Distinct Autoimmune Syndromes in Morphea a Review of 245 Adult and Pediatric Cases

STUDY Distinct Autoimmune Syndromes in Morphea A Review of 245 Adult and Pediatric Cases

Justin J. Leitenberger, BS; Rachael L. Cayce, BS; Robert W. Haley, MD; Beverley Adams-Huet, MS; Paul R. Bergstresser, MD; Heidi T. Jacobe, MD

Objective: To determine the prevalence of extracuta- fected less frequently than expected. The prevalence of neous manifestations and autoimmunity in adult and pe- concomitant autoimmunity in the generalized subtype diatric patients with morphea. of morphea was statistically significantly greater than that found in all other subtypes combined (P=.01). Fre- Design: A retrospective review of 245 patients with quency of a family history of autoimmune disease showed morphea. a trend in favor of generalized and mixed subgroups. The linear subtype showed a significant association with neu- Setting: University of Texas Southwestern Medical Cen- rologic manifestations, while general systemic manifes- ter–affiliated institutions. tations were most common in the generalized subtype. Antinuclear antibody positivity was most frequent in Patients: Patients with clinical findings consistent with mixed and generalized subtypes. morphea. Conclusions: High prevalences of concomitant and fa- Main Outcome Measures: Prevalence of concomitant autoimmune diseases, prevalence of familial auto- milial autoimmune disease, systemic manifestations, and immune disease, prevalence of extracutaneous manifes- antinuclear antibody positivity in the generalized and pos- tations, and laboratory evidence of autoimmunity sibly mixed subtypes suggest that these are systemic au- (antinuclear antibody positivity). Secondary outcome toimmune syndromes and not skin-only phenomena. This measures included demographic features. has implications for the management and treatment of patients with morphea. Results: In this group, adults and children were affected nearly equally, and African Americans were af- Arch Dermatol. 2009;145(5):545-550

ORPHEA IS CHARACTER- than-expected prevalence of familial auto- ized by sclerosis of the immune disease.6,7 The same pediatric study skin and, in some found systemic manifestations outside the cases, underlying tis- areaofmorpheain22.4%ofchildren,includ- sue. It is generally ing arthralgias and esophageal dysmotility. thoughtM to be an autoimmune disorder af- Datainadultsarelimited,with1case-control fecting a single organ—the skin. Conse- study5 of 50 adult white women reporting in- quently, current classification schemes di- creased personal and familial autoimmunity, vide morphea into categories based solely but antinuclear antibody (ANA) status and on cutaneous morphology, without refer- systemic manifestations were not reported. ence to systemic disease or autoimmune Many autoantibodies have been reported in Author Affiliations: phenomena. This classification is likely in- patientswithmorphea,includingANA,anti– Departments of Dermatology (Mr Leitenberger, Ms Cayce, complete. single-stranded DNA, antihistone, anti– and Drs Bergstresser and Rather, morphea may be a systemic au- topoisomerase II␣, antiphospholipid, and Jacobe), Internal Medicine toimmune condition.1-8 For example, case rheumatoid factor.9-12 The clinical and prog- (Dr Haley and reports describe morphea coexisting with nostic significance of these autoantibodies Ms Adams-Huet), and Clinical other autoimmune diseases, including sys- remains unclear. Sciences (Drs Haley and temic lupus erythematosus, vitiligo, pri- Taken together, these reports suggest that Bergstresser and Ms Adams-Huet), mary biliary cirrhosis, autoimmune hepa- morphea is an autoimmune disease with a The University of Texas titis, Hashimoto thyroiditis, and myasthenia spectrum of manifestations ranging from 1-5,8,9 Southwestern Medical Center gravis. A retrospective review of 750 only skin to multiple organ involvement. at Dallas. children with morphea revealed a greater- What remains unclear is the prevalence of

(REPRINTED) ARCH DERMATOL/ VOL 145 (NO. 5), MAY 2009 WWW.ARCHDERMATOL.COM 545

©2009 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/02/2021 ter review of the medical records, all patients who were con- Table 1. Main Demographic Characteristics of 245 Patients sidered to have met the clinical criteria for morphea With Morphea (histopathologic results were reviewed when present but were not necessary for inclusion) were included for analysis. Pa- Characteristic Valuea tients who were found on review of the medical record to have Age at onset, y, mean (SD) lichen sclerosus alone or sclerosing skin conditions other than Adults 44.1 (17.9) morphea were excluded (186 of 431). Children 8.7 (3.3) For the 245 patients who met the inclusion criteria, data were Sex, No. entered electronically into an Access database (Microsoft, Red- Overall mond, Washington). Data extraction included demographic in- Female 199 formation, clinical variables, histologic data, laboratory data, Male 46 medical history of rheumatic or other autoimmune disease, fam- F/M ratio 4:1 ily history of rheumatic or other autoimmune diseases in first- Adults 123 and second-degree relatives, and review of systems. Female 103 Morphea subtypes were determined from the medical rec- Male 20 ord when specified or from the physical examination findings F/M ratio 5:1 when not specified in the record. The subtypes investigated were Children 122 plaque, linear (including linear scleroderma, en coup de sa- Female 96 bre, and progressive hemifacial atrophy), generalized, mixed Male 26 (defined by the presence of 2 subtypes in the same patient; how- F/M ratio 3:1 ever, if 1 subtype was generalized, patients were categorized Race/ethnicity, No. (%) as generalized), lichen sclerosus/morphea overlap, guttate, bul- Overall 245 lous, and deep, using established clinical criteria; none of the White 178 (72.7) 14,15 Hispanic/Latino 34 (13.9) patients had bullous morphea. Asian 6 (2.4) Summary statistics were used for reporting demographic, African American 11 (4.5) clinical, and laboratory characteristics. The Fisher exact test Other 16 (6.5) was used to test the association between the 5 subtypes of mor- Adults 123 phea and the presence or absence of autoimmune disease, fam- White 93 (71.5) ily history of autoimmune disease, and systemic findings. Be- Hispanic/Latino 10 (7.7) cause ANAs were tested in relatively few patients, results are Asian 3 (2.3) reported as frequency counts. PϽ.05 was considered statisti- African American 7 (5.4) cally significant. SAS software version 9.1.3 (SAS Institute, Cary, Other 10 (7.7) North Carolina) was used for data analysis. Unavailable 7 (5.4) Children 122 White 85 (69.7) RESULTS Hispanic/Latino 24 (19.7) Asian 3 (2.5) DEMOGRAPHICS African American 4 (3.3) Other 6 (4.9) Table 1 shows the demographic characteristics for the Abbreviations: F, female; M, male. patients. Similar to prior reports, there were 199 fe- a Percentages do not total 100 because of rounding. males (81.2%) and 46 males (18.8%), with an overall female to male ratio of 4.2:1. The racial distribution in all of the patients with mor- autoimmune and systemic disease in morphea and its sub- phea was 72.7% white, 13.9% Hispanic, 4.5% African types and how these findings correlate with the autoanti- American, 2.4% Asian, and 6.5% other (Pacific Islander, body profile. This uncertainty negatively affects patient care Native American, etc). Of the 245 patients, 123 (50.2%) because many patients are untreated or are treated only with had onset of morphea as an adult and 122 (49.8%) de- skin-directed therapy and may require systemic therapy. veloped morphea as a child. The mean age of onset was This study aims to address this gap in knowledge by quan- 8.7 years for children compared with 44.1 years for adults. tifying the prevalence of autoimmunity and systemic manifestations in adults and children with morphea. SUBTYPE DISTRIBUTION

METHODS Table 2 shows subtype distribution among the 245 patients with morphea. The most common subtype over- Approval was obtained from the institutional review board to all was plaque at 35.9% (n=88) and then linear at 25.7% review the medical records of patients with morphea seen from (n=63). Deep and guttate subtypes were seen in less than June 1, 2001, to June 1, 2007, at University of Texas South- 1.2% (3 each) of all patients and are not included in sub- western Medical Center at Dallas–affiliated institutions (Uni- sequent analyses. Subtype distribution differed between versity of Texas Southwestern Medical Center at Dallas uni- adults and children. versity-based faculty practice, Texas Scottish Rite Hospital, Children’s Medical Center in Dallas, and Parkland Memorial Hospital). A total of 431 potential adult and pediatric patients Adults were identified on the basis of the International Classification of Diseases, Ninth Revision diagnosis code for morphea/ The most common subtypes were plaque-type morphea localized scleroderma and lichen sclerosus (code 701.0).13 Af- in 43.9% (54 of 123) of adult patients and generalized

(REPRINTED) ARCH DERMATOL/ VOL 145 (NO. 5), MAY 2009 WWW.ARCHDERMATOL.COM 546

Morphea Subtype, No. (%) of Patients

Group Overall Plaque Linear Generalized Deep Guttate Lichen Sclerosus/Morphea Mixed Overall 245 88 (35.9) 63 (25.7) 37 (15.1) 3 (1.2) 3 (1.2) 18 (7.3) 33 (13.5) Adults 123 54 (43.9) 12 (9.8) 29 (23.6) 2 (1.6) 3 (2.4) 18 (14.6) 5 (4.1) Children 122 34 (27.9) 51 (41.8) 8 (6.6) 1 (0.8) 0 0 28 (23.0)

a Percentages do not total 100 because of rounding.

in 23.6% (29 of 123) of adult patients. The mixed subtype was seen in 4.1% (5 of 123) of adult patients. Table 3. Rheumatic Disease and Other Autoimmune Disorders in Patients With Morphea and Their Relatives Children No. (%) of Patients

The most common subtype was linear in 41.8% (51 of Concomitant Familial 122) of pediatric patients, followed by plaque and mixed Disorder Disorder subtypes at 27.9% (34 of 122) and 23.0% (28 of 122), Overall (N=245) 43 (17.6) 40 (16.3) respectively. Among the 33 adults and children with mixed Adults (n=123) 37 (30.1)a,b 13 (10.6)c Children (n=122) 6 (4.9)a,b 29 (23.8)c subtypes, 84.8% (28 of 33) had plaque and linear le- b,c sions develop over time; the remainder had generalized Disease Autoimmune disease and linear subtypes of morphea. Psoriasis 5 (11.6) 7 (17.5) Vitiligo 3 (7.0) 2 (5.0) CONCOMITANT RHEUMATIC DISEASE Alopecia areata 5 (11.6) . . . AND OTHER AUTOIMMUNE DISORDERS Inflammatory bowel disease 2 (4.7) 1 (2.5) Type 1 diabetes mellitus 3 (7.0) 1 (2.5) Table 3 shows the details of rheumatic and other auto- Autoimmune thyroiditis 1 (2.3) 1 (2.5) Meniere disease 1 (2.3) . . . immune disorders present in patients with morphea or their Celiac disease 1 (2.3) . . . families. Among the 245 patients, 43 (17.6%) reported con- Multiple sclerosis 1 (2.3) 5 (12.5) comitant rheumatic or other autoimmune disorder. Con- Rheumatic disease comitant rheumatic or other autoimmune disorder was sig- Systemic lupus erythematosus 3 (7.0) 5 (12.5) nificantly more prevalent among adults (30.1% [37 of 123]) Spondyloarthropathy 3 (7.0) 1 (2.5) than among children (4.9% [6 of 122]) (P Յ.001). When Rheumatoid arthritis 3 (7.0) 7 (17.5) Sjögren syndrome 2 (4.7) 1 (2.5) analyzed by subtype, the frequency of concomitant auto- Antiphospholipid syndrome 1 (2.3) . . . immune disorders among adult and pediatric patients with Still disease 1 (2.3) . . . generalized morphea was 45.9% (17 of 37), which was sig- Mixed connective tissue disease 2 (4.7) . . . nificantly greater (P=.01) than the prevalence in the other Morphea . . . 5 (12.5) subtypes combined (9.6%). Scleroderma . . . 3 (7.5) Subtype d FAMILIAL RHEUMATIC DISEASE AND OTHER Generalized 17 (45.9) 8 (21.6) Linear 3 (4.8) 10 (15.9) AUTOIMMUNE DISORDERS Plaque 10 (11.4) 13 (14.8) Lichen sclerosus/morphea 4 (22.2) 1 (5.6) Among all patients, 16.3% (40 of 245) reported a family Mixed 3 (9.1) 8 (24.2) history of autoimmune disorders. Significantly more children (23.8%) than adults (10.6%) had a family history Abbreviation: ellipses, not calculated. a PϽ.001 for difference between children and adults with concomitant of a rheumatic or other autoimmune disorder in a first- autoimmune/rheumatic disease. or second-degree relative (PϽ.001). In adults, the gen- b Five patients had 2 concomitant diseases, for a total of 37. eralized subtype had the highest frequency of familial dis- c PϽ.001 for difference between children and adults with a family history order (21.6%). In children, those with mixed (24.2%) and of autoimmune disease. d P=.01 for association between generalized morphea and autoimmune generalized (21.6%) subtypes had the highest fre- disease vs other subtypes. There was no statistically significant association quency of family history of autoimmunity. Despite trends of family history of autoimmune and morphea subtypes, but there was a in favor of an association of generalized and mixed sub- trend in favor of an association in generalized and mixed subtypes. types with familial autoimmunity, this was not statistically significant (P=.29). Of note, 5 children (2.0% of all side the areas affected by morphea were common in pa- patients) reported a family history of morphea in a first- tients with generalized morphea, with the highest or second-degree relative. frequency compared with the other subtypes of dysphagia (14.3% [5 of 35], PϾ.05), dyspnea (20.0% [7 of 35], EXTRACUTANEOUS MANIFESTATIONS P=.03), and vascular complaints (Raynaud phenomenon) (31.4% [11 of 35], P=.03). Patients with the mixed Table 4 lists the extracutaneous manifestations of mor- subtype had a high level of musculoskeletal manifesta- phea experienced by the study patients. Symptoms out- tions (6.1%-36.4% related to the distribution of the le-

(REPRINTED) ARCH DERMATOL/ VOL 145 (NO. 5), MAY 2009 WWW.ARCHDERMATOL.COM 547

Morphea Subtype, No. (%) of Patientsa

Overall Plaque Linear Generalized Mixed Manifestation (n=246) (n=86) (n=86) (n=35) (n=33) Constitutional Fatigue 29 (11.8) 13 (15.1) 7 (8.1) 5 (14.3) 1 (3.0) Gastrointestinal Dysphagia 24 (9.8) 8 (9.3) 5 (5.8) 5 (14.3) 4 (12.1) Musculoskeletal Arthralgia 48 (19.5) 17 (19.8) 8 (9.3) 9 (25.7) 8 (24.2) Joint swelling 21 (8.5) 9 (10.5) 1 (1.2) 5 (14.3) 2 (6.1) Myalgia 32 (13.0) 12 (14.0) 6 (7.0) 5 (14.3) 4 (12.1) Limb contracture 19 (7.7) 2 (2.3) 6 (7.0) 4 (11.4) 6 (18.2) Limited range of motion 48 (19.5) 7 (8.1) 16 (18.6) 10 (28.6) 12 (36.4) Neurologic Vision changes 12 (4.9) 3 (3.5) 7 (8.1)b . . . 1 (3.0) Epileptic seizures 6 (2.4) . . . 4 (4.6)b 1 (2.9) 1 (3.0) Headaches 33 (13.4) 13 (15.1) 14 (16.3)b 3 (8.6) 2 (6.1) Vascular malformations ...... 1 (1.2) ...... Calcifications 1 (0.4) . . . 1 (1.2) ...... Respiratory Dyspnea 17 (6.9) 1 (1.2) 4 (4.6) 7 (20.0)c 2 (6.1) Vascular Cold fingers 25 (10.2) 8 (9.3) 5 (5.8) 7 (20.0) 4 (12.1) Raynaud phenomenon 5 (2.0) 1 (1.2) 1 (1.2) 3 (8.6) . . . Vasculitis ...... 1 (2.9) . . .

Abbreviation: ellipses, not calculated. a Six patients with lichen schlerosus morphea overlap were not included owing to small numbers. b Statistically significant association between linear morphea and visual changes, seizures, and headaches (P=.009, P=.04, and P=.03, respectively). c Association with generalized subtype statistically significant (P=.03 for both dyspnea and vascular manifestations).

Table 5. Autoantibodies in Patients With Morphea Subtypesa

Morphea Subtype, No. (%) of Patients

Autoantibody Overall Plaque Linear Generalized Mixed ANAb,c 35/89 (39.3) 8/25 (32.0) 6/21 (28.6) 11/19 (57.9) 7/13 (53.8) Anti–Scl-70 1/39 (2.6) 0/12 0/9 1/8 (12.5) 0/7 Anti-dsDNA 3/19 (15.8) 0/4 1/3 (33.3) 1/6 (16.7) 1/3 (33.3) Antiphospholipidd 3/8 (37.5) 2/3 (66.7) 0/1 0/2 . . . Anti-centromere 1/11 (9.1) 0/4 0/3 1/2 (50) . . . Anti-ssDNA 1/26 (3.8) 0/8 0/6 1/6 (16.7) 0/2 Anti-snRNP 1/21 (4.8) 0/6 0/5 1/5 (20.0) 0/2

Abbreviations: ANA, antinuclear antibody; anti-dsDNA, anti–double-stranded DNA; anti-ssDNA, anti–single-stranded DNA; anti-snRNP, anti–small nuclear ribonucleoprotein particle; ellipses, not calculated. a Owing to the relatively low numbers of patients sampled, statistical analysis was not performed and data are presented as frequency counts. b One of 3 patients with systemic lupus erythematosus reported a positive ANA titer (this patient also had generalized morphea). All patients with systemic lupus erythematosus either tested negative for autoantibodies or did not have the test performed. c Titer of 1:160 or higher. d No patient tested had a diagnosis of antiphospholipid syndrome.

sions). The linear subtype was associated with neuro- [7 of 13], respectively). Adults had a higher frequency logic (31.4% [27 of 86]) and ophthalmologic (8.1% [7 of ANA positivity with 52.8% (19 of 36) testing positive of 86]) complications related to the affected site on the compared with 30.2% (16 of 53) of children. The most face or scalp (both P=.05) (Table 4). common ANA patterns observed were speckled and nucleolar. ANA AND OTHER ANTIBODIES In contrast to prior reports,10-12 only1of26of patients tested had anti–single-stranded DNA antibod- Antinuclear antibodies were tested in 36.3% (89 of 245) ies (generalized morphea). One of 39 patients (with of patients and were positive (titer Ն1:160) in 39.3% (35 generalized morphea) was positive for anti–Scl-70 of 89) of those tested (Table 5). The percentage of pa- antibodies. This patient did not develop any signs or tients with ANA positivity was highest in the general- symptoms of scleroderma during a 9-year follow-up ized and mixed subtypes (57.9% [11 of 19] and 53.8% period. Three of 19 patients tested (14%) had anti–

(REPRINTED) ARCH DERMATOL/ VOL 145 (NO. 5), MAY 2009 WWW.ARCHDERMATOL.COM 548

©2009 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/02/2021 double-stranded DNA antibodies: 1 each for linear, the pediatric patients). Thus, the frequency of a positive generalized, and mixed subtypes (prior reports10-12 link family history is likely underestimated in adults. Of note, the presence of anti–double-stranded DNA to the gen- although not statistically significant, the generalized sub- eralized subtype). None of these patients developed type represented the highest frequency of familial dis- any signs or symptoms of systemic lupus erythemato- ease in adults and the second highest in children (lack sus after 1 year of follow-up. of statistical significance may be the result of relatively low numbers). Zulian et al6 reported similar findings in a pediatric cohort with a statistically significant associa- COMMENT tion of familial autoimmunity in the generalized subtype (23.5% vs 12.5%, 12.3%, and 8.8% in patients with This study represents the largest collection of adults and deep, linear, and plaque-type morphea, respectively). Fa- children with morphea to address the prevalence of au- milial autoimmune diseases also occurred more fre- toimmunity and systemic disease with stratification by quently in the mixed subtype (both adults and chil- morphea subtype. As in prior reports,7,15 there was a fe- dren), which was not reported by Zulian et al. Thus, male predominance and age-dependent difference in sub- familial autoimmunity is most prevalent in generalized type distributions. Of note, morphea occurred nearly and mixed subtypes. equally in adults and children (with a prior report15 cit- Similar disorders to those reported in patients with ing the disease is favored in childhood by 2:1). Al- morphea were reported in their family members. Rheu- though this is not a population-based study, it likely rep- matoid arthritis, systemic lupus erythematosus, and resents a fair estimation of distribution between adults psoriasis were observed with the highest frequency in first- and children because patients were enrolled from all uni- and second-degree relatives of patients with morphea over- versity-affiliated institutions, including 2 children’s medi- all. These findings are similar to those reported by cal centers, and included patients seen by all specialists. Zulian et al.6 Taken together, the increased frequency of Prior reports may be biased in favor of pediatric patients personal and familial autoimmunity in the generalized who present with linear lesions more frequently and are subtype may indicate a common susceptibility locus for subsequently diagnosed and brought to medical atten- this group of disorders, as was recently described in viti- tion more frequently. The racial distribution of mor- ligo and its association with NALP1.20 phea has not been widely addressed. In this cohort, mor- We found that 1% (4 families) of patients with morphea was less frequent in African Americans. The phea had first- or second-degree relatives with mor- percentage affected in this cohort (4.5%) is lower than phea, representing a higher risk than present in the gen- expected based on the racial distribution of metropoli- eral population (compared with an estimated prevalence tan Dallas, Texas (20.9% African American, 2005 esti- reported by Peterson et al21 of 0.2% [odds ratio, 6.8; 95% mate of the US Census Bureau16) and the racial distribu- confidence interval, 1.15-40.56; P=.01]). Multicase fami- tion of patients seen at the University of Texas lies with morphea have been described independently in Southwestern Medical Center at Dallas–affiliated insti- the literature22,23 and, as in our study, were in a non- tutions. Population-based studies are needed to fully ad- Mendelian pattern suggestive of a multifactorial, poly- dress age and racial distribution in morphea. genic inheritance. There was no family history of sclero- Overall, 17.6% of patients had a concomitant rheu- derma in this cohort, although this has been reported in matic or other autoimmune disorder, an occurrence 4-fold the pediatric group described by Zulian et al.6,7 higher than that in the general population, including all The systemic symptoms and signs observed in the pa- races and socioeconomic strata.17,18 But when analyzed tients in this study are similar to those in prior reports in by subtype, generalized morphea had a statistically sig- terms of organ systems of involvement.6,7,24 A novel obser- nificant association with autoimmune disease, with 45.9% vation, however, is that manifestations occurring outside of patients with generalized morphea affected, represent- the area affected by morphea (eg, dysphagia, joint pain, and ing 12 times the risk of the general population. Chil- Raynaud phenomenon) are most common in patients with dren with morphea were relatively spared of concomi- the generalized subtype, while patients with linear disease tant autoimmune disease (only 4.9% vs 30.1% of adults). often have neurologic and ophthalmologic manifestations It is unknown whether children with morphea will de- correlating with the area of morphea involvement.6,7 Prior velop autoimmune disorders at an increased rate as they reports24 describe morphea as a relatively painless disor- reach adulthood. der; this cohort, however, reported pain in areas affected Very little is known whether specific disorders aggre- by morphea, particularly in the generalized subgroup. This gate with morphea. In this cohort, several disorders oc- underscores the need to address pain as a potential symp- curred with greater frequency than expected compared with tom in patients with morphea. published population-based prevalence estimates, includ- An ANA positivity rate of 39% (of patients tested) is ing psoriasis (1.5- to 4.5-fold increase), systemic lupus ery- comparable with a prior study.11 Patients with the gen- thematosus (58-fold increase), multiple sclerosis (7- to eralized and mixed subtypes had the highest frequency 8-fold increase), and vitiligo (3.5-fold increase).17-19 (58% and 54%, respectively). In contrast to prior re- The statistically significant difference in family his- ports11,12 linking anti–single-stranded DNA to linear and tories of autoimmune disorders between children (24%) generalized morphea (50%), only 1 of 26 patients tested and adults (11%) with all subtypes of morphea is likely had a positive result. Prior reports10-12 indicate no clear confounded by differences in data acquisition at the study association between subtype and ANA pattern; this co- sites (family history was collected more thoroughly in hort, however, demonstrated a trend toward segrega-

(REPRINTED) ARCH DERMATOL/ VOL 145 (NO. 5), MAY 2009 WWW.ARCHDERMATOL.COM 549

©2009 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/02/2021 tion by subtype (generalized with a homogeneous pat- Cayce) and The Dermatology Foundation (Dr Jacobe). tern, linear and plaque with a speckled pattern, and mixed Disclaimer: The sponsors had no role in the design and with a nucleolar pattern). This may indicate that mor- conduct of the study; in the collection, analysis, and in- phea subtypes are distinct phenomena with distinct an- terpretation of data; or in the preparation, review, or ap- tigenic targets. Because ANAs were not tested in all pa- proval of the manuscript. tients and may have been tested only in more severely affected patients, conclusions are limited. Limitations of the study are those inherent in a ret- REFERENCES rospective review, including ascertainment bias in ac- 1. Gonza´lez-Lo´pez MA, Drake M, Gonzalez-Vela MC, Armesto S, Llaca HF, Val- quisition of medical and family histories and laboratory Bernal JF. Generalized morphea and primary biliary cirrhosis coexisting in a male testing, as well as recall bias in reporting of information patient. J Dermatol. 2006;33(10):709-713. by patients and families. Subtype classification was de- 2. Khalifa M, Ben Jazia E, Hachfi W, Sriha B, Bahri F, Letaief A. Autoimmune hepa- pendent on the treating physician and may not reflect stan- titis and morphea: a rare association. Gastroenterol Clin Biol. 2006;30(6-7): dard criteria. Conclusions regarding autoantibodies are 917-918. 3. Dervis E, Acbay O, Barut G, Karaoglu A, Ersoy L. Association of vitiligo, mor- limited by the relatively low numbers of patients who were phea, and Hashimoto’s thyroiditis. Int J Dermatol. 2004;43(3):236-237. tested. Prospective longitudinal studies examining these 4. Parra V, Driban N, Bassotti A. Localized morphea and myasthenia gravis. JAm findings in greater detail are indicated. Acad Dermatol. 2008;49(5):1. In conclusion, there is a strong association between 5. Harrington CI, Dunsmore IR. An investigation into the incidence of autoimmune disorders in patients with localized morphoea. Br J Dermatol. 1989; generalized morphea and, to a lesser extent, mixed mor- 120(5):645-648. phea with markers of a systemic rather than skin- 6. Zulian F, Athreya BH, Laxer R, et al; Juvenile Scleroderma Working Group of the limited autoimmune process (high frequency of con- Pediatric Rheumatology European Society (PRES). Juvenile localized sclero- comitant autoimmune disease, systemic findings, and derma: clinical and epidemiological features in 750 children: an international study. positive ANAs), and plaque-type morphea may repre- Rheumatology (Oxford). 2006;45(5):614-620. 7. Zulian F, Vallongo C, Woo P, et al; Juvenile Scleroderma Working Group of the sent a single-organ (skin only) autoimmune process. This Pediatric Rheumatology European Society (PRES). Localized scleroderma in child- has implications for the management of these patients. hood is not just a skin disease. Arthritis Rheum. 2005;52(9):2873-2881. Specifically, results of this study suggest that patients with 8. Bonifati C, Impara G, Morrone A, Pietrangeli A, Carducci M. Simultaneous oc- generalized and mixed subtypes should be aggressively currence of linear scleroderma and homolateral segmental vitiligo. J Eur Acad Dermatol Venereol. 2006;20(1):63-65. monitored for the presence of autoimmune disorders and 9. Majeed M, Al-Mayouf SM, Al-Sabban E, Bahabri S. Coexistent linear sclero- systemic manifestations and treated with systemic im- derma and juvenile systemic lupus erythematosus. Pediatr Dermatol. 2000; munosuppressive agents when indicated. These find- 17(6):456-459. ings may warrant a reexamination of the current classi- 10. Rosenberg AM, Uziel Y, Krafchik BR, et al. Antinuclear antibodies in children with fication of morphea from the perspective of inclusion of localized scleroderma. J Rheumatol. 1995;22(12):2337-2343. 11. Takehara K, Moroi Y, Nakabayashi Y, Ishibashi Y. Antinuclear antibodies in lo- extracutaneous disorders. calized scleroderma. Arthritis Rheum. 1983;26(5):612-616. 12. Falanga V, Medsger TA Jr, Reichlin M. Antinuclear and anti-single-stranded DNA Accepted for Publication: September 4, 2008. antibodies in morphea and generalized morphea. Arch Dermatol. 1987;123 Correspondence: Heidi T. Jacobe, MD, JA5.521, Depart- (3):350-353. 13. World Health Organization. International Classification of Diseases, Ninth Revi- ment of Dermatology, UT Southwestern Medical Cen- sion (ICD-9). Geneva, Switzerland: World Health Organization; 1977. ter, 5323 Harry Hines Blvd, Dallas, TX 75390-9069. 14. Okun MR, Edelstein LM, Fisher BK. Gross and Microscopic Pathology of the Skin. ([email protected]). 2nd ed. Canton, MA: Dermatopathology Foundation Press; 1988. Author Contributions: Mr Leitenberger, Ms Cayce, and 15. Vierra E, Cunningham BB. Morphea and localized scleroderma in children. Se- Drs Bergstresser and Jacobe had full access to all of the min Cutan Med Surg. 1999;18(3):210-225. 16. Census Bureau, Dallas Regional Office. US Census Bureau, 2005-2007 Ameri- data in the study and take responsibility for the integ- can Community Survey. http://www.census.gov. Accessed February 23, 2009. rity of the data and the accuracy of the data analysis. Study 17. Spritz RA. The genetics of generalized vitiligo and associated autoimmune diseases. concept and design: Leitenberger, Haley, Bergstresser, and Pigment Cell Res. 2007;20(4):271-278. Jacobe. Acquisition of data: Leitenberger. Analysis and in- 18. Lawrence RC, Helmick CG, Arnett FC, et al. Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States. Arthritis Rheum. terpretation of data: Leitenberger, Cayce, Haley, Adams- 1998;41(5):778-799. Huet, Bergstresser, and Jacobe. Drafting of the manu- 19. Hoffman RW, Greidinger EL. Mixed connective tissue disease. Curr Opin Rheumatol. script: Leitenberger, Cayce, and Jacobe. Critical revision 2000;12(5):386-390. of the manuscript for important intellectual content: 20. Jin Y, Mailloux CM, Gowan K, et al. NALP1 in vitiligo-associated multiple auto- Leitenberger, Cayce, Haley, Adams-Huet, and Jacobe. Sta- immune disease. N Engl J Med. 2007;356(12):1216-1225. 21. Peterson LS, Nelson AM, Su WP, Mason T, O’Fallon WM, Gabriel SE. The epi- tistical analysis: Haley and Jacobe. Obtained funding: Ad- demiology of morphea (localized schleroderma) in Olmsted County, 1960-1993. ams-Huet. Administrative, technical, and material sup- J Rheumatol. 1997;24(1):73-80. port: Leitenberger, Cayce, Bergstresser, and Jacobe. Study 22. Wadud MA, Bose BK, Al NT. Familial localised scleroderma from Bangladesh: supervision: Haley and Jacobe. two case reports. Bangladesh Med Res Counc Bull. 1989;15(1):15-19. 23. Rees RB, Bennett J. Localized scleroderma in father and daughter. AMA Arch Financial Disclosure: None reported. Derm Syphilol. 1953;68(3):360. Funding/Support: This study was supported by The Doris 24. Bolognia JL, Jorizzo JL, Rapini RP, eds. Dermatology. 2nd ed. London, En- Duke Charitable Foundation (Mr Leitenberger and Ms gland: Mosby Elsevier; 2008:1469-1484.

(REPRINTED) ARCH DERMATOL/ VOL 145 (NO. 5), MAY 2009 WWW.ARCHDERMATOL.COM 550