Medial scapular body (MSB) Goutallier classification: MRI-based reliability and validity of evaluation of the Goutallier classification for grading fatty infiltration of the rotator cuff.

ABSTRACT

BACKGROUND: The degree of atrophy and fatty infiltration of rotator cuff muscle belly is a key predictor for cuff repairability. Traditionally, Goutallier grading of fatty infiltration is assessed at sagittal scapular Y-view. Massive rotator cuff tears are associated with tendon retraction and medial retraction of cuff musculature, resulting in medialization of the muscle bulk. Thus, standard Y-view can misrepresent the region of interest and may misguide clinicians when assessing repairability. It is hypothesized that by assessing the muscle belly with multiple medial sagittal magnetic resonance imaging (MRI) sections at the medial scapular body, the Medial Scapular Body-Goutallier Classification (MSB-GC) will improve reliability and repeatability, giving a more representative approximation to the degree of fatty infiltration, as compared with the original Y-view. METHODS: Fatty infiltration of the rotator cuff muscles were classified based on the Goutallier grade (0-4) at 3 defined sections section 1, original Y-view; section 2, level of suprascapular notch; and section 3, 3 cm medial to the suprascapular notch on MRI scans. Six subspecialist fellowship-trained shoulder surgeons and 3 musculoskeletal radiologists independently evaluated deidentified MRI scans of included patients. RESULTS: Of 80 scans, 78% (n = 62) were massive cuff tears involving the supraspinatus, infraspinatus, and subscapularis tendons. Interobserver reliability (consistency between observers) for Goutallier grade was excellent for all 3 predefined sections (range 0.87-0.95). Intraobserver reliability (repeatability) for Goutallier grade was excellent for all 3 sections and 4 rotator cuff muscles (range 0.83-0.97). There was a moderate to strong positive correlation of Goutallier grades between sections 1 and 3 and between sections 2 and 3 and these were statistically significant (P < .001). There was a reduction in the severity of fatty infiltration on the Goutallier classification from sections 1 to 3 across all muscles 42.5% of both supraspinatus and infraspinatus were downgraded by 1, 20% of supraspinatus and 3.8% of infraspinatus were downgraded by 2, and 2.5% of supraspinatus were downgraded by 3. CONCLUSION: This study found that applying the Goutallier classification to more medial MRI sections (MSB-GC) resulted in assignment of lower grades for all rotator cuff muscles. Additionally, this method demonstrated excellent test-retest reliability and repeatability. Inclusion of a more medial view or whole scapula on MRI, especially in advanced levels of tear retraction, could be more reliable and representative for assessment of the degree of fatty infiltration within the muscle bulk that could help predict tear repairability and therefore improve clinical decision making, which should be studied further in clinical studies.