Multiple courses of antenatal corticosteroids for preterm birth (MACS): a randomised controlled trial

Articles

Multiple courses of antenatal corticosteroids for preterm birth (MACS): a randomised controlled trial Kellie E Murphy, Mary E Hannah, Andrew R Willan, Sheila A Hewson, Arne Ohlsson, Edmond N Kelly, Stephen G Matthews, Saroj Saigal, Elizabeth Asztalos, Susan Ross, Marie-France Delisle, Kofi Amankwah , Patricia Guselle, Amiram Gafni, Shoo K Lee, B Anthony Armson, for the MACS Collaborative Group*

Summary Background One course of antenatal corticosteroids reduces the risk of respiratory distress syndrome and neonatal death. Weekly doses given to women who remain undelivered after a single course may have benefits (less respiratory morbidity) or cause harm (reduced growth in utero). We aimed to find out whether multiple courses of antenatal corticosteroids would reduce neonatal morbidity and mortality without adversely affecting fetal growth. Methods 1858 women at 25–32 weeks’ gestation who remained undelivered 14–21 days after an initial course of antenatal corticosteroids and continued to be at high risk of preterm birth were randomly assigned to multiple courses of antenatal corticosteroids (n=937) or placebo (n=921), every 14 days until week 33 or delivery, whichever came first. The primary outcome was a composite of perinatal or neonatal mortality, severe respiratory distress syndrome, intraventricular haemorrhage (grade III or IV), periventricular leucomalacia, bronchopulmonary dysplasia, or necrotising enterocolitis. Analysis was by intention to treat. All patients and caregivers were unaware of the treatment given. This trial is registered as number ISRCTN2654148. Findings Infants exposed to multiple courses of antenatal corticosteroids had similar morbidity and mortality to those exposed to placebo (150 [12·9%] vs 143 [12·5%]). Those receiving multiple doses of corticosteroids also weighed less at birth than those exposed to placebo (2216 g vs 2330 g, p=0·0026), were shorter (44·5 cm vs 45·4 cm, p<0·001), and had a smaller head circumference (31·1 cm vs 31·7 cm, p<0·001). Interpretation Multiple courses of antenatal corticosteroids, every 14 days, do not improve preterm-birth outcomes, and are associated with a decreased weight, length, and head circumference at birth. Therefore, this treatment schedule is not recommended. Funding Canadian Institutes of Health Research.

Introduction Preterm birth is a worldwide health-care problem contributing greatly to neonatal morbidity and mortality. The administration of one course of antenatal corticosteroids to women who are at high risk of giving birth prematurely reduces the risk of neonatal mortality, respiratory distress syndrome, and intraventricular haemorrhage.1 However, women who receive one course may remain undelivered for weeks afterwards. Basic science and clinical research have suggested that the benefits of one course might diminish over time. Thus, multiple courses of corticosteroids every 7–14 days have been administered even before completion of randomised controlled trials.2–5 Several trials have investigated the short-term benefits of weekly courses of antenatal corticosteroids versus placebo in women who had already received one course of corticosteroids.6–8 Overall, initial small trials showed no benefit.6,8 However, the National Institutes of Child Health and Human Development (NICHD) Maternal Fetal Medicine Units Network trial8 showed a trend towards improved composite outcome for infants who were exposed to weekly courses of antenatal corticosteroids born before 32 weeks’ gestation (23% of infants on antenatal corticosteroids vs 39% on placebo, p=0·08). The Australasian www.thelancet.com Vol 372 December 20/27, 2008

Collaborative Trial of Repeat Doses of Steroids (ACTORDS)7 enrolled 982 women and showed a benefit of weekly courses of antenatal corticosteroids. Fewer infants in the treatment group than in the placebo group had respiratory distress syndrome (33% vs 41%; relative risk [RR] 0·82 [95% CI 0·71–0·95], p=0·01) and severe lung disease (12% vs 20%; 0·60 [0·46–0·79], p=0·0003). A Cochrane systematic review, which included the results of these trials, suggested that weekly courses of antenatal corticosteroids are associated with reduced occurrence [0·82 (0·72–0·93)] and severity [0·60 (0·48–0·75)] of neonatal lung disease, and serious infant morbidity [0·79 (0·67–0·93)].9 However, multiple courses of antenatal corticosteroids may have adverse effects10–13 such as decreased fetal growth.8,14,15 Such treatment might also have long-term adverse effects; indeed, adverse neurological outcomes have been shown in follow-up studies of children given dexamethasone after birth.16 Our aim was to see if a less frequent intervention (a course every 14 days in our trial vs every 7 days in other steroid trials) would show short-term respiratory benefits, and reduce to a minimum the risk of short-term and long-term adverse effects.

Lancet 2008; 372: 2143–51 See Comment page 2094 *Members listed at end of paper Department of Obstetrics and Gynaecology, Mount Sinai Hospital, University of Toronto, Toronto, ON, Canada (K E Murphy MD); Department of Obstetrics and Gynaecology, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, ON, Canada (Prof M E Hannah MDCM); Programme in Child Health Evaluative Sciences, SickKids Research Institute, Department of Public Health Sciences, University of Toronto, Toronto, ON, Canada (Prof A R Willan PhD); Maternal, Infant and Reproductive Health Research Unit at the Women’s College Research Institute, University of Toronto, Toronto, ON, Canada (S A Hewson BA, P Guselle MSc); Department of Paediatrics, Mount Sinai Hospital, University of Toronto, Toronto, ON, Canada (Prof A Ohlsson MD, E N Kelly MB); Departments of Physiology, Obstetrics and Gynaecology and Medicine, University of Toronto, Toronto, ON, Canada (Prof S G Matthews PhD); Department of Paediatrics, McMaster University Medical Centre, Hamilton, ON, Canada (Prof S Saigal MD); Department of Newborn and Developmental Paediatrics, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, ON, Canada (E Asztalos MD); Department of Obstetrics and Gynaecology, University of Calgary, Calgary, AB, Canada (Prof S Ross PhD); Department of Obstetrics and Gynaecology, BC Women’s Hospital, University of British Columbia, Vancouver, BC, Canada (M-F Delisle MD); Department of Gynaecology and Obstetrics, Women’s and Children’s Hospital, State University of New York at Buffalo, Buffalo, NY, USA (Prof K Amankwah MD); Centre for Health Economics and Policy Analysis,

2143

Articles

Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, ON, Canada (Prof A Gafni DSc); Department of Paediatrics and the Integrated Centre for Care Advancement through Research (iCARE), University of Alberta, Edmonton, AB, Canada (Prof S K Lee MBBS); and Department of Obstetrics and Gynaecology, IWK Health Centre, Dalhousie University, Halifax, NS, Canada (Prof B A Armson MD) Correspondence to: Dr Kellie E Murphy, Mount Sinai Hospital, Maternal Fetal Medicine, Department of Obstetrics and Gynaecology, Room 3726, Ontario Power Generation Building, 700 University Avenue, Toronto, ON, M5R 1X5, Canada [email protected]

Methods Participants Multiple courses of antenatal corticosteroids for preterm birth study (MACS) is an international, multicentre, double-blind, randomised controlled trial. 1858 women between 25 and 32 weeks of gestation who remained undelivered 14–21 days after an initial course of antenatal corticosteroids (either betamethasone or dexamethasone) and continued to be at high risk of preterm birth (table 1) were enrolled in 80 centres in 20 countries. The study protocol was explained and consenting participants were randomly assigned to a study group after informed consent. Women were not eligible if they had a contraindication to corticosteroids, needed chronic doses of these drugs, had evidence of chorioamnionitis, carried a fetus with a known lethal congenital abnormality, had an initial course of corticosteroids before 23 weeks’ gestation, or previously participated in MACS. Women with a multiple pregnancy were judged eligible for MACS if a fetus had died before 13 weeks’ gestational age; however, the dead fetus was not included as an outcome. Women were not eligible if they

Maternal age (years)

Antenatal corticosteroids (N=935)

Placebo (N=918)

29·1 (6·23)

29·1 (6·18)

Number of fetuses 1

737 (79%)

726 (79%)

2

162 (17%)

158 (17%)

3

36 (4%)

34 (4%)

Number of previous pregnancies 0

263 (28%)

252 (27%)

1–4

577 (62%)

571 (62%)

>4

95 (10%)

91 (10%)

65 (7%)

69 (8%)

322 (34%)

334 (36%)

53 (6%)

60 (7%)

Previous second-trimester abortion (14–19 weeks) Previous preterm delivery (20–<37 weeks) History of a previous pregnancy with intrauterine growth restriction Method of gestational age calculation*

31 (3%)

41 (4%)

Ultrasound±clinical

Clinical only

896 (96%)

874 (95%)

Estimated fetal weight by ultrasound at randomisation (g)†

1203 (12·9)

1211 (12·8)

Mean gestational age at randomisation (weeks)

29·3 (2·0)

29·4 (2·0)

Gestational age at randomisation <25 weeks

1 (<1%)

0

25–27 weeks

256 (27%)

255 (28%)

28–32 weeks

678 (73%)

661 (72%)

>32 weeks Fetal anomalies

0

2 (<1%)

2 (<1%)

5 (<1%)

Agenesis of right kidney

1 (<1%)

··

Cardiac abnormalities

1 (<1%)

2 (<1%)

Cleft lip and palate

··

1 (<1%)

Diaphragmatic hernia

··

1 (<1%)

Hydrocephaly

··

1 (<1%) (Continues on next page)

2144

experienced a fetal death after 13 weeks’ gestation. MACS received ethical approval from the University of Toronto, Mount Sinai Hospital, and from all collaborating institutions.

Procedure and outcomes Randomisation was done with a 24-h telephone service after patient eligibility and baseline information were recorded. A study number was assigned, corresponding to a box of study medication at the participating centre. The study was stratified by gestational age and centre. Women randomly assigned to the antenatal corticosteroids group received two doses of 12 mg betamethasone—a combination of 6 mg betamethasone sodium phosphate and 6 mg betamethasone sodium acetate (Celestone, Schering-Plough Corporation, Madison, NJ, USA)—intramuscularly 24 h apart. Women randomly assigned to placebo received similarly appearing intramuscular injections, containing a dilute concentration of aluminum monostearate (Eminent Services Corporation, Frederick, MD, USA). This substance is used as a filler in many pharmaceutical preparations and is inert. 901 (96%) women in the treatment group and 875 (95%) in the placebo group had an ultrasound within 21 days before randomisation; 20 (2%) in the treatment group and 27 (3%) in the placebo group had an ultrasound more than 21 days before randomisation, and 10 (1%) in the treatment group and 14 (2%) in the placebo group had an ultrasound after randomisation. Women who remained at increased risk of preterm birth after the first course of study medication—after study enrolment and after their first course of either betamethasone or placebo—continued to receive two doses of 12 mg betamethasone or placebo, 24 h apart, every 14 days until 33 weeks’ gestation or birth, whichever happened first. For women who had preterm rupture of the membranes, the recommendation was that investigators would stop the study medication at 32 weeks’ gestation. All patients and caregivers were unaware of the treatment given. Women were asked to complete a structured questionnaire 3 months after giving birth to assess the occurrence of postpartum depression and other maternal side-effects. The primary outcome was perinatal or neonatal mortality, or neonatal morbidity. Perinatal or neonatal mortality was defined as stillbirth or neonatal death during the first 28 days of life or before hospital discharge, whichever happened later. Clinically significant neonatal morbidity was defined as one or more of the following: (i) severe respiratory distress syndrome—ie, needing assisted ventilation via endotracheal tube and supplemental oxygen both within the first 24 h of life and for 24 h or more, and either a radiographic scan compatible with respiratory distress syndrome or surfactant given between the first 2–24 h of life; (ii) bronchopulmonary dysplasia—ie, needing oxygen at a postmenstrual age of 36 completed www.thelancet.com Vol 372 December 20/27, 2008

Articles

weeks and radiographic scan compatible with bronchopulmonary dysplasia; (iii) intraventricular haemorrhage grade III or IV, diagnosed by cranial ultrasound with Papile and colleagues’ classification;17 (iv) cystic periventricular leucomalacia—ie, periventricular cystic changes in the white matter, excluding subependymal and choroid plexus cysts, diagnosed by cranial ultrasound; (v) necrotising enterocolitis—either perforation of intestine, pneumatosis intestinalis, or air in the portal vein, diagnosed by radiographic scan or at surgery.18 For infants with a birthweight of less than 1500 g, participating centres were encouraged to arrange at least two cranial ultrasounds before hospital discharge to look for evidence of intraventricular haemorrhage and cystic periventricular leucomalacia. Other neonatal outcomes were weight, length, and head circumference at birth, neonatal infection, retinopathy of prematurity, stay in neonatal intensive care unit, use of ventilation with intubation, and patent ductus arteriosus needing pharmacological treatment or surgery. Neonatal infection was defined as clinical sign of infection, and one or more of the following: a positive culture of blood or cerebrospinal fluid, a gram-positive stain of cerebrospinal fluid, or chest radiographic findings compatible with pneumonia. Maternal outcomes were clinical chorioamnionitis— defined as maternal temperature of 38°C or more before delivery, and one or more of the following signs: maternal tachycardia (120 beats per minute [bpm] or more), white blood-cell count of 20 000 μL or more, fetal tachycardia (>160 bpm), uterine tenderness, or foul smelling amniotic fluid—and maternal infection after delivery— defined as one or more of the following signs: endometritis (ie, postpartum maternal temperature of 38°C or more and tender fundus without other source of infection); pneumonia (ie, maternal temperature of 38°C or more and signs of pneumonia on radiographic scan); wound infection or breakdown; pyelonephritis (ie, maternal temperature of 38°C or more, positive urine culture, and costal vertebral angle tenderness); or maternal sepsis (ie, maternal temperature of 38°C or more and a positive blood culture). We obtained information about drug administration and co-interventions (table 2). After the first 600 women were randomised, random samples of study medication were sent to a sham site for testing. The study drug was analysed to ensure that study numbers corresponding to placebo and betamethasone were correct. Overall, 160 vials were tested and the assigned study number allocation corresponded with the actual study medication 100% of the times.

Statistical analysis We calculated a sample size of 1900 women (950 per group) to have 80% probability of achieving a significant difference between the two groups with a two-tailed, type I error of 0·05, if multiple courses of antenatal www.thelancet.com Vol 372 December 20/27, 2008

Antenatal corticosteroids (N=935)

Placebo (N=918)

(Continued from previous page) Prestudy course of ACS Betamethasone

810 (87%)

786 (86%)

Dexamethasone

125 (13%)

131 (14%)

15·0 (14·0, 21·0)

15·0 (13·0, 21·0)

Time from first dose of prestudy corticosteroids to randomisation (days)‡ Medical and obstetrical problems at enrolment Uterine contractions within previous week

520 (56%)

522 (57%)

Short cervical length or cervical dilation

457 (49%)

450 (49%)

Antepartum vaginal bleeding

129 (14%)

130 (14%)

Preterm rupture of membranes

149 (16%)

142 (15%)

Intrauterine growth restriction

85 (9%)

74 (8%)

Pre-eclampsia

43 (5%)

53 (6%)

108 (12%)

93 (10%)

Smoking Substance abuse

7 (<1%)

5 (<1%)

Hypertension needing treatment

66 (7%)

70 (8%)

Maternal diabetes

50 (5%)

39 (4%)

Controlled by diet only

31 (62%)

24 (62%)

Insulin dependent

19 (38%)

15 (38%)

Maternal treatments during previous 2 weeks Antibiotics

331 (35%)

292 (32%)

Tocolytics

465 (50%)

439 (48%)

266 (57%)

256 (58%)

Magnesium sulphate (iv)

Betamimetics (iv)

81 (17%)

89 (20%)

Indomethacin (po or pr)

61 (13%)

50 (11%)

150 (32%)

135 (31%)

99 (21%)

107 (24%)

Calcium channel blocker (po) Other§ Principal reason(s) for study participation Signs or symptoms of preterm labour

773 (83%)

777 (85%)

Fetal anomalies or pathologies

115 (12%)

103 (11%)

Maternal medical condition

199 (21%)

190 (21%)

Multiple pregnancy

191 (20%)

179 (19%)

History of obstetrical complications

287 (31%)

280 (31%)

National perinatal mortality rate ≤10 in 1000

623 (67%)

612 (67%)

>10–20 in 1000

239 (26%)

238 (26%)

73 (8%)

68 (7%)

>20 in 1000

Data are mean (SD) or n (%), unless otherwise stated. ACS=antenatal corticosteroid. iv=intravenously. po=orally. pr=rectally *Some data from some of the variables were never obtained and are therefore missing. †Cluster analysis was used to adjust for the interdependency of multiple births. Data are mean (SE). ‡Data are median (5th, 95th centile). §Oral betamimetics, atosiban, theophylline, terbutaline, nitroglycerine, hexoprenaline, oral magnesium sulphate, progesterone, spasmolytics, with the addition of monothioglycerol and diazepam in the placebo group.

Table 1: Baseline characteristics of enrolled women

corticosteroids reduced the risk of respiratory distress syndrome from 12% to 8%. We did an interim analysis of all data on the first 800 patients enrolled after complete data were obtained. We presented the results to an independent data safety monitoring board. The a-priori stopping rule was finding a higher rate of the primary outcome in the treatment group than in the placebo group with a one-tailed, type I error of 0·002. 2145

Articles

Antenatal corticosteroids (N=935)

Placebo (N=918)

p value

Number of courses of study drug* 0

5 (<1%)

5 (<1%)

1

385 (41%)

365 (40%)

2

305 (33%)

273 (30%)

3

150 (16%)

169 (18%)

90 (10%)

104 (11%)

4 Fully compliant with study drug†

747 (80%)

722 (79%)

Partly compliant with study drug

181 (19%)

189 (21%)

Non-compliant, incorrect, or no study drug given Gestational age at birth (weeks)

7 (<1%)

6 (<1%)

34·5 (3·6)

34·9 (3·6)

Gestational age at birth‡ <28 weeks

39 (4%)

27 (3%)

28–32 weeks

280 (30%)

254 (28%)

33–36 weeks

338 (36%)

319 (35%)

≥37 weeks

278 (30%)

318 (35%)

Time of delivery after repeated study drug exposure§ <48 h

92 (10%)

91 (10%)

48 h to <7 days

153 (16%)

131 (14%)

≥7 days

685 (73%)

689 (75%)

32 (3%)

28 (3%)

15 (47%)

11 (39%)

90 (10%)

79 (9%)

Worsening of pre-existing or onset of diabetes after randomisation§ Women treated with insulin Worsening of pre-existing or onset of hypertension after randomisation§ Women treated with anti-hypertensive drugs Tocolytics given after randomisation Betamimetics (iv)

73 (81%)

62 (78%)

323 (35%)

339 (37%)

168 (52%)

172 (51%)

MgSO4 (iv)

71 (22%)

79 (23%)

Indomethacin (po or pr)

27 (8%)

32 (9%)

Calcium channel blocker (po)

133 (41%)

136 (40%)

Other¶

101 (31%)

108 (32%)

Corticosteroids administered in addition to the study drug after randomisation§

13 (1%) 9 (69%)

Dexamethasone

2 (15%)

3 (38%)

Prednisone

2 (15%)

0

Reasons for giving corticosteroids in addition to the study drug after randomisation

13 (1%) 3 (23%)

2 (25%)

4 (31%)

4 (50%)

Maternal medical indication||

4 (31%)

2 (25%)

Not stated

2 (15%)

0 503 (55%)

Before labour

251 (50%)

247 (49%)

During labour or at delivery

375 (74%)

363 (72%)

Clinical chorioamnionitis§

0·6 (0, 1009·3)

Results

8 (<1%)

Clinical decision

Duration of rupture of membranes (h)**

MACS was funded by the Canadian Institutes of Health Research (CIHR). CIHR had no role in the design, management, data collection, analysis, or interpretation of the data. CIHR had no role in the writing of the manuscript or in the decision to submit for publication. All authors had full access to the data. KEM had final responsibility for the decision to submit the paper for publication.

5 (63%)

Staff error

Antibiotics after randomisation, and before or at delivery 505 (54%)

Role of funding source

8 (<1%)

Betamethasone

0·8 (0, 1069·0)

22 (2%)

20 (2%)

Vaginal delivery

396 (42%)

415 (45%)

Caesarean section

537 (57%)

501 (55%)

0·80

Method of delivery

Unknown Antibiotics after delivery§

2 (<1%)

2 (<1%)

276 (30%)

255 (28%) (Continues on next page)

2146

Final analysis of the outcome data was by intention to treat. We used a logistic regression model, with a random effect for multiple pregnancy to adjust for the dependence of observation within the pregnancy, to estimate the adjusted odds ratios (ORs) and measure the 95% CI for the comparison of the two groups in relation to the primary and other outcomes. Significance for the primary outcome was p<0·05 and for the other outcomes was p<0·01. We analysed interactions between treatment group and selected baseline variables for the primary outcome. Although not planned a priori, the rate of the primary outcome was calculated for the two groups in the subgroups of infants born at 32 weeks’ gestation or earlier, and those born within 7 days of repeated study drug administration. We did the first subanalysis because previous trials suggested that multiple courses of antenatal corticosteroids might benefit infants born before 32 weeks’ gestational age, and the second in an effort to compare findings of MACS with previously published trials of weekly courses of antenatal corticosteroids.6–8

The figure shows the trial profile. Eight women (five from the antenatal corticosteroid group and 3 from the placebo group), whose one or more fetuses from a multiple gestation died in utero after 13 weeks but before randomisation, were randomly assigned. Live fetuses from these pregnancies were excluded from the analysis (five from the antenatal corticosteroid group and four from the placebo group). After intention-to-treat analysis, data from the eight women were obtained and included in the analysis. Table 1 shows the baseline characteristics of the women recruited for the trial. Randomisation started on April 9, 2001, and finished on Aug 31, 2006. Countries with a national perinatal mortality rate of 10 in 1000 or less were Canada, Chile, Denmark, Germany, Hungary, Israel, Netherlands, Poland, Spain, Switzerland, UK, and USA. Countries with a national perinatal mortality rate of 10–20 in 1000 were Argentina, Brazil, and Peru; and countries with a national perinatal mortality rate of more than 20 in 1000 were Bolivia, China, Colombia, Jordan, and Russia.19 The study drug was given according to the protocol. 1469 women were fully compliant with study drug. Partial compliance was defined as first course of study drug being given less than 14 or more than 21 days after initial www.thelancet.com Vol 372 December 20/27, 2008

Articles

treatment (n=31 in treatment group and n=36 in placebo group); one or more course of study drug being given less than 14 days apart (n=28 in treatment group and n=17 in placebo group); one or more course of study drug being given more than 14 days apart (n=67 in treatment group and n=70 in placebo group); study drug stopped more than 14 days before delivery or before 33 weeks of gestational age (in women with intact membranes) or before 32 weeks (in women with preterm rupture of the membranes), whichever came first, secondary to patient or physician request, or other reasons including patient did not return to hospital or subsequent course date miscalculated in the antenatal corticosteroid group, and patients did not return to the hospital, subsequent course date miscalculated, rash, or vaginal infection in the placebo group (n=72 in the treatment group and n=78 in the placebo group). 370 women were partly compliant. Non-compliance was defined as incorrect study drug given (n=2 in the treatment group and n=1 in the placebo group) or no study drug given (n=5 in the treatment group and n=5 in the placebo group). 13 women were non-compliant. Of those women who stopped the study drug early, 110 (73%) stopped after 31 weeks’ gestational age, whereas the remaining 40 (27%) stopped before 31 weeks’ gestational age. One participant’s treatment in the antenatal corticosteroid group was unmasked at the request of her private physician; however, she completed the trial, and her and her infant’s data were obtained. Gestational age was confirmed by ultrasound examination for most women in the study (96% and 95% in the treatment and placebo groups, respectively). During the trial, eight serious adverse events were reported, two from the treatment group (neonate with a respiratory syncitial virus infection in a woman with diabetes [n=1] and neonate admitted to intensive care in a woman with pre-eclampsia [n=1]), and six from the placebo group (temporary loss of consciousness of the mother [n=1], maternal and neonatal Escherichia coli infection followed by neonatal death [n=1], neonatal thrombocytopenia [n=1], neonatal subarachnoid haemorrhage [n=1], maternal Bell’s palsy [n=1], and neonatal lung aspiration and neonatal death [n=1]). The composite primary outcome of perinatal or neonatal mortality, severe respiratory distress syndrome, bronchopulmonary dysplasia, intraventricular haemorrhage (grade III or IV), periventricular leucomalacia, or necrotising enterocolitis did not substantially differ between treatment and placebo groups (p=0·83) (table 3). For the primary outcome, we assessed interactions between treatment group and gestational age at randomisation, preterm rupture of the membranes, and single versus multiple pregnancies. We identified no significant interactions, which suggested that the risk of primary outcomes was similar for the treatment and the placebo groups. At birth, infants who received multiple courses of antenatal corticosteroids weighed less than those who received www.thelancet.com Vol 372 December 20/27, 2008

Antenatal corticosteroids (N=935)

Placebo (N=918)

p value

(Continued from previous page) Maternal infection after delivery§

34 (4%)

Endometritis

21 (62%)

25 (3%)

0·26

9 (36%)

Pneumonia

3 (9%)

2 (8%)

Pyelonephritis

4 (12%)

4 (16%)

Sepsis

4 (12%)

4 (16%)

Wound infection resulting in wound breakdown

7 (21%)

10 (40%)

Serious maternal complications before discharge§††

29 (3%)

20 (2%)

Duration in hospital after delivery (h)**

78·8 (25·8, 236·1)

76·8 (25·4, 260·0)

Data are n (%) or mean (SD), unless otherwise stated. iv=intravenously. po=orally. pr=rectally *Number unknown for two women in the placebo group. †Compliance was unknown for one woman in the placebo group. ‡In case of a multiple pregnancy, the earliest gestational age at delivery was taken in consideration. §One or two values are missing from this group; percentages are calculated on the data available. ¶Oral betamimetics, atosiban, theophylline, terbutaline, nitroglycerine, betamimetics (route not specified) hexoprenaline, oral magnesium sulphate, spasmolytics, diazepam, with the addition of methyldopa and progesterone in the treatment group. Some patients received more than one tocolytic drug. ||Maternal medical indications included facial nerve paresis (n=1), allergy treatment (n=1), colitis (n=1), and asthmatic crisis (n=1) in the treatment group; facial nerve paresis (n=1) and allergy treatment (n=1) in the placebo group. **Data are median (5th, 95th centile). ††Serious maternal complications were: postpartum haemorrhage, placenta accreta, or hysterectomy (n=15); pregnancy induced hypertension or haemolysis; high concentration of liver enzymes, low platelet number, and pre-eclampsia (n=5); thromboembolic and vascular event (n=4); spinal headache (n=1); mastectomy (n=1); cardiomyopathy (n=1); sponge left in (n=1); pancreatitis (n=1) in the treatment group; postpartum haemorrhage, placenta accreta, or hysterectomy (n=10); pregnancy induced hypertension or haemolysis, high concentration of liver enzymes, low platelet number, and pre-eclampsia (n=1), thromboembolic and vascular event (n=1); anaphylactic shock (n=1); ureteric obstruction (n=1); toxic shock (n=1); bladder haematoma (n=1); bladder injury (n=1); nephrotic syndrome (n=1); postpartum depression or domestic violence (n=1); placenta percreta or death (n=1) in placebo group.

Table 2: Maternal treatments and outcomes

1858 pregnant women randomised

937 women allocated to repeated ACS (1171 fetuses)

921 women allocated to placebo (1147 fetuses)

2 women (2 fetuses) Lost to follow-up

935 had maternal outcomes analysed

3 women (3 fetuses) Lost to follow-up

918 had maternal outcomes analysed

5 fetuses/infants excluded*

1164 infants had perinatal outcomes analysed

4 fetuses/infants excluded*

1140 infants had perinatal outcomes analysed

Figure: Trial profile ACS=antenatal corticosteroid.*Stillbirths within multiple pregnancies that took place before randomisation.

placebo (p=0·0026), were shorter (p<0·001), and had a smaller head circumference (p<0·001) (table 3). Neonatal respiratory and other outcomes were comparable between groups (table 3). Rates of the composite primary outcome and components were similar between the groups for infants who were born before 32 weeks and for those born within 7 days of repeated study drug administration (table 4). 2147

Articles

Total number of infants Composite primary outcome*†

Antenatal corticosteroids

Placebo

1164

1140

150 (13%)

143 (13%)

Singletons

88/737 (12%)

83/726 (11%)

Multiples

62/427 (15%)

60/414 (15%)

43 (4%)

40 (4%)

Stillbirth or neonatal death ≤28 days after birth or before discharge, whichever happened later Number of surviving infants‡

1121

Mean difference (95% CI)

p value

1·04 (0·77 to 1 ·39)

0·83

1·08 (0·67 to 1·66)

0·82

1100

Severe RDS

87 (8%)

77 (7%)

1·14 (0·80 to 1·58)

0·51

BPD

19 (2%)

11 (1%)

1·50 (0·68 to 2 ·95)

0·37

IVH (grade III or IV)

6 (<1%)

9 (<1%)

0·92 (0·37 to 1 · 88)

0·68

Cystic PVL

9 (<1%)

10 (<1%)

1·07 (0·41 to 2 ·33)

0·95

12 (1%)

1·03 (0·38 to 2 · 29)

0·87

NEC Total number of infants

10 (<1%) 1164

1140

Male

616 (53%)

Female§

546 (47%)

540 (48%)

2216 (28·3)

2330 (28·7)

44·5 (0·2)

45·4 (0·2)

–0·9 (0·25) (–1·34 to –0·37)

<0·001

31·1 (0·1)

31·7 (0·1)

–0·6 (0·15) (–0·90 to –0·32)

<0·001

196 (17%)

158 (14%)

Birthweight (g) Length at birth (cm) Mean head circumference (cm) Birthweight less than 10th centile for gestational age20§ Birthweight less than 3rd centile for gestational age20§ Number of surviving infants‡ Cord blood pH<7·0§¶

64 (6%) 1121 3 (<1%)

Apgar score <7 at 5 mins after birth||

40 (4%)

Major life-threatening anomaly§

10 (<1%)

598 (53%)

59 (5%) 4 (<1%) 49 (5%) 8 (<1%)

1 (<1%)

··

Atrial septal defect

1 (<1%)

··

Hypoplastic right ventricle

1 (<1%)

··

Tuberous sclerosis

1 (<1%)

··

Deformed calvarium

1 (<1%)

··

Agenesis right kidney

1 (<1%)

··

Malformed lung vessels

1 (<1%)

··

Diaphragmatic hernia

1 (<1%)

··

Trisomy 21

1 (<1%)

2 (<1%)

Congenital cytomegalovirus infection

1 (<1%)

··

Cleft lip and palate

··

1 (<1%)

Tetralogy of fallot

··

1 (<1%)

Congenital heart anomaly (hypoplastic left heart)

··

1 (<1%)

Transposition of major vessels

··

1 (<1%)

Polycystic kidney

··

1 (<1%)

Lower urinary outlet obstruction Duration of supplemental oxygen (h)**†† Intubation and ventilation via endotracheal tube Duration of endotracheal ventilation (h)**†† Other ventilatory support without intubation Duration of other ventilatory support (h)**†† Surfactant given

.. 410 (37%) 57·6 (1·5, 1696·2) 175 (16%) 68·3 (5·0, 1123·8) 274 (24%) 57·6 (1·1, 1117·0) 122 (11%)

0·0026

1100

Patent ductus arteriosus

Receiving supplemental oxygen after initial resuscitation

–113·1 (37·3) (–187·0 to –41·17)

1 (<1%) 427 (39%) 68·1 (2·1, 1091·0) 204 (19%)

0·84 (0·63 to 1·09)

0·19

48·4 (3·5, 624·3) 291 (27%) 68·3 (4·0, 995·8) 141 (13%) (Continues on next page)

Discussion In MACS, after an initial course of antenatal corticosteroids, infants born to women who received multiple courses of treatment every 14 days had similar risk of morbidity and mortality to those who were born to 2148

women receiving placebo. These findings differ from those of the Australian trial ACTORDS7 in which women who received weekly courses of treatment showed short-term neonatal benefits over women receiving placebo. In MACS, the lack of improvement in respiratory www.thelancet.com Vol 372 December 20/27, 2008

Articles

Antenatal corticosteroids

Placebo

Mean difference (95% CI)

p value

0·96 (0·76 to 1·20)

0·71

1·23 (0·88 to 1 67)

0·24

1·24 (0·74 to 1·98)

0·46

(Continued from previous page) Any ventilatory support Postnatal corticosteroids given to the infant Neonatal infection‡‡ Sepsis

449 (40%)

495 (45%)

15 (1%)

8 (<1%)

105 (9%)

91 (8%)

67 (64%)

46 (51%)

Meningitis

7 (7%)

2 (2%)

Pneumonia

45 (43%)

55 (60%)

Seizure before discharge Patent ductus arteriosus needing treatment

8 (<1%)

10 (<1%)

44 (4%)

42 (4%)

37 (84%)

28 (67%)

Treatment with ligation

8 (18%)

12 (29%)

Treatment with other

6 (14%)

Treatment with indomethacin

Retinopathy of prematurity§§ Admitted to neonatal intensive care unit Length of stay in neonatal intensive care unit (h)††

6 (14%)

46 (4%)

34 (3%)

1·04 (0·55 to 1·81)

0·99

465 (42%)

464 (42%)

1·03 (0·85 to 1·24)

0·80

192 (24·0, 1392·0)

192 (24·0, 1200·0)

Data are n (%) or mean (SE), unless otherwise stated. ACS=antenatal corticosteroids. RDS=respiratory distress syndrome. BPD=bronchopulmonary dysplasia. IVH=intraventricular haemorrhage. PVL=periventricular leucomalacia. NEC=necrotising enterocolitis. *One or more of the following: neonatal mortality, severe respiratory distress syndrome, bronchopulmonary dysplasia, intraventricular haemorrhage (grade III or IV), cystic periventricular leucomalacia, and necrotising enterocolitis. †No significant interactions existed between the primary outcome and gestational age at randomisation (p=0·34), preterm prelabour rupture of membranes (p=0·97), or type of pregnancy (singleton vs multiple, p=0·76). ‡Surviving infants defined as alive at 28 days or surviving hospital discharge, whichever happened later. §Few values are missing from this variable; percentages are calculated on data available. ¶Umbilical blood gases from 411 (37%) in the ACS group and 403 (37%) in the placebo group were not obtained or were unknown. ||13 (1%) in the ACS group and 14 (1%) in the placebo group are unknown. **Duration was defined as the time from starting treatment to stopping treatment. ††Data are median (5th, 95th centile). ‡‡Neonatal infection was either sepsis (positive blood culture), meningitis (positive cerebral spinal fluid culture or gram stain), or pneumonia (confirmed by radiography). §§773 (69%) in the ACS group and 821 (75%) in the placebo group were not reported or unknown because ophthalmological examination was not done.

Table 3: Primary and other neonatal outcomes

morbidity might be due to the diminuishing effect of antenatal corticosteroids on fetal pulmonary type II pneumocytes after 7 days. This hypothesis is supported by laboratory evidence, which suggests that the induction of surfactant in the fetal lung is reversible.4,21 However, we showed no benefit of multiple courses in infants delivered within 7 days of repeated study drug exposure, and thus our data do not support this hypothesis. We are therefore uncertain as to why the results of MACS differed from those of the ACTORDS trial. In MACS, infants exposed to multiple courses of antenatal corticosteroids, every 14 days, were significantly smaller at birth than those in the placebo group. They weighed on average 113 g less, were 0·9 cm shorter, and their head circumferences were 0·6 cm smaller. These findings are concerning and are consistent with those from randomised controlled trials of weekly courses of antenatal corticosteroids. The NICHD trial8 showed that more infants in the treatment group than in the placebo group had a birthweight under the 10th percentile (24% vs 15%, p=0·02), and the mean birthweight of infants exposed to four or more courses of antenatal corticosteroids was significantly lower in the treatment group than in the placebo group (2400 g vs 2561 g, p=0·01). In the ACTORDS trial,7 mean Z scores (SD) for weight (–0·40 [1·05] antenatal corticosteroids vs –0·27 [1·14] placebo, p=0·04) and head circumference (–0·30 [1·22] antenatal corticosteroids vs –0·14 [1·28] placebo, p=0·03) were lower in the treatment group than in the placebo group at www.thelancet.com Vol 372 December 20/27, 2008

Antenatal Placebo corticosteroids Number of infants born at <32 weeks‘ gestational age

321

282

Death or serious neonatal morbidity: composite primary outcome*

131 (41%)

110 (39%)

Stillbirth or neonatal death ≤28 days after birth or before discharge, whichever happened later

37 (12%)

33 (12%)

Number of surviving† infants born <32 weeks‘ gestational age

284

249

Severe RDS

79 (28%)

57 (23%)

BPD

19 (7%)

10 (4%)

IVH (grade III or IV)

5 (2%)

7 (3%)

Cystic PVL

6 (2%)

9 (4%)

NEC Number of infants born <7 days after last study drug exposure

8 (3%) 333

9 (4%) 290

Death or serious neonatal morbidity: composite primary outcome*

97 (29%)

84 (29%)

Stillbirth or neonatal death ≤28 days after birth or before discharge, whichever happened later

25 (8%)

28 (10%)

Number of surviving† infants born <7 days after last study drug exposure

308

262

Severe RDS

57 (19%)

43 (16%)

BPD

16 (5%)

9 (3%)

IVH (grade III or IV)

5 (2%)

5 (2%)

Cystic PVL

6 (2%)

6 (2%)

NEC

7 (2%)

6 (2%)

Data are n (%). ACS=antenatal corticosteroids. RDS=respiratory distress syndrome. BPD=bronchopulmonary dysplasia. IVH=intraventricular haemorrhage. PVL=periventricular leukomalacia. NEC=necrotising enterocolitis. *One or more of the following: neonatal mortality, severe respiratory distress syndrome, bronchopulmonary dysplasia, intraventricular haemorrage (grade III or IV), cystic periventricular leukomalacia, and necrotising enterocolitis. †Surviving infants defined as alive at 28 days or surviving hospital discharge, whichever happened later.

Table 4: Primary neonatal outcome in infants born before 32 weeks’ gestation and within 7 days of repeated study drug administration

2149

Articles

birth. An explanation for less effect on fetal growth in ACTORDS was that women in this trial received less treatment than those in other trials. Although most women in the ACTORDS and MACS trials received only one or two courses (66% for ACTORDS and 72% for MACS), in the NICHD trial 63% of women received four or more courses. A repeated course of antenatal corticosteroids in the ACTORDS trial consisted of only one dose of betamethasone at each dosing time, whereas in the other trials (Guinn, NICHD, and MACS) a repeated course consisted of two doses of betamethasone 24 h apart.6,8 The clinical relevance of decreased fetal size due to multiple courses of antenatal corticosteroids is not clear. Long-term follow-up of these infants is therefore important. Children of the NICHD study were followed up to 2–3 years of age. Physical and neurocognitive measures did not differ between groups, but a slight increased risk of cerebral palsy was seen in those exposed to weekly courses of treatment (6 [3%] vs 1 [<1%] placebo).22 In the ACTORDS trial, the rate of survival without major disabilities was similar between the treatment and placebo groups at 2 years of age. However, children exposed to weekly courses of antenatal corticosteroids (n=31) were more likely to warrant an assessment for attention deficits than were controls (n=17; p=0·04).23 These results are concerning and caution is needed in the use of weekly courses of antenatal corticosteroids. 18–24-month follow-up assessments for children in MACS and 5-year follow-up assessments for children in the MACS and ACTORDS trials are continuing. Similar to the other trials, 32% of women in MACS gave birth at term. Their infants did not need antenatal corticosteroids for pulmonary maturation, and therefore they were potentially negatively affected by study medication. Furthermore, they were exposed to more courses of treatment because they remained in utero longer than preterm infants. The mean number of courses of drug was 2·2 for infants who were born at term and 1·8 for those born before 37 weeks of gestation. Overall, multiple courses of antenatal corticosteroids, given every 14 days, are associated with decreased growth in utero and no neonatal benefits compared with one course of antenatal corticosteroids. Therefore, in women who remain at increased risk of preterm birth after receiving an initial course of antenatal corticosteroids, multiple courses every 14 days are not recommended. Contributors KEM, MEH, ARW, SAH, AO, SGM, SS, EA, SR, KA, and BAA participated in the design, methodological issues, implementation, conduct, monitoring, analysis, interpretation of the study, and have written, seen, and approved the final version of the manuscript. ENK participated in the original discussions regarding the development and design of the MACS study, took part in steering committee meetings, and had been involved in assessments of outcome at 2 years of the children recruited at Mount Sinai Hospital. M-FD participated in the implementation and recruitment of Canadian centres, and had seen and approved the final version of the manuscript. PG participated as a

2150

member of the steering committee on the MACS trial, and assisted with methodological and implementation issues. AG participated in the design and monitoring of the MACS study, interpretation of results, and reviewing and commenting on drafts during the writing of the manuscript, and had seen and approved the final version of the manuscript. SKL participated in data interpretation, reviewing of the paper, and had seen and approved the final version of the manuscript. MACS Collaborative Group Steering committee: K E Murphy, K Amankwah, B A Armson, El Asztalos, M-F Delisle, A Gafni, P Guselle, M E Hannah, S A Hewson, E N Kelly, S K Lee, S G Matthews, A Ohlsson, S Ross, J Rovet, S Saigal, R Sananes, I Schmid, A R Willan. Collaborators: Argentina—L Kwiatkowski, S M Tortorella (Hospital Fiorito, Avellaneda); M S Bertin, J L Castaldi, C Deguer, M Klun, C Besegato (Hospital Penna, Bahia Blanca); G Izbizky, M C Vaneri, C A Fustinana, L Otano (Hospital Italiano, Buenos Aires); M S F Palermo, E Javier Murua, D Montes Valera, H Sampietro, A Monaco (Hospital Posadas, Buenos Aires); R Savransky, A Dunaiewsky (Hospital Ramos Mejia, Buenos Aires); M N Basualdo, E Andina, I Di Marco (Hospital Sarda, Buenos Aires); M Rivero, M C Feu, S Garcia (Hospital Angela Iglesia de Llano, Corrientes); J D Aguirre, E M Morales, L E Ayala, M T De Sagastizabal, G Abreo (Hospital JR Vidal, Corrientes); A Uranga (Hospital Italiano de La Plata, La Plata); R de Lourdes Martin (Hospital LC Lagomaggiore, Mendoza); C Arias (Hospital JM Cullen, Santa Fe); R Abalos Gorostiaga, M Curioni, J Alvarado (Hospital Ramon Carrillo, Santiago del Estero); Bolivia—C Fuchtner, D Mostajo Flores (Instituto de Salud Reproductiva/UDIME, Santa Cruz); Brazil—D M Tonoli Tessari, J M Madi, D R Soares de Lorenzi, M do Carmo Mattana, C Brunstein (Hospital Geral, Caxias do Sul); A Trapani Jr (UFSC, Florianopolis); L Schmaltz, G Ribero de Souza, M E de Assis, I A Melo Melgaco (Hospital Materno Infantil, Goiania); R A Moreira de Sa, R Guerios Bornia (Maternidade Escola da UFRJ, Rio de Janeiro); Canada—N N Demianczuk, E Penttinen (Royal Alexandra Hospital, Edmonton); K Butt, K Hay, V Sandwich (Dr Everett Chalmers Regional Hospital, Fredericton); B A Armson, M Vincer, V Allen, C Fanning (IWK Health Centre, Halifax); R Kulkarni, J Laplante (North Bay General Hospital, North Bay); G D Carson, S Williams, S Holfeld (Regina General Hospital, Regina); F Olatunbosun, S Dalton, A Henry, J Haughian (Royal University Hospital, Saskatoon); J-M Moutquin, D Blouin, S Kocsis Bédard (CHUS Fleurimont, Sherbrooke); K Murphy, A Ohlsson, E Kelly, A Jordan, J Shapiro (Mount Sinai Hospital, Toronto); E Asztalos, J Barrett, H Cohen, L Andrews, H Owen (Sunnybrook Health Sciences Centre, Toronto); M-Fr Delisle, V Popovska, S Soanes (Children’s and Women’s Health Centre of BC, Vancouver); M E Helewa, D Kenny-Lodewyks (St Boniface General Hospital, Winnipeg); Chile—R Gomez, K Silva (Hospital Dr. Sotero del Rio, Puente Alto); J Figueroa Poblete, P Ferrand (Hospital Clinico San Borja Arriaran, Santiago); C Belmar, C Vera (Universidad Catolica, Santiago); China—Q F Su, W Gu, Z W Liu (Peace Maternity, Shanghai); Colombia—M Marrugo Flores, C Malabet Santoro (Universidad del Norte, Barranquilla); E I Ortiz, J Torres, A Rodriguez (CEMIYA, Cali); Denmark— L Hvidman, A Mouritzen, J Vikre-Jørgensen (Aarhus University Hospital, Aarhus); Germany—H Hopp, A Nonnenmacher, U Braig (CUB— Benjamin Franklin, Berlin); C Berg, G Bizjak, U Gembruch, V Schwarzer (Bonn University, Bonn); U B Hoyme, H-J Bittrich, B Oletzky, J Schneider (Helios Klinikum, Erfurt); B Hollwitz, K Oehler, F Dressler (MHH Hannover, Hannover); A Kubilay Ertan, J Hentschel, A Mack, W Schmidt (University of Saarland, Department Of obstetrics and gyneacology, Homburg/Saar); R Faber, H Stepan (Universitatsklinikum Leipzig, Leipzig); M Kuhnert, S Stiller (University Hospital Geissen and Marburg, Marburg); B Kuschel, K T M Schneider, A Zimmermann (TU Munchen, Munich); M Krause, H Gröbe, N Terzioglu (Klinikum Nurnberg Sud, Nurnberg); B Seelbach-Goebel, A Falkert (St Hedwig Hospital, Regensburg); K Mueller, H Voss (Dr Horst-Schmidt-Kliniken, Wiesbaden); Hungary—T Major, K Zoárd, T Bartha, P Bea, J Zsadányi (University of Debrecen, Debrecen); Israel—Z Nachum, M Peniakov (HaEmek Medical Centre, Afula); M Hallak, A Harlev (Soroka Medical Centre, Beer Sheva); L Harel, B Chayen, S Siev (Ma’ayney HaYeshua Medical Centre, Bnei Brak); I Samberg, L Wolff (Bnai Zion Medical Centre, Haifa); O Sadan, A Elyassi, C Baider, D Kohelet, A Golan (Edith Wolfson Medical Centre, Holon); D Mankuta, B Bar-Oz, D Combs (Hadassah Medical Centre— Ein Kerem, Jerusalem); D J D Rosen, H Y Kaneti, T Tzachi, J Zausmer

www.thelancet.com Vol 372 December 20/27, 2008

Articles

(Meir Medical Centre, Kfar-Saba); M Maman, T Perri, S Taitelboum (Rabin Medical Centre, Petach Tikva); M Simchen, G Shalev (Sheba Medical Centre, Ramat-Gan); M Goldinfeld, O Levine (Poriya Government Hospital, Tiberias); Jordan—M Y El-Zibdeh, L T Al-Faris, H A Ayyash (Islamic Hospital, Amman); Peru—P Saona Ugarte (Hospital Nacional Cayetano Heredia, Lima); Poland—K Preis, I Domzalska-Popadiuk, M Swiatkowska-Freund, I Janczewska (Medical University of Gdansk, Gdansk); J Wilczynski, M Krekora, M Kesiak, E Gulczyńska (Polish Mothers Memorial Hospital Research, Lodz); M Ropacka, Ma Madejczyk, J Rozycka, G H Breborowicz, M Szymankiewicz (University School of Medical Sciences, Poznan); Russia—N Borzova, L V Posiseeva (Research Institute, Ivanovo); Z Khodjaeva, E Vikhlyaeva (Research Centre of Obstetrics, Gynecology and Perinatology, Moscow); Spain—M Palacio, D Salvia, Fr Botet, M Massanes, M Lopez (Hospital Clinic-University of Barcelona, Barcelona); Switzerland—I Hösli, W Holzgreve, C A Voekt (University Women’s Hospital, Basel); F Belhia, P Hohlfeld, E Prince-dit-Clottu (CHU Vaudois, Lausanne); E Beinder, U von Mandach, J-C Fauchere (University Hospital, Zurich); Netherlands—F JME Roumen, M Pieters (Atrium Medical Centre, Heerlen); F Smits (Academisch Ziekenhuis Maastricht, Maastricht); United Kingdom—I I Bolaji, P Adiotomre (Diana, Princess of Wales Hospital, Grimsby); H Al-Taher, H Barnes (The Queen Elizabeth Hospital, King’s Lynn); R Abdul-Kadir, C Chi, V Van Someren (Royal Free Hospital, London); USA—S Dexter, R Samelson, M J Horgan, C Valentini (Albany Medical Center, Albany); S Pardanani, M Bebbington, C Chazotte (Montefiore Medical Center, Bronx); S Kilpatrick, Je L Drahos (University of Illinois Medical Center at Chicago, Chicago); L Saldana, Be Mount, B Warner, K Wedig (Bethesda North Hospital and Good Samaritan Hospital, Cincinnati); A Lysikiewicz (Winthrop-University Hospital, Mineola); F Bsat, J Fleming, A Lee, D Hoffman (Baystate Health System, Springfield). Data safety monitoring board: M Bracken, P Crowley, Al Donner, L Duley, J Tyson. Conflict of interest statement We declare that we have no conflict of interest. Acknowledgments We thank all participants in the MACS trial; Caroline Crowther and Ron Wapner for their collaboration and support; the members of our data safety monitoring board for their guidance; and all the staff at the Maternal, Infant and Reproductive Health Research Unit in Toronto for their dedication and hard work. MACS was supported by a grant from the Canadian Institutes for Health Research (grant number MCT 38142). The data coordinating centre was supported by grants from Sunnybrook Health Sciences Centre, Women’s College Hospital and the Department of Obstetrics and Gyneacology at the University of Toronto. Betamethasone and placebo were purchased from Schering-Plough Corporation (Madison, NJ, USA) and Eminent Services Corporation (Gaithersburg, MD, USA), respectively. References 1 Crowley P, Chalmers I, Keirse MJNC. The effects of corticosteroid administration before preterm delivery: an overview of the evidence from controlled trials. Br J Obstet Gynaecol 1990; 97: 11–25. 2 Liggins GC, Howie RN. A controlled trial of antepartum glucocorticoid treatment for prevention of the respiratory distress syndrome in premature infants. Pediatrics 1972; 50: 515–25. 3 Taeusch HW, Brown E, Torday JS, Nielsen HC. Magnitude and duration of lung response to dexamethasone in fetal sheep. Am J Obstet Gynecol 1981; 140: 452–55.

www.thelancet.com Vol 372 December 20/27, 2008

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18 19

20

21

22 23

Ballard PL, Ertsey R, Gonzales LW, Gonzales J. Transcriptional regulation of human pulmonary surfactant proteins SP-B and SP-C by glucocorticoids. Am J Respir Cell Mol Biol 1996; 14: 599–607. Quinlivan JA, Evans SF, Dunlop SA, Beazley LD, Newnham JP. Use of corticosteroids by Australian obstetricians—a survey of clinical practice. Aust N Z J Obstet Gynaecol 1998; 38: 1–7. Guinn DA, Atkinson MW, Sullivan L, et al. Single vs weekly courses of antenatal corticosteroids for women at risk of preterm delivery: a randomized controlled trial. JAMA 2001; 286: 1581–87. Crowther CA, Haslam RR, Hiller JE, Doyle LW, Robinson JS, for ACTORDS. Neonatal respiratory distress syndrome after repeat exposure to antenatal corticosteroids: a randomised controlled trial. Lancet 2006; 367: 1913–19. Wapner RJ, Sorokin Y, Thom EA, et al. Single vs weekly courses of antenatal corticosteroids: evaluation of safety and efficacy. Am J Obstet Gynecol 2006; 195: 633–42. Crowther CA, Harding JE. Repeat doses of prenatal corticosteroids for women at risk of preterm birth for preventing neonatal respiratory disease. Cochrane Database Syst Rev 2007; 3: CD003935. Kapoor A, Petropoulos S, Matthews SG. Fetal programming of the hypothalamic pituitary-adrenal (HPA) axis function and behavior by synthetic glucocorticoids. Brain Res Rev 2008; 57: 586–95. French NP, Hagan R, Evans SF, Mullan A, Newnham JP. Repeated antenatal corticosteroids: effects on cerebral palsy and childhood behavior. Am J Obstet Gynecol 2004; 190: 588–95. Aghajafari F, Murphy K, Matthews S, Ohlsson A, Amankwah K, Hannah M. Repeated doses of antenatal corticosteroids in animals: a systematic review. Am J Obstet Gynecol 2002; 186: 843–49. Peltoniemi OM, Kari A, Tammela O, et al. Randomized trial of a single repeat dose of prenatal betamethasone treatment in imminent preterm birth. Pediatrics 2007; 119: 290–98. Jobe AH, Wada N, Berry LM, Ikegami M, Ervin MG. Single and repetitive glucocorticoid exposure reduce fetal growth in sheep. Am J Obstet Gynecol 1998; 178: 880–85. Quinlivan JA, Archer MA, Dunlop SA, Evans SF, Beazley LD, Newnham JP. Fetal growth retardation, particularly within lymphoid organs, following repeated maternal injections of betamethasone in sheep. J Obstet Gynecol Res 1998; 24: 173–82. Halliday HL, Ehrenkranz RA, Doyle LW. Early postnatal (<96 hours) corticosteroids for preventing chronic lung disease in preterm infants. Cochrane Database Syst Rev 2003; 3: CD001146. Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weight less than 1500 grams. J Pediatr 1978; 92: 529–34. Walsh MC, Kliegman RM. Necrotizing enterocolitis: treatment based on staging criteria. Pediatr Clin North Am 1986; 33: 179–201. World Health Organization. Neonatal and perinatal mortality country, regional and global estimates. WHO, Geneva 2000; http://www.who. int/making_pregnancy_safer/publications/neonatal.pdf (accessed March 1, 2007). Kramer MS, Platt RW, Wen SW, et al. A new and improved population-based Canadian reference for birth weight for gestational age. Pediatrics 2001; 108: E35. Brown ER, Nielsen H, Torday JS, Tauesch HW. Reversible induction of surfactant production in fetal lambs treated with glucocorticoids. Pediatr Res 1979; 13: 491. Wapner RJ, Sorokin Y, Mele L, et al. Long-term outcomes after repeat doses of antenatal corticosteroids. N Engl J Med 2007; 357: 1190–98. Crowther CA, Doyle L, Haslam RR, et al. Outcomes at 2 years of age after repeat doses of antenatal corticosteroids. N Engl J Med 2007; 357: 1179–89.

2151