The purpose of this study was to determine if the regression formula developed for the 20-m shuttle run test (20 MST) for children and adolescents with mild mental retardation (MR), used to predict cardiovascular fitness (V̇O_2peak), is valid for adolescents with Down syndrome (DS). Twenty-six adolescents (mean age = 15.3 ± 2.7 yr) with DS (15 males, 11 females) completed a maximal treadmill protocol (measured V̇O_2peak) and a 20 MST (predicted V̇O_2peak). There was a significant difference (p < .01) between the means of the measured (25.5 ± 5.2 ml·kg^-1-·min^-1) and the predicted (33.5 ± 3.9 ml·kg^-1·min^-1) V̇O_2peak, respectively. In addition, there was a low relationship between measured and predicted values (r = .54). The results of this study indicate that the regression formula developed for children and adolescents with MR to predict V̇O_2peak was not valid in this sample of adolescents with DS.