Surface Electromyography of the Pelvic Floor Musculature: Reliability and Validity of a Novel Electrode Design
MetadataShow full item record
Purpose: Intravaginal probes used for recording electromyography (EMG) from the pelvic floor muscles (PFMs) likely record activity from nearby muscles (crosstalk), and move during functional tasks, causing motion artifact data contamination, threatening the validity of results obtained. This study investigated the test-retest reliability and validity of surface EMG recordings from the PFMs using a novel, theoretically superior electrode in comparison to a commercially available intravaginal probe, the FemiscanTM. Methods: Healthy subjects (n=20) performed tasks with each vaginal electrode in situ: i) PFM maximal voluntary contractions (MVC), ii) coughs, iii) unilateral hip adductor/external rotator contractions at 25%MVC, 50%MVC, and MVC while keeping the PFMs relaxed or maximally contracted, and iv) transversus abdominis contractions (TrA; recorded using fine-wires) at 25%MVC, 50%MVC, MVC. Analyses: i) Intraclass correlation coefficients (ICC), ii) t-tests of proportions (α=0.05), iii) repeated measures ANOVAs and Tukey’s post-hoc testing (α=0.05) and iv) cross-correlation functions between peaks of transversus abdominis and PFM activity were used to determine the between-trial and between-day reliability of each vaginal electrode, a difference in prevalence of motion artifact contamination between electrodes, and the presence of crosstalk from the hip and TrA, respectively. Results: Between-trial reliability of both vaginal electrodes was excellent (ICC(3,1)=0.943-0.974). Between-day reliability was less consistent (ICC(3,1)=0.788-0.924 and 0.648-0.715 for the FemiscanTM and novel electrode, respectively). No significant difference in the proportion of files contaminated with motion artifact using each electrode existed. At submaximal intensities of hip muscle contractions, the FemiscanTM recorded significantly higher EMG amplitudes compared to what it recorded when the hip was relaxed, whereas the novel electrode did not, indicating that the FemiscanTM recorded crosstalk from the hip musculature. Low cross-correlation coefficients (<0.90) and large time delays (≥ 0.5 milliseconds) between peaks of PFM and TrA activity indicated that neither vaginal electrode recorded crosstalk from the TrA. Conclusion: The novel electrode is a promising tool to record EMG from the PFMs, as it records less crosstalk from the hip musculature than current technology while maintaining a high degree of reliability when comparing results collected within the same session; however, this electrode should not be used to compare one’s muscle activity between days.