Introduction: it is frequent and common the use of percutaneous catheter device (PICC) in the neonates admitted at the neonatal intensive care unit for implementation drug therapy. The installation of PICC device is a painful procedure but rarely is adopted an analgesic treatment. Objective: To compare the analgesic efficacy of combining oral supply of 25% glucose solution and non-nutritive sucking (NNS) with only oral supply of 25% glucose solution in the PICC insertion procedure in neonates. Method: a randomized clinical trial of superiority carried out at the Children Institute Nursery of the Faculty of Medicine, University of São Paulo. Data were collected from October 2011 to February 2013. The sample consisted of 85 neonates randomly allocated into two groups, experimental group (EG) consisting of 43 neonates and control group (CG) with 42 neonates. Both groups received oral supply of 25% glucose solution, the volume (dose) ranging from 0.5 ml to 2.0 ml, according to the newborn weight. To the neonates of EG was offered oral supply of 25% glucose solution and also pacifier for NNS stimulation that was kept during all time of catheter insertion procedure. The dose of 25% glucose solution was fractionated into three parts and offered in three periods: 1-pre-venipuncture, the glucose solution was offered two minutes before venipuncture and it was offered the half volume of 25% glucose solution prescribed; 2-during venipuncture procedure, it was offered ¼ of 25% glucose volume prescribed and 3-during the catheter progression, it was offered two minutes after venipuncture and the dose was ¼ of oral glucose 25% volume prescribed. The primary outcome was the pain score assessed with Premature Infant Pain Profile (PIPP) in the first 30 seconds after venipuncture and 180 seconds during the progression of the PICC. Secondary outcomes were related to the duration of facial expressions (brow bulge, eye squeeze and nasolabial furrow) changes and variability in the heart rate (HR), oxygen saturation (O 2 Sat) length of neonate crying and adverse events. The newborn face and the display of monitor equipment that captured the HR and O 2 Sat were filmed and later analyzed. Data were stored in the Excel spreadsheet. Data analysis were processed in the app Minitab, version 15.1. The identification of EG (G1) and CG (G2) occurred only after has concluded statistical analysis to ensure blinding. For testing the normal distribution of quantitative data it was used Kolmogorov-Smirnov analysis, for qualitative data was used the Chi-square test or Fisher exact test. The groups were compared by t test or Mann-Whitney test. To analyse data related to the catheter progression was used ANOVA and to the missing data were adopted intention to treat analysis. The research proposal received approval by the Ethics Committee of the Hospital, where data were collected and after it was registered in the Brazilian Registry of Clinical Trials. Results: from the total of 85 neonates recruited, 19 (22.4%) were not evaluated for painful response du...