This pilot study used a fully randomized, controlled, double-blinded, double-simulation research method. The study was registered in the China Clinical Trial Registration Center (registration number:ChiCTR2100054072 23/08/2021 ) and reviewed and approved by the Ethics Committee of the first affiliated hospital of Zhejiang Chinese Medical University with approval number (ethics number:2021-K-491-01 19/07/2021). All subjects signed an informed consent before the study. All methods were carried out in accordance with the tenets of the Declaration of Helsinki as well as the relevant guidelines and regulations.
Eligibility criteria: All participants were diagnosed with CEH according to the diagnostic criteria for CEH in the third edition of the International Classification of Headache Disorders (ICHD-3) in 2018. They were examined at the outpatient clinic of the first affiliated hospital of Zhejiang Chinese Medical University from January 2022 to December 2022
Inclusion criteria were as follows: Meet the diagnostic criteria; Age 18–70 years old, right-handed; Duration of disease ≥ 1 week; and VAS score ≥ 3/10.
Exclusion criteria: History of cervical spine surgery or history of severe cervical spine trauma; Combination of severe osteoporosis, bone tumor, and bone tuberculosis; Combination of cardiovascular, cerebrovascular, digestive, hematologic and other serious medical diseases; Combined with mental diseases, such as claustrophobia and mental retardation; Pregnant or lactating women; People with language comprehension and expression disorders; Contraindications to magnetic resonance, such as metal foreign bodies in the body; and pacemaker implantation.
Elimination criteria: Enrolled cases should be excluded if they meet any of the following criteria: those who are unable to complete the entire treatment and observation process; those who are unable to fully understand the forms and questionnaires provided as part of the research process or who require assistance from others to complete them; those who are receiving other therapies at the same time during the treatment process; and those who have poor adherence to the study.
Dropout Criteria: Cases that have been enrolled but have not completed the clinical program should be considered to have dropped out in the following cases: The patient withdraws from the study on his/her own; Subjects with poor compliance cannot come for treatment as required, and the patient does not cooperate; The treatment is effective during treatment but cannot complete the entire observation program.; The treatment is effective during treatment, but the patient cannot complete the entire observation program; The subject does not withdraw from the study but no longer accepts the treatment and loses the visit; In the course of the study, other diseases occur, and the treatment is stopped for more than one week in the middle of the study.
Procedures
First, all participants signed an informed consent form and were randomly allocated to the Mulligan manipulative therapy group and the control group. The randomization method achieved allocation concealment using randomization envelopes that were opaque and sealed randomization envelopes. The allocation scheme of two groups was placed in envelopes with the serial number of the group on the outside and the group printed on the inside of the envelope. The envelopes with numerical identifiers were sequentially unsealed in accordance with the order of patients who were still not being attended to and the administration of treatment was carried out according to the specified interventions outlined within the envelopes and treatment was administered according to the interventions requested within the envelopes. The treatment regimen received by each subject was determined by the generated randomized sequence. The Clinical Evaluation Centre of the first affiliated hospital of Zhejiang Chinese Medical University was responsible for the efficacy evaluation to achieve blinding of patients and statistical evaluators. A sample expansion of 20% was proposed, given the possibility of attrition in the study (Fig. 1).
Before starting of the study, individuals in both experimental groups underwent the Visual Analogue Scale (VAS) in order to evaluate the perception of pain. Additionally, the Hamilton Anxiety Scale (HAMA) and the Hamilton Depression Scale (HAMD) were administered to assess the emotional aspects of pain. Furthermore, resting-state magnetic resonance imaging (fMRI) was conducted on all participants to evaluate brain functional activity using the ReHo score, a specialized metric for measuring local brain functional activity. At the end of the study, VAS, HAMA, HAMD, and fMRI scans were repeated to assess efficacy.
Interventions
The manipulations used were carried out by a therapist possessing over six years of expertise who underwent training in both the subject matter of investigation and quality control before the beginning of the trial. The patients in the treatment group received one Mulligan maneuver per day, with each session lasting at least 20 min for ten sessions. The health promotion group delivered the health promotion content to all subjects via the WeChat group daily and answered questions about the promotion content for ten days.
The educational content encompasses various aspects such as the definition of CEH, the primary causes of diseases, daily preventive measures, lifestyle habits, and dietary recommendations. It emphasizes the importance of avoiding excessive strain, exposure to wind and cold, and prolonged usage of electronic devices such as computers and mobile phones. However, it does not encompass the treatment of cervical vertebrae or exercise methods for rehabilitation within the scope of health education.
Mulligan Manipulation:
Step 1: Assessment
CEH is associated with abnormal proprioceptive input in the upper cervical segment (C1-C2). Therefore, confirmation of the key site for treatment influences the outcome.To verify the primary stimulus of the provocation movement, if the occurrence of headache is triggered by flexion and extension, it is recommended to carefully manipulate the spinous process of the second cervical vertebra and engage in posterior extension in collaboration with the patient. Confirmation of cervical 2 as the primary location is established when the headache is relieved.Whenever the patient experiences a headache due to rotation, it is recommended to gently move the transverse process of the ipsilateral side of cervical 1 in a sliding motion from posterior to anterior, while also assisting the patient in performing rotational movements. In case that the headache persists, it is recommended to manipulate the transverse process of the contralateral side of cervical vertebra 1 in a sliding motion, while also engaging the patient in rotational movements. In the event that the headache persists, it is recommended to gently manipulate the ipsilateral spinous process of cervical 2 in a sliding motion, while concurrently engaging the patient in rotational movements(Fig. 2).
Step 2: Treatment
Flexion-extension headache: In the seated position, the operator applies a constant posterior-anterior thrust at the assessed site of application (posterior border of the C2 spinous process), This thrust is accompanied by a compressive force of approximately 50N, directed towards the patient's eyes. The operator instructs the patient to perform a cervical lordosis maneuver while the remaining fingers of the therapist immobilize the patient's cheeks. This procedure is repeated three times.
Rotational headache: In the seated position, the operator applied a steady posterior-anterior thrust at the assessed site of application (left or right side of the C1 transverse process), maintaining a compressive force of approximately 50N and directed towards the patient's eyes. Simultaneously, the patient was instructed to turn his head to the same side, and the procedure was repeated three times.
Step 3: Exercise
Flexion-extension headache: The patient is seated, and the middle or index finger of patient is in contact with the posterior aspect of posterior border of C2, exerting and maintaining a posterior-anterior slide while actively extending the neck posteriorly.
Rotational headache: The patient takes a sitting position, the middle finger or index finger of patient touches the posterior aspect of C1 transverse process on the symptomatic side, and the finger continuously pushes the cervical vertebrae to slide forward while actively turning the head to the same side.
Image acquisition
A GE Discovery MR750 3.0T MRI scanner was used for imaging. The subjects underwent structural and BOLD functional imaging, and the scanning range was from the frontal-parietal to the base of skull, covering the entire brain. The scanning parameter settings: The structural image was acquired using the weighted fast scrambled phase gradient echo sequence technique. This allowed for acquiring the axial weighted images of T1W1 and T2W1. The scanning parameters were as follows: a repetition time of 1,900 ms, a field of view set to 256×256 cm, an echo time of 2.26 ms, a matrix size of 256×256, and a scanning time of 7 min. Additionally, BOLD functional imaging was performed using the gradient echo sequence technique with a single excitation planar echo to acquire the information function. Scanning parameters: The repetition time is 2000 ms, the time point is 240, the echo time is 30 ms, the number of layers is 30, the layer thickness is 5 mm, the matrix is set to 64×64, the field of view is set to 256×256 cm, the flip angle is 90°, and the scanning time is 8 min.
Data preprocessing
Preprocessing was performed using the Statistical Paretric Mapping12 (SPM12) software package based on the MATLAB R2014a platform, which mainly includes removing time points, temporal layer correction and head movement correction, spatial normalization, de-linearization drift, and noise filtration. After preprocessing, the data were used to calculate the local consistency of each voxel with its neighboring 27 voxels in the same time series by applying Kendall's concordance coefficient. After preprocessing, Kendall's concordance coefficient was applied to calculate the local coherence between each voxel and its neighboring 27 voxels in the same time series. Subsequently, the whole-brain ReHo images of subjects' whole brains were generated. Finally, the whole-brain ReHo images were spatially smoothed using a smoothing kernel with a full-width and half-height of 6 mm to filter out the noise.
Sample size calculation
PASS26 software was used for sample size estimation. This study was a randomized controlled trial wherein two groups were established based on the pre-experiment outcomes. These groups were referred to as the experimental group and the control group. The treatment effectiveness of the two groups was expected to be 90% and 36.67%, respectively, with two-sided α = 0.05 and confidence level 1-α = 0.95 and the test efficacy 1-\({\beta }\) not less than 0.8.The allocation of participants into the test and control groups followed a 1:1 ratio. Each group initially consisted of 16 cases, and the sample size was increased by 20% to account for potential factors including loss of visits and attrition. Consequently, a minimum of 20 cases per group was required for the study subjects. The sample size estimation formula for comparing the two sample rates is as follows:
n1 = n2 =\(\frac{{\left[{u}_{\alpha }\sqrt{2p(1-p)}+{u}_{\beta }\sqrt{{p}_{1}(1-{p}_{1})+{p}_{2}(1-{p}_{2})}\right]}^{2}}{{({p}_{1}-{p}_{2})}^{2}}\)
n is the required sample size for each group, Eq.\({p}_{1}\)and\({p}_{2}\)are the estimated two-sample frequencies, respectively, and\(p\) the two-sample combined frequency.
Statistical analysis
Data were statistically analyzed using the SPSS26 software package (SPSS Inc. USA). Measurement data was described using \(\stackrel{-}{x}\)±s and count data were described using constitutive ratios and frequencies. Independent samples t-test and paired rank sum test (Z-test) were used to compare the differences in indicators between the two groups of CEH patients. The RESTplusv1.24 software package was used for statistical analysis of the data. The smReHo plots of CEH patients in the two groups were subjected to paired samples t-test, corrected for multiple comparisons by the GRF method (P < 0.01 for voxel level, P < 0.05 for clusters). The ReHo values of different brain regions were extracted and analyzed by Pearson correlation analysis with clinical behavior indices (VAS, HAMA, HAMD), and the differences were considered statistically significant at P < 0.05.