Objective: To evaluate the performance of boys with DMD in the ten meter walking test (10 MWT) in order to predict gait loss. Method: This longitudinal study consisted of consecutive evaluations, minimum of 3 and maximum of 12, conducted every 4 months, during 33 months, depending on time of inclusion in the study. Ambulant boys with DMD (n=18), ages 4 to 13 years, mean 7 (SD=2.22), were assigned to Ambulatory group (A; n=11) or Non-Ambulatory group (NA; n=7) according to their status at the end of the study. Diagnosis was based on the absence of dystrophin in a muscle biopsy and/or identification of a mutation of the dystrophin-gene. The main outcome measures were: 10 MWT total time and rates between two consecutive sessions. Secondary measures included: functional status and muscle strength of the hip, knee and ankle. Results: The 10 MWT total time for the NA group oscillated over time, while remaining steady for the A group. The NA group showed mean of 16.18 s (CI 95% 14.38-17.98) and the A group showed mean of 10.2 s (CI 95% 9.08-11.24). The difference between groups was estimated as-5.98 s (CI 95%-8.11;-3.89). The linear model of mixed effects identified significant increase in 10 MWT time for the NA group and decrease for the A group. The rates were >1.25 for participants who became wheelchair users, indicating increased time to perform 10 MWT overtime. Conclusion: Rates ≥ 1.25 indicate the borderline between independent gait and wheelchair confinement and are useful for predicting gait loss.