Studies in adults with rheumatoid arthritis reported low serum ghrelin that increased following anti-tumor necrosis factor (TNF) infusion. Data on juvenile idiopathic arthritis (JIA) are lacking. The aim of this pilot study was to explore serum ghrelin levels in patients with JIA and the possible association with anti-TNF treatment, disease activity, and nutritional status. Fifty-two patients with JIA (14/52 on anti-TNF treatment) were studied. Juvenile idiopathic arthritis was inactive in 3 of 14 anti-TNF-treated patients and in 11 of 38 non-anti-TNF-treated patients. The nutritional status, energy intake/requirements, appetite, and fasting serum ghrelin levels were assessed. Ghrelin control values were obtained from 50 individuals with minor illness matched for age, sex, and body mass index. Ghrelin levels in patients with JIA were significantly lower than in controls (P < .001, confidence interval [CI] = -101 to -331). Analysis according to anti-TNF treatment and disease activity showed that ghrelin levels were comparable to control values only in 3 patients with anti-TNF-induced remission. Ghrelin in non-anti-TNF-treated patients in remission was low. Multiple regression analysis showed that disease activity (P = .002, CI = -84.16 to -20.01) and anti-TNF treatment (P = .003, CI = -82.51 to -18.33) were significant independent predictors of ghrelin after adjusting for other potential confounders. Ghrelin did not correlate with nutritional status, energy balance, and appetite. Serum ghrelin is low in patients with JIA and is restored to values similar to those in controls following anti-TNF-induced remission. Our study provides evidence that TNF blockade is independently associated with serum ghrelin, which possibly contributes to anti-TNF-induced remission. These preliminary results could form the basis for future research.