Abstract
Objective:
Trial of labor after cesarean delivery (TOLAC) has been mostly studied in the setting of one prior cesarean delivery, whereas controversy remains regarding the risks and benefits of TOLAC for women with two prior cesarean deliveries. This study aimed to examine utilization, success rate, and maternal and neonatal outcomes of TOLAC in this population.
Methods:
Using linked hospital discharge and birth certificate data, we retrospectively analyzed a cohort of mothers with non-anomalous, term, singleton live births in California during 2010-2012 who had two prior cesarean deliveries and no clear contraindications for trial of labor. We measured whether they attempted labor and if so, whether they delivered vaginally. Association of patient and hospital characteristics with the likelihood of attempting labor and TOLAC success was examined using multivariable regressions. We compared composite severe maternal morbidities and composite severe newborn complications in those who underwent TOLAC versus planned cesarean delivery using a propensity score matching approach.
Results:
Among 42,771 women meeting sample eligibility criteria, 1,228 (2.9%) attempted labor of whom 484 (39.4%) delivered vaginally. There was no significant difference in the risk of composite severe maternal morbidities, but a modest increase in the risk of composite severe newborn complications among women who attempted labor compared to those who did not (2.0% versus 1.4%, p=0.04). After accounting for differences in patient and hospital characteristics, propensity score matched analysis showed no significant association between TOLAC and the risk of composite severe maternal morbidities ( odds ratio [OR]=1.16, 95% confidence interval [CI]: 0.70-1.91), but TOLAC was associated with a higher risk for the composite of severe newborn complications (OR=1.78, 95% CI: 1.04-3.04).
Conclusion:
Among women with two prior cesarean deliveries, trial of labor was rarely attempted and was successful in 39.4% of attempts. TOLAC in this population was associated with a modest increase in severe neonatal morbidity.
Précis:
Among women giving birth in California with two prior cesarean deliveries, trial of labor was successful in 39% of the attempts and was associated with a modest increase in severe newborn morbidity.
INTRODUCTION
Repeated cesarean deliveries are associated with increased risk of maternal morbidities, such as placenta accreta, organ injury, and hysterectomy.1,2 Trial of labor after cesarean delivery (TOLAC) plays an important role in reducing repeat cesarean delivery and associated maternal morbidities.
TOLAC has been mostly studied in the setting of one prior cesarean delivery, with few studies examining TOLAC use after two prior cesarean deliveries. Hence, although the American College of Obstetricians and Gynecologists (ACOG) recommends that “Most women with one previous cesarean delivery with a low-transverse incision are candidates for and should be counseled about and offered TOLAC”, it states that there is “limited or inconsistent scientific evidence” regarding TOLAC in women with two previous cesarean deliveries.3
Existing research comparing outcomes of planned cesarean versus TOLAC after two cesarean deliveries was primarily based on data from the 1980s-early 2000s and showed modestly increased maternal morbidities associated with TOLAC.4–7 However, data regarding neonatal outcomes are limited and there is little information about factors that might influence the utilization and success of TOLAC after two prior cesarean deliveries.
The purpose of this study was to provide a more up-to-date evaluation of maternal and neonatal outcomes associated with TOLAC among women with two prior cesarean deliveries using a retrospective cohort of women delivering in the state of California. We also sought to identify maternal, neonatal, and institutional factors that are associated with the utilization and success of TOLAC in this cohort.
METHODS
This study used a comprehensive dataset obtained from the California Office of Statewide Health Planning and Development (OSHPD) consisting of linked information from hospital discharge records (mother and infant) and birth certificates for births in California during 2010-2012.8 This study period reflects the most recent years in which the linked hospital discharge-birth certificate dataset is available, and the three years were combined to form our study cohort. We limited our sample to women with two prior cesarean deliveries who met the following criteria: 1) had live singleton births at 37-42 weeks gestation, 2) had no congenital anomalies or clear contraindications to trial of labor (malpresentation, genital herpes, placenta previa, vasa previa), 3) had no missing or erroneous information on parity (e.g., parity <2 that contradicts two prior cesarean deliveries), and 4) had successful linkage across maternal hospital discharge record, newborn hospital discharge record, and birth certificate. Patients whose disposition status indicates that they left the hospital against medical advice were excluded since their experience did not reflect usual care. To ensure a reasonable volume of relevant births at study hospitals, we further limited our sample to hospitals with at least 25 eligible births in the study period. This study was approved by the Yale University Human Investigation Committee and the California Committee for the Protection of Human Subjects (CPHS).
A woman was determined as undergoing a trial of labor if her delivery method on the birth certificate was noted as vaginal delivery or trial of labor attempted, or if the International Classification of Diseases Ninth Revision (ICD-9) procedure codes on maternal hospital discharge record indicated process of labor, or if at least two of the following information sources suggested process of labor: prolonged labor on birth certificate, induction of labor on birth certificate, augmentation of labor on birth certificate, third or fourth degree perineal laceration on birth certificate, or ICD-9 diagnosis codes on maternal hospital discharge record indicated process of labor (see Appendix 1, available online at http://links.lww.com/xxx). Women who did not meet these criteria for trial of labor were assumed to have undergone a planned cesarean delivery.
Maternal outcomes included length of stay and a composite indicator for severe maternal morbidities. This composite indicator followed the definition by the Centers for Disease Control and Prevention and encompassed 16 diagnoses and 5 procedures suggestive of severe maternal complications such as eclampsia, air and thrombotic embolism, hysterectomy, and blood transfusion.9,10 These conditions were identified based on maternal diagnosis and procedure codes, length of stay, and hospital disposition status.9,10 We could not separately examine the individual maternal morbidities in this composite or uterine rupture because these conditions were rare and our study is subject to small cell size suppression requirement by the California OSHPD and CPHS in order to protect patient confidentiality (i.e., suppressing report of numbers with ≤15 patients).
Neonatal outcomes included length of stay and five binary variables: a composite indicator for any severe unexpected newborn complications, a composite indicator for any moderate unexpected newborn complications, a composite indicator for total unexpected newborn complications (severe or moderate), a composite indicator for any unexpected newborn respiratory complications, and neonatal intensive care unit (NICU) admission. The composite indicators for unexpected newborn complications were measured following the definition developed by the California Maternal Quality Care Collaborative based on infant diagnosis and procedure codes, length of stay, and hospital disposition status.11 Examples of severe newborn complications included intracranial hemorrhage, asphyxia, and severe sepsis; examples of moderate newborn complications included fetal inflammatory response syndrome and transient tachypnea. An extensive list of the conditions included in the unexpected newborn complication measures is available elsewhere.12 NICU admission was measured using birth certificate data. Since the standard birth certificate reporting guidelines define NICU admission as “Admission into a facility or unit staffed and equipped to provide continuous mechanical ventilator support for a newborn,” which approximates level III or IV nursery,13,14 we limited our assessment of NICU admission to the subset of births delivered at hospitals with a level III or IV nursery.
Patient characteristics assessed in this study included maternal age, race, ethnicity, parity, number of prior vaginal deliveries, adequacy of prenatal care (based on the Kotelchuck adequacy of prenatal care utilization index15), body mass index, hypertensive disorders of pregnancy, diabetes or abnormal glucose tolerance, anemia, smoking status, substance use, birthweight, and small or large for gestational age. These characteristics were assessed based on birth certificate data elements and diagnosis codes on maternal and newborn hospital discharge records.
Hospital characteristics assessed in this study included teaching status, type of ownership, affiliation with a multi-hospital system, average annual delivery volume in our study period, level of neonatal care capacity, availability of blood bank service in-house, availability of 24-hour on-premises coverage of anesthesia, proportion of all live births at the hospitals attended by a midwife, safety-net burden (i.e., proportion of mothers of all live births with Medicaid or no insurance), and local malpractice insurance premium for obstetrics and gynecology. These measures were obtained from multiple sources, including the 2011 American Hospital Association Annual Survey, the 2011 California OSHPD hospital annual financial data, the California Perinatal Quality Care Collaborative, Medical Liability Monitor 2011 rate survey of malpractice insurance premiums, and our own analysis of the linked hospital discharge and birth certificate data (regarding birth volume, birth attendant, and patient insurance type).16–19
We characterized patient and hospital characteristics, as well as maternal and neonatal outcomes, using descriptive statistics. Differences in these characteristics and outcomes between births attempting TOLAC and births with a planned cesarean delivery were compared using chi-square test for categorical variables, Student’s t test for continuous variables that were normally distributed, and Wilcoxon rank-rum test for continuous variables that were not normally distributed. Multivariable logistic regression using a hierarchical generalized linear model (HGLM) (with a logit link function and binomial distribution) was used to identify patient and hospital factors that were associated with the likelihood of attempting labor and the likelihood of success among those who attempted, respectively. The model used a hospital random effect to account for clustering of births by hospitals.
Similar HGLM models were used to examine the association between TOLAC and maternal and neonatal outcomes while adjusting for patient and hospital characteristics. Binary outcomes were analyzed using a logit link function and binomial distribution, while length of stay was analyzed using a log link function and Poisson distribution. Their corresponding results were interpreted as odds ratios (OR) and mean ratios, respectively. We also used a propensity score matching approach to more rigorously account for differences in patient and hospital characteristics between the TOLAC and planned cesarean delivery group when comparing maternal and neonatal outcomes. First, we used the HGLM model for trial of labor delineated in the preceding paragraph as a propensity score model to predict a woman’s likelihood of attempting TOLAC based on the corresponding patient and hospital characteristics. Each woman who attempted TOLAC was matched to up to three women who did not attempt TOLAC using the Greedy nearest neighbor matching method with caliper width of 0.025.20 Then, we used HGLM models to estimate maternal and neonatal outcomes using the propensity score matched sample while accounting for clustering of data by hospital and matched patients. Separate propensity score matching was performed for the overall sample, as well as the sub-sample in analysis of NICU admissions (limited to births at hospitals with level III-IV nurseries) and the sub-sample in analysis of unexpected newborn complications (limited to births that met eligibility criteria for these measures such as birthweight ≥2500 grams and no maternal drug use).11,14
P values less than 0.05 were considered statistically significant. All analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC).
RESULTS
A total of 42,771 women met our sample eligibility criteria with a mean age of 31.1 years (Table 1). Of these, 1,228 (2.9%) underwent trial of labor with 484 of them (39.4%) successfully delivered vaginally. Comparison of patient characteristics showed that women who attempted TOLAC differed significantly from those who underwent planned cesarean delivery by gestational age, parity, race, ethnicity, prenatal care, and comorbidities (Table 1). They also differed significantly by characteristics of the hospital where they delivered.
Table 1.
Patient and hospital characteristics, overall and by trial of labor status
| Characteristics | Overall (N=42,771) |
By TOLAC Status |
||
|---|---|---|---|---|
| Planned Cesarean Delivery (N=41,543) |
TOLAC (N=1,228) |
P Value | ||
| Maternal Characteristics | ||||
| Maternal age (years), mean±SD | 31.1 ± 5.3 | 31.1 ± 5.3 | 31.3 ± 5.4 | 0.12 |
| Gestational age (weeks), mean±SD | 38.7 ± 0.8 | 38.7 ± 0.8 | 38.8 ± 1.1 | <0.001 |
| Parity | <0.001 | |||
| 2 | 34774 (81.3%) | 33943 (81.7%) | 831 (67.7%) | |
| 3 | 5276 (12.3%) | 5046 (12.1%) | 230 (18.7%) | |
| ≥4 | 2721 (6.4%) | 2554 (6.1%) | 167 (13.6%) | |
| Number of previous vaginal deliveries, median (IQR) | 0 (0-0) | 0 (0-0) | 0 (0-1) | <0.001 |
| Race and ethnicity | <0.001 | |||
| Non-Hispanic white | 8499 (19.9%) | 8191 (19.7%) | 308 (25.1%) | |
| Non-Hispanic black | 2469 (5.8%) | 2372 (5.7%) | 97 (7.9%) | |
| Hispanic | 27773 (64.9%) | 27073 (65.2%) | 700 (57.0%) | |
| Other | 4030 (9.4%) | 3907 (9.4%) | 123 (10.0%) | |
| Kotelchuck Adequacy of Prenatal Care Utilization Index | <0.001 | |||
| Inadequate | 4536 (10.6%) | 4379 (10.5%) | 157 (12.8%) | |
| Intermediate | 5873 (13.7%) | 5687 (13.7%) | 186 (15.2%) | |
| Adequate | 21746 (50.8%) | 21208 (51.1%) | 538 (43.8%) | |
| Adequate plus | 9523 (22.3%) | 9207 (22.2%) | 316 (25.7%) | |
| Unknown | 1093 (2.6%) | 1062 (2.6%) | 31 (2.5%) | |
| Body mass index (kg/m2) | 0.21 | |||
| <25 | 14066 (32.9%) | 13643 (32.8%) | 423 (34.4%) | |
| 25 - 29.9 | 12218 (28.6%) | 11853 (28.5%) | 365 (29.7%) | |
| ≥30 | 14551 (34.0%) | 14168 (34.1%) | 383 (31.2%) | |
| Unknown | 1936 (4.5%) | 1879 (4.5%) | 57 (4.6%) | |
| Hypertensive disorders of pregnancy (prepregnancy or gestational) | 2643 (6.2%) | 2546 (6.1%) | 97 (7.9%) | 0.01 |
| Diabetes or abnormal glucose tolerance (prepregnancy or gestational) | 6172 (14.4%) | 5966 (14.4%) | 206 (16.8%) | 0.02 |
| Anemia | 4424 (10.3%) | 4298 (10.3%) | 126 (10.3%) | 0.92 |
| Smoking status | 1463 (3.4%) | 1400 (3.4%) | 63 (5.1%) | <0.001 |
| Substance use | 703 (1.6%) | 659 (1.6%) | 44 (3.6%) | <0.001 |
| Neonatal Characteristics | ||||
| Birth weight (grams), mean±SD | 3439.6 ± 456.7 | 3439.5 ± 456.3 | 3442.4 ± 470.0 | 0.82 |
| Small for gestational age | 2772 (6.5%) | 2691 (6.5%) | 81 (6.6%) | 0.87 |
| Large for gestational age | 6523 (15.3%) | 6343 (15.3%) | 180 (14.7%) | 0.56 |
| Institutional Characteristics | ||||
| Teaching Status | 15232 (35.6%) | 14546 (35.0%) | 686 (55.9%) | <0.001 |
| Hospital ownership | <0.001 | |||
| Government | 7555 (17.7%) | 7311 (17.6%) | 244 (19.9%) | |
| Private for-profit | 7672 (17.9%) | 7557 (18.2%) | 115 (9.4%) | |
| Private nonprofit | 27544 (64.4%) | 26675 (64.2%) | 869 (70.8%) | |
| Affiliation with a multi-hospital system | 29855 (69.8%) | 28923 (69.6%) | 932 (75.9%) | <0.001 |
| Average annual delivery volume, median (IQR) | 2648.7 (1795.3-4050.7) | 2648.7 (1779.0-4050.7) | 2764.0 (1997.7-4226.3) | <0.001 |
| Neonatal care capacity | <0.001 | |||
| Level I | 7198 (16.8%) | 7036 (16.9%) | 162 (13.2%) | |
| Level II | 9696 (22.7%) | 9477 (22.8%) | 219 (17.8%) | |
| Level III | 22565 (52.8%) | 21895 (52.7%) | 670 (54.6%) | |
| Level IV | 3312 (7.7%) | 3135 (7.5%) | 177 (14.4%) | |
| Blood bank service in house | 34836 (81.4%) | 33830 (81.4%) | 1006 (81.9%) | 0.66 |
| 24-hour in house anesthesia | 26885 (62.9%) | 26079 (62.8%) | 806 (65.6%) | 0.04 |
| Proportion of live births attended by a midwife (%), median (IQR) | 0.01 (0.00-4.5) | 0.01 (0.00-2.9) | 0.03 (0.01-19.5) | <0.001 |
| Safety-net burden (%), median (IQR) | 37.3 (21.8-56.2) | 37.3 (21.8-56.2) | 32.2 (20.3-56.3) | <0.001 |
| Malpractice insurance premium ($), median (IQR) | 59269 (34256-66039) | 59269 (34256-66039) | 34256 (32846-63434) | <0.001 |
IQR = interquartile range; SD = standard deviation; TOLAC = trial of labor after prior cesarean delivery.
Statistics were reported as N (%) unless otherwise specified.
After adjusting for patient and hospital characteristics in multivariable regression analysis, older maternal age, early-term, maternal obesity, and anemia were significantly associated with a lower likelihood of TOLAC attempt (Table 2). In contrast, late-term gestation, more previous vaginal deliveries, greater utilization of prenatal care, and maternal substance use were associated with a higher likelihood of TOLAC attempt (Table 2).
Table 2.
Patient and hospital characteristics associated with utilization and success of trial of labor
| Characteristics | Utilization of TOLAC | Success of TOLAC |
|---|---|---|
| Adjusted OR (95% CI)* | Adjusted OR (95% CI)*† | |
| Maternal Characteristics | ||
| Maternal age (years) | 0.14 (0.05-0.34) | 1.56 (0.26-9.53) |
| Gestational age (weeks) | ||
| 37 | 0.47 (0.37-0.60) | 0.35 (0.19-0.63) |
| 38 | 0.41 (0.34-0.50) | 0.50 (0.31-0.81) |
| 39 | 0.24 (0.20-0.28) | 0.42 (0.28-0.65) |
| 40 | Ref | Ref |
| 41-42 | 2.00 (1.45-2.75) | 0.46 (0.23-0.92) |
| Number of previous vaginal deliveries | 1.35 (1.28-1.43) | 1.62 (1.38-1.89) |
| Race and ethnicity | ||
| Non-Hispanic white | Ref | Ref |
| Non-Hispanic black | 0.95 (0.74-1.23) | 0.83 (0.44-1.56) |
| Hispanic | 0.87 (0.74-1.01) | 1.02 (0.68-1.51) |
| Other | 0.82 (0.65-1.03) | 1.08 (0.61-1.92) |
| Kotelchuck Adequacy of Prenatal Care | ||
| Utilization Index | ||
| Inadequate | 1.03 (0.84-1.26) | 1.57 (0.96-2.57) |
| Intermediate | 1.06 (0.88-1.28) | 1.37 (0.88-2.12) |
| Adequate | Ref | Ref |
| Adequate plus | 1.20 (1.03-1.41) | 1.31 (0.87-1.97) |
| Unknown | 1.07 (0.71-1.60) | NA‡ |
| Body mass index category (kg/m2) | ||
| <25 | Ref | Ref |
| 25 - 29.9 | 0.95 (0.81-1.10) | 0.79 (0.55-1.15) |
| ≥30 | 0.77 (0.66-0.89) | 0.56 (0.38-0.83) |
| Unknown | 0.76 (0.56-1.03) | 1.48 (0.69-3.16) |
| Hypertensive disorders of pregnancy (prepregnancy or gestational) | 1.06 (0.84-1.33) | 0.35 (0.18-0.66) |
| Diabetes or abnormal glucose tolerance (prepregnancy or gestational) | 1.07 (0.90-1.27) | 0.74 (0.49-1.14) |
| Anemia | 0.76 (0.62-0.92) | 0.67 (0.41-1.11) |
| Smoking status | 0.90 (0.67-1.22) | 0.84 (0.38-1.83) |
| Substance use | 1.52 (1.06-2.17) | 1.44 (0.55-3.80) |
| Neonatal Characteristics | ||
| Small for gestational age | 0.93 (0.73-1.18) | 1.33 (0.73-2.43) |
| Large for gestational age | 0.91 (0.77-1.08) | 0.95 (0.62-1.47) |
| Institutional Characteristics | ||
| Teaching status (Yes vs. No) | 1.59 (1.07-2.37) | 0.96 (0.51-1.83) |
| Hospital ownership | ||
| Government (non-federal) | 1.22 (0.75-1.98) | 1.55 (0.69-3.47) |
| Private for-profit | 1.04 (0.63-1.73) | 1.08 (0.44-2.60) |
| Private nonprofit | Ref | Ref |
| Affiliation with a multi-hospital system (Yes vs. No) | 1.12 (0.74-1.68) | 1.04 (0.51-2.12) |
| Average annual delivery volume | ||
| Small (<1000) | Ref | Ref |
| Medium (1000-2500) | 1.03 (0.61-1.74) | 0.42 (0.15-1.16) |
| Large (>2500) | 0.78 (0.44-1.41) | 0.41 (0.14-1.21) |
| Neonatal care capacity | ||
| Level I | Ref | Ref |
| Level II | 1.17 (0.68-2.03) | 1.84 (0.64-5.27) |
| Level III | 1.79 (1.03-3.11) | 2.97 (1.10-8.08) |
| Level IV | 3.27 (1.54-6.91) | 3.22 (0.92-11.19) |
| Blood bank service in house (Yes vs. No) | 1.29 (0.83-2.02) | 0.60 (0.28-1.26) |
| 24-hour in house anesthesia (Yes vs. No) | 1.02 (0.72-1.44) | 1.72 (0.95-3.11) |
| Proportion of live births attended by a midwife (in 10 percentage points) | 1.14 (1.02-1.28) | 1.28 (1.06-1.55) |
| Safety-net burden (in 10 percentage points) | 0.92 (0.84-1.001) | 0.90 (0.78-1.03) |
| Malpractice insurance premium (in $10,000s) | 0.80 (0.72-0.90) | 1.02 (0.85-1.23) |
CI = confidence interval; OR = odds ratio; TOLAC = trial of labor after prior cesarean delivery.
All variables reported in this table were included in the adjustment model.
Among births that attempted TOLAC.
Patients with unknown Kotelchuck Adequacy of Prenatal Care Utilization index were excluded from this regression analysis due to small numbers.
Multivariable regression analysis also showed that number of previous vaginal deliveries was positively associated with patients’ likelihood of successful trial of labor (Table 2). In contrast, early- and late-term gestation, maternal obesity and hypertensive disorders of pregnancy were associated with lower likelihood of successful TOLAC (Table 2).
Importantly, we found several hospital characteristics that were significantly associated with TOLAC attempt and TOLAC success even after adjusting for patient characteristics (Table 2). Teaching hospitals, hospitals with a level III or IV NICU, and hospitals that had a higher proportion of births attended by a midwife were also more likely to attempt TOLAC, whereas hospitals located in areas with higher malpractice insurance premiums were less likely to attempt TOLAC (Table 2). Among women who attempted TOLAC, those delivering at hospitals with Level III NICU or a higher proportion of births attended by midwives were more likely to deliver vaginally (Table 2).
For evaluation of maternal and neonatal outcomes, propensity score matching generated well-balanced comparison groups. In all propensity score matched samples, women who attempted TOLAC were comparable to those who underwent planned cesarean delivery on all patient and hospital characteristics assessed. Propensity score matched analysis showed no significant difference in the risk of composite severe maternal morbidity, NICU admission, or moderate unexpected newborn complication between women who attempted TOLAC and those who underwent planned cesarean delivery (Table 3). However, TOLAC was associated with a higher risk for the composites of severe unexpected newborn complications (OR=1.78, 95% CI: 1.04-3.04) and total unexpected newborn complications (OR=1.66, 95% CI: 1.11-2.49), as well as neonatal respiratory complications (OR=2.08, 95% CI: 1.17-3.71). Since 39.4% of the births attempting TOLAC delivered vaginally while only a small proportion experienced adverse outcomes, maternal and neonatal length of stay was generally shorter for those who attempted TOLAC (mean ratio=0.89, 95% CI: 0.85-0.92; and 0.85, 95% CI: 0.82-0.89; respectively). These findings are largely consistent with results from analysis using the adjustment method (Table 3).
Table 3.
Birth outcomes by trial of labor status
| Outcomes | Overall Sample (unadjusted analysis) |
Overall Sample (adjusted analysis) |
Propensity Score Matched Sample |
|||
|---|---|---|---|---|---|---|
| N | Planned Cesarean Delivery | TOLAC | P Value | Odds Ratio or Mean Ratio* (95% CI) | Odds Ratio or Mean Ratio† (95% CI) | |
| Maternal Outcomes | ||||||
| Severe maternal morbidity | 42771‡ | 577/41543 (1.4%) | 25/1228 (2.0%) | 0.06 | 1.51 (0.99-2.32) | 1.16 (0.70-1.91) |
| Length of stay (days), median (IQR), mean ± SD | 42771‡ | 3 (2-3); 2.9 ± 1.0 | 2 (2-3); 2.6 ± 1.3 | <0.001 | 0.91 (0.87-0.94) | 0.89 (0.85-0.92) |
| Neonatal Outcomes | ||||||
| NICU Admission | 25877§ | 742/25030 (3.0%) | 41/847 (4.8%) | 0.002 | 1.46 (1.03-2.06) | 1.35 (0.91-1.99) |
| Unexpected newborn complication | ||||||
| Total | 41299‖ | 870/40128 (2.2%) | 41/1171 (3.5%) | 0.002 | 1.57 (1.12-2.19) | 1.66 (1.11-2.49) |
| Severe | 41299‖ | 544/40128 (1.4%) | 24/1171 (2.0%) | 0.04 | 1.69 (1.10-2.61) | 1.78 (1.04-3.04) |
| Moderate | 41299‖ | 326/40128 (0.8%) | 17/1171 (1.5%) | 0.02 | 1.42 (0.85-2.36) | 1.50 (0.81-2.75) |
| Respiratory subcategory | 41299‖ | 338/40128 (0.8%) | 22/1171 (1.9%) | <0.001 | 1.79 (1.13-2.82) | 2.08 (1.17-3.71) |
| Length of stay (days), median (IQR), mean ± SD | 42771‡ | 3 (2, 3); 2.9 ± 1.5 | 2 (2, 3); 2.5 ± 1.3 | <0.001 | 0.85 (0.82-0.88) | 0.85 (0.82-0.89) |
CI = confidence interval; IQR = interquartile range; NICU = neonatal intensive care unit; SD = standard deviation; TOLAC = trial of labor after prior cesarean delivery.
Odds ratio (or mean ratio) and 95% CI were estimated based on multivariable regression analysis in the overall sample with adjustment for the same set of patient and hospital characteristics as those included in Table 2, along with a hospital random effect.
Odds ratio (or mean ratio) and 95% CI were estimated based on regression analysis in the propensity score matched sample with a random effect for hospital and a random effect for matched patients.
Included all births in the original sample. The corresponding propensity score matched sample included N=3509 patients undergoing planned cesarean delivery and N=1210 patients attempting TOLAC.
Limited to births at hospitals with level III-IV nursery. The corresponding propensity score matched sample included N=2377 patients undergoing planned cesarean delivery and N=824 patients attempting TOLAC.
Limited to births that met the eligibility criteria for measuring unexpected newborn complications (e.g., birthweight ≥2500 grams and no maternal drug use). The corresponding propensity score matched sample included N=3353 patients undergoing planned cesarean delivery and N=1153 patients attempting TOLAC.
DISCUSSION
In this study, we showed that 2.9% of women attempted trial of labor after two prior cesarean deliveries in California in 2010-2012, and among those, 39.4% delivered vaginally. There was no significant difference in the risk of composite severe maternal morbidity between women who attempted labor and those who did not. However, we found a modest increase in the risk of composite severe newborn complications among women who attempted labor compared to those who did not (2.0% versus 1.4%). After accounting for differences in patient and hospital characteristics, propensity score matched analysis showed that TOLAC was associated with an odds ratio of 1.78 in the risk of composite severe newborn complications.
Several prior studies have examined outcomes of women who attempted TOLAC after two prior cesarean deliveries.4–7 Using data from the 1980s-early 2000s, these studies showed that 27-54% of the patients attempted TOLAC, and 62-75% of those who attempted TOLAC were able to deliver vaginally,4–7 which were higher than the rates of TOLAC utilization and TOLAC success found in our cohort of California women in 2010-2012. This likely reflects national trends in reduced utilization of TOLAC as a whole in recent years (e.g., birth certificate data from 41 states and the District of Columbia showed that only 6.8% of the mothers with two or more previous cesarean deliveries attempted TOLAC in 2013 with a 51.4% success rate).21,22 The less TOLAC-friendly environment in more recent years might also lower the threshold when cesarean delivery was initiated. Other factors that might explain the lower success rate observed in our study include potential differences in patient population and the type of hospitals involved. For instance, in our study, 32.3% of the patients attempting TOLAC had prior vaginal deliveries and 44.1% of the TOLAC attempts occurred at non-teaching hospitals. In contrast, one of the prior studies reported 51.4% of the patients attempting TOLAC had prior vaginal deliveries with all births occurring at large academic centers.5
Although we did not find a significant association between TOLAC and the risk for composite of severe maternal morbidities, we were not able to examine individual maternal morbidities due to rarity of such events, such as uterine rupture. Prior studies reported a significantly higher risk of uterine rupture among women attempting labor after two or more prior cesarean deliveries (0.9-3.7%) compared to women who did not attempt labor (0-0.03%).4–7,23 These findings highlight the need for careful patient selection when attempting TOLAC. In this regard, our analysis showed that women who were obese, had fewer previous vaginal deliveries, had a hypertensive disorder, or were in the early- or late-term may be less appropriate candidates as they had lower odds of successful TOLAC. This is consistent with risk factors identified in prior studies regarding likelihood of TOLAC success after prior cesarean delivery.24–26 Meanwhile, the risks of TOLAC failure and neonatal morbidity should be balanced against a woman’s planned family size as a possible fourth cesarean in subsequent pregnancy carries a different risk profile than a third cesarean.1 For instance, the risk of placenta previa, placenta accreta, and hysterectomy reached 2.3%, 2.1% and 2.4%, respectively, for a fourth cesarean delivery, compared to 1.1%, 0.6% and 0.9%, respectively, for a third cesarean delivery.1
We also found important institutional characteristics that were associated with TOLAC utilization and success. In particular, hospitals with teaching status, higher levels of NICU, more midwife involvement, and less litigious environments were more likely to have patients using TOLAC. Success rate of TOLAC was also higher at hospitals with higher levels of NICU or more midwife involvement. These findings suggest a potential need to better address concerns of lower-capacity hospitals or hospitals with high medical-legal pressure when managing patients with two prior cesarean deliveries. For instance, prior research has reported fear for liability and considerations for meeting the ACOG guidelines about required resources (e.g., immediate availability of emergency cesarean delivery) as major factors influencing providers’ practice regarding offering TOLAC.27,28 On the other hand, more midwifery involvement in care may suggest an overall culture of low-interventional approaches to intrapartum care, which has been associated with higher rates of vaginal birth in women after prior cesarean delivery in a recent study.29
We recognize several limitations of this study. First, we used data from a single state and hence the findings may not generalize to patients elsewhere. Second, our study relied on retrospective analysis of administrative data. Although we applied a careful algorithm to identify women who attempted TOLAC by integrating information from both the birth certificate and hospital discharge records, these information sources might not accurately capture intended mode of delivery and we may have over- or under-estimated the occurrence of TOLAC. Likewise, administrative data might lack sufficient detail regarding patient clinical risk factors, limiting our ability of risk-adjustment and propensity score matching when comparing maternal and neonatal outcomes between patients who underwent TOLAC versus those who did not.
Reducing morbidities from repeated cesarean deliveries is an important goal and requires improved understanding of the risks and benefits associated with TOLAC. Our data showed that trial of labor after two prior cesareans was used only by 2.9% of women and resulted in vaginal delivery in 39.4% of them. TOLAC in this population was associated with a modest increase in severe neonatal morbidity. However, the overall risk of neonatal morbidities is relatively low and should be considered in conjunction with the potential benefit of TOLAC in preventing cesarean delivery in subsequent pregnancies. For women with two prior cesarean deliveries, more research is needed to determine the optimal candidates and hospital environments to safely avoid a third cesarean delivery.
Supplementary Material
Acknowledgments
Funding: This project was supported by grant number R01HS023801 from the Agency for Healthcare Research and Quality. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality.
Footnotes
Financial Disclosure
Henry C. Lee reports receiving consulting funds from Lansinoh Laboratories, Inc. as a member of a clinical advisory board. The other authors did not report any potential conflicts of interest.
Presented at the AcademyHealth Annual Research Meeting in Washington, DC, June 2-4, 2019.
REFERENCES
- 1.Silver RM, Landon MB, Rouse DJ, et al. Maternal morbidity associated with multiple repeat cesarean deliveries. Obstet Gynecol. 2006;107(6):1226–1232. [DOI] [PubMed] [Google Scholar]
- 2.Sondgeroth KE, Wan L, Rampersad RM, et al. Risk of Maternal Morbidity with Increasing Number of Cesareans. Am J Perinatol. 2019;36(4):346–351. [DOI] [PubMed] [Google Scholar]
- 3.ACOG Practice Bulletin No. 205: Vaginal Birth After Cesarean Delivery. Obstet Gynecol. 2019;133(2):e110–e127. [DOI] [PubMed] [Google Scholar]
- 4.Caughey AB, Shipp TD, Repke JT, Zelop CM, 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] [PubMed] [Google Scholar]
- 5.Landon MB, Spong CY, 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] [PubMed] [Google Scholar]
- 6.Macones GA, Cahill A, Pare E, et al. Obstetric outcomes in women with two prior cesarean deliveries: is vaginal birth after cesarean delivery a viable option? Am J Obstet Gynecol. 2005;192(4):1223–1228; discussion 1228-1229. [DOI] [PubMed] [Google Scholar]
- 7.Miller DA, Diaz FG, Paul RH. Vaginal birth after cesarean: a 10-year experience. Obstet Gynecol. 1994;84(2):255–258. [PubMed] [Google Scholar]
- 8.State of California. Office of Statewide Health Planning and Development (OSHPD). Types of OSHPD Patient-Level Data. https://www.oshpd.ca.gov/HID/Data_Request_Center/Types_of_Data.html Accessed May 1, 2018.
- 9.Callaghan WM, Creanga AA, Kuklina EV. Severe maternal morbidity among delivery and postpartum hospitalizations in the United States. Obstet Gynecol. 2012;120(5):1029–1036. [DOI] [PubMed] [Google Scholar]
- 10.Centers for Disease Control and Prevention. Severe maternal morbidity in the United States. https://www.cdc.gov/reproductivehealth/MaternalInfantHealth/SevereMaternalMorbidity.html Accessed May 15, 2017.
- 11.California Maternal Quality Care Collaborative. Unexpected Complications in Term Newborns. https://www.cmqcc.org/focus-areas/quality-metrics/unexpected-complications-term-newborns Accessed March 16, 2016.
- 12.National Perinatal Information Center. V.12.4 Special Report: Unexpected Newborn Complications Updated (NQF #716 “Healthy Term Newborn”). http://www.npic.org/wp-content/uploads/2018/11/DBAnalysis_Unxepected_NB_Complications_2013-07.pdf Accessed November 25, 2019.
- 13.Harrison W, Goodman D. Epidemiologic Trends in Neonatal Intensive Care, 2007-2012. JAMA Pediatr. 2015;169(9):855–862. [DOI] [PubMed] [Google Scholar]
- 14.Centers for Disease Control and Prevention National Center for Health Statistics. Guide to Completing the Facility Worksheets for the Certificate of Live Birth and Report of Fetal Death. May 2016. https://www.cdc.gov/nchs/data/dvs/GuidetoCompleteFacilityWks.pdf Accessed November 1, 2019.
- 15.Kotelchuck M An evaluation of the Kessner Adequacy of Prenatal Care Index and a proposed Adequacy of Prenatal Care Utilization Index. Am J Public Health. 1994;84(9):1414–1420. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.American Hospital Association. AHA Annual Survey Database. https://www.ahadataviewer.com/additional-data-products/AHA-Survey/ Accessed May 1, 2018.
- 17.California Perinatal Quality Care Collaborative. Stanford, CA: https://www.cpqcc.org/ Accessed May 1, 2018. [Google Scholar]
- 18.Medical Liability Monitor. Annual Rate Survey Issue. Oak Park, IL: Medical Liability Monitor; October 2011. Vol 36 No. 10 https://medicalliabilitymonitor.com/rate-survey/. [Google Scholar]
- 19.State of California. Office of Statewide Health Planning and Development (OSHPD). Hospital Annual Financial Data. https://www.oshpd.ca.gov/HID/Hospital-Financial.html Accessed October 20, 2017.
- 20.Austin PC. A comparison of 12 algorithms for matching on the propensity score. Stat Med. 2014;33(6):1057–1069. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.American College of Obstetrician and Gynecologists, Society for Maternal-Fetal M. Obstetric care consensus no. 1: safe prevention of the primary cesarean delivery. Obstet Gynecol. 2014;123(3):693–711. [DOI] [PubMed] [Google Scholar]
- 22.Curtin SC, Gregory KD, Korst LM, Uddin SF. Maternal Morbidity for Vaginal and Cesarean Deliveries, According to Previous Cesarean History: New Data From the Birth Certificate, 2013. Natl Vital Stat Rep. 2015;64(4):1–13, back cover. [PubMed] [Google Scholar]
- 23.Asakura H, Myers SA. More than one previous cesarean delivery: a 5-year experience with 435 patients. Obstet Gynecol. 1995;85(6):924–929. [DOI] [PubMed] [Google Scholar]
- 24.Wu Y, Kataria Y, Wang Z, Ming WK, Ellervik C. Factors associated with successful vaginal birth after a cesarean section: a systematic review and meta-analysis. BMC Pregnancy Childbirth. 2019;19(1):360. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Xu X, Lee HC, Lin H, et al. Hospital variation in utilization and success of trial of labor after a prior cesarean. Am J Obstet Gynecol. 2019;220(1):98 e91–98 e14. [DOI] [PubMed] [Google Scholar]
- 26.Grobman WA, Lai Y, Landon MB, et al. Development of a nomogram for prediction of vaginal birth after cesarean delivery. Obstet Gynecol. 2007;109(4):806–812. [DOI] [PubMed] [Google Scholar]
- 27.Cox KJ. Providers’ perspectives on the vaginal birth after cesarean guidelines in Florida, United States: a qualitative study. BMC Pregnancy Childbirth. 2011;11:72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Wells CE. Vaginal birth after cesarean delivery: views from the private practitioner. Semin Perinatol. 2010;34(5):345–350. [DOI] [PubMed] [Google Scholar]
- 29.Lundsberg LS, Main EK, Lee HC, Lin H, Illuzzi JL, Xu X. Low-interventional approaches to intrapartum care: hospital variation in practice and associated factors. J Midwifery Womens Health. 2020;65(1):33–44. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.