Reliability and educational potential of a computational model of vacuum-assisted delivery on a birth simulator

Vacuum-assisted delivery is widely used in operative vaginal deliveries but carries a risk of rare yet severe complications. Computational simulations offer a promising way to improve our understanding of the mechanisms underlying these injuries. However, these simulations remain insufficiently validated experimentally and have not yet been applied to vacuum extraction. This study aims to evaluate the reliability of these methods using a simplified training dummy and to demonstrate their educational potential by analyzing key parameters of the clinical procedure.

Reliability was assessed using a birth simulator by comparing predicted extraction forces (derived from computational simulations using the finite element method) with experimental measurements obtained from vacuum extractions performed using a Kiwi® cup on a training dummy mounted on a force platform (N = 16). The model was then used to investigate how prescribed trajectory, suction cup position, and fetal head size influence both extraction forces and maximum strain in external deformable tissues.

The maximal extraction force predicted by computational simulations was 42.3 N, while it ranged between 46.1 N and 85.3 N (mean 67.5 ± 13.3 N, N = 16) experimentally. Similar trends in the force components were obtained through the whole extraction trajectory. The most downward trajectory was shown to minimize extraction force and maximal perineal tissue strain, as well as achieving the most occipital cup placement, in line with reported recommendations.

Results demonstrated that computational simulations could reliably predict extraction forces during birth simulations on a training dummy, while also revealing associated uncertainties and a tendency to underestimate extraction forces. Simulations of various scenarios provided valuable insights that could enhance obstetrics training and complement existing methods. Although not intended for direct clinical translation, this study evaluates the relevance of computational modelling in a controlled, simulation-based training context.