Psychopharmacology (2002) 165:29–36 DOI 10.1007/s00213-002-1165-4

ORIGINAL INVESTIGATION

Geoffrey E. Ott · Uma Rao · Innocencia Nuccio · Keh-Ming Lin · Russell E. Poland

Effect of bupropion-SR on REM sleep: relationship to antidepressant response Received: 15 January 2002 / Accepted: 29 May 2002 / Published online: 6 November 2002
Springer-Verlag 2002

Abstract Rationale: The effects of antidepressant (AD) drugs on sleep in depressed patients and their relationship to AD response have been investigated previously. However, newer AD agents, which appear to have different effects on sleep, have not been evaluated systematically for their usefulness in predicting treatment response. Objectives: To examine the effect of bupropion sustained release (SR) (Wellbutrin SR) on sleep macroarchitecture, and to assess whether the observed electroencephalographic (EEG) sleep changes in response to a single dose of bupropion are associated with treatment response to the AD. Methods: Twenty patients with unipolar major depressive disorder received EEG sleep assessments prior to treatment. Subjects were studied twice for 2 consecutive nights, with each 2-night session approximately 1 week apart. Baseline EEG sleep and the EEG sleep responses to placebo (baseline sleep) and a single dose of bupropion SR (150 mg, PO) were measured using a randomized, double-blind, crossover design. The participants then received open-label treatment with bupropion SR for about 8 weeks. Results: No relationship was observed between baseline EEG sleep measures and response to treatment with bupropion. However, a statistically significant relationship was found between G.E. Ott · U. Rao · R.E. Poland ()) Interdepartmental Training Program for Neuroscience, UCLA, Los Angeles, CA 90024, USA e-mail: [email protected] Tel.: +1-310-4233533 Fax: +1-310-4230888 U. Rao · K.-M. Lin · R.E. Poland Department of Psychiatry, UCLA Neuropsychiatric Institute, 760 Westwood Plaza, Room 68-237, Los Angeles, CA 90024-1759, USA U. Rao · I. Nuccio · K.-M. Lin · R.E. Poland Department of Psychiatry, Harbor-UCLA Research and Education Institute, Los Angeles, CA 90509, USA U. Rao · R.E. Poland Department of Psychiatry, Cedars-Sinai Medical Center, 8730 Alden Drive, Room E-135, Los Angeles, CA 90048, USA

latency to the onset of rapid eye movement (REM) sleep following a single dose of bupropion and clinical response to treatment with bupropion. Responders showed an increase in REM latency following bupropion challenge, whereas non-responders showed a decrease. Moreover, the REM latency change in response to bupropion challenge correlated with change in depression ratings as a result of treatment. Conclusions: These findings suggest that bupropion’s effect on REM latency and its AD action might be linked, possibly via dopamine (D2) receptor-mediated effects, or by noradrenergic mechanism(s). Keywords Depression · Sleep · Antidepressant · Bupropion SR · Wellbutrin SR · Treatment

Introduction Despite the demonstrated efficacy of AD compounds for the treatment of depression, many patients do not manifest an optimal or even acceptable response (Thase and Rush 1995; Charney et al. 1998). It is recognized that about one-third of patients treated for major depression do not respond adequately to the AD drug of first choice, and approximately 10% of patients remain depressed despite multiple interventions (Paykel 1994). In addition, it often takes up to 8 weeks or longer to detect a notable clinical response (Rush et al. 1977; Quitkin et al. 1996). Consequently, many patients continue to be treated with ineffective drug(s) for prolonged periods. Hence, a pretreatment measure related to response would be of considerable clinical importance, as well as having great public health significance. Since the vast majority of patients with depression have sleep disturbances, and because depression is linked to some specific changes in EEG sleep (Benca et al. 1992; Cartwright 1993), researchers have attempted to employ sleep measures as potential predictors of response to various types of treatment for depression. Reduced REM latency is the most widely replicated EEG sleep variable

30

associated with major depression (Reynolds and Kupfer 1987; Benca et al. 1992), and data suggest that patients with reduced REM latency at baseline may respond poorly to placebo (Coble et al. 1979; Zammit et al. 1988; Heiligenstein et al. 1994). Other investigators reported that “normal” EEG sleep patterns may be associated with a better response to psychosocial interventions (Buysse et al. 1992; Thase et al. 1996, 1997). These findings suggest that patients having biological disruptions (including EEG sleep manifestations) might require pharmacological intervention(s). With respect to pharmacological treatment, both uncontrolled and controlled studies indicate that patients with shortened REM latency responded more favorably to AD treatment (Coble et al. 1979; Akiskal et al. 1980; Svendson and Christensen 1981; Rush et al. 1985, 1989). In contrast to these findings, Kupfer et al. (1981) found that reduced REM latency was associated with poor AD response, whereas Heiligenstein et al. (1994) reported that baseline REM latency had no predictive value for AD response. In addition to baseline EEG sleep measures, the relationship between acute effects of AD drugs on sleep and clinical response to treatment has been studied. The results from initial reports suggested that REM sleep suppression may be associated with better AD response (Wyatt et al. 1969; Dunleavy and Oswald 1973; Kupfer et al. 1976, 1981; Gillin et al. 1978; Hochli et al. 1986; Vogel et al. 1990; Reynolds et al. 1991). However, other studies did not replicate this finding (Mendlewicz et al. 1991; van Bemmel et al. 1992, 1993; Staner et al. 1995; Staedt et al. 1998; Wilson et al. 2000; Landolt et al. 2001). For both baseline and provoked measures of sleep, methodological differences among studies (including clinical characteristics of the target population, AD dose, and timing of EEG sleep assessment in relation to duration of AD treatment) likely contribute to the disparate findings. Additionally, the inconsistent results might be related to the type of AD drug used (Berger and Riemann 1993; Thase 1998). Previous studies have utilized primarily tricyclic agents (TCA), monoamine oxidase inhibitors (MAOI) or selective serotonin reuptake inhibitors (SSRI), all of which have pronounced effects on sleep, and on REM sleep in particular. The newer “atypical” AD compounds, which appear to have different effects on REM sleep (Sharpley et al. 1992; Ware et al. 1994; Nofzinger et al. 1995; Rush et al., 1998), have not been assessed for the predictability of treatment response. With these agents, it might be possible to discern more subtle effects on sleep that are more closely linked to their AD activity. Bupropion is an atypical AD drug (Preskorn and Othmer 1984). It differs from other AD agents in that its major pharmacological effect is on the dopamine (DA) transporter. It has some effects on norepinephrine uptake, but shows little activity on serotoninergic (5-HT) systems (Cooper et al. 1980; Ferris et al. 1982; Preskorn and Othmer 1984). In contrast to previous reports of TCA,

MAOI or SSRI effects on sleep, Nofzinger et al. (1995) observed reduced REM latency and increased REM sleep in seven depressed patients who received short-term treatment with bupropion (immediate release). This investigation was undertaken to extend the preliminary findings of bupropion’s effect on REM sleep, and to evaluate whether sleep changes following administration of a single-dose of bupropion might predict subsequent treatment response.

Materials and methods The study was performed in accordance with ethical standards laid down in the 1964 Declaration of Helsinki. The study protocol was approved by the Institutional Review Board at Harbor-UCLA Medical Center where the study was conducted, and all participants signed the written informed consent form prior to performing research procedures. Participants Subjects were recruited from the outpatient clinic at Harbor-UCLA Medical Center and through advertisements in local newspapers. All potential participants were assessed using the Structured Clinical Interview for DSM-IV (SCID; First et al. 1994) for the identification of major depressive disorder and comorbid conditions. Severity of depressive symptoms was determined by the first 17 items of the Hamilton Depression Rating Scale (HAM-D; Hamilton, 1960). Patients should have been free from AD drugs and other psychotropic agents for at least 4 weeks (8 weeks for fluoxetine) for eligibility to participate in the study. A minimum HAM-D score of 15 was required for acceptance into the study. All subjects were medically healthy, as determined by physical examination, full chemistry panel, thyroid function tests, electrocardiogram and urine drug screens. Exclusion criteria included prior use of bupropion for the treatment of depression or for other conditions (e.g. smoking), history of seizure disorder or other neurological conditions, active suicidal ideation or recent suicide attempt, and current or previous diagnosis of anorexia/bulimia nervosa, primary anxiety disorder, bipolar disorder or psychotic disorder. Also, potential subjects with substance use disorder diagnosis in the previous 6 months, patients with a personal history of sleep disorder(s), and women with suspected pregnancy were excluded from the study. Sleep protocol and scoring of sleep records Each participant was studied on two separate sessions for 2 consecutive nights during each session, with approximately 1 week between sessions. Conventional EEG electrodes were attached by 9:00 p.m., and sleep recordings were made from 11:00 p.m. (lights out) to 7:00 a.m. On the morning of night 2, subjects were given either placebo or bupropion SR (150 mg, PO) in a randomized, double-blind, cross-over fashion. The first night of each sleep study was considered an adaptation night, and only data from the second night were utilized in statistical analyses. Sleep measures following placebo administration were considered as baseline values. (Unless stated otherwise, bupropion SR will be referred to as bupropion.) The International 10-20 System was used for EEG electrode placement, electromyogram, electrooculogram and electrocardiogram. In order to rule out the presence of major sleep disorders, a full sleep polysomnography was performed on the first night, including respiratory, oximetry and leg movement measurements. Bilateral EEG recordings were obtained from left (C3) and right (C4) central leads referenced to the opposite mastoid, A2 and A1, as well as to a linked reference (A1+A2).

31 Sleep records were coded and scored “blindly” according to standard criteria (Rechtschaffen and Kales 1968). REM latency was defined as the time between sleep onset (first minute of stage 2 or deeper sleep, followed by at least 9 min of stage 2 or deeper sleep, interrupted by no more than 1 min of waking or stage 1 and the first REM period ‡3 min in length). Other REM sleep measures, including REM activity and REM density, and additional sleep variables were scored according to the criteria of Kupfer (1976), as was done previously (Poland et al. 1989, 1997). REM activity was scored on a scale ranging from 0 to 8 units. Treatment of depression with bupropion SR After the second 2-night sleep polysomnography session, patients began standard clinical treatment with bupropion SR under the care of a psychiatrist for approximately 8 weeks (mean=55.1 days; SEM=2.1 days), with weekly monitoring of symptoms and side effects. The protocol required 8 weeks of treatment. However, due to scheduling difficulties for some subjects, the final assessment was not obtained exactly at week 8. Thus, treatment duration ranged from 7 to 9 weeks. Dose adjustments were made based on reports of depressive symptoms and side effects. The final dosage ranged from 150 to 400 mg/day. Subjects who showed ‡50% reduction in HAM-D score in response to bupropion treatment were classified as responders. In order to determine change in HAM-D score in response to treatment, the final HAM-D score was subtracted from the baseline (pre-treatment) value. For post-hoc analyses, remission from depression (a final HAM-D score of £8 and £7) was used as a secondary outcome measure. Statistics Descriptive statistics were derived for all variables. Scrutiny of the W statistic revealed whether variables were suitable for parametric tests. Logarithmic transformations were performed for variables that did not meet normal distribution. Repeated measures ANOVAs were carried out over the two nights on all major EEG sleep variables. Student’s t-tests were utilized for group comparisons between responders and non-responders, and paired t-tests were used for within-subject comparisons. Pearson-product moment correlations were used for assessing relationships between variables, and multiple regression was employed to determine the association between sleep variables and treatment response.

Results Demographic and clinical parameters Twenty subjects (ten men, ten women) were studied. Of these, 11 were classified as treatment responders. Demo-

Table 1 Demographic and clinical characteristics (mean€ SEM) of the total sample, and in responders and non-responders to bupropion treatment

graphic and clinical characteristics of the entire group, and separately for responders and non-responders, are outlined in Table 1. There were no significant differences between responders and non-responders with respect to age, gender, baseline HAM-D score, duration of index depressive episode, recurrent versus non-recurrent depression, number of treatment days, body mass index (BMI), or final dose of bupropion. However, as expected, the final HAM-D score in the responders was significantly lower compared with the corresponding score in non-responders [t(10)=3.10, P£0.05]. In total, nine subjects met remission criteria at the end of treatment with a HAM-D £8 versus five with a HAM-D £7 (n=5, if a final HAM-D score of £7 was considered as criteria for remission). Age did not correlate significantly with baseline HAM-D score, final HAM-D score, or change in HAMD score in response to treatment. Duration of treatment did not correlate significantly either with the final HAMD score or with change in HAM-D score. Finally, bupropion dose did not correlate significantly with change in HAM-D score, or with the BMI. Men and women did not differ significantly with respect to age, BMI, treatment duration, or any of the HAM-D scores. However, by the end of treatment, men were taking a higher dose of bupropion than women [315.0€10.7 versus 265.0€19.8 mg, t(18)=2.22, P£0.05]. Nevertheless, the response rate was not significantly different in two groups (response rate=60% in males, and 50% in females). Effect of bupropion on EEG sleep Polysomnographic variables for all 20 patients following placebo and single-dose bupropion administration are shown in Table 2. Bupropion had no significant effect on any of the sleep continuity measures. With respect to nonREM sleep, bupropion did not affect stage 1 sleep. Bupropion significantly reduced stage 2 sleep time [F(1,19)=6.33, P£0.05], while it increased slow-wave sleep [F(1,19)=4.57, P£0.05]. Bupropion had no effect on any of the REM sleep measures. Order of placebo and drug administration (placebo versus bupropion administration during the first session) did not have an effect on any sleep or outcome variables.

Age (years) Gender (M/F) Body mass index Baseline HAM-D score Final HAM-D score Duration of index depressive episode (weeks) Recurrent depression (%) Duration of treatment (days) Final bupropion dose (mg/day) *P£0.05

Total sample (n=20)

Responders (n=11)

Non-responders (n=9)

46.2€2.8 10/10 27.6€1.0 20.3€0.8 10.6€1.3 69.7€9.4 13 (65.0) 55.1€2.1 290.0€12.4

44.7€4.2 6/5 27.3€0.9 21.1€0.9 7.5€0.8* 72.2€12.4 7 (63.6) 55.3€2.7 300.0€17.8

47.9€2.0 4/5 28.0€2.0 19.3€1.4 14.3€2.0 66.7€15.1 6 (66.7) 54.8€3.5 277.8€16.9

32 Table 2 Sleep polysomnography variables (mean€SEM) following placebo and sustained-release bupropion (150 mg, PO) in depressed patients

Sleep continuity Sleep latency (min) Total sleep time (min) Sleep efficiency (%) Awake time (min) Number of awakenings Number of arousals Sleep architecture Stage 1 sleep (%) Stage 2 sleep (%) Slow-wave sleep (%) REM sleep (%) Stage 1 sleep (min) Stage 2 sleep (min) Slow-wave sleep (min) First REM episode REM latency (min) REM activity (units) REM density (units/min) REM duration (min) All REM episodes REM activity (units) REM density (units/min) REM duration (min) Number of REM episodes

Placebo

Bupropion

F(1,19)

32.5€8.2 401.6€10.4 84.9€2.1 69.1€10.2 17.3€1.8 42.6€4.0

30.2€8.5 394.6€12.1 83.5€2.6 75.2€12.5 17.2€1.6 47.7€4.3

0.27 1.84 1.84 1.56 0.00 2.29

6.5€1.0 55.7€1.6 12.2€1.5 25.5€1.5 25.8€3.9 223.9€8.8 49.4€6.1

5.9€0.9 51.2€2.3 17.4€2.3 25.4€1.2 22.3€2.8 201.8€11.0 71.1€10.5

0.88 4.15 5.17* 0.01 1.82 6.33* 4.57*

54.6€5.2 53.0€11.2 2.3€0.3 22.4€3.3

52.1€6.1 44.6€7.8 2.6€0.3 17.0€2.5

0.15 0.39 0.80 1.99

263.3€17.9 2.6€0.2 102.6€4.4 4.3€0.2

259.9€19.9 2.6€0.2 99.4€4.8 4.5€0.2

0.03 0.03 0.45 0.88

Fig. 1 Effect of single-dose bupropion administration on REM latency in responders (n=11) and non-responders (n=9)

*P£0.05

Association between bupropion’s effect on REM sleep and clinical response There was a significant relationship interaction between bupropion’s effect on REM latency and clinical response [F(1,18)=5.71, P£0.05]. As depicted in Fig. 1, following single-dose bupropion administration, treatment responders had significantly longer REM latency compared with non-responders [66.4€7.1 versus 34.6€6.9 min, t(18)=3.18, P£0.005]. Baseline REM latency was not significantly different between the two groups [56.2€9.1 and 52.6€3.9 min in responders and non-responders, respectively; t(18)=0.74, NS]. Remitted subjects (HAMD£8) also had a significantly longer REM latency during the bupropion session than the non-remitted group [65.1€8.2 versus 41.4€7.6 min, t(18)=2.12, P£0.05], but the two groups did not differ significantly during the placebo session. Using a HAM-D£7 for remission, a similar trend was noted. Examination of the change in REM latency from placebo to bupropion administration within responder and non-responder groups revealed no significant change in responders [56.2€9.1 versus 66.4€7.0 min, t(10)=1.26, NS]. However, there was a non-significant trend for the non-responders to show a reduction in REM latency with following bupropion administration [52.6€3.9 versus 34.6€6.9, t(8)=2.11, P£0.10]. REM latency value following bupropion administration correlated significantly with change in HAM-D score

Fig. 2 Correlation between REM latency following administration of a single dose of bupropion and change in HAM-D score in response to treatment (positive value=improvement)

in response to bupropion treatment (r=0.68, P£0.001; see Fig. 2). Higher REM latency value following bupropion administration was associated with greater improvement in HAM-D score following treatment. The correlation remained significant when two individuals with extreme values were excluded (r=0.55, P£0.05). Using multiple regression analysis, a combination of REM latency and total REM activity measures following bupropion administration accounted for 62% of the variance for change in HAM-D score [F(2,17)=16.33, P£0.0001]. No other EEG sleep variables, following either placebo or bupropion, discriminated responders and non-responders.

Discussion To the best of our knowledge, this is the first report on the effects of a single dose of bupropion on EEG sleep in

33

patients with major depressive illness. Compared with placebo, a single dose of bupropion SR administration reduced stage 2 sleep and increased slow-wave sleep. No significant effects of bupropion on REM sleep were observed. Following short-term treatment with immediate release bupropion (mean duration=16.4 weeks), Nofzinger et al. (1995) reported a reduction in REM latency and increased REM sleep. The lack of effect of a single dose of bupropion on REM sleep was different from that which occurred following a challenge with a TCA (Kupfer et al. 1981, 1982; Hochli et al. 1986; Feuillade et al. 1992), SSRI (Oswald and Adam 1986; Saletu et al. 1991; van Bemmel et al. 1993), venlafaxine (Salin-Pascual et al. 1997), or trazodone (van Bemmel et al. 1992; Ware et al. 1994). However, the results were consistent with findings from studies of other atypical AD agents, such as mirtazapine (Winokur et al. 2000), as well as some (Sharpley et al. 1996; Vogel et al. 1998), but not other investigations using nefazodone (Sharpley et al. 1992). No normal controls were assessed in this study for comparison purposes. However, baseline REM latency values in both responders and non-responders were consistent with the findings in our previous studies of depressed patients, being significantly lower compared with controls (see Poland et al. 1997). Although bupropion did not significantly affect REM sleep in the patient sample as a whole, differences in REM latency emerged when the group was categorized as responders and nonresponders to treatment. Although baseline REM latency did not differentiate the two groups, responders showed an increase in REM latency after single-dose bupropion administration, whereas non-responders showed a decrease. Furthermore, REM latency correlated with the degree of clinical improvement as measured by the change in HAM-D score. These findings suggest that there might be a common mechanism underlying the effects of bupropion on REM latency and subsequent clinical response to the drug. As mentioned previously, bupropion has been shown to have an inhibitory effect on the DA transporter (Cooper et al. 1980; Ferris et al. 1982; Preskorn and Othmer 1984). Studies on the pharmacology of REM sleep have found that D1 (Monti et al. 1990; Gaillard et al. 1994), as well as D2 (Monti et al. 1988), receptor activation suppresses REM sleep. Similarly, both D1 (Sampson et al. 1991; Baamonde et al. 1992) and D2 (Sampson et al. 1991) receptor activation have been associated with AD effects. Thus, patients who are more sensitive to D1 and/or D2 activation are likely to show greater REM latency effects from bupropion and better treatment response to the drug. This differential sensitivity could be caused by DNA variants in the D2 receptor (Propping and Nothen 1995), or polymorphisms in the DA transporter (Gill et al. 1997; Wisor et al. 2001), both of which could mediate the different degrees of DA receptor-coupled activation involved in both sleep and mood responses to the drug. Neuroanatomical and pharmacological research suggests some possible mechanisms for DA’s effects on

REM sleep and mood. Since activation of D2 receptors in the dorsal raphe nuclei results in increased 5-HT levels (Monti et al. 1999), and the dorsal raphe, in turn, regulates cholinergic REM-generating systems in the pedunculopontine tegmental nucleus (McCarley et al. 1995), the increase in DA neurotransmission induced by bupropion potentially could suppress the appearance of REM sleep through activation of 5-HT neurotransmission. Other evidence suggests that rewarding events activate the mesocorticolimbic DA system, which in turn innervates the amygdala, hippocampus, prefrontal cortex and nucleus accumbens. These latter areas innervated by the mesocorticolimbic DA system are all associated with mood regulation (Willner 1995), and might be involved in bupropion’s AD effects. Consistent with the observed effects of DA systems on mood regulation and REM sleep in animal studies, clinical studies using positron emission tomography have shown altered metabolism in brain regions innervated by mesolimbic and mesocortical DA systems in depressed patients during sleep deprivation or during REM sleep (Nofzinger et al. 1999; Wu et al. 1999). Preliminary evidence suggests that AD response to sleep deprivation, or to AD drugs, may reverse these changes (Buchsbaum et al. 1997; Mayberg et al. 1997; Nofzinger et al. 2001; Wu et al. 2001). The finding that a single dose of bupropion increases slow-wave sleep at the expense of stage 2 sleep is inconsistent with the idea that bupropion is an “activating” drug (Soroko et al. 1977; van Wyck Fleet et al. 1983; Settle 1998), which more likely would decrease slowwave sleep. However, the increase in slow-wave sleep is consistent with the findings by Monti et al. (1988) who showed a low dose of the D2 receptor agonist, apomorphine, increased slow-wave sleep, while higher doses produced the opposite effect. The response to bupropion was similar to that elicited by low-dose apomorphine. Nofzinger et al. (1995), using a high immediate release dose of bupropion (mean dose=428.6 mg/day) found that slow-wave sleep was reduced during chronic treatment, similar to the effects produced by a higher dose of apomorphine. In addition to its effect on DA neurotransmission, bupropion also affects norepinephrine uptake (Cooper et al. 1980; Preskorn and Othmer 1984). Because noradrenergic systems also have been shown to affect REM sleep, the REM latency changes observed in the current study also could result from bupropion-induced changes in noradrenergic neurotransmission as opposed to a DA mechanism(s). Future investigations, using selective antagonists at DA or noradrenergic receptors, should be able to clarify this issue. Our observation is consistent with some (Kupfer et al. 1976, 1981; Gillin et al. 1978; Hochli et al. 1986), but not all (Staner et al. 1995), previous studies on the relationship between REM latency response to acute AD administration and subsequent clinical response. The discrepancy in findings may, be at least partly, accounted for by the type and dose of AD used for the challenge. For example, some AD drugs, such as amitriptyline, are

34

pharmacologically non-selective and can influence REM sleep by different mechanisms (e.g. 5-HT and anticholinergic effects). Consequently, the REM sleep response might or might not necessarily be linked to systems involved in mediating clinical response to the drug. Similarly, previous studies have used doses of AD agents which produced robust effects on REM sleep in all patients (Kupfer et al. 1981; Hochli et al. 1986; Staner et al. 1995). In studies with amitriptyline, clomipramine and paroxetine challenges, REM latency was delayed for up to 197.5 min (range 129.5–197.5). This is a much larger response than that produced by bupropion, which had no overall effect in the total patient sample studied. It is possible that the magnitude of REM latency response to the other AD challenges overwhelmed individual variability. Differences in sensitivity that might have been related to subsequent treatment response possibly were lost due to a “ceiling” effect. It remains to be determined if the REM latency response to lower doses of TCAs and SSRIs might be related to treatment response, similar to that observed herein with bupropion. In the current study, REM latency was further shortened in patients who did not respond to treatment with bupropion. The mechanisms underlying the “worsening” of REM latency in non-responders are not clear. One potential mediator might be through the hypothalamic-pituitary-adrenal (HPA) axis (Ehlers and Kupfer 1987; Poland et al. 1989; Steiger and Holsboer 1997). Consistent with this hypothesis, the non-responders showed higher levels of urinary free cortisol following single dose bupropion administration compared with the responders (Rao et al., unpublished data). Further investigation of the relationships among REM sleep regulation, HPA activity and AD treatment, in conjunction with the use of corticotrophin releasing hormone or glucocorticoid antagonists as adjunctive agents, might be able to shed light on this issue. In summary, a more favorable AD response to treatment with bupropion was related to a longer REM latency following a single dose of the drug. This association was strengthened by also taking total REM activity into account, with greater clinical improvement occurring in subjects with reduced REM activity after bupropion administration. Bupropion also significantly increased slow-wave sleep at the expense of stage 2 sleep, although this effect was not related to treatment outcome. The degree of REM latency and mood-elevating response to bupropion, as well as the slow-wave sleep increase, potentially could be accounted for by differential sensitivity of the DA systems, possibly D2 receptor-coupled responses, although noradrenergic mechanism(s) cannot be ruled out at this time. Acknowledgements This study was supported in part by the Research Center on the Psychobiology of Ethnicity (MH47193), MH34471, by the NIH General Clinical Research Center (RR00425), the Multi-Site Sleep Training Grant (MH18825), by an NIMH Scientist Development Award to U. Rao (MH01419), and by the Glaxo Wellcome Company.

References Akiskal HS, Rosenthal TL, Haykal RF, Lemmi H, Rosenthal RH, Scott-Strauss A (1980) Characterological depressions: clinical and sleep EEG findings separating subaffective dysthymias from character spectrum disorders. Arch Gen Psychiatry 37:777–783 Baamonde A, Dauge V, Ruiz-Gayo M, Fulga IG, Turcaud S, Fourni-Zaluski MC, Roques BP (1992) Antidepressant-type effects of endogenous enkephalins protected by systemic RB 101 are mediated by opioid delta and D1 receptor stimulation. Eur J Pharmacol 216:157–166 Benca RM, Obermeyer WH, Thisted RA, Gillin JC (1992) Sleep and psychiatric disorders: a meta-analysis. Arch Gen Psychiatry 49:651–668 Berger M, Riemann D (1993) REM sleep in depression – an overview. J Sleep Res 2:211–223 Buchsbaum MS, Wu J, Siegel BV, Hackett E, Trennary M, Abel L, Reynolds C (1997) Effect of sertraline on regional metabolic rate in patients with affective disorder. Biol Psychiatry 41:15– 22 Buysse DJ, Kupfer DJ, Frank E, Monk TH, Ritenour A, Ehlers CL (1992) Electroencephalographic sleep studies in depressed outpatients treated with interpersonal psychotherapy: I. baseline studies in responders and non-responders. Psychiatry Res 40:13–26 Cartwright RD (1993) Sleep problems. In: Costell GG (ed) Symptoms of depression. Wiley, New York, pp 243–257 Charney DS, Berman RM, Miller HL (1998) Treatment of Depression. In: Schatzberg AF, Nemeroff CB (eds) The American Psychiatric Press textbook of psychopharmacology, 2nd edn. American Psychiatric Press, Washington D.C. pp 705– 731 Coble PA, Kupfer DJ, Spiker DG, Neil JF, McPartland RJ (1979) EEG sleep in primary depression: a longitudinal placebo study. J Affect Disord 1:131–138 Cooper BR, Hester TJ, Maxwell RA (1980) Behavioral and biochemical effects of the antidepressant bupropion (Wellbutrin): evidence for selective blockade of dopamine in vivo. J Pharmacol Exp Ther 215:127–134 Dunleavy DLF, Oswald I (1973) Phenelzine: mood response and sleep. Arch Gen Psychiatry 28:353–356 Ehlers CL, Kupfer DJ (1987) Hypothalamic peptide modulation of EEG sleep in depression: a further application of the S-process hypothesis. Biol Psychiatry 22:513–517 Ferris RM, Maxwell RA, Cooper BR, Soroko FE (1982) Neurochemical and neuropharmacological investigations into the mechanisms of action of bupropion HCl – a new atypical antidepressant agent. In: Costa E, Racagni G (eds) Advances in biochemical psychopharmacology: 1. typical and atypical antidepressants: molecular mechanisms. Raven Press, New York, pp 277–286 Feuillade P, Pringuey D, Belugou JL, Robert P, Darcourt G (1992) Trimipramine: acute and lasting effects on sleep in healthy and major depressive subjects. J Affect Disord 24:135–146 First MB, Spitzer RL, Gibbon M, Williams JB (1994) Structured clinical interview for DSM-IV axis I disorders: SCID-I clinician version. American Psychiatric Association, Washington D.C. Gaillard J-M, Nicholson AN, Pascoe PA (1994) Neurotransmitter systems. In: Kryger M, Roth T, Dement W (eds) Principles and practices of sleep medicine, 2nd edn. Saunders, Philadelphia, pp 338–348 Gill M, Daly G, Heron S, Hawi Z, Fitzgerald M (1997) Confirmation of association between attention deficit hyperactivity disorder and a dopamine transporter polymorphism. Mol Psychiatry 2:311–313 Gillin JC, Wyatt RJ, Fram D, Snyder F (1978) The relationship between changes in REM sleep and clinical improvement in depressed patients treated with amitriptyline. Psychopharmacology 59:267–272

35 Hamilton M (1960) A rating scale for depression. J Neurol Neurosurg Psychiatry 25:56–62 Heiligenstein JH, Faries DE, Rush AJ, Andersen JS, Pande AC, Roffwarg HP, Dunner D, Gillin JC, James SP, Lahmeyer H, Zajecka J, Tollefson GD, and Gardner, DM (1994) Latency to rapid eye movement sleep as a predictor of treatment response to fluoxetine and placebo in nonpsychotic depressed outpatients. Psychiatry Res 52:327–339 Hochli D, Riemann D, Zulley J, Berger M (1986) Initial REM sleep suppression by clomipramine: a prognostic tool for treatment response in patients with a major depressive disorder. Biol Psychiatry 21:1217–1220 Kupfer DJ (1976) REM latency: a psychobiologic marker for primary depressive disease. Biol Psychiatry 11:159–174 Kupfer DJ, Foster G, Reich L, Thompson KS, Weiss B (1976) EEG sleep changes as predictors in depression. Am J Psychiatry 133:622–626 Kupfer DJ, Spiker DG, Coble PA, Neil JF, Ulrich R, Shaw DH (1981) Sleep and treatment prediction in endogenous depression. Am J Psychiatry 138:429–434 Kupfer DJ, Spiker DG, Rossi A, Coble PA, Shaw D, Ulrich R (1982) Nortriptyline and EEG sleep in depressed patients. Biol Psychiatry 17:535–546 Landolt H-P, Raimo EB, Schnierow BJ, Kelsoe JR, Rapaport MH, Gillin JC (2001) Sleep and sleep electroencephalogram in depressed patients treated with phenelzine. Arch Gen Psychiatry 58:268–276 Mayberg HS, Brannan SK, Mahurin RK, Jerabek PA, Brickman JS, Tekell JL, Silva JA, McGinnis S, Glass TG, Martin CC, Fox PT (1997) Cingulate function in depression: a potential predictor of treatment response. Neuroreport 8:1057–1061 McCarley RW, Greene RW, Rainnie D, Portas CM (1995) Brainstem neuromodulation and REM sleep. Semin Neurosci 7:341–354 Mendlewicz J, Kempenaers C, de Maertelaer V (1991) Sleep EEG and amitriptyline treatment in depressed inpatients. Biol Psychiatry 30:691–702 Monti JM, Hawkins M, Jantos H, D’Angelo L, Fernandez M (1988) Biphasic effects of dopamine D-2 receptor agonists on sleep and wakefulness in rat. Psychopharmacology 95:395–400 Monti JM, Fernandez M, Jantos H (1990) Sleep during acute dopamine D1 agonist SKF 38393 and D1 antagonist SCH 23900 administration in rats. Neuropsychopharmacology 3:153–162 Monti JM, Ponzoni A, Jantos H, Lagos P, Silveira R, Banchero P (1999) Effects of accumbens m-chlorophenylbiguanide microinjections on sleep and waking in intact and 6-hydroxydopamine-treated rats. Eur J Pharmacol 364:89–98 Nofzinger EA, Reynolds CF III, Thase ME, Frank E, Jennings JR, Fasiczka AL, Sullivan LR, Kupfer DJ (1995) REM sleep enhancement by bupropion in depressed men. Am J Psychiatry 152:274–276 Nofzinger EA, Nichols TE, Meltzer CC, Price J, Steppe DA, Miewald JM, Kupfer DJ, Moore RY (1999) Changes in forebrain function from waking to REM sleep in depression: preliminary analyses of (18F)FDG PET studies. Psychiatry Res 91:59–78 Nofzinger EA, Berman S, Fasiczka A, Mielwald JM, Meltzer CC, Price JC, Sembrat RC, Wood A, Thase ME (2001) Effects of bupropion SR on anterior paralimbic function during waking and REM sleep in depression: preliminary findings using (18F)FDG PET. Psychiatry Res 106:95–111 Oswald I, Adam K (1986) Effects of paroxetine on human sleep. Br J Clin Pharmacol 22:97–99 Paykel ES (1994) Epidemiology of refractory depression. In: Nolen WA, Zohar I, Roose SP (eds) Refractory depression: current strategies and future directions. Wiley, New York, pp 3– 18 Poland RE, Tondo L, Rubin RT, Trelease RB, Lesser IM (1989) Differential effects of scopolamine on nocturnal cortisol secretion, sleep architecture, and REM latency in normal volunteers: relation to sleep and cortisol abnormalities in depression. Biol Psychiatry 25:403–412

Poland RE, McCracken JT, Lutchmansingh P, Lesser IM, Tondo L, Edwards C, Boone KB, Lin K-M (1997) Differential response of rapid eye movement sleep to cholinergic blockade by scopolamine in currently depressed, remitted, and normal control subjects. Biol Psychiatry 41:929–938 Preskorn SH, Othmer SC (1984) Evaluation of bupropion hydrochloride: the first of a new class of atypical antidepressants. Pharmacotherapy 4:20–34 Propping P, Nothen MM (1995) Genetic variation of CNS receptors – a new perspective for pharmacogenetics. Pharmacogenetics 5:318–325 Quitkin FM, McGrath PJ, Stewart JW, Ocepek-Welikson K, Taylor BP, Nunes E, Deliyannides D, Agosti V, Donovan SJ, Petkova E, Klein DF (1996) Chronological milestones to guide drug change. When should clinicians switch antidepressants? Arch Gen Psychiatry 53:785–792 Rechtschaffen A, Kales A (1968) A manual of standardized terminology, techniques, and scoring system for sleep stages of human subjects. NIH Publication 204, Bethesda, Md. Reynolds CF III, Kupfer DJ (1987) Sleep research in affective illness: state of the art circa 1987. Sleep 10:199–215 Reynolds CF III, Hoch CC, Buysse DJ, George CJ, Houck PR, Mazumdar S, Miller M, Pollock BG, Rifai H, Frank E, Cornes C, Moryez RK, Kupfer DJ (1991) Sleep in late-life recurrent depression: changes during early continuation therapy. Neuropsychopharmacology 5:85–96 Rush AJ, Beck T, Kovacs M, Hollen SD (1977) Comparative efficacy of cognitive therapy and pharmacotherapy in the treatment of depressed outpatients. Cognit Ther Res 1:17–37 Rush AJ, Erman MK, Schlesser MA, Roffwarg HP, Vasavada N, Khatami M, Fairchild C, Giles DE (1985) Alprazolam vs amitriptyline in depressions with reduced REM latencies. Arch Gen Psychiatry 42:1154–1159 Rush AJ, Giles DE, Jarrett RB, Feldman-Koffler F, Debus JR, Weissenburger J, Orsulak PJ, Roffwarg HP (1989) Reduced REM latency predicts response to tricyclic medication in depressed outpatients. Biol Psychiatry 26:61–72 Rush AJ, Armitage R, Gillin JC, Yonkers KA, Winokur A, Moldofsky H, Vogel GW, Kaplita SB, Fleming JB, Montplaisir J, Erman MK, Albala BJ, McQuade RD (1998) Comparative effects of nefazodone and fluoxetine on sleep in outpatients with major depressive disorders. Biol Psychiatry 44:3–14 Saletu B, Frey R, Krupka M, Anderer P, Grunberger J, See WR (1991) Sleep laboratory studies on the single-dose effects of serotonin reuptake inhibitors paroxetine and fluoxetine on human sleep and awakening qualities. Sleep 14:439–447 Salin-Pascual RJ, Garcia-Polo L, Drucker-Colin R (1997) Sleep changes after four continuous days administration of venlafaxine in normal volunteers. Sleep Res 26:126 Sampson D, Willner P, Muscat R (1991) Reversal of antidepressant action by dopamine antagonists in an animal model of depression. Psychopharmacology 104:491–495 Settle EC (1998) Bupropion sustained release: side effect profile. J Clin Psychiatry 59:32–36 Sharpley AL, Walsh AES, Cowen PJ (1992) Nefazodone – a novel antidepressant – may increase REM sleep. Biol Psychiatry 31:1070–1073 Sharpley AL, Williamson DJ, Attenburrow MEJ, Pearson G, Sargent P, Cowen PJ (1996) The effects of paroxetine and nefazodone on sleep: a placebo controlled trial. Psychopharmacology 126:50–54 Soroko FE, Mehta NB, Maxwell RA, Ferris RM, Schroeder DH (1977) Bupropion hydrochloride (€) a-t-butylamino-3-chloropropiophenone HCl: a novel antidepressant agent. J Pharm Pharmacol 29:767–770 Staedt J, Hunerjager H, Ruther E, Stoppe G (1998) Sleep cluster arousal analysis and treatment response to heterocyclic antidepressants in patients with major depression. J Affect Disord 49:221–227 Staner L, Kerkhofs M, Detroux D, Leyman S, Linkowski P, Medlewicz J (1995) Acute, subchronic, and withdrawal sleep

36 EEG changes during treatment with paroxetine and amitriptyline: a double-blind randomized trial in major depression. Sleep 18:470–477 Steiger A, Holsboer (1997) Neuropeptides and human sleep. Sleep 20:1038–1052 Svendson K, Christensen PG (1981) Duration of REM sleep latency as predictor of effect of antidepressant therapy. Acta Psychiatr Scand 64:238–243 Thase ME (1998) Depression, sleep and antidepressants. J Clin Psychiatry 59:55–65 Thase ME, Rush AJ (1995) Treatment-resistant depression. In: Bloom FE, Kupfer DJ (eds) Psychopharmacology: the fourth generation of progress. Raven Press, New York, pp 1081–1097 Thase ME, Simons AD, Reynolds CF III (1996) Abnormal electroencephalographic sleep profiles in major depression: association with response to cognitive behavior therapy. Arch Gen Psychiatry 53:99–108 Thase ME, Buysse DJ, Frank E, Cherry CR, Cornes CL, Mallinger AG, Kupfer DJ (1997) Which depressed patients will respond to interpersonal psychotherapy? The role of abnormal EEG sleep profiles. Am J Psychiatry 154:502–509 van Bemmel AL, Havermans AG, van Diest R (1992) Effects of trazodone on EEG sleep and clinical state in major depression. Psychopharmacology 107:569–574 van Bemmel AL, van den Hoofdakker RH, Beersma DGM, Bouhuys AL (1993) Changes in sleep polygraphic variables and clinical state in depressed patients during treatment with citalopram. Psychopharmacology 113:225–230 van Wyck Fleet J, Manberg PJ, Miller LL, Harto-Truax N, Sato T, Fleck RJ, Stern WC, Cato AE (1983) Overview of clinically significant adverse reactions to bupropion. J Clin Psychiatry 44:191–196 Vogel GW, Buffenstein A, Minter K, Hennessey A (1990) Drug effects on REM sleep and on endogenous depression. Neurosci Biobehav Rev 14:49–63

Vogel G, Cohen J, Mullis D, Kensler T, Kaplita S (1998) Nefazodone and REM sleep: how do antidepressant drugs decrease REM sleep? Sleep 21:70–77 Ware JC, Rose FV, McBrayer RH (1994) The acute effects of nefazodone, trazodone and buspirone on sleep and sleep-related penile tumescence in normal subjects. Sleep 17:544–550 Willner P (1995) Dopaminergic mechanisms in depression and mania. In: Bloom FE, Kupfer DJ (eds) Psychopharmacology: the fourth generation of progress. Raven Press, New York, pp 921–931 Winokur A, Sateia MJ, Hayes JB, Bayles-Dazet W, MacDonald MM, Gary KA (2000) Acute effects of mirtazapine on sleep continuity and sleep architecture in depressed patients: a pilot study. Biol Psychiatry 48:75–78 Wilson SJ, Bell C, Coupland NJ, Nutt DJ (2000) Sleep changes during long-term treatment of depression with fluoxamine – a home based study. Psychopharmacology 149:360–365 Wisor JP, Seiji N, Sora I, Uhl GH, Mignot E, Edgar DM (2001) Dopaminergic role in stimulant-induced wakefulness. J Neurosci 21:1987–1794 Wu J, Buchsbaum MS, Gillin JC, Tang C, Cadwell S, Wiegand M, Najafi A, Klein E, Hazen K, Bunney WE (1999) Prediction of antidepressant effects of sleep deprivation by metabolic rates in the ventral anterior cingulate and medial prefrontal cortex. Am J Psychiatry 156:1149–1158 Wu JC, Buchsbaum M, Bunney WE (2001) Clinical neurochemical implications of sleep deprivation’s effects on the anterior cingulate of depressed responders. Neuropsychopharmacology 25:S74–S78 Wyatt RJ, Kupfer DJ, Scott J, Robinson DS, Snyder F (1969) Longitudinal studies of the effect of monoamine oxidase inhibitors on sleep in man. Psychopharmacologia 15:236–244 Zammit Z, Rosenbaum A, Stokes P, Davis J, Zorick F, Roth T (1988) Biological differences in endogenous depressive placebo responders versus non-responders: dexamethasone suppression test and sleep EEG data. Biol Psychiatry 24:97–101