Abstract
BACKGROUND: Macroscopic alterations of the affected rotator cuff (RC) are undoubtedly linked to microscopic changes, but they may underestimate the actual degree of the disease. Moreover, it remains unclear whether preoperative structural RC changes may alter clinical outcomes. METHODS: Supraspinatus tendon and muscle samples were collected from 47 patients undergoing RC surgery. Tendons were evaluated histologically according to the Bonar score; fatty infiltration and muscle atrophy were quantified using a software for biomedical image analysis (ImageJ) in percentage of area affected in the observed muscle section. Preoperative shoulder ROM and pain were evaluated. Radiological muscle atrophy was evaluated with the Tangent Sign and Occupation Ratio; fatty infiltration was assessed according to the Goutallier classification. Correlations between histological, radiological and clinical outcomes were assessed. Statistics were performed using the Spearman correlation coefficient. Intraobserver and interobserver agreement was calculated. RESULTS: Histopathologic fatty infiltration (r = 0.007, p = 0.962), muscle atrophy (r = 0.003, p = 0.984) and the total Bonar score (r = 0.157, p = 0.292) were not correlated to preoperative shoulder pain. Muscle atrophy showed a significant but weak negative correlation with the preoperative movement of abduction (r = -0.344, p = 0.018). A significant but weak positive correlation was found between muscle atrophy and the total Bonar score (r = 0.352, p = 0.015). No correlation between histological and radiological evaluation was found for both fatty infiltration (r = 0.099, p = 0.510) and muscle atrophy (Tangent Sign: r = -0.223, p = 0.131; Occupation Ratio: r = -0.148, p = 0.319). Our histological evaluation showed a modal value of 3 (out of 3) for fatty infiltration and an equal modal value of 2 and 3 (out of 3) for muscle atrophy. In contrast, the modal value of the Goutallier score was 1 (out of 4) and 28 patients out of 47 showed a negative Tangent sign. At histology, intraobserver agreement ranged from 0.59 to 0.81 and interobserver agreement from 0.57 to 0.64. On the MRI intraobserver agreement ranged from 0.57 to 0.71 and interobserver agreement ranged from 0.53 to 0.65. CONCLUSIONS: Microscopic muscle atrophy appeared to negatively correlate with the movement of abduction leading to functional impairment. Shoulder pain did not show any relationship with microscopic changes. Radiological evaluation of the supraspinatus muscle alterations seemed to underestimate the degree of the same abnormalities evaluated at histology.