No long-term data are available on the efficacy of glargine insulin in comparison with continuous sc insulin infusion (CSII) in children and adolescents affected by Type 1 diabetes (T1D). Our aim was to compare the 2-yr efficacy of the 2 insulin approaches, in order to know how to best supply basal insulin in these patients. Thirty-six 9 to 18-yr-old consecutive children with at least 3 yr previous T1D diagnosis were enrolled. As part of routine clinical care, the patients consecutively changed their previous insulin scheme (isophane insulin at bedtime and human regular insulin at meals) and were randomly selected in order to receive either multiple daily injections (MDI) treatment with once-daily glargine and human regular insulin at meals, or CSII with aspart or lispro insulin. Both groups showed a significant decrease in glycosylated hemoglobin (HbA1c) values during the 1st year of therapy, though only in the CSII treated children was the decrease also observed during the 2nd year. The overall insulin requirement significantly decreased only in the CSII group and exclusively during the 1st year, while no significant differences were observed concerning body mass index SD score, severe hypoglycemic episodes and basal insulin supplementation. The work illustrates the first long-term study comparing the efficacy of CSII to MDI using glargine as basal insulin in children. Only with CSII were better HbA1c values obtained for prolonged periods of time, so that CSII might be considered the gold standard of intensive insulin therapy also for long-term follow-ups.