Morphea in adults and children cohort II: Patients with morphea experience delay in diagnosis and large variation in treatment Weilan Johnson, MD, a and Heidi Jacobe, MD, MSCS b Dallas, Texas Background: Little is known about the diagnosis, evaluation, and therapy of morphea (localized scleroderma) in the United States. Delays in diagnosis and initiation of appropriate therapy, if present, may negatively affect patient care. Further, this gap in knowledge hinders planning for clinical trials and therapeutic guidelines. The morphea in adults and children (MAC) cohort is designed to address this gap. Objective: We sought to determine the duration between morphea onset and diagnosis, specialty of the diagnosing provider, and initial evaluation and therapy in the MAC cohort. Methods: This was a cross-sectional survey of the inception cohort of the MAC study. Results: In all, 63% (n = 141 of 224) of patients were given the diagnosis more than 6 months after onset. Dermatologists diagnosed and treated the majority of patients (83.5%, n = 187). Rheumatologists diagnosed and treated the more severe forms of morphea (linear and generalized). The most commonly prescribed therapy was topical corticosteroids (63%). Dermatologists predominantly prescribed topical treatments or phototherapy (P \.0001, P = .0018, respectively), even to patients with linear and generalized morphea. In contrast, rheumatologists predominantly prescribed systemic immunosuppressives and physical therapy (P \ .0001, P = .0021, respectively). Limitations: Referral bias and recall bias may affect patterns of evaluation/therapy and ascertainment of disease duration before diagnosis. Conclusions: Patients with morphea experience delay in diagnosis, which likely impacts outcome. Therapeutic decision making is largely determined by the specialty of the provider rather than disease characteristics and many treatments with little or no proven efficacy are used, whereas others with proven efficacy are underused. This underscores the need for a collaborative, multispecialty approach in designing therapeutic trials and guidelines. ( J Am Acad Dermatol 2012;67:881-9.) Key words: drug; evaluation; localized scleroderma; morphea; sclerosing skin conditions; specialty; therapy; treatment. M orphea, or localized scleroderma, is char- acterized by inflammation and sclerosis of the dermis and sometimes subcutaneous tissue, and may produce significant morbidity and systemic disease (arthritis). 1,2 Little is known about the time to initial diagnosis and types of evaluation and treatment prescribed for patients with morphea, particularly among different specialties and morphea subtypes. A recent study in the United Kingdom found significant delay in the diagnosis and initiation From the Departments of Dermatology at Texas Health Presbyte- rian Hospital Dallas a and University of Texas Southwestern Medical Center. b Both authors contributed to this manuscript equally. Supported by grant No. NIH K23AR056303-4 from the National Institutes of Health. Conflicts of interest: None declared. Accepted for publication January 8, 2012. Reprints not available from the authors. Correspondence to: Heidi Jacobe, MD, MSCS, Department of Dermatology, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd, Dallas, TX 75390-9069. E-mail: [email protected]. Published online March 1, 2012. 0190-9622/$36.00 Ó 2012 by the American Academy of Dermatology, Inc. doi:10.1016/j.jaad.2012.01.011 881
Transcript
Morphea in adults and children cohort II: Patients withmorphea experience delay in diagnosis and large
variation in treatment
Weilan Johnson, MD,a and Heidi Jacobe, MD, MSCSb
Dallas, Texas
From
ria
M
Both
Supp
In
Conf
Acce
Repr
Background: Little is known about the diagnosis, evaluation, and therapy of morphea (localizedscleroderma) in the United States. Delays in diagnosis and initiation of appropriate therapy, if present, maynegatively affect patient care. Further, this gap in knowledge hinders planning for clinical trials andtherapeutic guidelines. The morphea in adults and children (MAC) cohort is designed to address this gap.
Objective: We sought to determine the duration between morphea onset and diagnosis, specialty of thediagnosing provider, and initial evaluation and therapy in the MAC cohort.
Methods: This was a cross-sectional survey of the inception cohort of the MAC study.
Results: In all, 63% (n = 141 of 224) of patients were given the diagnosis more than 6 months after onset.Dermatologists diagnosed and treated the majority of patients (83.5%, n = 187). Rheumatologists diagnosedand treated the more severe forms of morphea (linear and generalized). The most commonly prescribedtherapy was topical corticosteroids (63%). Dermatologists predominantly prescribed topical treatments orphototherapy (P\.0001, P = .0018, respectively), even to patients with linear and generalized morphea. Incontrast, rheumatologists predominantly prescribed systemic immunosuppressives and physical therapy(P\ .0001, P = .0021, respectively).
Limitations: Referral bias and recall bias may affect patterns of evaluation/therapy and ascertainment ofdisease duration before diagnosis.
Conclusions: Patients with morphea experience delay in diagnosis, which likely impacts outcome.Therapeutic decision making is largely determined by the specialty of the provider rather than diseasecharacteristics and many treatments with little or no proven efficacy are used, whereas others with provenefficacy are underused. This underscores the need for a collaborative, multispecialty approach in designingtherapeutic trials and guidelines. ( J Am Acad Dermatol 2012;67:881-9.)
Morphea, or localized scleroderma, is char-acterized by inflammation and sclerosis ofthe dermis and sometimes subcutaneous
tissue, and may produce significant morbidity andsystemic disease (arthritis).1,2 Little is known about
the Departments of Dermatology at Texas Health Presbyte-
n Hospital Dallasa and University of Texas Southwestern
edical Center.b
authors contributed to this manuscript equally.
orted by grant No. NIH K23AR056303-4 from the National
stitutes of Health.
licts of interest: None declared.
pted for publication January 8, 2012.
ints not available from the authors.
the time to initial diagnosis and types of evaluationand treatment prescribed for patients with morphea,particularly among different specialties andmorpheasubtypes. A recent study in the United Kingdomfound significant delay in the diagnosis and initiation
Correspondence to: Heidi Jacobe, MD, MSCS, Department of
Dermatology, University of Texas Southwestern Medical
Center, 5323 Harry Hines Blvd, Dallas, TX 75390-9069. E-mail:
of treatment within their health care system.3 Thestatus of patients in the United States, however, ispoorly described. The situation is similar for thetypes of therapy prescribed for patients with mor-phea.3 A recent cross-sectional survey of pediatricrheumatologists demonstrated most used a combi-nation of methotrexate4 and systemic corticosteroids
CAPSULE SUMMARY
d Little is known about how providers indifferent specialties evaluate and treatpatients with morphea.
d Our results indicate that there is delay indiagnosis of morphea, which maynegatively impact patient care. Theevaluation and treatment of morpheadiffered depending on the specialty ofthe treating physician, and less ondisease subtype. Also, treatments withlittle evidence for efficacy werefrequently used, whereas others withproven efficacy were underused.
d This underscores the need for thedevelopment of multidisciplinarytherapeutic trials and guidelines.
with a limited role for topicalmedications as adjunctivetherapy. The approach ofother providers who care forpatients withmorphea, partic-ularly dermatologists, hasnot been studied. This is aproblem because understan-ding patterns of diagnosis,evaluation, and therapy arecrucial to identify relevantstakeholders for developmentof therapeutic guidelinesand prioritize treatments forinclusion in clinical trials.
This study, referred to asmorphea in adults and chil-dren (MAC) cohort, wasdesigned to prospectivelyexamine demographic, clini-cal (including disease onset,time of diagnosis, evaluation,
and therapy before study entry), immunogenetic,and autoimmune features of adults and children withmorphea in a comprehensive fashion.
As part of the MAC study, we determined timeof morphea onset versus initial diagnosis, specialtyof diagnosing physician, types of evaluation(eg, imaging, skin biopsy) performed, and thera-pies prescribed by the diagnosing physician. Inaddition to patient report, medical records wereobtained for verification. Our overarching hypoth-esis was that patients with morphea have signifi-cant delay in diagnosis and considerable variationexists in therapeutic decision making based on thespecialty of the provider, independent of diseasecharacteristics.
METHODSPatients
The institutional review boardeapproved MACcohort contains 224 patients with morphea enrolledbetween September 2007 and July 2010 who metcriteria for inclusion in this study. The registry, aprospective cohort study, enrolls patients age 3 yearsor older with clinical and/or histologic features ofmorphea. The cohort was designed to capture prev-alent and incident cases of morphea. Patients are
recruited from within the University of TexasSouthwestern Medical Center at Dallas system en-compassing two dedicated pediatric care facilities, acounty hospital, a faculty-based practice, and otheracademic facilities. In addition, patients are routinelyenrolled through regional referrals from private prac-titioners (dermatologists and rheumatologists, both
pediatric and adult). This rep-resents a conscious effort toenroll patients of varied dis-ease severity, subtypes, andsocioeconomic background.
After patients (or guard-ians) sign consent, all dataare abstracted using a com-prehensive clinical reportform designed before thestudy, and medical recordsare obtained and reviewedfor confirmation of patient-reported findings includingspecialty of diagnosing pro-vider, evaluation, and therapy.All patients are examined bythe one examiner with signif-icant expertise in morphea(H. J.) who assigns subtypeand clinical scores (Table I).Phlebotomy, saliva, or both
are obtained for immunologic and immunogeneticstudies.Allpatients enrolled in theMACcohortwith thefollowing data available were included in this study:date/age of morphea onset, date of diagnosis, infor-mation about diagnosing physician sufficient to obtainrecords, morphea subtype, and medications and eval-uations prescribed.
VariablesThe comprehensive MAC database contains the
following domains: demographic, clinical, medicalhistory (including interval between patient-/parent-determined disease onset and diagnosis, priormorphea evaluation, and therapy), immunologic,immunogenetic, Dermatology Life Quality Index,physician-based determinants of disease severity(modified Rodnan skin score and LocalizedScleroderma Severity Index/Localized SclerodermaDamage Index), and functional status (presence/absence of limited range of motion, contracture, limblength discrepancy). These variables were assessedby direct physician interview/examination and useof validated questionnaires. Immunologic and im-munogenetic parameters will not be discussed fur-ther in this article, as they were not included infurther analysis.
Table I. Proposed classification of morpheasubtypes
Circumscribed
Superficial Single or multiple oval/round lesionslimited to epidermis and dermis
Deep Single or multiple oval/round lesionsinvolving subcutaneous tissue,fascia, or muscle
LinearTrunk/limbs Plaques in linear distribution or
continuous linear lesions; mayinvolve underlying tissue
Head En coup de sabre, progressive facialhemiatrophy, linear lesions of face(may involve underlying bone)
Generalized1. Coalescentplaque
$ 4 Plaques in at least 2 of 7anatomic sites (head-neck, right/left upper extremity, right/leftlower extremity, front/back oftrunk)
2. Pansclerotic Circumferential involvement ofmajority of body surface area(sparing fingertips and toes),affecting skin, subcutaneoustissue, muscle, or bone; nointernal organ involvement
Mixed Combination of any above subtype:example: linear-circumscribed
Data from Laxer and Zulian.18
Table II. Demographic characteristics of 224patients with morphea
Characteristic Value
Total no. of patients 224Patients age # 17 y at enrollment 95Patients age $ 18 y at enrollment 129Overall age at onset, yMean 30.3 6 21.6Median 25Range 3-78
GenderF, % 84.0F:M ratio 5.26:1F:M ratio, age # 17 y 3.07:1F:M ratio, age $ 18 y 7:1
*Remaining 4 patients had indeterminate or mixed subtype.
J AM ACAD DERMATOL
VOLUME 67, NUMBER 5Johnson and Jacobe 883
Subtypes of morphea were determined usingestablished clinical criteria and assigned by oneinvestigator (H. J.) with significant experience inmorpea.1 Subtype classifications used in this inves-tigation are presented in Table I.
Statistical analysisDescriptive statistics including frequency counts
and tables were used for demographics, specialty ofdiagnosing physician, evaluation modalities, andprescribed treatments.
The x2 tests (or Fisher exact test if there were\5responses in a cell) were used to examine differencesin proportions and the strength of association be-tweenpediatric and adult onset, disease subtypes, andphysician specialties. A P value of less than .05 was setas the cutoff for statistical significance. All analyseswere performed using software (GraphPad Prism,GraphPad Software, La Jolla, CA).
RESULTSDemographics
Data were available for 224 patients in theMorphea Registry and DNA Repository, 95 children
and 129 adults. Differences in distribution of sub-types based on age of onset were present as de-scribed in other studies.5 Details of demographicand subtype characteristics are available in Tables IIand III.
Time from patient-reported disease onset todiagnosis
Details of time from onset to diagnosis are in Fig 1.Pediatric-onset morphea was diagnosed earlier thanadult-onset morphea: 45.8% (n = 44) of pediatricpatients were given the diagnosis within the first 6months whereas 30.5% (n = 39) of adults were giventhe diagnosis in the same time period (P = .05).Patients with linear subtype were usually given thediagnosis within the first 6 months (42.2%, n = 43),whereas 33.8% (n = 26) of patients with generalizedsubtype and 33.3% (n = 8) of patients with plaquesubtype were given the diagnosis in the same timeperiod (P = .341). A total of 25.5% (n = 57) of patients
Fig 1. Morphea. Time from onset to diagnosis by age group (A) and subtype (B).
J AM ACAD DERMATOL
NOVEMBER 2012884 Johnson and Jacobe
overall were given the diagnosis more than 2 yearsafter disease onset. Patients reported that they wereoften initially evaluated by primary care providers(pediatric and adult), who did not render a diagnosisor rendered a diagnosis other than morphea.
Specialty of diagnosing physicianIn all, 83.5% (n = 187) of patients were given the
diagnosis by a dermatologist. Dermatologists diag-nosed and treated the vast majority of plaque (95.8%,n = 23), linear (82.0%, n = 82), and generalized
(87.7%, n = 71) morphea. Rheumatologists diag-nosed linear, generalized, and eosinophilic fasciitisbut not plaque morphea.
Prescribed treatments before enrollmentTreatment choice based on subtype and age
of the patient. There were a total of 531 pre-scribed treatments (patients often received[1 ther-apy), 218 (41.1%) in children and 313 (59.0%) inadults (Fig 2). Overall, the most common treatmentsin descending order were topical corticosteroids
Fig 2. Morphea. Prescribed treatments, by specialty of prescribing physician overall (A), age17 years or younger (B), age 18 years or older (C).
J AM ACAD DERMATOL
VOLUME 67, NUMBER 5Johnson and Jacobe 885
(63.4%, n = 142), other topicals (41.5%, n = 93),phototherapy (27.2%, n = 61), systemic corticoste-roids (21.4%, n = 48), methotrexate (21.0%, n = 47),and antibiotics/antimalarials (19.6%, n = 44) (Fig 2,A). There were no differences in use of a particulartherapy for a specific disease subtype (P [ .05)(Fig 3).
Methotrexate was prescribed to more childrenthan adults (P = .037), and antibiotics and
antimalarials were prescribed to more adults thanchildren (P \ .001). Frequency of the remainingprescribed treatments did not differ between chil-dren and adults. Treatments prescribed based onmorphea subtype are listed in Fig 3.
Therapy choice based on specialty. Of thetotal 531 prescribed treatments shown in Fig 3, 460(86.8%) had an associated specialty of prescribingphysician recorded. Specialties of treating physicians
Fig 3. Morphea. Treatments prescribed by dermatologist (A) and rheumatologist (B) andsubtype of morphea. EF, Eosinophilic fasciitis.
J AM ACAD DERMATOL
NOVEMBER 2012886 Johnson and Jacobe
included dermatologist (n = 346), rheumatologist(n = 96), family physician (n = 6), and other (n = 12).
The most common treatments prescribed by adermatologist included topical corticosteroids(40.7%, n = 125), topicals (25.1%, n = 77), andphototherapy (16.3%, n = 50). The most commontreatments prescribed by a rheumatologist includedmethotrexate (34.1%, n = 29), systemic corticoste-roids (30.6%, n = 26), and antibiotics and antimalar-ials (14.1%, n = 12). Differences in prescribedtreatments between specialties were statisticallysignificant.
Fig 3 shows the distribution of prescribed treat-ments by prescribing specialty and subtype of mor-phea. Dermatologists prescribed treatments for all
subtypes, whereas rheumatologists prescribed treat-ments to linear (n = 49) (Table IV) and generalized(n = 29) subtypes. In treating the linear subtype,dermatologists prescribed significantly more topicalcorticosteroids (P\ .0001), topicals (P = .003), andphototherapy (P = .003) compared with rheumatol-ogists who prescribed systemic corticosteroids (P\.0001), methotrexate (P = .0001), and physical ther-apy (P = .003). In treating the generalized subtype,dermatologists prescribed significantly more topicalcorticosteroids (P = .015) compared with rheumatol-ogists, and rheumatologists prescribed significantlymore antibiotics and antimalarials (P = .024), sys-temic corticosteroids (P = .0005), and methotrexate(P = .003).
Tables IV to VI illustrate use of different evalua-tion modalities.
Overall, 76.3% (n = 171) of patients had a skinbiopsy (Table IV). Biopsies were performed lessfrequently in children (60.0%, n = 57) (Table V) thanadults (88.4%, n = 114) (P = .008).
In all, 48.2% (n = 108) of patients had laboratoryexamination (42.1% [n = 40] of children and 52.7%[n = 68] of adults). Eosinophilic fasciitis had thehighest percentage of patients with laboratory as-sessment (76.9%, n = 10). The frequency of labora-tory examination did not differ significantly based onage of onset (children vs adults) (P = .259) or diseasesubtype (P = .444). A total of 67 imaging studies wereperformed at similar frequency in adults and children(Table VI) (P = .918). Rates of magnetic resonanceimaging differed significantly among subtypes (P\.0001). A total of 8.9% (n = 20) of patients overall hadmagnetic resonance imaging (53.9% [n = 7] of eosin-ophilic fasciitis subtype, 9.9% [n = 10] of linearsubtype, and 3.7% [n = 3] of generalized subtype).
DISCUSSIONVery little is known about how patients with
morphea are evaluated and treated in the UnitedStates. This impedes development of clinical trialsand therapeutic guidelines. We undertook a cross-sectional survey of patients enrolled in the MAC
cohort to determine time to diagnosis, specialty ofdiagnosing physician, and the types of evaluationsand therapies prescribed across specialties and dis-ease subtypes.
Delay in diagnosis and, consequently, therapywas common. Of patients, 63% were given thediagnosis more than 6 months after disease onset,and 25.5% (n = 57) of patients were given thediagnosis more than 2 years after disease onset.This is similar to reports in pediatric morphea and theUnited Kingdom that report similar delays betweendisease onset and diagnosis.1,3,6 Children and thosewith linear morphea (which predominates in chil-dren) were given the diagnosis earlier than adults.Reasons for this discrepancy may include greaterawareness of morphea in the pediatric communityand greater inclination among parents to seek treat-ment for their children.
Delay in diagnosis and, consequently, therapymay negatively impact outcome. Although data arelacking for the timing of a therapeutic window inmorphea, the prevailing expert opinion is thatmorphea has an active phase that is amenable totreatment followed by a relatively irreversibledamage phase that is not.7-9 It follows that delayin diagnosis and treatment of 6 months or moremay result in more patients presenting in thedamage phase. Studies identifying causes for delayin diagnosis and the impact on outcome areindicated.
Denominator used to calculate percentage was total number of patients in that respective age group and subtype.
Compared with overall X-ray rate, rates of X-ray use in each subtype was not significantly different (P = .923).
EF, Eosinophilic fasciitis; EMG, electromyography; MRI, magnetic resonance imaging.
J AM ACAD DERMATOL
NOVEMBER 2012888 Johnson and Jacobe
Choice of therapy was at least partially driven bythe specialty of the diagnosing physician rather thanindividual morphea characteristics. Although thepreferential use of systemic immunosuppressantsby rheumatologists is partially attributable to thepredominance of more severe subtypes under theircare, our data indicate that patients with a givenmorphea subtype may receive entirely differenttherapy depending on the specialty of the treatingphysician. For example, 68.2% of patients with linearmorphea received topical steroids as the mainstay oftreatment if they were seen by a dermatologist (only4% received methotrexate), whereas 38.8% of pa-tients with linear morphea received methotrexate ifthey were seen by a rheumatologist (only 8% re-ceived topical steroids). These findings are similar tothose reported by Li et al,4 in which pediatricrheumatologists predominantly prescribed systemicimmunosuppressants. Furthermore, phototherapywas used almost exclusively by dermatologists (al-beit infrequently), despite extensive literature sup-porting its use in morphea.10-16 This implies that atleast some patients with morphea are receivingsuboptimal therapy. To date, treatment recommen-dations are based on minimal evidence and largelyexpert opinion.7,17 However, therapies with thegreatest evidence for efficacy are methotrexateand/or systemic corticosteroids, which has led ex-pert opinion to suggest these treatments are indi-cated for progressive or severe morphea (linear andgeneralized).7,17 This underscores the need for amultidisciplinary approach for treatment guidelines,and comparative effectiveness studies examiningmany commonly prescribed morphea treatments.
In addition to prescribing differences based onphysician specialty, there were differences in pre-scribing patterns by age of disease onset.Dermatologists rarely prescribed systemic immuno-suppressives to children, whereas rheumatologistsprescribed systemic immunosuppressives to morechildren than adults. These results suggest thatrheumatologists are more aggressive than dermatol-ogists in treating children. Although outcomes stud-ies have not been performed, current expertconsensus particularly among pediatric rheumatolo-gists is that children with linear lesions affecting theface or crossing joints should receive systemic im-munosuppressives.7 Therefore, dermatologists maynot be adequately treating pediatric morphea,especially the linear subtype. These differencesmay be because the pediatric rheumatology com-munity is very actively developing consensus re-garding clinical evaluation, therapy, and outcomemeasures through the Childhood Arthritis andRheumatology Research Alliance, which are pre-sented at society meetings and submitted for publi-cation.9 Dermatologists may not be aware of thesedevelopments. If so, similar efforts are needed acrossspecialties and for adults with morphea.
Many commonly prescribed treatments and eval-uations were not supported by current evidence. Forexample, despite the lack of evidence, 48% ofpatients did have laboratory-based assessment(which might have been ordered to differentiatemorphea from scleroderma). The most commonlyprescribed therapy, topical corticosteroids (63% ofpatients), does not have evidence for efficacy inmorphea.7,17
J AM ACAD DERMATOL
VOLUME 67, NUMBER 5Johnson and Jacobe 889
There are several limitations to this study. Resultsare based on the participants in the MAC cohort,which is a referral-based cohort. Although everyeffort is made to enroll diverse patients with mor-phea from a large variety of providers, referral biaslikely exists. Time from disease onset to diagnosisrelied on patient recall, which may have impactedaccuracy, although medical records were obtainedand reviewed.
In summary, we identified several factors thatimpact the care of patients with morphea. Theseinclude delay in diagnosis and treatment, largevariation in evaluation and therapy based on thespecialty of the provider, and widespread use ofevaluations and therapy with little evidence fortheir efficacy. We also identified dermatologistsand rheumatologists as relevant stakeholders.Consensus-based, multidisciplinary outcome mea-sures, outcomes studies, and comparative clinicalstudies are needed to improve patient care.
REFERENCES
1. Zulian F, Vallongo C, Woo P, Russo R, Ruperto N, Harper J, et al.
Localized scleroderma in childhood is not just a skin disease.
Arthritis Rheum 2005;52:2873-81.
2. Zulian F, Athreya BH, Laxer R, Nelson AM, Feitosa de Oliveira
SK, Punaro MG, et al. Juvenile localized scleroderma: clinical
and epidemiological features in 750 children: an international
