Rugby shoulder dislocations, the injury that tends to come back

On 21 September 2026 the All Blacks confirmed that their captain, Ardie Savea, needed surgery after dislocating his shoulder during the four-Test series in South Africa. Reuters reported that recovery from that operation normally takes four to six months, enough to rule him out of the Bledisloe Cup and the rest of the year. For one of the best-known players in the world, a single joint had ended a season.

The surgery is not mainly about the dislocation that has already happened. It is about the next one. In a French randomised trial of young adults with a first dislocation, 79% of those treated with a sling and physiotherapy had dislocated again within about six years, against 22% of those who had an early keyhole repair. Shoulder instability is rarely the commonest injury in rugby, but it is among the costliest.

This article covers what the tackle does to the shoulder joint, how many days the injury costs, who is most at risk of a first and a second dislocation, what surgery and rehabilitation have achieved in trials and rugby cohorts, where that evidence is thin, and why the sensible response is to prepare the shoulder and treat a first dislocation properly rather than give up the game.

What the tackle does to a shoulder

The shoulder trades stability for range: a large ball of humerus sits on a shallow socket, deepened by a rim of cartilage called the labrum and held by the capsule and rotator cuff. When the arm is levered backwards and outwards, the ball can be forced out of the front. On the way it usually tears the labrum from the front of the socket, a Bankart lesion, and may dent the back of the ball, a Hill-Sachs lesion, or break off part of the socket rim.

Rugby supplies the lever. A 2012 video study of 24 elite players, led by the shoulder surgeon Lennard Funk, described three mechanisms. The “try scorer” lands on an outstretched arm forced overhead. The “tackler” has his abducted arm dragged behind him by a ball carrier running through it. The third is a direct blow to an arm held at the side. The first two lever the ball out of its socket and mostly produced dislocations.

A 2019 analysis of 39 dislocations filmed in top professional and international matches found those three patterns in 67% of cases. It added a fourth, the “poach” position, when a player is bent over the ball at the breakdown trying to win it, which accounted for 18%. The remaining 15% fitted no category. Across studies the tackle dominates: it caused 66% of shoulder injuries over 16 seasons of the English Premiership.

How rugby shoulders are injured

Share of shoulder injuries, or of filmed dislocations, attributed to each event or mechanism.

Irish schools: shoulder injuries in the tackle81%

English Premiership: shoulder injuries in the tackle66%

Pro dislocations: try scorer, tackler or direct impact67%

Pro dislocations: poach position at the breakdown18%

Pro dislocations: no recognised mechanism15%

Sources: Leahy et al., Orthopaedic Journal of Sports Medicine (2021); West et al., British Journal of Sports Medicine (2020); Montgomery et al., American Journal of Sports Medicine (2019), 39 dislocations.

The cost is counted in days. In the English Premiership surveillance reported by Headey and colleagues in 2007, dislocation and instability were the most severe shoulder injuries, at a mean of 81 days’ absence, and caused 42% of all days lost to shoulder injury. A 2026 meta-analysis of 37 studies put the average shoulder injury at 57 days. For comparison, the average Premiership match injury in 2002–04 cost 18. Head injuries and the tackle laws are covered in our earlier article on rugby’s tackle.

Days lost per injury

Mean days of absence per injury. Different surveillance systems and definitions, so compare orders of magnitude rather than exact gaps.

Shoulder dislocation or instability, English Premiership81

Any shoulder injury, pooled across 10 cohorts57

Any shoulder injury, Irish schools Senior Cup47

Any shoulder injury, Super Rugby 2005–201037

Any match injury, English Premiership 2002–0418

Sources: Headey et al., American Journal of Sports Medicine (2007), as reported by Lynch et al. (2013) and Watura et al. (2025); Doyle et al., Archives of Orthopaedic and Trauma Surgery 146:273 (2026); Leahy et al. (2021); Usman et al., Journal of Science and Medicine in Sport (2015); Brooks et al., BJSM 39(10):757 (2005).

Where the risk concentrates

The tackler carries most of the risk. In two seasons of Irish schools Senior Cup rugby, 81% of match shoulder injuries happened in the tackle, and tacklers were injured at 7.5 per 1,000 match hours against 2.3 for ball carriers, more than three times the rate. Shoulder injuries made up 23% of time-loss match injuries but 35% of the days missed, and dislocations and subluxations, mostly suffered by backs, were the single most burdensome type.

Level of play matters as well. Over five seasons the French federation’s insurer recorded 1,345 shoulder dislocations and subluxations among licensed players, about ten per 10,000 men each season. Among senior professionals the rate was 83 per 10,000, four to five times that of senior amateurs and juniors, and almost none occurred in children’s rugby schools. Professionals tackle harder and more often, and their injuries are probably reported more completely.

Shoulder dislocations by level, French rugby

Dislocations and subluxations reported per 10,000 licensed players per season, 2008–2013, from insurance declarations.

Senior professionals83

Juniors21

Senior amateurs17

Cadets12

Rugby school (children)<1

Source: Bohu et al., British Journal of Sports Medicine (2014), 1,345 episodes. Rates depend on reporting to the insurer, which may be more complete in the professional game.

Then comes the second dislocation. In a cohort of 252 people aged 15 to 35, treated with a sling and physiotherapy after a first dislocation, 56% had further instability within two years and 67% within five. Age dominated: 87% of those aged 15 to 20 at the first dislocation had recurrence within five years, against 31% of those aged 31 to 35. Men were affected far more often than women.

Recurrence after a first dislocation without surgery

Share with further instability (dislocation or subluxation) after non-operative treatment. Rugby rows come from player-reported and insurance-linked cohorts.

Aged 15–20 at first dislocation, within 5 years86.6%

Aged 21–25, within 5 years73.8%

Aged 26–30, within 5 years48.8%

Aged 31–35, within 5 years30.7%

French rugby players after a first episode (any treatment)68.8%

Japanese high-school players, reinjury in the same season54.3%

Sources: Robinson et al., Journal of Bone and Joint Surgery (Am) (2006), age figures as reported by Gurney-Dunlop et al. (2017); Gérometta et al., Arthroscopy abstract (2017); Kawasaki et al., Journal of Shoulder and Elbow Surgery (2014).

Rugby players fit the worst profile: young, male and returning to repeated collisions. Among French players who reported a first episode to the federation’s insurer in 2012–13, 69% had another, and half eventually had a stabilisation operation. In seven elite Japanese high-school teams, 15% of 378 players already had a dislocation behind them, and 54% of shoulders that had dislocated were injured again during the season.

What has been measured to work

The strongest evidence compares early surgery with a sling and rehabilitation after a first dislocation. The French trial randomised 40 people aged 18 to 25 to an arthroscopic Bankart repair within two weeks or to non-operative care, with the same immobilisation and physiotherapy. At two years recurrent instability had affected 10% of the surgical group and 65% of the others. By about six years it was 28% against 90%, and dislocation itself 22% against 79%.

Function followed. In the same trial 82% of the surgical group returned to sport at their previous level or higher, against 21% of the non-operative group, and nearly half of the non-operative patients went on to have stabilisation surgery anyway. Older and smaller trials pointed the same way: in one of 40 patients under 30, redislocation at two years was 16% after early repair and 47% after immobilisation; in another of young athletes, recurrent instability was 11% against 75%.

Randomised trials after a first dislocation

Recurrence in the surgical and non-operative arms. Small trials of young adults; none was limited to rugby players.

France, dislocation by ~6 years: sling and rehab (19)79%

France, dislocation by ~6 years: arthroscopic repair (18)22%

Kirkley trial, redislocation at 2 years: immobilisation47%

Kirkley trial, redislocation at 2 years: arthroscopic repair15.9%

Bottoni trial, instability at ~3 years: non-operative (12)75%

Bottoni trial, instability at ~3 years: arthroscopic repair (9)11.1%

Sources: Pougès et al., American Journal of Sports Medicine 53(10):2289 (2025); Kirkley et al., Clinical Journal of Sport Medicine abstract (1997); Bottoni et al., American Journal of Sports Medicine (2002).

For rugby players the question is often which operation. The Latarjet moves a piece of bone from the front of the shoulder blade to the socket rim, which helps when the rim is already damaged; in one series of 37 rugby shoulders, 73% had a bony glenoid lesion. A French cohort of 62 competitive players found recurrence in 19% after open Bankart repair and 3% after open Latarjet, with return to rugby at about eight months against six.

That is less settled than it looks. The French comparison set a 1990s Bankart group against a 2010s Latarjet group, assessed with better imaging, newer implants and different rehabilitation, and the authors flagged this themselves. A 2015 review of surgery in elite rugby players found only five studies, all retrospective, with poor methodological scores. The 2026 meta-analysis found no randomised comparison of operations at all, and non-operative care in rugby rested on a single study.

Even the trial result carries a caveat. In the French trial, every dislocation in the surgical group came after the second year, so the gap narrows with time; across 21 studies and 5,142 young patients, pooled redislocation was 16% after surgery and 25% without it, a smaller difference than the best trials suggest. And prevention is unproven: in a trial in 41 English community clubs, a warm-up that cut concussion saw more shoulder injuries, 1.7 against 1.0 per 1,000 hours, though fewer dislocations.

Why the answer is to play carefully, not to stop

Most players who have a shoulder stabilised go back to rugby. The 2026 meta-analysis pooled 775 operations and found a mean return to play of about six months, and a review of arthroscopic Bankart repair in collision athletes put return to sport at 90%, with 81% back at the same level. A first dislocation calls for assessment, imaging and an informed talk about surgery, not retirement.

Before injury, the levers are technique and strength. Tackling with the arm left trailing behind the body and landing on an outstretched arm are the patterns to coach out, and the rotator cuff and shoulder-blade muscles can be strengthened. A shoulder-specific warm-up for community rugby has been piloted in England, but players completed only 58% of its exercises, against 97% for a general programme, and it has not yet been tested for injury reduction.

After injury, the risk is going back too soon on an unstable shoulder. Repeated episodes are linked with more joint damage: in one study of shoulders awaiting surgery, early arthritis was closely related to the number and frequency of dislocations. Rugby’s shoulder evidence is a set of small trials and selected cohorts, but it points one way: young contact players do better when a first dislocation is taken seriously. Savea’s surgeons will be working to keep him playing for years, not one more match.

Questions people ask

Will a dislocated shoulder heal on its own?

It can settle with a sling and physiotherapy, but in young men who return to contact sport it usually dislocates again. In one cohort, 87% of people aged 15 to 20 had further instability within five years without surgery.

Should a young rugby player have surgery after a first dislocation?

Trials in young adults found far fewer redislocations after early arthroscopic repair, but they were small and not limited to rugby. The decision depends on age, bone damage on scans and timing in the season, and is one for a shoulder specialist.

What is the difference between a Bankart repair and a Latarjet?

A Bankart repair reattaches the torn labrum to the socket, usually by keyhole surgery. A Latarjet transfers a piece of bone to the front of the socket and is often chosen when the rim is damaged. Rugby cohorts favour the Latarjet for recurrence, but no trial has compared them.

How long does it take to return to rugby after shoulder surgery?

Studies of rugby and collision athletes report averages of about five to eight months, depending on the operation. Reuters reported that such surgery normally needs four to six months of recovery.

Can exercises prevent a shoulder dislocation?

No trial has yet shown that a shoulder programme prevents rugby dislocations. Strength work and tackle technique are sensible, and general warm-ups reduce other injuries.

The short version

• Shoulder dislocation and instability cost a mean of 81 days per injury in English professional rugby, the most of any shoulder injury.
• Most rugby shoulder injuries happen in the tackle; in Irish schools rugby the tackler was injured about three times as often as the ball carrier.
• Without surgery, 87% of people aged 15 to 20 had further instability within five years of a first dislocation.
• In a small French trial, 79% redislocated within about six years after a sling and physiotherapy, against 22% after early arthroscopic repair.
• Evidence comparing operations is retrospective, and shoulder-specific prevention is untested. Treat a first dislocation seriously and get back to playing.

This is a summary of published research, not medical advice. If you have dislocated a shoulder, or it feels loose or slips when you tackle, see a doctor or sports physiotherapist, and ask to be referred to a shoulder specialist before returning to contact.

Further reading: Doyle et al., Archives of Orthopaedic and Trauma Surgery (2026), for the latest pooled evidence on rugby shoulder injuries. Crichton, Jones and Funk, British Journal of Sports Medicine (2012), for the three injury mechanisms. Pougès et al., American Journal of Sports Medicine (2025), for the six-year trial.

Sources

1. Mulvenney N. All Blacks captain Savea facing shoulder surgery and long spell out. Reuters, via Irish Examiner and rugby.com.au, 21 September 2026.
2. Pougès C, Boutry M, Maynou C, et al. Arthroscopic Bankart repair versus immobilization for a first episode of an anterior shoulder dislocation before the age of 25 years: a randomized controlled trial with 6-year follow-up. American Journal of Sports Medicine 53(10):2289–2297 (2025).
3. Crichton J, Jones DR, Funk L. Mechanisms of traumatic shoulder injury in elite rugby players. British Journal of Sports Medicine 46(7):538 (2012).
4. Montgomery C, O’Briain DE, Hurley ET, et al. Video analysis of shoulder dislocations in rugby: insights into the dislocating mechanisms. American Journal of Sports Medicine (2019), doi:10.1177/0363546519882412.
5. West SW, Starling L, Kemp S, et al. Trends in match injury risk in professional male rugby union: a 16-season review of 10 851 match injuries in the English Premiership. British Journal of Sports Medicine (2020), doi:10.1136/bjsports-2020-102529.
6. Leahy TM, Kenny IC, Campbell MJ, et al. Epidemiology of shoulder injuries in schoolboy rugby union in Ireland. Orthopaedic Journal of Sports Medicine (2021), doi:10.1177/23259671211023431.
7. Headey J, Brooks JHM, Kemp SPT. The epidemiology of shoulder injuries in English professional rugby union. American Journal of Sports Medicine (2007); figures as reported by Lynch et al., Pakistan Journal of Medical Sciences (2013) and Watura et al., Rugby injury imaging (2025), doi:10.1055/s-0045-1807753.
8. Doyle JH, Bellamy M, Fowler K, et al. Shoulder injuries in rugby union: a systematic review and meta-analysis. Archives of Orthopaedic and Trauma Surgery 146:273 (2026), doi:10.1007/s00402-026-06445-7.
9. Brooks JHM, Fuller CW, Kemp SPT, Reddin DB. Epidemiology of injuries in English professional rugby union: part 1 match injuries. British Journal of Sports Medicine 39(10):757 (2005).
10. Usman J, McIntosh AS, Quarrie K, Targett S. Shoulder injuries in elite rugby union football matches: epidemiology and mechanisms. Journal of Science and Medicine in Sport (2015), doi:10.1016/j.jsams.2014.07.020.
11. Bohu Y, Klouche S, Lefevre N, et al. The epidemiology of 1345 shoulder dislocations and subluxations in French rugby union players: a five-season prospective study from 2008 to 2013. British Journal of Sports Medicine (2014), doi:10.1136/bjsports-2014-093718.
12. Robinson CM, Howes J, Murdoch H, et al. Functional outcome and risk of recurrent instability after primary traumatic anterior shoulder dislocation in young patients. Journal of Bone and Joint Surgery (Am) (2006), doi:10.2106/00004623-200611000-00002; age-specific figures, and Ogawa et al. on arthritis before surgery, as reported by Gurney-Dunlop et al., Annals of Joint (2017), doi:10.21037/aoj.2017.10.14.
13. Gérometta A, Klouche S, Lefevre N, et al. Should we stabilize surgically the athlete’s shoulder after the first episode of instability? Comparative study in 138 rugby players. Arthroscopy conference abstract (2017), doi:10.1016/j.arthro.2017.04.090.
14. Kawasaki T, Ota C, Urayama S, et al. Incidence of and risk factors for traumatic anterior shoulder dislocation: an epidemiologic study in high-school rugby players. Journal of Shoulder and Elbow Surgery (2014), doi:10.1016/j.jse.2014.05.007.
15. Kirkley A, Griffin S, Richards C, Mohtadi NGH. Prospective randomized clinical trial comparing immediate arthroscopic stabilization versus immobilization and rehabilitation in first traumatic anterior dislocations of the shoulder. Clinical Journal of Sport Medicine abstract (1997).
16. Bottoni CR, Wilckens JH, DeBerardino TM, et al. A prospective, randomized evaluation of arthroscopic stabilization versus nonoperative treatment in patients with acute, traumatic, first-time shoulder dislocations. American Journal of Sports Medicine (2002), doi:10.1177/03635465020300041801.
17. Neyton L, Young A, Dawidziak B, et al. Surgical treatment of anterior instability in rugby union players: clinical and radiographic results of the Latarjet-Patte procedure with minimum 5-year follow-up. Journal of Shoulder and Elbow Surgery (2012), doi:10.1016/j.jse.2012.01.023.
18. Bonnevialle N, Mattési L, Martinel V, et al. Outcomes at long-term follow-up after open Latarjet versus open Bankart repair in rugby players. Orthopaedic Journal of Sports Medicine (2023), doi:10.1177/23259671231184394.
19. Sabharwal S. Surgical interventions for anterior shoulder instability in rugby players: a systematic review. World Journal of Orthopedics 6(4):400 (2015).
20. Cutteridge J, Dixon J, Garrido P, et al. A systematic review and meta-analysis of operative versus non-operative management for first time traumatic anterior shoulder dislocation in young adults. Shoulder & Elbow (2024), doi:10.1177/17585732241254693.
21. Attwood MJ, Roberts SP, Trewartha G, et al. Efficacy of a movement control injury prevention programme in adult men’s community rugby union: a cluster randomised controlled trial. British Journal of Sports Medicine 52(6):368 (2018).
22. Pasqualini I, Turan O, Hurley ET, et al. Return to sports following arthroscopic Bankart repair in collision athletes: a systematic review. Shoulder & Elbow (2024), doi:10.1177/17585732241249959.
23. Singh V, Stokes K, McKay CD. Feasibility and acceptability of a new shoulder-specific warm-up programme to prevent injuries in community youth rugby union as compared to the FIFA 11+ (2023), doi:10.1177/17479541231192170.