Analysis of Published Criteria for Clinically Inactive Disease in a Large Juvenile Dermatomyositis Cohort Shows That Skin Disease Is Underestimated

Abstract

Objective: The Pediatric Rheumatology International Trials Organisation (PRINTO) recently published criteria for classification of patients with juvenile dermatomyositis (DM) as having clinically inactive disease. The criteria require that at least 3 of 4 conditions be met, i.e., creatine kinase level ≤150 units/liter, Childhood Myositis Assessment Scale score ≥48, Manual Muscle Testing in 8 muscles score ≥78, and physician's global assessment of overall disease activity (PGA) ≤0.2. The present study was undertaken to test these criteria in a UK cohort of patients with juvenile DM.

Methods: We assessed 1,114 patient visits for the 4 items in the PRINTO criteria for clinically inactive disease. Each visit was analyzed to determine whether skin disease was present. The Disease Activity Score (DAS) for juvenile DM was determined in 59 patients.

Results: At 307 of the 1,114 visits, clinically inactive disease was achieved based on the 3 muscle criteria (but with a PGA of >0.2); rash was present at 65.8% of these visits and nailfold capillary abnormalities at 35.2%. When PGA ≤0.2 was one of the 3 criteria that were met, the frequency of skin signs was significantly lower (rash in 23.1% and nailfold capillary abnormalities in 8.7%). If PGA was considered an essential criterion for clinically inactive disease (P-CID), patients with active skin disease were less likely to be categorized as having clinically inactive disease (a median DAS skin score of 0 [of a possible maximum of 9] in visits where the PGA was ≤0.2, versus a median DAS skin score of 4 in patients meeting the 3 muscle criteria [with a PGA of >0.2]; P < 0.001). Use of the P-CID led to improvements in the positive predictive value and the positive likelihood ratio (85.4% and 11.0, respectively, compared to 72.9% and 5.1 with the current criteria).

Conclusion: There was a high frequency of skin disease among patients with juvenile DM who did not meet the PGA criterion for inactive disease but met the other 3 criteria. Incorporating PGA as an essential criterion for clinically inactive disease helps prevent the misclassification of patients with active skin disease.

Figure 1

Flow chart of 1,114 visits among 258 patients from the UK Juvenile Dermatomyositis Cohort and Biomarker Study, according to the Pediatric Rheumatology International Trials Organisation (PRINTO) criteria for clinically inactive disease. PGA = physician's global assessment of overall disease activity; CK = creatine kinase; CMAS = Childhood Myositis Assessment Scale; MMT‐8 = Manual Muscle Testing in 8 muscles.

Figure 2

Analysis of disease activity and laboratory results in the 411 patient visits at which 3 of the 4 criteria for clinically inactive disease were met. Each panel shows data analyzed when 1 of the 4 criteria was not met (i.e., CK in A, CMAS in B, MMT‐8 in C, and PGA in D). Each symbol represents an individual visit; horizontal lines show the median. See Figure 1 for definitions.

Figure 3

Scores according to the Disease Activity Score (DAS) for juvenile dermatomyositis in 59 patients meeting criteria for clinically inactive disease. DAS total score (A), DAS skin score (B), and DAS muscle score (C) in patients who met all 4 criteria for clinically inactive disease (group I), patients who met 3 of the 4 criteria, including the physician's global assessment (PGA) ≤0.2 criterion (group II), patients who met 3 of the 4 criteria but not including the PGA ≤0.2 criterion (group III), and patients who met 0 or 1 of the criteria (active disease [group IV]) are shown. Each symbol represents an individual patient; horizontal lines show the median. ∗ = P < 0.05; ∗∗∗ = P < 0.001.