- Email: [email protected]
Outcomes in twin pregnancies reduced to singleton pregnancies compared with ongoing twin pregnancies
SMFM Papers
ajog.org
Outcomes in twin pregnancies reduced to singleton pregnancies compared with ongoing twin pregnancies Simi Gupta, MD; Nathan S. Fox, MD; Jessica Feinberg, BS; Chad K. Klauser, MD; Andrei Rebarber, MD OBJECTIVE: Multifetal pregnancy reduction has been shown to
improve outcomes in triplet and higher-order multiple pregnancies. The data for fetal reduction of twin pregnancies are limited. The purpose of this study was to compare adverse pregnancy outcomes in ongoing twin pregnancies compared with twin pregnancies reduced to singletons. STUDY DESIGN: This was a retrospective cohort study comparing dichorionic diamniotic twin pregnancies with dichorionic diamniotic twin pregnancies reduced to singleton gestations between 11 and 24 weeks’ gestation in a single maternal-fetal medicine practice over a 9 year period. Adverse pregnancy outcomes after 24 weeks were compared, with a value of P < .05 used for significance. RESULTS: Five hundred one ongoing twin pregnancies and 63 twin
pregnancies reduced to singletons were included. Patients with
reductions to singletons had a significantly lower risk of preterm delivery before 37 weeks’ gestation (10% vs 43%; P < .001) but no difference in the risk of preterm delivery before 34 weeks’ or 28 weeks’ gestation. Patients with reductions to singletons also had a lower risk of infant birthweight less than the 10% (23% vs 49%; P < .001) but no difference in the risk of infant birthweight less than the 5%. There was no difference in the risk of fetal demise after 24 weeks’ gestation. CONCLUSION: Fetal reduction of twin pregnancies decreases the risk of late preterm birth and birthweight less than the 10% but not the risks of more severe complications such as early preterm birth or birthweight less than the 5%.
Key words: multifetal pregnancy reduction, selective reduction, twin pregnancy
Cite this article as: Gupta S, Fox NS, Feinberg J, et al. Outcomes in twin pregnancies reduced to singleton pregnancies compared with ongoing twin pregnancies. Am J Obstet Gynecol 2015;213:xx-xx.
M
ultifetal pregnancy reduction was developed to reduce higherorder multiple pregnancies to singleton or twin pregnancies in the attempt to improve several pregnancy outcomes but most importantly preterm birth.1-3 With improvements in infertility treatments, the number of triplet and higher-order pregnancies has dropped 9% from 2011
From Maternal-Fetal Medicine Associates, PLLC (all authors), and the Department of Obstetrics, Gynecology, and Reproductive Science, Icahn School of Medicine at Mount Sinai (Drs Gupta, Fox, Klauser, and Rebarber), New York, NY. Received March 3, 2015; revised May 6, 2015; accepted June 3, 2015. The authors report no conflict of interest. Presented at the 35th annual meeting of the Society for Maternal-Fetal Medicine, San Diego, CA, Feb. 2-7, 2015. Corresponding author: Simi Gupta, MD. [email protected] 0002-9378/$36.00 ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajog.2015.06.018
to 2012 and more than a third since its peak in 1998.4 Twin births, however, have remained stable over the last several years.4 The benefit of fetal reduction from twins to singletons remains controversial. Compared with singleton pregnancies, twin pregnancies are associated with an increase in fetal, infant, and maternal morbidity. Infant morbidity includes an increased risk of intrauterine growth restriction and preterm delivery with the associated complications of prematurity. Maternal morbidity includes an increased risk of gestational diabetes, hypertension, hemorrhage, and cesarean delivery.5 Multifetal reduction from twin pregnancies to singleton pregnancies may be performed for a number of reasons including an increased risk for infant or maternal complications based on maternal or obstetric history, to decrease the known complications of twin pregnancies, or for social reasons. Selective reduction may also be performed
because of a genetic or congenital anomaly in one of the fetuses. Fetal reduction to singletons in 52 twins by a single operator and 82 twins by multiple operators was shown to decrease the risk of preterm delivery and low birthweight when compared with national databases of ongoing twin pregnancies.6,7 However, in one small retrospective study comparing 35 ongoing twin pregnancies with 32 reduced pregnancies in a single unit, the authors found no difference in outcomes between the cohorts.8 However, this study was underpowered, given the small numbers in both the study and the control groups used. In addition, fetal reduction is not without risks. In a recent large study, the complete loss rate prior to 24 weeks in pregnancies reduced from 2 to 1 was 2.1%.9 The purpose of this study was to compare the risks of adverse pregnancy outcomes between ongoing twin pregnancies and twin pregnancies reduced to singleton pregnancies in a single practice.
MONTH 2015 American Journal of Obstetrics & Gynecology
1.e1
SMFM Papers M ATERIALS
AND
M ETHODS
This study is a retrospective cohort study of dichorionic diamniotic twin pregnancies between 2005 and 2013. Biomedical Research Alliance of New York Institutional Review Board approval was obtained. We compared outcomes between all ongoing dichorionic twin pregnancies and all dichorionic twin pregnancies that underwent fetal reduction to a singleton pregnancy in the first or second trimester. These reductions included both selective terminations (ie, because of a fetal abnormality) and multifetal pregnancy reductions (done solely to reduce the number of fetuses). For selective reductions, the anomalous fetus was reduced, and for multifetal pregnancy reductions, the fetus selected for reduction was chosen based on the technical ease of the procedure. Pregnancies were excluded if the data on birth outcome were not available. Baseline characteristics and pregnancy outcomes were obtained from the computerized medical record or from the referring physician. All of the ongoing twin pregnancies were managed by one single maternal-fetal medicine practice. The reduced pregnancies were managed either by the same practice or by the referring physician, with ongoing ultrasound and consultative services being provided by the maternal-fetal medicine practice. Fetal reductions were performed by maternal-fetal medicine attending physicians during the first and second trimesters. The reductions were performed according to standard technique by injecting 3-10 cc of 2 mEq/mL of potassium chloride into the fetal heart or thorax with a 20-gauge spinal needle until asystole was obtained. These were performed under real-time ultrasound guidance. Genetic screening with the first-trimester combined test of nuchal translucency and serum markers or diagnosis with chorionic villus sampling was performed prior to reduction. Patients with ongoing twin pregnancies and those with reduction to singleton pregnancies were managed similarly. Serial growth scans and ultrasounds for cervical length with fetal fibronectin testing were performed.
ajog.org Baseline characteristics evaluated were age, body mass index, conception with in vitro fertilization, ethnicity, obstetric history, and mullerian anomalies. Pregnancy outcomes evaluated were spontaneous preterm birth, low birthweight percentile,10 and intrauterine fetal demise after 24 weeks. We did not compare the rate of second trimester loss because outcome data for every ongoing twin pregnancy seen in the second trimester was not available. Therefore, we report loss rates only for the reduction cohort. Data were compared between ongoing twin pregnancies and twin pregnancies reduced to singletons with c2, MannWhitney U, or Student’s t test as appropriate. Logistic regression was performed to control for differences in baseline characteristics between the 2 groups. Incidences and odds ratios were reported with a value of P < .05 used for significance. The primary outcome was a composite outcome of spontaneous preterm birth less than 37 weeks’ gestation or a birthweight less than the 10%. Because we planned a priori on including all patients over the course of the study period, the power analysis was performed post hoc.
R ESULTS Between 2005 and 2013, the maternalfetal medicine practice managed 501 patients with ongoing dichorionic diamniotic twin pregnancies longer than 24 weeks’ gestation. All ongoing twin pregnancies had complete data for analysis. During this same time period, the practice performed 71 fetal reductions from dichorionic-diamniotic twin pregnancies to singleton pregnancies. Seven patients were excluded because outcome data were not available. One patient was not included in the data analysis because termination of the remaining twin was performed for severe intrauterine growth restriction prior to 24 weeks’ gestation. In this patient, fetal reduction was performed for a chromosomal anomaly in 1 twin and the severe intrauterine growth restriction
1.e2 American Journal of Obstetrics & Gynecology MONTH 2015
in the remaining twin was initially diagnosed before the procedure but continued to worsen after the procedure. This left 63 patients with twin pregnancies reduced to singleton pregnancies available for analysis. Twenty-two patients were from our own practice and 41 patients were referred from other providers. For the patients who underwent fetal reduction, 43 (68.3%) had their procedures performed in the first trimester and the remaining 20 (31.7%) in the second trimester. Approximately half of the procedures performed (n ¼ 32, 50.8%) were multifetal pregnancy reductions in which the terminated fetus was chosen randomly or based on the position in the uterus. Six of these 32 patients had a maternal or obstetrical history that contributed to their decision for a fetal reduction, specifically a mullerian anomaly or a prior preterm delivery. The remaining 31 of the reductions (49.2%) were selective terminations. Twenty-five of the 31 patients had a fetus diagnosed with a genetic or congenital anomaly. The other 6 of the 31 patients had a fetus diagnosed with oligohydramnios or intrauterine growth restriction. There were no losses of the remaining fetus after any of the reductions (0%; 95% confidence interval [CI], 0e5.8%). There were no differences in the adverse pregnancy outcomes evaluated between patients who underwent fetal reduction in the first trimester vs the second trimester (composite outcome of spontaneous preterm birth or low birthweight, 26% vs 35%; P ¼ .441) or for multifetal pregnancy reduction vs selective termination (composite outcome, 31% vs 26%; P ¼ .633). The baseline characteristics between the patients with ongoing twin pregnancies and those with reductions to singleton pregnancies are shown in Table 1. Patients with reductions to singleton pregnancies were older (37 years vs 34 years; P < .001) and more likely to be of a nonwhite ethnicity (22% vs 12%; P ¼ .042). Patients with reductions to singleton pregnancies were also more likely to have a history of
SMFM Papers
ajog.org a mullerian anomaly (8% vs 3%; P ¼ .042). There were no other significant differences noted. The risks of adverse pregnancy outcomes between the patients with ongoing twin pregnancies and those with reductions to singleton pregnancies are shown in Table 2. Patients with reductions to singletons had a significantly lower risk of the composite outcome of spontaneous preterm birth or low birthweight than patients with ongoing twin pregnancies (28.6% vs 69.6%; P < .001). Patients with reductions to singletons had a significantly lower risk of preterm delivery before 37 weeks’ gestation (10% vs 43%; P < .001) but no significant difference in the risk of preterm birth before 34 weeks’ gestation (5% vs 12%; P ¼ .088) or before 28 weeks’ gestation (3% vs 2%; P ¼ .637). Patients with reductions to singletons also had a lower risk of infant birthweight less than the 10% (23% vs 49%; P < .001) but no difference in the risk of infant birthweight less than the 5% (16% vs 28%; P ¼.079). There was no difference in the risk of intrauterine fetal demise between the 2 groups (odds ratio, 0.996; 95% CI, 0.990e1.002). Logistic regression analysis for the association between adverse pregnancy outcomes and multifetal pregnancy reduction is shown in Table 3. The adjusted odds ratio controls for the significant differences in age, ethnicity, and presence of mullerian anomalies between the 2 groups of patients. The differences in the composite outcome, spontaneous preterm birth less than 37 weeks, or birthweight less than 10% remained significant, even after controlling for differences in baseline characteristics. Similarly, there remained no significant differences in spontaneous preterm birth less than 34 weeks or less than 28 weeks or in birthweight less than 5%. Our post hoc power analysis showed that we had at least 80% power to show with a 5% alpha error rate a difference in the primary outcome of spontaneous preterm birth less than 37 weeks or birthweight less than 10% from 69.6% in patients with ongoing twin pregnancies to 28.6% in patients with a reduction from twins to singletons.
TABLE 1
Baseline characteristics compared between patients with ongoing twin pregnancies and those with reductions from twin to singleton pregnancies Demographic characteristics
Ongoing twins (n [ 496)
2 to 1 reduction (n [ 63)
P value
Age, y (mean SD)
34 6.63
37 4.65
.001
Body mass index, kg/m (mean SD)
23.71 4.65
22.60 3.00
.303
In vitro fertilization
71%
69%
.770
White ethnicity
88%
78%
.042
Prior term births
33%
45%
.096
Prior preterm births
7%
10%
.473
Mullerian anomaly
3%
8%
.042
2
Gupta. Fetal reduction in twin pregnancies. Am J Obstet Gynecol 2015.
C OMMENT In this study, we found that fetal reduction from dichorionic-diamniotic twin pregnancies to singleton pregnancies is associated with a decreased risk of spontaneous preterm delivery less than 37 weeks but not preterm delivery less than 34 weeks’ gestation or less than 28 weeks’ gestation. We also found that fetal reduction in twins is associated with a decreased risk of infant birthweight less than the 10% but not birthweight less than the 5%. These differences are clinically important because infant morbidity and mortality is higher in early preterm delivery vs late preterm delivery and in birthweight less than the 5% than birthweight less than the 10%. This information may be useful to patients who are considering fetal reduction to decrease the risk of significant morbidity or mortality in their infants. This study also showed no difference in the risk of intrauterine fetal demise after 24 weeks’ gestation between the 2 groups and no loss of the remaining fetus before 24 weeks’ gestation in patients who underwent fetal reduction. This information should be reassuring to patients who are choosing to undergo selective termination for a fetal anomaly. There have been only 3 previous studies that directly compared pregnancy outcomes in ongoing twin pregnancies vs twin pregnancies reduced to singletons.
The first compared the 2 groups within the same practice, as we did in our study, but found no difference in the risk of preterm delivery or intrauterine growth restriction between ongoing twins and twins reduced to singletons. This may be because of the small sample size of their study (35 ongoing twins and 32 reductions).8 Two other studies compared the risks of adverse pregnancy outcomes in twins reduced to singletons by a single and multiple operators with national databases of ongoing twins and showed similar results.6,7 There have been multiple studies looking at the loss rates of selective reduction and multifetal pregnancy reduction in twin pregnancies. Whereas older and smaller studies showed loss rates that ranged from 4.76% to 9.09%,8,11,12 the largest and most recent studies showed a loss rate between 2.1% and 2.5%.6,9 Our loss rate of 0% with a 95% CI of 0.0e0.575 is consistent with these recent data. The major strength of this study is that it is the largest study comparing pregnancy outcomes in ongoing dichorionic diamniotic twin pregnancies with twin pregnancies reduced to singletons with primary or consultative management by a single maternal-fetal medicine practice. This information may be useful for patients and clinicians considering the advantages and disadvantages of multifetal pregnancy reduction of twin
MONTH 2015 American Journal of Obstetrics & Gynecology
1.e3
SMFM Papers
ajog.org
TABLE 2
Pregnancy outcomes compared between patients with ongoing twin pregnancies and those with reductions from twin to singleton pregnancies Ongoing twins (n [ 496)
2 to 1 reduction (n [ 63)
P value
Composite outcome of spontaneous preterm delivery <37 wks’ gestation or birthweight <10%
69.6%
28.6%
< .001
Spontaneous preterm delivery <37 wks’ gestation
43%
10%
< .001
Spontaneous preterm delivery <34 wks’ gestation
12%
Spontaneous preterm delivery <28 wks’ gestation
2%
Outcome
Birthweight <10%
49%a
Birthweight <5%
28%
Intrauterine fetal demise >24 wks’ gestation a
5%
.088
3% a b
0.4%
.637
23%
< .001
16%
.079
0%
.613
Birthweight of at least 1 twin less than the 10% or 5% for the gestational age at the time of delivery; b Intrauterine fetal demise of at least 1 twin.
Gupta. Fetal reduction in twin pregnancies. Am J Obstet Gynecol 2015.
pregnancies vs continuing with an ongoing twin pregnancy. It also provides further evidence of the maternal safety of the procedure. One limitation of the study is that our sample size of reduced pregnancies may have prevented us from finding significant differences in some of the less common outcomes. Twin to singleton reductions are still rare procedures and therefore metaanalyses or collaborative data of both ongoing twin pregnancies and reduction data are necessary to further evaluate this issue for outcomes with low incidences to better quantify risk.
A second limitation is the use of infant birth data as a surrogate for neonatal outcome and the lack of long-term outcomes. This was necessary because the postnatal care of the infants took place at a number of different facilities. However, because there is a known association between gestational age at delivery and birthweight with neonatal outcomes, these data can still provide useful information. The proportion of patients undergoing consultation for twin to singleton reductions is steadily increasing. Some of this is due to improvements with in vitro fertilization with less triplet and
higher-order pregnancies and more twin pregnancies as well as increased pregnancy rates in women of advanced maternal age.13 Although some patients are considering this procedure because of a complicated medical or obstetrical history, many patients are just trying to optimize outcomes for their pregnancy. This study should aid in the counseling of these patients as to the likely benefits and risks of undergoing a reduction vs continuing with a twin pregnancy. Future studies should focus on collaborative multicenter prospective data of multifetal pregnancy reductions in twin pregnancies with an appropriate
TABLE 3
Logistic regression analysis for the association between adverse pregnancy outcomes and multifetal pregnancy reduction Unadjusted odds ratio (95% CI) for MFPR
Adjusted odds ratio (95% CI) for MFPRa
Composite outcome of spontaneous preterm delivery <37 wks’ gestation or birthweight <10%
0.175 (0.098e0.312)
0.193 (0.099e0.374)
Spontaneous preterm delivery <37 wks’ gestation
0.150 (0.063e0.357)
0.150 (0.056e0.401)
Spontaneous preterm delivery <34 wks’ gestation
0.368 (0.112e1.213)
0.253 (0.058e1.104)
Spontaneous preterm delivery <28 wks’ gestation
1.443 (0.312e6.662)
0.654 (0.076e5.652)
Birthweight <10%
0.310 (0.166e0.578)
0.377 (0.189e0.753)
Birthweight <5%
0.513 (0.253e1.040)
0.704 (0.327e1.517)
Intrauterine fetal demise >24 wks’ gestation
0.996 (0.990e1.002)
0.000
Outcome
CI, confidence interval; MFPR, multifetal pregnancy reduction. a
Adjusted for maternal age, ethnicity, and presence of mullerian anomalies.
Gupta. Fetal reduction in twin pregnancies. Am J Obstet Gynecol 2015.
1.e4 American Journal of Obstetrics & Gynecology MONTH 2015
SMFM Papers
ajog.org comparison group to more rigorously answer some of the concerns raised. In conclusion, reduction from a twin to a singleton pregnancy appears to be a safe procedure and reduces the risk of preterm birth less than 37 weeks and intrauterine growth restriction less than the 10th percentile. However, it does not appear to be associated with any improvement in more severe adverse outcomes. REFERENCES 1. Dodd JM, Crowther C. Multifetal pregnancy reduction of triplet and higher order multiple pregnancies to twins. Fertil Steril 2004;81:1420-2. 2. Dodd J, Crowther CA. Reduction of the number of fetuses for women with a multiple pregnancy. Cochrane Database Syst Rev 2012: CD003932.
3. Van de Mheen L, Everwijn SM, Knapen MF, et al. The effectiveness of multifetal pregnancy reduction in trichorionic triplet gestation. Am J Obstet Gynecol 2014;211:536.e1-6. 4. Martin J, Hamilton B, Osterman M, Curtin S, Mathews TJ. Births: final data for 2012. Natl Vital Stat Rep 2013;62:1-68. 5. American College of Obstetricians and Gynecologists. Multifetal gestations: twin, triplet, and higher-order multifetal pregnancies. ACOG Practice bulletin no. 144. Obstet Gynecol 2013;123:1118-32. 6. Evans MI, Kaufman MI, Urban AJ, Britt DW, Fletcher JC. Fetal reduction from twins to a singleton: a reasonable consideration? Obstet Gynecol 2004;104:102-9. 7. Yaron Y, Johnson KD, Bryant-Greenwood PK, Kramer RL, Johnson MP, Evans MI. Selective termination and elective reduction in twin pregnancies: 10 years experience at a single center. Hum Reprod 1998;13:2301-4. 8. Hasson J, Shapira A, Many A, Jaffa A, HarToov J. Reduction of twin pregnancy to
singleton: does it improve pregnancy outcome? J Matern Fetal Neonatal Med 2011;24:1362-6. 9. Stone J, Ferrara L, Kamrath K, et al. Contemporary outcomes with the latest 1000 cases of multifetal pregnancy reduction (MPR). Am J Obstet Gynecol 2008;199:406.e1-4. 10. Oken E, Kleinman KP, Rich-Edwards J, Gillman MW. A nearly continuous measure of birth weight for gestational age using a United States national reference. BMC Pediatr 2003;3:6. 11. Antsaklis A, Souka AP, Dashalakis G, et al. Pregnancy outcome after multifetal pregnancy reduction. J Matern Fetal Neonatal Med 2004;16:27-31. 12. Evans MI, Berkowitz RL, Wapner RJ, et al. Improvement in outcomes of multifetal pregnancy reduction with increased experience. Am J Obstet Gynecol 2001;184:97-103. 13. Evans MI, Andriole S, Britt DW. Fetal reduction: 25 years’ experience. Fetal Diagn Ther 2014;35:69-82.
MONTH 2015 American Journal of Obstetrics & Gynecology
1.e5
ajog.org
Outcomes in twin pregnancies reduced to singleton pregnancies compared with ongoing twin pregnancies Simi Gupta, MD; Nathan S. Fox, MD; Jessica Feinberg, BS; Chad K. Klauser, MD; Andrei Rebarber, MD OBJECTIVE: Multifetal pregnancy reduction has been shown to
improve outcomes in triplet and higher-order multiple pregnancies. The data for fetal reduction of twin pregnancies are limited. The purpose of this study was to compare adverse pregnancy outcomes in ongoing twin pregnancies compared with twin pregnancies reduced to singletons. STUDY DESIGN: This was a retrospective cohort study comparing dichorionic diamniotic twin pregnancies with dichorionic diamniotic twin pregnancies reduced to singleton gestations between 11 and 24 weeks’ gestation in a single maternal-fetal medicine practice over a 9 year period. Adverse pregnancy outcomes after 24 weeks were compared, with a value of P < .05 used for significance. RESULTS: Five hundred one ongoing twin pregnancies and 63 twin
pregnancies reduced to singletons were included. Patients with
reductions to singletons had a significantly lower risk of preterm delivery before 37 weeks’ gestation (10% vs 43%; P < .001) but no difference in the risk of preterm delivery before 34 weeks’ or 28 weeks’ gestation. Patients with reductions to singletons also had a lower risk of infant birthweight less than the 10% (23% vs 49%; P < .001) but no difference in the risk of infant birthweight less than the 5%. There was no difference in the risk of fetal demise after 24 weeks’ gestation. CONCLUSION: Fetal reduction of twin pregnancies decreases the risk of late preterm birth and birthweight less than the 10% but not the risks of more severe complications such as early preterm birth or birthweight less than the 5%.
Key words: multifetal pregnancy reduction, selective reduction, twin pregnancy
Cite this article as: Gupta S, Fox NS, Feinberg J, et al. Outcomes in twin pregnancies reduced to singleton pregnancies compared with ongoing twin pregnancies. Am J Obstet Gynecol 2015;213:xx-xx.
M
ultifetal pregnancy reduction was developed to reduce higherorder multiple pregnancies to singleton or twin pregnancies in the attempt to improve several pregnancy outcomes but most importantly preterm birth.1-3 With improvements in infertility treatments, the number of triplet and higher-order pregnancies has dropped 9% from 2011
From Maternal-Fetal Medicine Associates, PLLC (all authors), and the Department of Obstetrics, Gynecology, and Reproductive Science, Icahn School of Medicine at Mount Sinai (Drs Gupta, Fox, Klauser, and Rebarber), New York, NY. Received March 3, 2015; revised May 6, 2015; accepted June 3, 2015. The authors report no conflict of interest. Presented at the 35th annual meeting of the Society for Maternal-Fetal Medicine, San Diego, CA, Feb. 2-7, 2015. Corresponding author: Simi Gupta, MD. [email protected] 0002-9378/$36.00 ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajog.2015.06.018
to 2012 and more than a third since its peak in 1998.4 Twin births, however, have remained stable over the last several years.4 The benefit of fetal reduction from twins to singletons remains controversial. Compared with singleton pregnancies, twin pregnancies are associated with an increase in fetal, infant, and maternal morbidity. Infant morbidity includes an increased risk of intrauterine growth restriction and preterm delivery with the associated complications of prematurity. Maternal morbidity includes an increased risk of gestational diabetes, hypertension, hemorrhage, and cesarean delivery.5 Multifetal reduction from twin pregnancies to singleton pregnancies may be performed for a number of reasons including an increased risk for infant or maternal complications based on maternal or obstetric history, to decrease the known complications of twin pregnancies, or for social reasons. Selective reduction may also be performed
because of a genetic or congenital anomaly in one of the fetuses. Fetal reduction to singletons in 52 twins by a single operator and 82 twins by multiple operators was shown to decrease the risk of preterm delivery and low birthweight when compared with national databases of ongoing twin pregnancies.6,7 However, in one small retrospective study comparing 35 ongoing twin pregnancies with 32 reduced pregnancies in a single unit, the authors found no difference in outcomes between the cohorts.8 However, this study was underpowered, given the small numbers in both the study and the control groups used. In addition, fetal reduction is not without risks. In a recent large study, the complete loss rate prior to 24 weeks in pregnancies reduced from 2 to 1 was 2.1%.9 The purpose of this study was to compare the risks of adverse pregnancy outcomes between ongoing twin pregnancies and twin pregnancies reduced to singleton pregnancies in a single practice.
MONTH 2015 American Journal of Obstetrics & Gynecology
1.e1
SMFM Papers M ATERIALS
AND
M ETHODS
This study is a retrospective cohort study of dichorionic diamniotic twin pregnancies between 2005 and 2013. Biomedical Research Alliance of New York Institutional Review Board approval was obtained. We compared outcomes between all ongoing dichorionic twin pregnancies and all dichorionic twin pregnancies that underwent fetal reduction to a singleton pregnancy in the first or second trimester. These reductions included both selective terminations (ie, because of a fetal abnormality) and multifetal pregnancy reductions (done solely to reduce the number of fetuses). For selective reductions, the anomalous fetus was reduced, and for multifetal pregnancy reductions, the fetus selected for reduction was chosen based on the technical ease of the procedure. Pregnancies were excluded if the data on birth outcome were not available. Baseline characteristics and pregnancy outcomes were obtained from the computerized medical record or from the referring physician. All of the ongoing twin pregnancies were managed by one single maternal-fetal medicine practice. The reduced pregnancies were managed either by the same practice or by the referring physician, with ongoing ultrasound and consultative services being provided by the maternal-fetal medicine practice. Fetal reductions were performed by maternal-fetal medicine attending physicians during the first and second trimesters. The reductions were performed according to standard technique by injecting 3-10 cc of 2 mEq/mL of potassium chloride into the fetal heart or thorax with a 20-gauge spinal needle until asystole was obtained. These were performed under real-time ultrasound guidance. Genetic screening with the first-trimester combined test of nuchal translucency and serum markers or diagnosis with chorionic villus sampling was performed prior to reduction. Patients with ongoing twin pregnancies and those with reduction to singleton pregnancies were managed similarly. Serial growth scans and ultrasounds for cervical length with fetal fibronectin testing were performed.
ajog.org Baseline characteristics evaluated were age, body mass index, conception with in vitro fertilization, ethnicity, obstetric history, and mullerian anomalies. Pregnancy outcomes evaluated were spontaneous preterm birth, low birthweight percentile,10 and intrauterine fetal demise after 24 weeks. We did not compare the rate of second trimester loss because outcome data for every ongoing twin pregnancy seen in the second trimester was not available. Therefore, we report loss rates only for the reduction cohort. Data were compared between ongoing twin pregnancies and twin pregnancies reduced to singletons with c2, MannWhitney U, or Student’s t test as appropriate. Logistic regression was performed to control for differences in baseline characteristics between the 2 groups. Incidences and odds ratios were reported with a value of P < .05 used for significance. The primary outcome was a composite outcome of spontaneous preterm birth less than 37 weeks’ gestation or a birthweight less than the 10%. Because we planned a priori on including all patients over the course of the study period, the power analysis was performed post hoc.
R ESULTS Between 2005 and 2013, the maternalfetal medicine practice managed 501 patients with ongoing dichorionic diamniotic twin pregnancies longer than 24 weeks’ gestation. All ongoing twin pregnancies had complete data for analysis. During this same time period, the practice performed 71 fetal reductions from dichorionic-diamniotic twin pregnancies to singleton pregnancies. Seven patients were excluded because outcome data were not available. One patient was not included in the data analysis because termination of the remaining twin was performed for severe intrauterine growth restriction prior to 24 weeks’ gestation. In this patient, fetal reduction was performed for a chromosomal anomaly in 1 twin and the severe intrauterine growth restriction
1.e2 American Journal of Obstetrics & Gynecology MONTH 2015
in the remaining twin was initially diagnosed before the procedure but continued to worsen after the procedure. This left 63 patients with twin pregnancies reduced to singleton pregnancies available for analysis. Twenty-two patients were from our own practice and 41 patients were referred from other providers. For the patients who underwent fetal reduction, 43 (68.3%) had their procedures performed in the first trimester and the remaining 20 (31.7%) in the second trimester. Approximately half of the procedures performed (n ¼ 32, 50.8%) were multifetal pregnancy reductions in which the terminated fetus was chosen randomly or based on the position in the uterus. Six of these 32 patients had a maternal or obstetrical history that contributed to their decision for a fetal reduction, specifically a mullerian anomaly or a prior preterm delivery. The remaining 31 of the reductions (49.2%) were selective terminations. Twenty-five of the 31 patients had a fetus diagnosed with a genetic or congenital anomaly. The other 6 of the 31 patients had a fetus diagnosed with oligohydramnios or intrauterine growth restriction. There were no losses of the remaining fetus after any of the reductions (0%; 95% confidence interval [CI], 0e5.8%). There were no differences in the adverse pregnancy outcomes evaluated between patients who underwent fetal reduction in the first trimester vs the second trimester (composite outcome of spontaneous preterm birth or low birthweight, 26% vs 35%; P ¼ .441) or for multifetal pregnancy reduction vs selective termination (composite outcome, 31% vs 26%; P ¼ .633). The baseline characteristics between the patients with ongoing twin pregnancies and those with reductions to singleton pregnancies are shown in Table 1. Patients with reductions to singleton pregnancies were older (37 years vs 34 years; P < .001) and more likely to be of a nonwhite ethnicity (22% vs 12%; P ¼ .042). Patients with reductions to singleton pregnancies were also more likely to have a history of
SMFM Papers
ajog.org a mullerian anomaly (8% vs 3%; P ¼ .042). There were no other significant differences noted. The risks of adverse pregnancy outcomes between the patients with ongoing twin pregnancies and those with reductions to singleton pregnancies are shown in Table 2. Patients with reductions to singletons had a significantly lower risk of the composite outcome of spontaneous preterm birth or low birthweight than patients with ongoing twin pregnancies (28.6% vs 69.6%; P < .001). Patients with reductions to singletons had a significantly lower risk of preterm delivery before 37 weeks’ gestation (10% vs 43%; P < .001) but no significant difference in the risk of preterm birth before 34 weeks’ gestation (5% vs 12%; P ¼ .088) or before 28 weeks’ gestation (3% vs 2%; P ¼ .637). Patients with reductions to singletons also had a lower risk of infant birthweight less than the 10% (23% vs 49%; P < .001) but no difference in the risk of infant birthweight less than the 5% (16% vs 28%; P ¼.079). There was no difference in the risk of intrauterine fetal demise between the 2 groups (odds ratio, 0.996; 95% CI, 0.990e1.002). Logistic regression analysis for the association between adverse pregnancy outcomes and multifetal pregnancy reduction is shown in Table 3. The adjusted odds ratio controls for the significant differences in age, ethnicity, and presence of mullerian anomalies between the 2 groups of patients. The differences in the composite outcome, spontaneous preterm birth less than 37 weeks, or birthweight less than 10% remained significant, even after controlling for differences in baseline characteristics. Similarly, there remained no significant differences in spontaneous preterm birth less than 34 weeks or less than 28 weeks or in birthweight less than 5%. Our post hoc power analysis showed that we had at least 80% power to show with a 5% alpha error rate a difference in the primary outcome of spontaneous preterm birth less than 37 weeks or birthweight less than 10% from 69.6% in patients with ongoing twin pregnancies to 28.6% in patients with a reduction from twins to singletons.
TABLE 1
Baseline characteristics compared between patients with ongoing twin pregnancies and those with reductions from twin to singleton pregnancies Demographic characteristics
Ongoing twins (n [ 496)
2 to 1 reduction (n [ 63)
P value
Age, y (mean SD)
34 6.63
37 4.65
.001
Body mass index, kg/m (mean SD)
23.71 4.65
22.60 3.00
.303
In vitro fertilization
71%
69%
.770
White ethnicity
88%
78%
.042
Prior term births
33%
45%
.096
Prior preterm births
7%
10%
.473
Mullerian anomaly
3%
8%
.042
2
Gupta. Fetal reduction in twin pregnancies. Am J Obstet Gynecol 2015.
C OMMENT In this study, we found that fetal reduction from dichorionic-diamniotic twin pregnancies to singleton pregnancies is associated with a decreased risk of spontaneous preterm delivery less than 37 weeks but not preterm delivery less than 34 weeks’ gestation or less than 28 weeks’ gestation. We also found that fetal reduction in twins is associated with a decreased risk of infant birthweight less than the 10% but not birthweight less than the 5%. These differences are clinically important because infant morbidity and mortality is higher in early preterm delivery vs late preterm delivery and in birthweight less than the 5% than birthweight less than the 10%. This information may be useful to patients who are considering fetal reduction to decrease the risk of significant morbidity or mortality in their infants. This study also showed no difference in the risk of intrauterine fetal demise after 24 weeks’ gestation between the 2 groups and no loss of the remaining fetus before 24 weeks’ gestation in patients who underwent fetal reduction. This information should be reassuring to patients who are choosing to undergo selective termination for a fetal anomaly. There have been only 3 previous studies that directly compared pregnancy outcomes in ongoing twin pregnancies vs twin pregnancies reduced to singletons.
The first compared the 2 groups within the same practice, as we did in our study, but found no difference in the risk of preterm delivery or intrauterine growth restriction between ongoing twins and twins reduced to singletons. This may be because of the small sample size of their study (35 ongoing twins and 32 reductions).8 Two other studies compared the risks of adverse pregnancy outcomes in twins reduced to singletons by a single and multiple operators with national databases of ongoing twins and showed similar results.6,7 There have been multiple studies looking at the loss rates of selective reduction and multifetal pregnancy reduction in twin pregnancies. Whereas older and smaller studies showed loss rates that ranged from 4.76% to 9.09%,8,11,12 the largest and most recent studies showed a loss rate between 2.1% and 2.5%.6,9 Our loss rate of 0% with a 95% CI of 0.0e0.575 is consistent with these recent data. The major strength of this study is that it is the largest study comparing pregnancy outcomes in ongoing dichorionic diamniotic twin pregnancies with twin pregnancies reduced to singletons with primary or consultative management by a single maternal-fetal medicine practice. This information may be useful for patients and clinicians considering the advantages and disadvantages of multifetal pregnancy reduction of twin
MONTH 2015 American Journal of Obstetrics & Gynecology
1.e3
SMFM Papers
ajog.org
TABLE 2
Pregnancy outcomes compared between patients with ongoing twin pregnancies and those with reductions from twin to singleton pregnancies Ongoing twins (n [ 496)
2 to 1 reduction (n [ 63)
P value
Composite outcome of spontaneous preterm delivery <37 wks’ gestation or birthweight <10%
69.6%
28.6%
< .001
Spontaneous preterm delivery <37 wks’ gestation
43%
10%
< .001
Spontaneous preterm delivery <34 wks’ gestation
12%
Spontaneous preterm delivery <28 wks’ gestation
2%
Outcome
Birthweight <10%
49%a
Birthweight <5%
28%
Intrauterine fetal demise >24 wks’ gestation a
5%
.088
3% a b
0.4%
.637
23%
< .001
16%
.079
0%
.613
Birthweight of at least 1 twin less than the 10% or 5% for the gestational age at the time of delivery; b Intrauterine fetal demise of at least 1 twin.
Gupta. Fetal reduction in twin pregnancies. Am J Obstet Gynecol 2015.
pregnancies vs continuing with an ongoing twin pregnancy. It also provides further evidence of the maternal safety of the procedure. One limitation of the study is that our sample size of reduced pregnancies may have prevented us from finding significant differences in some of the less common outcomes. Twin to singleton reductions are still rare procedures and therefore metaanalyses or collaborative data of both ongoing twin pregnancies and reduction data are necessary to further evaluate this issue for outcomes with low incidences to better quantify risk.
A second limitation is the use of infant birth data as a surrogate for neonatal outcome and the lack of long-term outcomes. This was necessary because the postnatal care of the infants took place at a number of different facilities. However, because there is a known association between gestational age at delivery and birthweight with neonatal outcomes, these data can still provide useful information. The proportion of patients undergoing consultation for twin to singleton reductions is steadily increasing. Some of this is due to improvements with in vitro fertilization with less triplet and
higher-order pregnancies and more twin pregnancies as well as increased pregnancy rates in women of advanced maternal age.13 Although some patients are considering this procedure because of a complicated medical or obstetrical history, many patients are just trying to optimize outcomes for their pregnancy. This study should aid in the counseling of these patients as to the likely benefits and risks of undergoing a reduction vs continuing with a twin pregnancy. Future studies should focus on collaborative multicenter prospective data of multifetal pregnancy reductions in twin pregnancies with an appropriate
TABLE 3
Logistic regression analysis for the association between adverse pregnancy outcomes and multifetal pregnancy reduction Unadjusted odds ratio (95% CI) for MFPR
Adjusted odds ratio (95% CI) for MFPRa
Composite outcome of spontaneous preterm delivery <37 wks’ gestation or birthweight <10%
0.175 (0.098e0.312)
0.193 (0.099e0.374)
Spontaneous preterm delivery <37 wks’ gestation
0.150 (0.063e0.357)
0.150 (0.056e0.401)
Spontaneous preterm delivery <34 wks’ gestation
0.368 (0.112e1.213)
0.253 (0.058e1.104)
Spontaneous preterm delivery <28 wks’ gestation
1.443 (0.312e6.662)
0.654 (0.076e5.652)
Birthweight <10%
0.310 (0.166e0.578)
0.377 (0.189e0.753)
Birthweight <5%
0.513 (0.253e1.040)
0.704 (0.327e1.517)
Intrauterine fetal demise >24 wks’ gestation
0.996 (0.990e1.002)
0.000
Outcome
CI, confidence interval; MFPR, multifetal pregnancy reduction. a
Adjusted for maternal age, ethnicity, and presence of mullerian anomalies.
Gupta. Fetal reduction in twin pregnancies. Am J Obstet Gynecol 2015.
1.e4 American Journal of Obstetrics & Gynecology MONTH 2015
SMFM Papers
ajog.org comparison group to more rigorously answer some of the concerns raised. In conclusion, reduction from a twin to a singleton pregnancy appears to be a safe procedure and reduces the risk of preterm birth less than 37 weeks and intrauterine growth restriction less than the 10th percentile. However, it does not appear to be associated with any improvement in more severe adverse outcomes. REFERENCES 1. Dodd JM, Crowther C. Multifetal pregnancy reduction of triplet and higher order multiple pregnancies to twins. Fertil Steril 2004;81:1420-2. 2. Dodd J, Crowther CA. Reduction of the number of fetuses for women with a multiple pregnancy. Cochrane Database Syst Rev 2012: CD003932.
3. Van de Mheen L, Everwijn SM, Knapen MF, et al. The effectiveness of multifetal pregnancy reduction in trichorionic triplet gestation. Am J Obstet Gynecol 2014;211:536.e1-6. 4. Martin J, Hamilton B, Osterman M, Curtin S, Mathews TJ. Births: final data for 2012. Natl Vital Stat Rep 2013;62:1-68. 5. American College of Obstetricians and Gynecologists. Multifetal gestations: twin, triplet, and higher-order multifetal pregnancies. ACOG Practice bulletin no. 144. Obstet Gynecol 2013;123:1118-32. 6. Evans MI, Kaufman MI, Urban AJ, Britt DW, Fletcher JC. Fetal reduction from twins to a singleton: a reasonable consideration? Obstet Gynecol 2004;104:102-9. 7. Yaron Y, Johnson KD, Bryant-Greenwood PK, Kramer RL, Johnson MP, Evans MI. Selective termination and elective reduction in twin pregnancies: 10 years experience at a single center. Hum Reprod 1998;13:2301-4. 8. Hasson J, Shapira A, Many A, Jaffa A, HarToov J. Reduction of twin pregnancy to
singleton: does it improve pregnancy outcome? J Matern Fetal Neonatal Med 2011;24:1362-6. 9. Stone J, Ferrara L, Kamrath K, et al. Contemporary outcomes with the latest 1000 cases of multifetal pregnancy reduction (MPR). Am J Obstet Gynecol 2008;199:406.e1-4. 10. Oken E, Kleinman KP, Rich-Edwards J, Gillman MW. A nearly continuous measure of birth weight for gestational age using a United States national reference. BMC Pediatr 2003;3:6. 11. Antsaklis A, Souka AP, Dashalakis G, et al. Pregnancy outcome after multifetal pregnancy reduction. J Matern Fetal Neonatal Med 2004;16:27-31. 12. Evans MI, Berkowitz RL, Wapner RJ, et al. Improvement in outcomes of multifetal pregnancy reduction with increased experience. Am J Obstet Gynecol 2001;184:97-103. 13. Evans MI, Andriole S, Britt DW. Fetal reduction: 25 years’ experience. Fetal Diagn Ther 2014;35:69-82.
MONTH 2015 American Journal of Obstetrics & Gynecology
1.e5