Documente Academic
Documente Profesional
Documente Cultură
OBJECTIVE: To compare perinatal and peripartum out- RESULTS: There were 1,189 pregnancies that met inclu-
comes of vanishing twin gestations with singleton and sion criteria (798 singleton, 291 twin, and 100 vanishing
dichorionic twin gestations in pregnancies conceived by twin). The mean gestational age at birth and birth
in vitro fertilization. weights were 38.662.3 weeks of gestation and
METHODS: We conducted a retrospective cohort study 3,2076644 g in singleton pregnancies, 35.562.7 weeks
of vanishing twin pregnancies after fresh and cryopre- of gestation and 2,5396610 g in twin pregnancies, and
served autologous in vitro fertilization cycles performed 38.561.8 weeks of gestation and 3,1756599 g in vanish-
at our institution from 2007 to 2015. Singleton, dichor- ing twin pregnancies. When compared with twins, those
ionic twin, and dichorionic twin pregnancies with spon- with a vanishing twin had lower odds of preterm delivery
taneous reduction to one by 14 weeks of gestation (OR 0.13, 95% CI 0.07–0.23; adjusted OR 0.12, 95% CI
(vanishing twins) were included. Analysis was restricted 0.07–0.22) and small-for-gestational-age birth weight
to patients with a live birth delivery at our institution at (OR 0.24, 95% CI 0.13–0.45; adjusted OR 0.14, 95% CI
or beyond 24 weeks of gestation. The primary outcomes 0.07–0.28).
were gestational age and birth weight at delivery; CONCLUSION: In pregnancies conceived by in vitro
secondary outcomes included peripartum morbidities. fertilization that progress to at least 24 weeks of
A subanalysis further differentiated the vanishing twin gestation, vanishing twin and singleton pregnancies had
pregnancies between those in which demise of the twin similar perinatal and peripartum outcomes. Both were
occurred before compared with after identification of significantly better than twin pregnancies.
fetal cardiac activity. Logistic regression models were (Obstet Gynecol 2018;131:1011–20)
used to estimate the adjusted odds ratio (OR) with a 95% DOI: 10.1097/AOG.0000000000002595
CI of outcomes.
1012 Romanski et al Pregnancy Outcomes After Vanishing Twin OBSTETRICS & GYNECOLOGY
VOL. 131, NO. 6, JUNE 2018 Romanski et al Pregnancy Outcomes After Vanishing Twin 1013
Among all vanishing twin pregnancies, the gesta- of patients by the furthest developmental stage docu-
tional age by IVF dating at demise ranged from 41 to 85 mented before identification of a vanishing twin were:
days with a mean of 59.2610.9 days. The distributions 15 patients (15%) had only a gestational sac; 21 (21%) had
1014 Romanski et al Pregnancy Outcomes After Vanishing Twin OBSTETRICS & GYNECOLOGY
a gestational sac with a yolk sac; nine (9%) had a fetal Perinatal outcomes in the three exposure groups
pole with no documented cardiac activity; and 55 (55%) of patients are shown in Table 2. The mean gesta-
had documented fetal cardiac activity. tional age at birth and birth weights were 38.662.3
VOL. 131, NO. 6, JUNE 2018 Romanski et al Pregnancy Outcomes After Vanishing Twin 1015
1016 Romanski et al Pregnancy Outcomes After Vanishing Twin OBSTETRICS & GYNECOLOGY
pregnancies in which the vanishing twin was identi- twins is likely explained by the age-related risk of mis-
fied after the presence of fetal cardiac activity (38% vs carriage, largely attributable to aneuploidy.22
18%; OR 2.86, 95% CI 1.12–7.30; adjusted OR 3.10, The increased perinatal and peripartum morbid-
95% CI 1.19–8.09). ity associated with twin pregnancies as compared with
vanishing twin pregnancies is consistent with known
risks of multifetal gestations as compared with single-
DISCUSSION ton births.23 Of note, our perinatal and peripartum
The most notable findings of this study are 1) findings indicate that demise of one twin in the first
vanishing twin pregnancies had comparable perinatal trimester returns the patient and the surviving twin to
and peripartum outcomes with singleton pregnancies the risk profile of a singleton pregnancy.
and improved outcomes compared with twin preg- Despite the theoretically increased vascularity
nancies and 2) occurrence of twin demise after and tissue mass of a twin demised at a more advanced
documentation of fetal cardiac activity was not developmental stage, timing of demise was not
associated with increased risk of adverse outcomes associated with increased risk for any adverse perina-
other than primary cesarean delivery rate. tal or peripartum outcome, apart from primary
Our rate of 25.6% vanishing of all twin pregnan- cesarean delivery rate. The etiology underlying the
cies is consistent with previous studies.2–4 Our demo- association between increased use of cesarean deliv-
graphic predictors of a vanishing twin pregnancy are ery when vanishing occurred after the presence of
in agreement with risk factors for multiple pregnancy fetal cardiac activity is unclear and warrants a more
and miscarriage. Additionally, that our vanishing twin detailed obstetric investigation. In the absence of any
patient population was older than either singletons or biological plausibility to explain this association, we
VOL. 131, NO. 6, JUNE 2018 Romanski et al Pregnancy Outcomes After Vanishing Twin 1017
Fetal outcomes
Female sex of the 27 (60.0) 28 (50.9) 0.70 (0.31–1.53) 0.68 (0.31–1.52)
surviving twin
Gestational age at 38.561.9 38.561.7 — —
delivery (wk)
Gestational age less 9 (20.0) 8 (14.6) 0.68 (0.24–1.94) 0.71 (0.25–2.03)
than 37 wk
Birth weight (g)† 3,1686686 3,1816522 — —
Less than 2,500 8 (17.8) 5 (9.1) 0.45 (0.14–1.50) 0.24 (0.05–1.26)
2,500–4,000 33 (73.3) 46 (83.6) 1.00 (referent) 1.00 (referent)
Greater than 4,000 4 (8.9) 4 (7.3) 0.72 (0.17–3.08) 0.92 (0.20–4.34)
Z-score‡
SGA 8 (17.8) 6 (10.9) 0.55 (0.17–1.74) 0.55 (0.17–1.77)
AGA 33 (73.3) 45 (81.8) 1.00 (referent) 1.00 (referent)
LGA 4 (8.9) 4 (7.3) 0.73 (0.17–3.15) 0.91 (0.19–4.27)
Peripartum outcomes
Gestational hypertensive 5 (11.1) 4 (7.3) 0.63 (0.16–2.49) 0.71 (0.17–2.89)
diseases
GDM 2 (4.4) 1 (1.8) 0.40 (0.04–4.54) 0.38 (0.03–4.39)
Abnormal placentation§ 2 (4.4) 3 (5.5) 1.24 (0.20–7.77) 1.35 (0.21–8.66)
Estimated blood loss 560.26352.7 550.06230.5 — —
Postpartum hemorrhagek 1 (2.2) 0 (0) — —
Abruption 2 (4.4) 2 (3.6) 0.81 (0.11–6.0) 0.78 (0.11–5.81)
Primary cesarean delivery 8 (17.8) 21 (38.2) 2.86 (1.12–7.30) 3.10 (1.19–8.09)
OR, odds ratio; SGA, small for gestational age; AGA, appropriate for gestational age; LGA, large for gestational age; GDM, gestational
diabetes mellitus.
Data are n (%) or mean6SD unless otherwise specified.
* Adjusted a priori for patient age at retrieval.
†
Birth weight measurements additionally adjusted for gestational age at delivery (weeks) and neonatal sex.
‡
Refer to the text for explanation of Z-score.
§
Placenta previa, with the placental edge coming within 2.0 cm of the internal os on final ultrasonogram; or placenta accreta, increta, or
percreta diagnosed clinically at the time of delivery or on pathology review.
k
Estimated blood loss greater than 1,500 mL or if the patient received a red blood cell transfusion.
conclude that perinatal and peripartum outcomes do definition of a vanishing twin may help increase study
not appear adversely affected by timing of vanishing power, it can have unintended effects on reported out-
during the first trimester. comes as a result of the occurrence of varying pathol-
Prior publications have reported inconsistent ogies at different points in a gestation. We therefore
associations between vanishing twin and singleton chose to limit the definition of a vanishing twin to
pregnancies for perinatal outcomes ranging from no demise of one twin occurring within the first trimester.
adverse outcome in the surviving twin to significantly Furthermore, many prior studies do not define a lower
reduced gestational age and birth weight.3,6,7,9 One limit of gestational age at which the surviving twin is
explanation for these varied observations is lack of delivered. This, in turn, may lead to bias in perinatal
a consistent definition of vanishing twin, which could outcome data, because previable deliveries are often
lead to misclassification of the exposure. The majority related to maternal anatomy or infections and cannot
of prior research has included demises occurring in be attributed solely to the vanishing twin. By includ-
the second and third trimesters.6,7,24 It is well estab- ing these patients, previously reported perinatal and
lished that first-trimester losses are most commonly maternal outcomes likely were influenced by multiple
associated with cytogenetic abnormalities; for losses pathologic processes unrelated to the vanished twin
beyond the first trimester, placental dysfunction, per se, thereby introducing substantial bias.
maternal disease, and infection become more com- We acknowledge several limitations of our study:
mon causes of demise.25,26 Although a more liberal 1) the medically complex patients referred to our
1018 Romanski et al Pregnancy Outcomes After Vanishing Twin OBSTETRICS & GYNECOLOGY
VOL. 131, NO. 6, JUNE 2018 Romanski et al Pregnancy Outcomes After Vanishing Twin 1019
1020 Romanski et al Pregnancy Outcomes After Vanishing Twin OBSTETRICS & GYNECOLOGY