Sampling matters: the interplay of physiological parameters and performance over 1, 2 and 3 km in well-trained male endurance runners and sprinters

Aim To examine characteristics associated with running performance in heterogeneous and endurance-only samples of well-trained male runners. Methods Twenty well-trained (V̇O 2max : 65.7 ± 5.5 ml·min -1 ·kg -1 ) male endurance runners (n=16) and sprinters (n=4), completed comprehensive laboratory testing and 100-m sprint as well as, 1-, 2-, and 3-km time-trials. Physiological markers included fractional utilization at maximal lactate steady-state (%V̇O 2max ), maximal fat oxidation rate (MFO), maximal lactate accumulation rate (ċLa max ) and peak post-exercise lactate following the ramp test (La RT ). Hierarchical regression and bivariate correlations with Benjamini-Hochberg correction were used to examine performance associations and the influence of excluding sprinters. Results Baseline models (V̇O 2max + running economy) explained <40% of 1000-m and >60% in 2000- and 3000-m performance in the pooled sample. MFO increased explained variance by 15.4% at 2000 m and +14.81% at 3000 m, while %V̇O 2max added 11.3% at 3000 m. In the pooled sample, oxidative characteristics became more strongly associated with performance as distance increased, whereas several non-oxidative characteristics were positively associated with sprint performance and negatively associated with longer-distance performance. After excluding sprinters, associations of V̇O 2max and MFO with 2000- and 3000-m performance were retained, whereas associations involving ċLa max and La RT were attenuated. ASR remained negatively associated with 3000-m performance in both samples. Conclusion In well-trained male runners, associations between physiological characteristics and running performance depend on event distance and sample composition. Oxidative characteristics were particularly relevant to 2000- and 3000-m performance, whereas the apparent relevance of several non-oxidative characteristics partly reflected the inclusion of sprinters. These sampling-related findings require confirmation in larger, balanced cohorts.