Examination of Cervical Thoracic Differentiation Testing in Individuals With Neck Pain

Overview

Neck pain is a very common diagnosis and physical therapy, specifically thoracic manipulation, has been shown to be an effective treatment. Thoracic manipulation, which involves a quick thrust type push to a segment of the midback region of the spine, has been shown to decrease pain and improve function in individuals with neck pain. However, it is not known if there is a test that will accurately identify individuals with neck pain who would respond more favorably to thoracic techniques. The Cervical-Thoracic Differentiation Test (CTDT) is a test used clinically to differentiate pain originating from either the cervical (neck) or thoracic (midback) regions of the spine in individuals with neck pain. It has also been speculated that specific unloading tests, where a tester selectively unloads a portion of the weight of the cervical or thoracic spine through manual distraction may help determine who will respond more favorably to thoracic treatment. To the best of our knowledge, despite widespread clinical use, these tests have not been tested for validity or reliability at this time. Therefore, the purposes of this research are to assess the reliability and criterion referenced validity of the CTDT and unloading tests, as well as examine the relationship between tests and baseline subject characteristics.

The research aims and hypotheses are: Aim 1: Evaluate the inter-tester reliability of the CTDT and regional (cervical or thoracic) unloading tests Ho1: The CTDT and regional unloading tests will not demonstrate reliability greater than κ=.75 Ha1: The CTDT and regional unloading tests will demonstrate reliability greater than κ=.75 Aim 2: Determine if there are differences in pain response following thoracic manipulation between subjects with positive and negative CTDT results. Ho2: There will be no difference in response (change in pain, VAS) following thoracic manipulation between participants with positive CTDT and negative CTDT results Ha2: There will be statistically significant differences in response (change in pain, VAS) between individuals with positive CTDT results compared to those with negative CTDT results Aim 3: Determine if a relationship exists between CTDT results and the results of cervical and thoracic unloading tests. Ho3: There will be no statistically significant correlation between CTDT results and results of unloading tests Ha3: There will be significant correlations between positive CTDT results and positive thoracic unloading results Aim 4: Determine if a relationship exists between overall result (responder/non-responder) and CTDT results Ho4: There will not be a significant relationship between responder status and CTDT results Ha4: There will be a significant relationship, Φ\>.06, between responder status and CTDT result. Aim 5: Determine if a relationship exists between overall result (responder/non-responder) and various demographic factors (Neck Disability Index (NDI) score, age, pain level, duration of symptoms) A total of 50 subjects will be recruited into this trial. All subjects will first complete a brief screening questionnaire to determine study eligibility. All eligible subjects will then undergo an informed consent process including the opportunity to have all questions answered. Once consent is obtained, subjects will then be screened by a licensed physical therapist to rule out any contraindications to study participation. Subjects will also complete the neck disability index (NDI). This questionnaire will be used as a baseline variable to ensure group equality for levels of self perceived disability due to their neck pain. This study will consist of two distinct, but concurrent, phases. The first will consist of a reliability phase. A total of 20 subjects will be tested by each of two trained examiners on the same day for both the CTDT and two unloading tests. These 20 subjects will be part of the larger subject pool, and will be a sample of convenience based on the availability of two examiners on the day of enrollment. Each investigator will perform the CTDT; additionally the investigator will perform a cervical distraction test and a separate thoracic distraction test. These findings will be recorded, and then a second investigator (blinded to the results of the first investigator) will repeat these tests to allow for determination of inter-rater reliability. Upon completion of the reliability testing, these subjects will proceed directly to the intervention phase of the study. Subjects enrolled on days with only one examiner available, or after 20 subjects have been assessed for reliability, will enroll directly into the second phase. The intervention phase will include the thoracic manipulation intervention. Subjects will first be asked to complete a visual analog pain scale, both at rest and while moving their head to their most painful position. In the event that a subject has equal pain with rotation to both sides, they will be asked to choose the most bothersome side as our reference criterion. The CTDT and unloading tests will be performed as described below. For all subjects, the investigator will then perform up to two supine thoracic thrust manipulations. Post-intervention, subjects will repeat the VAS in both the resting and provocative positions. Change in pain on the VAS will be compared to the subject's FINAL trial of the CTDT and unloading tests for data analysis. Specific tests and interventions: VAS Pain will be assessed for all subjects using a 100mm VAS, both at rest and in their most provocative position. This will be administered both pre and post manipulation as the measure of change. Subject ratings of pain on the VAS will be measured by a blinded examiner who will not be aware of CTDT group assignment. Cervico-thoracic differentiation test (CTDT) Subjects will be tested via CTDT as described by Evjenth \& Gloeck. The test reportedly differentiates between pain originating in the cervical and upper thoracic regions. The test will be performed with subjects sitting in erect posture as described by Dunleavy and Goldberg, with an investigator facing them. Subjects will be asked to move through full cervical range of motion (ROM) in each direction (flexion, extension, right rotation, left rotation) to find their most painful direction. They will then be asked to move into their most painful cervical motion until their pain comes on. The investigator will note this position and it will be considered to be the motion of interest for the outcome analysis, and the subject will complete a VAS for their pain experienced while in this position. The subject will then return the head/neck to the resting posture; the thoracic spine will be positioned opposite the provocative cervical direction, with the participant asked to repeat the painful cervical motion while the examiner holds the subject in the new thoracic position. This position of the cervical spine will again be noted by the investigator, and the investigator will ask the subject: "Are your symptoms 'better', 'worse', or 'the same'"? A positive test includes a response of 'better', as well as 'the same' when a significant improvement in cervical motion (operationally defined as more than 10 degrees) is obtained. A negative test includes a response of 'worse', as well as 'the same' when a significant improvement in cervical motion is not observed. Cervical unloading The cervical unloading test will be performed as described by Kaltenborn. This test involves unloading of the cervical and upper thoracic vertebrae. The test will be performed with subjects sitting in erect posture with an investigator standing behind them. If the subjects have pain in the neutral cervical position, cervical unloading will be performed by the investigator in the neutral posture. However, if the subjects do not have pain in this neutral cervical position, cervical unloading will be performed with the cervical spine in the position identified as most provocative. In both scenarios, cervical unloading consists of the investigator gently lifting upward on the subject's head to unload his/her cervical spine. The subject will then be asked to report if his/her pain is 'better', 'worse', or 'the same'. A positive test is indicated if the subject reports 'better', while a negative test is indicated if the subject reports 'worse' or 'the same'. Thoracic unloading The thoracic unloading test will be performed as described by Kaltenborn. This test involves unloading of the thoracic vertebrae. The test will be performed with subjects sitting in erect posture with an investigator standing behind them. If the subjects have pain in the neutral position, thoracic unloading will be performed by the investigator in the neutral posture. However, if the subjects do not have pain in this neutral position, thoracic unloading will be performed with the cervical spine in the position identified as the concordant sign. In all scenarios, the subjects will be asked to cross their arms by putting their hands on their opposite upper arms. Thoracic unloading consists of the investigator gently lifting upward on the subject's elbows to unload his/her thoracic spine. The subject will then be asked to report if his/her pain is 'better', 'worse', or 'the same'. A positive test is indicated if the subject reports 'better', while a negative test is indicated if the subject reports 'worse' or 'the same'. Grade V thrust manipulation Supine grade V thrust manipulation will be performed as described by Cleland and Karas. The supine technique was selected as it has been shown to elicit a greater change in pain when compared to seated techniques, and is more readily applied to the upper thoracic region than prone techniques. First, the examiner will assess joint play of the segments of the upper thoracic spine. The most hypomobile (stiff) and/or painful segment between T1-T4 will be determined at the examiner's discretion, and this is the segment to which the examiner will apply the thrust. The manipulation will be performed with subjects lying supine. The examiner will use a 'pistol grip' to apply pressure on the articular pillars of the selected thoracic segment bilaterally in order to block the inferior vertebrae of the hypomobile segment. The subject will cross his/her arms over his/her chest to the point where his/her elbows are aligned. This will allow the examiner to apply pressure through both elbows in an anterior-to-posterior direction. The examiner will assure the direction of force is felt into his/her stabilizing hand. The subject will be asked to take a deep breath in and then exhale completely. A grade V thrust will be applied at the end of the subject's exhalation. If a cavitation is not heard or felt by either the subject or examiner, a second thrust will be performed. After completion of this manipulation, subjects will be returned to an erect sitting position and asked to complete a second VAS, both at rest and in the predetermined provocative position. A-prioi levels of significance will be established at the p=.05 level. Data will be analyzed quantitatively in aggregate form. Inter rater reliability will be assessed with the Kappa statistic. Groups (CTDT +,-) will be analyzed for baseline differences for age, gender, NDI score, direction of concordant sign, and chronicity. Tests of difference will be utilized to assess for statistically significant differences between the positive and negative CTDT response groups. This will include t-tests or non-parametric equivalent tests for within and between group change; selection will be based on normality of data/meeting parametric assumptions. Criterion referenced validity of the CTDT and regional unloading tests will be determined using responder analysis. The reference criterion will be pain relief meeting/not meeting the minimal clinically important difference (MCID) on the VAS following thoracic manipulation (the intervention purportedly indicated as the result of a positive CTDT). Subjects will be identified as responders/non-responders based on a change in VAS score in the concordant position, pre-post manipulation. The classification of responder will be determined by exceeding the established MCID of the 100mm VAS, reported to be change of at least 15mm or 20%. The statistical relationship between responder status and CTDT test response will be analyzed as dichotomous nominal variables, assessed via Fisher's Exact Test.