Rapid Responses to:

RESEARCH: David J Field, Jon S Dorling, Bradley N Manktelow, and Elizabeth S Draper Survival of extremely premature babies in a geographically defined population: prospective cohort study of 1994-9 compared with 2000-5 BMJ 2008; 336: 1221-1223 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Improving survival is good, but preventing extremely preterm birth is better

Lisa Hilder   (18 May 2008)

Survival rates in very preterm babies in the Marche region

Roberta Buglia, Annalisa Pedini, Roberta Freddara and Virgilio P. Carnielli   (13 June 2008)

Survival of extremely preterm babies, supersedes earlier response by Lisa Hilder, who has withdrawn it

Alison J Macfarlane, Lisa Hilder   (8 July 2008)

Improving survival is good, but preventing extremely preterm birth is better 18 May 2008
Lisa Hilder, Senior Research Fellow Department of Midwifery, City University, London Send response to journal: Re: Improving survival is good, but preventing extremely preterm birth is better	I read with interest the analysis of the trends in survival of babies born very preterm to mothers normally resident in Trent (1). The study found no change in survival of babies born at 22 and 23 weeks and better survival when birth occurs at 24 and 25 weeks gestation. The rate of survival to discharge of babies live born over five years from 2000-2005 at 22, 23,34 and 25 weeks gestational age in the period 2000-2005 can be calculated as 0%, 11%, 18% and 49%. The one year survival rates for babies born in England & Wales in 2005, the last year of the studied by Field and his colleagues, were observed to be 5%, 16%, 42% and 65% for babies born respectively at 22, 23, 24 and 25 weeks gestational age (2). Although not comparable directly, these rates of survival over a longer time from birth, for a much larger area, and obtained with different methods have the same intriguing pattern of apparent improved survival over time for babies born after 24 weeks, but little difference for babies born at earlier gestations. With such large differences in survival from one week to the next the effect of gestational age misclassification is an important consideration. In the national study concern about the accuracy of recording of gestational ages was raised, especially as this affects indices at the lower extreme of gestational age at birth (2). In the Trent study measures were taken to reduce the possibility of transcription errors and the method of gestational age ascertained was described. The validity of nominating certain LMP dates as a more reliable method of assessing gestational age than early ultrasound scanning is questionable(3), but allows for a more consistent assessment of gestational age over the course of t he study where the availability of dating ultrasound scans changed from ‘perhaps 50%’ to ‘most’. The starting point for the analysis undertaken by Field et al was babies alive at the onset of labour. In many more pregnancies ending at 22-25 weeks fetal death occurred before this, or the timing of fetal demise is not recorded. Fetal losses at this advanced stage of pregnancy are as traumatic for women and their families as fetal or infant deaths later in the pregnancy. One important consequence of efforts to provide gestational age statistics for England & Wales is the potential to contribute to high quality surveillance and timely recognition of changes in survival at the limits of viability. There are as many births in England & Wales annually as in the twelve years needed for the analysis of trends in extremely preterm babies in Trent. A particular challenge for such study is the fact that no one system reliably captures all the relevant data needed for surveillance and to monitor strategies aimed at preventing these very preterm births. References 1. Field DJ, Dorling JS, Manktelow BN, Draper ES. Survival of extremely preterm babies in a geographically defined population: prospective cohort study of 1994-9 compared with 2000-5. BMJ http://www.bmj.com/cgi/content/full/bmj.39555.670718.BE. Accessed 9 May 2008. 2. Moser K, Macfarlane A, Chow YH, Hilder L, Dattani N. Introducing new data on gestation-specific infant mortality among babies born in 2005 in England and Wales. Health Statistics Quarterly 2007;35:13-27 www.statistics.gov.uk/downloads/theme_health/HSQ35.pdf 3. Bricker L, Garcia J, Henderson J, Mugford M, Neilson J, Roberts M, Martin M-A. Ultrasound screening in pregnancy: a systematic review of the clinical effectiveness, cost-effectiveness and women’s views. Health Technology Assessment 2000; Vol. 4: No. 16 Competing interests: None declared
Survival rates in very preterm babies in the Marche region 13 June 2008
Roberta Buglia, SpR in Pediatrics Division of Neonatology Salesi Hospital Polytechnical University of Marche, 60123 Ancona, Italy, Annalisa Pedini, Roberta Freddara and Virgilio P. Carnielli Send response to journal: Re: Survival rates in very preterm babies in the Marche region	We read with great interest the article by David J. Field and coll. (1) about survival of extremely premature babies in a geographically defined population and the letter by Lisa Hilder (2) stressing among other points the importance of assessing gestational age by early ultrasound. We would like to underscore the importance of area based data collection and we wish to report our experience from the Marche region (centre Italy) where a prenatal network was started in 2003 with a comprehensive data collection system including maternal, foetal and neonatal variables. Data collection was performed for all pregnancies from 22 weeks and 0 days onwards by the local obstetrician; the local neonatologists and verified by an external reviewer. In our region gestational age was assessed by early ultrasound in more than 85% of pregnancies. In 2003-2007 of the 69025 live births of the Marche region 109 occurred between 22-26 weeks, 70 (64%) were the infants alive at the onset of labour and 56 (51%) were admitted alive to Neonatal Intensive Care Unit. Stillbirth rate (death before labour) was 31%. Deaths in labour were 6% (9% of infants alive at the onset of labor), and death in the delivery room 7% (13% of live births). These figures appear to be lower than those reported for the Trent region where death in labour was 18% and death at delivery were 27% (1). It is thus of importance that these figures should be given when evaluating regional survival rates. Infants from 22 to 25 wks who survived to discharge were 48% with survival rates of 0, 10, 78, 52% survival rates at 22, 23, 24 and 25 weeks respectively. The EPICure study (3) report 34% and 52% survival rates (% of infants admitted in Neonatal intensive Care Unit) at 24 and 25 weeks gestation. The EPIpage study (4) report 64% and 79 % survival (% of infants admitted in Neonatal intensive Care Unit) at 24 and 25 week repectively however deaths in the delivery room was 38% (22-25 weeks). We underscore the importance of area based data collection, the importance of having gestational age assessed by early ultrasound, and we suggest that any epidemiological studies involving infants at the limit of viability should contain data on intrauterine death, deaths during delivery, and deaths in delivery room. 1.Field DJ, Dorling JS, Manktelow BN, Draper ES. Survival of extremely preterm babies in a geographically defined population: prospective cohort study of 1994-9 compared with 2000-5. British Medical Journal 2008; 336: 1221-1223 2. Hilder L., Improving survival is good, but preventing extremely preterm birth is better. British Medical Journal 2008; Rapid responses18 May 2008 3. Costeloe K, Hennessy E, Gibson AT, Marlow N, Wilkinson AR. The EPICure study: outcomes to discharge from hospital for infants born at the threshold of viability. Pediatrics 2000;106:659-71 4. Larroque B, Breart G, Kaminski M, Dehan M, Andre M, Burguet A, et al. Survival of very preterm infants: Epipage, a population based cohort study. Arch Dis Child Fetal Neonatal Ed 2004;89:F139-44 Competing interests: None declared
Survival of extremely preterm babies, supersedes earlier response by Lisa Hilder, who has withdrawn it 8 July 2008
Alison J Macfarlane, Professor of Perinatal Health City University, 20 Bartholomew Close, London EC1A 7QN, Lisa Hilder Send response to journal: Re: Survival of extremely preterm babies, supersedes earlier response by Lisa Hilder, who has withdrawn it	Survival of extremely preterm babies We read with interest the analysis of the trends in survival of babies born extremely preterm to mothers normally resident in the Trent Region.1 Although there are differences in detail, our analysis of survival rates to the age of one year among all 645,887 babies born alive in England and Wales in 2005 give a consistent picture, as the Table shows.2,3 This compares, as far as possible, the data from the two analyses, omitting deaths during labour from the Trent totals. The differences between reported survival rates can be seen to be compatible with random variation. As 2005 was the first year for which data about gestational age have been available for England and Wales no data are available for comparison with past years as was the case in the Trent study. It is notable, however, that survival rates among babies born at 24 and 25 weeks of gestation are higher than those reported from UK studies undertaken in the 1990s.4,5 As in the Trent study, survival rates were much lower at earlier gestational ages. With such wide differences in survival, the potential effect of gestational age misclassification is an important consideration. In the Trent Region study, measures were taken to reduce the possibility of transcription errors. The strategy for ascertaining gestational age, was consistent over the course of the study, although no detailed results were provided. In contrast, our data were derived from linkage of data derived from routine clinical and civil registration practice. The gestational ages were those reported by maternity units to the NHS Numbers for Babies system.6 The accuracy of recorded gestational ages of 22 and 23 weeks in our study is a particular concern. Checks carried out found internal inconsistencies in the data for babies recorded as being born before 22 weeks of gestational age. Births where the gestational age was recorded as less than 22 weeks and birthweights as 1,000g or over were therefore excluded from our analyses. This was not done for births with gestational ages of 22 weeks and above, however, so similar errors may therefore remain. The effect of including these births is to increase apparent survival and is likely to be more marked in the 22 and 23 week groups where the numbers of live births were relatively small. In addition, some of our registered deaths may have failed to link with births, and so our survival rates could be slightly inflated. Despite their very poor prospects, we would argue that larger multiple ascertainment studies are needed before accepting that survival is impossible for babies born at 22 weeks. We have no data to validate the vital status or condition of the babies who appeared to have survived in our study. Many of the data in the Trent Region relate to small numbers of births, even though data were aggregated for six year periods. Larger population aggregates are needed to provide more timely recognition of changes. The Office for National Statistics is now compiling these data routinely to monitor trends in future years and this system has the potential to contribute to high quality national surveillance of changes in survival at the limits of viability. It will also include fetal deaths at 24 or more weeks of gestation registrable as stillbirths, but there is also a need for continuing voluntary monitoring of late fetal deaths at earlier gestations. This is necessary to shift the emphasis of such monitoring to include prevention of very preterm labour. Table Comparison of survival rates of very preterm babies born alive in the Trent Region and in England and Wales as a whole. Alison Macfarlane Lisa Hilder Department of Midwifery, City University London, 20 Bartholomew Close, London EC1A 7QN References 1. Field DJ, Dorling JS, Manktelow BN, Draper ES. Survival of extremely preterm babies in a geographically defined population: prospective cohort study of 1994-9 compared with 2000-5. BMJ http://www.bmj.com/cgi/content/full/bmj.39555.670718.BE. Accessed 9 May 2008. 2. Moser K, Macfarlane A, Chow YH, Hilder L, Dattani N. Introducing new data on gestation-specific infant mortality among babies born in 2005 in England and Wales. Health Statistics Quarterly 2007;35:13-27 www.statistics.gov.uk/downloads/theme_health/HSQ35.pdf 3. Office for National Statistics. Preterm births, England and Wales, 2005, London: Office for National Statistics, 2007. Available online at: www.statistics.gov.uk/StatBase/Product.asp? vlnk=14882&Pos=&ColRank=1&Rank=272. 4. Saigal S, Doyle LW. An overview of mortality and sequelae of preterm birth from infancy to adulthood. Lancet 2008; 371: 261-269. 5. Moser K, Macfarlane A, Dattani N. Survival rates in very preterm babies in England and Wales. Letter. Lancet 2008; 371: 896-7.897. 6. Moser K, Hilder L. Assessing the quality of NHS numbers for babies data and providing gestational age statististics. Health Statistics Quarterly 2008; 37: 15-23. Competing interests: None declared