Kjær et al (2021)

Early vs Delayed Active Exercises after Rotator Cuff Repair

Early active loading is not superior to delayed loading

Early loading did not increase retear rates


The prevalence of rotator cuff (RC) tears lies around 11-13% of people in their fifth decade, up to 80% in their eighth. Yet, only 50% of tears in the fifth decade are symptomatic. Surgical repair is recommended for full- and/or partial-thickness tears exceeding 50% of the transversal or longitudinal tendon size. Biomechanical studies suggested early immobilization (4-6 weeks) postoperatively to avoid gap formation between the tendon and bone. Clinical studies indicate increased ROM at 3-6 months when early passive motion is immediately started postoperatively while not increasing retears. There is some evidence suggesting early active mobilization might result in greater retear rates. However, there’s not a lot of literature to hang one’s hat on. This study hypothesized that patients who receive 12 weeks of progressive active exercise therapy (PR) would benefit more with respect to improved shoulder function, pain reduction, and quality of life than those receiving passive exercise therapy, or usual care (UC). Let’s dive in.


This two-center outcome assessor-blinded RCT recruited patients with a minimum age of 18 with a clinical diagnosis of a traumatic full-thickness RC tear, involving the supraspinatus. Patients with a confirmed full-thickness supraspinatus tear, confirmed by the surgeon peri-operatively, were included and randomized. Individuals who had previous shoulder surgery, shoulder OA, rheumatoid arthritis, or peri-arthrosis were excluded.

The PR group started active assisted exercises at week 2, while this being done in week 4 for the UC group. The PR group was scheduled three times weekly, the UC group once weekly, both supplemented with daily home exercises. At weeks 6-12, both groups were confluent with physical therapist-supervised exercise therapy twice a week. Exercises were progressive according to levels of shoulder function and tailored to each patient. The exact mobilization program was as follows:

Table 1 early active shoulder after cuff repair
From: AJSM, Kjær et al 2021

The primary outcome was the Western Ontario Rotator Cuff Index (WORC) at 12 weeks post-op. This questionnaire measures pain, functional activity level, and health-related quality of life. Secondary measures were WORC at 6 weeks and 1 year. Other measures at 6 and 12 weeks, and 1 year were: DASH, GRS, NPRS, active and passive ROM, and MVIC. Intervention adherence was measured via exercise logbooks.


A priori sample size calculations showed a total of 82 patients were required. The authors aimed for 100 patients in total to account for dropouts. All post-operative examinators were blinded to group allocation and an external statistical consultant performed the analysis on the primary outcome data, blinded. Groups were similar at baseline. The interval between trauma and surgery ranged between 3 and 28 weeks. Half underwent arthroscopic repair, half deltoid split repair (open).

There were no significant differences between groups in WORC scores. Additionally, no between-group differences were found for any of the secondary outcome measures. Nonetheless, both groups improved clinically relevant. There were no differences in retear rates between groups.

Fig 2 early active shoulder after cuff repair
From: AJSM, Kjær et al 2021

Questions and Thoughts

This is a highly pragmatic study which is something we love. You can easily implement this into your daily practice. The researchers note that their results are somewhat in line with previously published literature. They acknowledge the fact that the discrepancy in loading between groups might not have been sufficient to result in meaningful changes.

Something that wasn’t measured but should be noted is that patients in the early loading group might have been more confident in using their shoulder in everyday life. This requires an additional study to investigate.

Not everyone is going to want to hear this but… Do patients need a physio in the first few weeks? What if we saw them once to educate them, explained self-mobilization exercises, and make sure they can contact us when they have any questions? This could be pretty interesting cost-benefit-wise for the healthcare system. Not saying this is non-inferior or superior, just thinking out loud — leave no stone unturned.

Talk Nerdy to Me

This was a pretty nice study. They asked a simple question — as it should be — built a study around it and got it answered. The methodology was alright and the fact that they used an external statistical consultant should be applauded. Stats are hard, we need professionals for that.

Some limitations are apparent. One is the obvious attention bias. Patients in the PR groups saw their physical therapist thrice as much as the UC group. One would think that this would result in a non-specific effects boost on subjective outcome measures, yet it didn’t result in false-positives since no differences were found. Frequent contact between clinician and patient might even increase adherence, but even here no differences were found. Another limitation is the huge load of secondary outcome measures. This usually results in an increase in random noise getting through as a statistically significant effect. An example might be the active scaption at 6 weeks, which was statistically in favor of the PR group. Results like this can be blatantly ignored due to the high false-positive rates with so many outcome measures. Being extra critical, one might say it is unusually coincidental that the researchers collected the exact number of patients needed for their intended statistical power. Not to point any fingers, it might really be a coincidence.

Early active loading might not be needed after rotator cuff repair, although more research — as always — is needed.

Take home messages

  • Early active loading did not result in superior results
  • Early active loading did not increase retear rates
  • The program difference between groups might not have been sufficient so more research is needed


Kjær, B. H., Magnusson, S. P., Henriksen, M., Warming, S., Boyle, E., Krogsgaard, M. R., … & Juul-Kristensen, B. (2021). Effects of 12 weeks of progressive early active exercise therapy after surgical rotator cuff repair: 12 weeks and 1-year results from the CUT-N-MOVE randomized controlled trial. The American journal of sports medicine49(2), 321-331.



Awarded World Leading Shoulder Expert Filip Struyf takes you on a 5-Day Video Course to Bust Lots of Shoulder Myths that Prevent you From Delivering the Best Care for your patients with Shoulder Pain

Free rcrsp webinar cta
Download our FREE app