Nine females and five males (mean age, 29.43±8.58 years) were recruited through an advertisement. The exclusion criteria were as follows: 1) females who were pregnant/nursing; 2) a history of serious medical problems; 3) currently taking medications that act on the nervous system, including sleeping pills; 4) DSM-5-based psychiatric diagnosis [20] within the last 3 months; 5) symptoms of a sleep disorder, such as insomnia, snoring, apnea, daytime sleepiness, or restlessness in the legs before bedtime; and 6) shift worker. This study was approved by the Institutional Review Boards of Seoul National University Hospital (approval no. 1806-089-951), and all procedures were performed after obtaining written informed consent from the participants.

All participants completed self-reported questionnaires to obtain demographic and sleep-related information, as well as medical, psychiatric, neurological, and family histories. The self-reported questionnaires included the Pittsburgh Sleep Quality Index (PSQI), Epworth Sleepiness Scale (ESS), Beck Depression Inventory (BDI), Beck Anxiety Inventory (BAI), Dysfunctional Beliefs and Attitudes about Sleep (DBAS), International Restless Legs Syndrome Study Group Severity Scale (IRLSSG), Morningness-Eveningness Questionnaire (MEQ), and the Berlin Questionnaire (BQ).

The PSQI has a total of 19 items, and scores range from 0 to 21 points; higher scores indicate poorer subjective sleep quality [21]. The ESS is useful for evaluating daytime sleepiness, where significant daytime sleepiness is indicated by scores ≥10 points [22]. The BDI comprises 21 items, and scores range from 0 to 63 points (0–9, not depressed; 10–23, moderately depressed; 24–63, severely depressed) [23]. BAI scores range from 0 to 63 points (22–26, anxious; 27–31, severe anxious; ≥32, extremely anxious) [24]. DBAS-16 scores range from 0 to 160; higher scores indicate more dysfunctional beliefs and attitudes toward sleep [25]. The IRLSSG is a standard instrument for measuring the degree of restless legs syndrome scored as follows: 0–10 points, mild severity; 11–20 points, moderate severity; 21–30 points, severity; and 31–40 points, highest severity [26]. The MEQ classifications are as follows: definitely morning, 70–86 points; moderately morning, 59–69 points; intermediate, 42–58 points; moderately evening, 31–41 points; and definitely evening, 16–30 points [27]. The BQ consists of three categories: category 1 concerns snoring, category 2 daytime sleepiness, and category 3 hypertension. A high risk of sleep apnea is indicated by a positive score in two or three categories [28].

In the morning following the each night of nocturnal polysomnography (nPSG), participants completed a survey on subjective sleep quality and their opinion of the sleepwear. They indicated how long it took to fall asleep [sleep latency (SL)], how long they slept for [total sleep time (TST)] in minutes, and the total number of awakenings. Sleep quality was scored from 1 (very tired and sleepy) to 4 points (refreshed).

In addition, participants indicated how deep their sleep was, how long they slept for, how often they woke, how many dreams they had, and how comfortable their sleep was, on a 7-point Likert scale (1: slept very lightly/slept very little/woke up very often/did not dream/very uncomfortable, respectively; 7: slept very deeply/slept for a very long time/did not wake up/had many dreams/very comfortable, respectively). They also scored their overall sleep satisfaction (1 point, very satisfied; 5 points, very dissatisfied) and indicated the reasons for satisfaction/dissatisfaction. The participants also indicated how uncomfortable they were due to the sleepwear, whether they experienced tension and stiffness, the level of discomfort felt in the neck, waist, and back, and the extent of discomfort caused by sleepwear while lying down and moving in bed (i.e., tossing and turning) (1 point, very comfortable; 4 points, very uncomfortable). Finally, they rated the sleepwear in terms of the material, size, tactile sensation, and overall satisfaction (1 point, very satisfied; 5 points, very dissatisfied). The evaluators did not interact with the participants when they were completing the questionnaire.

All participants underwent nPSG for three consecutive nights (adaptation night followed by two nPSG sessions to prevent the first night effect) [29]. The coin flip was used to randomize the order of dressing: when the front side appeared, participants were assigned to wear ES on the second day and CS on the third day; when the back of the coin came out, they were assigned to wear sleepwear in reverse order. They were advised that they could engage in routine activities during the daytime (i.e., outside of the nPSG sessions), but tobacco, alcohol, and caffeine were prohibited for 4 days before the adaptation night. The temperature, humidity, and lighting conditions of the sleep laboratory were kept constant.

ProFusion EEG software (ver. 3.4; Compumedics Ltd., Melbourne, Australia) was used for the nPSG recordings. The electrodes and sensors were attached to the participants in a standard manner. Electroencephalogram, electrooculogram, chin electromyogram, electrocardiogram, snoring sounds, oral and nasal airflow, chest movement, abdominal movement, limb movement, body position, and arterial oxygen saturation were continuously recorded during nighttime sleep. The PSG results were scored based on American Academy of Sleep Medicine (AASM) criteria [30]. Sleep architecture and related variables (time in bed, TST, wake after sleep onset, SL, sleep efficiency, the proportions of N1, N2, N3, and REM sleep, and apnea-hypopnea index, the periodic limb movement index) were analyzed.

The ES and CS were identical so that participants could not identify them (gray short-sleeved tops and long gray pants). The order of wear was randomized. All clothes were supplied by Venex Co., Ltd. (Atsugi, Japan).

Data were analyzed using SPSS software (ver. 25.0; IBM Corp., Armonk, NY, USA). The paired t-test was performed to compare objective and subjective sleep parameters between the CS and ES. Any effect of adaptation was evaluated by comparing the data between the second and third nights. Groups distinguished based on a PSQI cutoff score of 5 were also compared, to explore whether the results differed according to subjective sleep quality. A p-value <0.05 was considered significant.

The demographic and baseline characteristics of all participants are summarized in Table 1. Table 2 shows the mean and standard deviation of objective and subjective sleep parameters for all participants, seven of whom wore the ES first and then the CS; the other seven wore the two types of sleepwear the reverse order.

The objective sleep variables did not differ significantly between the two sleepwear (Table 2). We divided groups using PSQI and performed the same analysis in order to see if these results differed according to the quality of sleep. Among participants with a PSQI score <5 who reported relatively high sleep quality, a tendency toward a longer TST was observed when wearing the ES compared to the CS (ES, 447.44±37.31 min; CS, 433.61±28.39 min; p=0.06) (Table 3). An analysis of the five participants who reported relatively poor sleep quality (PSQI≥5) showed no significant differences in any parameter between the ES and CS.

Participants reported dreaming more frequently (ES, 5.00±1.84; CS, 3.39±1.42; p=0.03), and less discomfort due to tossing and turning (ES, 3.43±0.76; CS, 3.71±0.47; p=0.04) when wearing the ES compared to the CS. No difference was seen in any other variable by sleepwear (Table 2).

The nine participants with a PSQI <5 had more dreams when wearing the ES compared to the CS (ES, 5.39±1.52; CS, 3.50±1.52; p=0.01) (Table 3). Among the five participants with a PSQI ≥5, no significant difference in any factor was seen by sleepwear.

Regardless of the type of sleepwear, N2 (%) stage sleep increased significantly on the third night (second night, 49.18±8.91%; third night, 56.13±7.93%; p=0.01), while for N3 (%) the opposite pattern was seen (second night, 16.60±3.97%; third night, 10.68±5.10%; p<0.001). Participants reported to sleep more comfortably during the third night than the second night (second night, 3.82±1.54; third night, 4.61±1.11; p=0.02).