epidural and maternal fever - fever and neonatal evaluation for sepsis

Fever and neonatal evaluation for sepsis

Some laboring women develop fever, which is significantly more common among those who have epidurals. Potentially maternal fever may subject newborns to unnecessary tests for sepsis and treatment with antibiotics. However, the development of fever depends on several other factors, most importantly parity and duration of labor.

Women who have epidural are more likely to develop significant fever – elevated body temperature above 38 degrees C – during labor. By itself is nothing more than maybe a small inconvenience. However, there is also a debate if this increase leads to the unnecessary blood tests and potentially harmful antibiotic treatment of the babies.

The increase of maternal temperature with epidural has been reported for the first time in 1989. In that study women who had labors longer than four hours had their temperature slightly increased, though none of them as high as 38 degrees. It was concluded that this increase in temperature is not clinically important. In 1991 another report confirmed this conclusion. Several studies conducted in the years to follow not only confirmed the association of labor epidurals with elevated body temperature, but found that it can reach significant levels.

Serious arguments about fever caused by epidural started after the report by the group headed by E.Lieberman published in 1997, in which it was found that not only epidural analgesia in labor can lead to fever, but that babies may be subjected to the unnecessary tests and investigations for sepsis. In that study 34 percent of the newborns born to women who received epidurals were investigated for sepsis compared to only 10 percent of those born by women without epidurals. And even though the actual number of babies who had to be investigated for sepsis was low, the authors of the study emphasized that women who request epidurals must be aware that by doing so they place their babies at increased risk of unnecessary tests and antibiotics, both of which may lead to complications. As it often happens, the media took great interest in this study, and headlines such as “When labor pain drugs cause fevers, babies face tests” and “Epidurals lead to more infant tests” appeared in newspapers and on television.

As the fever associated with labor epidural is potentially harmful to the mother and the baby we will take a closer look at this issue. The first question that should be asked is: how can epidural cause fever?

In order to answer this a brief refresher of physiology of temperature regulation is required. Human body normally functions within very narrow temperature range, around 37 degrees C (98.6 degrees F). The organ responsible for regulation of body temperature is the area in the brain called hypothalamus, which also maintains many other important physiological functions. Hypothalamus processes the temperature of the blood and neural signals from the temperature receptors on the skin. If temperature gets out of acceptable range, hypothalamus triggers appropriate physiological responses.

When body temperature gets below acceptable levels peripheral blood vessels constrict, this decreases heat loss through the skin and causes all too familiar cold feeling in the toes and fingers. There are several conscious responses: the person moves away from the cold area and puts on warm clothes. Physical activity increases, which results in increasing metabolic rate and increased heat production.

Physiological responses to fever are vasodilatation and sweating. Dilatation of blood vessels leads to increased heat loss through the body surface. Evaporation of sweat from the skin is associated with significant heat loss, and is efficient in lowering body temperature. Hypothalamus also sends impulses to the cerebral cortex which results in the sensation of being hot and appropriate behavior changes: shedding of clothes, moving to colder areas and other appropriate conscious steps. Increasing the rate of breathing is another way of losing heat.

In brief, hypothalamus is the thermostat of the body, with its set point around normal temperature. However in some situations the temperature set point changes. The most common example of such change is immune response to infection.

It was also observed as long ago as in 1875 that labor is associated with fever in the absence of infection. The contractions of the uterus lead to very small increases in temperature which can accumulate over hours of labor and cause the body temperature increase up to two degrees Celsius. Obviously, with longer labor the chance of temperature increase due to uterine contractions increases. Usually it is counteracted by normal physiological responses such as sweating and panting. Women in labor also receive cold intravenous fluids, which also helps preventing fever.

Epidural analgesia may interfere with normal mechanisms of temperature regulation in several ways. It blocks sympathetic nerves that regulate sweat glands. This abolishes sweating in large part of the body affected by labor epidural: the lower torso and lower limbs. The blood vessels in the areas affected by epidural dilate which leads to heat loss that may eventually lead to shivering and overproduction of heat. At the same time blood vessels in the rest of the body constrict which leads to the conservation of heat. By efficiently blocking pain labor epidural effectively eliminates another mechanism of heat loss, panting. The combination of these effects of epidural and prolonged labor may lead to increases in body temperatures which in most patients are not clinically significant. This was exactly what first studies reported: the temperature increased in women who were in labor for longer than 12 hours. More of those women whose temperature increased had epidural than those whose did not.

The increase in temperature by itself at worst is an inconvenience. However, it is the main symptom for diagnosing chorioamnionitis – infection of the placenta – potentially serious complication of labor. The question therefore arises, if the rise in body temperature is associated with infection, and is this infection possibly caused by epidural?

This question was investigated by Dashe and colleagues in 1999 who studied the placentas of delivered women. They confirmed that fever was more common among women with epidural analgesia (46% versus 25%) and it was more common where inflammation of the placenta was present (35% versus 15%). In those without placental inflammation the fever was uncommon and was no different between those who had epidural – 11% – and those who did not – 9%. The authors concluded that fever is caused by inflammation rather than epidural itself.

The connection between epidural analgesia, fever and chorioamnionitis is worrying. On the surface, fever is more common among those receiving epidurals and so is placental inflammation. The logical question that follows therefore is: can epidural cause inflammation? From physiological point of view there is no possible explanation for this. There is, however, more plausible explanation.

The study by Phillips and his colleagues observed 715 women in labor, 68 of whom had fever. Among those 47% were nulliparous – gave childbirth for the first time. Most of them had labors of 12 hours or longer. Women who had shorter labors had a very low incidence of fever, 4%, as did multiparous women. After analyzing the data the authors came to the conclusion that three factors were associated with fever in labor: epidural analgesia, nulliparity and prolonged labor. Importantly, it was found that fever associated with labor epidurals was only observed in nulliparous women with prolonged labors.

That, in turn, explains the initial association of epidural analgesia and fever. The studies that found the association were observational, not randomized, and epidural were given to patients on request. As nulliparous women are known to have longer and more painful labors, they are also more likely to ask for epidural analgesia. Women with developing chorioamnionitis are also more likely to have more pain in labor and, therefore, are also more likely to request epidurals. All this means that it is likely that other factors are more responsible for fever than the epidural itself.

There is one possibility of the epidural directly associated with labor. One recent randomized study by S Sharma and co-authors looking into the association of epidural analgesia and the risk of caesarean section also found that epidural prolongs the first stage of labor by about 40 and the second stage by 10 minutes. This lengthening of labor may, at least theoretically, result in higher incidence of fever in those receiving epidural. Importantly, most parameters reflecting the wellbeing of the newborns were generally better in the epidural group, compared with those receiving pethidine (meperidine) injections. No adverse effects were observed in newborns delivered by women with labor epidurals.

It is worth mentioning that the study of 1997, the one that demonstrated that labor epidurals lead to unnecessary investigations of newborns for sepsis and potentially harmful treatment with antibiotics, has been heavily criticized in medical literature.

First of all, the study was retrospective, and the database of patients for analysis was taken from another study that compared different approaches of managing labor. Epidural analgesia was not even part of that study and was given to women on request who, as noted earlier, are much more likely to have difficult and prolonged labors and higher likelihood of placental infection.

Second, two thirds of infants – the majority – who were evaluated for sepsis were born to mothers who did not have fever, and there is no explanation in the text of the study how newborns were selected for testing. Apparently there were other criteria according to which babies underwent additional tests for sepsis, such as low grade fever, prolonged rupture of membranes, positive vaginal cultures for Streptococcus B. For some reason, however, this has not been discussed in the study, even though the authors were specifically asked about it in one of the comments by other researchers. All that was said in the letter to the editor of the journal is that “re-review of the medical records is planned”. To my knowledge, this “re-review” never happened.

Most importantly, even though more infants in the epidural group were treated with antibiotics, the number of confirmed cases of sepsis among neonates was very low, 0.2 – 0.3%.

Existing data suggests that epidural analgesia in labor is associated with increased incidence of fever among laboring women. The mechanism of this increase in body temperature is determined by several factors. Studies have shown that chances of fever are higher in longer labors and in primiparas. This fever does not cause problems in vast majority of mothers and neonates. If it places the babies at risk of unnecessary tests and treatment with antibiotics depends on particular hospital and the policies of their pediatric departments. Increased maternal temperature will prompt many pediatricians to pay more attention to the baby, and some of them will proceed with taking bloods and giving antibiotics. On the other hands, as epidurals become more and more popular, many of our colleagues become more familiar with this phenomenon, feel more relaxed when it happens and do not rush testing. The possibility of fever and its potential consequences should be discussed with the obstetrician before deciding which method of relieving labor pain to choose.


1. JM. Alexander. Epidural Analgesia for Labor Pain and Its Relationship to Fever. Clin Perinatol 32 (2005): 777–787.

2. W Camann. Intrapartum Epidural Analgesia and Neonatal Sepsis Evaluations: A Casual or Causal Association? Anesthesiology – Volume 90, Issue 5 (May 1999). 

3. E Lieberman, J Lang, F Frigoletto, D Richardson, S Ringer, A Cohen. P Epidural Analgesia, Intrapartum Fever, and Neonatal Sepsis Evaluation. Pediatrics 1997;99:415-419.

4. SK Sharma, JM Alexander, G Messick, SL Bloom, DD McIntire, J Wiley, KJ Leveno. A Randomized Trial of Epidural Analgesia versus Intravenous Meperidine Analgesia during Labor in Nulliparous Women. Anesthesiology 2002; 96:546–51.



Dr. Eugene Smetannikov is a practicing anesthesiologist with the interest in obstetric anesthesia. He is the author of the most comprehensive book on the subject, The Truth About Labor Epidural