Downloaded from http://adc.bmj.com/ on June 14, 2017 - Published by group.bmj.com
Archives of Disease in Childhood 1993; 68: 326-329
326
PERSONAL VIEW
Should we look after babies less than 800 g?
N R C Roberton
It is a decade since Professors David Milner
and Richard Beard produced data to show
that 39% of babies born at 25 weeks' gestation and 16% of those born at 24 weeks'
gestation were surviving in the major neonatal
intensive care units (NICUs) in Britain.'
Translated into birth weight, this and other
contemporaneous data suggested that at least
20-30% of babies weighing 600-800 g were
surviving (V Y H Yu, unpublished data).2 3
With survival figures like that it was doubtful,
even in 1982, whether the title of this article
was a justifiable question. With the considerable improvements in neonatal care in the
subsequent decade, I hope to show that such
a question is now totally unacceptable.
In the past 12 months several papers have
been published adding strength to this view as
they address not only the number of babies
surviving at different birth weights and gestations, but also give data on the neurological
sequelae in those who survive.
Survival
Alberman and Botting gave very low birthweight survival figures by 100 g cohort
between 1983 and 1987, and I am very grateful to them for providing me with an update
on their data for 1988 and 1989.4 The great
merit of their data is that it is national data
for the whole of England and Wales. Table 1
shows that between 1983 and 1989 the number of babies registered, weighing less than
800 g, had increased, though this number
may now be levelling off. The cause of the
increase is a matter of speculation: it may
be that, rather than being recorded as abortions, they have been recognised as live births
and treated as such. It may also be that
obstetricians, increasingly responsive to the
skills of neonatologists, are delivering electively more and more extremely preterm
babies. Not only are more of these babies
being delivered, but more of them are
Table 2 Neonatal survival per 1000 live births in
England and Wales
Birth weight (g)
600-699
Year
500-599
700-799
1983
1984
1985
1986
1987
1988
1989
78
131
150
84
198
163
215
311
331
352
383
403
459
468
145
198
204
253
194
296
367
Data of Alberman and Botting4 and personal communication.
surviving with an overall improvement in
survival of about 90% (table 2).
In a study from Australia, Kitchen et al
reported the survival rate of all babies born
in the state of Victoria and weighing under
1000 g, again by 100 g cohort.5 They found
that, compared with 1977-82, their results in
1985-7 showed no survivors under 600 g, but
about a 50% improvement in babies between
600 and 699 g and 700 and 799 g. National
rates for survival in extremely low birthweight
(ELBW) neonates have been reported very
recently from Sweden.6 Like the British data
100 r
80 -
T
601.E
2
nl
401
T
T
20 [
-
Table 1 Number of live births registered in England and
Wales 1983-9
Birth weight (g)
Rosie Maternity
Hospital, Cambridge
Correspondence to:
Dr N R C Roberton,
Addenbrooke's Hospital,
Hills Road,
Cambridge CB2 2QQ.
Year
500-599
600-699
700-799
Total
1983
154
152
147
178
198
178
172
234
253
304
336
324
361
313
309
332
361
400
412
414
438
697
737
812
914
934
953
923
1984
1985
1986
1987
1988
1989
Data of Alberman and Botting4 and personal communication.
0
Birth weight (g)
Figure 1 Average survival at different birth weights from
the data of Hack et al. 7 The bars represent the range of
survival at different birth weights in the units contributing
to the study.
Downloaded from http://adc.bmj.com/ on June 14, 2017 - Published by group.bmj.com
Should we look after babies less than 800g?
327
they show a marked increase in the absolute
number of ELBW babies in recent years and a
steady decline in mortality with, in 1988, over
30% survival at 600-699 g and 50% at
700-799 g. Only 10% survived at 500-599 g,
the only birthweight cohort in which there has
been no significant improvement in the
Swedish survival rate in the last five years.
In another important study, Hack et al
looked at ELBW survival in the seven NICUs
which comprise the National Institute of
Child Health and the Human Development
Intensive Care Network in the USA.7 The
defect of their paper compared with those of
Alberman and Botting,4 Kitchen et al,5 and
Ericson et al,6 which look at geographically
defined areas, is that the American study just
looked at the inpatient population of level 3
neonatal units, a notoriously unrepresentative
sample. Nevertheless the merit of their paper
compared with the others is that they gave a
range of survival at different birth weights and
gestations within their units, and also gave
data by gestation. Figure 1 shows their data
for birth weight. The overall figures are not
dissimilar to the overall data for England and
Wales in 1989 (table 2), though in general
they are 5-15% (that is, 50-150/1000 live
births better). More interesting is the fact that
some units have very good survival-up to
40% at 501-600 g and 75% at 701-800 g.
These better figures are similar to our own
(ungeographical!) results from the level 3
NICU in Cambridge.8 Precisely what causes
this variation in survival rate is not clear from
their paper, though they do report wide differences in the incidence of several major complications of neonatal intensive care between
units, and furthermore, where North America
is concerned, there is considerable data to
show that low birthweight black babies do
better than white ones.9
More interesting in the data of Hack et al,7
is the analysis by gestation (fig 2). The open
columns show that they are recording significant survival at 23 weeks, and even some survival at gestations of 22 weeks, with as many
as 30% of babies of this gestation surviving in
500-599
600-699
700-799
10-40%
15-50% -
50-75%
*The left hand figure is from national data and the right hand
figure is from the results in a level 3 NICU (see text).
at least one unit. Those are the data based on
the obstetric assessment of gestational age.
When the data are presented in terms of gestational age assessed by neonatal examination
(fig 2, hatched columns) there are no survivors below 24 weeks' gestation.
The data in fig 2 have considerable clinical
implications. They imply that, on the basis of
the obstetric assessment of gestational age
which is, after all, the only figure neonatologists have available when they are summoned
to the labour ward to attend the delivery of a
very premature baby, survival at 22 and 23
weeks' gestation is not uncommon. Furthermore, as neonatal assessment of gestational
age is a singularly inaccurate activity, with a
tendency to overestimate the baby's gestational
age, and a standard deviation of two weeks on
each side of the true age,10 11 the open
columns in fig 2, based on obstetric data, may
actually give a more accurate picture than the
neonatal data.
In summary, table 3 gives survival figures
for babies less than 800 g that can be gleaned
from the literature with the lower figures
representing national data,4 6 and the higher
ones representing the results from individual
high class level 3 NICU.7 8 Readers can draw
their own conclusions.
Furthermore, do not let me give the impression that we should not apply intensive care to
babies less than 500 g. After all, survival at
283 g was reported before the last war,'2
somebody even smaller has recently survived,13 and both were normal on follow up;
many babies weighing 400-499 g at birth are
now surviving.'4
40
Quality of survival
I hope I do not have to explain to readers of
Archives of Disease in Childhood that it is not
the absolute numbers of surviving children
that matters, though this is obviously important, but the quality of their survival. All follow up studies suffer from the unresolvable
problem that a two year follow up may miss
subtle educational deficits, yet by the time
8-10 year follow up is reported the data apply
to a standard of neonatal care that is obsolete
20
Table 4 Outcome indices in ELBW survivors in
Victoria'5
100
F
80
EObstetric
data
T
60
cn
Table 3 Survival figures for ELBW neonates
Birth weight (g)
Survival range*
0-
1979-80
Live births <I 000 g
No (/o):
2 year survivors
Cerebral palsy
Gestation (weeks)
Figure 2 Survival by gestation in the data of Hack et al. 7
The bars represent the range of survival at different
gestations at the units contributing to the study.
Blind
Deaf
Mean MDI
No (%/6) with MDI <69
351
89 (25-4)
12 (13-5)
6 (6-7)
3 (3-4)
88-1
12 (13-5)
MDI=Mental development index.
1985-7
560
212 (37-9)
15 (7-1)
9 (4-2)
1 (05)
98-6
12 (5 7)
Downloaded from http://adc.bmj.com/ on June 14, 2017 - Published by group.bmj.com
Roberton
328
'it
LO£LO
Ir-
LO
100 r
80 F-
60k
40 H
20
0L
600-699
700-799
Weight (g)
Survived
to 2 years
Disability
Figure 3 Outcome in
babies 6fQ799 g in the
Victori4$n)tudy
(figure
modified).'5 For
each birth weight the left
hand side of the
histogram refers to 1979-80
and the right hand to
1985-7.
slightly
or obsolescent. At that age it is
impossible,
whatever the pundits say, to tease out the
effects of social deprivation that affects so
many of these babies from the genuine long
term sequelae of perinatal brain damage.
Bearing these caveats in mind, valuable
data have recently been reported. A study
of all ELBW babies born in the state of
Victoria (table 4, fig 3) shows that with their
increased survival, the absolute number of
children with cerebral palsy and blindness
has also increased, but the proportion of survivors with these unfortunate sequelae has
been reduced.15 This is shown in graphic form
in fig 3 for birth weights less than 800 g.
These data are compatible with much of
what has been published in the last decade
that show that, in general, for babies less
than 1500 g between 10-20% of all survivors
have some disability on long term follow up
with slightly fewer than half having severe
problems.'6 What all studies do show, however, is that neonatal intensive care does
increase the numbers of survivors and
reduce the number of deaths, but has had
little impact on the 10-20% of all live births
at these gestations who end up with disabilities. The most encouraging feature
of the Victorian data is, therefore, that it
shows that with the improved survival as a
result of more aggressive application of intensive care, there has been a reduction
in the proportion of handicapped babies,
and most important of all there has been
a very large increase in the neurologically
normal survivors.
In another population based survey in
Ontario, Canada, Saigal et al'7 showed that
babies weighing less than 1000 g had a similar
incidence of severe neurological handicap to
that reported from Victoria.'5 They also found
the mean developmental quotient of their
patients was about 10 points below the control term population but that this effect was
most dramatic in babies from socioeconomic
class 5, confirming much from the original
studies from Hammersmith 20 years ago.'8
The third recent population based study, a
British one with a high level of ascertainment,'9 also showed that the majority of babies
surviving at gestations less than 28 weeks were
neurologically intact.
As with all data there are different ways of
interpreting it. In fig 3 I have interpreted the
data as showing that 10% of all babies born
show handicap, but of course pessimists and
depressives would interpret the 1979/80
figures as showing that the majority of survivors under 700 g were handicapped and that
even now as many as a quarter to a third of
such patients showed some handicap. For the
optimists, the data in fig 3 clearly show that
with high calibre neonatal intensive care, the
number of handicapped survivors does seem
to be falling, and that, as in all studies, as
neonatal intensive care becomes more successful, the absolute number of neurologically
normal potentially wage earning and marrying
survivors is increasing.
Clinical conclusions
On the basis of these data I believe the case is
proved that we must look after, properly, all
babies of a birth weight greater than 500 g
or of a gestation of 24 weeks or more.
Furthermore, for babies of 22 or 23 weeks'
gestation who are in a satisfactory condition at
delivery, or those whose birth weight is just
below 500 g and who are also in a satisfactory
condition, a long term neurologically intact
survivor is also a distinct possibility, and such
babies should also receive appropriate levels
of care.
Perhaps even more important is the fact that
the courts have accepted figures of survival
similar to those I have given above; stillbirths
must now be registered at 24 weeks and above,
and when this is combined with the implication
of the new Human Fertilization and Embryology Act 1990, that beyond the 24th week of
pregnancy (that is, 168 days) the baby is capable of being born alive and surviving, we have
a clear obligation to such a baby that is no different from one weighing 3500 g at full term.
Practical implications
To achieve survival figures with handicap
rates as low as those I have reported is
demanding on time and resources. British
figures suggest that the cost of looking
after these babies is not prohibitive, about
/10 000-£15 000 per survivor,20 and compares very favourably with the price of a new
Ford Escort or modern high technology
medical activities such as a triple coronary
artery bypass, and is phenomenally cheap
compared with a year of haemodialysis, or
hormone treatment for the menopausal
middle aged.
The political message emanating from these
data is also I believe very clear. Our job as paediatricians, whether we are neonatal specialists
or not, is to fight our corner to ensure that
adequate funding is provided for the care of
these patients. If we do not act as their advocates, it is very unlikely that anyone else will.
Sadly, I believe that the British Paediatric
Association has shown a lamentable lack of
leadership in this area; it is a sad comment on
their contribution to the debate-or indeed
their lack of it that in the last decade the
most useful statements on provision for low
birthweight babies have come either from parliament2l 22 or from the Royal College of
Physicians.23 This journal could also do better.
Editorials implying that resources are limited,
and it is our job as paediatricians to be good
boys and girls and not empire build or demand
more money,24 25 have not helped the cause of
ill low birthweight babies; hopefully this review
will go some way to redress the balance!
While, theoretically of course, funds ultimately are limited, the extent to which the
limitation now applies is a political and not a
medical decision. While we are part of a
health service that spends less on medical care
as a proportion of the nation's gross national
product than almost any other member of the
European Community,26 I do not believe that
Downloaded from http://adc.bmj.com/ on June 14, 2017 - Published by group.bmj.com
Should we look after babies less than 800g?
the ephemeral end point of absolute limitation
of resources has yet been reached, or is even
remotely in sight!
In the new and increasingly unpleasant
health service in which we practice, we are
forced to come to terms with the fact that we
will have to fight for even the most basic and
clearly required services; there will be no place
for those who are not prepared to fight their
corner and to do so vigorously and at the
highest levels.
I hope the data in this paper makes it clear
to those representing paediatricians within the
corridors of power, be they from an association, a college, or merely the humble hard
working executive of the British Association of
Perinatal Medicine, that they have no alternative now but to insist that we are provided
with adequate funds to look after babies
between 500 and 800 g who clearly have an
excellent chance of neurologically intact survival, and for whom we have a clinical, moral,
and indeed legal responsibility.
This paper is based on a presentation to a plenary session at
the British Paediatric Association annual meeting University of
Warwick, April 1992.
1 Milner RDG, Beard RW. Limit of fetal viability. Lancet
1984;i: 1079.
2 Bennett FC, Robinson NM, Sells CJ. Growth and development of infants weighing less than 800 grams at birth.
Pediatrics 1983;71:319-23.
3 Roberton NRC. Annual reports. Cambridge: Cambridge
Matenity Hospital, 1981-1983.
4 Alberman E, Botting B. Trends in prevalence and survival
of very low birthweight infants, England and Wales
1983-7. Arch Dis Child 199 1;66:1304-8.
5 Kitchen WH, Doyle LW, Ford GW, et al. Changing two
year outcome of infants weighing 500-999 grams at birth.
A hospital study. Jf Pediatr 199 1;118:938-43.
6 Ericson A, Gunnarskog J, Kallen B, Olausson PO. A registry study of VLBW live born infants in Sweden.
1973-1988. Acta Obstet Gynecol Scand 1992;71:104-1 1.
329
7 Hack M, Horbar JD, Malloy M, Tyson JE, Wright E,
Wright L. Very low birthweight outcomes of the National
Institutes of Child Health and Human Development
Neonatal Network. Pediatrics 1991 ;87:587-9.
8 Annual reports. Cambridge: Rosie Maternity Hospital,
1989, 1990, 1991.
9 Wilcox A, Russell I. Why small black infants have a lower
mortality rate than small white infants: the case for population-specific standards for birthweight. Jf Pediatr 1990;
116:7-10.
10 Sanders M, Allen M, Alexander GR. Gestational age
assessment in preterm neonates weighing less than
1500 g. Pediatrics 199 1;88:542-6.
11 Gagliardi L, Scimone F, Del Prete A, et al. Precision of
gestational age assessment in the neonate. Acta Paediatr
Scand 1992;81:95-9.
12 McWhirter ND. Guinness book of records. 31 st Ed. London:
Guinness Books, 1985:14-5.
13 Muraskas JK, Carlson NJ, Halsey C, Frederiksen MC,
Sabbagha RE. Survival of a 280 g infant. N Engl Jf Med
199 1;324: 1598-9.
14 Fomufod AK. The lower limits of viability: outcome of
infants with birthweights less than 500 G. Proceedings of
the XI congress of perinatal medicine. Rome, Italy, 1988:
260.
15 Victoria Infant Collaborative Group. Improvement of outcome for infants of birth weight under 1000 g. Arch Dis
Child 1991;66:765-9.
16 Escobar GJ, Littenberg B, Petitti DB. Outcome among
surviving very low birthweight infants: a meta-analysis.
Arch Dis Child 199 1;66:204-1 1.
17 Saigal S, Szatmari P, Rosenbaum P, Campbell D, King S.
Cognitive abilities and school performance of extremely
low birthweight children and matched term control children at age 8 years. A regional study. J Pediatr 1991;
118:751-60.
18 Francis-Williams J, Davies PA. Very low birthweight and
later intelligence. Dev Med Child Neurol 1974;16:709-28.
19 Wariyar U, Richmond S, Hey E. Pregnancy outcome at
24-31 weeks gestation-neonatal survivors. Arch Dis
Child 1989;64:678-86.
20 Tubman TRJ, Halliday HL, Norman C. Cost of surfactant
treatment for severe neonatal respiratory distress syndrome. A randomised controlled trial. BMJ 1990;301:
842-5.
21 Social Services Committee. Perinatal and neonatal mortality.
2nd report. London: HMSO, 1980.
22 Social Services Committee. Perinatal and neonatal mortality.
3rd report. London: HMSO, 1984.
23 Royal College of Physicians. Medical care of the newborn in
England and Wales. London: RCP, 1988.
24 Davies PA. Perinatal mortality. Arch Dis Child 1980;55:
833-7.
25 Campbell AGM. '... officiously to keep alive'.
Commentary. Arch Dis Child 1988;63:565-6.
26 Anonymous. BMA rises to challenge of Europe. BMA news
review 1992; Jan: 20-1.
Downloaded from http://adc.bmj.com/ on June 14, 2017 - Published by group.bmj.com
Should we look after babies less than 800 g?
N R Roberton
Arch Dis Child 1993 68: 326-329
doi: 10.1136/adc.68.3_Spec_No.326
Updated information and services can be found at:
http://adc.bmj.com/content/68/3_Spec_No/326.citation
These include:
Email alerting
service
Receive free email alerts when new articles cite this article. Sign up in the
box at the top right corner of the online article.
Notes
To request permissions go to:
http://group.bmj.com/group/rights-licensing/permissions
To order reprints go to:
http://journals.bmj.com/cgi/reprintform
To subscribe to BMJ go to:
http://group.bmj.com/subscribe/