Positive family history of psoriasis does not affect the clinical expression and course of juvenile idiopathic arthritis patients with oligoarthritis
In the 1997 revision of the International League of Associations for Rheumatology (ILAR) criteria for juvenile idiopathic arthritis (JIA), a family history of psoriasis is an exclusion for the oligoarthritis category. We investigated whether psoriasis in a first or second degree relative influences the clinical expression and course of JIA patients with oligoarthritis.
In a cross-sectional study, consecutive oligoarticular-onset JIA patients were investigated. Clinical evaluations included confirmation of a family history of psoriasis and assessment of nail abnormalities, dactylitis, psoriatic rash, variables of JIA activity, and laboratory indicators of inflammation. Retrospective assessments included sex, onset age, disease duration, antinuclear antibodies, HLA–B27, uveitis, ocular complications, second-line therapies, intraarticular corticosteroid injections, radiographic joint lesions, joint involvement over time, and laboratory investigations at disease presentation and first observation.
A total of 185 patients were included. Thirty-three had a positive family history of psoriasis (group 2) and 139 did not (group 1). Thirteen patients fulfilled the ILAR criteria for juvenile psoriatic arthritis (group 3). Patients in groups 1 and 2 were comparable for all parameters, except for a higher frequency of females in group 1 (P = 0.04). As compared with group 2, patients in group 3 were less frequently antinuclear antibody positive and had a more severe arthritis and a different distribution of joint involvement.
We found close similarities in the clinical features and course among patients with oligoarthritis who had a positive family history for psoriasis and those who did not. These findings argue against the exclusion of the former patients from the oligoarthritis category of JIA.
Childhood chronic arthritis is a heterogeneous group of conditions of unknown etiology whose classification is based on clinical characteristics. In the past 3 decades, a number of classification systems have been used (reviewed in1), including the American College of Rheumatology criteria for juvenile rheumatoid arthritis, the European League Against Rheumatism criteria for juvenile chronic arthritis, the European Spondyloarthropathy Study Group criteria for spondyloarthropathy, and the Vancouver criteria for juvenile psoriatic arthritis (JPsA). However, none of these traditional classifications of childhood arthritis has been universally embraced. A Classification Taskforce of the Pediatric Standing Committee of the International League of Associations for Rheumatology (ILAR) was, therefore, created with the aim of developing new criteria that would enable identification of more homogeneous groups of children with chronic arthritis to facilitate international communication and research. The committee proposed (2) and revised (3) criteria for the classification of the juvenile idiopathic arthritis (JIA), the term adopted as an umbrella term to indicate all chronic childhood arthritides of unknown cause. Eight different categories were recognized (systemic arthritis, oligoarthritis, extended oligoarthritis, polyarthritis rheumatoid factor negative, polyarthritis rheumatoid factor positive, JPsA, enthesitis- related arthritis, and other arthritis) and definitions, descriptors (i.e., information for further subclassification), and specific exclusions for each category were provided. According to the committee recommendation, the criteria, as now they stand, should be regarded as a “work in progress” and members of the pediatric rheumatology research community are urged to participate in the evaluation process by making their opinions known and by analyzing the proposed criteria on patient series (1).
In recent years, the applicability of ILAR criteria has been tested in different pediatric rheumatology patient populations, which had been previously classified according to other established criteria (4–10). Although the proposed ILAR system appeared to be an improvement over the traditional criteria in that the patients within each group seemed to be more homogeneous, a common findings in these studies was that a significant proportion of patients failed to meet ILAR criteria for specific arthritides and, therefore, became unclassifiable, falling into the category termed “other arthritis.” One of the leading reasons that prevented appropriate categorization of these patients was the choice of the ILAR committee to include in the list of exclusions for the oligoarthritis category the presence of a positive family history of psoriasis in at least 1 first or second degree relative. However, it is unclear whether inclusion of children who have a positive family history of psoriasis, and are not eligible for the JPsA category because they have no psoriatic rash, dactylitis, or nail changes, would dilute the homogeneity of the population in the oligoarthritis category (11).
In the present study, we evaluated a cohort of patients with oligoarticular-onset JIA, not fulfilling the ILAR criteria for JPsA, to determine whether there were differences in the clinical features and course between patients who had a positive family history of psoriasis and those who did not. In addition, the characteristics of the former group of patients were contrasted to those of consecutive patients with oligoarthritis seen in the same period who fulfilled the ILAR criteria for JPsA.
PATIENTS AND METHODS
All consecutive patients attending the department of pediatrics of the University of Pavia, Italy between September 2000 and December 2001 who met the ILAR criteria for JIA (arthritis of unknown etiology persisting for at least 6 weeks with onset before the 16th birthday) (3) and had an oligoarticular onset (arthritis affecting 1–4 joints during the first 6 months of disease) were included in the study. Patients who met the ILAR criteria for systemic arthritis, polyarthritis (rheumatoid factor positive or negative), enthesitis-related arthritis, and HLA–B27-positive males with onset of arthritis after 8 years of age were excluded. The latter group of patients was excluded because the ILAR Committee believes that most children with these characteristics probably have an enthesitis-related arthritis and, thus, suggests to exclude them from the oligoarthritis category (3).
At the study visit, a family history of psoriasis confirmed by a dermatologist in any first or second degree relative was carefully elicited. Clinical evaluations included the search for nail abnormalities (pitting, defined as a minimum total of 2 pits on one or more nails, or onycholysis), dactylitis (defined as swelling of 1 or more digits that extends beyond the joint margins), or psoriatic rash. In addition, we assessed the following clinical variables of JIA activity: physician global assessment of overall disease activity measured on a 10-cm visual analog scale (VAS; 0 best, 3 worst); parent global assessment of overall well being measured on 10-cm VAS (0 best, 3 worst); parent assessment of pain in the past week measured on a 10-cm VAS (0 best, 3 worst); number of joints with active arthritis (defined as the number of joints with swelling, pain on motion, or tenderness and limited range of motion); number of joints with pain on motion or tenderness; number of joints with swelling; number of joints with limited range of motion; and overall articular severity score, calculated as the sum of the severity ratings obtained for the scores of pain on motion or tenderness, swelling, and limited range of motion, as previously reported (12). Parents of each patient were asked to complete the Italian versions of a questionnaire for the assessment of functional ability (the Childhood Health Assessment Questionnaire [CHAQ]), and a generic health instrument designed to capture the physical and psychosocial well being of children ≥ 5 years of age (the Childhood Health Questionnaire [CHQ]) (13, 14). The CHAQ disability index ranges from 0 (best) to 3 (worst), whereas the CHQ yields 2 summary scores, the physical and psychosocial scores, whose mean ± SD norm-based score is 50 ± 10. Laboratory indicators of systemic inflammation included white blood cell count, hemoglobin level, platelet count, erythrocyte sedimentation rate, and C-reactive protein.
For each patient, the following information was recorded by reviewing medical records: sex, age at onset of JIA, disease duration, age at the time of cross-sectional assessment, antinuclear antibody (ANA) and HLA–B27 determinations, occurrence of anterior uveitis and ocular complications, previous identification of nail abnormalities (pitting or onycholysis), dactylitis, psoriatic rash, second line drug administration, number of intraarticular corticosteroid injections, occurrence of radiographic joint lesions (defined as the presence of joint space narrowing and/or bone erosions in at least 1 joint), and type of joints involved at disease presentation and at 6 months, 12 months, and then every 12 months after disease onset. We were facilitated in gathering joint involvement information at specific intervals because we perform and record prospectively on standardized sheets a detailed joint assessment in each patient every 6 months since first observation. Based on the distribution of joint involvement, patients were evaluated for number and type of joints involved over time, occurrence of polyarticular extension (involvement of a cumulative total of more than 4 joints), involvement limited to large or small joints, and limb predominance. Retrospective evaluations included the registration of the following laboratory investigations at disease presentation (if available) and at first observation at our center: white blood cell count, hemoglobin level, platelet count, erythrocyte sedimentation rate, and C-reactive protein. Patients were considered ANA positive if they had ≥75% of positive results in the test for the determination of ANA during the whole disease course. The mean (±SD) number of ANA determinations in the whole cohort was 4.4 (±3.4).
Based on the results of the clinical assessments, patients were divided into 3 groups: group 1 = patients with a negative family history of psoriasis; group 2 = patients with a positive family history of psoriasis; group 3 = patients fulfilling the ILAR criteria for JPsA.
Comparisons of quantitative variables among the 3 groups of patients were made by means of the one-way analysis of variance; the Scheffé test was chosen as a posterior test. Comparisons of qualitative data between group 1 and group 2, and between group 2 and group 3 were performed by means of the chi-square test or the Fisher's exact test in case of expected frequencies less than 5. All the statistical tests were two sided; a P value less then 0.05 was considered statistically significant.
A total of 185 consecutive children with oligoarticular-onset JIA were included in the study. Of the 172 patients who did not fulfill the ILAR criteria for JPsA, 33 had a positive family history of psoriasis in a first (42%) or second degree (58%) relative (group 2) and 139 did not (group 1). Thirteen patients fulfilled the ILAR criteria for JPsA (group 3).
Tables 1–3 show the comparison of the clinical and laboratory features and course and distribution of joint involvement over time among the 3 patient groups. Patients in groups 1 and 2 were comparable for all parameters, with the exception of a slightly higher frequency of females in group 1 (P = 0.04). The 2 groups were also comparable for all variables of JIA activity and severity measured at the time of cross-sectional assessment (Table 4).
Table 1. Demographic and clinical characteristics of the study patients*
|Mean onset age, years||3.81||3.73||4.48|| || |
|Mean age at cross-sectional assessment, years||8.00||8.18||9.67|| || |
|Sex, no. (%)|| || || || || |
| Female||118 (84.9)||23 (69.7)||7 (53.8)||0.04|| |
| Male||21 (15.1)||10 (30.3)||6 (46.2)|| || |
|Mean disease duration, years||4.24||4.48||5.26|| || |
|ANA positive, no. (%)||129 (92.8)||30 (90.9)||8 (61.5)|| ||0.02|
|HLA–B27, no. positive/tested (%)||8/70 (11.4)||3/15 (20.0)||1/8 (12.5)|| || |
|Mean ESR at first observation, mm/hour||34.0||35.2||25.8|| || |
|Radiographic joint lesions, no./tested||16/68 (23.5)||5/17 (29.4)||2/6 (33.3)|| || |
|Iridocyclitis, no. (%)||27 (19.4)||6 (18.2)||3 (23.1)|| || |
| Monolateral||14 (51.8)||4 (66.7)||1 (33.3)|| || |
| Bilateral||13 (48.2)||2 (33.3)||2 (66.7)|| || |
|Ocular complications, no. (%)||16 (11.6)||5 (15.1)||2 (15.4)|| || |
|Previous second-line therapy, no. (%)||41 (29.5)||12 (36.4)||9 (69.2)|| ||0.04|
|Previous IAC injections, no. (%)||123 (88.5)||32 (97.0)||10 (76.9)|| || |
Table 2. Joint involvement over time
|Mean no. joints at disease presentation||1.33||1.36||2.00|| ||0.04|
|Mean cumulative no. joints at 6 months||2.29||2.30||2.54|| || |
|Mean cumulative no. joints at 12 months||3.63||2.90||3.43|| || |
|Mean cumulative no. joints at 24 months||4.07||4.00||4.71|| || |
|Mean cumulative no. joints in disease course||5.21||6.94||12.92|| ||0.006|
|Polyarticular extension, no. (%)||64 (46.0)||13 (39.4)||10 (76.9)|| ||0.02|
|Large joints only, no. (%)||67 (48.2)||15 (45.5)||4 (30.8)|| || |
|Small joints only, no. (%)||0 (0)||0 (0)||0 (0)|| || |
|Large and small joints, no. (%)||72 (51.8)||18 (54.5)||9 (69.2)|| || |
|Limb predominance, no. (%)|| || || || || |
| Upper||28 (20.1)||6 (18.2)||8 (61.5)|| ||0.002|
| Lower||93 (66.9)||26 (78.8)||3 (23.1)|| || |
| No limb predominance||18 (13.0)||1 (3.0)||2 (15.4)|| || |
|Dactylitis, no. (%)||2 (1.4)||1 (3.0)||6 (46.2)|| ||—‡|
Table 3. Distribution of joint involvement throughout the disease course
|Temporomandibular, no. (%)||8 (5.8)||0 (0)||0 (0)|| || |
|Cervical spine, no. (%)||9 (6.5)||3 (9.1)||4 (30.8)|| || |
|Shoulder, no. (%)||5 (3.6)||2 (6.1)||1 (7.7)|| || |
|Elbow, no. (%)||32 (23.0)||7 (21.2)||5 (38.5)|| || |
|Wrist, no. (%)||37 (26.6)||11 (33.3)||7 (53.8)|| || |
|Metacarpophalangeal, no. (%)||24 (17.3)||5 (15.2)||7 (53.8)|| ||0.02|
|Proximal interphalangeal, no. (%)||51 (36.7)||12 (36.4)||8 (61.5)|| || |
|Distal interphalangeal, no. (%)||8 (5.8)||1 (3.0)||5 (38.5)|| ||0.005|
|Hip, no. (%)||10 (7.2)||4 (12.1)||4 (30.8)|| || |
|Knee, no. (%)||131 (94.2)||32 (97.0)||11 (84.6)|| || |
|Ankle, no. (%)||97 (69.8)||24 (72.7)||7 (53.8)|| || |
|Metatarsophalangeal, no. (%)||10 (7.2)||5 (15.2)||1 (7.7)|| || |
|Interphalangeal of feet, no. (%)||12 (8.6)||4 (15.1)||3 (23.1)|| || |
|Sacroiliac, no. (%)||2 (1.4)||0 (0)||3 (23.1)|| ||0.02|
Table 4. Mean values of the clinical variables of JIA activity at cross-sectional evaluation*
|Physician's global assessment of disease severity||3.68||3.48||4.65|| || |
|Parent's global assessment of overall well being||2.04||1.65||2.79|| || |
|Parent's assessment of pain||1.71||1.98||3.07|| || |
|CHAQ disability index||0.29||0.37||0.49|| || |
|CHQ physical score||44.8||47.4||43.3|| || |
|CHQ psychosocial score||49.1||47.8||52.0|| || |
|No. joints with active arthritis||2.28||3.10||7.54|| ||0.006|
|Global articular severity score||6.22||10.03||21.54|| ||0.02|
|White blood cells (× 109/liter)||8,310.6||8,012.7||7,692.2|| || |
|Hemoglobin (g/dl)||12.4||12.4||12.3|| || |
|Platelets (× 109/liter)||346.4||362.1||322.8|| || |
|Erythrocyte sedimentation rate (mm/hour)||17.5||19.8||17.4|| || |
|C-reactive protein (mg/dl)||1.34||1.78||2.27|| || |
As compared with group 2, patients in group 3 were less frequently ANA positive (P = 0.03), more commonly received second-line drug therapies (P = 0.04), had a greater number of joints involved at disease presentation (P = 0.04) and in the whole disease course (P = 0.006), had a more common upper limb predominance (P = 0.002), experienced a more frequent polyarticular extension (P = 0.02), and more frequent involvement of metacarpophalangeal joints (P = 0.02), distal interphalangeal joints (P = 0.005), and sacroiliac joints (P = 0.02). The higher frequency of dactylitis in group 3 was expected because this feature is part of the ILAR criteria for JPsA. At cross-sectional assessment, patients in group 3 had a significantly greater number of joints with active arthritis and global articular severity score than patients in group 2, whereas all the other variables were comparable across the 2 groups (Table 4).
The choice of the ILAR Committee to exclude patients with a positive family history of psoriasis was dictated by the concern that they could “contaminate” the oligoarthritis category of JIA, thus compromising the search for homogeneity within groups (9). Recent studies of the applicability of the ILAR criteria in different pediatric rheumatology populations, however, have shown that this exclusion has a major impact on patient classification, preventing appropriate grouping of most of the patients who fell into the category “other arthritis.” The 5 studies in which this information is available (4, 5, 7, 9, 10) included a total of 806 patients, 126 (16%) of whom fit in the category “other arthritis.” If the positive family history of psoriasis was removed from the exclusions for the oligoarthritis category, 70 additional patients, representing 56% of those unclassifiable and 9% of the whole cohort, would have been classifiable as having oligoarthritis by the ILAR criteria. However, it is unclear whether inclusion of children with a positive family history of psoriasis would dilute the homogeneity of the population in the oligoarthritis category.
To answer this question, we examined all consecutive children with oligoarticular-onset JIA not fulfilling the ILAR criteria for JPsA seen at our department over a 15- month period. We compared the clinical and laboratory features and the disease course over time between those who had a positive family history of psoriasis and those who did not. To validate the history of psoriasis in first or second degree relatives, we did not rely on information written on patient records, but asked parents at the time of cross-sectional assessment, requiring that psoriasis was diagnosed by a dermatologist. Our analysis revealed that, with the exception of a slightly higher frequency of females in the group with a positive family history of psoriasis, the 2 groups were comparable regarding the demographic, clinical, and laboratory features; the distribution and course of joint involvement over time; and the clinical variables of JIA activity and severity recorded at the time of cross-sectional assessment. Furthermore, the frequency of dactylitis, which is regarded as a distinctive feature of joint involvement in JPsA, was comparable across the 2 groups.
To obtain further information on the clinical relationship between patient groups, the features of patients with a positive family history of psoriasis were contrasted to those of consecutive patients with oligoarthritis seen in the study period who fulfilled the ILAR criteria for JPsA. This analysis showed several significant clinical differences between the 2 groups. With respect to patients with a family history of psoriasis who did not fulfill criteria for JPsA, patients meeting these criteria were less frequently ANA positive and appeared to have a more aggressive arthritis, as shown by the greater number of joints affected throughout the disease course, the worse articular severity variables at cross-sectional assessment, and the higher frequency of previous second-line drug therapy. The distribution of joint involvement was also different, with patients fulfilling criteria for JPsA having experienced more upper limb predominance and involvement of metacarpophalangeal, distal interphalangeal, and sacroiliac joints. The clinical features recorded in our patients meeting the ILAR criteria for JPsA were similar to those of JPsA patients reported in the literature (reviewed in15).
We acknowledge the limitations associated with the cross-sectional design of our study. Because the patient sampling reflects both prevalent cases and incident cases, the differences in the length of followup might have led to bias. Furthermore, it is possible that patients in long-term remission were not seen during the study period, leading to an overrepresentation of patients with a more active disease. However, since the duration of patient recruitment was 15 months, it is likely that the large majority of patients followed were included, including those with a mild disease who were attending the clinic just for their annual review.
In conclusion, we found close similarities in the clinical features and course among patients with oligoarthritis who had a positive family history and those who did not. Furthermore, the comparison of the former patients with those meeting the ILAR criteria for JPsA revealed several distinctive clinical differences. These findings argue against the exclusion of patients with a positive family history of psoriasis who do not fulfill the ILAR criteria for JPsA from the oligoarthritis category of JIA.