Data were first analyzed considering the habitual body postures
(i.e., sitting during the day and supine at night). The 24-hour profile
of the mean IOP for all 21 subjects is presented in
Figure 1 . The lowest IOP occurred at the last measurement in the light/wake
period (9:30 PM), and the highest IOP appeared at the first measurement
in the dark period (11:30 PM). The difference between the trough and
the peak was 8.6 ± 0.8 mm Hg (mean ± SEM,
n = 21;
P < 0.001, paired
t-test). The mean
supine IOP in the dark period, 23.1 ± 0.7 mm Hg, was
significantly higher (
P < 0.001, paired
t-test) than the mean sitting IOP in the light/wake period
(16.9 ± 0.5 mm Hg). The four IOP mean values in the dark period
were relatively close to each other, and a sharp reduction occurred
around the lights-on transition (7 AM). During the light/wake period,
mean IOP was high in the early morning and continuously decreased
toward the end of the light/wake period.
When the cosine fits of each individual’s IOP data (
Fig. 2 ; solid circles) were examined, acrophases appeared between midnight and
8 AM for 20 of the 21 subjects. One subject had an unusual acrophase
near 6 PM and a rather small amplitude of the cosine fit. The mean
acrophase was at 4:14 AM (± 0.79 hour). Visual inspection of
Figure 2 indicated that the acrophases were not random around the clock. The
null hypothesis of a random circular distribution of group acrophases
was rejected by the Rayleigh test (
P < 0.001).
Therefore, a significant 24-hour rhythm of IOP was detected. The mean
amplitude of the fitted cosine was 3.6 ± 0.4 mm Hg.
During the light/wake period, the mean supine IOP was 21.7 ± 0.5
mm Hg (
n = 21). It was significantly higher
(
P < 0.001, paired
t-test) than the mean
sitting IOP of 16.9 ± 0.5 mm Hg, obtained at comparable times. To
exclude the postural influence, 24-hour IOP profile was examined again
using only the supine IOP data. Among the 12 time points, the trough
and peak IOP occurred at 9:30 and 11:30 PM
(Fig. 1) , the same trough
and peak times found when sitting IOP data from the light/wake period
were considered. The trough-peak IOP difference in supine observations
was 3.4 ± 0.7 mm Hg (
P < 0.001, paired
t-test), which was smaller than the trough-peak difference
when sitting-supine IOP data were analyzed. A paired
t-test
showed that the average supine IOP in the 8-hour dark period was not
significantly higher than the average supine IOP in the 16-hour
light/wake period (
P > 0.05,
n = 21).
However, when IOP data were divided into three 8-hour time blocks of 7
AM to 3 PM, 3 PM to 11 PM, and 11 PM to 7 AM, repeated-measures ANOVA
showed a significant difference among the 3 IOP averages of these time
blocks. Post-hoc Bonferroni multiple comparisons indicated that the
only significant difference (an increase) was between the late
light/wake period (3–11 PM) and the dark period (11 PM–7 AM).
Cosine fits of the 24-hour supine IOP data showed that 19 of 21
acrophases appeared between 1 AM and 3 PM (
Fig. 3 ; solid triangles), and the mean acrophase was at 8 AM (±1.17 hour).
Once postural influences were removed these acrophases were more
dispersed than those in
Figure 2 . The Rayleigh test did not reject the
null hypothesis of a random circular distribution of acrophases
(0.1 >
P > 0.05). This indicated that there was
no significant 24-hour IOP rhythm of cosine shape when only the supine
IOP data were considered. The mean amplitude of these cosine fits was
2.1 ± 0.3 mm Hg.