PURPOSE OF THE STUDY Radial head fractures in paediatric patients account for 5-19% of all elbow injuries and approximately 1% of all fractures in children. Non-displaced fractures are treated with plaster cast fixation. If the fracture is displaced, we proceed to closed reduction, or to osteosynthesis in case of unstable fragments. If closed reduction fails, we opt for open reduction and osteosynthesis. The prospective randomised clinical study aims to compare the two methods of minimally invasive osteosynthesis using the pre-bent Kirschner wire or Prévot nail and to identify differences between them. MATERIAL AND METHODS The prospective randomised clinical study was conducted in 2015-2019. The final cohort included 32 patients who met the inclusion criteria. The patients in whom other osteosynthesis implants had been used or in whom open reduction had to be performed were excluded from the study. Also excluded were the patients with serious concomitant injuries of elbow. For patients included in the cohort, demographic data, precise evaluation of the displacement and location of the fracture as well as the duration of plaster cast fixation and osteosynthesis implants used were recorded. In the clinical part, the methods were compared based on the achievement of full range of motion in dependence on the degree of original displacement, use of osteosynthesis implant, and occurrence of early and delayed complications. Clinical and radiological outcomes were compared. In both types of minimally invasive osteosynthesis, Métaizeau surgical technique was used. RESULTS Based on the clinical trial criteria, 26 (81%) excellent, 4 (13%) good and 2 (6%) acceptable outcomes were achieved. In 3 cases the loss of rotation was up to 20°, in 1 case the loss of flexion was up to 10°. In one patient the loss of flexion was 15° and rotation up to 30°. In another patient the loss of rotation was up to 40°. The radiological assessment showed 14 (44%) excellent outcomes, 15 good (47%) and 3 (9%) acceptable outcomes. The statistical analysis of both the groups of the cohort using non-parametric tests revealed no statistically significant differences in individual demographic parameters. The comparisons of both types of osteosynthesis in dependence on the degree of displacement by non-parametric Fisher's exact test showed no statistically significant difference in the radiologic or clinical results. The only statistically significant difference was observed in the duration of metal implant placement. DISCUSSION Comparable studies report excellent or good clinical outcomes in 80-95% of cases (1,13,16). In our cohort, excellent or good clinical outcomes were achieved in 30 patients (94%). In two patients, in whom Prévot nail was used, the outcomes were acceptable. Nonetheless, this fact did not result in any statistical significance when comparing the two methods separately or in comparisons based on the degree of displacement. CONCLUSIONS The comparison of the two methods of minimally invasive osteosynthesis revealed no statistically significant difference, namely not even when both the methods were compared based on the degree of displacement. When Kirschner wire is used, the possibility to remove the metal implant in the outpatient setting is considered to be an advantage. The drawback consists in potential penetration of the sharp Kirschner wire in the radiocapitellar joint, which we did not encounter when the second technique of osteosynthesis was used. The advantage of Prévot nail includes a lower risk of pin-tract infection. Key words: minimally invasive osteosynthesis, radial head, fracture, child.