Objective
We sought to examine the extent to which a temporal increase in preterm cesarean delivery is associated with gestational age–specific changes in perinatal survival in preterm gestations.
Study Design
We utilized data on singleton births in the United States (1990 through 2004) delivered between 24-36 weeks’ gestation. Associations between changes in cesarean delivery at preterm gestations and trends in the risk of preterm stillbirth, and neonatal and perinatal mortality were estimated before and after adjustments for a variety of potential confounders.
Results
From 1990 through 2004, cesarean delivery rates increased by 50.6%, 40.7%, and 35.8% at 24-27, 28-33, and 34-36 weeks, respectively. The largest incremental effect of cesarean was associated with a reduction in stillbirths by 5.8%, 14.2%, and 23.1% at 24-27, 28-33, and 34-36 weeks, respectively, leading to an 11.4%, 4.9%, and 0.6% reduction in perinatal deaths at 24-27, 28-33, and 34-36 weeks, respectively.
Conclusion
Increasing rates of preterm cesarean were associated with improved perinatal survival. This association was evident largely because of dramatic incremental declines in stillbirths.
The recent decade has witnessed a dramatic increase in the rate of cesarean delivery in most industrialized societies. The rate of cesarean delivery in the United States (32% in 2007 ) has increased by >50% over the last decade, with consequent decline in the rate of vaginal birth following a previous cesarean. Such increases in cesarean delivery have been observed in both low-risk and high-risk populations, including among women who delivered very low birthweight (<1500 g) and macrosomic infants, breech presentations, multiple births, and in pregnancies with an array of maternal and fetal indications for cesarean.
Concurrent with an increase in cesarean deliveries is the temporal increase in the rate of preterm births in the United States and most other industrialized countries. Obstetrical interventions at preterm gestational ages are performed in the setting of impending fetal compromise (or maternal indications warranting such interventions), and these interventions include either a labor induction or a prelabor cesarean or both. The increase in preterm births in the United States has occurred with a concurrent temporal increase in the rate of indicated preterm births. For instance, from 1989 through 2000, indicated preterm births at <37 weeks among singleton gestations in the United States increased by approximately 50%. In fact, some investigators have interpreted the increasing preterm birth rate to be linked to the increase in cesarean deliveries. Despite these general trends, however, the extent to which temporal changes in cesarean delivery, especially at preterm gestations, are associated with gestational age–specific changes in perinatal mortality remain poorly understood.
We evaluated temporal changes in cesarean deliveries at preterm gestations in the United States, and investigated the extent to which such changes are associated with changes in preterm stillbirth and perinatal mortality rates. We carried out this study in a large population-based setting of singleton preterm births in the United States from 1990 through 2004.
Materials and Methods
We utilized the US-linked natality and infant deaths data files composed of births from 1990 through 2004, and the corresponding fetal death files for the same period. These data files correspond to data abstracted from birth certificates of live-born infants and from fetal and infant death certificates and assembled by the National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention. Infant deaths from 1992 through 1994 were not linked to the corresponding live births.
Gestational age, reported in completed weeks, was based on menstrual dates. In about 5% of births that did not contain a valid date for the menstrual period, or when the menstrual-based gestational age was incompatible with birthweight, a clinical estimate of gestation, also contained on these data files, was used instead. If a valid month and year of the last menstrual period was available but the day was missing, gestational age was imputed by the NCHS. The replacement of menstrual-based gestational age by a clinical estimate, as well as imputation of missing gestational age, was performed consistently by the NCHS for all years (1990 through 2004).
Statistical analysis
The primary variable for analysis pertains to the “method of delivery” data contained on the birth certificates that included both primary and repeat cesarean deliveries. We examined changes (1990 through 2004) in cesarean delivery rates at preterm gestations grouped as 24-27, 28-33, and 34-36 weeks. We then estimated the extent to which changes in cesarean delivery rates from 1990 through 2004 were associated with changes in stillbirth and perinatal mortality rates over the same period. Perinatal mortality included stillbirths and neonatal deaths. This study was restricted to singleton pregnancies that ended at preterm (24-36 weeks) gestations.
These preliminary analyses were followed by a multivariable adjustment for a variety of confounding factors. This was accomplished by fitting a series of log binomial regression models with mortality as the dependent variable and an indicator for period (1990 or 2004). In the first model, we estimated the crude change in mortality rates from 1990 (reference) through 2004. This was followed by a sequential adjustment for confounders (listed below) and, lastly, a model with confounders and an indicator for cesarean delivery. In all models, the indicator for period was the primary variable of interest, with risk ratio for mortality associated with period (denoting change in mortality rates from 1990 through 2004) expressed as a “relative change” in mortality. This was calculated as [exp(β)−1]*100 where β corresponds to the log risk ratio for the period effect (an indicator for year of birth coded 1 for 2004 and 0 for 1990).
We estimated the associations of changes in preterm cesarean delivery on trends in stillbirth and perinatal mortality at preterm gestational ages overall, as well as in the absence of ischemic placental disease (IPD). We defined IPD as the presence of ≥1 of the following conditions: pregnancy-induced hypertension, small for gestational age (SGA), or placental abruption. The latter analysis was deemed necessary to evaluate if there are any variations in the effect of preterm cesareans on mortality according to the underlying pathophysiology or, possibly, indications for cesarean delivery. We defined SGA as newborns with birthweight <10th centile for gestational age, with standards derived from all births in this study.
The confounders considered for adjustment included maternal age (grouped as <20, 20-24, 25-29, 30-34, 35-39, and ≥40 years), primiparity, single marital status, and maternal race (whites, blacks, and other races). In addition, to account for variations in cesarean delivery rates across states, we further adjusted the analyses for geographic region of birth. Geographic regions were classified as Northeast (Maine, New Hampshire, Vermont, Rhode Island, Massachusetts, Connecticut, New York, New Jersey, and Pennsylvania), Midwest (Michigan, Ohio, Illinois, Indiana, Wisconsin, Minnesota, Iowa, Missouri, North Dakota, South Dakota, Nebraska, and Kansas), South (Delaware, Maryland, District of Columbia, Virginia, West Virginia, North Carolina, South Carolina, Georgia, Florida, Kentucky, Tennessee, Alabama, Mississippi, Arkansas, Louisiana, Oklahoma, and Texas), and West (Montana, Idaho, Wyoming, Colorado, New Mexico, Arizona, Utah, Nevada, Washington, Oregon, California, Hawaii, and Alaska). The choice of confounders was based on a conservative change of 10% between the unadjusted and the confounder-adjusted risk ratios. Missing data on confounders were estimated to be <2%.
We also derived the number needed to treat to provide an indication of the number of cesarean deliveries that need to be performed to prevent a stillbirth or perinatal death. Although the number needed to treat provides a causal interpretation regarding the relationship between cesarean delivery and mortality, we use this statistic purely as a measure to understand the associations better.
Sensitivity analysis
We performed a sensitivity analysis to examine the effect of primary and repeat cesarean deliveries at preterm gestational ages on preterm stillbirth and perinatal mortality.
Statistical analysis was accomplished using the SAS system (version 9.2; SAS Institute, Cary, NC). This study was approved by the institutional review board of University of Medicine and Dentistry of New Jersey-Robert Wood Johnson Medical School, New Brunswick, NJ.
Results
Temporal changes in preterm cesarean deliveries and perinatal mortality
Temporal trends in rates of preterm cesarean delivery from 1990 through 2004 and the corresponding relative changes in rates with 1990 as the base period are shown in Figure 1 . Relative to the rates in 1990, preterm cesarean delivery increased by 37% (25.5% in 1990 to 35.0% in 2004). This increase in preterm cesarean was greatest at 24-27 weeks ( Table 1 ). There was an overall 20.2% decline in preterm stillbirths from 1990 through 2004 ( Figure 2 ); however, the greatest decline occurred at 34-36 weeks (26.0% decline) ( Table 2 ). A similar pattern of decline was observed for neonatal and perinatal deaths.
| Variable | Rates of preterm cesarean delivery, % | Change 1990 through 2004, % (95% CI) | |||
|---|---|---|---|---|---|
| 1990 | 1995 | 2000 | 2004 | ||
| Total births a | 406,591 | 382,671 | 406,349 | 435,599 | |
| Preterm cesarean | |||||
| 24-27 wk | 30.8 | 34.8 | 40.2 | 46.4 | 50.6 (47.1–54.1) |
| 28-33 wk | 29.9 | 31.5 | 35.5 | 42.0 | 40.7 (39.1–42.5) |
| 34-36 wk | 23.5 | 22.6 | 25.3 | 31.9 | 35.8 (34.6–36.9) |
| Preterm primary cesarean | |||||
| 24-27 wk | 25.1 | 28.6 | 32.7 | 37.0 | 47.5 (43.5–51.6) |
| 28-33 wk | 22.1 | 23.7 | 26.1 | 30.0 | 35.5 (33.5–37.6) |
| 34-36 wk | 15.1 | 14.7 | 15.9 | 19.3 | 27.6 (26.2–29.1) |
| Preterm repeat cesarean | |||||
| 24-27 wk | 5.7 | 6.3 | 7.5 | 9.4 | 63.9 (53.7–74.9) |
| 28-33 wk | 7.7 | 7.8 | 9.4 | 12.0 | 55.6 (51.4–60.0) |
| 34-36 wk | 8.4 | 7.9 | 9.4 | 12.6 | 50.4 (48.1–52.8) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
