Background We report functional outcomes at six years in patients with varying degrees of fatty infiltration and atrophy of the rotator cuff muscles who have undergone anatomic total shoulder replacement. Methods A retrospective analysis of case notes and magnetic resonance imaging scans of patients undergoing total shoulder replacement for primary glenohumeral arthritis was performed. Patients were grouped based upon their pre-operative magnetic resonance imaging findings for fatty infiltration, muscle area and tendinopathy. Post-operative functional outcomes were assessed using the Oxford Shoulder Score and Quick Disabilities of the Arm, Shoulder and Hand score. Post-operative measurements were made for active shoulder movements. Results Thirty-two patients were reviewed at a mean of 67 months following surgery. All patients demonstrated fatty infiltration on their pre-operative magnetic resonance imaging scan. Muscle atrophy was shown in 22 patients and 12 had tendinopathy. Multiple regression analysis showed no correlation between the Oxford Shoulder Score (p = 0.443), the Quick Disabilities of the Arm, Shoulder and Hand score (p = 0.419), forward flexion (p = 0.170), external rotation (p = 0.755) and any of the pre-operative independent variables. Discussion The degree of fatty infiltration, muscle atrophy and tendinopathy of the rotator cuff muscle on pre-operative magnetic resonance imaging scanning is not associated with functional outcome score or functional movement at medium-term follow-up following total shoulder replacement. Level of evidence IV