Intended for healthcare professionals

Rapid response to:

Research

Risk of respiratory morbidity in term infants delivered by elective caesarean section: cohort study

BMJ 2008; 336 doi: https://doi.org/10.1136/bmj.39405.539282.BE (Published 10 January 2008) Cite this as: BMJ 2008;336:85

Rapid Response:

Full dataset

Sir,

Sutchfield and Whittaker charge me with guesswork on the meaning of their published data but they needed to check their original data to show that my conclusion was incorrect. I would not accept that guesswork is the correct term and, without the full dataset available, it was impossible for me to reach another conclusion. This example confirms the potential value of my proposal to provide raw data with all publications (1) and demonstrated on the website(2). Providing raw data can easily be done with epublications.

It should be noted that the ASTECS trial was not blinded as Blinding participants and professionals was therefore neither desirable nor possible. Similarly the use of placebos was not practical. Furthermore the primary outcome, respiratory distress requiring admission to special care baby unit, is hardly susceptible to maternal influence. While respiratory distress requiring admission to the special care baby unit may not be susceptable to maternal influence, professionals making the decision to admit could well have been influenced by knowledge of whether or not the mother had steroids.

I fully agree that clinical management can be influenced by many factors including unsubstantiated statements, powerful and biased teaching and intransigent opinion. It could be argued that these factors have been responsible for the risk of immediate cord clamping being unrecognised for the past 50 years! Landau et al (3) working in the USA recognised that respiratory problems after caesarean delivery could be largely prevented by avoiding immediate cord clamping. They used a technique to allow the placental circulation to remain intact till long after delivery. I quote from their paper “Since instituting this technique our results with our cesarean babies have been much better than formerly. For this reason we have not felt justified in running a control series.” Similar conclusions of Professor Peter Dunn working in Bristol have also been virtually ignored for the last 20 years.(4) I am at a loss to understand why a pharmacological approach is promoted and the physiological approach has never been investigated further. Both have a logical basis and indeed could well be complimentary approaches particularly in the preterm baby.(5)

I fully agree that elective delivery by caesarean section should always be delayed until at least 39 weeks if possible. When earlier delivery is necessary there is an increasing risk of respiratory distress. Giving antenatal steroids reduces the risk that a professional will make a diagnosis of RDS which requires admission to a special care unit. In advising parents about the safety of antenatal steroids, professionals should advise that although the 30 year follow-up did not show any adverse effects neither did it show any benefit.(6)

If Stuchfield and Whittaker had been able to include raw data in their original paper (7) it would remain available in the foreseeable future for other workers to analyse further. The data could be copied into other applications and statistical analysis checked by peer reviewers and readers. In addition the raw data may make any future metanalysis more accurate.

Finally they state above that Our study was not powered to make conclusions about admissions other than for respiratory distress. In a previous response they stated that they stated 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.(8) However according to Hewett et al such a statement will always be true given a large enough study.(9) Although total numbers of days admitted to special care (with its implications of cost, morbidity, and separation) for the babies was reduced by 50%, they do not tell us if this is a statistically significant difference. Using simply the number of days, I calculate a statistically significant odds ratio of 0.4 (95% confidence 0.3 - 0.6) in favour of steroids preventing admission.

References

1. Hutchon DJR BMJ Information in practice Infopoints
Publishing raw data and real time statistical analysis on e-journals 2001;322:530

2. Demonstration real time statistical analysis.

3. Landau D B, Goodrich H B, Franka WF, Burns FR. Journal of Pediatrics (1950) 36:421-426

4. Dunn P M. Reservations about the methods of assessing at birth the predictive value of intrapartum fetal monitoring including premature interruption of the feto-placental circulation. (1986) In Fetal Physiology Measurements. Report of the 2nd International Conference, Oxford, 2-4 April 1984. Ed. P Rolfe. Publ Butterworths, London pp 130-137

5. Reynolds GJ. Beyond sweetness and warmth: transition of the preterm infant Arch. Dis. Child. Fetal Neonatal Ed. 2008; 93: F2-F3.

6. Dalziel SR, Walker NK, Parag V, Mantell C Rea HH, Rodgers A, Harding JE. Cardiovascular risk factors after antenatal exposure to betamethasone: 30- year follow-up of a randomised controlled trial. Lancet, 2005; 365:1856 – 62

7. Stutchfield P, Whitaker R, Russell I. Antenatal betamethasone and incidence of neonatal respiratory distress after elective caesarean section: pragmatic randomised trial. BMJ,doi:10.1136/bmj.38547.416493.06 (2005).

8. Stutchfield and Whittaker Rapid response Admissions to Special Care Baby Unit in the Antenatal Steroids for Term Elective Caesarean Section Trial 14 December 2005

9. Luis C Silva rapid response About p-values and confidence intervals: habitual misconceptions. 12 January 2008

David J R Hutchon FRCOG
Consultant Obstetrician

Competing interests:
None declared

Competing interests: No competing interests

04 March 2008
David JR Hutchon
Consultant Obstetrician and Gynaecologist
Darlington Memorial Hospital, DL3 6HX