Imaging modality for measuring the presence and extent of the labral lesions of the shoulder: a systematic review and meta-analysis

The glenoid labrum, composed of fibro-cartilage, is a ring or band structure that effectively increases the depth of the glenoid fossa [1]. Lesions of the glenoid labrum, occurring with glenohumeral instability, result in serious shoulder pain because of the destruction of free nerve endings located in the peripheral part of the glenoid labrum and the subacromial bursae [2, 3].

Based on their location and lesion features in imaging [4], disorders of the glenoid labrum have been broadly categorized as superior, posterior, inferior or anterior lesions [5]. Specifically, superior labral anterior-posterior tears (SLAP), initially described by Andrews et al. in 1985, have been an ongoing diagnostic challenge in the clinic [6]. Additionally, Bankart lesions are one kind of injury on the anteroinferior aspect of the glenoid labral complex, which are thought to predispose shoulders to recurrent dislocation [7, 8]. The integrity of the labrum and whether any bone has been avulsed or missing from the bony glenoid determined the different treatment strategies. For example, SLAP lesions were usually managed by arthroscopy at present [9] while detached Labra is often treated by open surgical repair [10]. Because of the serious pain associated with these injuries and the limitations they place on participation in high-level activities, the need to evaluate accuracy, efficiency, and economics of diagnostic tests for labral damage is increasingly important [9]. In addition, reorganization of the integrity of the glenoid labrum is an essential factor for clinicians to consider when making treatment decisions (i.e., to use conservative vs. surgical strategies) [10]. Medical imaging technologies not only provide rich and useful information to support findings from the medical history and physical examination but also demonstrate the pathoanatomy of shoulder dysfunction of the shoulder [11]. Therefore, a suitable choice of imaging technique could help to establish an appropriate treatment strategy.

Many imaging methods, including arthrography, computed tomography arthrography (CTA), magnetic resonance imaging (MRI), direct MR arthrography (D-MRA) and indirect MR arthrography (I-MRA) have been used to image the glenoid labrum as well as the associated structures of the capsular mechanism [12]. Shoulder MRI is becoming quite popular as a screening examination for the detection of labral abnormalities [13]. However, intra-articular structures of the shoulder are not well imaged by MRI when insufficient fluid is present to outline the glenoid contour [14].

MRA of shoulders mainly included indirect shoulder magnetic resonance arthrography (I-MRA) and direct shoulder magnetic resonance arthrography (D-MRA). D-MRA, which involves intra-articular administration of contrast agent, has become an established imaging modality for assessing different types of labral lesions [14]. Additionally, an alternative and less invasive technique, I-MRA, were intravenously administered contrast enhances the joint space and indirectly produces an arthrographic effect [15]. MRA is considered to have higher accuracy than MRI in the detection of glenoid labral tears, but it is invasive [15]. CTA does not have advantages in the evaluation of soft tissue injuries such as labral damage over MRI and MRA; however, it was proven to have much higher diagnostic accuracy for detecting bony defects of the glenoid [16]. With the development of MRI technologies, the diagnostic sensitivity and specificity of 3-Tesla (T) MRI versus MRA for assessing labral abnormalities is controversial to a certain extent [17]. A previous meta-analysis [18] suggested that MRA had greater diagnostic accuracy than MRI for the overall detection of glenoid labral lesions. The opposite result was obtained when diagnosing anterior glenoid labral lesions. Another meta-analysis demonstrated that MRA was superior to MRI for the detection of SLAP lesions [19]. A recent meta-analysis from 2018, involving 10 studies, revealed that 3.0 T MRA improved sensitivity for the diagnosis of anterior and posterior labral tears, but reduced specificity in the diagnosis of SLAP tears [20].

Recently, multiple high-quality studies [14, 21‐23] were published, most of which used relatively high resolution for CTA and relatively high-field strength magnets and multidimensional imaging for MRI and MRA. Moreover, no studies have compared the diagnostic performance of MRI, D-MRA, I-MRA and CTA using side-by-side analysis in a single study for the detection of labral lesions. Therefore, an updated meta-analysis is warranted to determine if the new data and improved technology have had an impact on the diagnostic accuracy of a given pool of data.

The primary objective of this study was to perform a meta-analysis on the diagnostic accuracy of MRI, MRA, CTA and CT in the assessment of glenoid labral lesions. The second objective was to compare the diagnostic accuracy of MRI and MRA for detecting different types of labral lesions, such as anterior, posterior or superior lesions. The third objective was to evaluate the effect of magnet strength on the diagnostic accuracy of MRI and MRA for glenoid labral lesions.

This meta-analysis was conducted based on the guidelines of the Preferred Reporting Items for a Systematic Review and Meta-analysis of Diagnostic Test Accuracy Studies (PRISMA-DTA) [24] statement. Patient informed consent and committee approval were not required for this study due to the use of published data.

Lesions of the glenoid labrum are critical factors causing shoulder pain and disability [40, 41], which can seriously affect the quality of patients’ lives if without suitable diagnostic techniques and proper treatment strategies. The decision to perform arthroscopy or open surgery [42], as the ultimate treatment option of labral disorders, depends not only on the patients’ clinical histories and physical examinations but also on their imaging results [37], and accurate positioning of the tears undergoing surgery are largely affected by the pre-operative imaging reports [43]. Diagnostic accuracy and effective use of imaging technology are the main concerns of clinicians and patients. Therefore, it is essential to compare the accuracy of MRA (I-MRA and D-MRA), MRI and CTA for labral diagnosis and to analyse their advantages and disadvantages under various specific conditions.

It has long been important to address the roles of MRI and MRA as imaging tools for detecting pathologic labral lesions [15, 44]. While there is a large body of literature suggesting that MRA is superior to conventional MRI for the diagnosis of labral lesions (even at 3-T) [20, 43], our pooled results considering the two techniques suggest that MRA enhances the sensitivity of the detection of labral disorders, while it is only marginally superior to MRI in terms of specificity. Although it is undeniable that MRA maximizes anatomic resolution and diagnostic confidence, the injection of contrast material may provoke several inevitable problems, such as invasion [45], ionizing radiation [46], adverse reactions and additional radiologist time and expertise [47]. Therefore, with regard to the option of MRI vs. MRA for detecting labral pathologic lesions, it seems that patient presentation is an often-neglected but crucial consideration in the choice of imaging tool [48, 49]. Patients with acute symptoms or unstable, severe, pathologic tears are more likely to have intrinsic image contrast in the form of effusion or soft-tissue changes that allow diagnosis and characterization without an invasive procedure [50, 51]. In contrast, those with chronic symptoms or a pathologic abnormality that is suspected to be more subtle on the basis of the clinical assessment more often require MRA [43].

MRA can be used directly with intra-articular contrast agent injection (D-MRA) or indirectly with intravenous (i.v.) contrast agent injection (I-MRA) [37, 47]. In this meta-analysis, we evaluated the diagnostic accuracy of labral lesions using I-MRA compared to D-MRA and found that D-MRA is superior to I-MRA. One of the greatest strengths of D-MRA lies in the benefit conferred by joint distension [52]. This distinguishes the redundant capsule from the adjacent labral tissue and allows further passage of contrast agents into the labral substance in the case of unstable labral lesions, as well as between the labrum and the glenoid in the case of labral detachment [37]. However, an obvious disadvantage of shoulder I-MRA is the absence of controlled joint capsule fluid distension, which many researchers feel is essential for improving the diagnostic accuracy of subtle detachments of the glenoid labrum [37]. This concern led to early recommendations that I-MRA should not be used for the evaluation of labral tears although it has been considered an alternative, less invasive method.

The field intensity of MRI may have an important effect on the diagnostic accuracy of diseases [36]. Therefore, we wanted to determine whether 3-T MRA provided more useful information to clinicians than conventional 3-T MRI. Our subgroup analysis based on field intensity showed that 3-T MRA had an increased sensitivity and specificity compared with 3-T MRI, which is consistent with our pooled results of 1.5-T MRA vs. 1.5-T MRI. Other subgroup analyses based on the location of labral lesions obtained the similar results. Even though MRA has an overwhelming advantage, we do not suggest that MRA should be performed on the shoulders of all patients to increase the accuracy of diagnosis. In the actual clinical work, the doctors make the diagnosis in combination with the patients’ medical history and various physical examinations, which is not as blind as research work, prompting the acknowledgement that MRA should not be a general recommendation in the diagnosis of acute labral lesions.

With regard to detecting the overall presence of labral tears, CTA had obviously less sensitivity and specificity compared with MRI and MRA in our meta-analysis. CTA was frequently used to evaluate the extent of soft and osseous tissue abnormalities, and the ability of CTA to show anteroinferior labral lesions as well as SLAP lesions has been established in a previous study [39]. However, the limited spatial resolution and soft-tissue contrast in reformatted scans from conventional CTA have led to its replacement by MRI and MRA imaging in the detection of labral lesions [28]. MRI and MRA provide superior soft-tissue contrast; therefore, no-detached labral tears can be seen as a signal extending from within the labrum to its surface [29]. However, with conventional CTA, a morphologic abnormality must be present on the surface of the labrum [29]. If there is a lack of surface contour abnormalities, false-negative results often occur when using conventional CTA. A retrospective study involving 83 patients revealed that labral damage was found in nearly all cases of recurrent anterior shoulder instability and proved that conventional CT was more important for pre-operative planning because of its detection of glenoid defects due to open or arthroscopic repair techniques that had been performed, mainly according to the bony integrity of the glenoid [23]. Therefore, CT should be a necessary supplemental imaging technique when there is highly suspected glenoid bone damage.

Several limitations exist in this meta-analysis. We assessed only the diagnostic value of the imaging modalities alone. The diagnostic performance of physical tests was not evaluated. Two or three methods, such as MRI + physical tests and MRA + physical tests, were also not analysed side-by-side. Several subgroup analyses were implemented based on insufficient data, which make certain results unreliable. In addition, the safety, cost-effectiveness, and application of these imaging techniques in clinical practice should be assessed systematically.

This meta-analysis of diagnostic tests, which included 14 studies involving 1216 patients with labral lesions, revealed that MRA had the highest sensitivity and specificity compared with those of MRI and CTA. However, MRA was just suggested for use in patients with chronic shoulder symptoms or a pathologic abnormality. MRI is by far the first choice recommendation for imaging modality for the detection of acute labral lesions. CT should be a necessary supplemental imaging technique when there is highly suspected glenoid bone damage.