Objective
This systematic review and meta-analysis aimed to compare the fetal survival rate and perinatal outcomes of triplet pregnancies after selective reduction to twin pregnancies vs singleton pregnancies.
Data Sources
PubMed, Web of Science, Scopus, and Embase were systematically searched from the inception of the databases to January 16, 2022.
Study Eligibility Criteria
Studies comparing the survival and perinatal outcomes between reduction to twin pregnancies and reduction to singleton pregnancies were included. The primary outcomes were fetal survival, defined as a live birth at >24 weeks of gestation. The secondary outcomes were gestational age at birth, preterm birth at <32 and <34 weeks of gestation, early pregnancy loss (<24 weeks of gestation), low birthweight, and rate of neonatal demise (up to 28 days after birth).
Methods
The random-effect model was used to pool the mean differences or odds ratios and the corresponding 95% confidence intervals. To provide a range of expected effects if a new study was conducted, 95% prediction intervals were calculated for outcomes presented in >3 studies.
Results
Of note, 10 studies with 2543 triplet pregnancies undergoing fetal reduction, of which 2035 reduced to twin pregnancies and 508 reduced to singleton pregnancies, met the inclusion criteria. Reduction to twin pregnancies had a lower rate of fetal survival (odds ratio, 0.61; 95% confidence interval, 0.40–0.92; P =.02; 95% prediction interval, 0.36–1.03) and comparable rates of early pregnancy loss (odds ratio, 0.89; 95% confidence interval, 0.58–1.38; P =.61; 95% prediction interval, 0.54–1.48) and neonatal demise (odds ratio, 0.57; 95% confidence interval, 0.09–3.50; P =.55) than reduction to singleton pregnancies. Reduction to twin pregnancies had a significantly lower gestation age at birth (weeks) (mean difference, −2.20; 95% confidence interval, −2.80 to −1.61; P <.001; 95% prediction interval, −4.27 to −0.14) than reduction to singleton pregnancies. Furthermore, reduction to twin pregnancies was associated with lower birthweight and greater risk of preterm birth at <32 and <34 weeks of gestation.
Conclusion
Triplet pregnancies reduced to twin pregnancies had a lower fetal survival rate of all remaining fetuses, lower gestational age at birth, higher risk of preterm birth, and lower birthweight than triplet pregnancies reduced to singleton pregnancies; reduction to twin pregnancies vs reduction to singleton pregnancies showed no substantial difference for the rates of early pregnancy loss and neonatal death.
Why was this study conducted?
This study aimed to systematically compare the perinatal outcomes of triplet pregnancies after selective reduction to twin (RTT) pregnancies vs reduction to singleton (RTS) pregnancies.
Key findings
Triplet pregnancies reduced to twin pregnancies had a lower rate of survival of the remaining fetuses after reduction, but the risks of early pregnancy loss and neonatal demise were comparable with triplet pregnancies reduced to singleton pregnancies. Triplet pregnancies reduced to twin pregnancies had a noticeable lower gestational age at birth, greater risk of preterm birth (PTB) at <32 and <34 weeks of gestation, and lower birthweight than triplet pregnancies reduced to singleton pregnancies.
What does this add to what is known?
The decision to undergo fetal reduction to twins or singletons can be challenging for parents with triplet pregnancies. Moreover, determining the perinatal outcome after the reduction is important. This systematic review and meta-analysis demonstrated a decreased fetal survival and an increased risk of PTB and low birthweight for RTT pregnancies compared with RTS pregnancies.
Introduction
Fetal reduction (FR) was developed in the 1980s because of the high incidence of pregnancy loss and complications from higher-order multiples whose incidence was rising substantially from infertility therapies. , As with many leading-edge technological advances, FR started at a limited number of centers that developed the technique, that published the technique, and that formed centers of excellence. In addition, there was a phase of diffusion when the technology expanded across the medical community, allowing increasing numbers of patients to have the service, but with wider variability of offerings and quality.
The need for such services expanded dramatically during the last 30 years. By 2015, triplet and higher-order pregnancies accounted for 103.6 per 100,000 births in the United States. Even with improvements in neonatal care, triplet pregnancies still had increased perinatal mortality and morbidity compared with twin or singleton pregnancies. Lower gestational age (GA) at birth and lower birthweight were the main factors of adverse outcomes. Triplet prematurity and low birthweight are known to contribute to 11-fold and 2.2-fold increases, respectively, in infant mortality rate in triplet pregnancies compared with singleton and twin prematurity pregnancies. Accordingly, all-cause healthcare expenses related to singleton pregnancies have been calculated to be an average of $21,500 (US dollars) compared with $407,200 for triplet or higher-order pregnancies.
FR was developed to decrease such adverse pregnancy outcomes associated with multiple pregnancies. The most frequently used method is transabdominal ultrasound-guided injection of potassium chloride (KCl) into the fetal heart, and the less commonly used method is ultrasound-guided transvaginal aspiration with no injection. A previous meta-analysis of observational studies demonstrated that FR of triplet pregnancies to twin pregnancies was associated with an improved perinatal outcome compared with nonreduced triplet pregnancies, including a lower risk of birth before 32 or 28 weeks of gestation. Another study on 298 triplet pregnancies compared the pregnancy outcomes between triplet pregnancies reduced to singleton or twin pregnancies and triplet pregnancies managed expectantly. FR of triplet pregnancies to singleton or twin pregnancies substantially improved pregnancy outcomes compared with expectant management. However, what was not completely clear was whether FR of triplet pregnancies to singleton pregnancies might further improve pregnancy outcome compared with reduction of triplet (RTT) pregnancies to twin pregnancies.
The main objective of this meta-analysis was to compare the survival and perinatal outcomes of triplet pregnancies after RTT pregnancies vs triplet pregnancies after reduction to singleton (RTS) pregnancies. In addition, network meta-analysis was performed to compare the survival and perinatal outcomes between reduced triplet pregnancies (RTT pregnancies and RTS pregnancies) and expectantly managed triplet pregnancies.
Methods and Materials
This systematic review and meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses checklist. The study protocol for this systematic review was registered in the International Prospective Register of Systematic Reviews (registration number: CRD42021275053).
Search strategy
A systematic literature search was performed by 2 independent authors (K.H. and A.A.N.) on PubMed, Web of Science, Scopus, and Embase from inception to January 16, 2022. The search was conducted using the following key words: “fetal reduction” OR “selective reduction” OR “feticide” OR “fetal termination” OR “pregnancy reduction” OR “multifetal reduction” OR “multifetal termination” OR “elective reduction” AND “triplet” OR “triplets” OR “multiple gestation” OR “multiple pregnancy” OR “high order pregnancy” OR “high order gestation.” References of relevant articles were manually reviewed, and eligible studies were added to results from an electronic literature search.
Inclusion and exclusion criteria
All studies reporting data that compared selective FR of triplet pregnancies to twin pregnancies vs singleton pregnancies were considered. No publishing date, language, or location limitation was imposed. The exclusion criteria were multiple pregnancies with >3 fetuses, monochorionic triplets, no clear description of the technique of reduction, and missing data regarding perinatal outcomes of reduced pregnancies. Moreover, case reports or case series, conference abstracts, and review articles were excluded. Initially selected studies were reviewed for eligibility by 2 independent authors (K.H. and A.A.N.); discrepancies between reviewers were resolved by consulting the third investigator (A.A.S.). Only a small fraction of identified studies met the criteria for inclusion, and only 1 study (Israel) came from the pioneering centers of FR.
Outcome measure
The primary outcome was complete fetal survival rate, defined as a live birth rate at >24 weeks of gestation, of both twins in case of RTT pregnancies and live birth rate of singletons in case of RTS pregnancies. The secondary outcomes included GA at birth, birthweight, incidence of preterm birth (PTB) at <32 and <34 weeks of gestation, early pregnancy loss (<24 weeks of gestation), and neonatal death (up to 28 days after birth).
Data extraction
Data abstraction of included articles using a standardized sheet was performed by 2 independent authors (K.H. and A.A.N.). Data included the following: first author’s name, publication year, country, study design, patients’ demographic data, prenatal variables (as defined in outcome measure), rate of in vitro fertilization or intracytoplasmic sperm injection, technique of FR, and GA at which reduction was performed.
Risk of bias assessment
The Newcastle-Ottawa Scale (NOS) was used to evaluate the quality of included studies and the risk of bias. NOS was composed of “participant selection,” “comparability of study groups,” and “assessment of outcome or exposure.” A score of >7 was considered high quality. Of note, 2 authors assessed all the studies. A group consensus was reached to resolve disagreements.
Certainty of evidence
The Grading of Recommendations Assessment, Development, and Evaluation (GRADE) assessment was used to assess the overall quality of evidence for each outcome by 2 authors (K.H. and S.M.) independently, and a “summary of findings” was designed and shown in Supplemental Table 1 . Disagreements between the 2 review authors on the quality assessment of the evidence were resolved by discussion, with the involvement of the third author (A.A.N.) whenever necessary. Based on the GRADE approach, the quality of evidence was assessed on the basis of the risk of bias, inconsistency, indirectness, imprecision, or publication bias. The quality of evidence was rated as follows:
- •
High certainty: We are very confident that the true effect lies close to that of the estimate of the effect.
- •
Moderate certainty: We are moderately confident in the effect estimate. The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.
- •
Low certainty: Our confidence in the effect estimate is limited. The true effect may be substantially different from the estimate of the effect.
- •
Very low certainty: We have very little confidence in the effect estimate. The true effect is likely to be substantially different from the effect estimate.
Statistical analysis
Statistical analysis was performed using the Cochrane Review Manager software (version 5.3; Nordic Cochrane Center, Cochrane Collaboration, Copenhagen, Denmark). Pooled odds ratio (OR) for dichotomous outcomes or mean difference (MD) for continuous outcomes, with 95% confidence intervals (CIs), were calculated using a random-effects model. I 2 test was performed to assess heterogeneity; a value of >50% was considered to represent low heterogeneity. To provide a range of expected effects if a new study was conducted, 95% prediction intervals (PIs) were calculated for variables if presented in >3 studies. Subgroup analyses were performed to assess the effect of the FR technique (KCl injection vs transvaginal embryo aspiration) and chorionicity (trichorionic triplet [TCT] pregnancy vs dichorionic triplet [DCT] pregnancy) on fetal survival after the procedure. Additional statistical analysis was performed using “netmeta” R package as part of the network meta-analysis to compare the reduced triplet pregnancies (RTT pregnancies and RTS pregnancies) to expectant triplet pregnancies.
Results
Search strategy and study characteristics
A total of 1146 articles were retrieved ( Figure 1 ). Of those articles, 435 were excluded for duplication. The remaining 711 studies were screened for eligibility. Title and abstract screening resulted in 28 potentially eligible studies. After a full-text assessment was performed, 20 studies were found to present data on survival or perinatal outcome of triplet pregnancies. Of the 20 studies, 10 directly compared RTT pregnancies with RTS pregnancies and the remaining studies (n=10) compared reduced triplet pregnancies, either RTT pregnancies or RTS pregnancies, to expectant management (no FR).
The characteristics of studies directly comparing RTT pregnancies and RTS pregnancies are presented in the Table . Included studies were published between 2012 and 2020. Of note, 5 studies were conducted in China, , , , , 2 in the United Kingdom, , , 2 in Israel, , and 1 in France. All studies had a retrospective design, except for 1 study in which the study design was not mentioned. A total of 2392 triplet pregnancies undergoing FR were eligible for meta-analysis, of which 1903 were reduced to twin pregnancies and 489 reduced to singleton pregnancies. In 8 studies (with 2324 patients), FR was performed transabdominally by ultrasound-guided intrathoracic KCl injection, , , , and in 2 studies (with 68 patients), , FR was performed by transvaginal aspiration of embryonic parts.
| First author, year, country | Design | No. of total reduced triplet pregnancies | TCT | DCT | Reduced totwin or singleton pregnancies | GA at reduction | Not conceived by ART | Methods of reduction | NOS |
|---|---|---|---|---|---|---|---|---|---|
| Kuhn-Beck et al, 2012, France | Retrospective | 180 | 180 | NA | 136/44 | 10–12 wk | 104 | KCl injection | 6 |
| Chaveeva et al, 2013, United Kingdom | Retrospective | 343 | 299 | 44 | 280/63 | 10–14 wk | 493 | KCl injection | 8 |
| Li et al, 2013, China | Retrospective | 35 | NA | 35 | 9/26 | 6–8 wk | 88 | Transvaginal aspiration | 6 |
| Lin et al, 2016, China | Retrospective | 33 | NA | 33 | 21/12 | 6–8 wk | 113 | Transvaginal aspiration | 7 |
| Morlando et al, 2015, United Kingdom | Retrospective | 22 | NA | 22 | 12/10 | 8–14 wk | 65 | KCl injection or RFA or laser | 7 |
| Drugan et al, 2017, Israel | NR | 140 a | NA | NA | 105/35 | 12–15 wk | NR | KCl injection | 6 |
| Liu et al, 2019, China | Retrospective | 704 | 632 | 72 | 660/44 | Day 45 to day 65 afterembryo transfer | 704 | KCl injection | 7 |
| Zemet et al, 2020, Israel | Retrospective | 285 | 251 | 34 | 223/62 | 11–14 wk | 152 | KCl injection | 8 |
| Yan et al, 2020, China | Retrospective | 417 | 311 | 106 | 308/109 | 11–14 and 15–24 wk | NR | KCl injection | 6 |
| Cai et al, 2020, China | Retrospective | 233 | NA | 233 | 149/84 | 11–14 wk | 233 | KCl injection | 7 |
Meta-analysis of reduction to twin vs reduction to singleton for triplets
RTT pregnancies had significantly lower survival rate of all fetuses than RTS pregnancies (OR, 0.61; 95% CI, 0.40–0.92, 95% PI, 0.36–1.03; P =.02; I 2 =0%) ( Figure 2 , A). Subgroup analyses assessed the role of chorionicity and FR technique on fetal survival rate. Fetal survival rate for DCT pregnancies after RTT pregnancies was significantly lower than RTS pregnancies (OR, 0.49; 95% CI, 0.28–0.88; 95% PI, 0.19–1.26; P =.02; I 2 =0%); for TCT pregnancies, there was no significant difference between RTT and RTS pregnancies (OR, 1.23; 95% CI, 0.45–3.37; P =.73; I 2 =13%) ( Figure 3 , A). In addition, 2 FR techniques of KCl injection (OR, 0.69; 95% CI, 0.39–1.20; 95% PI, 0.24–1.97; P =.19; I 2 =12%) and transvaginal aspiration (OR, 0.37; 95% CI, 0.08–1.73; P =.21; I 2 =0%) did not significantly affect fetal survival ( Figure 3 , B).
RTT and RTS pregnancies had comparable risk of early pregnancy loss (<24 weeks of gestation) (OR, 0.89; 95% CI, 0.58–1.38; 95% PI, 0.54–1.48; P =0.61; I 2 =0%) and neonatal death (up to 28 days of birth) (OR, 0.56; 95% CI, 0.10–3.21; P =.55; I 2 =28%) ( Figure 2 , B and C).
Triplet pregnancies reduced to twin pregnancies had significantly lower GA at birth (weeks) (MD, −2.20; 95% CI, −2.80 to −1.61; 95% PI, −4.27 to −0.14; P <.001; I 2 =79%) ( Figure 4 , A) and lower birthweight (OR, 8.76; 95% CI, 5.56–13.80; 95% PI, 4.60–16.67; P <.001; I 2 =0.0%) ( Figure 4 , B) than triplet pregnancies reduced to singleton pregnancies. Furthermore, RTT pregnancies were associated with greater risk of PTB at <34 weeks of gestation (OR, 3.04; 95% CI, 1.45–6.36; 95% PI, 0.54–17.18; P =.003; I 2 =22%) ( Figure 4 , C) and PTB at <32 weeks of gestation (OR, 2.14; 95% CI, 1.02–4.49; 95% PI, 0.64–7.13; P =.04; I 2 =0%) ( Figure 4 , D). The GRADE approach indicated that the quality of evidence for the comparisons between RTT and RTS pregnancies was very low for neonatal demise and GA at delivery; low for fetal survival, early pregnancy loss, and PTB at <32 and <34 weeks of gestation; and moderate for low birthweight ( Supplemental Table 1 ).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
