Objective Data
The purpose of this study was to conduct a systematic review and meta-analyses of literature regarding the risk of preterm birth in singleton pregnancies after a preterm twin birth.
Study
We conducted a literature search of Embase, Ovid Medline, and Cochrane from inception until February 28, 2019. Studies that evaluated women with a previous twin birth followed by a singleton birth were included.
Study Appraisal and Synthesis Methods
Data were abstracted in duplicate, and summary odds ratios and confidence intervals were calculated with the use of random effects model. Risk of bias was assessed with the use of the Newcastle-Ottawa Scale, and quality of evidence was evaluated with the use of the Grades of Recommendation, Assessment, Development, and Evaluation approach. The primary outcome was a preterm singleton birth (<37 weeks gestational age); secondary outcomes included risk of late preterm birth (34–36+6 weeks gestation), preterm birth between 30 and 33+6 weeks gestation, and preterm birth at <30 weeks gestation. Subanalysis of risk of singleton preterm birth after spontaneous twin birth was also performed. A priori, the protocol was developed and registered with PROSPERO (2016; registration number: CRD42017053382).
Results
We included 8 cohort studies at low risk-of-bias met inclusion criteria for the systematic review and 6 for the meta-analysis. Compared with women with previous term twin births, women who had previous preterm (<37 weeks gestation) twin births were at increased odds of preterm singleton birth in subsequent pregnancy (odds ratio, 4.34; 95% confidence interval, 2.83–6.65). Gestational age at birth of previous twin pregnancy was an effect modifier. Compared with previous term twin births, the odd ratios of subsequent preterm singleton birth were 2.13 (95% confidence interval, 1.21–3.74) if twins were born between 34 and 36+6 weeks gestation, 5.18 (95% confidence interval, 2.78–9.64) if twins were born between 30 and 33+6 weeks gestation, and 9.78 (95% confidence interval, 4.99–18.98) if twins were born at <30 weeks gestation. A similar trend was seen for the risk of singleton preterm birth after spontaneous twin preterm birth.
Conclusion
A history of preterm twin birth is associated with higher odds of subsequent preterm singleton birth. The odds increase with decreasing gestational age of previous twin birth.
Prematurity, defined as birth at <37 completed weeks gestational age, is a leading cause of infant morbidity and death in North America and worldwide. More than 10% of infants in the United States are born preterm and account for 15–16% of all deaths under the age of 5 years. Many risk factors for preterm birth (PTB) have been identified and include multiple pregnancy and previous PTB. Multiple pregnancy is increasingly more prevalent because of more women of advanced maternal age and the use of artificial reproductive technology. ,
Why was this study conducted?
The purpose of this study was to examine the risk of preterm birth in a singleton pregnancy after previous twin preterm birth.
Key findings
A previous preterm twin birth increases the risk of premature birth in a subsequent singleton pregnancy.
What does this add to what is known?
A history of premature birth, whether it is from a previous singleton or twin birth, confers an increased risk in a subsequent singleton pregnancy
Having a history of PTB in a singleton pregnancy confers increased risk in each subsequent pregnancy. The recurrence of spontaneous preterm labor in singletons is estimated to be 7–30%. This phenomenon is likely due to the intrinsic risk of PTB in that woman. However, when there is a pregnancy with an independent risk factor for PTB, such as multiple pregnancy, is there a recurrence risk? Several cohort studies have sought to answer this question with retrospective methods; the degree of risk is unclear. Currently, a controversy exists whether a previous twin PTB confers an increased risk of subsequent singleton PTB.
The aim of this systematic review and meta-analysis was to assess the risk of PTB in singleton pregnancy after previous twin PTB.
Methods
Eligibility criteria, information sources, search strategy
An information specialist (D.H.) performed a comprehensive search of the Medline, Embase, and Cochrane databases (latest search February 28, 2019) for relevant literature regarding recurrence of prematurity in singleton pregnancies after a preterm twin pregnancy. The search included combinations of controlled vocabulary and key words. The detailed search strategy can be viewed in Appendix A . The clinicaltrials.gov website was also searched for relevant literature. In addition, the first 200 hits on Google (Google LLC, Mountain View, CA) were searched for relevant literature. No geographic or language restrictions were applied. References of studies that were excluded after we screened full text, as described in Figure 1 , are listed in Appendix B .
Study selection
A priori, this systematic review was registered in the PROSPERO database (2016; registration number: CRD42017053382). A study variation to include secondary analyses by gestational age and spontaneous twin PTB was submitted for amendment in Prospero. The methods were performed in accordance with the PRISMA guidelines ( http://www.prisma-statement.org ).
Inclusion criteria included prospective cohort, retrospective cohort, and randomized controlled trials with a previous twin birth and subsequent singleton birth in the same woman. Study population, age, or other demographic data were not used to limit the relevance of studies. Exclusion criteria included case series, case reports, nonhuman studies, review papers, posters, and abstracts.
Data extraction
Authors (R.M. and A.L.K.L.) independently screened all titles and abstracts. Relevant studies were then pulled for full text review. These 2 authors independently scrutinized each full text selected article for inclusion criteria, and data were extracted. Agreement was achieved for all selected studies; consensus by a third author was not required.
Assessment of risk of bias
Risk of bias was assessed with the use of the Newcastle-Ottawa Scale. The Newcastle-Ottawa Scale assessed study selection (representative of exposed cohort, selection of nonexposed cohort, ascertainment of exposure, incident disease), comparability, and outcome (assessment of outcome, length of follow up, adequacy of follow up).
The primary outcome was the risk of singleton PTB after previous twin PTB. Secondary analyses were performed for the risk of singleton PTB after previous twin birth at 34–36+6, 30–33+6, and <30 weeks gestation. Analyses were also performed for each of these gestational ages for risk of singleton PTB after previous spontaneous twin PTB. Original data from our team’s 2019 publication was extracted to facilitate secondary analysis of spontaneous singleton PTB.
Data synthesis
Data were analyzed with the use of Review Manager 5.3 software. Completed analysis was assessed for accuracy by 2 authors (R.M. and A.L.K.L.). Meta-analysis was performed with a Mantel-Haenszel random effects model because statistical homogeneity was present between studies. Odds ratios (OR) with 95% confidence interval (CI) were calculated. The GRADE approach was used to rate the quality of evidence.
Results
Study selection
The results of the literature search revealed 3524 citations. After the review, 8 studies fulfilled inclusion criteria and were included in the systematic review ( Figure 1 ).
Study characteristics
All included studies used a retrospective cohort method that used hospital or national databases ( Table 1 ). Characteristics of included studies are displayed in Table 1 .
| Study | Year | Type of study | Setting | Population | Exposure assessment | Outcome assessed | Twin preterm birth type specified |
|---|---|---|---|---|---|---|---|
| Menard et al | 1996 | Retrospective cohort study | South Carolina | Medical University of South Carolina (1981–1993), n=144 | Hospital medical records | Gestational age in singleton subsequent delivery | No |
| Rydhstroem | 1998 | Retrospective cohort study | Sweden | Swedish Birth Registry (1973–1993), n=1216 | Registry log | Gestational age in singleton subsequent delivery | No |
| Bloom et al | 2001 | Retrospective cohort study | Texas | Parkland Hospital (1988–1999), n=82 | Hospital medical ecords | Gestational age in singleton subsequent delivery | No |
| Facco et al | 2007 | Retrospective cohort study | Illinois | Northwestern Memorial Hospital (1995–2005), n=167 | Hospital medical records | Gestational age in singleton subsequent delivery | Yes |
| Rafael et al | 2012 | Retrospective cohort study | Delaware | Christiana Health Care Systems Hospital (1996–2010), n=255 | Hospital medical records | Gestational age in singleton subsequent delivery | Yes |
| Schaaf et al | 2012 | Retrospective cohort study | The Netherlands | Netherlands Birth Registry (1999–2007), n=1957 | Registry log | Gestational age in singleton subsequent delivery | Yes |
| Ouh et al | 2018 | Retrospective cohort study | Korea | Korean National Health Insurance Claims Database (2007–2014), n=475 | Database log | Gestational age in singleton subsequent delivery | No |
| Menzies et al | 2019 | Retrospective cohort study | Canada | Mount Sinai Hospital & Sunnybrook Health Sciences Centre (2000–2016), n=378 | Hospital medical records | Gestational age in singleton subsequent delivery | Yes |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
