Abstract

BACKGROUND: Accurate knowledge of fetal presentation at term is vital for optimal antenatal and intrapartum care. The primary objective was to compare the impact of routine third trimester ultrasound or point-of-care ultrasound (POCUS) with standard antenatal care, on the incidence of overall and proportion of all term breech presentations that were undiagnosed at term, and on the related adverse perinatal outcomes. METHODS AND FINDINGS: This was a retrospective multicentre cohort study where we included data from St. George's (SGH) and Norfolk and Norwich University Hospitals (NNUH). Pregnancies were grouped according to whether they received routine third trimester scan (SGH) or POCUS (NNUH). Women with multiple pregnancy, preterm birth prior to 37 weeks, congenital abnormality, and those undergoing planned cesarean section for breech presentation were excluded. Undiagnosed breech presentation was defined as follows: (a) women presenting in labour or with ruptured membranes at term subsequently discovered to have a breech presentation; and (b) women attending for induction of labour at term found to have a breech presentation before induction. The primary outcome was the proportion of all term breech presentations that were undiagnosed. The secondary outcomes included mode of birth, gestational age at birth, birth weight, incidence of emergency cesarean section, and the following neonatal adverse outcomes: Apgar score <7 at 5 minutes, unexpected neonatal unit (NNU) admission, hypoxic ischemic encephalopathy (HIE), and perinatal mortality (including stillbirths and early neonatal deaths). We employed a Bayesian approach using informative priors from a previous similar study; updating their estimates (prior) with our own data (likelihood). The association of undiagnosed breech presentation at birth with adverse perinatal outcomes was analyzed with Bayesian log-binomial regression models. All analyses were conducted using R for Statistical Software (v.4.2.0). Before and after the implementation of routine third trimester scan or POCUS, there were 16,777 and 7,351 births in SGH and 5,119 and 4,575 in NNUH, respectively. The rate of breech presentation in labour was consistent across all groups (3% to 4%). In the SGH cohort, the percentage of all term breech presentations that were undiagnosed was 14.2% (82/578) before (years 2016 to 2020) and 2.8% (7/251) after (year 2020 to 2021) the implementation of universal screening (p < 0.001). Similarly, in the NNUH cohort, the percentage of all term breech presentations that were undiagnosed was 16.2% (27/167) before (year 2015) and 3.5% (5/142) after (year 2020 to 2021) the implementation of universal POCUS screening (p < 0.001). Bayesian regression analysis with informative priors showed that the rate of undiagnosed breech was 71% lower after the implementation of universal ultrasound (RR, 0.29; 95% CrI 0.20, 0.38) with a posterior probability greater than 99.9%. Among the pregnancies with breech presentation, there was also a very high probability (>99.9%) of reduced rate of low Apgar score (<7) at 5 minutes by 77% (RR, 0.23; 95% CrI 0.14, 0.38). There was moderate to high probability (posterior probability: 89.5% and 85.1%, respectively) of a reduction of HIE (RR, 0.32; 95% CrI 0.0.05, 1.77) and extended perinatal mortality rates (RR, 0.21; 95% CrI 0.01, 3.00). Using informative priors, the proportion of all term breech presentations that were undiagnosed was 69% lower after the initiation of universal POCUS (RR, 0.31; 95% CrI 0.21, 0.45) with a posterior probability greater of 99.9%. There was also a very high probability (99.5%) of a reduced rate of low Apgar score (<7) at 5 minutes by 40% (RR, 0.60; 95% CrI 0.39, 0.88). We do not have reliable data on number of facility-based ultrasound scans via the standard antenatal referral pathway or external cephalic versions (ECVs) performed during the study period. CONCLUSIONS: In our study, we observed that both a policy of routine facility-based third trimester ultrasound or POCUS are associated with a reduction in the proportion of term breech presentations that were undiagnosed, with an improvement in neonatal outcomes. The findings from our study support the policy of third trimester ultrasound scan for fetal presentation. Future studies should focus on exploring the cost-effectiveness of POCUS for fetal presentation.