Too cold, too hot or just right? Temperature regulation during caesarean section.

Abstract

Winter invokes thoughts of cozy scarves, plump down comforters and roaring wood fires marshalled against frigid temperatures, treacherously icy roads and arctic snow squalls. Thermostats in homes, office suites and operating rooms are the site of quotidian conflicts on what is considered the ‘optimal temperature’. These skirmishes also occur internally, as human thermoregulation, which is controlled by the pre-optic/anterior hypothalamus, responds to information relayed from internal (core) and surface (skin) sensors. Heat conservation, generation or dissipation is controlled through a series of feedback loops, with afferent temperature stimuli from the core being significantly more influential than that from the skin; in contrast, efferent thermal responses are mostly dependent on cutaneous vasodilation and constriction to effect change. During exercise or heat exposure, skin blood flow can increase from a resting baseline of 250 ml.min−1 up to 8 l.min−1 through increases in cardiac output, redistribution of blood flow from other areas such as the splanchnic region and sweating to dissipate heat effectively 1. In cold environments, heat conservation is reliant on cutaneous vasoconstriction, which decreases skin blood flow and heat transfer from the core to the periphery. Despite a dynamic relationship between the hypothalamus, core and surface sensors and blood flow, the accompanying science appears ‘frozen’ by limitations in instrumentation or imagination. Our understanding of the thermal influences of pregnancy, labour, neuraxial analgesic and anaesthetic agents and mode of delivery remains nascent. Although a few centrally-located temperature monitoring sites (e.g. urinary bladder, oesophageal, uterine and rectal) have been studied in pregnancy, intra-arterial thermistors (particularly within the pulmonary artery, since 1972) represent the gold, but seldom used, standard 2. Peripheral thermometers (e.g. tympanic membrane, temporal artery, axillary or oral), although more convenient for clinical use, suffer from poor diagnostic accuracy, especially in their sensitivity for detecting low-grade fever 2. The thermal effects of pregnancy hormones and commonly used agents further confuse matters. Oestrogen, oxytocin and neuraxial morphine are associated with hypothermia, although the mechanisms for these effects remain speculative 3. The use of neuraxial analgesia has been associated with an increase in temperature in approximately 20% of labouring women. The hyperthermia likely represents a non-infectious, systemic inflammatory process and is typically modest (simultaneously obtained mean temperature at baseline to labour maximum by site: oral, 36.6 °C–37 °C; tympanic, 36.8 °C–37 °C and intrauterine, 37.2 °C–37.6 °C) 4, but occasionally can progress to a clinical fever (38 °C; labour epidural analgesia-associated fever (LEAF)) 5. In contrast, most women undergoing neuraxial anaesthesia during caesarean section will experience a 0.5 °C decrease in temperature over the first 30 min, and in women receiving spinal morphine, a 1.0 °C decrease per hour may persist after delivery and result in severe hypothermia 6. Shivering in this setting appears ineffective at mitigating the hypothermia. The aetiology and impact of these observations, including the thermal transition from labour analgesia to caesarean anaesthesia, remain unknown. Seemingly in response, Mullington et al. in a recent issue of Anaesthesia, studied 20 pregnant actively-labouring women with effective epidural analgesia provided by levobupivacaine 0.1% with fentanyl 2 μg.ml−1, who required a category-2 or -3 caesarean section 7. Core (tympanic) and skin (seven sites) temperatures, as well as cutaneous heat loss (five sites) and skin blood flow (laser Doppler flowmetry probe; two sites), were measured continuously from conversion of epidural labour analgesia to caesarean anaesthesia until the end of surgery. The investigators observed that before delivery, core body temperature did not change, and although skin temperature increased, skin blood flow did not change. Following delivery, core body temperature decreased to below baseline, skin temperature remained higher than baseline and skin blood flow increased. Thus, heat production exceeded heat loss until birth, after which they were balanced. These findings, and the attention applied to measuring thermodynamics, are novel and valuable; they faithfully transcribe what we observe clinically and focus our vigilance towards the postpartum heat loss that occurs during caesarean delivery. However, they also establish a conundrum of pre-partum heat gain and postpartum heat loss with unexpected skin temperature and blood flow results. Pre-partum heat gain, for example, should be met with increased skin temperature and blood flow. So, why do these changes happen? The investigators suggest, but did not directly measure, that the reduction in cutaneous heat loss results from ‘blockade of active vasodilation instead of invoking thermoregulatory (vasodilator), non-thermoregulatory (vasoconstrictor), or baroreceptor-mediated reflex vasoconstrictive processes’ 7. Such a statement deserves a more nuanced explanation, as it is likely that each of these processes are invoked to some (yet to be determined) degree. The vasoconstrictive system is tonically active in thermoneutral environments, and subtle changes in this system are most responsible for maintenance of normothermia during changes in activity or environment. Of interest, withdrawal of vasoconstrictor activity is responsible for 10–20% of the cutaneous vasodilation observed during hyperthermia 1. Thus, this system is designed to maintain normothermia; postpartum, if this system is inhibited by anaesthetic blockade (which should be the focus of future studies!), it is likely that skin temperature and blood flow would be increased. In contrast, the sympathetic active vasodilator system is tonically inactive in normothermia, and invoked only with increases in internal temperature, such as during exercise or moderate environmental heat exposure. Activation of the vasodilator nerves is responsible for 80–90% of the cutaneous vasodilation observed during hyperthermia. Very small increases in skin blood flow, such as an 8 ml per 100 m.min−1 over the entire body surface, can double the heat transfer to the environment 1. This system has been demonstrated to be inhibited by anaesthetic blockade. Thus, in the pre-partum period, increasing temperature could be accompanied by diminished skin temperature and blood flow. The mechanism by which anaesthetic blockade exhibits its effects remains unclear, but most likely includes local pre-synaptic inhibition of cholinergic nerves, bradykinin and nitric oxide 1. Finally, the baroreceptor-mediated cutaneous vasoconstriction and vasodilation is affected by blood pressure; activation of this reflex is minimal during normothermia, but maximal with hyperthermia 1. So where does this leave us? Principally, we now understand that further investigations are needed to more fully explain the thermal influences during pregnancy, labour, neuraxial analgesia and anaesthetic techniques and agents and operative events. Even a single intravenous dose of phenylephrine or ephedrine, used to treat spinal anaesthesia-induced hypotension, increases and decreases skin blood flow, respectively, without influencing sympathetic blockade. Adrenaline, such as that placed within the epidural catheter mixed with local anaesthetic agents, has been observed to decrease skin blood flow, in comparison with increases in skin blood flow observed with norepinephrine and dobutamine 8. Moreover, the methods of investigation should continue to be challenged and developed. Doppler velocimetry is restricted to a relatively small area of measurement and is optimally measured during maximal vasodilation, which can be produced by local warming of the skin. The human cutaneous circulation is uniquely controlled by two populations of sympathetic nerves, the well-known adrenergic vasoconstrictor and the less well-understood vasodilator nerves 1. Glabrous skin (palm, soles and lips) are solely innervated by sympathetic vasoconstrictor nerves and have numerous arteriovenous anastomoses. Arteriovenous anastomoses are thick-walled, low resistance conduits that allow high flow rates directly from arterioles to venules. In contrast, nonglabrous skin is innervated by both sympathetic vasoconstrictor and vasodilator nerves, and have few, if any arteriovenous anastomoses. The influence of these two systems can and should be directly compared and contrasted. True baseline temperatures should be obtained at the start of labour, such as in women having labour induced, and at the initiation of labour analgesia. Moreover, because neuraxial opioids are known to affect thermal regulation, their presence in the epidural mixture should be considered and controlled. Methods for faithful core and peripheral temperature measurements should be developed and validated; tympanic sensors are poorly tolerated and subject to poor contact, and consequently reflect external auditory meatus temperature (e.g. peripheral rather than core temperatures) 9. Skin sites, although showing a rise in temperature consistent with vasodilation, demonstrate such a high degree of variability as to be unreliable in comparison with intra-uterine temperatures 9. Finally, we should attempt to quantify the thermal contributions of the basic elements of heat transfer, that is, radiation, convection, conduction and evaporation. The miracle of Winter is that it yields a fuller appreciation of Spring, when new discoveries are revealed. In terms of understanding the thermal influences of analgesia and anaesthesia in the parturient, we are just scratching the frozen tundra. However, as indicated by the study by Mullington et al. 7, we are slowly, and softly snow-shoeing our way towards progress. No conflict of interest declared.