The 60-second version
Shoulder impingement - or subacromial pain syndrome - responds well to structured exercise rehab, with no advantage to surgery over supervised exercise in the landmark Beard 2018 CSAW trial. The Tipton 2003 connective-tissue framework sets the timescale: 8-12 weeks for measurable tendon adaptation. The Cook 2015 loading model permits pain up to 3/10 if it settles within 24 hours. Start with isometrics and scapular control, progress to eccentrics in weeks 3-8, and reintroduce overhead movement in weeks 8-12. Red flags - sudden trauma, night pain not settling, significant weakness, neurological signs - mean see a physiotherapist first.
The most common shoulder complaint in adults who lift, swim, paddle, or work at a desk is some version of it hurts when I raise my arm overhead. The umbrella term is shoulder impingement - or, in the more careful modern literature, subacromial pain syndrome. Either way, the encouraging news is that the strongest evidence base supports exercise-based self-rehab over surgery for most mild-to-moderate cases, with comparable long-term outcomes and no surgical downside Beard 2018 Holmgren 2012.
The discouraging news is that exercise rehab is a slower, more disciplined process than most people give it. Tipton's foundational 2003 review of connective-tissue adaptation established what every tendon-rehab framework since has worked downstream of: tendon and ligament tissue remodels in response to graded loading on a timescale of 8 to 12 weeks, not days Tipton 2003. There is no shortcut.
What we actually mean by impingement
The classic description - rotator-cuff tendon and subacromial bursa pinched between the head of the humerus and the underside of the acromion - is mechanically intuitive but clinically less useful than it sounds. Modern reviews led by Lewis 2009 and the Dutch multidisciplinary guideline (Diercks 2014) prefer the term subacromial pain syndrome because imaging rarely identifies a single tissue culprit, and the rehab response is comparable across underlying tissues Lewis 2009 Diercks 2014.
The practical mental model is that the rotator-cuff tendons and the bursa beneath the acromion are irritated, the scapula is moving suboptimally during arm elevation, and the surrounding muscles - upper trap, pec minor, levator scapulae - have built compensatory tension that locks the scapula in a position that worsens the pinch. Ludewig's 2009 work mapped the scapular-kinematic patterns associated with impingement and is the operating manual most rehab protocols are scaffolded from Ludewig 2009.
When self-rehab is not the answer
Before any exercise, screen for the red flags. These are the cases that need a physiotherapist or physician first, not later:
- Sudden trauma onset - a fall on the shoulder, a tackle, a heavy lift gone wrong. Rotator-cuff tears, labral injuries, and AC-joint injuries have different rehab needs than insidious-onset impingement.
- Night pain that won't settle - waking up because of the shoulder, lying-on-the-side intolerance not improving over 2 weeks.
- Significant weakness - inability to lift a coffee mug, hold a phone to your ear, raise the arm against gravity.
- Numbness or tingling down the arm - this is a cervical-spine or brachial-plexus sign, not impingement.
- Failure to improve over 6 to 8 weeks of careful self-rehab.
Outside those flags, the evidence supports starting with structured exercise. Beard 2018 - the landmark CSAW placebo-controlled surgical trial - found no advantage to subacromial decompression over either placebo surgery or supervised exercise rehab at 6 and 12 months Beard 2018.
Phase 1: Isometrics and scapular control (weeks 1 to 3)
The starting point in most modern protocols is isometric loading - holding a contraction without joint movement - because it produces analgesic effects in tendinopathy without provoking the painful range of motion. Cook 2015's loading framework is the operating model: pain up to about 3 out of 10 during an exercise is acceptable if it settles to baseline within 24 hours Cook 2015.
Daily exercises in this phase:
- Isometric external rotation at the side, elbow at 90 degrees, pressing the back of the hand into a doorframe at 30 to 50 percent of maximum effort. 5 sets of 30 to 45 seconds, 1 minute rest. Once per day.
- Wall slides - back against the wall, arms in a W, slide the arms slowly up to a Y within the pain-free range. 3 sets of 10. Focus on the lower trap pulling the scapula down and back, not just the deltoid raising the arm.
- Scapular squeeze - standing or seated, gently retract the scapulae (think wallet pockets) and hold for 10 seconds. 3 sets of 10. This trains the postural set that takes the scapula out of the impingement-prone position Ludewig 2009 described.
- Pec-minor and upper-trap soft tissue work - 3 to 5 minutes per side on a firm ball or foam roller, before the strength work.
Phase 2: Eccentric and concentric loading (weeks 3 to 8)
Once isometrics are tolerated and pain has settled below 4 out of 10 during the day, the protocol introduces movement under load. Camargo 2014's review of eccentric training for rotator-cuff tendinopathy summarised the evidence for slow, controlled eccentrics as a remodelling stimulus Camargo 2014. Kuhn 2009 synthesised these findings into a graded-exercise protocol that has informed most clinical guidelines since Kuhn 2009.
Sample weekly structure:
- Side-lying external rotation with a 1 to 2 kg dumbbell or light band. 3 sets of 12 to 15, slow eccentric (3 to 4 seconds lowering). Three times per week.
- Banded external rotation at 90/90 (shoulder abducted to 90 degrees, elbow bent to 90) - only once side-lying ER is comfortable. 3 sets of 12.
- Prone Y-T-W - face down, light dumbbells (1 to 2 kg), raising arms in Y, T, and W shapes. 2 sets of 8 each.
- Serratus push-up plus - in a plank or wall position, push through the floor at end range to protract the scapula. 3 sets of 10.
- Pulling exercises - rows, face-pulls - twice per week. Cumulative pulling volume is one of the more reliable predictors of impingement recovery Hanratty 2012.
Phase 3: Loaded overhead and return to activity (weeks 8 to 12+)
By weeks 8 to 12 the rotator-cuff tendons have had time for the adaptive remodelling Tipton 2003 mapped, and most users have built tolerance to mid-range loading. The final phase reintroduces overhead movement and sport-specific loading.
- Landmine press - the angled bar trajectory limits overhead range and is well tolerated by most rehabbing shoulders. 3 sets of 8 to 10.
- Half-kneeling cable or band press - vertical pressing in the pain-free range. Build range as tolerated.
- Cuban press with light dumbbells - high-quality rotator-cuff coordination drill.
- Pull-ups, lat pull-downs, swimming, paddling, throwing - reintroduced incrementally, starting at 30 to 50 percent of pre-flare volume and progressing weekly.
Things that stall recovery
The four most common mistakes in self-rehab, in rough order of frequency:
- Skipping the isometric phase and going straight to dynamic loading. The pain doesn't settle and the protocol stalls.
- Too much volume in the first 3 weeks. Cook 2015's framework is the safety net: pain that lingers into the next day means too much yesterday.
- Ignoring the desk-posture background. Eight hours of forward-rounded shoulders undoes 30 minutes of rehab. The thoracic-extension work and pec-minor release are not optional.
- Avoiding pain entirely. Hanratty 2012 and the broader rehab literature consistently find that the avoidance-only path produces worse outcomes than graded loading with mild, settling pain Hanratty 2012.
Practical takeaways
- The evidence backs exercise rehab over surgery for most cases - Beard 2018 found no surgical advantage over supervised exercise.
- Start with isometrics - pain-free, postural, daily. Build scapular control before loading the cuff.
- Progress to eccentrics in weeks 3 to 8. Slow, controlled, light load.
- Reintroduce overhead movement in weeks 8 to 12, starting with landmine and half-kneeling variations.
- See a physiotherapist if there's traumatic onset, night pain not settling, significant weakness, neurological signs, or no improvement after 6 to 8 weeks.
References
Tipton 2003Tipton CM, Matthes RD, Maynard JA, Carey RA. (2003) The influence of physical activity on ligaments and tendons. Med Sci Sports. 7(3):165-75. View source →Lewis 2009Lewis JS. (2009) Rotator cuff tendinopathy / subacromial impingement syndrome: is it time for a new method of assessment? Br J Sports Med. 43(4):259-64. View source →Holmgren 2012Holmgren T, Bjornsson Hallgren H, Oberg B, Adolfsson L, Johansson K. (2012) Effect of specific exercise strategy on need for surgery in patients with subacromial impingement syndrome: randomised controlled study. BMJ. 344:e787. View source →Kuhn 2009Kuhn JE. (2009) Exercise in the treatment of rotator cuff impingement: a systematic review and a synthesized evidence-based rehabilitation protocol. J Shoulder Elbow Surg. 18(1):138-60. View source →Cook 2015Cook JL, Rio E, Purdam CR, Docking SI. (2015) Revisiting the continuum model of tendon pathology: what is its merit in clinical practice and research? Br J Sports Med. 50(19):1187-91. View source →Camargo 2014Camargo PR, Alburquerque-Sendin F, Salvini TF. (2014) Eccentric training as a new approach for rotator cuff tendinopathy: review and perspectives. World J Orthop. 5(5):634-44. View source →Diercks 2014Diercks R, Bron C, Dorrestijn O, et al. (2014) Guideline for diagnosis and treatment of subacromial pain syndrome: a multidisciplinary review. Acta Orthop. 85(3):314-22. View source →Beard 2018Beard DJ, Rees JL, Cook JA, et al. (2018) Arthroscopic subacromial decompression for subacromial shoulder pain (CSAW): a multicentre, pragmatic, parallel group, placebo-controlled, three-group, randomised surgical trial. Lancet. 391(10118):329-38. View source →Hanratty 2012Hanratty CE, McVeigh JG, Kerr DP, et al. (2012) The effectiveness of physiotherapy exercises in subacromial impingement syndrome: a systematic review and meta-analysis. Semin Arthritis Rheum. 42(3):297-316. View source →Ludewig 2009Ludewig PM, Reynolds JF. (2009) The association of scapular kinematics and glenohumeral joint pathologies. J Orthop Sports Phys Ther. 39(2):90-104. View source →