In the article by Williams et al (2013) ‘A preliminary evaluation of surface electromyography as a tool to measure muscle fatigue in the National Hunt racehorse’ published in The Veterinary Nurse 4(9): 566–71, the data analysis protocol reported for mean frequency contained an error. The correct analysis protocol, subsequent amended results and conclusions are reported below. The authors apologise for this error.
Method: evaluation of mean amplitude frequency (Hz) between runs and to trainer perceived fitness level
Following application of the inbuilt Delsys’ filter to remove noise, initially the mean and median amplitude frequencies of the remaining signal were calculated via EMG Works™. Each of these parameters are considered valuable in the detection of fatigue, but as they are highly correlated, only one is required to assess fatigue changes (Kamen and Gabriel, 2010). Mean frequency was selected for analysis; a consistent decrease in mean frequency implies a reduction in muscle workload corresponding to fatigue (Hanon et al, 2005). Therefore the mEMGF for individual horses were calculated and were plotted over time at 0.125 s intervals with 0.0625 window overlap, for each piece of canter work to visually assess fitness and fatigue levels for the duration of the training periods that were recorded.
Results: fitness and fatigue
Interval training in this cohort of National Hunt racehorses recorded a range of amplitude frequencies between 26 and 120 Hz. Analysis of mEMGF between runs found that significant lateral differences were present across the cohort (p<0.05), in contrast to what was previously reported. Within the individual horses significant lateral differences were present in 67% of the population: horse 1 (p<0.001), horse 2 (p<0.006), horse 3 (p<0.01), horse 5 (p<0.05), horse 7 (p<0.03) and horse 8 (p<0.01).
In contrast, mEMGF did not vary between runs across the cohort (p>0.05) but significant differences were exposed between the first and second runs in three horses (p<0.001) and the first and third runs in two horses (p<0.05). Within individual horses, a significant difference was exposed in mEMGF between the first and second runs in three horses (p<0.001) and between the first run and third run in two horses (p<0.01). Interestingly, no relationship between fitness level (as predetermined by the trainer) and mEMGF was exposed (p>0.05).
Discussion and conclusion
mEMGF varied throughout runs suggesting increased and decreased muscle effort occurred during training and there was no evidence that fatigue was present in any of the horses. The conclusions drawn in the original article remain valid.