This randomized controlled trial investigated whether providing multiple courses of antenatal corticosteroids every 14 days reduced neonatal morbidity and mortality without negatively impacting fetal growth. 1858 women between 25-32 weeks gestation who remained at risk of preterm birth after an initial corticosteroid course were assigned to receive multiple corticosteroid courses or placebo until 33 weeks or delivery. The trial found no difference in neonatal outcomes between groups. Infants exposed to multiple corticosteroid courses weighed less and had smaller head circumferences at birth, suggesting the treatment negatively impacted fetal growth. The study concludes multiple corticosteroid courses every 14 days is not recommended due to lack of benefit and potential for decreased fetal growth.
1. 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 Lancet 2008; 372: 2143–51
death. Weekly doses given to women who remain undelivered after a single course may have benefits (less respiratory See Comment page 2094
morbidity) or cause harm (reduced growth in utero). We aimed to find out whether multiple courses of antenatal *Members listed at end of paper
corticosteroids would reduce neonatal morbidity and mortality without adversely affecting fetal growth. Department of Obstetrics and
Gynaecology, Mount Sinai
Methods 1858 women at 25–32 weeks’ gestation who remained undelivered 14–21 days after an initial course of Hospital, University of Toronto,
Toronto, ON, Canada
antenatal corticosteroids and continued to be at high risk of preterm birth were randomly assigned to multiple courses (K E Murphy MD); Department
of antenatal corticosteroids (n=937) or placebo (n=921), every 14 days until week 33 or delivery, whichever came first. of Obstetrics and Gynaecology,
The primary outcome was a composite of perinatal or neonatal mortality, severe respiratory distress syndrome, Sunnybrook Health Sciences
intraventricular haemorrhage (grade III or IV), periventricular leucomalacia, bronchopulmonary dysplasia, or Centre, University of Toronto,
Toronto, ON, Canada
necrotising enterocolitis. Analysis was by intention to treat. All patients and caregivers were unaware of the treatment (Prof M E Hannah MDCM);
given. This trial is registered as number ISRCTN2654148. Programme in Child Health
Evaluative Sciences, SickKids
Research Institute, Department
Findings Infants exposed to multiple courses of antenatal corticosteroids had similar morbidity and mortality to those
of Public Health Sciences,
exposed to placebo (150 [12·9%] vs 143 [12·5%]). Those receiving multiple doses of corticosteroids also weighed less at University of Toronto, Toronto,
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 ON, Canada (Prof A R Willan PhD);
had a smaller head circumference (31·1 cm vs 31·7 cm, p<0·001). Maternal, Infant and
Reproductive Health Research
Unit at the Women’s College
Interpretation Multiple courses of antenatal corticosteroids, every 14 days, do not improve preterm-birth outcomes, Research Institute, University of
and are associated with a decreased weight, length, and head circumference at birth. Therefore, this treatment Toronto, Toronto, ON, Canada
schedule is not recommended. (S A Hewson BA, P Guselle MSc);
Department of Paediatrics,
Mount Sinai Hospital, University
Funding Canadian Institutes of Health Research. of Toronto, Toronto, ON, Canada
(Prof A Ohlsson MD, E N Kelly MB);
Introduction Collaborative Trial of Repeat Doses of Steroids Departments of Physiology,
Obstetrics and Gynaecology and
Preterm birth is a worldwide health-care problem contri- (ACTORDS)7 enrolled 982 women and showed a benefit Medicine, University of Toronto,
buting greatly to neonatal morbidity and mortality. The of weekly courses of antenatal corticosteroids. Fewer Toronto, ON, Canada
administration of one course of antenatal corticosteroids infants in the treatment group than in the placebo group (Prof S G Matthews PhD);
to women who are at high risk of giving birth prematurely had respiratory distress syndrome (33% vs 41%; relative Department of Paediatrics,
McMaster University Medical
reduces the risk of neonatal mortality, respiratory distress risk [RR] 0·82 [95% CI 0·71–0·95], p=0·01) and severe Centre, Hamilton, ON, Canada
syndrome, and intraventricular haemorrhage.1 However, lung disease (12% vs 20%; 0·60 [0·46–0·79], p=0·0003). (Prof S Saigal MD); Department
women who receive one course may remain undelivered A Cochrane systematic review, which included the results of Newborn and Developmental
for weeks afterwards. Basic science and clinical research of these trials, suggested that weekly courses of antenatal Paediatrics, Sunnybrook Health
Sciences Centre, University of
have suggested that the benefits of one course might corticosteroids are associated with reduced occurrence Toronto, Toronto, ON, Canada
diminish over time. Thus, multiple courses of corti- [0·82 (0·72–0·93)] and severity [0·60 (0·48–0·75)] of (E Asztalos MD); Department of
costeroids every 7–14 days have been administered even neonatal lung disease, and serious infant morbidity [0·79 Obstetrics and Gynaecology,
before completion of randomised controlled trials.2–5 (0·67–0·93)].9 University of Calgary, Calgary,
AB, Canada (Prof S Ross PhD);
Several trials have investigated the short-term benefits However, multiple courses of antenatal corticosteroids Department of Obstetrics and
of weekly courses of antenatal corticosteroids versus may have adverse effects10–13 such as decreased fetal Gynaecology, BC Women’s
placebo in women who had already received one course growth.8,14,15 Such treatment might also have long-term Hospital, University of British
Columbia, Vancouver, BC,
of corticosteroids.6–8 Overall, initial small trials showed no adverse effects; indeed, adverse neurological outcomes
Canada (M-F Delisle MD);
benefit.6,8 However, the National Institutes of Child Health have been shown in follow-up studies of children given Department of Gynaecology
and Human Development (NICHD) Maternal Fetal dexamethasone after birth.16 and Obstetrics, Women’s and
Medicine Units Network trial8 showed a trend towards im- Our aim was to see if a less frequent intervention Children’s Hospital, State
University of New York at
proved composite outcome for infants who were exposed (a course every 14 days in our trial vs every 7 days in other
Buffalo, Buffalo, NY, USA
to weekly courses of antenatal corticosteroids born before steroid trials) would show short-term respiratory benefits, (Prof K Amankwah MD); Centre
32 weeks’ gestation (23% of infants on antenatal corti- and reduce to a minimum the risk of short-term and for Health Economics and
costeroids vs 39% on placebo, p=0·08). The Australasian long-term adverse effects. Policy Analysis,
www.thelancet.com Vol 372 December 20/27, 2008 2143
2. Articles
Department of Clinical Methods experienced a fetal death after 13 weeks’ gestation. MACS
Epidemiology and Participants received ethical approval from the University of Toronto,
Biostatistics, McMaster
University, Hamilton, ON,
Multiple courses of antenatal corticosteroids for preterm Mount Sinai Hospital, and from all collaborating
Canada (Prof A Gafni DSc); birth study (MACS) is an international, multicentre, institutions.
Department of Paediatrics and double-blind, randomised controlled trial. 1858 women
the Integrated Centre for Care between 25 and 32 weeks of gestation who remained Procedure and outcomes
Advancement through
Research (iCARE), University of
undelivered 14–21 days after an initial course of antenatal Randomisation was done with a 24-h telephone service
Alberta, Edmonton, AB, corticosteroids (either betamethasone or dexamethasone) after patient eligibility and baseline information were
Canada (Prof S K Lee MBBS); and and continued to be at high risk of preterm birth (table 1) recorded. A study number was assigned, corresponding
Department of Obstetrics and were enrolled in 80 centres in 20 countries. The study to a box of study medication at the participating centre.
Gynaecology, IWK Health
Centre, Dalhousie University,
protocol was explained and consenting participants were The study was stratified by gestational age and centre.
Halifax, NS, Canada randomly assigned to a study group after informed Women randomly assigned to the antenatal corti-
(Prof B A Armson MD) consent. costeroids group received two doses of 12 mg
Correspondence to: Women were not eligible if they had a contraindication betamethasone—a combination of 6 mg betamethasone
Dr Kellie E Murphy, Mount Sinai to corticosteroids, needed chronic doses of these drugs, sodium phosphate and 6 mg betamethasone sodium
Hospital, Maternal Fetal
Medicine, Department of
had evidence of chorioamnionitis, carried a fetus with a acetate (Celestone, Schering-Plough Corporation,
Obstetrics and Gynaecology, known lethal congenital abnormality, had an initial course Madison, NJ, USA)—intramuscularly 24 h apart. Women
Room 3726, Ontario Power of corticosteroids before 23 weeks’ gestation, or previously randomly assigned to placebo received similarly
Generation Building, participated in MACS. Women with a multiple pregnancy appearing intramuscular injections, containing a dilute
700 University Avenue,
Toronto, ON, M5R 1X5, Canada
were judged eligible for MACS if a fetus had died before concentration of aluminum monostearate (Eminent
kmurphy@mtsinai.on.ca 13 weeks’ gestational age; however, the dead fetus was not Services Corporation, Frederick, MD, USA). This
included as an outcome. Women were not eligible if they substance is used as a filler in many pharmaceutical
preparations and is inert.
Antenatal corticosteroids Placebo (N=918) 901 (96%) women in the treatment group and 875 (95%)
(N=935)
in the placebo group had an ultrasound within 21 days
Maternal age (years) 29·1 (6·23) 29·1 (6·18) before randomisation; 20 (2%) in the treatment group
Number of fetuses and 27 (3%) in the placebo group had an ultrasound more
1 737 (79%) 726 (79%) than 21 days before randomisation, and 10 (1%) in the
2 162 (17%) 158 (17%) treatment group and 14 (2%) in the placebo group had an
3 36 (4%) 34 (4%) ultrasound after randomisation.
Number of previous pregnancies Women who remained at increased risk of preterm
0 263 (28%) 252 (27%) birth after the first course of study medication—after
1–4 577 (62%) 571 (62%) study enrolment and after their first course of either
>4 95 (10%) 91 (10%) betamethasone or placebo—continued to receive two
Previous second-trimester abortion (14–19 weeks) 65 (7%) 69 (8%) doses of 12 mg betamethasone or placebo, 24 h apart,
Previous preterm delivery (20–<37 weeks) 322 (34%) 334 (36%) every 14 days until 33 weeks’ gestation or birth, whichever
History of a previous pregnancy with intrauterine 53 (6%) 60 (7%) happened first. For women who had preterm rupture of
growth restriction the membranes, the recommendation was that investi-
Method of gestational age calculation* gators would stop the study medication at 32 weeks’
Clinical only 31 (3%) 41 (4%) gestation. All patients and caregivers were unaware of the
Ultrasound±clinical 896 (96%) 874 (95%) treatment given. Women were asked to complete a
Estimated fetal weight by ultrasound at 1203 (12·9) 1211 (12·8) structured questionnaire 3 months after giving birth to
randomisation (g)†
assess the occurrence of postpartum depression and other
Mean gestational age at randomisation (weeks) 29·3 (2·0) 29·4 (2·0)
maternal side-effects.
Gestational age at randomisation
The primary outcome was perinatal or neonatal mor-
<25 weeks 1 (<1%) 0 tality, or neonatal morbidity. Perinatal or neonatal mortality
25–27 weeks 256 (27%) 255 (28%) was defined as stillbirth or neonatal death during the first
28–32 weeks 678 (73%) 661 (72%) 28 days of life or before hospital discharge, whichever
>32 weeks 0 2 (<1%) happened later. Clinically significant neonatal morbidity
Fetal anomalies 2 (<1%) 5 (<1%) was defined as one or more of the following: (i) severe
Agenesis of right kidney 1 (<1%) ·· respiratory distress syndrome—ie, needing assisted
Cardiac abnormalities 1 (<1%) 2 (<1%) ventilation via endotracheal tube and supplemental oxygen
Cleft lip and palate ·· 1 (<1%) both within the first 24 h of life and for 24 h or more, and
Diaphragmatic hernia ·· 1 (<1%) either a radiographic scan compatible with respiratory
Hydrocephaly ·· 1 (<1%) distress syndrome or surfactant given between the first
(Continues on next page) 2–24 h of life; (ii) bronchopulmonary dysplasia—ie,
needing oxygen at a postmenstrual age of 36 completed
2144 www.thelancet.com Vol 372 December 20/27, 2008
3. Articles
weeks and radiographic scan compatible with broncho-
Antenatal corticosteroids Placebo (N=918)
pulmonary dysplasia; (iii) intraventricular haemorrhage (N=935)
grade III or IV, diagnosed by cranial ultrasound with
(Continued from previous page)
Papile and colleagues’ classification;17 (iv) cystic
Prestudy course of ACS
periventricular leucomalacia—ie, periventricular cystic
Betamethasone 810 (87%) 786 (86%)
changes in the white matter, excluding subependymal and
choroid plexus cysts, diagnosed by cranial ultrasound; Dexamethasone 125 (13%) 131 (14%)
(v) necrotising enterocolitis—either perforation of intes- Time from first dose of prestudy corticosteroids to 15·0 15·0
randomisation (days)‡ (14·0, 21·0) (13·0, 21·0)
tine, pneumatosis intestinalis, or air in the portal vein,
Medical and obstetrical problems at enrolment
diagnosed by radiographic scan or at surgery.18
Uterine contractions within previous week 520 (56%) 522 (57%)
For infants with a birthweight of less than 1500 g,
Short cervical length or cervical dilation 457 (49%) 450 (49%)
participating centres were encouraged to arrange at least
two cranial ultrasounds before hospital discharge to look Antepartum vaginal bleeding 129 (14%) 130 (14%)
for evidence of intraventricular haemorrhage and cystic Preterm rupture of membranes 149 (16%) 142 (15%)
periventricular leucomalacia. Intrauterine growth restriction 85 (9%) 74 (8%)
Other neonatal outcomes were weight, length, and Pre-eclampsia 43 (5%) 53 (6%)
head circumference at birth, neonatal infection, retino- Smoking 108 (12%) 93 (10%)
pathy of prematurity, stay in neonatal intensive care Substance abuse 7 (<1%) 5 (<1%)
unit, use of ventilation with intubation, and patent Hypertension needing treatment 66 (7%) 70 (8%)
ductus arteriosus needing pharmacological treatment Maternal diabetes 50 (5%) 39 (4%)
or surgery. Neonatal infection was defined as clinical Controlled by diet only 31 (62%) 24 (62%)
sign of infection, and one or more of the following: a Insulin dependent 19 (38%) 15 (38%)
positive culture of blood or cerebrospinal fluid, Maternal treatments during previous 2 weeks
a gram-positive stain of cerebrospinal fluid, or chest Antibiotics 331 (35%) 292 (32%)
radiographic findings compatible with pneumonia. Tocolytics 465 (50%) 439 (48%)
Maternal outcomes were clinical chorioamnionitis— Betamimetics (iv) 266 (57%) 256 (58%)
defined as maternal temperature of 38°C or more before Magnesium sulphate (iv) 81 (17%) 89 (20%)
delivery, and one or more of the following signs: maternal Indomethacin (po or pr) 61 (13%) 50 (11%)
tachycardia (120 beats per minute [bpm] or more), white Calcium channel blocker (po) 150 (32%) 135 (31%)
blood-cell count of 20 000 μL or more, fetal tachycardia Other§ 99 (21%) 107 (24%)
(>160 bpm), uterine tenderness, or foul smelling Principal reason(s) for study participation
amniotic fluid—and maternal infection after delivery— Signs or symptoms of preterm labour 773 (83%) 777 (85%)
defined as one or more of the following signs: endometritis Fetal anomalies or pathologies 115 (12%) 103 (11%)
(ie, postpartum maternal temperature of 38°C or more Maternal medical condition 199 (21%) 190 (21%)
and tender fundus without other source of infection); Multiple pregnancy 191 (20%) 179 (19%)
pneumonia (ie, maternal temperature of 38°C or more History of obstetrical complications 287 (31%) 280 (31%)
and signs of pneumonia on radiographic scan); wound National perinatal mortality rate
infection or breakdown; pyelonephritis (ie, maternal ≤10 in 1000 623 (67%) 612 (67%)
temperature of 38°C or more, positive urine culture, and >10–20 in 1000 239 (26%) 238 (26%)
costal vertebral angle tenderness); or maternal sepsis >20 in 1000 73 (8%) 68 (7%)
(ie, maternal temperature of 38°C or more and a positive
blood culture). 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
We obtained information about drug administration and
was used to adjust for the interdependency of multiple births. Data are mean (SE). ‡Data are median (5th,
co-interventions (table 2). After the first 600 women were 95th centile). §Oral betamimetics, atosiban, theophylline, terbutaline, nitroglycerine, hexoprenaline, oral magnesium
randomised, random samples of study medication were sulphate, progesterone, spasmolytics, with the addition of monothioglycerol and diazepam in the placebo group.
sent to a sham site for testing. The study drug was analysed
Table 1: Baseline characteristics of enrolled women
to ensure that study numbers corresponding to placebo
and betamethasone were correct. Overall, 160 vials were
tested and the assigned study number allocation corre- corticosteroids reduced the risk of respiratory distress
sponded with the actual study medication 100% of the syndrome from 12% to 8%.
times. We did an interim analysis of all data on the first
800 patients enrolled after complete data were obtained.
Statistical analysis We presented the results to an independent data safety
We calculated a sample size of 1900 women (950 per monitoring board. The a-priori stopping rule was finding
group) to have 80% probability of achieving a significant a higher rate of the primary outcome in the treatment
difference between the two groups with a two-tailed, group than in the placebo group with a one-tailed, type I
type I error of 0·05, if multiple courses of antenatal error of 0·002.
www.thelancet.com Vol 372 December 20/27, 2008 2145
4. Articles
Antenatal Placebo p
Final analysis of the outcome data was by inten-
corticosteroids (N=918) value tion to treat. We used a logistic regression model, with a
(N=935) random effect for multiple pregnancy to adjust for the
Number of courses of study drug* dependence of observation within the pregnancy, to
0 5 (<1%) 5 (<1%) estimate the adjusted odds ratios (ORs) and measure the
1 385 (41%) 365 (40%) 95% CI for the comparison of the two groups in relation
2 305 (33%) 273 (30%) to the primary and other outcomes. Significance for the
3 150 (16%) 169 (18%) primary outcome was p<0·05 and for the other outcomes
4 90 (10%) 104 (11%) was p<0·01.
Fully compliant with study drug† 747 (80%) 722 (79%) We analysed interactions between treatment group and
Partly compliant with study drug 181 (19%) 189 (21%) selected baseline variables for the primary outcome.
Non-compliant, incorrect, or no study drug given 7 (<1%) 6 (<1%) Although not planned a priori, the rate of the primary
Gestational age at birth (weeks) 34·5 (3·6) 34·9 (3·6) outcome was calculated for the two groups in the
Gestational age at birth‡ subgroups of infants born at 32 weeks’ gestation or earlier,
<28 weeks 39 (4%) 27 (3%) and those born within 7 days of repeated study drug
28–32 weeks 280 (30%) 254 (28%)
administration. We did the first subanalysis because
33–36 weeks 338 (36%) 319 (35%)
previous trials suggested that multiple courses of antenatal
≥37 weeks 278 (30%) 318 (35%)
corticosteroids might benefit infants born before 32 weeks’
Time of delivery after repeated study drug exposure§
gestational age, and the second in an effort to compare
<48 h 92 (10%) 91 (10%)
findings of MACS with previously published trials of
weekly courses of antenatal corticosteroids.6–8
48 h to <7 days 153 (16%) 131 (14%)
≥7 days 685 (73%) 689 (75%)
Role of funding source
Worsening of pre-existing or onset of diabetes after 32 (3%) 28 (3%)
randomisation§ MACS was funded by the Canadian Institutes of Health
Women treated with insulin 15 (47%) 11 (39%) Research (CIHR). CIHR had no role in the design,
Worsening of pre-existing or onset of hypertension 90 (10%) 79 (9%) management, data collection, analysis, or interpretation
after randomisation§ of the data. CIHR had no role in the writing of the
Women treated with anti-hypertensive drugs 73 (81%) 62 (78%) manuscript or in the decision to submit for publication.
Tocolytics given after randomisation 323 (35%) 339 (37%) All authors had full access to the data. KEM had final
Betamimetics (iv) 168 (52%) 172 (51%) responsibility for the decision to submit the paper for
MgSO4 (iv) 71 (22%) 79 (23%) publication.
Indomethacin (po or pr) 27 (8%) 32 (9%)
Calcium channel blocker (po) 133 (41%) 136 (40%) Results
Other¶ 101 (31%) 108 (32%) The figure shows the trial profile. Eight women (five
Corticosteroids administered in addition to the study 13 (1%) 8 (<1%) from the antenatal corticosteroid group and 3 from the
drug after randomisation§ placebo group), whose one or more fetuses from a
Betamethasone 9 (69%) 5 (63%) multiple gestation died in utero after 13 weeks but before
Dexamethasone 2 (15%) 3 (38%) randomisation, were randomly assigned. Live fetuses
Prednisone 2 (15%) 0 from these pregnancies were excluded from the analysis
Reasons for giving corticosteroids in addition to the 13 (1%) 8 (<1%) (five from the antenatal corticosteroid group and four
study drug after randomisation from the placebo group). After intention-to-treat analysis,
Staff error 3 (23%) 2 (25%) data from the eight women were obtained and included
Clinical decision 4 (31%) 4 (50%) in the analysis. Table 1 shows the baseline characteristics
Maternal medical indication|| 4 (31%) 2 (25%) of the women recruited for the trial. Randomisation
Not stated 2 (15%) 0 started on April 9, 2001, and finished on Aug 31, 2006.
Antibiotics after randomisation, and before or at delivery 505 (54%) 503 (55%) Countries with a national perinatal mortality rate
Before labour 251 (50%) 247 (49%) of 10 in 1000 or less were Canada, Chile, Denmark,
During labour or at delivery 375 (74%) 363 (72%) Germany, Hungary, Israel, Netherlands, Poland, Spain,
Duration of rupture of membranes (h)** 0·6 (0, 1009·3) 0·8 (0, 1069·0) Switzerland, UK, and USA. Countries with a national
Clinical chorioamnionitis§ 22 (2%) 20 (2%) 0·80 perinatal mortality rate of 10–20 in 1000 were Argentina,
Method of delivery Brazil, and Peru; and countries with a national perinatal
Vaginal delivery 396 (42%) 415 (45%) mortality rate of more than 20 in 1000 were Bolivia,
Caesarean section 537 (57%) 501 (55%) China, Colombia, Jordan, and Russia.19
Unknown 2 (<1%) 2 (<1%) The study drug was given according to the protocol.
Antibiotics after delivery§ 276 (30%) 255 (28%) 1469 women were fully compliant with study drug. Partial
(Continues on next page) compliance was defined as first course of study drug
being given less than 14 or more than 21 days after initial
2146 www.thelancet.com Vol 372 December 20/27, 2008
5. Articles
treatment (n=31 in treatment group and n=36 in placebo
Antenatal Placebo p
group); one or more course of study drug being given corticosteroids (N=918) value
less than 14 days apart (n=28 in treatment group and (N=935)
n=17 in placebo group); one or more course of study drug (Continued from previous page)
being given more than 14 days apart (n=67 in treatment
Maternal infection after delivery§ 34 (4%) 25 (3%) 0·26
group and n=70 in placebo group); study drug stopped
Endometritis 21 (62%) 9 (36%)
more than 14 days before delivery or before 33 weeks of
Pneumonia 3 (9%) 2 (8%)
gestational age (in women with intact membranes) or
Pyelonephritis 4 (12%) 4 (16%)
before 32 weeks (in women with preterm rupture of the
Sepsis 4 (12%) 4 (16%)
membranes), whichever came first, secondary to patient
Wound infection resulting in wound breakdown 7 (21%) 10 (40%)
or physician request, or other reasons including patient
Serious maternal complications before discharge§†† 29 (3%) 20 (2%)
did not return to hospital or subsequent course date
Duration in hospital after delivery (h)** 78·8 (25·8, 236·1) 76·8 (25·4, 260·0)
miscalculated in the antenatal corticosteroid group, and
patients did not return to the hospital, subsequent course Data are n (%) or mean (SD), unless otherwise stated. iv=intravenously. po=orally. pr=rectally *Number unknown for
date miscalculated, rash, or vaginal infection in the 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
placebo group (n=72 in the treatment group and n=78 in
group; percentages are calculated on the data available. ¶Oral betamimetics, atosiban, theophylline, terbutaline,
the placebo group). 370 women were partly compliant. nitroglycerine, betamimetics (route not specified) hexoprenaline, oral magnesium sulphate, spasmolytics, diazepam, with
Non-compliance was defined as incorrect study drug the addition of methyldopa and progesterone in the treatment group. Some patients received more than one tocolytic
given (n=2 in the treatment group and n=1 in the placebo 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.
group) or no study drug given (n=5 in the treatment **Data are median (5th, 95th centile). ††Serious maternal complications were: postpartum haemorrhage, placenta
group and n=5 in the placebo group). 13 women were accreta, or hysterectomy (n=15); pregnancy induced hypertension or haemolysis; high concentration of liver enzymes, low
non-compliant. Of those women who stopped the study 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,
drug early, 110 (73%) stopped after 31 weeks’ gestational placenta accreta, or hysterectomy (n=10); pregnancy induced hypertension or haemolysis, high concentration of liver
age, whereas the remaining 40 (27%) stopped before enzymes, low platelet number, and pre-eclampsia (n=1), thromboembolic and vascular event (n=1); anaphylactic shock
31 weeks’ gestational age. One participant’s treatment in (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.
the antenatal corticosteroid group was unmasked at the
request of her private physician; however, she completed Table 2: Maternal treatments and outcomes
the trial, and her and her infant’s data were obtained.
Gestational age was confirmed by ultrasound examin-
ation for most women in the study (96% and 95% in the 1858 pregnant women randomised
treatment and placebo groups, respectively).
During the trial, eight serious adverse events were
reported, two from the treatment group (neonate with a
937 women allocated to repeated ACS 921 women allocated to placebo
respiratory syncitial virus infection in a woman with (1171 fetuses) (1147 fetuses)
diabetes [n=1] and neonate admitted to intensive care in a
woman with pre-eclampsia [n=1]), and six from the 2 women (2 fetuses) 3 women (3 fetuses)
placebo group (temporary loss of consciousness of the Lost to follow-up Lost to follow-up
mother [n=1], maternal and neonatal Escherichia coli
infection followed by neonatal death [n=1], neonatal 935 had maternal outcomes analysed 918 had maternal outcomes analysed
thrombocytopenia [n=1], neonatal subarachnoid haemor-
rhage [n=1], maternal Bell’s palsy [n=1], and neonatal
5 fetuses/infants 4 fetuses/infants
lung aspiration and neonatal death [n=1]). excluded* excluded*
The composite primary outcome of perinatal or
neonatal mortality, severe respiratory distress syndrome, 1164 infants had perinatal 1140 infants had perinatal
bronchopulmonary dysplasia, intraventricular haem- outcomes analysed outcomes analysed
orrhage (grade III or IV), periventricular leucomalacia,
or necrotising enterocolitis did not substantially differ Figure: Trial profile
between treatment and placebo groups (p=0·83) (table 3). ACS=antenatal corticosteroid.*Stillbirths within multiple pregnancies that took place before randomisation.
For the primary outcome, we assessed interactions be-
tween treatment group and gestational age at placebo (p=0·0026), were shorter (p<0·001), and had a
randomisation, preterm rupture of the membranes, and smaller head circumference (p<0·001) (table 3). Neonatal
single versus multiple pregnancies. We identified no respiratory and other outcomes were comparable between
significant interactions, which suggested that the risk of groups (table 3). Rates of the composite primary outcome
primary outcomes was similar for the treatment and the and components were similar between the groups for
placebo groups. infants who were born before 32 weeks and for those born
At birth, infants who received multiple courses of ante- within 7 days of repeated study drug administration
natal corticosteroids weighed less than those who received (table 4).
www.thelancet.com Vol 372 December 20/27, 2008 2147
6. Articles
Antenatal corticosteroids Placebo Mean difference (95% CI) p value
Total number of infants 1164 1140
Composite primary outcome*† 150 (13%) 143 (13%) 1·04 (0·77 to 1 ·39) 0·83
Singletons 88/737 (12%) 83/726 (11%)
Multiples 62/427 (15%) 60/414 (15%)
Stillbirth or neonatal death ≤28 days after birth or before 43 (4%) 40 (4%) 1·08 (0·67 to 1·66) 0·82
discharge, whichever happened later
Number of surviving infants‡ 1121 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
NEC 10 (<1%) 12 (1%) 1·03 (0·38 to 2 · 29) 0·87
Total number of infants 1164 1140
Male 616 (53%) 598 (53%)
Female§ 546 (47%) 540 (48%)
Birthweight (g) 2216 (28·3) 2330 (28·7) –113·1 (37·3) (–187·0 to –41·17) 0·0026
Length at birth (cm) 44·5 (0·2) 45·4 (0·2) –0·9 (0·25) (–1·34 to –0·37) <0·001
Mean head circumference (cm) 31·1 (0·1) 31·7 (0·1) –0·6 (0·15) (–0·90 to –0·32) <0·001
Birthweight less than 10th centile for gestational age20§ 196 (17%) 158 (14%)
Birthweight less than 3rd centile for gestational age20§ 64 (6%) 59 (5%)
Number of surviving infants‡ 1121 1100
Cord blood pH<7·0§¶ 3 (<1%) 4 (<1%)
Apgar score <7 at 5 mins after birth|| 40 (4%) 49 (5%)
Major life-threatening anomaly§ 10 (<1%) 8 (<1%)
Patent ductus arteriosus 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 .. 1 (<1%)
Receiving supplemental oxygen after initial resuscitation 410 (37%) 427 (39%)
Duration of supplemental oxygen (h)**†† 57·6 (1·5, 1696·2) 68·1 (2·1, 1091·0)
Intubation and ventilation via endotracheal tube 175 (16%) 204 (19%) 0·84 (0·63 to 1·09) 0·19
Duration of endotracheal ventilation (h)**†† 68·3 (5·0, 1123·8) 48·4 (3·5, 624·3)
Other ventilatory support without intubation 274 (24%) 291 (27%)
Duration of other ventilatory support (h)**†† 57·6 (1·1, 1117·0) 68·3 (4·0, 995·8)
Surfactant given 122 (11%) 141 (13%)
(Continues on next page)
Discussion women receiving placebo. These findings differ from
In MACS, after an initial course of antenatal corti- those of the Australian trial ACTORDS7 in which women
costeroids, infants born to women who received multiple who received weekly courses of treatment showed
courses of treatment every 14 days had similar risk of short-term neonatal benefits over women receiving
morbidity and mortality to those who were born to placebo. In MACS, the lack of improvement in respiratory
2148 www.thelancet.com Vol 372 December 20/27, 2008
7. Articles
Antenatal corticosteroids Placebo Mean difference (95% CI) p value
(Continued from previous page)
Any ventilatory support 449 (40%) 495 (45%) 0·96 (0·76 to 1·20) 0·71
Postnatal corticosteroids given to the infant 15 (1%) 8 (<1%)
Neonatal infection‡‡ 105 (9%) 91 (8%) 1·23 (0·88 to 1 67) 0·24
Sepsis 67 (64%) 46 (51%)
Meningitis 7 (7%) 2 (2%)
Pneumonia 45 (43%) 55 (60%)
Seizure before discharge 8 (<1%) 10 (<1%)
Patent ductus arteriosus needing treatment 44 (4%) 42 (4%) 1·24 (0·74 to 1·98) 0·46
Treatment with indomethacin 37 (84%) 28 (67%)
Treatment with ligation 8 (18%) 12 (29%)
Treatment with other 6 (14%) 6 (14%)
Retinopathy of prematurity§§ 46 (4%) 34 (3%) 1·04 (0·55 to 1·81) 0·99
Admitted to neonatal intensive care unit 465 (42%) 464 (42%) 1·03 (0·85 to 1·24) 0·80
Length of stay in neonatal intensive care unit (h)†† 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
Antenatal Placebo
morbidity might be due to the diminuishing effect of corticosteroids
antenatal corticosteroids on fetal pulmonary type II
Number of infants born at <32 weeks‘ gestational age 321 282
pneumocytes after 7 days. This hypothesis is supported
Death or serious neonatal morbidity: composite primary outcome* 131 (41%) 110 (39%)
by laboratory evidence, which suggests that the induction
Stillbirth or neonatal death ≤28 days after birth or before discharge, 37 (12%) 33 (12%)
of surfactant in the fetal lung is reversible.4,21 However,
whichever happened later
we showed no benefit of multiple courses in infants
Number of surviving† infants born <32 weeks‘ gestational age 284 249
delivered within 7 days of repeated study drug exposure,
Severe RDS 79 (28%) 57 (23%)
and thus our data do not support this hypothesis. We are
BPD 19 (7%) 10 (4%)
therefore uncertain as to why the results of MACS
IVH (grade III or IV) 5 (2%) 7 (3%)
differed from those of the ACTORDS trial.
Cystic PVL 6 (2%) 9 (4%)
In MACS, infants exposed to multiple courses of ante-
NEC 8 (3%) 9 (4%)
natal corticosteroids, every 14 days, were significantly
Number of infants born <7 days after last study drug exposure 333 290
smaller at birth than those in the placebo group. They
Death or serious neonatal morbidity: composite primary outcome* 97 (29%) 84 (29%)
weighed on average 113 g less, were 0·9 cm shorter, and
Stillbirth or neonatal death ≤28 days after birth or before discharge, 25 (8%) 28 (10%)
their head circumferences were 0·6 cm smaller. These
whichever happened later
findings are concerning and are consistent with those
Number of surviving† infants born <7 days after last study drug exposure 308 262
from randomised controlled trials of weekly courses of
Severe RDS 57 (19%) 43 (16%)
antenatal corticosteroids. The NICHD trial8 showed that
BPD 16 (5%) 9 (3%)
more infants in the treatment group than in the placebo
IVH (grade III or IV) 5 (2%) 5 (2%)
group had a birthweight under the 10th percentile (24% vs
Cystic PVL 6 (2%) 6 (2%)
15%, p=0·02), and the mean birthweight of infants
NEC 7 (2%) 6 (2%)
exposed to four or more courses of antenatal corticosteroids
was significantly lower in the treatment group than in the Data are n (%). ACS=antenatal corticosteroids. RDS=respiratory distress syndrome. BPD=bronchopulmonary dysplasia.
placebo group (2400 g vs 2561 g, p=0·01). In the ACTORDS IVH=intraventricular haemorrhage. PVL=periventricular leukomalacia. NEC=necrotising enterocolitis. *One or more of
the following: neonatal mortality, severe respiratory distress syndrome, bronchopulmonary dysplasia, intraventricular
trial,7 mean Z scores (SD) for weight (–0·40 [1·05]
haemorrage (grade III or IV), cystic periventricular leukomalacia, and necrotising enterocolitis. †Surviving infants
antenatal corticosteroids vs –0·27 [1·14] placebo, p=0·04) defined as alive at 28 days or surviving hospital discharge, whichever happened later.
and head circumference (–0·30 [1·22] antenatal
Table 4: Primary neonatal outcome in infants born before 32 weeks’ gestation and within 7 days of
corticosteroids vs –0·14 [1·28] placebo, p=0·03) were
repeated study drug administration
lower in the treatment group than in the placebo group at
www.thelancet.com Vol 372 December 20/27, 2008 2149
8. Articles
birth. An explanation for less effect on fetal growth in member of the steering committee on the MACS trial, and assisted with
ACTORDS was that women in this trial received less methodological and implementation issues. AG participated in the
design and monitoring of the MACS study, interpretation of results, and
treatment than those in other trials. Although most reviewing and commenting on drafts during the writing of the
women in the ACTORDS and MACS trials received only manuscript, and had seen and approved the final version of the
one or two courses (66% for ACTORDS and 72% for manuscript. SKL participated in data interpretation, reviewing of the
MACS), in the NICHD trial 63% of women received four paper, and had seen and approved the final version of the manuscript.
or more courses. A repeated course of antenatal MACS Collaborative Group
corticosteroids in the ACTORDS trial consisted of only 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,
one dose of betamethasone at each dosing time, whereas S K Lee, S G Matthews, A Ohlsson, S Ross, J Rovet, S Saigal, R Sananes,
in the other trials (Guinn, NICHD, and MACS) a repeated I Schmid, A R Willan.
course consisted of two doses of betamethasone 24 h Collaborators: Argentina—L Kwiatkowski, S M Tortorella (Hospital Fiorito,
Avellaneda); M S Bertin, J L Castaldi, C Deguer, M Klun, C Besegato
apart.6,8
(Hospital Penna, Bahia Blanca); G Izbizky, M C Vaneri, C A Fustinana,
The clinical relevance of decreased fetal size due to L Otano (Hospital Italiano, Buenos Aires); M S F Palermo, E Javier Murua,
multiple courses of antenatal corticosteroids is not clear. D Montes Valera, H Sampietro, A Monaco (Hospital Posadas, Buenos
Long-term follow-up of these infants is therefore im- Aires); R Savransky, A Dunaiewsky (Hospital Ramos Mejia, Buenos Aires);
M N Basualdo, E Andina, I Di Marco (Hospital Sarda, Buenos Aires);
portant. Children of the NICHD study were followed up
M Rivero, M C Feu, S Garcia (Hospital Angela Iglesia de Llano,
to 2–3 years of age. Physical and neurocognitive measures Corrientes); J D Aguirre, E M Morales, L E Ayala, M T De Sagastizabal,
did not differ between groups, but a slight increased risk G Abreo (Hospital JR Vidal, Corrientes); A Uranga (Hospital Italiano de La
of cerebral palsy was seen in those exposed to weekly Plata, La Plata); R de Lourdes Martin (Hospital LC Lagomaggiore,
Mendoza); C Arias (Hospital JM Cullen, Santa Fe); R Abalos Gorostiaga,
courses of treatment (6 [3%] vs 1 [<1%] placebo).22 In the
M Curioni, J Alvarado (Hospital Ramon Carrillo, Santiago del Estero);
ACTORDS trial, the rate of survival without major Bolivia—C Fuchtner, D Mostajo Flores (Instituto de Salud
disabilities was similar between the treatment and placebo Reproductiva/UDIME, Santa Cruz); Brazil—D M Tonoli Tessari, J M Madi,
groups at 2 years of age. However, children exposed to D R Soares de Lorenzi, M do Carmo Mattana, C Brunstein (Hospital Geral,
Caxias do Sul); A Trapani Jr (UFSC, Florianopolis); L Schmaltz,
weekly courses of antenatal corticosteroids (n=31) were G Ribero de Souza, M E de Assis, I A Melo Melgaco (Hospital Materno
more likely to warrant an assessment for attention Infantil, Goiania); R A Moreira de Sa, R Guerios Bornia (Maternidade
deficits than were controls (n=17; p=0·04).23 These results Escola da UFRJ, Rio de Janeiro); Canada—N N Demianczuk, E Penttinen
are concerning and caution is needed in the use of weekly (Royal Alexandra Hospital, Edmonton); K Butt, K Hay, V Sandwich
(Dr Everett Chalmers Regional Hospital, Fredericton); B A Armson,
courses of antenatal corticosteroids. 18–24-month M Vincer, V Allen, C Fanning (IWK Health Centre, Halifax); R Kulkarni,
follow-up assessments for children in MACS and 5-year J Laplante (North Bay General Hospital, North Bay); G D Carson,
follow-up assessments for children in the MACS and S Williams, S Holfeld (Regina General Hospital, Regina); F Olatunbosun,
ACTORDS trials are continuing. S Dalton, A Henry, J Haughian (Royal University Hospital, Saskatoon);
J-M Moutquin, D Blouin, S Kocsis Bédard (CHUS Fleurimont,
Similar to the other trials, 32% of women in MACS gave Sherbrooke); K Murphy, A Ohlsson, E Kelly, A Jordan, J Shapiro (Mount
birth at term. Their infants did not need antenatal Sinai Hospital, Toronto); E Asztalos, J Barrett, H Cohen, L Andrews,
corticosteroids for pulmonary maturation, and therefore H Owen (Sunnybrook Health Sciences Centre, Toronto); M-Fr Delisle,
they were potentially negatively affected by study V Popovska, S Soanes (Children’s and Women’s Health Centre of BC,
Vancouver); M E Helewa, D Kenny-Lodewyks (St Boniface General
medication. Furthermore, they were exposed to more Hospital, Winnipeg); Chile—R Gomez, K Silva (Hospital Dr. Sotero del
courses of treatment because they remained in utero Rio, Puente Alto); J Figueroa Poblete, P Ferrand (Hospital Clinico
longer than preterm infants. The mean number of courses San Borja Arriaran, Santiago); C Belmar, C Vera (Universidad Catolica,
of drug was 2·2 for infants who were born at term and Santiago); China—Q F Su, W Gu, Z W Liu (Peace Maternity, Shanghai);
Colombia—M Marrugo Flores, C Malabet Santoro (Universidad del Norte,
1·8 for those born before 37 weeks of gestation. Barranquilla); E I Ortiz, J Torres, A Rodriguez (CEMIYA, Cali); Denmark—
Overall, multiple courses of antenatal corticosteroids, L Hvidman, A Mouritzen, J Vikre-Jørgensen (Aarhus University Hospital,
given every 14 days, are associated with decreased Aarhus); Germany—H Hopp, A Nonnenmacher, U Braig (CUB—
growth in utero and no neonatal benefits compared Benjamin Franklin, Berlin); C Berg, G Bizjak, U Gembruch, V Schwarzer
(Bonn University, Bonn); U B Hoyme, H-J Bittrich, B Oletzky, J Schneider
with one course of antenatal corticosteroids. Therefore, (Helios Klinikum, Erfurt); B Hollwitz, K Oehler, F Dressler (MHH
in women who remain at increased risk of preterm Hannover, Hannover); A Kubilay Ertan, J Hentschel, A Mack, W Schmidt
birth after receiving an initial course of antenatal (University of Saarland, Department Of obstetrics and gyneacology,
corticosteroids, multiple courses every 14 days are not Homburg/Saar); R Faber, H Stepan (Universitatsklinikum Leipzig,
Leipzig); M Kuhnert, S Stiller (University Hospital Geissen and Marburg,
recommended. Marburg); B Kuschel, K T M Schneider, A Zimmermann (TU Munchen,
Contributors Munich); M Krause, H Gröbe, N Terzioglu (Klinikum Nurnberg Sud,
KEM, MEH, ARW, SAH, AO, SGM, SS, EA, SR, KA, and Nurnberg); B Seelbach-Goebel, A Falkert (St Hedwig Hospital,
BAA participated in the design, methodological issues, implementation, Regensburg); K Mueller, H Voss (Dr Horst-Schmidt-Kliniken, Wiesbaden);
conduct, monitoring, analysis, interpretation of the study, and have Hungary—T Major, K Zoárd, T Bartha, P Bea, J Zsadányi (University of
written, seen, and approved the final version of the manuscript. ENK Debrecen, Debrecen); Israel—Z Nachum, M Peniakov (HaEmek Medical
participated in the original discussions regarding the development and Centre, Afula); M Hallak, A Harlev (Soroka Medical Centre, Beer Sheva);
design of the MACS study, took part in steering committee meetings, L Harel, B Chayen, S Siev (Ma’ayney HaYeshua Medical Centre, Bnei
and had been involved in assessments of outcome at 2 years of the Brak); I Samberg, L Wolff (Bnai Zion Medical Centre, Haifa); O Sadan,
children recruited at Mount Sinai Hospital. M-FD participated in the A Elyassi, C Baider, D Kohelet, A Golan (Edith Wolfson Medical Centre,
implementation and recruitment of Canadian centres, and had seen and Holon); D Mankuta, B Bar-Oz, D Combs (Hadassah Medical Centre—
approved the final version of the manuscript. PG participated as a Ein Kerem, Jerusalem); D J D Rosen, H Y Kaneti, T Tzachi, J Zausmer
2150 www.thelancet.com Vol 372 December 20/27, 2008
9. Articles
(Meir Medical Centre, Kfar-Saba); M Maman, T Perri, S Taitelboum 4 Ballard PL, Ertsey R, Gonzales LW, Gonzales J. Transcriptional
(Rabin Medical Centre, Petach Tikva); M Simchen, G Shalev (Sheba regulation of human pulmonary surfactant proteins SP-B and SP-C
Medical Centre, Ramat-Gan); M Goldinfeld, O Levine (Poriya by glucocorticoids. Am J Respir Cell Mol Biol 1996; 14: 599–607.
Government Hospital, Tiberias); Jordan—M Y El-Zibdeh, L T Al-Faris, 5 Quinlivan JA, Evans SF, Dunlop SA, Beazley LD, Newnham JP. Use
H A Ayyash (Islamic Hospital, Amman); Peru—P Saona Ugarte (Hospital of corticosteroids by Australian obstetricians—a survey of clinical
Nacional Cayetano Heredia, Lima); Poland—K Preis, practice. Aust N Z J Obstet Gynaecol 1998; 38: 1–7.
I Domzalska-Popadiuk, M Swiatkowska-Freund, I Janczewska (Medical 6 Guinn DA, Atkinson MW, Sullivan L, et al. Single vs weekly courses of
University of Gdansk, Gdansk); J Wilczynski, M Krekora, M Kesiak, antenatal corticosteroids for women at risk of preterm delivery:
E Gulczyńska (Polish Mothers Memorial Hospital Research, Lodz); a randomized controlled trial. JAMA 2001; 286: 1581–87.
M Ropacka, Ma Madejczyk, J Rozycka, G H Breborowicz, 7 Crowther CA, Haslam RR, Hiller JE, Doyle LW, Robinson JS, for
M Szymankiewicz (University School of Medical Sciences, Poznan); ACTORDS. Neonatal respiratory distress syndrome after repeat
exposure to antenatal corticosteroids: a randomised controlled trial.
Russia—N Borzova, L V Posiseeva (Research Institute, Ivanovo);
Lancet 2006; 367: 1913–19.
Z Khodjaeva, E Vikhlyaeva (Research Centre of Obstetrics, Gynecology
8 Wapner RJ, Sorokin Y, Thom EA, et al. Single vs weekly courses of
and Perinatology, Moscow); Spain—M Palacio, D Salvia, Fr Botet,
antenatal corticosteroids: evaluation of safety and efficacy.
M Massanes, M Lopez (Hospital Clinic-University of Barcelona, Am J Obstet Gynecol 2006; 195: 633–42.
Barcelona); Switzerland—I Hösli, W Holzgreve, C A Voekt (University
9 Crowther CA, Harding JE. Repeat doses of prenatal corticosteroids for
Women’s Hospital, Basel); F Belhia, P Hohlfeld, E Prince-dit-Clottu women at risk of preterm birth for preventing neonatal respiratory
(CHU Vaudois, Lausanne); E Beinder, U von Mandach, J-C Fauchere disease. Cochrane Database Syst Rev 2007; 3: CD003935.
(University Hospital, Zurich); Netherlands—F JME Roumen, M Pieters 10 Kapoor A, Petropoulos S, Matthews SG. Fetal programming of the
(Atrium Medical Centre, Heerlen); F Smits (Academisch Ziekenhuis hypothalamic pituitary-adrenal (HPA) axis function and behavior by
Maastricht, Maastricht); United Kingdom—I I Bolaji, P Adiotomre (Diana, synthetic glucocorticoids. Brain Res Rev 2008; 57: 586–95.
Princess of Wales Hospital, Grimsby); H Al-Taher, H Barnes (The Queen 11 French NP, Hagan R, Evans SF, Mullan A, Newnham JP. Repeated
Elizabeth Hospital, King’s Lynn); R Abdul-Kadir, C Chi, V Van Someren antenatal corticosteroids: effects on cerebral palsy and childhood
(Royal Free Hospital, London); USA—S Dexter, R Samelson, M J Horgan, behavior. Am J Obstet Gynecol 2004; 190: 588–95.
C Valentini (Albany Medical Center, Albany); S Pardanani, 12 Aghajafari F, Murphy K, Matthews S, Ohlsson A, Amankwah K,
M Bebbington, C Chazotte (Montefiore Medical Center, Bronx); Hannah M. Repeated doses of antenatal corticosteroids in
S Kilpatrick, Je L Drahos (University of Illinois Medical Center at animals: a systematic review. Am J Obstet Gynecol 2002; 186: 843–49.
Chicago, Chicago); L Saldana, Be Mount, B Warner, K Wedig (Bethesda 13 Peltoniemi OM, Kari A, Tammela O, et al. Randomized trial of a
North Hospital and Good Samaritan Hospital, Cincinnati); A Lysikiewicz single repeat dose of prenatal betamethasone treatment in imminent
(Winthrop-University Hospital, Mineola); F Bsat, J Fleming, A Lee, preterm birth. Pediatrics 2007; 119: 290–98.
D Hoffman (Baystate Health System, Springfield). 14 Jobe AH, Wada N, Berry LM, Ikegami M, Ervin MG. Single and
Data safety monitoring board: M Bracken, P Crowley, Al Donner, L Duley, repetitive glucocorticoid exposure reduce fetal growth in sheep.
J Tyson. Am J Obstet Gynecol 1998; 178: 880–85.
15 Quinlivan JA, Archer MA, Dunlop SA, Evans SF, Beazley LD,
Conflict of interest statement Newnham JP. Fetal growth retardation, particularly within lymphoid
We declare that we have no conflict of interest. organs, following repeated maternal injections of betamethasone in
Acknowledgments sheep. J Obstet Gynecol Res 1998; 24: 173–82.
We thank all participants in the MACS trial; Caroline Crowther and 16 Halliday HL, Ehrenkranz RA, Doyle LW. Early postnatal (<96 hours)
Ron Wapner for their collaboration and support; the members of our corticosteroids for preventing chronic lung disease in preterm
infants. Cochrane Database Syst Rev 2003; 3: CD001146.
data safety monitoring board for their guidance; and all the staff at the
Maternal, Infant and Reproductive Health Research Unit in Toronto for 17 Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution
of subependymal and intraventricular hemorrhage: a study of infants
their dedication and hard work. MACS was supported by a grant from
with birth weight less than 1500 grams. J Pediatr 1978; 92: 529–34.
the Canadian Institutes for Health Research (grant number MCT
18 Walsh MC, Kliegman RM. Necrotizing enterocolitis: treatment based
38142). The data coordinating centre was supported by grants from
on staging criteria. Pediatr Clin North Am 1986; 33: 179–201.
Sunnybrook Health Sciences Centre, Women’s College Hospital and the
19 World Health Organization. Neonatal and perinatal mortality country,
Department of Obstetrics and Gyneacology at the University of Toronto.
regional and global estimates. WHO, Geneva 2000; http://www.who.
Betamethasone and placebo were purchased from Schering-Plough int/making_pregnancy_safer/publications/neonatal.pdf (accessed
Corporation (Madison, NJ, USA) and Eminent Services Corporation March 1, 2007).
(Gaithersburg, MD, USA), respectively. 20 Kramer MS, Platt RW, Wen SW, et al. A new and improved
References population-based Canadian reference for birth weight for gestational
1 Crowley P, Chalmers I, Keirse MJNC. The effects of corticosteroid age. Pediatrics 2001; 108: E35.
administration before preterm delivery: an overview of the evidence 21 Brown ER, Nielsen H, Torday JS, Tauesch HW. Reversible induction
from controlled trials. Br J Obstet Gynaecol 1990; 97: 11–25. of surfactant production in fetal lambs treated with glucocorticoids.
2 Liggins GC, Howie RN. A controlled trial of antepartum Pediatr Res 1979; 13: 491.
glucocorticoid treatment for prevention of the respiratory distress 22 Wapner RJ, Sorokin Y, Mele L, et al. Long-term outcomes after repeat
syndrome in premature infants. Pediatrics 1972; 50: 515–25. doses of antenatal corticosteroids. N Engl J Med 2007; 357: 1190–98.
3 Taeusch HW, Brown E, Torday JS, Nielsen HC. Magnitude and 23 Crowther CA, Doyle L, Haslam RR, et al. Outcomes at 2 years of age
duration of lung response to dexamethasone in fetal sheep. after repeat doses of antenatal corticosteroids. N Engl J Med 2007;
Am J Obstet Gynecol 1981; 140: 452–55. 357: 1179–89.
www.thelancet.com Vol 372 December 20/27, 2008 2151