Hyperthermia after epidural analgesia in obstetrics

Learning objectives.

After reading this article you should be able to:

• •Describe the epidemiology of intrapartum hyperthermia.
• •Discuss the possible consequences of epidural hyperthermia.
• •Explain the potential mechanisms underlying epidural hyperthermia.
• •Illustrate the management of epidural hyperthermia.

Key points.

• •Epidural analgesia increases the risk of intrapartum hyperthermia.
• •Epidural hyperthermia is associated with an increased risk of needing treatment with antibiotics, operative delivery, and possibly neonatal brain injury.
• •It is not possible to differentiate epidural hyperthermia and intrapartum infection.
• •Two leading mechanisms for epidural hyperthermia have emerged: sympathetic blockade and immunomodulation.
• •Antibiotics and paracetamol do not prevent epidural hyperthermia.

Epidural analgesia is a widely used form of labour analgesia, particularly in developed countries. In the UK it is chosen by 25% of parturients during labour, whereas in the USA it is selected by more than 70% of parturients.^1,2 It is the most effective form of labour analgesia and it may be extended to provide anaesthesia for intrapartum Caesarean delivery, thus avoiding the need for general anaesthesia.³ In addition, in high-risk parturients, such as those with pre-eclampsia or cardiac disease, it may improve maternal and neonatal outcome.

Epidural hyperthermia, also known as epidural-related fever, refers to the situation in which a parturient who has an epidural for labour analgesia develops an increased body temperature. Hyperthermia is the preferred term, as fever implies an inflammatory mechanism and this has not been proven.⁴ It was first reported in 1989 when vaginal temperature was compared between parturients who received either pethidine or epidural analgesia: the temperature did not change in those who received pethidine but increased at 0.15°C h⁻¹ in those who received epidural analgesia (Fig. 1).⁵ Thirty years later it remains a somewhat enigmatic disorder. Its underlying mechanism is still not understood and there are no effective treatments. However, there is increasing recognition that it is not a benign condition.

Fig 1.

Vaginal temperature during labour in women receiving epidural analgesia (n=19) and women receiving intramuscular pethidine (n=15). Epidural analgesia was instituted after 4.2 h. Reproduced from Fusi and colleagues, with permission.⁵

Intrapartum hyperthermia is the umbrella term for any increased maternal temperature during labour. During an uncomplicated labour, maternal metabolic rate increases by 25%, but this does not result in a significant change in body temperature. Consequently, a commonly used definition of intrapartum hyperthermia is: ‘a core temperature during labour of ≥38°C on one occasion or ≥37.5°C on two consecutive occasions two hours apart’.⁶ Epidural hyperthermia and intrapartum infection are the two major aetiologies. The incidence of intrapartum hyperthermia is 20% in parturients with epidural analgesia and 5% in those without.^3,7 In parturients without epidural analgesia, intrapartum hyperthermia is almost exclusively secondary to infection.⁷ In parturients with epidural analgesia intrapartum, hyperthermia may be secondary to infection or a direct result of the epidural itself (Fig. 2).⁷ Epidural analgesia does not increase the risk of intrapartum infection and so approximately 75% of cases of epidural hyperthermia are caused by the epidural.⁷ However, currently it is not possible to differentiate cases of epidural hyperthermia that are caused by intrapartum infection from cases that are a consequence of the epidural itself.

Fig 2.

The causes of intrapartum hyperthermia. Proportions are based upon a population in which one third of parturients have epidural analgesia and the incidences of epidural hyperthermia and intrapartum infection are 20% and 5%, respectively.1, 2, 3^,7

Consequences of epidural hyperthermia

Epidural hyperthermia should not be considered a benign condition. Evaluation and investigation for sepsis, and antibiotic treatment are both established consequences.⁷ It is also associated with an increased risk of both Caesarean and instrumental delivery, although these correlations may be coincidental.⁸ Of greater concern, preliminary evidence points to a link with an increased risk of neonatal brain injury.⁹

The associations with evaluation for maternal and neonatal sepsis, and with antibiotic treatment have been recognised for more than 20 yrs.⁷ However, these links are likely a consequence of the difficulty in clinically differentiating epidural hyperthermia and intrapartum infection, rather than an indicator that epidural analgesia increases the risk of intrapartum infection.⁷ Tachypnoea, tachycardia, and leukocytosis may all occur in the absence of intrapartum infection, and as a result intrapartum hyperthermia alone is considered sufficient evidence to initiate maternal sepsis evaluation and antibiotic administration.^6,10 In addition, owing to the significant morbidity and mortality associated with early onset neonatal infection, maternal intrapartum antibiotic treatment is a trigger for neonatal sepsis evaluation and antibiotic treatment.¹¹ In the majority of cases of epidural hyperthermia there is no objective evidence of infection, and so this management strategy likely leads to over-investigation and overtreatment of intrapartum hyperthermia, which has economic and antibiotic stewardship consequences.⁷ However, given the serious consequences of undertreating intrapartum infection, there is currently no viable alternative approach.

Epidural hyperthermia is associated with a two-fold increase in the risk of Caesarean delivery and instrumental vaginal delivery.⁸ However, it is not clear if this association signifies a causal link or is a coincidence. Epidural hyperthermia increases the risk of labour arrest, rather than the risk of maternal or fetal distress, and so it is unlikely that the increased incidence of operative delivery is attributable to clinical concerns for intrapartum infection.⁸ Chorioamnionitis impairs myometrial contractility and thus increases the risk of labour arrest and operative delivery.¹² It is possible that epidural hyperthermia has a similar effect on myometrial contractility, but this explanation is untested. An alternative explanation is that the association between epidural hyperthermia and operative delivery is confounded by the duration of labour. Epidural hyperthermia is a time-dependent condition; that is maternal temperature increases at a constant rate after epidural insertion.⁵ Parturients with longer labours have greater durations of epidural analgesia and are therefore at greater risk of developing an increased body temperature.⁵ By definition operative delivery for arrest of labour also occurs after a prolonged labour, and so it is possible that the increased risk of operative delivery is coincidental. Further mechanistic work is required to establish which explanation is correct.

The potential association with an increased risk of neonatal brain injury is the most concerning consequence of epidural hyperthermia. It is recognised that intra- and postpartum hyperthermia increases the risk of injury to the infant brain. Intrapartum hyperthermia is a risk factor for neonatal encephalopathy, and clinical chorioamnionitis is associated with the development of cerebral palsy.¹³ Furthermore, in term infants with hypoxic ischaemic encephalopathy, postpartum hyperthermia is associated with poor long-term neurological outcome and therapeutic hypothermia improves neurodevelopmental outcome.¹⁴ However, it is not known if the risk of brain injury is confined to cases with intrapartum infection or if it is present with intrapartum hyperthermia of any cause. Preliminary work, demonstrating that epidural hyperthermia increases the need for resuscitation at birth and is a risk factor for seizures during the neonatal period, gives credence to the latter explanation, but further work is required to determine whether epidural hyperthermia has an impact on long-term neurodevelopmental outcome.⁹

Mechanisms of epidural hyperthermia

The mechanism underlying epidural hyperthermia is uncertain. A number of mechanisms have been hypothesised, including shivering, oxytocin, and a lack of opioids, but two leading theories have emerged: sympathetic blockade and immunomodulation.^5,15, 16, 17, 18 Both are supported by preliminary evidence, but neither mechanism is proven.

The sympathetic blockade hypothesis is that epidural analgesia limits cutaneous heat loss via blockade of active cutaneous vasodilation and sweating (Fig. 3). Mean body temperature is dependent upon heat production and heat loss; thus if heat loss decreases and heat production remains constant, body temperature will increase. Heat loss is regulated by two pathways of the sympathetic nervous system: the noradrenergic and the cholinergic pathways. The noradrenergic pathway is responsible for active cutaneous vasoconstriction and the cholinergic pathway is responsible for active cutaneous vasodilation and sweating.¹⁹ In thermoneutral conditions and during cold stress, heat loss is regulated solely by the noradrenergic pathway. However, during heat stress heat loss is almost exclusively regulated by the cholinergic pathway (Fig. 3). Consequently, the effect of neuraxial blockade on body temperature is dependent upon body temperature before block placement. Before elective Caesarean delivery or non-obstetric surgery, patients are in a thermoneutral state and so neuraxial blockade blocks active vasoconstriction resulting in increased heat loss and a reduction in body temperature (Fig. 3).²⁰ By contrast, neural blockade during established hyperthermia blocks active vasodilation and sweating resulting in decreased heat loss.¹⁹ Labour is a heat stress, and so it is plausible that the sympathetic blockade associated with intrapartum epidural analgesia would limit heat loss. It has been demonstrated that epidural anaesthesia for intrapartum Caesarean delivery limits heat loss and as a result body temperature increases.¹⁵ However, it remains to be tested whether or not a similar effect occurs during epidural analgesia for labour.

Fig 3.

Schematic layout of the effects of neuraxial blockade upon cutaneous heat permeability in different thermal conditions. In thermoneutral conditions, such as elective Caesarean delivery (ECD), neuraxial blockade blocks active vasoconstriction resulting in increased cutaneous heat permeability, increased cutaneous heat loss and a fall in body temperature.²⁰ However, during heat stress, such as labour, neuraxial blockade blocks active vasodilation resulting in decreased cutaneous heat permeability, decreased cutaneous heat loss and a rise in body temperature.¹⁵

The immunomodulation hypothesis is that epidural administration of local anaesthetic during labour provokes a sterile febrile response. This is the source of the synonym for epidural hyperthermia: epidural-related fever. The term ‘hyperthermia’ refers to any increase in body temperature above the normal range. ‘Fever’ is a form of hyperthermia that is secondary to an increase in the body's set-temperature.⁴ Body set-temperature is regulated by the hypothalamus via a cascade of pro'- and anti-inflammatory cytokines, notably interleukin-1, interleukin-1 receptor antagonist, and interleukin-6 (Fig. 4). It is recognised that epidural hyperthermia occurs in conjunction with a proinflammatory state, but it is unclear if this proinflammatory state is induced by epidural analgesia or if it is a confounder: the proinflammatory state is pre-existent, before epidural placement, in parturients who later become hyperthermic.

Fig 4.

Schematic of proposed immunomodulation mechanism. Reproduced from Del Arroyo and colleagues, with permission.¹⁷ In parturients without epidural analgesia, activation of the pyrogenic pathway via activated caspase-1 (casp1), interleukin-1β (IL-1β), and interlukin-6 (IL-6) is attenuated by the parallel release of the antipyrogenic cytokine interleukin-1 receptor antagonist (IL1-ra). In parturients receiving epidural analgesia, inhibition of caspase-1 activity reduces the release of IL1-ra. Loss of this mechanism against fever increases the chances of hyperthermia.

The origin of the link between epidural hyperthermia and inflammation is the finding that titres of the pyrogenic cytokine, interleukin-6, are greater in parturients with epidural hyperthermia, compared to those with epidural analgesia and a normal body temperature.¹⁶ Subsequently, a randomised double-blind placebo-controlled trial also found that the incidence of epidural hyperthermia is reduced by high doses of methylprednisolone.²¹ These results demonstrated an association between epidural hyperthermia and inflammation, but they did not provide evidence of causation. To date, two causative mechanisms have been investigated. The first is that intrapartum epidural analgesia increases the risk of intrapartum infection and the second is that, during the intrapartum period, epidural local anaesthetic has a pro-inflammatory immunomodulatory effect. Infection is a common trigger for intrapartum inflammation, but it is unlikely that epidural analgesia either increases the risk of intrapartum infection or is utilised more frequently in parturients that are more likely to develop infection (selection bias).⁷ It has been consistently demonstrated that epidural analgesia does not increase the risk of culture positive infection for either the mother or the neonate even when mode of labour analgesia is randomised.⁷ In addition, a double-blind placebo-controlled trial found that prophylactic antibiotics are ineffective at preventing epidural hyperthermia.²² There is preliminary evidence to support the immunomodulation hypothesis. One study demonstrated that interleukin-1 receptor antagonist/interleukin-1β ratio is reduced in intrapartum serum of parturients with epidural analgesia compared with those without other forms of analgesia.¹⁷ Interleukin-1 receptor antagonist is anti-pyrogenic and interleukin-1β is pyrogenic, and so this result suggests that intrapartum epidural analgesia induces a pyrogenic state, possibly via inhibition of leucocyte caspase-1 activity (Fig. 4).¹⁷ A further study found that in vitro ropivacaine induces release of the propyrogenic cytokines interlukin-6 and prostaglandin E from placental trophoblasts.¹⁸

At present it is not certain if either sympathetic blockade or immunomodulation is responsible for epidural hyperthermia. Neither mechanism is mutually exclusive, and further research is warranted to determine if either or both mechanisms have a role in the aetiology of hyperthermia.

Management of epidural hyperthermia

The management of epidural hyperthermia includes the consent process for an intrapartum epidural and the management of intrapartum hyperthermia. Given that currently the impact of epidural hyperthermia is not fully understood, it is reasonable to question how much information should be provided to parturients during the consent process for an intrapartum epidural. In the UK, informed consent requires that doctors ‘take reasonable care to ensure that the patient is aware of any material risks involved in any treatment’;.a ‘material risk’ being one in which ‘a reasonable person in the patient's position would be likely to attach significance to the risk’.²³ It is likely that parturients would attach significance to any of the potential risks of epidural hyperthermia, and so the question of whether to inform patients depends on the certainty of a causative link. A high threshold should be set during this assessment of certainty, because dissuading parturients from choosing intrapartum epidural analgesia may in itself cause harm – particularly in high-risk pregnancies, such as in those with pre-eclampsia or cardiac disease. At present the Obstetric Anaesthetists' Association advises that fever is an adverse effect of intrapartum epidural analgesia, but it does not elaborate on the potential consequences.²⁴ The evidence supporting a causative link with maternal and neonatal sepsis evaluation and antibiotic treatment is well established, and so in our opinion it is reasonable to inform parturients of this risk.⁷ By contrast causative links to operative delivery and neonatal brain injury are suspected rather than proven, and so there is no consensus to provide this information. Nonetheless, medical professionals should be aware of all the potential consequences of epidural hyperthermia and be prepared to discuss them with parturients if directly questioned.

At present there is no treatment for epidural hyperthermia, and it is not possible to differentiate epidural hyperthermia from intrapartum infection. Intrapartum infection has potentially devastating maternal and neonatal consequences, and so the management of intrapartum hyperthermia should be the same regardless of whether the parturient has epidural analgesia (Table 1).⁶ This includes paracetamol, blood cultures, antibiotic administration, and supportive measures.⁶ In addition, the neonate should be evaluated for sepsis (blood cultures and C-reactive protein) and treated empirically with intravenous antibiotics.¹¹ Early delivery is only warranted if there is evidence of fetal compromise (e.g. cardiotocogram abnormality).⁶

Table 1.

Diagnosis and management of intrapartum infection and early onset neonatal infection.^6,11 ∗Core temperature ≥38°C on one occasion or ≥37.5°C on two consecutive occasions 2 h apart. ^†Systolic blood pressure <90 mmHg; mean arterial pressure <70 mmHg or systolic blood pressure decrease >40 mmHg. PaO₂, arterial oxygen partial pressure; FIO₂, fraction of inspired oxygen.

Diagnostic criteria	Management
Maternal
One or more of -Hyperthermia∗ -Hypothermia (<36°C) -Tachycardia (>100 beats min−1) -Tachypnoea (>20 breaths min−1) -Hypoxia (Pao2/Fio2 <40 kPa) -Hypotension† -Oliguria (urine output <0.5 ml kg−1 h−1) -Impaired consciousness	-Blood cultures -Arterial blood gas -Imaging (to confirm source) -Paracetamol -Intravenous broad spectrum antibiotics
Neonatal
One or more of: -Invasive group B streptococcal infection in a previous baby -Maternal group B streptococcal colonisation, bacteriuria or infection in the current pregnancy -Rupture of membranes before labour -Preterm birth after spontaneous labour (before 37 weeks' gestation) -Suspected or confirmed rupture of membranes for more than 18 h in a preterm birth -Intrapartum fever higher than 38°C, or confirmed or suspected chorioamnionitis -Parenteral antibiotic treatment given to the woman for confirmed or suspected invasive bacterial infection at any time during labour, or in the 24 h periods before and after the birth -Suggested or confirmed infection in another baby in the case of a multiple pregnancy	-Blood cultures -C-reactive protein -Lumbar puncture (if strong clinical suspicion of infection) -Intravenous antibiotics (benzylpenicllin and gentamicin)

Three treatments for epidural hyperthermia have been investigated, but to date none have been found to be effective. Two randomised double-blind placebo-controlled trials have found that paracetamol and prophylactic broad-spectrum antibiotics (cefoxitin) are ineffective, and a randomised double-blind placebo-controlled trial found that high dose corticosteroids (methylprednisolone 100 mg every 4 h) reduced the incidence of hyperthermia, but increased the risk of neonatal bacteraemia.^16,21,22 Clearly therefore there is a need for further research into treatments for epidural hyperthermia; however, consideration should also be given to preventative strategies. For example the optimal time of epidural insertion and the optimal temperature of labour rooms are potential areas of investigation. Epidural hyperthermia is a time-dependent process, and as a result early epidural insertion may increase the risk of hyperthermia.⁵ Also, as neuraxial blockade impairs the regulation of heat loss, the current recommended labour room temperature (minimum 25°C) may increase the risk of epidural hyperthermia.²⁵

Summary

Epidural hyperthermia is a common condition, which is increasingly recognised to have adverse consequences, particularly for the neonate. The mechanism is not fully understood, but two leading theories have emerged: sympathetic blockade and immunomodulation. Currently there are no effective treatments, and it is not possible to differentiate epidural hyperthermia from other causes of intrapartum hyperthermia. As a result, the management of epidural hyperthermia is the same as the management for intrapartum infection. Further research is necessary to fully elucidate the consequences and the mechanism of epidural hyperthermia and to develop strategies for prevention and treatment.

Declaration of interests

The authors declare that they have no conflicts of interest.

MCQs

The associated MCQs (to support CME/CPD activity) are accessible at www.bjaed.org/cme/home by subscribers to BJA Education.

Biographies

Chris Mullington BSc (Hons) FRCA PhD is a locum consultant anaesthetist at Imperial College NHS Trust and an honorary clinical research fellow at Imperial College London. His PhD thesis investigated thermoregulation and obstetric regional anaesthesia.

Surbhi Malhotra FRCA is a consultant anaesthetist who has worked in the UK and Australia. Her clinical and research interests are obstetric anaesthesia and women's health. Currently she is director of research and joint lead for obstetric anaesthesia at Fiona Stanley Hospital in Murdoch, Western Australia. She is a past honorary secretary to the Obstetric Anaesthetists' Association and a past editor of Anaesthesia.

Matrix codes: 1A01, 2B01, 3B00