A diagnostic algorithm to evaluate elevated C-reactive protein levels in juvenile systemic lupus erythematosus

C-reactive protein (CRP) levels are generally not correlated with systemic lupus erythematosus (SLE) disease activity. We aimed to develop an algorithm to evaluate juvenile SLE (JSLE) patients with elevated CRP. JSLE patients diagnosed at <18 years were included. Each episode of CRP elevation was evaluated separately. Of 190 JSLE patients (F/M:4/1), 88 (46.3%) never had an elevated CRP, while 102 (53.7%) had 174 episodes of CRP elevation. Causes were infection (n = 139), arthritis (n = 15), macrophage activation syndrome (MAS) (n = 9), MAS and infection (n = 3), serositis (n = 6), and vasculitis (n = 2). MAS was more common in the SLE disease activity index (SLEDAI)>4 group, while infections were more frequent in the SLEDAI ≤ 4 group. MAS episodes were more prevalent among patients with CRP >2x the upper limit of normal. We developed an algorithm to prioritize etiology in JSLE patients with elevated CRP. It led to the correct etiology in 164 of 165 episodes (99.4%) in the primary cohort. In an external JSLE cohort including 37 patients with 68 elevated CRP episodes, the algorithm led to the correct etiology in 67 (98.5%). Our algorithm could assist physicians while evaluating elevated CRP episodes in JSLE patients. Validation in larger cohorts may improve its performance.