Different corticosteroids and regimens for accelerating fetal lung maturation for women at risk of preterm birth

  • Review
  • Intervention

Authors

  • Fiona C Brownfoot,

    Corresponding author
    1. Mercy Hospital for Women, Heidelberg, Australia
    Search for more papers by this author
  • Daniela I Gagliardi,

    1. The University of Adelaide, ARCH: Australian Research Centre for Health of Women and Babies, The Robinson Institute, Discipline of Obstetrics and Gynaecology, Adelaide, Australia
    Search for more papers by this author
  • Emily Bain,

    1. The University of Adelaide, ARCH: Australian Research Centre for Health of Women and Babies, The Robinson Institute, Discipline of Obstetrics and Gynaecology, Adelaide, Australia
    Search for more papers by this author
  • Philippa Middleton,

    1. The University of Adelaide, ARCH: Australian Research Centre for Health of Women and Babies, The Robinson Institute, Discipline of Obstetrics and Gynaecology, Adelaide, Australia
    Search for more papers by this author
  • Caroline A Crowther

    1. The University of Adelaide, ARCH: Australian Research Centre for Health of Women and Babies, The Robinson Institute, Discipline of Obstetrics and Gynaecology, Adelaide, Australia
    2. The University of Auckland, Liggins Institute, Auckland, New Zealand
    Search for more papers by this author

Abstract

Background

Despite the widespread use of antenatal corticosteroids to prevent respiratory distress syndrome in preterm infants, there is currently no consensus as to the type of corticosteroid to use; nor the dose, frequency, timing of use or the route of administration.

Objectives

To assess the effects of different corticosteroid regimens for women at risk of preterm birth.

Search methods

We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (13 February 2013).

Selection criteria

All identified published and unpublished randomised controlled trials or quasi-randomised control trials comparing any two corticosteroids (dexamethasone or betamethasone or any other corticosteroid that can cross the placenta), comparing different dose regimens (including frequency and timing of administration) in women at risk of preterm birth were included. We planned to exclude cross-over trials and cluster-randomised trials. We included studies published as abstracts only along with studies published as full-text manuscripts

Data collection and analysis

Two review authors independently assessed study eligibility, extracted data and assessed the risk of bias of included studies. Data were checked for accuracy.

Main results

For this update, 12 trials (1557 women and 1661 infants) were included. Dexamethasone was associated with a reduced risk of intraventricular haemorrhage (IVH) compared with betamethasone (risk ratio (RR) 0.44, 95% confidence interval (CI) 0.21 to 0.92; four trials, 549 infants). No statistically significant differences were seen for other primary outcomes: respiratory distress syndrome (RDS) (RR 1.06, 95% CI 0.88 to 1.27; five trials, 753 infants) and perinatal death (neonatal death RR 1.41, 95% CI 0.54 to 3.67; four trials, 596 infants). Similarly, very few differences were seen for secondary outcomes such as rate of admission to the neonatal intensive care unit (NICU) although in one trial, those infants exposed to dexamethasone, compared with betamethasone, had a significantly shorter length of NICU admission (mean difference (MD) -0.91 days, 95% CI -1.77 to -0.05; 70 infants). Results for biophysical parameters were inconsistent, but mostly no clinically important differences were seen.

Compared with intramuscular dexamethasone, oral dexamethasone significantly increased the incidence of neonatal sepsis (RR 8.48, 95% CI 1.11 to 64.93) in one trial of 183 infants. No statistically significant differences were seen for other outcomes reported.

Apart from a reduced maternal postpartum length of stay for women who received betamethasone at 12-hourly intervals compared to 24-hourly intervals in one trial (MD -0.73 days, 95% CI -1.28 to -0.18; 215 women), no differences in maternal or neonatal outcomes were seen between the different betamethasone dosing intervals assessed. Similarly, no significant differences in outcomes were seen when betamethasone acetate and phosphate was compared with betamethasone phosphate in one trial.

Authors' conclusions

It remains unclear whether one corticosteroid (or one particular regimen) has advantages over another.

Dexamethasone may have some benefits compared with betamethasone such as less IVH, and a shorter length of stay in the NICU. The intramuscular route may have advantages over the oral route for dexamethasone, as identified in one small trial. Apart from the suggestion that 12-hour dosing may be as effective as 24-hour dosing of betamethasone based on one small trial, few other conclusions about optimal antenatal corticosteroid regimens were able to be made. No long-term results were available except for a small subgroup of 18 month old children in one trial. Trials comparing the commonly used corticosteroids are most urgently needed, as are trials of dosages and other variations in treatment regimens.

Résumé scientifique

Différents corticostéroïdes et schémas thérapeutiques pour accélérer la maturation pulmonaire fœtale chez les femmes présentant un risque d'accouchement prématuré

Contexte

Malgré la pratique courante de l'administration prénatale de corticoïdes pour prévenir le syndrome de détresse respiratoire chez les nouveau-nés prématurés, il n'existe actuellement aucun consensus sur le type de corticostéroïde à utiliser, ni sur la dose, la fréquence, le moment ou la voie d'administration.

Objectifs

Évaluer les effets de différents schémas posologiques de corticostéroïdes pour les femmes présentant un risque d'accouchement prématuré.

Stratégie de recherche documentaire

Nous avons consulté le registre des essais cliniques du groupe Cochrane sur la grossesse et la naissance (le 13 février 2013).

Critères de sélection

Tous les essais contrôlés, randomisés ou quasi-randomisés, publiés et non publiés, comparant deux corticostéroïdes au choix (dexaméthasone, bétaméthasone ou tout autre corticoïde qui peut traverser le placenta), comparant différents schémas posologiques (y compris la fréquence et le moment d'administration) chez des femmes présentant un risque d'accouchement prématuré ont été inclus. Nous avons exclu les essais croisés et les essais randomisés en grappes. Nous avons inclus des études publiées uniquement sous forme de résumés ainsi que des études publiées sous forme de manuscrit intégral.

Recueil et analyse des données

Deux auteurs de la revue ont indépendamment évalué l’admissibilité des études, extrait les données et évalué le risque de biais dans les études incluses. L'exactitude des données a été vérifiée.

Résultats principaux

Pour cette mise à jour, 12 essais (1 557 femmes et 1 661 nourrissons) ont été inclus. La dexaméthasone était associée à une réduction du risque d'hémorragie intraventriculaire (HIV) par rapport à la bétaméthasone (risque relatif (RR) 0,44, intervalle de confiance (IC) à 95 % 0,21 à 0,92 ; quatre essais, 549 nouveau-nés). Aucune différence statistiquement significative n'a été observée pour les autres critères de jugement principaux : syndrome de détresse respiratoire (SDR) (RR 1,06, IC à 95 % 0,88 à 1,27 ; cinq essais, 753 nourrissons) et décès périnatal (décès néonatal RR de 1,41, IC à 95 % 0,54 à 3,67 ; quatre essais, 596 nourrissons). De même, très peu de différences ont été observées pour les critères de jugement secondaires tels que les taux d'admission en unité néonatale de soins intensifs (UNSI) ; cependant, dans un essai, la durée d'admission en UNSI était significativement plus courte chez les nourrissons exposés à la dexaméthasone, par rapport à ceux exposés à la bétaméthasone (différence moyenne (DM) -0,91 jours, IC à 95 % -1,77 à -0,05 ; 70 nourrissons). Les résultats pour les paramètres biophysiques étaient contradictoires, mais dans l'ensemble aucune différence cliniquement importante n'a été observée

Dans un essai sur 183 nouveau-nés, une augmentation significative de l'incidence de septicémie néonatale a été observée lors de l'administration de la dexaméthasone par voie orale par rapport à la voie intramusculaire (RR 8,48, IC à 95 % 1,11 à 64,93). Aucune différence statistiquement significative n'a été observée pour les autres critères de jugement rapportés.

À l'exception d'une réduction de la durée d'hospitalisation post-partum pour les femmes ayant reçu de la bétaméthasone à intervalles de 12 heures par rapport à celles l'ayant reçu à intervalle de 24 heures dans un essai (DM de -0,73 jours, IC à 95 % -1,28 à -0,18 ; 215 femmes), aucune différence dans les résultats maternels et néonataux n'a été observée entre les différentes posologies de bétaméthasone évaluées. De même, dans un essai, aucune différence significative dans les critères de jugement n'a été observée lors de l'utilisation d'acétate et phospate de bétaméthasone comparée à l'utilisation de phosphate de bétaméthasone.

Conclusions des auteurs

On ignore encore si un corticostéroïde (ou un schéma posologique spécifique) présente des avantages par rapport à un autre.

La dexaméthasone peut avoir certains effets bénéfiques par rapport à la bétaméthasone, tels qu'une incidence moindre de HIV et une plus courte durée du séjour en UNSI. Pour la dexaméthasone, l'administration intramusculaire pourrait avoir des avantages par rapport à l'administration par voie orale, comme identifié dans un essai de petite taille. Hormis l'hypothèse, fondée sur un essai de petite taille, selon laquelle l'administration de la bétaméthasone à intervalles de 12 heures pourrait être aussi efficace que l'administration à intervalles de 24 heures, peu d'autres conclusions sur le traitement prénatal optimal par corticostéroïdes ont pu être tirées. Aucun résultat à long terme n'était disponible, sauf dans un essai, pour un petit sous-groupe d'enfants âgés de 18 mois. Il est urgent de mener des essais comparant les corticostéroïdes utilisés couramment, ainsi que des essais portant sur les dosages et autres variations dans les schémas thérapeutiques.

Plain language summary

Corticosteroid treatments before early birth for reducing death, lung problems and brain haemorrhage in babies

Babies born early are at risk of death, lung problems (respiratory distress syndrome) and bleeding of the brain (intraventricular haemorrhage). Corticosteroids are given to the mother to help stop these problems occurring and there is high-quality evidence that they are effective in preventing many of these problems. These drugs work by maturing the baby's lungs before birth. There are different types of corticosteroids and they can be given in different ways and in different doses. Since there is no clear or agreed best type or dose, hospitals may vary in how they give this drug.

Most trials have compared the two most commonly used corticosteroids before early birth, dexamethasone and betamethasone. In this review of 12 trials (involving 1557 women and 1661 infants) of moderate quality, 10 trials compared dexamethasone and betamethasone; one trial compared two different ways of giving dexamethasone and one trial compared two different ways of giving betamethasone. We found that dexamethasone and betamethasone showed similar results, although there was less bleeding of the brain and a shorter length of neonatal intensive care unit hospital stay for dexamethasone compared with betamethasone. On the basis of one trial, giving dexamethasone by injection (intramuscularly) may be better than giving the drug to the mother by mouth (orally). Usually the drug is given in two doses 24 hours apart and one trial showed that this interval could perhaps be reduced to 12 hours if required. We need more studies to establish which is the best drug and what is the best way to give it, and babies in these trials need to be followed up over a long period to monitor any effects on child and adult development.

Résumé simplifié

Traitements par corticostéroïdes avant un accouchement prématuré pour réduire la mortalité, les problèmes pulmonaires et les hémorragies cérébrales chez les nouveau-nés

Chez les bébés nés prématurément, il existe un risque de décès, de problèmes pulmonaires (syndrome de détresse respiratoire) et d'hémorragies du cerveau (hémorragie intraventriculaire). Des corticostéroïdes sont administrés à la mère pour tenter de remédier à ces problèmes et il existe des preuves de grande qualité indiquant que les corticostéroïdes sont efficaces dans la prévention de la plupart de ces problèmes. Ces médicaments accélèrent la maturation des poumons du bébé avant la naissance. Il existe différents types de corticostéroïdes et ils peuvent être administrés de différentes manières et à des doses différentes. Aucun consensus clair ne se dégage concernant le choix du meilleur médicament ou de la meilleure dose, les hôpitaux peuvent donc choisir entre différentes manières d'administrer les corticostéroïdes.

La plupart des essais ont comparé les deux corticostéroïdes les plus souvent utilisés avant un accouchement prématuré, la dexaméthasone et la bétaméthasone. Dans cette revue de 12 essais de qualité modérée (impliquant 1 557 femmes et 1 661 nourrissons), 10 essais comparaient la dexaméthasone à la bétaméthasone ; un essai comparait deux différents modes d'administration de la dexaméthasone et un essai comparait deux différents modes d'administration de la bétaméthasone. Nous avons constaté que la dexaméthasone et la bétaméthasone ont des résultats similaires, bien qu'il y ait moins de saignements du cerveau et une plus courte durée de séjour à l'hôpital en unité néonatale de soins intensifs quand la dexaméthasone est utilisée que lorsque la bétaméthasone est utilisée. Les résultats d'un essai semblent indiquer que l'administration de dexaméthasone par injection intramusculaire pourrait être plus efficace que l'administration à la mère par voie orale. En général, le médicament est administré en deux doses espacées de 24 heures et un essai a montré que cet intervalle pourrait être ramené à 12 heures si besoin. D'autres études sont nécessaires pour déterminer quel est le meilleur médicament et quel est le meilleur mode d'administration. Les bébés ayant fait partie de ces essais devront être suivis sur une période plus longue afin de surveiller les effets des traitements sur le développement de l'enfant et de l'adulte.

Notes de traduction

Traduit par: French Cochrane Centre 14th January, 2014
Traduction financée par: Ministère du Travail, de l'Emploi et de la Santé Français

Background

Preterm birth (less than 37 weeks' gestation) poses a significant health burden affecting approximately 5% to 18% of all babies born globally (Goldenberg 2007; Haram 2003; March of Dimes 2012), with over 60% of all preterm births occurring in Africa and South Asia (March of Dimes 2012). Preterm infants, especially those born before 32 weeks' gestation, are at high risk of respiratory distress syndrome (RDS), a serious complication that remains the primary cause of early neonatal death and disability (Haram 2003). Those infants born preterm who do survive the neonatal period are at a significantly increased risk of long-term neurological disability (Johnson 1993; Saigal 2007). RDS develops as a consequence of surfactant deficiency and immature lung development. The risk of RDS and neonatal mortality reduces as gestation increases, reflecting maturity of organ systems (Doyle 2001; Moise 1995; Saigal 2007). Treatments that may reduce the incidence of respiratory distress syndrome (RDS) in infants born preterm, including antenatal corticosteroids, have therefore received considerable attention (Roberts 2006).

Corticosteroids

Corticosteroids act by altering gene expression resulting in glucocorticoid effects, including gluconeogenesis, proteolysis, lipolysis, suppression of immune responses and mineralocorticoid effects, including hypertension, sodium and water retention and potassium loss (AMH 2006). In the fetal lung, the action of corticosteroids leads to an increase in protein production, biosynthesis of phospholipids and the appearance of surfactant (Ballard 1995).

Liggins 1969 demonstrated that the lungs of lambs born preterm became functionally mature following antenatal corticosteroid administration. Following these initial animal studies, Liggins and other investigators conducted several clinical trials to assess the effects of corticosteroids before preterm birth in humans.

The Cochrane review 'Antenatal corticosteroids for accelerating fetal lung maturation for women at risk of preterm birth' showed that a single course of antenatal corticosteroids significantly reduced the incidence of RDS (risk ratio (RR) 0.66, 95% confidence interval (CI) 0.59 to 0.73; 21 trials, 4038 infants) (Roberts 2006). Other beneficial effects included a reduction in neonatal death, cerebroventricular haemorrhage, necrotising enterocolitis, infectious morbidity, need for respiratory support and neonatal intensive care unit admission. For the mother, corticosteroid use was not shown to increase the risk of death, chorioamnionitis or puerperal sepsis (Roberts 2006). Contrary to the concern that corticosteroid treatment may increase infection in those with preterm prelabour rupture of membranes (Imseis 1996), or increase the rate of stillbirth in those with pregnancy-related hypertension (Liggins 1976), this Cochrane review confirmed that antenatal corticosteroid treatment is effective in women at risk of preterm birth with these complications (Roberts 2006). Corticosteroids have become the standard of care for women at risk of preterm birth before 32 to 34 weeks' gestation in many countries (Jobe 2004; NIH 1995).

Despite their widespread use, there is currently variation in clinical practice as to the type of corticosteroid used, the dose and frequency given, and the route of administration of corticosteroid doses.

Corticosteroid type

Currently either betamethasone or dexamethasone are the recommended corticosteroid regimens used in clinical practice (NIH 1995). Betamethasone is available in two different forms: betamethasone sodium phosphate, a solution with a short biological half-life of 36 to 72 hours; and betamethasone acetate, a suspension with a relatively long half-life (Jobe 2004; Katzung 2004; NNF6 2011). These forms of betamethasone are often used in combination to maximize the drug's efficiency while reducing the number of injections given to the mother (NNF6 2011). Dexamethasone generally comes in the form of dexamethasone sodium phosphate, a solution with a short biological half-life of 36 to 72 hours (Ballard 1995; Jobe 2004; Katzung 2004; NNF6 2011).

Both betamethasone and dexamethasone are able to cross the placenta in their active form and have comparable properties (NNF6 2011). The chemical composition of betamethasone and dexamethasone are virtually identical except for the configuration of a methyl group in position 16 (Bar-Lev 2004; NNF6 2011). Some dexamethasone preparations contain a sulphite preservative (NNF6 2011). Sulphites have been linked to neurotoxicity in the newborn especially when in combination with peroxy nitrite (Bar-Lev 2004; Baud 1999; Goldenberg 2001; Walfisch 2001).

The optimal type of corticosteroid to use for prenatal treatment remains unclear. The indirect subgroup comparison of betamethasone and dexamethasone in the Roberts 2006 Cochrane review indicated similar short-term neonatal outcomes for both drugs. Maternal outcomes were also similar although the risk of puerperal sepsis was higher in the dexamethasone versus placebo or no treatment group, while betamethasone did not show an increase in puerperal sepsis over placebo or no treatment (Roberts 2006). The results from observational studies are not always consistent with the results from randomised trials. For instance, a National Institute of Child Health and Human Development Neonatal Research Network (NICHD NRN) cohort study of over 300 infants reported a link between betamethasone and reduced risk of neonatal death, whereas dexamethasone was associated with an increased risk of neonatal death (Lee 2006). In contrast, the Roberts 2006 Cochrane review showed a reduced risk for fetal and neonatal death for both the betamethasone and dexamethasone groups compared with placebo/no treatment. In a later NICHD NRN report of part of this cohort, Lee 2008 reported reduced adverse childhood neurological outcomes at 18 to 22 months for dexamethasone but not for betamethasone.

The long-term outcomes related to corticosteroid use have largely been positive. Within the Roberts 2006 Cochrane review, overall antenatal corticosteroid treatment was shown to be associated with less developmental delay in childhood, and a trend towards fewer children having cerebral palsy when compared with no corticosteroid treatment. It is not known if the long-term outcomes vary by type of corticosteroid used. While follow-up at 30 years following use showed no clinical differences in adults who were exposed or not exposed to betamethasone in utero (Dalziel 2005), there have been no similar long-term follow-up studies reported on dexamethasone use. There are no published data on the long-term effects of antenatal betamethasone compared directly with dexamethasone.

Corticosteroid dose, timing and frequency

The optimal corticosteroid dose to use, timing of use and frequency of administration similarly remains unclear. The common regimens of two doses of 12 mg of betamethasone given intramuscularly 24 hours apart and the treatment of four doses of 6 mg of dexamethasone given intramuscularly 12 hours apart was recommended by the National Institutes of Health (NIH) Consensus Development Panel on the Effect of Corticosteroids for Fetal Maturation on Perinatal Outcomes (NIH 1995). This dose corresponds to a high occupancy of steroid receptors in fetal tissues. While the benefits of corticosteroids are well-known, there is concern regarding their potential for adverse effects, particularly at high doses. There is a suggestion that the current antenatal doses used may be higher than needed (Jobe 2004). Similarly, the rationale for two doses of betamethasone and four doses of dexamethasone and the effects of using different formulations for the initial and subsequent injections remain unclear (Jobe 2004; NNF6 2011).

Corticosteroid route

The optimal route of administration of antenatal betamethasone and dexamethasone is also uncertain. Both drugs may be administered as intramuscular injections. Betamethasone can be given intra-amniotically (Lefebvre 1976; Murphy 1982) and intravenously (Petersen 1983) and dexamethasone can be given orally (Egerman 1998).

Repeat doses of corticosteroid

The reduction in the incidence of RDS by antenatal corticosteroid therapy has been shown to be effective up to seven days after treatment (Roberts 2006). A single dose of antenatal corticosteroid does not prevent RDS if it is administered seven days or more prior to birth (Crowther 2011; Roberts 2006). Whether antenatal corticosteroids for women who remain at risk of preterm birth need to be repeated seven days after the initial course is assessed in another Cochrane review (Crowther 2011); therefore, this review will not cover repeat steroid doses compared with single doses.

Why it is important to do this review

This review updates a previously published Cochrane review on different corticosteroids and regimens for accelerating fetal lung maturation for women at risk of preterm birth (Brownfoot 2008). This review found that while dexamethasone may have some benefits compared to betamethasone such as less intraventricular haemorrhage, it may also be associated with a higher rate of NICU admission (seen in one trial). Few other conclusions about optimal antenatal corticosteroid regimens could be made, and thus the review concluded that high-quality evidence from randomised trials was urgently needed in this area.

Despite the widespread use of antenatal corticosteroids to prevent RDS in preterm infants, there is still no consensus as to the type of corticosteroid to use; nor the dose, frequency, timing of use or the route of administration. This review assesses studies making a head-to-head comparison of different regimens of corticosteroid type, dose, timing, frequency of dose per treatment course and route of administration. Other corticosteroid Cochrane reviews have examined inter-study differences between drug regimens, in subgroup analysis. We have assessed these indirect comparisons comparing any corticosteroid with placebo following the methods outlined in an appendix accompanying Song 2003, and by performing subgroup interaction tests on the trials from the Roberts 2006 review (Table 1).

Table 1. Comparison of direct and indirect estimates and interaction tests
  1. CI: confidence interval
    IVH: intraventricular haemorrhage
    MD: mean difference
    NA: not applicable
    RDS: respiratory distress syndrome
    RR: risk ratio
    v: versus

Outcome Direct comparison Indirect comparison Discrepancy Interaction test
Fetal/neonatal deathDirect comparison from this review: dexamethasone v betamethasone RR 1.41, 95% CI 0.54 to 3.67 (4 trials, 596 infants).Indirect comparison from Roberts 2006 Cochrane review: RR 0.92, 95% CI 0.57 to 1.49.Nonsignificant discrepancy between direct and indirect results (MD 0.42, 95% CI -0.65 to 1.49).Test for subgroup differences in trials in Roberts 2006 Cochrane review was non significant (Chi² statistic: 0.00 and P value: 0.98, I² value: 0%).
RDSDirect comparison from this review: dexamethasone v betamethasone: RR 1.06, 95% CI 0.88 to 1.27 (5 trials, 753 infants).Indirect comparison from Roberts 2006 Cochrane review: RR 1.40, 95% CI 1.02 to 1.90.Nonsignificant discrepancy between direct and indirect results (MD -0.28, 95% CI -0.64 to 0.08).Test for subgroup differences in trials in Roberts 2006 Cochrane review was significant (Chi² statistic: 4.68 and P value 0.03, I² value: 78.6%).
IVH (any)Direct comparison from this review: dexamethasone v betamethasone: RR 0.44, 95% CI 0.21 to 0.92 (4 trials, 549 infants).Indirect comparison from Roberts 2006 Cochrane review: RR 0.96, 95% CI 0.35 to 2.66.Nonsignificant discrepancy between direct and indirect results (MD -0.78, 95% CI -2.04 to 0.48).Test for subgroup differences in trials in Roberts 2006 Cochrane review was non significant (Chi² statistic: 0.78 and P value: 0.38, I² value: 0%).
ChorioamnionitisNo direct comparison (outcome not reported in any trials included in this review).Indirect comparison from Roberts 2006 Cochrane review: RR 1.90, 95% CI 1.10 to 3.28.NA.Test for subgroup differences in trials in Roberts 2006 Cochrane review was significant (Chi² statistic: 5.41 and P value 0.02, I² value: 81.5%).
Puerperal sepsisNo direct comparison (outcome not reported in any trials included in this review).Indirect comparison from Roberts 2006 Cochrane review: RR 1.68, 95% CI 0.60 to 4.66.NA.Test for subgroup differences in trials in Roberts 2006 Cochrane review was non significant (Chi² statistic: 0.99 and P value: 0.32, I² value: 0%).

Objectives

To assess the effects on fetal and neonatal morbidity and mortality, on maternal morbidity and mortality, and on the child and adult in later life, of administering different types of corticosteroids (dexamethasone or betamethasone), or different corticosteroid dose regimens, including timing, frequency and mode of administration.

Methods

Criteria for considering studies for this review

Types of studies

All identified published and unpublished randomised controlled trials or quasi-randomised control trials comparing any two corticosteroids (dexamethasone or betamethasone or any other corticosteroid that can cross the placenta), comparing different dose regimens (including frequency and timing of administration) in women at risk of preterm birth were included. We planned to exclude cross-over trials and cluster-randomised trials. We included studies published as abstracts only along with studies published as full-text manuscripts.

Types of participants

Women with a singleton or multiple pregnancy expected to give birth preterm (before 37 weeks) as a result of either spontaneous preterm labour, preterm prelabour rupture of membranes or elective preterm birth.

Types of interventions

  • Different types of corticosteroids including dexamethasone, betamethasone, hydrocortisone or any other corticosteroid that can cross the placenta.

  • Different corticosteroid regimens including dose, frequency, timing and route of administration.

Trials which tested the effect of corticosteroids with other interventions have been excluded. Trials assessing repeat corticosteroid doses versus a single corticosteroid dose have also been excluded.

Types of outcome measures

These cover outcomes of maternal morbidity, perinatal morbidity and mortality, child morbidity and mortality, child as adult morbidity and mortality and the use of health services by the mother and by the neonate or child.

They are divided into primary outcomes, thought to be the most clinically relevant, and secondary outcomes of importance, including possible complications and also additional measures of effectiveness. Groups include: women; fetuses/neonates; children; children as adults; health services.

Primary outcomes
For the woman
  • Death;

  • chorioamnionitis (however defined by authors);

  • puerperal sepsis (however defined by authors).

For the fetus/neonate
  • Death (fetal or neonatal);

  • respiratory distress syndrome (RDS);

  • intraventricular haemorrhage (IVH) (diagnosed by ultrasound, diagnosed by autopsy).

For the child
  • Death;

  • neurodevelopmental disability at follow-up (blindness, deafness, moderate/severe cerebral palsy (however defined by authors), or developmental delay/intellectual impairment (defined as developmental quotient or intelligence quotient less than -2 standard deviations below population mean) or variously defined).

For the child as adult
  • Death;

  • neurodevelopmental disability at follow-up (blindness, deafness, moderate/severe cerebral palsy (however defined by authors), or developmental delay/intellectual impairment (defined as developmental quotient or intelligence quotient less than -2 standard deviations below population mean) or variously defined.

Secondary outcomes
For the woman
  • Fever after trial entry requiring the use of antibiotics;

  • intrapartum fever requiring the use of antibiotics;

  • postnatal fever requiring the use of antibiotics;

  • admission to intensive care unit;

  • adverse effects of therapy;

  • glucose intolerance (however defined by authors);

  • hypertension (however defined by authors).

For the fetus/neonate
  • Apgar score less than seven at five minutes;

  • interval between trial entry and birth;

  • birthweight;

  • low birthweight;

  • mean length at birth;

  • mean head circumference at birth;

  • mean skin fold thickness at birth;

  • small-for-gestational age (however defined by authors);

  • mean placental weight;

  • neonatal blood pressure;

  • admission to neonatal intensive care;

  • need for inotropic support (days);

  • need for mechanical ventilation/continuous positive airways pressure;

  • mean duration of mechanical ventilation/continuous positive airways pressure (days);

  • air leak syndrome;

  • need for oxygen supplementation;

  • duration of oxygen supplementation (days);

  • surfactant use;

  • severity of RDS;

  • chronic lung disease (need for continuous supplemental oxygen at 28 days postnatal age or 36 weeks' postmenstrual age, whichever was later);

  • bronchopulmonary dyplasia (variously defined);

  • severe IVH;

  • periventricular leukomalacia;

  • systemic infection in first 48 hours of life (neonatal sepsis);

  • proven infection while in the NICU;

  • necrotising enterocolitis;

  • retinopathy of prematurity;

  • patent ductus arteriosus;

  • hypothalamo-pituitary-adrenal (HPA) axis function (however defined by authors);

  • biophysical parameters (however defined by the authors).

For the child
  • Mean weight;

  • mean head circumference;

  • mean length;

  • mean skin fold thickness;

  • abnormal lung function (however defined by authors);

  • mean blood pressure;

  • glucose intolerance (however defined by authors);

  • HPA axis function (however defined by authors);

  • dyslipidaemia (however defined by authors);

  • any neurodisability;

  • visual impairment (however defined by authors);

  • hearing impairment (however defined by authors);

  • developmental delay (defined as developmental quotient less than -2 standard deviations below population mean);

  • intellectual impairment (defined as intelligence quotient less than -2 standard deviations below population mean);

  • cerebral palsy (however defined by authors);

  • behavioural/learning difficulties (however defined by authors).

For the child as an adult
  • Mean weight;

  • mean head circumference;

  • mean length;

  • mean skin fold thickness;

  • abnormal lung function (however defined by authors);

  • mean blood pressure;

  • glucose intolerance (however defined by authors);

  • HPA axis function (however defined by authors);

  • dyslipidaemia (however defined by authors);

  • mean age at puberty;

  • bone density (however defined by authors);

  • educational achievement (completion of high school, or however defined by authors);

  • any neurodisability;

  • visual impairment (however defined by authors);

  • hearing impairment (however defined by authors);

  • intellectual impairment (defined as intelligence quotient less than -2 standard deviations below population mean);

  • behavioural/learning difficulties (however defined by authors).

For health services
  • Mean length of antenatal hospitalisation for women (days);

  • mean length of postnatal hospitalisation for women (days);

  • mean length of neonatal hospitalisation (days).

Search methods for identification of studies

Electronic searches

We contacted the Trials Search Co-ordinator to search the Cochrane Pregnancy and Childbirth Group’s Trials Register (13 February 2013). 

The Cochrane Pregnancy and Childbirth Group’s Trials Register is maintained by the Trials Search Co-ordinator and contains trials identified from:

  1. monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL);

  2. weekly searches of MEDLINE;

  3. weekly searches of Embase;

  4. handsearches of 30 journals and the proceedings of major conferences;

  5. weekly current awareness alerts for a further 44 journals plus monthly BioMed Central email alerts.

Details of the search strategies for CENTRAL, MEDLINE and Embase, the list of handsearched journals and conference proceedings, and the list of journals reviewed via the current awareness service can be found in the ‘Specialized Register’ section within the editorial information about the Cochrane Pregnancy and Childbirth Group.

Trials identified through the searching activities described above are each assigned to a review topic (or topics). The Trials Search Co-ordinator searched the register for each review using the topic list rather than keywords. 

We did not apply any language restrictions.

Data collection and analysis

Selection of studies

Two review authors independently assessed for inclusion all the potential studies we identified as a result of the search strategy. We resolved any disagreement through discussion or, if required, we consulted a third person.

Data extraction and management

We designed a form to extract data. For eligible studies, at least two review authors extracted the data using the agreed form. We resolved discrepancies through discussion or, if required, we consulted a third person. We entered data into Review Manager software (RevMan 2011) and checked for accuracy.

When information regarding any of the above was unclear, we attempted to contact authors of the original reports to provide further details.

Assessment of risk of bias in included studies

Two review authors independently assessed risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We resolved any disagreement by discussion or by involving a third author.

(1) Random sequence generation (checking for possible selection bias)

We described for each included study the methods used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.

We assessed the methods as:

  • low risk of bias (any truly random process, e.g. random number table; computer random number generator);

  • high risk of bias (any non-random process, e.g. odd or even date of birth; hospital or clinic record number);

  • unclear risk of bias.   

(2) Allocation concealment (checking for possible selection bias)

We described for each included study the method used to conceal the allocation sequence and determined whether intervention allocation could have been foreseen in advance of, or during recruitment, or changed after assignment.

We assessed the methods as:

  • low risk of bias (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes);

  • high risk of bias (open random allocation; unsealed or non-opaque envelopes, alternation; date of birth);

  • unclear risk of bias.   

(3.1) Blinding of participants and personnel (checking for possible performance bias)

We described for each included study, the methods, if any, used to blind study participants and personnel from knowledge of which intervention a participant received. We considered studies to be at a low risk of bias if they were blinded, or if we judged that the lack of blinding would be unlikely to affect results. We assessed blinding separately for different outcomes or classes of outcomes.

We assessed the methods as:

  • low, high or unclear risk of bias for participants;

  • low, high or unclear risk of bias for personnel.

(3.2) Blinding of outcome assessment (checking for possible detection bias)

We described for each included study the methods used, if any, to blind outcome assessors from knowledge of which intervention a participant received. We assessed blinding separately for different outcomes or classes of outcomes.

We assessed methods used to blind outcome assessment as:

  • low, high or unclear risk of bias.

(4) Incomplete outcome data (checking for possible attrition bias due to the amount, nature and handling of incomplete outcome data)

We described for each included study and for each outcome or class of outcomes,the completeness of data including attrition and exclusions from the analysis. We stated whether attrition and exclusions were reported, the numbers included in the analysis at each stage (compared with the total randomised participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. Where sufficient information was reported or was supplied by the trial authors, we included missing data in the analyses which we undertook.

 We assessed the methods as:

  • low risk of bias (e.g. where there was no missing data or where reasons for missing data were balanced across groups);

  • high risk of bias (e.g. numbers or reasons for missing data imbalanced across groups; 'as treated' analysis done with substantial departure of intervention received from that assigned at randomisation);

  • unclear risk of bias.

(5) Selective reporting bias (checking for reporting bias)

We described for each included study how the possibility of selective outcome reporting bias was examined by us and what we found.

We assessed the methods as:

  • low risk of bias (where it was clear that all of the study’s pre-specified outcomes and all expected outcomes of interest to the review had been reported);

  • high risk of bias (where not all the study’s pre-specified outcomes had been reported; one or more reported primary outcomes were not pre-specified; outcomes of interest were reported incompletely and so could not be used; study failed to include results of a key outcome that would have been expected to have been reported);

  • unclear risk of bias.

(6) Other sources of bias (checking for bias due to problems not covered by (1) to (5) above)

We described for each included study any important concerns we had about other possible sources of bias. We assessed whether each study was free of other problems that could put it at risk of bias:

  • low risk of other bias;

  • high risk of other bias;

  • unclear whether there is risk of other bias.

(7) Overall risk of bias

We made explicit judgements about whether studies were at a high risk of bias, according to the criteria given in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). With reference to (1) to (6) above, we assessed the likely magnitude and direction of the bias and whether we considered it is likely to impact on the findings. We planned to explore the impact of the level of bias through undertaking sensitivity analyses - see Sensitivity analysis

Measures of treatment effect

Dichotomous data

For dichotomous data, we presented results as risk ratio with 95% confidence intervals.

Continuous data

For continuous data, we used the mean difference when outcomes were measured in the same way between trials. If necessary, we would have used the standardised mean difference to combine trials that measured the same outcome, but used different methods.  

Unit of analysis issues

We considered cross-over trials and cluster-randomised trials inappropriate for this research question.

Dealing with missing data

For included studies, we noted levels of attrition. We planned to explore the impact of including studies with high levels of missing data in the overall assessment of treatment effect by sensitivity analysis.

For all outcomes, we carried out analyses, as far as possible, on an intention-to-treat basis, i.e. we attempted to include all participants randomised to each group in the analyses, and all participants were analysed in the group to which they were allocated, regardless of whether or not they received the allocated intervention. The denominator for each outcome in each trial was the number randomised minus any participants whose outcomes were known to be missing.

Assessment of heterogeneity

We assessed statistical heterogeneity in each meta-analysis using the T², I² and Chi² statistics. We regarded heterogeneity as substantial if the I² was greater than 30% and either the T² was greater than zero, or there was a low P value (less than 0.10) in the Chi² test for heterogeneity. 

Assessment of reporting biases

In future updates of this review, if there are 10 or more studies in the meta-analysis, we will investigate reporting biases (such as publication bias) using funnel plots. We will assess funnel plot asymmetry visually. If asymmetry is suggested by a visual assessment, we will perform exploratory analyses to investigate it.

Data synthesis

We carried out statistical analysis using the Review Manager software (RevMan 2011). We used fixed-effect meta-analysis for combining data where it was reasonable to assume that studies were estimating the same underlying treatment effect: i.e. where trials were examining the same intervention, and the trials’ populations and methods were judged sufficiently similar. Where there was clinical heterogeneity sufficient to expect that the underlying treatment effects differed between trials, or where substantial statistical heterogeneity was detected, we used random-effects meta-analysis to produce an overall summary if an average treatment effect across trials was considered clinically meaningful. The random-effects summary was treated as the average range of possible treatment effects and we have discussed the clinical implications of treatment effects differing between trials. If the average treatment effect was not clinically meaningful, we would not have combined trials.

Where we have used random-effects analyses, we have presented the results as the average treatment effect with its 95% confidence interval, and the estimates of T² and I².

Subgroup analysis and investigation of heterogeneity

If we had identified substantial heterogeneity, we planned to investigate it using subgroup analyses and sensitivity analyses. We planned to consider whether an overall summary was meaningful, and if it was, use random-effects analysis to produce it.

We performed separate comparisons for different types of corticosteroids; and different preparations, timings and routes of administration.

We planned the following subgroup analyses:

  • singleton versus multiple pregnancy;

  • preterm prelabour rupture of membranes (at trial entry: yes versus no);

  • gestational age at trial entry (24 to 26 weeks, 27 to 29 weeks, 30 to 34 weeks, 35 to 37 weeks);

  • pregnancy-induced hypertension syndrome (yes or no).

However, we were only able to perform a subgroup analysis based on gestational age at trial entry for one included trial, and we were not able to perform the other subgroup analyses due to paucity of data. We used only primary outcomes in the subgroup analysis.

We assessed subgroup differences by interaction tests available in within RevMan (RevMan 2011). We have reported the results of the subgroup analysis quoting the Chi² statistic and P value, and the interaction test I² value.

Sensitivity analysis

We planned sensitivity analyses to explore the effect of trial quality assessed by concealment of allocation, by excluding studies with clearly inadequate allocation of concealment, rated at 'high risk of bias' for this component. However, only one quasi-randomised trial was included in this version of the review and since no other trials reported the same outcomes as this trial, a sensitivity analysis by adequacy of allocation could not be carried out.

Results

Description of studies

Results of the search

The updated searches of the Pregnancy and Childbirth Group Trials Register identified three trial reports (Danesh 2012; Khandelwal 2012; Shanks 2010). We have included two trials in the review (Danesh 2012; Khandelwal 2012), and have excluded the other trial (Shanks 2010).

Therefore, of the 20 trials that were identified for possible inclusion, 12 trials met our pre-selected inclusion criteria in that they compared any two corticosteroids (dexamethasone or betamethasone or any other corticosteroid that can cross the placenta), or compared different dose regimens and timing or frequency and route of administration in women at risk of preterm birth (Chen 2005; Danesh 2012; Egerman 1998; Elimian 2007; Khandelwal 2012; Magee 1997; Mulder 1997; Mushkat 2001; Rotmensch 1999; Senat 1998; Subtil 2003; Urban 2005).

We excluded eight trials from the review (Egerman 1997; Kurz 1993; Liu 2006; Romaguera 1997; Salzer 1982; Shanks 2010; Vytiska 1985; Whitt 1976); one trial is awaiting classification (Romejko-Wolniewicz 2013); and one trial is ongoing (Crowther 2010). See Characteristics of studies awaiting classification and Characteristics of ongoing studies.

Included studies

Description of interventions used in the included trials

In the 12 trials included in this review, 1557 women and 1661 infants were recruited (1159 women and 1213 infants in the 10 trials that compared dexamethasone with betamethasone). Four different corticosteroid regimens were used:

  • six trials compared 24 mg dexamethasone (6 mg, 12 hourly, four doses) and 24 mg betamethasone (12 mg, 24 hourly, two doses) (Chen 2005; Danesh 2012; Elimian 2007; Rotmensch 1999; Subtil 2003; Urban 2005);

  • two trials compared 24 mg dexamethasone (12 mg, 12 hourly, two doses) and 24 mg betamethasone (12 mg, 12 hourly, two doses) (Magee 1997; Mushkat 2001);

  • one trial compared 16 mg dexamethasone (4 mg, 12 hourly, four doses) and 24 mg betamethasone (6 mg,12 hourly, four doses) (Senat 1998);

  • one trial compared 24 mg dexamethasone (12 mg, 12 hourly, two doses) and 24 mg betamethasone (12 mg, 24 hourly, two doses) (Mulder 1997).

One trial of 170 women and 188 infants compared 32 mg oral dexamethasone (8 mg, 12 hourly, four doses) and 24 mg intramuscular dexamethasone (6 mg, 12 hourly, four doses) (Egerman 1998), and one trial of 228 mothers and 260 fetuses compared dosing intervals of betamethasone; 12 mg, 24 hourly, two doses versus 12 mg, 12 hourly, two doses (Khandelwal 2012).

The Subtil 2003 trial compared two forms of betamethasone (acetate and phosphate versus phosphate alone) as a third arm.

Four of the trials allowed repeat weekly doses of the allocated corticosteroid (Egerman 1998; Magee 1997; Mushkat 2001; Senat 1998).

The gestational age at trial entry varied widely between trials (from 23 to 35 weeks of gestation). All women were at increased risk of preterm birth or had a medical indication for birth at a preterm gestational age (see Characteristics of included studies).

The included studies came from a range of healthcare systems. Three of the trials were conducted in the USA (Egerman 1998; Elimian 2007, Khandelwal 2012), two in France (Senat 1998; Subtil 2003), two in Israel (Mushkat 2001; Rotmensch 1999) and one in each of Taiwan (Chen 2005), UK (Magee 1997), Netherlands (Mulder 1997), Poland (Urban 2005) and Iran (Danesh 2012). The trials were conducted over more than two decades from 1990 to 2012.

The primary outcomes varied between the trials. The primary outcomes of six of the trials focused on neonatal outcomes including RDS, IVH and death and at times, child outcomes of neurodisability at 18 months (Chen 2005; Egerman 1998; Elimian 2007; Khandelwal 2012; Senat 1998; Subtil 2003), while the five other trials concentrated on biophysical parameters of the fetus (Magee 1997; Mulder 1997; Mushkat 2001; Rotmensch 1999; Urban 2005). One trial focused on effects on maternal serum indicators of infection (Danesh 2012).

Excluded studies

Eight trials were excluded: three because L-carnitine added to a corticosteroid was compared against a corticosteroid to assess the effect of L-carnitine (Kurz 1993; Salzer 1982; Vytiska 1985); one trial was excluded as thyroxine added to a corticosteroid was compared against a corticosteroid to assess the effects of thyroxine (Romaguera 1997); one trial was excluded as vitamin K added to a corticosteroid was compared against a corticosteroid, vitamin K and no treatment to assess the effects of vitamin K (Liu 2006); one trial was excluded because it was a cross-over trial (Egerman 1997); one trial was excluded because it compared corticosteroids (either dexamethasone or betamethasone) with placebo (Shanks 2010); and one trial was not randomised (Whitt 1976). For further details see Characteristics of excluded studies.

Risk of bias in included studies

Overall, the trials were judged to be at a moderate risk of bias, see Figure 1 and Figure 2.

Figure 1.

'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 2.

'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.

For further details on the risk of bias in individual studies, see Characteristics of included studies.

Allocation

Methods of sequence generation and allocation concealment were both considered adequate in five of the 12 trials (Danesh 2012; Elimian 2007; Khandelwal 2012; Magee 1997; Urban 2005).

In four further trials, while methods of random sequence generation were adequate, methods to conceal allocation were unclear (Egerman 1998; Rotmensch 1999; Senat 1998; Subtil 2003). Two trials were judged to be at an unclear risk of selection bias, with methods of sequence generation and allocation concealment being unclear (Chen 2005; Mulder 1997).

The remaining trial, Mushkat 2001 was quasi-randomised and assessed as having a high risk of selection bias.

Blinding

Three trials were judged to be at a low risk of performance bias, with blinding of women and personnel (Elimian 2007; Magee 1997; Mushkat 2001). For four trials, the risk of performance bias was judged to be unclear, as blinding of women and personnel was not detailed (Chen 2005; Mulder 1997; Rotmensch 1999; Urban 2005). The remaining five trials were judged to be at a high risk of performance bias, with no blinding of women and study personnel (or blinding considered unfeasible) (Danesh 2012; Egerman 1998; Khandelwal 2012; Senat 1998; Subtil 2003).

Four trials were judged to be at a low risk of detection bias, with blinding of outcome assessment (Egerman 1998; Elimian 2007; Khandelwal 2012; Magee 1997). One trial was judged to be at a high risk of detection bias, with no blinding of outcome assessors (Subtil 2003); for the other seven trials, the risk of detection bias was judged to be unclear (Chen 2005; Danesh 2012; Mulder 1997; Mushkat 2001; Rotmensch 1999; Senat 1998; Urban 2005).

Incomplete outcome data

Five trials were judged to be at low risk of attrition bias (Danesh 2012; Egerman 1998; Khandelwal 2012; Senat 1998; Urban 2005). 

Losses to follow-up were unclear and not documented in the Rotmensch 1999 and Mushkat 2001 trials. Mulder 1997 reported a 17% loss to follow-up (6/30 from the dexamethasone group and 4/30 from the betamethasone group), and in Elimian 2007 less than 60% of all infants were assessed for IVH and PVL. In Magee 1997 and Subtil 2003, losses were greater than 50% at the end of follow-up for some of the biophysical parameters.

The Chen 2005 trial was judged at a high risk of attrition bias, as data were excluded for 16% of women, with no indication from which groups the losses had occurred.

Selective reporting

There was no obvious risk of selective reporting in three trials (Elimian 2007; Khandelwal 2012; Subtil 2003).

While two trials pre-specified their outcomes in the manuscript methods, the risk of reporting bias was judged to be unclear, with outcome data reported incompletely for some clinical outcomes, e.g."gestational age, birthweight and Agpar score at five minutes did not differ between the two groups" (Egerman 1998; Mushkat 2001).

The remaining seven trials reported some important clinical outcomes however with no access to a trial protocol it was difficult to confidently assess selective reporting; we therefore judged these studies to be at an unclear risk of reporting bias (Chen 2005; Danesh 2012; Magee 1997; Mulder 1997; Rotmensch 1999; Senat 1998; Urban 2005).

Other potential sources of bias

For eight trials, groups were comparable at baseline and there were no other obvious sources of bias identified (Danesh 2012; Egerman 1998; Elimian 2007; Magee 1997; Mulder 1997; Mushkat 2001; Rotmensch 1999; Subtil 2003; Urban 2005). 

Other bias was judged as unclear in the Chen 2005 and Khandelwal 2012 trials due to some baseline imbalances between groups regarding age and ethnicity.

Effects of interventions

Twelve trials involving 1557 women and 1661 babies were included.

The results are presented by type of corticosteroid or method of administration compared:

1. Dexamethasone versus betamethasone

Ten trials with 1159 women and 1213 infants were included in this comparison.

Primary outcomes
Women

No primary outcomes for women were reported in any of the included trials.

Infants

No statistically significant differences between those exposed to dexamethasone or betamethasone were seen for neonatal death (risk ratio (RR) 1.41, 95% confidence interval (CI) 0.54 to 3.67; four trials, 596 infants) (Analysis 1.1) or respiratory distress syndrome (RDS) (RR 1.06, 95% CI 0.88 to 1.27; five trials; 753 infants) (Analysis 1.2).

Danesh 2012 reported on the "number of infants admitted to the neonatal intensive care unit (NICU) because of respiratory distress syndrome" for women with intact membranes; 9/60 in the dexamethasone group were admitted versus 12/60 in the betamethasone group. It was unclear however as to whether this represented all cases of RDS or only those admitted to the NICU, and thus these data have not been included in the review meta-analysis for RDS.

Dexamethasone significantly decreased the risk of intraventricular haemorrhage (IVH) compared with betamethasone (RR 0.44, 95% CI 0.21 to 0.92; four trials, 549 infants) (Analysis 1.3).

Children

Out of a small subgroup assessed at 18 months in the Subtil 2003 trial, one child in the dexamethasone group was recorded as having a neurosensory disability (RR 1.67, 95% CI 0.08 to 33.75; one trial, 12 infants - Subtil 2003) (Analysis 1.4). Death in childhood was not reported as an outcome in any of the included trials.

For the child as adult

No primary outcomes for the child as an adult were reported in any of the included trials.

Secondary outcomes
Women

No secondary outcomes for women were reported in any of the included trials. While in Danesh 2012 no data regarding adverse effects were reported for inclusion in a meta-analysis, it was stated that "Both, dexamethasone and betamethasone treatment was tolerated well and most of the adverse events reported were mild in severity".

Infants

No statistically significant differences between those exposed to dexamethasone or betamethasone were seen for Apgar score less than seven at five minutes (RR 0.97, 95% CI 0.43 to 2.18; two trials, 207 infants) (Analysis 1.5), Apgar score at five minutes (mean difference (MD) 0.23, 95% CI -0.23 to 0.70; two trials, 307 infants) (Analysis 1.6, random-effects, T² = 0.11; I² = 64%), mean birthweight (MD 0.01 kg, 95% CI -0.11 to 0.12; five trials, 734 infants) (Analysis 1.8), low birthweight less than 2500 g (RR 0.89, 95% CI 0.65 to 1.24; one trial, 105 infants) (Analysis 1.10), or head circumference (MD -0.50 cm, 95% CI -1.55 to 0.55; one trial, 157 infants) (Analysis 1.11).

Overall, therefore, there was no significant difference in the risk of NICU admission (RR 1.72, 95% CI 0.44 to 6.72; two trials, 345 infants) (Analysis 1.12). As we identified substantial statistical heterogeneity for this outcome (T² = 0.80; I² = 81%) a random-effects model was used. It is possible that the differences in reasons for risk of preterm birth, or other clinical factors, between these trials contributed to this level of heterogeneity.

In one of the trials, significantly more infants were admitted to the NICU in the dexamethasone group compared with the betamethasone group. Seven of the eight infants in the dexamethasone group were transferred to the NICU because of respiratory distress and the remaining infant was transferred due to suspected infection. The reason for all four infants in the betamethasone group being transferred to NICU was because of respiratory distress. In the other trial of 240 infants (Danesh 2012), however, there was no significant difference in NICU admissions with 34 admissions in the dexamethasone group, compared with 36 in the betamethasone group. It appears that all 70 infants were transferred to NICU due to respiratory distress, however this was unclear.

No significant differences between the dexamethasone and betamethasone groups were seen for the outcomes vasopressor use (RR 0.44, 95% CI 0.17 to 1.11; one trial, 359 infants) (Analysis 1.13), bronchopulmonary dysplasia (RR 2.50, 95% CI 0.10 to 61.34; two trials, 464 infants) (Analysis 1.14), severe IVH (RR 0.40, 95% CI 0.13 to 1.24; four trials, 549 infants) (Analysis 1.15), periventricular leukomalacia (RR 0.83, 95% CI 0.23 to 3.03; four trials, 703 infants) (Analysis 1.16), neonatal sepsis (RR 1.30, 95% CI 0.78 to 2.19; two trials, 516 infants) (Analysis 1.17), necrotising enterocolitis (RR 1.29, 95% CI 0.38 to 4.40; three trials, 598 infants) (Analysis 1.18), retinopathy of prematurity (RR 0.93, 95% CI 0.59 to 1.47; two trials, 516 infants) (Analysis 1.19), or patent ductus arteriosus (RR 1.19, 95% CI 0.56 to 2.49; one trial, 359 infants) (Analysis 1.20). Considering the outcome bronchopulmonary dysplasia, there was substantial statistical heterogeneity (T² = 4.40; I² = 80%), and thus a random-effects model was used.

Some differences in biophysical parameters were seen:

  • the dexamethasone group had a significantly lower fetal heart rate than the betamethasone group at day two (MD -4.20 beats per minute, 95% CI -7.17 to -1.23; one trial, 46 infants - Rotmensch 1999) (Analysis 1.21), in Senat 1998 (see 'Other data' table (Analysis 1.22)); but not in Magee 1997;

  • the dexamethasone group had a significantly higher level of fetal movements detected via ultrasound than the betamethasone group (MD 7.0 movements per hour, 95% CI 2.15 to 11.85; one trial, 33 infants - Mushkat 2001) (Analysis 1.27);

  • at day two, the dexamethasone group had significantly higher breathing times than the betamethasone group (MD 32.00 more seconds per 30 minutes, 95% CI 4.37 to 59.63; one trial, 46 infants - Rotmensch 1999) (Analysis 1.30).

No statistically significant differences between those exposed to betamethasone or dexamethasone were seen for other biophysical parameters including accelerations per hour (MD 2.80, 95% CI -0.15 to 5.75; one trial, 46 infants - Rotmensch 1999) (Analysis 1.23), fetal movements in 30 minutes (MD 2.30, 95% CI -0.74 to 5.34; one trial, 46 infants - Rotmensch 1999) (Analysis 1.25), fetal movement via maternal perception (MD 3.00 movements per hour, 95% CI -3.20 to 9.20; one trial, 33 infants - Mushkat 2001) (Analysis 1.26), and fetal breathing movements per hour (MD 0.00, 95% CI -2.05 to 2.05; one trial, 33 infants - Mushkat 2001) (Analysis 1.29) .

Some additional data in the form of median and interquartile ranges are shown in 'Other data' tables, with no differences seen between dexamethasone and betamethasone except for a lower heart fetal rate in Senat 1998 for dexamethasone (Analysis 1.22).

A range of fetal heart rate indicators were measured in Subtil 2003 but only reported in graphical form; the trial authors reported that none of the indicators showed significant differences between dexamethasone and betamethasone.

Where statistical heterogeneity has not been reported for outcomes discussed above (if more than one trial was included in the meta-analysis), there was no heterogeneity observed between trials (I² = 0).

Children

No secondary outcomes for children were reported in any of the included trials.

For the child as adult

No secondary outcomes for children as adults were reported in any of the included trials.

Health services

Mean length of antenatal admission to birth (days) was reported in one trial of 240 women (Danesh 2012), and no significant difference was observed overall between the dexamethasone and betamethasone groups (MD 3.48 days, 95% CI -3.38 to 10.34; 240 women) (Analysis 1.31). The Danesh 2012 trial reported data separately for women with intact and ruptured membranes, and a substantial level of heterogeneity was observed for this outcome across these subgroups (T² = 61.91; I² = 98%). Women with intact membranes who received betamethasone were significantly more likely to have a shorter length of admission to birth than women who received dexamethasone; this was not the case for women with ruptured membranes.

The Danesh 2012 trial also reported on mean length of NICU admission (for the 70 infants admitted to the NICU), and found a significant reduction in length of stay for infants who had been exposed to dexamethasone, as compared with betamethasone (MD -0.91 days, 95% CI -1.77 to -0.05; 70 infants) (Analysis 1.32).

Indirect comparisons and subgroup interaction tests

Using the methods outlined in the additional material accompanying Song 2003, we calculated indirect comparisons of the trials of betamethasone versus placebo/no treatment, and dexamethasone versus placebo/no treatment that were included in the Roberts 2006 Cochrane review (see Table 1). We also performed subgroup interaction tests on these trials from the Roberts 2006 Cochrane review (see Table 1).

Considering fetal/neonatal death, the indirect estimate and subgroup interaction test were compatible with the findings of this review, indicating no significant difference between dexamethasone and betamethasone for this outcome.

For the outcome RDS, while the discrepancy between the direct estimate from this review and the indirect estimate was not significant, the indirect comparison did indicate a significant difference between dexamethasone and betamethasone in favour of betamethasone (RR 1.40, 95% CI 1.02 to 1.90) that was inconsistent with the non significant difference shown in this review. In addition, the subgroup interaction test for RDS was inconsistent with the direct comparison in this review, indicating a significant difference between subgroups and a possible differential effect between the two steroids in favour of betamethasone (Chi² statistic: 4.68 and P value: 0.03, I² value: 78.6%).

While the indirect estimate for IVH did not reveal a significant difference between the two steroids as was shown in the direct comparison in this review (in favour of dexamethasone), the discrepancy between the indirect and direct estimates was non significant. The subgroup interaction test was however inconsistent with the direct estimate from this review, failing to show a significant difference between the two steroids for IVH (Chi² statistic: 0.78 and P value: 0.38, I² value: 0%).

Although chorioamnionitis and puerperal sepsis were not reported in any trials directly comparing betamethasone and dexamethasone included in this review, we estimated the indirect estimates and tested for subgroup differences in the trials from the Roberts 2006 review. While the indirect estimate and subgroup interaction test suggested no significant difference between dexamethasone and betamethasone for puerperal sepsis, both the indirect estimate (RR 1.90, 95% CI 1.10 to 3.28) and the subgroup interaction test (Chi² statistic: 5.41 and P value 0.02, I² value: 81.5%) indicated a significant difference between the two steroids for chorioamnionitis, in favour of betamethasone.

2. Dexamethasone (oral versus intramuscular injection)

One trial with 170 women and 188 infants (Egerman 1998) was included in this comparison.

Primary outcomes
Women

No primary outcomes for women were reported in this trial.

Infants

No statistically significant differences between oral or intramuscular dexamethasone were seen for neonatal death (RR 1.48, 95% CI 0.45 to 4.90) (Analysis 2.1), or RDS (RR 1.15, 95% CI 0.75 to 1.77) (Analysis 2.2). No significant difference was seen between oral and intramuscular dexamethasone for IVH (RR 4.24, 95% CI 0.96 to 18.33), although this did reach statistical significance in favour of the intramuscular route when looking at only the less than 34 weeks' gestation at birth subgroup (RR 4.92, 95% CI 1.12 to 21.55) (Analysis 2.3). All instances of IVH (10 in oral group and two in the intramuscular group) occurred in babies born before 34 weeks.

Children

No primary outcomes for children were reported in this trial.

For the child as adult

No primary outcomes for children as adults were reported in this trial.

Secondary outcomes
Women

No secondary outcomes for women were reported in this trial.

Infants

There was no statistically significant difference between oral and intramuscular dexamethasone seen for birthweight (MD -0.05 kg, 95% CI -0.17 to 0.27) (Analysis 2.4) or necrotising enterocolitis (RR 5.09, 95% CI 0.63 to 41.45) (Analysis 2.6).

Treatment with oral dexamethasone was associated with an increase in neonatal sepsis compared with intramuscular dexamethasone (RR 8.48, 95% CI 1.11 to 64.93) (Analysis 2.5) with all neonatal sepsis cases occurring in the less than 34 weeks' gestation subgroup (RR 9.84, 95% CI 1.30 to 74.60).

Children

No secondary outcomes for children were reported in this trial.

For the child as adult

No secondary outcomes for children as adults were reported in this trial.

3. Betamethasone acetate + phosphate versus betamethasone phosphate

One trial with 69 infants (Subtil 2003) was included in this comparison.

Primary outcomes
Women

No primary outcomes for women were reported in this trial.

Infants

No statistically significant differences between those exposed to betamethasone acetate and phosphate versus betamethasone phosphate were seen for neonatal death (RR 0.32, 95% CI 0.01 to 7.69) (Analysis 3.1), RDS (RR 0.19, 95% CI 0.01 to 3.91) (Analysis 3.2), or IVH (RR 0.32, 95% CI 0.01 to 7.69) (Analysis 3.3).

Children

None of the children were reported to have a neurodevelopmental disability at 18 months (Analysis 3.4).

For the child as adult

No primary outcomes for children as adults were reported in this trial.

Secondary outcomes
Infants

No statistically significant differences between those exposed to betamethasone acetate and phosphate versus betamethasone phosphate were seen for birthweight (MD -0.10 kg, 95% CI -0.44 to 0.24) (Analysis 3.5) or low birthweight (RR 1.21, 95% CI 0.86 to 1.72) (Analysis 3.6). No infants from the betamethasone acetate and phosphate group were transferred to NICU compared with four in the betamethasone phosphate group, all four due to respiratory distress (RR 0.11, 95% CI 0.01 to 1.93) (Analysis 3.7); this difference was not statistically significant. No instances of bronchopulmonary dysplasia (Analysis 3.8) or periventricular leukomalacia (Analysis 3.9) were reported in this trial.

A range of fetal heart rate indicators were measured but only reported in graphical form. The trial authors reported that none of the indicators showed significant differences between the different betamethasone formulations.

4. Betamethasone dosing interval: 12 hourly versus 24 hourly

One trial with 260 infants (Khandelwal 2012) was included in this comparison.

Primary outcomes
Women

No primary outcomes for women were reported in this trial.

Infants

No statistically significant differences were seen when a 12-hour dosing interval of betamethasone (12 mg) was compared to a 24-hour dosing interval for perinatal mortality (fetal and neonatal mortality were not reported separately) (RR 0.93, 95% CI 0.46 to 1.87) (Analysis 4.1), RDS (RR 0.98, 95% CI 0.69 to 1.40) (Analysis 4.2), or IVH (RR 1.40, 95% CI 0.76 to 2.56) (Analysis 4.3).

Children

No primary outcomes for children were reported in this trial.

For the child as adult

No primary outcomes for children as adults were reported in this trial.

Subgroup analysis

The Khandelwal 2012 trial considered separately women at the following gestational age categories at trial entry: 23+1 to 26+0 weeks' gestation; 26+1 to 29+0 weeks' gestation; 29+1 to 32+0 weeks' gestation; and 32+1 to 34+0 weeks' gestation.

No significant differences were shown for the three infant primary outcomes that were reported (perinatal death, RDS and IVH) for any of the subgroups, and subgroup interaction tests were not significant for any of the three outcomes, indicating no differential treatment effect by gestational age at trial entry (perinatal mortality: Chi² statistic: 0.93 and P value: 0.92, I² value: 0%; Analysis 4.1) (RDS: Chi² statistic: 5.62 and P value: 0.23, I² value: 28.8%; Analysis 4.2) (IVH: Chi² statistic: 2.42 and P value: 0.66, I² value: 0%; Analysis 4.3).

Secondary outcomes
Women

Maternal fever (defined as greater than 100.4°F) was not significantly different between groups (RR 0.71, 95% CI 0.25 to 2.02) (Analysis 4.4).

Infants

No statistically significant differences between groups were seen for birthweight (MD 84.00 g, 95% CI -144.63 to 312.63) (Analysis 4.5), small-for-gestational age (RR 0.61, 95% CI 0.36 to 1.05) (Analysis 4.6), chronic lung disease (RR 0.79, 95% CI 0.49 to 1.26) (Analysis 4.8), neonatal sepsis (RR 1.15, CI 0.47 to 2.81) (Analysis 4.9), neonatal antibiotic use of more than five days (RR 0.94, 95% CI 0.61 to 1.46) (Analysis 4.10) or retinopathy of prematurity (RR 0.94, 95% CI 0.53 to 1.66) (Analysis 4.12). A trend towards reduced admission to NICU for the 12-hourly regimen group as compared with the 24-hourly regimen group was observed (P = 0.05) (RR 0.89, 95% CI 0.79 to 1.00) (Analysis 4.7).

No infants in the 24-hour group developed necrotising enterocolitis while 10 infants in the 12-hour group developed necrotising enterocolitis, however this was not statistically significant (RR 9.20, 95% CI 0.55 to 154.92) (Analysis 4.11).

Health services

Women who received the 12-hour dosing interval of betamethasone had a significantly shorter mean maternal postpartum length of stay than women who received the 24-hour dosing interval (MD -0.73 days, 95% CI -1.28 to -0.18) (Analysis 4.13).

Discussion

That antenatal corticosteroids are effective in preventing neonatal morbidity is not in dispute (Roberts 2006) and this life-saving therapy is now widely used throughout the world (Chow 2013; Brocklehurst 1999; Foix-L'Helias 2008; NIH 2000; Quinlivan 1998; Saengwaree 2005). However, it is not yet clear which corticosteroid and which regimens perform best. Determining the optimal corticosteroid and optimal regimens is very important since most pregnant women at risk of preterm birth will be considered candidates for antenatal corticosteroid treatment (NIH 2000) and these numbers may continue to grow as rates of preterm birth are increasing in a number of countries (Chow 2013; Goldenberg 2007).

There is considerable variation reported between countries as to whether dexamethasone or betamethasone is preferred by health practitioners, with many likely reasons for these differences including availability and costs (dexamethasone is cheaper than betamethasone so is widely used in low- and middle-income countries) (Henderson-Smart 2007; Saengwaree 2005), impact of inconsistent findings from observational studies (Baud 1999; Lee 2006) and influence of opinion leaders (Jobe 2004).

Although we were able to include 12 trials (involving 1557 women and 1661 infants) rated of moderate quality, including one quasi-randomised trial in this review (Mushkat 2001), our ability to reach conclusions was limited by the small number of comparisons of different antenatal steroid regimens. Most of the data available focused on the type of corticosteroid used, with 10 of the studies comparing the two most commonly used corticosteroids, dexamethasone and betamethasone (with some variation in frequency and timing of administration).

The results of this review are broadly consistent with results of the Roberts 2006 Cochrane review of antenatal corticosteroids when they are recalculated as indirect comparisons of dexamethasone versus betamethasone (see Table 1) (with no significant discrepancies between the direct and indirect estimates found). However, the suggestion of increased benefit of dexamethasone over betamethasone from this review for IVH is not sufficient evidence to support dexamethasone over betamethasone; while the discrepancy between the indirect and direct estimates for IVH was not significant, the indirect comparison from the Roberts 2006 review did not support the significant difference between the two steroids for IVH shown in this review, and neither did the subgroup interaction test. A recent observational study, which reported reduced adverse neurological outcomes at 18 to 22 months for betamethasone but not for dexamethasone, highlights the persisting uncertainty, stating that "to elucidate more fully predictive or causative neonatal or neurodevelopmental outcomes, a randomised clinical trial comparing dexamethasone and betamethasone should be performed" (Lee 2008). Such a trial would need to measure long-term effects, particularly for dexamethasone, as there has been no long-term follow-up reported in studies that have used this type of corticosteroid. One such trial has just completed recruitment (Crowther 2010).

Very few maternal outcomes were reported in the trials included in this review, with none of the review's primary or secondary maternal review outcomes reported in the 10 trials that compared dexamethasone and betamethasone. As noted in the results section, while the indirect comparison of dexamethasone and betamethasone from the Roberts 2006 review for puerperal sepsis suggested no significant difference between the two steroids, both the indirect estimate and subgroup interaction test suggested a significant difference for chorioamnionitis (in favour of betamethasone), which requires further evaluation. This review has not been able to provide any further evidence as none of the included trials reported on chorioamnionitis.

Although extensively reported in several of the included trials (Magee 1997; Mushkat 2001; Rotmensch 1999; Senat 1998; Subtil 2003), the clinical significance of differences in biophysical parameters such as fetal heart rate and respiratory rate is not clear (Rotmensch 1999). Overall, these trials generally showed few differences between dexamethasone and betamethasone, except for a significantly lower heart rate at day two, a longer duration of breathing time at day two, and more fetal movements detected via ultrasound for the dexamethasone group. Some authors suggest that the influence of antenatal corticosteroids on parameters such as fetal heart rate is not clinically important, being a transient physiological response (Magee 1997; Rotmensch 1999; Subtil 2003).

Evidence about optimal doses, timing and frequency of administration of specific antenatal corticosteroids was even more sparse than that for type of corticosteroid, with three of the 12 trials contributing data to three separate comparisons. In regards to oral versus intramuscular dexamethasone, some benefits were shown for intramuscular administration, in regards to less neonatal sepsis, and a reduction in IVH for infants born before 34 weeks' gestation. No differences were seen for any of the reported outcomes when betamethasone acetate and phosphate was compared with betamethasone phosphate alone, although information was only available from one trial of 69 infants. In the trial that compared 12-hourly versus 24-hourly betamethasone administration, no differences were shown between regimens for the review's primary outcomes, however a reduction in maternal postpartum length of stay for women who received betamethasone at 12-hourly intervals was observed, and a trend towards reduced NICU admissions was also seen for infants in the 12-hourly group.

Authors' conclusions

Implications for practice

Dexamethasone may have some benefits compared with betamethasone such as less IVH, possibly some improved biophysical parameters and a shorter length of NICU stay. Apart from the superiority of an intramuscular compared with an oral route for dexamethasone in one trial and the similar efficacy of 24-hour versus 12-hour dosing of betamethasone, so far there is little evidence to guide clinical practice about optimal antenatal corticosteroid regimens.

Implications for research

Further trials directly comparing the type, dose, frequency and route of betamethasone with dexamethasone for women at risk of preterm birth are required. They should be of high quality, large enough to assess morbidity and mortality of the fetus/infant, long-term outcomes and maternal outcomes; one such study has just completed recruitment (Crowther 2010). High-quality trials are needed to establish which of the commonly used corticosteroids (dexamethasone or betamethasone) is most effective and causes least harm (including at longer-term, assessments), as are trials of dosages and other variations in regimens.

Acknowledgements

For this update, we thank Frances Kellie, Leanne Jones, Denise Atherton and Lynn Hampson of the Cochrane Pregnancy and Childbirth Group for their support.

Special thanks to Sonja Henderson, Denise Atherton and Lynn Hampson for their support and guidance throughout previous versions of this review. Thanks to Dr Reinaldo Figueroa and Dr Laura Magee, authors of included trials, for providing additional information for the previous update of this review.

As part of the pre-publication editorial process, this review has been commented on by three peers (an editor and three referees who are external to the editorial team) and the Group's Statistical Adviser.

The National Institute for Health Research (NIHR) is the largest single funder of the Cochrane Pregnancy and Childbirth Group. The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the NIHR, NHS or the Department of Health.

Data and analyses

Download statistical data

Comparison 1. Dexamethasone versus betamethasone
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Neonatal death4596Risk Ratio (M-H, Fixed, 95% CI)1.41 [0.54, 3.67]
1.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)2464Risk Ratio (M-H, Fixed, 95% CI)1.12 [0.38, 3.33]
1.2 Dexamethasone (16 mg - 4 x 4 mg; 12 hourly) v betamethasone (24 mg - 4 x 6 mg; 12 hourly)182Risk Ratio (M-H, Fixed, 95% CI)3.15 [0.13, 75.05]
1.3 Dexamethasone (24 mg - 2 x 12 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)150Risk Ratio (M-H, Fixed, 95% CI)3.24 [0.14, 75.91]
2 Respiratory distress syndrome5753Risk Ratio (M-H, Fixed, 95% CI)1.06 [0.88, 1.27]
2.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)3621Risk Ratio (M-H, Fixed, 95% CI)1.08 [0.89, 1.30]
2.2 Dexamethasone (16 mg - 4 x 4 mg; 12 hourly) v betamethasone (24 mg - 4 x 6 mg; 12 hourly)182Risk Ratio (M-H, Fixed, 95% CI)0.35 [0.01, 8.34]
2.3 Dexamethasone (24 mg - 2 x 12 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)150Risk Ratio (M-H, Fixed, 95% CI)0.36 [0.02, 8.43]
3 Intraventricular haemorrhage4 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
3.1 Intraventricular haemorrhage (any dose)4549Risk Ratio (M-H, Fixed, 95% CI)0.44 [0.21, 0.92]
3.2 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)3467Risk Ratio (M-H, Fixed, 95% CI)0.44 [0.21, 0.92]
3.3 Dexamethasone (16 mg - 4 x 4 mg; 12 hourly) v betamethasone (24 mg - 4 x 6 mg; 12 hourly)182Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
4 Neurosensory disability as a child (18 months)112Risk Ratio (M-H, Fixed, 95% CI)1.67 [0.08, 33.75]
4.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)112Risk Ratio (M-H, Fixed, 95% CI)1.67 [0.08, 33.75]
5 Apgar score < 7 at 5 minutes2207Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.43, 2.18]
5.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)1157Risk Ratio (M-H, Fixed, 95% CI)1.07 [0.45, 2.54]
5.2 Dexamethasone (24 mg - 2 x 12 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)150Risk Ratio (M-H, Fixed, 95% CI)0.54 [0.05, 5.60]
6 Apgar score at 5 minutes2307Mean Difference (IV, Random, 95% CI)0.23 [-0.23, 0.70]
6.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)167Mean Difference (IV, Random, 95% CI)-0.20 [-0.89, 0.49]
6.2 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly); intact membranes1120Mean Difference (IV, Random, 95% CI)0.60 [0.26, 0.94]
6.3 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly); ruptured membranes1120Mean Difference (IV, Random, 95% CI)0.10 [-0.37, 0.57]
7 Apgar score at 5 minutes  Other dataNo numeric data
8 Birthweight (kg)5734Mean Difference (IV, Fixed, 95% CI)0.01 [-0.11, 0.12]
8.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)5734Mean Difference (IV, Fixed, 95% CI)0.01 [-0.11, 0.12]
9 Birthweight (kg)  Other dataNo numeric data
10 Low birthweight1105Risk Ratio (M-H, Fixed, 95% CI)0.89 [0.65, 1.24]
10.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)1105Risk Ratio (M-H, Fixed, 95% CI)0.89 [0.65, 1.24]
11 Head circumference (cm)1157Mean Difference (IV, Fixed, 95% CI)-0.5 [-1.55, 0.55]
11.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)1157Mean Difference (IV, Fixed, 95% CI)-0.5 [-1.55, 0.55]
12 Neonatal intensive care unit admission2345Risk Ratio (M-H, Random, 95% CI)1.72 [0.44, 6.72]
12.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)2345Risk Ratio (M-H, Random, 95% CI)1.72 [0.44, 6.72]
13 Vasopressor use1359Risk Ratio (M-H, Fixed, 95% CI)0.44 [0.17, 1.11]
13.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)1359Risk Ratio (M-H, Fixed, 95% CI)0.44 [0.17, 1.11]
14 Bronchopulmonary dysplasia2464Risk Ratio (M-H, Random, 95% CI)2.50 [0.10, 61.34]
14.1 Dexamethasone (24 mg - 4 x 6 mg, 12 hourly) v betamethasone (24 mg - 2 x 12 mg, 24 hourly)2464Risk Ratio (M-H, Random, 95% CI)2.50 [0.10, 61.34]
15 Severe intraventricular haemorrhage4549Risk Ratio (M-H, Fixed, 95% CI)0.40 [0.13, 1.24]
15.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)3467Risk Ratio (M-H, Fixed, 95% CI)0.40 [0.13, 1.24]
15.2 Dexamethasone (16 mg - 4 x 4 mg; 12 hourly) v betamethasone (24 mg - 4 x 6 mg; 12 hourly)182Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
16 Periventricular leukomalacia4703Risk Ratio (M-H, Fixed, 95% CI)0.83 [0.23, 3.03]
16.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)3621Risk Ratio (M-H, Fixed, 95% CI)0.83 [0.23, 3.03]
16.2 Dexamethasone (16 mg - 4 x 4 mg; 12 hourly) v betamethasone (24 mg - 4 x 6 mg; 12 hourly)182Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
17 Neonatal sepsis2516Risk Ratio (M-H, Fixed, 95% CI)1.30 [0.78, 2.19]
17.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)2516Risk Ratio (M-H, Fixed, 95% CI)1.30 [0.78, 2.19]
18 Necrotising enterocolitis3598Risk Ratio (M-H, Fixed, 95% CI)1.29 [0.38, 4.40]
18.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)2516Risk Ratio (M-H, Fixed, 95% CI)1.29 [0.38, 4.40]
18.2 Dexamethasone (16 mg - 4 x 4 mg; 12 hourly) v betamethasone (24 mg - 4 x 6 mg; 12 hourly)182Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
19 Retinopathy of prematurity2516Risk Ratio (M-H, Fixed, 95% CI)0.93 [0.59, 1.47]
19.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)2516Risk Ratio (M-H, Fixed, 95% CI)0.93 [0.59, 1.47]
20 Patent ductus arteriosus1359Risk Ratio (M-H, Fixed, 95% CI)1.19 [0.56, 2.49]
20.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)1359Risk Ratio (M-H, Fixed, 95% CI)1.19 [0.56, 2.49]
21 Fetal heart rate, bpm (day 2)146Mean Difference (IV, Fixed, 95% CI)-4.20 [-7.17, -1.23]
21.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)146Mean Difference (IV, Fixed, 95% CI)-4.20 [-7.17, -1.23]
22 Fetal heart rate (day 2)  Other dataNo numeric data
23 Accelerations per hour146Mean Difference (IV, Fixed, 95% CI)2.80 [-0.15, 5.75]
23.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg, 24 hourly)146Mean Difference (IV, Fixed, 95% CI)2.80 [-0.15, 5.75]
24 Accelerations per hour  Other dataNo numeric data
25 Fetal movements in 30 minutes146Mean Difference (IV, Fixed, 95% CI)2.3 [-0.74, 5.34]
25.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)146Mean Difference (IV, Fixed, 95% CI)2.3 [-0.74, 5.34]
26 Fetal movements per hour (maternal perception)133Mean Difference (IV, Fixed, 95% CI)3.0 [-3.20, 9.20]
26.1 Dexamethasone (24 mg - 2 x 12 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 12 hourly)133Mean Difference (IV, Fixed, 95% CI)3.0 [-3.20, 9.20]
27 Fetal movements per hour (ultrasound)133Mean Difference (IV, Fixed, 95% CI)7.00 [2.15, 11.85]
27.1 Dexamethasone (24 mg - 2 x 12 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 12 hourly)133Mean Difference (IV, Fixed, 95% CI)7.00 [2.15, 11.85]
28 Fetal movements per hour  Other dataNo numeric data
29 Fetal breathing movements per hour133Mean Difference (IV, Fixed, 95% CI)0.0 [-2.05, 2.05]
29.1 Dexamethasone (24 mg - 2 x 12 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 12 hourly)133Mean Difference (IV, Fixed, 95% CI)0.0 [-2.05, 2.05]
30 Duration of breathing time at 2 days (seconds in 30 minutes)146Mean Difference (IV, Fixed, 95% CI)32.0 [4.37, 59.63]
30.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly)146Mean Difference (IV, Fixed, 95% CI)32.0 [4.37, 59.63]
31 Length of admission to birth (days)1240Mean Difference (IV, Random, 95% CI)3.48 [-3.38, 10.34]
31.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly); intact membranes1120Mean Difference (IV, Random, 95% CI)7.0 [5.56, 8.44]
31.2 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly); ruptured membranes1120Mean Difference (IV, Random, 95% CI)0.0 [-0.99, 0.99]
32 Neonatal intensive care unit stay (days)170Mean Difference (IV, Fixed, 95% CI)-0.91 [-1.77, -0.05]
32.1 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly); intact membranes121Mean Difference (IV, Fixed, 95% CI)-2.2 [-4.52, 0.12]
32.2 Dexamethasone (24 mg - 4 x 6 mg; 12 hourly) v betamethasone (24 mg - 2 x 12 mg; 24 hourly); ruptured membranes149Mean Difference (IV, Fixed, 95% CI)-0.70 [-1.63, 0.23]
Analysis 1.1.

Comparison 1 Dexamethasone versus betamethasone, Outcome 1 Neonatal death.

Analysis 1.2.

Comparison 1 Dexamethasone versus betamethasone, Outcome 2 Respiratory distress syndrome.

Analysis 1.3.

Comparison 1 Dexamethasone versus betamethasone, Outcome 3 Intraventricular haemorrhage.

Analysis 1.4.

Comparison 1 Dexamethasone versus betamethasone, Outcome 4 Neurosensory disability as a child (18 months).

Analysis 1.5.

Comparison 1 Dexamethasone versus betamethasone, Outcome 5 Apgar score < 7 at 5 minutes.

Analysis 1.6.

Comparison 1 Dexamethasone versus betamethasone, Outcome 6 Apgar score at 5 minutes.

Analysis 1.7.

Comparison 1 Dexamethasone versus betamethasone, Outcome 7 Apgar score at 5 minutes.

Apgar score at 5 minutes
StudyDexamethasoneBetamethasone
Magee 199724 mg - 2 x 12 mg; 12 hourly24 mg - 2 x 12 mg; 12 hourly
Magee 1997median 10: IQR 8.8 to 10 (n = 22)median 10: IQR 10 to 10 (n = 20)
Analysis 1.8.

Comparison 1 Dexamethasone versus betamethasone, Outcome 8 Birthweight (kg).

Analysis 1.9.

Comparison 1 Dexamethasone versus betamethasone, Outcome 9 Birthweight (kg).

Birthweight (kg)
StudyDexamethasoneBetamethasone
Magee 199724 mg - 2 x 12 mg; 12 hourly24 mg - 2 x 12 mg; 12 hourly
Magee 1997median 1.843 IQR 1.211 to 3.363 (n = 20)median 2.001 IQR 1.150 to 3.121 (n = 22)
Mulder 199724 mg - 2 x 12 mg; 12 hourly24 mg - 2 x 12 mg; 24 hourly
Mulder 1997median 2.02 IQR 0.75 to 3.5 (n = 24)median 1.79 IQR 0.84 to 3.75 (n = 26)
Senat 199816 mg - 4 x 4 mg; 12 hourly24 mg - 4 x 6 mg; 12 hourly
Senat 1998median 2.55 IQR 1.13 to 3.91 (n = 40)median 2.5 IQR 0.98 to 3.84 (n = 42)
Analysis 1.10.

Comparison 1 Dexamethasone versus betamethasone, Outcome 10 Low birthweight.

Analysis 1.11.

Comparison 1 Dexamethasone versus betamethasone, Outcome 11 Head circumference (cm).

Analysis 1.12.

Comparison 1 Dexamethasone versus betamethasone, Outcome 12 Neonatal intensive care unit admission.

Analysis 1.13.

Comparison 1 Dexamethasone versus betamethasone, Outcome 13 Vasopressor use.

Analysis 1.14.

Comparison 1 Dexamethasone versus betamethasone, Outcome 14 Bronchopulmonary dysplasia.

Analysis 1.15.

Comparison 1 Dexamethasone versus betamethasone, Outcome 15 Severe intraventricular haemorrhage.

Analysis 1.16.

Comparison 1 Dexamethasone versus betamethasone, Outcome 16 Periventricular leukomalacia.

Analysis 1.17.

Comparison 1 Dexamethasone versus betamethasone, Outcome 17 Neonatal sepsis.

Analysis 1.18.

Comparison 1 Dexamethasone versus betamethasone, Outcome 18 Necrotising enterocolitis.

Analysis 1.19.

Comparison 1 Dexamethasone versus betamethasone, Outcome 19 Retinopathy of prematurity.

Analysis 1.20.

Comparison 1 Dexamethasone versus betamethasone, Outcome 20 Patent ductus arteriosus.

Analysis 1.21.

Comparison 1 Dexamethasone versus betamethasone, Outcome 21 Fetal heart rate, bpm (day 2).

Analysis 1.22.

Comparison 1 Dexamethasone versus betamethasone, Outcome 22 Fetal heart rate (day 2).

Fetal heart rate (day 2)
StudyDexamethasoneBetamethasonep-value
Magee 199724 mg - 2 x 12 mg; 12 hourly24 mg - 2 x 12 mg; 12 hourly 
Magee 1997median change from baseline (bpm) 1.5: IQR -5.9 to 9.0 (n=22)median change from baseline (bpm) -2.0: IQR -6.0 to 5.0 (n=20)pns
Senat 199816 mg - 4 x 4 mg; 12 hourly24 mg - 4 x 6 mg; 12 hourly 
Senat 1998median bpm 142 IQR 137 to 149 (n=40)median bpm 147 IQR 141 to 154 (n=42)p=0.01 (for change data)
Analysis 1.23.

Comparison 1 Dexamethasone versus betamethasone, Outcome 23 Accelerations per hour.

Analysis 1.24.

Comparison 1 Dexamethasone versus betamethasone, Outcome 24 Accelerations per hour.

Accelerations per hour
StudyDexamethasoneBetamethasone
Magee 199724 mg - 2 x 12 mg; 12 hourly24 mg - 2 x 12 mg, 12 hourly
Magee 1997median change -1.0 IQR -4.0 to 0 (n=22)median change 0.5 IQR -2.5 to 2.5 (n=20)
Senat 199816 mg - 4 x 4 mg; 12 hourly24 mg - 4 x 6 mg; 12 hourly
Senat 1998median 5.0 IQR 2.2 to 10 (n=40)median 7.0 IQR 3.0 to 10 (n=42)
Analysis 1.25.

Comparison 1 Dexamethasone versus betamethasone, Outcome 25 Fetal movements in 30 minutes.

Analysis 1.26.

Comparison 1 Dexamethasone versus betamethasone, Outcome 26 Fetal movements per hour (maternal perception).

Analysis 1.27.

Comparison 1 Dexamethasone versus betamethasone, Outcome 27 Fetal movements per hour (ultrasound).

Analysis 1.28.

Comparison 1 Dexamethasone versus betamethasone, Outcome 28 Fetal movements per hour.

Fetal movements per hour
StudyDexamethasoneBetamethasone
Magee 199724 mg - 2 x 12 mg; 12 hourly24 mg - 2 x 12 mg; 12 hourly
Magee 1997median 2.5 IQR -27.0 to 32.0 (n=22)median 4.0 IQR -15.0 to 18.5 (n=20)
Analysis 1.29.

Comparison 1 Dexamethasone versus betamethasone, Outcome 29 Fetal breathing movements per hour.

Analysis 1.30.

Comparison 1 Dexamethasone versus betamethasone, Outcome 30 Duration of breathing time at 2 days (seconds in 30 minutes).

Analysis 1.31.

Comparison 1 Dexamethasone versus betamethasone, Outcome 31 Length of admission to birth (days).

Analysis 1.32.

Comparison 1 Dexamethasone versus betamethasone, Outcome 32 Neonatal intensive care unit stay (days).

Comparison 2. Dexamethasone: oral versus intramuscular
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Neonatal death1 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
1.1 Dexamethasone: oral (32 mg - 4 x 8 mg, 12 hourly) v IM (24 mg - 4 x 6 mg, 12 hourly)1183Risk Ratio (M-H, Fixed, 95% CI)1.48 [0.45, 4.90]
1.2 < 34 weeks' gestation1125Risk Ratio (M-H, Fixed, 95% CI)1.72 [0.53, 5.59]
2 Respiratory distress syndrome1 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
2.1 Dexamethasone: oral (32 mg - 4 x 8 mg, 12 hourly) v IM (24 mg - 4 x 6 mg, 12 hourly)1183Risk Ratio (M-H, Fixed, 95% CI)1.15 [0.75, 1.77]
2.2 < 34 weeks' gestation1125Risk Ratio (M-H, Fixed, 95% CI)1.26 [0.85, 1.86]
3 Intraventricular haemorrhage1 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
3.1 Dexamethasone: oral (32 mg - 4 x 8 mg, 12 hourly) v IM (24 mg - 4 x 6 mg, 12 hourly)1183Risk Ratio (M-H, Fixed, 95% CI)4.24 [0.96, 18.83]
3.2 < 34 weeks' gestation1125Risk Ratio (M-H, Fixed, 95% CI)4.92 [1.12, 21.55]
4 Birthweight (kg)1 Mean Difference (IV, Fixed, 95% CI)Subtotals only
4.1 Dexamethasone: oral (32 mg - 4 x 8 mg, 12 hourly) v IM (24 mg - 4 x 6 mg, 12 hourly)1183Mean Difference (IV, Fixed, 95% CI)0.05 [-0.17, 0.27]
5 Neonatal sepsis1 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
5.1 Dexamethasone: oral (32 mg - 4 x 8 mg, 12 hourly) v IM (24 mg - 4 x 6 mg, 12 hourly)1183Risk Ratio (M-H, Fixed, 95% CI)8.48 [1.11, 64.93]
5.2 < 34 weeks' gestation1125Risk Ratio (M-H, Fixed, 95% CI)9.84 [1.30, 74.60]
6 Necrotising enterocolitis1 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
6.1 Dexamethasone: oral (32 mg - 4 x 8 mg, 12 hourly) v IM (24 mg - 4 x 6 mg, 12 hourly)1183Risk Ratio (M-H, Fixed, 95% CI)5.09 [0.63, 41.45]
6.2 < 34 weeks' gestation1125Risk Ratio (M-H, Fixed, 95% CI)4.92 [0.59, 40.92]
Analysis 2.1.

Comparison 2 Dexamethasone: oral versus intramuscular, Outcome 1 Neonatal death.

Analysis 2.2.

Comparison 2 Dexamethasone: oral versus intramuscular, Outcome 2 Respiratory distress syndrome.

Analysis 2.3.

Comparison 2 Dexamethasone: oral versus intramuscular, Outcome 3 Intraventricular haemorrhage.

Analysis 2.4.

Comparison 2 Dexamethasone: oral versus intramuscular, Outcome 4 Birthweight (kg).

Analysis 2.5.

Comparison 2 Dexamethasone: oral versus intramuscular, Outcome 5 Neonatal sepsis.

Analysis 2.6.

Comparison 2 Dexamethasone: oral versus intramuscular, Outcome 6 Necrotising enterocolitis.

Comparison 3. Betamethasone acetate + phosphate versus betamethasone phosphate
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Neonatal death169Risk Ratio (M-H, Fixed, 95% CI)0.32 [0.01, 7.69]
1.1 24 mg beta a+p (2 x 12 mg, 24 hourly) v 24 mg beta p (4 x 6 mg, 12 hourly)169Risk Ratio (M-H, Fixed, 95% CI)0.32 [0.01, 7.69]
2 Respiratory distress syndrome169Risk Ratio (M-H, Fixed, 95% CI)0.19 [0.01, 3.91]
2.1 24 mg beta a+p (2 x 12 mg, 24 hourly) v 24 mg beta p (4 x 6 mg, 12 hourly)169Risk Ratio (M-H, Fixed, 95% CI)0.19 [0.01, 3.91]
3 Intraventricular haemorrhage169Risk Ratio (M-H, Fixed, 95% CI)0.32 [0.01, 7.69]
3.1 24 mg beta a+p (2 x 12 mg, 24 hourly) v 24 mg beta p (4 x 6 mg, 12 hourly)169Risk Ratio (M-H, Fixed, 95% CI)0.32 [0.01, 7.69]
4 Neurodevelopmental disability169Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
4.1 24 mg beta a+p (2 x 12 mg, 24 hourly) v 24 mg beta p (4 x 6 mg, 12 hourly)169Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
5 Birthweight (kg)169Mean Difference (IV, Fixed, 95% CI)-0.10 [-0.44, 0.24]
5.1 beta a+p (2 x 12 mg, 24hrly) and beta p (4 x 6 mg,12hrly)169Mean Difference (IV, Fixed, 95% CI)-0.10 [-0.44, 0.24]
6 Low birthweight169Risk Ratio (M-H, Fixed, 95% CI)1.21 [0.86, 1.72]
6.1 24 mg beta a+p (2 x 12 mg, 24 hourly) v 24 mg beta p (4 x 6 mg, 12 hourly)169Risk Ratio (M-H, Fixed, 95% CI)1.21 [0.86, 1.72]
7 Neonatal intensive care unit admission169Risk Ratio (M-H, Fixed, 95% CI)0.11 [0.01, 1.93]
7.1 24 mg beta a+p (2 x 12 mg, 24 hourly) x 24 mg beta p (4 x 6 mg, 12 hourly)169Risk Ratio (M-H, Fixed, 95% CI)0.11 [0.01, 1.93]
8 Bronchopulmonary dysplasia169Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
8.1 24 mg beta a+p (2 x 12 mg, 24 hourly) v 24 mg beta p (4 x 6 mg, 12 hourly)169Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
9 Periventricular leukomalacia169Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
9.1 24 mg beta a+p (2 x 12 mg, 24 hourly) v 24 mg beta p (4 x 6 mg, 12 hourly)169Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
Analysis 3.1.

Comparison 3 Betamethasone acetate + phosphate versus betamethasone phosphate, Outcome 1 Neonatal death.

Analysis 3.2.

Comparison 3 Betamethasone acetate + phosphate versus betamethasone phosphate, Outcome 2 Respiratory distress syndrome.

Analysis 3.3.

Comparison 3 Betamethasone acetate + phosphate versus betamethasone phosphate, Outcome 3 Intraventricular haemorrhage.

Analysis 3.4.

Comparison 3 Betamethasone acetate + phosphate versus betamethasone phosphate, Outcome 4 Neurodevelopmental disability.

Analysis 3.5.

Comparison 3 Betamethasone acetate + phosphate versus betamethasone phosphate, Outcome 5 Birthweight (kg).

Analysis 3.6.

Comparison 3 Betamethasone acetate + phosphate versus betamethasone phosphate, Outcome 6 Low birthweight.

Analysis 3.7.

Comparison 3 Betamethasone acetate + phosphate versus betamethasone phosphate, Outcome 7 Neonatal intensive care unit admission.

Analysis 3.8.

Comparison 3 Betamethasone acetate + phosphate versus betamethasone phosphate, Outcome 8 Bronchopulmonary dysplasia.

Analysis 3.9.

Comparison 3 Betamethasone acetate + phosphate versus betamethasone phosphate, Outcome 9 Periventricular leukomalacia.

Comparison 4. Betamethasone 12 hour versus 24 hour dosing
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Perinatal death1 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
1.1 Betamethasone: 12 mg 12 hourly v 12 mg 24 hourly1255Risk Ratio (M-H, Fixed, 95% CI)0.93 [0.46, 1.87]
1.2 23+1 to 26+0 weeks at trial entry156Risk Ratio (M-H, Fixed, 95% CI)0.80 [0.43, 1.50]
1.3 26+1 to 29+0 weeks at trial entry143Risk Ratio (M-H, Fixed, 95% CI)1.93 [0.24, 15.71]
1.4 29+1 to 32+0 weeks at trial entry181Risk Ratio (M-H, Fixed, 95% CI)1.45 [0.06, 34.35]
1.5 32+1 to 34+0 weeks at trial entry175Risk Ratio (M-H, Fixed, 95% CI)1.75 [0.09, 35.00]
2 Respiratory distress syndrome1 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
2.1 Betamethasone: 12 mg 12 hourly v 12 mg 24 hourly1242Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.69, 1.40]
2.2 23+1 to 26+0 weeks at trial entry149Risk Ratio (M-H, Fixed, 95% CI)0.83 [0.69, 0.99]
2.3 26+1 to 29+0 weeks at trial entry140Risk Ratio (M-H, Fixed, 95% CI)2.73 [0.97, 7.67]
2.4 29+1 to 32+0 weeks at trial entry181Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.43, 2.06]
2.5 32+1 to 34+0 weeks at trial entry172Risk Ratio (M-H, Fixed, 95% CI)1.43 [0.17, 12.04]
3 Intraventricular hemorrhage1 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
3.1 Betamethasone: 12 mg 12 hourly v 12 mg 24 hourly1135Risk Ratio (M-H, Fixed, 95% CI)1.40 [0.76, 2.56]
3.2 23+1 to 26+0 weeks at trial entry138Risk Ratio (M-H, Fixed, 95% CI)1.57 [0.76, 3.24]
3.3 26+1 to 32+0 weeks at trial entry131Risk Ratio (M-H, Fixed, 95% CI)3.27 [0.48, 22.52]
3.4 29+1 to 32+0 weeks at trial entry145Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.29, 3.45]
3.5 32+1 to 34+0 weeks at trial entry121Risk Ratio (M-H, Fixed, 95% CI)0.31 [0.02, 4.14]
4 Maternal fever > 100.4 F1213Risk Ratio (M-H, Fixed, 95% CI)0.71 [0.25, 2.02]
4.1 Betamethasone: 12 mg 12 hourly v 12 mg 24 hourly1213Risk Ratio (M-H, Fixed, 95% CI)0.71 [0.25, 2.02]
5 Birthweight (g)1255Mean Difference (IV, Fixed, 95% CI)84.0 [-144.63, 312.63]
5.1 Betamethasone: 12 mg 12 hourly v 12 mg 24 hourly1255Mean Difference (IV, Fixed, 95% CI)84.0 [-144.63, 312.63]
6 Small-for-gestational age1255Risk Ratio (M-H, Fixed, 95% CI)0.61 [0.36, 1.05]
6.1 Betamethasone: 12 mg 12 hourly v 12 mg 24 hourly1255Risk Ratio (M-H, Fixed, 95% CI)0.61 [0.36, 1.05]
7 Neonatal intensive care unit admission1247Risk Ratio (M-H, Fixed, 95% CI)0.89 [0.79, 1.00]
7.1 Betamethasone: 12 mg 12 hourly v 12 mg 24 hourly1247Risk Ratio (M-H, Fixed, 95% CI)0.89 [0.79, 1.00]
8 Chronic lung disease1230Risk Ratio (M-H, Fixed, 95% CI)0.79 [0.49, 1.26]
8.1 Betamethasone: 12 mg 12 hourly v 12 mg 24 hourly1230Risk Ratio (M-H, Fixed, 95% CI)0.79 [0.49, 1.26]
9 Neonatal sepsis1236Risk Ratio (M-H, Fixed, 95% CI)1.15 [0.47, 2.81]
9.1 Betamethasone: 12 mg 12 hourly v 12 mg 24 hourly1236Risk Ratio (M-H, Fixed, 95% CI)1.15 [0.47, 2.81]
10 Neonatal antibiotic use (> 5 days)1236Risk Ratio (M-H, Fixed, 95% CI)0.94 [0.61, 1.46]
10.1 Betamethasone: 12 mg 12 hourly v 12 mg 24 hourly1236Risk Ratio (M-H, Fixed, 95% CI)0.94 [0.61, 1.46]
11 Necrotising enterocolitis1231Risk Ratio (M-H, Fixed, 95% CI)9.20 [0.55, 154.92]
11.1 Betamethasone: 12 mg 12 hourly v 12 mg 24 hourly1231Risk Ratio (M-H, Fixed, 95% CI)9.20 [0.55, 154.92]
12 Retinopathy of prematurity1109Risk Ratio (M-H, Fixed, 95% CI)0.94 [0.53, 1.66]
12.1 Betamethasone: 12 mg 12 hourly v 12 mg 24 hourly1109Risk Ratio (M-H, Fixed, 95% CI)0.94 [0.53, 1.66]
13 Postpartum maternal length of stay (days)1215Mean Difference (IV, Fixed, 95% CI)-0.73 [-1.28, -0.18]
13.1 Betamethasone: 12 mg 12 hourly v 12 mg 24 hourly1215Mean Difference (IV, Fixed, 95% CI)-0.73 [-1.28, -0.18]
Analysis 4.1.

Comparison 4 Betamethasone 12 hour versus 24 hour dosing, Outcome 1 Perinatal death.

Analysis 4.2.

Comparison 4 Betamethasone 12 hour versus 24 hour dosing, Outcome 2 Respiratory distress syndrome.

Analysis 4.3.

Comparison 4 Betamethasone 12 hour versus 24 hour dosing, Outcome 3 Intraventricular hemorrhage.

Analysis 4.4.

Comparison 4 Betamethasone 12 hour versus 24 hour dosing, Outcome 4 Maternal fever > 100.4 F.

Analysis 4.5.

Comparison 4 Betamethasone 12 hour versus 24 hour dosing, Outcome 5 Birthweight (g).

Analysis 4.6.

Comparison 4 Betamethasone 12 hour versus 24 hour dosing, Outcome 6 Small-for-gestational age.

Analysis 4.7.

Comparison 4 Betamethasone 12 hour versus 24 hour dosing, Outcome 7 Neonatal intensive care unit admission.

Analysis 4.8.

Comparison 4 Betamethasone 12 hour versus 24 hour dosing, Outcome 8 Chronic lung disease.

Analysis 4.9.

Comparison 4 Betamethasone 12 hour versus 24 hour dosing, Outcome 9 Neonatal sepsis.

Analysis 4.10.

Comparison 4 Betamethasone 12 hour versus 24 hour dosing, Outcome 10 Neonatal antibiotic use (> 5 days).

Analysis 4.11.

Comparison 4 Betamethasone 12 hour versus 24 hour dosing, Outcome 11 Necrotising enterocolitis.

Analysis 4.12.

Comparison 4 Betamethasone 12 hour versus 24 hour dosing, Outcome 12 Retinopathy of prematurity.

Analysis 4.13.

Comparison 4 Betamethasone 12 hour versus 24 hour dosing, Outcome 13 Postpartum maternal length of stay (days).

What's new

DateEventDescription
1 May 2013New citation required but conclusions have not changedReview updated - the overall conclusions have not changed. However, in this update, the increase in NICU admission for the infants in the dexamethasone group (compared with the betamethasone group), is no longer statistically significant, with the inclusion of the Danesh 2012 trial.
13 February 2013New search has been performedSearch updated. Two new trials included (Danesh 2012; Khandelwal 2012); and one new report excluded (Shanks 2010).

History

Protocol first published: Issue 4, 2007
Review first published: Issue 4, 2008

DateEventDescription
9 May 2008AmendedConverted to new review format.

Contributions of authors

For this update of the review, Daniela Gagliardi, Fiona Brownfoot and Emily Bain extracted data and assessed the risk of bias for the two new trials, and all authors contributed to the final draft of the review. Daniela Gagliardi and Emily Bain updated the indirect comparison estimates, and performed the subgroup interaction tests.

Fiona Brownfoot researched and wrote the initial draft of the previous version of the review, and extracted data together with Philippa Middleton. Caroline Crowther and Philippa Middleton edited the subsequent drafts.

Declarations of interest

Two of the review authors (Caroline Crowther and Philippa Middleton) are investigators on the A*STEROID trial, and Daniela Gagliardi is a Research Officer with the A*STEROID trial, which may be considered for inclusion in this review on completion and publication (see Ongoing studies).

Sources of support

Internal sources

  • ARCH, The Robinson Institute, Discipline of Obstetrics and Gynaecology, The University of Adelaide, Australia.

External sources

  • Department of Health and Ageing, Australia.

  • National Health and Medical Research Council, Australia.

Differences between protocol and review

In this update of the review: we have clarified that cross-over trials and cluster-randomised trials will be excluded, and that studies reported as abstracts only will be included.

We have reduced the number of primary outcomes for the baby to three (death, RDS and IVH) and have re-classified the remaining outcomes as secondary outcomes.

In previous versions of the review: we added the outcome of any neurodisability for the child, and for the child as an adult.

We have included a previously excluded subgroup (see below) in the review methods of this update as we believe that this maternal characteristic may affect health outcomes. However we were unable to perform this subgroup analysis due to insufficient data available in this update:

  • preterm prelabour rupture of membranes (at trial entry: yes versus no).

In the previous update of this review, we deleted the following subgroup from the review methods, as it may be influenced by the intervention:

  • gestational age at birth (24 to 26 weeks, 27 to 29 weeks, 30 to 34 weeks, 35 to 37 weeks).

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Chen 2005

MethodsRandomised controlled trial.
Participants

168 women "received antenatal corticosteroids" so we have assumed 168 was the number randomised.

After exclusions 140 women, who gave birth to 157 infants (17 sets of twins), were included.

Setting: Mackay Memorial Hospital, Taiwan (from December 2001 to September 2003).

Inclusion criteria: preterm prelabour rupture of membranes between 24 and 32 weeks and preterm labour between 24 and 34 weeks.

Interventions

Dexamethasone group (n = 76 infants)

24 mg: 4 x 6 mg doses IM 12 hours apart.
Betamethasone group (n = 81 infants)

24 mg: 2 x 12 mg doses IM 24 hours apart.

Outcomes

Infant:
RDS; IVH (grade 3 to 4); PVL; birthweight; Apgar score < 7 at 5 mins; head circumference; neonatal sepsis; NEC; ROP.

Mother:

Caesarean birth.

NotesFunding source was not detailed.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskQuote: "These women were randomly divided into two groups,"
Allocation concealment (selection bias)Unclear riskAs above, no further detail provided.
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskUnclear, not reported.
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskUnclear, not reported.
Incomplete outcome data (attrition bias)
All outcomes
High riskData were excluded for 28/168 (16%) women (1 intrauterine fetal death, 15 women gave birth after 37 weeks' gestation, 4 neonates died immediately after delivery, 8 women gave birth at another hospital); it was not reported in which groups (dexamethasone or betamethasone) the losses occurred.
Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement (while a number of important outcomes were reported, no protocol was available, and thus it was difficult to assess selective reporting).
Other biasUnclear riskWhile groups were comparable at baseline, women in the betamethasone group were approximately 3 years older compared to women in the dexamethasone group.

Danesh 2012

MethodsRandomised controlled trial.
Participants

240 women randomised.

Setting: Isfahan University of Medical Sciences Obstetrics Department, Iran (from February to November 2011).

Inclusion criteria: pregnant women of low parity; 16 to 45 years of age; between 24 and 34 weeks gestation; hospitalised because of high risk of preterm birth that justified preventative corticosteroid therapy; with or without intact membranes; low Bishop score ≤ 5; non-smokers; singleton pregnancy; residence in Isfahan; hospitalisation planned to last at least 3 days.

Preterm rupture of membranes was diagnosed in the presence of a gush of fluid from the vagina, followed by persistent, uncontrolled leakage, or pooling of fluid on speculum examination, with positive nitrazine and Fern testing. Preterm labour was diagnosed in the presence of uterine contractions of 4 in 20 minutes or 8 in 60 minutes, plus progressive change in the cervix, cervical dilatation greater than 1 cm and cervical effacement 80% or greater.

Exclusion criteria: women were excluded if they had evidence of fetal distress, substantial abnormalities in neurological, psychiatric, cardiac, endocrinological, haematologic, hepatic, renal or metabolic function; had signs of infection; positive urine culture; vaginal bleeding due to placental praevia or abruption.

Interventions

Betamethasone group (n = 120)

2 IM injections of 12 mg betamethasone sodium at 24-hour intervals.

Dexamethasone group (n = 120) 

4 IM injections of 6 mg dexamethasone phosphate at 12-hour intervals.

Outcomes

Infant:

Fetal plasma glucose; Apgar scores (1 minute and 5 minute); NICU stay.

Mother:

Maternal WBC and differential count; ESR; maternal fasting plasma glucose; length of admission to birth; preterm birth; gestational age at birth.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskQuote: “women were randomised with a list of computer-generated numbers".
Allocation concealment (selection bias)Low riskGroup assignments were concealed in an opaque sealed envelope until just before entry into the study. It was not stated whether the envelopes were consecutively numbered.
Blinding of participants and personnel (performance bias)
All outcomes
High riskQuote: "due to the different injection frequency and treatment period of dexamethasone and betamethasone, it was not possible to ensure patients were blinded".
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskUnclear, not reported.
Incomplete outcome data (attrition bias)
All outcomes
Low riskQuote: "all 240 women who completed treatment were available for follow-up at delivery". It was assumed that there was no attrition nor were there any exclusions.
Selective reporting (reporting bias)Unclear riskWith no access to a trial protocol, the review authors cannot confidently assess selective reporting. In regards to NICU admission, it is unclear as to whether the values in the table represents all admissions, or only those due to RDS (the text suggests it represents only those due to RDS). In regards to adverse events, data is not reported, quote "Both, dexamethasone and betamethasone treatment was tolerated well and most of the adverse events reported were mild in severity".
Other biasLow riskThe study appears to be free of other sources of bias. Groups were comparable at baseline.

Egerman 1998

MethodsRandomised controlled trial.
Participants

170 women randomised.

Setting: Tennessee, USA (from July 1996 to July 2007).

Inclusion criteria: preterm birth between 24 to 33 weeks' gestation, preterm labour, preterm rupture of membranes, medical indication for delivery.

Exclusion criteria: received corticosteroids during the pregnancy (except immediately before transfer), anticonvulsant therapy, rifampin, infection other than cystitis or cervicitis, advanced cervical dilatation, fetal pulmonary maturity.

Interventions

Oral dexamethasone group (n = 92 women; 99 infants)

32 mg oral dexamethasone: 4 x 8 mg, 12 hourly.

IV dexamethasone group (n = 78 women; 84 infants)
24 mg IM dexamethasone: 4 x 6 mg IM, 12 hourly.

Repeated weekly until 34 weeks' gestation if birth had not yet occurred.

Outcomes

Infant:
Death; RDS; IVH; birthweight; sepsis; NEC; gestational age at birth.

Mother:
Gestational age at entry (weeks); dilatation (cm); latency; caesarean birth; antibiotic use.

NotesThe study was discontinued at 39% enrolment (170 women) after a blinded review of available outcomes.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskParticipants were randomly assigned by computer-generated numbers (placed in sealed envelopes).
Allocation concealment (selection bias)Unclear riskSealed envelopes were used; no further details provided.
Blinding of participants and personnel (performance bias)
All outcomes
High riskDue to the nature of the intervention, it is considered unlikely that participants and personnel were blinded.
Blinding of outcome assessment (detection bias)
All outcomes
Low riskAn investigator not involved in recruitment or patient care performed the analysis.
Incomplete outcome data (attrition bias)
All outcomes
Low risk5/170 (3%) women were unavailable for follow-up.
Selective reporting (reporting bias)Unclear riskWhile the pre-specified primary and secondary outcomes from the methods were reported, no protocol was available to assess selective reporting, and furthermore, clinical outcomes were reported incompletely, for example quote "No differences in Apgar scores at 1 and at 5 minutes were noted between the oral and intramuscular groups".
Other biasLow riskNo other obvious sources of bias identified. Groups were comparable at baseline.

Elimian 2007

MethodsRandomised controlled trial.
Participants

299 women randomised.
Setting: Stony Brook University Hospital, New York, USA (from August 2002 to July 2004).

Inclusion criteria: risk of preterm birth between 24 to 33 weeks' gestation.
Exclusion criteria: clinical chorioamnionitis, major fetal structural abnormalities, major fetal chromosomal abnormalities, prior antenatal corticosteroid exposure, use of betamethasone or dexamethasone for other medical indications and quadruplets.

Interventions

Dexamethasone group (n = 149 women; 178 infants)

24 mg dexamethasone: 4 x 6 mg doses of dexamethasone IM 12 hours apart.

Betamethasone group (n = 150 women; 181 infants).
24 mg betamethasone: 2 x 12 mg doses of betamethasone IM 24 hours apart (placebo 2 doses given at 12-hour intervals).

Outcomes

Infant:
Death; IVH (diagnosed by ultrasound, diagnosed by autopsy); RDS; PVL; birthweight; BPD; NEC; neonatal sepsis; surfactant use; ROP; neonatal blood pressure; need for inotropic support; mean duration of inotropic support (days); PDA; need for a vasopressor.

Women:
Chorioamnionitis; fever after trial entry requiring the use of antibiotics; intrapartum fever requiring the use of antibiotics.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated random numbers were used.
Allocation concealment (selection bias)Low riskAllocation was conducted by pharmacy.
Blinding of participants and personnel (performance bias)
All outcomes
Low riskParticipants and health care providers were reported as being blind. Syringes were covered with "opaque material". A placebo was used to ensure that the regimens for administration did not differ.
Blinding of outcome assessment (detection bias)
All outcomes
Low riskOutcome assessors were blind to group assignment.
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskNo loss to follow-up; however only 105/178 infants in the dexamethasone group and 100/181 in the betamethasone group were assessed for IVH and PVL.
Selective reporting (reporting bias)Low riskData for all pre-specified outcomes (from the methods) and many of the expected outcomes were reported.
Other biasLow riskGroups were comparable at baseline. The study appears to be free of other sources of bias.

Khandelwal 2012

MethodsRandomised controlled trial.
Participants

228 mothers, 260 fetuses were randomised.

Setting: Cooper University Hospital, Camden, New Jersey, USA (from July 2006 to May 2009).

Inclusion criteria: steroids administered for any indication between 23 to 34 weeks. gestation

Exclusion criteria: < 23 or > 34 weeks' gestation; elapsed time > 12 hours since administration of the first dose of betamethasone; known drug allergy to betamethasone; given steroids other than betamethasone for lung maturation; any contraindication to steroid therapy were excluded.

Interventions

12 hourly group (n = 161 women; 180 fetuses)

24 mg betamethasone: 2 x 12 mg doses, 12 hourly.

24 hourly group (n = 67 women; 80 fetuses)

24 mg betamethasone: 2 x 12 mg doses; 24 hourly.

No rescue steroid doses were offered in this trial.

Outcomes

For the fetus/neonate:

Death (fetal/neonatal); RDS; neonatal sepsis; antibiotic use; admission to NICU; mean birthweight; mode of delivery LUSCS; SGA; NEC; ROP; IVH; CLD.

For the mother: 

Maternal fever; maternal postpartum length of stay.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk

Participants were allocated into four groups stratified by 3 gestational age categories:

Group A: 23.1 to 26.0 weeks;

Group B: 26.1 to 29.0 weeks;

Group C: 29.1 to 32.0 weeks;

Group D: 32.1 to 34.0 weeks.

The participants in each group were randomly assigned to either the 12-hour arm or 24-hour arm. Randomisation sequence was generated in Excel using its 'random function' in 4 groups of 50 each. Randomisation was performed in 2:1 ratio (for 12:24-hour dosing interval).

Allocation concealment (selection bias)Low riskStudy assignments placed in sealed opaque envelopes with a study number and group number on each envelope.
Blinding of participants and personnel (performance bias)
All outcomes
High riskParticipants and personnel were not blinded.
Blinding of outcome assessment (detection bias)
All outcomes
Low riskOutcome assessors were blinded; neonatologists were informed if women received steroids and whether 1 or 2 doses, but not informed of the dosing interval.
Incomplete outcome data (attrition bias)
All outcomes
Low risk7 infants 3% (7/260) were lost to follow-up: incomplete information was available for 4 neonates in the 12-hour dosing group and 3 in the 24-hour dosing group (7/260: 3%).
Selective reporting (reporting bias)Low riskOutcome data for all pre-specified outcomes (from the methods and trial registration) were reported.
Other biasUnclear risk

There were some baseline imbalances between groups: approximately 70% (161/228 mothers; 180/260 fetuses) were assigned to the 12-hour arm and only 30% (67/228 mothers; 80/260 fetuses) were assigned to the 24-hour arm.

Women in the 24-hour arm were about 2 years older than in the 12-hour arm; In the 24-hour cohort, there were more African American women at risk for delivery 26 weeks and more Caucasians at risk in the 32 to 34 week group.

Magee 1997

MethodsRandomised controlled trial.
Participants

59 women were randomised.

Setting: John Radcliffe Hospital, Oxford, UK (from April 1995 to March 1996).

Inclusion criteria: women with singleton pregnancies at risk of birth between 26 to 34 weeks' gestation inclusive, who had not received steroids in the preceding week.

Interventions

Dexamethasone group (n = 29)

24 mg dexamethasone: 2 x 12 mg IM; 12 hourly.

Betamethasone group (n = 29)
24 mg betamethasone: 2 x 12 mg IM; 12 hourly.

Outcomes Infant:
Biophysical parameters (day 0, 1, 2); FHR; LTV/STV; number of movements/hour; number of accelerations; number of decelerations; Apgar score at 5 mins; caesarean birth.
NotesFunding source not mentioned.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandom number table was used to generate the random sequence.
Allocation concealment (selection bias)Low riskConsecutively numbered, sealed, opaque envelopes were used. The IM vials were labelled by the hospital pharmacy with A or B.
Blinding of participants and personnel (performance bias)
All outcomes
Low riskParticipant and clinicians were blinded.
Blinding of outcome assessment (detection bias)
All outcomes
Low riskData were analysed in "a blinded fashion".
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskLosses to follow-up: > 20% in total: 1/30 post-randomisation exclusion in the betamethasone group (a woman with a twin pregnancy was enrolled in error); then 9/29 losses from the betamethasone group at day 2 (1 transfer, 3 early births, 5 early discharges); 7/29 losses from the dexamethasone group (1 self-discharge, 3 early births, 3 early discharges).
Selective reporting (reporting bias)Unclear riskThe trial focused on the effects of steroids on FHR, and very few clinical outcomes were reported. With no access to a trial protocol it is difficult to assess selective reporting.
Other biasLow riskThe study appears to be free of other sources of bias. Groups were comparable at baseline.

Mulder 1997

MethodsRandomised controlled trial.
Participants

60 women were randomised.

Setting: University Hospital, Utrecht, The Netherlands.

Inclusion criteria: women with premature contractions or at risk of preterm labour, between 26 to 33 weeks' gestation, SGA (estimated fetal size < 5th centile), premature contractions, placenta praevia or other cause of vaginal blood loss, preterm rupture of membranes without evidence of intrauterine infection, pre-eclampsia, essential hypertension, poor obstetrical history, or leiomyoma.

Exclusion criteria: cervical dilatation > 5 cm, signs of intrauterine infection, ritodrine hydrochloride treatment for < 4 days at the start of the study.

Interventions

Dexamethasone group (n = 24)
24 mg dexamethasone: 2 x 12 mg IM, 12 hourly.

Betamethasone group (n = 26)
24 mg betamethasone: 2 x 12 mg IM, 24 hourly.

Outcomes Infant:
Birthweight; biophysical parameters - FHR, LTV/STV, breathing movement, breathing bout length, number of breaths, breath-to-breath interval, body movement incidence, body movement number of bursts, body movement burst length; Apgar score less than 7 at 5 mins; mode of birth.
Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskQuote: "eligible women were randomised (sealed envelope method)". No further detail provided.
Allocation concealment (selection bias)Unclear riskAs above.
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskUnclear, not reported; however considered unlikely in view of the intervention (12-hourly dexamethasone versus 24-hourly betamethasone).
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskUnclear, not reported.
Incomplete outcome data (attrition bias)
All outcomes
Unclear risk17%; 6 of 30 from the dexamethasone group and 4 of 30 from the betamethasone group were lost to follow-up.
Selective reporting (reporting bias)Unclear riskThe trial focused on the effects of steroids on FHR and behaviour and clinical outcomes were not reported. With no access to a trial protocol it is difficult to assess selective reporting.
Other biasLow riskThe study appears to be free of other sources of bias. Groups were comparable at baseline.

Mushkat 2001

MethodsQuasi-randomised controlled trial.
Participants

33 women were randomised.

Setting: Israel.

Inclusion criteria: women with preterm labour between 26 to 33 weeks' gestation.
Exclusion criteria: chronic or acute hypertension, gestational diabetes, vaginal bleeding due to placenta praevia or abruption placenta, IUGR, fetal distress.

Interventions

Dexamethasone group (n = 16)
24 mg dexamethasone: 2 x 12 mg IM, 12 hourly.

Betamethasone group (n = 17)
24 mg betamethasone: 12 mg betamethasone sodium and 12 mg betamethasone acetate IM, divided into 2 doses 12 hours apart.

Outcomes Infant:
Biophysical parameters (0, 6, 12, 18, 36 hours): maternal perception of fetal movements, body movement, breathing movements; gestational age.
NotesFunding source not documented.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskThe participants were randomly assigned to 2 groups according to computer-generated randomisation schedule.
Allocation concealment (selection bias)High riskQuasi-randomised design "each consecutive candidate got an even or an uneven number drawn out of performed random computer-generated list. Even numbers were assigned to betamethasone treatment, while uneven numbers were assigned to dexamethasone treatment".
Blinding of participants and personnel (performance bias)
All outcomes
Low riskParticipants and clinicians were blinded, with treatment regimens being similar, and both drugs being "identical in appearance".
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskWhile the study was described as "double blind" it was not made clear as to whether outcome assessment was performed blind.
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskLosses to follow-up were unclear.
Selective reporting (reporting bias)Unclear riskWhile the trial focused on fetal biophysical parameters, it also reported a number of clinical outcomes incompletely, example: "gestational age, birthweight and Agpar score at 5 min did not differ between the 2 groups".
Other biasLow riskGroups were comparable at baseline for age, parity and gestational age. The study appears to be free of other sources of bias.

Rotmensch 1999

MethodsRandomised controlled trial.
Participants

46 women were randomised.

Setting: Israel and Italy (from June 1995 to March 1997).

Inclusion criteria: women with preterm birth at 27 to 34 weeks' gestation, preterm premature rupture of membranes with no clinical evidence of infection, pregnancy-induced hypertension syndromes, IUGR, third trimester bleeding due to placenta praevia.

Interventions

Dexamethasone group (n = 24)
24 mg dexamethasone: 2 x 12 mg IM, 24 hourly.

Betamethasone group (n = 22)
24 mg betamethasone: 2 x 12 mg IM, 24 hourly.

Outcomes Infant:
Birthweight; biophysical parameters (0, 2 ,4 days): FHR, acceleration > 10 bpm, deceleration > 10 bpm, LTV/STV, breathing time (sec in 30 mins), movement in 30 mins.
NotesFunding source not documented.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandomised by "computer-generated randomisation tables".
Allocation concealment (selection bias)Unclear riskUnclear, details were not reported.
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskUnclear if participants were blinded. "Examiners were blinded to the administered drugs."
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskAs above; unclear if outcome assessors were blinded.
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskAttrition or exclusions were not documented.
Selective reporting (reporting bias)Unclear riskThe trial focused on the effects of steroids on FHR and biophysical activities and clinical outcomes were not reported. With no access to a trial protocol it is difficult to assess selective reporting.
Other biasLow riskGroups were comparable at baseline.The study appears to be free of other sources of bias.

Senat 1998

MethodsRandomised controlled trial.
Participants

82 women were randomised.

Setting: Clamart, France (between October 1994 to October 1995).

Inclusion criteria: women with preterm labour < 34 weeks' gestation.

Exclusion criteria: uncertain pregnancy history, clinical infection in women, vaginal bleeding, suspicion of premature rupture of membranes.

Interventions

Dexamethasone group (n = 40 women, 44 babies - 39 babies analysed)
16 mg dexamethasone: 4 x 4 mg IM, 12 hourly.

Betamethasone group (n = 42 women, 53 babies - 42 babies analysed).
24 mg betamethasone: 4 x 3 mg betamethasone sodium and 3 mg betamethasone acetate IM, 12 hourly.

Outcomes Infant:
Death; RDS; IVH; PVL; birthweight; NEC; biophysical parameters (0, 24-48 hours, 4-7 days): FHR, LTV/STV, high/low variation, acceleration > 10 bpm; deceleration > 10 bpm; uterine contractions; gestational age; CTG.
NotesIn the case of multiple pregnancy, one fetus was randomly selected for analysis.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskA table of random numbers was used.
Allocation concealment (selection bias)Unclear riskThe table of random numbers was "held by an independent investigator. " Women were allocated to either one of two different corticosteroid regiments in a "non blinded fashion."
Blinding of participants and personnel (performance bias)
All outcomes
High riskParticipants and clinicians were not blinded.
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskUnclear if outcome assessors were blinded.
Incomplete outcome data (attrition bias)
All outcomes
Low risk1 neonate from the dexamethasone group died; this case was excluded from the study and statistics were therefore derived from the analysis of 39 pregnancies.
Selective reporting (reporting bias)Unclear riskThe trial focused on the effects of steroids on FHR variability in preterm labour; while some clinical outcomes were reported in addition, with no access to a trial protocol it is difficult to assess selective reporting.
Other biasUnclear riskIn the case of multiple pregnancy, one fetus was randomly selected for analysis.

Subtil 2003

MethodsRandomised controlled trial.
Participants

105 women were randomised.

Setting: Lille, France (from November 1997 to February 1999).

Inclusion criteria: women at high risk of preterm birth, recruited at 27 to 35 weeks' gestation, with singleton pregnancies.

Exclusion criteria: imminent birth, multiple pregnancy, previously participated in the protocol, received corticosteroid therapy < 10 days prior.

Interventions

Betamethasone acetate and phosphate group (n = 35)
24 mg betamethasone acetate and phosphate: 2 x 12 mg IM, 24 hourly.

Betamethasone phosphate group (n = 34)
24 mg betamethasone phosphate: 4 x 6 mg IM, 12 hourly.

Dexamethasone phosphate group (n = 36)

24mg dexamethasone phosphate: 4 x 6 mg IM, 12 hourly.

Outcomes

Infant:
Death; RDS; BPD; IVH (grade 1 and 2); severe IVH (grade 3 and 4); hyperechoic > 10 days; PVL; birthweight; NICU admission; GA at delivery; biophysical parameters: duration of tracing; STV/LTV; FHR; acceleration/deceleration number per hr; movement number per hr.
For the child (18 months):
Neurodevelopmental disability at follow-up.

Other:

Tests of the specific drug effect, time effect, and interaction by analysis of variance.

NotesFunding source not mentioned.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskA random number table was used.
Allocation concealment (selection bias)Unclear riskWomen were allocated randomly to treatment groups through a "system of envelopes" that had been numbered from 1 to 105. The randomisation was performed with a random number table, by random distribution of the 3 formulations.
Blinding of participants and personnel (performance bias)
All outcomes
High riskThe midwives and participants were not blinded. The manuscript details that physicians were blinded, however it is unclear as to whether this would have been successfully achieved, quote "physicians were blinded to the product; both midwives and patients were instructed that no physician was to be informed of the rhythm of injections or of the product that was being administered".
Blinding of outcome assessment (detection bias)
All outcomes
High riskOutcome assessment was not blinded.
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskHigh numbers lost to follow-up for FHR (due to discharge and birth): none at day 0; 1 of 105 (1%) at day 1; 16 (15%) at day 2, 44 (42%) at day 3; and 57 (54%) at day 4 (due to discharge and birth). The percentage of missing recordings for the relevant time period averaged 8.6% and remained stable from day 0 through to day 4. The percentage of women who went home or gave birth before day 4 (and were lost to follow-up) did not differ significantly between groups.
Selective reporting (reporting bias)Low riskWhile this trial focused on the effects of steroids on FHR, the manuscript also reports the pre-specified and expected clinical outcomes.
Other biasLow riskGroups were comparable at baseline. The study appears to be free of other sources of bias.

Urban 2005

  1. a

    BPD: biparietal diameter
    bpm: beats per minute
    CTG: cardiotocography
    ESR: erythrocyte sedimentation rate
    FHR: fetal heart rate
    GA: gestational age
    hr: hour
    IM: intramuscular
    IUGR: intrauterine growth restriction
    IVH: intraventricular haemorrhage
    LUSCS: lower uterine segment caesarean section
    LTV: long-term variation
    MCA PI: middle cerebral artery pulsatility index
    mins: minutes
    NEC: necrotising enterocolitis
    NICU: neonatal intensive care unit
    PDA: patent ductus arteriosus
    PVL: periventricular leukomalacia
    RDS: respiratory distress syndrome
    ROP: retinopathy of prematurity
    secs: seconds
    SGA: small-for-gestational age
    STV: short-term variation
    UA PI: umbilical artery pulsatility index
    WBC: white blood cell

MethodsRandomised controlled trial.
Participants

67 women were randomised.

Setting: Medical Academy of Bialystok, Poland.

Inclusion criteria: preterm contractions of the uterus, preterm premature rupture of membranes, cervical length less than 20 mm, placenta praevia before 34 weeks, singleton pregnancy.

Exclusion criteria: fetal major structural malformations or abnormal karyotype.

Interventions

Dexamethasone group (n = 34)
24 mg dexamethasone: 4 x 6 mg IM, 12 hourly.

Betamethasone group (n = 33)
24 mg betamethasone: 2 x 12 mg IM, 24 hourly.

Outcomes Infant:
Birthweight; UA PI; MCA PI; abnormal FHR patterns (at 0, 24 and 72 hours); Apgar score at 1 and 5 minutes; umbilical cord artery pH; base deficit.
NotesFunding sources not mentioned.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskA computer-generated randomisation table was used.
Allocation concealment (selection bias)Low risk"Consecutively numbered and sealed opaque envelopes" were used.
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskBlinding not stated; considered unlikely for participants and personnel in view of the different timings of administration of the dexamethasone versus betamethasone.
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskBlinding of outcome assessors was not detailed.
Incomplete outcome data (attrition bias)
All outcomes
Low riskIt was assumed that there were no exclusions or attrition.
Selective reporting (reporting bias)Unclear riskThe trial focused on the effects of steroids on fetal Doppler flow velocity waveforms; while some important clinical outcomes were reported in addition, with no access to a trial protocol it is difficult to assess selective reporting.
Other biasLow riskThe study appears to be free of other sources of bias. The groups were comparable at baseline for maternal age and gestational age.

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    IM: intramuscular
    IVH: intraventricular haemorrhage
    RDS: respiratory distress syndrome

Egerman 1997Assessed bioavailability of dexamethasone post oral or IM administration; parallel cross-over trial.
Kurz 1993L-carnitine was added to betamethasone and compared against betamethasone alone to assess the effect it had on RDS.
Liu 2006Vitamin K was added to dexamethasone and compared with dexamethasone injection alone, vitamin K injection alone or no treatment to determine which treatment was most effective in reducing the incidence of IVH.
Romaguera 1997Intra-amniotic thyroxine and intramuscular betamethasone versus betamethasone alone to assess the effect it had on maturity.
Salzer 1982Carnitine and dexamethasone versus dexamethasone to assess the effect it had on lung maturity.
Shanks 2010Corticosteroids (either betamethasone or dexamethasone) were compared with placebo.
Vytiska 1985Carnitine and betamethasone versus betamethasone to assess the effects on RDS prophylaxis.
Whitt 1976Trial not randomised.

Characteristics of studies awaiting assessment [ordered by study ID]

Romejko-Wolniewicz 2013

MethodsRandomised controlled trial.
Participants121 women who gave birth by gestational week 35 and within 7 days after the completion of a full course of steroid treatment (24 mg).
Interventions6 4 mg doses of betamethasone every 8 hours was compared with 2 12 mg doses of betamethasone separated by 24 hours.
OutcomesMaternal rheological parameters and C-reactive protein concentrations; leucocyte blood counts; gestational age; method of delivery; neonatal birthweight, status, and complications.
NotesRandomisation unclear: "The dosage regimen was chosen in a randomized manner"; awaiting contract from trialists.

Characteristics of ongoing studies [ordered by study ID]

Crowther 2010

  1. a

    SD: standard deviation
    IVH: intraventricular haemorrhage
    PVL: periventricular leukomalacia
    ROP: retinopathy of prematurity
    PDA: patent ductus arteriosus
    RDS: respiratory distress syndrome
    CLD: chronic lung disease

Trial name or titleAustralasian Antenatal Study To Evaluate the Role of Intramuscular Dexamethasone versus Betamethasone prior to preterm birth to increase survival free of childhood neurosensory disability- a randomised controlled trial (A*STEROID).
MethodsThe randomised schedule will use balanced variable blocks and will be created using computer software (computerised sequence generation) by researchers not involved in clinical care. Assignment to either group will be stratified for collaborating centre, gestational age (< 28 weeks, > 28 weeks' gestation), and number of fetuses (1 or 2).
ParticipantsWomen at risk of preterm birth at less than 34 weeks' gestation, who have a singleton or twin pregnancy, have no contraindications to the use of antenatal corticosteroids and give informed consent.
Interventions

Intervention:

Dexamethasone (antenatal corticosteroid),

2 syringes of 12 mg dexamethasone sodium phosphate- a non-sulphite containing preparation

Administered as 2 intramuscular injections, 24 hours apart.

Control:

Betamethasone (antenatal corticosteroid),

2 syringes of 11.4mg betamethasone (as Clestone Chronodose 11.4 mg)

Administered as 2 intramuscular injections, 24 hours apart.

Outcomes

Primary outcome:

Composite of incidence of death (defined as stillbirths, deaths from live born infants before and after hospital discharge) or any neurosensory disability in the children (includes cerebral palsy, blindness, deafness and any developmental delay defined as a standardised score more than 1 SD below the mean (< -1 SD).

Secondary outcomes:

Neonatal outcomes:

IVH; severe IVH; PVL; ROP requiring treatment; PDA requiring treatment; use of inotropes; RDS; Severity of any neonatal lung disease; CLD; use of mechanical ventilation; confirmed infection within the first 48 hours; infection after the first 48 hours; birthweight.

Childhood outcomes:

Developmental domains as measured by Ages & Stages Questionnaire; body size; general health (including use of health services since primary hospitalisation; childhood respiratory morbidity; blood pressure z scores and proportions in hypertensive ranges and behaviour.

Maternal outcomes:

Maternal perinatal infectious morbidity (defined as clinical chorioamnionitis requiring intrapartum antibiotics, use of postpartum antibiotics).

Starting date1/12/2008
Contact information

Professor Caroline Crowther

Australian Research Centre for Health of Women and Babies (ARCH)

Discipline of Obstetrics and Gynaecology

The University of Adelaide

Women's and children's Hospital

King William Road

North Adelaide

South Australia 5006

Australia

Tel: +61 8 8161 7747

Fax: +61 8 8161 7652

Notes 

Ancillary