Study aim
The study aim is to determine the effects of a treatment series consisting of stretching and SMT vs stretching alone on HRV and pain in a clinical setting in a population of patients with recurrent or persistent NP. A secondary aim is to test CPM as a predictor of treatment outcome in terms of pain.
Setting
This multicentre randomized controlled clinical trial will be carried out in multidisciplinary primary care clinics where physiotherapists and chiropractors are consulted for musculoskeletal pain. These types of clinics are selected to minimize bias from patients having expectations towards a specific treatment modality.
A total of 6 clinics will take part in this study, and each clinic will include 20 subjects, resulting in 120 subjects in total, 60 in each treatment arm.
Two research trained clinicians will conduct all the measurements.
The treatments (both intervention and control) will be delivered by clinicians (licenced chiropractors) in the participating clinics.
Eligibility criteria
Inclusion criteria: minimum 18 years old, able to read and understand Swedish. The presence of recurrent (at least one previous episode) and persistent (duration more than 6 months) NP, and no chiropractic treatment during the previous 3 months. This interval was chosen as research have shown that similar treatments seem to have little effect beyond three months (24).
Exclusion criteria: conditions or medications that will affect the HRV measurements such as cardiovascular disease, hypertension, diabetes, pregnancy, obesity (BMI >30), currently using pain-reducing medication on a daily basis, steroids, beta-blockers or antidepressants. All contraindications to SMT, i.e. anything that could seriously aggravate the pain (e.g. inflammatory conditions) or be indicative of cerebrovascular injuries (previous drop attacks or a recent episode of a new headache or dizziness) will exclude the patient from the study.
Procedure (Please see flow chart, fig. 1 and 2)
Recruitment
Patients in this study are self-referred after hearing about the study from another health care provider, or reading about the study in an advertisement or newsletter. A research assistant calls the patient and assess eligibility using a standardized form. The patients are informed about the aim of the study and the study procedures. If eligible, the patient is scheduled for all study visits during this call. Logistical details of this recruitment stage will be adapted to individual clinics as some clinics have newsletters and some use social media to inform their patients about clinic news. Different local newspapers are also used to recruit patients, as the individual clinics are located in and around the Stockholm area.
Study visit
On their first visit, patients will sign an informed consent form and have their baseline measurements taken. Then, they will be randomized to one of the treatment arms and treated accordingly. On the day of the study visit, the participants will refrain from caffeine, nicotine, alcohol, and from performing strenuous exercise.
A standardized protocol is followed on the day of the first study visit: After consenting to participate, the subjects will start by filling out a baseline questionnaire for demographic information. They will also answer questions concerning their NP (duration, episodes, intensity and frequency) as well as questions regarding pain levels and the affective quality of pain using the NRS-11 scale (25, 26), STarT Back (27, 28) and the short-form McGill Pain Questionnaire (29, 30). Pain will be measured as average pain over the last 24 hours. This information is collected on paper on the first visit, and transferred to a secure server at Karolinska Institutet (KI) by a research assistant. The follow up questionnaires are digital, administered through Karolinska Institutet, managed by Survey & Report by Artologic (https://www.artologik.com/en/SurveyAndReport.aspx).
Measurements
The equipment used to measure HRV is called FirstBeat (https://www.firstbeat.com/en/). It is applied by the research clinician, and the participants will rest quietly for 5 minutes with the equipment attached before baseline resting values are ascertained over a period of 5 minutes. After this, CPM will be tested with a structured CPM test (31). This test includes mechanical pressure point intensity and a cold pressor test (22). Reported pain measurements during the CPM test are noted in a paper form and transferred to a secure KI server by a research assistant. When patients leave, the HRV-equipment will still be attached to their chest, so that a measurement can be done the following night to record HRV at their deepest sleep (32). Data collected from the FirstBeat monitors are downloaded to a secure computer administered by Karolinska Institutet.
After the measurements, study subjects are randomized into one of the two treatment arms. Their allocated clinician will conduct a standard anamnesis and examination procedure including neuro-orthopaedic tests to further assess exclusion criteria. The treatment protocol is then initiated according to allocation. All subjects will be scheduled to a treatment series consisting of 5 visits over 2 weeks.
Data on normal treatment reactions (tiredness/fatigue and pain/tenderness) are collected the day after the first treatment using SMS (https://www.sms-track.com/) to ascertain the type and level of reactions to the interventions (33, 34). The data is automatically stored in a secure cloud, accessible only by authorized researchers. For analysis, the data is transferred to a secure KI server.
Before the subjects’ third treatment, a second measurement of HRV and CPM will be conducted, and measurements of pain will be ascertained, using the standardized protocol used at the initial visit.
Before the patients’ fifth treatment, or at least two days after the fourth visit, the final measurements will be conducted, this time the HRV-equipment will be taken off directly after the measurement. Again, the standard questionnaire measuring pain will be answered.
After the study period (4 visits), the clinicians are free to select any other treatment modality for the patients. However, patients will be monitored with questionnaires every other week (via email) regarding their affective quality of pain and pain levels for 2 months after their final measurement at the clinic. The clinicians will also report what treatment modalities were used after the initial two weeks of the study.
Patients who do not complete the full treatment plan will be asked to complete all measurements in order to study attrition and to complete a drop-out analysis.
Randomization procedure
Consecutively numbered opaque envelopes containing group allocation are created off-site at the research centre by a statistician. A 1:1 allocation ratio in randomly permuted blocks of different sizes according to a randomization schedule is used. The envelopes are arranged in batches of 20 and distributed to the clinics at the start of each data collection period. SPSS v20 is used to generate the randomization code. The envelopes are opened consecutively by the treating clinician.
Blinding
The subjects will be unaware of what treatment the other group is receiving, as they will be told that the study is testing two different treatment modalities with similar clinical benefit to examine the effect over two weeks on physiological parameters and pain. Subjects in both treatment arms should feel that complying with their treatment plan during two weeks is a necessity for their improvement.
It will not be possible to blind the clinicians performing the treatments. The research clinicians who will collect the data in the experiment will be blinded to the treatment allocation. The statistical analysis will be performed with the treatment allocation blinded.
Sample size
Log root mean squared successive differences in RR intervals (RMSSD) is the primary measurement of HRV. We will also explore other aspects of HRV according to Task Force Standards (35) to get an overall impression of the subjects’ HRV. In a recent study that examined the reliability of HRV measures, the sample size was estimated to 20 subjects in each group to detect a mean change of 20% in RMSSD, and 20-50 subjects in each group to detect a change of 10% (36). A difference of 10-20% has been considered to be clinically important (36). This value has also been used by other researchers investigating changes in HRV from manual treatment (37). With a significance level of 5%, it was estimated that 60 subjects were needed in each treatment arm to reach a power of 80%. This is also in line with the general recommendations to detect a medium effect size (38). A high number of dropouts is not expected in this study as it is conducted using an effective practise-based research network with established and tested routines developed to minimize the burden on participating patients.
Treatment arms
SMT in this study is defined both as a High-Velocity-Low -Amplitude (HVLA) thrust applied to the target joint, and also spinal mobilization (MOB) where the application of manual force to the spinal joints is within the passive range of joint motion and does not involve a thrust (39). The type of techniques applied will be decided upon and described by the participating clinicians (chiropractors), and both HVLA and MOB will be considered manual treatments as they have been found to have similar effects on several pain parameters in a recent multicentre study (40). This also provides the possibility for the chiropractor to adapt the force applied to the individual patient, which is normally done in the clinical encounter.
As the participating chiropractors have similar educational backgrounds and have received the same instructions concerning their limited choice of treatment techniques, we expect that they will have a similar approach to SMT. Data on the specific interventions will be collected.
The common modality used in both treatment arms is a program of home stretching exercises. Both groups will receive both verbal and written instructions describing the home stretching exercises that have been recommended as a low-cost first instance intervention for NP (additional file no.1) (16). Patients will be instructed to keep an exercise diary to monitor their exercise frequency (additional file no.1) (16).
The testing of HRV and CPM will be conducted by two research clinicians only, to ensure consistency throughout the study. The two research clinicians will meet several times in advance of the study to test and calibrate the examination procedures and their communication with the study subjects.
Outcome measures
Experimental measures
The primary outcome is log root mean squared successive differences in RR intervals (RMSSD). The variation in beat-to-beat heart rate is an indicator of parasympathetic and sympathetic modulation of the heart rhythm. Deviations in HRV have been found in patients with both acute and chronic pain. Patients with various chronic pain conditions show reduced parasympathetic activity at rest, the proposed mechanism behind central sensitisation (41). Thus, vagus activity, assessed through HRV, is suggested to correlate with pain severity and could possibly be used as a proxy for treatment efficacy in patients with chronic pain (41). There are some studies showing that SMT influence HRV, but the quality is questioned (42).
In this study, Conditioned Pain Modulation consists of the evaluation of a painful test stimulus followed by a second evaluation after the painful conditioning stimulus has been withdrawn (sequential stimuli) (21). CMP is a well-known concept in modern medicine, particularly when it comes to prediction of post-operative pain (43). It has been suggested that a dysfunctional CPM response can be a pathogenic factor in the development of chronic pain, but also that a dysfunctional CPM response can be the result of chronic pain, hence a possible bi-directional relationship (44).
There is a growing body of evidence suggesting that CPM may be an important biomarker of chronic pain and a predictor of treatment response (21). One may suggest that in patients with chronic pain and a dysfunctional CPM response, treatments with approaches that address the central nervous system mechanisms (e.g., pharmacological and cognitive) could be the first choice of treatment. Patients with chronic pain that demonstrate a dysfunctional CPM might also be particularly sensitive to interventions that help to reduce specific local nociceptive input (eg, physical medicine and manual treatment). However, standardization of CPM testing is lacking (31).
Our study will use a structured CPM testing protocol with a standardized clamp pressing on the thumb nail for 10 seconds as the test stimulus, and cold water (0-2˚C) as the conditioning stimulus, previously tested and validated by O’Neill 2015 (22). A NRS 11 point scale will record pain associated with both stimuli. This allow us to examine if the CPM responses are predictive of treatment outcomes after SMT and stretching exercises.
Patient- reported outcome measures:
Secondary measurements such as disability and health related quality of life will be collected. The subjective pain experience will also be evaluated, as this is important when considering chronic pain (45). These will be collected using the standard instruments described below:
The neck disability index is an instrument designed to measure disability, and has been shown to be reliable and valid in Swedish (46).
Pain intensity is measured with a validated NRS-11 scale where the subjects grades their perceived pain level using the anchors “no pain” and “worst possible pain”(25, 26).
Measures of self-rated health is assessed by EQ-5D, a translated (Swedish) and validated questionnaire with five domains and three answer options in each (47, 48).
To assess the affective quality of pain, the Swedish version of the short-form McGill pain questionnaire-2 will be used. This is a validated questionnaire (29, 30) used in clinical trials designed to measure the subjective pain experience.
These questionnaires will be given at baseline, and again at the 2nd and 3rd measurements. In addition, they will be administered every other week the following 2 months after the study period has ended. Pain intensity (NRS-11) (25, 26) will be collected daily during the two week study period using text-messages (SMS) and every other week the following 2 months using emailed questionnaires.
The most common side-effects following SMT are local tenderness and tiredness of a short duration (19). In this study, the reactions to both treatment arms will be monitored using SMS sent to the participants one day after the first treatment.
Time line
The data collection is expected to commence in January 2019 and to be finished within 12 months.
Analysis
Intention to treat analysis will be applied.
Univariate multiple regression analysis (one outcome) and/or multivariate multiple regression analysis (more than one outcome) will be used to analyse the primary and secondary outcomes of the trial.
If appropriate age and sex will be included as covariates and controlled for in the model.
To investigate whether CPM is a predictor of treatment outcome we will look at the statistical interaction between CPM and SMT with regards to their effects on the primary outcome.
Ethical aspects
SMT is applied clinically in musculoskeletal health care and has been examined in a variety of research studies. Serious complications are very rare (49, 50). The present study examines two commonly used treatment protocols in a clinical environment, which means that the subjects will not be subjected to a treatment that they would not normally receive when consulting for care. All test methods are well-established procedures commonly used in research practice. If a subject experience an unexpected reaction to treatment or testing procedure, the subject will be taken out from the study by the research assistant, and undertake an individual treatment plan.
The clinicians (chiropractors) who performs the treatments all have an academic degree, are licensed by the Swedish National Board of Health and Welfare (the national Patient Safety Act applies) and has a personal liability insurance through their professional federation (https://www.lkr.se/) (Nordic Insurances). Thus, the subjects are insured in case of adverse events from treatment.
When screening for eligibility, both written and verbal information about the purpose of the study, treatments, measurements and SMS procedures will be provided. At the study visit the patient will have an opportunity to ask the research clinician relevant questions, but will also be provided with a telephone number where a part of the research group not involved in the data collection can answer all the questions they may have. All study subjects will sign informed consent forms before entering the trial.
Upon registration in the study, each subject will receive a subject identification number (ID), replacing personal identity number and name, which all measurement data and patient reported data will be linked to. The key that match the subject ID with personal identity number and name will be securely stored in a locked fireproof cabinet at Karolinska Institutet in accordance with the National Board of Health and Welfare's requirements for storage of journal documents.
During the data collection, data are recorded and processed by the research clinicians, and all entries in the databases are recorded using the subject ID only. During the analyses, data will be completely anonymised and only the involved researchers will have access to the data, which will be stored electronically at Karolinska Institutet in accordance with local rules and European GDPR regulations.
All reporting will be done at the group level without the possibility of identifying any individual study subjects. The results of the study will be published in open access journals, and will be communicated through several professional channels nationally and internationally.
Central ethical approval has been confirmed from the Regional Ethical Review Board in Stockholm (ref approval no. 2018/2137-31) and has approved participation of all individual centres in the trial.