Computerized cardiotocography and obstetric cholestasis: is it a useful tool?

Intrahepatic cholestasis of pregnancy (ICP) is associated with an increased risk of adverse perinatal outcomes, including stillbirth. While computerized cardiotocography (cCTG) is a standard tool for fetal surveillance, the specific impact of maternal biochemical severity on cCTG parameters remains poorly defined. Specifically, the qualitative aspects of fetal heart rate accelerations-rather than just their frequency-may provide deeper insights into fetal wellbeing under conditions of metabolic stress induced by elevated bile acids.

To evaluate the association between maternal bile acid levels and cCTG parameters in pregnancies complicated by ICP, with a primary focus on the quality and distribution of accelerations (small vs. large) as a potential marker of fetal status.

We conducted a retrospective observational cohort study pooling two tertiary referral centers. The pooled cohort included 94 pregnancies with ICP, contributing 368 Sonicaid cCTG recordings across gestation (50 at <32 weeks, 55 at 32 to <34 weeks, and 263 at ≥34 weeks). For each pregnancy, peak bile acids (peak BA), defined as the maximum total serum bile-acid concentration recorded during pregnancy, were used as a proxy of biochemical severity and modeled as log2(peak BA), so effects represent a doubling in bile acids; a sensitivity analysis used peak BA ≥ 40 μmol/L. From each Sonicaid™ cCTG tracing we extracted short-term variability (STV), acceleration counts, and software-defined small and large accelerations. Acceleration quality was summarized as Small-share = small/(small + large) and Small/Large = small/large. Associations were tested using generalized linear models with robust standard errors clustered by pregnancy, stratified by gestational age (<32 weeks, 32 to < 34 weeks, and ≥ 34 weeks) and adjusted for gestational age within each band.

Across the 368 cCTG recordings, bile acids were not consistently associated with STV or total acceleration rate. A gestational-age-specific signal emerged for acceleration quality: in the 32 to less than 34-week window, higher bile acids were associated with a shift toward smaller accelerations (OR 1.146 per bile-acid doubling; 95% CI 1.029-1.275; p = 0.013), whereas no clear associations were observed at <32 weeks or ≥34 weeks. The association remained directionally consistent in sensitivity analyses using peak BA ≥40 micromol/L (OR 1.406; 95% CI 1.073-1.843; p = 0.014). Neonatal outcomes were variably available and were summarized descriptively.

In ICP, higher maternal bile acids are associated with a qualitative shift toward smaller cCTG accelerations specifically between 32 and 34 weeks. Acceleration quality may represent an underexplored cCTG dimension for risk stratification in ICP, but prospective validation with time-matched biochemistry and complete neonatal outcomes is needed.