Rotator cuff tears are one of the most common causes of pain and disability in the upper extremity. With the use of an animal model, we studied the healing response of a controlled defect in the normal supraspinatus tendon and in a tendon with a reduced intrinsic healing capacity. In 36 Sprague-Dawley rats, defects (2 mm x 2 mm) were created in the supraspinatus tendons bilaterally. To model a tendon with an intrinsically reduced capacity to heal, the tissue adjacent to the defect area in the left shoulder was treated with in situ freezing. The contralateral tendon was not frozen. After 3 (n = 12), 6 (n = 12), and 12 (n = 12) weeks, animals were killed and underwent histologic (n = 4 from each group) and biomechanical (n = 8 from each group) evaluation. An additional group of untreated animals served as a normal control group. On histologic evaluation 78% of tendons had persistent defects (defined as incomplete closure of the defect site). Over time, the tissue from both groups demonstrated an improved histologic grade but did not reach normal levels, even at 12 weeks. No histologic differences were found between defect healing in normal tendons and in those treated with in situ freezing. On biomechanical evaluation there were also no significant differences between treatment groups. Over time, an improvement occurred in tissue properties, indicating that some healing of the defects had occurred. However, these tissue properties remained an order of magnitude lower than those of normal control tendons. These findings indicate that there is an active but inadequate repair response to the defect in the rat supraspinatus tendon, which is not significantly worsened by in situ freezing of the tissue around the defect. This model has applications toward the study of techniques to improve or accelerate cuff defect healing.