Endogenous circadian rhythm in vasovagal response to head-up tilt

Abstract

Background: The incidence of syncope exhibits a daily pattern with more occurrences in the morning, possibly as a result of influences from the endogenous circadian system and/or the daily pattern of behavioral/emotional stimuli. This study tested the hypothesis that the circadian system modulates cardiovascular responses to postural stress, leading to increased susceptibility to syncope at specific times of day.

Methods and results: Twelve subjects underwent a 13-day in-laboratory protocol in which subjects' sleep-wake cycles were adjusted to 20 hours for 12 cycles. A 15-minute tilt-table test (60° head-up) was performed ≈4.5 hours after scheduled awakening in each cycle so that 12 tests in each subject were distributed evenly across the circadian cycle. Of 144 tests, signs/symptoms of presyncope were observed in 21 tests in 6 subjects. These presyncope events displayed a clear circadian rhythm (P=0.028) with almost all cases (17/21) occurring in the half of the circadian cycle corresponding to the biological night (10:30 pm to 10:30 am). Significant circadian rhythms were also observed in hemodynamic and autonomic function markers (blood pressure, heart rate, epinephrine, norepinephrine, and indices of cardiac vagal tone) that may underlie the circadian rhythm of presyncope susceptibility.

Conclusions: The circadian system affects cardiovascular responses to postural stress, resulting in greater susceptibility to presyncope during the night. This finding suggests that night-shift workers and people with disrupted sleep at night may have greater risk of syncope as a result of their exposure to postural stress during the biological night.

Figure 1

A Graphical representation of the forced desynchrony protocol. Solid black areas indicate scheduled sleep, light-grey is wakefulness in dim light (~1.8 lux), hatched is wakefulness on baseline days in normal room light (~90 lux), and dark-grey bars indicate scheduled tilt tests.

Figure 2

Systolic blood pressure (SBP), diastolic blood pressure (DBP), heart rate (HR), stroke volume (SV), cardiac output (CO) and total peripheral resistance (TPR) estimated via finger plethysmography in one subject during a tilt test with presyncope. Data are shown for the 3-minute baseline, the tilt test (light-grey region), and the 3-minute recovery period. Two dark-grey highlighted regions are the transitions during tilting up and tilting down. Data are missing at ~15-45 seconds due to a recalibration. The sequence of changes during tilt showed four distinct phases, separated by dashed vertical lines: I (0-555 seconds). Immediately upon tilting HR increased and SV decreased while SBP, DBP and TPR were relatively stable. All variable were relatively stable during this phase; II (555-630 seconds). TPR, SBP and DBP were steadily decreasing with maintained HR, SV and CO; III (630-672 seconds). SV and CO decreased abruptly, TPR increased rapidly and then decreased steadily, HR was maintained, and SBP and DBP were steadily decreasing; IV (672-710 seconds). SBP, DBP, HR and CO decreased abruptly with no compensatory increase in SV. Note that TPR continued decreasing precipitously at the beginning of the transition of tilting down when HR increased/recovered quickly. After returning to the supine posture TPR quickly recovered and remained elevated above baseline for ~10 minutes (3 minutes shown); HR and DBP normalized within 30 seconds; and SBP, SV and CO normalized after ~ 2 minutes.

Figure 3

Circadian variations in susceptibility to presyncope. A. SBP recordings (finger plethysmography) of a representative subject with presyncope. Data are shown for the 10 minutes of baseline, the tilt test (grey highlighted region), and the recovery period. Four presyncope events (Cycles 2, 8-10) occurred at 300° (1 case), 0° (1 case), and 60° (2 cases). All presyncope cases were associated with precipitous SBP drops. B. Probability of presyncope occurrences across all circadian phases. Results were obtained from the generalized linear mixed model (Supplemental Table I), and are double plotted to better visualize rhythmicity with circadian phase on the lower abscissa and the corresponding habitual time of day on the upper abscissa. Error bars are standard errors. Grey bars indicate the average habitual sleep period in home environment.

Figure 4

Physiological responses to head-up tilt, and their differences between the non-presyncopal (circles) and the presyncopal (squares) groups. Data are presented as Mean±SE. Shown are P values for tilt effects, mean group differences, and the interaction between group and tilt stressor. Results were obtained from the mixed model ANOVAs (Supplemental Table II). Here “NS” indicates P >0.1.

Figure 5

Physiological responses to head-up tilt, and their differences between 51 completed trials (squares) and 21 aborted trials (triangles) within the presyncopal group. Data are presented as Mean±SE, where Mean was obtained by averaging the individual means for presyncopal trials and non-presyncopal trials separately and SE indicates between-subject error. Shown are P values for the effects of presyncope and its interaction with tilt effect. Results were obtained from the mixed model ANOVAs (Supplemental Table III). Here “NS” indicates P >0.1.

Figure 6

Circadian influences on indices of hemodynamics and autonomic activity and their responses to head-up tilt. The data (symbols) and the cosinor fits (lines) are plotted separately for baseline (squares and continuous lines) and head-up tilt (circles and dashed lines). Grey bars indicate the average habitual sleep period in home environment. Data are presented as Mean±SE across subjects. Shown are the P values for circadian influences and interaction between tilt and circadian influences. Results were obtained from the cosinor analyses using mixed-model ANOVAs (see Supplemental Methods) and are double plotted to better visualize rhythmicity with circadian phase on the lower abscissa and the corresponding habitual time of day on the upper abscissa.