Progesterone for prevention of preterm birth in women with short cervical length: 2‐year infant outcomes

Abstract

Objective To evaluate the long‐term outcomes of children born to women with a short cervix and otherwise low risk for preterm birth, after antenatal exposure to vaginal progesterone vs placebo. Methods This was a follow‐up study of the Triple P trial, which randomized 80 low‐risk women with a short cervix (≤ 30 mm) at 18–22 weeks' gestation to progesterone (n = 41) or placebo (n = 39). At 2 years of corrected age, children were invited for a neurodevelopmental assessment, using the Bayley Scales of Infant and Toddler Development, third edition (BSID‐III), and a neurological and physical examination by an assessor blinded to the allocated treatment. Parents filled out the Ages and Stages Questionnaire, the Child Behavior Checklist (CBCL) and a general‐health questionnaire. The main outcome of interest was mean BSID‐III cognitive and motor scores. Additionally, a composite score of mortality and abnormal developmental outcome, including BSID‐III ≤–1 SD, CBCL score in the clinical range and/or parental reported physical problems (at least two operations or at least two hospital admissions in the previous 2 years), was evaluated. Our sample size, dictated by the original sample of the Triple P trial, provided 80% power to detect a mean difference (MD) of 15 points (1 SD) between groups for the BSID‐III tests. Results Of the 80 children born to the randomized women, one in the progesterone group and two in the placebo group died in the neonatal period. Follow‐up data were obtained for 59/77 (77%) children and BSID‐III outcomes in 57 children (n = 28 in the progesterone group and n = 29 in the placebo group) born at a median gestational age of 38 + 6 weeks (interquartile range (IQR), 37 + 3 to 40 + 1 weeks) with a median birth weight of 3240 g (IQR, 2785–3620 g). In the progesterone vs placebo groups, mean BSID‐III cognitive development scores were 101.6 vs 105.0 (MD, –3.4 (95% CI, –9.3 to 2.6); P = 0.29) while mean motor scores were 102.4 vs 107.3 (MD, –4.9 (95% CI, –11.2 to 1.4); P = 0.13). No differences were seen between the two groups in physical (including genital and neurological examination), behavioral and health‐related outcomes. Conclusion In this sample of children born to low‐risk women with a short cervix at screening, no relevant differences in neurodevelopmental, behavioral, health‐related and physical outcomes were found between offspring exposed to vaginal progesterone and those exposed to placebo. © 2020 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

The long‐term effect of prenatal progesterone treatment on child development, behaviour and health: a systematic review

“…29,[31][32][33]35 Furthermore, congenital malformations, chromosomal anomalies or genital abnormalities reported in follow up did not differ between groups. 29,31,34,35 Due to severe heterogeneity in definitions regarding health/malformations/abnormalities, we were unable to aggregate results in meta-analyses. There were no outcomes reported regarding cardiovascular or mental health, sexual or gender orientation, or pubertal development.…”

“…The conference abstract 39 of the TripleP follow up by Cuijpers et al 29 was retrieved from our systematic search and, after corresponding with the authors, we were able to use data from their submitted manuscript ( n = 59 children). After corresponding with Rode et al 34 we were able to use unpublished data on the number of children (total n = 1050 at 6 months and n = 991 at 18 months of age) assessed per treatment group and the number of children with an Ages and Stages Questionnaire score below a cut‐off of 115 points at 18 months (total n = 37).…”

“…Neurodevelopment was assessed using the Bayley Scales of Infant Development-III (Bayley-III) in two studies (n = 890 measurements), 29,32 four studies (n = 2827 measurements) used the Ages and Stages Questionnaire (ASQ) 29,30,[33][34][35] and one study (n = 324 measurements) 31 . The Bayley-III mean cognitive composite score in two studies 29,32 could be included in meta-analysis, comprising 438 children in progesterone group versus 452 children in placebo group. No difference was found in Bayley-III scores between children exposed to progesterone and those exposed to placebo during pregnancy (standardised mean difference [SMD] of À0.04, 95% CI À0.26 to 0.19) ( Figure 2).…”

“…Composite outcome of death or moderate/severe neurodevelopmental impairment Two studies 29,32 reported a composite of moderate/severe neurodevelopmental impairment at 2 years of age, with and without mortality rates ( Table 2). The first study 32 defined neurodevelopmental impairment as 'individual component of disability (motor, cognitive, function, hearing, speech and language, vision, respiratory, gastrointestinal and renal) or hospital admission', and the second study 29 defined it as 'Bayley-III score < À1 SD, or CBCL score in the clinical range, or >1 hospital admission, or >1 surgery in the past 2 years. As imputations for missing data between studies were difficult to compare, only results of complete cases could be used for reporting of the composite outcome.…”

“…Two studies assessed behavioural development. 29,32 One study (n = 597 measurements) used the Strengths and Difficulties Questionnaire (SDQ) 32 and the other study (n = 54 measurements) the Child Behavior Checklist (CBCL) 29 . Meta-analysis was not possible due to the use of different questionnaires.…”