Sleep Breath DOI 10.1007/s11325-015-1152-8
ORIGINAL ARTICLE
Effect of sertraline on breathing in depressed patients without moderate-to-severe sleep-related breathing disorders Bin Zhang & Yanli Hao & Fujun Jia & Xueli Li & Yi Tang & Huirong Zheng & Wuhan Liu
Received: 9 January 2015 / Accepted: 25 February 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Background Previous studies have reported that selective serotonin reuptake inhibitors (SSRIs) might improve sleeprelated breathing disorders (SRBDs). However, the effects of SSRIs on breathing are not evaluated in subjects without moderate-to-severe SRBDs. Further, many symptoms of depression and SRBDs overlap, and so, it is interesting whether there are interactions between breathing and psychopathologic symptoms during SSRI treatment for depression. Methods Data were taken from an open-label 8-week trial of sertraline in depressed patients with insomnia (n=31). The depressed patients were administered 50 mg sertraline at 8 AM on the first day, and the dosage was subsequently titrated up to a maximum of 200 mg/day during the 8-week trial. All the patients were tested by repeated polysomnography (PSG) (baseline, 1st day, 14th day, 28th day, and 56th day). Sleep-disordered breathing events were categorized as apneas, hypopneas, and respiratory event-related arousals (RERAs). Results The clinical responses and PSG characteristics improved continuously during the 8-week trial. From the 14th day on, the RERA index during all-night and nonrapid eye movement (NREM) sleep became stable and
Clinical trial registry An 8-week, open-label study to evaluate the effect of sertraline on the polysomnogram of depressed patients with insomnia (http://clinicaltrials.gov/ct2/show/NCT01032434). Registry identifier: NCT01032434 Bin Zhang and Yanli Hao contributed equally to this work. B. Zhang (*) : F. Jia : X. Li : Y. Tang : H. Zheng : W. Liu Guangdong General Hospital, Guangdong Academy of Medical Sciences, Guangdong Mental Health Center, Guangzhou 510120, China e-mail: [email protected] Y. Hao Department of Human Anatomy, Guang Zhou Medical University, Guangzhou 510182, China
significantly higher than baseline and the first day (RERA index 7.3±2.2 at baseline, 7.3±2.5 on the 1st day, 4.4± 1.9 on the 14th day, 3.9±1.3 on the 28th day, 4.2±2.0 on the 56th day, F=5.71, P=0.02; NREM-RERA index 6.2± 2.0 at baseline, 6.3±2.3 on the 1st day, 3.2±1.5 on the 14th day, 3.5±0.9 on the 28th day, 3.2±1.7 on the 56th day, F=4.92, P=0.03). Additionally, the NREM-apnea index showed a similar pattern to that of the RERA index and reached a significant difference between baseline (1.0 ±0.5) and the 14th day (0.5±0.4) (KW=4.28, P=0.047). Compared to the no-improvement group, the improvement group with a decreasing score rate of the respiratory disturbance index (RDI) greater than or equal to −50 % had a more positive decreasing score rate of slow wave sleep (SWS) (439.0±78.2 vs 373.2±77.9 %, T=3.46, P= 0.04) and a more negative decreasing score rate on the arousal index (−43.7 ± 16.7 vs −26.6 ± 9.7 %, T = 9.16, P =0.01), Pittsburgh Sleep Quality Index (PSQI) scores (−65.1 ± 33.7 vs −49.6 ± 21.4 %, T = 4.74, P = 0.05), and Epworth Sleepiness Scale (ESS) scores (−55.7±21.3 vs −36.4±17.5 %, T=6.44, P=0.02). Discussion This research indicates that SRBDs could be improved to some extent by sertraline treatment, which might be more common in patients with relatively more severe sleepdisordered breathing (e.g., RDI≥10 in the current study). Although the sertraline-induced SRBD improvement seems not to have a significant clinical effect, the SRBD improvement group with decreasing score rate of RDI greater than or equal to −50 % has better subjective and objective sleep aspects than the no-improvement group. Thus, the fact that the SRBDs’ improvement was related to SSRIs might have a potential clinical benefit in the antidepressant treatment.
Keywords Sertraline . Sleep-related breathing disorders (SRBDs) . Depression . Eight-week trial
Sleep Breath
Introduction Sleep-related breathing disorders (SRBDs) are highly prevalent and underdiagnosed diseases associated with substantial morbidity and increased mortality [1]. The abrupt state-related (wake to sleep) decrease in upper airway muscle activity is a major factor in the development of upper airway obstruction, and so, the positive airway pressure (PAP) is currently the most efficacious treatment for SRBDs [1]. However, PAP cannot completely alleviate sleep-disordered breathing events in all patients and cannot be tolerated by some patients [2]. In this situation, medications that target the neurochemical events that underlie the state-dependent obstruction are hypothesized to be the putative therapies for SRBDs [3]. Some animal experiments have shown that serotonin could alleviate SRBDs by stimulating hypoglossal motor nuclei to maintain upper airway potency [4, 5], stimulating respiration, and/or suppressing rapid eye movement (REM) sleep [6, 7]. A recent study in the Chinese Han population reported that the serotonin transporter (5-HTT) gene might be involved in susceptibility to obstructive sleep apnea (OSA), which is the most common subtype of SRBDs [8]. Most antidepressants have some serotoninergic property, and the selective serotonergic reuptake inhibitors (SSRIs) are purer serotoninergic agents. So, SSRIs might provide more insight into the effect of serotonin on the respiratory system. To date, only two SSRIs (fluoxetine and paroxetine) have been evaluated in humans [9–12]. One 4-week study reported that fluoxetine significantly decreased the proportion of REM sleep time and decreased the number of apneas or hypopneas in non-REM (NREM) sleep. It is noted that some patients did not improve their breathing in this trail, which might suggest individual variability [10]. Another prospective study reported that combined treatment with fluoxetine and ondansetron was well tolerated and yielded a potentially therapeutic response in some subjects with SRBDs [12]. On the other hand, the acute administration of paroxetine could augment peak inspiratory genioglossus activity during NREM sleep, but this effect was insufficient to decrease the frequency of obstructive apnea [11]. Another double-blind, placebo-controlled crossover study reported that paroxetine improved breathing during NREM sleep in OSA patients, which was poorly related to effects on sleep architecture or daytime symptoms [9]. These few studies cannot elucidate the effect of SSRIs on SRBDs. Further, there are some flaws of these studies. They compared only sleep-disordered breathing events before and after pharmacotherapy, so the night-to-night variability of respiratory events might be underappreciated. In addition, they only evaluated subjects with moderate-to-severe SRBDs. However, the sleep heart health study has shown that cardiovascular risk is as substantial for a respiratory disturbance index (RDI) of 5 as it for an RDI of 20 or 40 [13]. As a result, it is crucial to evaluate the effects of SSRIs on breathing in
subjects without moderate-to-severe SRBDs. On the other hand, it is noted in both clinical and community samples that there is a higher prevalence of depression in people with SRBDs and many of the symptoms of depression and SRBDs overlap, such as insomnia, sleepiness, and fatigue [14]. Thus, it is interesting whether there are interactions between breathing and psychopathologic symptoms during SSRI treatment. The main purpose of this study is to characterize the effect of sertraline on SRBDs and psychopathologic symptoms in depressed patients in an 8-week clinical trial with repeated polysomnography (PSG) assessment.
Methods Patients and study design The protocol of this study was approved by the Independent Ethics Committee (IEC) of Guangdong Provincial Mental Health Center. Written informed consent forms were signed prior to participation. All the patients were enrolled from the inpatient population of Guangdong Provincial Mental Health Center. If a patient was diagnosed with a single or recurrent type of major depressive disorder according to the Diagnostic and Statistical Manual of Mental Disorders Fourth Edition (DSM-IV) upon admission, the patient’s diagnosis would be ascertained by one of the authors (BZ, psychiatrist) using the second version of the Structured Clinical Interview for DSM-IVAxis I Disorders (SCID-II) [15]. None of the patients included in the study fulfilled any other current or lifetime diagnostic criteria of DSM-IV Axis I disorders. The patients were 18- to 65-yearold males and females with a Hamilton Rating Scale for Depression (HRSD) score ≥18 and a sleep disturbance factor score in the HRSD ≥3 [16], which reflects a moderate-tohigh level of illness severity (depression and insomnia). Possible concurrent medical disorders were ruled out by a thorough medical examination and laboratory tests (electroencephalography (EEG), electrocardiograph [ECG], computed tomography [CT], blood analysis, and urinary analysis). Patients were excluded if they had experienced serious adverse events while taking sertraline, currently had significant suicidal or homicidal tendencies (medical history or item 3 “suicide” in HRSD ≥4), were currently pregnant or breastfeeding, were currently shift workers, currently had a significant sleep disorder (e.g., OSA, periodic limb movement syndrome (PLMS), and restless leg syndrome), or had a serious medical condition in the previous 3 months. After a 7-day screen and/or washout phase and the baseline PSG assessment, the patients received sertraline for 8 weeks. At baseline and during the four visits (1st day, 14th day, 28th day, and 56th day), the patients were assessed by HRSD (clinical improvement), the Treatment Emergent Symptom Scale
Sleep Breath
(TESS-Severity [TESS-S] and TESS-Treatment [TESS-T]: side effects) [17], the Epworth Sleepiness Scale (ESS: sleepiness) [18], and the Pittsburgh Sleep Quality Index (PSQI: sleep quality) [19]. All of these scales have validated Chinese versions [20]. Remitted patient was defined as HRSD ≤7 at the end of the treatment period, and Suicidal ideation was defined as item 3 “suicide” in HRSD ≥2. Because SRBDs might lead to hypertension [21], systolic and diastolic blood pressures (BP, mmHg) were recorded at 8 AM at every visit. Fifty milligrams of sertraline was administered at 8 AM on the first day. It was then titrated according to clinical efficacy and side effects, with a maximum dosage of 200 mg/day. Concomitant use of central nervous system medications during the trial, especially benzodiazepines and sedatives, was prohibited. Polysomnography and sleep-disordered breathing events The related procedure for nocturnal PSG assessments and multiple sleep latency test (MSLT) has been described in our previous paper [22]. All of the sleep variables were derived from the visual scoring of recordings using standard criteria and were divided into two groups: sleep continuity indices and sleep architecture indices. Sleep continuity indices included the total recording time (TRT, “lights out” to “lights on” in minutes), total sleep time (TST), sleep efficiency (SE, the TST divided by the TRT), sleep latency (SL, “lights out” to the first epoch of any sleep in minutes), REM latency (sleep onset to the first epoch of REM stage in minutes), wake after sleep onset (WASO, stage W during TRT, minus SL, in minutes), and arousal index (the number of arousals divided by TST). The sleep architecture indices included the percentages a patient was in each stage (the time in stage 1, stage 2, stage 3, and stage REM divided by the TST) [23]. The five-nap MSLT was performed according to the standard recommendation to determine the SL [24]. All the computerized sleep data were further edited by an experienced PSG technologist, and this technologist was blind to this research. Sleep stages, respiratory events, and periodic limb movements were scored according to American Academy of Sleep Medicine (AASM) criteria at 30-s intervals [23]. As per the AASM criteria, sleep-disordered breathing events were categorized as apneas (a decrease in airflow ≥90 % from baseline for ≥10 s), hypopneas (decrease in airflow ≥30 % for ≥10 s and followed by a desaturation ≥3 % from the pre-event baseline), and respiratory event-related arousals (RERAs) (a sequence of breaths lasting ≥10 s, which was associated with flattening of the nasal pressure waveform leading to an arousal from sleep (not meeting criteria for apnea or hypopnea)). Because central sleep-disordered breathing events are very rare in the current study, we did not divide sleep-disordered breathing events into central or obstructive events. The total number of apneas, hypopneas, and RERAs
divided by the hours of sleep was expressed as the RDI. Similarly, the AHI, the apnea index (AI), the hypopnea index (HI), and the RERA index were calculated. The mean nighttime SpO2 and minimum SpO2 (lowest values recorded during sleep) were recorded. The total number of oxygen desaturation (≥3 %) events divided by the hours of sleep was expressed as the oxygen desaturation index (ODI). In addition, all of these indexes were calculated separately in NREM sleep and REM sleep [23]. At the first night PSG assessment, subjects with moderateto-severe PLMS (PLMI ≥15) or moderate-to-severe OSA (AHI ≥15) would be excluded from the study. Recruitment process Fifty-five patients with major depressive disorder were initially enrolled in this study. After the clinical interview and the first night of PSG assessment, 31 depressed patients with insomnia were enrolled in this study (Fig. 1). Nine patients discontinued treatment during the trial period, and 22 patients completed this trial. This recruitment process has been described in our previous paper [22]. Statistical analysis The data were presented as the mean±standard deviation ðχ SÞ for continuous variables and as numbers or percentages for categorical variables. Parametric and nonparametric data were compared using an independent t test (two groups). A one-way analysis of variance (ANOVA) and Kruskal-Wallis test were performed for comparing parametric and nonparametric data (≥3 groups). Significant effects in ANOVA were further examined with post hoc tests using the least significant difference method with a Bonferroni correction for multiple comparisons. Mann–Whitney U tests with adjusted P values (significant at P=0.005) were used for multiple pairwise comparisons. A chi-squared test was used to analyze the differences in the categorical variables, and multiple chi-squared tests with adjusted P values (significant at P=0.005) were used for multiple pairwise comparisons. A two-sided 5 % level of significance was considered statistically significant. All the statistical procedures were performed by using Statistical Package for the Social Sciences 17.0 for Windows (SPSS, Inc., Chicago, IL).
Results Demographic and clinical characteristics All the enrolled patients were Chinese with Han ethnicity. They were predominantly young (32.7±9.2 years old) and female (61.3 %) subjects. Of these patients, eight (25.8 %)
Sleep Breath Fig. 1 Flow diagram documenting the recruitment and treatment of depressed patients with insomnia. PSG polysomnogram, DSM-IV Diagnostic and Statistical Manual of Mental Disorders fourth edition, HRSD Hamilton Rating Scale for Depression, OSA obstructive sleep apnea, PLMS periodic limb movement during sleep
reported snoring during sleep, and five (16.1 %) were current smokers. Their demographic and clinical characteristics are presented in Table 1. Table 1 (n=31)
Demographic and clinical characteristics of depressed patients Mean±standard derivation (range) or number
Demographic characteristics Age (years) Gender (male/female) BMI (kg/m2) Snoring (yes/no) Smoking history (nonsmokers, ex-smokers, current smokers) Marriage (married/single/divorced or widowed) Occupation (full-time/part-time/no job or retired) Education (university or above/middle school/primary school or below) Resident (city/town/country) Clinical characteristics Age onset (years) Total duration of illness (years) Single type/recurrent type Number of illness episodes Length of current illness (weeks) BMI body mass index
32.7±9.2 (18–57) 12/19 23.2±6.2 (19.4–25.3) 8/23 19/7/5 17/9/5 16/7/8
Clinical assessment The mean daily doses for sertraline were 126.9±25.4 mg (100–150) on the 14th day, 144.0±30.0 mg (100–200) on the 28th day, and 134.1±28.4 mg (100–200) on the 56th day. In addition, there were only limited side effects (TESS) during the 8-week trial. The HRSD scores started to improve from the 14th day of treatment. From the 14th day on, the proportion of remitted patients and patients without suicidal ideation increased steadily. At the treatment endpoint, 10 (45.5 %) of the patients were classified as remitted and 18 (81.8 %) patients reported no suicidal ideation (item 3 “suicide” in HRSD ≥2). The HRSD-sleep disturbance score became significantly lower after the 28th day. The scores of the PSQI and ESS decreased gradually during this trial, and both questionnaires on the 14th, 28th, and 56th days were significant lower than baseline. The systolic and diastolic BP (mmHg) of these patients was stable throughout the trial (Table 2).
11/16/4 13/10/8 23.9±8.0 (15–33) 9.7±10.4 (0–27) 8/23 2.7±1.9 (1–7) 6.6±5.0 (2–12)
Polysomnographic assessment and sleep-disordered breathing events Table 3 shows selected polysomnographic measures and sleep-disordered breathing events. There were no significant differences in the TRT during the trial. From the 14th day on, the TST and SE became longer and higher than the baseline or 1st day, respectively. From the 14th day on, the SL and WASO decreased significantly and the SL reached a normal range
Sleep Breath Table 2
Clinical measures across the sertraline treatment in depressed patients Baseline (n=31) 1st day (n=31) 14th day (n=26) 28th day (n=25) 56th day (n=22) Statistics
Dosage (mg/day) HRSD HRSD-sleep disturbance factor Remitteda (yes/no) Suicidal ideationb (yes/no) TESS-S TESS-T PSQI ESS Systolic BP (mm Hg) Diastolic BP (mm Hg)
22.4±5.3 a 4.1±3.3 a 0/31 a 18/13 a
13.5±6.2 a 7.2±4.5 a 116.6±23.9 74.3±10.6
50.0 a 23.1±5.3 a 4.0±3.6 a 0/31 a 17/14 a 0.8±1.5 0.6±1.6
120.3±24.1 77.8±11.7
126.9±25.4 b 14.5±4.1 b 3.5±3.1 ab 5/21 ab 11/15 ab 0.7±0.7 0.6±1.0 7.9±4.7 b 5.3±3.9 b 123.4±25.6 75.3±9.9
144.0±30.0 b 9.7±2.6 bc 2.7±1.4 b 9/16 b 8/17 ab 0.5±0.6 0.4±0.5 6.3±3.4 b 3.8±4.1 b 118.7±20.3 73.1±12.6
134.1±28.4b 6.9±1.9 c 2.5±1.5 b 10/12 b 4/18 b 0.5±0.6 0.4±0.4 6.0±3.5 b 4.0±3.5 b 122.3±21.0 76.1±10.4
F=103.90, P
ORIGINAL ARTICLE
Effect of sertraline on breathing in depressed patients without moderate-to-severe sleep-related breathing disorders Bin Zhang & Yanli Hao & Fujun Jia & Xueli Li & Yi Tang & Huirong Zheng & Wuhan Liu
Received: 9 January 2015 / Accepted: 25 February 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Background Previous studies have reported that selective serotonin reuptake inhibitors (SSRIs) might improve sleeprelated breathing disorders (SRBDs). However, the effects of SSRIs on breathing are not evaluated in subjects without moderate-to-severe SRBDs. Further, many symptoms of depression and SRBDs overlap, and so, it is interesting whether there are interactions between breathing and psychopathologic symptoms during SSRI treatment for depression. Methods Data were taken from an open-label 8-week trial of sertraline in depressed patients with insomnia (n=31). The depressed patients were administered 50 mg sertraline at 8 AM on the first day, and the dosage was subsequently titrated up to a maximum of 200 mg/day during the 8-week trial. All the patients were tested by repeated polysomnography (PSG) (baseline, 1st day, 14th day, 28th day, and 56th day). Sleep-disordered breathing events were categorized as apneas, hypopneas, and respiratory event-related arousals (RERAs). Results The clinical responses and PSG characteristics improved continuously during the 8-week trial. From the 14th day on, the RERA index during all-night and nonrapid eye movement (NREM) sleep became stable and
Clinical trial registry An 8-week, open-label study to evaluate the effect of sertraline on the polysomnogram of depressed patients with insomnia (http://clinicaltrials.gov/ct2/show/NCT01032434). Registry identifier: NCT01032434 Bin Zhang and Yanli Hao contributed equally to this work. B. Zhang (*) : F. Jia : X. Li : Y. Tang : H. Zheng : W. Liu Guangdong General Hospital, Guangdong Academy of Medical Sciences, Guangdong Mental Health Center, Guangzhou 510120, China e-mail: [email protected] Y. Hao Department of Human Anatomy, Guang Zhou Medical University, Guangzhou 510182, China
significantly higher than baseline and the first day (RERA index 7.3±2.2 at baseline, 7.3±2.5 on the 1st day, 4.4± 1.9 on the 14th day, 3.9±1.3 on the 28th day, 4.2±2.0 on the 56th day, F=5.71, P=0.02; NREM-RERA index 6.2± 2.0 at baseline, 6.3±2.3 on the 1st day, 3.2±1.5 on the 14th day, 3.5±0.9 on the 28th day, 3.2±1.7 on the 56th day, F=4.92, P=0.03). Additionally, the NREM-apnea index showed a similar pattern to that of the RERA index and reached a significant difference between baseline (1.0 ±0.5) and the 14th day (0.5±0.4) (KW=4.28, P=0.047). Compared to the no-improvement group, the improvement group with a decreasing score rate of the respiratory disturbance index (RDI) greater than or equal to −50 % had a more positive decreasing score rate of slow wave sleep (SWS) (439.0±78.2 vs 373.2±77.9 %, T=3.46, P= 0.04) and a more negative decreasing score rate on the arousal index (−43.7 ± 16.7 vs −26.6 ± 9.7 %, T = 9.16, P =0.01), Pittsburgh Sleep Quality Index (PSQI) scores (−65.1 ± 33.7 vs −49.6 ± 21.4 %, T = 4.74, P = 0.05), and Epworth Sleepiness Scale (ESS) scores (−55.7±21.3 vs −36.4±17.5 %, T=6.44, P=0.02). Discussion This research indicates that SRBDs could be improved to some extent by sertraline treatment, which might be more common in patients with relatively more severe sleepdisordered breathing (e.g., RDI≥10 in the current study). Although the sertraline-induced SRBD improvement seems not to have a significant clinical effect, the SRBD improvement group with decreasing score rate of RDI greater than or equal to −50 % has better subjective and objective sleep aspects than the no-improvement group. Thus, the fact that the SRBDs’ improvement was related to SSRIs might have a potential clinical benefit in the antidepressant treatment.
Keywords Sertraline . Sleep-related breathing disorders (SRBDs) . Depression . Eight-week trial
Sleep Breath
Introduction Sleep-related breathing disorders (SRBDs) are highly prevalent and underdiagnosed diseases associated with substantial morbidity and increased mortality [1]. The abrupt state-related (wake to sleep) decrease in upper airway muscle activity is a major factor in the development of upper airway obstruction, and so, the positive airway pressure (PAP) is currently the most efficacious treatment for SRBDs [1]. However, PAP cannot completely alleviate sleep-disordered breathing events in all patients and cannot be tolerated by some patients [2]. In this situation, medications that target the neurochemical events that underlie the state-dependent obstruction are hypothesized to be the putative therapies for SRBDs [3]. Some animal experiments have shown that serotonin could alleviate SRBDs by stimulating hypoglossal motor nuclei to maintain upper airway potency [4, 5], stimulating respiration, and/or suppressing rapid eye movement (REM) sleep [6, 7]. A recent study in the Chinese Han population reported that the serotonin transporter (5-HTT) gene might be involved in susceptibility to obstructive sleep apnea (OSA), which is the most common subtype of SRBDs [8]. Most antidepressants have some serotoninergic property, and the selective serotonergic reuptake inhibitors (SSRIs) are purer serotoninergic agents. So, SSRIs might provide more insight into the effect of serotonin on the respiratory system. To date, only two SSRIs (fluoxetine and paroxetine) have been evaluated in humans [9–12]. One 4-week study reported that fluoxetine significantly decreased the proportion of REM sleep time and decreased the number of apneas or hypopneas in non-REM (NREM) sleep. It is noted that some patients did not improve their breathing in this trail, which might suggest individual variability [10]. Another prospective study reported that combined treatment with fluoxetine and ondansetron was well tolerated and yielded a potentially therapeutic response in some subjects with SRBDs [12]. On the other hand, the acute administration of paroxetine could augment peak inspiratory genioglossus activity during NREM sleep, but this effect was insufficient to decrease the frequency of obstructive apnea [11]. Another double-blind, placebo-controlled crossover study reported that paroxetine improved breathing during NREM sleep in OSA patients, which was poorly related to effects on sleep architecture or daytime symptoms [9]. These few studies cannot elucidate the effect of SSRIs on SRBDs. Further, there are some flaws of these studies. They compared only sleep-disordered breathing events before and after pharmacotherapy, so the night-to-night variability of respiratory events might be underappreciated. In addition, they only evaluated subjects with moderate-to-severe SRBDs. However, the sleep heart health study has shown that cardiovascular risk is as substantial for a respiratory disturbance index (RDI) of 5 as it for an RDI of 20 or 40 [13]. As a result, it is crucial to evaluate the effects of SSRIs on breathing in
subjects without moderate-to-severe SRBDs. On the other hand, it is noted in both clinical and community samples that there is a higher prevalence of depression in people with SRBDs and many of the symptoms of depression and SRBDs overlap, such as insomnia, sleepiness, and fatigue [14]. Thus, it is interesting whether there are interactions between breathing and psychopathologic symptoms during SSRI treatment. The main purpose of this study is to characterize the effect of sertraline on SRBDs and psychopathologic symptoms in depressed patients in an 8-week clinical trial with repeated polysomnography (PSG) assessment.
Methods Patients and study design The protocol of this study was approved by the Independent Ethics Committee (IEC) of Guangdong Provincial Mental Health Center. Written informed consent forms were signed prior to participation. All the patients were enrolled from the inpatient population of Guangdong Provincial Mental Health Center. If a patient was diagnosed with a single or recurrent type of major depressive disorder according to the Diagnostic and Statistical Manual of Mental Disorders Fourth Edition (DSM-IV) upon admission, the patient’s diagnosis would be ascertained by one of the authors (BZ, psychiatrist) using the second version of the Structured Clinical Interview for DSM-IVAxis I Disorders (SCID-II) [15]. None of the patients included in the study fulfilled any other current or lifetime diagnostic criteria of DSM-IV Axis I disorders. The patients were 18- to 65-yearold males and females with a Hamilton Rating Scale for Depression (HRSD) score ≥18 and a sleep disturbance factor score in the HRSD ≥3 [16], which reflects a moderate-tohigh level of illness severity (depression and insomnia). Possible concurrent medical disorders were ruled out by a thorough medical examination and laboratory tests (electroencephalography (EEG), electrocardiograph [ECG], computed tomography [CT], blood analysis, and urinary analysis). Patients were excluded if they had experienced serious adverse events while taking sertraline, currently had significant suicidal or homicidal tendencies (medical history or item 3 “suicide” in HRSD ≥4), were currently pregnant or breastfeeding, were currently shift workers, currently had a significant sleep disorder (e.g., OSA, periodic limb movement syndrome (PLMS), and restless leg syndrome), or had a serious medical condition in the previous 3 months. After a 7-day screen and/or washout phase and the baseline PSG assessment, the patients received sertraline for 8 weeks. At baseline and during the four visits (1st day, 14th day, 28th day, and 56th day), the patients were assessed by HRSD (clinical improvement), the Treatment Emergent Symptom Scale
Sleep Breath
(TESS-Severity [TESS-S] and TESS-Treatment [TESS-T]: side effects) [17], the Epworth Sleepiness Scale (ESS: sleepiness) [18], and the Pittsburgh Sleep Quality Index (PSQI: sleep quality) [19]. All of these scales have validated Chinese versions [20]. Remitted patient was defined as HRSD ≤7 at the end of the treatment period, and Suicidal ideation was defined as item 3 “suicide” in HRSD ≥2. Because SRBDs might lead to hypertension [21], systolic and diastolic blood pressures (BP, mmHg) were recorded at 8 AM at every visit. Fifty milligrams of sertraline was administered at 8 AM on the first day. It was then titrated according to clinical efficacy and side effects, with a maximum dosage of 200 mg/day. Concomitant use of central nervous system medications during the trial, especially benzodiazepines and sedatives, was prohibited. Polysomnography and sleep-disordered breathing events The related procedure for nocturnal PSG assessments and multiple sleep latency test (MSLT) has been described in our previous paper [22]. All of the sleep variables were derived from the visual scoring of recordings using standard criteria and were divided into two groups: sleep continuity indices and sleep architecture indices. Sleep continuity indices included the total recording time (TRT, “lights out” to “lights on” in minutes), total sleep time (TST), sleep efficiency (SE, the TST divided by the TRT), sleep latency (SL, “lights out” to the first epoch of any sleep in minutes), REM latency (sleep onset to the first epoch of REM stage in minutes), wake after sleep onset (WASO, stage W during TRT, minus SL, in minutes), and arousal index (the number of arousals divided by TST). The sleep architecture indices included the percentages a patient was in each stage (the time in stage 1, stage 2, stage 3, and stage REM divided by the TST) [23]. The five-nap MSLT was performed according to the standard recommendation to determine the SL [24]. All the computerized sleep data were further edited by an experienced PSG technologist, and this technologist was blind to this research. Sleep stages, respiratory events, and periodic limb movements were scored according to American Academy of Sleep Medicine (AASM) criteria at 30-s intervals [23]. As per the AASM criteria, sleep-disordered breathing events were categorized as apneas (a decrease in airflow ≥90 % from baseline for ≥10 s), hypopneas (decrease in airflow ≥30 % for ≥10 s and followed by a desaturation ≥3 % from the pre-event baseline), and respiratory event-related arousals (RERAs) (a sequence of breaths lasting ≥10 s, which was associated with flattening of the nasal pressure waveform leading to an arousal from sleep (not meeting criteria for apnea or hypopnea)). Because central sleep-disordered breathing events are very rare in the current study, we did not divide sleep-disordered breathing events into central or obstructive events. The total number of apneas, hypopneas, and RERAs
divided by the hours of sleep was expressed as the RDI. Similarly, the AHI, the apnea index (AI), the hypopnea index (HI), and the RERA index were calculated. The mean nighttime SpO2 and minimum SpO2 (lowest values recorded during sleep) were recorded. The total number of oxygen desaturation (≥3 %) events divided by the hours of sleep was expressed as the oxygen desaturation index (ODI). In addition, all of these indexes were calculated separately in NREM sleep and REM sleep [23]. At the first night PSG assessment, subjects with moderateto-severe PLMS (PLMI ≥15) or moderate-to-severe OSA (AHI ≥15) would be excluded from the study. Recruitment process Fifty-five patients with major depressive disorder were initially enrolled in this study. After the clinical interview and the first night of PSG assessment, 31 depressed patients with insomnia were enrolled in this study (Fig. 1). Nine patients discontinued treatment during the trial period, and 22 patients completed this trial. This recruitment process has been described in our previous paper [22]. Statistical analysis The data were presented as the mean±standard deviation ðχ SÞ for continuous variables and as numbers or percentages for categorical variables. Parametric and nonparametric data were compared using an independent t test (two groups). A one-way analysis of variance (ANOVA) and Kruskal-Wallis test were performed for comparing parametric and nonparametric data (≥3 groups). Significant effects in ANOVA were further examined with post hoc tests using the least significant difference method with a Bonferroni correction for multiple comparisons. Mann–Whitney U tests with adjusted P values (significant at P=0.005) were used for multiple pairwise comparisons. A chi-squared test was used to analyze the differences in the categorical variables, and multiple chi-squared tests with adjusted P values (significant at P=0.005) were used for multiple pairwise comparisons. A two-sided 5 % level of significance was considered statistically significant. All the statistical procedures were performed by using Statistical Package for the Social Sciences 17.0 for Windows (SPSS, Inc., Chicago, IL).
Results Demographic and clinical characteristics All the enrolled patients were Chinese with Han ethnicity. They were predominantly young (32.7±9.2 years old) and female (61.3 %) subjects. Of these patients, eight (25.8 %)
Sleep Breath Fig. 1 Flow diagram documenting the recruitment and treatment of depressed patients with insomnia. PSG polysomnogram, DSM-IV Diagnostic and Statistical Manual of Mental Disorders fourth edition, HRSD Hamilton Rating Scale for Depression, OSA obstructive sleep apnea, PLMS periodic limb movement during sleep
reported snoring during sleep, and five (16.1 %) were current smokers. Their demographic and clinical characteristics are presented in Table 1. Table 1 (n=31)
Demographic and clinical characteristics of depressed patients Mean±standard derivation (range) or number
Demographic characteristics Age (years) Gender (male/female) BMI (kg/m2) Snoring (yes/no) Smoking history (nonsmokers, ex-smokers, current smokers) Marriage (married/single/divorced or widowed) Occupation (full-time/part-time/no job or retired) Education (university or above/middle school/primary school or below) Resident (city/town/country) Clinical characteristics Age onset (years) Total duration of illness (years) Single type/recurrent type Number of illness episodes Length of current illness (weeks) BMI body mass index
32.7±9.2 (18–57) 12/19 23.2±6.2 (19.4–25.3) 8/23 19/7/5 17/9/5 16/7/8
Clinical assessment The mean daily doses for sertraline were 126.9±25.4 mg (100–150) on the 14th day, 144.0±30.0 mg (100–200) on the 28th day, and 134.1±28.4 mg (100–200) on the 56th day. In addition, there were only limited side effects (TESS) during the 8-week trial. The HRSD scores started to improve from the 14th day of treatment. From the 14th day on, the proportion of remitted patients and patients without suicidal ideation increased steadily. At the treatment endpoint, 10 (45.5 %) of the patients were classified as remitted and 18 (81.8 %) patients reported no suicidal ideation (item 3 “suicide” in HRSD ≥2). The HRSD-sleep disturbance score became significantly lower after the 28th day. The scores of the PSQI and ESS decreased gradually during this trial, and both questionnaires on the 14th, 28th, and 56th days were significant lower than baseline. The systolic and diastolic BP (mmHg) of these patients was stable throughout the trial (Table 2).
11/16/4 13/10/8 23.9±8.0 (15–33) 9.7±10.4 (0–27) 8/23 2.7±1.9 (1–7) 6.6±5.0 (2–12)
Polysomnographic assessment and sleep-disordered breathing events Table 3 shows selected polysomnographic measures and sleep-disordered breathing events. There were no significant differences in the TRT during the trial. From the 14th day on, the TST and SE became longer and higher than the baseline or 1st day, respectively. From the 14th day on, the SL and WASO decreased significantly and the SL reached a normal range
Sleep Breath Table 2
Clinical measures across the sertraline treatment in depressed patients Baseline (n=31) 1st day (n=31) 14th day (n=26) 28th day (n=25) 56th day (n=22) Statistics
Dosage (mg/day) HRSD HRSD-sleep disturbance factor Remitteda (yes/no) Suicidal ideationb (yes/no) TESS-S TESS-T PSQI ESS Systolic BP (mm Hg) Diastolic BP (mm Hg)
22.4±5.3 a 4.1±3.3 a 0/31 a 18/13 a
13.5±6.2 a 7.2±4.5 a 116.6±23.9 74.3±10.6
50.0 a 23.1±5.3 a 4.0±3.6 a 0/31 a 17/14 a 0.8±1.5 0.6±1.6
120.3±24.1 77.8±11.7
126.9±25.4 b 14.5±4.1 b 3.5±3.1 ab 5/21 ab 11/15 ab 0.7±0.7 0.6±1.0 7.9±4.7 b 5.3±3.9 b 123.4±25.6 75.3±9.9
144.0±30.0 b 9.7±2.6 bc 2.7±1.4 b 9/16 b 8/17 ab 0.5±0.6 0.4±0.5 6.3±3.4 b 3.8±4.1 b 118.7±20.3 73.1±12.6
134.1±28.4b 6.9±1.9 c 2.5±1.5 b 10/12 b 4/18 b 0.5±0.6 0.4±0.4 6.0±3.5 b 4.0±3.5 b 122.3±21.0 76.1±10.4
F=103.90, P
