Standardization of the sonographic measurement of the head to perineum distance: A pilot study comparing techniques with and without perineal compression

Objectives: To evaluate the relationship between head-to-perineum distance (HPD) measurements obtained with (compressed HPD) and without (uncompressed HPD) compression of the perineum, and to establish a conversion rule between these measurements. Additionally, we investigated whether the body mass index (BMI) of the participant impacted these two types of measurements. Study design: Prospective cohort study conducted at the University Hospital of Parma, Italy, involving a non-consecutive cohort of singleton pregnancies with cephalic presenting fetus at 37 weeks or more. HPD was sonographically assessed using two different techniques: 1) applying compression on the soft tissues of the perineum (compressed HPD), or 2) without compressing the perineum (uncompressed HPD). The main outcome was the development of a conversion model between both HPD measurements. The secondary outcome was the relationship between BMI and HPD measurements. We identified the best regression model by systematically exploring polynomial regression (orders 1 and 2) adjusted with restricted cubic splines or fractional polynomials, applying different variable transformations. Results: From July 2023 to June 2024, we included 109 participants, of which 78.9 % were nulliparous with a mean age of 32.4 ± 5.2 years, and a body mass index at inclusion of 26.7 ± 4.2. The best model was a log–log transformation of both HPD variables, fitted with a 2nd-degree polynomial and restricted cubic splines with 2 knots at the 33rd and 66th percentiles. The interaction between BMI and compressed HPD did not significantly alter the model (ΔR2 = 0.001 and ΔF = −19.9, p = 0.25). Subsequently, we constructed a conversion table that provides estimated uncompressed HPD values (in mm) for a range of compressed HPD measurements (in mm), using the best regression model. Conclusion: In this pilot study, we developed a regression model using restricted cubic splines to convert between both HPD measurements. BMI did not seem to significantly influence the relationship between the two HPD measurements, suggesting the model's applicability across different body compositions. Further validation of our model in a larger cohort is needed before clinical implementation.