OBJECTIVE: Ultraviolet (UV) radiation is considered to be an important environmental factor in the clinical course of children with juvenile dermatomyositis (DM). We aimed to evaluate the association between UV radiation and severe disease outcomes in juvenile DM. METHODS: This is a cross-sectional study of patients with juvenile DM enrolled in the US multicenter Childhood Arthritis and Rheumatology Research Alliance (CARRA) Legacy Registry from 2010 to 2015. The mean UV index (UVI) in the calendar month prior to symptom onset in each subject's zip code was calculated from daily satellite solar noon measurements. Multivariable logistic regression was used to model the relationship between the mean UVI and calcinosis as well as other outcomes of severe disease. Covariates included sex, race, age, time to diagnosis, disease duration, and latitude. RESULTS: In a multivariable model, there was no association between the mean UVI and calcinosis. African American race was associated with a 3-fold greater odds of calcinosis. However, there was a significant statistical interaction between race and mean UVI. Accounting for this interaction, the odds of calcinosis markedly decreased in African American subjects and steadily increased in non-African American subjects over a range of increasing the mean UVI. Higher mean UVI was associated with decreased odds of using biologics or nonmethotrexate disease-modifying antirheumatic drugs and skin ulceration. CONCLUSION: We described a novel association between UV radiation, calcinosis, and race in a large cohort of patients with juvenile DM. This study furthers our knowledge of the role of UV radiation in the clinical course of juvenile DM and highlights the complex interplay between genes and environment in the clinical phenotypes and development of calcinosis in children with juvenile DM.
OBJECTIVE: Ultraviolet (UV) radiation is considered to be an important environmental factor in the clinical course of children with juvenile dermatomyositis (DM). We aimed to evaluate the association between UV radiation and severe disease outcomes in juvenile DM. METHODS: This is a cross-sectional study of patients with juvenile DM enrolled in the US multicenter Childhood Arthritis and Rheumatology Research Alliance (CARRA) Legacy Registry from 2010 to 2015. The mean UV index (UVI) in the calendar month prior to symptom onset in each subject's zip code was calculated from daily satellite solar noon measurements. Multivariable logistic regression was used to model the relationship between the mean UVI and calcinosis as well as other outcomes of severe disease. Covariates included sex, race, age, time to diagnosis, disease duration, and latitude. RESULTS: In a multivariable model, there was no association between the mean UVI and calcinosis. African American race was associated with a 3-fold greater odds of calcinosis. However, there was a significant statistical interaction between race and mean UVI. Accounting for this interaction, the odds of calcinosis markedly decreased in African American subjects and steadily increased in non-African American subjects over a range of increasing the mean UVI. Higher mean UVI was associated with decreased odds of using biologics or nonmethotrexate disease-modifying antirheumatic drugs and skin ulceration. CONCLUSION: We described a novel association between UV radiation, calcinosis, and race in a large cohort of patients with juvenile DM. This study furthers our knowledge of the role of UV radiation in the clinical course of juvenile DM and highlights the complex interplay between genes and environment in the clinical phenotypes and development of calcinosis in children with juvenile DM.