Placental lesions in stillbirths: A case-control study using the Amsterdam criteria and predictive models at a UK tertiary unit

Introduction

The UK stillbirth rate remains higher than in many high-income countries, with placental disorders -particularly maternal vascular malperfusion (MVM) lesions -linked to adverse maternal and fetal outcomes. This study examines placental lesions in stillbirth at one of the largest maternity units in the UK using the Amsterdam criteria for histological classification. It also retrospectively examines whether women with global/partial MVM – where most maternal decidual vessels show pathological changes but are only partially occluded- would have received aspirin and further surveillance if their placental dysfunction risk had been assessed using the Fetal Medicine Foundation (FMF) algorithm from the Tommy’s app in their first trimester.

Materials and methods

We conducted a case-control study of spontaneous non-anomalous stillbirths (≥24 weeks) at Sheffield maternity unit from 2018 to 2021 (n = 83). We then compared singleton stillbirths at term with matched livebirths. Placental lesions were categorised with the Amsterdam criteria. Using the FMF’s algorithm which has only been recently introduced in our unit, we then retrospectively calculated the risk for placental dysfunction in women who experienced preterm PET stillbirth and also in those whose placentas showed global/partial MVM.

Results

MVM was the most common placental lesion in stillbirths, significantly more frequent than in livebirths (p < 0.001). The FMF algorithm had higher predictive accuracy for PET than the traditional NICE model in stillbirths [AUC: 0.76 (95% CI 0.65–0.86) vs 0.51 (95% CI 0.39–0.63), p = 0.03], but only when at least one continuous variable such as PAPP-A was included. In women with stillbirth and whose placentas showed global/partial MVM, first-trimester placental risk assessment using the FMF algorithm during the first trimester would have identified most of them as high risk [FMF AUC: 0.7 (0.58–0.80), p = 0.02].

Conclusion

MVM is frequently found in stillbirths. Our retrospective placental dysfunction risk assessment suggests that Tommy’s algorithm would have more accurately identified women who went onto experience stillbirth with significant MVM lesions as high risk, leading to aspirin treatment and closer monitoring. Further research is needed to confirm these findings and potentially enhance placental dysfunction screening to reduce stillbirth rates.