Traditional Chinese medicine (TCM) collaborative care for patients with axial spondyloarthritis (AcuSpA): protocol for a pragmatic randomized controlled trial

This is a two-arm, pragmatic, randomized controlled trial to evaluate the clinical effectiveness, safety, and cost-effectiveness of TCMCMC, in particular acupuncture, for patients with AxSpA with inadequate response to NSAIDs [21‐24]. It is anchored in the Pragmatic Explanatory Continuum Indicator Summary Framework-2 (PRECIS-2) criteria and the extended Consolidated Standards of Reporting Trials (CONSORT) guidelines for pragmatic trials as well as the Standards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA) statement for acupuncture. The two arms are (1) TCMCMC (including acupuncture) plus usual rheumatologic care and (2) usual rheumatologic care alone. Participants will be recruited from dedicated clinics in a tertiary hospital setting. Recruitment is mainly through doctors’ referrals and recruitment posters and brochures. Patients will be randomly allocated to receive usual rheumatologic care or the intervention (TCMCMC) with a 1:1 allocation ratio via random permuted block randomization. The trial work plan is summarized in Fig. 2. This paper is based on protocol version 3.0, 30 August 2018.

We aim to recruit patients with AxSpA who have spinal pain and active disease despite standard medical therapy. Patients are eligible for the study if they are 21 years of age or older; have AxSpA, diagnosed according to the 2009 Assessment of Spondyloarthritis International Society (ASAS) criteria [25]; have active disease based on Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) score ≥ 4 on a 11-point numerical rating sScale (NRS) and spinal pain score ≥ 4 on a 11-point NRS [26]; have failed two sequential NSAIDs (including cyclooxygenase-2 inhibitor) at maximal tolerated doses for ≥4 weeks in total; and have received no biologic therapy (i.e., tumor necrosis factor blocker or anti-interleukin 17) within the past 3 months. Patients who are on current treatment with concomitant methotrexate (MTX) or sulfasalazine (SSZ) at study entry must be on the drug for ≥ 12 weeks and at stable dose for ≥ 4 weeks prior to randomization. Patients who are on non-biologic disease-modifying antirheumatic drugs (DMARDs) other than MTX or SSZ must discontinue the DMARD 4 weeks prior to randomization, except for leflunomide, which has to be discontinued for 8 weeks prior to randomization unless a cholestyramine washout has been performed. Patients taking systemic corticosteroids have to be on a stable dose of ≤10 mg/day prednisolone or equivalent for at least 2 weeks before randomization. Patients with a BASDAI 50% response to NSAIDs will be recruited in one block, while patients who did not have a BASDAI 50% response to NSAIDs will be recruited in another block.

We will exclude female patients who are pregnant or breastfeeding; on antiplatelet agents (i.e., aspirin, clopidogrel, dipyridamole, etc.) or anticoagulants (i.e., warfarin, enoxaparin, rivaroxaban, dabigatran, etc.); have bleeding disorders; or have blood-borne communicable diseases (e.g., hepatitis B, hepatitis C, human immunodeficiency virus, etc.).

The attending rheumatologist will identify eligible patients according to the inclusion and exclusion criteria. Informed consent will be taken by the attending rheumatologist, and participants will be referred to a research coordinator who will randomly assign them to the intervention or control arm. The research coordinator will have custody of the randomization list that was pre-generated by the biostatistician using a computerized random number generator, and will assign treatment accordingly. The attending rheumatologist who will enroll participants will not be aware of the allocation sequence.

The randomization list is kept by the biostatistician and research coordinator until the end of the study to ensure allocation concealment; therefore, the data analysts will be kept blinded to the allocation. The participants will be instructed not to disclose the allocation to the attending rheumatologist.

Usual rheumatologic care, guided by the management guidelines developed by ASAS, consists of a referral to the physiotherapist for therapeutic exercise, medications including NSAIDs, and regular monitoring for complications which may arise from AxSpA such as uveitis, interstitial lung disease, and risk factors for cardiovascular disease and osteoporosis [27]. In this trial, the attending rheumatologist will see each patient at regular intervals (ranging from 6 weeks to 6 months) depending on the patient’s condition as per routine care. At each session, the attending rheumatologist will conduct a thorough physical examination and monitor the disease activity using validated patient-reported outcome measures. The rheumatologist will be allowed to prescribe the full range of medications, including biological agents, as per routine care and according to local treatment guidelines. There are no concomitant or prohibited interventions during this trial. The rheumatologist will remind patients not to visit any TCM physician and not to seek alternative therapy for the duration of the study.

TCM physicians registered with the Singapore Traditional Chinese Medicine Practitioners Board with at least 3 years of experience will participate in the study. Prior to the treatment sessions, all the acupuncturists will undergo training to ensure standardization of acupuncture techniques. The intervention arm will involve the TCM physicians in the management of the patients, in addition to usual rheumatologic care. The clinical interventions were designed and agreed on by senior TCM physicians in consultation with rheumatologists. The clinical interventions to be carried out by the TCM physicians include counseling, diagnosis based on TCM clinical syndromes, and the prescription of acupuncture.

The main acupuncture points of interest have been identified a priori as Jiaji, Shenshu, Yaoyangguan, Mingmen, Huantiao, and Ashixue. We will use sterile disposable stainless-steel needles of 0.25 mm diameter, 25 mm or 40 mm long and 0.30 mm diameter, 50 mm or 70 mm long, depending on the acupuncture points. In addition to the main acupuncture points specific for the treatment of AxSpA, the TCM physicians will be allowed to make minor adjustments to the acupuncture points in view of the differing constitution of the patients as per the holistic treatment philosophy of TCM as shown in Fig. 3. Therefore, the number of needles to be inserted per subject per session will differ. The needles will be inserted from 0.25 to 1 in. deep depending on the acupuncture points. After eliciting the de qi response, the needles will be left in place for 30 min. The needles will be stimulated manually every 10 min. The acupuncture treatment will consist of a total of 10 sessions (or 2 courses) in total. Each course of treatment will consist of 5 acupuncture sessions held over 2 weeks, each session lasting 30 min. The patient will have a break of at least 3 days and up to 1 week in between each course of acupuncture. TCM physicians will document components of treatment and adherence in standardized logbooks. The reporting of the intervention is guided by STRICTA guidelines (Additional file 1: Table S2).

The outcomes selection was guided by the ASAS and Outcome Measures in Rheumatology (OMERACT) core domains for SpA [27]. Spinal pain score at week 6 is the primary outcome of this study, since pain is what matters most to patients with AxSpA. The 11-point pain NRS (range 0–10), which is self-administered and validated across many settings, will be used [28]. The 6-week time-point was selected to investigate if there is short-term efficacy in the intervention. The primary and secondary outcome measures are shown in Table 1.

Spinal pain score at week 24 is the main secondary outcome in this study. This time-point was selected to investigate if there is sustained long-term effectiveness for the intervention.

We will also collect secondary clinical outcomes, quality of life outcomes, and economic outcomes. Secondary clinical outcomes include spinal pain at week 12 and week 24 and clinical parameters, including BASDAI [29], Bath Ankylosing Spondylitis Functional Index (BASFI) [30], Bath Ankylosing Spondylitis Global score (BAS-G) [31], and Health Assessment Questionnaire (HAQ) [32, 33] at weeks 6, 12, and 24.

We will assess quality of life (QoL) using the 36-item Short Form Health Survey (SF-36) [34] and the Ankylosing Spondylitis Quality of Life (ASQoL) questionnaire [35, 36], which will be administered at baseline and at weeks 6, 12, and 24. Both patient-reported outcomes have been validated in the Singapore population [34].

The spinal pain, BASDAI, BASFI, BAS-G, HAQ, SF-36, and ASQoL results, rheumatologist’s consultation fees, cost of laboratory procedures, and number of inpatient days at week 52 will serve as exploratory outcomes.

BASDAI (ranging from 0 to 10) is an English, self-administered, disease-specific questionnaire to measure disease activity, with higher values indicating more active disease [37]. BASFI (ranging from 0 to 10) is a disease-specific questionnaire used to measure physical functioning, with higher values indicating worse functioning [37]. BAS-G (ranging from 0 to 100) is a disease-specific questionnaire to give a global assessment of well-being, with higher score reflecting poorer well-being [31].

HAQ (ranging from 0 to 3) is a self-administered, generic questionnaire used to assess disability, with a higher score reflecting worse disability [38]. It includes eight domains: dressing, arising, eating, walking, hygiene, reach, grip, and daily activities. The English version of HAQ is a valid and reliable measure for use in patients with AxSpA in Singapore [33].

SF-36 version 2 is a self-administered, generic questionnaire used to measure QoL in eight areas of perceived health. The 36 individual questions make up 8 subscales, with lower scores reflecting poorer QoL. Norm-based scores were used whereby 50 represents the mean and 10 represents the standard deviation. The eight subscales are physical functioning, role limitation due to physical problems, bodily pain, general health, vitality, social functioning, and role limitation due to emotional problem and mental health. These eight subscales were summed with different weights to give two summary scores: physical component summary (PCS) and mental component summary (MCS) [39]. The English version of SF-36 has been shown to be valid and reliable in patients with AxSpA in Singapore [34].

ASQoL is a self-administered, patient-derived, and disease-specific measure of QoL for AS. It consists of 18 items with a “yes” (scored as 1) or “no” (scored as 0) response to each item. All item scores are summed to a total score ranging from 0 to 18 [40], with higher scores indicating worse QoL. The English version of ASQoL has been shown to be a valid and reliable measure for use in patients with AxSpA, also in Singapore [36].

Adherence to treatment defined as the number of acupuncture sessions attended by the patients in the intervention arm will be assessed. We will also record the number of acupuncture and rheumatology consultations and the number of needles used in each acupuncture session. The TCM physicians will complete a checklist to ensure standardization of acupuncture treatment across all sessions (Additional file 1: Table S3).

For this study, an adverse event (AE) is any untoward medical occurrence in a patient which does not necessarily have a causal relationship with the treatment. An AE can therefore be any unfavorable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the treatment. A serious adverse event (SAE) is any untoward medical occurrence that at any dose results in death, is life-threatening, requires inpatient hospitalization or prolongation of existing hospitalization, results in persistent or significant disability/incapacity, or is a congenital anomaly/birth defect. However, pre-planned hospitalizations before recruitment will not be recorded as AEs.

AEs will be recorded throughout the study up to week 52. Acupuncture-related AEs are defined as symptoms or complications related to acupuncture that begin or worsen after the first session of acupuncture through the last session of acupuncture [41]. We adapted the AEs of interest from a systematic review of AEs of acupuncture treatment, including broken needles, fainting during a session, and local infections at the site of acupuncture.

AEs related to usual rheumatologic care were adapted from previously conducted randomized controlled trials performed in patients with AxSpA [42, 43]. Adverse effects related to usual rheumatologic care include abdominal distension and pain, nausea, diarrhea, gastrointestinal disease, headache, giddiness, upper respiratory tract infection, nasopharyngitis, malignancy, infections, and hepatic-related AEs. These AEs will be recorded.

As this is the first study to explore a collaborative model between TCM physician and rheumatologists in the field of AxSpA, we based our sample size calculation on a study by Meng et al. [11]. With a conservative estimate of 0.6 point difference on a 11-point scale in spinal pain score between the two arms, and assuming a standard deviation of 1.2 for both arms, approximately 64 patients are needed for each arm to obtain a statistical power of 80% (two-sided type I error rate of 0.05) based on a 1:1 treatment allocation. Allowing for a dropout rate of 20%, a total of 160 patients with 80 patients per arm will be needed. According to Cohen, this effect size is considered “moderate” [44].

To ensure the accuracy of outcome assessments and data collection, the investigators and research coordinator will attend a training workshop before the start of the trial. All attendees will be provided with a protocol and will discuss the topics they may feel confused about until everyone is totally clear about the procedures. Data will be entered into SingHealth Research Electronic Data Capture (REDCap) [45], which is a password-protected software application designed to collect and manage data for research studies. All hardcopy documents will be kept in a locked cabinet. Range checks will be implemented for data values in REDCap to promote data quality. There is no plan to promote participant retention and complete follow-up.

A data and safety monitoring board, independent from the sponsor and competing interests, will meet annually to provide interim monitoring of the safety and efficiency data for the study. The data and safety monitoring board, comprising senior clinical experts and external biostatisticians, will help ensure the availability of appropriate expertise in trial design, execution, interim monitoring, analysis, and reporting. After each interim analysis, the data and safety monitoring board will determine whether it is necessary to continue, modify, or terminate the collection of these outcome data.

The statistical analysis will be performed on an intention-to-treat (ITT) basis. The primary outcome of interest is the difference in pain score at week 6 between the intervention and control arms. The baseline characteristics will be shown as the mean ± standard deviation (or median and interquartile range where adequate) for continuous variables (e.g., age) and n (%) for categorical variables (e.g., gender). Primary outcome of pain score at week 6 will be compared between the arms using Student’s t test. Further adjustment will be made with baseline pain score using analysis of covariance (ANCOVA). Multiple imputations will be used to account for missing values. All evaluations will be made assuming a two-sided type I error rate set at 0.05.

For secondary outcomes with repeated measurements (at weeks 6, 12, 24, and 52), we will use a linear mixed model to account for within-individual correlation among measurements and the sandwich estimator to obtain robust standard error estimates. The intervention indicator and time factor will be included among the linear predictors adjusting for baseline covariates.

To assess the safety of the intervention, we will present the frequency of AEs that occurred, expressed in frequency and percentages, in both the intervention and control arms within the study period of 52 weeks.

The economic evaluation will be conducted from both the healthcare system and societal perspectives. Both a cost-effectiveness analysis (i.e., cost of reduction in 1 pain score point) and cost-utility analysis (i.e., cost of reduction in 1 quality-adjusted life year saved) will be performed [46, 47]. Costs will include direct healthcare-related costs of the TCM physician, rheumatologist, hospital stays, and any drugs. The indirect costs caused by lost workdays will also be considered. Health utility will be measured using the Short-Form Six-Dimension (SF-6D) scale, which is a derivation of the SF-36 scale [48]. The cost-effectiveness and cost-utility analyses will be performed by calculating the incremental cost-effectiveness ratio and incremental cost-utility ratio respectively. The analysis period will be at weeks 6, 12, 24, and 52.

The incremental cost-effectiveness ratio will be calculated by dividing the between-group difference in costs by the between-group difference in effects (i.e., costs per pain score reduced). The incremental cost-utility ratio will be calculated by dividing the between-group difference in costs by the between-group difference in utility (i.e., costs per SF-6D unit improvement). Sensitivity analysis will be conducted on the most important cost drivers to assess the robustness of the results.

One interim analysis will be performed by the trial statistician. The statistician will report to the independent data and safety monitoring board, which will provide recommendations without revealing any detailed results on treatment effect. The principal investigator then decides on the continuation of the trial and will report to the ethics committee. There is no formal stopping rule for this study.