Summary
Background
Short interpregnancy interval (IPI) following caesarean delivery is associated with uterine rupture in subsequent pregnancies. However, the interval required to minimise this risk is unknown. We investigated how the interval between pregnancies and induction or augmentation of labour affect the likelihood of uterine rupture among parturients with one previous livebirth by caesarean delivery who had a subsequent trial of labour.
Methods
In this population-based cohort study, we used data from U.S National Vital Statistics System from 2011 to 2021. Multiple pregnancies and births of infants with congenital abnormalities were excluded. A linear spline logistic regression with one knot was used to assess the relationship between uterine rupture and interpregnancy interval for spontaneous and for induced/augmented labours. Multivariable logistic regression was performed with multiple imputation and stepwise backward elimination to adjust for maternal demographic and clinical factors including maternal age, height, and BMI and gestational age. The predicted risk of uterine rupture was tabulated for interpregnancy intervals between zero and 21 months. Adverse outcomes were compared between labours with and without uterine rupture.
Findings
We examined 491,998 trials of labour among parturients with one previous livebirth by caesarean delivery and no previous vaginal births. The odds ratio (OR) of uterine rupture per three months interpregnancy interval was 0.91 (95% CI 0.88–0.94) between zero and 21 months after adjusting for confounders, with no further change in risk detected beyond 21 months. The OR was 2.51 (95% CI 2.27–2.78) for induced or augmented labours compared with spontaneous labours. Other factors associated with uterine rupture included older maternal age, shorter maternal height, more advanced gestational age (from 35 to 43 weeks), and heavier birthweight. Predicted rates of uterine rupture ranged from 0.36% at zero to 0.19% at 21 months’ interpregnancy interval for spontaneous labours and from 0.91% to 0.47% for induced/augmented labours for parturients with a typical clinical and demographic background. When uterine rupture occurred, the rates of unplanned hysterectomy, intrapartum or neonatal death, and neonatal seizures were 4.0% (95% CI 3.2–5.1%), 3.7% (95% CI 2.7–5.1%), and 2.6% (95% CI 1.8–3.3%) respectively.
Interpretation
The risk of uterine rupture progressively decreases as IPI increases until about 21 months and then stabilises. Counselling should advise that for women choosing between a planned TOLAC or a planned caesarean delivery after one previous caesarean delivery and no previous vaginal births waiting until 21 months or longer after a prior low transverse caesarean delivery might minimise the risk of uterine rupture. The absolute risk of certain serious maternal and fetal/neonatal complications such as unplanned hysterectomy and perinatal death is low.
Funding
No funding.
Keywords: Uterine rupture, Interpregnancy interval, Caesarean section, Vaginal birth after, Caesarean, Trial of labour after caesarean
Research in context.
Evidence before this study
Several studies have categorised the interpregnancy interval (IPI), between a caesarean delivery and a subsequent trial of labour after caesarean (TOLAC), into two or more groups to compare rates of uterine rupture for short versus long IPIs. Some of these studies found that shorter IPIs are associated with an increased risk of uterine rupture, and on this basis guidelines in Australia, Canada, and the United States recommend a minimum IPI of nine months before a TOLAC.
We searched Medline for relevant studies, from its inception to June 2023 using a strategy that included but was not limited to: (TOLAC.tw OR trial of lab∗.tw OR vbac.tw) AND (caesar∗.tw OR cesar∗.tw OR exp Cesarean Section/) AND (exp Uterine Rupture/OR (ruptur∗ adj5 uter∗).tw AND interval.tw). We reviewed relevant original research and review articles and checked their references for additional relevant sources. Our search identified six observational studies with sample sizes ranging from 170 to 17,890 TOLACs that examined IPI as a risk factor for uterine rupture. Some larger studies showed an association between shorter IPIs and uterine rupture with cut-off points for categorising the length of IPI ranging between about six and 15 months. While most evidence supports an association between short IPI and uterine rupture, there is uncertainty around the magnitude of any effect and the optimal cut-off for defining a “short” IPI.
Added value of this study
To the best of our knowledge, this was the largest observational study assessing the relationship between IPI and uterine rupture. It is the only study to estimate the IPI required to minimise the risk of uterine rupture and is the only study to estimate month-by-month rates of uterine rupture by IPI. It provides added insight into the relationship between uterine rupture and other risk factors including increasing gestational age, maternal age, and birthweight as well as decreasing maternal height and body mass index. It provides estimates of the risks of complications of uterine rupture including stillbirth or neonatal death, unplanned hysterectomy and neonatal seizures.
Implications of all the available evidence
More specific advice can be given to those who are choosing between a planned TOLAC or a planned caesarean delivery after one previous caesarean delivery and no previous vaginal births. Such women may be advised that aiming for a longer IPI can reduce their risk of uterine rupture, with the risk reaching a minimum at around 21 months. Estimates of the risk of uterine rupture and other severe adverse outcomes can be provided.
Introduction
In the United States and Australia, 32%–37% of all mothers gave birth by caesarean delivery in 2021,1,2 and 31% gave birth by caesarean delivery in the United Kingdom in 2019.3 Recent reports demonstrate that having a previous caesarean delivery is now the most common indication for a caesarean delivery in the current pregnancy.4,5 Conversely, despite a trial of labour after caesarean (TOLAC) being considered safe for most women,6,7 only 12%–14% of women with one or more previous caesarean deliveries had a vaginal birth.1,2
Uterine rupture, often defined as any full thickness separation of the myometrial layer and the visceral peritoneum,6 typically occurs at the site of a previous caesarean scar but can also occur following other uterine surgery such as myomectomy. The cardinal risk factor for uterine rupture is a TOLAC, as rupture in the absence of uterine contractions is rare.6 Consequently, uterine rupture may be avoided in most cases if a caesarean delivery is performed before labour. Hence, when deciding on the planned mode of birth, the risks of a planned caesarean delivery must be balanced against the risks of a TOLAC, primarily uterine rupture. To deepen understanding of the risks of the latter choice, we focus only on TOLAC outcomes, and particularly the primary serious risk which is uterine rupture. Uterine rupture in turn, is associated with serious complications including perinatal mortality, neonatal hypoxic ischaemic encephalopathy, and maternal hysterectomy.6, 7, 8
Risk factors for uterine rupture among women undergoing a TOLAC include shorter interpregnancy interval (IPI, the time from the index caesarean section to conception), post-dates pregnancy, fetal macrosomia, induction or augmentation of labour, and lower pre-labour Bishop score.6 Although IPI is perhaps the most modifiable of these, data around the importance of IPI as a risk factor for uterine rupture are conflicting, both with respect to the relative risk and the length of the interval required to minimise the risk. Several smaller studies have shown shorter IPIs among women undergoing a TOLAC to be associated with uterine rupture, with an arbitrary cut-off used to define ‘short IPI’ ranging from six to 24 months,9, 10, 11, 12, 13, 14, 15 while others have not shown a difference, potentially due to small sample sizes.16,17 One large study included women who had a prelabour caesarean delivery but did not report rates of uterine rupture among women undergoing a TOLAC.18 As a result, considerable uncertainty exists about the relationship between IPI and uterine rupture, leaving open questions about the optimal gap needed for a safe trial of labour after caesarean in a subsequent pregnancy.
Using a large U.S population-based dataset, we aimed to answer the study question: Among women with one previous caesarean section and no previous vaginal deliveries who undergo a TOLAC, what is the relationship between IPI and the risk of uterine rupture, with and without induction or augmentation of labour?
Methods
Study design
This was a retrospective population-based observational study where the primary exposure variables were IPI and induction or augmentation of labour and the primary outcome measure was uterine rupture. IPI was defined as the number of months between the previous birth and the estimated date of the last menstrual period in the current pregnancy. Induction of labour was commencing the labour by artificial means and augmentation was stimulating contractions by drugs or manipulation after labour had commenced. The study protocol is available at https://www.slhd.nsw.gov.au/pdfs/inter-pregnancy-study-protocol.pdf.
Ethics
Ethics approval was not required nor sought from any ethical committee, as publicly available data were used.
Data
Demographic and health data were obtained from the U.S National Vital Statistics System of the National Centre for Health Statistics (https://www.cdc.gov/nchs/data_access/vitalstatsonline.htm). Birth data covered years where uterine rupture was available (years Jan 2011 to Dec 2021). Fetal death data covered the years Jan 2014 to Dec 2021.
To examine the outcome of neonatal death, we used period linked birth-infant death data for years Jan 2014 to Dec 2019, with 2019 being the most recent year available.
Interpregnancy interval was calculated by subtracting the gestational age at birth in the current pregnancy from the interdelivery interval (IDI, the time from the index caesarean delivery to the time of birth in the current pregnancy).
Uterine rupture was collected as a tick-box answer to the question in the Facility Worksheet for the Live Birth Certificate and was defined as a full-thickness disruption of the uterine wall involving the overlaying visceral peritoneum. Cases where the fetus, placenta and umbilical cord remained contained within the uterine cavity and silent or incomplete ruptures or asymptomatic separations were excluded. Although this definition differs from that in other studies, it is the only one available in the data.
Other clinical and demographic variables collected were maternal (‘delivery’) weight at the time of birth, age, height, race (Black, White, American Indian or Alaskan Native, Other), pre-pregnancy smoking (yes/no), year of birth, payment type (Medicaid, self-pay and ‘other’ grouped as ‘Public’, and Private Insurance grouped separately), maternal level of education (five categories as described in Supplementary Materials, Part 1), diabetes (no diabetes, gestational diabetes, pre-existing diabetes), hypertension in pregnancy (yes/no), onset of antenatal care (first, second, third trimester, or no antenatal care), induction of labour, augmentation of labour, plurality, gestational age at birth (using the obstetric estimate—see Supplementary Materials Part 1), birthweight, and 5-min Apgar score.
Other maternal outcome variables collected were unplanned hysterectomy, blood transfusion, and intensive care unit (ICU) admission. Other fetal/neonatal outcome variables collected were stillbirth (fetal death), neonatal death within 28 days of birth (available for 2014 to 2019 data only), 5-min Apgar score <4, neonatal seizures, admission to the neonatal intensive care unit (NICU), immediate neonatal ventilation (at birth), and neonatal ventilation for more than 6 h.
Population
The population consisted of all reported livebirths in the United States from 2011 to 2021 for women recorded as having one previous livebirth by caesarean delivery and no previous vaginal births using the 2003 revision of the U.S. Certificate of Live Birth. Due to the available data fields, it was not possible to exclude rare cases where there could have been both a stillbirth and livebirth in the past, one of which was a vaginal birth.
The inclusion criteria were vaginal birth or birth by caesarean delivery with a trial of labour attempted from 17 weeks gestational age onwards. These criteria represented women undergoing a TOLAC. Parturients with spontaneous, augmented and induced labours were included to facilitate counselling around the risks of these types of labour. Exclusion criteria were triplets or higher birth order, and congenital abnormalities in the infant. There were post-hoc decisions to exclude twins as there were only two uterine ruptures in this group, and to exclude gestational ages greater than 43 or greater than the interdelivery interval, as these were more likely to result from data entry errors or are of reduced relevance to typical clinical circumstances. Additionally, a post-hoc decision was made, to conduct a complete case sensitivity analysis, for all livebirths and stillbirths compared with livebirths only for the years 2014–2021. This decision was prompted by the unavailability of uterine rupture data for stillbirths before 2014. For the combined outcome of fetal death (stillbirth) or neonatal death within 28 days of birth, data from stillbirths and period linked data were utilised, spanning 2014 to 2019.
Thus, we used three datasets:
-
(1)
All livebirths from 2011 to 2021 for the main analysis.
-
(2)
All livebirths and stillbirths (2014–2021) were compared with livebirths only (2014–2021) in a sensitivity analysis.
-
(3)
For the combined outcome of fetal death (stillbirth) and neonatal death within 28 days of birth, analysis was restricted to 2014 to 2019.
Outcomes
The pre-specified primary outcome was tabulated predictions for uterine rupture risk in typical parturients according to number of months of IPI, and whether labour onset was spontaneous. Risk of secondary adverse outcomes were also assessed among the rupture group.
Statistics
In a preliminary analysis, the frequency of uterine rupture was examined for IPIs grouped into intervals (as described in Supplementary Materials Part 2) and visually examined. The results suggested that the simplest model of the relationship between IPI and uterine rupture was a linear spline logistic regression with one knot.
An algorithm was pre-specified to determine the best knot position for the linear spline logistic regression of uterine rupture on IPI. This algorithm is described in Supplementary Materials Part 2. The algorithm was repeated in 10,000 bootstrapped samples to obtain 80% confidence bounds around the best knot position.
Overall missingness when including IPI was over 5% (see Table 1 below), so following knot selection, multiple imputation with fully conditional specification was used to create five imputed datasets. All variables utilised in all subsequent modelling, including IPI, explanatory variables, derived terms and maternal/perinatal outcome variables were used for the multiple imputation using the MICE package in R. These are listed in Supplementary Materials Part 1. Standard fully conditional regression forms were used for each variable, according to the type of variable, such as logistic regression for dichotomous variables. Dependencies amongst variables were handled with passive imputation.
Table 1.
Demographic and clinical characteristics among 491,998 parturients who underwent a trial of labour in the United States following one previous caesarean section and no previous vaginal births.
| Factor | Missing data n (%) | |
|---|---|---|
| Interpregnancy interval (months) (median, IQR) | 28 (17–51) | 29,485 (6.0) |
| Maternal BMI (kgm−2) (median, IQR) | 30.9 (27.5–35.3) | 11,002 (2.2) |
| Maternal height (cm) (mean, SD) | 162 (7.2) | 4018 (0.8) |
| Maternal age (years) (median, IQR) | 30 (26–33) | None |
| Gestational age (weeks) (median, IQR) | 39 (38–40) | None |
| Race (n %) | ||
| White | 345,854 (70.3) | |
| Black | 83,321 (17.0) | None |
| American Indigenous or Alaskan Native | 3859 (0.8) | |
| Other | 58,964 (12.0) | |
| Pre-pregnancy smoking (n %) | ||
| Non-smoker | 454,105 (92.3) | 8026 (1.6) |
| Smoker | 29,867 (6.1) | |
| Year of birth (n %) | ||
| 2011 | 32,945 (6.7) | |
| 2012 | 36,230 (7.4) | |
| 2013 | 38,618 (7.9) | |
| 2014 | 43,146 (8.8) | |
| 2015 | 46,378 (9.4) | |
| 2016 | 49,904 (10.1) | None |
| 2017 | 49,012 (10.0) | |
| 2018 | 49,676 (10.1) | |
| 2019 | 49,468 (10.0) | |
| 2020 | 47,752 (9.7) | |
| 2021 | 48,869 (9.9) | |
| Payment type (n %) | ||
| Public | 223,421 (45.4) | 3774 (0.8) |
| Private | 264,803 (53.8) | |
| Education (n %) | ||
| 12th grade or less | 52,066 (10.6) | |
| High school graduate, tertiary credit or | ||
| Associate degree | 234,931 (47.8) | |
| Bachelor’s degree | 119,067 (24.2) | 5767 (1.2) |
| Master’s degree | 61,615 (12.5) | |
| PhD or Professional degree | 18,852 (3.8) | |
| Diabetes (n %) | ||
| No diabetes | 455,453 (92.6) | |
| Gestational diabetes | 31,804 (6.5) | None |
| Pre-existing diabetes | 4741 (1.0) | |
| Hypertension during pregnancy (n %) | None | |
| No | 464,406 (94.4) | |
| Yes | 27,592 (5.6) | |
| Onset of antenatal care (n %) | ||
| First trimester | 217,926 (44.3) | |
| Second trimester | 218,055 (44.3) | 13,696 (2.8) |
| Third trimester | 35,488 (7.2) | |
| No antenatal care | 6833 (1.4) | |
| Birthweight (g) (mean, SD) | 3313 (568) | 451 (0.1) |
| Onset of labour type (n %) | ||
| Spontaneous onset | 301,389 (61.3) | 134 (<0.1) |
| Labour induced or augmented | 190,475 (38.7) |
IQR = interquartile range; OR = odds ratio; SD = standard deviation.
To complement the primary multivariate logistic regression modelling, univariable linear spline logistic regression curves fitted to the imputed data were plotted along with the frequency of uterine rupture for IPIs grouped into the intervals described in Supplementary Materials Part 2.
Multivariate logistic regression modelling was then conducted via backward stepwise elimination in the imputed datasets. Important covariates were identified from amongst the available data fields for inclusion in the base model as potential predictors of uterine rupture: Interpregnancy interval, labour induced/augmented (yes/no), maternal age, gestational age (as a spline with one knot as described below), maternal height, maternal body mass index (BMI), birthweight, hypertension during pregnancy (yes/no), diabetes (pre-existing, gestational, none), maternal race (Black, White, Indigenous American or Alaskan Native, Other), pre-pregnancy cigarette smoking (yes/no), year of birth (11 categories from 2011 to 2021), payment type (public/private), maternal education (five categories as described in Supplementary Materials, Part 1), and trimester of onset of maternity care (four categories: first, second or third trimester, or no antenatal care). A further interaction term was included in the base model allowing the slope of the spline for IPI to differ for spontaneous versus induced/augmented labours.
The modelling plan included a check of the significance of the slope in the IPI spline segment after the knot, and fitting a model with zero slope in this segment, if that slope was insignificant.
For the continuous explanatory variables, linearity with the log-odds of uterine rupture was assessed by including squared terms in the backward stepwise elimination. If the squared term was eliminated, then only a linear model was considered thereafter.
On clinical grounds, the relationship between gestational age and log-odds of uterine rupture was considered unlikely to be linear across the entire range of gestational ages in the data set,18,19 so we conducted a preliminary analysis (Supplementary Materials Part 2). Based on this, we fitted gestational age as a linear spline with a single knot at 35 weeks. The cases with gestational age below 35 weeks made up less than 6% of the data.
To support clinical counselling, the final fitted multivariate model was used to tabulate predicted probabilities and risk upper bounds for two typical cases by IPI (see Tables 2 and 3 below, and details in Supplementary Materials Part 2).
Table 2.
Predicted probabilities of uterine rupture for a parturient undergoing a trial of labour after one previous caesarean section and no previous vaginal births, with spontaneous onset of labour.a
| IPI (months) | Probabilitya (upper limit 95% CI) | ‘Risk’a (upper limit 95% CI) | IPI (months) | Probabilitya (upper limit 95% CI) | ‘Risk’a (upper limit 95% CI) |
|---|---|---|---|---|---|
| 0 | 0.36% (0.48%) | 1 in 275 (1 in 208) | 11 | 0.26% (0.32%) | 1 in 388 (1 in 309) |
| 1 | 0.35% (0.46%) | 1 in 283 (1 in 216) | 12 | 0.25% (0.31%) | 1 in 400 (1 in 320) |
| 2 | 0.34% (0.45%) | 1 in 292 (1 in 224) | 13 | 0.24% (0.30%) | 1 in 413 (1 in 331) |
| 3 | 0.33% (0.43%) | 1 in 302 (1 in 233) | 14 | 0.23% (0.29%) | 1 in 426 (1 in 342) |
| 4 | 0.32% (0.41%) | 1 in 311 (1 in 241) | 15 | 0.23% (0.28%) | 1 in 439 (1 in 353) |
| 5 | 0.31% (0.40%) | 1 in 321 (1 in 250) | 16 | 0.22% (0.27%) | 1 in 453 (1 in 365) |
| 6 | 0.30% (0.39%) | 1 in 331 (1 in 260) | 17 | 0.21% (0.27%) | 1 in 468 (1 in 377) |
| 7 | 0.29% (0.37%) | 1 in 342 (1 in 269) | 18 | 0.21% (0.26%) | 1 in 483 (1 in 389) |
| 8 | 0.28% (0.36%) | 1 in 353 (1 in 279) | 19 | 0.20% (0.25%) | 1 in 498 (1 in 401) |
| 9 | 0.27% (0.35%) | 1 in 364 (1 in 289) | 20 | 0.19% (0.24%) | 1 in 514 (1 in 413) |
| 10 | 0.27% (0.33%) | 1 in 376 (1 in 299) | 21+ | 0.19% (0.24%) | 1 in 530 (1 in 425) |
Predicted probabilities (also expressed as ‘risk’) are for a parturient aged 30 years, with body mass index 30 kgm−2, height 165 cm, birthweight 3500 g, gestational age 40 weeks, year of birth 2015, and first trimester onset of antenatal care, in the absence of induction or oxytocin augmentation of labour.
Table 3.
Predicted probabilities of uterine rupture for a parturient undergoing a trial of labour after one previous caesarean section and no previous vaginal births, with induced onset or oxytocin augmentation of labour.a
| IPI (months) | Probabilitya (upper limit 95% CI) | ‘Risk’a (upper limit 95% CI) | IPI (months) | Probabilitya (upper limit 95% CI) | ‘Risk’a (upper limit 95% CI) |
|---|---|---|---|---|---|
| 0 | 0.91% (1.2%) | 1 in 110 (1 in 84) | 11 | 0.65% (0.81%) | 1 in 155 (1 in 124) |
| 1 | 0.88% (1.15%) | 1 in 113 (1 in 87) | 12 | 0.63% (0.78%) | 1 in 160 (1 in 128) |
| 2 | 0.85% (1.11%) | 1 in 117 (1 in 90) | 13 | 0.61% (0.75%) | 1 in 165 (1 in 133) |
| 3 | 0.83% (1.07%) | 1 in 121 (1 in 93) | 14 | 0.59% (0.73%) | 1 in 170 (1 in 137) |
| 4 | 0.8% (1.03%) | 1 in 125 (1 in 97) | 15 | 0.57% (0.71%) | 1 in 176 (1 in 142) |
| 5 | 0.78% (1.00%) | 1 in 128 (1 in 100) | 16 | 0.55% (0.68%) | 1 in 181 (1 in 146) |
| 6 | 0.75% (0.96%) | 1 in 133 (1 in 104) | 17 | 0.54% (0.66%) | 1 in 187 (1 in 151) |
| 7 | 0.73% (0.93%) | 1 in 137 (1 in 108) | 18 | 0.52% (0.64%) | 1 in 193 (1 in 156) |
| 8 | 0.71% (0.89%) | 1 in 141 (1 in 112) | 19 | 0.5% (0.62%) | 1 in 199 (1 in 161) |
| 9 | 0.69% (0.86%) | 1 in 146 (1 in 116) | 20 | 0.49% (0.60%) | 1 in 205 (1 in 165) |
| 10 | 0.67% (0.83%) | 1 in 150 (1 in 120) | 21+ | 0.47% (0.59%) | 1 in 212 (1 in 170) |
Predicted probabilities (also expressed as ‘risk’) are for a parturient aged 30 years, with body mass index 30 kgm−2, height 165 cm, birthweight 3500 g, gestational age 40 weeks, year of birth 2015, and first trimester onset of antenatal care, with induction or oxytocin augmentation of labour.
The final multivariate fitted model from the imputed datasets was replicated in the complete data in a sensitivity analysis. To identify the potential influence of including the stillbirth dataset on the overall result, a multivariable logistic regression was fitted in the combined live/still birth dataset using variables common to both datasets in a post-hoc analysis. For this, the subset of the raw (non-imputed) data was used, where all rows had complete data, including for the secondary outcome variables.
To investigate secondary outcomes, IPI was then categorised into two groups based on the knot position in the linear spline regression and outcomes were compared by IPI group (IPI ≤ knot position v IPI > knot position), in the uterine rupture cohort, using χ2 tests for all dichotomous outcome variables and the Wilcoxon Rank Sum test for Apgar score. If no significant relationship between IPI and the outcome was found, the outcome was reported by whether uterine rupture occurred. The secondary outcomes were unplanned hysterectomy, maternal blood transfusion, maternal ICU admission, intrapartum or neonatal death, 5-min Apgar score <4, neonatal seizures, NICU admission, immediate neonatal ventilation and neonatal ventilation for >6 h.
All analyses were performed using R (Version 4.4.0).
Role of funding source
There was no funding source for this study.
Results
Among 42,519,424 recorded livebirths in the U.S between 2011 and 2021, 491,998 met our inclusion criteria, having undergone a TOLAC in the current pregnancy following one previous caesarean delivery and no vaginal births in the past, after excluding multiple pregnancies and births of infants with congenital anomalies (Fig. 1). Table 1 displays baseline maternal demographic and clinical characteristics. The median IPI was 28 months, and the median age was 30 years. Missing data varied from none to 2.8% for all variables, except for IPI, which had 6.0% missing data. Complete cases, including secondary outcomes constituted 87.1% of the data.
Fig. 1.
Study flow chart.
Among singleton pregnancies, the overall rate of uterine rupture was 0.34% (1647/491,602) after excluding 396 births where rupture status was not recorded. Among 4317 twin births, there were two uterine ruptures for a rupture rate of 0.05% (95% confidence interval [CI] 0.01–0.17%). Due to the low event rate in twin pregnancies, twins were excluded from any further analysis.
The rate of uterine rupture was 0.20% (608/301,124) for spontaneous labours and 0.55% (1038/190,372) for induced/augmented labours after excluding 502 cases where uterine rupture and/or onset of labour was unknown. We were unable to separate induced and augmented labours as there were 19,857 births where the labour was labelled as both induced and augmented, despite the definition of augmented labour specifically excluding induced labours. Induced and augmented were therefore combined into a single group (with uterine rupture rates of 0.62% and 0.50% for induced-only and augmented-only labours respectively).
The rate of vaginal birth was 64.1% (315,142/491,998) among parturients undergoing a TOLAC.
The knot locating algorithm found the best linear spline regression knot at 21 months’ IPI. The bootstrapped 80% confidence range for the knot position was 19–24 months, although this is not intended as a reliable confidence interval but as an indication of the statistical uncertainty in our knot fitting procedure. Fig. 2 shows the raw rates of uterine rupture by IPI in the complete data with fitted curves in the univariate linear spline regression in the imputed datasets superimposed. There was a linear decline in the log-odds of uterine rupture between zero and 21 months for spontaneous and induced/augmented labours.
Fig. 2.
Univariate fitted (solid curves with 95% Confidence Regions) and observed (point estimates with 95% Confidence Intervals) for uterine rupture rates versus inter-pregnancy interval, grouped by onset of labour type among 491,998 trials of labour for parturients with one previous caesarean delivery and no vaginal births.
Stepwise backward elimination on the multiply imputed data resulted in the following variables being eliminated: hypertension during pregnancy, diabetes, maternal race, pre-pregnancy cigarette smoking, payment type, and maternal education. The interaction term of IPI below 21 months with augmented/induced was also eliminated, as were all quadratic terms, indicating no evidence of non-linearities.
The odds ratio per month for the spline segment for IPI above 21 months was 1.0 (95% CI 0.99 to 1.0; p = 0.13). This was not statistically different from one, indicating no observed relationship between IPI and uterine rupture from 21 months onwards. As specified in the modelling plan, the spline segment above 21 months was therefore fixed to be constant and the model re-estimated with stepwise backward elimination. All variables previously eliminated were again eliminated.
Table 4 shows the results of the final multivariable logistic regression with multiple imputation for the primary outcome of uterine rupture. The adjusted odds ratio for uterine rupture was 0.91 (95% CI 0.88 to 0.94) for every three-month increase in IPI between zero and 21 months. The adjusted odds ratio was 2.51 (95% CI 2.27 to 2.78) for induced/augmented labours compared with spontaneous labours. In the final model, uterine rupture increased with shorter maternal height, higher birthweight, older maternal age and higher gestational age (between 35 and 43 weeks) and decreased with higher maternal BMI (Table 4).
Table 4.
Factors associated with uterine rupture among 491,998 parturients who underwent a trial of labour in the United States following one previous caesarean section and no previous vaginal births in the multivariable logistic regression after stepwise elimination with multiple imputation.
| ORadj (95% CI)a | p | |
|---|---|---|
| Primary exposures | ||
| Interpregnancy interval (per 3 months from zero to 21 months) | 0.91 (0.88–0.94) | <0.0001 |
| Labour induced or augmentedb | 2.51 (2.27–2.78) | <0.0001 |
| Covariatesc | ||
| Maternal height (per 5 cm) | 0.89 (0.86–0.92) | <0.0001 |
| Birthweight (per 500 g) | 1.18 (1.12–1.25) | <0.0001 |
| Maternal age (per 5 years) | 1.13 (1.08–1.19) | <0.0001 |
| Gestational age (per week from 35 to 43 weeks)d | 1.12 (1.09–1.16) | <0.0001 |
| Maternal BMI (per 5 kgm−2) | 0.94 (0.90–0.98) | 0.0018 |
| Onset of antenatal care | ||
| First trimester | 1.0e | |
| Second trimester | 1.15 (1.04–1.28) | 0.026 |
| Third trimester | 1.12 (0.93–1.36) | |
| No antenatal care | 0.79 (0.46–1.34) | |
CI = confidence interval; ORadj = odds ratio adjusted for the parameters in the table and year of birth.
Adjusted for all variables in the table, and also for year of delivery (fitted odds ratios for birth year are in Supplementary Materials Part 2).
Compared with spontaneous onset of labour.
Adjusted odds ratios for covariates should be interpreted with caution due to potential causal pathways (the ‘Table 2 Fallacy’). All of these odds ratios are direct effects and must only be considered in combination. For example, the odds ratio for gestational age cannot be considered in isolation because an increase of one week of gestational age is expected to be accompanied by an increase in birthweight due to fetal growth, leading to a larger total effect.
Linear spline with a knot 35 weeks’ gestational age (spline for 35+ weeks reported here).
Referent group.
The results of the multivariable logistic regressions in the multiply imputed datasets and complete cases were similar (Supplementary Fig. S1).
For years 2014–2021, the overall rate of uterine rupture was 0.36% (1414/389,831) in the combined livebirth-stillbirth data and 0.36% (1378/383,951) in the livebirth data alone. The results of the multivariable logistic regressions for complete cases were also similar in these two datasets (Supplementary Fig. S2). Due to differences between the livebirth and stillbirth datasets the augmented/induced variable was omitted, and the pre-pregnancy BMI was used. Given the stillbirths had minimal impact on the observed rate of uterine rupture, stillbirths were not included in the main analysis, allowing a larger sample size (from 2011 to 2021).
Tables 2 and 3 show predicted probabilities for spontaneous and induced/augmented labours respectively in the multivariable regression. Predicted probabilities are for a parturient aged 30 years with a BMI of 30 kgm−2, 165 cm tall, birthweight 3500 g and 40 weeks’ gestational age. For parturients with a spontaneous labour, the predicted risk of rupture decreased from 0.36% (1 in 274) at 0 months IPI to 0.19% (1 in 524) at 21 or more months. For induced/augmented labours, these probabilities ranged from 0.91% (1 in 110) to 0.47% (1 in 212) over the same range of IPI values.
Except for unplanned hysterectomy, all comparisons of maternal and fetal/neonatal outcomes with categorised IPI (≤21 months v > 21 months) were not statistically significant (using χ2 tests for categorical variables and the Wilcoxon Rank Sum test for Apgar score). Therefore, rates of each outcome are reported in Table 5 by ruptured versus no ruptured uterus. Unplanned hysterectomy was mildly associated with longer IPI (p-value 0.029) so rates are shown for both categories of IPI. The risk of unplanned hysterectomy was 4.0% when there was a ruptured uterus compared with 0.05% when there was no uterine rupture. These figures were 5.9% versus 0.17% for maternal ICU admission, 3.7% versus 0.62% for intrapartum fetal death or neonatal death within 28 days, 2.5% versus 0.04% for neonatal seizures, 10.1% versus 0.36% for 5-min Apgar <4, and 7.1% versus 0.69% for neonatal ventilation more than 6 h.
Table 5.
Raw rates of adverse outcomes among 491,998 parturients who underwent a trial of labour in the United States following one previous caesarean section, with and without a ruptured uterus.
| Factor | Ruptured uterus % (95% CI) | ‘Risk’ (95% CI) | No rupture % (95% CI) | ‘Risk’ (95% CI) |
|---|---|---|---|---|
| Maternal: | ||||
| Unplanned hysterectomy (total)a | 4.0% (3.2%–5.1%) (66/1647) | 1 in 25 (20–32) | 0.05% (0.04%–0.06%) (239/489,995) | 1 in 2050 (1806–2327) |
| IPI ≤ 21 months | 2.6% (1.6%–4.1%) (17/654) | 1 in 38 (24–61) | 0.04% (0.03%–0.05%) (68/168,859) | 1 in 2483 (1959–3148) |
| IPI > 21 months | 4.9% (3.7%–6.5%) (44/895) | 1 in 20 (15–27) | 0.05% (0.05%–0.06%) (157/291,754) | 1 in 1858 (1590–2173) |
| Blood transfusiona | 12.9% (11.4%–14.6%) (213/1647) | 1 in 8 (7–9) | 0.51% (0.49%–0.53%) (2503/489,955) | 1 in 196 (188–204) |
| Maternal ICU admissiona | 5.9% (4.9%–7.1%) (97/1647) | 1 in 17 (14–21) | 0.17% (0.16%–0.19%) (857/489,955) | 1 in 572 (535–611) |
| Fetal/neonatal: | ||||
| Intrapartum fetal death or neonatal deathb | 3.7% (2.7%–5.1%) (36/973) | 1 in 27 (20–37) | 0.62% (0.59%–0.65%) (1784/286,522) | 1 in 161 (153–168) |
| 5-min Apgar score < 4c | 10.1% (8.7%–11.7%) (156/1546) | 1 in 10 (9–12) | 0.36% (0.34%–0.38%) (1590/440,846) | 1 in 277 (264–291) |
| Neonatal seizuresc | 2.5% (1.8%–3.4%) (39/1553) | 1 in 40 (29–54) | 0.04% (0.03%–0.04%) (158/441,599) | 1 in 2795 (2392–3266) |
| NICU admissionc | 27.7% (25.5%–30%) (430/1553) | 1 in 4 (3–4) | 5.51% (5.44%–5.58%) (24,331/441,599) | 1 in 18 (18–18) |
| Neonatal ventilation (immediate)c | 28% (25.8%–30.3%) (435/1553) | 1 in 4 (3–4) | 3.63% (3.58%–3.69%) (16,047/441,599) | 1 in 28 (27–28) |
| Neonatal ventilation >6 hc | 7.1% (6.0%–8.5%) (111/1553) | 1 in 14 (12–17) | 0.69% (0.67%–0.72%) (3059/441,599) | 1 in 144 (139–150) |
CI = confidence interval; TOLAC = trial of labour after caesarean.
2011 to 2021 maternal outcomes (denominators omit 396 TOLACs where rupture status was missing). Unplanned hysterectomy is stratified by IPI categories (with a further 29,440 records omitted due to IPI missingness), as a chi-squared test showed a mild association (p = 0.029). This association should be treated with caution due to multiple tests of statistical significance.
2014 to 2019 period-linked dataset only (n = 293,448). Denominators omit 5705 TOLACs occurring prior to 28 weeks, and 248 where rupture status and/or still birth and/or neonatal death prior to 28 days missing.
Fetal/neonatal outcomes other than death (denominators omit 48,636 TOLACs where labour occurred prior to 37 weeks or where gestational age, rupture status, or neonatal outcome of interest were missing). For neonatal outcomes other than 5-min Apgar score<4, an additional 210 cases where the adverse neonatal outcome status was missing (all in the no rupture group) were omitted. For 5-min Apgar score <4, there were 970 cases where the 5-min Apgar score was missing: 7 in the rupture and 963 in the no rupture group.
These data may be combined with an estimate of risk of uterine rupture. For example, if the risk of uterine rupture is 1 in 400 (0.25%) for a given parturient, then the risk of uterine rupture with intrapartum death or fetal death is 0.25% × 3.7% or approximately 0.009% (1 in 10,810).
Discussion
We found an inverse relationship between IPI and uterine rupture rates, showing progressively lower rates of uterine rupture from zero to 21 months, but not beyond 21 months. Other factors that increased the risk of uterine rupture include older maternal age, shorter maternal height, advancing gestational age, heavier birthweight, and induced/augmented labours. Severe complications of a ruptured uterus were uncommon with unplanned hysterectomy following 4.0% of uterine ruptures, and intrapartum fetal death or neonatal death within 28 days in 3.7%.
Our findings suggest that the effect of IPI may persist for longer than previously understood. In three smaller studies, lower rates of uterine rupture were observed among women attempting TOLAC beyond arbitrary thresholds of 9, 12, and 15 months IPI compared with IPIs below these thresholds. This is reflected in international guidelines which recommend a minimum IPI of nine months.7,19,20 In contrast, our study benefits from finer statistical resolution, indicating that the maximum risk reduction for uterine rupture may be achieved at 21 months. Notably, clinically important risk reductions may still be achieved earlier (Tables 2 and 3). Our findings are not comparable to the large study by Dong et al. which included women with a planned caesarean delivery and did not report on rates of uterine rupture following a TOLAC.18
Shorter IPIs potentially contribute to other adverse outcomes such as preterm birth, stillbirth and fetal growth restriction. However, the evidence is mixed and may vary by outcome and maternal age.21, 22, 23 For women with a previous caesarean delivery, it is important to consider not only uterine rupture but also risks of other adverse outcomes when making informed decisions regarding obstetric care and management.
We observed an overall rate of uterine rupture of 0.34%, lower than in most,8,24, 25, 26, 27, 28 but not all previous studies.29, 30, 31 The rate was 0.20% for spontaneous labours compared with 0.12%–0.52% reported elsewhere, and 0.55% for induced or augmented labours compared with 0.30%–1.02% reported elsewhere.8,25,26,28,31, 32, 33
In one highly cited publication, reported rates of uterine rupture were 0.36% (95% CI 0.22%–0.50%) and 0.87% (95% CI 0.63%–1.10%) respectively for spontaneous and induced labours.8 Uterine rupture was defined as “disruption or tear of the uterine muscle and visceral peritoneum or a separation of the uterine muscle with extension to the bladder or broad ligament”.8 In our study, the definition additionally included (i) that the placenta, umbilical cord or fetus must be protruding beyond the uterus and (ii) that the full thickness disruption cannot be asymptomatic. This stricter definition could explain the lower rates of uterine rupture in our study. Variations in reported rates of uterine rupture are also likely explained by differences in obstetric practices and populations over time and variations in practices between different hospitals and countries.
Our study identified gestational age as an additional risk factor. Some smaller studies found that gestational ages >40 weeks,34,35 or ≥41 weeks,36 independently increase the risk of uterine rupture in women attempting TOLAC while others found no significant difference.37,38 We found an increasing risk of uterine rupture for every week between 35 weeks and 43 weeks with on odds ratio of 1.12 (95% CI 1.09 to 1.16) per week of gestational age. This may provide useful information for women beyond 41 weeks gestation who are weighing the options between continued expectant management and caesarean delivery. Our results are also consistent with smaller studies reporting increased rates of uterine rupture for maternal age ≥ 35 years,39 height of ≤160 cm,40 birthweight of ≥4.5 kg,40 and induced onset of labour.8 The rate of uterine rupture in twin pregnancies was low (2/4317), but there were insufficient data to make any conclusions. Our study also identified a mild protective effect from increasing BMI, possibly due to weaker contractions in mothers with higher BMI.41 Others have found no association,42 or a higher risk of uterine rupture with increasing BMI without adjusting for confounders such as maternal height.43,44
Our finding of low rates of perinatal death following uterine rupture is consistent with the study by Landon et al. who reported two intrapartum stillbirths or neonatal deaths among 114 uterine ruptures (1.8%) consistent with our rate of 3.7%.8 Our 2.5% rate of neonatal seizures is consistent with the 6.2% (7/114) rate of hypoxic ischaemic encephalopathy (HIE) reported by Landon et al., as not all infants with HIE have seizures. Furthermore, we found that 10.1% of infants had a 5-min Apgar score <4 compared with 14% with an Apgar score ≤5 in the Landon study, and a rate of admission to the neonatal intensive care unit of 27.7% compared with 40.4% (46/114; 95% CI 31–50%).
Future studies should aim to further evaluate maternal, fetal and intrapartum factors which influence the risk of uterine rupture among women attempting a TOLAC and explore changes over time. Further research is also warranted to assess the short and long-term risks and benefits of attempting a TOLAC compared with planned caesarean delivery and the impact of providing individualised counselling about these risks. Development of a clinical prediction model for the risk of uterine rupture would facilitate individualised counselling and decision-making. It is important to engage community members, particularly with regards to the development and implementation of guidelines around counselling and informed decision-making.
The primary strength of this study lies in its large sample size which enabled us to estimate the risk of uterine rupture for IPI as a continuous variable with relatively narrow confidence intervals, and to estimate an IPI threshold beyond which the risk of uterine rupture stabilises. Other strengths include multiple imputation which can reduce the risk of bias and enhances statistical power. Limitations include the retrospective study design using administrative data, limiting the available variables. Misclassification and under-reporting of complications are potential concerns, although uterine rupture is unlikely to be missed by reporting clinicians and the birth certificate worksheet provided a clear definition for guidance. We have no reason to suspect any misclassification of IPI. As reported rates of uterine rupture are relatively consistent, our main results may be generalisable to most settings, but secondary outcomes such as perinatal death and unplanned hysterectomy may vary between high and low-resource settings.
In conclusion, we found that the risk of uterine rupture progressively decreases as IPI increases until about 21 months and then stabilises. Labours that are induced or augmented carry a higher risk of rupture. Importantly, the absolute risk of certain serious maternal and fetal/neonatal complications such as unplanned hysterectomy and perinatal death is low. Based on our results, women may be advised that aiming for a longer IPI can effectively reduce their risk of uterine rupture, and that this risk becomes progressively lower until it reaches a minimum at around 21 months, while on the other hand, induction or augmentation increases this risk. Furthermore, clinicians can provide estimates of risk associated with different IPIs to facilitate informed decision-making.
Contributors
Pejman Adily∗: Conducted all analyses, revised manuscript and Supplementary Materials, drafted data cleaning and preprocessing methods and parts of the statistical analysis, chaired author meetings.
Travis Bettison: Drafted introduction and discussion, revised the manuscript, attended author meetings, contributing to study design.
Mark Lauer: Provided advice about the statistical analysis, supervised the analysis, developed the analysis plan, drafted data analysis section of Supplementary Materials and parts of the statistics section, detailed revision of the manuscript, attended author meetings.
Rajit Narayan: Supervised drafting the introduction and discussion, revised the manuscript, attended author meetings, contributing to study design.
Adam Mackie: Revised the manuscript, attended author meetings, contributing to study design.
Hala Phipps: Revised the manuscript, attended author meetings, contributing to study design.
Vincenzo Berghella: Revised the manuscript, attended author meetings, contributing to study design.
Marjan Mosalman Haghighi: Revised the manuscript, contributed to study reporting and study design.
Katelyn Perren: Revised the manuscript, attended author meetings, contributing to study design.
George Johnson: Revised the manuscript, attended author meetings, contributing to study design.
Bradley de Vries∗: Original study concept and design, preliminary analyses, drafted the study protocol, attended author meetings, overall supervision, attended author meetings.
All authors have read and approved the final version of the manuscript.
∗These authors verified the underlying data.
Data sharing statement
Microdata files for public use and user’s guides are publicly available from https://www.cdc.gov/nchs/data_access/vitalstatsonline.htm.
The study protocol is available at https://www.slhd.nsw.gov.au/pdfs/inter-pregnancy-study-protocol.pdf. More details of the statistical analysis are included in Supplementary Materials Part 2—Data Analysis. The statistical analysis plan is included in the study protocol. The authors are happy to answer questions about the analysis and provide R Code on request by email.
Declaration of interests
Bradley de Vries declared that he is a member of a data safety and monitoring board of an unrelated study, and was financially supported by Sydney Local Health District to present data from this study at the Society for Maternal Fetal Medicine ASM, National Harbor, Marylands, 14th February 2024. There were no other declarations.
Hala Phipps declared that she is a member of a data safety monitoring Committee for an unrelated study (a cluster RCT on Midwifery workforce burnout).
Acknowledgements
We would like to thank the anonymous reviewers for their valuable feedback.
Footnotes
Supplementary data related to this article can be found at https://doi.org/10.1016/j.eclinm.2025.103071.
Appendix A. Supplementary data
References
- 1.Australian Institute of Health and Welfare Australia’s mothers and babies 2021 Web Report. https://www.aihw.gov.au/reports/mothers-babies/australias-mothers-babies/contents/about Last updated 13/12/2023. Downloaded from:
- 2.Osterman M.J.K., Hamilton B.E., Martin J.A., Driscoll A.K., Valenzuela C.P. Births: final data for 2021. Natl Vital Stat Rep. 2023;72(1):1–53. [PubMed] [Google Scholar]
- 3.The Organisation for Economic Co-operation and Development (OECD) Caesarean sections. https://data.oecd.org/healthcare/caesarean-sections.htm Data accessed from:
- 4.Morton R., Burton A.E., Kumar P., et al. Cesarean delivery: trend in indications over three decades within a major city hospital network. Acta Obstet Gynecol Scand. 2020;99(7):909–916. doi: 10.1111/aogs.13816. [DOI] [PubMed] [Google Scholar]
- 5.Jiandani F., Somalwar S., Bhalerao A. Frequency of caesarean section classified by robson's ten group classification system: a scoping review. Cureus. 2023;15(6) doi: 10.7759/cureus.41091. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Royal College of Obstetricians and Gynaecologists . 2015. Green-top Guideline No. 45. Birth after previous caesarean birth. [Google Scholar]
- 7.ACOG Practice Bulletin No. 205 Vaginal birth after cesarean delivery. Obstet Gynecol. 2019;133(2):e110–e127. doi: 10.1097/AOG.0000000000003078. [DOI] [PubMed] [Google Scholar]
- 8.Landon M.B., Hauth J.C., Leveno K.J., et al. Maternal and perinatal outcomes associated with a trial of labor after prior cesarean delivery. N Engl J Med. 2004;351(25):2581–2589. doi: 10.1056/NEJMoa040405. [DOI] [PubMed] [Google Scholar]
- 9.Esposito M.A., Menihan C.A., Malee M.P. Association of interpregnancy interval with uterine scar failure in labor: a case-control study. Am J Obstet Gynecol. 2000;183(5):1180–1183. doi: 10.1067/mob.2000.109046. [DOI] [PubMed] [Google Scholar]
- 10.Shipp T.D., Zelop C.M., Repke J.T., Cohen A., Lieberman E. Interdelivery interval and risk of symptomatic uterine rupture. Obstet Gynecol. 2001;97(2):175–177. doi: 10.1016/s0029-7844(00)01129-7. [DOI] [PubMed] [Google Scholar]
- 11.Bujold E., Mehta S.H., Bujold C., Gauthier R.J. Interdelivery interval and uterine rupture. Am J Obstet Gynecol. 2002;187(5):1199–1202. doi: 10.1067/mob.2002.127138. [DOI] [PubMed] [Google Scholar]
- 12.Landon M.B., Spong C.Y., Thom E., et al. Risk of uterine rupture with a trial of labor in women with multiple and single prior cesarean delivery. Obstet Gynecol. 2006;108(1):12–20. doi: 10.1097/01.AOG.0000224694.32531.f3. [DOI] [PubMed] [Google Scholar]
- 13.Stamilio D.M., DeFranco E., Pare E., et al. Short interpregnancy interval: risk of uterine rupture and complications of vaginal birth after cesarean delivery. Obstet Gynecol. 2007;110(5):1075–1082. doi: 10.1097/01.AOG.0000286759.49895.46. [DOI] [PubMed] [Google Scholar]
- 14.Bujold E., Gauthier R.J. Risk of uterine rupture associated with an interdelivery interval between 18 and 24 months. Obstet Gynecol. 2010;115(5):1003–1006. doi: 10.1097/AOG.0b013e3181d992fb. [DOI] [PubMed] [Google Scholar]
- 15.Cunningham S., Algeo C.E., DeFranco E.A. Influence of interpregnancy interval on uterine rupture. J Matern Fetal Neonatal Med. 2021;34(17):2848–2853. doi: 10.1080/14767058.2019.1671343. [DOI] [PubMed] [Google Scholar]
- 16.Huang W.H., Nakashima D.K., Rumney P.J., Keegan K.A., Jr., Chan K. Interdelivery interval and the success of vaginal birth after cesarean delivery. Obstet Gynecol. 2002;99(1):41–44. doi: 10.1016/s0029-7844(01)01652-0. [DOI] [PubMed] [Google Scholar]
- 17.Kessous R., Sheiner E. Is there an association between short interval from previous cesarean section and adverse obstetric and perinatal outcome? J Matern Fetal Neonatal Med. 2013;26(10):1003–1006. doi: 10.3109/14767058.2013.765854. [DOI] [PubMed] [Google Scholar]
- 18.Dong H., Chi J., Wang W., Liu L. Association between interpregnancy interval and maternal and neonatal adverse outcomes in women with a cesarean delivery: a population-based study. BMC Pregnancy Childbirth. 2023;23(1):284. doi: 10.1186/s12884-023-05600-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Dy J., DeMeester S., Lipworth H., Barrett J. No. 382-Trial of labour after caesarean. J Obstet Gynaecol Can. 2019;41(7):992–1011. doi: 10.1016/j.jogc.2018.11.008. [DOI] [PubMed] [Google Scholar]
- 20.Royal Australian and New Zealand College of Obstetricians and Gynaecologists . 2019. Birth after previous caesarean section. Best Practice Statement.https://ranzcog.edu.au/wp-content/uploads/2022/05/Birth-after-previous-caesarean-section.pdf Downloaded from: [Google Scholar]
- 21.Hanley G.E., Hutcheon J.A., Kinniburgh B.A., Lee L. Interpregnancy interval and adverse pregnancy outcomes: an analysis of successive pregnancies. Obstet Gynecol. 2017;129(3):408–415. doi: 10.1097/AOG.0000000000001891. [DOI] [PubMed] [Google Scholar]
- 22.Haight S.C., Hogue C.J., Raskind-Hood C.L., Ahrens K.A. Short interpregnancy intervals and adverse pregnancy outcomes by maternal age in the United States. Ann Epidemiol. 2019;31:38–44. doi: 10.1016/j.annepidem.2018.12.002. [DOI] [PubMed] [Google Scholar]
- 23.Schummers L., Hutcheon J.A., Hernandez-Diaz S., et al. Association of short interpregnancy interval with pregnancy outcomes according to maternal age. JAMA Intern Med. 2018;178(12):1661–1670. doi: 10.1001/jamainternmed.2018.4696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Caughey A.B., Shipp T.D., Repke J.T., Zelop C.M., Cohen A., Lieberman E. Rate of uterine rupture during a trial of labor in women with one or two prior cesarean deliveries. Am J Obstet Gynecol. 1999;181(4):872–876. doi: 10.1016/s0002-9378(99)70317-0. [DOI] [PubMed] [Google Scholar]
- 25.Lydon-Rochelle M., Holt V.L., Easterling T.R., Martin D.P. Risk of uterine rupture during labor among women with a prior cesarean delivery. N Engl J Med. 2001;345(1):3–8. doi: 10.1056/NEJM200107053450101. [DOI] [PubMed] [Google Scholar]
- 26.Smith G.C., Pell J.P., Pasupathy D., Dobbie R. Factors predisposing to perinatal death related to uterine rupture during attempted vaginal birth after caesarean section: retrospective cohort study. BMJ. 2004;329(7462):375. doi: 10.1136/bmj.38160.634352.55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Cahill A.G., Stamilio D.M., Odibo A.O., et al. Is vaginal birth after cesarean (VBAC) or elective repeat cesarean safer in women with a prior vaginal delivery? Am J Obstet Gynecol. 2006;195(4):1143–1147. doi: 10.1016/j.ajog.2006.06.045. [DOI] [PubMed] [Google Scholar]
- 28.Dekker G.A., Chan A., Luke C.G., et al. Risk of uterine rupture in Australian women attempting vaginal birth after one prior caesarean section: a retrospective population-based cohort study. BJOG. 2010;117(11):1358–1365. doi: 10.1111/j.1471-0528.2010.02688.x. [DOI] [PubMed] [Google Scholar]
- 29.Flamm B.L., Newman L.A., Thomas S.J., Fallon D., Yoshida M.M. Vaginal birth after cesarean delivery: results of a 5-year multicenter collaborative study. Obstet Gynecol. 1990;76(5 Pt 1):750–754. doi: 10.1097/00006250-199011000-00004. [DOI] [PubMed] [Google Scholar]
- 30.McMahon M.J., Luther E.R., Bowes W.A., Jr., Olshan A.F. Comparison of a trial of labor with an elective second cesarean section. N Engl J Med. 1996;335(10):689–695. doi: 10.1056/NEJM199609053351001. [DOI] [PubMed] [Google Scholar]
- 31.Fitzpatrick K.E., Kurinczuk J.J., Alfirevic Z., Spark P., Brocklehurst P., Knight M. Uterine rupture by intended mode of delivery in the UK: a national case-control study. PLoS Med. 2012;9(3) doi: 10.1371/journal.pmed.1001184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Zelop C.M., Shipp T.D., Repke J.T., Cohen A., Caughey A.B., Lieberman E. Uterine rupture during induced or augmented labor in gravid women with one prior cesarean delivery. Am J Obstet Gynecol. 1999;181(4):882–886. doi: 10.1016/s0002-9378(99)70319-4. [DOI] [PubMed] [Google Scholar]
- 33.Al-Zirqi I., Daltveit A.K., Forsen L., Stray-Pedersen B., Vangen S. Risk factors for complete uterine rupture. Am J Obstet Gynecol. 2017;216(2):165.e1–165.e8. doi: 10.1016/j.ajog.2016.10.017. [DOI] [PubMed] [Google Scholar]
- 34.Barger M.K., Weiss J., Nannini A., Werler M., Heeren T., Stubblefield P.G. Risk factors for uterine rupture among women who attempt a vaginal birth after a previous cesarean: a case-control study. J Reprod Med. 2011;56(7–8):313–320. [PubMed] [Google Scholar]
- 35.Kiran T.S., Chui Y.K., Bethel J., Bhal P.S. Is gestational age an independent variable affecting uterine scar rupture rates? Eur J Obstet Gynecol Reprod Biol. 2006;126(1):68–71. doi: 10.1016/j.ejogrb.2005.07.021. [DOI] [PubMed] [Google Scholar]
- 36.Hammoud A., Hendler I., Gauthier R.J., Berman S., Sansregret A., Bujold E. The effect of gestational age on trial of labor after Cesarean section. J Matern Fetal Neonatal Med. 2004;15(3):202–206. doi: 10.1080/14767050410001668329. [DOI] [PubMed] [Google Scholar]
- 37.Coassolo K.M., Stamilio D.M., Pare E., et al. Safety and efficacy of vaginal birth after cesarean attempts at or beyond 40 weeks of gestation. Obstet Gynecol. 2005;106(4):700–706. doi: 10.1097/01.AOG.0000179389.82986.50. [DOI] [PubMed] [Google Scholar]
- 38.Zelop C.M., Shipp T.D., Cohen A., Repke J.T., Lieberman E. Trial of labor after 40 weeks' gestation in women with prior cesarean. Obstet Gynecol. 2001;97(3):391–393. doi: 10.1016/s0029-7844(00)01175-3. [DOI] [PubMed] [Google Scholar]
- 39.Shipp T.D., Zelop C., Repke J.T., Cohen A., Caughey A.B., Lieberman E. The association of maternal age and symptomatic uterine rupture during a trial of labor after prior cesarean delivery. Obstet Gynecol. 2002;99(4):585–588. doi: 10.1016/s0029-7844(01)01792-6. [DOI] [PubMed] [Google Scholar]
- 40.Hesselman S., Hogberg U., Ekholm-Selling K., Rassjo E.B., Jonsson M. The risk of uterine rupture is not increased with single- compared with double-layer closure: a Swedish cohort study. BJOG. 2015;122(11):1535–1541. doi: 10.1111/1471-0528.13015. [DOI] [PubMed] [Google Scholar]
- 41.Zhang J., Bricker L., Wray S., Quenby S. Poor uterine contractility in obese women. BJOG. 2007;114(3):343–348. doi: 10.1111/j.1471-0528.2006.01233.x. [DOI] [PubMed] [Google Scholar]
- 42.Kaczmarczyk M., Sparen P., Terry P., Cnattingius S. Risk factors for uterine rupture and neonatal consequences of uterine rupture: a population-based study of successive pregnancies in Sweden. BJOG. 2007;114(10):1208–1214. doi: 10.1111/j.1471-0528.2007.01484.x. [DOI] [PubMed] [Google Scholar]
- 43.Zwart J.J., Richters J.M., Ory F., de Vries J.I., Bloemenkamp K.W., van Roosmalen J. Uterine rupture in The Netherlands: a nationwide population-based cohort study. BJOG. 2009;116(8):1069–1078. doi: 10.1111/j.1471-0528.2009.02136.x. discussion 78-80. [DOI] [PubMed] [Google Scholar]
- 44.Zhan W., Zhu J., Hua X., Ye J., Chen Q., Zhang J. Epidemiology of uterine rupture among pregnant women in China and development of a risk prediction model: analysis of data from a multicentre, cross-sectional study. BMJ Open. 2021;11(11) doi: 10.1136/bmjopen-2021-054540. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.