Efficacy and Safety of Auricular Therapy for Depression

Abstract

Background:

Researchers have found applications for auricular therapy (AT) for treating depression. There is, however, a lack of comprehensive evaluation of AT's effects and safety.

Objective:

The objective of this review was to evaluate the safety and efficacy of AT for treating depression.

Methods:

Ten databases were used to search for randomized controlled and quasi-randomized controlled trials relating to AT and depression. Related available articles up to January 2016 were used for the research. Rev Man 5.2 software was used for data analysis with the effect estimate presented as relative risk and mean difference with a 95% confidence interval.

Results:

Interventions of trials included in this study were either ear buried seeds (EBS) or transcutaneous vagus nerve stimulation (tVNS). Compared to sham, tVNS produced significant differences on the Hamilton Depression (HAM-D) rating scale, Beck's Depression Inventory (BDI), Self-Rate Depression Scale (SDS), and Self-Rate Anxiety Scale (SAS), but no significant difference were seen on the Hamilton Anxiety (HAM-A) scale. Compared to other treatment, tVNS produced significant differences on the HAM-D, Clinical Global Impression Scale (CGI-S) and Clinical Global Impression—Improvement (CGI-I). tVNS also had an advantage with respect to reducing side-effects. When EBS combined with other treatments was compared to other treatments, there were also significant differences on the HAM-D, SDS, SAS, remission, and Patient Assessment of Constipation Quality of Life (PAC-QOL) questionnaire, but there was no significant difference seen on the Nerve Function Defect Score Scale Barthel Index (BI).

Conclusions:

tVNS and EBS forms of AT could alleviate depression symptoms measured by the HAM-D. It is suggested that the designs of AT trials should follow the recommendations of the Standards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA) guidelines. AT clinicians should use the HAM-D to gauge depression, and somatic symptom reduction should be observed separately.

Introduction

Depression is a common mental disorder in modern society. Approximately 400 million people suffer from depression worldwide, with more women than men affected. Depression is characterized by sadness, loss of interest or pleasure, feelings of guilt, poor concentration, or low self-worth.¹ In addition to mental symptoms, there are somatic symptoms, such as disturbed sleep or appetite, feelings of tiredness, pain, digestive irregularities, chest tightness, nausea, and dizziness. Treatment options include pharmacotherapy, psychotherapy, electroconvulsive therapy, and exercise.²

Auricular therapy (AT) is defined as, “a health care modality whereby the external surface of the ear, or auricle, is stimulated to alleviate pathological conditions in other parts of the body.”³ It is said that rudimentary forms of AT probably appeared during the Stone Age. The therapy was first described in France in 1950 by Paul Nogier, MD, who put forward the recognition of the homunculus and anatomical correlation of the “inverted fetus” in the ear.⁴ Dr. Nogier's first great insight was the representation and anatomical correlation of the inverted fetus on the ear. He then noticed that there was a distinct change that was repeatable and measurable, in the amplitude and dimension of the pulse when certain points on the auricle were stimulated; he called this phenomenon the vascular autonomic signal (VAS, also called the auricular cardiac reflex). Practitioners then used the VAS to determine the location of a point precisely and made judgments regarding whether or not there was a pathology in the region of the body represented on the ear and if certain substances were indicated for treatment. Although the mechanisms of transcutaneous electrical nerve stimulation (TENS) have not yet been elucidated, Pilurzi et al., in a recent study, found that TENS at the trigeminal nerve causes neuroplastic changes at the brainstem level.⁵

Transcutaneous vagus nerve stimulation (tVNS) is the stimulation of a special location on the ear where the vagus nerve is distributed. There are various modalities used in AT worldwide, including auricular acupuncture, acupressure, moxibustion, medical-syringe injection, and tVNS.^6,7 As far back as 1990, the World Health Organization confirmed that AT was one of the most popular complementary therapies.⁸

Although AT has been used to treat depression, the effect of AT has not been investigated sufficiently. The safety and effect of AT needs to be evaluated. Hence, this systematic review was conduction to help address this lack.

Materials and Methods

Protocol and Registration

A protocol for this systematic review was previously published in an article entitled “Auricular Therapy for Depression: A Systematic Review and Meta-Analysis.”⁹

Inclusion and Exclusion Criteria

This study included randomized controlled trials (RCTs) and quasi-randomized controlled trials (Q-RCTs) of AT, which involved comparing AT to a control condition (sham, placebo, other treatment, or no treatment). There was no restriction on gender, race, or region. There were no limitations on sample size, instruments, durations of treatment, or selections of auricular points. The interventions had to include stimulation on the external surface of the ear or auricle. Trials were excluded if they related to any acupoint therapies other than AT. This exclusion was established in order to eliminate the influences of different acupoint therapies. Crossover trials and historical controlled trials were also excluded.

Adults with depression (or depression with comorbidity) were diagnosed using one of the following: the Diagnostic and Statistical Manual of Mental Disorders, 5^th Edition¹⁰; Research Diagnostic Criteria (RDC) for a Selected Group of Functional Disorders, 3rd Edition¹¹; The ICD-10 Classification of Mental and Behavioural Disorders: Diagnostic Criteria for Research¹²; or The Chinese Classification of Mental Disorders [CCMD-3], 3rd Edition.¹³

Identification and Selection of Studies

Bibliographic or electronic databases were searched to identify relevant studies for the current review. English databases included the Cochrane Central Register of Controlled Trials (CENTRAL), PubMed, MEDLINE,^® OVID, SCI, and Springer. Chinese databases included the Chinese National Knowledge Infrastructure Database (CNKI), the Chinese Science and Technology Periodical Database (VIP), the Chinese Biomedical Literature Database (CBM), and the Wan Fang Database. All searches ended on January 2016.

The following search terms were used separately or in combination: auricular pressure; auricular-plaster therapy; auricular acupuncture; auriculotherapy; auricular application pressure; auricular electroacupuncture; auricular acupressure; vagus nerve stimulation; auricular treatment; transcutaneous electrical nerve stimulation; clinical trial; randomized; depression; major depressive disorder; depressive disorder; depressive symptoms; and depressive mood. Chinese words/phrases included: er xue liao fa; er xue ya dou; er xue tie fu; er zhen; er; mi zou shen jing ci ji; jing pi dian ci ji; yi yu; sui ji; and lin chuang yan jiu. Language, publication year, and publication status were not limited. All eligible articles obtained from the reference lists were checked for additional studies. In cases of missing information, all unpublished and ongoing trials were also collected for the research. All reference lists were checked for related subject topics, articles, and reviews. The authors of these articles were contacted for any unpublished data.

Data Extraction and Quality Assessment

Five of the current authors conducted the literature search (Z.-X.Z. and Z.-H.B.), study selection (Z.-X.Z. and Z.-H.B.), and data extraction (Q.J. and C.-R.L.), independently. The extracted data included authors and title of studies, years of publication, study sizes, ages and genders of the participants, primary diseases, designs, interventions, duration outcomes, and adverse events for each study. Disagreement was resolved by discussion and consensus reached through a third party (L.R.).

Types of Outcomes

Primary outcomes included results on the Hamilton Depression (HAM-D) scale, Self-Rating Depression Scale (SDS), remission, and adverse events. Meanwhile, Beck's Depression Inventory (BDI), the Nerve Function Defect Score Scale Barthel Index (BI), the Hamilton Anxiety (HAM-A) scale, Self-Rate Anxiety Scale (SAS), the Clinical Global Impression Scale (CGI-S), the Clinical Global Impression—Improvement (CGI-I), and the Patient Assessment of Constipation Quality of Life (PAC-QOL) questionnaire were included as secondary outcomes, but any outcome relating to comorbidities was ignored.

Data Analysis

Rev Man 5.2 software was used for data analysis with the effect estimate presented as relative risk (RR) or mean difference (MD) with a 95% confidence interval (CI). If a sufficient number of randomized trials were identified, subgroup analyses of AT vagus nerve stimulation (VNS) and EBS would be carried out for these interventions. Meta-analysis was performed if the trials had good homogeneity for study design, participants, interventions, control, and outcome measures, which were assessed by examining I² (a quantity that approximately describes the proportion of variation in point estimates caused by heterogeneity rather than being a sampling error). Heterogeneity between studies has been investigated by considering the I² statistic, which quantifies inconsistency across studies to assess the impact of heterogeneity on a meta-analysis. If an I² was larger than 50%, it indicated the possibility of heterogeneity.¹⁴ If there was a possibility of statistical heterogeneity among the trials, both a fixed-effect model and a random-effect model were used. The fixed-effect model was used for meta-analysis in the case of I² being <50%.

Statistical analysis was performed according to the statistical guidelines referenced in the newest version of the Cochrane Handbook for Systematic Reviews of Interventions.¹⁵ In cases when the data were not available, attempts were made to obtain missing data from the original trial authors. If a sufficient number of randomized trials were identified, sensitivity analyses were performed to explore the influence of trial quality on effect estimates. The quality components of methodology included generation adequacy of allocation sequences, concealment of allocations, double-blinding, and use of intention-to-treat (yes or no).

Results

Description of Studies

Twelve randomized trials^16–27 were included in this review. Two trials were reported in English journals,^22,26 and the remaining ten trials were published in Chinese journals.^{16–21,23–25,27} The flowchart shown in Figure 1 depicts the search process and study selection. The six RCTs^{17,20,22,23,26,27} and six Q-RCTs,^{16,17,19,21,24,25} with a total of 759 patients who had depression, were included in this review (63.25 patients per trial). Differences in comorbid diseases and stimulation were ignored. Four subgroups were created in accordance with the differing interventions of EBS, tVNS, combined EBS treatment, and combined tVNS treatment. The controls included sham AT, routine treatment and no treatment.

FIG. 1.

Study flow diagram. RCT, randomized controlled trial; Q-RCT, quasi-randomized controlled trial.

There were a number of different outcomes. The primary outcome, HAM–D, was reported in nine trials.^16–23,26 Remission^{16–18,23,27} and SDS ^23–27 were reported in eight trials. The secondary outcomes, BDI²² and BI,²¹ were reported in two trials, while HAM-A²⁶ was reported in one trial. One trial²⁵ observed changes in PAC-QOL at the end of treatment. CGI-S and CGI-I were reported in one trial.¹⁶ The characteristics of all included studies are presented in Table 1.

Table 1.

Characteristics of All Included Studies

		Participants
								Intervention or control conditions
Study ID (1st author, year, ref.)	Study type	Sample size (T/C, male/female)	Age (T & C)	Random method	Blinding method	Disease/comorbidity disease	Design	AT	Control	Duration	Outcomes	Follow-up	Comorbidity disease outcome	Adverse events (T & C)
Xu, Wenwei 2009¹⁶	Q-RCT	36/36, Unclear	Unclear	Medical order	Unclear	Depression/	Tvns + SSRIs vs. SSRIs	tVNS	Western medicine	12 weeks	Remission (HAM-D) CGI-I CGI-S	Unclear	Unclear	4.9 + 5.1/9.7 + 5.5
Fan, Chun 2011¹⁷	RCT	34/36, 21/49	Unclear	Random number table	Unclear	Depression/	EBS vs. no treatment	EBS	No treatment	8 weeks	Remission (HAM-D)	Unclear	Unclear	Unclear
Zheng, Jie 2012¹⁸	Q-RCT	40/40, 49/31	T: 45–79 (72) C: 49–78 (70)	Unclear	Unclear	Depression/stroke	EBS + SHT vs. SHT	EBS + SHT	SHT	3 weeks	HAM-D remission	Unclear	Unclear	Pain,^b nausea,^c insomnia,^d loose stools,^e tachycardia^e/nausea,^f,g loose stools,^b,e dizziness,^c insomnia,^d & tachycardia^g
Zhang, Shumei 2012¹⁹	Q-RCT	47/47, 52/42	T: (41–75) C: (42–73)	Unclear	Unclear	Depression/stroke	EBS + EC vs. EC	EBS + EC	EC	5 weeks	HAM-D	Unclear	Unclear	Unclear
Zheng, Qiujiao 2012²⁰	RCT	33/30, unclear	Unclear	Pocock & Simon Minimization Method	Unclear	Depression/hypertension	EBS + RTN vs. RTN	(EBS + RTN	RTN	4 weeks	HDRS ( = HAM-D)	6-month	BP	Unclear
Zheng, Lihong 2013²¹	Q-RCT	30/30, unclear	Unclear	Unclear	Unclear	Depression/functional constipation	EBS + RTN vs. RTN	EBS + RTN	RTN	30 days	HAM-D, BI	Unclear	Unclear	Unclear
Hein, Ernst 2013²²	RCT	11/11, 9/13	Unclear	Random number table	Blinded	Depression/	Auricular TENS vs. sham	Auricular TENS	Sham stimulation	2 weeks	HAMD, BDI	Unclear	Unclear	Unclear
Hu, Yufang 2013²³	RCT	32/32, 34/30	T: 60.5 ± 9.8 C: 62.2 ± 8.9	Admission sequence and random number table	Unclear	Depression/Stroke	EBS plus RTN vs. RTN	EBS plus RTN	RTN	4 weeks	SDS, remission (HAM-D)	Unclear	Unclear	Unclear
Tian, Meixin 2014²⁴	Q-RCT	40/40, 46/34	T: 67.07 ± 10.24 C: 67.08 ± 10.12	Admission sequence	Unclear	Depression/Stroke	EBS + RTN vs. RTN	EBS + RTN	RTN	4 weeks	SDS	Unclear	Unclear	Unclear
Zeng, Lirong, 2014²⁵	Q-RCT	30/30, 31/29	T: 42.37 ± 10.2/ C: 42.41 ± 10.31	Medical order	Unclear	Depression/constipation	EBS + CT vs. CT	EBS + CT	Conventional therapy	3 weeks	SDS, SAS, PAC-QOL	Unclear	Unclear	Unclear
Fang, Jiliang 2016²⁶	RCT	18/16, unclear	Unclear	Random number table	Double-blinded	Depression/	tVNS vs. sham tVNS	tVNS	Sham tVNS	4 weeks	HAM-D, HAM-A, SDS, SAS	Unclear	Unclear	Unclear
Lv, Xiao'ai 2015²⁷	RCT	30/30, unclear	Unclear	Medical order & random number table	Unclear	Depression/breast cancer	EBS + Deanxit^a vs. Deanxit	EBS + Deanxit	Deanxit	4 weeks	SDS, remission	1-month	Unclear	Unclear/dizziness,^h constipation,ⁱ & insomnia^e

Deanxit is a flupentixol + melitracen combination.

Adverse events: ^bPer HAM-D, ^cper BDI; ^dwith EBS; ^ewith SSRIs; ^fwith AT; ^gper SDS; ^hwith EC; ⁱwith RTN.

Scales: HAM-D, Hamilton Depression rating scale; CGI-I, Clinical Global Impression—Improvement scale; CGI-S, Clinical Global Impression Scale; HDRS, Hamilton Depression Rating Scale ( = HAM-D); BI, Barthel Index (Nerve Function Defect Score Scale); BDI, Beck's Depression Inventory; SDS, Self-rate Depression Scale; SAS, Self-Rate Anxiety Scale; PAC-QOL, Patient Assessment of Constipation Quality of Life questionnaire; HAM-A, Hamilton Anxiety scale.

ID, identification; T, treatment; C, control; AT, auricular therapy, Q-RCT; quasirandomized controlled trial; RCT, randomized controlled trial; tVNS; transcutaneous vagus nerve stimulation; SSRI, selective serotonin inhibitors (5-hydroxytryptamine reuptake inhibitor); EBS, ear buried seeds; SHT, sertraline hydrochloride tablets; EC, emotional care; TENS, transcutaneous electrical nerve stimulation; RTN, routine treatment and nursing; CT, conventional therapy.

Risk of Bias in Included Studies

Overall, twelve trials contributed data to the meta-analysis. The majority of included trials were of poor quality. Six trials^{17,20,22,23,26,27} used a random number table or the Pocock and Simon Minimization Method to allocate treatment. Two trials^24,25 used a Q-RCT format, which allocated patients alternately according to the admission sequence or medical order. Four trials^16,18,19,21 did not describe the details of sequence generation. Neither adequate concealment nor a blinding method was reported in these trials.^{17–21,23–25,27} Risk of bias is summarized in Figure 2.

FIG. 2.

Risk of bias graph: Review authors' judgments about each-risk-of-bias item presented as percentages across all included studies.

Effects of Interventions

Results of the meta-analysis, with respect to the intervention effects, are shown in Figures 3 –18.

FIG. 3.

Forest plot of comparison for transcutaneous vagus nerve stimulation (tVNS) versus sham treatment, outcome: Hamilton Depression rating scale. tVNS, transcutaneous vagus nerve stimulation; SD, standard deviation; CI, confidence interval.

FIG. 4.

Forest plot of comparison for transcutaneous vagus nerve stimulation (tVNS) versus sham treatment, outcome: Beck's Depression Inventory. tVNS, transcutaneous vagus nerve stimulation; SD, standard deviation; CI, confidence interval.

FIG. 5.

Forest plot of comparison for transcutaneous vagus nerve stimulation (tVNS) versus sham treatment, outcome: Self-rating Depression Scale. SD, standard deviation; CI, confidence interval.

FIG. 6.

Forest plot of comparison for transcutaneous vagus nerve stimulation (tVNS) versus sham treatment, outcome: Hamilton Anxiety scale. SD, standard deviation; CI, confidence interval.

FIG. 7.

Forest plot of comparison for transcutaneous vagus nerve stimulation (tVNS) versus sham treatment, outcome: Self-Rate Anxiety Scale. SD, standard deviation; CI, confidence interval.

FIG. 8.

Forest plot of comparison for transcutaneous vagus nerve stimulation (tVNS) plus other treatment versus other treatment, outcome: Hamilton Depression rating scale. M–H, Mantel-Haenszel; CI, confidence interval.

FIG. 9.

Forest plot of comparison for transcutaneous vagus nerve stimulation (tVNS) plus other treatment versus other treatment, outcome: Clinical Global Impression Scale. M–H, Mantel-Haenszel; CI, confidence interval.

FIG. 10.

Forest plot of comparison for transcutaneous vagus nerve stimulation (tVNS) plus other treatment versus other treatment, outcome: Clinical Global Impression—Improvement. M–H, Mantel-Haenszel; CI, confidence interval.

FIG. 11.

Forest plot of comparison for transcutaneous vagus nerve stimulation (tVNS) plus other treatment versus other treatment, outcome: adverse events. SD, standard deviation; CI, confidence interval.

FIG. 12.

Forest plot of comparison for ear buried seeds versus no treatment, outcome: remission. M–H, Mantel-Haenszel; CI, confidence interval.

FIG. 13.

Forest plot of comparison for ear buried seeds plus other treatment versus other treatment, outcome: Hamilton Depression rating scale. SD, standard deviation; CI, confidence interval.

FIG. 14.

Forest plot of comparison for ear buried seeds plus other treatment versus other treatment, outcome: Self-rate Depression Scale. SD, standard deviation; CI, confidence interval.

FIG. 15.

Forest plot of comparison for ear buried seeds plus other treatment versus other treatment, outcome: Self-Rate Anxiety Scale. SD, standard deviation; CI, confidence interval.

FIG. 16.

Forest plot of comparison for ear buried seeds plus other treatment versus other treatment, outcome: remission. SD, standard deviation; CI, confidence interval. M–H, Mantel-Haenszel; CI, confidence interval.

FIG. 17.

Forest plot of comparison for ear buried seeds plus other treatment versus other treatment, outcome: Nerve Function Defect Score Scale Barthel Index. SD, standard deviation; CI, confidence interval.

FIG. 18.

Forest plot of comparison for ear buried seeds plus other treatment versus other treatment, outcome: and Patient Assessment of Constipation Quality of Life Questionnaire. SD, standard deviation; CI, confidence interval.

Comparison Between tVNS and Sham Treatment

HAM-D change at the end of treatment

Two trials^22,26 measured the outcome of HAM-D at the end of the treatment for 56 patients. There was a significant difference between tVNS and sham treatment (MD: −5.68; 95% CI: −11.17 to −0.20; I² = 66%; random; Fig. 3).

BDI change at the end of treatment

One trial²² measured BDI change for 22 patients. The difference was significant between tVNS and sham treatment (MD: −11.80; 95% CI: −21.85 to −1.75; Fig. 4).

SDS change at the end of treatment

One trial²⁶ applied SDS at the end of treatment for 34 patients. The difference was significant between tVNS and sham treatment (MD: −10.63; 95% CI: −18.05 to −3.21; Fig. 5).

HAM-A change at the end of treatment

One trial²⁶ used HAM-A at the end of treatment for 34 patients. The difference was not significant between tVNS and sham treatment (MD: −1.77; 95% CI: −5.50 to 1.96; Fig. 6).

SAS change at the end of treatment

One trial²⁶ used SAS at the end of treatment for 34 patients. The difference was significant between tVNS and sham treatment (MD: −11.61; 95% CI: −16.96 to −6.26; Fig. 7).

Comparison Between tVNS Plus Other Treatment and Other Treatment

HAM-D change at the end of treatment

One trial¹⁶ measured changes in adverse events for 72 patients. The difference was significant between tVNS plus other treatment and other treatment (MD: 4.38; 95% CI: 1.57 to 12.19; Fig. 8).

CGI-S changes at the end of treatment

One trial¹⁶ measured changes in adverse events for 72 patients. The difference was significant between tVNS plus other treatment and other treatment (MD: 4.55; 95% CI: 1.69 to 12.25; Fig. 9).

CGI-I change at the end of treatment

One trial¹⁶ measured change in adverse events for 72 patients. The difference was significant between tVNS plus other treatment and other treatment (MD: 2.23: 95% CI: 0.79 to 6.26; Fig. 10).

Adverse events change at the end of treatment

One trial¹⁶ measured change in adverse events for 72 patients. The difference was significant between tVNS plus other treatment and other treatment (MD: −4.80, 95% CI, −7.25 to −2.35; Fig. 11).

Comparison Between EBS and No Treatment

Remission changes at the end of treatment

One trial¹⁷ observed remission change at the end of treatment for 70 patients. The difference was significant between EBS and no treatment (Risk Difference: 3.57; 95% CI: −1.89 to 6.74; Fig. 12).

Comparison Between EBS Plus Other Treatment and Other Treatment

HAM-D change at the end of treatment

Four trials^18–20,25 measured HAM-D change at the end of treatment for 297 patients. The difference was significant between EBS plus other treatment and other treatment (MD: −2.95; 95% CI: −4.94 to −0.97; I² = 83%; random; Fig. 13).

SDS change at the end of treatment

Four trials^23–25,27 measured SDS change at the end of treatment for 264 patients. The difference was significant between EBS plus other treatment and other treatment (MD: −8.66; 95% CI: −9.96 to −7.37; I² = 21%; fixed; Fig. 14).

SAS change at the end of treatment

One trial²⁵ measured SAS change at the end of treatment for 60 patients. The difference was significant between EBS plus other treatment and other treatment (MD: −6.20; 95% CI: −11.22 to −1.18; Fig. 15).

Remission change at the end of treatment

Four trials^16,18,23,27 measured remission change at the end of treatment for 180 patients. The difference was significant between EBS plus other treatment and other treatment (MD: 3.87; 95% CI: 2.10 to 7.13; I² = 0%; fixed; Fig. 16).

BI change at the end of treatment

One trial²⁵ measured change of BI for 60 patients. The difference was not significant between EBS plus other treatment and other treatment (MD: 3.34; 95% CI: −1.05 to 7.73; Fig. 17).

PAC-QOL change at the end of treatment

One trial²⁵ measured PAC-QOL change for 60 patients. The difference was significant between EBS plus other treatment and other treatment (MD: −10.71; 95% CI: −13.02 to −8.40; Fig. 18).

Discussion

AT Research

At present, the stimulation methods of AT can be diverse, but the stimulation locations will often be chosen according to the auricular points or the anatomical characteristics of the patient's ear. There were two important and different acupuncture systems for mapping the auricular points; these were (1) the European ear acupuncture system and (2) the Chinese ear acupuncture system.⁶ To detect the difference between the European ear acupuncture system and the Chinese ear acupuncture system, Alimi et al. used functional magnetic resonance imaging (fMRI) and found topographic evidence supporting that the European auricular system was better than that of the Chinese auricular system.²⁸ However, there are many varying diagrams worldwide with little agreement regarding point locations in the ear. The reflex systems do not correlate with modern knowledge of anatomy.

The concept of an inverted fetus map on the external ear was usually the same across the two groups with clinical experience informing the establishment of this map.^6,29 A previous fMRI study showed that tVNS provoked brain activation with patterns sharing features arising from changes brought about from invasive vagus nerve stimulation.³⁰ That study revealed that the central vagal pathway could be stimulated by tVNS. As a new therapy, tVNS is now being used for the treatment of drug-resistant epilepsy,³¹ chronic tinnitus,³² and depressive disorder. Meanwhile, many instruments have been developed and became available to guide acupuncturists given that such devices are needed to detect low skin resistance on the ear. More clinical research and observation are still required for better therapeutic evaluation.³³

Differences Among the Trials

The current review included twelve trials with a total of 759 participants. EBS and tVNS were included in this review. An intervention in one trial used transcutaneous electrical nerve stimulation (TENS). The researchers in that trial considered TENS to be a form of tVNS.²² No trial mentioned or applied the VAS for the accuracy of point placement on the ear. The eleven different evaluation criteria used included the HAM-D, SDS, adverse events occurrence rates, SAS, HAM-A, remission, CGI-I, CGI-S, BDI, BI, and PAC-QOL.

There were three trials^16,22,26 involving tVNS treatment. tVNS produced significant improvement in HAM-D, BDI, SDS and SAS scores compared to sham, but these studies indicated that tVNS produced no significant improvement in HAM-A (Fig. 6). One trial showed that anxiety symptoms of different degrees among the participants coexisted with major depressive disorder.²⁶ Anxiety and depression could coexist in the same individual concurrently or at different times. The severity of a patient's anxiety could be indicated by a HAM-A score and it has been included in the psychologic measures used in many depression studies.³⁴

Compared with other forms of treatment, tVNS also produced significant improvement on HAM-D, CGI-S, and CGI-I scores while also inducing a lower occurrence rate of adverse events (Figs. 3, 5, 7–11). Furthermore, when EBS was combined with other treatments versus other treatments alone, there were significant improvements on scores for HAM-D, SDS, SAS, remission, and PAC-QOL (Figs. 13 –16, 18). Compared to no treatment, EBS was a valid form of treatment for depression in remission, according to the HAM-D reductive ratio (Fig. 12). However, when EBS treatment plus other treatment was compared with the corresponding treatment, no significant difference in BI change was produced (Fig. 17).

The HAM-D has been the standard for the assessment of depression for >50 years, and it is still a reliable instrument. Eight studies that included EBS and tVNS used the HAM-D for depression assessment; the data showed that AT could reduce psychologic symptoms and relieve somatic symptoms. It is suggested that, for AT studies, clinicians should learn more and apply the HAM-D for assessing depression rather than relying on self-reporting by subjects or participants.

Heterogeneities (I² = 61% and I² = 83%) were apparent when tVNS was compared with sham treatment or EBS against other treatment forms, respectively. The source of heterogeneity noted in this review was from a lack of trail-to-trail standards, such as stimulation intensity of treatments or varying primary diseases of the participants. With respect to stimulation intensities, there was fixed intensity (130 μA) and varied intensity (4–6 mA) based on the tolerance of the patients, which could have affected heterogeneity among the tVNS studies. Meanwhile, differences among comorbidities of participants, such as stroke, hypertension, breast cancer and functional constipation, may have affected the heterogeneity of the EBS trials.

Somatic symptom reduction was one of the foci of this review. One trial mentioned that somatic symptoms had been in participants, but the details were lacking.²⁵

Quality of the Evidence

There were six trials of relatively high quality that supported the effect of AT for relieving depression. The random method in six trials was unclear, and two trials described follow-ups while three trials mentioned adverse events. Meanwhile, no trial used an allocation concealment or blinding method, with the exception of two trials.. There were defects in the designs of sham treatment, because clamp pressure could also cause stimulation even with a lack of an electrical current.^22.23 In addition, it is suggested that the VAS could be considered as a guide for auricular therapy according to Nogier's theory. Physicians who provide AT to their patients should know about the uses of the VAS.

Trials of AT for depression should follow the Standards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA) regarding acupuncture rationales and details on interventions, treatment regimens, controls, or comparator interventions, etc. It is also suggested that follow-up should be included in any further research. Effects should be assessed if the variations of AT—such as stimulation method, quantity of stimulus, duration, and stimulation site—can cause any significant differences in depression. Studies are especially needed to confirm whether or not there are differences between tVNS and EBS treatment modalities.

Some research provides preliminary evidence that AT is a relatively safe approach.^30,35 The most frequently reported adverse events were tenderness or pain at insertion, dizziness, local discomfort, minor bleeding, and nausea. No significant effects on heart rate, blood pressure, or peripheral microcirculation can be detected during stimulation procedures, so it appears that patients with depression would tend to derive benefits from AT against the consideration of the safety and satisfaction involved in long treatment procedures. Somatic symptom reduction is one of the foci of this review. Only one trial mentioned this parameter. It is suggested that statistics for, and descriptions of somatic symptoms should be taken using the HAM-D, SDS, BDI, or PAC-QOL in further research.

Conclusions

The tVNS and EBS of AT could reduce psychologic symptoms and relieve somatic symptoms consistently, as measured by the HAM-D. It is suggested that future auriculotherapy trials should be reported according to the STRICTA guidelines. Somatic symptom reduction should be observed separately in accordance with the HAM-D.

Footnotes

Acknowledgments

This work was supported by funding from the National Natural Science Foundation of China (No. 81373718) and the Shenyang Science and Technology Bureau (No. F13-318-1-67). The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the Liaoning University of Traditional Chinese Medicine. The authors appreciate the revision from Timothy Helland, MD on this article.

Author Disclosure Statement

The authors declare that they have no conflicts of interest connected with the research.

References

World Health Organization. Health Topics: Depression. Online document at: www.who.int/topics/depression/en/ Accessed September 25, 2016.

Manning

, Jackson

. Treating depression in primary care: Initial and follow-up treatment strategies. J Clin Psychiatry. 2015; 76(2):e5.

Oleson

. Auriculotherapy Manual: Chinese and Western Systems of Ear Acupuncture, 3th ed. Edinburgh: Churchill Livingstone; 2008:1.

Gori

, Firenzuoli

. Ear acupuncture in European traditional medicine. Evid Based Complement Alternat Med. 2007; 4(suppl1):13–16.

Pilurzi

, Mercante

, Ginatempo

, Follesa

, Tolu

, Deriu

. Transcutaneous trigeminal nerve stimulation induces a long-term depression-like plasticity of the human blink reflex. Exp Brain Res. 2016; 234(2):453–461.

Wang

. Ear acupuncture. In: Micro-acupuncture in practice. St. Louis: Churchill Livingstone, 2008:99–138.

Kember

, Ardell

, Armour

, Zamir

. Vagal nerve stimulation therapy: What is being stimulated?. PLoS One. 2014; 9(12):e114498.

World Health Organization (WHO). WHO Report of the Working Group on Auricular Nomenclature. Lyon, France: WHO; 1990.

Zhang

Z-X

, Bai

Z-H

, Kim

, Li

Ch-Ri

. Auricular Therapy for Depression: A Systematic Review And Meta-Analysis. University of York Centre for Reviews and Dissemination. Online document at: www.crd.york.ac.uk/PROSPERO/display_record.asp?ID=CRD42015022176) Accessed September 25, 2016.

10.

American Psychiatric Association (APA). Diagnostic and Statistical Manual of Mental Disorders, 5th ed. Washington, DC: APA; 2013:155–156.

11.

Spitzer

, Endicott

, Robins

. Research Diagnostic Criteria (RDC) for a Selected Group of Functional Disorders, 3rd ed. New York: Biometric Research; 1977.

12.

World Health Organization (WHO). The ICD-10 Classification of Mental and Behavioural Disorders: Diagnostic Criteria for Research. Geneva: WHO; 1993.

13.

Chinese Psychiatric Society. The Chinese Classification of Mental Disorders [CCMD-3], 3rd ed. Shandong: Shandong Publishing House of Science and Technology, 2001; 91–92.

14.

Higgins

, Thompson

. Quantifying heterogeneity in a meta-analysis. Stat Med, 2002; 21(11):1539–1558.

15.

Higgins

JPT

, Green

, eds. Cochrane Handbook for Systematic Reviews of Interventions. New York: Wiley; 2008.

16.

, Li

, Cai

, Wang

. Efficacy improved on treatment of depression. J Clin Psychiatry. 2009; 19(3):198–200.

17.

Fan

, Liang

, Liu

, Li

. Clinical study of auricular point sticking-pressing in treating subsyndromal symptomatic depression. New Chin Med. 2011; 43(8):110–111.

18.

Zheng

, Wang

, Zhou

, Cao

. The clinical efficacy research of auricular therapy combined with sertraline hydrochloride for the treatment of stroke with depression. Seek Medical. 2012; 10(11):1035–1036.

19.

Zhang

, Chang

, Qu

, Yuan

. Observation on effect of emotional care and ear buried seeds for stroke patient at acute phase complicated with depression. Chin Nurs Res. 2012; 26(6):1503–1504.

20.

Zheng

, Xu

, Gao

, Zhu

. The effect of auricular therapy on primary hypertension complicated with depression. J Zhejiang University Chin Med. 2012; 36(10):1147–1150.

21.

Zheng

, Chen

, Zhong

. The influence of auricular-plaster therapy on brain stroke patients with depression. Chin General Pract Nurs. 2013; 11(9):2324–2325.

22.

Hein

, Nowak

, Kiess

, Biermann

, Bayerlein

, Kornhuber

, Kraus

. Auricular transcutaneous electrical nerve stimulation in depressed patients: A randomized controlled pilot study. J Neural Transm (Vienna). 2013; 120(5):821–827.

23.

. The effect of auricular application pressure on stroke patients with depression. New Chin Med. 2013; 45(11):187–188.

24.

Tian

, Yang

. Effect of auricular point sticking on the depressive emotions in patients with stroke involving meridians. Chin J Nurs Educ. 2014; 11(6):409–411.

25.

Zeng

. The influence of auricular application pressure on the quality of life and depression of functional constipation patients. Chin J Coal Industry Med. 2014; 17(10):1706–1708.

26.

Fang

, Rong

, Hong

, et al. Transcutaneous vagus nerve stimulation modulates default mode network in major depressive disorder. Biol Psychiatry. 2016; 79(4):266–273.

27.

, Wang

, Chen

, Ye

. Clinical observation of depression after breast cancer operation treated with auricular point sticking therapy. Chin Acupunct Moxibustion. 2015; 35(5):447–450.

28.

Alimi

. Geissmann A, Gardeur D, Bahr F. Study in fMRI of the stimulation of the auricular areas of the knee as the French-German and Chinese localizations. Photon J Radiol. 2014; 125:133–141.

29.

Hou

, Hsu

, Lin

, Tang

, Cheng

, Hsieh

. The history, mechanism, and clinical application of auricular therapy in Traditional Chinese Medicine. Evid Based Complement Alternat Med. 2015; 2015:495684.

30.

Kraus

, Hösl

, Kiess

, Schanze

, Kornhuber

, Forster

. BOLD fMRI deactivation of limbic and temporal brain structures and mood enhancing effect by transcutaneous vagus nerve stimulation. J Neural Transm (Vienna). 2007; 114(11):1485–1493.

31.

Bauer

, Baier

, Baumgartner

, et al. Transcutaneous vagus nerve stimulation (tVNS) for treatment of drug-resistant epilepsy: A randomized, double-blind clinical trial (cMPsE02). Brain Stimul. 2016; 9(3):356–363.

32.

Shim

, Kwak

, An

, Kim

. Feasibility and safety of transcutaneous vagus nerve stimulation paired with notched music therapy for the treatment of chronic tinnitus. J Audiol Otol. 2015; 19(3):159–167.

33.

Wang

, Maranets

, Lin

, DeZinno

. Is commercially available point finder accurate and reliable in detecting active auricular acupuncture points?. J Altern Complement Med. 2012; 18(9):860–863.

34.

Hranov

. Comorbid anxiety and depression: Illumination of a controversy. Int J Psychiatry Clin Pract. 2007; 11(3):171–189.

35.

Tan

, Molassiotis

, Wang

, Suen

. Adverse events of auricular therapy: A systematic review. Evid Based Complement Alternat Med. 2014; 2014:506758.