Between November 2016 and July 2017, 159 consecutive patients were screened for this cross-sectional study from a university based tertiary headache clinic in Ribeirão Preto, Brazil. It was included in the study males and females between the ages 18 and 55, diagnosed by neurologists specialists in headache according to the International Classification of Headache Disorders (ICHD-III).11
Exclusion criteria encompassed any of the following: self-report of neck injury/trauma, patients who have undergone cervical/cranial nerve block in the past three months, pregnancy, history of infection, with a migraine attack during the assessment, diagnosis of other concomitant headaches, and patients with any other rheumatic or chronic disease such as fibromyalgia, neuralgias or self-report of vestibular disease.
Among the 159 patients, 34 (21.3%) were excluded due to presence of concomitant headache diagnosis (n=12, 7.5%), fibromyalgia, diabetes and psychiatric disorders (n=10; 6.2%), and abortive medication overuse (n=5; 3.1%). Furthermore, 7 patients (4.4%) were additionally excluded for not completing all the assessment procedures required for this study. Therefore, 125 subjects were included in this study and allocated in one of the two groups: migraine without aura (MoA, n= 88) and migraine with aura (MA, n=37) (Figure 1). This study was approved by the Local Ethics Committee (process num.: 16692/2012) and all patients provided written consent before the data collection.
Initially, subjects filled a structured questionnaire containing general information such as sex, age, headache frequency (days/month), attacks intensity (numeric visual scale, NVS), attack average duration (hours), headache onset (years), self-report of neck pain. Furthermore, they completed the neck disability index (NDI) questionnaire.37 The NDI is composed by 10 questions measuring the influence of neck pain on daily activities including personal care aspects, sleep quality and reading, among others. The scores are from 0 to 50 and the patient can be classified in one of the disability levels as follows: mild (5 to 14 points), moderate (15 to 24 points), severe (25 to 35 points) and complete disability (36 or more points).37
Subsequently, a blinded physiotherapist assessed the patients through two physical tests, the Flexion Rotation Test (FRT) and the Craniocervical Flexion Test (CCFT). Both tests are considered part of a cluster suggested in the assessment of patients with headaches.38
The Flexion Rotation Test (FRT) aims to assess the passive range of motion (ROM) of the superior cervical segment (C1-C2).39 Patients were instructed to lay on their back, and with maximal flexion of the cervical spine (supported by the assessor), were submitted to a passive rotation for both sides. The ROM was measured by the CROM® (CROM Performance Attainment Associates, St. Paul, MN, USA). The test was interrupted when the assessor felt a block sensation during the range of motion or when the patients reported pain. A positive test was verified when the ROM was lower than 34o, 10o less than the maximal ROM (44o).39, 40
The CCFT assesses the neuromotor performance of the deep flexors of the cervical spine (Longus Capitis and Colli).41 Patients were instructed to lay on their back and perform an active craniocervical flexion during 10 seconds, monitored by pressure feedback unit – Stabilizer Pressure Biofeedback®, Chatanooga, Hixson, TN, USA. The test is composed by the following 5 progressive stages of pressure, with 10 seconds of duration each one: (1) 20 mmHg, (2) 22 mmHg, (3) 24 mmHg, (4) 28 mmHg, and (5) 30 mmHg.41, 42 Patients performed a familiarization trial before the test in order to ensure the correct performance of the head movement, in order to avoid compensation. The test was interrupted when patients reported pain or demonstrated activation of the superficial flexors of the cervical spine (Sternocleidomastoid and the Anterior Scalene), perceived by the assessor through palpation.42
A priori sample size calculation was performed based on a pilot study with 10 participants in each group. To detect a minimal difference of 6 degrees in the FRT (ES: 0.81), with power of 80% and α=5%, it was required a minimum of 24 participants in each group. For a power of 80% in the detection of stages differences in the CCFT, it was required at least 35 subjects in each group (ES: 0.82). The sample size calculation was performed in the GPower 3.0.10 software.
Descriptive analysis of the data was performed by means of percentage, average and standard derivation (SD). The Shapiro-Wilk test was used to check normal distribution of the data, and it was verified just for the FRT outcome (p=NS).
Prevalence of neck pain among groups were verified through the Qui-square test, and mean scores of the NDI questionnaire were compared by the Mann-Whitney test. The severity categories of both NDI was also compared among groups through the Qui-square test. For the FRT, independent t-Student test was employed to compare the average of groups. The Qui-square test was used to compare the prevalence of a positive FRT test and the stages of the CCFT test among groups. The statistical analysis was performed in the SPSS 20.0 software considering a significance level of 5%.