AAOS Now Editor-in-chief Robert M. Orfaly, MD, MBA, FAAOS, sat down with John E. Kuhn, MD, FAAOS, chief of the Division of Sports Medicine at Vanderbilt University Medical Center, to discuss these findings and their implications for practice.


Published 12/20/2023
Rebecca Araujo

Physical Therapy Demonstrates Long-term Efficacy for Treatment of Atraumatic Rotator Cuff Tears

A study presented at the AAOS 2023 Annual Meeting reported the long-term results of the MOON Shoulder Group study, which sought to investigate the outcomes of physical therapy as nonoperative treatment for atraumatic rotator cuff repairs. The study was presented by John E. Kuhn, MD, FAAOS, chief of the Division of Sports Medicine at Vanderbilt University Medical Center. At 10-year follow-up, more than 70 percent of patients had not undergone subsequent shoulder surgery, with no significant change in patient-reported outcomes over the study period.

AAOS Now Editor-in-chief Robert M. Orfaly, MD, MBA, FAAOS, sat down with Dr. Kuhn to discuss these findings and their implications for practice.

The MOON study
In 2007, the MOON Shoulder Group conducted a multicenter, prospective study of 452 patients with atraumatic, symptomatic, full-thickness rotator cuff tears. The researchers sought to identify predictors of failure of nonoperative treatment to help define surgical indications. The results of that study, which were published in the Journal of Shoulder and Elbow Surgery in 2013, demonstrated that physical therapy was highly effective at 2-year follow-up.

“When I was young, we started operating on every rotator cuff tear that we saw. The condition was the indication for surgery,” Dr. Kuhn stated. “We were surprised, when we did this study 10 years ago, that many, many people do get better with physical therapy.”

In the original study, patients were enrolled between January 2007 and September 2010, and 452 patients underwent a standard evidence-based physical therapy protocol for 6 to 12 weeks. Afterward, follow-up was performed at 1, 2, 5, 7, and 10 years. Incidence and details regarding surgical treatment were recorded during follow-up.

To assess the durability of physical therapy, Dr. Kuhn and colleagues analyzed 10-year follow-up data from that cohort. Thirty-one patients withdrew prior to the 10-year follow-up, leaving 421 available for long-term analysis. Thirty-seven patients (9.1 percent) died before 10 years; 40 (9.0 percent) were lost to follow-up. Overall, 116 patients (28 percent) underwent surgery during the long-term follow-up period.

“Quite honestly, we expected the success rate to decline over time. We know that rotator cuff tears do get larger as time goes on, and we thought more and more patients would have surgery,” he said. “But as it turns out, that wasn’t the case. The effect of doing physical therapy is very durable, at least to 10 years. Those are really impressive results.”

Surgical details were available for 104 of the patients. Most (97 percent; n = 101) underwent rotator cuff repair, including one partial repair and one repair augmented with grafting. Thirty-seven patients (36 percent) underwent biceps surgery: 23 tenodeses, 12 tenotomies, and two debrided ruptured tendon stump. Twenty patients (19 percent) had distal clavicle excisions, and one patient (0.2 percent) had reverse total shoulder arthroplasty. Four patients had lysis of adhesion that required manipulation under anesthesia.

Data on surgical indications were available for 89 of 104 surgical patients. Ten patients underwent surgery due to a traumatic event injuring the rotator cuff. The most common indications for surgery were failure of nonoperative treatment (n = 67), pain (42), and weakness and/or functional loss (11). Three surgeries were due to biceps problems, two for adhesive capsulitis, and two for tear enlargement.

Patient expectations and surgical timing
According to Dr. Kuhn, the reasons for surgery differed depending on the time frame. “If someone decided to have surgery in the first 6 months, those people generally had their surgery because they didn’t think physical therapy would work,” he stated. Sixty-five patients (56.5 percent) underwent surgery within 6 months of study enrollment.

To understand the relationship between surgical timing and patient expectations, “We looked at the patient’s expectations about whether or not physical therapy would treat their problems and recorded that,” Dr. Kuhn said. “The patients who thought physical therapy would not work were much more likely to have surgery early.” In the group of patients who had surgery 6 months to 10 years after nonoperative treatment (43.5 percent), workers’ compensation status and shoulder activity were more important predictors of surgery than patient expectations of physical therapy efficacy.

Dr. Kuhn noted that, initially, the researchers did not anticipate the role of patient expectations on the choice to proceed with surgery. “When we started the study, we didn’t know about this effect of patient expectations, so we made no effort to modify it. And now in practice, of course, we do,” he said.

Now, in clinical practice, he said, “We explain to patients that physical therapy does work, and it works for the majority of patients. In our study, over 70 percent of the people at 10 years never had surgery. So we tell that to patients, and we kind of have to change their expectations regarding physical therapy. And usually what happens is they’ll go to do therapy for 6 weeks, and even if they didn’t think it would work, they come back saying, ‘Hey, this is getting better. This is working. Let’s keep going.’”

The authors acknowledged several limitations to the study, including the analysis of only patients with atraumatic rotator cuff tears. “I think there’s a general consensus that if someone has a traumatic event [leading to a rotator cuff tear] and has significant pain and disability, that patient probably is better treated with surgery,” Dr. Kuhn said. “If someone comes in with a complaint of pain, physical therapy is highly effective at treating pain. If someone has weakness or functional problems, that makes me think this patient may be better served with a surgical treatment if I can get the repair to heal.”

Other limitations include potential selection bias, as patients who enrolled in the study may have had a particular interest in avoiding surgery, as well as the lack of imaging at the 10-year follow-up, which prevented the researchers from assessing whether there were patients whose tears enlarged over the 10-year period and whether that change in anatomy was associated with patient outcomes.

Dr. Kuhn’s coauthors of “Full-thickness Rotator Cuff Tears: 10-year Follow-up of the MOON Shoulder Cohort” are Warren Dunn, MD, MPH, FAAOS; Rosemary Sanders, BA; Keith M. Baumgarten, MD, FAAOS; Julie Young Bishop, MD, FAAOS; Robert H. Brophy, MD, FAAOS; James L. Carey, MD, FAAOS; G. Brian Holloway, MD, FAAOS; Grant L. Jones, MD, FAAOS; ChunBong Benjamin Ma, MD, FAAOS; Robert G. Marx, MD, FAAOS; Eric Cleveland McCarty, MD, FAAOS; Sourav K. Poddar, MD; Matthew V. Smith, MD; Edwin E. Spencer, MD, FAAOS; Armando Felipe Vidal, MD, FAAOS; Brian R. Wolf, MD, MS, FAAOS; and Rick W. Wright, MD, FAAOS.

Rebecca Araujo is the managing editor of AAOS Now. She can be reached at raraujo@aaos.org.