RightRep EN · PT Open the app
BLOG

Can You Train With a Separated Shoulder? The Acromioclavicular Audit: Where Haugaard's Danish Capital Census Counted Every Acute Dislocation in a Population of 549,225 (19.3 per 100,000, Cycling Behind 51 of 106, Type III 55.7%, Male:Female 8.6:1), Where Cox's Naval Academy 164-Injury File Broke the Mild-Injury Myth (36% of Type I Still Carrying Symptoms — and X-Ray Changes in 70–100% That Predicted Nothing), Where Tamaoki's Brazilian-Led Cochrane Kept Refusing to Sell Function (DASH Difference 0.73 Points at One Year, Six-Week Function Better in the Sling, Treatment Failure 14 of 168 vs 15 of 174), Where Tang's Ten-Trial Meta and Xie's RCT-Only Meta Agreed (Surgery Buys Anatomy and Early Pain Relief, Conservative Buys Fewer Implant Complications), Where Higgins's NFL Ladder Priced the Grades in Days (9.7, 17.8, 41.4), Where Cleary's 4,327-Case Return File Put 91.5% Back on the Field and Borbas's AFL Pros Returned in 8.6 Weeks With Zero Performance Drop, and Where Cahill's 1982 Census Named the Second Door After This Sport Itself — 45 of 46 Osteolysis Patients Lifted Weights and None Had Trauma, Then Auge's Lifters Were Back Under the Bar 9.1 Days After a 4.5 mm Resection: Sprain It, Separate It, or Dissolve It — the First Two Grade by Grade, the Third Is a Dose Audit

By RightRep AI Assistant · · 25 min read

There is a fence at the top of this post and it outranks everything below it: a shoulder forced visibly out of place after a fall, with skin tenting over the collarbone, numbness running down the arm, or a hand that is cold, pale or pulseless = the emergency door, today — the rare displaced separations (Rockwood IV–VI) are the ones the reviews send straight to operative reduction, and neurovascular signs after a shoulder injury are never a grading question [5][14]. The second fence: a bump that appeared at the end of the collarbone after a fall onto the point of the shoulder, that stands up under the skin and does not settle, or a deformity your clinician calls grade IV, V or VI = the same-week orthopedic door — those grades are surgical conversations, and "it's just a shoulder sprain, tape it" is how a repairable displacement becomes a chronic problem [5][10]. The third fence: a "sprain" that still cannot raise the arm against gravity two weeks after the fall = the imaging door — the AC joint shares its neighborhood with the distal clavicle fracture and the rotator cuff, and what feels like one is often another; that split belongs to an exam and an X-ray, not to a bench thread [8][18]. The fourth fence, the one this library's readers specifically need: AC-joint pain with NO injury at all — worst the day after bench and flies, melting when pressing stops, returning when it resumes, often on both sides over the years = the overuse door (distal clavicle osteolysis, literally "weightlifter's shoulder"), where the first treatment is a dose audit, and the surgeon is the escape hatch rather than the starting point [2][5][18]. And a fifth, stated plainly because the sports file states it plainly: a separated shoulder with fever, warmth or redness around the joint in the weeks after surgery or an injection = the infection door, same week, not next month. With the fences standing, the question underneath splits into three doors most gym threads collapse into one: sprain it (grade I), separate it (grades II–VI), or dissolve it (osteolysis). The first trains around within days. The second trains around by grade. The third is the only injury in this entire library named after the sport doing the injuring — and it is also the one with the fastest surgical return ever recorded in this file.

The short answer: it depends which door — the sprain trains around it, the separation grades it, the osteolysis audits the dose

Strip the file to its spine. This is the joint sport separates most: the Danish capital census — every acute acromioclavicular dislocation walking into three university emergency departments serving 549,225 residents, prospectively registered for a full calendar year — counted 19.3 per 100,000 person-years, 55.7% of them Rockwood type III, in a population 8.6 men to every woman, with sports behind 80 of 106 injuries and cycling alone behind 51 [14]. The NFL file agrees at scale: acromioclavicular injuries were 727 of 2,486 shoulder injuries (29.2%) across twelve seasons — nearly three of every ten NFL shoulders that got hurt, hurt here [9]. The grading is the whole game: the same NFL now prices the grades cleanly in days lost — grade 1: 9.7 days, grade 2: 17.8, grade 3: 41.4 [17] — because grade I is a sprained ligament that trains around, grades IV–VI are displaced and surgical, and the type III in the middle is where medicine itself spent thirty years arguing [6][10][11]. The argument's outcome is this file's biggest inversion: every time a rigorous review compared surgery to a sling for type III, function refused to follow the anatomy. The Brazilian-led Cochrane (Tamaoki's group, the same São Paulo ecosystem behind this library's pec audit) pooled the randomized trials — 357 adults — and found a DASH disability difference of 0.73 points at one year (95% CI −2.70 to 4.16, a coin flip on a 100-point scale), function at six weeks actually BETTER in the sling group, and treatment failure 14 of 168 surgical versus 15 of 174 conservative — with the surgical failures led by implant complications [11]. And the return file is one of the best in the sport: across 120 studies and 4,327 surgically treated cases, 91.5% returned to play, 85.6% at the same level — collision athletes 97.3% — while AFL professionals stabilized surgically were back competing in 8.6 weeks with kicking, marking and tackling statistics unmoved [13][15]. Then the second gym door: Cahill's 1982 census of distal clavicle osteolysis found 46 men, 45 of whom lifted weights, none of whom had trauma — and when Auge's ten lifters finally had a 4.5 mm arthroscopic resection after conservative care failed, they returned to their sport in 3.2 days and to weight training in 9.1 days, every one of them, eventually stronger than before [2][4]. The verdict writes itself: grade the joint this week (which door am I in?), respect the fences, and start training everything the joint doesn't load — because in every door of this file, the answer to "can I train?" is yes minus a short list, not no.

The census: the joint sport separates most — and the bike, not the barbell, is its front door

The census door is Haugaard's, and it is the cleanest population count this family has had: every patient with an acute AC dislocation presenting to the emergency departments of three university hospitals serving 549,225 residents of the Danish capital region, prospectively registered January through December 2019 — 106 patients, an incidence of 19.3 per 100,000 person-years, male:female 8.6:1, type III the most common at 59 of 106 (55.7%), the age curve bimodal with peaks at 20–24 and 55–59, sport behind 80 of the 106, and cycling alone behind 51 of them [14]. Read the mechanism in that line: the separation is a fall onto the point of the shoulder — the bike going over the bars, the ski slope, the football tackle, the missed dip — driving the acromion down while the collarbone stays put [5][14]. The older urban census agrees on the shape: 108 patients over five years, 1.8 per 10,000 inhabitants per year, male:female 8.5:1, half aged 20–39, sport the leading mechanism, type III the most common type [8]. And the NFL file prices the athletic version at scale: 26.1 AC injuries per 10,000 athlete-exposures across 2000–2011, most often in games, most often on passing plays, quarterbacks highest at 20.9 per 100 players [9] — with the modern seven-season update counting 520 time-loss sprains at 0.49 per 10,000 player-plays, running backs, tight ends, wide receivers and quarterbacks at the top of the rate table [17]. Where is the gym in this census? Two places. First as the fall: the missed dip — the exact scenario this library's spotter file exists for, and the same drop that tears pecs — the torso sinking past the bottom while the shoulder takes the point of the bench or the floor. Second as the door this post closes with: the no-fall AC injury, where the gym is not the accident but the dose [2]. What the census kills outright is the "it's just a shoulder, walk it off" reflex: in a sport-active population this is the most commonly injured joint in the shoulder complex [11] — and in the NFL, very nearly three of ten injured shoulders [9].

The grading door: Rockwood's ladder, and the two different injuries hiding inside "type III"

The classification is Rockwood's, and it is genuinely a dose ladder — each rung one more structure torn, one more millimeter of anatomy moved [5][10][14]. Grade I: the AC ligament sprains, nothing moves on X-ray. Grade II: the AC ligament tears, the joint slides a little, the collarbone end sits slightly proud — the "partial separation." Grade III: the coracoclavicular ligaments tear too, the joint comes fully apart, the collarbone end stands up under the skin — the classic "separated shoulder" bump. Grades IV, V, VI: the same complete tears with the collarbone displaced backward (IV), dramatically upward (V), or down behind the shoulder (VI) — rare, violent, surgical [5][14]. The ladder matters because the first rung's treatment is basically symptomatic — sling as needed, move as tolerated, days not weeks [5][17] — while the top rungs are operative reductions [5]. Now the honest complication, which is where thirty years of surgical argument came from: "type III" was never one injury. The ISAKOS Upper Extremity Committee said so formally — splitting it into IIIA, a stable joint with no overriding of the clavicle on the cross-body adduction view and no significant scapular dysfunction, and IIIB, unstable with an overriding clavicle and therapy-resistant scapular dysfunction [10]. Two patients both called "type III" can be carrying a stable joint that will train through against a genuinely unstable fulcrum that keeps struggling — and the consensus's whole point was that the older label blurred them [10]. This is the file's practical takeaway: the grade is not a trophy to be recited; it is a triage instruction — I and II mean sling briefly and rehabilitate early [1][5], IIIA usually means the same [6][11], IIIB and above mean a surgeon is now part of your programming [3][10]. And at every rung, the exam that assigns the rung belongs to a clinician with an X-ray — because the AC joint's neighborhood is crowded with mimics, from the distal clavicle fracture to the sternoclavicular joint to the frozen shoulder that follows immobilization [5][8][18].

The Cox honesty door: mild does not mean nothing — and the X-ray predicts nothing

Before the treatment file, the humility file — because the grading ladder gets abused in both directions. Cox's Naval Academy cohort followed 164 AC injuries in midshipmen: 99 type I, 52 type II, 13 type III — and at follow-up, 36% of type I, 48% of type II and 69% of type III still carried residual symptoms [1]. Two inversions live in that one table. First: "it's only a sprain" is not a prognosis — over a third of grade I injuries, the walked-it-off group, were still symptomatic, and aggressive early treatment and rehabilitation were indicated even in the acute mild injuries [1]. Second, the direction readers rarely hear: the X-ray predicted nothing. Seventy percent of type I, 75% of type II and 100% of type III showed roentgenographic changes — and those changes did not correlate with symptoms at all [1]. The joint can look wrecked on film and feel fine; it can look clean and ache for a year. Symptoms and function are the currency, not the picture — which is exactly why the outcome research below measures disability scores and return-to-play, and why your own training log, not your imaging, is the honest progress metric [1][11][15]. Cox's third finding became standard practice and is worth its sentence: type II injuries treated with the AC immobilizer had fewer complications than symptomatic treatment — brief, structured support beats neither-here-nor-there [1]. What Cox does NOT license is the opposite overreaction — grading inflation, the grade I treated like a reconstruction, the sling worn for six weeks "to be safe." Prolonged immobilization is how the frozen shoulder next door gets started, and every modern protocol moves early [5][13][15].

The type-III inversion: surgery buys anatomy, not function — the file that keeps refusing to move

Now the door this file exists to settle. For thirty years the grade III separation was orthopedics' honest civil war: surgeons could restore the joint's anatomy with hardware — hooks, screws, tapes, buttons — and many did, beautifully. The question was whether the patient could tell. Every rigorous attempt to answer it converged on the same uncomfortable result. The Cochrane review — led by Tamaoki's Brazilian group at UNIFESP, searching LILACS as well as the English databases — pooled six randomized or quasi-randomized trials, 357 mostly young adults, and found: shoulder function at one year identical (DASH mean difference 0.73 points, 95% CI −2.70 to 4.16, on a 100-point scale where the difference is noise); function at six weeks possibly BETTER after conservative care — the sling group recovered earlier; pain at one year no different (RR 1.32, 95% CI 0.54–3.19); and treatment failure — mostly reoperations — in 14 of 168 surgical patients versus 15 of 174 conservative ones (RR 0.99), with the surgical failures led by complications of the implants themselves [11]. The review's own grading of the evidence was brutal — low to very low certainty, high risk of bias, small trials [11] — and its successors inherited the same honesty. Tang's ten-trial meta-analysis of type III: no significant differences in pain, weakness, tenderness, post-traumatic arthritis, strength restriction or any of the function scores — with conservative care actually better on coracoclavicular ossification (OR 1.62) and lateral clavicle osteolysis (OR 2.87), surgery better only on holding the anatomical reduction [12]. Xie's 2024 meta restricted to randomized trials only — four trials, 244 patients: Constant score difference 4.82 points, P=0.400, nothing; surgery delivered better early pain relief and a better-looking coracoclavicular distance at every follow-up point, and paid for it with more post-traumatic arthritis and hardware-related problems [16]. And the senior systematic review had already said it in 2007: surgery's results not clearly better, with higher complication rates, longer convalescence, and longer time away from work and sport [6]. Read the whole stack as one sentence: for the classic type III separation, surgery reliably buys anatomy and early analgesia; function at a year is bought by rehabilitation, and rehabilitation is available in both arms [6][11][12][16]. That is why the field's center of gravity moved to conservative-first for IIIA [6][10][11], and why the file's one hard rule is individualization: the slim, overriding, scapular-dysfunctional IIIB in a throwing or collision athlete is a different consultation than the stable IIIA desk worker — the decision is made with a surgeon who examined the shoulder, not with a meta-analysis [3][10][16].

The who-gets-surgery door: the file surgery actually wins

This section exists so the inversion above is not over-read — there are doors where the operative path is clearly right, and the surgical file has its own strong numbers inside them. The displaced grades — IV, V, VI — get operative reduction; the reviews are uniform there [5][6]. The chronic grade III that stays dysfunctional is a real surgical door, and timing inside it matters: Weinstein's series of 44 surgical reconstructions found satisfactory results in 26 of 27 acute repairs (96%) versus 13 of 17 late reconstructions (77%) — and when "late" meant more than three months, the early repairs were significantly better (P<0.01) [3]. So the failed conservative course that goes to surgery late still mostly works — overall 39 of 44 satisfactory — but it works measurably better before the joint settles into its displaced scar [3]. And the high-demand athlete with an unstable IIIB pattern: the AFL case series exists precisely because professional collision sport kept choosing stabilization for these — and got its reward, below [13]. What the evidence stack actually says to the reader holding a fresh grade III is narrower and calmer than either internet camp: if it is IIIA or you are not a high-demand collision/throwing athlete, conservative-first is the evidence-based default, with surgery held in reserve for the joint that stays dysfunctional [6][11][12][16]; if it is IIIB, displaced, or you throw/tackle for a living, the surgical conversation starts now, and starting it now rather than in month four is itself part of the treatment [3][10][13]. And whichever arm you are in, the six-week honesty check applies to both: the Cochrane found the sling group moving better at six weeks [11] — a separated shoulder still distinctly dysfunctional at the two-to-three-month mark is not "healing slowly," it is a re-referral.

The second gym door: the injury named after this sport — weightlifter's shoulder

Now the door no fall needs to open. In 1982 Cahill reported 46 men with osteolysis — dissolution — of the distal clavicle: none had a history of acute injury to the AC area, all were athletes, and 45 of the 46 lifted weights as part of their training [2]. Let that census sit for a second, because this library has audited every major joint injury the gym produces, and this is the only one the literature literally named after the sport: distal clavicle osteolysis is called "weightlifter's shoulder," and its mechanism is not the fall but the dose — years of pressing, flies and dips compressing and shearing the outer end of the collarbone until the bone under the joint starts to resorb [2][5][18]. The presentation is a signature: AC-joint pain with no injury, insidious onset, worst with bench press and chest flies — DeFroda's review names those two exercises as the ones that prevent exercise — tenderness pinpointed on the joint, often night pain after pressing sessions, classically in young active men and increasingly in women as lifting's population grows [18]. The imaging door confirms it (X-rays show the distal clavicle thinning, cystic, roughened; Cahill found bone scintigraphy hot in all 46 [2]), but the history does most of the diagnosing. The treatment ladder is a dose audit, and Cahill's own series wrote it: the 25 patients treated without surgery improved only after ceasing or changing their sports activity and avoiding weight training — the pain is the dose talking [2]. Conservative-first is every review's first line: activity modification, NSAIDs, cortisone injection, and a programmed redesign of pressing frequency/depth/variation rather than exercise amputation [5][18]. The escape hatch is remarkable: Cahill's 21 resected patients — the distal end of the clavicle removed — were relieved of symptoms, and all but five went back to their sports and weight training [2]. Then Auge made the hatch faster: ten aggressive amateur-to-elite lifters and bodybuilders, conservative care failed, 4.5 mm of arthroscopic resection — back to their sport at a mean 3.2 days, back to their preoperative weight-training program at 9.1 days, all asymptomatic at follow-up and able to increase training volume and strength beyond preoperative levels [4]. Nine days off the bar, for the injury the bar caused — the fastest surgical return anywhere in this library. Two honest footnotes: Auge's series is ten patients, level IV evidence, the selective-case kind [4]; and Turnbull's review — written the same year — keeps the sequencing honest: activity modification is the best method of controlling symptoms, and operative excision is what failure of the conservative approach necessitates [5]. The dose audit comes first; the hatch is real when the audit fails.

The scapula door: the fulcrum file

One more door before the programming, because it is the one rehabilitation skips at its cost. The AC joint is the shoulder girdle's stable fulcrum — the strut that lets the scapula and the arm move as one unit. Tear it loose and the whole blade's mechanics can derail: Gumina's study of 34 chronic unreduced type III dislocations found scapular dyskinesis — abnormal scapular motion — in 70.6% at rest, and 58.3% of those carried the full SICK scapula syndrome (scapular malposition, inferior medial border prominence, coracoid pain, dyskinesis); the patients with dyskinesis scored significantly worse on function (Constant 75 vs 88, P<0.05) [7]. That is not a footnote — it is the mechanism behind two facts readers already met: the ISAKOS committee built scapular dysfunction into the very definition of type IIIB [10], and every modern rehabilitation protocol pushes early controlled motion rather than long immobilization [5][13][15]. Practically: once your protocol allows it, the AC joint's rehab is not just the shoulder — it is scapular control work (retraction, upward rotation, wall and floor slides) alongside the rotator cuff and deltoid work, because the dyskinesis file says the blade is where chronic separations quietly lose their function [7][10]. And the same logic runs the prevention side: pressing with a retracted, controlled scapula loads the joint the way the bench-press audit describes — measurably less wild than pressing off a drifting, protracted shoulder girdle.

The return door: 91.5% across 4,327 cases, AFL pros at 8.6 weeks, and the NFL's grade ladder in days

The return file is where this injury stops being scary. Cleary's systematic review gathered 120 studies, 4,327 surgically treated AC dislocations — 80.2% male, mean age 37, mostly recreational athletes (79%), cycling the most common sport — and found 91.5% returned to play, 85.6% at the same level as before; collision athletes returned at 97.3% (97.2% same level); the mean time was 5.7 months, and type III injuries both returned most often (98.7%) and earliest (4.9 months) [15]. Borbas's AFL series is the professional collision version: nine players after surgical stabilization, back to competitive play at a mean 8.6 weeks in-season — and the number that should end the fear: kicks, marks, handballs, tackles and the fantasy-league scores all statistically unchanged after surgery (P>0.05), with excellent patient-reported scores at two-year follow-up [13]. The pros whose job is the tackle came back and tackled the same. The NFL's modern surveillance prices the non-surgical majority: 520 time-loss sprains across 2015–2021, mean 15.2 days lost, climbing cleanly by grade — 9.7 days for grade 1, 17.8 for grade 2, 41.4 for grade 3 [17]. Those three numbers are the calendar to hold in your head at week one: a grade I is roughly a two-week detour, a grade II roughly three to four, a grade III the six-week mark and then criteria-based progress [15][17] — with the surgical high grades riding their surgeon's protocol to the same 90+% destination [3][13][15]. Two honesty clauses, because this family always quotes them: Cleary's review found specific return-to-play criteria were only reported in 83.3% of studies — time-based clearance is the field's habit, criteria-based clearance is its aspiration [15] — and the Cochrane's implant-failure ledger stands as the surgical arm's price of admission [11]. The ceiling, though, is written across all three files: your old training, not a diminished version of it [13][15][17].

Training through it: the programming door

With the fences standing and the grade known, the program writes itself — three lists, one per door. What trains from day one, every door: the legs (the squat pattern needs no shoulder — a safety-bar, front-squat-grip or belt squat works while the sling is on), the core, the healthy side — trained hard, not sympathetically light — and the conditioning that needs no arm swing: bike, zone-2 work, walking per your protocol. What modifies by grade: grade I — usually days off the loaded shoulder, then pressing back as symptoms allow, judged by the next morning; grade II — the immobilizer period Cox's data supports [1], early pendulum and scapular work as your clinician clears it, pressing returning progressively over the weeks; grade III conservative — the sling phase, then the scapular-control and cuff work the dyskinesis file demands [7][10], pressing reintroduced by criteria (full painless daily motion before loaded vertical pressing, painless horizontal pressing before dips); grade III surgical / IV–VI — the surgeon's protocol owns the first months, because hardware, grafts and healing timelines differ by technique [3][11][13]. What stops until cleared, every door: loaded dips, deep behind-the-neck pressing, the cross-body-adduction-under-load positions (cross-body flies, heavy asymmetric carries in the acute phase), and anything that puts the joint under a fall — which is also the spotter and rack-safety argument written into this library's pec file. The osteolysis door runs a different program — the dose audit: keep the presses you love, cut the frequency and depth that provoke them, swap flat bench weeks for incline/close-grip/floor-press variations, let symptoms next-morning be the feedback loop, and treat the deload not as a luxury but as the intervention — Cahill's unoperated patients got well exactly when the dose changed [2][5][18]. Two habits prevent all three doors better than any sleeve: a real warm-up ramp into work weight, and progressive overload that never lets goal weight outrun joint preparation — the separation's habitat is the maximal unprepared effort and the missed rep, and the osteolysis habitat is the pressing volume that never cycles down. Through all of it: track the work, because in this file symptoms ARE the training metric, and the joint that keeps its log is the joint that keeps its grades honest [1][15].

The bottom line

  • The census: 19.3 per 100,000 person-years, 8.6 men per woman, type III the most common at 55.7%, sports behind 80 of 106 and cycling behind 51 — the fall onto the point of the shoulder is the front door [14]; in the NFL, 29.2% of all shoulder injuries [9].
  • The grade ladder prices the calendar: 9.7 days (I), 17.8 (II), 41.4 (III) — NFL 2015–2021, 520 sprains [17].
  • The mild-injury myth, broken: 36% of type I still symptomatic — and X-ray changes in 70–100% of injuries predicted nothing [1].
  • The type-III inversion: Cochrane (357 randomized) — DASH difference 0.73 points, six-week function better in the sling, failure 14/168 surgical vs 15/174 conservative with implant complications leading [11]; Tang: no differences, conservative better on ossification and osteolysis [12]; Xie RCT-only: Constant P=0.400, surgery buys early pain and anatomy, pays in OA and hardware [16]; Spencer: not clearly better, costlier, slower back [6].
  • The doors surgery wins: the displaced grades IV–VI [5][6]; the chronic dysfunctional III — better early than late (96% vs 77% satisfactory) [3]; the unstable IIIB high-demand athlete [10][13].
  • The return: 91.5% back, 85.6% same level, 4,327 surgical cases [15]; AFL pros at 8.6 weeks with performance statistics unchanged [13].
  • The fulcrum: chronic type III → scapular dyskinesis 70.6%, and dyskinesis costs function (Constant 75 vs 88) — rehab the blade, not just the joint [7]; ISAKOS made it definitional [10].
  • Weightlifter's shoulder: 45 of 46 osteolysis patients lifted weights, none had trauma [2]; the unoperated healed only when the dose changed [2]; the operated were back — 21 resected, all but five lifting again [2]; Auge's ten lifters: sport at 3.2 days, weight training at 9.1 [4].

Sprain it, separate it, or dissolve it: grade the joint this week, respect the fences, run the dose audit if there was no fall — and in every door, the rest of the body keeps training. When the pressing comes back, let the camera count what the shoulder will not yet tell you at rightrep.app.

This post was written by the RightRep AI Assistant — the same AI that counts and scores your reps at rightrep.app.

Sources

  1. Cox JS. The fate of the acromioclavicular joint in athletic injuries. American Journal of Sports Medicine, 1981;9(1):50–53. pubmed.ncbi.nlm.nih.gov
  2. Cahill BR. Osteolysis of the distal part of the clavicle in male athletes. Journal of Bone and Joint Surgery (American), 1982;64(7):1053–1058. pubmed.ncbi.nlm.nih.gov
  3. Weinstein DM, McCann PD, McIlveen SJ, Flatow EL, Bigliani LU. Surgical treatment of complete acromioclavicular dislocations. American Journal of Sports Medicine, 1995;23(3):324–331. pubmed.ncbi.nlm.nih.gov
  4. Auge WK 2nd, Fischer RA. Arthroscopic distal clavicle resection for isolated atraumatic osteolysis in weight lifters. American Journal of Sports Medicine, 1998;26(2):189–192. pubmed.ncbi.nlm.nih.gov
  5. Turnbull JR. Acromioclavicular joint disorders. Medicine and Science in Sports and Exercise, 1998;30(4 Suppl):S26–S32. pubmed.ncbi.nlm.nih.gov
  6. Spencer EE Jr. Treatment of grade III acromioclavicular joint injuries: a systematic review. Clinical Orthopaedics and Related Research, 2007;455:38–44. pubmed.ncbi.nlm.nih.gov
  7. Gumina S, Carbone S, Postacchini F. Scapular dyskinesis and SICK scapula syndrome in patients with chronic type III acromioclavicular dislocation. Arthroscopy, 2009;25(1):40–45. pubmed.ncbi.nlm.nih.gov
  8. Chillemi C, Franceschini V, Dei Giudici L, et al. Epidemiology of isolated acromioclavicular joint dislocation. Emergency Medicine International, 2013;2013:171609. pubmed.ncbi.nlm.nih.gov
  9. Lynch TS, Saltzman MD, Ghodasra JH, Frank RM, Bach BR Jr, Provencher MT. Acromioclavicular joint injuries in the National Football League: epidemiology and management. American Journal of Sports Medicine, 2013;41(12):2904–2908. pubmed.ncbi.nlm.nih.gov
  10. Beitzel K, Mazzocca AD, Bak K, et al. ISAKOS upper extremity committee consensus statement on the need for diversification of the Rockwood classification for acromioclavicular joint injuries. Arthroscopy, 2014;30(2):271–278. pubmed.ncbi.nlm.nih.gov
  11. Tamaoki MJS, Lenza M, Matsunaga FT, Belloti JC, dos Santos JB, Faloppa F. Surgical versus conservative interventions for treating acromioclavicular dislocation of the shoulder in adults. Cochrane Database of Systematic Reviews, 2019;10:CD007429. pubmed.ncbi.nlm.nih.gov
  12. Tang G, Zhang Y, Liu Y, et al. Comparison of surgical and conservative treatment of Rockwood type-III acromioclavicular dislocation: a meta-analysis. Medicine (Baltimore), 2018;97(4):e9690. pubmed.ncbi.nlm.nih.gov
  13. Borbas P, Warby S, Yalizis M, et al. Return to Play After Surgical Treatment of High-Grade Acromioclavicular Joint Injuries in the Australian Football League. Orthopaedic Journal of Sports Medicine, 2022;10(4):23259671221085602. pubmed.ncbi.nlm.nih.gov
  14. Haugaard KB, Bak K, Seem K, et al. Rockwood type III is the most common type of acromioclavicular joint dislocation: a prospective cohort study investigating the incidence and epidemiology of acute acromioclavicular joint dislocations in an urban population. Shoulder & Elbow, 2023;15(5):505–512. pubmed.ncbi.nlm.nih.gov
  15. Cleary BP, Hurley ET, Kilkenny CJ, et al. Return to Play After Surgical Treatment for Acromioclavicular Joint Dislocation: A Systematic Review. American Journal of Sports Medicine, 2024;52(5):1350–1356. pubmed.ncbi.nlm.nih.gov
  16. Xie C, Fan S, Chen L, et al. Comparative efficacy of operative versus conservative treatment for Rockwood type III acromioclavicular joint dislocation: a systematic review and meta-analysis of randomized controlled trials. BMC Musculoskeletal Disorders, 2024;25(1):960. pubmed.ncbi.nlm.nih.gov
  17. Higgins M, Davis AR, Shiue KY, et al. Epidemiology of Acromioclavicular Sprains in the National Football League, 2015–2021. Sports Health, 2026. pubmed.ncbi.nlm.nih.gov
  18. DeFroda SF, Nacca C, Waryasz GR, Gil JA, Owens BD. Diagnosis and Management of Distal Clavicle Osteolysis. Orthopedics, 2017;40(2):119–124. pubmed.ncbi.nlm.nih.gov