Article Text

Download PDFPDF

  1. Agna Neto1,2,
  2. Ana Filipa Mourão1,2,
  3. Filipa Oliveira-Ramos3,4,
  4. José Melo Gomes5,6,
  5. Maria Jose Santos7,
  6. Raquel Campanilho-Marques3,4,
  7. Daniela Piotto8,
  8. Clovis Artur Silva8,
  9. Paula Estanqueiro9,
  10. Joao Eurico Fonseca3,4,
  11. Maria T. Tererri8,
  12. Helena Canhão10
  1. 1Rheumatology, CHLO, Lisbon, Portugal
  2. 2CEDOC, NOVA Medical School, Lisbon, Portugal
  3. 3Rheumatology and Metabolic Bone Diseases, CHLN, Lisbon, Portugal
  4. 4Rheumatology Research Unit, IMM, Centro Académico de Medicina de Lisboa, Lisbon, Portugal
  5. 5Instituto Português de Reumatologia, Lisbon, Portugal
  6. 6Clínica Dr. Melo Gomes, Lisbon, Portugal
  7. 7Rheumatology, Hospital Garcia de Orta, Lisbon, Portugal
  8. 8Pediatric Rheumatology Unit, Universidade Federal de São Paulo, São Paulo, Brazil
  9. 9Pediatrics, Hospital Pediátrico Carmona da Mota, Coimbra, Portugal
  10. 10CEDOC, EpiDoC Unit, NOVA Medical School, Lisbon, Portugal


Background: The influence of body mass index (BMI) on Juvenile Idiopathic Arthritis (JIA) disease activity is poorly understood. In adults with Rheumatoid Arthritis, obesity has been associated with higher disease activity, while in JIA patients, a previous study has failed to find any association.

Objectives: To investigate the relationship between BMI and JIA disease activity.

Methods: This is an international, multicenter, observational, cross-sectional study. JIA patients (according to ILAR criteria) aged ≤ 18 years, registered at in Portugal and Brazil were included. Data was analysed upon records from the first registered visit. Age- and sex-specific BMI percentiles (P) were calculated based on WHO growth standard charts and categorized into underweight (P<3), normal weight (3≤P≤85), overweight (85<P≤97) and obesity (P>97). Disease activity was assessed by Juvenile Arthritis Disease Activity Score (JADAS-27). Univariate linear regression was used to examine the association of JADAS-27 with BMI categories. Two multivariate regression models were performed a) adjusting for age, gender, race, country, disease duration and JIA category (model 1); b) adjusting for those covariates plus use of DMARDs (model 2).

Results: 255 patients included, mean age 10.1±4.7 years, mean disease duration 6.3±4.9 years; 62% female; 85% Caucasian. Thirty-two percent were persistent oligoarticular, 9% extended oligoarticular, 34% polyarticular RF+, 6% systemic, 13% enthesitis-related arthritis, 5% psoriatic arthritis and 1% undifferentiated arthritis. The prevalence of underweight, normal weight, overweight and obesity was 7.5%, 65.9%, 15.7% and 11%, respectively. In the univariate linear regression, underweight was significantly associated with higher JADAS-27, compared to normal weight (B=-9.563, p<0.001), overweight (B=-10.661, p<0.001) and obesity (B=-7.422, p=0.004). Lower age (B=-0.299, p=0.012), shorter disease duration (B=-0.396, p=0.001), black race (B=6.852, p=0.033), RF+ polyarthritis (B=7.101, p<0.001), living in Brazil (B=5.357, p=0.002) and the absence of DMARD therapy (B=4.831, p<0.001) were also associated with higher JADAS-27.

In the model 1 of multivariate analysis, the same variables, except the country, remained significantly associated with higher disease activity. When DMARD therapy was added to the model (model 2), RF+ polyarthritis (B=4.447, p=0.001) and living in Brazil (B=4.728, p=0.013) were associated with higher JADAS-27. Patients with normal weight (B=-9.964, p<0.001), overweight (B=-10.316, p<0.001) and obesity (B=-9.502, p=0.001) had significantly lower activity disease, compared to underweight patients, as well as those under DMARD therapy (B=-4.858, p<0.001).

Conclusion: Despite the lack of adjustment for corticosteroids use, there seems to be an independent association between underweight and higher disease activity in JIA patients. Importantly, these results suggest that active disease can impair child’s weight gain. Further studies are needed to confirm these findings and understand the underlying mechanisms of this association.

Disclosure of Interests: None declared

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.