Research Review by Dr. Rob Rodine©

Date:

Nov 2008

Study Title:

Diagnostic utility of clinical tests for SLAP lesions: a systematic literature review

Authors:

Powell J, Huijbregts PA, Jensen R

Authors’ Affiliations: University of St. Augustine for Health Sciences, Florida, USA

Publication Information:

The Journal of Manual & Manipulative Therapy 2008; 16 (3): E58-E79.

Summary:

The glenoid labrum is a fibrocartilaginous ring that serves to deepen the glenoid and increase shoulder stability. It also serves as an anchoring point for other anatomical structures such as the long head of the biceps.

The prevalence of glenoid labrum injuries seems highly dependent on the population being studied, and is reported to be between 6-26% of those undergoing shoulder arthroscopy.

Injury mechanisms for labral pathology are variable, and can include:
  • falling onto an outstretched arm
  • excessive traction or tension through the long head of the biceps
  • hyperabduction and external rotation positions which result in a ‘peel-back’ mechanism of the long biceps tendon
A Superior Labrum Anterior-Posterior (SLAP) lesion is described as a specific type of labral pathology affecting the superior portion of the labrum. This injury has been classified into four categories:
  1. Type I – degenerative fraying with an intact biceps insertion
  2. Type II – biceps detachment
  3. Type III – bucket-handle tearing within the labrum with an intact biceps insertion
  4. Type IV – bucket-handle tearing with an intra-substance tear of the biceps
Three subtype classifications of SLAP type II have been labeled, however this is advanced knowledge which is not clinically useful for most manual therapists.

Study Methods:

Pubmed, Proquest and CINAHL were searched using search terms relating to SLAP injuries and the glenoid labrum. The search was limited to studies published between 1985 and June 2007. The reference lists of retrieved articles were hand searched for further citations.

The search strategy yielded 977 articles. Hand searching yielded no further results. A total of 17 studies met the inclusion criteria. Nineteen clinical tests were studied in their accuracy for diagnosing SLAP lesions.

Methodological quality was assessed via the QUADAS 14 item tool, where a score of 10 or less indicates poor design conditions. Previous literature has indicated an inter-rater reliability score of 90% using this scoring item, however scores have ranged from 50-100%.

QUADAS scores for the 17 studies included in this review ranged from 3-10, where only three studies scored 10. Therefore, in an effort to generate the best evidence synthesis, only three articles were reviewed.

This review of clinical testing stands out compared previous reviews as results are accompanied by methodological assessment via the QUADAS criterion. This methodological review resulted in several important considerations.

SLAP lesion as a pathology requiring surgery:
  • SLAP lesions are complex disorders that frequently involve rotator cuff tears, loss of integrity to the long biceps tendon and associated instability.
  • This array of associated injuries will play a role in response to both provocative testing as well as response to conservative therapy.
  • Published literature frequently reiterates that all SLAP injuries require surgical correction. This is based on the extent of anatomical disruption as well as the avascular nature of the labrum which contributes to a poor response to conservative therapy. Contrary to this however, some authors point out that Type I lesions may be more suited to conservative therapy versus surgical intervention.
SLAP lesion as a diagnosis guiding conservative management
  • Currently, literature regarding SLAP injuries follows a patho-anatomical abnormality construct, rather than a mechanism-based classification system that would aid the manual therapist.
  • A mechanism-based system would demonstrate aspects of aberrant joint mechanics to a clinician. As the consistent and specific aspects of joint dysfunction remain unknown in SLAP lesions, clinical prediction rules and treatment guidelines for conservative therapy are currently unavailable.
Definition of a SLAP lesion:
  • It is important to note that the operational definition for a SLAP lesion varies across the literature. As a result, the diagnostic utility of clinical testing is questioned when compared across studies. Consensus should be reached.
Imaging or arthroscopy
  • Arthroscopic evaluation is considered the gold standard for SLAP diagnosis.
  • Evaluations of clinical testing has been performed however using special imaging such as MRI, MRA, CT or ultrasound as the reference test.
  • One study included in this review evaluated MRA as the reference standard. Imaging methods such as this can result in false-positive diagnosis due to the presence of congenital malformation. While imaging such as MRA frequently demonstrates a high correlation to arthroscopic findings, it is never perfect.
  • Reference testing standards should always be taken into consideration.
Notes on studies included in this review:
  • The arthroscopic procedure performed in approximately half of the described studies was not detailed. As a result, procedures could not be replicated by other surgical teams. Therefore, these procedures and results cannot be accurately compared across populations.
  • The importance of this is acknowledged when one considers that one procedure may identify SLAP lesions more accurately than another.
  • Only two studies appropriately blinded surgeons to the results of clinical tests.
  • Time delay between clinical and reference test: The time lapse between clinical testing and reference standards (arthoscopy) varied across the literature, ranging from 24 hours to 48 weeks. It is highly probable that the SLAP lesion will change over time. Therefore, this time delay can result in an invalid correlation between testing procedures.
  • Spectrum bias: Populations examined typically were presented to orthopaedic or sports medicine clinics. This study setting is not representative of clinical practice within a manual therapy environment. Also, some of the studies carried out in these environments have reported prevalence rates as high as 73%. Exaggerated prevalence rates due to study setting lead to lowered opportunity to detect false positive and true negatives results, which reduces validity.
  • Operational definition of test performance and interpretation: Six of the eight studies evaluating the active compression test used an identical definition. Other testing performance definitions and interpretations were not as consistent across studies. This should be taken into consideration when interpreting the results.
Pertinent results for the various clinical tests included in this review are as follows:

Anterior Apprehension Maneuver (evaluated in 2 studies)
  • Sensitivity ranged from 0.04 – 0.83 whereas specificity ranged from 0.40 – 0.87.
  • Accuracy was rated at 0.59 in one study only.
Active Compression (also called O’Brien’s test) (evaluated in 7 studies)
  • Sensitivity ranged from 0.47 – 1.0 whereas specificity ranged from 0.111 – 0.985.
  • Accuracy ranged from 0.54 – 0.988.
Anterior Slide (evaluated in 4 studies)
  • Sensitivity ranged from 0.05 – 0.784 whereas specificity ranged from 0.815 – 0.93.
  • Accuracy ranged from 0.54 – 0.858.
Biceps Groove Tenderness (evaluated in 2 studies)
  • Sensitivity ranged from 0.25 to 0.44 whereas specificity ranged from 0.40-0.80.
  • One study reported the accuracy to be 0.56, whereas the second study did not report an accuracy rating.
Biceps Load test I (evaluated in 1 study)
  • Sensitivity was rated as 0.909 and specificity at 0.969.
  • Accuracy was rated at 0.96.
Biceps Load test II (evaluated in 1 study)
  • Sensitivity was rated as 0.897 and specificity at 0.966.
  • Accuracy was rated at 0.944.
Clunk test (evaluated in 1 study):
  • Sensitivity was rated as 0.44 and specificity at 0.68.
  • Accuracy was rated at 0.57.
Compression Rotation test (evaluated in 2 studies):
  • Sensitivity ranged from 0.24 – 0.25 whereas specificity ranged from 0.76 – 1.0.
  • Accuracy ranged from 0.63 – 0.71.
Crank test (evaluated in 6 studies):
  • Sensitivity ranged from 0.125 – 0.906 whereas specificity ranged from 0.56 – 0.933.
  • Accuracy ranged from 0.66 – 0.919.
Forced Abduction test (evaluated in 1 study):
  • Sensitivity and specificity both equated to 0.67.
  • Accuracy was rated at 0.67.
Hawkins Impingement sign (evaluated in 2 studies):
  • Sensitivity ranged from 0.50 – 0.675 whereas specificity ranged from 0.304 – 0.67.
  • Accuracy was reported at 0.59 in one study only.
Neer Impingement sign (evaluated in 2 studies):
  • Sensitivity ranged from 0.33 – 0.50 whereas specificity ranged from 0.522 – 0.60.
  • Accuracy was reported at 0.48 in one study only.
Jobe Relocation test (evaluated in 4 studies):
  • Sensitivity ranged from 0.44 – 1.0 whereas specificity ranged from 0.4 – 0.87.
  • Accuracy was reported at 0.56 in one study only.
Pain Provocation test (evaluated in 2 studies):
  • Sensitivity ranged from 0.15 – 1.0 whereas specificity ranged from 0.90 – 0.902.
  • Accuracy ranged from 0.97 – 1.0.
Passive Compression test (evaluated in 1 study):
  • Sensitivity was rated as 0.818 and specificity at 0.857.
  • Accuracy was rated at 0.836.
Resisted Supination External Rotation test (evaluated in 1 study):
  • Sensitivity was rated as 0.828 and specificity at 0.818.
  • Accuracy was rated at 0.825.
SLAPrehension test (evaluated in 1 study):
  • Sensitivity was rated as 0.819
  • Specificity and accuracy were not reported.
Speed’s test (evaluated in 4 studies):
  • Sensitivity ranged from 0.04 – 0.32 whereas specificity ranged from 0.674 – 1.0.
  • Accuracy ranged from 0.56 – 0.57.
Yergason’s test (evaluated in 4 studies):
  • Sensitivity ranged from 0.09 – 0.43 whereas specificity ranged from 0.79 – 1.0.
  • Accuracy ranged from 0.61 – 0.63.
Multiple Testing Regimes (evaluated in 3 studies):
  • Using the results of multiple tests is more applicable to clinical practice settings.
  • Using positive results from the anterior apprehension, relocation, load & shift, sulcus and crank tests provided a sensitivity of 0.90, a specificity of 0.85 and an accuracy of 0.89.
  • Using two positive results from the Jobe and active compression test resulted in a sensitivity of 0.41 and a specificity of 0.91. Accuracy was not reported.
  • Using two positive results from the Jobe and apprehension tests or active compression and anterior apprehension test produced a sensitivities of 0.38 and produced respective specificities of 0.93 and 0.82.
  • Three positive results from the Jobe, active compression and anterior apprehension tests resulted in a specificity of 0.91, however a low sensitivity of 0.34.

Conclusions & Practical Application:

It is important to note that labral pathology rarely occurs in isolation. Therefore, one should expect positive clinical testing for impingement and rotator cuff pathology. In addition, it is important to consider clinical tests in conjunction, rather than in isolation, which more closely resembles the habits of clinical practice. Regimes such as this often display high specificity, while the sensitivities are often much lower.

The authors conclude that the best evidence indicates that a negative finding from the passive compression test provides the most definitive result that a SLAP lesion is absent and that a positive result provides the most definitive result that a SLAP lesion is present. This finding is based on the best evidence summary, whereby only three studies showed sufficient methodological quality to be considered. Of all tests considered in the three studies evaluated, the passive compression test demonstrated the greatest diagnostic accuracy.

The author’s earlier statements however should be considered: ‘a clinical test that shows sufficient diagnostic utility in more than one study is more relevant for clinical diagnosis that a test that has only shown sufficient diagnostic utility in one study'.