For many older surgical procedures that are associated with good provider and patient experiences—including rotator cuff tear repair—the strength of the literature supporting their use is weak because it was often deemed unethical to compare treatments that work with placebo. With their high prevalence and potentially large cost burden, rotator cuff repairs are an ideal topic for a clinical practice guideline. However, when the American Academy of Orthopaedic Surgeons (AAOS) Board of Directors recently sought to develop a guideline, they found that the strength of the literature was poor. Therefore, appropriate use criteria (AUC) were developed based on available evidence from the literature and sound clinical judgment of expert clinicians.

Examining Best Evidence

In developing the AUC, an AAOS writing panel first identified 432 clinical scenarios that clinicians would likely face among patients with rotator cuff tears. “These were based on factors such as symptom severity and the patient’s overall health,” explains James O. Sanders, MD, AUC Section Leader. “They were also based on whether alcoholism or drug abuse was present, if workman’s compensation could affect the outcome, the size of the tear and whether it was old or new, and the patient’s response to prior treatment.” Potential treatments to consider for each scenario included non-surgical modalities, partial repair and/or debridement, repair, reconstruction, and arthroplasty.

After the 2,160 total voting items in the AAOS document were modified, a voting panel reviewed the strength of evidence for each scenario and discussed how they would approach it in practice before voting on the appropriateness of the treatments for these scenarios. Panel members assigned each of the voting items a level of appropriateness, from 1 to 9. Levels 1 to 3 were designated as “rarely appropriate,” 4 to 6 as “may be appropriate,” and 7 to 9 as “highly appropriate.” More than one-quarter (26%) of the voting items received a level 1 vote, whereas only 2% received a level 9 vote. “In general, there was a high level of agreement among the voting panel,” Dr. Sanders says.

Take a Conservative Approach

For many rotator cuff tears, non-operative treatment was deemed most appropriate. Only 1% of scenarios for non-operative treatment were voted as rarely appropriate. “For most patients,” Dr. Sanders says, “shoulder arthroplasty is not necessary. The few exceptions were with specific scenarios in which patients didn’t respond to conservative care, were healthy, had large or massive tears, and showed chronic changes on MRI.”

Dr. Sanders adds that physicians should use the AUC to help their patients make treatment decisions because rotator cuff repair is a procedure that clinically works well, and high-quality evidence is probably not feasible. “Overall,” he says, “the patient’s response to conservative care should dictate any next steps.”


AAOS. Appropriate use criteria for optimizing the management of full-thickness rotator cuff tears. Available at

Yamaguchi K, Tetro A, Blam O, et al. Natural history of asymptomatic rotator cuff tears: a longitudinal analysis of asymptomatic tears detected sonographically. J Shoulder Elbow Surg. 2001;10:199-203.

U.S.Department of Labor BoLS. Injuries, Illnesses, and Fatalities. Available at

Sher J, Uribe J, Posada A, et al. Abnormal findings on magnetic resonance images of asymptomatic shoulders. J Bone Joint Surg Am. 1995;77:10-15.