![]() |
|||||||||||||||||||||
|
|||||||||||||||||||||
Infant and perinatal outcomes of triplet pregnancy in
Auckland: better than expected?
Malcolm Battin, Michelle Wise, Anne DeZoete, Peter Stone
The triplet birth rate has generally increased worldwide
since the 1970s.1,2 Two important factors
associated with this are the tendency towards increased maternal age, which may
be associated with higher rates of spontaneous multiple
births,3 and the rising use of medical
assistance to become pregnant.4–7
Locally, there has certainly been a significant increase over time in the
proportion of multiple pregnancies conceived following fertility
treatments.8
A triplet pregnancy has significant implications for the
mother, the infant, the family,9 and society as
a whole.10 Triplet pregnancies are reported to
have high rates of complications,11–13
the most common of which is preterm
delivery.12,13 Indeed, the recent literature on
triplet pregnancies reports delivery to occur at a mean gestation of
approximately 32–34 weeks.4–6,12,13
Other important pregnancy complications that occur more
frequently are preeclampsia,13,14 excessive
postnatal haemorrhage11,13 and growth
restriction, which is associated with an increased rate of perinatal
mortality.15 In addition there is a two-fold
increase in risk of maternal mortality over singleton pregnancies.
Postnatally, there are reports of an increased rate of
neonatal morbidity including respiratory distress syndrome
(RDS),5,12 intraventricular haemorrhage
(IVH)5,12 and retinopathy of prematurity
(ROP).5,16 Lastly triplet pregnancy carries an
increased risk of cerebral palsy (CP).17 In
twins the highest risks for CP are in monochorionic twins, especially in
association with discordant growth or fetal demise of a
co-twin.18 Most triplet pregnancies are
polyzygotic and polychorionic / polyamniotic but some contain a monochorionic
pair.
The combination of the growing number of triplet pregnancies
and the potential for problems make it important to review the available data on
perinatal and neonatal outcomes. Although there have been reports of neonatal
outcome in recent cohorts of triplets
published,16,19-21 a recent review highlighted
the need for quality neurodevelopmental outcome
data.18 Particularly, there is a lack of data
on neurodevelopmental follow up of triplets born with a birth weight below 1500
grams, i.e. those who may be expected to have the highest mortality and
morbidity.
Accordingly, we have reviewed the neonatal, maternal and
perinatal outcomes of triplet pregnancies born at NWH in Auckland during the
period 1995 to 2005 inclusive. There were two broad aims. Firstly, to provide
local outcome data that would be useful in counselling prospective parents of
triplets. Hence a basic analysis of gestation, birth weight and survival data
are presented for all triplet pregnancies during this period. Secondly, to
address the deficit in accurate contemporary data on neurodevelopmental outcome
and neonatal morbidity for those infants weighing less than 1500 g at birth.
MethodsAll women who had given birth to triplets, at National
Women’s Hospital (subsequently National Women’s Health, Auckland
City Hospital), Auckland, New Zealand during the years 1995–2005, were
identified from the Healthware database. A triplet pregnancy for the purpose of
this study was defined as a pregnancy beyond 20 weeks that lead to registration
of at least one birth. The mothers’ and babies’ paper and electronic
medical records were searched for clinical and demographic details including
pregnancy and neonatal histories.
Pregnancy details included maternal age at conception,
delivery method, gestational age at delivery, conception details, maternal
complications, and antenatal steroid use. Method of conception was defined as
spontaneous or with fertility treatments. Details of any fertility treatment
were recorded including ovulation induction (OI) with Clomiphene Citrate or
Gonadotrophins, Assisted Reproductive Technologies (ART) such as in
vitro fertilisation (IVF) with embryo transfer or other IVF techniques such
as intracytoplasmic sperm injection (ICSI) and gamete intra-fallopian transfer
(GIFT).
Neonatal data included details of resuscitation, sex,
birth weight, neonatal morbidity, neonatal follow up and outcome. All identified
infants were cross referenced with the neonatal database to confirm data. Major
neonatal morbidity
included chronic lung disease (CLD), confirmed
necrotising enterocolitis (NEC), retinopathy of
prematurity (ROP) stage 3 or 4, intraventricular
haemorrhage (IVH) grade 3 or 4, periventricular
leucomalacia, porencephalic cyst, or
hydrocephalus. CLD was defined as the need for
respiratory support (oxygen, CPAP, or
ventilation) at 36 weeks of corrected postmenstrual
age.
Infants with a birth weight below 1500 g are enrolled
in a follow-up programme when they graduate from the NICU, including physical
examination and formal developmental assessment using the Bayley scales of
infant development22 at 18 months of age. In
addition to Mental Developmental Index (MDI) and Psychomotor Developmental Index
(PDI), outcome category is formally allocated on a scale of 1-4, according to
published criteria,23 where 4 is normal and 1
severely abnormal.
The outcome for infants with a birth weight above 1500
g is generally good thus their follow up arrangements will vary and includes:
neonatologist clinic, at a tertiary centre; local paediatric follow; GP; and
Bayley assessment is only performed if requested. Infants undergo routine formal
hearing assessments and ophthalmological assessments as indicated from gestation
and clinical course.
Data are presented as mean and standard deviation if
normal or median (range) if not normally distributed. Incidences were compared
by Chi squared or Fisher’s exact test as appropriate. Pregnancy and
delivery data is presented as a percent of total triplet pregnancies whilst
neonatal data is presented as a percent of the number of babies (either total
number or number of neonatal admissions). For the calculations based on the
number of babies (such as birth weight, neonatal morbidity, and perinatal
mortality) any fetus born before 20 weeks gestation was excluded.
ResultsFor the study period, 1995 to 2005 inclusive, 55 triplet
pregnancies were identified and the charts reviewed. The mean age of the women
was 32 years (Table 1). Twenty-five pregnancies (45.4%) were known to be
spontaneously conceived. Thirty-three women (60%) had no major complications
other than multiple pregnancy. All but one set of triplets delivered in the same
way. Eight women (14%) delivered vaginally, and of these, 3 were less than 24
weeks. The number of infants delivered each year varied from 6 to 33 with a peak
in 2000 (Figure 1).
Table 1. Maternal, pregnancy and birth
characteristics
* Includes the set of triplets where Triplet A
delivered vaginally and Triplets B and C by CS; APH=antepartum haemorrhage,
PPH=Post partum haemorrhage (>1000 mL), GDM=gestational diabetes mellitus,
LFTs=liver function tests, LSCS=lower segment caesarean section.
Figure 1. Distribution of triplet pregnancies
by year and showing proportion that were spontaneously conceived.
![]() *Conception data unavailable for 7 triplet
pregnancies
Of the 165 fetuses, 10 died prior to admission to the
neonatal unit. These included three infants from one pregnancy who were liveborn
at 22 weeks gestation but died in the delivery suite; three were stillborn, one
at 27 and two at 29 weeks; and one 23 week infant who weighed 475 grams was not
resuscitated and died in delivery suite. In addition, there were three fetuses
from one pregnancy that spontaneously aborted prior to 22 weeks gestation.
Although two delivered at 19 weeks plus 6 days the third was
delivered at 20 plus 5 days so the pregnancy was included in the current report.
There was also one set of triplets born at 37 weeks gestation who did not
require admission to the neonatal unit. The overall sex distribution was even
with 84 (51%) males and 81(49%) females.
Of the infants actively resuscitated and/or admitted to NICU
(n=152), the median gestational age at birth was 32 (range 23–-35) weeks
and birth weight 1620 g (530–2780) g respectively. Two infants were born
prior to 24 weeks gestation, 17 infants between 24 and 27 weeks plus 6 days, 49
infants between 28 weeks and 31 weeks plus 6 days and 84 infants between 32
weeks and 35 weeks 6 days. At birth 19 (12.5 %) infants required intubation for
resuscitation and the median (range) Apgar score, for the liveborn infants, was
8 (2–10) and 10 (4–10) for 1 and 5 minutes respectively.
There were five neonatal deaths (3.3%), including the one
23-week infant who died in delivery suite, with each of these occurring in
infants born at 25 weeks gestation or less. Thus the overall number (i.e.
including stillbirths, spontaneous abortions and pre viable live born infants)
of perinatal deaths was 14/165, which equated to a crude perinatal mortality
rate for triplet pregnancies of 83/1000 and a neonatal mortality rate of 33/1000
births respectively.
Neonatal course and morbidity—In 35
infants the main respiratory diagnosis was respiratory distress syndrome (RDS);
of these, 28 infants received surfactant. There were also 11 cases of transient
tachypnoea and one case of pulmonary hypertension. The median (range) duration
of ventilation was 0 days (0–70) and median duration of CPAP was 2
(0–317) days.
Respiratory morbidity included six infants who developed air
leak, five with chronic lung disease and one who required home oxygen. Other
neonatal morbidities included one case of proven necrotising enterocolitis and
serious cerebral ultrasound abnormalities in nine infants that included various
combinations of grade 4 intraventricular haemorrhage (five infants),
ventriculomegaly / hydrocephalus (six infants) and cyst formation (five
infants).
Congenital anomalies were present in six infants (4%) and
included one case each of polycystic dysplastic kidney, arachnoid cyst, two
vessel cord, massive facial haemangioma, tracheal oesophageal atresia and
hydrops.
Outcome data—Infants below 1500 g are
routinely followed in clinic and undergo developmental review including Bayley
assessment. Fifty-three infants weighed less than 1500 g at birth, with a mean
gestation of 29 (23–35) weeks and birth weight 1040 (530–1480) g. In
this group, 39 infants had received a complete antenatal steroid course, 10 an
incomplete course and in only four cases no antenatal steroids had been
received. Six infants died, including one at 7 months unrelated to neonatal
period; three developed cerebral palsy, two cases of hemiplegia and one spastic
diplegia; one infant had marked motor delay and one infant hearing impairment
requiring aids.
Bayley scores were performed in 30 infants with a median MDI
95 (71–105) and 94 (65–110). Outcomes were categorised for all
surviving infants below 1500 g as normal (4) in 31/47 (66%) of survivors, mild
abnormality in 8/47 (17%), moderate abnormality in 7/47 (15%) and severely
abnormal in only 1/47 (2%) infants. The one infant born prior to 24 weeks
gestation (23 + 6) who survived the period and had a good neurodevelopmental
outcome on follow-up.
Of the 64 infants above 1500 g some limited follow up
information is available. Twenty-four were normal when last reviewed at a median
of 14 (range 3–24) months and two had died from conditions presumed not to
be related to the neonatal course (SIDS and aspiration secondary to epilepsy
associated with an arachnoid cyst). The median birth weight of this group was
1795 g and median gestation was 33.5 weeks, which would normally be expected to
be associated with a reassuring outcome.
DiscussionIn this paper we report the outcome, including mortality and
neonatal morbidity, for an 11 year cohort of
165 triplets born at a single large perinatal centre in New Zealand. The overall
fetal losses for pregnancies that reached 20 weeks gestation were low with only
9 fetuses resulting in stillbirth, spontaneous abortion or pre-viable delivery.
Furthermore, there were only five neonatal deaths, all of which occurred before
one week of age.
Consistent with our expectation, mortality was focused in
those pregnancies that delivered at the very premature end of the spectrum,
indeed 1/3 cases in infants born before 24 weeks gestation died. In general the
crude perinatal mortality rate of 83/1000 and a neonatal mortality rate of
33/1000 for triplet births compares favourably to reported rates of
41-121/10005,12,13,16,23,24 and
51–595,6 respectively.
Even though neonatal mortality was quite low for a group of
premature infants there was morbidity. Approximately 6% had cerebral ultrasound
abnormality and 3% developing chronic lung disease but below 1% of infants
required home oxygen or developed proven necrotising enterocolitis. These
morbidities with potential long-term consequences were also largely focused in
those pregnancies below 28 weeks gestation. Congenital anomalies occurred
sporadically with a rate that was consistent with that reported from other
studies.21,26
Review of the neurodevelopmental outcome for triplet infants
<1500g revealed over 80 % of the survivors, categorised as normal or only
mildly abnormal (66%+ 17%) and only 1/47 (2%) infant categorised as severely
abnormal. These results were comparable with those reported from a cohort of
triplets from Israel;27 however, that study
reported outcome of a slightly more mature group of triplets with a mean BW of
1660 g and gestation of 32 weeks.
In the New Zealand data the scores encompass a broad range
and the values of 65 and 71 for minimum MDI and PDI respectively represent a
significant degree of impairment. Also three of these infants developed cerebral
palsy with two cases of hemiplegia and one spastic diplegia plus one infant had
marked motor delay and one infant with hearing impairment requiring aids.
There are very limited data available on neurodevelopmental
follow up of triplets with a birth weight below 1500 g but the one report from
Zagreb described the outcome to be normal in 15 infants, CP in seven and minimal
abnormality on neurological testing in seven
infants.19 Although it should be noted that in
the centre reporting this data there was not a specific follow up program for
infants born at this weight and findings were from neurological assessments by a
neurologist so there may be some referral and / or ascertainment bias in that
data.
Although the data described in the current study are from a
single high risk centre so represent a fairly homogeneous pattern of management
there are some limitations. Specifically, there are quite small numbers that
mean it is not possible to perform meaningful comparisons within the cohort such
as the outcomes of triplet pregnancies conceived spontaneously versus those
conceived using assisted techniques.
The other potential limitation is the lack of complete
follow up data on those infants who were not enrolled in the follow up program.
Nevertheless a minimum data set of gestation, birth weight, neonatal survival
and neonatal morbidity were established. Furthermore, if the clinician following
the child considers development to be abnormal then formal assessment can be
performed on request.
The role of fertility treatment in triplet pregnancies is
well recognised.8 Locally there have been
strong moves to decrease the number of multiple births secondary to assisted
fertility.28 During the study period the number
of triplet pregnancies fluctuated from none to 11 per year and there is the
suggestion that numbers are declining since the peak in 2000. This experience is
consistent with that of others with recent UK data on triplet and higher order
births demonstrating that the rate has decreased by one-quarter since
1998.29 NZ has already taken steps to reduce
the number of multiple births following ART by implementing policy limiting the
number of embryos transferred.
It is of interest to explore the relative contribution of
multiple pregnancy and prematurity to any adverse outcome for the infant. Some
authors have compared data obtained from triplets to singletons of same
gestation and report survival and major short term morbidity to be very similar
once controlled for appropriate
variables.4,16,21
Although neonatal stay is reported to be generally longer
for triplets than twins4,30 and in one study
16 there was still an increased rate of mild
IVH and severe ROP associated with triplet pregnancy. A logical extension to
this is to examine the role of selective fetal reduction on subsequent
neurodevelopmental outcome. Although it is well described in the
literature,31–35 the published results in
terms of neurodevelopmental outcome are somewhat contradictory and this
procedure was not performed on any of the triplet pregnancies in the current
cohort.
The Cochrane Review36
concludes there is no strong evidence about the effects of reducing the number
of fetuses in women pregnant with triplets or higher order multiples. A
subsequent systematic review37 suggests that
embryo reduction reduces the rate of preterm delivery, with a number needed to
treat 7 (95%CI: 5–9) but there is an increased rate of miscarriage, with a
number needed to treat of 26 (95%CI: 14–193).
Although this suggests embryo reduction may have a role in
reducing prematurity, there is still a lack of data on the effect, if any, on
subsequent long term neurodevelopmental outcome. One retrospective study has
examined cerebral palsy (CP) rates in a large cohort of trichorionic triplets,
where the decision to have ER or not was determined by parent
choice.38
The mean gestation at birth was 35.6 weeks versus 33.8 weeks
and CP rate was 13.8 versus 18 per 1000 live births respectively for those
managed with reduction to twins or expectant management. The authors suggested
that there was a significant difference in gestation at birth but the rates of
cerebral palsy were similar and concluded there was a need for further
data.38
In summary, this study provides local data on the outcome of
New Zealand triplets. This will be useful not only for counselling prospective
parents of triplets but also in planning services and potentially as a baseline
for monitoring impact of ongoing developments in the care of triplet
pregnancies. In addition, the detailed neurodevelopmental follow up data may be
used to counsel the parents of preterm triplet infants admitted to the neonatal
unit.
Competing interests: None known.
Author information: Malcolm Battin,
Neonatologist, Newborn Services, National Women’s Health, Auckland City
Hospital, Auckland, New Zealand; Michelle Wise, Specialist in Obstetrics and
Gynaecology, The Scarborough Hospital, Toronto, Ontario, Canada; Anne DeZoete,
Psychologist, Newborn Services, National Women’s Health, Auckland City
Hospital, Auckland, New Zealand; Peter Stone, Professor, Department of
Obstetrics & Gynaecology, University of Auckland, Auckland, New
Zealand
Acknowledgements: We thank
Shirley Beer and Marjet Pot for help searching the Healthware databases
as well as Gemma Bryning for data collection and data input.
Correspondence: Dr M Battin, Neonatal
Paediatrics, 9th Floor, Support Building, Auckland City Hospital, Private Bag
92024, Auckland, New Zealand. Fax: +64 (0)9 3754973; email: malcolmb@adhb.govt.nz
References:
|
|||||||||||||||||||||
| Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |