Agreement between lactate threshold and critical speed in trained, highly trained, and World Class runners

To determine whether differences between laboratory-based lactate and field-based critical speed estimates of the heavy-to-severe intensity threshold vary by athlete caliber, and to assess agreement between methods.

Thirty-one (52% female) trained, highly-trained and World Class runners (McKay Tier 2 vs. 3 vs. 4/5; peak aerobic velocity = 15.5 ± 0.9 vs. 17.5 ± 1.1 vs. 19.8 ± 1.2 km/h) completed two exercise tests, ~4 days apart.

A laboratory treadmill protocol included 6-10, 4-minute stages increasing 0.6 km/h until blood lactate was >4 mmol/L followed by a 1% incline increase each minute until volitional fatigue. Threshold speeds were determined from individualized blood lactate responses using modified D-max and the onset of blood lactate. Critical speed was calculated from a linear distance-time regression using maximal efforts over 3200 m, 2000 m, and 1000 m.

There were main effects of test type (p = 0.001) and athlete caliber (p < 0.001) on threshold speeds, with no interaction (p = 0.282). Speeds increased from trained to highly-trained and World Class (p < 0.001) and highly-trained to World Class (p = 0.001). Critical speed was 0.47 [0.16-0.78 km/h, p = 0.001] and 0.36 [0.05-0.67 km/h, p = 0.016] faster than modified D-max and onset of blood lactate. Mean biases [95% limits of agreement] versus critical speed were 0.45 [-0.54-1.43 km/h] for modified D-max and 0.28 [-1.40-1.96 km/h] for onset of blood lactate.

Critical speed yielded consistently faster threshold estimates than lactate-based methods irrespective of athlete caliber, suggesting these tests not be used interchangeably.