A prospective evaluation of survivorship of asymptomatic degenerative rotator cuff tears

Transcription

1 Washington University School of Meicine Digital Open Access Publications 2014 A prospective evaluation of survivorship of asymptomatic egenerative rotator cuff tears Jay D. Keener Washington University School of Meicine in St. Louis Leesa M. Galatz Washington University School of Meicine in St. Louis Sharlene A. Teefey Washington University School of Meicine in St. Louis William D. Mileton Washington University School of Meicine in St. Louis Karen Steger-May Washington University School of Meicine in St. Louis See next page for aitional authors Follow this an aitional works at: Recommene Citation Keener, Jay D.; Galatz, Leesa M.; Teefey, Sharlene A.; Mileton, William D.; Steger-May, Karen; Stobbs-Cucchi, Georgia; Patton, Rebecca; an Yamaguchi, Ken,,"A prospective evaluation of survivorship of asymptomatic egenerative rotator cuff tears." The journal of bone an joint surgery.97, (2014). This Open Access Publication is brought to you for free an open access by Digital Commons@Becker. It has been accepte for inclusion in Open Access Publications by an authorize aministrator of Digital Commons@Becker. For more information, please contact engeszer@wustl.eu.

2 Authors Jay D. Keener, Leesa M. Galatz, Sharlene A. Teefey, William D. Mileton, Karen Steger-May, Georgia Stobbs- Cucchi, Rebecca Patton, an Ken Yamaguchi This open access publication is available at Digital

3 89 COPYRIGHT Ó 2015 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED A Prospective Evaluation of Survivorship of Asymptomatic Degenerative Rotator Cuff Tears Jay D. Keener, MD, Leesa M. Galatz, MD, Sharlene A. Teefey, MD, William D. Mileton, MD, Karen Steger-May, BA, Georgia Stobbs-Cucchi, RN, Rebecca Patton, MA, an Ken Yamaguchi, MD Investigation performe at the Shouler an Elbow Service, Department of Orthopaeic Surgery, Washington University, St. Louis, Missouri Backgroun: The purpose of this prospective stuy was to report the long-term risks of rotator cuff tear enlargement an symptom progression associate with egenerative asymptomatic tears. Methos: Subjects with an asymptomatic rotator cuff tear in one shouler an pain ue to rotator cuff isease in the contralateral shouler enrolle as part of a prospective longituinal stuy. Two hunre an twenty-four subjects (118 initial full-thickness tears, fifty-six initial partial-thickness tears, an fifty controls) were followe for a meian of 5.1 years. Valiate functional shouler scores were calculate (visual analog pain scale, American Shouler an Elbow Surgeons [ASES], an simple shouler test [SST] scores). Subjects were followe annually with shouler ultrasonography an clinical evaluations. Results: Tear enlargement was seen in 49% of the shoulers, an the meian time to enlargement was 2.8 years. The occurrence of tear-enlargement events was influence by the severity of the final tear type, with enlargement of 61% of the full-thickness tears, 44% of the partial-thickness tears, an 14% of the controls (p < 0.05). Subject age an sex were not relate to tear enlargement. One hunre subjects (46%) evelope new pain. The final tear type was associate with a greater risk of pain evelopment, with the new pain eveloping in 28% of the controls, 46% of the shoulers with a partialthickness tear, an 50% of those with a full-thickness tear (p < 0.05). The presence of tear enlargement was associate with the onset of new pain (p < 0.05). Progressive egenerative changes of the supraspinatus muscle were associate with tear enlargement, with supraspinatus muscle egeneration increasing in 4% of the shoulers with a stable tear compare with 30% of the shoulers with tear enlargement (p < 0.05). Nine percent of the shoulers with a stable tear showe increase infraspinatus muscle egeneration compare with 28% of those in which the tear ha enlarge (p = 0.07). Conclusions: This stuy emonstrates the progressive nature of egenerative rotator cuff isease. The risk of tear enlargement an progression of muscle egeneration is greater for shoulers with a full-thickness tear, an tear enlargement is associate with a greater risk of pain evelopment across all tear types. Level of Evience: Prognostic Level II. See Instructions for Authors for a complete escription of levels of evience. Peer Review: This article was reviewe by the Eitor-in-Chief an one Deputy Eitor, an it unerwent bline review by two or more outsie experts. It was also reviewe by an expert in methoology an statistics. The Deputy Eitor reviewe each revision of the article, an it unerwent a final review by the Eitor-in-Chief prior to publication. Final corrections an clarifications occurreuring one or more exchanges between the author(s) an copyeitors. Quantifying the risk an rate of rotator cuff tear progression is of funamental importance for formulating appropriate surgical inications. Despite numerous reports pertaining to the prevalence of asymptomatic rotator cuff tears 1-7, we are not aware of any long-term longituinal stuies efining the natural history of egenerative rotator cuff tears. Previous stuies have suggeste the progressive nature of egenerative rotator cuff tears in both asymptomatic an painful shoulers 8-12 ; however, these stuies were often retrospective with short-term follow-up. Disclosure: One or more of the authors receive payments or services, either irectly or inirectly (i.e., via his or her institution), from a thir party in support of an aspect of this work. In aition, one or more of the authors, or his or her institution, has ha a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomeical arena that coul be perceive to influence or have the potential to influence what is written in this work. Also, one or more of the authors has ha another relationship, or has engage in another activity, that coul be perceive to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitte by authors are always provie with the online version of the article. J Bone Joint Surg Am. 2015;97:

4 90 TABLE I Baseline an Final Tear Types in the Stuy Cohort (N = 219)* Tear Type at Baseline Control Partial-Thickness Full-Thickness Total Final tear type Control Partial-thickness Full-thickness Total *Baseline tear types were reclassifie to a new, final tear type when a control was converte to a partial or full-thickness tear or a partial-thickness tear was converte to a full-thickness tear uring the follow-up perio. The inications for the surgical treatment of rotator cuff tears have varie wiely 13. Early surgical intervention for painful tears is intene to prevent the tear enlargement an muscle egeneration that can occur with nonoperative treatment as well as the iminishe potential for tenon healing with elaye surgery A more thorough unerstaning of the natural history of rotator cuff isease can increase the ability to ientify patients an shoulers at risk for isease progression an refine surgical inications. Furthermore, patients can be more effectively counsele regaring the potential risks of nonoperative treatment. Although recent stuies have improve our unerstaning of the evolution of rotator cuff isease 8-11, many questions remain unanswere. Accurate characterization of the natural history of egenerative cuff isease requires a well-controlle prospective longituinal follow-up stuy esign with well-efine en TABLE II Demographics an Baseline Characteristics of the Stuy Cohort ä Variable (Baseline) Entire Cohort (N = 224*) Control (N = 50*) Partial-Thickness Tear (N = 56*) Age (yr) 62.0 ± ± ± 10 Female sex (no. [%]) 92 (41%) 22 (44%) 26 (46%) Dominant sie asymptomatic (no. [%]) 81 (36%) 14 (28%) 17 (30%) Full-thickness tear size (no.) Tear length (mm) Not applic. Not applic. 6.0 (2.5) Tear with (mm) Not applic. Not applic. 9.0 (4.0) Tear area (mm 2 ) Not applic. Not applic (49) Muscle egeneration (no. [%]) Supraspinatus 13/162 (8%) 1/46 (2%) 0/36 (0%) Infraspinatus 11/163 (7%) 1/46 (2%) 0/37 (0%) Supraspinatus or infraspinatus 16/162 (10%) 1/46 (2%) 0/36 (0%) VAS pain score 1.0 (0) 1.0 (0) 1.0 (0) ASES score (points) 98.3 (10), n = (0) 98.3 (10) SST score (points) 91.7 (27), n = (0) 91.7 (33) Elevation strength (N) 55.8 ± 28, n = ± ± 28 External rotation strength (N) 68.8 ± 32, n = ± ± 33 *When ata were available for less than the entire cohort, the sample size (n) or the enominator is also provie. The values are given as the mean an stanareviation. The values are given as the meian with the interquartile range in parentheses. Fatty infiltration was ientifie by rating architecture an echogenicity on a 0 to 2-point scale an summing the two scores. Values of >0 inicate the presence of fatty infiltration. #Analysis of variance was use to compare the tear-type groups an, when the overall moel was significant (p < 0.05), Tukey-ajuste pairwise comparisons were performe. **The analysis was performe with use of rank-transformeata. The analysis was performe with the Fisher exact test.

5 91 points, particularly since the factors relate to tear progression an the relationships of tear progression to pain evelopment have not been clearly efine. Asymptomatic egenerative rotator cuff tears represent an ieal cohort for stuying the natural history of cuff isease, as clinical intervention is not necessitate by pain, an the recognize potential for later symptomatic change 8,9 suggests an appropriate moel. The purpose of this stuy was to report the risk of symptom evelopment, tear size progression, an muscle egeneration in a cohort of subjects with an asymptomatic rotator cuff tear an analyze factors associate with these changes. Materials an Methos Our university institutional review boar approve this stuy. This stuy cohort consiste of subjects with an asymptomatic rotator cuff tear who presente for evaluation of shouler pain seconary to rotator cuff isease in the contralateral shouler 8. All subjects ha physical finings in the painful shouler that were consistent with rotator-cuff-base pain an nearly all ha an ultrasoun-proven cuff tear in the painful shouler. Inclusion criteria were (1) bilateral shouler ultrasonography performe to investigate unilateral shouler pain, (2) painful rotator cuff isease in the symptomatic shouler, (3) a rotator cuff tear in the asymptomatic shouler at the time of stuy enrollment, an (4) no history of trauma to either shouler an no traumatic episoe throughout the stuy perio. Control subjects with no ultrasoun evience of a rotator cuff tear in one shouler an painful cuff isease in the contralateral shouler were also enrolleuring the same time perio. Exclusion criteria were (1) any past or current pain in the asymptomatic shouler, (2) continuous use of narcotic or nonsteroial anti-inflammatory rugs (NSAIDs) in the three months prior to enrollment, (3) a traumatic episoe affecting the asymptomatic shouler, (4) inflammatory arthritis, (5) raiographic evience of osteoarthritis in the asymptomatic shouler, (6) upper-extremity weight-bearing emans, (7) a subscapularis tenon tear in the asymptomatic shouler, an (8) a very small (<5 mm) partial-thickness tear in the asymptomatic shouler. Stuy Protocol Subjects were enrolle from the clinical practices of three surgeons over a thirtymonth perio. A priori power calculations were performe to etermine the sample size neee for acceptably precise estimate rates of tear enlargement an pain evelopment in shoulers with an asymptomatic tear. With an enrollment of 265 subjects, a 30% ropout rate (185 subjects proviing ata), an the assumption that 39% of the subjects woul experience tear enlargement uring the follow-up perior 20,the95%confience bouns aroun the estimate enlargement rate range from 32% to 46%, meaning that the estimates woul be within 7%ofthetruevalue.Withtheseassumptions,thesamplesizeprovieenlargement estimates within 8.5% an 11.5% of the true value for each group, respectively. Assuming that 51% of the subjects woulevelop pain, the target sample size woul again achieve 95% bouns with 7% of the true rate (44% to 58%). A traine research nurse performe a comprehensive physical examination of the asymptomatic shouler consisting of goniometric assessment of its active range of motion. Assessment of isometric shouler elevation strength TABLE II (continue) P Value Full-Thickness Tear (N = 118*) Comparison Across Tear Groups Control Vs. Full-Thickness Partial Vs. Full-Thickness Control Vs. Partial-Thickness 63.8 ± 9 44 (37%) 50 (42%) 31 small, 74 meium, 10 large, 3 massive 11.0 (8.0), n = 114 <0.05#** 10.0 (7.0), n = #** 108 (164), n = 111 <0.05#** 12/80 (15%) <0.05 <0.05 < /80 (13%) < < /80 (19%) <0.05 <0.05 < (0) 0.30#** 96.3 (12) <0.05#** < (33) <0.05#** < ± # 65.7 ± #

6 92 TABLE III Risk Factors for Tear Progression* Risk Factor No Enlargement (N = 111) Enlargement (N = 108) P Value from Cox Regression for Enlargement Tear type P < 0.05 Full vs. control: p < 0.05, HR = 4.17 Partial vs. control: p < 0.05, HR = 2.73 Full vs. partial: p = 0.07, HR = 1.53 Control 31 (86%) 5 (14%), time to enlargement: meian, 2.2 yr (IQR, 1.7) Partial-thickness 30 (56%) 24 (44%), time to enlargement: meian, 3.3 yr (IQR, 3.1) Full-thickness 50 (39%) 79 (61%), time to enlargement: meian, 2.3 yr (IQR, 3.7) Mean baseline age (±SD) (yr) 62.4 ± ± 10 Sex HR = 0.99 ( ) Male 61 (48%) 67 (52%) Female 50 (55%) 41 (45%) Stuy sie ominant P < 0.05, HR = 1.53 No 82 (58%) 59 (42%) Yes 29 (37%) 49 (63%) Smoking status Previous vs. never: HR = 0.97 ( ) Current vs. never: HR = 0.90 ( ) Never 59 (48%) 63 (52%) Previous 41 (51%) 39 (49%) Current 11 (65%) 6 (35%) *Five massive tears were exclue from the analysis because they were too large for enlargement to be ocumente accurately. HR = hazar ratio. The numbers in parentheses after the hazar ratios represent the 95% confience intervals for the hazar ratios. The values are given as the number of subjects with the percentage of the risk-factor group in parentheses. A p value was not assigne because of a lack of power analysis for the variable. (90 elevation in the scapular plane) an external rotation strength (with the shouler aucte an in neutral rotation) was performe with an IsoBex ynamometer (Meical Device Solutions, Oberburg, Switzerlan), with the final measurement the average of three repetitions. Patient questionnaires consisting of a visual analog scale (VAS) ranging from 0 to 10 to rate average aily pain with use of whole integers, the American Shouler an Elbow Surgeons (ASES) score, an the simple shouler test (SST) with the score normalize to 100 points were complete. The patient was assesse for shouler pain at stuy enrollment anuring longituinal followup, with the efinition of pain consisting of any of the following: (1) shouler pain 3 on the 10-point scale lasting six weeks or longer, (2) pain greater than that experience as a part of aily living, (3) pain requiring narcotic or NSAID meications, (4) pain prompting evaluation by a physician, an (5) night pain affecting sleep. Annual (plus or minus three months) examinations were performe with all components of the initial evaluation repeate, incluing ultrasonography. Subjects were aske to contact the stuy coorinator if new shouler pain evelope between visits, an this prompte immeiate repeat evaluation. Subjects were retaine for ongoing evaluation regarless of pain evelopment or tear enlargement. Shouler Ultrasonography an Tear Enlargement Criteria Each subject unerwent a stanarize shouler ultrasoun stuy (see Appenix) 21,22. The maximum anteroposterior imension of the tear was measure on transverse views (perpenicular to the long axis of the cuff) anesignate as the tear with. The maximum egree of retraction was measure on longituinal views (parallel to the long axis of the cuff) an esignate as the tear length. Tear length was measure from the tenon efect to the lateral ege of the normal tenon footprint on the greater tuberosity. In this stuy, tear type refers to the control, partial-thicknesstear, or full-thickness-tear category. Tear enlargement was etermine on the basis of sequential annual ultrasoun reports. A full-thickness cuff tear was consiere to have enlarge if its size ha increase by 5 mm in any imension compare with baseline. Five millimeters was chosen to account for the inherent variability in the accuracy of ultrasonography 22 an to represent clinically meaningful enlargement. If the change in tear imension met the criterion for enlargement but was a borerline value, subsequent ultrasoun stuies were reviewe to confirm that the size change was consistent with that seen on the previous ultrasoun stuy. If the subsequent value i not confirm an increase from baseline values, the tear was not consiere to have enlarge. Once a tear was confirme to have enlarge, the new tear size was set as the new baseline for future comparisons. A partial-thickness tear was consiere to have enlarge when it ha converte to a full-thickness efect, efine as a complete isruption of tenon continuity at the insertion. In the control shoulers, enlargement was efine as the evelopment of a partial or full-thickness efect of at least 5 mm in any imension. Tear-size values were also analyze to ientify a ecrease in size over time. A tear was consiere to be smaller when a ecrease of 5 mminanyimensionwas ientifie an was corroborate on subsequent ultrasoun examinations. Five subjects who ha a massive tear that was too large to accurately measure at baseline were exclue from the tear-enlargement analysis.

7 93 TABLE IV Association Between Final Tear Type an the Development of Pain Final Tear Type* Variable Control (N = 36) Partial-Thickness (N = 54) Full-Thickness (N = 129) P Value from Cochran- Armitage Tren Test New pain 10 (28%) 25 (46%) 65 (50%) <0.05 Remaine asymptomatic (ref.) 26 (72%) 29 (54%) 64 (50%) *The values are given as the number of subjects with the percentage of the tear-type group in parentheses. Tear enlargement was efine as a change in the tear type or an increase in the size from baseline values. Once a tear progresse into a more avance category (from control to partial or from partial to full), that tear was inclue in the more avance group (now efine as the final tear type) for further monitoring of pain evelopment (Table I). Statistical Analysis Categorical baseline characteristics were compare across baseline tear groups with use of the Fisher exact test. Continuous baseline characteristics were compare with analysis of variance followe by Tukey-ajuste post-hoc pairwise comparisons between tear groups. Variables that were not normally istribute were rank-transforme prior to analysis. To account for ifferent urations of follow-up, separate univariable Cox proportional hazar regression moels were use to etermine if the risk factors for new pain evelopment or egenerative rotator cuff muscle changes were associate with the time between stuy entry an the occurrence of tear enlargement. Subjects with a stable tear who unerwent surgery were censore at the time of surgery, an subjects with a stable tear who were lost to follow-up were censore at their final visit. Hazar regression analysis was use to examine the association of emographic variables, tear type, new pain evelopment, an progression of muscle egeneration with tear enlargement. The reference group for the hazar regression analysis for each categorical risk factor was the group of patients without the risk factor. Hazar regression analysis for continuous risk factors is expresse for a one-unit increase in the risk factor. Without regar for the temporal relationship of events, the Cochran- Armitage tren test was use to etermine if avancing category of final tear type was associate with new pain evelopment. A Cox moel was use to etermine if the occurrence of tear enlargement was associate with the time between stuy entry an new pain evelopment. Kaplan-Meier survival curves were generate to escribe the annual rate at which (1) tear enlargement occurre accoring to tear type an (2) new pain occurre. Annual survival rates are reporte for the time between stuy entry an the outcome at each follow-up point through year five. For patients who evelope pain before the occurrence of tear enlargement, an unpaire t test or the Wilcoxon signe rank test was performe to etermine if a change in function occurre between baseline an the first occurrence of pain. Summary ata for normally istribute variables are presente as the mean an stanareviation (SD), anata for variables that are not normally istribute are presente as the meian an interquartile range (IQR). Categorical variables are presente as the number of subjects an the percentage of the specifie group. P values of <0.05 were consiere significant. Source of Funing The stuy was fune by the National Institutes of Health (R01 AR051026). Results Baseline Demographics Two hunre an fifty subjects were enrolle, an twentysix of them were lost to follow-up or later exclueuring the stuy. Of the 224 subjects who remaine, 118 (53%) ha a full-thickness tear, fifty-six (25%) ha a partial-thickness tears, an fifty (22%) were control subjects. The age an sex istributions were similar among the baseline tear types (Table II). At baseline, the full-thickness tears were significantly larger than the partial-thickness tears (meian, 11.0 mm versus 6.0 mm, p < 0.05) an significantly larger in area (p < 0.05); however, the ifference in tear with i not reach significance (meian, 10.0 mm versus 9.0 mm, p = 0.052). A greater proportion of full-thickness tears were associate with egenerative changes in the supraspinatus (p < 0.05) an infraspinatus (p < 0.05) muscles. Baseline SST (p < 0.05) an ASES (p < 0.05) scores worsene with avancing tear type; however, no significant ifferences were seen in baseline pain scores or strength among the tear groups. Tear Progression The meian uration of follow-up was 5.1 years (range, 0.3 to 10.0 years). The meian uration of stuy inclusion was longer for the tears that enlarge (5.9 years) than for the stable tears TABLE V Association Between Development of New Pain an Tear Enlargement Variable No Enlargement* (N = 111) Enlargement* (N = 108) P Value from Cox Regression for Enlargement New pain 37 (37%) 63 (63%): 41 at/before 1st enlargement, 22 after 1st enlargement Remaine asymptomatic (ref.) 74 (62%) 45 (38%) P < 0.05, HR = 1.66 *The values are given as the number of subjects with the percentage of the pain group in parentheses. HR = hazar ratio.

8 94 TABLE VI Association Between Degenerative Rotator Cuff Muscle Changes an Tear Enlargement Variable No Enlargement* (N = 77) Enlargement* (N = 82) P Value from Cox Regression for Enlargement Increase muscle egeneration Supraspinatus <0.05, HR = 2.00 No (ref.) 74 (96%) 57 (70%) Yes 3 (4%) 25 (30%) Infraspinatus 0.07 No (ref.) 70 (91%) 59 (72%) Yes 7 (9%) 23 (28%) Supraspinatus or infraspinatus <0.05, HR = 1.91 No (ref.) 69 (90%) 47 (57%) Yes 8 (10%) 35 (43%) *The values are given as the number of subjects with the percentage of the no-enlargement or enlargement group in parentheses. HR = hazar ratio. (4.9 years, p < 0.05). Tear progression was seen in 49% of the shoulers, an the meian time to enlargement was 2.8 years. The risk of tear enlargement was significantly influence by the severity of the final tear type (Table III), with tear-enlargement rates of 61% in the full-thickness-tear group, 44% in the partialthickness-tear group, an 14% in the control group (p < 0.05). Cox regression analysis showe the risk of enlargement in the full-thickness-tear group to be 4.2 times greater than that in the control group an 1.5 times larger than that in the partialthickness-tear group. Three partial-thickness an six fullthickness tears ecrease in size compare with baseline. Four of these tears (all full-thickness) later enlarge compare with the original values. If the stuy shouler was the ominant extremity, there was a greater likelihoo of tear enlargement (63% in ominant shoulers compare with 42% in nonominant shoulers, p < 0.05). We foun no clinically relevant ifferences in age, sex, or smoking status between stable an enlarge tears. Fig. 1 Kaplan-Meier annual survival curves for tear enlargement by final tear type.

9 95 TABLE VII Changes in Function Between Baseline an the Onset of New Pain Visit* Entire Cohort (N = 80: 10 Controls, 20 Partial-Thickness Tears, 50 Full-Thickness Tears) P Value VAS score for pain (0 = no pain, 10 = incapacitating pain) <0.05 Baseline 1.0 (0.0) 1st occurrence of pain 5.0 (3.0) Delta 3.0 (3.0) SST <0.05 Baseline 83.3 (34.8) 1st occurrence of pain 69.7 (33.3) Delta (33.2) ASES <0.05 Baseline 96.5 (11.7) 1st occurrence of pain 60.0 (26.5) Delta (27.0) External rotation strength# (N) <0.05 Baseline 68.4 ± 34 1st occurrence of pain 58.2 ± 30 Delta ± 25 Elevation strength# (N) <0.05 Baseline 55.4 ± 26 1st occurrence of pain 33.3 ± 20 Delta ± 18 Forwar elevation active# (eg) <0.05 Baseline 154 ± 14 1st occurrence of pain 145 ± 20 Delta 29.2 ± 25 External rotation at 90 active# (eg) <0.05 Baseline 91.1 ± 12 1st occurrence of pain 82.2 ± 13 Delta 28.9 ± 16 Internal rotation in abuction active# (eg) 0.24 Baseline 65.9 ± 20 1st occurrence of pain 62.7 ± 17 Delta 23.2 ± 24 External rotation with arm at sie active# (eg) <0.05 Baseline 72.4 ± 18 1st occurrence of pain 62.8 ± 14 Delta 29.6 ± 23 *Delta for each patient was calculate by subtracting the baseline value from the value at the first occurrence of pain. The average time between these visits was 987 ± 705 ays (range, 161 to 2881 ays). The p values were erive by comparing elta with zero in paire t tests unless otherwise inicate. The values are given as the meian with the interquartile range in parentheses. The p value was erive with the Wilcoxon signe rank test. #The values are given as the mean an stanareviation. Survivorship Analysis The meian time to enlargement was 2.2, 3.3, an 2.3 years for the control, partial-thickness, an full-thickness groups, respectively. Figure 1 shows the five-year survivorship of the three groups with tear enlargement as the en point. Survivorship was 100% at one an two years, 90% at three years, 86% at four years, an 81% at five years in the control group. The yearly survivorship values (for years one through five) were 94%, 89%, 80%, 71%, an 65% in the partial-thickness group an 95%, 78%, 66%, 57%, an 50% in the full-thickness group.

10 96 Symptom Progression One hunre shoulers (46%) evelope new pain (Table IV). The meian time to pain evelopment was 2.6 years. In the entire cohort, the annual survivorship values (for years one through five) with the evelopment of new pain as the en point were 94%, 82%, 72%, 63%, an 56%. A greater risk for pain evelopment was associate with a more avance final tear type (p < 0.05). Twenty-eight percent of the controls, 46% of the patients with a partial-thickness tear, an 50% of those with a full-thickness tear evelope new pain. Tear enlargement was a significant risk factor for the evelopment of shouler pain (p < 0.05) (Table V). The tear enlarge in 38% of the shoulers that remaine asymptomatic compare with 63% of those that evelope pain. Cox regression analysis emonstrate a 1.66 times higher prevalence of tear enlargement in shoulers that evelope pain compare with shoulers that remaine asymptomatic. There was a 1.69 times higher prevalence of new pain in the shoulers in which the tear enlarge compare with the shoulers with a stable tear (p < 0.05). Of the sixty-three shoulers in which the tear enlarge an became painful, forty-one became painful at or before the time of the initial enlargement an twenty-two, at a later time point. Muscle Degenerative Changes Complete assessment of muscle egenerative changes was one for 159 shoulers. Tear enlargement was significantly associate with progressive egenerative changes in the supraspinatus muscle (Table VI). Four percent of the stable tears were associate with an increase in supraspinatus egeneration (of at least one grae) compare with 30% of the tears that enlarge (p < 0.05). Nine percent of the shoulers with a stable tear showe an increase in infraspinatus egeneration compare with 28% of the shoulers with tear enlargement (p = 0.07). Shouler Function Nearly all shoulers with new pain showe a significant ecline in function from baseline values (Table VII). The meian VAS pain score increase by 3.0 points (p < 0.05), an the meian ASES an SSTscores ecrease by 31.9 (p < 0.05) an 14.8 (p < 0.05) points, respectively. Discussion The results of this prospective longituinal stuy emonstrate a high risk of rotator cuff tear enlargement over a relatively short time perio. The most clinically relevant relationship was between a greater risk of tear progression an a more severe tear type. Full-thickness tears were 4.2 times an 1.5 times more likely to enlarge than controls an partialthickness tears, respectively. Our ata suggest that activity level may influence tear progression given the higher rate of enlargement seen in ominant shoulers. We foun a greater risk of pain evelopment in shoulers in which the tear enlarge compare with those in which the tear remaine stable. Natural history stuies are important for efining the risks an rates of isease progression over time, an they allow better assessment of the appropriate timing of intervention an ientification of specific at-risk groups in which early intervention may be most beneficial. Natural history information about egenerative rotator cuff isease is lacking in comparison with the large volume of ata eicate to the outcomes of various treatments, usually surgical intervention. Repeate longituinal assessments provie clearer estimates of the timeline for tear progression an allow closer monitoring of symptom progression an subsequent changes in shouler function. The rate of progression of asymptomatic cuff tears in this stuy is similar to rates in two previous stuies of conservatively treate painful cuff tears 10,11. Using ultrasoun an similar efinitions of tear enlargement, Safran et al. showe that 49% of painful full-thickness cuff tears enlarge after a mean of twenty-nine months of follow-up 11. Maman et al. examine tear progression with magnetic resonance imaging (MRI) an note that 48% of painful tears reexamine at more than eighteen months ha enlarge compare with 19% of tears followe for less than eighteen months 10. Similar to our finings, the risk of progression was lower for partial-thickness tears than for full-thickness tears 10. The increase risk of progression of full-thickness tears highlights the nee for either close surveillance or surgical intervention for painful shoulers in this higher-risk group. The relatively low prevalence of muscle egeneration in our stuy may be a reflection of the small size of the typical fullthickness tear in this cohort. Nevertheless, the results emonstrate that progressive muscle egeneration is associate with progression of even smaller tears. In two previous stuies, no correlation was seen between enlargement of painful tears an progression of muscle egeneration, although trens were evient 10,12. In both series the number of subjects, particularly those who evelope atrophy, was small. It is likely that our finings iffer because of our larger cohort, longer follow-up perio, an prospective stuy esign. The progression of muscle egeneration is a very relevant consieration when counseling subjects with an enlarging rotator cuff tear as both factors are associate with lower rates of tenon healing following surgery. The onset of pain was associate with a clinically meaningful ecline in shouler function base on establishe minimal clinically important ifferences for shouler pain an the ASES score; however, the ecrease in the SST score was just below the establishe minimal clinically important ifference for patients with rotator cuff isease 23. A more severe tear type was associate with a higher risk of pain evelopment. We chose to analyze the risk of pain evelopment on the basis of the final tear type to more accurately reflect how these risks change as a tear evolves. In our previous stuy of this cohort, 23% of the subjects evelope new pain, which was relate to tear progression 8. In the present stuy, we showe that tear progression was a significant risk factor for pain evelopment, noting an even stronger relationship than was seen in our earlier stuy. These finings are similar to those in a recent stuy by Moosmayer et al., in which 36% of the subjects with an

11 97 asymptomatic full-thickness tear evelope pain over the course of short-term follow-up 9. They also note a greater progression of tear size in shoulers that became painful. The temporal relationship between enlargement an pain evelopment warrants further analysis as pain occurre after the ientification of initial tear enlargement in 35% of our patients who ha both enlargement an pain evelopment. We believe that the collective finings of these stuies emphasize that symptom progression shoul be carefully evaluate as a possible clinical sign of tear progression. Our stuy ha several limitations. Our subjects ha painful rotator cuff isease in the contralateral shouler; therefore, finings may iffer from those in subjects with unilateral isease. Nevertheless, we believe that the results of this stuy may represent the natural history of symptomatic cuff isease given the recognize bilateral nature of tears; our cohort was a true at-risk population. Another limitation is that the chronicity of the tears was unknown an it is possible that unknown ifferences in isease uration can influence the rates of tear progression an symptom evelopment. Also, we i not analyze the potential relationships of other variables, such as occupation, meical comorbiities, an the tear size in the contralateral shouler, to the risks of tear progression an pain evelopment. The evaluation of muscle egeneration may have been biase as muscle evaluation was not performe in the earlier part of the stuy. We began collecting muscle egeneration ata after ultrasoun finings were valiate through a comparison with MRI finings 24. Despite this shortcoming, we were able to clearly emonstrate progression of supraspinatusegenerationanatrentowarprogressionof infraspinatus egeneration in association with full-thickness tears that enlarge. The strengths of this stuy are primarily relate to its esign, as it involve a large cohort of patients followe prospectively at well-efine time intervals to allow better efinition of time-epenent tear progression. Use of a control group similar to the stuy group was an important reference with which to compare risks of tear progression in shoulers with tears. This stuy inclue a stanarize an valiate annual evaluation performe by traine research nurses, an raiologists conucte all ultrasoun examinations to minimize observer bias. Strict efinitions of tear enlargement an symptom evelopment were esignate prior to stuy initiation. Finally, categorization of final tear type better characterize the risks of tear progression an pain evelopment as an asymptomatic tear evolves over time. In conclusion, this prospective, longituinal stuy emonstrate the progressive nature of asymptomatic egenerative rotator cuff tears in subjects with a contralateral symptomatic rotator cuff tear. The risk of enlargement is greater for more avance tears, an tear enlargement is associate with a greater risk of cuff muscle egeneration an of pain evelopment. New pain may be a clinical sign of tear progression. Appenix A etaileescription of the ultrasoun stuies of the rotator cuff tears an the evaluation of the fatty egeneration of the rotator cuff muscles is available with the online version of this article as a ata supplement at jbjs.org. n Jay D. Keener, MD Leesa M. Galatz, MD Georgia Stobbs-Cucchi, RN Rebecca Patton, MA Ken Yamaguchi, MD Shouler an Elbow Service, Department of Orthopaeic Surgery, Washington University, Campus Box 8233, 660 South Eucli Avenue, St. Louis, MO aress for J.D. Keener: keenerj@wustl.eu Sharlene A. Teefey, MD William D. Mileton, MD Mallinckrot Institute of Raiology, Washington University, St. Louis, MO Karen Steger-May, BA Division of Biostatistics, Washington University School of Meicine, St. Louis, MO References 1. Moosmayer S, Smith HJ, Tariq R, Larmo A. Prevalence an characteristics of asymptomatic tears of the rotator cuff: an ultrasonographic an clinical stuy. J Bone Joint Surg Br Feb;91(2): Kim HM, Teefey SA, Zelig A, Galatz LM, Keener JD, Yamaguchi K. Shouler strength in asymptomatic iniviuals with intact compare with torn rotator cuffs. J Bone Joint Surg Am Feb;91(2): Yamaguchi K, Ditsios K, Mileton WD, Hilebolt CF, Galatz LM, Teefey SA. The emographic an morphological features of rotator cuff isease. A comparison of asymptomatic an symptomatic shoulers. J Bone Joint Surg Am Aug;88 (8): Schibany N, Zehetgruber H, Kainberger F, Wurnig C, Ba-Ssalamah A, Herneth AM, Lang T, Gruber D, Breitenseher MJ. Rotator cuff tears in asymptomatic iniviuals: a clinical an ultrasonographic screening stuy. Eur J Raiol Sep;51 (3): Tempelhof S, Rupp S, Seil R. Age-relate prevalence of rotator cuff tears in asymptomatic shoulers. J Shouler Elbow Surg Jul-Aug;8(4): Milgrom C, Schaffler M, Gilbert S, van Holsbeeck M. Rotator-cuff changes in asymptomatic aults. The effect of age, hanominance an gener. J Bone Joint Surg Br Mar;77(2): Sher JS, Uribe JW, Posaa A, Murphy BJ, Zlatkin MB. Abnormal finings on magnetic resonance images of asymptomatic shoulers. J Bone Joint Surg Am Jan;77(1): Mall NA, Kim HM, Keener JD, Steger-May K, Teefey SA, Mileton WD, Stobbs G, Yamaguchi K. Symptomatic progression of asymptomatic rotator cuff tears: a prospective stuy of clinical an sonographic variables. J Bone Joint Surg Am Nov 17;92(16): 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 Jul 17;95(14): Maman E, Harris C, White L, Tomlinson G, Shashank M, Boynton E. Outcome of nonoperative treatment of symptomatic rotator cuff tears monitore by magnetic resonance imaging. J Bone Joint Surg Am Aug;91(8):

12 Safran O, Schroeer J, Bloom R, Weil Y, Milgrom C. Natural history of nonoperatively treate symptomatic rotator cuff tears in patients 60 years ol or younger. Am J Sports Me Apr;39(4): Epub 2011 Feb Fucentese SF, von Roll AL, Pfirrmann CW, Gerber C, Jost B. Evolution of nonoperatively treate symptomatic isolate full-thickness supraspinatus tears. J Bone Joint Surg Am May 2;94(9): Dunn WR, Schackman BR, Walsh C, Lyman S, Jones EC, Warren RF, Marx RG. Variation in orthopaeic surgeons perceptions about the inications for rotator cuff surgery. J Bone Joint Surg Am Sep;87(9): Goutallier D, Postel JM, Gleyze P, Leguilloux P, Van Driessche S. Influence of cuff muscle fatty egeneration on anatomic an functional outcomes after simple suture of full-thickness tears. J Shouler Elbow Surg Nov-Dec;12(6): Liem D, Lichtenberg S, Magosch P, Habermeyer P. Magnetic resonance imaging of arthroscopic supraspinatus tenon repair. J Bone Joint Surg Am Aug;89 (8): Glastone JN, Bishop JY, Lo IK, Flatow EL. Fatty infiltration an atrophy of the rotator cuff o not improve after rotator cuff repair an correlate with poor functional outcome. Am J Sports Me May;35(5): Epub 2007 Mar Fuchs B, Gilbart MK, Holer J, Gerber C. Clinical an structural results of open repair of an isolate one-tenon tear of the rotator cuff. J Bone Joint Surg Am Feb;88(2): Gerber C, Schneeberger AG, Hoppeler H, Meyer DC. Correlation of atrophy an fatty infiltration on strength an integrity of rotator cuff repairs: a stuy in thirteen patients. J Shouler Elbow Surg Nov-Dec;16(6): Epub 2007 Oct Tashjian RZ, Hollins AM, Kim HM, Teefey SA, Mileton WD, Steger-May K, Galatz LM, Yamaguchi K. Factors affecting healing rates after arthroscopic ouble-row rotator cuff repair. Am J Sports Me Dec;38(12): Epub 2010 Oct Yamaguchi K, Tetro AM, Blam O, Evanoff BA, Teefey SA, Mileton WD. Natural history of asymptomatic rotator cuff tears: a longituinal analysis of asymptomatic tears etecte sonographically. J Shouler Elbow Surg May-Jun;10(3): Teefey SA, Hasan SA, Mileton WD, Patel M, Wright RW, Yamaguchi K. Ultrasonography of the rotator cuff. A comparison of ultrasonographic an arthroscopic finings in one hunre consecutive cases. J Bone Joint Surg Am Apr;82 (4): Teefey SA, Rubin DA, Mileton WD, Hilebolt CF, Leibol RA, Yamaguchi K. Detection an quantification of rotator cuff tears. Comparison of ultrasonographic, magnetic resonance imaging, an arthroscopic finings in seventy-one consecutive cases. J Bone Joint Surg Am Apr;86(4): Tashjian RZ, Deloach J, Green A, Porucznik CA, Powell AP. Minimal clinically important ifferences in ASES an simple shouler test scores after nonoperative treatment of rotator cuff isease. J Bone Joint Surg Am Feb;92(2): Wall LB, Teefey SA, Mileton WD, Dahiya N, Steger-May K, Kim HM, Wessell D, Yamaguchi K. Diagnostic performance an reliability of ultrasonography for fatty egeneration of the rotator cuff muscles. J Bone Joint Surg Am Jun 20;94(12): e Strobel K, Holer J, Meyer DC, Pfirrmann CW, Pirkl C, Zanetti M. Fatty atrophy of supraspinatus an infraspinatus muscles: accuracy of US. Raiology Nov;237(2): Epub 2005 Sep 28.