Neck-shoulder pain – is it possible to use commonly used provocative tests and functional scores for evaluation? Large scale prospective cohort study


 Background. According to International Association for the Study of Pain (IASP), the prevalence of neck-shoulder pain in general population is the fourth leading cause of disability. Neck-shoulder pain is known to be caused by different conditions. The aim of this study was to determine and statistically justify the correlation between shoulder girdle injuries and development of neck-shoulder pain syndrome. Methods. The study enrolled 1402 patients with hospital-treated shoulder girdle injuries (with consequences of upper limb trauma). Detailed protocol and questionnaire for clinical examination after discharge from the hospital were conducted. The prevalence of neck-shoulder pain among patients was calculated. The diagnostic value (sensitivity and specificity) of tests used in the study was assessed. Results. The prevalence of neck-shoulder pain was 66,9%. According to questionnaire results (VAS, ODI, SF-36, UCLA, SPADI, DASH), the tendency of increasing pain intensity, progressive upper extremity functional loss and, therefore, quality of life impairment in patients with shoulder girdle injury was noticed. Conclusion. 3 types of neck-shoulder pain syndrome in patients with shoulder girdle injury can be determined: articular, vertebrogenic and mixed (myofascial), which correlate with anatomical localization of the injury.


Background
According to the International Association for the Study of Pain (IASP), the prevalence of neck-shoulder pain (NSP) in the general population varies between 30% and 50% with higher prevalence in middle-aged women [3,7]. It is also considered to be the fourth leading cause of disability [2], in 15-30% of the working population followed by postural biomechanics disorder and myofascial syndrome [18]. The annual incidence of NSP in the working-age population ranges from 20-40%, while about two-thirds of people experience pain episodes at least once across their lifetime [11,5]. Also it can be demonstrated a decrease in pain assessment and disability scales rates by 53,8% and 48,4%, respectively [12]. NSP can be caused by different conditions like pathological changes in the spine, vertebrogenic algic syndrome, muscle and ligament disorders, pathology of thoracic viscera, and psychological disorders [6,8]. The most common non-speci c causes of pain are cervical spine osteochondrosis, cervical spondylosis, cervical facet joint blocks [15,16]. The pain occurs after muscle overstretching, prolonged poor posture with musculoskeletal distortion of the neck, and upper back and facet arthritis.
There is a vast amount of publications dedicated to cervical spine injury-related NSP. At the same time, there is a lack of studies investigating other etiologies [5,10,17,19]. Above these are myofascial pain syndrome, diffuse idiopathic skeletal hyperostosis, connective tissue diseases, referred pain (e.g. esophagitis, thyroiditis), bromyalgia, psychogenic pain, and shoulder girdle injury aftermath [9,13].
According to J. Dubousset's cone of economy, the vertebral column maintains standing posture and balance in 3 planes (coronal, axial, and sagittal) without external support providing minimal muscular effort4. Therefore, functional changes of the cervical spine and shoulder girdle are acting in concert to keep this balance. This means that shoulder girdle injuries can underline NSP. Much interest deserves the group of disorders, which affect the sagittal balance without structural changes (contracture of the shoulder joint and shoulder instability). Thus, the presence of structural pathology of the cervical spine in patients after shoulder girdle injury remains unclear [14].
Following this rationale, understanding of clinical picture of NSP in complex with shoulder girdle injury and anatomical localization of the injury appear to be quite promising in creation of treatment strategy.
The study's objective is to investigate and statistically justify the correlation between shoulder girdle injury (and its treatment) and NSP in patients without previous pathology.

Assessment methods
The patients were divided into 3 study groups according to localization of the injury.
Since in our research we examined patients who had late posttraumatic/postoperative complications, special scales, survey questionnaires, and physical examination were conducted. Methods were performed at three points of the study: 1 year, 1,5 years and 2 years after discharge from the hospital.
On the rst point (1 year after discharge) physical examination of all the patients and questionnaire assessments were performed. In some cases, the questionnaires were conducted via mobile phone with the subsequent visit for physical examination. At 6-8 months after the rst visit, the next point of the research was performed. The questionnaires used in point 1 were administered. The patients who demonstrated a decrease in questionnaire scales rates were invited for clinical re-examination 1,7 − 2,2 years after the injury.
Pain intensity was evaluated using the following scales: VAS (visual analogue scale), UCLA (University of California Los Angeles) and SPADI (The Shoulder Pain and Disability Index).
Humeral articulation function was evaluated using following scales: UCLA (University of California Los Angeles), SPADI (The Shoulder Pain and Disability Index) and DASH (Disabilities of the Arm, Shoulder and Hand).
Shoulder function was evaluated using following scales: UCLA (University of California Los Angeles), SPADI (The Shoulder Pain and Disability Index) and DASH (Disabilities of the Arm, Shoulder and Hand).
ODI (Oswestry Disability Index) and SF-36 (Short Form-36 Health Status Survey) were used for disability and life quality assessment, respectively.
We set commonly used clinical tests to conduct the protocol for shoulder joint and cervical spine investigation. To investigate the shoulder girdle a number of tests was performed: stress test, range of motion test, test for shoulder bursitis, test for shoulder instability, test for rotator cuff injury, tests for acromioclavicular joint and long head of the biceps tendon pathology, etc. Cervical spine examination included range of motion, exibility, pain localization and sagittal balance assessment. As a result, the protocol consisted of 15 tests for examination of the shoulder girdle and 10 tests for assessment of the cervical spine: Test (T) 1

Statistical data processing
For the statistical analysis of results, Statistica 10.0 for Windows (StatSoft Inс., USA) was used.
Quantitative variables were described using standard variation statistical methods, for which the arithmetic mean (М), standard deviation (d), 25th and 75th percentiles, and median patient number were calculated. The average values were presented as М ± d. The qualitative variables were described as absolute and relative frequency ratios (percentages). Differences were considered signi cant at р < 0.05. To evaluate results, the statistical analysis methods used included: Student's t-test and nonparametric tests for variable samplings inconsistent with the normal distribution law (MannWhitney test, the Wilcoxon test, the Chi-square test). Different activity scores were compared with the use of contingency table analysis. Raw data from questionnaires and physical examinations was analyzed using JanGO Patients System software. To evaluate diagnostic e ciency, we used ROC curve analysis by calculation area under ROC curve (AUC). Optimal threshold was chosen to maximize the sum of sensitivity and speci city.

Results
The questionnaire and physical examination of 1402 patients after shoulder girdle injury showed the total prevalence of NSP -938 patients (66,9%). The questionnaires assessment results after 1 and 2 years after the shoulder girdle injury were shown in Tables 2 and 3, respectively.  Mean ODI rate after 1 and 2 years after the injury was 29,90 ± 3,85 and 44,35 ± 2,38, respectively. SF-36 also showed negative dynamics: quality of life impairment and disability rate increase were noticed in all 3 study groups.
The optimal method of medical diagnostics should demonstrate high speci city as well as sensitivity. Unfortunately, such accuracy cannot be always performed by all the tests, so ROC curve creation was necessary to compute the prognostic value.
Understandably, almost all highly sensitive clinical tests demonstrate overdiagnosis, however this fact minimizes the risk of unnoticed disease.
Tests Т12 and Т22 showed high sensitivity (> 75%) with relatively low speci city (≤ 50%), so this combination of tests was considered the most accurate for making a diagnosis. The area under the curve (AUC) was 0,967, which indicated high diagnostic performance of these tests.
During the clinical test analysis non-normal distribution of symptoms frequencies was found. In 1st group-the prevalence of articular component of NSP was noticed. In 2nd group -cervical syndromes as well as shoulder syndromes were observed, i.e. mixed (myofascial) type of NSP. In 3rd group, which comprised patients with injuries of subacromial (suprahumeral) region the vertebrogenic NSP predominated.

Discussion
The purpose of this study was to investigate and statistically justify the correlation between shoulder girdle injury (and its treatment) and NSP in patients without previous pathology. Shoulder and neck pain are common musculoskeletal complaints for which identi cation of the causes may be challenging [20]. Hattrup et al proposed the presence of biomechanical connections which can mediated by irritation or dysfunction of the suprascapular or axillary nerves, both of which carry bers from the C5 and C6 nerve roots [21]. Zhang et al conducted a retrospective cohort study by reviewing the diagnoses in medical care records. They found that patients with a diagnosis of rotator cuff pathology were more likely to have a Cspine problem than an L-spine problem [22].The frequent revealing of NSP in patients with various consequences of shoulder injuries and the in uence on the remote functional results of treatment give the present research the modern scienti c urgency. Even though at present it is still unclear whether there are structural disorders of the cervical spine in patients with shoulder injuries, we consider the development of clinical and radiological methods of diagnostics of various types of NSP in this group of patients to be an important task.
The creation of a clinical classi cation of NSP based on the clinical signs is also an important and integral task in the way of treatment of patients with shoulder injuries consequences.
To carry out these tasks, it seems important to de ne and statistically substantiate the connection between shoulder injuries and the development of NSP in patients without vertebrogenous pathology. Then to analyze the diagnostic value (sensitivity and speci city) of the most used and well known cervical and shoulder clinical tests and stress tests. Based on the data and the statistical distribution of results, identify the most used clinical types of NSP, and create a working classi cation of it. All patients included in this study were clinically tested with the same set of stress functional tests and symptoms.
These tests are widely known and common in the daily practice of physicians for the clinical diagnosis of cervical and shoulder injuries. However, the justi cation for these tests and their combinations in patients with shoulder injury consequences for the diagnosis of NSP has not been done before. To determine the possibility and eligibility of using standard shoulder and cervical tests to diagnose NSP it is necessary to provide statistical substantiation of their clinical signi cance.
Wei-Ting Wu et al reported in patients with shoulder and neck pain, no association existed between rotator cuff tendon tears and cervical foraminal stenosis at the C5 and C6 levels. Aging seemed to be the main contributor to this coexistence of both disorders. When patients present with undifferentiated shoulder and neck pain, physicians should obtain a detailed history and perform physical examination and imaging studies for both the neck and shoulder regions [23].
Our uni ed protocol for the clinical assessment of the shoulder and the cervical spine includes stress guideline tests, as well as tests to clarify the range of motion and the presence of in ammation, tests to detect instability of the shoulder joint, rotator cuff pathology, the pathology of the acromioclavicular joint and tendon of the long head biceps.
In our study the total prevalence of NSP in patients was 66,9%. NSP developed more frequently after clavicle fracture, acromioclavicular joint, and sternoclavicular joint injury. There was a tendency of increasing pain intensity, progressive upper extremity functional loss, and, therefore, quality of life impairment in patients with NSP after shoulder girdle injury.

Conclusion
As a result of statistical processing of the obtained results, the most sensitive and speci c tests were singled out from the clinical examination protocol. And it is possible to make suggestion that they can be used in NSP identi cation. A combination of overly sensitive tests was found, indicating the presence of CHF in patients with a high probability of shoulder injuries. Identi cation of positive tests, which demonstrated the highest sensitivity and speci city, as well as the presence of a combination of positive tests T12 and T22 make it possible to consider the clinical diagnosis of splinters the most reliable in these patients.
By clinical picture analysis, 3 types of NSP in patients with shoulder girdle injury can be determined: articular, vertebrogenic, and mixed (myofascial). The clinical picture of NSP in patients with shoulder girdle injury correlates with anatomical localization of the injury.

Consent to publish
The participants in this study provided their written informed consent to undergo the study and to have their data used in the study.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding authors on reasonable request.