Materials and Methods
This study is a retrospective cohort study from the Maternal-Fetal Medicine Units Network cesarean registry, which was a previously reported prospective multicenter cohort study. The initial study was conducted from 1999 through 2002 at 19 centers. Each participating center and the data-coordinating center received institutional review board approval prior to the study. This secondary analysis was reviewed and determined exempt by The University of North Carolina institutional review board.
Trained study nurses at each institution reviewed medical records of women enrolled to obtain maternal demographic information, medical and obstetrical history, as well as intrapartum and postpartum details. Infant records were reviewed and data were collected up to 120 days after delivery or until hospital discharge. All pertinent information was entered onto data forms that were sent weekly to the data-coordinating center at George Washington University Biostatistics Center. The data were edited on a regular basis.
Women were included in the present study if they had a singleton; delivered at ≥24 gestational weeks; underwent TOLAC, but failed; and underwent repeat cesarean in the current pregnancy. Women not planning TOLAC but who labored prior to their scheduled cesarean were included. Women were excluded if they underwent an elective or indicated repeat cesarean delivery and did not labor. Due to the nature of the database, women with a successful VBAC were excluded. Study groups were defined by the indication for repeat cesarean as determined by the provider. Women with primary indication for repeat cesarean documented as NRFS were defined as the exposure group and were compared to women who underwent cesarean for other indications.
The primary outcomes were uterine rupture and uterine dehiscence. Secondary outcomes included 5-minute Apgar score <7 and neonatal intensive care unit (NICU) admission. 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 bladder or broad ligament. Uterine dehiscence was defined as a disruption of uterine muscle with intact serosa. Cases of uterine rupture were reviewed centrally to confirm the diagnosis. Uterine rupture and dehiscence cannot be confirmed after a vaginal delivery, which was partial motivation for excluding these women from the current study.
Baseline characteristics between women with a failed trial of labor who underwent repeat cesarean for NRFS were compared to those who underwent repeat cesarean for other indications using the Student t test or Mann-Whitney U test for continuous variables and the χ 2 or Fisher exact test for categorical variables as appropriate. Incidences of the primary and secondary outcomes were compared between study groups and the unadjusted relative risks with 95% confidence intervals (CIs) were estimated. Potentially confounding factors and effect modifiers were identified via bivariable and stratified analyses. Logistic regression models were developed to estimate the independent risk of NRFS for each outcome adjusting for confounding factors that were identified historically and in the bivariable and stratified analyses. Backward step-wise selection was performed to reduce the number of variables in the regression model by assessing the magnitude of change in the effect size of remaining covariates. Differences in the explanatory model were tested using the likelihood ratio test or Wald test. Only statistically significant variables or those historically known to be important covariates were included in the final regression models and adjusted odds ratios (aORs) with 95% CI were estimated ( P < .05 was considered significant). The final models were tested with the Hosmer-Lemeshow goodness-of-fit test. Planned subanalyses were performed and included only term deliveries, only preterm deliveries, and NRFS as the only indication (as opposed to the primary indication) for repeat cesarean. Statistical analyses were performed using software (STATA 13.0, Special Edition; StataCorp LP, College Station, TX).
Results
Of 45,805 singletons with a history of a cesarean who delivered at ≥24 weeks, 28,065 (61.3%) were excluded for indicated or elective repeat cesarean without trial of labor. Of the 17,740 women who underwent a trial of labor, 12,986 (73.2%) had a successful vaginal birth and were excluded. Vaginal births were excluded secondary to the nature of the database (the presence or lack of NRFS was not documented in women with a successful VBAC, nor could uterine rupture or dehiscence be confirmed). The remaining 4754 (26.8%) women had a failed TOLAC, and underwent repeat cesarean. In all, 1516 (31.9%) women underwent repeat cesarean for the primary indication of NRFS, comprising the exposed study group. In addition, 3238 (68.1%) women with a different primary indication comprised the unexposed study group. Among the 1516 women with NRFS as the primary indication, 997 (65.8%) had no additional indications for the cesarean ( Figure ).
The 2 groups were similar in terms of maternal age (mean 28.6 ± 5.8 years for the exposure group and 28.6 ± 5.6 years for the nonexposure group) and type of prior uterine incision. The study groups differed in several characteristics. Women in the NRFS group were more often of advanced maternal age and black race, had a prior vaginal delivery, and had gestational hypertension. They were more likely to smoke cigarettes, use alcohol, have pregestational diabetes, and have spontaneous labor compared to the unexposed study group. On average, exposure group patients had lower body mass index, higher gravidity, higher number of prior cesareans, shorter duration since the last cesarean, lower gestational age at delivery, and lower birthweight ( Table 1 ). The primary indications for cesarean are listed in Table 2 . The most prevalent cesarean indications other than NRFS were cephalopelvic disproportion and failure to progress.
| Characteristic | Cesarean delivery for NRFS n = 1516 | Cesarean delivery for other indication n = 3238 | P value | Missing data |
|---|---|---|---|---|
| Advanced maternal age (≥35 y) | 298 (19.66%) | 550 (16.99%) | .03 | 0 (0%) |
| BMI, kg/m 2 | 33.19 ± 7.15 | 33.78 ± 7.23 | .01 | 323 (6.8%) |
| BMI ≥30 kg/m 2 | 871 (62.30%) | 2025 (66.77%) | < .01 | 323 (6.8%) |
| Black race | 679 (44.79%) | 1135 (35.05%) | < .01 | 0 (0%) |
| Gravidity | 3.34 ± 1.69 | 3.08 ± 1.54 | < .01 | 0 (0%) |
| No. of prior cesareans | 1.10 ± 0.01 | 1.07 ± 0.01 | < .01 | 7 (<1.0%) |
| Years since last cesarean | 5.01 ± 3.70 | 4.76 ± 3.25 | < .01 | 194 (4.1%) |
| History of classic | 6 (0.40%) | 20 (0.62%) | .33 | 0 (0%) |
| History of low vertical | 9 (0.59%) | 14 (0.43%) | .46 | 0 (0%) |
| History of T or J incision | 4 (0.26%) | 5 (0.15%) | .48 | 0 (0%) |
| History of unknown scar | 201 (13.26%) | 424 (13.09%) | .88 | 0 (0%) |
| History of vaginal delivery | 498 (33.05%) | 761 (23.57%) | < .01 | 19 (<1.0%) |
| Gestational age at delivery, wk | 38.41 ± 3.19 | 39.15 ± 2.41 | < .01 | 8 (<1.0%) |
| Birthweight, g | 3089.57 ± 802.46 | 3433.21 ± 671.05 | < .01 | 3 (<1.0%) |
| Labor | .02 | 490 (10%) | ||
| Induced | 412 (34.30%) | 1121 (36.60%) | ||
| Augmented | 410 (34.14%) | 1108 (36.17%) | ||
| Spontaneous | 379 (31.56%) | 834 (27.23%) | ||
| Alcohol use | 82 (5.42%) | 102 (3.15%) | < .01 | 4 (<1.0%) |
| Tobacco use | 268 (17.70%) | 425 (13.13%) | < .01 | 4 (<1.0%) |
| DM | < .01 | 0 (0%) | ||
| None | 1376 (90.83%) | 2985 (92.22%) | ||
| Pregestational DM | 46 (3.04%) | 50 (1.54%) | ||
| Gestational DM | 93 (6.14%) | 202 (6.24%) | ||
| Gestational hypertension or preeclampsia | 188 (12.40%) | 317 (9.79%) | < .01 | 0 (0%) |
| Chronic hypertension requiring medications | 59 (3.89%) | 92 (2.84%) | .05 | 0 (0%) |
| Variable | n (%) |
|---|---|
| NRFS | 1516 (31.9) |
| Cephalopelvic disproportion | 449 (9.4) |
| Failure to progress | 1810 (38.1) |
| Cord prolapse | 55 (1.2) |
| Abnormal presentation | 14 (0.3) |
| Suspected macrosomia | 26 (0.5) |
| Abruption | 46 (1.0) |
| Previa with hemorrhage | 2 (0.04) |
| Preeclampsia/hypertension | 57 (1.2) |
| Herpes | 1 (0.02) |
| Prior classic/vertical/T/J | 13 (0.3) |
| Unknown incision | 11 (0.2) |
| Elective | 253 (5.3) |
| Nonreassuring antenatal testing | 20 (0.4) |
| Failed induction | 304 (6.4) |
| Other | 73 (1.5) |
| 4754 (100) |
NRFS as the primary indication for repeat cesarean in women who failed a TOLAC was associated with a 3.3-fold increased risk of uterine rupture, 1.7-fold increased risk of uterine dehiscence, 1.6-fold increased risk of low 5-minute Apgar score, and a 1.8-fold increased risk of NICU admission after delivery ( Table 3 ). The increased risk of uterine rupture and uterine dehiscence remained significant after adjusting for obesity, history of a vaginal delivery, type of labor, and number of years since the last cesarean (rupture aOR, 3.32; 95% CI, 2.21–5.00 and dehiscence aOR, 1.70; 95% CI, 1.09–2.65). The increased risk of 5-minute Apgar score <7 (aOR, 1.55; 95% CI, 1.14–2.12) and NICU admission (aOR, 1.79; 95% CI, 1.52–2.12) remained significant after adjusting for black race, obesity, and gestational age at delivery.
| Variable | NRFS (primary indication) n = 1516 | Other indication n = 3238 | RR (95% CI) n = 4754 | aOR (95% CI) n = 3802 | P value |
|---|---|---|---|---|---|
| Uterine rupture | 64 (4.22%) | 45 (1.39%) | 3.04 (2.08–4.43) | 3.32 a (2.21–5.00) | < .01 |
| Uterine dehiscence | 36 (2.37%) | 57 (1.76%) | 1.35 (0.89–2.04) | 1.70 a (1.09–2.65) | .02 |
| n = 1513 | n = 3235 | n = 4748 | n = 4419 | ||
| 5-min Apgar score <7 | 101 (6.67%) | 101 (3.12%) | 2.14 (1.63–2.80) | 1.55 b (1.14–2.12) | < .01 |
| n = 1508 | n = 3226 | n = 4734 | n = 4404 | ||
| NICU admission | 496 (32.89%) | 585 (18.14%) | 1.84 (1.64–2.01) | 1.79 b (1.52–2.12) | < .01 |
a Adjusted for obesity, history of vaginal delivery, labor type, and years since last cesarean
b Adjusted for black race, obesity, gestational age at delivery.
In planned subanalyses, NRFS as the only indication for repeat cesarean after failed TOLAC was associated with a 1.8-fold increased risk of uterine rupture that remained significant after adjusting for obesity, history of a vaginal delivery, type of labor, and number of years since the last cesarean (aOR, 1.84; 95% CI, 1.18–2.88). There was also a 1.4-fold increased risk of NICU admission that remained significant after adjusting for black race, obesity, and gestational age at delivery (aOR, 1.36; 95% CI, 1.13–1.65) ( Table 4 ). All regression models were well fit, except the NICU models. We were unable to explain lack of fit after investigating subsets of variables, but the poor fit may be the result of important missing unknown covariates. Importantly, adding uterine rupture to the model did not account for the association between NICU admission and NRFS as the primary indication, suggesting that unaccounted factors other than uterine rupture were contributing to NICU admission and that uterine rupture was not a dominant causative factor.
| Variable | NRFS (only indication) n = 997 | Other indication n = 3757 | RR (95% CI) n = 4754 | aOR (95% CI) n = 3802 | P value |
|---|---|---|---|---|---|
| Uterine rupture | 36 (3.61%) | 73 (1.94%) | 1.85 (1.25–2.75) | 1.84 a (1.18–2.88) | < .01 |
| Uterine dehiscence | 21 (2.11%) | 72 (1.92%) | 1.10 (0.68–1.79) | 1.20 a (0.71–2.02) | .49 |
| n = 995 | n = 3753 | n = 4748 | n = 4419 | ||
| 5-min Apgar score <7 | 56 (5.62%) | 146 (3.89%) | 1.45 (1.07–1.95) | 1.21 b (0.85–1.73) | .28 |
| n = 991 | n = 3743 | n = 4734 | n = 4404 | ||
| NICU admission | 282 (28.46%) | 799 (21.35%) | 1.33 (1.19–1.50) | 1.36 b (1.13–1.65) | < .01 |
a Adjusted for obesity, history of vaginal delivery, labor type, years since last cesarean
b Adjusted for black race, obesity, gestational age at delivery.
In preterm deliveries (24-37 gestational weeks), given small outcome numbers aOR were not calculated for uterine rupture or uterine dehiscence. However, we observed more uterine rupture and dehiscence among patients with NRFS as the primary indication, which is consistent with the overall analysis. There was an increased risk of NICU admission (aOR, 2.02; 95% CI, 1.32–3.11) that remained significant after adjusting for black race, obesity, and gestational age at delivery ( Table 5 ). In women with preterm deliveries with NRFS as the only indication, the differences in primary and secondary outcomes did not reach statistical significance ( Table 6 ).
| Variable | NRFS (primary indication) n = 303 | Other indication n = 342 | RR (95% CI) n = 645 | aOR (95% CI) | P value |
|---|---|---|---|---|---|
| Uterine rupture | 4 (1.32%) | 3 (0.88%) | 1.50 (0.34–6.67) | N/A | N/A |
| Uterine dehiscence | 3 (0.99%) | 2 (0.58%) | 1.69 (0.28–10.06) | N/A | N/A |
| n = 302 | n = 342 | n = 644 | n = 600 | ||
| 5-min Apgar score <7 | 40 (13.20%) | 46 (13.45%) | 0.98 (0.66–1.46) | 0.70 a (0.41–1.18) | .18 |
| n = 302 | n = 336 | n = 638 | n = 593 | ||
| NICU admission | 240 (79.47%) | 205 (61.01%) | 1.30 (1.18–1.44) | 2.02 a (1.32–3.11) | < .01 |
a Adjusted for black race, obesity, gestational age at delivery.
| Variable | NRFS (only indication) n = 160 | Other indication n = 485 | RR (95% CI) n = 645 | aOR (95% CI) | P value |
|---|---|---|---|---|---|
| Uterine rupture | 2 (1.25%) | 5 (1.03%) | 1.21 (0.24–6.19) | N/A | N/A |
| Uterine dehiscence | 0 | 5 (1.03%) | N/A | N/A | N/A |
| n = 160 | n = 484 | n = 644 | n = 600 | ||
| 5-min Apgar score <7 | 18 (11.25%) | 68 (14.02%) | 0.80 (0.49–1.31) | 0.55 a (0.29–1.06) | .08 |
| n = 159 | n = 479 | n = 638 | n = 593 | ||
| NICU admission | 119 (74.84%) | 326 (68.06%) | 1.10 (0.99–1.22) | 1.31 a (0.80–2.15) | .28 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
