Participant demographic characteristics
A total of 287 individuals, 96 males and 191 females, were included in this study, and the ratio of isolated TSH elevation was 58.54% (168/287). No obvious difference was found in the ratio of isolated TSH elevation between males and females (59.35% vs. 58.12%, p=0.838). As shown in Table 1, we further divided all 287 subjects into isolated TSH elevation group and normal TSH group, and similarly there was no significant difference in sex between the two groups (p>0.05, Table 1). Nevertheless, no statistically significant differences in FT3 or FT4 were observed between the two groups, as shown in Table 1.
Sleep status and PSQI scores
As shown in Figure 1, the proportions of patients with poor sleep (Group A), good sleep with occasional poor sleep (Group B-1), and persistent good sleep (Group B-2) were 70.24% (118/168), 9.52% (16/168), and 20.24% (34/168), respectively, among patients with isolated TSH elevation and 49.58% (59/119), 1.68% (2/119), and 48.74% (58/119), respectively, among subjects with normal TSH. A significantly higher ratio of poor sleep and good sleep with occasional poor sleep was observed among patients with isolated TSH elevation than normal TSH (70.24% vs. 49.58%, p=0.001; 9.52% vs. 1.68%, p=0.006).
Compared with subjects with normal TSH, patients with isolated TSH elevation had higher scores for multiple PSQI components. We observed significant differences in the PSQI scores of subjective sleep quality, sleep latency, sleep duration, and habitual sleep efficiency between the two groups (all p<0.05, Table 2). However, no significant differences were observed in the other three PSQI component scores of sleep disturbance, need for sleep medications, and daytime dysfunction (all p>0.05, Table 2). The incidence of a later sleep time was significantly higher among patients with isolated TSH elevations than normal TSH (p=0.003, Table 2).
Sleep status and PSQI scores of non-autoimmune thyroiditis subjects
Excluding individuals with chronic autoimmune thyroiditis, the baseline characteristics were not significantly different between the two groups (Table 3, all p>0.05), and there was a higher ratio of poor sleep among patients with isolated TSH elevation than normal TSH [Figure 1, 67.88% (93/137) vs. 50.43% (58/115), p=0.007; 9.49% (13/137) vs. 1.74% (2/115), p=0.014].
Similarly, when excluding individuals with chronic autoimmune thyroiditis, patients with isolated TSH elevation had higher scores for multiple PSQI components, including subjective sleep quality, sleep latency, sleep duration, and habitual sleep efficiency, than subjects with normal TSH. No significant differences in the other three PSQI component scores, namely, sleep disturbance, need for sleep medications, or daytime dysfunction, were found between the two groups (all p>0.05, Table 4). Additionally, the late sleep time incidence was significantly higher among patients with isolated TSH elevation than normal TSH (p=0.003, Table 4).
Follow-up study of patients with isolated TSH elevation
A total of 105 patients, 79 in Group A, 14 in Group B-1, and 12 in Group B-2, were reexamined for thyroid hormone levels until Feb. 13, 2017 (Figure 2).
A total of 48 subjects in Group A slept better than before, and the remaining 31 participants still slept poorly (22 of whom had slept poorly for more than one year). As shown in Table 5 and Figure 2, the concentration of TSH, the proportion of different TSH decline, and the ratio of TSH normalization among the patients who slept better was significantly higher than that among those who still slept poorly (all p<0.05, Table 5 and Figure 2).
A total of 12 subjects in Group B-1 had normal TSH levels when their sleep had improved.
In Group B-2, five patients experienced good sleep but had late sleep times (beyond 00:30). When they fell asleep before 23:00, 4 of the five subjects were reexamined, and normal TSH levels were found.
The follow-up study of non-autoimmune thyroiditis patients with isolated TSH elevations
A total of 79 non-autoimmune thyroiditis patients, 58 in Group C, 11 in Group D-1, and 9 in Group D-2, were reexamined for thyroid hormone levels until Feb. 13, 2017.
A total of 39 patients in Group C slept better than they had previously, and the remaining 19 subjects still slept poorly (22 of whom had slept poorly for more than one year). The ratio of TSH normalization among patients who slept better was significantly higher than that among those who still slept poorly (89.74% vs. 5.26%, p<0.05).
All of the patients in Group D-1 had normal TSH levels after their sleep improved.
In Group D-2, five patients experienced good sleep but had late sleep times (beyond 00:30). When they fell asleep before 23:00, 4 of the five subjects were reexamined, and normal TSH levels were found.