Very-Low-Birthweight Neonates: Do

Outcomes Differ in Multiple Compared with

Singleton Gestations?
Edward J. Hayes, M.D.,1 David Paul, M.D.,2 Amen Ness, M.D.,1
Amy Mackley, R.N.,2 and Vincenzo Berghella, M.D.1

ABSTRACT

The purpose of this study is to determine if outcomes for very-low-birthweight

(VLBW) neonates differ in multiple versus singleton gestations. This is a retrospective
cohort study of neonates weighing less than 1500 g admitted to a neonatal intensive care
unit from 1993 to 2004. Outcome variables were necrotizing enterocolitis, death, and/or
severe intraventricular hemorrhage (IVH). Statistical analysis included univariate and
multivariate analysis. During the study period, 1769 VLBW infants including 465 multiples and 1304 singletons were identified. Gestational age and birthweight were similar;
conversely white race (68% multiples versus 43% singletons), maternal age (28.7
5.7
versus 26.1
6.5 years), born at facility (95% versus 86%), antenatal steroids (74% versus
58%), preeclampsia (14% versus 24%), and preterm labor (74% versus 62%) were
significantly different. Correcting for these, VLBW multiples had a higher odds ratio
(OR) of death and/or severe IVH, OR 1.4 (1.031.95). In our population, VLBW multiple
gestations were at elevated odds for death and/or severe IVH compared with VLBW
singletons.
KEYWORDS: Very low birthweight, multiple gestations, twins, intraventricular

hemorrhage

he percentage of live births that are multiple

gestations in the United States has increased significantly over the last 20 years, with a 65% increase in twins
and a 500% increase in triplets and higher order births.1
This rise has been linked to the enhanced use of assisted
reproductive technologies and the increasing maternal
age at the time of pregnancy. Preceding this recent
increase, the natural incidence of twins and triplets in
the United States as reported in 1973 was 1 in 80 and 1
in 800 respectively.2 Currently, multiple gestations comprise 3.3% of the total births, of which 3.1% of births are
twins and 0.2% are triplets or higher order multiples.3

Although accounting for less than 5% of total births,

multiples account for 30% of very-low-birthweight
(VLBW) infants and nearly 20% of infant mortality.4
Infants of multiple gestations who survive the neonatal
period have a higher incidence of long-term morbidity.
The rate of cerebral palsy increases in proportion to the
number of fetuses, varying from 1.6 per 1000 in singletons to 7.3 and 28 per 1000 for twins and triplets,
respectively.5
Preterm birth is the principal cause of the increased morbidity and mortality associated with multiple gestations. The average gestational age at delivery

1
Division of Maternal Fetal Medicine, Department of Obstetrics and
Gynecology, Thomas Jefferson University, Obstetrics and Gynecology,
Philadelphia, Pennsylvania; 2Division of Neonatology, Department of
Pediatrics, Christiana Care Health Services, Newark, Delaware.
Address for correspondence and reprint requests: Dr. Edward J.
Hayes, 834 Chestnut Street, Suite 400, Philadelphia, PA 19107.

Am J Perinatol 2007;24:373376. Copyright # 2007 by Thieme

Medical Publishers, Inc., 333 Seventh Avenue, New York, NY
10001, USA. Tel: +1(212) 5844662.
Accepted: March 28, 2007. Published online: June 13, 2007.
DOI 10.1055/s-2007-981852. ISSN 0735-1631.

373

374

AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 24, NUMBER 6

decreases as the number of fetuses increase, from 40

weeks for singletons, to 35.4 weeks for twins, to 32.0
weeks for triplets.1 Iatrogenic preterm delivery is also
more common in multiple gestations due to a higher
rate of maternal and fetal complications. Many pregnancy-related disease states, such as preeclampsia, are
seen more frequently as fetal number increases,6 and
diseases unique to multiple gestations, such as twin-totwin transfusion syndrome, may result in a preterm
delivery.
Because multiple gestations have become an increasingly greater proportion of premature births, it is
crucial to determine if the outcomes of these premature
infants at a given birthweight parallel singleton neonates. Weight-based neonatal outcomes play an important role in patient counseling and may have implications
in clinical management. Therefore, the goal of this study
was to determine if outcomes differ for VLBW neonates
at a given birthweight in multiple and singleton gestations.

MATERIALS AND METHODS

We conducted a retrospective cohort study of neonates
with a birthweight of less than 1500 g admitted to the
single level-three nursery, Christiana Medical Center, in
the state of Delaware. Data were obtained from a
computerized database maintained on all neonatal intensive care unit (NICU) admissions. Data were prospectively entered into the database from the medical
record by trained reviewers. The Institutional Review
Board approved the formation of a VLBW infant database derived from the neonatal database for all infants
admitted to the intensive care nursery (ICN) since July
1993. This database is periodically checked for accuracy.
All VLBW infants admitted to the ICN from July
1993 until July 2004 were analyzed. Twin-to-twin transfusion was determined by clinical diagnosis based on
ultrasound findings of monochorionic twins with asymmetric fluid distribution and growth. This study specifically focused on neonatal death, severe intraventricular
hemorrhage (IVH), and necrotizing enterocolitis
(NEC). Death in this population was defined as death
before discharge from this unit. Intraventricular hemorrhage was determined via cranial ultrasounds routinely
obtained on the fourth day of life and then monthly until
discharge. Cranial ultrasounds were obtained more frequently if clinically indicated. Ultrasounds were done
using a 7.5-MHz transducer, and studies were interpreted by a pediatric radiologist. Intraventricular hemorrhage was graded using the classification system of
Papile.7 Severe IVH was considered grade IIIIV. Because death and severe IVH are competing variables, the
outcome of death and/or severe IVH was studied.
Necrotizing enterocolitis was defined as a minimum of
Bell stage 2.8

2007

Statistical analysis included both univariate and

multivariate analyses. Univariate analysis included oneway analysis of variance (ANOVA) for continuous data
that were normally distributed, Mann-Whitney U test
for ordinal data or continuous data, which was not
normally distributed, and the chi-square test for dichotomous variables. Multivariate analysis included
logistic regression. Any variable with P < 0.15 on univariate analysis, or variables that are known to be
confounding with the dependent variables were entered
into the models. A P < 0.05 was considered significant.
All data are expressed as mean
standard deviation
unless otherwise noted. All statistical calculations were
performed using Statistica v7.0 (StatSoft Inc., Tulsa,
OK).

RESULTS
The study sample included 1769 VLBW infants admitted from July 1993 until July 2004. In the study
sample, 465 were multiple gestations (353 twins,
75 triplets, 32 quadruplets, and 5 quintuplets) and
1304 were singletons. Of the 465 multiple gestations,
206 (43%) weighed < 1000 g, and 86 (18%) weighed
< 750 g. The 1304 singletons had similar proportions,
with 578 (44%) weighing < 1000 g and 262 (20%)
weighing < 750 g. An analysis of maternal and infant
demographics revealed that mothers of multiple gestations were more likely to be white, to be of advanced
maternal age, to have received antenatal steroids, and to
have been diagnosed with preterm labor, whereas singleton mothers were more likely to be diagnosed with
preeclampsia. The neonates of multiple gestations had a
higher probability of being born at Christiana Medical
Center than a smaller community hospital in comparison
to the VLBW singletons (Table 1). Among the multiples, 29 babies were diagnosed with twin-to-twin transfusion syndrome before delivery, of which nine
subsequently died.
Table 1 Maternal and Infant Demographics*
Multiple
Gestations

Singletons

P Value

28.2
2.9

28.0
2.9

Birthweight (g)

1039
277

1035
289

0.80

White

68%

43%

< 0.01y

EGA (wks)

0.15

26.1
6.5

< 0.01y

Born at CMC

95%

86%

< 0.01y

Antenatal steroids

74%

58%

7%

9%

14%
74%

24%
62%

Maternal age (y)

Chorioamnionitis
Preeclampsia
Preterm labor

28.7
5.7

< 0.01y
0.12
< 0.01y
< 0.01y

*Comparisons were made with the chi-square test for categorical

variables and with analysis of variance for continuous variables.
y
Significant.
EGA, estimated gestational age; CMC, Christiana Medical Center.

COMPARISON OF MULTIPLE TO SINGLETON NEONATE GESTATIONS/HAYES ET AL

Table 2 Univariate Analysis of Main Outcomes*

Multiple
Gestations (%)

Singletons
(%)

Odds Ratio
(95% CI)

Death

15

14

1.1 (0.91.5)

Severe IVH

12

11

1.1 (0.81.6)

24

21

1.1 (0.91.5)

1.1 (0.71.7)

(grade 34)
Death and/or
severe IVH
NEC

*Chi-square test was used for the analysis.

CI, confidence interval; IVH, intraventricular hemorrhage; NEC,
necrotizing enterocolitis.

Univariate analysis revealed no significant difference between the two groups in the occurrence in NEC,
death, severe IVH, or severe IVH and/or death
(Table 2). However, after controlling for those demographic variables determined to be significant between
the two populations as reported in Table 1, multiple
gestation infants weighing < 750 g had significantly
increased odds of death (54% (46/86) versus 42% (112/
262)). When the combined competing variables of death
and severe IVH were analyzed, all weight categories
showed a higher odds ratio of this outcome (Table 3).
A secondary analysis was performed after eliminating higher order multiples (triplets and above) and
included 353 twins compared with 1304 singletons.
After controlling for potential confounding variables,
all twins remained at increased odds of death and/or
severe IVH (Table 3).

COMMENT
Multiple gestations comprise a large proportion of
VLBW infants in todays ICNs. In our population of

VLBW infants, multiple gestation infants were at significantly increased risk of death and/or severe IVH
compared with singletons. Furthermore, the subgroup
of infants weighing < 750 g had an increased occurrence
of death after adjusting for potential confounding variables.
These findings are in contrast to several earlier
studies of morbidity and mortality of multiple gestations compared with singletons. Spellacy et al,9 in 1990,
reported that twins born at  2500 g had lower perinatal mortality rates compared with same-weight singleton gestations, leading him to conclude that twins
mature at an earlier gestational age.9 However, several
more recent studies have shown conflicting results. In
1997, Friedman et al10 reported that twins born between 24 and 34 weeks gestation had similar outcomes
to singletons. They concluded that twin infants do not
have accelerated maturation nor improved neonatal
outcome compared with matched singletons. Friedmans conclusion was recently supported by Garite
and colleagues11 via their large multicenter study,
which showed that premature singletons, twins, and
triplets all had similar neonatal outcomes. The finding
of these two recent studies are significant, but limited in
application to VLBW neonates because the average
weight in both studies was > 1500 g. Nevertheless,
their findings do demonstrate that recently born multiple-gestation infants do not have accelerated maturation, leading to improved neonatal outcomes, compared
with singletons.
Consistent with our study, Nielsen and colleagues12 showed that multiple-gestation infants born
at 27 to 29 weeks were more likely than weight-matched
singletons to have one of the following complications
of prematurity: patent ductus arteriosus, IVH, NEC,
or retinopathy. They therefore concluded that at

Table 3 Multivariate Analysis, Logistic Regression Models*

Outcome

Weight

Odds Ratio
(95% CI) Multiples
Versus Singletons

Death

< 1500 g

1.3 (0.91.9)

1.3 (0.91.9)

< 1000 g

1.5 (0.92.2)

1.5 (1.02.3)

< 750 g

1.9 (1.13.5)y

1.9 (1.13.4)y

< 1500 g

1.2 (0.81.7)

1.2 (0.91.8)

< 1000 g
< 750 g

1.1 (0.81.7)
1.1 (0.62.1)

1.2 (0.81.8)
1.3 (0.62.2)

< 1500 g

1.4 (1.02.0)y

1.4 (1.02.0)y

< 1000 g

1.7 (1.12.4)y

Severe IVH (grade 34)

Death and/or severe IVH

NEC

1.6 (1.12.3)

Odds Ratio
(95% CI) Twins
Versus Singletons

< 750 g

2.3 (1.34.2)

2.4 (1.34.3)y

< 1500 g

1.1 (0.71.8)

1.0 (0.42.7)

< 1000 g

1.0 (0.61.7)

0.8 (0.41.6)

< 750 g

0.8 (0.32.1)

0.8 (0.32.1)

*Models were controlled for gestational age, inborn status, steroids, race, preeclampsia, and preterm labor.
y
Significant.
CI, confidence interval; IVH, intraventricular hemorrhage; NEC, necrotizing enterocolitis.

375

376

AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 24, NUMBER 6

these gestational ages multiples have significantly higher

morbidity than age-matched singletons.
The strengths of our study include the relatively
large population of VLBW neonates, 1769 overall,
which provided a sufficient number to examine multiple
outcomes while attempting to control for confounding
variables. This study also benefits from the unique
relationship this tertiary center has with the state of
Delaware. Because it is the only level-three neonatal
center for the state of Delaware providing care to both
urban centers and rural communities, it inherently provides a diverse population of neonates. Also, because all
neonates in our study were cared for in this single
institution, the neonatal management between the two
groups can be assumed to be similar.
A retrospective cohort study inherently has its
intrinsic weaknesses, which we attempted to control. To
minimize selection bias, all infants meeting weight
criteria admitted to the NICU in the defined time period
were included. However, the database contained only
neonates who were admitted to the ICN. Those infants
who did not survive from the delivery room were not
identified and were therefore excluded from the analysis.
These neonates may have changed our outcomes. In our
study design, attempts were made to identify and control
for multiple confounding variables. Nevertheless there
may have been other important obstetrical differences
between the group that were not recognized or controlled for during our analysis.
The reason for the higher rate of death and/or
IVH seen in multiple-gestation infants weighing
< 1500 g and death in those weighing < 750 g could
not be determined from this study. Whether VLBW
multiple-gestation neonates are inherently more fragile, as in the case of twin-to-twin transfusion, or are
less responsive to therapeutic intervention, either in the
antenatal or the neonatal period, could not be determined. Several recent theories in the literature may
explain our outcome difference between multiple and
singleton gestations. One proposed theory by Linder
et al13 is that fertility treatments, particularly in vitro
fertilization, which is responsible for a substantial
proportion of multiples, may be a risk factor for IVH
in the VLBW premature infant. Other theories implicate the possible differential effect of antenatal corticosteroids on premature singletons compared with
multiples. These vary in speculating that multiples are
not stressed enough to mount a beneficial response to
antenatal corticosteroids14 or there may be decreased
bioavailaility15 of betamethasone in multiples due to
the increased number of fetuses. Whatever the mech-

2007

anism for the increase in adverse outcomes noted in our

study, given the increasing proportion of VLBW neonates resulting from multiple gestations, further research is required to determine the causes of these
differences and whether they are amenable to therapeutic intervention.

REFERENCES
1. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Menacker
F, Munson ML. Births: final data for 2002. Natl Vital Stat
Rep 2003;52:1113
2. Benirschke K, Kim CK. Multiple pregnancy. 1. N Engl J
Med 1973;288:12761284
3. PeriStats: Your online source for perinatal statistics. Available
at: www.MarchofDimes.com/peristats. Accessed August 1,
2006
4. Magee BD. Role of multiple births in very low birth weight
and infant mortality. J Reprod Med 2004;49:812816
5. Petterson B, Nelson KB, Watson L, Stanley F. Twins,
triplets, and cerebral palsy in births in Western Australia in
the 1980s. BMJ 1993;307:12391243
6. Wen SW, Demissie K, Yang Q, Walker MC. Maternal
morbidity and obstetric complications in triplet pregnancies
and higher-order multiple pregnancies. Am J Obstet Gynecol
2004;191:254258
7. Papile LA, Munsick-Bruno G, Schaefer A. Relationship of
cerebral intraventricular hemorrhage and early childhood
handicaps. J Pediatr 1983;103:273277
8. Bell MJ, Ternberg JL, Feigin RD, et al. Neonatal necrotizing
enterocolitis. Therapeutic decisions based upon clinical
staging. Ann Surg 1978;187:17
9. Spellacy WN, Handler A, Ferre CD. A case-control study of
1253 twin pregnancies from 19821987 perinatal data base.
Obstet Gynecol 1990;75:168171
10. Friedman SA, Schiff E, Kao L, Kuint J, Sibai BM. Do twins
mature earlier than singletons? Results from a matched
cohort study. Am J Obstet Gynecol 1997;176:11931199
11. Garite TJ, Clark RH, Elliott JP, Thorp JA. Twins and
triplets: the effect of plurality and growth on neonatal
outcome compared with singleton infants. Am J Obstet
Gynecol 2004;191:700707
12. Nielsen HC, Harvey-Wilkes K, MacKinnon B, Hung S.
Neonatal outcome of very premature infants from multiple
and singleton gestations. Am J Obstet Gynecol 1997;177:
653659
13. Linder N, Haskin O, Levit O, et al. Risk factors for
intraventricular hemorrhage in very low birth weight
premature infants: a retrospective case-control study. Pediatrics 2003;111(5 Pt 1):e590e595
14. Jobe AH. Antenatal steroids in twins. [letter]. Am J Obstet
Gynecol 2003;188:856
15. Murphy DJ, Caukwell S, Joels LA, Wardle P. Cohort study
of the neonatal outcome of twin pregnancies that were
treated with prophylactic or rescue antenatal corticosteroids.
Am J Obstet Gynecol 2002;187:483488