send blood samples from their children forevaluation of autoantibodies. Initial screening of the sera forantibodies reactive with SSA/Ro and/or SSB/La was done byenzyme-linked immunosorbent assay (ELISA) using a com-mercial kit (Diamedix, Miami, FL). Although the SSA/Roantigen in this kit does not include the 52-kd component, allsera from mothers enrolled in the Registry who have anti-SSA/Ro antibodies are positive by this ELISA, including thosewith only anti52-kd SSA/Ro on immunoblot. We presumethat the commercial ELISA includes native 60-kd SSA/Ro.Therefore, sera were evaluated by immunoblot only if theELISA results were positive. Antinuclear antibodies (ANA)were evaluated using HEp-2 cell substrate (Kallestad Quan-tafluor kit; Bio-Rad, Hercules, CA).

Statistical analysis. Comparisons between groupswere calculated using Fishers exact test. Pvalues less than 0.05were considered significant.

RESULTS

Demographics. As of October 26, 2000, a total of254 mothers and their 300 children were enrolled in the

2378 MARTIN ET AL

7/29/2019 10638_ftp

3/7

Research Registry for Neonatal Lupus. Based on find-ings from the enrollment questionnaire, review of themedical records, and verification of maternal autoanti-bodies to SSA/Ro and/or SSB/La, 241 children wereclassified as having neonatal lupus. The distribution of

clinical manifestations was as follows: 141 children hadAV node block only, 27 had both cardiac and cutaneousmanifestations, 2 had isolated cardiomyopathy, 66 hadrashes alone, and 5 had miscellaneous manifestations,including hematologic or hepatic abnormalities in theabsence of congenital heart block (CHB) or rash.

Eighty-five mothers with anti-SSA/Ro and/or anti-SSB/La antibodies whose children met the inclusioncriteria were sent a letter inviting them to participate inthe study. Sixty-nine mothers consented and were thensent questionnaires. Of these, 55 mothers returned 94questionnaires: 49 concerning their children with neo-natal lupus and 45 concerning their unaffected children.In addition, these 55 mothers referred 53 control moth-ers who completed the identical questionnaire for theirchildren.

The majority of the study mothers who com-pleted the questionnaires were Caucasian (89%). None

were African American, 8% were Hispanic, and 3%were Asian. The ethnic proportions were similar for thecontrol mothers. The only significant difference in eth-nic distribution between the study group, the group of 30invited mothers who did not respond to the studyinvitation, and the Registry itself is in the percentage of

African Americans: 0 in the study group, 10% in the

nonresponse group, and 10% in the Registry (P

0.04for the study group versus the nonresponse group).Disease status of the mothers. Maternal health

status at birth of the child with neonatal lupus and atfollowup was ascertained by questionnaire, interview,and review of medical records. Twelve of the mothers(22%) were initially asymptomatic; 6 of them progressedto develop SLE (in 2), SS (in 3), and SLESS overlap (in1). Eight mothers (15%) had UAS at the birth of theaffected child; of these, 1 developed SLE and 2 devel-oped SS. Seventeen mothers (31%) were initially diag-nosed as having SS; 4 of them developed symptoms ofSLE. Ten mothers (18%) initially presented with SLE; 1

of them progressed to develop SLESS overlap. Fourmothers (7%) had SLESS at the birth of the affectedchild. In 4 mothers (7%) whose initial diagnosis wasunknown, 3 developed SLESS overlap and 1 developedSLE.

The current rheumatologic diagnoses of the 30mothers who did not respond to the study invitation areas follows: 3 (10%) asymptomatic, 3 (10%) UAS, 13

(43%) SS, 7 (23%) SLE, 3 (10%) SLESS overlap, and1 (3%) unknown. This distribution was not significantlydifferent from that of the followup diagnoses of themothers who did respond to the study invitation.

Demographics of the children and manifesta-

tions of neonatal lupus. The mean SD age of theaffected children was 14.6 6 years, of the unaffectedchildren 15.7 5 years, and of the controls 13.9 6

years (Figure 1). The majority of the affected childrenwere between the ages of 8 years and 16 years: 49% were812 years old, and 27% were 1316 years old. Thepercentages diminished for each subsequent age group.The age distribution was similar in the unaffected sib-lings and the controls. Four children in the affectedgroup are currently older than 25 years.

The manifestations of neonatal lupus in the 49affected children were as follows: 27 (55%) had CHBalone, 19 (39%) had only cutaneous manifestations, and3 (6%) had both cardiac and skin involvement. Of the 27children with CHB alone, 17 (63%) were male; of the 19children with rash alone, 7 (37%) were male. However,utilizing Fishers exact test, there was no significantdifference in the sex distribution between these twomanifestations (P 0.13), which parallels the distribu-tion observed in the entire Registry. The 3 children withboth manifestations were female. The sex distribution ofthe unaffected siblings was 25 males and 20 females, andthat of the controls was 26 males and 27 females.

The 30 mothers who did not respond to the studyinvitation had 17 affected children (9 males, 8 females)

and 42 unaffected children (19 males, 23 females). Themean age of the affected children was 17.82 years (Pnotsignificant versus the study group), and the mean age ofthe unaffected children was 16.4 years (P not significant

versus the study group). Of the 17 affected children, 13(6 males, 7 females) had CHB alone, 2 (1 male, 1female) had rash alone, and 2 (both males) had bothCHB and rash (Pnot significant versus the study group).

Rheumatic disease manifestations in the chil-

dren. The questionnaire was the initial source of infor-mation regarding medical followup of the children in thestudy group and the control group. Mothers were spe-cifically instructed to report only on symptoms that were

present after the children were 1 year of age. Specifi-cally, malar rash, thrombocytopenia, leukopenia, or liverabnormalities present at birth could be considered partof the spectrum of neonatal lupus.

As shown in Figure 2, the majority of positiveresponses were to the questions related to joint symp-toms. Overall, 10 of the affected children (20%), 11 ofthe unaffected siblings (24%), and 8 of the controls

LONG-TERM FOLLOWUP IN NEONATAL LUPUS 2379

7/29/2019 10638_ftp

4/7

(15%) had joint complaints (P 0.5 by Fishers exacttest). Given the importance and confusion of jointcomplaints in a child, further elucidation of this symp-tom was obtained (insert in Figure 2). Specifically,mothers reported that 8 of their affected children (16%)and 6 of their unaffected children (13%) complained of

joint stiffness. Joint stiffness was not reported in the

controls. Arthralgias were reported in 8 of the affectedchildren (16%), 9 of the unaffected siblings (20%), and8 of the controls (15%). Frank swelling of the joints wasreported in 5 children in the affected group; swelling wasof sufficient severity to warrant medical consultation in2 of them (see below). Interestingly, 3 children in the

unaffected group had joint swelling but were not

Figure 1. Age distribution of the children with neonatal lupus, their unaffected siblings,and the controls at the time of followup. Numbers at the top of the bars are the number of

subjects. Information in the boxes to the right of the bars indicates the subjects withrheumatic/autoimmune diseases and serologic abnormalities (see Table 1 for age at disease

onset). JRA juvenile rheumatoid arthritis; ANA antinuclear antibody.

Figure 2. Rheumatic disease manifestations in the children with neonatal lupus, their unaffected siblings,

and the controls, based on responses to the questionnaire. Numbers at the top of the bars are the numberof subjects reporting each symptom. The inset at upper right provides detailed data regarding joint

symptoms.

2380 MARTIN ET AL

7/29/2019 10638_ftp

5/7

brought to medical attention (see below). No child in thecontrol group had joint swelling. Except for joint stiff-ness, which was significantly increased in affected chil-dren and their unaffected siblings versus controls (P0.008), no other symptoms reached statistical signifi-cance.

In all groups, there were isolated reports ofphotosensitivity, malar rash, Raynauds phenomenon,sicca symptoms, and palatal ulcers. Telephone inter-

views with each of the mothers revealed that in themajority of cases, these complaints did not prompt a visitto the childs pediatrician. Moreover, those children who

were seen by a pediatrician were not considered to havesignificant disease.

Based on the questionnaires and medical records,there were 6 substantiated cases of rheumatic/auto-

immune diseases, all of which occurred in the childrenwith neonatal lupus (Table 1). Two females, each withcutaneous manifestations of neonatal lupus, developed

juvenile rheumatoid arthritis: one (diagnosed at age 2years) had polyarticular disease, and the other (diag-nosed at age 5 years) had pauciarticular disease. Both oftheir mothers had SS. The third child (female), also withcutaneous neonatal lupus, developed Hashimoto thy-roiditis at age 7 years. Her mother had SS. The fourthchild (female) with CHB developed psoriasis at age 13

years, and 8 months later, she developed iritis. Hermother had UAS. The fifth child (female) with CHB wasdiagnosed as having type 1 diabetes mellitus at age 7

years and psoriasis at age 8 years. Her mother had SS.The sixth child (male) with CHB was diagnosed ashaving congenital hypothyroidism at age 5 weeks anddeveloped nephrotic syndrome (minimal change dis-ease) at age 3 years. His mother had SLESS overlap.The current ages of these affected children are shown inFigure 1.

There were no cases of SLE, SS, ankylosing

spondylitis, scleroderma, polymyositis, dermatomyositis,inflammatory bowel disease, or primary biliary cirrhosisin any group. No children in the unaffected or control

groups were diagnosed as having a rheumatic or auto-immune disease.Serologic results. Fifty-five blood samples (33

from children with neonatal lupus and 22 from theirunaffected siblings) were sent to the Registry for sero-logic evaluation. Four sera were found to have ANA.One sample (ANA titer 1:80 in a homogeneous pattern)

was from a 9-year-old boy with a history of cutaneousneonatal lupus. He has experienced intermittent jointpains, but this has not been brought to medical atten-tion. The second sample (ANA titer 1:160 in a homog-enous pattern) was from the affected child who devel-oped pauciarticular juvenile rheumatoid arthritis. Thethird sample (ANA titer 1:80 in a speckled pattern) wasfound in an unaffected female (age 21 years) who iscurrently completely asymptomatic. The fourth sample(ANA titer 1:160 in a homogenous pattern) was from anunaffected male (age 15 years) who is currently com-pletely asymptomatic. None of the childrens sera testedcontained antibodies to either SSA/Ro or SSB/La.

DISCUSSION

The Research Registry for Neonatal Lupus pro-vides an excellent resource with which to address the

clinical and serologic progression of autoimmunity inchildren born to women with anti-SSA/Ro and/or anti-SSB/La antibodies. Such information is important notonly for family counseling, but also for addressing thepotential role of maternal factors in the subsequentdevelopment of autoimmunity in their children. Thesefactors may be genetically encoded, such as HLA, andinherited by the offspring, and/or they may be related to

Table 1. Children with rheumatic/autoimmune diseases*

Patient/

sex

Neonatal lupus

manifestation

Diagnosis at followup

(age at disease onset)

Mothers

diagnosis

1/F NLE rash Polyarticular JRA (2 years) SS2/F NLE rash Pauciarticular JRA (5 years) SS

3/F NLE rash Hashimoto thyroiditis (7 years) SS4/F CHB Psoriasis and iritis (13 years) UAS5/F CHB Type 1 diabetes mellitus (7 years)

Psoriasis (8 years)SS

6/M CHB Congenital hypothyroidism (5 weeks)Nephrotic syndrome (3 years)

SLESS

* NLE neonatal lupus erythematosus; JRA juvenile rheumatoid arthritis; SS Sjogrens syndrome;CHB congenital heart block; UAS undifferentiated autoimmune syndrome; SLESS systemiclupus erythematosusSjogrens syndrome overlap.

LONG-TERM FOLLOWUP IN NEONATAL LUPUS 2381

7/29/2019 10638_ftp

6/7

the maternal immunologic presence (cellular or hu-moral) in the fetus.

Genetic factors are undoubtedly important in thepathogenesis of autoimmunity and, in particular, SLE.Familial aggregation of SLE has been demonstrated

(37). Hochberg and colleagues (3) studied the occur-rence of SLE among first-degree relatives of 77 patients

with SLE and age-, sex-, and race-matched controls whohad no history of a rheumatic disease. Ten percent of theSLE probands had 1 or more first-degree relatives withSLE, compared with only 1% of the controls (relativerisk 8, P 0.03). SLE occurred in 9 (7 females) of 541first-degree relatives of the SLE probands, compared

with only 1 (female) of the controls. In another study,Lawrence et al (4) reported SLE in 4% of the first-degree relatives of SLE probands, compared with only1% of 128 age- and sex-matched first-degree relatives ofthe controls (P 0.001). Further support for a heredi-tary predisposition to SLE is the finding of a greaterconcordance of SLE among monozygotic than dizygotictwin pairs (57). Interestingly, inheritance may be re-flected in the acquisition of autoantibodies and notnecessarily in the expression of overt disease (8).

Previous studies have demonstrated that autoan-tibodies are commonly found in the sera of first-degreerelatives of children with SLE (12,13). Provocative, butthus far unexplained, is the observation by Lehman et al(14) that the development of SLE in children youngerthan age 10 or in males younger than age 18 is stronglycorrelated with maternal antibodies to SSA/Ro. That

study included 71 children with SLE and their 188unaffected first-degree relatives. The mothers of themale children were all clinically well. None of thechildren diagnosed as having early-onset SLE had neo-natal lupus, at least no cardiac manifestations, sincechildren with CHB were excluded. The authors specu-lated that maternal antibodies to SSA/Ro might initiateantiidiotypic antibodies in the fetus or neonate, withsubsequent amplification by the idiotypeantiidiotypenetwork.

Viewed from a different perspective, one cantake as the inclusion group mothers with anti-SSA/Roantibodies, and study the incidence of SLE or other

autoimmune diseases in their offspring. To date, theliterature on the long-term followup of children withneonatal lupus is limited to anecdotal reports, andfollowup of unaffected siblings is nonexistent. Therehave been 7 cases published in which an autoimmunedisease developed in a child who had manifestations ofneonatal lupus. All of the children were female, and allexcept one of their mothers had SLE.

Specifically, Esscher and Scott (15) described afemale with CHB who developed SLE at age 15 years.Jackson and Gulliver (16) described an infant withcutaneous neonatal lupus who developed SLE at age 13,and Fox et al (17) described a similar patient who

developed SLE at age 19 years. In a patient described byWaterworth (18), CHB was identified at age 6 years, andSLE was diagnosed at age 13 years; no informationabout the mother was provided. Lanham et al (19)described 2 children with CHB; one subsequently devel-oped primary SS at age 23 years, and the other subse-quently developed arthritis, ANA, and antibodies todouble-stranded DNA at age 19 years. Hubscher et al(20) reported the development of scleroderma of theface, puffy hands, and Raynauds phenomenon in a13-year-old girl with CHB. At age 15, she was found tobe seropositive for anti-SSA/Ro and U1 RNP. It shouldbe emphasized that maternal anti-SSA/Ro or anti-SSB/La antibodies were not actually documented for anyof these mothers, a requisite for inclusion in our study.In one small series, Brucato et al (21) found that at amean followup of 18 years, none of 13 children withCHB developed clinical symptoms or serologic abnor-malities suggestive of an immune disease.

To date, the data presented herein represent thelargest number of children with neonatal lupus and theirsiblings whose cases have been followed until the chil-dren were at least 8 years of age. Most of the positiveresponses to items on the questionnaire were referableto the joints. Specifically, arthralgias, which could be

attributed to growing pains, were reported for severalchildren in all groups. The joint stiffness and arthritisreported for only the patients in families with neonatallupus could reflect a heightened concern on the part ofmothers who are more aware of rheumatic diseases,although in 2 cases, bona fide disease was established.Based on medical records, we identified 6 of 49 children

with neonatal lupus who had a definable rheumatic/auto-immune disease; all but 1 were female. Three of thesechildren were born with CHB, and 3 presented withcutaneous manifestations. In contrast, none of the 45unaffectedsiblingsdevelopedadefinablerheumatic/auto-immune disease. Although these data suggest that vul-

nerability to neonatal lupus might be a risk factor(whether simply a clinical marker or related to anas-yet-unidentified important molecular mechanism) forthe development of a subsequent rheumatic/auto-immune disease in an offspring of a mother with anti-bodies to SSA/Ro and/or SSB/La, the small numberspreclude firm conclusions. None of 55 sera tested (in-clusive of children with and without neonatal lupus) had

2382 MARTIN ET AL

7/29/2019 10638_ftp

7/7

antibodies to SSA/Ro or SSB/La, although 4 werepositive for ANA.

A contribution of maternal health status to thedevelopment of disease in the children is suggestive, butrestricted by the observation that of the mothers who

consented to participate in the study, only 6 remainedasymptomatic during the followup period. Of the 6mothers whose children subsequently developed arheumatic/autoimmune disease, 5 (83%) had or devel-oped symptoms or signs of SS, either alone or inassociation with SLE.

These data suggest that children with neonatallupus require continued followup, especially prior toadolescence and if the mother herself has an auto-immune disease. While there was no apparent increasedrisk of SLE, the development of some form of auto-immune disease (systemic or organ-specific) in earlychildhood may be of concern. During adolescence and

young adulthood, individuals with neonatal lupus andtheir unaffected siblings do not appear to have anincreased risk of developing systemic rheumatic dis-eases. The maintenance of a national registry, withenrollment of patients during their infancy, provides theoptimum means of confirming these important observa-tions. Followup as the cohort ages to adulthood will be amajor resource for definitively assessing the risk ofdeveloping an autoimmune disease.

ACKNOWLEDGMENTS

The authors thank all the families enrolled in theResearch Registry for Neonatal Lupus, Drs. Kathleen Hainesand Lisa Rider for reviewing the study questionnaire from thepediatric rheumatologists perspective, and Ann Rupel forassistance in preparing the manuscript and figures.

REFERENCES

1. Buyon JP. Neonatal lupus syndromes. In: Lahita RG, editor.Systemic lupus erythematosus. 3rd ed. New York: Academic Press;1998. p. 33759.

2. Lee LA. Maternal autoantibodies and pregnancy. II. The neonatallupus syndrome. Baillieres Clin Rheumatol 1990;4:6984.

3. Hochberg M, Florsheim P, Scott J, Arnett F. Familial aggregationof systemic lupus erythematosus [abstract]. Am J Epidemiol1985;122:526 7.

4. Lawrence JS, Martins L, Drake G. A family survey of lupuserythematosus. 1. Heritability. J Rheumatol 1987;14:91321.

5. Block SR, Winfield JB, Lockshin MD, DAngelo WA, ChristianCL. Studies of twins with systemic lupus erythematosus: a reviewof the literature and presentation of 12 additional sets. Am J Med1975;59:53352.

6. Arnett FC, Shulman LE. Studies in familial systemic lupus ery-thematosus. Medicine (Baltimore) 1976;55:31322.7. Deapen D, Escalante A, Weinrib L, Horwitz D, Bachman B,

Roy-Burman P, et al. A revised estimate of twin concordance insystemic lupus erythematosus. Arthritis Rheum 1992;35:3118.

8. Shoenfeld Y, Slor H, Shafrir S, Krause I, Granados J, VillarrealGM, et al. Diversity and pattern of inheritance of autoantibodiesin families with multiple cases of systemic lupus erythematosus.

Ann Rheum Dis 1992;51:6118.9. Buyon JP, Hiebert R, Copel J, Craft J, Friedman D, Katholi M, et

al. Autoimmune-associated congenital heart block: mortality, mor-bidity, and recurrence rates obtained from a national neonatallupus registry. J Am Coll Cardiol 1998;31:165866.

10. Neiman AR, Lee LA, Weston WL, Buyon JP. Isolated cutaneousmanifestations of neonatal lupus: characteristics of mothers andchildren enrolled in a national registry. J Pediatr 2000;37:674 80.

11. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield

NF, et al. The 1982 revised criteria for the classification of systemiclupus erythematosus. Arthritis Rheum 1982;25:12717.

12. Lehman TJA, Hanson V, Zvaifler N, Sharp G, Alspaugh M.Antibodies to nonhistone nuclear antigens and anti-lymphocyteantibodies among children and adults with systemic lupus ery-thematosus and their relatives. J Rheumatol 1984;11:6447.

13. Lehman TJA, McCurdy DM, Bernstein BH, King KK, Hanson V.Systemic lupus erythematosus in the first decade of life. Pediatrics1989;83:2359.

14. Lehman TJA, Reichlin M, Santner TJ, Silverman E, Petty RE,Spencer CH, et al. Maternal antibodies to Ro (SS-A) are associ-ated with both early onset of disease and male sex among children

with systemic lupus erythematosus. Arthritis Rheum 1989;32:141420.

15. Esscher E, Scott JS. Congenital heart block and maternal systemiclupus erythematosus. Br Med J 1979;1:12358.

16. Jackson R, Gulliver M. Neonatal lupus erythematosus progressinginto systemic lupus erythematosus: a 15 year follow-up. Br JDermatol 1979;101:816.

17. Fox RJ, McCuistion CH, Schoch EP Jr. Systemic lupus erythem-atosus: association with previous neonatal lupus erythematosus[abstract]. Arch Dermatol 1979;115:340.

18. Waterworth RF. Systemic lupus erythematosus occurring withcongenital complete heart block. N Z Med J 1980;92:3112.

19. Lanham JG, Walport MJ, Hughes GR. Congenital heart block andfamilial connective tissue disease. J Rheumatol 1983;10:8235.

20. Hubscher O, Carrillo D, Reichlin M. Congenital heart block andsubsequent connective tissue disorder in adolescence. Lupus 1997;6:2834.

21. Brucato A, Gasparini M, Vignati G, Riccobono S, De Juli E,Quinzanini M, et al. Isolated congenital complete heart block:longterm outcome of children and immunogenetic study. J Rheu-matol 1995;22:5413.

LONG-TERM FOLLOWUP IN NEONATAL LUPUS 2383