This classification has been used to identify the risk of or the actual condition of obesity in populations.9 To assess cardiorespiratory

fitness, aerobic power (VO2max) was estimated using a progressive submaximal protocol on a mechanical exercise bike (Monark® Ergomedic model 828E) with 8-minute duration, with a warm-up period (4 Selumetinib clinical trial minutes) and a workload calculated based on body weight (4 minutes). The values of blood pressure, heart rate, and subjective perceived exertion were measured at rest and at each minute of the test. The children were instructed to pedal at a speed of 50 RPM, and the bikes were previously calibrated. The same bikes were used in the four periods. VO2max data are shown in absolute values (L.min−1), as well as in values relative to body mass (mL.kg−1.min−1), thus allowing for a more accurate comparison between students with different PCI-32765 order body compositions. To calculate the absolute and relative VO2max, Åstrand’s nomogram17 was used, considering the heart rate in the last minute of the exertion load. All measurements and tests followed the same standardization in all assessed periods.14 The objectivity and reproducibility of the measurements

obtained from a subsample of 30 randomly selected schoolchildren in each evaluation were calculated and used as an internal quality criterion of all measurements and tests. Variations

of objectivity and reproducibility were observed over the four time periods analyzed, as follows: body weight, 0.96 and 0.99; height, 0.97 and 0.99; and aerobic power, 0.58 and 0.88, respectively. Descriptive statistics, means, standard deviation, frequency, and percentage were used in the statistical analysis. The variation between 2008/2010 and 1978/1988 was assessed by the delta percent (Δ%). Data distribution was verified by the Kolmogorov-Smirnov test.18 A comparison among the four cardiorespiratory fitness evaluations was performed by analysis of variance with three factors (gender, nutritional Silibinin status, and decade), followed by Bonferroni multiple comparison.18 The calculations were performed using the Statistical Package for the Social Sciences (SPSS), release 18.0, and the level of significance was set at p < 0.01.18 From a database with over 16,000 evaluations from 1978 to 2011, 1,291 students met the inclusion criteria. In both genders, in the four evaluations performed during the 30-year period with ten-year intervals, the number of schoolchildren with normal weight (n = 789, 61%) was higher than overweight (n = 502). In males, the number of students with normal weight was higher than the number of those with overweight in all evaluations. In females, the same was observed in the evaluations performed in 1998/2000 and 2008/2010 (Fig. 1).