Bartholomew, M. Lisa MD; Ashkin, Eric MD, MS; Schiffman, Amy MD, MHS; Larsen, John W. MD
The importance of body temperature assessment in medicine has long been appreciated. The current concept of normal basal body temperature, defined as 98.6F (37.0C), originated from the work of Wunderlich in 1868.1 This value was based on the study of 25,000 patients and over one million observations. He described fever as a temperature above 100.4F (38.0C); however, the statistical methods employed to determine this value were not clearly documented in his historic treatise. Although Wunderlich did not specifically study temperatures of the laboring gravida, the obstetric community has adopted his conclusions. This may be due, in part, to the United States Joint Committee of Maternal Welfare, who in 1935, defined puerperal morbidity to be “a fever of 100.4F (38.0C) or over on any two of the first 10 days postpartum, exclusive of the first 24 hours.”2 This definition has been applied to both the antepartum and intrapartum periods.
With regard to pregnancy and parturition, published temperature data are sparse and have been limited by sample size or number of temperature observations per patient.3–5 Data addressing the definition of fever in labor are equally scant. Although most authors have used 38.0C as the limit for fever in the gravida, one prospective investigation involving fever in labor caused by group B streptococcus used a threshold of 37.5C to define the action point for antibiotic therapy.6
The bulk of obstetric literature concerning maternal temperature addresses several specific aspects of obstetric care. These include risk factors for development of fever or chorioamnionitis in labor,7–11 the maternal‐fetal temperature relationship,12–16 effects of epidural anesthesia on temperature,17–22 diurnal variation of temperatures,23,24 elevated temperature related to mode of delivery,25 and maternal hypothermia as a side effect of magnesium sulfate therapy.26,27 The interpretation of temperature and fever in these articles is based primarily on Wunderlich's classic definitions.1
The purpose of this study is to document normal values as well as normal trends in maternal temperature during labor. The traditional definition of normal temperature and fever deserves critical reappraisal in the setting of obstetrics because small temperature changes can change the clinical management of mother, fetus, and newborn.
MATERIALS AND METHODS
A retrospective chart review of 189 consecutive singleton deliveries occurring from September 1 through October 1, 1996 at The George Washington University Hospital was conducted. The institutional review board approved the project. Using a preprinted data sheet, trained personnel abstracted the data. Patient demographics, medical history, antenatal history, and delivery data were collected. All sublingual temperatures, from the time of admission to the time of discharge from the labor and delivery area, were collected from the available records.
Patients were included if the delivery resulted in a live singleton at 37 weeks' gestation or longer. Induced labors and patients requiring cesarean delivery were included. Those patients receiving epidural analgesia were also included. The clinicians managing labor administered epidural analgesia when clinically indicated upon patient request. Patients were excluded for medical or antenatal problems associated with infectious morbidity (human immunodeficiency virus, acquired immunodeficiency syndrome, pregestational and insulin‐dependent gestational diabetes mellitus, connective tissue disease, renal disease, sickle cell anemia, intrauterine fetal demise, and cerclage) and clinical chorioamnionitis. Those who received acetaminophen during labor were also excluded.
Temperatures were included either from the onset of labor or from the time of admission if the patient was admitted in spontaneous labor. Onset of labor was defined as the time that regular uterine contractions began. The frequency of temperature recordings was based on the protocol of the labor and delivery unit in 1996. Intrapartum temperatures were recorded every 4 hours when membranes were intact and every 2 hours after amniorrhexis. During the first 2 hours of the recovery period, one temperature was recorded. Patients not transferred within 2 hours after delivery had temperatures recorded every 4 hours. All temperatures were taken sublingually with IVAC Temp Plus II digital thermometers (IVAC Corp., San Diego, CA, error ± 0.1C).
Analysis was performed in several parts. To evaluate the possible confounding variable of prophylactic antibiotic therapy on temperature during labor, the patients were assigned to one of two groups: those who did not receive antibiotics in labor (group 1), and those who received antibiotic prophylaxis for subacute bacterial endocarditis, group B streptococcus, or prolonged rupture of membranes (group 2). Within each group, temperatures were also divided temporally: those taken during labor and those taken during recovery.
The mean of the maximum temperatures for each patient during labor (labor Tmax) and recovery (recovery Tmax) was used for inter‐ and intragroup comparisons. Based on the lack of significant difference between them, groups 1 and 2 were combined to form the group of normal parturients. For patients with at least two observations in labor, the temperature curve (C per hour) for each normal parturient was estimated.
EpiInfo (Centers for Disease Control and Prevention, Atlanta, GA) was used for data entry and SAS (SAS Inc., Cary, NC) was used for all data analysis. Student t tests were used to compare patients who did not receive antibiotics during labor (group 1) and patients who did receive them (group 2). P values < .05 were considered to be significant. Simple linear regression was used to calculate temperature slopes for each patient.
A total of 147 (77.7%) parturients met the inclusion criteria, and 138 patients had available labor temperature data with an average of 2.7 observations per patient. The average length of labor was 9.7 hours. The overall average of the mean labor temperatures was 36.8 ± 0.33C, and 95% of the observations were between 36.2C and 37.5C (see Figure 1). For group 1, the average of the mean labor temperatures was 36.8 ± 0.36C, and for group 2, the average of the mean labor temperatures was 36.8 ± 0.28C. A total of 100 patients had greater than one observation in labor, and 19 patients had greater than one observation in recovery.
Demographics are presented in Table 1. The incidence of risk factors for fever in labor is presented in Table 2. The distribution of temperatures for groups 1 and 2 are presented in Figure 1.
The mean Tmax for each group in labor and recovery are presented in Table 3. There was no significant difference between the labor and recovery Tmax within groups, nor were there significant differences between groups 1 and 2 for either labor Tmax or recovery Tmax. The mean admission temperatures were 36.7 ± 0.44C and 36.8 ± 0.42C for groups 1 and 2, respectively, and 36.7 ± 0.40C for all parturients. Thirty‐eight patients had only one labor observation. The Tmax for these patients was 36.7 ± 0.40C (n = 38), and their average length of labor was 6.8 hours.
Once groups 1 and 2 were combined, these normal parturients were stratified by the use of epidural analgesia, and 81.6% of all patients received epidural analgesia. During labor, those who received epidural analgesia had a mean Tmax of 37.0C compared with 36.8C (not significant) for those who did not have epidural. During recovery, those who received epidural analgesia had a mean Tmax of 37.1C compared with 36.8C (P < .05) for those who did not have epidural.
Temperature slopes of parturients with at least two observations during labor (n = 100) ranged from −0.30C per hour to 0.44C per hour. The mean slope was 0.01 ± 0.12C per hour, and 82% of slopes ranged from −0.10C per hour to +0.10C per hour. Eight percent were greater than 0.10C per hour, and 10% were less than −0.10C per hour.
Distribution of labor temperatures over the course of a 24‐hour period showed a midday nadir and a peak in late evening.
There were no cases of neonatal sepsis or death. Apgar scores were similar in both groups and are shown in Table 1.
Previous work has suggested that rigorous physical exercise may increase body temperature.28 Similar conclusions have been mentioned regarding labor.3,4 Our study suggests that in the normal parturient without clinical chorioamnionitis, labor does not significantly increase body temperature. The overall average of each patient's mean labor temperature was 36.8 ± 0.33C. Furthermore, we found that 95% of temperature observations fell in the range of 36.2C to 37.5C. From the perspective of using Tmax data, 95% of temperature observations fell in the range of 36.2C to 37.8C. Postpartum temperatures also remained in this normal range.
In the subgroup of patients who received epidural analgesia, there was a statistically significant higher Tmax in recovery than in the subgroup without an epidural. However, this was not a clinically significant change because both values are well within the normal range. In the literature, epidural analgesia has been shown to be associated with elevated maternal17,18,20 and fetal20 temperature. A large number of our patients (81.6%) received epidural analgesia, and in those patients, the mean labor Tmax was nearly 0.2C higher (not significant) than in those patients without epidurals. It has been proposed that the essential sympathectomy below the waist prohibits thermoregulatory mechanisms (sweating) from fully functioning in over one‐half of the body when an epidural is in place. Heat dissipation during the high metabolic state of labor is, therefore, compromised and temperature may increase.
Comparing our data with the limited studies on temperature in parturition is difficult particularly because there is diversity in the sites used to record temperatures. Marx and Loew,4 for example, recorded continuous tympanic temperatures in 11 laboring gravidas because tympanic temperatures were thought to closely represent brain stem temperatures. They found that tympanic temperatures increased progressively during the active phase of labor (mean increase 1.46 ± 0.55C in primigravidas and 0.51 ± 0.16C in multigravidas). Marx and Loew4 also found an increase of 1.4C in those patients who received any type of analgesia. Temperature increases were attributed to increased oxygen consumption during contraction of uterine and skeletal muscle; however, their data were derived from a small number of patients with no published information about the starting temperatures or the possibility of intra‐amniotic infection.
Our data are supported in part by the findings of Goodlin and Chapin.5 They used multiple sites (axilla, breast fold, rectum, tympanic membrane, and vagina) to record temperature during labor in 50 patients. Although few data on oral temperatures were reported, all were below 37.2C. Goodlin and Chapin's patients without pain control demonstrated decreases in temperature (across all sites) thought to be caused by hyperventilation and perspiration. The patients with pain control demonstrated little variation in temperature (all sites) over time with a slight rise to a maximum of 37.4C measured vaginally. The slight rise is similar to that in our study as noted above for the patients receiving epidurals.
We selected oral thermometry because it is part of common clinical practice in most labor and delivery units. Oral thermometry may not detect the increase in core body temperature seen with other types of probes. Internal temperature monitoring requires special equipment and was considered too invasive for the current protocol. The large number of patients with epidural analgesia in our population may be considered a confounding factor in the determination of temperature change during labor. Nevertheless, the popularity of epidural analgesia is a reality of modern obstetric care, and in practical terms represents a common patient situation.
Our data coincide with other studies that have looked at temperatures and circadian rhythms. Acker et al's study of temperature at the time of admission showed a diurnal temperature distribution with a nadir between 11 AM and noon and an apogee between midnight and 2 AM.3 Our study had similar findings.
Our study has several limitations. As a retrospective chart review, there was potential for mistakes during recording of the data. The hospital temperature protocol was not always followed exactly. Furthermore, 2‐ and 4‐hour intervals between temperature recordings yield fewer data points than might be planned for in a prospective study. Patients commonly drink fluids and take ice chips during labor, which may have lowered temperatures in some patients. Because the diagnosis of chorioamnionitis is a clinical one, subclinical chorioamnionitis may have gone undiagnosed in this study. The use of amniocentesis and or placental pathology would have more clearly elucidated the meaning of a specific temperature change, and given more laboratory data relevant to the cause of fever in individual cases.
A further limitation of this study is the limited sensitivity of standard clinical thermometry. Commonly used thermometers have 0.1‐degree increments. Nonetheless, modern battery powered digital thermometers are reliable and give reproducible measurements within their stated error. We believe that the observations of this study are reliable for normal parturients, although a university hospital study population may not always extrapolate to some community practice settings. Follow‐up studies should be directed towards patients who develop chorioamnionitis to study nuances of changes in temperature as fever develops.
Our results have significant implications for treatment of patients at risk for developing chorioamnionitis. A temperature increase above a threshold of 37.5C may be an early indicator of evolving chorioamnionitis compared with waiting for a threshold of 38C. We propose that patients who reach the 37.5C threshold have temperatures recorded more often and be closely monitored for other signs of chorioamnionitis, such as fetal tachycardia. In our study, 92% of the temperatures taken from normal patients were less than 37.5C, and 99% were less than 37.8C. If future studies show that chorioamnionitis patients frequently have sentinel temperature readings in the 37.5C to 38.0C range, this would permit an early start for antibiotic therapy.
Although the threshold of 38C or higher has been used to define fever in guidelines from the Centers for Disease Control,29 the American Academy of Pediatrics,30 and the American College of Obstetricians and Gynecologists31 regarding group B streptococcal infection in labor, and a prospective study of early onset group B streptococcal infection in neonates by Boyer and Gotoff6 used a lower threshold of 37.5C. A recent retrospective study by Petrova et al32 concluded that temperature in labor greater than 38C was a predictor of neonatal morbidity and infection‐related mortality. The possibility of a lower threshold was not considered in their study design.
Early antibiotic treatment may significantly decrease intrapartum and puerperal infectious morbidity. In our study, the effectiveness of early antibiotic treatment was seen in the group B streptococcus prophylaxis patients. Twenty‐four of 27 group B streptococcus‐positive patients received intrapartum antibiotic prophylaxis; none developed chorioamnionitis or neonatal sepsis.
We have shown that temperature does not increase significantly in normal labor. To more clearly define the meaning of fever, prospective studies are needed to analyze trends when temperatures reach 37.5C and make comparisons with objective markers for chorioamnionitis.
1. Wunderlich CA, Seguin E. Medical thermometry and human temperature. New York: William Wood and Coe, 1871.
2. Adair FL. The American Committee of Maternal Welfare, Inc: Chairman's Address. Am J Obstet Gynecol 1935;30:868–71.
3. Acker DB, Schulman EB, Ransil BJ, Sachs BP, Friedman EA. The normal parturient's admission temperature. Am J Obstet Gynecol 1987;157:308–11.
4. Marx GF, Loew DA. Tympanic temperature during labour and parturition. Br J Anaesth 1975;47:600–2.
5. Goodlin RC, Chapin JW. Determinants of maternal temperature during labor. Am J Obstet Gynecol 1982;143:97–103.
6. Boyer KM, Gotoff SP. Prevention of early-onset neonatal group B streptococcal disease with selective intrapartum chemoprophylaxis. N Engl J Med 1986;314:1665–9.
7. Soper DE, Mayhall CG, Dalton HP. Risk factors for intraamniotic infection: A prospective epidemiologic study. Am J Obstet Gynecol 1989;161:562–8.
8. Newton ER, Prihoda TJ, Gibbs RS. Logistic regression analysis of risk factors for intra-amniotic infection. Obstet Gynecol 1989;73:571–5.
9. Herbst A, Wolner-Hanssen P, Ingemarsson I. Risk factors for fever in labor. Obstet Gynecol 1995;86:790–4.
10. Soper DE, Mayhall CG, Froggatt JW. Characterization and control of intraamniotic infection in an urban teaching hospital. Am J Obstet Gynecol 1996;175:304–10.
11. Seaward PG, Hannah ME, Myhr TL, Farine D, Ohlsson A, Wang EE, et al. International multicentre term prelabor rupture of membranes study: Evaluation of predictors of clinical chorioamnionitis and postpartum fever in patients with prelabor rupture of membranes and term. Am J Obstet Gynecol 1997;177:1024–9.
12. Macaulay JH, Randall NR, Bond K, Steer PJ. Continuous monitoring of fetal temperature by non-invasive probe and its relationship to maternal temperature, fetal heart rate, and cord arterial oxygen and pH. Obstet Gynecol 1992; 79:469–74.
13. Walker DW, Wood C. Temperature relationship of the mother and fetus during labor. Am J Obstet Gynecol 1970;107:83–7.
14. Morishima HO, Glaser B, Niemann WH, James LS. Increased uterine activity and fetal deterioration during maternal hyperthermia. Am J Obstet Gynecol 1975;121:531–8.
15. Zilianti M, Cabello F, Chacon NR, Wincon CS, Salazar JR. Fetal scalp temperature during labor and its relationship to acid base balance and condition of the newborn. Obstet Gynecol 1983;61:474–9.
16. Morishima HO, Yeh MN, Niemann WH, James LS. Temperature gradient between fetus and mother as an index for assessing maternal and fetal condition. Am J Obstet Gynecol 1977;129:443–8.
17. Macaulay JH, Bond K, Steer PJ. Epidural analgesia in labor and fetal hyperthermia. Obstet Gynecol 1992;80:665–9.
18. Fusi L, Steer PJ, Moresh MJ, Beard RW. Maternal pyrexia associated with the use of epidural analgesia in labor. Lancet 1989;1(8649):1250–2.
19. Webb PJ, James EM, Wheeler AS. Shivering during epidural analgesia in women in labor. Anesthesia 1981;55:706–9.
20. Camann WR, Hortvet LA, Hughes N, Bader AM, Datta S. Maternal temperature regulation during extradural analgesia during labor. Br J Anaesth 1991;67:565–8.
21. Alexander PJ, Sharma SK, Levino KJ, McIntire DD, Wiley J. Epidural anesthesia during labor and maternal fever. Anesthesia 1999;90:1271–5.
22. 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.
23. Malik J, Suk K, Brestak M, Maly V. Daily rhythms of leukocytes, blood pressure, pulse rate and temperature during pregnancy. Ann NY Acad Sci 1962;98:1018–27.
24. Kaiser IH, Halberg F. Circadian periodic aspects of birth. Ann NY Acad Sci 1962;98:1056–63.
25. Lieberman E, Cohen A, Lang J, Frigoletto F, Goetzl L. Maternal intrapartum temperature elevation as a risk factor for cesarean delivery and assisted vaginal delivery. Am J Public Health 1999;89:506–10.
26. Parsons MT, Owens CA, Spellacy WN. Thermic effects of tocolytic agents: Decreased temperature with magnesium sulfate. Obstet Gynecol 1987;69:88–90.
27. Rodis JF, Vintzileos AM, Campbell WA, Deaton JL, Nochimson DJ. Maternal hypothermia: An unusual complication of magnesium sulfate. Am J Obstet Gynecol 1987;156:435–6.
28. Gisolfi CV. Temperature regulation during exercise. Med Sci Sports 1983;15:15–20.
29. Centers for Disease Control and Prevention. Prevention of perinatal group B streptococcal disease: A public health perspective. MMWR 1996;45:1–24.
30. American Academy of Pediatrics. Revised guidelines for prevention of early-onset group B streptococcal (GBS) infection. Pediatrics 1997;99:489–96.
31. American College of Obstetricians and Gynecologists. Prevention of early-onset group B streptococcal disease in newborns. ACOG committee opinion no. 173. Washington, DC: American College of Obstetricians and Gynecologists, 1996.
32. Petrova A, Dejmissie K, Rhoads G, Smulian JC, Marcella S, Ananth C. Association of maternal fever during labor with neonatal and infant morbidity and mortality. Obstet Gynecol 2001;98:20–7.