Purpose The purpose of this prospective non-randomised study was to compare the efficacy of two opposed methods, operative and conservative. Our hypothesis was that if the method was selected correctly, on an individual basis, the results should be approximately equal. Methods The study included 37 adolescents aged between 12 and 16 years, with a mean follow up of 6.1 years. The presence of a significant loose body, confirmed by precise imaging, was the key for selecting operative or arthroscopic treatment. In both groups of patients, we evaluated functional knee scores and the incidence of residual patellofemoral disorders.Results We confirmed our hypothesis using the t-test to compare functional results and a test for comparison of proportions for incidence of residual disorders. There was no statistically significant difference (p=0.091) between operatively and conservatively treated groups with regard to functional results. The same statistical outcome emerged when comparing incidences of re-dislocation (p=0.854), or other major patellar instabilities (p=0.856), between the groups. Conclusions The results obtained should not promote a non-operative method on the basis of lower risk, but do support an individual approach based on precise diagnosis and defined criteria.