OBJECTIVE The purpose of this study was to evaluate the changes in fatty degeneration and atrophy of rotator cuff muscles after arthroscopic repair. We further assessed the factors affecting the functional outcomes and integrity of the rotator cuff. METHODS One hundred and two prospectively followed patients who underwent single-row arthroscopic repair for full-thickness rotator cuff tears between 2008 and 2010 in our institution were included. All patients underwent shoulder MRI examination before the arthroscopic repair and at the final follow-up at least 2 years after the surgical repair. Supraspinatus muscle atrophy was measured and evaluated according to the Thomazeau classification. The fatty degeneration of the cuff muscles was graded according to the Goutallier classification. Functional outcomes were assessed with the Constant shoulder score. The changes in fatty degeneration and atrophy were analyzed during the treatment period. Correlation coefficients (Pearson r) and stepwise, multiple linear regression were used to determine the relationship between the outcome variables (final Constant score and integrity of the cuff), and the predictor variables, age, sex, follow-up duration, initial muscle atrophy, final muscle atrophy, initial fatty degeneration and final fatty degeneration. RESULTS Of the 102 patients reviewed, 87 patients responded and concluded the final clinical follow-up and MRI examination (85.2 % follow-up rate). There were 67 females and 20 males with a mean age of 62.5 ± 8.3 years (range 40-80 years). Mean follow-up period was 30.1 ± 5.8 months (range 24-43 months). At the final follow-up, the mean Constant shoulder score was 94.2 ± 8.2 (range 70-100), and 66 (75.9 %) patients rated as excellent, 14 (16.1 %) as good, and 7 (8.0 %) as fair. No patient had poor results. There was re-rupture in 26 (29.9 %) patients on final MRI examination. No patient had improvement in muscle atrophy and fatty degeneration. The atrophic changes between intact tendon and re-rupture cases were statistically similar (p = 0.300). The deterioration of fatty degeneration was significantly higher in the re-rupture group (p = 0.0001). The Constant shoulder score was significantly lower in patients with re-rupture (97.4 ± 5.0 versus 86.6 ± 9.3, p = 0.001).Multiple stepwise regression analysis showed that the Constant score was dependent on the final integrity of the tendon and the size of the tear (R (2) 0.420, p 0.001). The final integrity of the tendon was dependent on the age of the patient, initial and final fatty degeneration of the cuff muscles and the size of the tear (R (2) 0.669, p 0.001). CONCLUSIONS Initial muscle atrophy and fatty degeneration did not improve even after a successful rotator cuff repair where the tendon anatomic integrity was maintained for at least 2 years. It may continue to deteriorate, and the best possibility was preservation of the preoperative status. On the other hand, in cases of re-rupture, fatty degeneration and atrophy continued to worsen significantly. The factors affecting tendon integrity were found to be the age of the patient, the size of the tear and the severity of preoperative fatty degeneration in the rotator cuff.