Chorioamnionitis occurs in approximately 0.5–1% of pregnancies.1 Risk factors for it include premature rupture of membranes, abnormal vaginal flora (vaginal Ureaplasma urealyticum colonization,2 bacterial vaginosis, sexually transmitted disease), obstetric manipulations (vaginal examination or internal fetal monitoring), and diminished host immune response (caused by smoking, drug abuse, or immunodeficiency).3 Although chorioamnionitis now rarely causes maternal death, it is associated with serious maternal morbidity, partly caused by high cesarean delivery rates among women with chorioamnionitis.4,5 It also is associated with increased risk of preterm labor,6 perinatal death,7,8 and neonatal morbidity, such as meningitis,5 pneumonia,9 sepsis,10 and cerebral palsy.11,12
The relationship between chorioamnionitis and uterine function is less clear. Several small studies have suggested that chorioamnionitis has an inhibitory effect on labor, characterized by decreased uterine contractility, decreased sensitivity to oxytocin stimulation, and subnormal cervical dilation.4,5,13 Although most women with chorioamnionitis in those studies received oxytocin and one third had cesareans, no control groups were included. One study also found a relationship between chorioamnionitis and increased blood loss.14
Despite the serious sequelae of chorioamnionitis and the common belief that it adversely affects the progress of labor and uterine function, its relationship to specific types of labor abnormalities has not been fully investigated. The purpose of this study was to examine the relationship between chorioamnionitis and abnormalities of uterine function, including labor abnormalities, need for oxytocin augmentation, cesarean delivery, and hemorrhage during parturition.
Materials and Methods
This retrospective cohort study examined data from the University of California, San Francisco Perinatal Database, which includes data from every delivery there since 1976. Antepartum and intrapartum variables were recorded at the time of delivery, and entries were crosschecked for logic errors before final inclusion in the database. Usage of the database for this study was approved by the University of California, San Francisco Committee on Human Research. From the Perinatal Database we identified women with singleton vertex presentation, who had labor with gestational ages of at least 24 weeks, and who delivered between January 1, 1986 and December 31, 1996. Onset of labor was defined as the time when regular, painful contractions began more frequently than every 5 minutes, leading to cervical change.15
For this study, low birth weight was defined as less than 2500 g. Preterm delivery was defined as fewer than 37 completed weeks of gestation. Prolonged first stage was defined as longer than 12 hours in multiparas and 15 hours in nulliparas.15 Prolonged second stage was defined as longer than 1 hour in multiparas and 2 hours in nulliparas.15 Those values represented the 95% confidence interval (CI) for normal labor lengths. Hemorrhage was defined as estimated blood loss greater than 1000 mL during cesarean deliveries or more than 500 mL during vaginal deliveries. Other extracted variables included maternal age, delivery date, parity, epidural usage, oxytocin usage, type of labor (spontaneous versus induced), cesarean delivery, indications for cesarean, operative delivery, duration of stage 3 labor, specific labor abnormalities, antibiotic treatment, and chorioamnionitis status. In the database, prolonged latent phase was defined as more than 12 hours in nulliparas and 6 hours in multiparas.
Chorioamnionitis was diagnosed by clinical signs of intra-amniotic infection, ie, maternal fever (temperature above 37.8C) and one or more of the following conditions: maternal tachycardia, uterine tenderness, purulent amniotic fluid (AF), fetal tachycardia, or peripheral blood leukocytosis. Clinical signs were used because AF culture was not routinely done in the study population. A randomly selected 10% sample of charts of women in the exposed (chorioamnionitis) group and a randomly selected 0.8% sample of charts of women in the unexposed (no chorioamnionitis) group were reviewed to confirm accuracy of coded diagnoses. Charts were selected for review based on a random numbers table matched to facility identification numbers. Women who did not meet the eligibility criteria for the chorioamnionitis group were included in the comparison group.
Using the Statistical Package for Social Sciences (SPSS, Inc., Chicago, IL), we analyzed outcome variables by chorioamnionitis exposure to produce χ2 statistics and relative risks (RRs) and 95% CIs for each dichotomous outcome. Student t test was used for analysis of continuous outcome variables. Adjustments for clinically relevant and statistically significant confounding factors were made through multiple logistic regression. If two confounding variables were significantly correlated, only one was chosen for the final logistic model on the basis of greater clinical significance. Adjusted estimates of the odds ratio (OR) value were calculated by taking the antilog of the beta coefficients (eβ) derived from logistic regression and calculating corresponding 95% CI.
From January 1986 through December 1996, 16,226 deliveries that met study eligibility criteria were entered into the Perinatal Database. Among those, 1379 (8.5%) had chorioamnionitis noted. Random chart reviews on 10% of deliveries coded as having chorioamnionitis (138 charts) showed 91% coding accuracy; chart reviews on 0.8% (115 charts) of deliveries without chorioamnionitis showed 96% coding accuracy. When a randomly reviewed chart had been coded in error, the coding for chorioamnionitis diagnosis was corrected for that subject only.
Table 1 presents the characteristics for the chorioamnionitis and nonchorioamnionitis groups. Statistically and clinically significant differences between the groups included a higher proportion of nulliparous women and preterm and low-birth-weight births in the chorioamnionitis group.
Table 2 presents labor and delivery characteristics for the two groups. There were several notable findings, including a significantly higher proportion of women in the chorioamnionitis group who had cesarean deliveries and hemorrhages. There was also a significantly higher proportion of women with labor abnormalities in the chorioamnionitis group, a relationship that held true for all first- and second-stage abnormalities investigated.
Table 3 presents the unadjusted and adjusted ORs and associated 95% CIs. Odds ratios rather than RRs were used because multiple logistic regression allows one to control for confounding variables, but it does not allow for direct calculation of adjusted RR values. It only allows calculation of adjusted OR values that serve as estimates of RR. For several outcomes, the RR and unadjusted OR were discrepant because the OR approaches the RR only when there is a rare outcome. Although the RR values in Table 2 indicate the true incidence of outcomes in that cohort, the unadjusted OR values are useful for comparison with the adjusted OR values when adjustments are made for confounding variables. The change in OR with adjustment allowed us to evaluate the effect of adjustment on the relationship between chorioamnionitis and outcome variables. After adjustment for clinically relevant and statistically significant confounding variables, a significant relationship remained between chorioamnionitis and increased epidural use, increased oxytocin use, prolonged first stage of labor, labor abnormalities (particularly in the first stage), increased cesearean rate, and hemorrhage.
Chorioamnionitis appears statistically significantly associated with abnormalities in labor, including increased use of oxytocin, increased cesarean rates, hemorrhage, and specific labor abnormalities, with adjusted OR greater than 2.0 for several first-stage abnormalities. We also observed that chorioamnionitis was associated with a significantly higher risk of hemorrhage in vaginal and cesarean deliveries.
Although it is commonly believed that chorioamnionitis adversely affects the progress of labor and uterine function, specific influences are not known. Duff et al4 found increased cesarean delivery rates when chorioamnionitis was present, and Silver et al13 found that highly virulent bacteria were associated with greater oxytocin requirements, perhaps because of adverse bacterial effects on the myometrium. In a more recent study, Satin et al16 found that rates of cesarean for dystocia increased only when chorioamnionitis presented late in labor, leading them to conclude that chorioamnionitis was not the cause of dysfunctional labor, but was a result of dysfunctional labor resulting in dystocia.
Time of chorioamnionitis onset was not recorded in our database and was difficult to extract from medical records, so we cannot comment on the possibility of differing effects of chorioamnionitis early or late in labor. However, it is clear that in a large sample of the population, the primary association between chorioamnionitis and labor or uterine function is adverse with respect to labor abnormalities, need for oxytocin augmentation, cesarean delivery rates, and hemorrhage.
Potential limitations of this study included validity and reliability of information in the database and medical records. Chart reviews confirmed the high accuracy of database coding. Another potential limitation was the slight possibility that within the study population there were women who delivered more than once, thereby violating the underlying assumption of independence of observations. We believe that in light of the large study population, that issue would be unlikely to affect our findings markedly. The retrospective nature of our study precluded our ability to make conclusions about timing. For example, although our results show that chorioamnionitis exposure is significantly associated with a prolonged first stage of labor, even after controlling for confounding factors, it is not possible to distinguish whether chorioamnionitis led to prolongation by uterine dysfunction or prolongation led to chorioamnionitis. Further study of the treatment and timing of its diagnosis would help us understand the etiology. The size of our study population and the use of a nonchorioamnionitis comparison group were the primary advantages of the present study over previous studies on the effects of chorioamnionitis on labor.
1. Gibbs RS, Castillo MS, Rodgers PJ. Management of acute chorioamnionitis. Am J Obstet Gynecol 1980;136:709–13.
2. Abele-Horn M, Peters J, Genzel-Boroviczeny O, Wolff C, Zimmermann A, Gottschling W. Vaginal Ureaplasma urealyticum
colonization: Influence on pregnancy outcome and neonatal morbidity. Infection 1997;25:286–91.
3. Casey BM, Cox SM. Chorioamnionitis and endometritis. Infect Dis Clin North Am 1997;11:203–22.
4. Duff P, Sanders R, Gibbs RS. The course of labor in term patients with chorioamnionitis. Am J Obstet Gynecol 1983;147:391–5.
5. Hauth JC, Gilstrap LC, Hankins GDV, Connor KD. Term maternal and neonatal complications of acute chorioamnionitis. Obstet Gynecol 1985;66:59–62.
6. Gomez R, Ghezzi F, Romero R, Munoz H, Tolosa JE, Rojas I. Premature labor and intra-amniotic infection. Clinical aspects and role of the cytokines in diagnosis and pathophysiology. Clin Perinatol 1995;22:281–342.
7. Quinn PA, Butany J, Chipman M, Taylor J, Hannah W. A prospective study of microbial infection in stillbirths and early neonatal death. Am J Obstet Gynecol 1985;151:238–49.
8. Gibbs RS, Castillo MS, Rodgers PJ. Management of acute chorioamnionitis. Am J Obstet Gynecol 1980;136:709–13.
9. de Araujo MC, Schultz R, Vaz FA, Massad E, Feferbaum R, Ramos JL. A case-control study of histological chorioamnionitis and neonatal infection. Early Hum Develop 1994;40:51–8.
10. Yancey MK, Duff P, Kubilis P, Clark P, Frentzen BH. Risk factors for neonatal sepsis. Obstet Gynecol 1996;87:188–94.
11. Grether JK, Nelson KB. Maternal infection and cerebral palsy in infants of normal birth weight. JAMA 1997;278:207–11.
12. O'shea TM, Klinepeter KL, Dillard RG. Prenatal events and the risk of cerebral palsy in very low birth weight infants. Am J Epidemiol 1998;147:362–9.
13. Silver RK, Gibbs RS, Castillo M. Effect of amniotic fluid bacteria on the course of labor in nulliparous women at term. Obstet Gynecol 1986;68:587–92.
14. Nicol B, Croughan-Minihane M, Kilpatrick SJ. Lack of value of routine postpartum hematocrit determination after vaginal delivery. Obstet Gynecol 1997;90:514–8.
15. Kilpatrick SJ, Laros RK. Characteristics of normal labor. Obstet Gynecol 1989;74:85–7.
16. Satin AJ, Maberry MC, Leveno LJ, Sherman L, Kline DM. Chorioamnionitis: A harbinger of dystocia. Obstet Gynecol 1992;79:913–5.