1133

Preliminary experience with a prospective protocol for planned

vaginal delivery of triplet gestations
Vito Alamia, Jr, MD, Anthony B. Royek, MD, Ronald K. Jaekle, MD, and Bruce A. Meyer, MD
Stony Brook, New York
OBJECTIVE: The objective of the study was to evaluate a protocol for vaginal delivery of triplet gestations.
STUDY DESIGN: All women with triplet gestations managed between January 1, 1995, and December 31,
1997, by University Medical Centers perinatal practice were offered enrollment in our vaginal delivery proto-
col. Our protocol offered attempt of vaginal delivery if triplet A was in vertex presentation, fetal monitoring
was possible, and there were no other obstetric contraindications. Twenty-three triplet gestations were identi-
fied; 8 achieved vaginal delivery. Outcome parameters investigated included neonatal mortality, Apgar
scores, neonatal intracranial hemorrhage, arterial cord pH, neonatal weight, and length of postpartum hospi-
tal stays of mother and neonates. All parameters were analyzed with analysis of variance and the Student t
test as appropriate with the JMP 3.1 statistics program (Cary, NC).
RESULTS: Twenty-three sets of triplets were enrolled. Eight sets were delivered vaginally. Eight of 9 patients
(88.9%) who attempted trial of labor were delivered vaginally, 1 of which was a vaginal birth after cesarean
section. The remaining triplet gestation failed to progress at 4-cm dilation. Twelve sets of triplets had a non-
vertex-presenting triplet and were delivered by the cesarean route. The remaining 2 triplet gestations were
delivered by the cesarean route because of inadequate fetal monitoring. Neonatal survivals were 100% for
both groups. No significant differences in neonatal mortality, Apgar scores, intracranial hemorrhage, arterial
cord blood pH, hospital or neonatal intensive care unit stay of neonate, neonatal weight, and change in ma-
ternal or neonatal blood cell count were noted. There were no cases of grade III or IV intraventricular hemor-
rhage in either group. A significant reduction in postpartum hospital stay of mother was noted in the vaginal
delivery group (2.8 vs 4.5 days, P < .001). The mean gestational age at delivery was significantly lower for
the vaginal delivery group (31.3 vs 34.0 weeks, P < .02). The mean neonatal weight for the vaginal delivery
group was significantly lower (1758 473 vs 2022 407 g, P < .02). There were no significant differences in
outcome parameters for the first, second, and third triplets within each group when compared with each
other or with the other study group. One patient who underwent vaginal delivery had retained products of
conception and required curettage. A single fetal death occurred at 22 weeks gestation from twin-twin trans-
fusion, with the remaining triplets being delivered vaginally at 35 weeks gestation. Cesarean hysterectomy
was required in 1 case for uncontrollable bleeding at the time of cesarean delivery. Perinatal complications
occurred in a large number of patients, with the incidence of premature labor 47.8% (n = 11), that of preterm
premature rupture of membranes 26.1% (n = 6), and that of preeclampsia 34.8% (n = 8).
CONCLUSION: In selected cases vaginal delivery of triplet gestations can be accomplished without in-
creased maternal or neonatal morbidity and mortality and may significantly decrease maternal hospital stay
and postoperative morbidity. (Am J Obstet Gynecol 1998;179:1133-5.)
Key words: Triplet gestation, vaginal delivery
The numbers of viable triplet gestations at our institu-
tion and at other medical centers around the world have
markedly increased, mainly as a result of the expanding
use of assisted reproductive technology.
1
Patients carrying
twins and higher-order multiple gestations are at an in-
creased risk for such perinatal events as preterm labor
and delivery, premature rupture of membranes, pre-
eclampsia, and intrapartum complications. Postpartum
complications, such as hemorrhage, have also been de-
scribed. In an effort to reduce intrapartum fetal compli-
cations there has been a worldwide tendency to deliver
high-order multiple gestations by the cesarean route. The
rationale for this practice has been to reduce the birth in-
terval between neonates and thereby decrease the likeli-
hood of hypoxia and birth-related injury for the second
and third triplets.
2
It has also been suggested that ab-
dominal delivery protects the mother against injury from
blind intrauterine manipulation related to internal ver-
sion and extraction because fetal position assumes less
importance during cesarean delivery.
3
A review of the re-
cent English language literature on human beings gives
From Division of Maternal-Fetal Medicine, Department of Obstetrics,
Gynecology, and Reproductive Sciences, State University of New York at
Stony Brook.
Presented at the Eighteenth Annual Meeting of the Society of Perinatal
Obstetricians, Miami, Florida, February 2-7, 1998.
Reprint requests: Vito Alamia, Jr, MD, Division of Maternal-Fetal
Medicine, Department of Obstetrics and Gynecology, Room 030, Level 9
HSC, SUNY at Stony Brook, Stony Brook, NY 11794-8091.
Copyright 1998 by Mosby, Inc.
0002-9378/98 $5.00 + 0 6/6/92575
few objective data to support this position. Additionally,
there are data to suggest that vaginal delivery may be a
safe alternative to elective cesarean delivery for experi-
enced operators.
4
Our clinical impression suggested that there was no
difference in fetal outcome between vaginal and ce-
sarean delivery for our triplet population. Patients un-
dergoing vaginal delivery seemed to have a shortened
hospital stay, less postpartum discomfort, and less diffi-
culty in caring for their babies. Vaginal delivery has been
shown to carry fewer maternal complications for both
singleton and twin gestations. We therefore designed a
prospective planned triplet vaginal delivery protocol at
University Medical Center to evaluate the safety and per-
formance of both delivery methods.
Method
All patients cared for by University Medical Centers
maternal-fetal medicine faculty January 1, 1995-December
31, 1997, were offered enrollment in our vaginal delivery
protocol. During this period 23 sets of triplets were en-
rolled. Inclusion criteria included a vertex presentation
for triplet A, the ability to monitor all 3 fetuses appropri-
ately, and the absence of maternal obstetric contraindica-
tions for vaginal delivery. All deliveries were performed by
experienced perinatologists with ultrasonographic guid-
ance as required for direct fetal visualization. Appropriate
neonatology support teams attended each delivery. All pa-
tients received epidural anesthesia, with an obstetric anes-
thesiologist present at the time of delivery. All deliveries
were carried out in an operating room, with a dual setup
of instruments in the event an emergency cesarean deliv-
ery needed to be performed.
The objective outcome parameters measured included
estimated gestational age, neonatal mortality rate, Apgar
scores, neonatal intracranial hemorrhage (documented
by cerebral ultrasonography), arterial cord blood pH,
neonatal weight, length of postpartum hospital stay of
mother, hospital and neonatal intensive care unit stays of
the neonates, changes in maternal hematocrit, and intra-
partum complications. These data were entered into our
database and analyzed with 1-way analysis of variance and
the Student t test as appropriate with the JMP 3.1 statisti-
cal program (Cary, NC). P value, R
2
, mean SE, and de-
mographic data were obtained.
Results
Twenty-three sets of triplets were enrolled in our pro-
tocol, with 8 sets being delivered vaginally. All triplet ges-
tations were diagnosed by ultrasonography in the first
trimester and serially evaluated for growth and fetal bio-
physical status before delivery. Eight of 9 patients
(88.9%) who attempted trial of labor were delivered vagi-
nally; 1 was a vaginal birth after cesarean section. The re-
maining triplet gestation failed to progress at 4-cm dila-
tion. Twelve sets of triplets had a nonvertex-presenting
triplet A and were delivered by the cesarean route. The
remaining 2 triplet gestations were delivered by the ce-
sarean route because of inadequate monitoring. Vaginal
delivery was accomplished by breech extraction of the
second and third triplets in 6 cases. The remaining 2
1134 Alamia et al November 1998
Am J Obstet Gynecol
Table I. Significant differences between vaginal and cesarean delivery
Vaginal delivery Cesarean delivery
Statistical
Mean SE Range Mean SE Range significance
Estimated gestational age (wk) 31.3 0.8 27.6-35.0 34.0 0.6 30.0-37.0 P < .02
Mean neonatal weight (g) 1758 98.7 2022 60.7 P < .02
Maternal postpartum hospital stay (d) 2.8 0.36 4.5 0.27 P < .001
Table II. Apgar scores at 1 and 5 minutes and arterial
cord pH, vaginal and cesarean deliveries
Vaginal Cesarean Statistical
(n = 23) (n = 45) R
2
significance
Apgar score at 1 min
Triplet A 7.37 0.61 7.13 0.44 0.005 NS
Triplet B 6.38 0.45 6.87 0.44 0.036 NS
Triplet C 7.00 0.31 7.80 0.22 0.175 NS
Apgar score at 5 min
Triplet A 8.50 0.29 8.47 0.22 0.001 NS
Triplet B 8.13 0.33 8.27 0.24 0.006 NS
Triplet C 8.29 0.23 8.53 0.16 0.039 NS
Arterial cord pH
Triplet A 7.31 0.02 7.33 0.02 0.025 NS
Triplet B 7.31 0.02 7.32 0.02 0.003 NS
Triplet C 7.26 0.03 7.31 0.03 0.125 NS
Values are mean standard error. NS, Not significant.
Table III. Neonatal intensive care unit and hospital stays
for neonates, vaginal and cesarean deliveries
Vaginal Cesarean
(n = 23) (n = 45) R
2
NICU stay (d)
Triplet A 28.4 7.3 9.4 5.3 0.174
Triplet B 22.5 5.8 10.4 4.4 0.123
Triplet C 28.1 6.1 10.8 4.3 0.210
Hospital stay (d)
Triplet A 28.6 7.1 11.1 5.2 0.160
Triplet B 22.8 5.5 12.1 4.2 0.106
Triplet C 28.1 5.9 12.7 4.0 0.188
Values are mean SE. NICU, Neonatal intensive care unit.
cases had a vertex presentation for all 3 triplets, and
these infants were delivered without intrauterine fetal
manipulation.
The mean gestational age at delivery and mean
neonatal weight were significantly lower for the vaginal
delivery group (Table I). Neonatal survival rate
was 100% for both groups. There were no cases of grade
III or IV intraventricular hemorrhage in either group.
No significant differences in Apgar scores, arterial
cord blood pH, hospital or neonatal intensive care
unit stay of neonate, neonatal weight, or change in ma-
ternal or neonatal blood cell count were noted be-
tween the 2 groups (Table II). There was no significant
difference in any outcome parameter for the first, sec-
ond, and third triplets within each group when com-
pared with each other or with the other study group
(Table III).
A significant reduction in postpartum hospital stay of
mother was noted in the vaginal delivery group (2.8 vs
4.5 days, P < .001). One woman who underwent vaginal
delivery had retained products of conception and re-
quired curettage. A single fetal death occurred at 22
weeks gestation from twin-twin transfusion, with the re-
maining triplets being delivered vaginally at 35 weeks
gestation. Cesarean hysterectomy was required in 1 case
because of uterine atony and uncontrollable bleeding at
the time of cesarean delivery. Perinatal complications oc-
curred in a large number of patients, with the incidence
of premature labor 47.8% (n = 11), that of preterm pre-
mature rupture of membranes 26.1% (n = 6), and that of
preeclampsia 34.8% (n = 8).
Comment
Elective abdominal delivery of triplet gestations has
traditionally been advocated because of concerns regard-
ing the safety of the fetus and mother. Vaginal delivery is
generally considered safer than cesarean delivery for sin-
gleton and selected twin gestations.
5
A recent study has
demonstrated the safety of nonvertex-presenting multi-
ple gestations and vaginal birth.
6
Higher-order multiple
gestations have been associated with increased maternal
and fetal morbidity and mortality rates; however, recent
studies have advocated vaginal delivery for triplet gesta-
tions in appropriately selected cases.
7
Our study supports
this recommendation and demonstrates no significant
increase in the chosen morbidity parameters of neonates
delivered vaginally when compared with our cesarean de-
livery group. It has been assumed that the second and
third triplets will be relatively compromised when com-
pared with triplet A because of the route and time delays
between delivery of each fetus. When the Apgar score
and arterial cord pH value of each triplet C were com-
pared both within each triplet set and for all triplets de-
livered, however, no significant differences were seen.
The average gestational age at delivery for the vaginal
group was 2.6 weeks earlier than that of the cesarean de-
livery group, and the average weight difference between
the 2 groups was 264 g. Interestingly, there was no signif-
icant intracranial bleeding or short-term morbidity, and
neonatal survivals were 100% for both groups.
Although the number of cases of vaginal delivery in
our study is low, there was sufficient power to show the re-
duction in hospital stay, estimated gestational age, and
neonatal weight. The power was high enough to show
that there was no significant difference between the
triplets Apgar scores and arterial cord pH value. The
sample size is too small to make conclusions regarding
neonatal intensive care unit or hospital stay. It is uncer-
tain why there is a difference in the mean gestational
ages associated with vaginal and cesarean deliveries.
Because our selection criteria included vertex presenta-
tion, it is possible that a greater number of vertex-pre-
senting triplets are seen at earlier gestations, or this may
be a type I error. A much larger population of 108 sets of
triplets would be needed to address this issue. Vaginal de-
livery clearly benefited the mothers by reducing hospital
stay and postpartum discomfort, thereby facilitating the
neonatal care of their triplets.
All triplet vaginal deliveries were carried out by experi-
enced perinatologists comfortable with intrauterine ma-
nipulation, instrumental and breech delivery, and ultra-
sonographic evaluation of fetal status. Vaginal births may
be considered a safe option for interested patients in a
controlled environment with appropriate neonatal and
anesthesia support personnel. We suggest that triplet vagi-
nal birth should be offered only when these conditions
are met and after thorough counseling with the patient.
Cesarean delivery should be performed for patients with-
out motivation for vaginal delivery and in the absence of
ideal circumstances. Further investigation is warranted to
determine the long-term outcomes of neonates delivered
vaginally with respect to those delivered by the cesarean
route and to determine the role of vaginal birth in non-
vertex-presenting multiple gestations.
REFERENCES
1. Luke B, Keith LG. The contribution of singletons, twins and
triplets to low birth weight, infant mortality, and handicap in the
United States. J Reprod Med 1992;37:661-6.
2. Wildschut HI, van Roosmalen J, van Leeuwen E, Keirse MJ.
Planned abdominal compared with planned vaginal birth in
triplet pregnancies. Br J Obstet Gynaecol 1995;104:292-6.
3. Holcberg G, Biale Y, Lewenthal H, Insler V. Outcome of preg-
nancy in 31 triplet gestations. Obstet Gynecol 1981;59:472-6.
4. Clarke JP, Roman JD. A review of 19 sets of triplets: the positive
results of vaginal delivery. Aust N Z J Obstet Gynaecol 1994;34:
50-3.
5. Houlihan C, Knuppel RA. Intrapartum management of multi-
ple gestations. Clin Perinatol 1996;23:91-116.
6. Blickstein I, Weissman A, Ben-Hur H, Borenstein R, Insler V.
Vaginal delivery of breech-vertex twins. J Reprod Med
1993;38:879-82.
7. Dommergues M, Mahieu-Caputo D, Mandelbrot L, Huon C,
Moriette G, Dumez Y. Delivery of uncomplicated triplet preg-
nancies: is the vaginal route safer? A case-control study. Am J
Obstet Gynecol 1995;172:513-7.
Volume 179, Number 5 Alamia et al 1135
Am J Obstet Gynecol