May 2011  << Back  

  Can't open the PDF? Click here for help.

 

 RESEARCH ARTICLE

Sleep disturbances in euthymic bipolar patients

Sharon Brill, BS

Mood Disorders Research Program, Department of Psychiatry, and Behavioral Science, University of Louisville School of Medicine, Louisville, KY, USA

Praveen Penagaluri, MD

Mood Disorders Research Program, Department of Psychiatry, and Behavioral Science, University of Louisville School of Medicine, Louisville, KY, USA

R. Jeannie Roberts, MD

Mood Disorders Research Program, Department of Psychiatry, and Behavioral Science, University of Louisville School of Medicine, Louisville, KY, USA

YongLin Gao, MD

Mood Disorders Research Program, Department of Psychiatry, and Behavioral Science, University of Louisville School of Medicine, Louisville, KY, USA

Rif S. El-Mallakh, MD

Mood Disorders Research Program, Department of Psychiatry, and Behavioral Science, University of Louisville School of Medicine, Louisville, KY, USA

BACKGROUND: Sleep disturbance has been implicated in both prodromal and syndromal phases of bipolar illness.

METHODS: Charts of bipolar disorder (BD) patients who had been euthymic for at least 2 months were reviewed for mood symptoms, Clinical Global Impression scores, Global Assessment of Functioning scores, and sleep.

RESULTS: Among 116 patients, 10 never achieved a euthymic interval of 2 months’ duration. Among the remaining 106 euthymic patients, 59 (55.6%) had BD I, 23 (21.7%) had BD II, and 24 (22.8%) had BD not otherwise specified (NOS). The mean age was 43.3±SD 14.6, and 35% were male. A total of 25 patients (23.6%) had a clinically significant ongoing sleep disturbance (27.1% of those with BD I, 21.7% of those with BD II, and 16.6% of those with BD NOS). Of 16 patients for whom a sleep description was available, 25% had difficulty falling asleep, 81.25% had middle insomnia (2 patients experienced both), and none had early morning awakening. Eleven patients (10.4%) received sleep aids, and 33 (31.1%) received sedating antipsychotics (3 patients received both).

CONCLUSIONS: Sleeping aids and sedating antipsychotics can potentially disguise an underlying sleep disturbance. Thus, it is possible that study patients taking these medications (n=58; 54.7%) suffer from a sleep disturbance that is being adequately or inadequately treated.

KEYWORDS: bipolar disorder, euthymia, sleep disturbance, insomnia, sleep aids, manic induction

ANNALS OF CLINICAL PSYCHIATRY 2011;23(2):113–116

  INTRODUCTION

Bipolar disorder (BD) is a severe psychiatric illness that manifests with episodes of mania and depression. The pathophysiology of this disorder is unknown, but recent evidence implicates disruption in natural rhythms.1 Circadian system abnormalities, such as blunting of rhythms’ amplitudes, advanced position (or even nonentrainment) of rhythms’ phases, and doubling of the sleep-wake cycle length from 24 to 48 hours may be involved.2 Alterations in the timing of sleep and wakefulness relative to other circadian rhythms may trigger the onset or offset of episodes of depression and mania.1 Disrupted sleep can trigger and intensify mania.3 Multiple genes of small effect are important in the generation and regulation of circadian rhythms and sleep system in BD, which includes Circadian Locomotor Output Cycles Kaput gene.4

The interepisode period of the bipolar cycle is poorly studied. A significant proportion of BD patients experience disabling symptoms between episodes. These ongoing symptoms can create dysfunction5 and lead to higher risk for relapse.6 Sleep disturbances are among the most prominent correlates of mood episodes.1,7

Small studies have shown that euthymic BD patients may have less stable and more variable circadian activity patterns than controls (n=19).8 The sleep of remitted BD outpatients measured using actigraphic and sleep parameters is characteristically different from controls in naturalistic studies; BD patients sleep longer, report longer latencies, and display greater variability across nights.9 The prevalence of sleep disturbances in treated outpatient euthymic BD patients is unclear. We performed a retrospective chart review to determine the frequency and nature of sleep patterns in BD patients attending an outpatient clinic.

  METHODS

Charts of patients attending an outpatient, university-based BD clinic were retrospectively reviewed. All patients with a BD diagnosis (I, II, or not otherwise specified [NOS]) were included if a 2-month euthymic interval could be identified. Patients were excluded if a clear euthymic period could not be found. Each clinic evaluation visit generated an outpatient form that quantified both depressive and (hypo)manic symptoms in a manner similar to that utilized in the Systematic Treatment Enhancement Program for Bipolar Disorder study.10 This allows identification and quantification of mood symptoms. Euthymia was defined as a score of <11 (out of a possible 42) for depression, and <10 (out of a possible 36) for hypomania or mania. Data were collected regarding mood symptoms, medications, sleep description and duration, Clinical Global Impression Severity (CGI-S) scores,11 and Global Assessment of Functioning (GAF) scores. The study was approved by the Institutional Human Studies Committee.

A total of 116 charts were reviewed. Diagnoses followed the DSM-IV criteria.12 Because clinical diagnosis may be more sensitive than structured interviews for the detection of hypomania,13 use of the clinical diagnosis is reasonable for retrospective review.

Sleep information was gleaned from clinical evaluations, which routinely included information regarding the quality of sleep and sleep duration over 24 hours, as well as a description of their sleep pattern over the 1 to 2 weeks preceding the outpatient visit. Additionally, data regarding the use of sleeping aids, sedating antipsychotics (specifically quetiapine, olanzapine, and clozapine), and stimulant medications were collected and correlated with sleep ratings. There were no objective measures of sleep.

The χ2 test was used to evaluate categorical statistical variables. Simple regression was utilized to examine relationships between continuous variables.

  RESULTS

Among the 116 patient charts reviewed, 10 never achieved a euthymic interval of 2 months’ duration. Among the remaining 106 euthymic patients, 59 (55.6%) had BD I, 23 (21.7%) had BD II, and 24 (22.8%) had BD NOS (TABLE). The mean age was 43.3±SD 14.6; 35% were male and 65% were female (TABLE). The patients were euthymic, with a mean CGI-S of 1.5±0.7, a mean mania rating of 2.3±SD 2.8 (out of 36), and a mean depression rating of 3.3±3.1 (out of 42). GAF scores were moderately low, with a mean of 68.8±10.9. Eighty-three patients (78.3%) received a mood stabilizer (lithium, valproate, carbamazepine, or lamotrigine), 44 (41.2%) received an antipsychotic (33 [31.1%] of which were sedating), 18 (17.0%) received a stimulant (methylphenidate or amphetamine), 14 (13.2%) received an antidepressant, 28 (26.4%) received a benzodiazepine (8 also received a sedating antipsychotic, 4 received a concomitant sleeping aid, and 5 received other anxiolytics), 7 (6.6%) received gabapentin, and 11 (10.4%) received a sleeping aid (zolpidem, eszopiclone, temazepam, or flurazepam), 3 of which also received a sedating antipsychotic.

A total of 25 patients (23.6%) had a clinically significant ongoing sleep disturbance (27.1% of those with BD I, 21.7% of those with BD II, and 16.6% of those with BD NOS) (TABLE). Of 16 patients for whom a sleep description was available, 25% had difficulty falling asleep, 81.25% had middle insomnia (2 patients experienced both), and none had early morning awakening. Overall sleep duration was 7.3±SD 1.7 hours. Among the 21 patients reporting a sleep disturbance for whom data were available, the average sleep duration was 6.7±2.5 hours (range 2 to 12 hours). For those sleeping <8 hours per night (n=14) the mean was 5.5±1.7 hours, while for those receiving >8 of sleep (n=7), the mean was 9.6±1.3 hours.

Half of the BD patients had a body mass index of >30 kg/m2, with an average weight of 189.9±59.6 lbs. There were no relationships between weights and the subjective sleep rating (r2=.007; P=.45), or sleep duration (r2=.004; P=.6). Weight also did not correlate with function as measured by GAF (r2=.0004; P=.9) or severity of subsyndromal symptoms as measured by CGI-S (r2=.02; P=.14).

Nine patients taking sleeping aids and 24 receiving sedating antipsychotics denied active sleep disturbance. However, if it is assumed that the sleeping aids are treating a sleep disturbance, then a total of 34 (32.1%) patients may be considered to have had underlying sleep difficulties. Similarly, if it is assumed that sedating antipsychotics and benzodiazepines prescribed for anxiety might accomplish a similar result, then 60 (56.6%) patients could potentially have a sleep disturbance that is being adequately or inadequately treated.


TABLE

Major findings in 106 bipolar disorder patients

Patient characteristic No. (%)
Diagnosis  
  Bipolar I disorder 59 (55.6%)
  Bipolar II disorder 23 (21.7%)
  NOS 24 (22.8%)
Sex  
  Male 37 (35%)
  Female 69 (65%)
Significant sleep disturbance 25 (23.6%)
Type of sleep disturbance (among 16 with adequate data)  
  Early insomnia 4 (25%)
  Middle insomnia 13 (81.25%)
  Early awakening 0
  Mean (SD)
Sleep duration overall 7.3 (1.7)
Sleep duration with sleep disturbance 6.7 (2.5)a
a For those sleeping <8 hours, (n=14) the mean was 5.5±1.7 hours. For those sleeping >8 hours (n=7), the mean was 9.6±1.3 hours.
NOS: not otherwise specified; SD: standard deviation.

  DISCUSSION

Persistent sleep disturbance appears to be relatively common in euthymic BD patients in treatment, with nearly 25% of the patients reporting some form of insomnia. However, this rate of sleep disturbances does not appear to be significantly different from the general population, in which 30% to 45% will experience insomnia in any given year.14 However, all patients in this study were in treatment and receiving medications. One tenth of the patients regularly consumed sleep aids, and an additional 43% were receiving sedating agents (benzodiazepines or sedating antipsychotics). This is greater than the prevalence of chronic insomnia in the general population of approximately 15% (25% in older patients).14 While a significant fraction of these patients denied current sleep disturbance, it is likely that these agents were chosen in part for their sedating properties. Consequently, more than half of the patients may have an adequately or inadequately treated sleep disturbance.

Interestingly, the majority of patients (>80%) reporting a sleep disturbance, reported middle insomnia. Disturbed sleep is associated with medical problems, such as pain, respiratory illness (eg, apnea, emphysema), and congestive heart failure. BD patients have notably more common medical problems.15 Furthermore, comorbid psychiatric conditions, such as posttraumatic stress disorder, which are associated with sleep disturbance,16 are much more frequent in BD patients.17 Additionally, middle insomnia may be a complication of prolonged antidepressant use in BD patients.18,19 All of these factors may contribute to the high rates of middle insomnia in BD patients.

Adequate sleep appears to be important in maintenance of euthymia in persons with BD.3 A sleep disturbance is one of the notable aspects of the prodromal phase that occurs a few months prior to a bipolar episode.20 Residual insomnia during euthymia may represent vulnerability to affective relapse in susceptible patients.7 Rapid-cycling BD patients may be especially vulnerable to mania/hypomania after disrupted sleep.21,22

Among the different sleep parameters, decreased sleep duration appears to be a good predictor of mania or hypomania the following day.23 Similarly, sleep, darkness, reduced activity, and/or endogenous rhythms may contribute to the tendency to switch into depression overnight.24 BD I patients identifying the “switches” emphasized that estimated average sleep time was significantly less the night prior to switching to mania compared with sleep time on the second, third, and fourth nights prior to switch.25

Normalization and entrainment of rhythms may partially underlie the effectiveness of interpersonal and social rhythms therapy (IPSRT).26,27 Analysis of outcome in these studies suggests that it is the maintenance of daily routines that mediates the effect of IPSRT.27 Part of this effect would be the regularity of sleep/wake patterns.

Design limitations reduce the generalizability of conclusions from the current study. This was a retrospective study in which data collection was limited by what was recorded by the clinician. While this was remediated by the use of a quantitative clinical evaluation form that followed previously validated clinical tools,10 the inherent limitations of retrospective studies remain. Additionally, there were no objective measures of sleep. This is important because the specific anomalies in the circadian cycle are needed to enlighten pathophysiologic knowledge in predicting relapse or interepisode dysfunction. A major limitation is the lack of systematically collected information about comorbid Axis I conditions. Comorbidity is common in BD patients28 and many of the comorbid conditions, such as anxiety and substance abuse, impact sleep patterns. Finally, while the overall sample size is reasonable, the number of patients in some cells is quite small (eg, only 11 patients received a sleeping aid), reducing our ability to study certain phenomena.

Despite these limitations, the study suggests that the rates of ongoing sleep disturbance in treated, euthymic BD patients do not appear to be significantly different from the general population. However, more than half of the BD patients are receiving medications that may have sedation as a direct or indirect effect, suggesting that basal (untreated) rates of sleep disturbance may be higher among BD patients than in the general population. Additionally, when a sleep disturbance is noted, it is overwhelmingly middle insomnia. Future studies examining somnographic studies are required to more fully examine sleep patterns in euthymic BD patients.

DISCLOSURES: Dr. El-Mallakh is on the speakers’ bureau of AstraZeneca, Bristol-Myers Squibb, Eli Lilly and Company, Merck, Novartis, and Pfizer. Ms. Brill and Drs. Penagaluri, Roberts, and Gao report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

    REFERENCES

  1. Harvey AG. Sleep and circadian rhythms in bipolar disorder: seeking synchrony harmony, and regulation. Am J Psychiatry. 2008;165:820–829.
  2. Wehr TA, Sack DA, Rosenthal NE. Sleep reduction as a final common pathway in the genesis of mania. Am J Psychiatry. 1987;144:201–204.
  3. Jackson A, Cavanagh J, Scott J. A systematic review of manic and depressive prodromes. J Affect Disord. 2003;74:209–217.
  4. Shi J, Wittke-Thompson JK, Badner JA, et al. Clock genes may influence bipolar disorder susceptibility and dysfunctional circadian rhythm. Am J Med Genet B Neuropsychiatr Genet. 2008;147B:1047–1055.
  5. MacQueen GM, Marriott M, Begin H, et al. Subsyndromal symptoms in longitudinal prospective follow-up of a cohort of patients with bipolar disorder. Bipolar Disord. 2003;5:349–355.
  6. Judd LL, Schettler PJ, Akiskal HS, et al. Residual symptom recovery from major affective episodes in bipolar disorders and rapid episode relapse/recurrence. Arch Gen Psychiatry. 2008;65:386–394.
  7. Plante DT, Winkelman JW. Sleep disturbance in bipolar disorder: therapeutic implications. Am J Psychiatry. 2008;165:830–843.
  8. Jones SH, Hare DJ, Evershed K. Actigraphic assessment of circadian activity and sleep patterns in bipolar disorder. Bipolar Disord. 2005;7:176–186.
  9. Millar A, Espie CA, Scott J. The sleep of remitted bipolar outpatients: a controlled naturalistic study using actigraphy. J Affect Disord. 2004;80:145–153.
  10. Sachs GS, Guille C, McMurrich SL. A clinical monitoring form for mood disorders. Bipolar Disord. 2002;4:323–327.
  11. Spearing MK, Post RM, Leverich GS, et al. Modification of the Clinical Global Impressions (CGI) scale for use in bipolar illness (BP): the CGI-BP. Psychiatry Res. 1997;73:159–171.
  12.  Diagnostic and statistical manual of mental disorders, 4th ed, text rev. Washington DC: American Psychiatric Association; 2004.
  13. Dunner DL, Tay LK. Diagnostic reliability of the history of hypomania in bipolar II patients and patients with major depression. Compr Psychiatry. 1993;34:303–307.
  14. Singer CM. Insomnia. In: Ferri FF ed. Ferri’s clinical advisor. Philadelphia, PA: Elsevier Mosby; 2006;458–459.
  15. Shah A, Shen N, El-Mallakh RS. Weight gain occurs after onset of bipolar illness in overweight bipolar patients. Ann Clin Psychiatry. 2006;18:239–241.
  16. Buljan R, Hrabrié K, Jukié V, et al. Disturbed sleep in war veterans according to overnight polysomnography [in Croatian]. Lijec Vjesn. 2008;130:101–103.
  17. El-Mallakh RS, Hollifield M. Comorbid anxiety in bipolar disorder alters treatment and prognosis. Psychiatr Q. 2008;79:139–150.
  18. El-Mallakh RS, Karippot A. Antidepressant-associated chronic irritable dysphoria (ACID) in bipolar disorder. J Affect Disord. 2005;84:267–272.
  19. El-Mallakh RS, Ghaemi SN, Sagduyu K, et al, and the STEP-BD Investigators. Antidepressant-associated chronic irritable dysphoria (ACID) in STEP-BD patients. J Affect Disord. 2008;111:372–377.
  20. Conus P, Ward J, Hallam KT, et al. The proximal prodrome to first episode mania—a new target for early intervention. Bipolar Disord. 2008;10:555–565.
  21. Kasper S, Wehr TA. The role of sleep and wakefulness in the genesis of depression and mania. Encephale. 1992;18:45–50.
  22. Riemann D, Voderholzer U, Berger M. Sleep and sleep-wake manipulations in bipolar depression. Neuropsychobiology. 2002;45(suppl):7–12.
  23. Leibenluft E, Albert PS, Rosenthal NE, et al. Relationship between sleep and mood in patients with rapid-cycling bipolar disorder. Psychiatry Res. 1996;63:161–168.
  24. Feldman-Naim S, Turner EH, Leibenluft E. Diurnal variation in the direction of mood switches in patients with rapid-cycling bipolar disorder. J Clin Psychiatry. 1997;58:79–84.
  25. Sitaram N, Gillin JC, Bunney WE Jr. The switch process in manic-depressive illness: circadian variation in time of switch and sleep and manic ratings before and after switch. Acta Psychiatr Scand. 1978;58:267–278.
  26. Frank E, Hlastala S, Ritenour A, et al. Inducing lifestyle regularity in recovering bipolar disorder patients: results from the maintenance therapies in bipolar disorder protocol. Biol Psychiatry. 1997;41:1165–1173.
  27. Frank E, Kupfer DJ, Thase ME, et al. Two-year outcomes for interpersonal and social rhythm therapy in individuals with bipolar I disorder. Arch Gen Psychiatry. 2005;62:996–1004.
  28. Merikangas KR, Akiskal HS, Angst J, et al. Lifetime and 12-month prevalence of bipolar spectrum disorder in the National Comorbidity Survey replication. Arch Gen Psychiatry. 2007;64:543–552. (Erratum in: Arch Gen Psychiatry. 2007;64:1039).

CORRESPONDENCE: Rif S. El-Mallakh, MD, Mood Disorders Research Program, Department of Psychiatry, and Behavioral Science, University of Louisville School of Medicine, MedCenter One, 501 East Broadway, Suite 340, Louisville, KY 40202 USA E-MAIL: rselma01@louisville.edu