OBJECTIVE: To investigate the role of infection and noninfectious inflammation in epidural analgesia-related fever.
METHODS: This was an observational analysis of placental cultures and serum admission and postpartum cytokine levels obtained from 200 women at low risk recruited during the prenatal period.
RESULTS: Women receiving labor epidural analgesia had fever develop more frequently (22.7% compared with 6% no epidural; P=.009) but were not more likely to have placental infection (4.7% epidural, 4.0% no epidural; P>.99). Infection was similar regardless of maternal fever (5.4% febrile, 4.3% afebrile; P=.7). Median admission interleukin (IL)-6 levels did not differ according to later epidural (3.2 pg/mL compared with 1.6 pg/mL no epidural; P=.2), but admission IL-6 levels greater than 11 pg/mL were associated with an increase in fever among epidural users (36.4% compared with 15.7% for 11 pg/mL or less; P=.008). At delivery, both febrile and afebrile women receiving epidural had higher IL-6 levels than women not receiving analgesia.
CONCLUSION: Epidural-related fever is rarely attributable to infection but is associated with an inflammatory state.
LEVEL OF EVIDENCE: II
At term, epidural-related fever is rarely attributable to infection but is associated with elevated interleukin-6 levels at admission.
From the Departments of Obstetrics and Gynecology, Brigham and Women's Hospital and Massachusetts General Hospital, Boston, Massachusetts; Department of Obstetrics and Gynecology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts; Department of Medicine, Clinical Microbiology Laboratory, and Department of Obstetrics and Gynecology, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts; Department of Pathology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts; Department of Newborn Medicine, Brigham and Women's Hospital, Boston, Massachusetts; Department of Anesthesiology, Perioperative and Pain Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts; Department of Anesthesia, Critical Care and Pain Medicine, Division of Obstetric Anesthesia, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts; Department of Obstetrics and Gynecology, Boston Medical Center, Boston University School of Medicine, Boston, Massachusetts.
Funding provided by the Maternal Child Health Bureau, Health Resources and Services Administration (R40 MC00248). Also supported by an NHLBI reentry award RO1HL081663 (ACC).
The authors thank the PACE and LABOR Study staff and doulas, especially study coordinator Gail Williams, data analysts Manuel Chinchilla and Claudia Stearns, the labor and delivery providers, laboratory personnel, and particularly, the women and their partners.
Presented in abstract form at the 29th Annual Meeting of the Society for Maternal-Fetal Medicine, January 26–31, 2009, San Diego, California.
Corresponding author: Laura Riley, MD, Department of Obstetrics and Gynecology, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114; e-mail: firstname.lastname@example.org.
Financial Disclosure The authors did not report any potential conflicts of interest.
Randomized trials and observational studies show that maternal fever occurs in approximately 15%–20% of laboring women after epidural analgesia.1–13 The pathogenesis of this fever is unclear.14 One theory is that fever is attributable to infection because women who choose epidural have longer labors with multiple interventions, increasing the risk for chorioamnionitis.15,16 Another theory is that epidural analgesia changes thermoregulatory mechanisms. Maternal heat is not dissipated when hyperventilation is inhibited and sweating is prevented by sympathetic blockade.1,4 Another theory is that fever is attributable to a noninfectious inflammatory process initiated or enhanced by labor epidural analgesia. Some studies have demonstrated elevated interleukin (IL)-6 levels associated with this fever,17–19 and other studies have utilized placental histology as indirect evidence of a noninfectious etiology.8,9,20 The aim of our study was to investigate the role of infection or noninfectious inflammation in epidural analgesia-related fever in a large number of women at low risk at term.
MATERIALS AND METHODS
The current study is an observational analysis of data from a randomized trial conducted from 2002 to 2005 to examine the etiology, physiologic correlates, and clinical consequences of epidural-related fever. The study was approved by the Partners Healthcare Institutional Review Board; written informed consent was obtained from study participants.
The study was conducted in two phases named Project Assessing Childbirth Epidural and Labor Assistance and Birth Outcomes Research. For both, we recruited women before 36 weeks of gestation from 14 clinical sites. Study participants were nulliparous with singleton low-risk pregnancies and planning to deliver at Brigham and Women's Hospital or Massachusetts General Hospital. Eligible women were 18 years or older, had a body mass index less than 40 at the first prenatal visit and spoke English or attended clinic visits with a translator. They had no history of pregnancy loss at 20 weeks of gestation or later, cerclage, chronic medical conditions such as hypertension or pregestational diabetes, contraindication to labor, psychiatric disorders requiring medication within 1 year, steroid use within 1 year, or illicit drug use within 1 year. In phase one (Project Assessing Childbirth Epidural, May 22, 2002–May 3, 2005), women were randomized to either the doula or the usual care group. Women in the doula group were asked to avoid epidural analgesia and were provided with doulas, who are assistants specially trained to offer support during labor. In phase two (Labor Assistance and Birth Outcomes Research, March 7, 2005–September 19, 2005), women were randomized to the doula or the usual care group but were not asked to avoid epidural analgesia. As part of the randomization scheme for both phases of the study, one third of women in each group were randomly designated to have biological samples evaluated, including admission, postpartum, and cord blood cytokines, placental cultures, and histology.
A study representative recorded maternal temperatures every 2 hours during labor and collected biologic samples for women in the usual care and doula groups. Information regarding labor, delivery, and neonatal events was abstracted from the medical record by trained abstracters. Women receiving penicillin, cephazolin, or clindamycin, alone were counted as receiving group B streptococci (GBS) prophylaxis. Women receiving a combination of antibiotics were assumed to be treated for clinical chorioamnionitis.
The epidural analgesia protocols were slightly different at Brigham and Women's Hospital and Massachusetts General Hospital. Eighteen percent of women (35/191) underwent a combined spinal epidural technique (details of anesthesia protocols available on request.) Because there were no differences in patient demographics, labor characteristics, or outcomes between women receiving combined spinal epidural and those receiving only epidural analgesia, the two groups were combined for analysis.
We drew blood samples from all participants at admission and within 1 hour after delivery and collected cord blood at delivery. These specimens were analyzed for IL-6 and IL-8 in the one third of the enrolled population preselected at randomization. Serum samples obtained from participants' peripheral blood and cord blood were flash-frozen at the time of collection and kept at −80°C until processed. Samples were analyzed for IL-6 and IL-8 concentrations by commercial enzyme-linked immunosorbent assays according to the manufacturer's protocol (BioSource). The values were converted to pg/mL by reference to a standard curve that was always generated in parallel to the test samples. The lower limits of sensitivity were 0.16 pg/mL for IL-6 and 0.39 pg/mL for IL-8.
For all participants after delivery, using sterile technique, placental samples were collected from the space between the amnion and chorion using a cotton-tipped swab by swabbing at the interface and placing the specimen in transport media.21,22 A sample of the placenta was collected for histologic examination. Placental swab samples were flash-frozen and maintained at −80°C until processed. The cryogenic vials containing swab samples obtained from the placenta for microbiologic culture were removed from the freezer and passed into an anaerobic chamber. Then, 1 mL of sterile phosphate-buffered saline was added to each swab sample and agitated on a vortex mixer for 1 minute. Serial dilutions of the sample were made in phosphate-buffered saline; aliquots of each dilution and the original sample were plated onto various selective and nonselective media. The culture medium for recovering anaerobes was prereduced Brucella-base agar with 5% sheep blood enriched with hemin and vitamin K1. Tryptic soy agar with 5% sheep blood was used for the recovery of aerobes and facultative anaerobes. Chocolate agar was used for the recovery of fastidious organisms (PML Microbiologicals). A-7 agar was used for the recovery of Ureaplasma and Mycoplasma (Northeast Laboratory). Prereduced Brucella-base agar with 5% sheep blood enriched with hemin and vitamin K1 and A-7 plates were incubated in an anaerobic chamber for a minimum of 120 hours at 35°C before enumeration. Tryptic soy agar with 5% sheep blood plates were incubated in air and chocolate agar plates in 5% carbon dioxide for 48 hours. After incubation, the various colony types were enumerated, isolated, and identified using established criteria. All estimates of population size were expressed as log10 colony-forming units per gram of sample.
For polymerase chain reaction (PCR) analysis, swab samples obtained from the placenta were transferred to a 2-mL Eppendorf Biopur tube and 1 mL of nuclease-free water was added. The tube was vortexed for 1 minute; the swab was removed and centrifuged at 7,500 rpm for 10 minutes. The supernatant was removed and DNA was extracted using the QIAamp DNA Mini Kit. A positive control and a reagent control were included with each run. After treating with PCR SuperMix and Taq DNA polymerase (Invitrogen) with DNase, PCR was performed using universal bacterial primers (10 pmol/microliter; Invitrogen) forward primer 5′-CCTACGGGAGGCAGCAGT-3′ and reverse primer 5′-ACGTCATCCCCACCTTCCT-3′. The reaction was allowed to run 60 cycles at 94°C for 30 seconds, 55°C for 30 seconds, and 72°C for 30 seconds (iCycler; BioRad). The presence of an 800-bp PCR product was considered a positive signal. To create these universal primers, we aligned the 16S rDNA of bacterial species from very different phyla. We then found forward and reverse primers 16S-F and 16S-R from regions of this gene that were conserved in all bacterial species. Before use for these studies, primers were tested with a variety of bacterial species to determine whether they were able to detect bacteria and Mycoplasma strains.
Infection was defined as 1,000 colony-forming units or greater of a single known pathogenic organism or at least a 2-log difference in the counts for a known pathogen compared with other organisms present in a mixed culture. Any signal on PCR was considered positive, although there were no PCR-positive/culture-negative infections identified.
The current report is an observational analysis of the one third of women whose biologic samples were randomly designated for analysis at enrollment (n=256). Demographic and clinical characteristics of this subsample were not significantly different from those of the overall enrolled population. For these analyses, we excluded women with delivery at less than 37 weeks of gestation (n=12), no trial of labor (n=6), a temperature more than 99.6°F within 1 hour of admission (n=3), delivery elsewhere (n=2), and absent recorded temperatures because of precipitous delivery (n=3). We also excluded 29 women with missing infection data, which occurred if a study representative was unavailable to collect samples. One additional woman was excluded because of an immunocompromised status. Our final study population consists of 200 women who had biologic data and at least one serum sample for cytokine analysis. Of these women, nine had infection documented by amnion–chorion culture. Analyses related to noninfectious fever were limited to the 191 women without infection. Of the 191 women, 163 had admission cytokines and 178 had delivery cytokines.
All analyses were performed using SAS 9.1. Characteristics and outcomes across groups were compared using χ2 for categorical variables. Continuous variables were compared using t tests for variables with normal distributions and the Wilcoxon rank-sum test for variables that were not normally distributed. Fever was defined as an oral temperature more than 100.4°F. Length of labor was defined as time from admission to delivery. To examine the association with outcomes, admission IL-6 levels were initially divided into terciles (less than 1 pg/mL, 1–11 pg/mL, and more than 11 pg/mL). Because the associations in the bottom two terciles were not different, they were combined, creating two categories (11 or less, more than 11).
A multiple logistic regression analysis performed to assess the predictors of noninfectious fever included factors for which the crude association with fever had P≤.1. Variables included in that model were epidural use, GBS status, length of labor (continuous), cm at admission (less than 3 cm, 3 cm or more), and IL-6 cytokine level at admission (11 pg/mL or less, more than 11 pg/mL). Of note, rupture of membranes was not included in the model because it strongly covaried with length of labor. Therefore, a separate model was performed when length of rupture of membranes was substituted for length of labor. Finally, a regression model that also included randomization group (doula, usual care), phase of the study (Project Assessing Childbirth Epidural/Labor Assistance and Birth Outcomes Research), and randomization site (Massachusetts General Hospital/Brigham and Women's Hospital) was performed; however, because in that model none of those factors significantly predicted fever and their inclusion did not substantively alter the associations undergoing study (data not shown), they were not included in the final model presented.
Overall, 18.5% (37/200) of women had a temperature more than 100.4°F develop during labor. However, only 4.5% (9/200) of women in this low-risk population had microbiologic-proven chorioamnionitis. Whereas women with epidural were far more likely to have intrapartum fever develop, 22.7% (34/150) with epidural compared with 6.0% (3/50) no epidural (P=.009), the use of epidural analgesia was not associated with infection. Infection was documented in 4.7% (7/150) of women receiving epidural and 4.0% (2/50) of women not receiving analgesia (P>.99).
In addition, the occurrence of infection was similar, regardless of the presence of fever. Among those with fever, 5.4% (2/37) were infected and among those without fever 4.3% (7/163) were infected (P=.7). Among women receiving epidural, the proportion of febrile women with infection was 5.9% (2/34) compared with 4.3% (5/116) without fever (P=.7). Among those not receiving an epidural, the rate of infection also was not different for women with fever 0% (0/3) and those without fever 4.3% (2/47; P>.99). However, interpretation is complicated by the small number of febrile women in this group.
We investigated whether antibiotics for GBS prophylaxis or for presumed chorioamnionitis may have masked microbiologic infection. We focused on antibiotic use among the 150 women who received epidural because there were few fevers in women without analgesia. Two of the seven women in the epidural group with infection were febrile and both received combination antibiotics in labor. Of the five afebrile women with infection, two received GBS prophylaxis and none received combination antibiotics. Organisms recovered included group B Streptococcus, Staphylococcus species, ureaplasma, and Propionibacterium. All of these organisms are susceptible to the antibiotics used.
Given that most fevers were not associated with infection, we investigated the physiologic correlates of noninfectious fever, specifically IL-6 and IL-8 levels at admission and delivery once women with positive cultures had been removed from the cohort. Among the 163 women with admission IL-6 levels, the median values did not differ between those later receiving and not receiving epidural (3.2 pg/mL epidural, 1.6 pg/mL no epidural; P=.2). Although not associated with later epidural use, higher IL-6 at admission did predict fever. Among women who received an epidural, 36.4% (16/44) with admission IL-6 levels greater than 11 pg/mL had fever develop as compared with only 15.7% (13/83) with lower IL-6 levels (P=.008, relative risk, 2.3, confidence interval [CI] 1.2–4.4). Among the 36 women not receiving epidural, the proportion of women with fever developing was also greater with admission IL-6 levels more than 11 pg/mL (18.2% [2/11] compared with 4% [1/25]) than with IL-6 levels 11 or less, but the difference did not reach statistical significance (P=.2). Women with fever developing, both within the epidural and no epidural groups, also had higher median IL-6 levels at admission (Table 1).
Median admission levels of IL-8 were the same for women who later received and who did not receive epidural analgesia (1.5 pg/mL epidural compared with 1.4 pg/mL no epidural; P=.7). Unlike IL-6, median admission IL-8 levels were similar among women who did and did not develop fever, both in the epidural and no epidural groups (Table 1).
At delivery, median IL-6 levels were significantly higher among women who had fever develop (384.7 pg/mL fever, 179.1 pg/mL afebrile; P<.001), and these differences were present among women receiving and not receiving epidural. Median delivery IL-8 levels were also significantly higher in women who had fever develop (5.1 pg/mL fever, 3.6 pg/mL afebrile; P=.01), but the difference reached statistical significance only in the epidural group (Table 1).
We also examined the association of epidural with IL-6 and IL-8 levels separately for women with and without fever. Among febrile women, there was no difference between those receiving or not receiving epidural at median admission IL-6 (17.8 pg/mL epidural, 21.6 pg/mL no epidural; P>.99) or IL-8 (1.7 pg/mL epidural, 1.0 pg/mL no epidural; P=.6). In the febrile group, epidural also was not associated with a difference in median delivery IL-6 (386.2 pg/mL epidural, 333.5 pg/mL no epidural; P=.4) or IL-8 (5.1 pg/mL epidural, 4.2 pg/mL no epidural; P=.3).
In contrast, among women without fever, median cytokine levels at delivery differed between those receiving and not receiving epidural. Among these women, there was no difference in median admission levels of IL-6 (2.4 pg/mL epidural, 1.4 pg/mL no epidural; P=.3) or IL-8 (1.4 pg/mL epidural, 1.6 pg/mL no epidural; P>.99). At delivery, however, afebrile women with epidural had higher median levels of both IL-6 (228.4 pg/mL epidural, 107.6 pg/mL, no epidural; P<.001) and IL-8 (epidural 4.3 pg/mL, no epidural 2.5 pg/mL; P<.001).
Finally, we examined the factors associated with noninfectious fever, including whether epidural remains a significant predictor of fever when accounting for potentially confounding factors. We found no significant differences in characteristics of women receiving and not receiving epidural, including cervical dilatation and IL-6 levels at admission, which might indicate an increased risk of intrapartum fever among women receiving epidural (Table 2). Women with fever developing were more likely to have rupture of membranes for more than 12 hours, to be GBS-positive, and to have an admission IL-6 level more than 11 pg/mL (Table 3).
We constructed a logistic regression model to evaluate predictors of noninfectious fever (Table 4). In that model, epidural was associated with a five-fold increase in the risk of fever (odds ratio [OR] 4.9, 95% CI 1.1–22.4). In addition, GBS carriage (OR 4.0, 95% CI 1.6–10.1), hours of labor (OR 1.1, CI 1.0–1.2), and a high admission IL-6 level (OR 8.9, 95% CI 3.1–25.8) remained significant predictors of fever. Substituting length of rupture of membranes in the model for length of labor did not substantially alter the associations noted.
Our study supports the hypothesis that most epidural-related fever represents noninfectious systemic inflammation. Others have suggested this fever is attributable to infection, asserting that women who choose epidural techniques are at greater risk for obstructed labors, increasing their risk of infection.7 However, this explanation was based on histologic findings alone without culture data.15,16 We found no increase in positive cultures among women who received epidural analgesia despite searching for evidence of infection using cultures and PCR. The prevalence of infection was the same in women with and without epidural; however, the rate of fever was markedly higher. In addition, placental examinations performed in our study population found histologic acute chorioamnionitis occurred more frequently than microbiologic infection and was present much more frequently in women with fever (70.6% fever, 27.2% no fever; P<.001).
Our overall incidence of infection is lower than that reported in other studies of laboring women at term.21,22 It is possible that our exclusion criteria defining the lowest-risk population may account for this difference. We collected samples utilizing previously validated amnion–chorion techniques and attempted to increase our recovery of even dead organisms utilizing PCR techniques. Onderdonk et al23 has utilized these same techniques in preterm labor populations also exposed to intrapartum antibiotics and found higher infection rates.
The association of elevated maternal IL-6 with histologic chorioamnionitis and neuraxial anesthesia has been previously described.17–19 One investigator found that prophylactic treatment with steroids before epidural placement decreased the rate of fever by 90% but increased asymptomatic neonatal bacteremia.24 We demonstrate that women who were admitted with high IL-6 levels were more than twice as likely to have fever develop after epidural as women with lower admission IL-6 levels. Women with high admission IL-6 levels not receiving epidural were also more likely to have fever develop, although the difference did not reach statistical significance. Given the small numbers of women not receiving epidural, our study only had 37% power to detect a difference of this magnitude. Although our study cannot determine whether cytokine activation is involved in the genesis of epidural analgesia-related fever, these findings suggest that women who have an “activated” immune system on admission may react differently to epidural analgesia, leading to fever.
Although maternal IL-6 levels on admission are associated with later development of fever, admission IL-8 levels are not. This may occur because IL-6 is the potent inflammatory cytokine in this process that stimulates production of IL-8, a chemotactic cytokine known to be a neutrophil attractant and activator in the placenta, accounting for the acute histologic chorioamnionitis observed.25,26 Multiple studies suggest a complex environment in which cytokine expression may differ depending on the site of origin and the population studied.27–30 We theorize that the compartment measured and the time in labor influences the measurement of IL-8.
Among women without fever, postpartum levels of IL-6 and IL-8 were higher among those who received an epidural. This finding provides additional support for the possibility that epidural analgesia plays a role in cytokine activation. This suggests a model in which the risk for intrapartum fever based on the woman's cytokine profile at admission is further stimulated by epidural, increasing the risk for fever in labor. However, we are unable to differentiate precisely which cytokine is the stimulant or whether there are other cytokines that catalyze fever in the presence of epidural analgesia.
Interestingly, only two of nine infected women had fever in labor. In our low-risk population, fever was an unreliable indicator of culture-proven infection, which is concerning because fever has been a principal sign of clinical chorioamnionitis.31,32 In this setting, many positive cultures were not predicted and antibiotics were not administered. This lack of association between fever and culture-proven infection warrants mention and more investigation.
The strength of our study was systematic microbiologic investigation using chorion–amnion culture, but PCR samples were not useful. Studies conducted after our protocol began suggest that placental constituents may limit the utility of PCR in this setting.23 Obtaining specimens from another site such as amniotic fluid may have yielded more positive cultures; however, there is no indication that this detection of organisms would have been different in those with and without epidural. Furthermore, we utilized a collection method that could be accomplished in most clinical settings of women in active labor. Serum cytokine levels varied according to the point in labor at which women were admitted to the hospital. However, because women receiving and not receiving an epidural did not differ in dilatation at admission, it is unlikely that this variation is responsible for our findings. Finally, given the high rate of epidural use and the low rate of fever without epidural, the number of febrile women without epidural was small, limiting our ability to evaluate associations in this group.
In summary, our study suggests epidural analgesia-related fever results from a noninfectious inflammatory process, and women with higher IL-6 levels on admission in labor are at increased risk for developing fever. Areas for future investigation include understanding the determinants of elevated IL-6 levels or “activated” immune system on admission. The role of GBS colonization in this immune activation also deserves further study. Determining the roles of serum IL-6 and IL-8 in the placenta may provide additional information about this systemic inflammatory process associated with epidural-related fever. Investigations of a larger cohort of women with epidural analgesia are needed to determine whether admission IL-6 levels can predict those at greatest risk for fever. Likewise, as our understanding of cytokine activation and inflammation increases, we may be able to investigate the efficacy of mediators of inflammation as a method to prevent intrapartum fever.
1. Fusi L, Steer PJ, Maresh MJA, Richard BW. Maternal pyrexia associated with the use of epidural analgesia in labour. Lancet 1989;1:1250–2.
2. Camann WR, Hortvet LA, Hughes N, Bader AM, Datta S. Maternal temperature regulation during extradural analgesia for labour. Br J Anaesth 1991;67:565–8.
3. Vinson DC, Thomas R, Kiser T. Association between epidural analgesia during labor and fever. J Fam Pract 1993;36:617–22.
4. Macaulay JH, Bond K, Steer PJ. Epidural analgesia in labor and fetal hyperthermia. Obstet Gynecol 1992;80:665–9.
5. Lieberman E, Lang JM, Frigoletto F Jr, Richardson DK, Ringer SA, Cohen A. Epidural analgesia, intrapartum fever, and neonatal sepsis evaluation. Pediatrics 1997;99:415–9.
6. Mayer DC, Chescheir NC, Spielman FJ. Increased intrapartum antibiotic administration associated with epidural analgesia in labor. Am J Perinatol 1997;14:83–6.
7. Philip J, Alexander JM, Sharma SK, Leveno KJ, McIntire DD, Wiley J. Epidural analgesia during labor and maternal fever. Anesthesiology 1999;90:1271–5.
8. Gonen R, Korobochka R, Degani S, Gaitini L. Association between epidural analgesia and intrapartum fever. Am J Perinatol 2000;17:127–30.
9. Negishi C, Lenhardt R, Ozaki M, Ettinger K, Bastanmehr H, Bjorksten AR, et al. Opioids inhibit febrile responses in humans, whereas epidural analgesia does not: an explanation for hyperthermia during epidural analgesia. Anesthesiology 2001;94:218–22.
10. Yancey MK, Zhang J, Schwarz J, Dietrich CS III, Klebanoff M. Labor epidural analgesia and intrapartum maternal hyperthermia. Obstet Gynecol 2001;98:763–70.
11. Sharma SK, Alexander JM, Messick G, Bloom SL, McIntire DD, Wiley J, et al. Cesarean delivery: a randomized trial of epidural analgesia versus intravenous meperidine analgesia during labor in nulliparous women. Anesthesiology 2002;96:546–51.
12. Mantha VRR, Vallejo MC, Ramesh V, Phelps AL, Ramanathan S. The incidence of maternal fever during labor is less with intermittent than with continuous epidural analgesia: a randomized controlled trial. Inst J Obstet Anesth 2008;17:123–9.
13. Goetzl L, Rivers J, Zighelboim I, Wali A, Badell M, Suresh MS. Intrapartum epidural analgesia and maternal temperature regulation. Obstet Gynecol 2007;109:687–90.
14. Hawkins JL. Epidural analgesia for labor and delivery. N Engl J Med 2010;362:1503–10.
15. Dashe JS, Rogers BB, McIntire DD, Leveno KJ. Epidural analgesia and intrapartum fever: placental findings. Obstet Gynecol 1999;93:341–4.
16. Vallejo MC, Kaul B, Adler LJ, Phelps AL, Craven CM, Macpherson TA, et al. Chorioamnionitis, not epidural analgesia, is associated with maternal fever during labour. Can J Anesth 2001;48:1122–6.
17. De Jongh RF, Bosmans EP, Puylaert MJ, Ombelet WU, Vandeput HJ, Berghmans RA. The influence of anaesthetic techniques and type of delivery on peripartum serum interleukin-6 concentrations. Acta Anaesthesiol Scand 1997;41:853–60.
18. Goetzl L, Evans T, Rivers J, Suresh MS, Lieberman E. Elevated maternal and fetal serum interleukin-6 levels are associated with epidural fever. Am J Obstet Gynecol 2002;187:834–8.
19. Smulian JC, Bhandari V, Vintzileos AM, Shen-Schwarz S, Quashie C, Lai-Lin YL, et al. Intrapartum fever at term: serum and histologic markers of inflammation. Am J Obstet Gynecol 2003;188:269–74.
20. Evron S, Parameswaran R, Zipori D, Ezri T, Sadon O, Koren R. Activin betaA in term placenta and its correlation with placental inflammation in parturients having epidural or systemic meperidine analgesia: a randomized study. J Clin Anesth 2007;19:168–74.
21. Hillier SL, Martius J, Krohn M, Kiviat N, Holmes KK, Eschenbach DA. A case-control study of chorioamniotic infection and histologic chorioamnionitis in prematurity. N Engl J Med 1988;319:972–8.
22. Dong Y, St. Clair PJ, Ramzy I, Kagan-Hallet KS, Gibbs RS. A microbiologic and clinical study of placental inflammation at term. Obstet Gynecol 1987;70:175–82.
23. Onderdonk AB, Delaney ML, DuBois AM. Allred EN, Leviton A. Detection of bacteria in placental tissues obtained from extremely low gestational age neonates. Am J Obstet Gynecol 2008;198:110e1–7.
24. Goetzl L, Zighelboim I, Badell M, Rivers J, Mastrangèlo MA, Tweardy D, et al. Maternal corticosteroids to prevent intrauterine exposure to hyperthermia and inflammation: a randomized, double-blind, placebo-controlled trial. Am J Obstet Gynecol 2006;195:1031–7.
25. Trautman MS, Dudley DJ, Edwin SS, Collmer D, Mitchell MD. Amnion cell biosynthesis of interleukin-8: regulation by inflammatory cytokines. J Cell Physiol 1992;153:38–43.
26. Laham N, Brennecke SP, Rice GE. Interleukin-8 release from human gestational tissue explants: the effects of lipopolysaccharide and cytokines. Biol Reprod 1997;57:616–20.
27. Elliott CL, Kelly RW, Critchley HO, Riley SC, Calder AA. Regulation of interleukin 8 production in the term human placenta during labor and by antigestagens. Am J Obstet Gynecol 1998;179:215–20.
28. Elliott CL, Loudon JA, Brown N, Slater DM, Bennett PR, Sullivan MH. IL-1beta and IL-8 in human fetal membranes: changes with gestational age, labor, and culture conditions. Am J Reprod Immunol 2001;46:260–7.
29. Maul H, Nagel S, Welsch G, Schäfer A, Winkler M, Rath W. Messenger ribonucleic acid levels of interleukin-1 beta, interleukin-6 and interleukin-8 in the lower uterine segment increased significantly at final cervical dilation during term parturition, while those of tumor necrosis factor alpha remained unchanged. Eur J Obstet Gynecol Reprod Biol 2002;102:143–7.
30. Keelan JA, Marvin KW, Sato TA, Coleman M, McCowan LM, Mitchell MD. Cytokine abundance in placental tissues: evidence of inflammatory activation in gestational membranes with term and preterm parturition. Am J Obstet Gynecol 1999;181:1530–6.
31. Gibbs RS. Diagnosis of intra-amniotic infection. Semin Perinatol 1977;1:71–7.
32. Sperling RS, Ramamurthy RS, Gibbs RS. A comparison of intrapartum compared with immediate postpartum treatment intra-amniotic infection. Obstet Gynecol 1987;7:861–5.
© 2011 The American College of Obstetricians and Gynecologists
Figure. No caption available.