Progressive retinal ganglion cell loss in primary open-angle glaucoma is associated with temperature circadian rhythm phase delay and compromised sleep

Advanced primary open-angle glaucoma (POAG) is characterized by progressive retinal ganglion cell complex (RGCC) damage that may cause subsequent disruption of the circadian rhythms. Therefore, we evaluated circadian body temperature (BT) rhythm and sleep characteristics of 115 individuals (38 men and 77 women) diagnosed with POAG. GLV (global loss volume; %), a measure of RGCC damage, was estimated by high-definition optical coherence tomography, and RGC functional ability was assessed by pattern electroretinogram amplitude (PERGA). Depending on dynamics of POAG progression criteria, two groups were formed that were distinctively different in GLV: Stable POAG group (S-POAG; GLV = 5.95 ± 1.84, n = 65) and Progressive POAG group (P-POAG; GLV = 24.27 ± 5.09, n = 50). S-POAG and P-POAG groups were not different in mean age (67.61 ± 7.56 versus 69.98 ± 8.15) or body mass index (24.66 ± 3.03 versus 24.77 ± 2.90). All subjects performed 21 around-the-clock BT self-measurements during a 72-h period and kept activity/sleep diaries. Results showed pronounced disruption of circadian physiology in POAG and its progression with increasing severity of the disease. The daily mean of BT was unusually low, compared to age-matched controls. Moreover, our results revealed distinctive features of BT circadian rhythm alterations in POAG development and POAG progression. S-POAG is associated with lowered BT circadian rhythm robustness and inter-daily phase stability compared to controls. In the P-POAG group, the mean phase of the circadian BT rhythm was delayed by about 5 h and phases were highly scattered among individual patients, which led to reduced group mean amplitude. Circadian amplitudes of individuals were not different between the groups. Altogether, these results suggest that the body clock still works in POAG patients, but its entrainment to the 24-h environment is compromised. Probably because of the internal desynchronization, bedtime is delayed, and sleep duration is accordingly shortened by about 55 min in P-POAG compared to S-POAG patients. In the entire POAG cohort (both groups), later sleep phase and shorter mean sleep duration correlate with the delayed BT phase (r = 0.215; p = 0.021 and r = 0.322; p = 0.0004, respectively). An RGCC GLV of 15% apparently constitutes a threshold above which a delay of the circadian BT rhythm and a shortening of sleep duration occur.