An investigation into morning light therapy, circadian rhythms and sleep-wake cycles in postoperative cardiac patients

Abstract

Hospitalisation, surgery and anaesthesia have been shown to cause disruption of sleep-wake cycles and circadian rhythms in animal models and in humans. These disturbances may adversely affect the well-being and recovery of patients. In this thesis, I describe a stepwise investigation to determine the effects of the clinical lighting environment on the lengths of hospital stay of postoperative cardiac patients, and to establish whether morning light therapy improves the postoperative disruption of sleep and circadian rhythms in the patients. First, the validity of the ambulatory measurement of light exposure in hospital using wrist-level light monitoring devices was assessed. Wrist-worn devices were found to estimate eye-level light exposure adequately for the purposes of research, when compared to eye-level light monitoring devices. There was agreement between the two devices, with differences of less than 10 lux at eye-level light intensities less than 5000 lux. This is the first study to evaluate the validity of wrist-level monitoring devices for the measurement of light levels in the clinical setting. The results support their continued use for research purposes. I then investigated the relationship between the hospital lighting environment and lengths of stay of patients in a 12-month prospective audit in 654 postoperative cardiac patients. Median daytime light levels in the ward ranged from 8 lux to 406.7 lux. There was no relationship between these light levels and patients’ lengths of stay in the ward (p = .99). Finally, I conducted the first randomised placebo controlled trial of the efficacy of morning light therapy for the amelioration of circadian and sleep disruption in postoperative cardiac patients. Sixty-one cardiac patients were randomised to receive either morning bright light therapy or placebo light therapy, administered between 7:30 a.m. and 9:30 a.m., for three days postoperatively. Sleep-wake cycles (monitored using actigraphy), circadian rhythms (monitored using 6-sulphatoxymelatonin sampling), mood (monitored using the Beck Depression Inventory) and patients’ lengths of hospital stay were assessed. There was no identifiable circadian rhythm in the mean 6-sulphatoxymelatonin excretion rates in the placebo group. In the light therapy group, there were identifiable rhythms in the mean postoperative 6-sulphatoxymelatonin excretion rates on postoperative days two and three (acrophases of 5:35 a.m. and 3:59 a.m., respectively). Postoperatively, placebo patients excreted significantly less 6-sulphatoxymelatonin overnight than they did preoperatively (mean difference 261.1 ng/h, 95 % CI = 18.4-503.9, p = .03). There was no significant difference in preoperative and postoperative excretion rates in the light therapy group (mean difference = 111, CI = -95.1-417.1, p = .7). However, I was unable to detect a significant improvement in postoperative sleep quality or quantity in the light therapy group. There was also no detectable difference in postoperative length of hospital stay and in mood scores between the two groups. These results indicate that in this poorly lit clinical setting, there is no relationship between length of hospital stay and ambient light levels. Morning light therapy may entrain the circadian rhythms of postoperative cardiac patients, but the results did not translate into obvious clinically important benefits. Further investigation of this therapy is warranted.