Epidural analgesia has emerged as a popular method of pain relief during labor during the latter half of the 20th century. Although epidural analgesia provides very effective pain relief during labor, its use may be associated with development of maternal fever (≥38°C) in >20% of women.1 Consequences of maternal fever may include increased neonatal evaluations for sepsis, increased use of antibiotics, and prolonged hospital stay.2–4 The precise mechanism of the development of maternal fever during epidural analgesia is unclear.5 However, possible mechanisms include (1) increased metabolic expenditure during labor, (2) decrease in evaporative heat loss through hyperventilation as a result of pain relief, (3) thermoregulatory changes induced by epidural analgesia through peripheral sensory (autonomic) blockade, (4) direct central nervous system effects of local anesthetic drugs due to the release of cytokines,6–8 (5) sweat gland paralysis resulting from neural blockade decreasing sweating and evaporative heat loss,9 (6) noninfectious systemic inflammatory process,10 and (7) intrapartum infection.11
Women who choose epidural analgesia may have prolonged and dysfunctional labor and hence may inherently be predisposed to infection of the amniotic cavity and fetus.12–15 Usual organisms isolated in the amniotic fluid of patients with chorioamnionitis include a mixture of aerobes and anaerobes. We hypothesized that if intrapartum infection is one of the causes of maternal fever during epidural analgesia, then prophylactic administration of antibiotics might reduce the occurrence of such infection-related fever attributed to epidural analgesia. This randomized, double-blind trial in low risk women at term was performed to determine whether prophylactic antibiotic administration before the initiation of epidural analgesia decreases the rate of epidural analgesia-related fever.
This randomized, double-blind, placebo-controlled study was developed by investigators from the Departments of Anesthesiology, Obstetrics and Gynecology, and Pathology at the University of Texas Southwestern Medical Center at Dallas. This study protocol was approved by the IRB at the University of Texas Southwestern Medical Center at Dallas, and was conducted between June 2002 and December 2005. This trial commenced and was completed before registration was required with ClinicalTrials.gov.
Healthy nulliparous women presenting to Parkland Hospital with singleton cephalic gestations at term and in spontaneous labor (cervical dilation 4–6 cm with uterine contractions) and requesting epidural analgesia were offered participation in this study. Women giving written informed consent were randomly assigned to receive either cefoxitin 2 g or an identical appearing placebo (normal saline) immediately preceding placement of an epidural catheter. The randomization sequence was computer-derived in blocks of 20 subjects and placed in sealed opaque envelopes. The envelope was opened at the time of request of epidural analgesia immediately before antibiotic administration. The study drug was disbursed by the Parkland Hospital Investigational Drug Service. A follow-up dose of study drug was given at 6 hours after the first dose in undelivered women. Women who developed an intrapartum fever ≥38°C were treated with IV ampicillin 2 g every 6 hours plus IV gentamicin (1. 5 mg/kg) every 8 hours. This therapy was continued after delivery until the patient was afebrile for 24 hours. Cefoxitin has been safely used for prophylaxis against infection with gram-positive and gram-negative organisms in patients undergoing gastrointestinal surgery, vaginal hysterectomy, abdominal hysterectomy, and cesarean delivery.16 Epidural analgesia was administered using a standardized method. A 500-mL bolus of lactated Ringer’s solution was administered, followed by placement of an epidural catheter into the lumbar epidural space through a 17-gauge Tuohy needle. After a test dose of 3 mL of 1.5% lidocaine with epinephrine 1:200,000, analgesia was achieved with 3-mL increments of 0.25% bupivacaine, titrated to a bilateral T-10 sensory level. A continuous epidural infusion of 0.125% bupivacaine with fentanyl 2 µg/mL was infused at 8 to 10 mL/h to maintain a T-10 sensory level.
Our labor management approach encourages amniotomy in active labor when the fetal head is applied to the cervix. Pelvic examinations were performed approximately every 2 hours to evaluate the progress of labor. Oxytocin augmentation of labor was begun when the rate of cervical dilation was <1 cm/h, and a hypotonic contraction pattern was diagnosed using internal electronic fetal monitoring. Maternal tympanic temperature was measured by midwives or research nurses hourly using a Genius® thermometer (Sherwood Medical, St. Louis, MO). A study comparing various temperature measurement sites during cardiac surgery concluded that the tympanic temperature using Genius® tympanic thermometer reflects rapid changes in body temperature better than axillary or rectal temperature.17 Intrapartum fever was defined as a maternal temperature of 38°C or higher. Neonates born to women with fever during labor were commonly evaluated for possible sepsis, which included a complete blood count and blood cultures, after which they were given antibiotics for 48 hours. Indications for a sepsis evaluation in neonates born to women without fever in labor included temperature instability, tachypnea, dusky spells, lethargy, or hypoglycemia. Neonates with suspected sepsis were empirically treated with antibiotics for 48 hours. This was the standard of practice during the interval of the study and was not part of the trial protocol.
Providers were asked to submit the placentas for histologic examination. Placentas were labeled with a study number at delivery and refrigerated immediately. A single senior pathologist blinded to study drug evaluated the placentas within 48 hours of delivery. The placenta was examined according to routine protocol. The following sections were submitted for each placenta: 2 sections of umbilical cord, 2 full-thickness sections of placental parenchyma, and a section of a membrane roll that included the point of membrane tear or “rupture.” All sections were evaluated for the presence and severity of neutrophilic infiltrates in the extraplacental membranes or chorionic plate (acute chorioamnionitis). There are many ways to define the severity of acute chorioamnionitis. We graded acute chorioamnionitis using a simple scheme previously reported.18,19 This scheme uses the intensity of the neutrophilic infiltrate in the area of greatest density on the free membranes or chorionic plate and categorizes the inflammation as mild, moderate, or severe. We also included “necrotizing chorioamnionitis” as a separate category, defined as inflammation with necrosis of the amnion and/or chorion.19 We modified the designation of severity of inflammation of umbilical cord by reporting funisitis as neutrophilic inflammation involving Wharton’s jelly and vasculitis by assigning categories based on the number of vessels involved by the inflammation. Similar to the extraplacental membranes, necrotizing funisitis required the presence of necrotizing neutrophilic inflammation involving Wharton’s jelly.
Sample size was based on the incidence of intrapartum fever in a prior study of similar women conducted at our hospital.1 We reported that 23% to 33% of women randomized to epidural analgesia developed fever during labor. An analysis based on a projected absolute reduction of fever with antibiotic treatment from 20% to 10% required 200 women enrolled in each study arm. This analysis used 80% power to detect a 2-tailed significance level α = 0.05.
Statistical analysis was performed using SAS version 9.2 (SAS Institute, Cary, NC). Data were analyzed on an intention-to-treat principle. Further analysis was performed according to the intended randomization for those deliveries in which placentas were available. We also compared histologic findings in the placentas and neonatal outcome in women who developed fever versus those who did not. Comparisons using Student t test or Wilcoxon rank sum test were assigned a priori under assumptions of statistical normality or not. Age, height, weight, birth weight, and temperature were assumed to satisfy statistical normality from prior experience as to the symmetry of their frequency distributions. Under the central limit theorem for sufficiently large sample sizes, such data satisfy the normality assumption for Student t test. Data such as duration of labor, duration of analgesia, length of membrane rupture, and cervical dilation at epidural placement which experience high degrees of asymmetry or positive skewness due to a zero lower bound and extreme possibilities were examined using the Wilcoxon rank sum test.
We further analyzed the fever outcome data for those with placentas available to determine the association between randomization arm and fever stratified by placental neutrophilic infiltration as well as the association between infiltration and fever stratified by randomization. This was accomplished using the Breslow-Day statistic to evaluate the difference in odds ratios between strata (interaction) and the Cochran-Mantel-Haenszel test for the association adjusting for the strata.
Risk differences are provided as the difference in the observed percentage and the corresponding 95% CIs of the difference, based on the normal approximation of the binomial distribution. Risk differences and CIs were estimated using a frequency procedure from SAS version 9.2 (PROC FREQ). A hyperbolic growth model of temperature versus hours is estimated with random effects models (hours being a random effect) for each of the 2 strata of patients, those who received cefoxitin and those who received placebo treatment. The treatment assignment was a fixed effect.
Four-hundred women were randomized in this study (Fig. 1). Maternal demographic characteristics are shown in Table 1. There were no significant differences between the study groups with the exception that there were significantly more African American women in the placebo than the cefoxitin arm.
A variety of labor characteristics were analyzed in relation to antibiotic or placebo study groups, (Table 2) and none was significantly different. The number of vaginal examinations during the epidural analgesia period was similar in both groups. Cesarean delivery was performed in 30 (15%) women randomized to cefoxitin and 31(15%) randomized to placebo (P = 0.89).
Approximately 40% of women in each study group developed fever ≥38°C during labor. The risk difference (95% confidence interval [CI]) for fever ≥38°C during labor (antibiotic versus placebo) was −2.0% (−11.5 to 7.5), and for fever >39°C during labor was −1.5% (−4.7 to 1.7). Figure 2 shows a hyperbolic growth model of epidural analgesia duration in hours versus tympanic temperature.
Histologic findings in the fetal membranes and placentas are summarized in Table 3 and 4. Placentas were available from 302 of the study cohort (Fig. 1). Administration of cefoxitin had no significant effect on any grade of neutrophilic infiltration. As shown in Figure 3, fever in women with neutrophilic inflammation was not reduced significantly by maternal cefoxitin prophylaxis (45% for cefoxitin and 48% for placebo, P = 0.703; risk difference 3% [95% CI, −13% to 19%]). Fever developed significantly more often in women with placental neutrophilic inflammation compared with those without such inflammation (73/158 (46%) vs 33/144 (23%), P < 0.001; risk difference 23.0% (95% CI, 13.0–34.0).
Infant outcomes in relation to maternal antibiotic treatment are summarized in Table 5. There were no significant differences in any outcomes between the antibiotic and placebo study groups. Sepsis, defined as a positive blood culture, was not diagnosed in any of the infants, and there were no neonatal deaths. Infant outcomes in relation to development of fever in the study cohort are summarized in Table 6. The percentage of infants with 1-minute Apgar scores ≤7 was significantly higher in the fever group compared with no fever group (risk difference 5.8% [95% CI, 0.8–10.9]) (Table 6). There were no significant differences in the percentage of infants with Apgar scores ≤7 at 5 minutes (risk difference 0.9% [95% CI, −1.1 to 2.9]). One infant in the febrile group was admitted to the intensive care unit due to suspected sepsis and low 1- and 5-minute Apgar scores with prolonged labor. There were no other significant differences in the infant outcome in the fever and no fever groups.
We further evaluated temperature trends after the first indication of fever in all 154 women who became febrile (≥38°C) during labor. In 33 (17%) women in the cefoxitin group and 38 (19%) women in the placebo group (P = 0.61), subsequent temperature readings after the first observed febrile reading continued to be ≥38°C throughout labor. Thirty-one (16%) women in the cefoxitin group and 31 (16%) women in the placebo group (P = 0.79) developed fever immediately preceding delivery. Finally, 11 (6%) women in the cefoxitin group and 10 (5%) women in the placebo group (P = 0.72) had a temperature range of 37.5°C to 37.9°C after at least 1 febrile readings of ≥38°C.
Antibiotic prophylaxis with cefoxitin did not significantly reduce the rate of fever defined as temperature ≥38°C during labor in women given epidural analgesia. This was also true for fever >38.5°C. However, based on the 95% CIs between groups, if antibiotic prophylaxis made a difference in the rate of fever, it was a relatively small difference. Cefoxitin administration had no effect on any neonatal outcome including work-ups for sepsis. Neutrophilic infiltration of the placenta and membranes was found in approximately 50% of both study groups. Such infiltration was associated with maternal fever, but cefoxitin administration did not reduce the fever rate in women with placental infiltration. These findings suggest that maternal fever in women receiving labor epidural analgesia probably cannot be attributed to maternal infection. Such fever was not shown to significantly impact either maternal or neonatal outcome.
Riley et al.10 in a recent observational study of labor epidural analgesia-related fever did not find evidence of a higher rate of infection using chorion–amnion culture among women who received epidural analgesia. In their study, despite the higher rate of fever in the epidural group (23% vs 6%), the prevalence of infection was the same in women with and without epidural analgesia (5% vs 4%). Also, the infection rate was similar in febrile (5.4%) and afebrile (4.3%) women. However, they found a higher rate of histologic acute chorioamnionitis in women with fever compared with women without fever.10 The authors concluded that most epidural-related fever is associated with a noninfectious inflammation of the placenta. This result is similar to our study. However, we did not attempt to culture the placentas in our study. The incidence of fever was somewhat higher in this study than was reported in our previous studies and other studies.1 This difference could be attributed to dysfunctional labor in some subjects. The Genius thermometer used in this study is better than other infrared systems.17 However, considerable variability from reading to reading has been reported with infrared tympanic membrane thermometry, especially when used by unskilled practitioners.20 The temperature in this study was recorded at a regular interval during labor by trained investigators, and Figure 2 shows a time-dependent increase in aural canal temperature in the study groups. Similar devices to measure aural canal temperature have also been used in several noticeable studies evaluating fever during labor epidural analgesia.21,22 Tympanic membrane thermometry eliminates the effects of oral intake and mouth breathing. However, we consider the use of infrared tympanic membrane thermometry in this study is a limitation of this study because significant variability in temperature measurement can occur even with skilled practitioners. Despite this limitation, the subsequent observation of fever after the first indication of fever (≥38°C) in women who became febrile during labor was not significantly different in the cefoxitin and placebo groups. Moreover, there were no significant differences in infant outcomes and histologic findings in the placentas in either group, which further supports the validity of temperature measurements in this study.
Some studies have found a strong association between maternal serum interleukin-6 levels and the development of fever during epidural analgesia.6–8 It has been suggested that the maternal inflammatory response is a primary source for the development of fever.7,8 Riley et al.10 reported that women with high initial maternal serum interleukin-6 levels were twice as likely to have fever after epidural analgesia compared with women with lower levels. They hypothesized that women who have an activated cytokine system on admission may react differently to labor epidural analgesia and further activate the cytokines leading to fever.10
Goetzl et al.21 administered steroid prophylaxis before epidural placement to reduce placental inflammation and thereby the rate of fever. They found that prophylactic treatment with steroids decreased the rate of fever by 90% but also increased neonatal bacteremia. Wang et al.23 in their recent trial used epidural dexamethasone infusion with epidural analgesia. They concluded that epidural dexamethasone alleviates maternal temperature increase after epidural analgesia. In another placebo-controlled study, prophylactic use of acetaminophen during labor epidural analgesia did not prevent maternal fever.22 It has also been hypothesized that an increased incidence of fever in women receiving epidural analgesia is due to the antipyretic effect of opioids in the nonepidural group. However, Gross et al.24 did not find a significant difference in the rate of fever in women whose labor analgesia was managed with opioids versus no opioids.
Epidural-related fever remains a significant problem because these fevers inevitably result in neonatal work-ups for infection. In our study although sepsis was suspected in neonates born to women with fever, sepsis was not diagnosed in any of the infants. Mecredy et al.25 in their literature review determined that 95% of asymptomatic, term neonates born to women with fever in labor have a benign course. They suggested that aggressive evaluation and management of these infants are unnecessary.
It is interesting to note that evaluation of neonates for possible sepsis has been reported with epidural analgesia in the absence of maternal fever. Lieberman et al.26 found a 3-fold increase in the rate of sepsis evaluations in infants born to afebrile women who had received epidural analgesia during labor. They concluded that the use of epidural analgesia during labor is strongly associated with neonatal sepsis evaluations. In contrast, Philip et al.13 reported that in the absence of maternal fever, epidural analgesia during labor has no bearing on the need for such neonatal management and therefore should not be considered a predictor per se for neonatal sepsis evaluations. These authors also concluded that association between epidural analgesia and maternal fever becomes apparent only in a small subset of women who are usually nulliparous and who experience prolonged labor.13
We were surprised by the results of our current study. The study was conceived with the expectation that fever associated with labor epidural analgesia was largely attributable to maternal bacterial infection and could therefore be prevented with prophylactic antibiotics. Clearly this was not the case in this trial. We thus conclude that epidural analgesia during labor is associated with fever which does not result from bacterial infection.
Name: Shiv K. Sharma, MD, FRCA.
Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.
Attestation: Shiv K. Sharma has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Beverly B. Rogers, MD.
Contribution: This author helped design and conduct the study and write the manuscript.
Attestation: Beverly B. Rogers has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: James M. Alexander, MD.
Contribution: This author helped design and conduct the study.
Attestation: James M. Alexander has seen the original study data and approved the final manuscript.
Name: Donald D. McIntire, PhD.
Contribution: This author helped design the study, analyze the data, and write the manuscript.
Attestation: Donald D. McIntire has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Kenneth J. Leveno, MD.
Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.
Attestation: Kenneth J. Leveno has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.
This manuscript was handled by: Cynthia A. Wong, MD.
1. Sharma SK, McIntire DD, Wile J, Leveno KJ. Labor analgesia and cesarean delivery: An individual patient meta-analysis of nulliparous women. Anesthesiology. 2004;100:142–8
2. Lieberman E, Lang J, Richardson DK, Frigoletto FD, Heffner LJ, Cohen A. Intrapartum maternal fever and neonatal outcome. Pediatrics. 2000;105:8–13
3. Goetzl L, Cohen A, Frigoletto F Jr, Ringer SA, Lang JM, Lieberman E. Maternal epidural use and neonatal sepsis evaluation in afebrile mothers. Pediatrics. 2001;108:1099–102
4. Goetzl L, Cohen A, Frigoletto F Jr, Lang JM, Lieberman E. Maternal epidural analgesia and rates of maternal antibiotic treatment in a low-risk nulliparous population. J Perinatol. 2003;23:457–61
5. Scott Segal. Labor epidural analgesia and maternal Fever. Anesth Analg. 2010;111:1467–75
6. DeJongh RF, Bosmans EP, Puylaert MJ, Ombelet WU, Vandeput HJ, Berghmans RA, M Maes, RJ Heylen. The influence of anesthetic techniques and type of delivery on peripartum serum interleukin-6 concentrations. Acta Anaesthesiol Scand. 1997;41:853–60
7. Smulian JC, Bhandari V, Vintzileos AM, Shen-Schwarz S, Quashie C, Lai-Lin YL, Ananth CV. Intrapartum fever at term: serum and histologic markers of inflammation. Am J Obstet Gynecol. 2003;188:269–74
8. 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
9. 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
10. Riley LE, Celi AC, Onderdonk AB, Roberts DJ, Johnson LC, Tsen LC, Leffert L, Pian-Smith MC, Heffner LJ, Haas ST, Lieberman ES. Association of epidural-related fever and noninfectious inflammation in term labor. Obstet Gynecol. 2011;117:588–95
11. Newton ER. Epidural analgesia, intrapartum fever, and neonatal outcomes. Birth. 2000;27:206–8
12. Dashe JS, Rogers BB, McIntire DD, Leveno KJ. Epidural analgesia and intrapartum fever: placental findings. Obstet Gynecol. 1999;93:341–4
13. Philip J, Alexander JM, Sharma SK, Leveno KJ, McIntire DD, Wiley J. Epidural analgesia during labor and maternal fever. Anesthesiology. 1999;90:1271–5
14. Camann W. Intrapartum epidural analgesia and neonatal sepsis evaluations. Anesthesiology. 1999;90:1250–2
15. Vallejo MC, Kaul B, Adler LJ, Phelps AL, Craven CM, Macpherson TA, Sweet RL, Ramanathan S. Chorioamnionitis, not epidural analgesia, is associated with maternal fever during labour. Can J Anaesth. 2001;48:1122–6
16. Counts GW. Cefoxitin: its role in treatment and prophylaxis of obstetric and gynecologic infections. Rev Infect Dis. 1988;10:76–91
17. Robinson J, Charlton J, Seal R, Spady D, Joffres MR. Oesophageal, rectal, axillary, tympanic and pulmonary artery temperatures during cardiac surgery. Can J Anaesth. 1998;45:317–23
18. Dexter SC, Pinar H, Malee MP, Hogan J, Carpenter MW, Vohr BR. Outcome of very low birth weight infants with histopathologic chorioamnionitis. Obstet Gynecol. 2000;96:172–7
19. Rogers BB, Alexander JM, Head J, McIntire D, Leveno KJ. Umbilical vein interleukin-6 levels correlate with the severity of placental inflammation and gestational age. Hum Pathol. 2002;33:335–40
20. Imamura M, Matsukawa T, Ozaki M, Sessler DI, Nishiyama T, Kumazawa T. The accuracy and precision of four infrared aural canal thermometers during cardiac surgery. Acta Anaesthesiol Scand. 1998;42:1222–6
21. Goetzl L, Zighelboim I, Badell M, Rivers J, Mastrangèlo MA, Tweardy D, Suresh MS. 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
22. Goetzl L, Rivers J, Evans T, Citron DR, Richardson BE, Lieberman E, Suresh MS. Prophylactic acetaminophen does not prevent epidural fever in nulliparous women: a double-blind placebo-controlled trial. J Perinatol. 2004;24:471–5
23. Wang LZ, Hu XX, Liu X, Qian P, Ge JM, Tang BL. Influence of epidural dexamethasone on maternal temperature and serum cytokine concentration after labor epidural analgesia. Int J Gynaecol Obstet. 2011;113:40–3
24. Gross JB, Cohen AP, Lang JM, Frigoletto FD, Lieberman ES. Differences in systemic opioid use do not explain increased fever incidence in parturients receiving epidural analgesia. Anesthesiology. 2002;97:157–61
25. Mecredy RL, Wiswell TE, Hume RF. Outcome of term gestation neonates whose mothers received intrapartum antibiotics for suspected chorioamnionitis. Am J Perinatol. 1993;10:365–8
© 2014 International Anesthesia Research Society
26. 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