Loss of signal (LoS) during external cardiotocography (CTG) can lead to misidentification of fetal heart rate abnormalities and misguided clinical decisions. Available quantitative data remain scarce and are more than fifteen years old.

We aimed to quantify LoS in routine care and relate it to patient characteristics, delivery mode, and neonatal outcomes.

We conducted a retrospective study of 303 external CTG recordings from February 2024 at Strasbourg University Hospital. An independent investigator blinded to outcomes measured all LoS ≥ 1mm on paper tracings, categorizing them as maternal heart rate acquisition or crude signal loss. Elevated LoS was defined as exceeding 20% of total CTG recording duration. Univariate analyses were complemented by multivariable logistic regression.

LoS occurred in 98.7% of recordings, ranging from 0.77% to 51.82%, with a mean ratio of 32% during active pushing. Elevated LoS was observed in 18% of deliveries (n=55), with no difference in neonatal outcomes between groups. Univariate associations with multiparity, absence of epidural analgesia, and shorter labor duration did not persist after multivariable adjustment for parity (OR 1.44, p=0.3), indicating confounding by labor profile. Obesity (OR 3.65, p=0.006) and absence of epidural analgesia (OR 6.44, p<0.001) were the only independent predictors.

LoS affects almost all external CTG recordings and reaches clinically significant levels during active pushing. Its association with physiological labor profiles reflects confounding rather than causation. Larger prospective studies with automated signal analysis are needed to assess its clinical impact.