Changes in 2000-m rowing ergometer performance physiological responses and biochemical markers during an 8-week training period in male rowers: a retrospective longitudinal study

Background 2000-m rowing ergometer performance, fixed-power physiological responses, and hematological and biochemical markers are commonly used in training monitoring, but whether these domains change in parallel within the same training period and correspond at the individual level remains unclear. Methods This retrospective longitudinal observational study analyzed routine monitoring data from 16 male rowers over 8 consecutive weeks. Before and after the training period, athletes completed a 2000-m rowing ergometer test and a six-stage fixed-power test (170–370 W) with heart rate and blood lactate measurements; hemoglobin (HGB), hematocrit (Hct), blood urea nitrogen (BUN), and creatine kinase (CK) were monitored weekly. Fixed-power heart rate, blood lactate and weekly blood markers were analyzed using linear mixed-effects models, with blood lactate analyzed after natural-log transformation. Individual associations were examined using Spearman correlations. Pre–Post differences in power at 4 mmol/L blood lactate (P4) were assessed post hoc using the Wilcoxon signed-rank test. Results: 2000-m completion time decreased from 370.39 ± 5.91 to 368.34 ± 6.25 s (Post−Pre: −2.04 s, 95% CI: −3.36 to −0.73 s, P = 0.005). The marginally averaged overall Pre–Post heart-rate difference across the six stages was +0.39 bpm (95% CI: −3.32 to 4.09 bpm, P = 0.828). For blood lactate across the six stages, the geometric mean ratio (Post/Pre) was 0.877 (95% CI: 0.728 to 1.057, P = 0.155). The estimated Post−Pre location shift in P4 was +10.3 W (95% CI: −6.2 to 19.3 W, P = 0.171). No statistically significant overall Week effect was observed for HGB, Hct, BUN, or CK. Spearman’s ρ values between individual monitoring changes and improvement in 2000-m performance ranged from −0.24 to 0.38, and all six bootstrap 95% CIs included 0. Conclusion Over the 8-week training period, 2000-m performance improved, whereas the other monitoring outcomes did not show similarly clear group-level changes, and correlations between individual monitoring changes and performance improvement had wide confidence intervals. Sport-specific performance should therefore be interpreted alongside, rather than inferred from, any single routine physiological or biochemical marker.