Outcome of twin gestations complicated by a single anomalous fetus

OBSTETRICS& GYNECO

Volume

88

July 1996

Outcome of Twin Gestations Single Anomalous Fetus FERGAL D. MALONE, MB, SABRINA MARY E. D’ALTON, MD

1990-1994, and excluded twin gestations with anomalies in both fetuses. The control twin group was composed of all other normal twin pregnancies followed and delivered at our center in the preceding 2 years. Results: We reviewed 24 twin gestations with at least one anomalous fetus. Five cases were excluded because of anomalies in both fetuses, and a further five pregnancies had selective termination or termination of the entire pregnancy. There were 14 ongoing twin pregnancies with one anomalous fetus, and their median gestational age at diagnosis was 18 weeks (range 16-20). All twin anomalies were correctly diagnosed antenatally. Gestational age at delivery and birth weight were significantly lower for twins complicated by an anomaly compared with control twins (P = .008 and P = .OOl, respectively). The cesarean delivery and perinatal mortality rates of twin pregnancies with anomalies were significantly higher than those of normal twins (P = .Ol and P < .OOl, respectively). ConcZusion: The presence of a single anomalous fetus in a

88, NO.

I, JULY

1996

Complicated

D. CRAZGO, MD, DAVID

Objective: To compare the outcome of twin gestations complicated by a single anomalous fetus with twin gestations with no fetal anomalies. Methods: The study included all patients with twin gestations diagnosed with a fetal anomaly in one fetus during

VOL.

Number

CHELMOW,

twin gestation significantly delivery compared with

increases nonanomalous

1

by a MD, AND

the

risk twin

of preterm gestations.

(Obstet Gynecol 2996;88:1-5)

The incidence of multiple gestations in the United States is increasing, primarily because of assisted reproductive technologies.’ When complications arise in a multiple gestation, management decisions are often extremely difficult for parents, as the fates of sibling fetuses are usually linked. Preterm labor and the risk of a premature delivery increase with increasing numbers of fetuses.2,3 The perinatal mortality rate of twins and higher order multiple gestations is increased over the rate of singletons, largely related to the frequency of complications of prematurity.J The incidence of congenital anomalies is higher in multiple gestations than in singleton pregnancies.’ The incidence of preterm labor in singleton gestations is increased if a congenital abnormality exists.“-’ At present, there is little information available to determine if the incidence of preterm labor in multiple gestations is further increased if an anomaly exists in one of the fetuses. The question of a possible additional

0029-7844/96/$15.00 r11 S0029-7R44(96)00112-3

1

risk imposed on the pregnancy by an anomalous fetus is pertinent, given the availability of selective termination. Selective termination has been shown to be a potential alternative in multiple gestations complicated by one anomalous fetus.’ The risk of preterm delivery in such a pregnancy is unclear when the option of selective termination is either unacceptable to prospective parents or when it is contraindicated. Using a cohort design, we attempted to determine whether the presence of an anomaly in a single fetus of a twin gestation increases the risk of preterm delivery over the risk attributable to the twin gestation alone.

Table

Materials and Methods

use of tocolytics and antenatal steroids up to 34 weeks when indicated. Outcomes assessed included gestational age at delivery, birth weight, cesarean delivery rate, and perinatal mortality rate. All anomalies were confirmed on postnatal examination and all placentas were examined for chorionicity. Statistical analysis was performed using Student t test and the 2 test, with relative risk (RR) and 95% confidence intervals (CI) calculated using EPI INFO 5.0 (Centers for Disease Control and Prevention, Atlanta, GA). A statistically significant difference was defined as P < .05.

We included patients with twin gestations diagnosed with a major fetal anomaly in only one fetus at our center during the 5-year period 1990-1994. Records were obtained and reviewed from our antenatal ultrasound data base, as well as from our delivery records and neonatal data base. Twin gestations with anomalies in both fetuses were excluded because the objective of our study was to assess the impact of one anomalous fetus on the outcome for a normal co-twin. We did not include twin gestations with minor fetal abnormalities, such as isolated choroid plexus cyst or minimal hydronephrosis. The control twin cohort comprised all twin gestations followed throughout their pregnancy and delivered at this center during the 2-year period 1993-1994. This time period was selected as our delivery center changed location during the study period and complete records of normal deliveries were not readily available from the original center. The same obstetric staff and management practices were maintained throughout the 5-year period. We excluded from the control group patients transferred to this tertiary referral center with obstetric complications, because of the inherently high incidence of preterm delivery in this group. There were no maternal transfer patients in the study group. All patients with a single anomalous fetus were counseled by perinatologists at Tufts University. Selective termination has been available at this institution since 1992. Sixteen of the 17 anomalous pregnancies managed since 1992 were counseled regarding the options of selective termination of the anomalous fetus if not contraindicated, termination of the entire pregnancy, and continuing the entire pregnancy. The remaining patient was diagnosed at 18 weeks but did not present for counseling until 24 weeks. The obstetric management for all patients was similar, regardless of the presence of an anomalous fetus. This included close surveillance for preterm labor, with

2 Malone

et al

Twin

Gestatims

With

Amxrzalies

1. Demographic Characteristics Control Twin Pregnancies Anomalous

Number of gestations Maternal age Nulliparous Race White Hispanic Black Asian Prior preterm delivery * Comparison

of white

twins

of Anomalous Control

twins

14 30.9 9 (64.3%)

78 29.0 46 (59.0%)

12 (85.8%) 1 (7.1%) 1(7.1%) 0 0

61 (78.2%) 9 (11.5%) 6 (7.7%) 2 (2.6%) 2 (2.6%)

us other

and P

.80 .71 .52+

.54

races.

Results During 1990-1994, 24 patients with twin gestations complicated by significant congenital anomalies were diagnosed and managed at this center. In this group, five patients were excluded because of anomalies in both fetuses. Of the 19 anomalous twin pregnancies, four patients chose second-trimester selective termination of the anomalous fetus (one cystic hygroma, one complex cardiac anomaly, and two open neural tube defects). These four pregnancies resulted in vaginal deliveries at 35,32,34, and 36 weeks, respectively, with all infants surviving without morbidity. One patient had a second-trimester termination of the entire pregnancy because of a cystic hygroma in one fetus. These five patients were excluded from the analysis because the objective was to assess perinatal outcome when a patient continues the entire pregnancy. Table 1 presents the demographic characteristics of the anomalous and control groups. No significant differences were noted in maternal age, race, parity, or incidence of previous preterm delivery. The study population included 14 sets of anomalous twin gestations. Table 2 presents the anomalies and the pregnancy details in this group. Two of the twin pregnancies with anomalies had polyhydramnios (twins 5b and 12b,

Obstetrics 6 Gynecology

Table

2. Details

of Twin

GA at diagnosis (wk)

Patient la lb 2a 2b 3a 3b 4a 4b 5a 5b 6a 6b 7a 7b 8a 8b 9a 9b 10a lob lla Ilb 12a 12b 13a 13b 14a 14b

18

Complicated

b>T One

Anomalous

Anomaly

19 20 16 18 17 20 18 18 17 16 18 19

age; DCDA = dichorionic-diamnionic; = monochorionic-monoamnionic;

NTD

88, NO.

1, JULY

GA at delivery (wk)

Birth weight (g)

Outcome

Placenta

36

2525 1950 1660 1350 1700 911 2895 1565 2100 1824 2640 2250 1375 1325 1840 1914 2241 3402 1660 1665 2640 2739 906 1012 820 568 2463 2247

Alive IUFD NND Alive Alive IUFD Alive IUFD Alive IUFD Alive Alive NND Alive AliW Alive NND Alive Alive Alive Alive Alive NND Alive Alive NND Alive Alive

DCDA

1996

29 32 37 33 36 29 36 38 34 37 28 29 36

1UFD = intrauterine fetal demise; NND = neural tube defect.

listed in Table 2). Three of the twin pregnancies with anomalies were complicated by intrauterine growth restriction (twin pregnancies 8, 9, and 13). Three of the twin pregnancies were monochorionic, one of which was monoamnionic. All anomalies were correctly diagnosed antenatally, with a median gestational age at diagnosis of 18 weeks (range 16-20). Thirteen of the 14 anomalous twins were diagnosed at our antenatal ultrasound unit, and one was diagnosed at an affiliated unit. No new anomalies were diagnosed in either twin group at postnatal examination. Table 3 describes the outcome for anomalous and control twin pregnancies. Gestational age at delivery and birth weight were significantly lower for anomalous twins. The risk of delivering at 36 weeks or less was 78.6% for anomalous twins versus 59.0% for control twins, resulting in a 19.6% attributable risk of preterm delivery based on the presence of an anomalous fetus. The cesareandelivery rate and the perinatal mortality rate were significantly higher for anomalous twin pregnancies. Of the nine cesareansin the anomalous twin group, four were for malpresentation in labor, three for

VOL.

Fetus

Normal Body stalk Cloaca1 extrophy Normal Normal Pentalogy of Cantrell Normal Edwards syndrome Normal Osteogenesis imperfecta Polycystic kidneys Normal Potter syndrome Normal I’olycystic kidneys Normal Anencephaly Normal Transposition great vessels Normal Down syndrome Normal Normal Down syndrome Normal Snenomelia Open neural tube defect Normal

16

GA = gestational diamnionic; MCMA

Gestations

= neonatal

demise;

DCDA MCDA DCDA DCDA DCDA DCDA MCDA DCDA MCMA DCDA DCDA DCDA DCDA

MCDA

= monochorionic-

nonreassuring fetal testing, one for malpresentation at term, and one for placenta previa. Of the 21 cesareansin the control twin group, seven were for malpresentation Table

3. Outcome

of Anomalous

and

Control

Twin

Pregnancies Anomalous twins 11 28

No. of gestations No. of fetuses Delivery at (wk) 24-28 29-32 33-36 ~36 Mean GA at delivery (wk) Cesarean section (wk) Mean birth weight (g) Intrauterine fetal demise Neonatal demise Perinatal mortality rate Perinatal death normal twins

et al

0 12 (15.4%) 34 (43.6%) 32 (41.0%) 35.6 ‘- 2.51 21 (26.9%) 2318 2 582

6% 333

mean

Tzrjin

P

78 156

1 (7.1?b) 3 (28.6%) 6 (42.9%) 3 (21.4?/,) 33.6 t 3.42 9 (64.3%) 1864 t- 672

GA = gestational age. Data are presented as PI, ?I (“;), * Includes one normal co-twin.

Malone

Control twins

6 1

2 standard

Gestntions

.OOS .Ol ,001

c.001 .28

deviation.

Wit)? Anormlies

3

in labor, ten for nonreassuring testing, three for malpresentation at term, and one was an elective repeat. There were nine deaths in the anomalous twin group, eight occurring in fetuses with anomalies. Mortality in twins with anomalies consisted of three intrauterine and five neonatal demises. One normal infant died from fulminant necrotizing enterocolitis at 28 weeks, but the co-twin with Down syndrome survived. Comparison of perinatal mortality between normal co-twins from the anomalous group and control twins revealed an RR of 5.57 (95% CI 0.37-83.9). Power analysis indicated that 875 controls and 157 cases would be necessary to detect an increase in perinatal mortality from 1 to 5% with 80% power.

Discussion Twin gestations complicated by a single anomalous fetus present difficulties in obstetric management. Options available to patients include continuing the entire pregnancy, selective termination of the anomalous fetus, or termination of the entire pregnancy. The feasibility of selective termination of an anomalous twin has been described, but it may be unsafe in monochorionic gestations.’ Little information is available regarding the impact of an anomalous fetus on the normal co-twin when a patient decides to continue the entire pregnancy. The only previous report addressing this question was limited by the gestational age at prenatal diagnosis (75% diagnosed after 24 weeks’ gestation), at which time the patient’s options for management were limited.” All anomalies in twins in our study population were correctly diagnosed antenatally, before 20 weeks’ gestation. This may reflect our policy of recommending tertiary level prenatal ultrasonography to all patients with multiple gestations. No additional anomalies were found in either twin group on postnatal examination. The lower birth weight in the anomalous twin group is due to the high incidence of prematurity in this group. Significant growth restriction was present in only three of the anomalous pregnancies and so is not responsible for the lower mean birth weight in the anomalous twin group. These data differ from those of Heydanus et al,” who suggested that the presence of an anomalous fetus did not significantly alter the prematurity rate when compared with normal twin pregnancies. However, their study matched 23 anomalous pregnancies to only 23 normal twins on the basis of maternal age, parity, and year of delivery. Our control group consisted of all twins managed throughout their pregnancy at this center, and is not as open to selection bias. We have

4 Malone

et al

Twin Gestations With Anomalies

demonstrated a significantly greater risk of preterm delivery for anomalous twin gestations in addition to that seen with normal twin gestations. The higher perinatal mortality rate for anomalous twin gestations reflects both the lethality of the anomalies themselves and the higher incidence of complications of prematurity. Only one of 14 normal co-twins died. Therefore, the lethality of the anomaly is the main contributor to the worse perinatal mortality for such pregnancies. One potential criticism of this study is that the obstetric staff knew that one of the fetuses had an anomaly and this may have affected obstetric management or even accounted for the higher cesarean delivery rate in the anomalous twin group. We believe this is unlikely as all twin pregnancies were subject to the same tocolysis eligibility criteria and the same vaginal twin delivery criteria regardless of the presence of an anomalous fetus. Decision for cesarean delivery was based only on accepted obstetric indications. The fact that obstetric staff knew about the presence of a lethal anomaly might even be expected to lower the rate of cesarean delivery as an ominous heart tracing might be disregarded in such a fetus. The higher cesarean delivery rate in the anomalous twin group reflects the higher incidence of malpresentations in such pregnancies, which is a well documented occurrence.” One of the limitations of this study is the wide range of congenital anomalies seen with twin gestations. It is unclear whether the additional risk of preterm delivery is applicable to only certain anomalies or to all twin gestations complicated by an anomalous fetus. Such information could only be derived from cohorts of twin pregnancies each complicated by specific anomalies. In our series of 19 twin gestations complicated by a single anomalous fetus, four patients opted for selective termination of the anomalous fetus, and all four had a successful outcome. However, we do not have sufficient numbers of selectively terminated anomalous pregnancies to make recommendations regarding the efficacy of this intervention. If a dichorionic pregnancy has been confirmed, we recommend that selective termination of the anomalous fetus should be included in the management options for patients faced with this problem. Our results indicate that the presence of a single anomalous fetus in a twin gestation significantly increases the risk of preterm delivery compared with normal twin gestations. This information may prove useful when counseling patients faced with this difficult dilemma, especially when the option of selective termination is not acceptable to prospective parents, or when there is a contraindication to selective termination.

Obstetrics & Gynecology

References 1. Jewel1 SE, Yip R. Increasing trends in plural births in the United States. Obstet Gynecol 1995;85:229-32. 2. Caspi E, Ronen J, Schreyer I’, Goldberg MD. The outcome of pregnancy after gonadotrophin therapy. Br J Obstet Gynaecol 1976;83:967-73. 3. Gardner MO, Goldenberg RL, Cliver SP, Tucker JM, Nelson KG, Copper RL. The origin and outcome of preterm twin pregnancies. Obstet Gynecol 1995;85:553-7. 4. Fowler MG, Kleinman JC, Kiely JL, Kessel SS. Double jeopardy: Twin infant mortality in the United States, 1983 and 1984. Am J Obstet Gynecol 1991;165:1!%22. 5. Kohl SG, Casey G. Twin gestation. Mt Sinai J Med 1975;42:523-39. 6. Honnebier WJ, Swaab DF. The influence of anencephaly upon intrauterine growth of fetus and placenta and upon gestation length. J Obstet Gynaecol Br Commonw 1973;80:577-88. 7. Ratten GJ, Beischer NA, Fortune DW. Obstetric complications when the fetus has Potters syndrome. Am J Obstet Gynecol 1973;115:890-6. 8. Greasy RK. Preterm labor and delivery. In: Creasy RK, Resnik R, eds. Maternal-fetal medicine: Principles and practice. 3rd ed. Philadelphia: Saunders, 1994:494-520. 9. Evans MI, Goldberg JD, Dommergues M, Wapner RJ, Lynch L, Dock OS, et al. Efficacy of second-trimester selective termination for fetal abnormalities: International collaborative experience among the world’s largest centers. Am J Obstet Gynecol 1994;171: 90-4.

VOL.

88, NO.

1, JULY

1996

10. Heydanus R, Santema JG, Stewart JW. Preterm delivery rate and fetal twin pregnancies: A retrospective Diagn 1993;13:155-62. 11. Brenner WE, Bruce RD, Hendricks perils of breech presentation. Am

Address

reprint

requests

PA, Mulder PGH, Wladimiroff outcome in structurally affected matched control stidy. Prenat CH. The characteristics and J Obstet Gynecol 1974;118:700-

to:

Fergal D. Malone, MB Department of Obstetrics and Gynecology New England Medical Center, Box #022 750 Washington Street Boston, MA 02111

Received October 10, 1995. Received in revised form january Accepted February 13, 1996.

17, 1996.

Copyright 0 1996 by The American College of Obstetricians Gynecologists. Published by Elsevier Science Inc.

Malone

et al

Twin Gestations With Anomalies

and

5