Abstract
Objective
To compare risk of uterine rupture in women with prior periviable cesarean versus prior term cesarean, independent of initial incision type.
Methods
We conducted a retrospective longitudinal cohort study using Washington State birth certificate data and hospital discharge records, identifying primary cesareans performed at 20-26 weeks and 37-41 weeks of gestation with subsequent delivery between 1989-2008. We compared subsequent uterine rupture risk in the two groups considering both primary incision type and subsequent labor indication and augmentation.
Results
We identified 456 women with index periviable cesarean and 10,505 women with index term cesarean. Women with index periviable cesarean were younger, more frequently of non-white race, more likely to smoke, and more likely to have hypertension. Women in the periviable group had more index classical incisions (42% versus 1%, p<0.001) and fewer subsequent inductions and augmentations (8% vs. 16%, p<0.001). Uterine rupture in the subsequent pregnancy occurred more frequently among women in the index periviable group than those in the index term group (8/456 [1.8%] versus 38/10,505 [0.4%], OR 4.9, 95% CI 2.3-10.6). This relationship persisted among women with a low transverse incision (4/228 [1.8%] versus 36/9,558 [0.4%], OR 4.7, 95% CI1.7 – 13.4).
Conclusion
Cesarean at periviability compared to term is associated with an increased risk for uterine rupture in a subsequent pregnancy, even after low transverse incision. These data support judicious use of cesarean at periviable gestational ages and inform subsequent counseling.
Introduction
Over the last several decades, improved neonatal survival at extremely premature gestational ages has repositioned the demarcation of “periviability.” With this shift, the acceptable gestational age for neonatal resuscitation is moving earlier in pregnancy(1). As a result, the rate of cesarean delivery at periviability has increased dramatically over recent years(2-5). This is in part due to a higher incidence of traditional indications for cesarean, including malpresentation and other fetal indications, at periviability than at term(6).
Preterm cesarean deliveries are associated with higher risks both immediately and in the subsequent delivery than term cesarean deliveries(7, 8). Much of the increased risk in the subsequent delivery is attributed to the type of uterine incision(9). Classical uterine incisions occur in approximately 30% of cesarean deliveries prior to 28 weeks of gestation(7). The risk of uterine rupture in women varies by prior incision type, with the risk of uterine rupture after classical cesarean approximating 1-12%(10).
Prior studies comparing uterine rupture risk after term cesarean delivery versus preterm cesarean delivery have focused on late preterm gestational ages. Some studies suggest increased risk of subsequent rupture after preterm cesarean delivery, estimating between a 1.6 and 5 fold increase(11, 12), while other studies show no association(13-15). The specific risk for uterine rupture after cesarean delivery at periviability remains unknown.
We sought to compare the risk of uterine rupture and its comorbidities after a prior periviable cesarean delivery compared to prior term cesarean delivery.
Materials and Methods
We conducted a longitudinal retrospective cohort study of primary singleton cesarean deliveries linked to subsequent singleton births by the same mother. We identified women according to Washington state birth certificate and fetal death certificate files for the years 1989-2008, provided by the Washington State Department of Health through the University of Washington. Data were linked to maternal and neonatal hospital discharge diagnosis International Classification of Diseases, 9th Revision, Clinical Modification (1CD-9-CM) for both the index and subsequent births and provided as a de-identified dataset. The Human Subjects Division at the University of Washington determined this study exempt from review due to use of de-identified data.
For inclusion, women were required to have two singleton deliveries in the state of Washington between 1989 and 2008. We refer to the exposure pregnancy as “index” and outcome pregnancy as “subsequent.” Subsequent pregnancies were the next documented pregnancy in the Washington State birth certificate files through 2008. Subjects with cesarean section prior to the index pregnancy were excluded. We broadly defined periviability to include gestational ages between 20 0/7 and 26 6/7 weeks of gestation, choosing a lower gestational age limit consistent with the Periviable Birth: Executive Summary(1). All qualifying women identified with a periviable cesarean delivery during the study period were included in the exposure (periviable) group. The comparison (term) group included randomly selected women with an index term cesarean between 37 0/7 and 41 6/7 weeks of gestation. As an additional factor to validate gestational age in the dataset, index periviable deliveries were limited to birth weights of 250 grams to 1500 grams, and index term delivery birth weights were limited to 2500-6500 grams. No limitations were placed on subsequent delivery gestational ages or birth weights.
Maternal and neonatal demographics and characteristics were evaluated for the index delivery as well as the subsequent delivery. The primary outcome was uterine rupture in the subsequent delivery (defined as birth certificate variable for uterine rupture or ICD-9-CM codes 665.0, 665.1). The cohort sample sizes of 456 periviable deliveries and 10,505 term deliveries were determined by including all qualifying deliveries during the time period. Secondary outcomes for the subsequent delivery included a composite measure of maternal morbidity and maternal length of stay. Composite morbidity included hemorrhage, infection, hysterectomy, obstetric injury and death (defined as birth certificate variable for transfusion, bleeding, coagulopathy, chorioamnionitis, sepsis, maternal infection, hysterectomy or maternal death or ICD-9-CM codes 641.3, 641.8, 641.9, 666.0, 666.1, 666.2, 666.3, 285.1, 286.6, 75.8, 99.0, 659.2, 659.3, 670, 672, 995.9, 68.3, 68.4, 68.8, 68.9, 665.3, 665.4, 665.5, 665.6, 665.8, 665.9, 998.2, 57.8, 69.29, 69.49, 75.5, 75.61, 761.6, or 798). Missing values for dichotomous outcome variables were assumed to represent negative values.
Differences in maternal demographics between groups according to gestational age at index cesarean delivery were assessed via t-tests and chi-squared tests. Binary outcomes were compared across groups via logistic regression models; continuous outcomes were analyzed via linear regression models. A logarithmic transformation was applied to length of hospital stay to accommodate modeling assumptions. Because our primary outcome is rare, we limited our covariates to two preset variables: index incision type and labor induction or augmentation. These two covariates were included one at a time in the primary outcome model to assess for confounding. A factor was defined as a confounder if there was a difference of 10% or more in the estimated coefficient of interest between the multivariable model including the factor and the model without it.
Index incision is determined by diagnosis code used for the index delivery and induction or augmentation is defined by ICD-9-CM codes for induction (73.1, 73.4, 73.99, 96.49) and birth certificate variables for induction and augmentation of labor in the subsequent pregnancy. For secondary outcomes with sufficient events we adjusted for confounding considering the a priori variables index incision type and labor induction or augmentation, as well as other potential confounders during the subsequent delivery including mother's race, mother's age, obesity, smoking status, diabetes, hypertension, birth weight, gestational age and delivery mode. Odds ratios (OR) are presented with 95% confidence intervals (CI). For all analyses, a two-sided significance level of <0.05 was considered statistically significant. Analyses were performed using SAS version 9 (SAS Institute, Inc., Cary, N.C.).
Results
We identified 456 index periviable primary cesarean deliveries and 10,505 index term primary cesarean deliveries. Index periviable deliveries occurred at 20 (N=5), 21 (N=10), 22 (N=9), 23 (N=27), 24 (N=94), 25 (N=144), 26 (N=167) weeks of gestation. Of women with an index periviable delivery, 163 (42%) underwent an index classical incision., Of women with an index term delivery, 61(1%) underwent an index classical incision.
Table 1 compares characteristics for the index and subsequent pregnancies among women in the periviable group and the term group. Women in the periviable group were younger at both pregnancies, more likely to be non-white, smokers, hypertensive, and more likely to have a lower income at the subsequent pregnancy than women in the term group. The distribution of number of prior live births (an approximation of parity) also varied significantly across groups.
Table 1. Demographic and pregnancy characteristics by gestational age group of index cesarean delivery.
| Periviable (20 – 26 weeks) N=456 |
Term (37 – 41 weeks) N=10,505 |
p-value1 | |
|---|---|---|---|
| Index pregnancy | |||
| Maternal age (years) | <0.001 | ||
| Mean (SD) | 23.7 (5.8) | 25.7 (6.1) | |
| Missing: N (%) | 0 (0) | 12 (<1) | |
| Maternal race: N (%)2 | <0.001 | ||
| White | 310 (68) | 8022 (76) | |
| Non-white | 125 (27) | 1986 (19) | |
| Missing | 21 (5) | 497 (5) | |
| Gestational age (weeks) | |||
| Median (IQR) | 25 (24 – 26) | 40 (39 – 40) | – |
| Missing N (%) | 0 (0) | 0 (0) | |
| Birth weight (grams) | |||
| Mean (SD) | 756 (202) | 3589 (499) | – |
| Missing N (%) | 0 (0) | 0 (0) | |
| Incision type: N (%) | <0.001 | ||
| Low transverse | 228 (50) | 9558 (91) | |
| Classical | 163 (36) | 61 (<1) | |
| Other | 1 (<1) | 11 (<1) | |
| Missing | 64 (14) | 875 (8) | |
| Subsequent pregnancy | |||
| Maternal age (years) | <0.001 | ||
| Mean (SD) | 26.7 (6.0) | 28.9 (6.0) | |
| Missing: N (%) | 0 (0) | 2 (<1) | |
| Maternal obesity: N (%) | 0.08 | ||
| No | 253 (55) | 6152 (58) | |
| Yes | 63 (14) | 1968 (19) | |
| Missing | 140 (31) | 2385 (23) | |
| Prior live births: N (%) | <0.001 | ||
| 0 | 71 (16) | 198 (2) | |
| 1 | 225 (49) | 8552 (81) | |
| 2 | 89 (20) | 1031 (10) | |
| 3 or more | 51 (11) | 508 (5) | |
| Missing | 20 (4) | 216 (2) | |
| Inter-delivery interval (months) | 0.07 | ||
| Median (IQR) | 28 (18 – 45) | 32 (23 – 47) | |
| Missing N (%) | 0 (0) | 0 (0) | |
| Maternal smoking: N (%) | <0.001 | ||
| No | 346 (76) | 8903 (85) | |
| Yes | 91 (20) | 1316 (12) | |
| Missing | 19 (4) | 286 (3) | |
| Maternal income (× 103) | <0.001 | ||
| Median (IQR) | 37.9 (30.7 – 47.6) | 42.5 (33.5 – 54.6) | |
| Missing: N (%) | 18 (4) | 215 (2) | |
| Hypertension: N (%) | 75 (16) | 670 (6) | <0.001 |
| Diabetes: N (%) | 0.64 | ||
| No | 423 (93) | 9712 (92) | |
| Gestational | 25 (5) | 651 (6) | |
| Established | 8 (2) | 142 (1) | |
| Gestational age (weeks) | <0.001 | ||
| Median (IQR) | 37 (35 – 38) | 39 (38 – 40) | |
| Missing: N (%) | 7 (2) | 150 (1) | |
| Birth weight (grams) | <0.001 | ||
| Mean (SD) | 2763 (841) | 3509 (542) | |
| Missing: N (%) | 5 (1) | 53 (<1) | |
| Induction/augment: N (%) | <0.001 | ||
| No | 420 (92) | 8824 (84) | |
| Yes | 36 (8) | 1681 (16) | |
| Delivery mode: N (%) | 0.01 | ||
| Vaginal | 92 (20) | 2681 (26) | |
| Cesarean | 364 (80) | 7822 (74) | |
| Missing | 0 (0) | 2 (<1) |
Analyses exclude missing values
SD = Standard deviation and IQR = Interquartile range.
Race was categorized as white or non-white using data collected at both first and subsequent pregnancies. We assumed that a person's race did not vary over time, so that if race was missing for a subject's first pregnancy we could use a non-missing race value from the subsequent pregnancy, and vice versa.
Data on maternal income at the subsequent pregnancy was missing and filled in with data from the first pregnancy for 273 patients, 259 (2.5%) in the term and 14 (3.6%) in the periviability delivery group (p=0.17).
Mean gestational age for the subsequent delivery was lower for the periviable group than the term group (36.0 weeks versus 38.8 weeks, respectively, p<0.001). Women in the periviable group were less likely to have had an induced or augmented subsequent delivery. The majority of women in this cohort delivered the subsequent pregnancy by repeat cesarean, over 70% in both exposure groups. There were fewer successful vaginal deliveries in the periviable group compared to the term group (20% versus 26%, respectively, p=0.01).
Our primary outcome, uterine rupture in the subsequent pregnancy, occurred more frequently among women in the periviable group than those in the term group (1.8% versus 0.4%, OR 4.9, 95%CI 2.3-10.6, p<0.001, Table 2). The relationship was not confounded by index incision type nor by subsequent induction or augmentation. Among the subset of women with a low transverse incision in the index pregnancy, uterine rupture remained more common in the periviable group compared to the term group (OR 4.7, 95% CI 1.7-13.4, p=0.004).
Table 2. Risk of uterine rupture by gestational age group of index cesarean delivery.
| Periviable | Term | Unadjusted | ||
|---|---|---|---|---|
|
| ||||
| (20-26 weeks) | (37-41 weeks) | OR (95% CI) | p-value | |
| All patients | 8/456 (1.8%) | 38/10,505 (0.4%) | 4.9 (2.3 – 10.6) | <0.001 |
| By incision type | ||||
| Classical | 4/163 (2.5%) | 0/61 (0%) | –1 | – |
| Low transverse | 4/228 (1.8%) | 36/9,558 (0.4%) | 4.7 (1.7 – 13.4) | 0.004 |
| Other2 | 0/1 (0%) | 0/11 (0%) | –1 | – |
| Missing | 0/64 (0%) | 2/875 (<0.1%) | –1 | – |
| By induction or augmentation | ||||
| No | 8/420 (1.9%) | 23/8,824 (0.3%) | 7.4 (3.3 – 16.7) | <0.001 |
| Yes | 0/36 (0%) | 15/1,681 (0.9%) | –1 | – |
OR could not be estimated for classical, other, and missing incision types or for womenwith induction/augmentation, due to lack of events.
other incision types as determined by ICD-9-CM procedure code 74.4 or 74.9
The incidence of our secondary maternal morbidity composite outcome was similar across groups (14.0% in the periviable group and 10.0% in the term group, p=0.68 in adjusted model). Analyses are presented in Table 3. The composite component maternal infection in the subsequent delivery occurred more often in the periviable group than the term group, although this difference was not statistically significant in an adjusted model (6.6% versus 3.8%, OR 1.1, 95% CI 0.6-1.9, p=0.74, Table 3). The composite components hemorrhage, hysterectomy, obstetric injury, and death were similar between the two groups (Table 3). The unadjusted mean maternal hospital stay was longer in the index periviable group than in the index term group (3.6 days, 95% CI 3.2 – 4.1 vs. 2.5 days, 95% CI 2.4 – 2.6, p<0.001).
Table 3. Results for secondary outcome: maternal composite morbidity by gestational age group of the index pregnancy.
| Periviable | Term | Adjusted2 | ||
|---|---|---|---|---|
|
| ||||
| N=456 | N=10,505 | OR (95% CI) | p-value | |
| Composite morbidity1 | 64 (14.0%) | 1047 (10.0%) | 1.1 (0.7 – 1.6) | 0.68 |
| Hemorrhage | 24 (5.3%) | 442 (4.2%) | 1.0 (0.5 – 1.7) | 0.89 |
| Infection | 30 (6.6%) | 403 (3.8%) | 1.1 (0.6 – 1.9) | 0.74 |
| Hysterectomy | 2 (0.4%) | 17 (0.2%) | 2.7 (0.6 – 11.8) | 0.18 |
| Obstetric injury | 15 (3.3%) | 271 (2.6%) | 1.3 (0.8 – 2.2) | 0.37 |
| Maternal death | 0 | 0 | – | – |
Maternal composite includes: death, hemorrhage, infection, obstetric injury, and hysterectomy.
Composite morbidity, hemorrhage, and infection models adjusted for maternal race, incision type, gestational age, induction or augmentation and mode of delivery in the subsequent delivery and excludes missing values (N=9,395). Analyses of hysterectomy, obstetric injury, and maternal death were unadjusted due to insufficient events.
Notably, uterine rupture cases in our cohort occurred with a substantial amount of morbidity. Of uterine ruptures occurring after index periviable cesarean, 7/8 (88%) had an associated morbidity or clinical sign such as hemorrhage, infection, obstetric injury (including bladder injury), abnormal fetal heart rate or fetal death. Among ruptures occurring after index term cesarean, 28/38 (74%) occurred with at least one of these associated clinical signs. Detailed clinical information on all uterine rupture cases is shown in Table 4 (included as supplementary digital content).
Table 4.
Uterine rupture case descriptions.
| Case | INDEX PREGNANCY | Inter- delivery Interval (month) |
SUBSEQUENT PREGNANCY | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Weeks of Gestation |
Indication | Incision | Birth weight (gm1) |
Complications | Weeks Gestation |
Intrapartum Risk Factors |
Birth weight (gm1) |
Obese | Complications | ||
| INDEX PERIVIABLE CASES | |||||||||||
| 1 | 21 | Preterm PROM1, cord prolapse, malpresentation | Low transverse | 312 | Perinatal stillbirth | 13 | 382 | 1814 | .3 | Maternal hemorrhage, obstetric injury, respiratory complication, retained placenta Neonatal respiratory disease |
|
| 2 | 22 | Preterm labor, puerperal infection, cord prolapse | Classical | 454 | Neonatal death, low Apgar4 | 12 | 34 | 2098 | Yes | Maternal placental abruption, obstetric injury, abnormal fetal heart rate Neonatal ventilation, umbilical vessel catheterization, respiratory disease, jaundice |
|
| 3 | 24 | Puerperal infection, previa with hemorrhage, cord prolapse, abnormal fetal heart rate | Low transverse | 623 | Maternal anemia, transfusion, retained foreign body Neonatal death, low Apgar3, ventilation, umbilical vessel catheterization, cutaneous hemorrhage |
20 | .3 | . | .3 | Maternal anemia, transfusion Perinatal stillbirth |
|
| 4 | 24 | Puerperal infection, placental abruption, abnormal fetal heart rate | Classical | 695 | Neonatal death, low Apgar3, ventilation | 12 | 36 | 3385 | Yes | Maternal anemia, hemorrhage, transfusion, puerperal infection, ileus, failed operative vaginal delivery Neonatal ventilation, hemolytic disease, jaundice, respiratory disease |
|
| 5 | 25 | Preterm PROM1, malpresentation | Classical | 822 | Neonatal death, respiratory disease, ventilation, umbilical vessel catheterization | 13 | 36 | 2665 | No | Maternal obstetric injury Neonatal temperature dysregulation, hypoglycemia, jaundice observation for infection |
|
| 6 | 25 | Preterm labor, puerperal infection, oligohydramnios, malpresentation | Low transverse | 790 | Neonatal ventilation, retinopathy, anemia, jaundice | 63 | 38 | 3075 | No | Maternal obstetric injury | |
| 7 | 25 | Puerperal infection, hypertensive disorder | Classical | 595 | Maternal pulmonary edema Neonatal death, ventilation, umbilical vessel catheterization, necrotizing enterocolitis, intraventricular hemorrhage, feeding problems, jaundice, electrolyte disturbance |
22 | 38 | 3275 | No | Maternal wound complication | |
| 8 | 26 | Preterm labor, puerperal infection malpresentation | Low transverse | 709 | Neonatal low Apgar3, ventilation | 56 | 223 | Intraamniotic infection | 1219 | No | Maternal placental abruption, puerperal infection, obstetric injury Neonatal ventilation, respiratory disease, pneumonia, parenteral nutrition, metabolic acidosis, fluid overload, electrolyte disturbance, jaundice |
| INDEX TERM CASES | |||||||||||
| 9 | 37 | Oligohydramnios malpresentation | Low transverse | 2807 | 33 | 38 | 2892 | No | |||
| 10 | 37 | PROM1, hypertensive disorder, malpresentation, abnormal fetal heart rate | Low transverse | 3334 | 31 | 38 | 3502 | No | Maternal wound complication, abnormal fetal heart rate | ||
| 11 | 37 | Abnormal fetal heart rate, failed operative vaginal delivery | Low transverse | 4110 | Neonatal ventilation | 55 | 38 | Induction | 3666 | No | |
| 12 | 37 | Previa with hemorrhage | Low transverse | 3543 | 22 | 41 | Induction | 3968 | No | Maternal obstetric injury, abnormal fetal heart rate Neonatal hypoglycemia, observation for infection |
|
| 13 | 38 | Labor dystocia, abnormal fetal heart rate | Low transverse | 2775 | Neonatal feeding problems, jaundice | 30 | 36 | 3655 | .3 | Placenta previa with hemorrhage Neonatal respiratory disease, observation for infection |
|
| 14 | 38 | Labor dystocia | Low transverse | 3713 | Neonatal respiratory disease, observation for infection | 15 | 37 | 3090 | No | Maternal hemorrhage, transfusion, obstetric injury, hysterectomy, bladder injury, respiratory complication, abnormal fetal heart rate Neonatal ventilation |
|
| 15 | 38 | Hypertensive disorder, suspected macrosomia, labor dystocia | Low transverse | 4054 | Maternal hemorrhage Neonatal temperature dysregulation, observation for infection, spinal tap |
21 | 38 | Induction, labor dystocia | 3883 | .3 | Maternal hypertensive disorder, complication of anesthesia, respiratory complication, vascular complication |
| 16 | 38 | PROM1, puerperal infection, abnormal fetal heart rate | Low transverse | 3742 | Neonatal respiratory disease, hypoglycemia | 25 | 39 | 3572 | No | Neonatal temperature dysregulation | |
| 17 | 38 | Labor dystocia | Low transverse | 3798 | Neonatal patent ductus arteriosus | 44 | 40 | 3402 | No | Maternal anemia, obstetric injury, abnormal fetal heart rate | |
| 18 | 39 | Low transverse | 3061 | 37 | 37 | Labor dystocia | 3231 | .3 | Maternal anemia, hemorrhage, obstetric injury, bladder injury, abnormal fetal heart rate | ||
| 19 | 39 | Hypertensive disorder, other abnormal labor, abnormal fetal heart rate | Low transverse | 3144 | 33 | 40 | Induction, labor dystocia | 3423 | .3 | Maternal anemia, hemorrhage, obstetric injury, hypertensive disorder, abnormal fetal heart rate | |
| 20 | 39 | Malpresentation | Low transverse | 3619 | 34 | 40 | 3785 | No | Maternal obstetric injury | ||
| 21 | 39 | Malpresentation | Low transverse | 3883 | Maternal anemia | 37 | 40 | Labor dystocia | 4167 | No | Maternal obstetric injury Neonatal metabolic acidosis |
| 22 | 40 | Labor dystocia | .3 | 3713 | 52 | 38 | 3118 | Yes | Abnormal fetal heart rate Neonatal observation for infection |
||
| 23 | 40 | Hypertensive disorder, labor dystocia | Low transverse | 3224 | 24 | 39 | 3314 | .3 | |||
| 24 | 40 | Hypertensive disorder, labor dystocia | Low transverse | 3459 | Maternal hepatorenal syndrome Neonatal umbilical vessel catheterization, respiratory disease |
19 | 39 | Labor dystocia, augmentation | 4706 | .3 | Neonatal hemorrhagic disease |
| 25 | 40 | Cord prolapse, abnormal fetal heart rate | Low transverse | 3856 | 26 | 39 | Augmentation | 2750 | No | ||
| 26 | 40 | Hypertensive disorder, labor dystocia, abnormal fetal heart rate, failed operative vaginal delivery | Low transverse | 3175 | Neonatal hemolytic disease, jaundice | 30 | 40 | Labor dystocia, augmentation | 3770 | No | Maternal anemia, hemorrhage |
| 27 | 40 | Hypertensive disorder, malpresentation | Low transverse | 2835 | 24 | 40 | Labor dystocia, augmentation | 3175 | No | Abnormal fetal heart rate | |
| 28 | 40 | Malpresentation | Low transverse | 3997 | 30 | 40 | Augmentation | 3798 | .3 | Abnormal fetal heart rate Neonatal low Apgar3, ventilation, respiratory disease, non-mechanical resuscitation, seizures, observation for infection |
|
| 29 | 40 | Puerperal infection, other abnormal labor | Low transverse | 3521 | 30 | 40 | 3827 | .3 | Maternal bladder injury, abnormal fetal heart rate | ||
| 30 | 40 | Puerperal infection, hypertensive disorder, labor dystocia, abnormal fetal heart rate | Low transverse | 3835 | 29 | 40 | Labor dystocia | 4110 | Yes | Abnormal fetal heart rate Neonatal jaundice |
|
| 31 | 40 | PROM1, labor dystocia | Low transverse | 3402 | 28 | 41 | Induction, prostaglandin use, labor dystocia, intraamniotic infection | 3969 | .3 | Maternal bladder injury, puerperal infection, abnormal fetal heart rate Neonatal respiratory disease, acute tubular necrosis, observation for infection |
|
| 32 | 40 | Previa with hemorrhage | Low transverse | 3373 | 19 | 41 | Labor dystocia, augmentation | 4167 | No | Maternal obstetric injury, puerperal infection, wound complication, abnormal fetal heart rate Neonatal jaundice |
|
| 33 | 40 | Labor dystocia | Low transverse | 3969 | Maternal obstetric injury, puerperal infection | 23 | 41 | intraamniotic infection | 3515 | No | Maternal puerperal infection, abnormal fetal heart rate Neonatal tachycardia |
| 34 | 40 | Malpresentation, abnormal fetal heart rate | Low transverse | 3374 | 18 | .3 | Labor dystocia | 3374 | .3 | Neonatal respiratory disease, non-mechanical resuscitation, scalp injury | |
| 35 | 41 | Puerperal infection, abnormal fetal heart rate | Low transverse | 3941 | Maternal 3rd degree laceration Neonatal ventilation, umbilical vessel catheterization |
17 | 38 | 3714 | .3 | Maternal obstetric injury, bladder injury | |
| 36 | 41 | Puerperal infection, labor dystocia | Low transverse | 3799 | 22 | 38 | Labor dystocia, augmentation, intraamniotic infection | 3289 | .3 | Maternal anemia, hemorrhage, puerperal infection, Abnormal fetal heart rate Neonatal observation for infection, spinal tap |
|
| 37 | 41 | Labor dystocia | Low transverse | 3968 | 15 | 38 | 2636 | .3 | Maternal hemorrhage, hysterectomy, abnormal fetal heart rate | ||
| 38 | 41 | Labor dystocia, abnormal fetal heart rate | Low transverse | 3770 | 64 | 39 | Augmentation | 2485 | No | Maternal anemia, hemorrhage, transfusion, postpartum exploratory laparotomy, obstetric injury, retained placenta, hypertensive disorder, precipitate labor Neonatal Temperature dysregulation, hypoglycemia, jaundice |
|
| 39 | 41 | Malpresentation | Low transverse | 3374 | 57 | 39 | Induction, prostaglandin use | 3395 | No | Abnormal fetal heart rate | |
| 40 | 41 | Labor dystocia, abnormal fetal heart rate | Low transverse | 3611 | Maternal anemia | 29 | 39 | 4140 | No | Maternal anemia | |
| 41 | 41 | Labor dystocia, abnormal fetal heart rate | Low transverse | 3856 | 21 | 39 | Induction, labor dystocia | 3771 | No | Maternal pelvic floor abnormality, abnormal fetal heart rate | |
| 42 | 41 | Suspected macrosomia | Low transverse | 4706 | 45 | 40 | Augmentation | 4195 | Yes | Maternal obstetric injury, puerperal infection, abnormal fetal heart rate Neonatal low Apgar3, ventilation, umbilical vessel catheterization |
|
| 43 | 41 | Herpes simplex virus | Low transverse | 3799 | 23 | 41 | Labor dystocia | 3742 | No | Maternal obstetric injury | |
| 44 | 41 | Labor dystocia | Low transverse | 3685 | 21 | 41 | Augmentation | 4309 | No | Maternal anemia, hemorrhage | |
| 45 | 41 | Labor dystocia, abnormal fetal heart rate | Low transverse | 4479 | 15 | 41 | Labor dystocia | 3713 | Yes | Maternal anemia, hemorrhage, obstetric injury, failed operative vaginal delivery, abnormal fetal heart rate Neonatal cutaneous hemorrhage, jaundice |
|
| 46 | 41 | .3 | .3 | 3572 | 37 | 42 | Induction | 3455 | No | Abnormal fetal heart rate | |
gm = grams, PROM = premature rupture of membranes.
Gestational age based on best clinical estimate. For this patient, we noted discordance with birth weight and concordance with alternative gestational age estimation by last menstrual period of 32 weeks.
. = missing data
Low Apgar defined as less than <5 at 5 or 10 minutes of life.
Discussion
Our data show an increased risk of uterine rupture after periviable cesarean delivery compared with term cesarean delivery. This risk was consistent across prior periviable classical and low transverse uterine incision types, reflected by a nearly fivefold increased risk after index periviable low-transverse cesarean delivery compared with index term low-transverse cesarean delivery. Overall, uterine rupture occurred in approximately 2% of women with a prior periviable cesarean delivery.
Our findings add to an extensive literature describing uterine rupture risk after cesarean delivery(9, 11-15). Some studies have demonstrated a higher risk of uterine rupture after preterm cesarean delivery, and our data support this risk after periviable cesarean. Sciscione et al reported higher uterine rupture risk after prior birth <37 weeks of gestation with an adjusted OR of 1.6 among index “nonclassic” cesarean deliveries(11). In this study, first delivery birth weight was considered a surrogate for gestational age and was not associated with subsequent uterine rupture risk. Similarly, Rochelson et al demonstrated an association of uterine rupture with preterm low transverse cesarean <36 weeks of gestation compared with term cesarean (OR 5.39)(12). However, the majority (80%) of women with a prior preterm cesarean delivery were >31 weeks of gestation, leaving uncertainty about risk after cesarean at earlier gestational ages.
Several additional studies found no heightened uterine rupture risk after preterm cesarean(13-15), though some studies were limited by relatively small cohort sizes(13, 14). Harper et al compared uterine rupture risk after cesarean delivery before or after 34 weeks of gestation and found no difference in risk among women undergoing a subsequent trial of labor. However, a subset analysis of women with a prior cesarean delivery at <28 weeks of gestation (n=55) compared to prior term cesarean showed a non-significant increased risk, with frequencies of uterine rupture of 1.8% and 0.9%, respectively (RR 2.1, 95% CI 0.3-15.0), similar in magnitude to our current findings.
While our study population is large and carefully defined, the limitations of our study design should be considered in interpretation of its findings. With a rare but important outcome, a population-based study is the most feasible way to identify important risk factors. This study uses birth certificate data which carry inherent limitations including concerns for accuracy and inconsistencies in data collection(16, 17). We carefully selected variables for clarity, prior validation, and conservative estimation of risk; for example we did not assess spontaneous labor, because a reliable variable was not available. Not all variables used in the current study were specifically assessed in validation studies, nor were we able to directly validate our findings with a chart review in the current study. However, while misclassification bias remains a concern in both directions, the additional information gleaned from hospital discharge data minimizes this issue. Specifically, our database of WA state birth certificate variables is enhanced by linked ICD-9-CM codes, which has been shown to improve accuracy in validation studies(18, 19). Lydon-Rochelle et. al. demonstrated that the true positive rate of several variables improves without increasing the false-positive rate when combining birth certificate data with hospital discharge data compared to medical chart review(18, 19). In addition, the association of uterine rupture with maternal and neonatal morbidity demonstrates clinical relevance and appropriate classification (Table 4).
We conclude that the risk of subsequent uterine rupture after periviable cesarean delivery, including low transverse uterine incisions, may be greater than previously estimated. Prospective studies are needed to confirm these findings and direct clinical management. However, in the absence of prospective studies, these data highlight the need for caution in the management of pregnancy after prior periviable cesarean delivery, regardless of incision type.
Acknowledgments
The authors thank the Washington Department of Health for data access and Mr. William O’Brien for programing assistance.
Supported by NICHD HD-067221.
Footnotes
Presented at the 2013 Society for Gynecologic Investigation (SGI) 60th Annual Scientific Meeting, March 20-23, 2013 in Orlando, FL.
Financial Disclosure: The authors did not report any potential conflicts of interest.
References
- 1.Raju TN, Mercer BM, Burchfield DJ, Joseph GF., Jr Periviable birth: executive summary of a joint workshop by the Eunice Kennedy Shriver National Institute of Child Health and Human Development, Society for Maternal-Fetal Medicine, American Academy of Pediatrics, and American College of Obstetricians and Gynecologists. Obstet Gynecol. 2014 May;123(5):1083–96. doi: 10.1097/AOG.0000000000000243. [DOI] [PubMed] [Google Scholar]
- 2.Fanaroff AA, Hack M, Walsh MC. The NICHD neonatal research network: changes in practice and outcomes during the first 15 years. Semin Perinatol. 2003 Aug;27(4):281–7. doi: 10.1016/s0146-0005(03)00055-7. [DOI] [PubMed] [Google Scholar]
- 3.MacDorman MF, Declercq E, Zhang J. Obstetrical intervention and the singleton preterm birth rate in the United States from 1991-2006. Am J Public Health. 2010 Nov;100(11):2241–7. doi: 10.2105/AJPH.2009.180570. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Malloy MH. Impact of cesarean section on neonatal mortality rates among very preterm infants in the United States, 2000-2003. Pediatrics. 2008 Aug;122(2):285–92. doi: 10.1542/peds.2007-2620. [DOI] [PubMed] [Google Scholar]
- 5.Wylie BJ, Davidson LL, Batra M, Reed SD. Method of delivery and neonatal outcome in very low-birthweight vertex-presenting fetuses. Am J Obstet Gynecol. 2008 Jun;198(6):640 e1–7. doi: 10.1016/j.ajog.2007.12.038. discussion e1-4. [DOI] [PubMed] [Google Scholar]
- 6.Hogberg U, Hakansson S, Serenius F, Holmgren PA. Extremely preterm cesarean delivery: a clinical study. Acta Obstet Gynecol Scand. 2006;85(12):1442–7. doi: 10.1080/00016340600969366. [DOI] [PubMed] [Google Scholar]
- 7.Evans LC, Combs CA. Increased maternal morbidity after cesarean delivery before 28 weeks of gestation. Int J Gynaecol Obstet. 1993 Mar;40(3):227–33. doi: 10.1016/0020-7292(93)90835-k. [DOI] [PubMed] [Google Scholar]
- 8.Lumley J. Method of delivery for the preterm infant. Bjog. 2003 Apr;110(Suppl 20):88–92. [PubMed] [Google Scholar]
- 9.ACOG Practice bulletin no. 115: Vaginal birth after previous cesarean delivery. Obstet Gynecol. 2010 Aug;116(2 Pt 1):450–63. doi: 10.1097/AOG.0b013e3181eeb251. [DOI] [PubMed] [Google Scholar]
- 10.Landon MB. Predicting uterine rupture in women undergoing trial of labor after prior cesarean delivery. Semin Perinatol. 2010 Aug;34(4):267–71. doi: 10.1053/j.semperi.2010.03.005. [DOI] [PubMed] [Google Scholar]
- 11.Sciscione AC, Landon MB, Leveno KJ, Spong CY, Macpherson C, Varner MW, et al. Previous preterm cesarean delivery and risk of subsequent uterine rupture. Obstet Gynecol. 2008 Mar;111(3):648–53. doi: 10.1097/AOG.0b013e318163cd3e. [DOI] [PubMed] [Google Scholar]
- 12.Rochelson B, Pagano M, Conetta L, Goldman B, Vohra N, Frey M, et al. Previous preterm cesarean delivery: identification of a new risk factor for uterine rupture in VBAC candidates. J Matern Fetal Neona. 2005 Nov;18(5):339–42. doi: 10.1080/14767050500275911. [DOI] [PubMed] [Google Scholar]
- 13.Kwee A, Smink M, Van Der Laar R, Bruinse HW. Outcome of subsequent delivery after a previous early preterm cesarean section. J Matern Fetal Neona. 2007 Jan;20(1):33–7. doi: 10.1080/14767050601036527. [DOI] [PubMed] [Google Scholar]
- 14.Estrade S, Schmitz T, Cabrol D, Huchon C, Goffinet F. History of cesarean before 32 weeks' gestation and trial of labor: what is the risk of uterine rupture? Acta Obstet Gynecol Scand. 2009;88(2):149–53. doi: 10.1080/00016340802491490. [DOI] [PubMed] [Google Scholar]
- 15.Harper LM, Cahill AG, Stamilio DM, Odibo AO, Peipert JF, Macones GA. Effect of gestational age at the prior cesarean delivery on maternal morbidity in subsequent VBAC attempt. Am J Obstet Gynecol. 2009 Mar;200(3):276 e1–6. doi: 10.1016/j.ajog.2008.10.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Ananth CV. Perinatal epidemiologic research with vital statistics data: validity is the essential quality. Am J Obstet Gynecol. 2005 Jul;193(1):5–6. doi: 10.1016/j.ajog.2005.05.064. [DOI] [PubMed] [Google Scholar]
- 17.Schoendorf KC, Branum AM. The use of United States vital statistics in perinatal and obstetric research. Am J Obstet Gynecol. 2006 Apr;194(4):911–5. doi: 10.1016/j.ajog.2005.11.020. [DOI] [PubMed] [Google Scholar]
- 18.Lydon-Rochelle MT, Holt VL, Nelson JC, Cardenas V, Gardella C, Easterling TR, et al. Accuracy of reporting maternal in-hospital diagnoses and intrapartum procedures in Washington State linked birth records. Paediatr Perinat Epidemiol. 2005 Nov;19(6):460–71. doi: 10.1111/j.1365-3016.2005.00682.x. [DOI] [PubMed] [Google Scholar]
- 19.Lydon-Rochelle MT, Holt VL, Cardenas V, Nelson JC, Easterling TR, Gardella C, et al. The reporting of pre-existing maternal medical conditions and complications of pregnancy on birth certificates and in hospital discharge data. Am J Obstet Gynecol. 2005 Jul;193(1):125–34. doi: 10.1016/j.ajog.2005.02.096. [DOI] [PubMed] [Google Scholar]