The use of a graded exercise test may be insufficient to quantify true changes in V̇o2max following exercise training in unfit individuals with metabolic syndrome.

We studied the accuracy of graded exercise testing (GXT) to assess improvements in maximal oxygen uptake (V̇o2max) with exercise training in unfit individuals with metabolic syndrome (MetS). Forty-four adults with MetS (58 ± 7 yr, 36% women, BMI 31.8 ± 4.8 kg/m-2) underwent 4 mo of supervised high-intensity interval exercise training. V̇o2max was assessed using GXT, followed by a constant-load verification test (VerT) at 110% of the maximal work rate achieved during GXT. V̇o2 data from GXT and VerT were compared using repeated-measures ANOVA. The mean improvement in V̇o2max following exercise training was similar when using GXT only or VerT. However, before training, 18 subjects achieved a higher V̇o2max during the verification test that was (+159 mLO2/min) higher than the GXT (P < 0.001). After training, the underestimation of V̇o2max by GXT was reduced but still present (+64 mLO2/min). As a result, improvements in V̇o2max following exercise training as assessed using GXT only almost doubled the "real" increase in V̇o2max as measured by VerT in these 18 individuals. In the remaining 26 subjects, GXT scored below VerT only after training (+54 mLO2/min, P = 0.046). As a consequence, GXT underestimated the actual V̇o2max increases (-49 mLO2/min, P = 0.013) in these individuals. Assessment of changes in V̇o2max following exercise training using only GXT over- or underestimates V̇o2max gains in unfit individuals with MetS. Thus, a verification test may be required to 1) identify the highest V̇o2max during a maximal exercise test on a cycle ergometer and 2) accurately quantify the true changes in cardiorespiratory fitness following exercise training in unfit individuals with MetS.NEW & NOTEWORTHY It is unclear whether the traditional GXT is suitable to assess V̇o2max changes in unfit individuals with metabolic syndrome. Mean changes in V̇o2max following exercise training were similar using GXT or VerT. However, we showed that the GXT overestimated V̇o2max improvements in 41% and underestimated V̇o2max improvements in 59% of subjects. Our data suggest the need for a verification test to appropriately determine training-induced improvements in V̇o2max in unfit individuals with metabolic syndrome.