Rotator cuff-related shoulder pain is usually treated with exercise. The honest read of the evidence: the type of exercise matters less than most programs claim. A systematic review with meta-analyses of 22 randomized controlled trials covered 1,281 adults. It found moderate-certainty evidence that motor control exercise programs (training the shoulder to move with better control) cut disability more than general exercise. But they did not significantly cut pain in the short term.1 That split between disability and pain is the key finding. Patients rarely hear it.
What the trials actually compared
The review sorted the evidence by the FITT principle: frequency, intensity, type and time. The goal was to see which of these the evidence can answer. Separate meta-analyses compared exercise type (specific versus nonspecific) and intensity (high versus low). Each was rated for certainty with GRADE.1
Type
Motor control exercise programs significantly cut disability compared with general programs in the short term (standardized mean difference −0.29, 95% CI −0.51 to −0.07; 323 participants, 7 trials) and the medium term (SMD −0.33, 95% CI −0.57 to −0.09; 286 participants, 5 trials). Short-term pain did not differ significantly (SMD −0.19, 95% CI −0.41 to 0.03).1
It is still unclear how eccentric and shoulder-blade programs compare with general exercise. The evidence there is low to very low certainty.1
Intensity, frequency and time
Evidence on intensity was low to very low certainty. And no trials at all compared different frequencies or session times.1
So maybe you were told a set number of sets, reps or sessions per week is the right dose. That number does not come from trials that compared options. It may still be fine. But it is convention (custom), not proof.
What the authors concluded
For adults with rotator cuff-related shoulder pain, motor control exercise programs were probably slightly better than general programs. But the authors added a key warning. It is unclear if the gains came from motor control exercise itself. They may have come from other parts of the program, such as progression and tailoring.1
That warning may be the most useful line in the paper. A program that steps up as you improve and fits you may be doing the work. The name of the exercises may not matter.
Where imaging and injection fit
Exercise comes first, but it is not the only question. When symptoms do not improve as expected, finding the source of the pain still matters. Image guidance allows an injection to go in a confirmed spot, not a rough guess.
When the pain comes from a partial tear rather than a tendon that is only irritated, the answer still starts with conservative care. Conservative care matches surgery for partial-thickness cuff tears, which are about twice as common as full-thickness ones. A 2026 systematic review by Trygonis and colleagues, covering 16 studies and 893 patients, found: “Conservative treatment, primarily physiotherapy-based, demonstrated comparable outcomes to surgery in partial-thickness tears or low-demand individuals.”2 Eubank and colleagues reached the same place in 2024: “Studies have shown partial-thickness rotator cuff tear (PTRCT) prevalence to be twice that of full-thickness tears.”3 For those tears we add guided injection and PRP into the tendon to the loading program, work on the metabolic terrain that slows healing, and work to avoid surgery.
The relevant pages are ultrasound-guided and fluoroscopic procedures, shockwave therapy and hydrodissection. When shoulder pain comes with neck symptoms, telling the two apart matters. See telling a wrist problem from a neck problem. When trigger points (tight knots) around the shoulder add to the pain, what the evidence supports for trigger points is relevant.
Frequently asked questions
Does the specific type of shoulder exercise matter?
Somewhat. Motor control programs were probably slightly better than general exercise for disability, on moderate-certainty evidence. But the edge may come from progression and tailoring, not the exercise type itself. See the range of pain treatments.
My function improved but the pain has not. Is that normal?
It fits the evidence. Pooled trials found significant gains in disability but no significant short-term difference in pain between exercise types. Function and pain are separate outcomes. They can move at different speeds. See pain management doctors in St. Louis.
How many sessions per week should I be doing?
No randomized trials have compared different frequencies or session times for this condition. So any set number is convention (custom), not proven evidence. Talk it over with whoever runs your program.
When is an injection considered?
When symptoms are not improving as expected and it helps to confirm and treat one exact structure. Image guidance means the injection goes where intended. See ultrasound-guided and fluoroscopic procedures.
Could my shoulder pain be coming from my neck?
It can, and that changes treatment entirely. See telling a wrist problem from a neck problem, or arrange an evaluation through the appointment request form.
To have shoulder pain evaluated, request an appointment, call (314) 481-5000, or text (314) 886-5902.
Sources
- Lafrance S, Charron M, Dubé MO, Desmeules F, Roy JS, Juul-Kristensen B, Kennedy L, McCreesh K. The Efficacy of Exercise Therapy for Rotator Cuff-Related Shoulder Pain According to the FITT Principle: A Systematic Review With Meta-analyses. Journal of Orthopaedic & Sports Physical Therapy. 2024;54(8):499–512. doi:10.2519/jospt.2024.12453. Retrieved via PubMed. DOI
- Trygonis N, Loupasis T, Hautz WE, Exadaktylos AK. Conservative management of degenerative rotator cuff tears: a systematic review of long-term clinical outcomes and cost effectiveness. Journal of Orthopaedics. 2026;75:232–241. 16 studies, 893 patients. PMID 41756190. DOI
- Eubank BHF, Sheps DM, Dennett L, et al. A scoping review and best evidence synthesis for treatment of partial-thickness rotator cuff tears. Journal of Shoulder and Elbow Surgery. 2024;33(3):e126–e152. PMID 38103720. DOI
Dr. Gurpreet Singh Padda, MD, MBA, MHP
