Abstract
The muscular characteristics of rugby players may make diagnosing the direction of shoulder instability and labral pathology challenging. This study aimed to assess the accuracy of clinical examination and specifically instability tests, in diagnosing the direction of shoulder instability in rugby players. One-hundred-and-forty rugby players, who had undergone a shoulder stabilization procedure, over a 55-month period, were included in this study. The mean age was 21.5 years with 137 males. Data collected included clinical examination and intraoperative findings. The two were compared to calculate the diagnostic accuracy of special tests for instability. The Anterior Apprehension Test had good sensitivity (82.7%), specificity (100%) and PPV (100%) but poor NPV (55.8%). All posterior instability tests demonstrated a sensitivity of over 85%, but all had a specificity of 25% or less. In 83.6% of cases the direction of instability was correctly identified from history and examination. Anterior instability was correctly diagnosed in 78.9% of cases and posterior in 100.0%. The poor NPV of the anterior apprehension test suggests that clinicians should be suspicious of anterior instability in rugby shoulders even in the light of negative examination findings. Positive posterior instability tests are highly suggestive of posterior instability in rugby players.
Introduction
Instability – epidemiology and mechanism
Rugby is a high impact collision sport that is associated with a significant risk of injury. Shoulder trauma resulting in instability is common, with a reported incidence of 2.9 cases of instability per 1000 player-hours. Such injuries result in an average of 81 days’ absence from the sport 1 with consequences for both player and club.
The mechanism of injury causing shoulder instability in rugby players has been studied previously by Crichton et al. 2 in a study that utilized video analysis of elite rugby players to categorize the mechanism of injury into four distinct entities; the “Tackler injury”, “Try-scorer injury”, “Direct Hit injury” and “Flexed Fall injury”. 3
The Tackler injury results from extension of an abducted arm during a tackle. The Try-Scorer injury involves hyperflexion of an outstretched arm whilst scoring. The Direct Hit Injury involves an impact to the arm whilst in a neutral or adducted position. The Flexed Fall injury sees the player fall elbow-first to the ground with both the elbow and shoulder joints flexed. 4
Whilst our understanding of shoulder instability, in rugby players, has developed in recent years, the clinical identification of the direction of shoulder instability remains challenging, in part due to their well-muscled physique. 3 that may enable them to compensate, masquerading the classic symptoms and signs normally associated with shoulder instability
No previous studies have investigated the diagnostic efficacy of clinical tests for shoulder instability in rugby players.
This study aims to address this gap in the literature and offer the reader information on the accuracy of diagnosing the direction of shoulder instability in rugby players.
Materials and methods
Data was collected prospectively, over a 55-month period, in a single specialist shoulder clinic. The inclusion criteria required patients to be competitive rugby union or rugby league players, both professional and amateur, that had undergone an examination under anesthesia, diagnostic arthroscopy and shoulder stabilization procedure as part of their care. Consequently, 140 patients were included in the present study. The procedures performed are summarized in Table 1.
Frequency of surgical procedures performed.
The mean age of the sample was 21.5 (range 14–42) years. The 140-shoulder sample consisted of 137 males and 3 females. The median time from initial appointment to surgery was 3 weeks with a range of 0 to 296 weeks.
The common clinical tests for anterior and posterior instability were documented as positive or negative in clinic. The senior clinician's opinion as to the direction of instability after physical examination was recorded as the clinical direction of instability. The clinical tests recorded included the anterior apprehension test with Jobe's relocation, the Wrightington Posterior Instability Test (WPIT, also known as the Modified O’Brien Test), 5 the Kim test 6 and the posterior impingement sign. 7 These particular tests were chosen, as they have previous reported high levels of sensitivity and/or specificity in athletic populations.
The direction(s) of instability identified during surgery was based on the pathological findings on arthroscopic examination. These included soft tissue and bony labral tears, capsular tears and Hill-Sachs or reverse Hill-Sachs lesions. The clinical and surgical directions of instability were compared.
The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of individual special tests for direction of shoulder instability were all calculated using 2 × 2 tables.
Results
On initial presentation 61.4% of shoulders were diagnosed with anterior shoulder instability (n = 86), and 38.6% with posterior instability (n = 54).
The direction of shoulder instability from clinical examination correlated with the surgical direction in 83.6% of all cases. Anterior shoulder instability was identified intraoperatively in 109 of all shoulders and posterior instability in 31 shoulders. Of the shoulders which had surgical evidence of isolated anterior instability, 78.9% were correctly diagnosed on clinical examination prior to surgery (n = 86). For those with surgical evidence of posterior instability, 100.0% (n = 31) were correctly diagnosed on clinical examination. The remaining 23 patients were initially diagnosed with posterior instability from examination, but went on to intraoperative findings consistent with anterior instability. A full breakdown of clinical versus surgical direction of instability is illustrated in Table 2.
Clinical versus intraoperative direction of shoulder instability.
Special tests
The sensitivity, specificity, PPV and NPV of the anterior apprehension test, WPIT, Kim test and the posterior impingement sign in the present cohort are presented in Table 3.
Diagnostic accuracy of tests for shoulder instability in rugby players (Sen - sensitivity, Spec - specificity, PPV – Positive Predictive Value, NPV – Negative Predictive Value).
The table shows that the anterior apprehension test was the most commonly performed special test, having been performed in 75.7% (n = 106) of all patients. The anterior apprehension test had excellent specificity (100%), sensitivity (82.7%) and positive predictive value (100%) but at 55.8% offered poor negative predictive value.
All tests for posterior instability had a sensitivity of over 85.0%, however none offered a specificity of greater than 25.0%. The WPIT result was recorded in 48 patients and had a sensitivity and negative predictive value of 100%. The PPV of the WPIT was 61.5% and specificity was 25.0%. The Kim test was recorded in 27 patients and offered little additional information with specificity of 18.2% and PPV of 64.0%. The Posterior Impingement Sign was the least frequently performed special test (n = 16) and had the poorest overall diagnostic efficacy with a sensitivity of 85.7%, specificity of 11.1%, PPV of 42.8% and NPV of 50.%.
Discussion
The results of this study would suggest that diagnosing the direction of shoulder instability in rugby players from physical examination findings is less challenging than previously suggested. Surgeons used special tests to correctly identify the direction of shoulder instability in 83.6% of cases when compared to intra-operative findings. The data presented above provides an insight into the unique physical examination findings of rugby shoulders, which will be discussed below.
Diagnosing anterior instability
Anterior instability proved more challenging to identify from examination findings than posterior instability. All patients with positive signs of anterior shoulder instability from special tests subsequently demonstrating isolated anterior instability at surgery. However, simply assessing patients with positive clinical examinations would be misleading. In total 109 patients in this study had evidence of isolated anterior instability at surgery and 78.9% were correctly identified at clinical examination. While those with examination findings of anterior instability were likely to go on to have anterior instability, 21.1% of patients who had surgical findings consistent with anterior instability were originally identified as having posterior instability. This finding highlights a significant proportion of rugby players in whom anterior instability is challenging to identify.
A potential explanation for this observation comes in the relatively poor negative predictive value of the anterior apprehension test in the study cohort. The anterior apprehension test is widely regarded as an accurate test for anterior glenohumeral joint instability. Hegedus et al.8,9 undertook a meta-analysis of studies focusing on special tests of the shoulder. They concluded that the anterior apprehension test had a mean specificity of 95.4% and sensitivity of 65.6%. Similar figures were presented by Jia et al. 10 who analyzed the accuracy of special tests in 1913 patients over thirteen years and demonstrated a specificity of 96.0% and sensitivity of 72.0% for the anterior apprehension test In the present study, the anterior apprehension test had a specificity of 100% and sensitivity of 82.7% in rugby players. These figures would suggest the anterior apprehension test is more accurate in rugby players than in non-rugby playing individuals. However, neither study reports positive or negative predictive values. The negative predictive value of the anterior apprehension test in the present study was only 55.8% which suggests that rugby players may be prone to false negative anterior apprehension signs.
The anterior apprehension test remains a useful tool for identifying anterior instability in rugby players because of its excellent specificity and positive predictive value. Clinicians can take confidence that a rugby player with a history suggestive of instability, who has a positive anterior apprehension test will likely go on to have anterior instability of the shoulder. However, the poor negative predictive value demonstrated in this study should urge clinicians to remain suspicious of anterior instability in rugby players when the history is suggestive, but they test negatively for anterior apprehension This study therefore has implications for all clinicians who assess the shoulders of rugby players.
Diagnosing posterior instability
Compared to anterior instability, posterior instability is regarded as difficult to identify in the general population from clinical examination.11–13 There is little consensus in the literature regarding which test is the most useful which somewhat explains why so many different tests have been described.10,14 In the preset study, all patients with intraoperative posterior instability were correctly identified at clinical examination. However, posterior instability was over-diagnosed, with only 57.4% of those identified as having posterior instability at clinical examination going on to have posterior instability at surgery. The remaining 42.6% had clinical examination findings consistent with posterior instability but went on to have anterior instability intraoperatively. This over-diagnosis is somewhat explained by the poor specificity of the WPIT and the Kim test in this cohort of rugby players.
The WPIT was originally described by Funk et al. 11 and subsequently validated by Owen et al. in a 2015 study of 74 patients. 5 The test involves placing the shoulder joint in 900 of flexion with an extended elbow joint, the arm is then adducted 10 to 15 degrees medial of the sagittal plane. The arm is internally rotated and then a downward force applied by the examiner – a positive result is indicated if weakness is observed such that the arm drops below 900 5. In the original study, the test had a sensitivity of 83% and specificity of 25%. No other studies have yet provided further validation of the WPIT. In the present study of rugby players, the WPIT was more sensitive (100.0%) and had the same specificity (25.0%) compared to the original study. The WPIT was designed with well-muscled athletes in mind, 11 which may go some way to explaining why the present study correlates with the original data. These results suggest that when posterior instability is present, the WPIT will be positive in 100% of rugby players therefore a negative WPIT makes posterior instability unlikely.
The Kim test offered little additional benefit to the WPIT in the study cohort. Despite having a previously reported sensitivity and specificity of 80% and 94% 6 in the present study the sensitivity was 100.0% and specificity only 18.8%. The considerable difference in specificity described should be interpreted with caution because there have been no other studies aside from the original by Kim et al. which have validated the Kim test Nevertheless, the Kim test shares the same issue as the WPIT in having poor specificity and a NPV of 53.3%.
Strengths and limitations
The two main strengths of this study are the sample size and the performance of all of examinations by a single examiner. The study included 140 patients all assessed within a single specialist shoulder centre. There does not appear to be any other study of this size which analyzes the assessment of the direction of shoulder instability in rugby players. As a comparison, the original Kim paper included 172 shoulders and the WPIT paper 74 patients.
Inter-examiner reliability is a well cited issue in all disciplines which rely on physical examination.15,16 A single senior shoulder surgeon examined all the patients in the study, effectively eliminating any issues with inter-examiner reliability. Whilst this provides confidence in the present study results, further studies are encouraged to examine whether similar results are replicated elsewhere. It is important to note that the senior authors extensive experience in examining rugby shoulders likely contributed to the high correlation between examination findings and surgical direction of instability. Rugby players may still offer less experienced clinicians a unique diagnostic challenge, which makes the results discussed above even more relevant.
A limitation of this study is the absence of a control group. The inclusion criteria of a shoulder stabilization procedure meant that all patients had some form of shoulder instability. Further studies which incorporate a control group would be welcomed. Finally, there is a plethora of special tests reported for identifying posterior instability of the shoulder. This study is limited in that it is only able to report on the accuracy of tests regularly performed by the senior author (LF).
Conclusion
This study provides clinicians with an overview of the clinical examination findings they can expect from rugby players suffering from different directions of instability of the shoulder joint. It has demonstrated the usefulness of the anterior apprehension test in confirming suspected anterior shoulder instability. However, an awareness of the poor negative predictive value should alert clinicians to remain suspicious of anterior instability in rugby players when the history suggests as such even if the anterior apprehension test is negative. This is relevant to the broad range of clinicians who have contact with regular rugby participants from community and team-based healthcare to specialist shoulder centres.
The challenge of diagnosing posterior shoulder instability was highlighted in this study. It was over-diagnosed, with only 57.4% of those with posterior instability on physical examination going on to have posterior instability at surgery. This was explained by the high sensitivity and low specificity of special tests for posterior instability. It also highlighted a significant discrepancy in the specificity of the Kim test in rugby players when compared to their original study.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
