Assessment of metabolic energy expenditure from indirect calorimetry is currently limited to sustained (>4 min) cyclic activities, because of steady-state requirements. This is problematic for patient populations who are unable to perform such sustained activities. Therefore, this study explores validity and reliability of a method estimating metabolic energy expenditure based on oxygen consumption (V̇O2) during short walking bouts. Twelve able-bodied adults twice performed six treadmill walking trials (1, 2 and 6 min at 4 and 5 km/h), while V̇O2 was measured. Total V̇O2 was calculated by integrating net V̇O2 over walking and recovery. Concurrent validity with steady-state V̇O2 was assessed with Pearson's correlations. Test-retest reliability was assessed using intra-class correlation coefficients (ICC) and Bland-Altman analyses. Total V̇O2 was strongly correlated with steady-state V̇O2 (r=0.91-0.99), but consistently higher. Test-retest reliability of total V̇O2 (ICC=0.65-0.92) was lower than or comparable to steady-state V̇O2 (ICC=0.83-0.92), with lower reliability for shorter trials. Total V̇O2 discriminated between gait speeds. Total oxygen uptake provides a useful measure to estimate metabolic load of short activities from oxygen consumption. Although estimates are less reliable than steady-state measurements, they can provide insight in the yet unknown metabolic demands of daily activities for patient populations unable to perform sustained activities.