Most surgeons agree that a healthy person with an acute cuff tear should consider a prompt repair. Not much of a debate there.
So, instead, let’s consider the more common situation, such as a 74 year old patient comes in with a painful shoulder and an MRI report that says "full-thickness supraspinatus tear."
He asks, “If I do not have this fixed, what will happen to my shoulder?”
This is a question about a natural history that has two parts: (1) do these tears get bigger? and (2) if the tear gets bigger, will he be worse off? The literature answers the first half clearly and the second half only loosely, and the gap between the two merits some thought.
The short answers:
(1) Yes, degenerative tears can enlarge over five to ten years.
(2) No, tear enlargement does not necessarily translate into worse comfort or worse function for the patient. Long-term cohort studies that have followed patients (rather than MRI or sonographic findings) have found stable or improving patient-reported outcomes over a decade.
A third question emerges: does repairing a degenerative tear change the patient’s clinical outcome?
This patient, eight years, no repair: the function of the shoulder with the MRI shown above after 8 years of non-operative management.
Question 1. Do these tears get bigger?
Tearing is a feature of aging
Pooled individual data from 30 studies and 6,112 shoulders show the rate of rotator cuff pathology rising from 9.7% in people 20 and under to 62% in people 80 and over, with a similar age gradient whether or not the shoulder hurts. The authors concluded that cuff degeneration is common enough in people without symptoms to be considered a feature of normal human aging [1]. For that reason, we often explain this to the patient as "cuff wear" rather than as "cuff tear" as in the difference between jeans that are worn through and jeans that got ripped
That single finding informs the discussion. The question is never whether there is a tear. It is whether the person is significantly bothered by the cuff deficiency and whether repair will lead to clinically significant improvement beyond what would be accomplished with non-operative management.
Rates of enlargement
● Washington University prospective sonographic cohort, 224 subjects with an asymptomatic tear in one shoulder, median 5.1 years: enlargement in 49%, median time to enlargement 2.8 years. Stratified by final tear type — 61% of full-thickness, 44% of partial-thickness, and 14% of controls with no tear. Age and sex were not related to enlargement [2].
● The same cohort extended to patients 65 and younger, 229 participants, median 7.1 years: enlargement in 60% [3].
● Systematic review of 8 studies and 411 full-thickness tears: progression in 40.6% of asymptomatic tears at a mean 46.8 months versus 34.1% of symptomatic tears at 37.8 months, p = .65 — no difference [4].
● Systematic review of 21 studies and 1,831 tears: partial-thickness progression 26.7% at 2.2 years, full-thickness 54.9% at 3.0 years [5]. Roughly a quarter of partial-thickness tears reach full thickness [6].
Risk factors
A full-thickness cuff defect is the strongest and most consistent predictor. Smoking, dominant-arm involvement, and medium tear size have been reported [7]. Shoulder activity level — measured directly and indirectly in a prospective cohort — was not associated with tear progression or with pain development [8].
What counts as progression?
A systematic review of 1,831 tears found five different definitions of progression in use across the literature [5]. Most studies use a 5 mm threshold on MRI or ultrasound, and that threshold sits close to the limit of measurement agreement. In the MOON interobserver study, fellowship-trained shoulder surgeons agreed better than 80% on whether a tear was full- or partial-thickness, but agreement fell to roughly 60% for the amount of retraction of a full-thickness tear; kappa for grading a partial-thickness tear was −0.11, which is worse than chance [9].
Demonstrated: tears enlarge over time in a large fraction of patients. Inferred and less secure: that the millimeter figures reported year over year represent biology rather than a mixture of biology and measurement variability.
Muscle changes
Fatty degeneration and atrophy progress along with enlargement and are essentially irreversible [10]. But the relationship to symptoms is weak. In the cohort of previously asymptomatic tears, progressive fatty degeneration was associated with neither pain development nor tear enlargement [11].
This matters because fatty infiltration is often invoked as the reason to operate early. It is a reasonable argument about reparability. It is not a demonstrated argument about the patient’s clinical outcome.
Question 2. If the tear gets bigger, will he be worse off?
Cross-sectionally, no
Tear size and thickness correlate poorly with pain. The MOON group established this cross-sectionally in 393 patients with a symptomatic atraumatic full-thickness tear: no measure of tear severity — size, retraction, superior humeral head migration, or muscle atrophy — was associated with the level of pain, while comorbidity, education, and race were [29]. At ten years, neither pain intensity nor duration of symptoms predicted failure of nonoperative treatment, and patients with smaller tears were in fact more likely to come to surgery than those with larger ones [12].
Longitudinally, a real but modest association
● When previously asymptomatic tears became painful, 18% of full-thickness tears had enlarged by more than 5 mm and 40% of partial-thickness tears had become full-thickness [11].
● In 50 asymptomatic tears followed 3 years, 18 became symptomatic, and the newly symptomatic shoulders showed larger size increases [14].
● In the long-term Washington University cohort, 46% of subjects developed new pain, and pain development rose with final tear type — 28% of controls, 46% of partial-thickness, 50% of full-thickness [2].
Enlargement and pain progress together. Causation is not established: the association is between groups rather than a prediction for the individual patient.
Over ten years, patients treated without repair do not seem to deteriorate
● MOON, 452 patients with symptomatic atraumatic full-thickness tears: physical therapy was effective in more than 70%; patient-reported outcomes improved after 12 weeks and did not decline over 10 years. Of those who had surgery, 56.5% (65 of 115) did so within the first 6 months [12,13].
● Calgary prospective cohort, chronic full-thickness tears treated with a structured nonoperative program: 75% successful at 3 months, and those numbers held at 2 and 5 years [15,31]. Success there was not a score threshold. It meant that at three months the patient and the surgeon agreed that surgery was no longer appropriate because the patient had improved considerably and was largely asymptomatic. Of ten baseline characteristics tested, only the patient’s own quality-of-life score predicted whether that would happen: age, sex, duration of symptoms, traumatic versus insidious onset, forward elevation, external rotation strength, smoking, and hand dominance did not, and tear size — missing in a third of the cases and so left out of the model — differed little between the successes and the failures, 15.5 versus 17 mm [31]. Only 6 of the 70 successes worsened over the next two years, and four of those six had fallen and reinjured the shoulder [31]. At a mean of 11.4 years, the nonoperative success group’s Rotator Cuff Quality of Life score had gone from 80 at 2 years to 82 at 5 years to 86 at 11 years, and only two patients crossed over to surgery between 5 and 10-plus years [16]. Nineteen shoulders were reimaged at that visit. Eight of the seventeen successes, 47%, had a tear that had extended by a centimeter or more or retracted further since the study began, and their mean Rotator Cuff Quality of Life score was 84 of 100, range 70 to 92 [16]. Success also did not mean needing nothing further: seven of the twenty-two were still doing something for the shoulder at eleven years — home exercises, physiotherapy, or an injection [16].
This last observation is the single most useful data point for the question as posed. Over eleven years the patients who did well without repair did not merely hold steady. Their scores drifted upward. And when some of their tendons were finally looked at, the ones that had visibly worsened belonged to shoulders scoring 84 out of 100.
These cohorts deserve the same scrutiny given below to the randomized trials. Of the 452 MOON patients, 20 (5%) withdrew, 37 (9%) died, and 40 (9%) were otherwise lost by ten years — 97 patients, just over one in five, which is more loss than Moosmayer’s 19% [12]. That said, MOON’s loss is better characterized, and its largest single component is death rather than refusal. And MOON’s central claim is not a between-group difference that a handful of absent patients could change; it is that the patients who stayed nonoperative did not decline. The Calgary cohort carries its own qualifications, and they run the other way. Its patients were held to the program by a research coordinator who telephoned each of them every week, and its authors say plainly that the 75% may not be reproducible in ordinary practice without that level of adherence. Workers’ compensation claimants, litigants, and elite athletes were excluded at entry. And although every patient had been symptomatic for at least three months, 49% dated the onset to an injury — a milder version of the objection raised below to Moosmayer’s trial, though in Calgary the mode of onset made no difference to whether nonoperative treatment succeeded, p = .69 [31]. The eleven-year report carries a heavier version of the same problem I have just raised against MOON. Of 88 patients contacted, 41 responded and 39 were analyzed; 38 could not be reached at all. The score of 86 rests on 22 patients out of the 104 originally enrolled. The authors attribute most of that to a follow-up window falling inside the COVID pandemic, when they judged it wrong to bring an elderly cohort in for examination, and they were refused permission to search the regional database to learn whether the missing had died, fallen ill, or had surgery elsewhere. What they could do, they did: baseline age, sex, tear size, motion, strength, mode of onset, and quality-of-life score were compared across the responders, the partial responders, and the unreachable, and did not differ [16]. That is reassurance about who is missing. It is not a substitute for their scores.
That framing is also MOON’s limitation. Patients who deteriorated left the cohort by having surgery, so MOON reports what happens to patients who remain nonoperative, not what happens to a nonoperative strategy. MOON also performed no imaging at 10 years, so it cannot say whether outcomes stayed stable despite enlargement or because there was no enlargement [12,13]. The Calgary series, which followed both its successes and its failures for eleven years, found no significant difference between them [16].
Demonstrated: the patients these cohorts were able to measure did not deteriorate over a decade. Inferred and less secure: that the same holds for everyone who entered them.
Question 3. Does repair change the patient’s clinical outcome?
The randomized trials point in two directions; the difference between them is methodological rather than biological.
Moosmayer: repair associated with significantly better outcomes at 10 and 15 years, but not at 2 or 5 years
One hundred and three patients with full-thickness tears not exceeding 3 cm were randomized to primary tendon repair or physiotherapy with optional secondary repair, with a blinded assessor and intention-to-treat analysis. Follow-up at 15 years was 81% [19].
The between-group difference in Constant score favoring repair grew from 3.0 points at 2 years to 6.8 at 5 years, 10.0 at 10 years, and 11.8 points at 15 years — 79.9 versus 68.5, 95% CI 5.1 to 18.5, p = 0.001 (all from Table IV of the 15-year report [19]). The widening from 2 to 15 years was itself significant. Secondary outcomes at 15 years all favored repair, and the proportion analyses were more striking than the means: 70% versus 43% met the patient acceptable symptom state, and 86% versus 53% met the threshold of maximal outcome improvement predicting satisfaction [19]. The original five-year report is worth reading alongside this: it put the five-year difference at 6.5 points (95% CI −0.7 to 13.6; p = 0.08) and the constant across-follow-up difference at 5.3 points (95% CI −0.05 to 10.7; p = 0.05), and its authors concluded that the differences were small and might be below clinical importance [17]. The 6.8 above is the same time point re-estimated from a model that now spans fifteen years.
This is a Level I trial with 81% retention at fifteen years, analyzed by intention to treat with a mixed model that draws on every patient’s earlier visits rather than only on those who appeared at the end. The proportion analyses are the better argument for clinical relevance, because they count patients who crossed a threshold rather than comparing group averages, and on that measure the separation is wide.
A word on the MCID, since the rest of this post leans on it. The figure for the Constant score is 10.4 points, derived by Kukkonen from 802 operated shoulders measured before surgery and at 3 months and 1 year, anchored to a single question: is the shoulder better or worse after the operation than it was before [21]? That estimates how much one patient’s score must change for that patient to notice. It is not a threshold for a difference in mean level between two groups fifteen years out, one of them largely unoperated, and the anchor question cannot even be put to a physiotherapy patient in the same form. Published thresholds for the Constant vary considerably across studies, so 10.4 should be considered with that in mind.
Two things limit the application of the 11.8 points difference at 15 years.
● The interval. The 11.8 points is a best estimate, not the whole answer. The 95% confidence interval runs from 5.1 to 18.5, meaning the data are equally consistent with a difference as small as 5.1 points or as large as 18.5. Anything under the 10.4 MCID falls short of the threshold, and roughly the bottom 40% of that interval lies there. The low end is close to the 5.6 points the two pooled analyses of the earlier randomized data reported [22,23]. The trial shows that repair came out ahead. How far ahead is less settled than the single number makes it sound.
● The missing twenty. Eighty-three of 103 attended: 43 of the 52 randomized to repair and 40 of the 51 randomized to physiotherapy [19]. The trial flowchart accounts for the other twenty, and it largely answers the objection: 13 of them had died (6 in the repair arm, 7 in the physiotherapy arm) and only 7 were living non-attenders (3 and 4). The authors also report that patients with missing values were on average five years older than those with complete data but did not differ on any other baseline variable. For a surviving-patient bias to pull 11.8 points below 10.4, the 4 absent physiotherapy patients would have to be scoring about 15 Constant points above their 40 attending peers. That is not a plausible amount. The mixed model still assumes that what is missing is missing at random, and a patient whose unrepaired shoulder is fine has less reason to travel for a fifteenth examination — but with only four such patients in the arm that matters, this is a limitation to name rather than a reason to discount the result.
Five additional things inside the Moosmayer report
● The SF-36 showed nothing: between-group differences did not reach significance on any of the eight component scales or either summary scale. A patient can carry an 11.8-point Constant deficit and a doubled tear and register no difference in general health [19].
● The authors’ own account of their physiotherapy group. In the 26 tears followed sonographically without repair, mean anteroposterior size went from 16.2 to 31.6 mm while the Constant score fell from 73.2 at 5 years to 62.5 at 15. The authors then note that these patients showed little interest in supplementary treatment at their final visit and may have developed coping strategies allowing them to live well despite restricted shoulder function [19]. The examiner measured decline. The patients were not asking for anything.
● The crossover group caught up. Twelve patients who failed physiotherapy and underwent secondary repair had a 15-year Constant score no different from the primary repair group. The authors flag this as new, since it contradicts the 5 and 10-year finding that delayed repair was penalized, and caution that n = 12 [19]. The "operate before it is too late" argument rested on that earlier finding.
● The retear rate was 37% and barely mattered. Intact repairs scored 82.2 versus 76.1 for retears — a 6.1-point difference that sits below the 10.4-point threshold [19,21]. Whether the repair held made little clinical difference, which sits awkwardly beside the claim that the repair is what produced the benefit.
● Strength fell in both arms after year 10, in patients now in their mid-seventies. Some of the late divergence is the calendar, and the Constant score — part examiner-measured strength and motion, part patient report — is sensitive to that in a way a pure patient-reported outcome is not [19].
Demonstrated: repair produced an advantage that appeared significant at fifteen years, and more patients in the repair arm reached an acceptable symptom state. Inferred and less secure: that the advantage is as large as 11.8 points in the population from which these patients came, and that a 10.4-point threshold built for within-patient change is the right yardstick for a between-group difference.
Kukkonen: no difference at all
One hundred and eighty shoulders in patients older than 55 with isolated small full-thickness supraspinatus tears were randomized to physiotherapy, acromioplasty plus physiotherapy, or repair plus acromioplasty plus physiotherapy. At a mean 6.2 years the mean change in Constant score was 18.5, 17.9, and 20.0 (p = .84). There was no difference in pain (p = .74) or satisfaction (p = .83), which was 88–92% in all three arms [20]. The parent report of the same trial is worth stating alongside this, because the early picture was not quite flat. At one year the total Constant score did not differ (74.1, 77.2, 77.9; p = .34), nor did satisfaction (87%, 96%, 95%; p = .14), but the pain and activities-of-daily-living subscores did favor the two operated arms (p = .03 and p < .001) while motion and strength did not [30]. Whatever that early separation represented, it was gone by 6.2 years. Two further details from the 6.2-year report belong beside the headline result. The responder analysis — the proportion of patients gaining at least the 10-point Constant threshold — ran 68%, 71%, and 76% across the three arms, a nonsignificant trend in the direction of the operated groups, p = .71. And 8 of the 51 physiotherapy patients, 16%, crossed over to repair during follow-up at a mean of 20 months; they finished with a mean Constant score of 78.9, and every one of them was satisfied [20]. Waiting cost them nothing that the trial was able to measure. The parent report is less comfortable reading on that point and should be acknowledged. Four of the five who had crossed over by one year are listed individually, with 12-month Constant scores of 56, 21, 81, and 21 — all measured within a few months of their repair [30]. Those are numbers from convalescence rather than outcome, and by 6.2 years the crossover group had reached 78.9. But anyone who pulls the parent paper will find them, and they are the strongest available argument that delay is not free. The same report records a cost on the other side of the ledger. At three months the repaired arm scored significantly lower than both non-repair arms, a fall the authors attribute — largely in the strength subscore — to postoperative immobilization and delayed rehabilitation. The gap had closed by six months [30]. Repair buys a period of being worse before any later advantage arrives, and that period is measurable while the advantage in this population never became so.
Reconciling the two trials
1. Different populations. Moosmayer included tears up to 3 cm and both traumatic and nontraumatic tears. Kukkonen restricted enrollment to isolated single-tendon supraspinatus tears involving less than 75% of the tendon insertion and averaging 9 to 10 mm on MRI, 13 to 15 mm when measured at arthroscopy in patients over 55 [20,30]. Small tears in older patients appear not to benefit; larger tears in younger patients may.
2. Different denominators. MOON reports what happens to patients who remain nonoperative — those who deteriorate leave the cohort by having surgery. Moosmayer’s intention-to-treat analysis keeps its crossovers in the physiotherapy arm and they drag it down. Both are correct. They answer different questions. Kukkonen ran it both ways. Regrouping his patients by the treatment actually received rather than by allocation gave one-year Constant scores of 75.4, 78.3, and 75.6, p = .69 — so his null result does not depend on where the crossovers are counted [30].
3. Anchor to the MCID, with its caveat. The published threshold for the Constant score is 10.4 points [21]. Moosmayer sits below it through 10 years and above it only at 15, at 11.8 [18,19]. Two meta-analyses pooling the earlier randomized data found a 5.6-point difference favoring repair — statistically significant, below the 10.4-point threshold [22,23]. The Cochrane review concluded that surgery may provide no benefit over conservative treatment for symptomatic full-thickness tears [24]. The advantage is late-arriving and modest, not immediate and large.
The summary: in tears up to 3 cm in patients around 60, primary repair is associated with an advantage by fifteen years that is clinically meaningful on shoulder-specific measures. It does not show up on general health, it does not require an intact repair to appear, and it does not appear at all in small isolated supraspinatus tears in patients over 55.
Which returns us to our patient described at the outset of this post: at 74, with a chronic isolated supraspinatus tear, he sits inside Kukkonen’s enrollment criteria almost exactly and outside Moosmayer’s: the trial that showed a fifteen-year advantage for repair enrolled patients fifteen years younger than he is, and included traumatic tears that his is not — in fact only 22 of 52 repair patients and 22 of 51 physiotherapy patients had chronic atraumatic tears, the rest being acute or acute-on-chronic. The trial that most closely resembles him found no difference at all.
Does repair alter the anatomic natural history?
This is the question behind the question, and the evidence is divided.
● Against. A systematic review of studies with minimum 5-year follow-up found that repair was associated with smaller final tear size and less future surgery after adjustment, but the likelihood of a recurrent defect after repair did not differ from the likelihood of tear extension after nonoperative treatment. The authors concluded that rotator cuff repair may not alter natural history [25].
● Against. In the Finnish trial, radiographic osteoarthritis progressed significantly across the whole trial population, but there was no between-group difference in osteoarthritis or in cuff tear arthropathy, and humeral head centering deteriorated similarly in all three arms. The authors state that degeneration and eccentricity of the glenohumeral joint cannot be prevented by supraspinatus repair [20].
● For. A comparative analysis from the Washington University longitudinal study found that patients undergoing repair had pain and ASES scores that were both statistically significant and clinically relevant compared with controls whose asymptomatic tears became painful and were managed nonoperatively, and that fatty degeneration was less prevalent in the surgical group — significantly so for the infraspinatus, 17% versus 34%, though the supraspinatus difference of 25% versus 41% is reported as significant in the abstract and as not significant in the results [26]. These are two parallel cohorts rather than a randomized comparison, so allocation was not random and the groups differed by the route that brought them there. The two arms were also followed for unequal periods after their index event — a median of 5.1 years after the onset of pain in the controls against 3.0 years after surgery in the repairs — and the authors themselves allow that the control group’s inferior results may be attributable in part to that longer follow-up.
● Numerically. Assuming an 8% retear rate at two years, approximately seven patients would need to undergo repair to prevent one tear from progressing radiographically [4]. That is a number needed to treat for an imaging endpoint, not a patient-reported endpoint.
The cuff tear arthropathy question
Cuff tear arthropathy is routinely invoked as the end stage of untreated progression, usually with no rate attached. Three figures put it in proportion.
● In MOON, over 10 years, exactly one patient of 452 underwent reverse arthroplasty — 0.2% of the cohort. Over the same period 37 patients (9%) died [12,13]. The competing risk of death was roughly thirty-seven times the risk of coming to a reverse.
● In the Calgary cohort at eleven years, 4 of 21 nonoperatively treated shoulders (19%) showed superior migration of the humeral head on radiographs, and 1 of those (5%) met the radiographic criteria for cuff tear arthropathy. The repaired shoulders were no better: 2 of 9 with migration and 1 of 9 with arthropathy. Neither finding was associated with the patient’s quality-of-life score, p = .32 and p = .33 [16].
● Radiographic arthritic change does progress in shoulders with degenerative tears [27], and Hamada grade correlated negatively with Constant score in the Finnish trial — but only weakly, and repair did not prevent it [20,28].
The framing: cuff tear arthropathy is real, and it is the reason we have the reverse. It is also rare enough as a ten-year outcome of a symptomatic degenerative tear that it cannot carry the weight it is often asked to carry in the consent conversation. Absence of evidence for a high rate is not evidence that the rate is zero, but the burden of proof sits with the person asserting the risk.
Ten lines we can use in discussions with our patients
1. Tearing of the cuff is a feature of aging. Sixty-two percent of people over 80 have one, and most of them are not coming in for care [1].
2. About half of full-thickness tears enlarge over five years, and about a quarter of partial-thickness tears become full-thickness. Full thickness is the strongest predictor of enlargement; age, sex, and activity level are not [2,3,8].
3. Asymptomatic and symptomatic tears progress at the same rate, which should give pause to anyone who believes pain is the signal that the tendon is failing [4].
4. The literature uses five different definitions of progression, and the 5 mm threshold most of them share is uncomfortably close to the limit of observer agreement [5,9].
5. Enlargement and new pain progress together, but the correlation is loose, it is between groups, and fatty degeneration — the change some fear — tracks with neither pain nor enlargement [2,11].
6. Physical therapy works for more than 70% of symptomatic atraumatic full-thickness tears, and their outcomes do not decline over ten years [12,13].
7. In the Calgary cohort, the nonoperative successes were scoring better at eleven years than they were at two [15,16,31]. Nothing about the tear predicted who they would be; the only baseline predictor of success was the patient’s own quality-of-life score [31]. Half of those reimaged at eleven years had a larger tear or more retraction, and they were scoring 84 out of 100 [16].
8. Moosmayer’s advantage widens with time and reaches 11.8 Constant points at fifteen years (95% CI 5.1 to 18.5), above the 10.4-point Constant threshold — but the same patients show no difference on any SF-36 scale [19,21].
9. Repair does not prevent glenohumeral degeneration, does not prevent loss of head centering, and produces recurrent defects at about the rate that unrepaired tears extend [20,25].
10. Over ten years, one MOON patient in 452 needed a reverse. Thirty-seven died [12,13].
Closing thought
The tendon and the patient are two different subjects, and we have been much more diligent about following the first. Every study above that measured the tendon found progression. The cohorts that measured the person with patient-reported instruments — MOON and Calgary — found stability or improvement over a decade. The one trial that measured decline used an examiner-weighted score, found nothing on general health, and reported that the declining patients were not asking for help.
Codman’s End Result Idea was asking one question of every patient: how are you doing? We have spent a great deal of the last two decades asking the tendon instead. The Calgary group, which did image its patients, arrived at the same place from the other direction: they recommend following these patients annually so that anyone who deteriorates is identified and offered repair, and they say that when a patient continues to do well clinically, they do not regard the imaging findings as important to management [16].
If we asked every patient with a chronic cuff tear how they are doing, at intervals, and reimaged only when that answer changed — what would we lose, and what would we stop doing?
Time for reflection
Nashville Warbler
Matsen Backyard
Follow on twitter/X: https://x.com/RickMatsen
Follow on facebook: https://www.facebook.com/shoulder.arthritis
Follow on LinkedIn: https://www.linkedin.com/in/rick-matsen-88b1a8133//
References
[1] Teunis T, Lubberts B, Reilly BT, Ring D. A systematic review and pooled analysis of the prevalence of rotator cuff disease with increasing age. J Shoulder Elbow Surg. 2014;23(12):1913-1921.
[2] Keener JD, Galatz LM, Teefey SA, Middleton WD, Steger-May K, Stobbs-Cucchi G, Patton R, Yamaguchi K. A prospective evaluation of survivorship of asymptomatic degenerative rotator cuff tears. J Bone Joint Surg Am. 2015;97(2):89-98.
[3] Torchia MT, Sefko JA, Steger-May K, Teefey SA, Middleton WD, Keener JD. Evaluation of survivorship of asymptomatic degenerative rotator cuff tears in patients 65 years and younger: a prospective analysis with long-term follow-up. J Shoulder Elbow Surg. 2023;32(7):1432-1444.
[4] Kwong CA, Ono Y, Carroll MJ, Fruson LW, More KD, Thornton GM, Lo IKY. Full-thickness rotator cuff tears: what is the rate of tear progression? A systematic review. Arthroscopy. 2019;35(1):228-234.
[5] Garcia MJ, Caro D, Velasquez Hammerle M, Villarreal JB, DeAngelis JP, Ramappa AJ, Nazarian A. Disparities in rotator cuff tear progression definitions and rates: a systematic review. JBJS Open Access. 2024;9(4):e24.00097.
[6] Tsuchiya S, Davison EM, Rashid MS, Bois AJ, LeBlanc J, More KD, Lo IKY. Determining the rate of full-thickness progression in partial-thickness rotator cuff tears: a systematic review. J Shoulder Elbow Surg. 2021;30(2):449-455.
[7] Yamamoto N, Mineta M, Kawakami J, Sano H, Itoi E. Risk factors for tear progression in symptomatic rotator cuff tears: a prospective study of 174 shoulders. Am J Sports Med. 2017;45(11):2524-2531.
[8] Keener JD, Skelley NW, Stobbs-Cucchi G, Steger-May K, Chamberlain AM, Aleem AW, Brophy RH. Shoulder activity level and progression of degenerative cuff disease. J Shoulder Elbow Surg. 2017;26(9):1500-1507.
[9] Spencer EE Jr, Dunn WR, Wright RW, Wolf BR, Spindler KP, McCarty E, Ma CB, Jones G, Safran M, Holloway GB, Kuhn JE. Interobserver agreement in the classification of rotator cuff tears using magnetic resonance imaging. Am J Sports Med. 2008;36(1):99-103.
[10] Hebert-Davies J, Teefey SA, Steger-May K, Chamberlain AM, Middleton W, Robinson K, Yamaguchi K, Keener JD. Progression of fatty muscle degeneration in atraumatic rotator cuff tears. J Bone Joint Surg Am. 2017;99(10):832-839.
[11] Mall NA, Kim HM, Keener JD, Steger-May K, Teefey SA, Middleton WD, Stobbs G, Yamaguchi K. Symptomatic progression of asymptomatic rotator cuff tears: a prospective study of clinical and sonographic variables. J Bone Joint Surg Am. 2010;92(16):2623-2633.
[12] Kuhn JE, Dunn WR, Sanders R, Baumgarten KM, Bishop JY, Brophy RH, Carey JL, Holloway BG, Jones GL, Ma CB, Marx RG, McCarty EC, Poddar SK, Smith MV, Spencer EE, Vidal AF, Wolf BR, Wright RW. The predictors of surgery for symptomatic, atraumatic full-thickness rotator cuff tears change over time: ten-year outcomes of the MOON shoulder prospective cohort. J Bone Joint Surg Am. 2024;106(17):1563-1572.
[13] Kuhn JE, Dunn WR, Sanders R, et al; Multicenter Orthopaedic Outcomes Network (MOON) Shoulder Group. 2024 Kappa Delta Ann Doner Vaughan Award: nonsurgical treatment of symptomatic, atraumatic full-thickness rotator cuff tears — a prospective multicenter cohort study with 10-year follow-up. J Am Acad Orthop Surg. 2024;32(23):1061-1073.
[14] Moosmayer S, Tariq R, Stiris M, Smith HJ. The natural history of asymptomatic rotator cuff tears: a three-year follow-up of fifty cases. J Bone Joint Surg Am. 2013;95(14):1249-1255.
[15] Boorman RS, More KD, Hollinshead RM, Wiley JP, Mohtadi NG, Lo IKY, Brett KR. What happens to patients when we do not repair their cuff tears? Five-year rotator cuff quality-of-life index outcomes following nonoperative treatment of patients with full-thickness rotator cuff tears. J Shoulder Elbow Surg. 2018;27(3):444-448.
[16] Boorman RS, More KD, Koles SL. What happens to patients in the long term when we do not repair their cuff tears? Ten-year rotator cuff quality of life index (RC-QOL) outcomes following nonoperative treatment of patients with full-thickness rotator cuff tears. JSES Int. 2025;9(1):268-273.
[17] Moosmayer S, Lund G, Seljom US, Haldorsen B, Svege IC, Hennig T, Pripp AH, Smith HJ. Tendon repair compared with physiotherapy in the treatment of rotator cuff tears: a randomized controlled study in 103 cases with a five-year follow-up. J Bone Joint Surg Am. 2014;96(18):1504-1514.
[18] Moosmayer S, Lund G, Seljom US, Haldorsen B, Svege IC, Hennig T, Pripp AH, Smith HJ. At a 10-year follow-up, tendon repair is superior to physiotherapy in the treatment of small and medium-sized rotator cuff tears. J Bone Joint Surg Am. 2019;101(12):1050-1060.
[19] Moosmayer S, Lund G, Sire Seljom U, Haldorsen B, Svege IC, Hennig T, Pripp AH, Smith HJ. Fifteen-year results of a comparative analysis of tendon repair versus physiotherapy for small-to-medium-sized rotator cuff tears: a concise follow-up of previous reports. J Bone Joint Surg Am. 2024;106(19):1785-1796.
[20] Kukkonen J, Ryösä A, Joukainen A, Lehtinen J, Kauko T, Mattila K, Äärimaa V. Operative versus conservative treatment of small, nontraumatic supraspinatus tears in patients older than 55 years: over 5-year follow-up of a randomized controlled trial. J Shoulder Elbow Surg. 2021;30(11):2455-2464.
[21] Kukkonen J, Kauko T, Vahlberg T, Joukainen A, Äärimaa V. Investigating minimal clinically important difference for Constant score in patients undergoing rotator cuff surgery. J Shoulder Elbow Surg. 2013;22(12):1650-1655.
[22] Ryösä A, Laimi K, Äärimaa V, Lehtimäki K, Kukkonen J, Saltychev M. Surgery or conservative treatment for rotator cuff tear: a meta-analysis. Disabil Rehabil. 2017;39(14):1357-1363.
[23] Piper CC, Hughes AJ, Ma Y, Wang H, Neviaser AS. Operative versus nonoperative treatment for the management of full-thickness rotator cuff tears: a systematic review and meta-analysis. J Shoulder Elbow Surg. 2018;27(3):572-576.
[24] Karjalainen TV, Jain NB, Heikkinen J, Johnston RV, Page CM, Buchbinder R. Surgery for rotator cuff tears. Cochrane Database Syst Rev. 2019;12:CD013502.
[25] Chalmers PN, Ross H, Granger E, Presson AP, Zhang C, Tashjian RZ. The effect of rotator cuff repair on natural history: a systematic review of intermediate to long-term outcomes. JBJS Open Access. 2018;3(1):e0043.
[26] Hill JR, Olson JJ, Sefko JA, Steger-May K, Teefey SA, Middleton WD, Keener JD. Does surgical intervention alter the natural history of degenerative rotator cuff tears? Comparative analysis from a prospective longitudinal study. J Shoulder Elbow Surg. 2025;34(2):430-440.
[27] Chalmers PN, Salazar DH, Steger-May K, Chamberlain AM, Stobbs-Cucchi G, Yamaguchi K, Keener JD. Radiographic progression of arthritic changes in shoulders with degenerative rotator cuff tears. J Shoulder Elbow Surg. 2016;25(11):1749-1755.
[28] Hamada K, Fukuda H, Mikasa M, Kobayashi Y. Roentgenographic findings in massive rotator cuff tears: a long-term observation. Clin Orthop Relat Res. 1990;(254):92-96.
[29] Dunn WR, Kuhn JE, Sanders R, An Q, Baumgarten KM, Bishop JY, Brophy RH, Carey JL, Holloway GB, Jones GL, Ma CB, Marx RG, McCarty EC, Poddar SK, Smith MV, Spencer EE, Vidal AF, Wolf BR, Wright RW. Symptoms of pain do not correlate with rotator cuff tear severity: a cross-sectional study of 393 patients with a symptomatic atraumatic full-thickness rotator cuff tear. J Bone Joint Surg Am. 2014;96(10):793-800.
[30] Kukkonen J, Joukainen A, Lehtinen J, Mattila KT, Tuominen EKJ, Kauko T, Äärimaa V. Treatment of non-traumatic rotator cuff tears: a randomised controlled trial with one-year clinical results. Bone Joint J. 2014;96-B(1):75-81.
[31] Boorman RS, More KD, Hollinshead RM, Wiley JP, Brett K, Mohtadi NG, Nelson AA, Lo IKY, Bryant D. The Rotator Cuff Quality-of-Life Index predicts the outcome of nonoperative treatment of patients with a chronic rotator cuff tear. J Bone Joint Surg Am. 2014;96(22):1883-1888.
