We read with great interest the recent analysis of the uterine electromyogram, most commonly referred to as electrohysterogram (EHG), that was proposed by Lucovnik et al.
We have previously advanced the hypothesis that the analysis of the propagation velocity (PV) in combination with other parameters that are derived by the EHG has potential value for discriminating between productive and unproductive uterine contractions. We believe that the impact of this study, which is the first to test this hypothesis on a broad population, is such that it deserves a thorough discussion on some fundamental methodologic aspects, which were not fully tackled in the article.
At a first glance, values and variability of PV that are found in the labor groups are surprisingly up to 10 times higher than those previously reported. Unfortunately, the authors did not comment on this.
We believe that the most plausible explanation to this discrepancy resides in the combination of the recording methods with the nature of uterine contractions. The article suggests that the PV was estimated by the analysis of 2 differential electrode couples; however, their relative position was not specified. The nature of uterine contractions is such that origin and direction of propagation of the EHG are a priori unknown. Therefore, EHG analysis poses demanding requirements for the number and relative position of the recording channels that are used for deriving the actual value of PV (ie, the value of velocity along the propagation direction). In general, the use of 2 recording channels may lead to PV values that do not correspond to the actual values.
Some possible EHG propagation patterns are represented schematically in the Figure . With reference to these examples, the approach of Lucovnik et al leads to the actual value of PV only for example 1. For the other cases (highly expected in EHG recording), we would erroneously estimate PV values even 10 times higher than the actual ones.
