The effect of general anaesthesia on the mammalian circadian clock and its clinical implications and potential treatment

Abstract

Patients frequently sleep poorly following an operation. This may be due, in part, to the effect of general anaesthesia on the circadian clock. The aim of this thesis was to understand the effect of general anaesthesia on the mammalian circadian clock and sleep, the relevance of this to the post-operative patient, and how these effects may be modified by intra-operative light administration. Previous data from our group’s work on the honey bee revealed that the general anaesthetic isoflurane causes phase delays of the circadian clock as a result of a shift in the expression of core circadian clock genes (cryptochrome and period). In addition, concurrent administration of light during anaesthesia in honey bees resulted in a reduction of the general anaesthesia-induced phase shift. In order to determine whether isoflurane causes robust and reproducible phase shifts in mammals, and whether these effects result from isoflurane acting directly on the suprachiasmatic nuclei (SCN), I developed a system to enable the examination of the effect of isoflurane on PER2::LUC mouse SCNs. PER2::LUC SCN tissue was cultured over eight days (n=36) and exposed to a six hour 2% isoflurane anaesthetic (or air control) after four cycles. Phase shifts and period changes were analysed using the peak expression time (acrophase) on each of the days following anaesthesia or control, compared to the days prior. Administration of a six hour isoflurane anaesthetic to SCNs between circadian times eight and 16 (CT8-CT16) resulted in an average phase delay of -6.99 hours (1.34 SEM) compared to controls at similar CTs that showed an average phase delay of only -1.33 hours (2.77 SEM, p=0.01). Neither isoflurane nor air administered at other CTs elicited significant phase shifts. The results from the SCNs and from the previous work on honey bees paved the way for a clinical trial in which I examined the efficacy of light administration in reducing anaesthesiainduced post-operative sleep and circadian disruption. I investigated this in a non-confounded patient population, kidney donors, who are hospitalised and undergo anaesthesia and surgery that they do not medically require. The work described in this thesis details the interim analysis of 20 patients (total study sample size=40) of whom 10 received phase shifting light and 10 placebo light during the operation. If humans were to respond like bees, then intra-operative bright light administration could represent a treatment strategy for combatting post-operative sleep and circadian misalignment. Actigraphy was recorded for seven pre-operative and seven post-operative days, and circadian markers (core body temperature and the urinary metabolite of melatonin, 6-sulphatoxymelatonin) were recorded for three pre-operative and three post-operative days. Subjective measures were also recorded throughout the study. Participants enrolled into the study were randomised to receive either placebo light or bright light intra-operatively. Operations occurred in the morning, the same subjective time at which anaesthesia elicited the largest phase shifts in SCNs and honey bees. All patients, irrespective of treatment group, showed disrupted sleep and circadian variables following hospitalisation, anaesthesia and surgery, with increased total sleep time and variability, and less stability. Non-parametric circadian rhythm analyses of actigraphic data showed higher interdaily stability (IS) post-operatively in those that received intra-operative bright light compared to those that received placebo light (p=0.08). Furthermore, the control group showed an average phase delay in core body temperature rhythms of -4.18 hours (1.74 SEM) compared to patients in the bright light treatment group who showed an average phase delay of -1.16 hours (0.62 SEM, p=0.07). Results from the interim analysis of the clinical trial indicate a trend towards intra-operative bright light improving sleep and circadian variables post-operatively. Having a general anaesthetic is essential for most major operations. The studies presented in this thesis show that general anaesthesia affects the mammalian (mouse) circadian clock and human circadian rhythms and sleep. In mice this occurs in a time-dependent manner which has not previously been recognised. Moreover, bright light administered intra-operatively may potentially reduce circadian and sleep disruption in the post-operative period.