Nocturnal sleep, daytime sleepiness, and napping among women with significant emotional behavioral premenstrual symptoms

Transcription

1 J. Sleep Res. (2007) 16, Nocturnal sleep, daytime sleepiness, and napping among women with significant emotional behavioral premenstrual symptoms LYNNE J. LAMARCHE 1, HELEN S. DRIVER 2, SABRINA WIEBE 1, LEAH CRAWFORD 1 and JOSEPH M. DE KONINCK 1 1 Sleep Laboratory, School of Psychology, University of Ottawa, Ottawa, ON, 2 Departments of Medicine and Psychology, Kingston General Hospital, QueenÕs University, Kingston, ON, Canada Accepted in revised form 20 June 2007; received 6 February 2007 SUMMARY The objective of this study is to examine daytime sleepiness and alertness and nap characteristics among women with significant emotional behavioral premenstrual symptoms, and to determine their relationship with nocturnal sleep. Participants spent one night during the follicular phase and two nights during the late-luteal phase, one of which occurred after a 40 min opportunity to nap, sleeping in the laboratory. Subjective measures of sleepiness and alertness were completed during the afternoon of each recording. Setting took place at the sleep laboratory at the University of Ottawa. A total number of participants were 10 women with significant and nine women with minimal emotional behavioral premenstrual symptoms (mean age 26 years). The results were compared with the follicular phase, both groups of women had less slow wave sleep and more stage 2 sleep at night, as well as a higher daytime and nocturnal mean and maximum temperature during the late-luteal phase. Women with significant symptoms were sleepier and less alert during the late-luteal phase and had a higher overall mean nocturnal temperature compared with women with minimal symptoms. No significant differences were found between the two groups on nap characteristics and nocturnal sleep characteristics. Results show that women with more severe premenstrual symptoms are sleepier during the late-luteal phase than women with minimal symptoms. The increased daytime sleepiness seems to be unrelated to nocturnal sleep or nap characteristics. keywords daytime sleepiness, napping, nocturnal sleep, premenstrual syndrome INTRODUCTION Sleep difficulty has been reported by up to 42% of healthy middle-aged women (Owens and Matthews, 1998) and complaints may be related to different phases of the menstrual cycle. Premenstrual symptoms affect between 10% (Mitchell et al., 1991) and 80% (Hamilton et al., 1984) of menstruating women. Women with severe premenstrual symptoms report sleep-related complaints (Hurt et al., 1992; Mauri et al., 1988) yet few studies have investigated the nature of these sleep disturbances. Studies that have compared sleep during the luteal and follicular phases have conflicting and equivocal findings. On Correspondence: Joseph De Koninck, Sleep Laboratory, School of Psychology, University of Ottawa, 125 University, Ottawa, ON K1N 6N5, Canada. Tel.: (ext.) 4315; fax: ; jdekonin@uottawa.ca balance, based on retrospective surveys and prospective subjective reports using diaries, women across a wide age range (18 50 years) report more sleep disturbances during the premenstrual week than at other times (Manber and Bootzin, 1997; Patkai et al., 1974), such as increased sleep onset latency (SOL) and number of awakenings (Manber and Bootzin, 1997) and decreased sleep efficiency (Manber and Bootzin, 1997). While some polysomnographic studies find a decrease in the amount of SWS (Cluydts, 1980; Ito et al., 1995) and an increase in stage 1 sleep (Parry et al., 1999) in the luteal phase compared with the follicular phase, other researchers report that differences in sleep parameters in relation to the phases of the menstrual cycle are minimal (Armitage and Yonkers, 1994; Driver et al., 1996; Lee et al., 1990). The severity of premenstrual symptoms experienced by women may have an adverse impact on their sleep architecture. 262 Ó 2007 European Sleep Research Society

2 Sleep among women with premenstrual symptoms 263 Women with more severe symptoms have reported more unpleasant dreams and a lower quality of sleep in the luteal phase (Mauri et al., 1998), and polysomnographic data from Lee et al.õs (1990) study show that they have a lower percentage of SWS and more stage 2 sleep in both the luteal and the follicular phase compared with women with minimal symptoms, as well as shorter SOL during the luteal phase. A smaller amount of rapid eye movement (REM) sleep throughout the menstrual cycle has also been noted among women with moderate-to-severe premenstrual depression, compared with women without significant premenstrual mood disturbances (Parry et al., 1989). However, other studies using objective (Chuong et al., 1997; Parry et al., 1999) and subjective (Manber and Bootzin, 1997) measures have failed to find a relationship between the severity of premenstrual symptoms and sleep disturbance. There may be a relationship between sleep and changes in body temperature during the different phases of the menstrual cycle (Driver et al., 1996) as well as with severity of premenstrual symptoms. For example, an increase in temperature during the luteal phase may be related to a poorer sleep quality at this time (Monroe, 1967) and reduced REM sleep time (Baker et al., 1999). Women with premenstrual symptoms have been found to have significantly higher mean temperatures as well as higher nocturnal minimum temperatures, but not daytime temperatures, throughout the entire menstrual cycle compared with women without premenstrual symptoms (Severino et al., 1991). Consistent with this finding, Parry et al. (1997) found that temperature mesors were higher during the luteal phase in women with more severe premenstrual symptoms compared with women with minimal symptoms. In addition to disturbed nocturnal sleep, women tend to have an increase in subjective daytime sleepiness during the late-luteal phase compared with the follicular phase (Manber and Bootzin, 1997; Shibui et al., 2000). The late-luteal phase daytime sleepiness (Manber and Bootzin, 1997) and fatigue (Mauri et al., 1998) have been found to be higher in women with more severe premenstrual symptoms, which has lead to the suggestion that daytime sleepiness may reflect an underlying need for more sleep among women with significant premenstrual symptoms (Manber and Bootzin, 1997). Alternatively they may have a more disturbed sleep (Mauri et al., 1998) or have a circadian phase-advance (Parry, 1993). In turn, more severe symptoms may decrease oneõs ability to effectively deal with disrupted sleep (Manber and Bootzin, 1997). Naps are quite common among healthy middle-aged women (Owens and Matthews, 1998). Naps are often used in research to measure levels of daytime sleepiness (Pollak, 1997), by examining the time it takes to fall asleep during a napping opportunity (Hartse et al., 1982). Shibui et al. (2000), in a study that investigated sleep duration within 10-min naps distributed across a 26-h period, found that subjective daytime sleepiness and the occurrence of SWS during the daytime naps in the luteal phase were increased, suggesting an increased demand for daytime SWS, compared with the follicular phase. The change in amount of SWS during the daytime was positively correlated with change in mean daily core body temperature. They suggested that there is an increased need for SWS in the luteal phase of the menstrual cycle, independent from nocturnal sleep. No studies have examined nocturnal sleep and nap characteristics among women with more severe premenstrual symptoms in an attempt to better understand the increase in daytime sleepiness noted among these women during the lateluteal phase. Given the findings that women with more severe symptoms have more disturbed nocturnal sleep and more daytime sleepiness during the late-luteal phase compared with women with minimal symptoms, we hypothesized that the severity of premenstrual symptoms is related to a greater need for sleep at this time. We therefore predicted that women with significant emotional behavioral premenstrual symptoms would have increased daytime sleepiness, a shorter SOL, a shorter SWS latency and more SWS during a late-luteal daytime nap compared with women with minimal symptoms. METHODS Participants Advertisements were posted at the University of Ottawa, including the campus health clinic, as well as at a Premenstrual Clinic of the Ottawa Gatineau region. In addition, an advertisement was placed in the Ottawa Sun newspaper. The final sample consisted of 19 women between the ages of 20 and 37 years. More specifically, 10 women (mean age 26.7 years, SD = 4.99) had significant emotional behavioral premenstrual symptoms, and nine women (mean age 26.6 years, SD = 6.62) had minimal to no emotional behavioral premenstrual symptoms. The Premenstrual Syndrome Self-Rating Scale (PMTS Self-Rating Scale) (Steiner et al., 1980), a measure with good reliability and validity (Steiner et al., 1999), was used to assess the severity of premenstrual symptoms and categorize women into the severe or minimal symptom groups. The PMTS Self-Rating Scale consists of a 36 yes no answer questions related to premenstrual tension syndrome (Steiner et al., 1999). Thirty-two of the questions relate to the presence of emotional behavioral symptoms, and four to physical symptoms (Haskett and Abplanalp, 1983). Based on the 32 emotional behavioral items only, women with a score of 14 or more during the late-luteal phase and a score of 5 or less during the follicular phase, were identified as experiencing significant premenstrual symptoms (Haskett and Abplanalp, 1983); a luteal phase score of 8 or lower identified women with minimal to no emotional behavioral premenstrual symptoms (Steiner et al., 1999). Women who qualified for the significant symptoms group scored on average 24.3 (SD = 4.90) during the late-luteal phase and 1.30 (SD = 1.25) during the follicular phase on the PMTS scale, while women in the minimal symptoms group scored on average 4.6 (SD = 3.24) during the late-luteal phase and 1.88 (SD = 2.95) during the follicular phase. Subject categorization remained consistent throughout the study, as measured by

3 264 L. J. Lamarche et al. the PMTS scale. It is important to note that participants were not diagnostically assessed for premenstrual syndrome (Steiner, 2000) or premenstrual dysphoric disorder (PMDD) (Freeman, 2003), which requires the prognostic presence of symptoms for a minimum of 2 months. The majority of the participants in this study (68%) were single. English or French was the first language of 63% of the sample. Most participants (95%) had at least some university or college education, and most were either students (68%) or employed (32%) at the time of the study. The majority of participants (63%) took a nap on a regular basis, while three participants from each group reported that they never napped. None of the participants was pregnant, or had been pregnant within the last year. Participants were not taking psychotropic medication or prescribed medication for their premenstrual syndrome. They were not taking any contraceptive medication, and had not been for at least 4 months prior to the beginning of the study. Participants did not have any history of, or current sleep disorders, or significant health or psychologic problems. Participants who were suffering from dysmenorrhoea were excluded from the study. Long (10 h or more of sleep) and short (6 h or less of sleep) sleepers, as well as women who went to bed after12:30 hours and woke up after 8:30 hours were also excluded from the study so as to ensure that the timing of the nap occurred in the mid-afternoon. All participants had regular sleep wake patterns. While body mass index (BMI) was not measured all participants were visually observed to be of normal weight. Prospective participants completed a number of questionnaires to determine their inclusion in the study, including the General Health Questionnaire, General Sleep Questionnaire, Pittsburgh Sleep Quality Index (PSQI) (Buysse et al., 1989), Sleep Log, Beck Depression Inventory (BDI) (Beck et al., 1961), as well as a questionnaire regarding the regularity of the menstrual cycles and the presence of dysmenorrhoea. The average length of the menstrual cycle was 28.5 days (SD = 1.89) for women with significant symptoms and days (SD = 1.06) for women with minimal symptoms. Participants with a score of 13 or lower (minimal to mild range) on the BDI were considered for the study. Qualifying participants then recorded their oral temperature for 3 4 weeks at approximately the same time every morning to track the cycle and to ensure they had ovulatory cycles, as evidenced by an increase in temperature following ovulation or a dip in temperature before this increase (Marshall, 1963). Women who did not have a regular (two participants) and ovulatory cycle (one participant), based on the biphasic temperature, were excluded from the remainder of the study. Participants were also asked to complete a questionnaire on their menstrual cycle and the PMTS Self-Rating Scale once during follicular and once during late-luteal phase to confirm the regularity of their cycle (consistent cycle lengths), group categorization, and absence of dysmenorrhoea. All women were also asked to complete a daily Sleep Log during this time to monitor regularity of sleep wake patterns. Informed consent was obtained from all women and a small monetary compensation was given for participation in the study. All procedures used in this study were approved by the Ethics Committee of the University of Ottawa ( ), and the treatment of participants was in accordance with the ethical standards of the Canadian and American Psychological Associations. Measures Measures for sleep and nap recordings and temperature Standard polysomnographic recordings (Rechtschaffen and Kales, 1968) were made for the night and nap recordings and included electroencephalogram (EEG), electro-oculogram, and electromyogram (EMG). During the adaptation night, women were screened for movement disorders by the inclusion of anterior tibialis EMG recording, and oral and nasal airflow was monitored with the use of an airflow sensor respiratory device to screen for sleep breathing disorders. Sleep recordings were scored using the Stellate Reviewer Eclipse program (Eclipse Ò Stellate Systems, Westmount, QC, Canada). Twenty-second epochs were scored according to standard criteria (Rechtschaffen and Kales, 1968). One consecutive minute of stage 1 sleep or the appearance of sleep other than stage 1 was considered to be sleep onset, as this criterion is used in recent research on napping (i.e. Brooks and Lack, 2006). Seven nocturnal EEG records were visually scored by 2 raters, with 92% inter-rater reliability. The remaining 50 records were scored by one rater. Digital oral thermometers (Formedica, Montreal, Canada) were used to measure oral temperature during screening, as well as throughout the study. An ambulatory rectal temperature monitoring apparatus and wrist activity recorder (Ambulatory Monitoring, Inc., version 3.01, New York, USA) was also used to measure and record core body temperature at every minute. A detailed analysis of the circadian distribution of core body temperature and nap and nocturnal sleep characteristics will be the subject of a separate report. Questionnaires The Stanford Sleepiness Scale (SSS) (Hoddes et al., 1972), consisting of seven statements from which participants identify current state of sleepiness, was used to measure subjective changes in sleepiness over any period of time (Hoddes et al., 1973). The Subjective Alertness Scale (SAS), a 10 cm Visual Analog Scale for alertness, identified the current level of alertness. Procedure Qualified participants first spent one night sleeping in the laboratory for adaptation and screening purposes. The women spent one night during the follicular phase and two nights with one napping session during the late-luteal phase sleeping in the laboratory; the two late-luteal phase nights and nap were

4 Sleep among women with premenstrual symptoms 265 counterbalanced. Nocturnal sleep was scheduled according to their habitual bedtime, and participants were asked to keep a regular sleep wake cycle for at least one week before the recordings. Participants were asked to refrain from consuming caffeine, drugs, and alcohol during the days of the sleep recordings, and were asked to withdraw from caffeine and nicotine 1 week prior to any sleep recordings. They were asked to refrain from napping outside of the laboratory during the days of the nocturnal recordings. The follicular nocturnal recording was performed 5 9 days after the onset of menses (average 7.3 days, SD = 1.37). Participants were asked to complete the SSS, SAS, and PMTS Self-Rating Scale during the mid-afternoon of that day. The timing of the two late-luteal phase recordings nap condition and no nap condition were calculated by counting back 1 6 days before the expected start of menstrual bleeding (one control participantõs study was 6.5 days before the onset of menses for the nap condition). The expected date of menstrual bleeding was determined based on regularity of the cycle and oral temperature, and was scheduled days (Marshall, 1963) after the determined time of ovulation to estimate onset of menstrual bleeding. The nap condition took place on average 3.4 days before the onset of menses (SD = 1.50); the no nap condition took place on average 3.3 days before onset of menses (SD = 1.66). For the nap condition, participants arrived at the sleep laboratory in the mid-afternoon. The nap was scheduled approximately 12 h after the mid-point of the womanõs habitual nocturnal sleep time (range of start of nap: between 13:00 and 15:45 hours) based on the low level of alertness in the mid-afternoon being approximately 180 out of phase with the minimum level of alertness during the night (Broughton, 1975). After completing the PMTS Self-Rating Scale, SSS and SAS (range of start of questionnaire completion: between 12:30 and 15:15 hours), they were allowed to take a nap. If the participant did not fall asleep (1 full minute of stage 1 or the appearance of any other stage) within 40 min, they were asked to get up. If the participant fell asleep within 40 min, the nap was ended 30 min later. Participants returned to the sleep laboratory that same night for a nocturnal sleep recording. For the no nap condition, participants came to the laboratory at the same time in the afternoon as during the napping condition, but remained awake. They sat and performed a quiet activity for this period of time (i.e. reading, studying), and were monitored to ensure that they did not nap. The same questionnaires were administered and participants returned to the sleep laboratory that night for nocturnal recording. In the case that participants completed the nap and no nap conditions during the same cycle (10 participants), at least 1 day separated the two conditions. Because of some unpredictability of cycles and time constraints, the nap and no nap conditions were sometimes separated by a month or a few months. In both of these cases, the PMTS scale was administered during the nap and no nap condition, to again ensure group categorization. Statistical analyses To compare the level of daytime sleepiness and alertness between women with significant and minimal premenstrual syndrome, independent t-tests were conducted for the SSS and SAS in the afternoon of the follicular and late-luteal phase, using the average of the nap and no nap late-luteal condition for the latter. To determine whether a difference existed in the level of sleepiness and alertness between the follicular and lateluteal phases of the cycle, paired t-tests were conducted. Independent t-tests were conducted to compare the nap characteristics of women with significant emotional behavioral premenstrual syndrome and women with minimal symptoms. A series of 2 (group) 2 (phase of cycle) mixed anovas were conducted to compare nocturnal sleep characteristics during the late-luteal (nap and no nap conditions) and follicular phases of the cycle, among both groups of women. RESULTS Daytime sleepiness and alertness in the late-luteal and follicular phase Based on the SSS and SAS, women with significant symptoms were more sleepy and less alert during the late-luteal phase compared with women with minimal symptoms, t (36) = 3.49, P < and t (36) = 3.02, P < 0.01 respectively (Figs 1 and 2). The two groups of women did not significantly differ in their level of sleepiness or alertness during the follicular phase, t (15) = 0.69, P = 0.50 and t (10.70) = 0.22, P = 0.83 respectively. Compared to during their follicular phase, women with significant premenstrual syndrome were marginally more sleepy and significantly less alert during the lateluteal phase t (8) = 2.26, P = 0.05, t (8) = 3.59, P < 0.01 whereas there was no significant difference between phases for women with minimal symptoms. SSS Follicular Significant Minimal Late-Luteal Menstrual cycle phase Figure 1. Mean (±SD) Stanford Sleepiness Scale (SSS) score for women with significant and minimal premenstrual symptoms during the follicular and late-luteal phases of the menstrual cycle.

5 266 L. J. Lamarche et al. SAS Late-luteal nap characteristics Significant Minimal Follicular Late-Luteal Menstrual cycle phase Figure 2. Mean (±SD) Subjective Alertness Scale (SAS) score for women with significant and minimal premenstrual symptoms during the follicular and late-luteal phases of the menstrual cycle. All the women, with the exception of one participant with minimal symptoms, fell asleep during the napping period. As shown in Table 1, the mean SOL for both groups was less than 10 min. Six women from each group entered SWS, and two women with significant premenstrual syndrome entered REM sleep. No significant group differences were found for any of the nap characteristics. Late-luteal versus follicular nocturnal sleep characteristics Compared to the follicular phase, as shown in Table 2, in the late-luteal phase night (nap and no nap conditions) both study groups had less stage 1 and SWS, but more stage 2 sleep. The proportion of REM sleep during the late-luteal (no nap) condition compared with the follicular phase was reduced yet with more REM periods and a reduced latency to REM sleep that was almost significant, [phase effect, F(1,17) = 3.96, P = 0.06, n 2 = 0.19]. No significant phase or group effects or interactions between phase and group were found for any other characteristics. No differences were found between the two groups on any of the nocturnal sleep variables. DISCUSSION As hypothesized, women with significant emotional behavioral symptoms had more subjective daytime sleepiness and less alertness than women with minimal symptoms during the late-luteal phase. This increased sleepiness was despite both groups having less SWS and stage 1 sleep but more stage 2 sleep during the late-luteal phase nocturnal sleep compared with the follicular phase, but no difference in nap sleep characteristics between the two groups. This finding offers support for the proposal that the increase in daytime sleepiness in women with more severe premenstrual symptoms is unrelated to a disturbed nocturnal sleeping pattern (Manber and Bootzin, 1997; Shibui et al., 2000). These results are also consistent with other objective (Chuong et al., 1997; Parry et al., 1999) and subjective (Manber and Bootzin, 1997) studies failing to find a relationship between the severity of premenstrual symptoms and sleep disturbance. All the women, with the exception of one participant from the minimal symptoms group, fell asleep during the napping period. On the nap, the mean SOL was <8 min for both groups (5.7 min for the severe symptoms group and 7.0 min for the minimal symptoms group). This short sleep onset suggests a high sleep need for both groups of young women during the late-luteal phase, possibly indicating chronic sleep restriction as may be expected in this age group (Klerman and Dijk, 2005). Based on this study, we cannot establish whether this is a menstrual phase specific sleep need because the nap characteristics of women were not studied during the follicular phase. It is also recognized that a more stringent criterion for scoring SOL may have lead to longer latencies among both Table 1 Mean and SD of nap characteristics for 30 min from sleep onset in a 40-min nap opportunity for women with significant and minimal premenstrual symptoms Characteristic Significant symptoms (n = 10) Minimal symptoms (n = 8)* t P-value df SOL 5.7 (5.0) 7.0 (5.1) ) Total sleep time (min) 27.3 (4.3) 25.7 (7.7) ) Stage 1 (min) 5.3 (2.8) 5.3 (2.8) Stage 2 (min) 13.2 (6.1) 13.6 (6.9) ) SWS (min) 8.6 (10.0) 7.5 (8.1) REM (min) 1.0 (2.2) n = 2 0 (0.0) n = Sleep efficiency 89.9 (10.3) 89.3 (6.8) No. awakenings 2.7 (2.4) 2.9 (2.1) ) Latency stage 2 (min) 2.8 (3.4) 2.9 (4.0) ) Latency SWS (min) 10.0 (6.5) n = (5.8) n = 6 ) Latency REM (min) 8.7 (7.6) N A N A N A N A SOL, sleep onset latency; SWS, slow wave sleep; REM, rapid eye movement. *One participant from the minimal symptoms group was not included in these analyses because she was not able to nap.

6 Sleep among women with premenstrual symptoms 267 Table 2 Mean and SD for selected nocturnal sleep characteristics in the late-luteal and follicular phase for women with significant (n = 10) and minimal (n = 9) premenstrual symptoms Significant Minimum Follicular Late-luteal (nap) Late-luteal (no nap) Follicular Late-luteal (nap) Late-luteal (no nap) Total sleep time (min) (36.6) (35.6) (37.3) (41.8) (33.2) (27.7) SOL (min) 4.2 (3.4) 12.1 (16.3) 4.2 (2.1) 9.2 (7.8) 13.5 (21.0) 6.5 (4.2) % Stage 1* 8.0 (2.6) 6.6 (1.8) 6.3 (1.7) 8.6 (2.0) 7.4 (1.2) 7.0 (1.6) % Stage (3.3) 51.0 (5.4) 51.8 (4.1) 44.6 (6.1) 50.0 (3.8) 50.5 (5.6) % SWS à 21.8 (5.0) 19.6 (6.6) 20.4 (3.1) 23.8 (5.3) 20.6 (4.7) 20.9 (4.7) % REM 23.9 (4.5) 22.8 (3.6) 21.5 (3.9) 23.0 (4.5) 21.9 (4.2) 21.6 (3.3) Latency stage 2 (min) 3.1 (4.8) 1.7 (2.1) 2.7 (2.2) 3.3 (3.4) 2.5 (2.9) 2.6 (2.7) Lat. SWS (min) 13.2 (6.1) 13.2 (2.9) 12.3 (5.6) 13.3 (5.5) 13.1 (5.0) 19.3 (19.2) Lat. REM (min) 96.6 (35.8) 91.4 (45.6) 76.7 (25.3) 97.3 (36.4) 72.0 (13.7) 82.6 (32.3) Sleep efficiency (%) 95.0 (2.8) 94.8 (2.1) 92.6 (6.0) 92.8 (6.7) (1.2) 95.5 (2.7) No. awakenings 21.3 (10.1) 19.0 (9.5) 19.6 (8.0) 17.7 (4.9) 20.1 (6.8) 20.9 (6.5) No. REM 4.2 (0.8) 4.2 (1.0) 4.5 (0.9) 3.8 (0.8) 4.6 (0.7) 4.6 (0.5) SOL, sleep onset latency; SWS, slow wave sleep; REM, rapid eye movement. *P 0.05 late-luteal phase (nap and no nap condition) compared to follicular phase, F(1,17) = 8.35, P < 0.05, n 2 = 0.33 and F(1,17) = 11.13, P < 0.005, n 2 = P 0.05 late-luteal phase (nap and no nap condition) compared to follicular phase, F(1,17) = 18.22, P < 0.005, n 2 = 0.52 and F(1,17) = 22.44, P < 0.001, n 2 = à P 0.05 late-luteal phase (nap and no nap condition) compared to follicular phase, F(1,17) = 5.04, P < 0.05, n 2 = 0.23 and F(1,17) = 4.56, P < 0.05, n 2 = P 0.05 late-luteal phase (no nap condition) compared to follicular phase, F(1,17) = P 0.05, n 2 = 0.24 and F(1,17) = 4.76, P < 0.05, n 2 = groups of women. There were no significant differences in the nap sleep time between the two groups though it is noteworthy that two of the women with severe premenstrual syndrome entered REM sleep within the 30-min sleep opportunity. Variations in sleep duration may reflect differences in either the sleep homeostatic process or the circadian process (Klerman and Dijk, 2005), and these variations may reflect differences in the rate of build-up for sleep ÔneedÕ during wakefulness, its rate of dissipation during sleep, the amount of sleep people select to take, or modulation by circadian, environmental, behavioural or social factors (Klerman and Dijk, 2005). Nocturnal sleep duration has been subjectively reported to be longer but more disturbed during the late-luteal phase compared with the follicular phase (Patkai et al., 1974). The women in this study kept the same sleep schedule at the two phases studied so we cannot establish whether, if given the opportunity, women with significant symptoms would have slept for a longer period of time during the late-luteal phase compared with women with minimal symptoms, as well as compared with their follicular phase. In addition, the possibility that the limited nap time (30 min) affected the results of this study should also be considered had the participants been given the opportunity to nap for a longer period of time, differences in nap characteristics between groups may have emerged. Furthermore, the high sleep need found in both groups, possibly because of age, time of day and menstrual cycle phase may have masked nocturnal sleep and nap characteristic differences between the two groups. Finally, the napping period occurring only during the late-luteal phase of the cycle limited our ability to fully understand the relationship between severity of premenstrual symptoms and daytime sleepiness. Examining the sleepiness level and nap characteristics among women with significant premenstrual symptoms during the follicular phase is necessary to help clarify this phenomenon. This study presents with certain limitations. Firstly, the women in the study were experiencing severe emotional behavioral premenstrual symptoms based on subjective questionnaires, which were not completed on a daily basis, and we did not conduct diagnostic interviews to establish whether they met criteria for PMDD. In addition, the present study focused specifically on emotional and behavioral symptoms of premenstrual symptoms. Women with suspected dysmenorrhea (painful uterine cramps, near and during menstruation) were intentionally excluded from this study, as they tend to experience a disturbed sleeping pattern that is different from that experienced by women without physical pain (Baker et al., 1999). Future research could include women who are suffering from physical symptoms and examine the relationship between nocturnal sleep, daytime sleepiness and nap characteristics among this population. Future research could also include spectral digital analyses of nap and nocturnal activity, as this may prove to be a more sensitive measure of sleep activity and may highlight further differences between groups and phases. In summary, women with significant emotional behavioral premenstrual symptoms have increased subjective sleepiness and lower alertness during the late-luteal phase of the cycle compared with women with minimal symptoms. Although this increased sleepiness may reflect an increased need for sleep, ad libitum sleep was not allowed for the nocturnal sleep or the nap. We therefore cannot rule out the possibility that

7 268 L. J. Lamarche et al. nocturnal or daytime sleep duration plays a role in this relationship. ACKNOWLEDGEMENTS This research was supported by Ontario Graduate Scholarship-Science and Technology. The funding agency had no role in design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript. REFERENCES Armitage, R. and Yonkers, K. A. Case report: menstrual-related very short REM latency in a healthy normal control. Sleep, 1994, 17: Baker, F. C., Driver, H. S., Rogers, G. G., Paiker, J. and Mitchell, D. High nocturnal body temperatures and disturbed sleep in women with primary dysmenorrhea. Am. J. Physiol., 1999, 277: Beck, A. T., Ward, C. H., Mendelson, M., Mock, J. and Erbaugh, J. An inventory for measuring depression. Arch. Gen. Psychiatry, 1961, 4: Brooks, A. and Lack, L. A brief afternoon nap following nocturnal sleep restriction: which nap duration is most recuperative? Sleep, 2006, 29: Broughton, R. Biorythmic variations in consciousness and psychological functions. Can. Psychol. Rev., 1975, 16: Buysse, D. J., Reynolds, C. F., Monk, T. H., Berman, S. R. and Kupfer, D. J. The pittsburgh sleep quality index: a new instrument for psychiatric practice and research. Psychiatry Res., 1989, 28: Chuong, C. J., Ryong, S., Taskin, O. and Karacan, I. Sleep pattern changes in menstrual cycles of women with premenstrual syndrome: a preliminary study. Am. J. Obstet. Gynecol., 1997, 177: Cluydts, V. Mood and sleep I: effects of the menstrual cycle. Waking Sleeping, 1980, 4: Driver, H. S., Dijk, D. J., Werth, E., Biedermann, K. and Borbély, A. A Sleep and sleep electroencephalogram across the menstrual cycle in young healthy women. J. Clin. Endocrinol. Metab., 1996, 81: Freeman, E. W. Premenstrual syndrome and premenstrual dysphoric disorder: definitions and diagnosis. Psychoneuroendoorinology 2003: Hamilton, J. A., Parry, B. L., Alagna, S., Blumenthal, S. and Herz, E. Premenstrual mood changes: a guide to evaluation and treatment. Psychiatr. Ann., 1984, 14: Hartse, K. M., Roth, T. and Zorick, J. Daytime sleepiness and daytime wakefulness: the effects of instruction. Sleep, 1982, S: S107 S118. Haskett, R. and Abplanalp, M. Premenstrual tension syndrome: diagnostic criteria and selection of research subjects. Psychiatry Res., 1983, 9: Hoddes, E., Dement, W. and Zarcone, V. The development and use of the stanford sleepiness scale (SSS). Psychophysiology, 1972, 9: 150. Hoddes, E., Smythe, Z., Philips, R. and Dement, C. Quantification of sleepiness: a new approach. Psychophysiology, 1973, 10: Hurt, S. W., Schnurr, P. P., Severino, S. K., Freeman, E. W., Gise, L. H., Rivera-Tovar, A. and Steege, J. F. Late-luteal phase dysphoric disorder in 670 women evaluated for premenstrual complaints. Am. J. Psychiatry, 1992, 149: Ito, M., Kohsaka, M., Homma, K., Fukuda, N., Honma, S., Katsuno, Y., Kawai, I., Honma, H., Morita, N., Miyamoto, T. et al. Changes in biological rhythm and sleep structure during the menstrual cycle in healthy women. Seishin Shinkeigaku Zasshi, 1995, 97: Klerman, E. B. and Dijk, D. Interindividual variation in sleep duration and its association with sleep debt in young adults. Sleep, 2005, 28: Lee, K. A., Shaver, J. F., Giblin, E. C. and Woods, N. F. Sleep patterns related to menstrual cycle phase and premenstrual affective symptoms. Sleep, 1990, 13: Manber, R. and Bootzin, R. R. Sleep and the menstrual cycle. Health Psychol., 1997, 16: Marshall, J. Thermal changes in the normal menstrual cycle. BMJ, 1963, 1: Mauri, M., Reid, R. L. and MacLean, A. W. Sleep in the premenstrual phase: a self-report study of PMS patients and normal controls. Acta Psychiatr. Scand., 1998, 78: Mitchell, E., Woods, N. and Lentz, M. Recognizing PMS when you see it: criteria for PMS sample selection. In: D. L. Taylor and N. F. Woods (Eds) Menstruation, Health, and Illness. Hemisphere, Washington, 1991: Monroe, L. J. Psychological and physiological differences between good and poor sleepers. J. Abnorm. Psychol., 1967, 72: Owens, J. F. and Matthews, K. A. Sleep disturbance in healthy middleaged women. Maturitas, 1998, 30: Parry, B., Mendelson, W., Duncan, W., Sack, D. and Wehr, T. Longitudinal sleep EEG, temperature, and activity measurements across the menstrual cycle in patients with premenstrual depression and in age-matched controls. Psychiatry Res., 1989, 30: Parry, B. L. Light therapy of premenstrual depression. In: R. W. Lam (ed) Seasonal effective disorder and beyond. American Psychiatric Press, Washington, 1993: Parry, B. L., Le Veau, B., Mostofi, N., Naham, H. C., Loving, R., Clopton, P. and Gillin, J. C. Temperature circadian rhythms during the menstrual cycle and sleep deprivation in premenstrual dysphoric disorder and normal comparison subjects. J. Biol. Rhythms, 1997, 12: Parry, B. L., Mostofi, N., Le Veau, B., Nahum, H. C., Golshan, S., Laughlin, G. A. and Gillin, J. C. Sleep EEG studies during early and late partial sleep deprivation in premenstrual dysphoric disorder and normal control subjects. Psychiatry Res., 1999, 85: Patkai, P., Johannson, G. and Post, B. Mood, alertness and sympathetic-adrenal medullary activity during the menstrual cycle. Psychosom. Med., 1974, 36: Pollak, C. P. How should the multiple sleep latency test be analyzed? Sleep, 1997, 20: Rechtschaffen, A. and Kales, A. A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects (National Institutes of Health Publication, No. 204). Public Health Service, Bethesda, MA, Severino, S. K., Wagner, D. R., Moline, M. L., Hurt, S. W., Pollak, C. P. and Zendell, S. High nocturnal body temperature in premenstrual syndrome and late-luteal phase dysphoric disorder. Am. J. Psychiatry, 1991, 148: Shibui, K., Uchiyama, M., Okawa, M., Kudo, Y., Kim, K., Kamei, Y., Kayakawa, T., Akamatsu, T., Ohta, K. and Ishibashi, K. Diurnal fluctuation of sleep propensity and hormonal secretion across the menstrual cycle. Biol. Psychiatry, 2000, 48: Steiner, M., Haskett, R. F. and Carroll, B. J. Premenstrual tension syndrome: the development of research diagnostic criteria and new rating scales. Acta Psychiatr. Scand., 1980, 62: Steiner, M., Streiner, D., Steinberg, S., Stewart, D., Carter, D., Berger, C., Reid, R. and Grover, D. The measurement of premenstrual mood symptoms. J. Affect. Disord., 1999, 53: Steiner, M. Premenstrual syndrome and premenstrual dysphoric disorder: Guidelines for management. J. Psychiatr. Neurosci, 2000: