- Email: [email protected]
The Incidence of Subsequent Surgery After Outpatient Arthroscopic Rotator Cuff Repair
The Incidence of Subsequent Surgery After Outpatient Arthroscopic Rotator Cuff Repair Siddharth A. Mahure, M.D., Brent Mollon, M.D., F.R.C.S.C., Steven D. Shamah, B.A., Joseph D. Zuckerman, M.D., Young W. Kwon, M.D., Ph.D., and Andrew S. Rokito, M.D.
Purpose: To quantify the incidence and risk factors associated with subsequent shoulder procedures in individuals undergoing outpatient arthroscopic rotator cuff repair (ARCR). Methods: We examined the New York Statewide Planning and Research Cooperative Systems outpatient database from 2003 through 2014 to identify patients undergoing isolated ARCR with or without concomitant acromioplasty. Patients were longitudinally followed up for a minimum of 2 years to determine the incidence of subsequent ipsilateral shoulder surgery. The impact of age, sex, insurance, concomitant acromioplasty, and tobacco use on reoperation was explored. Results: Between 2003 and 2012, 30,430 patients underwent isolated ARCR. The mean age was 56.6 11.5 years, and 55.1% were male patients. A total of 1,826 patients (6.0%) underwent subsequent ipsilateral outpatient shoulder surgery a mean of 24.3 27.1 months after the initial ARCR. Of patients who underwent repeat surgery, 57.3% underwent a revision cuff repair. Patients who underwent additional outpatient shoulder surgery were significantly younger (53.7 10.9 years v 56.8 11.5 years, P < .001). Tobacco use was associated with an increased rate of subsequent surgery (7.3% v 5.9%, P ¼ .044) and accelerated time to reoperation (16.9 months v 24.7 months, P < .001). Independent risk factors for subsequent ipsilateral surgery after initial ARCR were presence of a Workers’ Compensation claim (odds ratio, 2.11; 95% confidence interval, 1.89-2.36; P < .001) and initial ARCR without acromioplasty (odds ratio, 1.20; 95% confidence interval, 1.09-1.34; P < .001). Conclusions: We identified a 6.0% incidence of repeat ipsilateral surgery after isolated ARCR. Although reasons for reoperation are likely multifactorial, younger age, Workers’ Compensation claim, and absence of acromioplasty at the time of initial ARCR remained independent predictors of subsequent outpatient procedures, whereas a history of tobacco use was associated with accelerated time to subsequent surgery. Level of Evidence: Level III, retrospective comparative study.
A
rthroscopic rotator cuff repair (ARCR) is recognized to provide good to excellent long-term clinical results in patients with symptomatic rotator cuff tears failing conservative therapy.1-4 Studies have consistently found improved pain, motion, and functional outcome scores across all age groups.5-8 Analyses of large national inpatient databases suggest arthroscopic rotator cuff procedures have increased by upward of 600% between 1996 and 2006.9 In New York State the incidence of rotator cuff repair (RCR) has
From the Department of Orthopaedic Surgery, NYU Hospital for Joint Diseases, New York, New York, U.S.A. The authors report that they have no conflicts of interest in the authorship and publication of this article. Received September 2, 2015; accepted January 15, 2016. Address correspondence to Siddharth A. Mahure, M.D., Department of Orthopaedic Surgery, NYU Hospital for Joint Diseases, 301 E 17th St, New York, NY 10003, U.S.A. E-mail: [email protected] Ó 2016 by the Arthroscopy Association of North America 0749-8063/15857/$36.00 http://dx.doi.org/10.1016/j.arthro.2016.01.039
increased by 238% over a 15-year period,10 with demand further increasing because of an aging population with increasing activity requirements and enhanced access to advanced imaging modalities.9-11 With the enhancement of arthroscopic techniques, greater familiarity among surgeons with the performance of these procedures, and improvements in the management of postoperative pain, most of these procedures are now performed on an outpatient basis.9,10,12,13 The factors involved in the success of ARCR are largely unclear. Although the technical goals of repair are to restore anatomy with a construct that is able to withstand bodily forces while bone-to-tendon healing occurs,14 other studies have questioned the relation between anatomic healing and functional outcomes.15-18 Multiple studies have found favorable results when only partial repairs or debridement procedures were possible because of tendon quality or extent of the tear.19-21 Furthermore, a high proportion of patients with retears after RCR still report excellent functional outcome scores that were much improved from their
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol
-,
No
-
(Month), 2016: pp 1-11
1
2
S. A. MAHURE ET AL.
preoperative status.15-17 Indeed, a recent systematic review found that repair integrity was significantly associated with improved strength but not functional outcome scores.22 Even less clear are the factors that predict the need for subsequent surgery after primary RCR. Overall reoperation rates after RCR remain unclear and have rarely been reported in recent series.23-26 To address this, we examined the Statewide Planning and Research Cooperative Systems (SPARCS) database of the New York State Department of Health. The purpose of this study was to quantify the incidence and risk factors associated with subsequent shoulder procedures in individuals undergoing outpatient ARCR. On the basis of existing literature,27-30 we hypothesized that age and the existence of a Workers’ Compensation claim would be associated with the need for subsequent surgical procedures.
Methods The New York State Department of Health established the SPARCS database in 1979. In 1986 the passage of the New York State Public Health Law mandated all nonfederal, licensed hospitals to report on discharges from their facilities. These requirements were subsequently expanded to include emergency departments and freestanding, licensed ambulatory surgery facilities. The database records the following information for all individuals who have undergone treatment at one of the aforementioned locations: age, sex, race, type of insurance, diagnoses, and procedures performed. Although not without certain limitations, the SPARCS database has been used to collect epidemiologic information for various diseases in previous peer-reviewed publications.10,31-33 For this study, diagnoses and procedures were classified by Current Procedural Terminology (CPT), fourth edition, and International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes. The American Medical Association officially introduced CPT code 29827 for ARCR in 2003. We queried the database from 2003 through 2014 to identify all isolated ARCRs performed. Any patient with additional shoulder CPT codes at the time of initial RCR was excluded, with the exception of subacromial decompression (CPT code 29826). We believed that this would allow us to best capture patients undergoing surgery primarily because of rotator cuff pathology and that the elimination of concomitant CPT codes would provide a more homogeneous sample. We identified 71,920 unique patients who underwent ARCR between 2003 and 2012. Elimination of patients with concomitant CPT codes other than acromioplasty and those with evidence of prior ipsilateral shoulder procedures yielded our final sample of 30,430 patients, representing
approximately 42% of all ARCRs performed. For this homogeneous sample of patients, all available patient demographic data were extracted. Similar to previous studies, our study used International Classification of Diseases, Ninth Revision (ICD-9) codes of 305.1 and V15.62 to identify patients with a history of tobacco use.34-36 A custom-designed computer algorithm using SAS software (version 9.3; SAS Institute, Cary NC) was used to identify and group patients in the SPARCS database. Patients who underwent RCR were subsequently tracked using an encrypted unique patient identifier and followed up for a minimum of 2 years (up to and including 2014). Thus, although we identified all ARCRs from 2003 through 2014, our longitudinal analysis only included patients from 2003 to 2012, thus ensuring a minimum of 2 years’ follow-up for all patients. The timing and nature of each additional shoulder procedure were recorded. For patients who underwent multiple subsequent surgical procedures, we used the first operation immediately after the initial ARCR for purposes of data analysis. Subsequent outpatient surgical procedures were identified using CPT codes for commonly performed arthroscopic and open shoulder procedures. Concomitant ICD-9 diagnosis codes for rotator cuff tears, shoulder instability, glenohumeral arthritis, acromioclavicular joint arthritis, acromioclavicular joint sprains, shoulder stiffness, or rotator cuff tendinitis/syndromes at the time of subsequent surgery were also recorded. Because of limitations in ICD-9 procedural codes used for inpatient procedures, we were unable to comment on patients who may have undergone a subsequent anatomic or reverse shoulder arthroplasty. Statistical analysis using SAS software (version 9.3) was performed to compare patients who underwent additional surgery with patients who did not. To explore the effect of age on the need for subsequent surgery, we conducted a subgroup analysis by stratifying patients into 3 age groups: younger than 45 years, between 45 and 65 years, and older than 65 years. Differences in continuous variables were evaluated with a Student t test and Kruskal-Wallis analysis. Differences in categorical variables were evaluated with the Fisher exact test (when possible) or c2 analysis. Multivariate logistic regression that controlled for age, sex, insurance, tobacco use, and acromioplasty at the time of initial surgery was performed to determine independent risk factors for subsequent surgery. P < .05 was considered statistically significant. Institutional review board approval was not required for this study because no human subjects were involved.
Results Between 2003 and 2014, 39,284 unique patients were recorded as undergoing isolated outpatient ARCR
SUBSEQUENT SURGERY AFTER ARCR
3
Fig 1. Total number of arthroscopic rotator cuff repairs (ARCRs) performed per year from 2003 through 2014 and proportion per year with and without acromioplasty. For each year, the percentage of ARCRs with and without acromioplasty is listed immediately to the right of the blue and red bars, respectively. For example, in 2003, 71.8% of ARCRs were performed with concomitant acromioplasty, whereas 28.2% were performed without acromioplasty. (RCR, rotator cuff repair.)
with or without acromioplasty in New York State. The number of procedures performed annually increased 301% over this 12-year period, from 1,200 in 2003 to 4,812 in 2014. The proportion of ARCRs performed with acromioplasty also increased from 71.8% in 2003 to 76.3% in 2014, although there has been a decreasing trend since 2011. Total volume of ARCRs per year can be found in Figure 1, along with the proportion of ARCRs per year that were performed with and without acromioplasty. Longitudinal analysis was only performed on the 30,430 patients with isolated ARCR between 2003 and 2012. Male patients comprised 55.1% of our sample (16,751 of 30,430). The mean age of our sample was 56.6 11.5 years, with male patients being significantly younger than female patients (55.3 11.7 years v 58.1 11.1 years, P < .001). Primary insurance was most commonly private (58.5%) or Medicare (21.6%), with 14.1% Workers’ Compensation, 3.9% Medicaid, 0.4% Veterans Affairs, and 1.6% other. For the vast majority of our sample (92.5%), New York was listed as the home state. Additional Outpatient Shoulder Surgery During the minimum 2-year follow-up, 6.0% of patients (1,826 of 30,430) underwent subsequent outpatient shoulder surgery on the same shoulder as the initial ARCR. Of those with subsequent surgery, 9.1% (166 of 1,826) underwent 2 or more subsequent ipsilateral operations.
The subsequently performed shoulder procedures included ARCR (45.7%; 833 of 1,826), subacromial decompression (43.8%; 799 of 1,826), arthroscopic debridement (24.5%; 448 of 1,826), open RCR (11.7%; 214 of 1,826), distal clavicle resection (14.3%; 261 of 1,826), and biceps tenodesis/tenotomy (8.1%; 147 of 1,826). Therefore, 3.4% of our initial sample (1,047 of 30,430), or 57.3% of those requiring additional surgery (1,047 of 1,826), underwent subsequent rotator cuff surgery. The mean time to subsequent intervention was 24.4 27.1 months, with 69.3% of patients (1,265 of 1,826) being operated on within the first 2 postoperative years (Fig 2). When the 1,826 patients who underwent subsequent ipsilateral outpatient shoulder operations were compared with the 28,604 who did not, those who underwent additional outpatient shoulder procedures were significantly more likely to be younger (53.7 10.9 years v 56.8 11.5 years, P < .001). Male patients tended to have a higher rate of subsequent surgery than female patients, although the difference was not statistically significant (6.2% v 5.7%, P ¼ .062). Workers’ Compensation insurance at the time of initial ARCR showed the greatest rate of subsequent surgery among all forms of insurance (11.2%, P < .001). Patients who had received an acromioplasty at the time of initial ARCR had statistically lower rates of subsequent surgery than those who had not (5.7% v 6.8%, P < .001). Patients who were
4
S. A. MAHURE ET AL.
Fig 2. Timing of repeat ipsilateral shoulder surgery after outpatient arthroscopic rotator cuff repair.
identified with a positive tobacco use history had a greater rate of subsequent shoulder surgery than those without such a history (7.3% v 5.9%, P ¼ .044). Table 1 lists complete comparisons.
Multivariate analysis identified that when compared with patients older than 65 years, age younger than 45 years (odds ratio [OR], 1.88; 95% confidence interval [CI], 1.58-2.23; P < .001) and age between 45
Table 1. Comparison of Patients Who Required Additional Ipsilateral Shoulder Surgery After Outpatient Arthroscopic Rotator Cuff Repair and Patients Who Did Not Over 2- to 12-Year Follow-up (N ¼ 30,340) Patients With No Additional Surgical Procedures 94.0% (28,604/30,430)
Total Age, mean SD, yr All patients 56.8 11.5 Male 55.6 11.7 Female 58.3 11.1 Sex Male 93.8% (15,707/16,751) Female 94.3% (12,897/13,679) Primary insurance Private 94.7% (16,848/17,795) Medicare 95.8% (6,284/6,562) Medicaid 92.7% (1,092/1,178) Workers’ Compensation 88.8% (3,801/4,280) VA/military 94.1% (112/119) Other 93.5% (464/496) Acromioplasty at time of initial RTC repair Yes 94.3% (21,241/22,526) No 93.2% (7,363/7,904) Tobacco use Yes 92.7% (1,165/1,257) No 94.1% (27,439/29,173)
Patients With Subsequent Outpatient Ipsilateral Shoulder Surgery 6.0% (1,826/30,430) 53.7 10.9 52.7 11.2 54.9 10.2
Significance
P P P P
< < < ¼
.001* .001* .001* .062
6.2% (1,044/16,751) 5.7% (782/13,679) 5.3% 4.2% 7.3% 11.2% 5.9% 6.5%
(947/17,795) (275/6,562) (86/1,178) (479/4,280) (7/119) (32/496)
P < .001*
5.7% (1,285/22,526) 6.8% (541/7,904)
P < .001*
7.3% (92/1,257) 5.9% (1,734/29,173)
P ¼ .044*
NOTE. Data are reported as percentages (proportions within particular demographic categories) unless otherwise indicated. Age listed at subsequent surgery reflects age at the initial RTC repair. When percentages were calculated, in the event of multiple reinterventions, only the first of a given procedure was used to formulate the numerator. Differences in age between groups were evaluated with an unpaired t test, differences in sex with the Fisher exact test, and differences in insurance status with c2 analysis. c2 Analysis was calculated for primary insurance by comparing the proportion of patients under Workers’ Compensation versus other types of insurance. RTC, rotator cuff; SD, standard deviation; VA, Veterans Affairs. *Statistically significant.
5
SUBSEQUENT SURGERY AFTER ARCR
Table 2. Univariate and Multivariate Analysis With Adjusted Odds Ratios Identifying Independent Risk Factors for Subsequent Ipsilateral Shoulder Surgery Risk Factors for Subsequent Ipsilateral Shoulder Surgery After ARCR With or Without SAD Univariate Analysis OR Age <45 yr 2.29 45-65 yr 1.79 >65 yr Reference Sex Male v female 1.10 Insurance Workers’ 2.32 Compensation v all others Acromioplasty at time of initial RCR No v yes 1.21 Smoking history Yes v no 1.25
Multivariate Analysis
95% CI
Significance
Adjusted OR
Adjusted 95% CI
Significance
1.93-2.72 1.56-2.05
P < .001* P ¼ .001*
1.88 1.58 Reference
1.58-2.23 1.38-1.82
P < .001* P ¼ .006*
0.99-1.21
P ¼ .062
1.01
0.91-1.11
P ¼ .883
2.08-2.60
P < .001*
2.11
1.89-2.36
P < .001*
1.10-1.35
P < .001*
1.20
1.09-1.34
P < .001*
1.01-1.55
P ¼ .045*
1.18
0.95-1.47
P ¼ .133
CI, confidence interval; ARCR, arthroscopic rotator cuff repair; OR, odds ratio; RCR, rotator cuff repair; SAD, subacromial decompression. *Statistically significant.
and 65 years (OR, 1.58; 95% CI, 1.38-1.82; P ¼ .006) were independent risk factors for subsequent shoulder surgery. In addition, Workers’ Compensation insurance (OR, 2.11; 95% CI, 1.89-2.36; P < .001) and not undergoing an acromioplasty at the time of initial ARCR (OR, 1.20; 95% CI, 1.09-1.34; P < .001) were independent risk factors for subsequent ipsilateral surgery. Although Medicaid insurance was associated with higher rates of reoperation, multivariate analysis showed that it was not an independent risk factor for additional surgery. When focusing only on the 1,047 patients who underwent a subsequent rotator cuff procedure, we found that not undergoing an acromioplasty during initial ARCR was an independent risk factor for reoperation (OR, 1.29; 95% CI, 1.13-1.47; P < .001). Table 2 includes a univariate analysis that represents baseline rates of association of variables with the need for subsequent surgery, as well as a multivariate analysis that reports adjusted ORs within 95% CIs, to determine which variables are independent risk factors for subsequent ipsilateral shoulder surgery. Table 3 reports results for age subgroup analysis (<45 years, 45 to 65 years, and >65 years). Patients younger than 45 years had a significantly greater incidence of subsequent outpatient shoulder surgery (8.0%) when compared with those aged 45 to 65 years (6.4%) or older than 65 years (3.7%, P < .001). For all 3 age subgroups, the most common subsequent surgical procedure was revision RCR. Among patients who underwent subsequent surgery, 65.4% of those older than 65 years underwent either an arthroscopic or open cuff repair, as compared with 58.1% of those aged
between 45 and 65 years and 48.7% of those younger than 45 years (P < .001). At the time of subsequent surgery, 48% of patients older than 65 years had a diagnosis involving a rotator cuff tear, as compared with 38.8% of those aged between 45 and 65 years and 38.0% of those younger than 45 years (P ¼ .018). The remaining diagnoses at the time of additional surgery were largely similar among groups, with the exception of SLAP lesions and instability, which were more prevalent in patients younger than 45 years (P < .001 and P ¼ .012, respectively). Compared with the 2 older-age subgroups, patients younger than 45 years were significantly more likely to undergo SLAP repair (P < .001) or arthroscopic capsulorrhaphy/Bankart repair (P < .001). Patients aged between 45 and 65 years were more likely than the other 2 age subgroups to undergo manipulation/lysis of adhesions (P < .001). Patients older than 65 years undergoing subsequent surgery were significantly more likely to do so earlier (mean, 21.5 25.7 months) than patients aged between 45 and 65 years (mean, 23.7 26.5 months) and patients younger than 45 years (mean, 28.9 30.0 months; P < .001). Female patients undergoing subsequent surgery tended to do so earlier than male patients (22.3 25.5 months v 25.9 28.2 months, P ¼ .006). A positive history of tobacco use resulted in a mean time to subsequent surgery of 16.9 15.4 months as compared with 24.7 27.6 months for patients without a tobacco use history (P < .001). No differences were observed in mean time to surgery regarding Workers’ Compensation insurance or the inclusion of acromioplasty at the time of initial ARCR (Table 4).
6
S. A. MAHURE ET AL.
Table 3. Age Subgroup Comparison of Procedures and Diagnoses for Patients Undergoing Subsequent Shoulder Surgery Patients Undergoing Subsequent Ipsilateral Shoulder Surgery Aged <45 yr % 69.1
n 2,899/4,195
Proportion who underwent SAD at time of initial ARCR Subsequent surgery overall 8.0 337/4,195 Subsequent RTC repair 3.9 164/4,195 Proportion of subsequent 48.7 164/337 surgical procedures that were RTC repair Additional diagnoses RTC tear 38.0 128/337 RTC tendinitis 29.7 100/337 Impingement syndrome 27.3 92/337 SLAP lesion 10.4 35/337 Shoulder instability 8.0 27/337 Acromioclavicular joint arthritis 5.3 18/337 Glenohumeral OA 12.8 43/337 Shoulder stiffness 0.90 3/337 Additional surgery Arthroscopic RTC repair 39.5 133/337 Open RTC repair 9.2 31/337 Subacromial decompression 46.3 156/337 Distal clavicle excision 16.0 54/337 Arthroscopic debridement 26.7 90/337 Arthroscopic synovectomy 3.6 12/337 SLAP repair 7.4 25/337 Arthroscopic biceps tenodesis 2.7 9/337 Open biceps tenodesis 5.1 17/337 Biceps tenotomy 0.0 0/337 Arthroscopic Bankart repair 4.2 14/337 Open instability repair 0.9 3/337 Manipulation/LOA 12.5 42/337 Timing of subsequent surgery in relation to initial ARCR <1 yr 42.7 144/337 Between 1 and 2 yr 20.2 68/337 Between 2 and 3 yr 10.7 36/337 Between 3 and 4 yr 5.9 20/337 Between 4 and 5 yr 4.5 15/337 >5 yr 16.0 54/337
Aged >65 yr
Aged 45-65 yr % 74.8
n 14,457/19,332
% 74.9
n 5,170/6,903
Significance P < .001*
6.4 3.7 58.1
1,235/19,332 717/19,332 717/1,235
3.7 2.4 65.4
254/6,903 166/6,903 166/254
P < .001* P < .001* P < .001*
38.8 31.7 27.4 4.2 4.0 4.5 11.9 1.90
479/1,235 391/1,235 338/1,235 52/1,235 49/1,235 55/1,235 147/1,235 2/1,235
48.0 30.7 24.8 3.9 4.3 4.3 15.0 0.80
122/254 78/254 63/254 10/254 11/254 11/254 38/254 2/254
P P P P P P P P
¼ ¼ ¼ < ¼ ¼ ¼ ¼
.018* .784 .711 .001* .012* .775 .385 .348
46.0 12.2 43.0 14.7 24.9 3.2 2.4 4.1 4.1 0.08 1.1 0.3 17.8
567/1,235 150/1,235 531/1,235 181/1,235 307/1,235 39/1,235 30/1,235 50/1,235 51/1,235 1/1,235 14/1,235 4/1,235 221/1,235
52.4 13.0 44.1 10.2 20.1 2.4 1.1 3.2 4.3 0.0 0.4 0.0 9.5
133/254 33/254 112/254 26/254 51/254 6/254 3/254 8/254 11/254 0/254 1/254 0/254 24/254
P P P P P P P P P P P P P
¼ ¼ ¼ ¼ ¼ ¼ < ¼ ¼ ¼ < ¼ <
.007* .248 .550 .106 .157 .753 .001* .493 .716 .789 .001* .224 .001*
49.2 20.8 9.0 5.9 3.6 11.4
608/1,235 257/1,235 111/1,235 73/1,235 45/1,235 141/1,235
58.3 15.8 5.5 5.1 5.1 10.2
148/254 40/254 14/254 13/254 13/254 26/254
P ¼ .018*
NOTE. P values were calculated with 2-tailed c2 analysis or the Fisher exact test. ARCR, arthroscopic rotator cuff repair; LOA, lysis of adhesions; OA, osteoarthritis; RCR, rotator cuff repair; RTC, rotator cuff; SAD, subacromial decompression. *Statistically significant.
Discussion Our study found a 6% incidence of repeat outpatient surgical intervention after isolated ARCR (with or without acromioplasty) in New York State. Across all age groups, we found that the most common reason (58.7%) for subsequent surgery was revision RCR, thus resulting in an overall rotator cuff revision rate of 3.4% within our population (1,071 of 30,430). Because of the paucity of literature on rates of revision surgery after ARCR, it is difficult to comment on how our data compare with those of other authors. For example, of the recent case series examining tendon integrity after RCR,8,23-26 only 1 study reported on the rate of subsequent surgery (22% [14 of 64] 2 years after primary
RCR).23 All patients in the aforementioned study had retears, and the factors predicting reoperation in this sample were not specifically addressed.23 In addition, our finding that 9.1% of patients who underwent subsequent surgery (166 of 1,826) ultimately underwent 2 or more procedures is comparable to previously published values of 8% to 13%.2,4 Sherman et al.37 previously attempted to use the SPARCS database to identify readmission and revision rates after RCR. They found a 4% rate of revision after RCR at 1 year that was associated with increasing age, medical comorbidities, and decreasing surgeon volume. Their study differs from ours in several significant respects. First, we followed up patients in the database for
SUBSEQUENT SURGERY AFTER ARCR Table 4. Mean Time Until Subsequent Surgery Based on Demographic Characteristics Time Until Subsequent Ipsilateral Shoulder Procedure (n ¼ 1,826) Mean, mo Age <45 yr 28.9 45-65 yr 23.7 >65 yr 21.5 Sex Male 25.9 Female 22.3 Insurance Workers’ 22.5 Compensation All others 25.0 Acromioplasty at time of initial RCR Yes 24.1 No 24.9 Tobacco history Yes 16.9 No 24.7
SD, mo
Significance
30.0 26.5 25.7
P < .001*
28.2 25.5
P ¼ .006*
23.3
P ¼ .081
28.4 27.6 26.1
P ¼ .604
15.4 27.6
P < .001*
RCR, rotator cuff repair; SD, standard deviation. *Statistically significant.
2 years or more. Second, whereas Sherman et al. examined for only revision RCRs, we examined for all subsequent forms of outpatient shoulder procedures. Finally, Sherman et al. identified both open RCR and ARCR from the inpatient and outpatient SPARCS databases between 1997 and 2002. We limited our initial cohort to ARCR identified in the outpatient database for several reasons: There has been a substantial shift toward arthroscopic and outpatient RCR,9,10,12,13 and we wanted our study to reflect current practice trends. In addition, the outpatient and inpatient databases use 2 different procedural coding systems: CPT and ICD9-CM, respectively. The ICD-9-CM system used in the inpatient database does not have a laterality indicator, and thus it is impossible to determine if any subsequent procedures were even on the same side as the original RCR. Sherman et al. listed the inability to determine laterality as a limitation in their results. Given that numerous studies have reported rates of contralateral rotator cuff tears ranging from 38.6% to 67.3%, we did not use the inpatient SPARCS database, thus avoiding the ambiguity regarding laterality.38,39 For this reason, we believe our data best reflect current practice patterns while also providing the most accurate assessment of subsequent shoulder surgery to date. Reasons for failure after RCR are likely multifactorial. Advancing age is believed to be an important risk factor for failed RCR, with poorer tissue quality and larger tears noted in these patients.29,30 However, small case series of patients older than 65 years have found postoperative results comparable to younger patients at
7
early- to mid-term follow-up.40-42 In contrast to existing literature,29,30,37 we found younger age to be an independent risk factor for subsequent outpatient shoulder surgery. In New York State, 8.0% of patients younger than 45 years treated with RCR underwent additional outpatient shoulder procedures compared with only 3.7% of those older than 65 years (P < .001). It is possible that the greater physical demands of younger patients motivated subsequent outpatient surgery or that there existed concomitant pathology not addressed during the initial RCR procedure. For example, we reported that SLAP repairs and arthroscopic capsulorrhaphy comprised 7.4% and 4.2% of subsequent surgical procedures, respectively, in patients younger than 45 years.43 These associations are in accord with the orthopaedic literature because many authors have reported on the association between glenohumeral instability episodes or labral pathology and concomitant rotator cuff pathology, particularly in younger patients.43-47 Although patients older than 65 years had the lowest overall subsequent surgery rate (3.7%) among all age groups, it is important to keep in mind that 65.4% of subsequent procedures performed on those older than 65 years were either open or arthroscopic revision cuff repair. Thus, compared with the 2 younger-age subgroups, these patients had the highest rate of failure of initial ARCR, which corroborates the existing literature. In addition, we believe that the low rate of subsequent outpatient shoulder surgery in this demographic of older patients may be due to age-related trends in management. It is plausible that older patients with a failed ARCR are managed with a shoulder arthroplasty instead of a revision outpatient shoulder procedure. In 2011 reverse total shoulder arthroplasty (RTSA) accounted for one-third of all shoulder arthroplasties performed in the United States, and 81.4% of patients receiving an RTSA were aged 65 years or older.48 Moreover, 52.5% of patients receiving an RTSA had a diagnosis of rotator cuff tear/syndrome.49 Thus it is possible that older patients with irreparable rotator cuff tears are being managed by arthroplasty and thus are not captured in our study. In addition to age, several other factors have been suggested to predict failure after RCR. Workers’ Compensation has been shown to predict poor outcomes of RCR, as well as other arthroscopic procedures.50,51 Our results support this because Workers’ Compensation was identified as an independent risk factor for subsequent surgery in our series (OR, 2.11; 95% CI, 1.89-2.36; P < .001). Although no differences were observed between the sexes regarding reoperation rates, subgroup analysis did show that female patients underwent subsequent surgery earlier than male patients (22.3 months v 25.9 months, P ¼ .006). Previous authors have reported
8
S. A. MAHURE ET AL.
that female sex is associated with higher pain scores, slower improvements in range of motion, and increased dissatisfaction after ARCR.52-54 Because the SPARCS database does not provide any of these clinical or subjective outcome measures, we cannot comment on whether the same causal dynamic exists in our cohort and we look forward to future studies that elaborate on the association between sex and reoperation. With these limitations of our data being kept in mind, it is possible that our findings that age older than 65 years was associated with earlier time until reoperation (Table 4) could be reflective of literature suggesting that advanced age is associated with greater retear rates and poor tissue quality.55-57 Although our multivariate analysis did not show tobacco use as an independent risk factor for subsequent ipsilateral surgery (OR, 1.18; 95% CI, 0.95-1.47; P ¼ .133), our results regarding time until subsequent surgery and tobacco use add to the existing literature. In our sample, patients with a history of tobacco use underwent subsequent surgery approximately 8 months earlier than those without a history of tobacco use (16.9 months v 24.7 months, P < .001). Several authors have reported on the impact of smoking and tobacco use as negative prognostic indicators in rotator cuff pathology and after RCR.29,58,59 Nicotine has been reported to be a potent vasoconstrictor, and the resultant hypoxia can result in remarkable inflammation and tissue destruction, particularly in hypovascular areas such as the soft tissue of the shoulder.60,61 Bishop et al.58 reported in their systematic review that smoking and tobacco use are associated with larger rotator cuff tears, poor patient outcome scores, and accelerated rotator cuff degeneration. As health care continues to be more intimately linked to quality and outcomes, it is important for orthopaedic surgeons and an interdisciplinary team to routinely screen patients for tobacco use and offer counseling and treatment modalities to encourage smoking cessation. We also identified lack of acromioplasty at the time of initial ARCR as an independent risk factor for subsequent ipsilateral shoulder surgery (OR, 1.20; 95% CI, 1.09-1.34; P < .001) and subsequent cuff-related procedures (OR, 1.29; 95% CI, 1.13-1.47; P < .001). Opinions regarding the utility of acromioplasty with RCR vary: Some authors have suggested that acromioplasty can damage the integrity of the coracoacromial arch with resultant pathology in the glenohumeral joint,62,63 whereas others have reported positive outcomes coupled with an increasing incidence of acromioplasty procedures.64-68 Our analysis showed that from 2003 through 2011, there was a consistent rise in the proportion of RCRs performed with acromioplasty, with the trend reversing between 2012 and 2014 (Fig 1). Future studies comparing the revision rates between these 2 periods are warranted to
provide further evidence regarding the utility of acromioplasty in RCR procedures. Other factors that have been suggested to influence failure after RCR that we were not able to explore in this study include genetics,69 method of repair (single v double row, simple v mattress sutures),29,30 results of magnetic resonance imaging (e.g., fatty atrophy, degenerative tendon),70 or postsurgical rehabilitation.71 Similarly, despite our results regarding acromioplasty, we were unable to account for acromial morphology, which has been implicated in the success of acromioplasty and rotator cuff pathology treatment.72-75 Despite some of the limitations of our study, our large sample size and ability to longitudinally follow up patients provide meaningful data to physicians regarding risk factors for reoperation after ARCR. Limitations There are inherent limitations when performing studies using large statewide databases.76 Studies like this one rely on complete and accurate reporting of data. We had no way of confirming if additional procedures were performed to address sequelae of the initial RCR repair. It remains conceivable that surgery was performed for an unrelated or new problem, thus overestimating the need for subsequent surgery. Conversely, we could not identify patients who sought out-of-state care for subsequent procedures. As such, we may be underestimating the incidence of subsequent surgery. However, our analysis identified New York State as the home state for 92.5% of our sample, potentially limiting the number of patients who sought care in other states. In addition, although we eliminated patients with prior ipsilateral shoulder procedures, the SPARCS database only captures procedures within New York State. Therefore, patients who may have previously undergone an ARCR performed outside of New York State could potentially have been included in our sample. We believe that the probability of this making a significant difference in results, however, is minimal given our large sample size and the fact that 92.5% of our sample listed New York as the home state. Another limitation is that we could not account for surgeon experience, technical skill, or advanced fellowship training. Finally, given the limitations of ICD-9 coding, we were unable to comment on patients who underwent subsequent inpatient rotator cuff repair or shoulder arthroplasty.
Conclusions We identified a 6.0% incidence of repeat ipsilateral surgery after isolated ARCR. Although reasons for reoperation are likely multifactorial, younger age, Workers’ Compensation claim, and absence of acromioplasty at the time of initial ARCR remained independent predictors of subsequent outpatient
SUBSEQUENT SURGERY AFTER ARCR
procedures, whereas a history of tobacco use was associated with accelerated time to subsequent surgery. 16.
References 1. Denard PJ, Jiwani AZ, Ladermann A, Burkhart SS. Longterm outcome of a consecutive series of subscapularis tendon tears repaired arthroscopically. Arthroscopy 2012;28:1587-1591. 2. Ladermann A, Denard PJ, Burkhart SS. Midterm outcome of arthroscopic revision repair of massive and nonmassive rotator cuff tears. Arthroscopy 2011;27:1620-1627. 3. Paxton ES, Teefey SA, Dahiya N, Keener JD, Yamaguchi K, Galatz LM. Clinical and radiographic outcomes of failed repairs of large or massive rotator cuff tears: Minimum ten-year follow-up. J Bone Joint Surg Am 2013;95:627-632. 4. Chuang MJ, Jancosko J, Nottage WM. Clinical outcomes of single-row arthroscopic revision rotator cuff repair. Orthopedics 2014;37:e692-e698. 5. Lin EC, Mall NA, Dhawan A, et al. Arthroscopic primary rotator cuff repairs in patients aged younger than 45 years. Arthroscopy 2013;29:811-817. 6. Bhatia S, Greenspoon JA, Horan MP, Warth RJ, Millett PJ. Two-year outcomes after arthroscopic rotator cuff repair in recreational athletes older than 70 years. Am J Sports Med 2015;43:1737-1742. 7. Moraiti C, Valle P, Maqdes A, et al. Comparison of functional gains after arthroscopic rotator cuff repair in patients over 70 years of age versus patients under 50 years of age: A prospective multicenter study. Arthroscopy 2015;31:184-190. 8. Pauly S, Stahnke K, Klatte-Schulz F, Wildemann B, Scheibel M, Greiner S. Do patient age and sex influence tendon cell biology and clinical/radiographic outcomes after rotator cuff repair? Am J Sports Med 2015;43: 549-556. 9. Colvin AC, Egorova N, Harrison AK, Moskowitz A, Flatow EL. National trends in rotator cuff repair. J Bone Joint Surg Am 2012;94:227-233. 10. Ensor KL, Kwon YW, Dibeneditto MR, Zuckerman JD, Rokito AS. The rising incidence of rotator cuff repairs. J Shoulder Elbow Surg 2013;22:1628-1632. 11. Zhang AL, Montgomery SR, Ngo SS, Hame SL, Wang JC, Gamradt SC. Analysis of rotator cuff repair trends in a large private insurance population. Arthroscopy 2013;29: 623-629. 12. Jain NB, Higgins LD, Losina E, Collins J, Blazar PE, Katz JN. Epidemiology of musculoskeletal upper extremity ambulatory surgery in the United States. BMC Musculoskelet Disord 2014;15:4. 13. Cullen KA, Hall MJ, Golosinskiy A. Ambulatory surgery in the United States, 2006. Natl Health Stat Report 2009:1-25. 14. Brown MJ, Pula DA, Kluczynski MA, Mashtare T, Bisson LJ. Does suture technique affect re-rupture in arthroscopic rotator cuff repair? A meta-analysis. Arthroscopy 2015;31:1576-1582. 15. Akpinar S, Uysal M, Pourbagher MA, Ozalay M, Cesur N, Hersekli MA. Prospective evaluation of the functional and anatomical results of arthroscopic repair in small and
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
9
medium-sized full-thickness tears of the supraspinatus tendon. Acta Orthop Traumatol Turc 2011;45:248-253. Kim KC, Shin HD, Cha SM, Kim JH. Repair integrity and functional outcomes for arthroscopic margin convergence of rotator cuff tears. J Bone Joint Surg Am 2013;95: 536-541. Deutsch A, Kroll DG, Hasapes J, Staewen RS, Pham C, Tait C. Repair integrity and clinical outcome after arthroscopic rotator cuff repair using single-row anchor fixation: A prospective study of single-tendon and twotendon tears. J Shoulder Elbow Surg 2008;17:845-852. Lubiatowski P, Kaczmarek P, Dzianach M, et al. Clinical and biomechanical performance of patients with failed rotator cuff repair. Int Orthop 2013;37:2395-2401. Franceschi F, Papalia R, Vasta S, Leonardi F, Maffulli N, Denaro V. Surgical management of irreparable rotator cuff tears. Knee Surg Sports Traumatol Arthrosc 2015;23: 494-501. Liem D, Lengers N, Dedy N, Poetzl W, Steinbeck J, Marquardt B. Arthroscopic debridement of massive irreparable rotator cuff tears. Arthroscopy 2008;24: 743-748. Verhelst L, Vandekerckhove PJ, Sergeant G, Liekens K, Van Hoonacker P, Berghs B. Reversed arthroscopic subacromial decompression for symptomatic irreparable rotator cuff tears: Mid-term follow-up results in 34 shoulders. J Shoulder Elbow Surg 2010;19:601-608. 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;96:265-271. Shamsudin A, Lam PH, Peters K, Rubenis I, Hackett L, Murrell GA. Revision versus primary arthroscopic rotator cuff repair: A 2-year analysis of outcomes in 360 patients. Am J Sports Med 2015;43:557-564. Cho NS, Rhee YG. The factors affecting the clinical outcome and integrity of arthroscopically repaired rotator cuff tears of the shoulder. Clin Orthop Surg 2009;1:96-104. Cole BJ, McCarty LP III, Kang RW, Alford W, Lewis PB, Hayden JK. Arthroscopic rotator cuff repair: Prospective functional outcome and repair integrity at minimum 2year follow-up. J Shoulder Elbow Surg 2007;16:579-585. Boileau P, Brassart N, Watkinson DJ, Carles M, Hatzidakis AM, Krishnan SG. Arthroscopic repair of fullthickness tears of the supraspinatus: Does the tendon really heal? J Bone Joint Surg Am 2005;87:1229-1240. Sperling JW, Cofield RH, Schleck C. Rotator cuff repair in patients fifty years of age and younger. J Bone Joint Surg Am 2004;86:2212-2215. Cuff DJ, Pupello DR. Prospective evaluation of postoperative compliance and outcomes after rotator cuff repair in patients with and without workers’ compensation claims. J Shoulder Elbow Surg 2012;21:1728-1733. Abtahi AM, Granger EK, Tashjian RZ. Factors affecting healing after arthroscopic rotator cuff repair. World J Orthop 2015;6:211-220. Tashjian RZ, Hollins AM, Kim HM, et al. Factors affecting healing rates after arthroscopic double-row rotator cuff repair. Am J Sports Med 2010;38:2435-2442. Onyekwelu I, Khatib O, Zuckerman JD, Rokito AS, Kwon YW. The rising incidence of arthroscopic superior
10
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
S. A. MAHURE ET AL. labrum anterior and posterior (SLAP) repairs. J Shoulder Elbow Surg 2012;21:728-731. Bansal A, Khatib ON, Zuckerman JD. Revision total joint arthroplasty: The epidemiology of 63,140 cases in New York State. J Arthroplasty 2014;29:23-27. Khatib O, Onyekwelu I, Zuckerman JD. The incidence of proximal humeral fractures in New York State from 1990 through 2010 with an emphasis on operative management in patients aged 65 years or older. J Shoulder Elbow Surg 2014;23:1356-1362. Siddiq F, Adil MM, Malik AA, Qureshi MH, Qureshi AI. Effect of carotid revascularization endarterectomy versus stenting trial results on the performance of carotid artery stent placement and carotid endarterectomy in the United States. Neurosurgery 2015;75:726-732. Movahed MR, Hashemzadeh M, Jamal MM. Increased prevalence of third-degree atrioventricular block in patients with type II diabetes mellitus. Chest 2005;128: 2611-2614. Macleod LC, Dai JC, Holt SK, Bassett JC, Wright JL, Gore JL. Underuse and underreporting of smoking cessation for smokers with a new urologic cancer diagnosis. Urol Oncol 2015;33:504.e1-e7. Sherman SL, Lyman S, Koulouvaris P, Willis A, Marx RG. Risk factors for readmission and revision surgery following rotator cuff repair. Clin Orthop Relat Res 2008;466:608-613. Liem D, Buschmann VE, Schmidt C, et al. The prevalence of rotator cuff tears: Is the contralateral shoulder at risk? Am J Sports Med 2014;42:826-830. Ro KH, Park JH, Lee SH, Song DI, Jeong HJ, Jeong WK. Status of the contralateral rotator cuff in patients undergoing rotator cuff repair. Am J Sports Med 2015;43: 1091-1098. Verma NN, Bhatia S, Baker CL III, et al. Outcomes of arthroscopic rotator cuff repair in patients aged 70 years or older. Arthroscopy 2010;26:1273-1280. Charousset C, Bellaiche L, Kalra K, Petrover D. Arthroscopic repair of full-thickness rotator cuff tears: Is there tendon healing in patients aged 65 years or older? Arthroscopy 2010;26:302-309. Osti L, Papalia R, Del Buono A, Denaro V, Maffulli N. Comparison of arthroscopic rotator cuff repair in healthy patients over and under 65 years of age. Knee Surg Sports Traumatol Arthrosc 2010;18:1700-1706. Gombera MM, Sekiya JK. Rotator cuff tear and glenohumeral instability: A systematic review. Clin Orthop Relat Res 2014;472:2448-2456. Araghi A, Prasarn M, St Clair S, Zuckerman JD. Recurrent anterior glenohumeral instability with onset after forty years of age: The role of the anterior mechanism. Bull Hosp Jt Dis 2005;62:99-101. Craig EV. The posterior mechanism of acute anterior shoulder dislocations. Clin Orthop Relat Res 1984: 212-216. Erickson J, Lavery K, Monica J, Gatt C, Dhawan A. Surgical treatment of symptomatic superior labrum anterior-posterior tears in patients older than 40 years: A systematic review. Am J Sports Med 2015;43:1274-1282. Goldberg JA, Chan KY, Best JP, Bruce WJ, Walsh W, Parry W. Surgical management of large rotator cuff tears
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
combined with instability in elite rugby football players. Br J Sports Med 2003;37:179-181; discussion 181. Schairer WW, Nwachukwu BU, Lyman S, Craig EV, Gulotta LV. National utilization of reverse total shoulder arthroplasty in the United States. J Shoulder Elbow Surg 2015;24:91-97. Jain NB, Yamaguchi K. The contribution of reverse shoulder arthroplasty to utilization of primary shoulder arthroplasty. J Shoulder Elbow Surg 2014;23:1905-1912. Henn RF III, Tashjian RZ, Kang L, Green A. Patients with workers’ compensation claims have worse outcomes after rotator cuff repair. J Bone Joint Surg Am 2008;90: 2105-2113. Razmjou H, Lincoln S, Axelrod T, Holtby R. Factors contributing to failure of rotator cuff surgery in persons with work-related injuries. Physiother Can 2008;60: 125-133. Cho CH, Ye HU, Jung JW, Lee YK. Gender affects early postoperative outcomes of rotator cuff repair. Clin Orthop Surg 2015;7:234-240. Razmjou H, Davis AM, Jaglal SB, Holtby R, Richards RR. Disability and satisfaction after rotator cuff decompression or repair: A sex and gender analysis. BMC Musculoskelet Disord 2011;12:66. Robinson PM, Wilson J, Dalal S, Parker RA, Norburn P, Roy BR. Rotator cuff repair in patients over 70 years of age: Early outcomes and risk factors associated with retear. Bone Joint J 2013;95-B:199-205. 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;43:491-500. Gill TJ, McIrvin E, Kocher MS, Homa K, Mair SD, Hawkins RJ. The relative importance of acromial morphology and age with respect to rotator cuff pathology. J Shoulder Elbow Surg 2002;11:327-330. Le BT, Wu XL, Lam PH, Murrell GA. Factors predicting rotator cuff retears: An analysis of 1000 consecutive rotator cuff repairs. Am J Sports Med 2014;42:1134-1142. Bishop JY, Santiago-Torres JE, Rimmke N, Flanigan DC. Smoking predisposes to rotator cuff pathology and shoulder dysfunction: A systematic review. Arthroscopy 2015;31:1598-1605. Santiago-Torres J, Flanigan DC, Butler RB, Bishop JY. The effect of smoking on rotator cuff and glenoid labrum surgery: A systematic review. Am J Sports Med 2015;43: 745-751. Tashjian RZ. Epidemiology, natural history, and indications for treatment of rotator cuff tears. Clin Sports Med 2012;31:589-604. Babayan RK. Wound healing and infection in surgery: The pathophysiological impact of smoking, smoking cessation, and nicotine replacement therapy: A systematic review. J Urol 2012;188:2243-2244. MacDonald P, McRae S, Leiter J, Mascarenhas R, Lapner P. Arthroscopic rotator cuff repair with and without acromioplasty in the treatment of full-thickness rotator cuff tears: A multicenter, randomized controlled trial. J Bone Joint Surg Am 2011;93:1953-1960. Budoff JE, Nirschl RP, Guidi EJ. Debridement of partialthickness tears of the rotator cuff without acromioplasty.
SUBSEQUENT SURGERY AFTER ARCR
64.
65.
66.
67.
68.
69.
70.
Long-term follow-up and review of the literature. J Bone Joint Surg Am 1998;80:733-748. Romeo AA, Hang DW, Bach BR Jr, Shott S. Repair of full thickness rotator cuff tears. Gender, age, and other factors affecting outcome. Clin Orthop Relat Res 1999:243-255. Vitale MA, Arons RR, Hurwitz S, Ahmad CS, Levine WN. The rising incidence of acromioplasty. J Bone Joint Surg Am 2010;92:1842-1850. Chin PY, Sperling JW, Cofield RH, Stuart MJ, Crownhart BS. Anterior acromioplasty for the shoulder impingement syndrome: Long-term outcome. J Shoulder Elbow Surg 2007;16:697-700. Frieman BG, Fenlin JM Jr. Anterior acromioplasty: Effect of litigation and workers’ compensation. J Shoulder Elbow Surg 1995;4:175-181. Ellman H. Arthroscopic subacromial decompression: Analysis of one- to three-year results. Arthroscopy 1987;3: 173-181. Gwilym SE, Watkins B, Cooper CD, et al. Genetic influences in the progression of tears of the rotator cuff. J Bone Joint Surg Br 2009;91:915-917. Gladstone JN, Bishop JY, Lo IK, Flatow EL. Fatty infiltration and atrophy of the rotator cuff do not
71.
72.
73.
74. 75.
76.
11
improve after rotator cuff repair and correlate with poor functional outcome. Am J Sports Med 2007;35: 719-728. Gallagher BP, Bishop ME, Tjoumakaris FP, Freedman KB. Early versus delayed rehabilitation following arthroscopic rotator cuff repair: A systematic review. Phys Sportsmed 2015;43:178-187. Balke M, Schmidt C, Dedy N, Banerjee M, Bouillon B, Liem D. Correlation of acromial morphology with impingement syndrome and rotator cuff tears. Acta Orthop 2013;84:178-183. Neer CS II. Anterior acromioplasty for the chronic impingement syndrome in the shoulder: A preliminary report. J Bone Joint Surg Am 1972;54:41-50. Neer CS II. Impingement lesions. Clin Orthop Relat Res 1983:70-77. Zuckerman JD, Kummer FJ, Cuomo F, Simon J, Rosenblum S, Katz N. The influence of coracoacromial arch anatomy on rotator cuff tears. J Shoulder Elbow Surg 1992;1:4-14. Memtsoudis SG. Limitations associated with the analysis of data from administrative databases. Anesthesiology 2009;111:449; author reply 450-451 (letter).
Purpose: To quantify the incidence and risk factors associated with subsequent shoulder procedures in individuals undergoing outpatient arthroscopic rotator cuff repair (ARCR). Methods: We examined the New York Statewide Planning and Research Cooperative Systems outpatient database from 2003 through 2014 to identify patients undergoing isolated ARCR with or without concomitant acromioplasty. Patients were longitudinally followed up for a minimum of 2 years to determine the incidence of subsequent ipsilateral shoulder surgery. The impact of age, sex, insurance, concomitant acromioplasty, and tobacco use on reoperation was explored. Results: Between 2003 and 2012, 30,430 patients underwent isolated ARCR. The mean age was 56.6 11.5 years, and 55.1% were male patients. A total of 1,826 patients (6.0%) underwent subsequent ipsilateral outpatient shoulder surgery a mean of 24.3 27.1 months after the initial ARCR. Of patients who underwent repeat surgery, 57.3% underwent a revision cuff repair. Patients who underwent additional outpatient shoulder surgery were significantly younger (53.7 10.9 years v 56.8 11.5 years, P < .001). Tobacco use was associated with an increased rate of subsequent surgery (7.3% v 5.9%, P ¼ .044) and accelerated time to reoperation (16.9 months v 24.7 months, P < .001). Independent risk factors for subsequent ipsilateral surgery after initial ARCR were presence of a Workers’ Compensation claim (odds ratio, 2.11; 95% confidence interval, 1.89-2.36; P < .001) and initial ARCR without acromioplasty (odds ratio, 1.20; 95% confidence interval, 1.09-1.34; P < .001). Conclusions: We identified a 6.0% incidence of repeat ipsilateral surgery after isolated ARCR. Although reasons for reoperation are likely multifactorial, younger age, Workers’ Compensation claim, and absence of acromioplasty at the time of initial ARCR remained independent predictors of subsequent outpatient procedures, whereas a history of tobacco use was associated with accelerated time to subsequent surgery. Level of Evidence: Level III, retrospective comparative study.
A
rthroscopic rotator cuff repair (ARCR) is recognized to provide good to excellent long-term clinical results in patients with symptomatic rotator cuff tears failing conservative therapy.1-4 Studies have consistently found improved pain, motion, and functional outcome scores across all age groups.5-8 Analyses of large national inpatient databases suggest arthroscopic rotator cuff procedures have increased by upward of 600% between 1996 and 2006.9 In New York State the incidence of rotator cuff repair (RCR) has
From the Department of Orthopaedic Surgery, NYU Hospital for Joint Diseases, New York, New York, U.S.A. The authors report that they have no conflicts of interest in the authorship and publication of this article. Received September 2, 2015; accepted January 15, 2016. Address correspondence to Siddharth A. Mahure, M.D., Department of Orthopaedic Surgery, NYU Hospital for Joint Diseases, 301 E 17th St, New York, NY 10003, U.S.A. E-mail: [email protected] Ó 2016 by the Arthroscopy Association of North America 0749-8063/15857/$36.00 http://dx.doi.org/10.1016/j.arthro.2016.01.039
increased by 238% over a 15-year period,10 with demand further increasing because of an aging population with increasing activity requirements and enhanced access to advanced imaging modalities.9-11 With the enhancement of arthroscopic techniques, greater familiarity among surgeons with the performance of these procedures, and improvements in the management of postoperative pain, most of these procedures are now performed on an outpatient basis.9,10,12,13 The factors involved in the success of ARCR are largely unclear. Although the technical goals of repair are to restore anatomy with a construct that is able to withstand bodily forces while bone-to-tendon healing occurs,14 other studies have questioned the relation between anatomic healing and functional outcomes.15-18 Multiple studies have found favorable results when only partial repairs or debridement procedures were possible because of tendon quality or extent of the tear.19-21 Furthermore, a high proportion of patients with retears after RCR still report excellent functional outcome scores that were much improved from their
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol
-,
No
-
(Month), 2016: pp 1-11
1
2
S. A. MAHURE ET AL.
preoperative status.15-17 Indeed, a recent systematic review found that repair integrity was significantly associated with improved strength but not functional outcome scores.22 Even less clear are the factors that predict the need for subsequent surgery after primary RCR. Overall reoperation rates after RCR remain unclear and have rarely been reported in recent series.23-26 To address this, we examined the Statewide Planning and Research Cooperative Systems (SPARCS) database of the New York State Department of Health. The purpose of this study was to quantify the incidence and risk factors associated with subsequent shoulder procedures in individuals undergoing outpatient ARCR. On the basis of existing literature,27-30 we hypothesized that age and the existence of a Workers’ Compensation claim would be associated with the need for subsequent surgical procedures.
Methods The New York State Department of Health established the SPARCS database in 1979. In 1986 the passage of the New York State Public Health Law mandated all nonfederal, licensed hospitals to report on discharges from their facilities. These requirements were subsequently expanded to include emergency departments and freestanding, licensed ambulatory surgery facilities. The database records the following information for all individuals who have undergone treatment at one of the aforementioned locations: age, sex, race, type of insurance, diagnoses, and procedures performed. Although not without certain limitations, the SPARCS database has been used to collect epidemiologic information for various diseases in previous peer-reviewed publications.10,31-33 For this study, diagnoses and procedures were classified by Current Procedural Terminology (CPT), fourth edition, and International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes. The American Medical Association officially introduced CPT code 29827 for ARCR in 2003. We queried the database from 2003 through 2014 to identify all isolated ARCRs performed. Any patient with additional shoulder CPT codes at the time of initial RCR was excluded, with the exception of subacromial decompression (CPT code 29826). We believed that this would allow us to best capture patients undergoing surgery primarily because of rotator cuff pathology and that the elimination of concomitant CPT codes would provide a more homogeneous sample. We identified 71,920 unique patients who underwent ARCR between 2003 and 2012. Elimination of patients with concomitant CPT codes other than acromioplasty and those with evidence of prior ipsilateral shoulder procedures yielded our final sample of 30,430 patients, representing
approximately 42% of all ARCRs performed. For this homogeneous sample of patients, all available patient demographic data were extracted. Similar to previous studies, our study used International Classification of Diseases, Ninth Revision (ICD-9) codes of 305.1 and V15.62 to identify patients with a history of tobacco use.34-36 A custom-designed computer algorithm using SAS software (version 9.3; SAS Institute, Cary NC) was used to identify and group patients in the SPARCS database. Patients who underwent RCR were subsequently tracked using an encrypted unique patient identifier and followed up for a minimum of 2 years (up to and including 2014). Thus, although we identified all ARCRs from 2003 through 2014, our longitudinal analysis only included patients from 2003 to 2012, thus ensuring a minimum of 2 years’ follow-up for all patients. The timing and nature of each additional shoulder procedure were recorded. For patients who underwent multiple subsequent surgical procedures, we used the first operation immediately after the initial ARCR for purposes of data analysis. Subsequent outpatient surgical procedures were identified using CPT codes for commonly performed arthroscopic and open shoulder procedures. Concomitant ICD-9 diagnosis codes for rotator cuff tears, shoulder instability, glenohumeral arthritis, acromioclavicular joint arthritis, acromioclavicular joint sprains, shoulder stiffness, or rotator cuff tendinitis/syndromes at the time of subsequent surgery were also recorded. Because of limitations in ICD-9 procedural codes used for inpatient procedures, we were unable to comment on patients who may have undergone a subsequent anatomic or reverse shoulder arthroplasty. Statistical analysis using SAS software (version 9.3) was performed to compare patients who underwent additional surgery with patients who did not. To explore the effect of age on the need for subsequent surgery, we conducted a subgroup analysis by stratifying patients into 3 age groups: younger than 45 years, between 45 and 65 years, and older than 65 years. Differences in continuous variables were evaluated with a Student t test and Kruskal-Wallis analysis. Differences in categorical variables were evaluated with the Fisher exact test (when possible) or c2 analysis. Multivariate logistic regression that controlled for age, sex, insurance, tobacco use, and acromioplasty at the time of initial surgery was performed to determine independent risk factors for subsequent surgery. P < .05 was considered statistically significant. Institutional review board approval was not required for this study because no human subjects were involved.
Results Between 2003 and 2014, 39,284 unique patients were recorded as undergoing isolated outpatient ARCR
SUBSEQUENT SURGERY AFTER ARCR
3
Fig 1. Total number of arthroscopic rotator cuff repairs (ARCRs) performed per year from 2003 through 2014 and proportion per year with and without acromioplasty. For each year, the percentage of ARCRs with and without acromioplasty is listed immediately to the right of the blue and red bars, respectively. For example, in 2003, 71.8% of ARCRs were performed with concomitant acromioplasty, whereas 28.2% were performed without acromioplasty. (RCR, rotator cuff repair.)
with or without acromioplasty in New York State. The number of procedures performed annually increased 301% over this 12-year period, from 1,200 in 2003 to 4,812 in 2014. The proportion of ARCRs performed with acromioplasty also increased from 71.8% in 2003 to 76.3% in 2014, although there has been a decreasing trend since 2011. Total volume of ARCRs per year can be found in Figure 1, along with the proportion of ARCRs per year that were performed with and without acromioplasty. Longitudinal analysis was only performed on the 30,430 patients with isolated ARCR between 2003 and 2012. Male patients comprised 55.1% of our sample (16,751 of 30,430). The mean age of our sample was 56.6 11.5 years, with male patients being significantly younger than female patients (55.3 11.7 years v 58.1 11.1 years, P < .001). Primary insurance was most commonly private (58.5%) or Medicare (21.6%), with 14.1% Workers’ Compensation, 3.9% Medicaid, 0.4% Veterans Affairs, and 1.6% other. For the vast majority of our sample (92.5%), New York was listed as the home state. Additional Outpatient Shoulder Surgery During the minimum 2-year follow-up, 6.0% of patients (1,826 of 30,430) underwent subsequent outpatient shoulder surgery on the same shoulder as the initial ARCR. Of those with subsequent surgery, 9.1% (166 of 1,826) underwent 2 or more subsequent ipsilateral operations.
The subsequently performed shoulder procedures included ARCR (45.7%; 833 of 1,826), subacromial decompression (43.8%; 799 of 1,826), arthroscopic debridement (24.5%; 448 of 1,826), open RCR (11.7%; 214 of 1,826), distal clavicle resection (14.3%; 261 of 1,826), and biceps tenodesis/tenotomy (8.1%; 147 of 1,826). Therefore, 3.4% of our initial sample (1,047 of 30,430), or 57.3% of those requiring additional surgery (1,047 of 1,826), underwent subsequent rotator cuff surgery. The mean time to subsequent intervention was 24.4 27.1 months, with 69.3% of patients (1,265 of 1,826) being operated on within the first 2 postoperative years (Fig 2). When the 1,826 patients who underwent subsequent ipsilateral outpatient shoulder operations were compared with the 28,604 who did not, those who underwent additional outpatient shoulder procedures were significantly more likely to be younger (53.7 10.9 years v 56.8 11.5 years, P < .001). Male patients tended to have a higher rate of subsequent surgery than female patients, although the difference was not statistically significant (6.2% v 5.7%, P ¼ .062). Workers’ Compensation insurance at the time of initial ARCR showed the greatest rate of subsequent surgery among all forms of insurance (11.2%, P < .001). Patients who had received an acromioplasty at the time of initial ARCR had statistically lower rates of subsequent surgery than those who had not (5.7% v 6.8%, P < .001). Patients who were
4
S. A. MAHURE ET AL.
Fig 2. Timing of repeat ipsilateral shoulder surgery after outpatient arthroscopic rotator cuff repair.
identified with a positive tobacco use history had a greater rate of subsequent shoulder surgery than those without such a history (7.3% v 5.9%, P ¼ .044). Table 1 lists complete comparisons.
Multivariate analysis identified that when compared with patients older than 65 years, age younger than 45 years (odds ratio [OR], 1.88; 95% confidence interval [CI], 1.58-2.23; P < .001) and age between 45
Table 1. Comparison of Patients Who Required Additional Ipsilateral Shoulder Surgery After Outpatient Arthroscopic Rotator Cuff Repair and Patients Who Did Not Over 2- to 12-Year Follow-up (N ¼ 30,340) Patients With No Additional Surgical Procedures 94.0% (28,604/30,430)
Total Age, mean SD, yr All patients 56.8 11.5 Male 55.6 11.7 Female 58.3 11.1 Sex Male 93.8% (15,707/16,751) Female 94.3% (12,897/13,679) Primary insurance Private 94.7% (16,848/17,795) Medicare 95.8% (6,284/6,562) Medicaid 92.7% (1,092/1,178) Workers’ Compensation 88.8% (3,801/4,280) VA/military 94.1% (112/119) Other 93.5% (464/496) Acromioplasty at time of initial RTC repair Yes 94.3% (21,241/22,526) No 93.2% (7,363/7,904) Tobacco use Yes 92.7% (1,165/1,257) No 94.1% (27,439/29,173)
Patients With Subsequent Outpatient Ipsilateral Shoulder Surgery 6.0% (1,826/30,430) 53.7 10.9 52.7 11.2 54.9 10.2
Significance
P P P P
< < < ¼
.001* .001* .001* .062
6.2% (1,044/16,751) 5.7% (782/13,679) 5.3% 4.2% 7.3% 11.2% 5.9% 6.5%
(947/17,795) (275/6,562) (86/1,178) (479/4,280) (7/119) (32/496)
P < .001*
5.7% (1,285/22,526) 6.8% (541/7,904)
P < .001*
7.3% (92/1,257) 5.9% (1,734/29,173)
P ¼ .044*
NOTE. Data are reported as percentages (proportions within particular demographic categories) unless otherwise indicated. Age listed at subsequent surgery reflects age at the initial RTC repair. When percentages were calculated, in the event of multiple reinterventions, only the first of a given procedure was used to formulate the numerator. Differences in age between groups were evaluated with an unpaired t test, differences in sex with the Fisher exact test, and differences in insurance status with c2 analysis. c2 Analysis was calculated for primary insurance by comparing the proportion of patients under Workers’ Compensation versus other types of insurance. RTC, rotator cuff; SD, standard deviation; VA, Veterans Affairs. *Statistically significant.
5
SUBSEQUENT SURGERY AFTER ARCR
Table 2. Univariate and Multivariate Analysis With Adjusted Odds Ratios Identifying Independent Risk Factors for Subsequent Ipsilateral Shoulder Surgery Risk Factors for Subsequent Ipsilateral Shoulder Surgery After ARCR With or Without SAD Univariate Analysis OR Age <45 yr 2.29 45-65 yr 1.79 >65 yr Reference Sex Male v female 1.10 Insurance Workers’ 2.32 Compensation v all others Acromioplasty at time of initial RCR No v yes 1.21 Smoking history Yes v no 1.25
Multivariate Analysis
95% CI
Significance
Adjusted OR
Adjusted 95% CI
Significance
1.93-2.72 1.56-2.05
P < .001* P ¼ .001*
1.88 1.58 Reference
1.58-2.23 1.38-1.82
P < .001* P ¼ .006*
0.99-1.21
P ¼ .062
1.01
0.91-1.11
P ¼ .883
2.08-2.60
P < .001*
2.11
1.89-2.36
P < .001*
1.10-1.35
P < .001*
1.20
1.09-1.34
P < .001*
1.01-1.55
P ¼ .045*
1.18
0.95-1.47
P ¼ .133
CI, confidence interval; ARCR, arthroscopic rotator cuff repair; OR, odds ratio; RCR, rotator cuff repair; SAD, subacromial decompression. *Statistically significant.
and 65 years (OR, 1.58; 95% CI, 1.38-1.82; P ¼ .006) were independent risk factors for subsequent shoulder surgery. In addition, Workers’ Compensation insurance (OR, 2.11; 95% CI, 1.89-2.36; P < .001) and not undergoing an acromioplasty at the time of initial ARCR (OR, 1.20; 95% CI, 1.09-1.34; P < .001) were independent risk factors for subsequent ipsilateral surgery. Although Medicaid insurance was associated with higher rates of reoperation, multivariate analysis showed that it was not an independent risk factor for additional surgery. When focusing only on the 1,047 patients who underwent a subsequent rotator cuff procedure, we found that not undergoing an acromioplasty during initial ARCR was an independent risk factor for reoperation (OR, 1.29; 95% CI, 1.13-1.47; P < .001). Table 2 includes a univariate analysis that represents baseline rates of association of variables with the need for subsequent surgery, as well as a multivariate analysis that reports adjusted ORs within 95% CIs, to determine which variables are independent risk factors for subsequent ipsilateral shoulder surgery. Table 3 reports results for age subgroup analysis (<45 years, 45 to 65 years, and >65 years). Patients younger than 45 years had a significantly greater incidence of subsequent outpatient shoulder surgery (8.0%) when compared with those aged 45 to 65 years (6.4%) or older than 65 years (3.7%, P < .001). For all 3 age subgroups, the most common subsequent surgical procedure was revision RCR. Among patients who underwent subsequent surgery, 65.4% of those older than 65 years underwent either an arthroscopic or open cuff repair, as compared with 58.1% of those aged
between 45 and 65 years and 48.7% of those younger than 45 years (P < .001). At the time of subsequent surgery, 48% of patients older than 65 years had a diagnosis involving a rotator cuff tear, as compared with 38.8% of those aged between 45 and 65 years and 38.0% of those younger than 45 years (P ¼ .018). The remaining diagnoses at the time of additional surgery were largely similar among groups, with the exception of SLAP lesions and instability, which were more prevalent in patients younger than 45 years (P < .001 and P ¼ .012, respectively). Compared with the 2 older-age subgroups, patients younger than 45 years were significantly more likely to undergo SLAP repair (P < .001) or arthroscopic capsulorrhaphy/Bankart repair (P < .001). Patients aged between 45 and 65 years were more likely than the other 2 age subgroups to undergo manipulation/lysis of adhesions (P < .001). Patients older than 65 years undergoing subsequent surgery were significantly more likely to do so earlier (mean, 21.5 25.7 months) than patients aged between 45 and 65 years (mean, 23.7 26.5 months) and patients younger than 45 years (mean, 28.9 30.0 months; P < .001). Female patients undergoing subsequent surgery tended to do so earlier than male patients (22.3 25.5 months v 25.9 28.2 months, P ¼ .006). A positive history of tobacco use resulted in a mean time to subsequent surgery of 16.9 15.4 months as compared with 24.7 27.6 months for patients without a tobacco use history (P < .001). No differences were observed in mean time to surgery regarding Workers’ Compensation insurance or the inclusion of acromioplasty at the time of initial ARCR (Table 4).
6
S. A. MAHURE ET AL.
Table 3. Age Subgroup Comparison of Procedures and Diagnoses for Patients Undergoing Subsequent Shoulder Surgery Patients Undergoing Subsequent Ipsilateral Shoulder Surgery Aged <45 yr % 69.1
n 2,899/4,195
Proportion who underwent SAD at time of initial ARCR Subsequent surgery overall 8.0 337/4,195 Subsequent RTC repair 3.9 164/4,195 Proportion of subsequent 48.7 164/337 surgical procedures that were RTC repair Additional diagnoses RTC tear 38.0 128/337 RTC tendinitis 29.7 100/337 Impingement syndrome 27.3 92/337 SLAP lesion 10.4 35/337 Shoulder instability 8.0 27/337 Acromioclavicular joint arthritis 5.3 18/337 Glenohumeral OA 12.8 43/337 Shoulder stiffness 0.90 3/337 Additional surgery Arthroscopic RTC repair 39.5 133/337 Open RTC repair 9.2 31/337 Subacromial decompression 46.3 156/337 Distal clavicle excision 16.0 54/337 Arthroscopic debridement 26.7 90/337 Arthroscopic synovectomy 3.6 12/337 SLAP repair 7.4 25/337 Arthroscopic biceps tenodesis 2.7 9/337 Open biceps tenodesis 5.1 17/337 Biceps tenotomy 0.0 0/337 Arthroscopic Bankart repair 4.2 14/337 Open instability repair 0.9 3/337 Manipulation/LOA 12.5 42/337 Timing of subsequent surgery in relation to initial ARCR <1 yr 42.7 144/337 Between 1 and 2 yr 20.2 68/337 Between 2 and 3 yr 10.7 36/337 Between 3 and 4 yr 5.9 20/337 Between 4 and 5 yr 4.5 15/337 >5 yr 16.0 54/337
Aged >65 yr
Aged 45-65 yr % 74.8
n 14,457/19,332
% 74.9
n 5,170/6,903
Significance P < .001*
6.4 3.7 58.1
1,235/19,332 717/19,332 717/1,235
3.7 2.4 65.4
254/6,903 166/6,903 166/254
P < .001* P < .001* P < .001*
38.8 31.7 27.4 4.2 4.0 4.5 11.9 1.90
479/1,235 391/1,235 338/1,235 52/1,235 49/1,235 55/1,235 147/1,235 2/1,235
48.0 30.7 24.8 3.9 4.3 4.3 15.0 0.80
122/254 78/254 63/254 10/254 11/254 11/254 38/254 2/254
P P P P P P P P
¼ ¼ ¼ < ¼ ¼ ¼ ¼
.018* .784 .711 .001* .012* .775 .385 .348
46.0 12.2 43.0 14.7 24.9 3.2 2.4 4.1 4.1 0.08 1.1 0.3 17.8
567/1,235 150/1,235 531/1,235 181/1,235 307/1,235 39/1,235 30/1,235 50/1,235 51/1,235 1/1,235 14/1,235 4/1,235 221/1,235
52.4 13.0 44.1 10.2 20.1 2.4 1.1 3.2 4.3 0.0 0.4 0.0 9.5
133/254 33/254 112/254 26/254 51/254 6/254 3/254 8/254 11/254 0/254 1/254 0/254 24/254
P P P P P P P P P P P P P
¼ ¼ ¼ ¼ ¼ ¼ < ¼ ¼ ¼ < ¼ <
.007* .248 .550 .106 .157 .753 .001* .493 .716 .789 .001* .224 .001*
49.2 20.8 9.0 5.9 3.6 11.4
608/1,235 257/1,235 111/1,235 73/1,235 45/1,235 141/1,235
58.3 15.8 5.5 5.1 5.1 10.2
148/254 40/254 14/254 13/254 13/254 26/254
P ¼ .018*
NOTE. P values were calculated with 2-tailed c2 analysis or the Fisher exact test. ARCR, arthroscopic rotator cuff repair; LOA, lysis of adhesions; OA, osteoarthritis; RCR, rotator cuff repair; RTC, rotator cuff; SAD, subacromial decompression. *Statistically significant.
Discussion Our study found a 6% incidence of repeat outpatient surgical intervention after isolated ARCR (with or without acromioplasty) in New York State. Across all age groups, we found that the most common reason (58.7%) for subsequent surgery was revision RCR, thus resulting in an overall rotator cuff revision rate of 3.4% within our population (1,071 of 30,430). Because of the paucity of literature on rates of revision surgery after ARCR, it is difficult to comment on how our data compare with those of other authors. For example, of the recent case series examining tendon integrity after RCR,8,23-26 only 1 study reported on the rate of subsequent surgery (22% [14 of 64] 2 years after primary
RCR).23 All patients in the aforementioned study had retears, and the factors predicting reoperation in this sample were not specifically addressed.23 In addition, our finding that 9.1% of patients who underwent subsequent surgery (166 of 1,826) ultimately underwent 2 or more procedures is comparable to previously published values of 8% to 13%.2,4 Sherman et al.37 previously attempted to use the SPARCS database to identify readmission and revision rates after RCR. They found a 4% rate of revision after RCR at 1 year that was associated with increasing age, medical comorbidities, and decreasing surgeon volume. Their study differs from ours in several significant respects. First, we followed up patients in the database for
SUBSEQUENT SURGERY AFTER ARCR Table 4. Mean Time Until Subsequent Surgery Based on Demographic Characteristics Time Until Subsequent Ipsilateral Shoulder Procedure (n ¼ 1,826) Mean, mo Age <45 yr 28.9 45-65 yr 23.7 >65 yr 21.5 Sex Male 25.9 Female 22.3 Insurance Workers’ 22.5 Compensation All others 25.0 Acromioplasty at time of initial RCR Yes 24.1 No 24.9 Tobacco history Yes 16.9 No 24.7
SD, mo
Significance
30.0 26.5 25.7
P < .001*
28.2 25.5
P ¼ .006*
23.3
P ¼ .081
28.4 27.6 26.1
P ¼ .604
15.4 27.6
P < .001*
RCR, rotator cuff repair; SD, standard deviation. *Statistically significant.
2 years or more. Second, whereas Sherman et al. examined for only revision RCRs, we examined for all subsequent forms of outpatient shoulder procedures. Finally, Sherman et al. identified both open RCR and ARCR from the inpatient and outpatient SPARCS databases between 1997 and 2002. We limited our initial cohort to ARCR identified in the outpatient database for several reasons: There has been a substantial shift toward arthroscopic and outpatient RCR,9,10,12,13 and we wanted our study to reflect current practice trends. In addition, the outpatient and inpatient databases use 2 different procedural coding systems: CPT and ICD9-CM, respectively. The ICD-9-CM system used in the inpatient database does not have a laterality indicator, and thus it is impossible to determine if any subsequent procedures were even on the same side as the original RCR. Sherman et al. listed the inability to determine laterality as a limitation in their results. Given that numerous studies have reported rates of contralateral rotator cuff tears ranging from 38.6% to 67.3%, we did not use the inpatient SPARCS database, thus avoiding the ambiguity regarding laterality.38,39 For this reason, we believe our data best reflect current practice patterns while also providing the most accurate assessment of subsequent shoulder surgery to date. Reasons for failure after RCR are likely multifactorial. Advancing age is believed to be an important risk factor for failed RCR, with poorer tissue quality and larger tears noted in these patients.29,30 However, small case series of patients older than 65 years have found postoperative results comparable to younger patients at
7
early- to mid-term follow-up.40-42 In contrast to existing literature,29,30,37 we found younger age to be an independent risk factor for subsequent outpatient shoulder surgery. In New York State, 8.0% of patients younger than 45 years treated with RCR underwent additional outpatient shoulder procedures compared with only 3.7% of those older than 65 years (P < .001). It is possible that the greater physical demands of younger patients motivated subsequent outpatient surgery or that there existed concomitant pathology not addressed during the initial RCR procedure. For example, we reported that SLAP repairs and arthroscopic capsulorrhaphy comprised 7.4% and 4.2% of subsequent surgical procedures, respectively, in patients younger than 45 years.43 These associations are in accord with the orthopaedic literature because many authors have reported on the association between glenohumeral instability episodes or labral pathology and concomitant rotator cuff pathology, particularly in younger patients.43-47 Although patients older than 65 years had the lowest overall subsequent surgery rate (3.7%) among all age groups, it is important to keep in mind that 65.4% of subsequent procedures performed on those older than 65 years were either open or arthroscopic revision cuff repair. Thus, compared with the 2 younger-age subgroups, these patients had the highest rate of failure of initial ARCR, which corroborates the existing literature. In addition, we believe that the low rate of subsequent outpatient shoulder surgery in this demographic of older patients may be due to age-related trends in management. It is plausible that older patients with a failed ARCR are managed with a shoulder arthroplasty instead of a revision outpatient shoulder procedure. In 2011 reverse total shoulder arthroplasty (RTSA) accounted for one-third of all shoulder arthroplasties performed in the United States, and 81.4% of patients receiving an RTSA were aged 65 years or older.48 Moreover, 52.5% of patients receiving an RTSA had a diagnosis of rotator cuff tear/syndrome.49 Thus it is possible that older patients with irreparable rotator cuff tears are being managed by arthroplasty and thus are not captured in our study. In addition to age, several other factors have been suggested to predict failure after RCR. Workers’ Compensation has been shown to predict poor outcomes of RCR, as well as other arthroscopic procedures.50,51 Our results support this because Workers’ Compensation was identified as an independent risk factor for subsequent surgery in our series (OR, 2.11; 95% CI, 1.89-2.36; P < .001). Although no differences were observed between the sexes regarding reoperation rates, subgroup analysis did show that female patients underwent subsequent surgery earlier than male patients (22.3 months v 25.9 months, P ¼ .006). Previous authors have reported
8
S. A. MAHURE ET AL.
that female sex is associated with higher pain scores, slower improvements in range of motion, and increased dissatisfaction after ARCR.52-54 Because the SPARCS database does not provide any of these clinical or subjective outcome measures, we cannot comment on whether the same causal dynamic exists in our cohort and we look forward to future studies that elaborate on the association between sex and reoperation. With these limitations of our data being kept in mind, it is possible that our findings that age older than 65 years was associated with earlier time until reoperation (Table 4) could be reflective of literature suggesting that advanced age is associated with greater retear rates and poor tissue quality.55-57 Although our multivariate analysis did not show tobacco use as an independent risk factor for subsequent ipsilateral surgery (OR, 1.18; 95% CI, 0.95-1.47; P ¼ .133), our results regarding time until subsequent surgery and tobacco use add to the existing literature. In our sample, patients with a history of tobacco use underwent subsequent surgery approximately 8 months earlier than those without a history of tobacco use (16.9 months v 24.7 months, P < .001). Several authors have reported on the impact of smoking and tobacco use as negative prognostic indicators in rotator cuff pathology and after RCR.29,58,59 Nicotine has been reported to be a potent vasoconstrictor, and the resultant hypoxia can result in remarkable inflammation and tissue destruction, particularly in hypovascular areas such as the soft tissue of the shoulder.60,61 Bishop et al.58 reported in their systematic review that smoking and tobacco use are associated with larger rotator cuff tears, poor patient outcome scores, and accelerated rotator cuff degeneration. As health care continues to be more intimately linked to quality and outcomes, it is important for orthopaedic surgeons and an interdisciplinary team to routinely screen patients for tobacco use and offer counseling and treatment modalities to encourage smoking cessation. We also identified lack of acromioplasty at the time of initial ARCR as an independent risk factor for subsequent ipsilateral shoulder surgery (OR, 1.20; 95% CI, 1.09-1.34; P < .001) and subsequent cuff-related procedures (OR, 1.29; 95% CI, 1.13-1.47; P < .001). Opinions regarding the utility of acromioplasty with RCR vary: Some authors have suggested that acromioplasty can damage the integrity of the coracoacromial arch with resultant pathology in the glenohumeral joint,62,63 whereas others have reported positive outcomes coupled with an increasing incidence of acromioplasty procedures.64-68 Our analysis showed that from 2003 through 2011, there was a consistent rise in the proportion of RCRs performed with acromioplasty, with the trend reversing between 2012 and 2014 (Fig 1). Future studies comparing the revision rates between these 2 periods are warranted to
provide further evidence regarding the utility of acromioplasty in RCR procedures. Other factors that have been suggested to influence failure after RCR that we were not able to explore in this study include genetics,69 method of repair (single v double row, simple v mattress sutures),29,30 results of magnetic resonance imaging (e.g., fatty atrophy, degenerative tendon),70 or postsurgical rehabilitation.71 Similarly, despite our results regarding acromioplasty, we were unable to account for acromial morphology, which has been implicated in the success of acromioplasty and rotator cuff pathology treatment.72-75 Despite some of the limitations of our study, our large sample size and ability to longitudinally follow up patients provide meaningful data to physicians regarding risk factors for reoperation after ARCR. Limitations There are inherent limitations when performing studies using large statewide databases.76 Studies like this one rely on complete and accurate reporting of data. We had no way of confirming if additional procedures were performed to address sequelae of the initial RCR repair. It remains conceivable that surgery was performed for an unrelated or new problem, thus overestimating the need for subsequent surgery. Conversely, we could not identify patients who sought out-of-state care for subsequent procedures. As such, we may be underestimating the incidence of subsequent surgery. However, our analysis identified New York State as the home state for 92.5% of our sample, potentially limiting the number of patients who sought care in other states. In addition, although we eliminated patients with prior ipsilateral shoulder procedures, the SPARCS database only captures procedures within New York State. Therefore, patients who may have previously undergone an ARCR performed outside of New York State could potentially have been included in our sample. We believe that the probability of this making a significant difference in results, however, is minimal given our large sample size and the fact that 92.5% of our sample listed New York as the home state. Another limitation is that we could not account for surgeon experience, technical skill, or advanced fellowship training. Finally, given the limitations of ICD-9 coding, we were unable to comment on patients who underwent subsequent inpatient rotator cuff repair or shoulder arthroplasty.
Conclusions We identified a 6.0% incidence of repeat ipsilateral surgery after isolated ARCR. Although reasons for reoperation are likely multifactorial, younger age, Workers’ Compensation claim, and absence of acromioplasty at the time of initial ARCR remained independent predictors of subsequent outpatient
SUBSEQUENT SURGERY AFTER ARCR
procedures, whereas a history of tobacco use was associated with accelerated time to subsequent surgery. 16.
References 1. Denard PJ, Jiwani AZ, Ladermann A, Burkhart SS. Longterm outcome of a consecutive series of subscapularis tendon tears repaired arthroscopically. Arthroscopy 2012;28:1587-1591. 2. Ladermann A, Denard PJ, Burkhart SS. Midterm outcome of arthroscopic revision repair of massive and nonmassive rotator cuff tears. Arthroscopy 2011;27:1620-1627. 3. Paxton ES, Teefey SA, Dahiya N, Keener JD, Yamaguchi K, Galatz LM. Clinical and radiographic outcomes of failed repairs of large or massive rotator cuff tears: Minimum ten-year follow-up. J Bone Joint Surg Am 2013;95:627-632. 4. Chuang MJ, Jancosko J, Nottage WM. Clinical outcomes of single-row arthroscopic revision rotator cuff repair. Orthopedics 2014;37:e692-e698. 5. Lin EC, Mall NA, Dhawan A, et al. Arthroscopic primary rotator cuff repairs in patients aged younger than 45 years. Arthroscopy 2013;29:811-817. 6. Bhatia S, Greenspoon JA, Horan MP, Warth RJ, Millett PJ. Two-year outcomes after arthroscopic rotator cuff repair in recreational athletes older than 70 years. Am J Sports Med 2015;43:1737-1742. 7. Moraiti C, Valle P, Maqdes A, et al. Comparison of functional gains after arthroscopic rotator cuff repair in patients over 70 years of age versus patients under 50 years of age: A prospective multicenter study. Arthroscopy 2015;31:184-190. 8. Pauly S, Stahnke K, Klatte-Schulz F, Wildemann B, Scheibel M, Greiner S. Do patient age and sex influence tendon cell biology and clinical/radiographic outcomes after rotator cuff repair? Am J Sports Med 2015;43: 549-556. 9. Colvin AC, Egorova N, Harrison AK, Moskowitz A, Flatow EL. National trends in rotator cuff repair. J Bone Joint Surg Am 2012;94:227-233. 10. Ensor KL, Kwon YW, Dibeneditto MR, Zuckerman JD, Rokito AS. The rising incidence of rotator cuff repairs. J Shoulder Elbow Surg 2013;22:1628-1632. 11. Zhang AL, Montgomery SR, Ngo SS, Hame SL, Wang JC, Gamradt SC. Analysis of rotator cuff repair trends in a large private insurance population. Arthroscopy 2013;29: 623-629. 12. Jain NB, Higgins LD, Losina E, Collins J, Blazar PE, Katz JN. Epidemiology of musculoskeletal upper extremity ambulatory surgery in the United States. BMC Musculoskelet Disord 2014;15:4. 13. Cullen KA, Hall MJ, Golosinskiy A. Ambulatory surgery in the United States, 2006. Natl Health Stat Report 2009:1-25. 14. Brown MJ, Pula DA, Kluczynski MA, Mashtare T, Bisson LJ. Does suture technique affect re-rupture in arthroscopic rotator cuff repair? A meta-analysis. Arthroscopy 2015;31:1576-1582. 15. Akpinar S, Uysal M, Pourbagher MA, Ozalay M, Cesur N, Hersekli MA. Prospective evaluation of the functional and anatomical results of arthroscopic repair in small and
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
9
medium-sized full-thickness tears of the supraspinatus tendon. Acta Orthop Traumatol Turc 2011;45:248-253. Kim KC, Shin HD, Cha SM, Kim JH. Repair integrity and functional outcomes for arthroscopic margin convergence of rotator cuff tears. J Bone Joint Surg Am 2013;95: 536-541. Deutsch A, Kroll DG, Hasapes J, Staewen RS, Pham C, Tait C. Repair integrity and clinical outcome after arthroscopic rotator cuff repair using single-row anchor fixation: A prospective study of single-tendon and twotendon tears. J Shoulder Elbow Surg 2008;17:845-852. Lubiatowski P, Kaczmarek P, Dzianach M, et al. Clinical and biomechanical performance of patients with failed rotator cuff repair. Int Orthop 2013;37:2395-2401. Franceschi F, Papalia R, Vasta S, Leonardi F, Maffulli N, Denaro V. Surgical management of irreparable rotator cuff tears. Knee Surg Sports Traumatol Arthrosc 2015;23: 494-501. Liem D, Lengers N, Dedy N, Poetzl W, Steinbeck J, Marquardt B. Arthroscopic debridement of massive irreparable rotator cuff tears. Arthroscopy 2008;24: 743-748. Verhelst L, Vandekerckhove PJ, Sergeant G, Liekens K, Van Hoonacker P, Berghs B. Reversed arthroscopic subacromial decompression for symptomatic irreparable rotator cuff tears: Mid-term follow-up results in 34 shoulders. J Shoulder Elbow Surg 2010;19:601-608. 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;96:265-271. Shamsudin A, Lam PH, Peters K, Rubenis I, Hackett L, Murrell GA. Revision versus primary arthroscopic rotator cuff repair: A 2-year analysis of outcomes in 360 patients. Am J Sports Med 2015;43:557-564. Cho NS, Rhee YG. The factors affecting the clinical outcome and integrity of arthroscopically repaired rotator cuff tears of the shoulder. Clin Orthop Surg 2009;1:96-104. Cole BJ, McCarty LP III, Kang RW, Alford W, Lewis PB, Hayden JK. Arthroscopic rotator cuff repair: Prospective functional outcome and repair integrity at minimum 2year follow-up. J Shoulder Elbow Surg 2007;16:579-585. Boileau P, Brassart N, Watkinson DJ, Carles M, Hatzidakis AM, Krishnan SG. Arthroscopic repair of fullthickness tears of the supraspinatus: Does the tendon really heal? J Bone Joint Surg Am 2005;87:1229-1240. Sperling JW, Cofield RH, Schleck C. Rotator cuff repair in patients fifty years of age and younger. J Bone Joint Surg Am 2004;86:2212-2215. Cuff DJ, Pupello DR. Prospective evaluation of postoperative compliance and outcomes after rotator cuff repair in patients with and without workers’ compensation claims. J Shoulder Elbow Surg 2012;21:1728-1733. Abtahi AM, Granger EK, Tashjian RZ. Factors affecting healing after arthroscopic rotator cuff repair. World J Orthop 2015;6:211-220. Tashjian RZ, Hollins AM, Kim HM, et al. Factors affecting healing rates after arthroscopic double-row rotator cuff repair. Am J Sports Med 2010;38:2435-2442. Onyekwelu I, Khatib O, Zuckerman JD, Rokito AS, Kwon YW. The rising incidence of arthroscopic superior
10
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
S. A. MAHURE ET AL. labrum anterior and posterior (SLAP) repairs. J Shoulder Elbow Surg 2012;21:728-731. Bansal A, Khatib ON, Zuckerman JD. Revision total joint arthroplasty: The epidemiology of 63,140 cases in New York State. J Arthroplasty 2014;29:23-27. Khatib O, Onyekwelu I, Zuckerman JD. The incidence of proximal humeral fractures in New York State from 1990 through 2010 with an emphasis on operative management in patients aged 65 years or older. J Shoulder Elbow Surg 2014;23:1356-1362. Siddiq F, Adil MM, Malik AA, Qureshi MH, Qureshi AI. Effect of carotid revascularization endarterectomy versus stenting trial results on the performance of carotid artery stent placement and carotid endarterectomy in the United States. Neurosurgery 2015;75:726-732. Movahed MR, Hashemzadeh M, Jamal MM. Increased prevalence of third-degree atrioventricular block in patients with type II diabetes mellitus. Chest 2005;128: 2611-2614. Macleod LC, Dai JC, Holt SK, Bassett JC, Wright JL, Gore JL. Underuse and underreporting of smoking cessation for smokers with a new urologic cancer diagnosis. Urol Oncol 2015;33:504.e1-e7. Sherman SL, Lyman S, Koulouvaris P, Willis A, Marx RG. Risk factors for readmission and revision surgery following rotator cuff repair. Clin Orthop Relat Res 2008;466:608-613. Liem D, Buschmann VE, Schmidt C, et al. The prevalence of rotator cuff tears: Is the contralateral shoulder at risk? Am J Sports Med 2014;42:826-830. Ro KH, Park JH, Lee SH, Song DI, Jeong HJ, Jeong WK. Status of the contralateral rotator cuff in patients undergoing rotator cuff repair. Am J Sports Med 2015;43: 1091-1098. Verma NN, Bhatia S, Baker CL III, et al. Outcomes of arthroscopic rotator cuff repair in patients aged 70 years or older. Arthroscopy 2010;26:1273-1280. Charousset C, Bellaiche L, Kalra K, Petrover D. Arthroscopic repair of full-thickness rotator cuff tears: Is there tendon healing in patients aged 65 years or older? Arthroscopy 2010;26:302-309. Osti L, Papalia R, Del Buono A, Denaro V, Maffulli N. Comparison of arthroscopic rotator cuff repair in healthy patients over and under 65 years of age. Knee Surg Sports Traumatol Arthrosc 2010;18:1700-1706. Gombera MM, Sekiya JK. Rotator cuff tear and glenohumeral instability: A systematic review. Clin Orthop Relat Res 2014;472:2448-2456. Araghi A, Prasarn M, St Clair S, Zuckerman JD. Recurrent anterior glenohumeral instability with onset after forty years of age: The role of the anterior mechanism. Bull Hosp Jt Dis 2005;62:99-101. Craig EV. The posterior mechanism of acute anterior shoulder dislocations. Clin Orthop Relat Res 1984: 212-216. Erickson J, Lavery K, Monica J, Gatt C, Dhawan A. Surgical treatment of symptomatic superior labrum anterior-posterior tears in patients older than 40 years: A systematic review. Am J Sports Med 2015;43:1274-1282. Goldberg JA, Chan KY, Best JP, Bruce WJ, Walsh W, Parry W. Surgical management of large rotator cuff tears
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
combined with instability in elite rugby football players. Br J Sports Med 2003;37:179-181; discussion 181. Schairer WW, Nwachukwu BU, Lyman S, Craig EV, Gulotta LV. National utilization of reverse total shoulder arthroplasty in the United States. J Shoulder Elbow Surg 2015;24:91-97. Jain NB, Yamaguchi K. The contribution of reverse shoulder arthroplasty to utilization of primary shoulder arthroplasty. J Shoulder Elbow Surg 2014;23:1905-1912. Henn RF III, Tashjian RZ, Kang L, Green A. Patients with workers’ compensation claims have worse outcomes after rotator cuff repair. J Bone Joint Surg Am 2008;90: 2105-2113. Razmjou H, Lincoln S, Axelrod T, Holtby R. Factors contributing to failure of rotator cuff surgery in persons with work-related injuries. Physiother Can 2008;60: 125-133. Cho CH, Ye HU, Jung JW, Lee YK. Gender affects early postoperative outcomes of rotator cuff repair. Clin Orthop Surg 2015;7:234-240. Razmjou H, Davis AM, Jaglal SB, Holtby R, Richards RR. Disability and satisfaction after rotator cuff decompression or repair: A sex and gender analysis. BMC Musculoskelet Disord 2011;12:66. Robinson PM, Wilson J, Dalal S, Parker RA, Norburn P, Roy BR. Rotator cuff repair in patients over 70 years of age: Early outcomes and risk factors associated with retear. Bone Joint J 2013;95-B:199-205. 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;43:491-500. Gill TJ, McIrvin E, Kocher MS, Homa K, Mair SD, Hawkins RJ. The relative importance of acromial morphology and age with respect to rotator cuff pathology. J Shoulder Elbow Surg 2002;11:327-330. Le BT, Wu XL, Lam PH, Murrell GA. Factors predicting rotator cuff retears: An analysis of 1000 consecutive rotator cuff repairs. Am J Sports Med 2014;42:1134-1142. Bishop JY, Santiago-Torres JE, Rimmke N, Flanigan DC. Smoking predisposes to rotator cuff pathology and shoulder dysfunction: A systematic review. Arthroscopy 2015;31:1598-1605. Santiago-Torres J, Flanigan DC, Butler RB, Bishop JY. The effect of smoking on rotator cuff and glenoid labrum surgery: A systematic review. Am J Sports Med 2015;43: 745-751. Tashjian RZ. Epidemiology, natural history, and indications for treatment of rotator cuff tears. Clin Sports Med 2012;31:589-604. Babayan RK. Wound healing and infection in surgery: The pathophysiological impact of smoking, smoking cessation, and nicotine replacement therapy: A systematic review. J Urol 2012;188:2243-2244. MacDonald P, McRae S, Leiter J, Mascarenhas R, Lapner P. Arthroscopic rotator cuff repair with and without acromioplasty in the treatment of full-thickness rotator cuff tears: A multicenter, randomized controlled trial. J Bone Joint Surg Am 2011;93:1953-1960. Budoff JE, Nirschl RP, Guidi EJ. Debridement of partialthickness tears of the rotator cuff without acromioplasty.
SUBSEQUENT SURGERY AFTER ARCR
64.
65.
66.
67.
68.
69.
70.
Long-term follow-up and review of the literature. J Bone Joint Surg Am 1998;80:733-748. Romeo AA, Hang DW, Bach BR Jr, Shott S. Repair of full thickness rotator cuff tears. Gender, age, and other factors affecting outcome. Clin Orthop Relat Res 1999:243-255. Vitale MA, Arons RR, Hurwitz S, Ahmad CS, Levine WN. The rising incidence of acromioplasty. J Bone Joint Surg Am 2010;92:1842-1850. Chin PY, Sperling JW, Cofield RH, Stuart MJ, Crownhart BS. Anterior acromioplasty for the shoulder impingement syndrome: Long-term outcome. J Shoulder Elbow Surg 2007;16:697-700. Frieman BG, Fenlin JM Jr. Anterior acromioplasty: Effect of litigation and workers’ compensation. J Shoulder Elbow Surg 1995;4:175-181. Ellman H. Arthroscopic subacromial decompression: Analysis of one- to three-year results. Arthroscopy 1987;3: 173-181. Gwilym SE, Watkins B, Cooper CD, et al. Genetic influences in the progression of tears of the rotator cuff. J Bone Joint Surg Br 2009;91:915-917. Gladstone JN, Bishop JY, Lo IK, Flatow EL. Fatty infiltration and atrophy of the rotator cuff do not
71.
72.
73.
74. 75.
76.
11
improve after rotator cuff repair and correlate with poor functional outcome. Am J Sports Med 2007;35: 719-728. Gallagher BP, Bishop ME, Tjoumakaris FP, Freedman KB. Early versus delayed rehabilitation following arthroscopic rotator cuff repair: A systematic review. Phys Sportsmed 2015;43:178-187. Balke M, Schmidt C, Dedy N, Banerjee M, Bouillon B, Liem D. Correlation of acromial morphology with impingement syndrome and rotator cuff tears. Acta Orthop 2013;84:178-183. Neer CS II. Anterior acromioplasty for the chronic impingement syndrome in the shoulder: A preliminary report. J Bone Joint Surg Am 1972;54:41-50. Neer CS II. Impingement lesions. Clin Orthop Relat Res 1983:70-77. Zuckerman JD, Kummer FJ, Cuomo F, Simon J, Rosenblum S, Katz N. The influence of coracoacromial arch anatomy on rotator cuff tears. J Shoulder Elbow Surg 1992;1:4-14. Memtsoudis SG. Limitations associated with the analysis of data from administrative databases. Anesthesiology 2009;111:449; author reply 450-451 (letter).