Intended for healthcare professionals

Letters

Prediction of survival for preterm births

BMJ 2000; 320 doi: https://doi.org/10.1136/bmj.320.7235.647 (Published 04 March 2000) Cite this as: BMJ 2000;320:647

This article has a correction. Please see:

Survival table was not easy to understand

  1. T H H G Koh, neonatologist,
  2. H Harrison, mother of a child born at 28 weeks,
  3. A Casey, mother of a baby born at 27 weeks
  1. Kirwan Hospital, Thuringowa, Queensland 4817, Australia
  2. 1144 Sterling Avenue, Berkeley CA 94708, USA
  3. Seaholme, Victoria 3018, Australia
  4. Department of Feto-Maternal Medicine, Leeds General Infirmary, Leeds LS1 3EX
  5. Department of Child Health and Obstetrics, University of Leeds, Leeds LS2 9JT
  6. Division of Newborn Services, Royal Women's Hospital, Carlton, Victoria 3053, Australia
  7. Victorian Perinatal Data Collection Unit, Department of Human Services, Melbourne, Victoria 3001, Australia

    EDITOR—The article by Draper et al deserves further comment.1 The objective of the study was to produce current data on survival of preterm infants, yet theirs was a retrospective study for 1994-7. Since it is now 1999 we believe that their data, albeit useful, are recent rather than current. Antenatal administration of steroids to the mothers was not mentioned as one of the additional factors affecting infants' survival.

    As a neonatologist and two mothers of extremely premature babies we do not find the tables described in the paper easy to understand. Draper et al may be interested in our paper published last month, which describes a table giving outcome for parents of babies at less than 28 weeks of gestation.2 This table contains information on survival rates, and also short term complications and treatment of the most recent cohort of such babies. This table could be unit based, updated annually, and be made available to the parents and each member of the perinatal team to promote consistency in the information given to the parents. We also suggest that parents be informed of 11 points, four of which are:

    • Outcome for the baby depends on many factors, not all obvious and including infection and maternal and fetal health.

    • A reasonably easy to remember guide is that the survival rate is about 40% for all babies born at 24 weeks' gestation, 50% for those born at 25 weeks, 60% for those born at 26 weeks, 70% for those born at 27 weeks, and 80% for those born at 28 weeks.3

    • Every baby is an individual, and the parents need to realise that their baby may be different from the average. The table is designed to share with parents risk factors that need to be understood within the unique context of the child and family.

    • Babies born at 25 weeks and less are at high risk of death, a long, tortuous journey through life, and disability. Some babies born at 24 and 25 weeks do, however, seem to be developing normally.

    The views of doctors, nurses, and parents with respect to such information must be sought out. We have just completed a study of 71 perinatologists, 35 neonatal nurses, and 48 parents of extremely premature babies, seeking their views of our outcome table. We found that parents were the most positive and accepting of the table, followed by nurses in second position, and finally by doctors.

    References

    1. 1.
    2. 2.
    3. 3.

    Curve presents information on survival and morbidity more clearly than tables

    1. Emma L Ferriman, subspecialty training fellow in feto-maternal medicine,
    2. M Assaf Doulah, ecturer in paediatrics,
    3. Nigel A B Simpson, senior lecturer in obstetrics and gynaecology
    1. Kirwan Hospital, Thuringowa, Queensland 4817, Australia
    2. 1144 Sterling Avenue, Berkeley CA 94708, USA
    3. Seaholme, Victoria 3018, Australia
    4. Department of Feto-Maternal Medicine, Leeds General Infirmary, Leeds LS1 3EX
    5. Department of Child Health and Obstetrics, University of Leeds, Leeds LS2 9JT
    6. Division of Newborn Services, Royal Women's Hospital, Carlton, Victoria 3053, Australia
    7. Victorian Perinatal Data Collection Unit, Department of Human Services, Melbourne, Victoria 3001, Australia

      EDITOR—We read with great interest the article by Draper et al on prediction of survival of preterm neonates.1 It provides valuable data on the mortality of preterm infants in the Trent region. However, the tables, although detailed, seem cumbersome to use and do not provide information on morbidity. We have used the data collected on 395 white neonates born at the two main teaching hospitals in Leeds from 1995 to 1997. These data are presented as a mortality/morbidity curve (figure). Morbidity is defined as babies developing any of the following pathologies: chronic lung disease, periventricular leukomalacia, intraventricular haemorrhage (grade 3/4) or necrotising enterocolitis.

      Figure1

      Survival and morbidity of 395 babies born in Leeds from 1995 to 1997

      Although the curve does not provide the same degree of accuracy we have found it easy to use as it clearly presents information on survival and significant morbidity, which is paramount for clinicians and parents in the decision making process. The curves are simple for any centre to construct from their own databases, and we hope further to improve our own data by including information on one year survival and morbidity.

      References

      1. 1.

      Data on the quality of survival are needed

      1. Lex W Doyle, paediatrician,
      2. Colin J Morley, paediatrician,
      3. Jane Halliday, epidemiologist
      1. Kirwan Hospital, Thuringowa, Queensland 4817, Australia
      2. 1144 Sterling Avenue, Berkeley CA 94708, USA
      3. Seaholme, Victoria 3018, Australia
      4. Department of Feto-Maternal Medicine, Leeds General Infirmary, Leeds LS1 3EX
      5. Department of Child Health and Obstetrics, University of Leeds, Leeds LS2 9JT
      6. Division of Newborn Services, Royal Women's Hospital, Carlton, Victoria 3053, Australia
      7. Victorian Perinatal Data Collection Unit, Department of Human Services, Melbourne, Victoria 3001, Australia

        EDITOR—In their paper Draper et al reported survival rates for infants of <33 weeks born during 1994-7 in the Trent health region.1 We wish to comment on several aspects of their report.

        Their survival rates seem lower than those we found over a similar period, so we have compared their data with reported data from Australia and New Zealand.25 The Australian state of Victoria has about 60 000 births annually, similar to the Trent region. Data are compiled for all births in Victoria from 20 weeks' gestation and linked to data reported on each death.2 We collated survival data to hospital discharge for 1994-7 from the Victorian perinatal data collection unit. Survival rates from all three sources were higher than those reported for Europeans by Draper et al, particularly at the lower gestational ages (table).

        Survival rates by gestational age. Data are percentage survival (with 95% confidence intervals)

        View this table:

        Survival rates may differ between the regions for the following reasons. The Trent region, with the same number of births as Victoria, has 16 neonatal units compared with four in Victoria. Cohorting preterm infants improves medical and nursing skills, which in turn improves survival rates. The four units in Victoria are staffed predominantly by full time neonatologists and junior medical staff. The nurse-patient ratios are more favourable in Australia than in the United Kingdom. Australia has a policy of transfer in utero to tertiary perinatal centres for infants expected to deliver at <33 weeks. The improving survival rates are reported to obstetricians, who decide whether a preterm infant will be born alive and whether the paediatric team will participate. Survival rates are possibly higher because more extremely preterm infants are actively treated in Victoria than in the Trent region.

        If the causes of death were known for infants in the Trent region, other differences between the systems in Australia and the United Kingdom might be identified. We concluded that few preventable causes of death could be identified in preterm infants in 1994-6.5 Not only survival data but also data on the quality of survival help decision making.

        References

        1. 1.
        2. 2.
        3. 3.
        4. 4.
        5. 5.
        View Abstract

        Log in

        Log in through your institution

        Subscribe

        * For online subscription