Neonatal Outcomes Following Exposure to Different Betamethasone Formulations

Antenatal corticosteroids (ACS) therapy reduces neonatal morbidity and mortality [1, 2]. Two pharmacokinetic formulations of betamethasone are used even though the potential differences in clinical effect have never been studied [1, 3]. Liggins and Howie used a 12 mg dose of a 1:1 mixture of betamethasone sodium-phosphate (BP), a salt form, and betamethasone acetate (BA), an ester derivative, in their seminal trial [4]. The formulation was based on sheep models. This BP/BA mixture results in an immediate exposure to free betamethasone through dephosphorylation of BP, while extended dosing is provided by the slower deacetylation of BA [3]. In contrast, BP alone has a shorter half-life compared with the BP/BA mixture because of the absence of BA [5]. BP alone is used as ACS in e.g., the UK, Japan, and Denmark [5]. Only one underpowered study (N = 69) has examined the potential clinical differences between BP alone and the BP/BA mixture, despite pharmacokinetic differences that may be clinically significant [1, 6]. In a historical cohort study based on routine collected data, we compared prenatal exposure to the standard BP/BA mixture versus BP alone by assessing neonatal outcomes, including birth parameters, RDS, and mortality. We used routinely collected individual level information from the electronic medical charts of Central Denmark Region, including mother-infant linkage by use of the unique identifiers from the Danish Civil Registration System. The study population comprised all singleton pregnancies, with a fetus alive at the 1st trimester scan from gestational age (GA) 12 + 0/7 onwards, approximately 15 000 births per year, corresponding to some 25% of all births in Denmark, with estimated conception dates from January 1, 2014, to December 31, 2021. Exposure was retrieved from the electronic patient record, identified by hospital administration of betamethasone (ATC H02AB01). The types of betamethasone could be coded owing to recorded trade names even though they share the ATC code. The exposure included the mixture of BP and BA (Celeston) and BP alone (Betapred). Neonatal mortality (within 14 days) and respiratory distress (ICD-10 codes: DP220, DP220A, DP228, DP229) were retrieved from the electronic patient records. Statistical analyses were performed with Stata (StataCorp/College Station/TX/USA). Logistic regression was used for binary outcomes. Student's t-test and linear regression were used for numerical outcomes with adjustment for GA at pregnancy end and calendar-year for head circumference (without outliers (< 15, > 40 cm)), then comparing the two treatment regimens. Access to the data was approved according to Danish legislation (Sundhedsloven §46, permission (1-45-70-88-22)) and data was stored according to the EU GDPR rules. The population included 117 159 singleton pregnancies. Of these, 2169 (1.9%) were exposed to betamethasone in the 2nd or 3rd trimester, median GA for 1st dose was 30 + 6 (range 22 + 5 to 38 + 4), 75% of the exposed delivered preterm. Table 1 illustrates the characteristics of the BP/BA mixture and BP alone exposed. Pregnancies exposed to both treatment regimens were excluded (n = 46). The overall betamethasone use was stable during the study period. Among the exposed, 1540 (71.0%) received two doses of betamethasone, while 112 (5.2%) received more (“rescue”). There was no increased risk of death (OR 0.83 (95% CI 0.39; 1.75)) or RDS (OR 1.02 (95% CI 0.81; 1.27)) for BP compared to BP/BA (Table 1). The prevalence of RDS (2.4%) was unchanged during the study period. Newborns exposed to BP had smaller heads compared to BP/BA (difference −0.5 cm (95% CI (−0.9; −0.1)), p = 0.03), attributed to differences in gestational age at delivery over time (−0.18 cm (95% CI −0.38; 0.01)). Adjustment for calendar year did not change these estimates. There were no other differences between the BP and BP/BA exposed within the power of the study (Table 1). We found no difference in RDS or neonatal death between newborns of women treated with BP/BA mixture compared to BP alone, with the limitation that obstetric and neonatal care have changed over time. However, we found a time-stable prevalence of RDS which indicates that confounding from changes in practice may be limited. These findings are in accordance with previous smaller studies and are reassuring clinically. The apparent reduced head circumference was explained by gestational age and changes in gestational age over time but may warrant further studies. L.K., I.C.Ö., and L.H.P. were responsible for study concept and design. L.K. and L.H.P. acquired and analysed the data. L.K., I.C.Ö., T.B.H., S.S., and L.H.P. drafted the manuscript and revised it for important intellectual content. This work was supported by Borregaard Clinical Ascending Investigator Grant from the Novo Nordisk Foundation (NNF18OC0054457). Wellcome Turst Clinical Career Development Fellowship 209560/Z/17/Z. Health Research Foundation of Central Denmark Region. Sarah Stock is Program Director for the Wellcome Leap In Utero program. She has received research funding, paid to her institution, from the Medical Research Council, the National Institute of Healthcare Research, the Scottish Chief Scientist Office and Wellcome. She has received consultancy fees from Natera, Oganon and Norgine, and honoraria for a talk (over which she had complete control of content) from Hologic. The other authors declare no conflicts of interest. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.