- Email: [email protected]
Parity and sleep patterns during and after pregnancy
Parity and Sleep Patterns During and After Pregnancy KATHRYN A. LEE, RN, PhD, MARY ELLEN ZAFFKE, RN, PhD, AND GEOFFRY MCENANY, RN, PhD Objective: To describe changes in women’s sleep patterns from prepregnancy to postpartum. Methods: Polysomnography was done in women’s homes for 2 consecutive nights. Forty-five women were studied during the follicular and luteal phases of their menstrual cycles, and 33 conceived and were studied during each trimester of pregnancy. Twenty-nine were studied at 1 and 3 months postpartum. Results: Compared with prepregnant sleep characteristics, significant changes in sleep patterns were evident by 11–12 weeks’ gestation, with a significant increase in total sleep time but less deep sleep and more awakening during sleep. By the third month postpartum, there was improvement in sleep characteristics; however, sleep efficiency remained significantly lower than baseline prepregnancy values. Conclusion: Sleep disturbance was greatest during the first postpartum month, particularly for first-time mothers. (Obstet Gynecol 2000;95:14 – 8. © 2000 by The American College of Obstetricians and Gynecologists.)
Childbearing affects women’s sleep in many ways. Complaints heard most often are disrupted sleep patterns that begin late in pregnancy and continue through the first year postpartum.1–3 A healthy young adult typically spends less than 5% of the night awake, about 55% of the night in light sleep, and 20% in deep sleep. Deep sleep is associated with feeling more rested the next day.4 Rapid eye movement sleep, typically 20 –25% of the night, is believed to be important for cognitive functioning.5 Hormonal changes and physical discomfort are known to affect sleep. To date, research on sleep during pregnancy has been limited by small samples, lack of baseline data, and artificial laboratory settings.6 –13 The purpose of this longitudinal study was to determine the
From the Department of Family Health Care Nursing, School of Nursing, University of California, San Francisco, San Francisco, California. Funded by the National Institute of Nursing Research, R29 NR02247.
14 0029-7844/00/$20.00 PII S0029-7844(99)00486-X
changes in women’s sleep patterns in their home environments as they experience pregnancy and motherhood. Differences in sleep characteristics are also described by parity.
Materials and Methods A convenience sample was recruited through newspaper and television advertisements and posted flyers on the university campus. Healthy women 25–39 years old who were planning a pregnancy within the next year were eligible. Women with diagnosed sleep problems or children not yet consistently sleeping through the night were ineligible. Women with histories of mental health problems or who were taking antidepressants were excluded. After each woman gave informed consent, polysomnography was done in her home for 2 consecutive nights by using ambulatory monitoring (Medilog 9000; Oxford Medical Inc., Clearwater, FL). A member of the research team arrived at each participant’s home after dinner and at least 1 hour before bedtime. Application of the electrodes took approximately 45 minutes, and they were applied according to standard criteria, using C3/A2, C4/A1, and the outer canthus of right and left eyes. The women were instructed to go to bed at their normal bedtimes. There were three phases of childbearing in this study. Phase I included 45 women planning pregnancies within the next year who were studied during follicular (days 4 –10) and luteal (days 16 –25) phases of their menstrual cycles. Phase II included 33 women from phase I who conceived and were studied at 11–12 weeks’, 23–24 weeks’, and 35–36 weeks’ gestation. Phase III included 29 women who completed phase II and were studied at 3– 4 weeks and 11–12 weeks postpartum. Subjects were paid $25.00 at each phase, with a bonus of $25.00 at the completion of all seven time points. At each of the seven time points, the women were monitored for 2 nights to collect sleep electroencepha-
Obstetrics & Gynecology
lographic data. Fasting blood samples were drawn at each time point for thyroid function tests, hemoglobin and hematocrit, progesterone, iron, ferritin, and folate. Beginning 2 days before sleep monitoring, subjects completed a 7-day diary that included self-reported nap times, daily stresses and symptoms, perceptions of sleep, and reasons for awakening during the night. Polysomnographic data were analyzed using standardized sleep scoring criteria.14 The sleep scorer had greater than 90% agreement in sleep scoring of epochs selected from a random sample of sleep records for interrater reliability and was masked to subjects’ pregnancy status. There were no significant differences in sleep characteristics between nights 1 and 2 at any time point in the study, and correlation coefficients between nights exceeded .80 on all sleep characteristics. However, the correlation was moderate at the follicular time point, indicating a trend for a first-night effect of sleep monitoring at the initial recording period. To allow for adaptation to the monitoring equipment, night 2 was used in the analyses for all seven time points. Repeatedmeasures analysis of variance was used to test for significant changes in sleep variables by time and parity.
Results The 33 women who conceived were primarily white (88%) and upper middle class, having a median annual income of $45,000. Six (18%) were homemakers not employed outside the home. All graduated from high school, and 23 (70%) had college education. There were 16 nulliparas and 17 multiparas, with no significant difference in mean age (30.5 ⫾ 3.7 and 31.5 ⫾ 3.9, respectively). Most were in stable, satisfying marriages, with a mean of 19.6 ⫾ 2.2 standard deviation on the Kansas Marital Satisfaction Scale,15 which ranges from 0 to 21 points. Most (96%) planned to breast feed. All subjects were within normal limits for laboratory tests before pregnancy and remained so during the first trimester. Results for the second night of sleep at the follicular phase compared with the first trimester are presented in Table 1 for the 33 women who conceived. Three women miscarried and one conceived again within 1 year and continued in the study. At 23–24 weeks’ gestation, eight of 33 had low hemoglobin (under 12.0 g/dL), and 11 had ferritin levels below 10.0 ng/mL. There were three births before 35 weeks’ gestation and those three women were excluded from further participation. One mother delivered twins. There were four cesarean births. The woman with twins and the women with cesarean births were not outliers in the group data, and there were no significant differences in sleep characteristics by type of delivery or
VOL. 95, NO. 1, JANUARY 2000
Table 1. Baseline Prepregnancy Follicular Phase and FirstTrimester Sleep Characteristics for Night 2
Sleep Characteristic Total sleep time (min) Sleep efficiency (%) Sleep onset latency to stage 2 (min) REM onset latency (min) Awake time (% sleep) Stage 1 (% sleep) Stage 2 (% sleep) Stage 3 ⫹ 4 (% sleep) REM (% sleep)
Prepregnancy follicular phase (n ⫽ 33)
Pregnancy 11–12 weeks (n ⫽ 33)
412 ⫾ 60.6 93 ⫾ 6.7 14 ⫾ 13.3
446 ⫾ 65.5* 91 ⫾ 5.5* 11 ⫾ 8.2
83 ⫾ 25.6 6 ⫾ 5.4 3 ⫾ 2.0 54 ⫾ 6.9 13 ⫾ 6.5 25 ⫾ 5.5
74 ⫾ 25.8 9 ⫾ 5.6† 3 ⫾ 1.2 54 ⫾ 6.1 9 ⫾ 3.2‡ 24 ⫾ 4.6
REM ⫽ rapid eye movement. Data presented as mean ⫾ standard deviation. * t ⫽ 2.2, P ⫽ .03 significant decrease from baseline. † t ⫽ 3.5, P ⫽ .002 significant increase from baseline. ‡ t ⫽ 2.6, P ⫽ .015 significant decrease from baseline.
infant feeding pattern. Twenty-nine of the women with uncomplicated normal labor and delivery were studied at 1 month and 3 months postpartum. All mothers continued to breast feed but supplemented with bottle feedings to different degrees. At the first postpartum month, only three women had low hemoglobin levels, but ferritin and thyroid function were normal. Third trimester compared with 1 month postpartum sleep data are presented in Table 2 for the 29 women with complete data at those two points. There was a significant change in total sleep time over time (F5,80 ⫽ 2.98, P ⫽ .016). As shown in Figure 1, sleep time during the first trimester was significantly higher than during prepregnancy baseline and the third trimester. Total sleep time for the sample averaged a high of 446 minutes during the first trimester to a low of 372 Table 2. Third-Trimester and Postpartum Sleep Characteristics for Night 2
Sleep Characteristic
Pregnancy 35–36 weeks (n ⫽ 29)
Postpartum 3– 4 weeks (n ⫽ 29)
415 ⫾ 64.5 89 ⫾ 5.8 13 ⫾ 11.1
379 ⫾ 78.5* 81 ⫾ 7.7† 11 ⫾ 10.7
87 ⫾ 42.9 11 ⫾ 5.8 4 ⫾ 1.2 56 ⫾ 5.7 8 ⫾ 3.8 21 ⫾ 5.1
69 ⫾ 27.4* 19 ⫾ 7.7† 4 ⫾ 2.4 44 ⫾ 7.1† 12 ⫾ 5.1† 21 ⫾ 4.3
Total sleep time (min) Sleep efficiency (%) Sleep onset latency to stage 2 (min) REM onset latency (min) Awake time (% sleep) Stage 1 (% sleep) Stage 2 (% sleep) Stage 3 ⫹ 4 (% sleep) REM (% sleep)
REM ⫽ rapid eye movement. Data presented as mean ⫾ standard deviation. * t ⫽ 2.0, P ⬍ .05 significant decrease from third trimester. † t ⫽ 4.2, P ⱕ .001 significant change from third trimester.
Lee et al
Sleep and Pregnancy
15
Figure 1. Mean total sleep time (minutes in stage 1, stage 2, stage 3– 4, and REM sleep) on night 2, from prepregnancy follicular phase baseline measurement to 3rd month postpartum (post) for 13 nulliparous first-time mothers (shaded bars) and 16 multiparous experienced mothers (black bars). Note that mean total sleep time decreased over time, particularly for first-time mothers. tri ⫽ trimester.
Figure 2. Mean sleep efficiency (proportion of time in bed that was actually spent asleep) on night 2, from prepregnancy follicular phase baseline measurement to 3rd month postpartum (post) for 13 nulliparous first-time mothers (shaded bars) and 16 multiparous experienced mothers (black bars). The lowest mean sleep efficiency is at 1 month postpartum, particularly for first-time mothers. tri ⫽ trimester.
minutes at 1 month postpartum. Although the 16 experienced mothers had more sleep than the 13 novice mothers at each phase, total sleep time was not significantly influenced by parity. The amount of time actually spent asleep is expressed as a percentage of the time in bed trying to sleep and reported as a sleep efficiency that can range from 0% (awake all night) to 100% (fell asleep immediately after lights out and no awakenings until final awakening). Sleep efficiency ranged from an average high of 93% before pregnancy to a low of 81% during the first month postpartum. As seen in Figure 2, there was a significant change in sleep efficiency over time (F5,80 ⫽ 6.99, P ⬍ .001), with 3 months postpartum significantly (F ⫽ 10.4, P ⫽ .006) less than baseline. There was also an effect of parity at baseline and at 1 month postpartum (F5,80 ⫽ 2.3, P ⫽ .05). Multiparas had lower sleep efficiency compared with nulliparas at all time points except 1 month postpartum. There was no significant change over time in latency to the first REM cycle. The percentage of total sleep time spent in REM sleep also did not change over time (P ⫽ .086), and REM sleep was not influenced by parity (Figure 3). Conversely, deep sleep (stages 3 and 4 delta waves) did change over time (F5,80 ⫽ 5.3, P ⫽ .002). Deep sleep was diminished throughout pregnancy compared with baseline and postpartum. There was no significant effect of parity on changes in deep sleep (Figure 4).
Discussion
16 Lee et al
Sleep and Pregnancy
Our results showed that sleep is different for nulliparas and multiparas before and after pregnancy. Even though multiparas had children sleeping through the night, their sleep before pregnancy was less efficient than nulliparas because of frequent brief awakenings. There were significant changes in sleep as early as
Figure 3. Mean rapid eye movement sleep (proportion of total sleep time [TST] spent in rapid eye movement) on night 2, from prepregnancy follicular phase baseline measurement to 3rd month postpartum (post) for 13 nulliparous first-time mothers (shaded bars) and 16 multiparous experienced mothers (black bars). tri ⫽ trimester.
Obstetrics & Gynecology
Figure 4. Mean deep sleep (proportion of total sleep time [TST] spent in stage 3 and stage 4 delta-wave sleep) on night 2, from prepregnancy follicular phase baseline measurement to 3rd month postpartum (post) for 13 nulliparous first-time mothers (shaded bars) and 16 multiparous experienced mothers (black bars). Note decreases in mean deep sleep in all trimesters (tri) of pregnancy.
weeks 11–12 of pregnancy. There was a decrease in restorative, deep sleep from baseline to the first trimester for both groups, which did not improve until after delivery. The increase in total sleep time and reduced sleep efficiency from the follicular phase to the first trimester support data from Karacan and colleagues,16 who studied three women throughout their pregnancies and through return of normal menses. Sleep characteristics remained fairly stable during pregnancy, but total sleep time and sleep efficiency continued declining throughout pregnancy regardless of parity. Prior research included nulliparas and multiparas in the study samples, but samples were too small to analyze by parity. In this study, 13 nulliparas had consistently less total sleep time during pregnancy than 16 multiparas, although group differences were not statistically significant in our small sample. Even when infants were 3– 4 weeks old, there were still significantly disturbed maternal sleep patterns. Compared with third-trimester values, there was significantly less total sleep time and poorer sleep efficiency caused by awakenings during the night. There was a significant increase in deep sleep but no change in amount of REM sleep. The decrease in sleep efficiency was especially evident in novice mothers compared with experienced mothers (Figure 2) as reported elsewhere.17 Nishihara and Horiuchi18 studied young first-time mothers and their infants at 1, 3, and 6 weeks postpartum and also found poor sleep efficiency (75–78%) primarily because of infant feeding. Poor postpartum sleep efficiency in the present sample indicates more
VOL. 95, NO. 1, JANUARY 2000
time awake during the night, however deep sleep was spared at the expense of less time in light sleep. Rather than spend time in lighter sleep stages, mothers were quickly able to get into the deeper, more restorative stages of sleep after awakenings. That was especially the case for more experienced multiparas than first-time mothers. Those findings are similar to findings for healthy controls in the study by Coble and colleagues19 and support earlier descriptive studies of differences between primiparas and multiparas regarding postpartum concerns.20 Sleep patterns of novice mothers were similar to those of experienced mothers by the third month postpartum. There was improvement in all aspects of sleep, but there was still evidence of sleep deprivation compared with prepregnancy baseline data. Total sleep time and amount of deep sleep returned to prepregnancy values, but sleep efficiency was still significantly lower than baseline because of the amount of time awake during the night compared with prepregnancy values. Although awakenings during the night reduced sleep efficiency, the amount of deep sleep in the third month postpartum was comparable to prepregnancy levels. Our findings support the restorative theory of deep sleep and findings from other studies in which deep sleep takes precedence over REM sleep during opportunities for recovery from sleep deprivation.21,22 What was most interesting was the finding at 3 months postpartum that novice mothers did not return to prepregnancy sleep patterns but more closely resembled baseline sleep characteristics for experienced multiparas.
References 1. Schweiger MS. Sleep disturbance in pregnancy. Am J Obstet Gynecol 1972;114:879 – 82. 2. Lee KA, DeJoseph JF. Sleep disturbances, vitality, and fatigue among a select group of employed childbearing women. Birth 1992;19:208 –13. 3. Hertz G, Fast A, Feinsilver SH, Albertario CL, Schulman H, Fein AM. Sleep in normal late pregnancy. Sleep 1992;18:246 –51. 4. Horne J. Why we sleep: The functions of sleep in humans and other mammals. Oxford: Oxford University Press, 1990. 5. Bonnet MH. Sleep deprivation. In: Kryger MH, Roth T, Dement WC, eds. Principles and practice of sleep medicine. Philadelphia: WB Saunders, 1994:60 – 6. 6. Karacan I, Heine W, Agnew HW, Williams RL, Webb WB, Ross JJ. Characteristics of sleep patterns during late pregnancy and postpartum periods. Am J Obstet Gynecol 1968;101:579 – 86. 7. Branchey M, Petre-Quandens O. A comparative study of sleep parameters during pregnancy. Acta Neurol Psychiatrica Belg 1968; 7:453–9. 8. Hoppenbrouwers T, Ugartechea JC, Combs D, Hodgman JE, Harper RM, Sterman MB. Studies of maternal-fetal interaction
Lee et al
Sleep and Pregnancy
17
9.
10.
11.
12. 13. 14.
15.
16.
17.
18.
during the last trimester of pregnancy; ontogenesis of the basic rest-activity cycle. Exp Neurol 1978;61:136 –53. Driver HS, Shapiro CM. A longitudinal study of sleep stages in young women during pregnancy and postpartum. Sleep 1992;15: 449 –53. Schorr SJ, Chawla A, Devidas M, Sullivan C, Naef R, Morrison J. Sleep patterns in pregnancy: A longitudinal study of polysomnography recordings during pregnancy. J Perinatol 1998;18:427–30. Brunner DP, Munch M, Biedermann K, Huh R, Borbely AA. Changes in sleep and sleep electroencephalogram during pregnancy. Sleep 1994;17:576 – 82. Lee K. Self-reported sleep disturbances in employed women. Sleep 1992;15:493– 8. Lee K. Alterations in sleep during pregnancy and postpartum: A review of 30 years of research. Sleep Med Rev 1998;4:231– 42. Rechtschaffen A, Kales A. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. Los Angeles: Brain Information Services/Brain Research Institute, University of California, 1968. Schumm WR, Paff-Bergen LA, Hatch RC, Obiorah FC, Copeland JM, Meens LD, et al. Concurrent and discriminant validity of the Kansas Marital Satisfaction Scale. J Marriage Fam 1986;48:381–7. Karacan I, Williams RL, Hursch CJ, McCaulley M, Heine MW. Some implications of the sleep patterns of pregnancy for postpartum emotional disturbances. Am J Psychiatry 1969;115:929 –35. Waters MA, Lee KA. Differences between primigravidae and multigravidae mothers in sleep disturbances, fatigue, and functional status. J Nurse Midwifery 1996;41:364 –7. Nishihara K, Horiuchi S. Changes in sleep patterns of young women from late pregnancy to postpartum: Relationships to their infant’s movements. Percept Motor Skills 1998;87:1043–56.
18 Lee et al
Sleep and Pregnancy
19. Coble PA, Reynolds CF, Kupfer DJ, Houck PR, Day NL, Giles DE. Childbearing in women with and without a history of affective disorder. I. Psychiatric symptomatology. Comp Psychiatry 1994; 35:215–24. 20. Tribotti S, Lyons N, Blackburn S, Stein M, Withers J. Nursing diagnoses for the postpartum woman. J Obstet Gynecol Neonatal Nurs 1988;17:410 –7. 21. Webb WB, Agnew HW. Sleep: Effects of a restricted regime. Science 1965;150:1745–7. 22. Tilley AJ, Wilkinson RT. The effects of a restricted sleep regime on the composition of sleep and on performance. Psychophysiology 1984;21:406 –12.
Address reprint requests to:
Kathryn A. Lee, RN, PhD Department of Family Health Care Nursing University of California, San Francisco Box 0606 San Francisco, CA 94143-0606 E-mail: [email protected]
Received February 26, 1999. Received in revised form June 16, 1999. Accepted July 15, 1999. Copyright © 2000 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
Obstetrics & Gynecology
Childbearing affects women’s sleep in many ways. Complaints heard most often are disrupted sleep patterns that begin late in pregnancy and continue through the first year postpartum.1–3 A healthy young adult typically spends less than 5% of the night awake, about 55% of the night in light sleep, and 20% in deep sleep. Deep sleep is associated with feeling more rested the next day.4 Rapid eye movement sleep, typically 20 –25% of the night, is believed to be important for cognitive functioning.5 Hormonal changes and physical discomfort are known to affect sleep. To date, research on sleep during pregnancy has been limited by small samples, lack of baseline data, and artificial laboratory settings.6 –13 The purpose of this longitudinal study was to determine the
From the Department of Family Health Care Nursing, School of Nursing, University of California, San Francisco, San Francisco, California. Funded by the National Institute of Nursing Research, R29 NR02247.
14 0029-7844/00/$20.00 PII S0029-7844(99)00486-X
changes in women’s sleep patterns in their home environments as they experience pregnancy and motherhood. Differences in sleep characteristics are also described by parity.
Materials and Methods A convenience sample was recruited through newspaper and television advertisements and posted flyers on the university campus. Healthy women 25–39 years old who were planning a pregnancy within the next year were eligible. Women with diagnosed sleep problems or children not yet consistently sleeping through the night were ineligible. Women with histories of mental health problems or who were taking antidepressants were excluded. After each woman gave informed consent, polysomnography was done in her home for 2 consecutive nights by using ambulatory monitoring (Medilog 9000; Oxford Medical Inc., Clearwater, FL). A member of the research team arrived at each participant’s home after dinner and at least 1 hour before bedtime. Application of the electrodes took approximately 45 minutes, and they were applied according to standard criteria, using C3/A2, C4/A1, and the outer canthus of right and left eyes. The women were instructed to go to bed at their normal bedtimes. There were three phases of childbearing in this study. Phase I included 45 women planning pregnancies within the next year who were studied during follicular (days 4 –10) and luteal (days 16 –25) phases of their menstrual cycles. Phase II included 33 women from phase I who conceived and were studied at 11–12 weeks’, 23–24 weeks’, and 35–36 weeks’ gestation. Phase III included 29 women who completed phase II and were studied at 3– 4 weeks and 11–12 weeks postpartum. Subjects were paid $25.00 at each phase, with a bonus of $25.00 at the completion of all seven time points. At each of the seven time points, the women were monitored for 2 nights to collect sleep electroencepha-
Obstetrics & Gynecology
lographic data. Fasting blood samples were drawn at each time point for thyroid function tests, hemoglobin and hematocrit, progesterone, iron, ferritin, and folate. Beginning 2 days before sleep monitoring, subjects completed a 7-day diary that included self-reported nap times, daily stresses and symptoms, perceptions of sleep, and reasons for awakening during the night. Polysomnographic data were analyzed using standardized sleep scoring criteria.14 The sleep scorer had greater than 90% agreement in sleep scoring of epochs selected from a random sample of sleep records for interrater reliability and was masked to subjects’ pregnancy status. There were no significant differences in sleep characteristics between nights 1 and 2 at any time point in the study, and correlation coefficients between nights exceeded .80 on all sleep characteristics. However, the correlation was moderate at the follicular time point, indicating a trend for a first-night effect of sleep monitoring at the initial recording period. To allow for adaptation to the monitoring equipment, night 2 was used in the analyses for all seven time points. Repeatedmeasures analysis of variance was used to test for significant changes in sleep variables by time and parity.
Results The 33 women who conceived were primarily white (88%) and upper middle class, having a median annual income of $45,000. Six (18%) were homemakers not employed outside the home. All graduated from high school, and 23 (70%) had college education. There were 16 nulliparas and 17 multiparas, with no significant difference in mean age (30.5 ⫾ 3.7 and 31.5 ⫾ 3.9, respectively). Most were in stable, satisfying marriages, with a mean of 19.6 ⫾ 2.2 standard deviation on the Kansas Marital Satisfaction Scale,15 which ranges from 0 to 21 points. Most (96%) planned to breast feed. All subjects were within normal limits for laboratory tests before pregnancy and remained so during the first trimester. Results for the second night of sleep at the follicular phase compared with the first trimester are presented in Table 1 for the 33 women who conceived. Three women miscarried and one conceived again within 1 year and continued in the study. At 23–24 weeks’ gestation, eight of 33 had low hemoglobin (under 12.0 g/dL), and 11 had ferritin levels below 10.0 ng/mL. There were three births before 35 weeks’ gestation and those three women were excluded from further participation. One mother delivered twins. There were four cesarean births. The woman with twins and the women with cesarean births were not outliers in the group data, and there were no significant differences in sleep characteristics by type of delivery or
VOL. 95, NO. 1, JANUARY 2000
Table 1. Baseline Prepregnancy Follicular Phase and FirstTrimester Sleep Characteristics for Night 2
Sleep Characteristic Total sleep time (min) Sleep efficiency (%) Sleep onset latency to stage 2 (min) REM onset latency (min) Awake time (% sleep) Stage 1 (% sleep) Stage 2 (% sleep) Stage 3 ⫹ 4 (% sleep) REM (% sleep)
Prepregnancy follicular phase (n ⫽ 33)
Pregnancy 11–12 weeks (n ⫽ 33)
412 ⫾ 60.6 93 ⫾ 6.7 14 ⫾ 13.3
446 ⫾ 65.5* 91 ⫾ 5.5* 11 ⫾ 8.2
83 ⫾ 25.6 6 ⫾ 5.4 3 ⫾ 2.0 54 ⫾ 6.9 13 ⫾ 6.5 25 ⫾ 5.5
74 ⫾ 25.8 9 ⫾ 5.6† 3 ⫾ 1.2 54 ⫾ 6.1 9 ⫾ 3.2‡ 24 ⫾ 4.6
REM ⫽ rapid eye movement. Data presented as mean ⫾ standard deviation. * t ⫽ 2.2, P ⫽ .03 significant decrease from baseline. † t ⫽ 3.5, P ⫽ .002 significant increase from baseline. ‡ t ⫽ 2.6, P ⫽ .015 significant decrease from baseline.
infant feeding pattern. Twenty-nine of the women with uncomplicated normal labor and delivery were studied at 1 month and 3 months postpartum. All mothers continued to breast feed but supplemented with bottle feedings to different degrees. At the first postpartum month, only three women had low hemoglobin levels, but ferritin and thyroid function were normal. Third trimester compared with 1 month postpartum sleep data are presented in Table 2 for the 29 women with complete data at those two points. There was a significant change in total sleep time over time (F5,80 ⫽ 2.98, P ⫽ .016). As shown in Figure 1, sleep time during the first trimester was significantly higher than during prepregnancy baseline and the third trimester. Total sleep time for the sample averaged a high of 446 minutes during the first trimester to a low of 372 Table 2. Third-Trimester and Postpartum Sleep Characteristics for Night 2
Sleep Characteristic
Pregnancy 35–36 weeks (n ⫽ 29)
Postpartum 3– 4 weeks (n ⫽ 29)
415 ⫾ 64.5 89 ⫾ 5.8 13 ⫾ 11.1
379 ⫾ 78.5* 81 ⫾ 7.7† 11 ⫾ 10.7
87 ⫾ 42.9 11 ⫾ 5.8 4 ⫾ 1.2 56 ⫾ 5.7 8 ⫾ 3.8 21 ⫾ 5.1
69 ⫾ 27.4* 19 ⫾ 7.7† 4 ⫾ 2.4 44 ⫾ 7.1† 12 ⫾ 5.1† 21 ⫾ 4.3
Total sleep time (min) Sleep efficiency (%) Sleep onset latency to stage 2 (min) REM onset latency (min) Awake time (% sleep) Stage 1 (% sleep) Stage 2 (% sleep) Stage 3 ⫹ 4 (% sleep) REM (% sleep)
REM ⫽ rapid eye movement. Data presented as mean ⫾ standard deviation. * t ⫽ 2.0, P ⬍ .05 significant decrease from third trimester. † t ⫽ 4.2, P ⱕ .001 significant change from third trimester.
Lee et al
Sleep and Pregnancy
15
Figure 1. Mean total sleep time (minutes in stage 1, stage 2, stage 3– 4, and REM sleep) on night 2, from prepregnancy follicular phase baseline measurement to 3rd month postpartum (post) for 13 nulliparous first-time mothers (shaded bars) and 16 multiparous experienced mothers (black bars). Note that mean total sleep time decreased over time, particularly for first-time mothers. tri ⫽ trimester.
Figure 2. Mean sleep efficiency (proportion of time in bed that was actually spent asleep) on night 2, from prepregnancy follicular phase baseline measurement to 3rd month postpartum (post) for 13 nulliparous first-time mothers (shaded bars) and 16 multiparous experienced mothers (black bars). The lowest mean sleep efficiency is at 1 month postpartum, particularly for first-time mothers. tri ⫽ trimester.
minutes at 1 month postpartum. Although the 16 experienced mothers had more sleep than the 13 novice mothers at each phase, total sleep time was not significantly influenced by parity. The amount of time actually spent asleep is expressed as a percentage of the time in bed trying to sleep and reported as a sleep efficiency that can range from 0% (awake all night) to 100% (fell asleep immediately after lights out and no awakenings until final awakening). Sleep efficiency ranged from an average high of 93% before pregnancy to a low of 81% during the first month postpartum. As seen in Figure 2, there was a significant change in sleep efficiency over time (F5,80 ⫽ 6.99, P ⬍ .001), with 3 months postpartum significantly (F ⫽ 10.4, P ⫽ .006) less than baseline. There was also an effect of parity at baseline and at 1 month postpartum (F5,80 ⫽ 2.3, P ⫽ .05). Multiparas had lower sleep efficiency compared with nulliparas at all time points except 1 month postpartum. There was no significant change over time in latency to the first REM cycle. The percentage of total sleep time spent in REM sleep also did not change over time (P ⫽ .086), and REM sleep was not influenced by parity (Figure 3). Conversely, deep sleep (stages 3 and 4 delta waves) did change over time (F5,80 ⫽ 5.3, P ⫽ .002). Deep sleep was diminished throughout pregnancy compared with baseline and postpartum. There was no significant effect of parity on changes in deep sleep (Figure 4).
Discussion
16 Lee et al
Sleep and Pregnancy
Our results showed that sleep is different for nulliparas and multiparas before and after pregnancy. Even though multiparas had children sleeping through the night, their sleep before pregnancy was less efficient than nulliparas because of frequent brief awakenings. There were significant changes in sleep as early as
Figure 3. Mean rapid eye movement sleep (proportion of total sleep time [TST] spent in rapid eye movement) on night 2, from prepregnancy follicular phase baseline measurement to 3rd month postpartum (post) for 13 nulliparous first-time mothers (shaded bars) and 16 multiparous experienced mothers (black bars). tri ⫽ trimester.
Obstetrics & Gynecology
Figure 4. Mean deep sleep (proportion of total sleep time [TST] spent in stage 3 and stage 4 delta-wave sleep) on night 2, from prepregnancy follicular phase baseline measurement to 3rd month postpartum (post) for 13 nulliparous first-time mothers (shaded bars) and 16 multiparous experienced mothers (black bars). Note decreases in mean deep sleep in all trimesters (tri) of pregnancy.
weeks 11–12 of pregnancy. There was a decrease in restorative, deep sleep from baseline to the first trimester for both groups, which did not improve until after delivery. The increase in total sleep time and reduced sleep efficiency from the follicular phase to the first trimester support data from Karacan and colleagues,16 who studied three women throughout their pregnancies and through return of normal menses. Sleep characteristics remained fairly stable during pregnancy, but total sleep time and sleep efficiency continued declining throughout pregnancy regardless of parity. Prior research included nulliparas and multiparas in the study samples, but samples were too small to analyze by parity. In this study, 13 nulliparas had consistently less total sleep time during pregnancy than 16 multiparas, although group differences were not statistically significant in our small sample. Even when infants were 3– 4 weeks old, there were still significantly disturbed maternal sleep patterns. Compared with third-trimester values, there was significantly less total sleep time and poorer sleep efficiency caused by awakenings during the night. There was a significant increase in deep sleep but no change in amount of REM sleep. The decrease in sleep efficiency was especially evident in novice mothers compared with experienced mothers (Figure 2) as reported elsewhere.17 Nishihara and Horiuchi18 studied young first-time mothers and their infants at 1, 3, and 6 weeks postpartum and also found poor sleep efficiency (75–78%) primarily because of infant feeding. Poor postpartum sleep efficiency in the present sample indicates more
VOL. 95, NO. 1, JANUARY 2000
time awake during the night, however deep sleep was spared at the expense of less time in light sleep. Rather than spend time in lighter sleep stages, mothers were quickly able to get into the deeper, more restorative stages of sleep after awakenings. That was especially the case for more experienced multiparas than first-time mothers. Those findings are similar to findings for healthy controls in the study by Coble and colleagues19 and support earlier descriptive studies of differences between primiparas and multiparas regarding postpartum concerns.20 Sleep patterns of novice mothers were similar to those of experienced mothers by the third month postpartum. There was improvement in all aspects of sleep, but there was still evidence of sleep deprivation compared with prepregnancy baseline data. Total sleep time and amount of deep sleep returned to prepregnancy values, but sleep efficiency was still significantly lower than baseline because of the amount of time awake during the night compared with prepregnancy values. Although awakenings during the night reduced sleep efficiency, the amount of deep sleep in the third month postpartum was comparable to prepregnancy levels. Our findings support the restorative theory of deep sleep and findings from other studies in which deep sleep takes precedence over REM sleep during opportunities for recovery from sleep deprivation.21,22 What was most interesting was the finding at 3 months postpartum that novice mothers did not return to prepregnancy sleep patterns but more closely resembled baseline sleep characteristics for experienced multiparas.
References 1. Schweiger MS. Sleep disturbance in pregnancy. Am J Obstet Gynecol 1972;114:879 – 82. 2. Lee KA, DeJoseph JF. Sleep disturbances, vitality, and fatigue among a select group of employed childbearing women. Birth 1992;19:208 –13. 3. Hertz G, Fast A, Feinsilver SH, Albertario CL, Schulman H, Fein AM. Sleep in normal late pregnancy. Sleep 1992;18:246 –51. 4. Horne J. Why we sleep: The functions of sleep in humans and other mammals. Oxford: Oxford University Press, 1990. 5. Bonnet MH. Sleep deprivation. In: Kryger MH, Roth T, Dement WC, eds. Principles and practice of sleep medicine. Philadelphia: WB Saunders, 1994:60 – 6. 6. Karacan I, Heine W, Agnew HW, Williams RL, Webb WB, Ross JJ. Characteristics of sleep patterns during late pregnancy and postpartum periods. Am J Obstet Gynecol 1968;101:579 – 86. 7. Branchey M, Petre-Quandens O. A comparative study of sleep parameters during pregnancy. Acta Neurol Psychiatrica Belg 1968; 7:453–9. 8. Hoppenbrouwers T, Ugartechea JC, Combs D, Hodgman JE, Harper RM, Sterman MB. Studies of maternal-fetal interaction
Lee et al
Sleep and Pregnancy
17
9.
10.
11.
12. 13. 14.
15.
16.
17.
18.
during the last trimester of pregnancy; ontogenesis of the basic rest-activity cycle. Exp Neurol 1978;61:136 –53. Driver HS, Shapiro CM. A longitudinal study of sleep stages in young women during pregnancy and postpartum. Sleep 1992;15: 449 –53. Schorr SJ, Chawla A, Devidas M, Sullivan C, Naef R, Morrison J. Sleep patterns in pregnancy: A longitudinal study of polysomnography recordings during pregnancy. J Perinatol 1998;18:427–30. Brunner DP, Munch M, Biedermann K, Huh R, Borbely AA. Changes in sleep and sleep electroencephalogram during pregnancy. Sleep 1994;17:576 – 82. Lee K. Self-reported sleep disturbances in employed women. Sleep 1992;15:493– 8. Lee K. Alterations in sleep during pregnancy and postpartum: A review of 30 years of research. Sleep Med Rev 1998;4:231– 42. Rechtschaffen A, Kales A. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. Los Angeles: Brain Information Services/Brain Research Institute, University of California, 1968. Schumm WR, Paff-Bergen LA, Hatch RC, Obiorah FC, Copeland JM, Meens LD, et al. Concurrent and discriminant validity of the Kansas Marital Satisfaction Scale. J Marriage Fam 1986;48:381–7. Karacan I, Williams RL, Hursch CJ, McCaulley M, Heine MW. Some implications of the sleep patterns of pregnancy for postpartum emotional disturbances. Am J Psychiatry 1969;115:929 –35. Waters MA, Lee KA. Differences between primigravidae and multigravidae mothers in sleep disturbances, fatigue, and functional status. J Nurse Midwifery 1996;41:364 –7. Nishihara K, Horiuchi S. Changes in sleep patterns of young women from late pregnancy to postpartum: Relationships to their infant’s movements. Percept Motor Skills 1998;87:1043–56.
18 Lee et al
Sleep and Pregnancy
19. Coble PA, Reynolds CF, Kupfer DJ, Houck PR, Day NL, Giles DE. Childbearing in women with and without a history of affective disorder. I. Psychiatric symptomatology. Comp Psychiatry 1994; 35:215–24. 20. Tribotti S, Lyons N, Blackburn S, Stein M, Withers J. Nursing diagnoses for the postpartum woman. J Obstet Gynecol Neonatal Nurs 1988;17:410 –7. 21. Webb WB, Agnew HW. Sleep: Effects of a restricted regime. Science 1965;150:1745–7. 22. Tilley AJ, Wilkinson RT. The effects of a restricted sleep regime on the composition of sleep and on performance. Psychophysiology 1984;21:406 –12.
Address reprint requests to:
Kathryn A. Lee, RN, PhD Department of Family Health Care Nursing University of California, San Francisco Box 0606 San Francisco, CA 94143-0606 E-mail: [email protected]
Received February 26, 1999. Received in revised form June 16, 1999. Accepted July 15, 1999. Copyright © 2000 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
Obstetrics & Gynecology