Antenatal betamethasone and incidence of neonatal respiratory distress after elective caesarean section: pragmatic randomised trial
BMJ 2005; 331 doi: https://doi.org/10.1136/bmj.38547.416493.06 (Published 22 September 2005) Cite this as: BMJ 2005;331:662
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
We recently reported (1) that antenatal betamethasone is effective in
reducing admissions to special care baby unit with respiratory distress
after elective caesarean section at term. The recent Cochrane review (2)
of 21 studies (3885 women and 4269 infants) concludes that a single course
of antenatal corticosteroids should be considered routine for preterm
delivery with few exceptions. A reduction in the number of babies with
respiratory distress syndrome was found with fewer babies dying and a
reduction in the incidence of cerebroventricular haemorrhage and
necrotising enterocolitis. There was no evidence of a negative effect on
the mother with good long term outcomes for mother and baby.
The recent “rapid” response by D Hutcheon 23 October 2006 to our
paper is misleading and incorrect. He reports that “The most recent update
of the Cochrane review (2) on antenatal corticosteroids showed a
significantly increased relative risk of neonatal death of 3.25 (95% CI
0.99 – 10.66) following antenatal steroids given after 36 weeks
gestation.” The report actually states this relative risk as a
“nonsignificant trend towards an increase in combined fetal and neonatal
death”. One may argue that this is but a semantic difference, so it is
educational to consider the evidence further. Of the 14 relative risks
quoted in the relevant paragraph of the Cochrane report, six showed
statistically significant benefits in reducing fetal and neonatal deaths,
and seven showed no difference. Cochrane concluded that the combined
fetal and neonatal death rate was significantly reduced in antenatal
corticosteroid treated infants before 32 weeks ( RR 0.71 95% CI 0.57 to
0.88 ),before 34 weeks RR 0.75, 95% CI 0.58 to 0.91 ) ,and before 36
weeks (RR 0.75, 95% CI 0.61 to 0.94 ).
The data which was combined to give the reported “nonsignificant
trend” in relative risk of neonatal death of 3.25 (95% CI 0.99 –
10.66),following antenatal corticosteroids after 36 weeks gestation, is
drawn from two studies. One of the two (3) included 37 babies and
witnessed only one event (RR 1.88 95% CI 0.08 to 43.10). The second trial
(4), of 211 cases and 250 controls dates back to a publication date of
1972, (RR 3.55 95% CI 0.97 to 12.96) reporting data which is now over 35
years old. They report 12 neonatal or fetal deaths in babies born at least
36 weeks (i.e. a 2.6% mortality). The very wide confidence limits in
Cochrane reflect this limited data. Our study included 932 babies with a
gestation 37 weeks or greater and observed a mortality rate of zero.
Relative risks alter over time as medical practice develops.
Our randomized trial shows that the widely demonstrated beneficial
use of antenatal corticosteroids in preterm babies extends to those 37
weeks and greater.
In planning an elective caesarean section, the risk of respiratory
distress should be considered and the likely benefit of antenatal
corticosteroids should be compared with those of delaying delivery until
39 weeks when possible. Outdated evidence should be viewed with caution,
and statistical evidence should be read for what it is, an estimate of
probability, not a guarantee of certainty.
References
1) Stutchfield P,Whitaker R ,Russell I. Antenatal betamethasone and
the incidence of neonatal respiratory distress after elective caesarean
section: pragmatic randomised trial. BMJ 2005;331:662 (24 September 2005)
2) Roberts D,Dalziel S. Antenatal corticosteroids for accelerating
fetal lung maturation for women at risk of preterm birth. Cochrane
Database of Systematic Reviews 2006,Issue 3 Art.No. CD 004454.
3)Doran TA, Swyer P,MacMurray B, Mahon W , Enhorning G, Bernstein A
et al. Results of a double blind controlled study on the use of
betamethasone in the prevention of respiratory distress syndrome American
Journal Obstetrics and Gynaecology 1980;136 : 313 -20.
4)Liggins GC,Howie RN, A controlled trial of antepartum
glucocorticosteroid treatment for the prevention of respiratory distress
syndrome in premature infants. Pediatrics 1972;50:515 - 25
Competing interests:
None declared
Competing interests: No competing interests
Sir,
Stutchfield et al state in their rapid response that they are confident that a larger study, designed and adequately powered would show an overall difference in all admissions to the special care baby unit following antenatal steroids. (1) The study involved mothers in whom elective caesarean section was planned at 37 weeks' gestation or beyond. The most recent update of the Cochrane review (2) on antenatal corticosteroids showed a significantly increased relative risk of neonatal death of 3.25 (95% CI 0.99 – 10.66) following antenatal steroids given after 36 weeks gestation. I wonder how this risk could be justified in any future trial.
References
1. BMJ Rapid response by Peter Roy Stutchfield, Rhiannon Whitaker on. Admissions to Special Care Baby Unit in the Antenatal Steroids for Term Elective Caesarean Section Trial 14 December 2005
2. Antenatal corticosteroids for accelerating fetal lung maturation for women at risk of preterm birth [Reviews] Roberts, D; Dalziel, S : 24-May-2006 Cochrane Pregnancy and Childbirth Group
Dr Devender Roberts,
Competing interests: No competing interests
The letter from Stutchfield (4) raises some very important issues
and I am grateful for the authors’ rapid response now more than three
months after I queried the figures. (1) My concern is to ensure that
neonatal morbidity after elective Caesarean Section is minimised. The
authors admit that the meaning of the study, that is reduced separation of
the mother and baby, less interference with bonding and decresased cost of
treatment in the special care baby unit, was not met overall by the
administration of steroids. The criterion of reduced morbidity was only
met for the babies admitted to special care with a diagnosis of
respiratory distress or transient tachypnoea. However the treatment group
was not blinded and there is a possibility of bias on the part of the
doctor making the diagnosis. The fact that there were nine babies admitted
to SCBU in the whole series with either no diagnosis or considered
“normal” shows how tenuous admission to special care may be. A similar
reduction in admissions to special care can be obtained by postponing
delivery to at least 39 weeks. (2)
The authors point out that there were four patients in the treatment
arm who did not receive steroids because of emergency Caesarean Section
before the betamethasone would have been completed. Although they do not
state the reason, most likely this was because of the onset of spontaneous
labour. The stress of labour in itself may reduce neonatal respiratory
morbidity. The authors also point out that the results were analysed
according to intention to treat as is the correct protocol. This mimics
the real world since no doubt there would be a number women who go into
labour before steroids could be effectively administered steroids before
elective caesarean section was standard practice.
I have raised the issue of minimising the pathological intervention
of Caesarean Section by ensuring that it is not compounded by immediate
cord clamping. (1) Delay of at least 30 seconds has been shown to have a
beneficial effect on lung function. (3) This should be practiced routinely
and of course can be applied even when caesarean section is an emergency.
(4)
A physiological approach must be preferable to a pharmacological
approach. A single course of antenatal steroids may be fairly safe and
justified to reduce the risk of respiratory morbidity in the small number
of neonates in which elective caesarean delivery is necessary before 39
completed weeks. However it should not be used to make a more convenient
earlier delivery.
I accept that Stutchfield et al have made the effort to organise a
randomised trial to investigate the use of steroids in elective caesarean
section. A similar trial to determine the effect of physiological
management of the third stage is urgently required. The end point must be
morbidity associated with neonatal admission as it would be ridiculous to
provide a solution for respiratory morbidity at the expense of overall
morbidity. Investigators must approach such studies without preconceived
opinions and with an open mind. Statements like “If a larger study was
designed and adequately powered to show an overall difference in all
admissions , we are confident that a reduction would be seen as a direct
result of fewer admitted with respiratory distress” (4) do not fit with
an open mind.
References.
1. A physiological approach to reducing neonatal morbidity in elective
Caesarean Section David J R Hutchon BMJ eletter 24 August 2005
2. Steer P J. Giving steroids before elective caesarean section. BMJ
2005; 331: 645-646.
3. Kinmond S, Aitchison TC, Holland BM, Jones JG, Turner TL, Wardrop
CA. Umbilical cord clamping and preterm infants: a randomised controlled
trial. BMJ 1993;306: 172-5.
4. Stutchfield P R and Whitaker R. Admissions to Special Care Baby
Unit in the Antenatal Steroids for Term Elective Caesarean Section Trial.
BMJ eletter 14 December 2005
Competing interests:
I beleive in physiology before pharmacology
Competing interests: No competing interests
The Antenatal Steroids for Term Elective Caesarean Section randomised
trial (1) was established to determine whether giving 2 doses of
betamethasone before delivery reduced the number of admissions of babies
with respiratory distress to Special Care Baby Unit. Using data from a 2
year review of outcome of term elective caesarean sections at our
hospital, we found that 8.9% of babies were admitted with respiratory
distress.
Recruiting 1100 mothers would have yielded 80% power of detecting a
reduction in the percentage of babies admitted to special baby unit with
respiratory distress from 8% to 4% using a 5% significance level. We
successfully demonstrated a 50% reduction (from 5% to 2.4%) in admissions
to special care for respiratory distress with the 942 patients we
recruited.
Mr Hutchon (2) correctly points out from Table 3 in the electronic
version of our paper that the total number of babies admitted in the
betamethasone group was 26/467 and in the control group 32/475, with no
statistically significant difference in the overall rate of admissions to
Special Care. The study was not powered or designed to show a difference
in the total admission rate, merely the difference in the numbers admitted
with respiratory distress. If a larger study was designed and adequately
powered to show an overall difference in all admissions , we are
confident that a reduction would be seen as a direct result of fewer
admitted with respiratory distress.
To assist readers to assess the results of the trial we supply the
following additional information. Of those babies admitted in the
betamethasone group, 11 had respiratory distress. The other 15 were
admitted with a range of diagnoses : 4 jaundice, 2 poor feeding, 1 birth
asphyxia, 1 hypoglycaemia, 2 normal neonates, 4 no diagnosis and 1
hyponatraemia. There were 32 admitted in the control group of whom 24 had
respiratory distress and 8 admitted for other reasons: 1 feeding
difficulties with hypoglycaemia, 1 dysmorphism – Prader Willi syndrome, 1
auto-immune thrombocytopenia, 1 jaundice, 1 unproven infection, 1 normal
neonate and 2 no diagnosis given. This demonstrates that the admission of
babies without respiratory distress was not related to antenatal
corticosteroids.
As is good practice our analysis was by intention to treat. Out of
the eleven babies admitted with respiratory distress in the treatment
group, four had not received betamethasone as they were delivered by
emergency caesarean section before the betamethasone would have been
administered. This further strenthens the findings of our study.
We trust that this further information clarifies the situation.
References
1. Stutchfield P, Whitaker R, Russell I. Antenatal Betamethasone and
Incidence of Neonatal Respiratory Distress after Elective Caesarean
Section: Pragmatic randomised trial. BMJ 2005: 331: 662-664.
(Electronic publication 22.8.05)
2. Hutchon D J R. A clearer explanation of no benefit for steroids.
BMJ e-letter 08.12.05.
Competing interests:
None declared
Competing interests: No competing interests
Is it surprising that the neonate sometimes fails to adapt properly
to extra-uterine life when it is suddenly removed from the uterus and then
its placental circulation immediately obstructed by a clamp? The authors
do not describe in any detail the method of delivery by Caesarean Section
so we can presume the method was that routinely used in England and Wales
at present. The cord is routinely clamped immediately a Caesarean Section
and, if resuscitation is required, the neonate is passed to a member of
the paediatric team. (personal communication)
A delay of at least 30 seconds is recommended before the cord is
clamped at delivery of the preterm infant.(1) Preterm infants typically
are at risk of developing respiratory distress syndrome. A delay of at
least 30 seconds in the preterm neonate reduces neonatal anaemia and the
need for transfusion as well as the risk of intraventricular haemorrhage.
(1) There is also a decrease in the severity of respiratory
distress.(2,3,4)
Although clamping the pulsating cord immediately at delivery may be
common practice, it is an intervention which requires informed consent. It
would appear that such informed consent was not obtained in this study.
David Hutchon
Refernces
1. Rabe H, Reynolds G, Diaz-Rossello J. Early versus delayed
umbilical cord clamping in preterm infants. The Cochrane Database of
Systematic Reviews 2004, Issue 3. Art. No.: CD003248.pub2. DOI:
10.1002/14651858.
2. Kinmond S, Aitchison TC, Holland BM, Jones JG, Turner TL, Wardrop
CA. Umbilical cord clamping and preterm infants: a randomised controlled
trial. BMJ 1993;306: 172-5.
3. Dunn P M. Stress failure of pulmonary capillaries at birth. The
Lancet 1993;342:120
4. Dunn P M. Postnatal placental respiration. Developmental Medicine
and Child Neurology. 1966;8:607
Competing interests:
None declared
Competing interests: No competing interests
The references for our rapid response dated 9th December entitled
Elective Caesarean Section at Term are detailed below.
1. Stutchfield P, Whitaker R, Russell I. Antenatal betamethasone and
incidence of neonatal respiratory distress after elective caesarean
section: pragmatic randomised trial. BMJ 2005: 331: 662-
664.(electronic publication 22.8.05)
2. Fiori H. Why should antenatal steroids prevent transient
tachypnoea of the newborn? – rapid response letter BMJ posted 04.10.05.
3. Steer P J. Giving steroids before elective caesarean section.
BMJ 2005; 331: 645-646.
4. Crowley P, Roberts D, Dalziel S, Shaw BNJ. Antenatal
corticosteroids to accelerate fetal lung maturation for women at risk of
preterm birth. (Protocol) The Cochrane Database of Systematic Reviews
2003, Issue 4. Art. No.: CD004454. DOI: 10.1002/14651858.CD004454.
5. Royal College of Obstetricians and Gynaecologists Scientific
Advisory Committee. RCOG guidelines. No 7: Antenatal corticosteroids to
prevent respiratory distress syndrome. 2nd Ed.London: RCOG Press, 2004:
1-9
6. Dalziel SR, Liang A, Parag V, Rodgers A, Hardin JE. Blood
pressure at six years of age after prenatal exposure to betamethasone:
follow up results of a randomised controlled trial. Pediatrics 2004; 114.
7. Dalziel SR, Lim VK, Lambert A, McCarthy D, Parag V, Rodgers A,
Harding JE Antenatal exposure to betamethasone: psychological functioning
and health related quality of life 31 years after inclusion in randomised
controlled trial
BMJ, Sep 2005; 331: 665
Competing interests:
None declared
Competing interests: No competing interests
I saw the response from the authors with the expectation of an
explanation or at least recognition of the main points of my previous
letters. The authors failed to point out in their original publication
that although there was a significant reduction in SCBU admissions for TTN
and RDS, there was no significant difference in the number of babies
admitted to SCBU overall. Perhaps there has been a misprint in which case
a response would have been easy.
I have copied and pasted the figures in the paper directly to ensure
there is no error.
Total No of babies
admitted in betamethosone group 26/467
Total No of babies
admitted in control group 32/475
I therefore repeat from my letter “From table 3 we can see that
there were 26 out of 467 babies admitted to the SCBU in the treatment
group and 32 out of 475 in the control group.
This gives an incidence of admission to SCBU of 0.067 in the control group
and 0.056 in the treatment group (relative risk 0.83, 95% confidence
interval 0.5 to 1.36).”
This is a non-significant difference. How does it fit into the
meaning of the study as stated by the authors.
David Hutchon
Competing interests:
None declared
Competing interests: No competing interests
We reported the outcome of a randomised trial of antenatal
betamethasone before elective caesarean section at term demonstrating
>50% reduction in admission with respiratory distress for those
delivered between 37-39 weeks (1). Of the 35 babies admitted with
respiratory distress, 19 control babies had transient tachypnoea (TTN), 5
with respiratory distress syndrome (RDS), compared to 10 babies with TTN
in the intervention group and 1 RDS. 14 control babies required intensive
care, 3 with RDS requiring ventilation for between 2 and up to 5 days with
length of stay 12-18 days, whereas only 2 in the intervention group
received intensive care.
We postulated that the reduced incidence of TTN with antenatal
betamethasone may result from an effect on the expression of the
epithelial channel gene allowing the lung to switch from fluid secretion
to fluid absorption . Professor Fiori’s rapid response(2) provides
evidence for an additional factor, enhanced surfactant production. The
presence of lung fluid is likely to delay surfactant production leading
further to a decrease in lung compliance seen in TTN.
Professor Steer in the editorial in the same issue (3) raises
concerns about the long term consequences of giving antenatal steroids by
reporting the outcome of follow-up studies where multiple 2 weekly courses
of antenatal steroids were given from 24 weeks gestation onwards and high-
dose intravenous postnatal courses of corticosteroid prescribed for the
very preterm to prevent chronic lung disease. He then raised the example
of thalidomide and diethylstilboestrol in pregnancy where mass prescribing
and the lack of long term follow up research lead to well reported serious
consequences. These drugs were used early in pregnancy during early
embryogenesis. Professor Steer concludes that “giving steroids whose long
term safety, even as a single course, remains questionable.”
Though we understand Prof Steer’s concern, the Cochrane review in
2003 (4) and the evidence based guideline of Royal College of Obstetrics
& Gynaecology (5) on the use of antenatal corticosteroids to prevent
respiratory distress syndrome, concludes that a single course of
antenatal corticosteroid has no adverse effect on physical growth,
neurological or cognitive outcome or infection in child or mother .The
Royal College’s guidelines synthesises five papers following more than
1500 survivors from randomised trials or cohort studies of antenatal
corticosteroids for up to 20 years. Concerned that Prof Steer might know
of a flaw in this guideline from his own College, we have reviewed all
five papers, together with another published since the RCOG guideline (6)
and Dalziel’s follow up
paper (7) which extends follow up to 31 years. However we find the
College guideline rigorous and appropriate.
The risk of Prof Steer’s recommendation is that he may discourage
others from discussing and offering a simple, safe and effective evidence
based preventative measure to women who, for whatever reason require or
plan an elective section before 39 weeks.
Competing interests:
None declared
Competing interests: No competing interests
The authors rightly state that admission to a special unit separates
mother and baby, potentially disrupting bonding, and increasing the cost
of care and the risk of complications. From table 3 we can see that there
were 26 out of 467 babies admitted to the SCBU in the treatment group and
32 out of 475 in the control group. This gives an incidence of admission
to SCBU of 0.067 in the control group and 0.056 in the treatment group
(relative risk 0.83, 95% confidence interval 0.5 to 1.36). Therefore
although there were slightly fewer babies admitted to special care in the
treatment group the difference was not statistically significant. The
meaning of the study of reducing the separation of the mother and baby,
together with the costs and risks of admitting a baby to special care,
were NOT met.
This fact was missed completely by the authors. Readers
should note that these figures are only clear in the electronic version of
the paper. It is true that the primary outcome, respiratory distress
requiring admission to special care is unlikely to be susceptible to
maternal influence, however there is little doubt such diagnosis and
admission are susceptible to bias by doctors.
My apologies for my rather confusing previous letter.
Competing interests:
None declared
Competing interests: No competing interests
Not an intentional error
I am grateful for the response by Stutchfield et al (1) and I did misinterpret the 95% confidence interval (0.99 – 10.66) as significant. However I still feel that considerable caution needs to be exercised in view of the NON-significant trend. For my comment I relied upon, and misinterpreted the actual figures provided by the review (2) which showed a relative risk of fetal or neonatal death of 3.25 with 95% confidence intervals that I now realise do touch/cross unity. I did not appreciate that the authors considered this a non-significant trend. In my defence I quote from another Cochrane review comparing inhaled and systemic steroids. In the main results “There was an increase in the incidence of CLD at 36 weeks CGA in the inhaled steroid group amongst all randomised infants, which was of borderline statistical significance: [RR 1.45 (95% CI 0.99,2.11) “ Although the confidence interval also just crosses/touches unity the authors describe the result as borderline statistical significance. I recommend readers to check the facts for themselves.
It is difficult to understand consistency as the word “trend” is also used three other times in the review(2). First in relation to proven neonatal infection ( RR 0.83, 95%CI 0.66-1.03), then in relation to a reduction in the need for surfactant (RR=0.72, 95%CI 0.51 – 1.03) and also in relation to having cerebral palsy (RR=0.60, 95%CI 0.34-1.03). The confidence intervals for all these conditions DO cross unity yet there is no mention of “non-significance” in these results. As we are all aware statistics can be manipulated to fit our prejudice. I agree that the numbers are small and the studies quite old. However data from the Auckland trial was recently analysed to demonstrate the long-term safety of antenatal steroids. If the data is good enough for this analysis it should be good enough to question the safety of corticosteroids after 36 weeks. Re-analysis of this trial data by Dalzeil et al showed, contrary to the original Liggins and Howe reports, “similar numbers of neonatal survivors with much the same perinatal morbidity in both … groups.” (3)
It is not good enough for Stutchfield et al to quote their own results. The principle of systematic review is to include all controlled trial results including looking for unpublished data. Perhaps when the next update is published the Stutchfield et al data will be included.
Stutchfield et al in their carefully worded rapid response referred to the reduced admissions to special care with respiratory distress after caesarean section, but as I previously pointed out there was no reduction in admissions overall. Surely we need to find a way of reducing all morbidity from caesarean section and I would welcome a response about cord management. A delay of a minute from birth to clamping the cord is now recommended by the UK neonatal resuscitation council. If this advice was implemented much of the neonatal morbidity might be avoided.
The correct spelling of my name is HUTCHON
References
1. BMJ rapid response by Stutchfield PR and Whittaker R on Reduction in neonatal death with corticosteroids. 14th December 2006
2. Antenatal corticosteroids for accelerating fetal lung maturation for women at risk of preterm birth [Reviews] Roberts, D; Dalzeil, S: 24-May-2006 Cochrane Pregnancy and Childbirth Group
3. Dalzeil RS, Walker NK, Parag V, Mantell C, Rea HH, Rodgers A and Harding J. Cardiovascular risk factors after antenatal exposure to betamethasone: 30-year follow-up of a randomised controlled trial. Lancet 2005;365:1856-62
Competing interests:
None declared
Competing interests: No competing interests