BACKGROUND To recognise the relationship between the needle tip and the median nerve during peripheral nerve block is of interest to avoid neural damage. However, signs of intraneural injection are not clearly established. The aim of this study was to define the changes observed in the peripheral nerve after the intraneural or perineural administration of 1ml of solution. METHODS Ultrasound guided median nerve blocks were performed in the forearm of 10 fresh cadavers on 60 occasions (3 per forearm). They were randomised into the intraneural (n=30) or perineural (n=30) location of the needle tip, after the consensus of location by 7 specialists. After 1ml of solution was injected an evaluation was made of the changes in the cross-sectional area of the nerve, as well as the displacement along the nerve. RESULTS The cross-sectional area of the median nerve was increased in both groups, however, the increase was significantly higher in the intraneural group (perineural 0.007±0.013cm2 vs. intraneural 0.032±0.021cm2, P<.0001). An increase of more than 27% of the area ensures an intraneural injection in the median nerve according to the ROC curve analysis. Both proximal and distal diffusion were observed more frequently in the intraneural group (proximal: 86% vs 14%, P<.0001, Distal: 43% vs 4%, P<.0001). CONCLUSIONS Based on this experimental study, it is concluded that the injection of a small volume (1ml) allows to discriminate the disposition of the intraneural vs perineural needle in a high percentage of cases. Therefore, it is suggested that this "dose test" should be considered in the safety algorithms if it is required to reduce the incidence of intraneural injection.