Evaluation of an antenatal acupuncture intervention as an adjunct therapy for antenatal depression (AcuAnteDep): study protocol for a pragmatic randomised controlled trial

A pragmatic, parallel-group RCT will be conducted to compare individually tailored, semi-standardised, depression-specific acupuncture with an equivalent attention progressive muscle relaxation (PMR) comparator and usual care. The morbidity of this population is sufficiently serious to maintain existing treatment and evaluate the interventions as adjunctive therapies [93]. The trial design incorporates individualised treatment flexibility [94], typical of a ‘whole systems’ approach [95, 96], within the framework of a semi-standardised protocol [97‐100] and is thereby reflective of the recent emphasis on effectiveness studies of acupuncture interventions to maintain ecological validity [101] as well as generalisability and interpretability [98, 99]. The non-specific, placebo-like effects resulting from intervention participation [102, 103] will be estimated via the attention comparator, whereas antenatal depression progression or remission and compliance to standard therapy will be monitored via the non-treatment control [100].

Ethical approval for this study was granted in February 2015 by the Research and Ethics Office of the New South Wales Department of Health, South West Sydney Local Health District (SWSLHD-HREC/14/LPOOL/400) and the Western Sydney University Human Research Ethics Committee (WSU-HREC/H10993). The trial is registered with the Australian New Zealand Clinical Trials Registry (ACTRN12615000250538). The protocol (version 1.0) has been designed in accordance with the SPIRIT (Standard Protocol Items: Recommendations for Interventional Trials) guidelines for interventional trials [104, 105] (see Additional file 1) and will be conducted in accordance with the Declaration of Helsinki (1964) and the International Conference on Harmonization Good Clinical Practice (1996). Any changes that need to be made to the trial protocol will be communicated to all investigators, the ethics committees and the trial registry.

Women will be eligible if they are at 24 weeks of gestation, ≥18 years of age, have mood disorders and score ≥13 on the Edinburgh Postnatal Depression Scale (EPDS) [106] (indicative of a high probability of current depression) [106, 107]. Women will be excluded if they are experiencing a medically diagnosed major depressive episode ≥2 years in duration and/or have psychotic or manic features rendering them incapable of consent, post-traumatic stress disorder with needle phobia, a current psychiatric assessment of suicidal risk, a condition necessitating bed rest, and/or other major obstetric risks. Participants must also agree not to receive acupuncture or PMR other than that provided within the study.

The study is being conducted at two hospital sites in South Western Sydney, Australia. Both sites provide antenatal outpatient services and are under the governance of the same health authority; however, birthing services are available only in the larger of the two. The region serviced includes suburbs considered to be of disadvantaged SES [108], high ethnic diversity and greater than state average indigenous population density [108, 109]. Pregnant women will be introduced to the study via flyers included in antenatal information packs. Antenatal staff will additionally supply a study poster to all women identified with elevated EPDS scores during routine antenatal screening. Details of these women will be provided to the researcher, who, upon contact, will forward participant information sheets to all interested potential recruits. Once agreeing to join the study, all women will be further screened and, if eligible, randomised after signed informed consent forms containing separate provisions for the collection and use of biological specimens have been obtained in every case (Additional file 2).

Participants and practitioners are not blinded to the study; however, data entry and analysis will be blinded. The randomisation schedule was computer-generated to contain three groups of random numbers, enabling stratification for the three different models of antenatal care (obstetrician, standard and caseload midwifery) as well as to randomly contain ‘blocks of three’ within each group, acupuncture, PMR or usual antenatal hospital care (1:1:1). The schedule was generated by the school statistician (Paul Fahey) and concealed in opaque envelopes by an independent researcher from the National Institute of Complementary Medicine (NICM) (Anthony Good).

The study period will run for 8 consecutive weeks from gestation week 24 to the end of week 31 (Fig. 1). At baseline, age, gestational age, number of previous pregnancies and/or births, relationship status and/or quality, education, employment status and country of birth will be collected, along with depression-related clinical data, including medical diagnoses, entry EPDS score, age of onset and length of index episode, number and severity of past depressive episodes, length and severity of current episode, family history of depression and current medication and/or psychotherapy use. Collected data will be de-identified and securely stored at the NICM.

Owing to the morbidity of this population, women will undergo a weekly psychological assessment with the Kessler 6 (K6) instrument [110], either during the weekly session (intervention groups) or by telephone (usual care). If notable worsening or a ‘cut-off’ score is reached, immediate referral will be made to hospital antenatal mental health services. Any adverse events arising from the acupuncture intervention will be documented and reported to Western Sydney University and the South West Sydney Local Health District, and prevalence will be calculated. Such events are reportedly rare (occurring approximately 1.3 times every 1000 treatments) and generally of a mild [111, 112] to moderate nature [112]. However, if such events do occur, participants will be reminded of their right to withdraw from the study at any time.

In the first session, a full case history will be taken and a traditional Chinese medicine (TCM) and meridian therapy diagnosis will be made. This will also occur for the usual care group upon trial entry. The intervention groups will then receive weekly 1-h treatments over the 8-week period. At the end of each session, 5 minutes will be spent gaining feedback regarding the session. Treatments will be conducted in the antenatal clinics of either hospital or at the NICM.

The acupuncture protocol draws upon TCM theoretical foundations and treatment strategies as well as some generalisable traditional East Asian medicine approaches. The three-stage semi-standardised flexible process (Fig. 2) incorporates a ‘root’- and ‘branch’-style treatment [113, 114] as well as auricular acupuncture. The ‘root’ treatment (step 1) aims to provide a fundamental harmonisation via the extraordinary meridian system, while the ‘branch’ treatment (step 2) focuses on remaining disruptions in a more symptomatic way [114‐116]. Use of the extraordinary meridian system as a root-style approach has been suggested to facilitate a longer-lasting, more penetrating effect [114, 116] by virtue of this system’s ability to regulate the ‘qi’ of all of the yin-and-yang channels [117], access and distribute ‘pre-heaven essence’ throughout the entire body [116, 118] and affect a person at a constitutional level, due to its embryological roots [116, 118, 119]. Specific functions relevant to depression include influences to the marrow [116], brain, spinal chord, hypothalamic–pituitary–ovarian axis [120], circulatory, hepatic, biliary [121] and endocrine [118] systems. Pathology within this system is thought to manifest in the mind or ‘shen’ [118], and as such treatment is recommended for mental and/or emotional problems [118, 120] as well as cases of mixed pattern complexity involving multiple meridians [118]. In step 1, diagnosis of the most disharmonious extraordinary meridian pair will be made on the basis of the presence of relevant mood disturbance symptomology [116] in combination with disease indicating palpatory findings along involved channel trajectories [117‐119]. If uncertainty remains regarding which of the 2 points of the meridian pair will be the master point (MP), each point will be tested with acupressure to see which one provides the greatest meridian improvement. After identification, the gender-appropriate MP and coupled point (CP) (Table 1) will be unilaterally needled with the meridian flow to ensure that the areas traversed by both vessels are influenced [118]. This method was recommended by Maciocia in 2006 [118] for problems of the head and internal organs, weakened body condition and anxiety. It is also suitable for pregnant women in side-lying position. Japanese-style practitioners typically apply polarity devices to the MP and CP to remedy detected pathologies in the extraordinary vessel system [119, 122]. However, for the purposes of generalisability, polarity in this case will be achieved by side of sex-directed unilateral needling and reverse-order withdrawal [116]. Whilst needles are retained for 10–15 minutes, further diagnosis will be performed by palpation and questioning. As many of the symptoms of depression correspond to TCM ‘yang deficiency’ such as lethargy, reduced libido and lack of motivation [116], it has been theorised that disruption to the major yang channel, the extraordinary meridian governor vessel (GV) that traverses the middle of the back and head, is implicated in depression [123]. Treatment recommended by Wang and Zhang in 2010 [123] is the needling of the most painful or obstructed points along this channel or on the slightly adjacent Huato Jia Ji (HJJ) points, serving a dual main channel and extraordinary vessel function [118]. Step 2a will involve needling of the two points located in this way, in combination with either GV 20 (bai hui) [124] + GV 16 (feng fu) [124], the ‘Sea of Marrow’, indicated for mania, suicidal tendencies and calming the spirit; or ‘Shi Shen Cong 4-4 Alert spirits’, indicated for mania, depression, insomnia and calming of the spirit [117]. Selection in either case will be based upon palpation tenderness and/or symptom differentiation. Whilst these needles are retained, step 2b will involve the needling of two to six additional tender points, if required, according to numerous theoretical possibilities (see Tables 2 and 3) such as the ability to clear reflex areas, for a combined total time of 15–20 minutes. A step 2b example is the association between depression and inflammation [125] viewed as ‘heat’ in TCM; hence, if detected at a ‘fire’ point on a meridian trajectory, a remedying treatment involving the needling of ‘water’ and ‘metal’ points’ [116] to extinguish the ‘fire’, may be selected. Step 3 will then comprise two stainless steel pellets (Helio Supply Co., San Jose, CA, USA) being placed on appropriate points in the ear (Table 4) to extend the treatment effect [122] for an additional 5 days. Lifestyle and dietary advice, contraindicated labour augmentation points [126] and painful needling techniques will be avoided [127]. Needles will be inserted into the most sensitive area within the specified point location or ‘live’ point [100] to a depth enabling retention (2–6 mm), either perpendicularly (even technique) or obliquely with the meridian flow (tonification) and taped in place to allow for adjustment of position during the session. AcuGlide (Helio Supply Co.) single-use stainless steel needles 0.16 mm × 30 mm will be used predominantly, with 0.14-mm or 0.12-mm needles selected for tenderer locations.

In our examination of recent reviews of acupuncture as a treatment for depression [35, 36, 128‐130], we identified three main treatment approaches previously used: (1) fixed points for mood disturbance alleviation, (2) fixed points of this function in combination with additional points selected according to TCM pattern differentiation and (3) flexible point selection based entirely upon TCM pattern discrimination. Variations included diagnosis of disrupted extraordinary vessels Chong and Ren in one study [131], an abdominal points focus in another [132] and scalp acupuncture in a third [133]. Many studies included auricular acupuncture as well as multiple points along the extraordinary GV meridian. Two research groups developed flexible manualised protocols based on pattern differentiation [98, 99], one of which also provided modifications for pregnancy [99]. Both have been drawn upon for therapeutic possibilities presented in Tables 2 and 3, in combination with those suggested by Matsumoto and Euler in 2002 [134] and 2008 [135], such as for the alleviation of commonly observed LR Qi stagnation [98]. It is theorised that the approach of extraordinary meridian treatment in combination with mental disturbance–oriented pattern differentiation and symptom alleviation may provide additional therapeutic outcomes. In 2009, Finston [136] used a similar strategy in patients with severe mental disorders and reported significant reduction or alleviation of psychotic symptoms [136].

The PMR comparator has been adapted from previous research [137] to achieve overall body relaxation in all sessions with an additional weekly focus as follows: introduction (week 1); lower legs and knees (week 2); upper legs and buttocks (week 3); lower back and pelvic floor (week 4); upper back and chest (week 5); arms and shoulders (week 6); head, face and neck (week 7); and integration (week 8). Participants, whilst lying in a comfortable position, will be guided through the PMR in a one-to-one format. A participant who is unable to attend will be offered either a recording- or telephone-based session for home administration.

All groups will continue standard antenatal care, which may include medication and or counselling as well as antenatal classes. In compensation for not receiving treatment, the usual care group will be offered four acupuncture sessions after collection of final data at 6 weeks post-natally.

The principal investigator of this trial (SMO) has 14 years of full-time clinical experience and will conduct all acupuncture sessions, unless unable, in which case a backup acupuncturist with 14 years of part-time clinical experience will be employed. Both practitioners have bachelor’s degrees in acupuncture and are registered with the Chinese Medicine Board of Australia and the Australian Acupuncture and Chinese Medicine Association. The principal investigator will also conduct all PMR sessions, unless unable, in which case a backup associate researcher will be employed.

Data and samples will be collected at various time points, including baseline, mid-intervention (week 4), end of intervention (week 8), birthing and hospital discharge (medical records) and 6 weeks post-natally. All questionnaires will be self-administered and collected by the principal investigator. The primary outcome endpoint, reduction in depression severity, will be assessed at baseline, at 4 and 8 weeks from trial entry and at 6 weeks post-natally via the EPDS, an extensively validated, clinically sensitive [138], 10-item self-report questionnaire. Additional secondary outcome endpoints will also be examined as follows :

Every effort will be made to minimise missing data by ensuring weekly contact with participants during the intervention period as well as robust follow-up at later data collection points. Missing data for the primary endpoint of depression will be estimated using multiple imputation data derived from the K6 instrument. All missing data will be reported with reasons given and patterns of occurrence explored.

The principal investigator will process samples as follows. 4 ml of saliva will be obtained via passive drool into sterile CELLSTAR polypropylene test tubes (15 ml, catalogue number 188271; Greiner Bio-One, Frickenhausen, Germany), transported on ice and frozen and stored at −20 °C. Thawed samples will later be batch-processed and concentrated using equilibrated Sep-Pak C18 200-mg resin cartridges (catalogue number WAT 054945; Waters Corporation, Milford, MA, USA) before refreezing at −20 °C in 5-ml sterile Eppendorf tubes (catalogue number 0030119401, Eppendorf, Hauppauge, NY, USA). Triplicate OT concentrations will be batch-calculated using an OT enzyme-linked immunosorbent assay (ELISA) kit (catalogue number ADI-901-153A-0001; Enzo Life Sciences, Farmingdale, NY, USA) and an ELISA plate reader. Blood (2.5 ml) collected into PAXgene blood tubes (catalogue number 762165; BD Diagnostics, Hunt Valley, MD, USA) will be stored at −20 °C no sooner than 2 h after collection. Batch extraction of mRNA will later be performed using PAXgene blood RNA kits (catalogue number 762174; BD Diagnostics) and stored at −20 °C before cDNA libraries are made using iScript Advanced cDNA Synthesis kits (170-8843 Bio-Rad Laboratories, Hercules, CA, USA) and mRNA expression is determined by qRT-PCR using human OTR PCR primers (catalogue number VHPS-6555; Applied Biosystems, Foster City, CA, USA).

Mean antenatal EPDS scores extracted from a recent Australian study [111], as well as estimated post-intervention between-groups differences derived from a meta-analysis of two antenatal depression acupuncture studies [9], were used for a power calculation for this study. Using a power of 80 % and two-sided testing at a 5 % significance level, detecting a significant difference in end of intervention mean ± standard deviation (SD) EPDS scores of 8.9 ± 5 for the acupuncture group plus usual hospital care versus 13 ± 6.5 for usual hospital care will require a total recruitment goal of 75 participants. As high attrition rates have been reported for this population [144], an additional recruitment of 30 % has been added, resulting in a total goal of 96, or 32 participants per group.

Descriptive statistics will be used to describe the characteristics of the study population. An intention-to-treat analysis will be undertaken [145], with between-groups differences explored using analysis of variance for the primary and secondary endpoints of depression, stress, anxiety, quality of life and OT/OTR. Effect sizes will be reported with 95 % confidence intervals, and results will be considered significant if p values are less than 0.05.