Poor reliability of heart rate monitoring to assess oxygen uptake during field training.

The aim of this study was to quantify the error associated with the assessment of oxygen uptake (VO2) by heart rate (HR) monitoring in 15 athletes freely performing their training session. A laboratory-derived equation was used on the running track to calculate VO2 (VO2est) from HR. Oxygen uptake was also assessed by means of a portable gas analyzer (VO2real). Bland and Altman statistics were carried out in order to evaluate agreement between VO2real and VO2est. The excess in carbon dioxide production (CO2excess) was measured to test the hypothesis that the subject with the highest CO2excess had the highest VO2est - VO2real difference. VO 2real was on average 0.14 l x min (-1) below VO2est. The limits of agreement of this difference were between + 0.77 and - 0.48 l x min (-1). Thus, on average VO2real was overestimated by VO2est. Spearman non-parametric statistics found a significant correlation between CO2excess and VO2est - VO2real difference (R = 0.55, p = 0.031). This study demonstrates that the use of HR monitoring to assess VO2 during field training overestimates VO2real, especially when a substantial increase in CO2excess occurs. Therefore, this method should be used with care when an excessive amount of CO2 is produced as occurs in activities that involve the lactic acid anaerobic source of energy production.