PFAPA syndrome is the most frequent periodic fever syndrome in non-Mediterranean patients. The pathogenesis is unclear and the treatment is purely symptomatic and not standardized. The aim of this study was to assess colchicine's efficacy as prophylactic treatment in PFAPA syndrome and to identify factors able to predict response to treatment.

We performed a retrospective, multicentric, cohort study of PFAPA patients under colchicine prophylaxis. PFAPA diagnosis was established according to Feder's criteria. Medical records were reviewed and analyzed for demographic, clinical and laboratory data. We distinguished one responder's group, defined as patients who had no more or twice fewer crises under colchicine and another one of non-responders. Subgroup analyses were performed using non-parametric Mann-Whitney test for quantitative data and calculating odds ratio and confidence interval for qualitative data. Difference between the two groups was considered significant for P-value<0.05 or a confidence interval different from 1.

Twenty children, 65% of whom were boys, were analyzed. Their mean age at disease onset was 2.3±1.5 years. Among the nine responder patients, five were MEFV (71%) heterozygotes: M694V mutation in four and V726A once. Heterozygous MEFV gene mutation tended to be more frequent in the responders group (71% versus 43%; OR=0.3 [0.03–2.7]). Non-responder patients had more chronic fatigue (82% versus 33%; OR=9 [1,14–71]) and had more oral aphtosis (82% versus 11%; OR=36 [1,7–141]) than the responders ones. Although not significant, colchicine treatment appeared more effective in patients with less complete PFAPA phenotype and MEFV heterozygosity.