A Comparison between Different Methods of Estimating Anaerobic Energy Production.

The present study aimed to compare four methods of estimating anaerobic energy production during supramaximal exercise. Twenty-one junior cross-country skiers competing at a national and/or international level were tested on a treadmill during uphill (7°) diagonal-stride (DS) roller-skiing. After a 4-minute warm-up, a 4 × 4-min continuous submaximal protocol was performed followed by a 600-m time trial (TT). For the maximal accumulated O deficit (MAOD) method the [Formula: see text]O-speed regression relationship was used to estimate the [Formula: see text]O demand during the TT, either including (4+Y, method 1) or excluding (4-Y, method 2) a fixed Y-intercept for baseline [Formula: see text]O. The gross efficiency (GE) method (method 3) involved calculating metabolic rate during the TT by dividing power output by submaximal GE, which was then converted to a [Formula: see text]O demand. An alternative method based on submaximal energy cost (EC, method 4) was also used to estimate [Formula: see text]O demand during the TT. The GE/EC remained constant across the submaximal stages and the supramaximal TT was performed in 185 ± 24 s. The GE and EC methods produced identical [Formula: see text]O demands and O deficits. The [Formula: see text]O demand was ~3% lower for the 4+Y method compared with the 4-Y and GE/EC methods, with corresponding O deficits of 56 ± 10, 62 ± 10, and 63 ± 10 mL·kg, respectively ( < 0.05 for 4+Y vs. 4-Y and GE/EC). The mean differences between the estimated O deficits were -6 ± 5 mL·kg (4+Y vs. 4-Y, < 0.05), -7 ± 1 mL·kg (4+Y vs. GE/EC, < 0.05) and -1 ± 5 mL·kg (4-Y vs. GE/EC), with respective typical errors of 5.3, 1.9, and 6.0%. The mean difference between the O deficit estimated with GE/EC based on the average of four submaximal stages compared with the last stage was 1 ± 2 mL·kg, with a typical error of 3.2%. These findings demonstrate a disagreement in the O deficits estimated using current methods. In addition, the findings suggest that a valid estimate of the O deficit may be possible using data from only one submaximal stage in combination with the GE/EC method.