Purpose To assess the natural evolution of the osseous reaction following arthroscopic double-row rotator cuf repair with PEEK anchors and to analyze its correlation with clinical shoulder function. Methods Between 2015 and 2017, 159 patients received arthroscopic double-row rotator cuf repair with PEEK anchors and underwent serial clinical and radiological follow-up (3, 6, 12, and 24 months). Radiological results were analyzed by tendon integrity, bone marrow edema, and peri-implant osteolysis. Clinical shoulder function was evaluated with the Constant score. Results One-hundred and seventeen patients were enrolled; among them, 63% demonstrated bone marrow edema around the anchors on postoperative 3-month MRI. The edema area percentage was 41% ± 7%. At 6 months, edema was only seen in 12% of cases, with an area percentage of 18% ± 5%. At 12 and 24 months, edema was rarely present. Fluid signals around the anchor were observed in 17.6%, 42.7%, 33.3%, and 21.0% of patients at 3, 6, 12, and 24 months, respectively; the tunnel widening values were 1.1 ± 0.4 mm, 1.8 ± 0.5 mm, 2.3 ± 0.6 mm, and 2.2 ± 0.7 mm at each follow-up, respectively. The sign of osteolysis was signiicantly more obvious around the lateral anchor than around the medial anchor. The presence of an osseous reaction was not correlated with worse clinical outcome. Conclusion Osseous reactions following arthroscopic rotator cuf repair are common and signiicant even with PEEK anchors. Bone marrow edema does not last more than 6 months in patients without complications. Peri-implant osteolysis is more evident around the lateral anchor than around the medial anchor and improves gradually over time. The sign of osteolysis is not correlated with clinical shoulder function. Based on these indings, surgeons should be cautious about bone marrow edema lasting more than 6 months following arthroscopic rotator cuf repair. Level of evidence Level IV.