Rotator cuff tears: correlation between geometric tear patterns on MRI and arthroscopy and pre- and postoperative clinical findings

Background: Magnetic resonance imaging (MRI) is considered to be the best non-invasive procedure for the evaluation of rotator cuff (RC) tendon tears. Burkhart's classification is a geometric classification of full-thickness RC tears on MRI. Purpose: To correlate MRI and arthroscopic geometric full-thickness RC tears according to the Burkhart's classification with pre- and postoperative clinical findings. Material and Methods: Patients who underwent arthroscopic RC repair between 2006 and 2010 were retrospectively evaluated. Preoperative MRI and arthroscopic surgical reports were reviewed for tear geometry (Burkhart's) by three (1 radiologist, 2 surgeons) and two (surgeons) readers. MRIs were also evaluated for tear size and change of tear size in successive sagittal sections and for muscle mass and fatty infiltration. Clinical examinations were performed preoperatively and at least 12 months afterwards. Postoperative function questionnaires were filled in by the patients. Results: Forty-six patients (35 men, 11 women; mean age, 57 years; range, 41-72 years) were evaluated. Tears depicted on MRIs were classified as crescent in 11 patients (24%), longitudinal in three (6.5%), massive contracted in 29 (63%), and cuff arthropathy in three (6.5%). Muscle changes were noted almost exclusively in patients with massive tears and cuff arthropathy (16/32 patients, P = 0.013). MRIs and arthroscopic geometric classifications were in close agreement. Tear type did not correlate with pre- and postoperative physical examination or with postoperative clinical questionnaires scores. Conclusion: Geometric RC tear characterizations on preoperative MRIs were closely associated with arthroscopic findings. Postoperative results were not affected by the geometric pattern of the tears.