Disorders of the rotator cuff are among the leading causes of musculoskeletal disability and physician visits.1–3 Despite the tremendous volume of literature on the treatment of rotator cuff tears, surgical indications and timing remain controversial.4,5 Factors that might influence the decision to undergo rotator cuff repair include tear properties such as thickness, location (articular- or bursal-sided), size, chronicity, amount of retraction, and muscle atrophy, as well as patient characteristics such as age, functional demands, smoking history, and being medically fit to undergo an operation.
The initial management for most rotator cuff disorders, particularly partial-thickness tears, tendonitis, chronic tears in older patients, and irreparable tears, is nonoperative.4 Even in cases that are more amenable to early rotator cuff repair, such as symptomatic acute full-thickness tears or chronic full-thickness tears in younger patients, surgery is often delayed for a number of reasons, such as reluctance to undergo surgery as well as timing with work- and family-related obligations. For acute full-thickness rotator cuff tears, several studies have demonstrated improved patient-reported outcomes, range of motion, and satisfaction with early repair within 3 to 4 months.6–8
Furthermore, tear progression, tendon retraction, and potentially irreversible degradation in muscle and tendon quality are additional risks associated with non-operative management.9–11 In addition, the tension required to repair the supraspinatus to its footprint may be significantly higher after as little as 16 weeks following injury.12 It is unclear whether delayed rotator cuff repair is thus associated with increased failure rates and need for subsequent revision rotator cuff surgery. Therefore, with the use of a national longitudinal insurance database, the purpose of this study was to determine the association between time from the diagnosis of rotator cuff tear to repair and the rate of subsequent revision surgery for re-tear. This study aimed to answer two questions: (1) What is the association between time from the diagnosis of rotator cuff tear to repair and the rate of subsequent revision surgery for re-tear? (2) What effect does age, sex, and comorbidity burden have on the rate of revision rotator cuff repair relative to time to repair? The authors hypothesized that there would be a higher rate of revision surgery for patients who underwent delayed rotator cuff repair beyond 12 months from initial diagnosis.
Materials and Methods
A retrospective query of the Humana national private-payer insurance claims database from 2007 to 2016 was performed through the PearlDiver Direct Research Program (PearlDiver Technologies Inc, Colorado Springs, Colorado). Humana is a full-claims database that pools both commercial and Medicare populations. This database contains deidentified patient records searchable by Current Procedural Terminology (CPT) and International Classification of Diseases, Ninth Revision (ICD-9) codes. Given the use of a deidentified insurance claims database, this study was exempt from institutional review board approval.
Patients with a diagnosis of complete or partial shoulder rotator cuff tear using ICD-9 codes 726.13 and 727.61 who subsequently underwent arthroscopic rotator cuff repair as noted by CPT code 29287 with a minimum of 5-year follow-up were included in the study. A total of 48,071 patients with a rotator cuff repair procedure code were identified, of whom 40,752 were excluded due to less than 5-year follow-up, 1101 were excluded due to lack of laterality coding, and 3459 were excluded due to lack of a rotator cuff tear diagnosis code. Therefore, 2759 patients were included in the analysis.
The time to surgery was defined as the period that elapsed between the first incidence of a rotator cuff tear diagnosis code and the first subsequent rotator cuff repair procedure code. On the basis of this time interval, patients were categorized into 3 cohorts: early repair (<6 weeks), routine repair (between 6 weeks and 12 months), or delayed repair (>12 months). Patient characteristics including age, sex, and overall medical comorbidity burden as measured with the Charlson Comorbidity Index (CCI)13 were also collected.
The primary study outcome was revision rotator cuff repair. This was determined by the presence of a second CPT code—29287—following the initial repair procedure, matched for laterality using laterality modifier codes. Bivariate analysis of revision repair rates between the time to repair cohorts was performed using Pearson's chi-square tests. A multivariate logistic regression model with time to rotator cuff repair, age, sex, and CCI as covariates was used to determine the independent association between time from the diagnosis of rotator cuff tear to repair and the risk of subsequent revision surgery. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. Statistical significance was set at P<.05.
Time From Rotator Cuff Tear Diagnosis to Repair and the Rate of Subsequent Revision Surgery for Re-tear
A total of 2759 patients with minimum 5-year follow-up were included in the analysis, with 1510 (54.7%) undergoing early repair within 6 weeks of diagnosis, 1104 (40.0%) undergoing routine repair between 6 weeks and 12 months after diagnosis, and 145 (5.3%) undergoing delayed repair (ie, more than 12 months after initial diagnosis of rotator cuff tear) (Table 1).
The overall revision surgery rate was 9.6% within 5 years. Across the time to surgery cohorts, the revision rate was 9.9% in the early repair group, 8.3% in the routine repair group, and 15.2% in the delayed repair group (Table 2). On bivariate analysis, the differences in revision rates between the delayed repair group and both the early and the routine repair groups were statistically significant (P=.048 and P=.007, respectively). The difference in revision rates between the early and routine repair groups was not statistically significant (P=.163).
Time to Rotator Cuff Repair and Revision Surgery Rates
On multivariate logistic regression analysis, delayed repair was associated with significantly increased odds of subsequent revision surgery (OR, 1.97; 95% CI, 1.16–3.21; P=.009) relative to routine repair. Also, relative to routine repair, early repair trended toward slightly increased odds of revision (OR, 1.22; 95% CI, 0.93–1.60; P=.160) but did not achieve statistical significance.
The Effect of Age, Sex, and Comorbidity Burden on the Rate of Revision Rotator Cuff Repair Relative to Time to Surgery
Patient sex and CCI were not associated with the risk of revision surgery. Finally, increasing age was associated with decreased risk of subsequent revision rotator cuff repair (Table 3). In the cohort of patients older than 75 years, there was a decreased odds of revision (OR, 0.41; 95% CI, 0.25–0.63; P<.001) relative to patients younger than 75 years.
Multivariate Logistic Regression Model for Revision Rotator Cuff Repair
The principal findings of this study indicated that at minimum 5-year follow-up, rotator cuff repair more than 12 months after the initial diagnosis was associated with nearly double the odds of undergoing subsequent revision surgery relative to routine repair between 6 weeks and 12 months after diagnosis. In the context of clinical and animal data supporting the deleterious effect of delayed repair on functional outcomes, muscle atrophy, tendon retraction, and tension required for repair, the results of this study affirmed the authors' hypothesis that delay to rotator cuff repair may be associated with increased revision surgery rates.
This study had several important limitations. First, the time point of rotator cuff tear diagnosis was determined by the presence of a diagnosis code, and not the time of injury. Therefore, the actual period from injury to repair was longer than the intervals used in this study. Second, due to likely differences in coding practices across physician specialties and care settings, a large number of patients in the database who underwent rotator cuff repair did not have a rotator cuff tear diagnosis code, thus introducing possible sampling bias to these results. Third, because tear size, tear characteristics, functional status, patient-reported outcomes, range of motion, and previous nonoperative treatment(s) were not available in the database, selection bias may have been introduced, as these factors may influence the decision to operate earlier. Fourth, because a longitudinal private-payer database was used, patients who switched insurance plans at any point between diagnosis of rotator cuff tear and 5 years following their rotator cuff repair were excluded from the analysis. Finally, patients who were converted to shoulder arthroplasty after rotator cuff repair failure were not captured in this study.
Rotator cuff repair is typically delayed in favor of nonoperative treatment, with the goal of symptomatic relief, restoration of function, and avoiding the risks associated with an operation. Nonoperative treatment, however, is not without its own set of risks. In a study of 59 shoulders with nonoperatively treated rotator cuff tears and followed with repeat magnetic resonance imaging at a mean of 20 months after an initial scan, Maman et al11 reported that 52% of full-thickness tears increased in size, and 24% demonstrated worsened muscle atrophy. Similarly, Safran et al14 sought to investigate the natural history of symptomatic rotator cuff tears in patients younger than 60 years who were treated nonoperatively. Using ultrasound examinations performed by the same ultrasonographer for 51 patients at a mean of 29 months apart, they found that 49% of tears increased in size, and that tear progression was correlated with pain. This is in agreement with the literature on the natural history of asymptomatic rotator cuff tears in that pain development is often an indicator of tear progression.15,16 Finally, in a study of 24 patients with isolated full-thickness supraspinatus tears who were offered rotator cuff repair but elected for nonoperative treatment instead, Fucentese et al17 found that, at a median follow-up of 42 months, the patients had a mean subjective shoulder score of 74% of a normal shoulder, a mean Constant score of 75, and 29% were either less satisfied or not satisfied with their shoulder. In terms of tear assessment with magnetic resonance imaging, 33% of tears increased in size, and 25% had increased fatty infiltration of the supraspinatus.
Furthermore, several animal studies have examined rotator cuff muscle and tendon physiology in delayed repair models. Using a rotator cuff injury model involving sheep, Coleman et al9 sought to determine the reversibility of muscle contraction weakness and fatty infiltration at various repair time points. They found that earlier tendon repair (6 weeks after injury) was associated with more rapid recovery of both muscle strength and fatty infiltration, and that tears repaired later (18 weeks after injury) had irreversible degradation in tendon elasticity. An additional concern with delaying rotator cuff repair is that progressive tendon retraction might lead to a more difficult repair later. In a rat model, Gimbel et al12 showed that the tension required to repair the supraspinatus back to its insertion increased with time to repair, and that at just 16 weeks from injury, the tension required was up to 5 times greater than the tension required at the time of injury. Moreover, this increased repair tension was associated with worsened failure properties and viscoelastic peak stress. Similarly, in a sheep model, Gerber et al10 found that rotator cuff tendon release led to muscle atrophy, fatty infiltration, and a 7-fold decrease in the elasticity of the musculotendinous unit at 40 weeks after tendon release. In a subsequent clinical study comparing magnetic resonance imaging data with intraoperative muscle stimulation and force measurements, Gerber et al18 found that maximal supraspinatus tension was strongly associated with muscle atrophy and fatty infiltration, which were irreversible with current repair techniques.
The clinical literature on the effect of time to rotator cuff repair and postoperative outcomes is relatively sparse and predominantly focused on acute traumatic tears. In a study of 32 patients who had acute traumatic full-thickness rotator cuff tears, Petersen and Murphy8 reported significantly worse patient-reported outcomes and active elevation in patients who underwent repair more than 4 months after the injury. Similarly, Björnsson et al7 reviewed 42 patients with traumatic full-thickness rotator cuff tears and no previous history of shoulder symptoms. All patients underwent repair within 3 months after the injury; within this time period, no association was found between the number of days to repair and postoperative outcomes. Conversely, among 35 patients with traumatic rotator cuff tears, Hantes et al6 found that acute repair within 3 weeks of injury was associated with improved patient-reported outcomes and range of motion.
To the authors' knowledge, this is the first study specifically examining the association between time to rotator cuff repair and the rate of subsequent revision repair surgery. The increased revision surgery rate in the delayed surgery cohort was consistent with previous clinical and animal data indicating progressive increases in tear size, muscle atrophy, fatty infiltration, degradation of tendon elasticity, increased requisite tendon repair tension, and decreased failure properties that might ultimately lead to re-tear and revision. Notably, however, the overall revision rate of 9.6% reported in the current study is lower than the weighted mean re-tear rate of 26.6% that was reported in a recent meta-analysis of rotator cuff repair outcomes.19 Evidently, a significant proportion of patients who have a re-tear may be asymptomatic and do not undergo revision rotator cuff repair, particularly older patients.20,21 To that point, the results of this multivariate regression showed that increased age was independently associated with decreased odds of revision rotator cuff repair surgery. While advanced age has also previously been shown to be associated with lower healing rates after rotator cuff repair,19,22 older patients may be less likely to undergo revision rotator cuff surgery because of having lower functional demands, being less symptomatic despite having a re-tear, or potentially having more glenohumeral arthritis to the point where they are offered reverse shoulder arthroplasty instead. Older patients, who are part of an already limited population, may be reluctant to undergo revision because of the pain and increased limitation associated with recovery.
Delayed rotator cuff repair beyond 12 months of diagnosis was associated with an increased risk of undergoing subsequent revision rotator cuff repair while controlling for age and comorbidity burden. For patients with symptomatic rotator cuff tears who have failed nonoperative management and are indicated for surgical repair, it may be beneficial to perform rotator cuff repair within 12 months of diagnosis if possible. Future research efforts should focus on prospective clinical studies that include additional data such as the exact time from injury to surgery, imaging findings, muscle atrophy, tendon retraction, and functional outcomes.
- Colvin AC, Egorova N, Harrison AK, Moskowitz A, Flatow EL. National trends in rotator cuff repair. J Bone Joint Surg Am. 2012;94(3):227–233. doi:10.2106/JBJS.J.00739 [CrossRef]. PMID:22298054
- Yamamoto A, Takagishi K, Osawa T, et al. Prevalence and risk factors of a rotator cuff tear in the general population. J Shoulder Elbow Surg. 2010;19(1):116–120. doi:10.1016/j.jse.2009.04.006 [CrossRef]. PMID:19540777
- Matsen FA III, . Clinical practice; rotator-cuff failure. N Engl J Med. 2008;358(20):2138–2147. doi:10.1056/NEJMcp0800814 [CrossRef]. PMID:18480206
- Tashjian RZ. Epidemiology, natural history, and indications for treatment of rotator cuff tears. Clin Sports Med. 2012;31(4):589–604. doi:10.1016/j.csm.2012.07.001 [CrossRef]. PMID:23040548
- Oh LS, Wolf BR, Hall MP, Levy BA, Marx RG. Indications for rotator cuff repair: a systematic review. Clin Orthop Relat Res. 2007;455(455):52–63. doi:10.1097/BLO.0b013e31802fc175 [CrossRef]. PMID:17179786
- Hantes ME, Karidakis GK, Vlychou M, Varitimidis S, Dailiana Z, Malizos KN. A comparison of early versus delayed repair of traumatic rotator cuff tears. Knee Surg Sports Traumatol Arthrosc. 2011;19(10):1766–1770. doi:10.1007/s00167-011-1396-1 [CrossRef]. PMID:21258779
- Björnsson HC, Norlin R, Johansson K, Adolfsson LE. The influence of age, delay of repair, and tendon involvement in acute rotator cuff tears: structural and clinical outcomes after repair of 42 shoulders. Acta Orthop. 2011;82(2):187–192. doi:10.3109/17453674.2011.566144 [CrossRef]. PMID:21434791
- Petersen SA, Murphy TP. The timing of rotator cuff repair for the restoration of function. J Shoulder Elbow Surg. 2011;20(1):62–68. doi:10.1016/j.jse.2010.04.045 [CrossRef]. PMID:20675154
- Coleman SH, Fealy S, Ehteshami JR, et al. Chronic rotator cuff injury and repair model in sheep. J Bone Joint Surg Am. 2003;85(12):2391–2402. doi:10.2106/00004623-200312000-00018 [CrossRef]. PMID:14668510
- Gerber C, Meyer DC, Schneeberger AG, Hoppeler H, von Rechenberg B. Effect of tendon release and delayed repair on the structure of the muscles of the rotator cuff: an experimental study in sheep. J Bone Joint Surg Am. 2004;86(9):1973–1982. doi:10.2106/00004623-200409000-00016 [CrossRef]. PMID:15342760
- Maman E, Harris C, White L, Tomlinson G, Shashank M, Boynton E. Outcome of nonoperative treatment of symptomatic rotator cuff tears monitored by magnetic resonance imaging. J Bone Joint Surg Am. 2009;91(8):1898–1906. doi:10.2106/JBJS.G.01335 [CrossRef]. PMID:19651947
- Gimbel JA, Van Kleunen JP, Lake SP, Williams GR, Soslowsky LJ. The role of repair tension on tendon to bone healing in an animal model of chronic rotator cuff tears. J Biomech. 2007;40(3):561–568. doi:10.1016/j.jbiomech.2006.02.010 [CrossRef]. PMID:16600252
- Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Epidemiol. 1994;47(11):1245–1251. doi:10.1016/0895-4356(94)90129-5 [CrossRef]. PMID:7722560
- Safran O, Schroeder J, Bloom R, Weil Y, Milgrom C. Natural history of nonoperatively treated symptomatic rotator cuff tears in patients 60 years old or younger. Am J Sports Med. 2011;39(4):710–714. doi:10.1177/0363546510393944 [CrossRef]. PMID:21310940
- Mall NA, Kim HM, Keener JD, et al. 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. doi:10.2106/JBJS.I.00506 [CrossRef] PMID:21084574
- Yamaguchi K, Tetro AM, Blam O, Evanoff BA, Teefey SA, Middleton WD. Natural history of asymptomatic rotator cuff tears: a longitudinal analysis of asymptomatic tears detected sonographically. J Shoulder Elbow Surg. 2001;10(3):199–203. doi:10.1067/mse.2001.113086 [CrossRef]. PMID:11408898
- Fucentese SF, von Roll AL, Pfirrmann CWA, Gerber C, Jost B. Evolution of nonoperatively treated symptomatic isolated full-thickness supraspinatus tears. J Bone Joint Surg Am. 2012;94(9):801–808. doi:10.2106/JBJS.I.01286 [CrossRef]. PMID:22552669
- Gerber C, Schneeberger AG, Hoppeler H, Meyer DC. Correlation of atrophy and fatty infiltration on strength and integrity of rotator cuff repairs: a study in thirteen patients. J Shoulder Elbow Surg. 2007;16(6):691–696. doi:10.1016/j.jse.2007.02.122 [CrossRef]. PMID:17931904
- McElvany MD, McGoldrick E, Gee AO, Neradilek MB, Matsen FA III, . Rotator cuff repair: published evidence on factors associated with repair integrity and clinical outcome. Am J Sports Med. 2015;43(2):491–500. doi:10.1177/0363546514529644 [CrossRef]. PMID:24753240
- Russell RD, Knight JR, Mulligan E, Khazzam MS. Structural integrity after rotator cuff repair does not correlate with patient function and pain: a meta-analysis. J Bone Joint Surg Am. 2014;96(4):265–271. doi:10.2106/JBJS.M.00265 [CrossRef]. PMID:24553881
- Kim HM, Caldwell JM, Buza JA, et al. Factors affecting satisfaction and shoulder function in patients with a recurrent rotator cuff tear. J Bone Joint Surg Am. 2014;96(2):106–112. doi:10.2106/JBJS.L.01649 [CrossRef]. PMID:24430409
- Tashjian RZ, Hollins AM, Kim HM, et al. Factors affecting healing rates after arthroscopic double-row rotator cuff repair. Am J Sports Med. 2010;38(12):2435–2442. doi:10.1177/0363546510382835 [CrossRef]. PMID:21030564
| ≤59 y||571 (20.7%)|
| 60–69 y||883 (32.0%)|
| 70–74 y||776 (28.1%)|
| ≥75 y||529 (19.2%)|
| Female||1359 (49.3%)|
| Male||1400 (50.7%)|
|Charlson Comorbidity Index|
| 0||896 (32.5%)|
| 1||644 (23.3%)|
| 2||390 (14.1%)|
| ≥3||829 (30.1%)|
|Time to rotator cuff repair|
| Early repair (<6 wk)||1510 (54.7%)|
| Routine repair (6 wk-12 mo)||1104 (40.0%)|
| Delayed repair (>12 mo)||145 (5.3%)|
|Revision surgery||264 (9.6%)|
Time to Rotator Cuff Repair and Revision Surgery Rates
|Characteristic||No. of Cases||No. of Revision Surgeries||Revision Rate|
|Time to rotator cuff repair|
| Early repair (<6 wk)||1510||150||9.9%|
| Routine repair (6 wk-12 mo)||1104||92||8.3%|
| Delayed repair (>12 mo)||145||22||15.2%|
| Delayed vs early repair||P=.048|
| Delayed vs routine repair||P=.007|
| Routine vs early repair||P=.163|
Multivariate Logistic Regression Model for Revision Rotator Cuff Repair
|Characteristic||Odds Ratio||95% Confidence Interval||P|
| ≤59 y||Reference|
| 60–69 y||0.89||0.65–1.22||.475|
| 70–74 y||0.76||0.54–1.06||.108|
| ≥75 y||0.41||0.25–0.63||<.001|
|Charlson Comorbidity Index|
|Time to rotator cuff repair|
| Early repair (<6 wk)||1.22||0.93–1.60||.160|
| Routine repair (6 wk-12 mo)||Reference|
| Delayed repair (>12 mo)||1.97||1.16–3.21||.009|