Juvenile-Onset Ankylosing Spondylitis IsAssociated With Worse Functional OutcomesThan Adult-Onset Ankylosing SpondylitisMILLICENT STONE,1 ROBERT W. WARREN,2 JANE BRUCKEL,3 DANIEL COOPER,4

DANIELA CORTINOVIS,5 AND ROBERT D. INMAN6

Objective. To compare functional outcome of patients with juvenile-onset ankylosing spondylitis (JoAS; defined as ASwith symptom onset before 16 years of age) with that of patients with adult-onset AS (AoAS) and to identify variablesassociated with a poor functional outcome of JoAS.Methods. A cross-sectional study was performed of 326 JoAS patients who participated in a postal survey conducted bythe Spondylitis Association of America. This cohort was compared with 2,021 AoAS patients who participated in thesame survey. Simple and multiple logistic regression analyses were performed to identify differences with respect toclinical features, demographic features, and functional outcome (defined by the Bath Ankylosing Spondylitis FunctionalIndex [BASFI]) between the 2 groups. A validation cohort of 255 AS patients was also surveyed.Results. The mean � SD BASFI score (controlled for disease duration) for JoAS was 51.3 � 1.5 compared with 46.4 �0.6 for AoAS (P < 0.0001). Multiple logistic regression identified only age (P < 0.0001) and income status (P < 0.0001)as factors associated with functional impairment.Conclusion. It appears that JoAS is a progressive disease and is associated with significant delay in diagnosis and worsefunctional outcome compared with AoAS. Furthermore, women do worse than men with JoAS. This would argue for theimportance of early diagnosis and treatment of AS, particularly in the subgroup of patients with JoAS.

KEY WORDS. Juvenile-onset ankylosing spondylitis; Adult-onset ankylosing spondylitis; Functional outcomes.

INTRODUCTION

Ankylosing spondylitis (AS) is a chronic inflammatoryrheumatic disease of the spine that affects 0.2–0.8% of thepopulation (1). Although AS typically presents in the lateteens or early twenties, it can present in childhood. Whensymptom onset occurs in individuals �16 years of age andindividuals go on to develop radiographic sacroiliitis at a

later stage, the disease is termed juvenile-onset AS (JoAS)(2). Among patients with AS, prevalence rates for juvenileonset vary from 9% to 21% in white populations, althoughprevalence rates �40% have been reported with MexicanMestizo and Korean AS patients (3).

Although adult-onset AS (AoAS) and JoAS share manycommon features, both being characterized by the pres-ence of radiographic sacroiliitis, they also differ in manyrespects. JoAS presents more commonly with peripheraljoint symptoms and adults are more likely to present withaxial manifestations. Differences in functional outcomehave also been reported that depend on the age of onset. Ina study comparing 24 JoAS with 71 AoAS patients, JoAShad worse functional outcome (4). In another study com-paring functional outcome in 22 JoAS with 22 AoAS pa-tients, no difference was observed (5). There are severalreasons for the discrepancies reported between the stud-ies, such as small sample size and the lack of a commondefinition of outcomes, making direct comparison betweenstudies difficult. In addition, several phenotypic manifes-tations of JoAS exist that may account for differences indetermining functional outcomes. Burgos-Vargas et al de-scribed a characteristic group of JoAS patients with “gen-

1Millicent Stone, MB, MRCP (UK), MSc: University ofToronto, St. Michael’s Hospital, Toronto, Ontario, Canadaand Royal National Hospital for Rheumatic Diseases, Bath,United Kingdom; 2Robert W. Warren, MD, PhD: Baylor Col-lege of Medicine, Houston, Texas; 3Jane Bruckel, BSN, RN:Spondylitis Association of America, Sherman Oaks, Califor-nia; 4Daniel Cooper, PhD: Harris Interactive Inc., Rochester,New York; 5Daniela Cortinovis, BSc, MPhil: St. Michael’sHospital, Toronto, Ontario, Canada; 6Robert D. Inman, MD,FRCPC: Toronto Western Research Institute, and Universityof Toronto, Toronto, Ontario, Canada.

Address correspondence to Millicent Stone, MB, MRCP(UK), MSc, University of Toronto, Staff Rheumatologist, St.Michaels Hospital, 30 Bond Street, Toronto, Ontario M5B1W8, Canada. E-mail: stonem@smh.toronto.on.ca.

Submitted for publication August 9, 2004; accepted inrevised form January 2, 2005.

Arthritis & Rheumatism (Arthritis Care & Research)Vol. 53, No. 3, June 15, 2005, pp 445–451DOI 10.1002/art.21174© 2005, American College of Rheumatology

ORIGINAL ARTICLE

445

uine adult-like AS” as opposed to patients who presentwith the syndrome of enthesopathic arthritis and evolvesubsequently into AS (3).

No comparison of functional outcome of JoAS withAoAS has been performed to date in a North Americancohort. The potential for biologic agents to improve func-tional outcome in AS has recently been demonstrated inseveral short-term randomized controlled trials (6–8). Forthis reason and because functional outcome has beenshown to be the most important predictor of total costs inAS (9), it is important to determine why some patientshave a poor functional outcome. The objective of thisstudy was to compare functional outcome of JoAS withAoAS and to identify predictors of a poor functional out-come of JoAS.

PATIENTS AND METHODS

Study population. A survey was mailed to subjects reg-istered in the Spondylitis Association of America (SAA)database. The database consisted of 1,500 current mem-bers of the SAA, 2,460 lapsed members, and 3,600 indi-viduals who had made enquiries to the SAA after 1992.The questionnaire was administered by Harris Interactivefrom July through September 2002. In all, there were 30questions relating to diagnosis, pattern of disease, treat-ment, effect of AS on quality of life, and demographicinformation, such as age, sex, education, and socioeco-nomic status. The questionnaire took �30 minutes to com-plete. Potential respondents were sent a letter a few daysprior to receiving the survey introducing the survey in thehope that they would be more likely to respond if theywere informed about the study and its aims in advance.

A validation cohort was also surveyed and consisted of255 AS patients with diagnosis confirmed by the treatingphysician. The validation cohort was recruited from thepractices of 44 physicians who were approached andasked to request the participation of AS patients in theirpractice. This group was completely separate from thesurvey sample. The study was approved by the local ethicsreview committee.

Patients with disease onset at �16 years of age wereclassified as having JoAS and those with symptom onset�16 years of age were classified as having AoAS. There-fore, these disorders are not clinically distinct and differonly in the age of onset of symptoms.

Primary outcome measure. The Bath Ankylosing Spon-dylitis Functional Index (BASFI) was used to measurefunctional impairment (10). This instrument comprises 8questions relating to the functional anatomy of AS patientsand 2 additional questions that assess the AS patient’sability to cope with everyday life. Each question is an-swered on a 100-mm visual analog scale (0 � easy, 100 �impossible). The mean of the 10 scales gives the BASFIscore. This tool has also been validated in a cohort of 163AS patients. It takes on average 47 seconds to complete.When distinguishing between patients with high and lowdisease severity, it has been shown to have a sensitivity of94% and specificity of 87% (10). The BASFI correlates

with other external criterion for disease activity, such aspatient and physician global assessment of disease activityand function. Severe functional impairment was definedas a BASFI score �50.

Secondary outcome measures. Individuals were con-sidered to have AS-related work disability if they wereever on either temporary or permanent work disabilityresulting from AS. Self-perceived spinal deformity wasassessed to obtain information about AS disease severityand damage. Patients were asked to what extent their AScaused them to stoop over. Patients were shown 4 graphicsrepresenting 0°, �25°, 25–45°, and �45° of deformity andasked to choose the one that best represented their posture(Figure 1). This instrument was developed for the purposeof this survey, to ensure it met the construct of face valid-ity for the assessment of self-perceived spinal deformity. Afocus group of individuals, some of whom had AS, met toplan the development of the instrument and ensure it metthe principles of truth and feasibility as outlined by theOutcome Measures in Rheumatology Clinical Trials filter.Subsequent to this, an AS patient, a methodologist, andseveral lay individuals who work at the SAA met with thegraphic design team at Harris Interactive to finalize thegraphics for the instrument.

Statistical analysis. Comparisons between the valida-tion cohort and total cohort of JoAS and AoAS patients fordemographic and clinical characteristics were performedusing descriptive statistics and parametric and nonpara-metric tests, as appropriate. To provide some correction tothe skew in the data, time since diagnosis (i.e., diseaseduration) was truncated at 35 years, delay in diagnosis wastruncated at 50 years, and the minimum for age of firstsymptoms (JoAS) was set at 9 years.

Simple and multiple logistic regression analyses wereperformed to identify significant variables associated withfunctional impairment. In the univariate analysis, themean BASFI for JoAS and AoAS patients was comparedand adjustments were made for disease duration using atwo–way analysis of variance test. Functional impairment

Figure 1. Bruckel instrument for patient self assessment of spinaldeformity. This was a tool developed for the purpose of this studyto assess the degree of spinal deformity by patient self report. Fourcartoon figures denote various stages of advanced spinal defor-mity. Study patients were asked to select the cartoon figure thatthey felt best represented the degree of spinal deformity theyperceived they had. The instrument was named after JaneBruckel, cofounder of the Spondylitis Association of America.

446 Stone et al

was the dependant variable and was considered a contin-uous variable in this analysis. Thus, the influence on func-tional impairment of age, disease duration, delay in diag-nosis, education, household income, sex, and age at onsetwas statistically assessed. Variables with P values � 0.05in the simple logistic regression analyses were entered tothe multiple regression analysis. All statistical analyseswere performed using SAS version 8.02 (SAS Institute,Cary, NC).

RESULTS

Response to survey. Questionnaires were mailed to7,500 individuals on the SAA database, of which 545could not be delivered and were returned by the postoffice. Of the 6,955 individuals who received the question-naire, 2,384 responded. There was broad representationfrom all the States in the United States, with a meannumber of 53 patients recruited per state and a range of4–410. The percentage response rate by category was asfollows: for current members, 877 of 1,498 (59%); forlapsed members, 678 of 2,174 (31%); and for those whoinquired after 1992, 684 of 3,600 (19%). Among the vali-dation cohort, 194 of 245 (75%) individuals responded.

Validation cohort. The validation cohort was similar tothe remainder of the cohort with respect to age, age atdisease onset, sex, ethnicity, delay in diagnosis, and func-tional impairment due to AS (data not shown).

Demographic features. There were 326 individuals whoreported symptom onset before 16 years of age and 2,021who had onset after 16 years of age. The diagnosis of ASwas made by a rheumatologist in 63% of individuals, by ageneralist in 15%, by an orthopedic surgeon in 13%, anophthalmologist in 5.4%, and a chiropractor in 4.5%.There was no difference in sex distribution between thegroups; 60% were male. The majority of patients (94%)sampled were white. However, a significantly greater pro-portion of AoAS patients (95%) were white as compared

with JoAS patients (89%) (Table 1). JoAS patients wereyounger at the time of the survey (current mean age 46.5 �0.76 versus 50.9 � 0.27 years; P � 0.001), were bettereducated (P � 0.003), less likely to be married or in a stablerelationship (P � 0.001), and less likely to be from a highincome category (P � 0.0002) when compared with AoASpatients at the time of the survey (Table 1).

Clinical characteristics. JoAS patients were more likelyto present with peripheral joint symptoms (46.6% versus33.2%; P � 0.001) (Table 2). A greater percentage of AoASpatients presented with axial symptoms, low back pain,and stiffness (71.5% versus 66.0%; P � 0.043). There wasno difference between the 2 groups in the frequency ofenthesopathic manifestations, iritis, or neck pain at pre-sentation. At the time of survey, JoAS patients had a meandisease duration of 18.3 � 0.70 years compared with13.4 � 0.23 years for AoAS patients (P � 0.001). JoASpatients had a significantly longer delay in diagnosis com-pared with patients with AoAS (15.3 � 0.79 versus 7.6 �0.2 years; P � 0.001).

Disability and functional impairment. Among JoAS pa-tients, 27.4% reported AS-related work disability, com-pared with 20.9% of individuals with AoAS (P � 0.012)(Table 2). Among JoAS patients, 17.2% reported spinaldeformity (25°–45°) compared with 10.8% of AoAS pa-tients. Only 26.2% of JoAS patients and 34.8% of AoASpatients reported no spinal deformity.

The distribution of the BASFI scores for the JoAS pa-tients was reasonably normal. The mean BASFI for JoASwas 51.3 � 1.5 with an interquartile range of 72–30 com-pared with 46.4 � 0.57 for AoAS patients with an inter-quartile range of 66–25 (Table 2). This held true afteradjustment for disease duration. This adjustment wasmade by running a 2-way analysis of variance (JoAS/AoAsand disease duration, BASFI as the dependant variable)with interaction. Both disease group and disease durationwere significant (P � 0.0001 and P � 0.028, respectively),but the interaction term was not (P � 0.814). Thus, JoAS

Table 1. Comparison of demographic characteristics of juvenile- versus adult-onsetdisease

VariableJuveniles(n � 326)

Adults(n � 2,021) P

Women, % 39.1 39.7 0.838Age, mean � SEM years 46.5 � 0.76 50.9 � 0.27 � 0.001Marital status, % � 0.001

Married 60.4 74.6Single 22.4 10.7Divorced/widow 12.6 12.8Living with partner 4.6 1.9

Education, % 0.003High school or less 10.5 17.2College or more 89.5 82.8

Race, % � 0.001White 89.3 94.6Other 10.7 5.4

High income, % 88.6 93.8 0.002

Juvenile-Onset Ankylosing Spondylitis and Functional Outcomes 447

patients had significantly greater functional impairmentafter controlling for disease duration as defined with thesegroups (Figure 2). In addition, there was a significant effectof disease duration as defined with these groups. However,the effect of one did not depend on the other.

Among JoAS patients, 171 were classified as havingsevere functional impairment and 160 did not meet thecriteria for severe functional impairment. There was nodifference in sex distribution, marital status, education,race, or age at disease onset between JoAS patients withsevere functional impairment and those without. How-ever, the cumulative percentage of JoAS patients withsevere functional impairment appeared to increase withlonger disease duration (Figure 3A). Furthermore, womenwith JoAS were more likely to have severe functional impair-ment, as defined by a BASFI score �50, earlier in the diseasecourse as compared with men (Figure 3B). JoAS patients withsevere functional impairment were also noted to have agreater delay in diagnosis, 16.7 � 0.9 versus 11.9 � 0.8 yearsas compared with JoAS patients without severe functionalimpairment (data not shown). There appeared to be a linearassociation between delay in diagnosis and severe functionalimpairment in JoAS (Figure 3C).

Figure 2. Comparison of functional impairment (defined by theBath Ankylosing Spondylitis Functional Index [BASFI]) versusdisease duration stratified by age at disease onset. Juvenile-onsetankylosing spondylitis (JoAS) patients had significantly greaterfunctional impairment after controlling for disease duration ascompared with adult-onset ankylosing spondylitis (AoAS) pa-tients. This adjustment for disease duration was made by runninga two-way analysis of variance (JoAS/AoAs and disease duration,BASFI as the dependant variable) with interaction. Both diseasegroup and disease duration were significant (P � 0.0001 and P �0.028, respectively) but the interaction term was not (P � 0.814).In addition, there was a significant effect of disease duration asdefined with these groups. However, the effect of one did notdepend on the other.

Table 2. Comparison of clinical characteristics for juvenile- versus adult-onset disease*

VariableJuveniles(n � 326)

Adults(n � 2,021) P

Age at symptom onset, years 13.6 � 0.14 30.1 � 0.25 � 0.001Delay in diagnosis, years 15.3 � 0.79 7.6 � 0.20 � 0.001Disease duration, years 18.3 � 0.70 13.4 � 0.23 � 0.001BASFI (0–100) 51.3 � 1.5 46.4 � 0.57 � 0.001†

Putting on socks/tights 3.7 � 0.16 3.2 � 0.06 0.013Bending to pick up pen 5.0 � 0.18 4.3 � 0.07 � 0.001Reaching to high shelf 4.2 � 0.17 3.7 � 0.07 0.004Getting out of chair 4.7 � 0.18 4.4 � 0.07 0.067Off floor from lying on back 5.6 � 0.18 5.4 � 0.07 0.171Standing unsupported for 10 min 5.2 � 0.19 4.7 � 0.07 0.016Climbing 12–15 steps 4.5 � 0.19 4.2 � 0.07 0.143Looking over shoulder 6.8 � 0.18 6.2 � 0.08 0.004Physically demanding activities 6.0 � 0.15 5.4 � 0.06 � 0.001Full day’s activities 5.5 � 0.16 5.1 � 0.07 0.001

AS-related work disability, % 27.4 20.9 0.012Spinal deformity, %‡ � 0.001

Not at all 26.2 34.8�25° 53.8 52.725°–45° 17.2 10.8�45° 2.8 1.6

Presenting symptoms, %Neck pain/stiffness 31.3 33.9 0.346Back pain/stiffness 66.0 71.5 0.043Stiffness in joints 46.6 33.2 � 0.001Heel pain 13.2 12.2 0.622Sternum pain 16.9 16.5 0.877Iritis 13.8 15.3 0.473Other 32.2 21.8 � 0.001

* Except where otherwise indicated, the values are mean � SEM. BASFI � Bath Ankylosing SpondylitisFunctional Index; min � minutes.† Adjusted for disease duration.‡ Self-perceived spinal deformity.

448 Stone et al

Multiple logistic regression. Simple logistic regressionanalyses indicated age, delay in diagnosis, disease dura-tion, and income status as having a significant effect onfunctional impairment (Table 3). These variables werethen included in the multiple regression analysis, whichidentified only age (P � 0.046) and income status (P �0.001) as factors associated with functional impairment.Both disease duration and delay in diagnosis were highlycorrelated with age, explaining why they were not signif-icantly associated with the outcome in the multiple regres-sion analysis.

DISCUSSION

This is the first North American study of AS patients thatattempts to compare functional outcome in JoAS patientswith that in AoAS patients, and is the largest publishedseries to do so. In this study, we found that JoAS patientsexperience a greater delay in diagnosis compared withAoAS patients and suffered more functional impairmentcompared with AoAS patients with the latter holding evenwhen controlled for disease duration. Women with JoASappear to develop functional impairment earlier than men

Figure 3. A, Cumulative percentage of juvenile-onset ankylosingspondylitis (JoAS) patients with severe functional impairment(defined as Bath Ankylosing Spondylitis Functional Index[BASFI] score �50) plotted against disease duration representedin years. B, Cumulative percentage of JoAS patients with severefunctional impairment (defined as BASFI score �50) plottedagainst disease duration represented in years. JoAS patients arestratified by sex (thin line � men, thick line � women). C, Cu-mulative percentage of JoAS patients with severe functional im-pairment (defined as BASFI score �50) plotted against delay indiagnosis represented in years.

Table 3. Predictors of functional impairment in juvenile-onset ankylosing spondylitis by univariate and multiple regressionanalysis*

Variable Univariate model P Multiple model P

Age, years b � 0.633; SE(b) � 0.100 � 0.001 b � 1.187; SE(b) � 0.591 0.046High income b � �28.716; SE(b) � 4.762 � 0.001 b � �27.606; SE(b) � 4.463 � 0.001Delay in diagnosis, years b � 0.458; SE(b) � 0.113 � 0.001Disease duration, years b � 0.382; SE(b) � 0.131 0.004

* Disease duration was truncated at 35 years, delay in diagnosis truncated at 35 years, and the minimum age of first symptoms was set at 9 years.

Juvenile-Onset Ankylosing Spondylitis and Functional Outcomes 449

with JoAS. Multivariate analysis demonstrated that ageand low income status were strongly associated withworse functional outcome in JoAS compared with AoAS.

One strength of this study lies in the large number ofJoAS patients (n � 326) who participated in the survey, thelargest reported in the literature to date. Because this sur-vey targeted patients in the population, it is likely that thiscohort represents a spectrum of AS disease severity com-pared with patients recruited only from a tertiary referralcenter. A unique aspect of the study was that there wasrepresentation from all states in the US, although the great-est numbers came from California and New York. Despitethe large number of respondents, it is acknowledged thatthe overall response rate to the survey was 30%. This maybe due to the fact that a large number of patients who weremailed questionnaires from the SAA database were thosewho had only made an inquiry to the SAA. On the otherhand, the response rate from current members of the SAAwas high, at 78%. It is possible that nonresponders in-cluded a high percentage of patients who in fact had neverbeen diagnosed with AS, particularly since the responserate for SAA inquiries was only 19%.

The large sample size facilitated a multivariate analysisto identify factors associated with outcome in JoAS. Thiswas not possible in previous studies because of smallersample sizes (4,5). However, a limitation of a cross-sec-tional study design is that it precludes drawing any defin-itive conclusion about prognostic indicators of outcome.Therefore we were only able to identify strongly associatedvariables that may be predictors of outcome. A prospectiveobservational cohort study with patients enrolled at incep-tion would be required to confirm if the identified vari-ables are true predictors of outcome. To date, a large epi-demiologic cohort of AS patients has not been compiled inNorth America. The low incidence of the disease and theprolonged delay in diagnosis pose logistic problems forconducting such a study.

In this study, the diagnosis of AS was made by selfreport and this does have limitations with regard to mis-classification of cases. It is possible that this would biasthe results by including some patients who did not havethe disease. The bias would, however, most likely be thesame for JoAS and AoAS. We did ask patients which healthprofessional made the diagnosis and there was no differencebetween groups. Almost 90% of individuals reported that thediagnosis was made by a physician. In addition, there wereno differences between the 2 groups and the validation co-hort where the diagnosis was confirmed by a physician. Thiswas the reason we decided a validation cohort was importantto include in the study design.

We acknowledge that questioning patients about symp-tom onset may be fraught by recall bias. However we feelthat the recall bias in this instance may have lead to anunderestimation of the true disease duration for many ASpatients. Patients are more likely to remember episodes ofdisease activity and forget good periods. Therefore we mayhave biased the results toward underestimation of diseaseduration or indeed delay in diagnosis. Prospective fol-lowup of patients from date of onset of symptoms is theonly means of avoiding recall bias.

No recent studies have been performed comparing JoAS

to AoAS. The existing publications are older studies, eachreporting on a small number of patients. The managementof AS has changed markedly in recent years, with biologictherapies increasingly prescribed for many patients, whichmakes comparison of patient outcomes between our studyand these older studies difficult. In addition, several meth-odologic differences further limit direct comparisons be-tween studies.

In one study, a survey was conducted of 3,000 AS pa-tients of whom 135 had JoAS (11). The results were com-pared with 135 respondents who had AoAS. Functionalimpairment was measured using the Arthritis Impact Mea-surement Score (AIMS) (12), which may not be as sensitiveas the BASFI for detecting disease-specific functional im-pairment in AS. The authors reported that functional im-pairment, disability, and impact of disease were lessmarked in JoAS compared with AoAS. However, the JoASpatients had a mean age of 37.3 years at the time of sam-pling compared with our cohort, which had a mean age of46.5 years at time of survey. In our study, age was identi-fied as being significantly associated with functional out-come. Therefore, this difference in age at the time of sam-pling between studies may account for the discrepancy inreported outcomes between studies.

In another study by the same group, 22 JoAS patientswere compared with 22 AoAS patients (5). There weresome similarities with our study; delay in diagnosis wasgreater in JoAS compared with AoAS, with women expe-riencing the longest delay in diagnosis. Patient numbersdid not allow an analysis stratified by sex or a multivariateanalysis. The results of this study may have been limitedby referral bias, as patients with milder disease may nothave been included because this study was based on pa-tients recruited from a tertiary referral center.

Garcia-Morteo et al (4) addressed the comparison of the2 clinical entities in a retrospective study. The authorsconcluded that functional impairment was more severe inthe JoAS group. Nineteen percent (n � 24) of the patientsin this study had JoAS, as compared with 14% of ourcohort. Seventy-one patients with AoAS were included inthe study. Of these, only 46 patients were evaluated forfunctional outcome. In the study, all 24 patients with JoASexhibited some degree of functional impairment, as de-fined by the American College of Rheumatology functionalclassification system, after mean disease duration of 15years. In comparison, 80% of patients with AoAS wereclassified as having functional impairment. The method-ology of the study was not adequately outlined to deter-mine why the remaining individuals were not reported onfor this outcome.

Our study had some important messages for clinicians.There persists a significant delay in diagnosis that is moremarked for JoAS than AoAS, and women had an evengreater delay. Furthermore, delay in diagnosis is associ-ated with worse functional outcomes in JoAS patients. Themagnitude of the difference between the 2 groups in func-tional outcomes that we noted was small. To date therehave been no studies performed to determine the func-tional significance of small differences in the BASFI scores.The BASFI total score is a mean of 10 subscales. To betterunderstand the differences between groups in functional out-

450 Stone et al

come, we compared the difference in the mean of each ofthese subscales between JoAS and AoAS patients. For each ofthese subscales, patients with JoAS scored higher than AoASpatients. The data presented in Table 2 indicates that thosesubscales that relate to lumbar spine and cervical spinemovements had the largest differences between groups. Thusthe clinical relevance of this may be that patients with JoAShave more severe axial symptoms than those with AoAS.

This observation is particularly important now, giventhat biologics are emerging as effective treatments for AS(13). Earlier diagnosis and treatment may improve long-term functional outcomes, but this awaits formal proof inprospective therapeutic trials with longer followup. Thereis a need for a higher index of suspicion among physiciansso that patients with JoAS are diagnosed as early as pos-sible. With the emerging role of magnetic resonance imag-ing as a diagnostic tool in AS, this may also facilitate amore prompt diagnosis.

There remain many unanswered questions that need tobe addressed in future research efforts. It will be difficultto study the natural history of AS in the future as treatmentwith tumor necrosis factor blocking agents becomes morewidespread. Therefore, identification of clinical prognos-tic markers will also be difficult. However, research effortsshould focus on stratifying patients into prognostic cate-gories so that therapy is initiated not only sooner, but alsoappropriately for each patient. Based on our results, itappears that JoAS is a progressive disease and is associatedwith worse functional outcome compared with AoAS.This would argue for a widespread campaign to educatephysicians about the importance of early diagnosis andtreatment of patients with JoAS. However, it remains un-known if treatment of patients early in the disease willprevent disease progression, and future studies shouldaddress this issue.

ACKNOWLEDGMENTThe authors would like to acknowledge the assistance ofSandy Appelbaum of Harris Interactive.

REFERENCES

1. Braun J, Bollow M, Remlinger G, Eggens U, Rudwaleit M,Distler A, et al. Prevalence of spondylarthropathies in HLA-B27 positive and negative blood donors. Arthritis Rheum1998;41:58–67.

2. Burgos-Vargas R, Pacheco-Tena C, Vazquez-Mellado J. Juve-nile-onset spondyloarthropathies. Rheum Dis Clin North Am1997;23:569–98.

3. Burgos-Vargas R, Vazquez-Mellado J, Cassis N, Duarte C,Casarin J, Cifuentes M, et al. Genuine ankylosing spondylitisin children: a case-control study of patients with early definitedisease according to adult onset criteria. J Rheumatol 1996;23:2140–7.

4. Garcia-Morteo O, Maldonado-Cocco JA, Suarez-Almazor ME,Garay E. Ankylosing spondylitis of juvenile onset: compari-son with adult onset disease. Scand J Rheumatol 1983;12:246–8.

5. Marks SH, Barnett M, Calin A. A case-control study of juve-nile- and adult-onset ankylosing spondylitis. J Rheumatol1982;9:739–41.

6. Braun J, Brandt J, Listing J, Zink A, Alten R, Golder W, et al.Treatment of active ankylosing spondylitis with infliximab: arandomised controlled multicentre trial. Lancet 2002;359:1187–93.

7. Davis JC Jr, van der Heijde D, Braun J, Dougados M, Cush J, CleggDO, et al. Recombinant human tumor necrosis factor receptor(etanercept) for treating ankylosing spondylitis: a randomized,controlled trial. Arthritis Rheum 2003;48:3230–6.

8. Gorman JD, Sack KE, Davis JC Jr. Treatment of ankylosingspondylitis by inhibition of tumor necrosis factor alpha.N Engl J Med 2002;346:1349–56.

9. Ward MM. Functional disability predicts total costs in pa-tients with ankylosing spondylitis. Arthritis Rheum 2002;46:223–31.

10. Calin A, Garrett S, Whitelock H, Kennedy LG, O’Hea J, Mal-lorie P, et al. A new approach to defining functional ability inankylosing spondylitis: the development of the Bath Anky-losing Spondylitis Functional Index. J Rheumatol 1994;21:2281–5.

11. Calin A, Elswood J. The natural history of juvenile-onsetankylosing spondylitis: a 24-year retrospective case-controlstudy. Br J Rheumatol 1988;27:91–3.

12. Meenan RF. The AIMS approach to health statusmeasurement: conceptual background and measurementproperties. J Rheumatol 1982;9:785–8.

13. Braun J, Brandt J, Listing J, Rudwaleit M, Sieper J. Biologictherapies in the spondyloarthritis: new opportunities, newchallenges. Curr Opin Rheumatol 2003;15:394–407.

Juvenile-Onset Ankylosing Spondylitis and Functional Outcomes 451