Effect of Electroacupuncture on Insomnia in Patients With Depression

Key Points Question What is the efficacy of electroacupuncture in treating insomnia among patients with depression? Findings In this randomized clinical trial including 270 patients with insomnia and depression who underwent 8 weeks of electroacupuncture with a 24-week observational follow-up, electroacupuncture with standard care significantly improved patients’ quality of sleep compared with sham acupuncture with standard care and standard care alone. Meaning This study found that electroacupuncture with standard care significantly alleviated insomnia among patients with depression.


23
Depression and its related sleep disorders are becoming serious public health problems affecting 24 people worldwide. The global point prevalence of MDD is 4.7% 1 , and the estimation of a 12-month 25 cumulative incidence of depression in China is 5.23% 2 , causing an urgent need to improve depressive 26 patients' health. Sleeping disorders including insomnia, hypersomnia and pavor nocturnus occur 27 frequently in patients with depression 3 . Insomnia may occur in 60-80% of patients with major 28 depressive disorders 4 ; it is one of the most frequent residual symptoms of depression 5 , and may persist 29 even after depressive mood symptoms have been relieved 6 . 30 Insomnia is characterized by persistent dissatisfaction with sleep quantity or quality for at least 4 31 weeks, with specific complaints of difficulty falling asleep, frequent nighttime awakenings, and/or 32 awakening earlier in the morning than desired 7 . Insomnia may be triggered by different factors 33 including psychiatric disorders, organic diseases and the intake of drugs or alcohol 8   Although selective serotonin reuptake inhibitors (SSRIs) and barbiturates have considerably improved 41 the efficacy and prognosis in the treatment of comorbid depression with insomnia, their side effects 42 such as nausea, vomiting, tolerance, addiction, excessive sedation and neurological toxicity cannot be 43 ignored [13][14][15] . What makes the pharmacotherapy more difficult is that some antidepressant drugs may 44 worsen insomnia or cause daytime sleepiness 16 , and high hypnotic dosages for insomnia is closely 45 associated with worsened depressive outcomes 17 . In these cases, a drug-free alternative intervention is 46 urgently needed as an effective and safe therapeutic approach for treating insomnia and depression. 47 Our previous study about acupuncture for primary insomnia demonstrated that acupuncture is an 48 effective treatment to improve patients' sleep efficacy, prolong total sleep time and relieve patients' 49 depressive mood 18 . The preliminary result of our pilot study 19 about the effect of electroacupuncture 50 (EA) for depression related insomnia showed that the Pittsburgh Sleep Quality Index (PSQI) score in 51 depression patients with electroacupuncture treatment obviously decreased (from 16.47±1.89 to 52 9.83±3.11), and there was significant difference between EA and sham EA (p<0.01). Meta-analysis 53 also suggested that acupuncture combined with SSRIs is an effective and well-tolerated therapy for 54 depression and adverse effects of antidepressants 20 . However, other studies showed that acupuncture is 55 not significantly effective in relieving residual insomnia associated with depression 21 22 . As a result, 56 randomized clinical trials in high quality are needed to evaluate the clinical effects and long-term 57 effectiveness of acupuncture in the treatment of depression related insomnia. 58 59 We planned this patient-blinded, multi-center, randomized and controlled trial with a sufficient 60 observation period in three healthcare centers in Shanghai, China. We aim to observe the effects of EA 61 treatment on sleep status, and eliminate the possible placebo effect by setting reasonable sham methods. 62

Study aims
The results will help to demonstrate if EA is an effective and safe therapy for improving sleep quality 63 in patients with depression. 64

Design 67
This is a multi-center, patient-assessor-blinded, randomized and controlled trial, aimed at evaluating 68 the efficacy and safety of electroacupuncture for insomnia in depression patients and comparing the 69 effects between electroacupuncture plus standard care, sham acupuncture plus standard care and simple 70 standard care. 71 The trial will be performed in three healthcare centers in Shanghai: the acupuncture department in 72 Shanghai Municipal Hospital of Traditional Chinese Medicine, the acupuncture department in 73 Changhai Hospital of Shanghai and the therapeutic department in Shanghai Mental Health Center. We 74 will recruit 270 patients who meet the inclusion criteria and randomly assign them to one of 3 groups, 75 receiving electroacupuncture, sham acupuncture and/or standard medical care. After a week baseline, 76 participants will enter an 8-month observation period in this trial. All treatments will be given 3 times a 77 week (every other day) for 8 weeks. Participants will be assessed at the following time points: the 78 baseline (1 week before treatment), the middle of the treatment (4 weeks after treatment starts), the end 79 of the treatment (8 weeks after treatment starts) and follow-up (1 month, 3 months and 6 months after 80 treatment finishes). All participants will complete the assessments by the PSQI, Actigraphy, HAMD, 81 SAS and TESS (detailed trial process seen in Table 1). We will follow the Standards for Reporting 82 Interventions in Clinical Trials of Acupuncture (STRICTA) 23 throughout the trial. 83 Table 1 Trial process chart 84 85

Randomization and blinding 86
An online random allocation system will be designed by the central randomization system with a 1:1:1 87 If the participant meets the inclusion criteria and agrees to join in the trial, a researcher who is not 96 involved in the intervention in each healthcare center will login in to the central randomization system 97 with his own username and password, enter the participant's personal information, and then get the 98 randomized number and the group assignment. The patients' personal information will be protected and 99 keep confidential to the acupuncturists and the assessors before, during and after the trial. 100 We will conduct a patient-assessor-blinded trial where participants are not aware of their group 101 assignments and acupuncturists will not be involved in the outcome assessment or data analysis. 102 Participants will be informed that they have an equal chance of allocation to the three groups. 103 Participants who are assigned to the electroacupuncture (EA) or sham electroacupuncture (SA) will be 104 treated in a closed unit to avoid communication. Furthermore, they will be asked to wear eye masks 105 before and during the trial. Since there are inserted needles around participants' wrist joints, they will 106 not be able to move their hands easily and cannot take off the eye masks. With these methods, 107 participants will not be aware of the difference between EA and SA. To test the success of blinding, all 108 participants in three centers will be asked by their acupuncturists whether they received EA or SA 109 treatment at the end of treatment. Except the acupuncturists, other researchers including the statisticians, 110 outcome assessors and data analysts are all blinded to the group assignments. All researchers will 111 receive training on the specifications of this research method before the trial and strictly adhere to the 112 task separation principle. 113 The flowchart of this trial is shown in Figure 1.

Patients 116
The study will include a total of 270 depression patients with insomnia. To ensure the precision of the 117 results, we developed the following eligibility criteria. 118

Inclusion criteria 119
Participants meeting the following criteria will be included: 3. Participants whose HAMD score is 20-35 (mild to moderate depression); 124 4. Participants who have taken the same antidepressants for more than 4 weeks or have not taken 125 antidepressants; 126 5. Participants who complained about insomnia during first screening; 127 6. Participants whose PSQI score is more than 7; 128 7. Participants who have not received acupuncture treatment for at least one year; 129 8. Participants who voluntarily agree with the investigation and sign a written informed consent 130 form for the clinical trial. 131

Exclusion criteria 132
Participants who report any of the following conditions will be excluded: 133

Recruitment 143
The participants will be recruited through hospital-based advertisements from outpatient clinics and 144 from official websites of all three healthcare centers. If depression patients have interest in participating 145 in the trial, they can take the phone screening first and then will be asked for face-to-face screening in 146 any of the three healthcare centers where they need to fill in some forms with guidance from 147 psychologists or doctors with professional training. Participants then will be asked to wear a wrist 148 actigraphy to monitor their sleep quality for 3 days. Once the participants meet the inclusion criteria, 149 they will be asked to sign the written informed consent form before intervention begins. 150

Sample size calculation 151
The sample calculation is based on changes in the primary outcome of this trial, the Pittsburgh Sleep 152 Quality Index (PSQI) score. In our previous trial, we also used PSQI score as the primary outcome to 153 evaluate and compare the effects between acupuncture, superficial acupuncture at sham points and 154 sham acupuncture on treating depression related insomnia 26 . According to the preliminary results, the 155 PSQI score of the acupuncture group at the end of the 8 weeks' intervention was 9.83±3.11 and that of 156 the sham acupuncture group was 13.93±3.22. We assumed 0.2 of the PSQI difference is the superior 157 effect. 158 H0: A-B<= Δ but H1: A-B> Δ 159 We used the following formula to calculate the sample size in this trial: 160 where is the difference between group, Δ is the assumed superior effect 161 threshold and N is the estimated sample size of each group. σ is the ［(S 1 2 +S 2 2 )／2］ 0.5

162
According to the previous study 27 , the minimal clinically important difference (MCID) of PSQI is 163 about 1.14-1.75. Since there will be a comparison between the Treatment group and the Control A 164 group as well as a comparison between the Treatment group and the Control B group, a sample size of 165 27 in each group will have a power of 90% to detect the superior effect of 1.5 of PSQI at an α-value of 166 0.025 and a β-value of 0.1. Assuming a 10% dropout rate, a sample size of 30 for each group is needed. 167 For a better power and quality control among centers, we decided the recruiting sample size to 30 for 168 each group in each healthcare center. As a result, the total number of participants needed to be 169 randomized is 270. 170

Intervention 171
Participants in Treatment group and Control A group will receive EA or SA treatment. Participants in 172 these two groups will receive 24 sessions of different treatments, 3 times a week for 8 consecutive 173 weeks. EA or SA treatment will be performed after skin cleansing, with patients wearing eye masks and 174 lying supine. Each treatment will last for 30 minutes. The temperature of the treatment room cannot be 175 lower than 25°C. administration of antidepressants, sedatives, hypnotics or anxiolytics during the trial. They must record 178 the dose, especially when they reduce the amount; and dose escalation will not be allowed unless the 179 patient has consulted the psychiatrist. The patients will not be withdrawn the trial by changing the dose 180 of the drug. 181

Treatment group 182
Participants in the Treatment group will receive electroacupuncture (EA) treatment. The acupuncture 183 method of each acupoint is shown in Table 2. The regular acupuncture method will be applied at Baihui 184 (GV20), Shenting (GV24), Yintang (GV29), bilateral Anmian (EX-HN22), Shenmen (HT7), Neiguan 185 (PC6) and SanYinjiao (SP6). After needle insertion, rotating manipulation or lifting-thrusting 186 manipulation will be applied for "Deqi" sensation. Two electrodes of the electro-stimulator (CMNS6-1, 187 Wuxi Jiajian Medical Device CO., LTD, China) will be connected to the needles at Baihui (GV20) and 188 Yintang (GV29) for 30 minutes, delivering a continuous wave. The frequency will be set to 30 Hz with 189 a current intensity of 0.1 to 1 mA during the treatment, based on the tolerance of each patient. 190

Sham acupuncture group 191
Participants in the sham acupuncture group will receive sham electroacupuncture treatment at the same 192 acupoints as the Treatment group. The acupuncture needles are produced by Asia-med GmbH&Co.KG 193 (seen Figure 2), with the same appearance as those used in sham acupuncture treatment. The 194 Streitberger Placebo needle 28 29 have been successfully used in our previous study 19 30 . When the tip of 195 the blunt needles touches to the skin, the patient will get a pricking sensation but there is no real needle 196 inserted into the skin. The electro-stimulator will be set beside the patients and two electrodes will be 197 connected to the needles at Baihui (GV20) and Yintang (GV29). Acupuncturists will turn on the 198 electro-stimulator, but all indicators will be set to "0". Participants will be informed when removing the 199 needles after 30 minutes. Acupuncturists will use dry cotton balls to press the acupoints so that patients 200 can feel the withdrawal of the 'needles'.

Control group 206
Standard care (also known as treatment-as-usual or routine care) in RCTs is frequently employed as the 207 control condition to establish if the intervention is a significant improvement over existing practice 31 . 208 In this trial, we set the control group as the standard care group to investigate the differences between 209 EA treatment group and the blank control group so that the effects of EA for insomnia and depression 210 will be observed more clearly. All 90 participants in three healthcare centers in the control group will 211 continue taking in their routine antidepressants and/or sedative-hypnotics as before from baseline to 8 212 weeks. During the intervention period, if the patients have changes in their mental state and need help, 213 the psychologists in Shanghai Mental Health Center will give professional advices and guidance. After 214 finishing all the required scales and actigraphy records, they will get 10 sessions of free acupuncture 215 treatment for insomnia.

The Hamilton Rating Scale for Depression (HAMD) is an observer-rating questionnaire with 17 232
items used to assess the symptoms of patients diagnosed as suffering from depressive states 33 . Each 233 item is rated in 3-or 5-point scales. A higher total score indicates a higher depression level. 234 3. The Self-rating Anxiety Scale (SAS) is primarily used as a measure of somatic symptoms associated 235 with anxiety 34 . In using the scale, the participant will be asked to rate each item from 0-3 points 236 according to how it applies to him or her within the past week. The standard score is the sum of the 237 integer part of 1.25 times the raw score of the 20 items. A standard score of more than 50 points means 238 the subject has anxious symptoms. A higher score indicates a more serious case of anxiety. 239 4. The dose dairy is a notebook where participants will be required to record their daily dose of 240 antidepressants or sedative-hypnotics from baseline to 6 months follow-up, as well as the dosage time. 241

Adverse events 242
Any adverse events (described as unfavorable or unintended signs, symptoms or diseases occurring 243 during the trial) related to the administration of antidepressant and sedative-hypnotics must be reported 244 by patients and practitioners. These adverse events will be -recorded in the Treatment Emergent 245 Symptom Scale (TESS) 35 . 246 For the adverse events related to the acupuncture treatment, the most common ones include bleeding, 247 faint, bruising ecchymoma, serious pain etc. These AE data will be assessed in terms of severity and 248 causality, and the incidence will also be determined. The 3-point grading categories will be applied: 249 grade 1, mild, grade 2, moderate, grade 3, severe or medically significant. The causality categories used 250 will be certain, probable/likely, possible, unlikely, conditional/unclassified and 251 unassessable/unclassifiable. The incidence of AEs was presented as the number of AEs per number of 252 acupuncture sessions (%). 253

Statistical analysis 254
The statistical analyst will be blinded to the participants' personal information and their group 255 assignment during the trial. The primary analysis will be a comparison of the changes of patients' PSQI 256 score among three groups at 8 weeks after inclusion (comparison of the primary endpoint). The 257 secondary analysis will be performed to assess the changes of the SE, TST and SA recorded in the 258 actigraphy, as well as the HAMD scores and SAS scores from baseline to 8 weeks after inclusion. We 259 will also count the number of patients who increase or decrease the drug dose, and then analyze the 260 differences among three groups. All analyses will be performed on the intention-to-treat (ITT) 261 population of participants who have at least one treatment. Missing data will be handled using the 262 multiple imputation method, on the assumption that values at each time point follow a specific 263 distribution calculated by the computer software R V.3.5. We will also perform a complete-case 264 analysis without imputation of missing data, to find out if the results are consistent. Data analyses will 265 be performed with the use of the statistical software SPSS V.20.0. The t-test will be used to compare 266 the measurement data between either two groups from the baseline to 6 months follow-up; the rank 267 sum test will be used for ranked data while the test will be used to analyze categorical data. The 268 significance level that will be used for statistical analysis with 2-tailed testing will be 2.5%. Data values 269 will mainly be presented as Mean±SD. The results of this study will be published in peer-reviewed journals or presented at academic 288 conferences. 289

290
Acupuncture has been used to treat insomnia and some mental disorders since antiquity in China. 291 According to the theory of traditional Chinese medicine, acupuncture provides balance to the body by 292 stimulating specific acupoints, helping the body to achieve a state of relative equilibrium (the harmony 293 of "yin-yang"), thereby restoring the normal sleep-wake cycle. Recent systematic reviews indicate that 294 acupuncture could be an alternative therapy to medication for treating insomnia but needs further 295 studies using large samples and a rigorous study design to confirm its role 36 37 . 296 Previous RCTs always focus on either the acupuncture treatment for insomnia or that for depression, 297 ignoring the relationship between these two diseases. Insomnia has been identified as the most common 298 sleep disorder comorbid to depressive disorders 38 ; so a reasonable acupuncture treatment program 299 should be developed to normalize sleep disturbance and to relieve depressive mood as well. At the time 300 of this writing, there are no similar RCTs about acupuncture for insomnia in depression patients that 301 included a large sample size and were conducted in multiple healthcare centers. Our trial intends to 302 present a strictly designed trial to study the effects of EA on insomnia in depression patients and to 303 overcome some existing limitations, including illogical design, imperfect blinding method and practical 304 difficulties in previous acupuncture clinical researches. With a long follow-up period, we will be able to 305 explore the persistent effects of acupuncture for insomnia and determine for how long the therapeutic 306 effect will last. 307 For patients in the EA group, we decided to use EA at Baihui (GV20) and Yintang (GV29), with the 308 frequency set to 30 Hz during the treatment. According to the TCM theory, GV20 is the convergent 309 point of six yang meridians as well as the foot Jueyin meridian; it is located on the top of the head, 310 governs yang qi of the body and is the key point of calming mind. GV29 promotes the circulation of qi 311 and blood in the head and restores the function of brain. EA at GV20 and GV29 enhances the effect of 312 soothing nerves. In addition, a functional connectivity MRI (fcMRI) study suggested that EA at GV20 313 and GV29 may have effect on mental disorders 39 . Using fcMRI to identify the key cerebral functional 314 region affected by EA at GV20 and GV29 found that the center of the cerebral network changed from 315 the caudate nucleus to the parahippocampal gyrus and hypothalamus. The network centered on the parahippocampal gyrus and hypothalamus primarily functioned in somatic movement, sensation, vision, 317 hearing and language. This finding may indicate a mechanism for treating depression using EA at 318 GV20 and GV29. 319 A frequency-specific neurochemical response in the central nervous system may be related to 320 differential response of the body to low-and high-frequency EA stimulation and different peripheral 321 and central pathways 40 . Previous research found that low frequency EA could be useful in clinical 322 settings to manage pain 41 while high-frequency stimulation has more potent effects on 5-HT activity 42 . 323 Thirty Hz separates the continuous wave of the electro-stimulator from disperse wave to dense wave 324 and we chose 30Hz based on an acupuncture textbook 43 . 325 Considering the complicated mental state of depression patients with insomnia, we will apply 326 standard medication instead of unified antidepressants or sedative-hypnotics in this trial. Participants in 327 all groups will continue taking in their individual routine dosage from baseline to 6 months follow-up. 328 If their conditions obviously change during the study, they will be free to consult our psychologists 329 from Shanghai Mental Health Center to adjust the dose. The use of standard care control groups has 330 been the subject of much debate, with some pointing out that what constitutes standard care is unclear 331 44 45 . For better implementation of the standard care, researchers in our trial will try to carry out proper 332 health education for all patients and supervise them in recording their daily medication dosage. 333 As a multi-center RCT conducted in a first-tier city, our study can provide more representative 334 results about the role and value of acupuncture as a complementary and alternative therapy for 335 insomnia and depressive moods than other single-center RCTs. 336 Considering the high prevalence of insomnia and depression in rural areas in China 46 47 , the correlated 337 heavy economic burden and serious public health problems cannot be underestimated. In future studies, 338 the focus might be on the acupuncture treatment for insomnia in nationwide healthcare centers. 339 340