- Email: [email protected]
Functional Outcome of Arthroscopic Extensor Carpi Radialis Brevis Tendon Release in Chronic Lateral Epicondylitis
SCIENTIFIC ARTICLE
Functional Outcome of Arthroscopic Extensor Carpi Radialis Brevis Tendon Release in Chronic Lateral Epicondylitis Ruby Grewal, MD, Joy C. MacDermid, PhD, Parag Shah, MD, Graham J.W. King, MD
Purpose To evaluate outcomes of arthroscopic tennis elbow release in a population of patients with chronic, recalcitrant symptoms, a large number of workers’ compensation claims, and high occupational demands using standardized outcome measures, including a detailed objective assessment of workplace demands. Methods We treated 36 patients with chronic lateral epicondylitis with an arthroscopic release. A standardized protocol was used to measure strength, motion, and outcomes (American Shoulder and Elbow Surgeons Elbow [ASES-e] score, Short Form-12, PatientRated Tennis Elbow evaluation [PRTEE], and Work Limitations Questionnaire-26). Results The mean duration of symptoms before surgery was 30 months. A total of 25 of 36 patients were employed in heavy or repetitive occupations and 23 of 36 were involved in a workers’ compensation claim. The final overall results were favorable, with 30 of 36 subjects reporting improvement with surgery. The final mean Mayo Elbow Performance Index score was 78.6 ⫾ 16.5 (22 ⫽ good to excellent, 9 ⫽ fair, and 5 ⫽ poor). The average total PRTEE was 26.2 ⫾ 24.3 out of 100. The average ASES-e pain score was 16.1 ⫾ 15.0 and the average ASES-e function score was 27.9 ⫾ 8.8. Patients in heavy or repetitive occupations and those with workers’ compensation claims had significantly worse outcome scores (Mayo Elbow Performance Index, ASES, and PRTEE). Based on Work Limitations Questionnaire-26 scores, patients with workers’ compensation claims had significantly greater difficulties with physical (36.8 vs 3.2, p ⬍ .001), output (40.8 vs 3.1, p ⫽ .002), mental (36.0 vs 9.0, p ⫽ .05), and social (27.7 vs 6.3, p ⫽ .05) workplace demands. Conclusions Arthroscopic tennis elbow release provides symptomatic improvement in most patients with lateral epicondylitis. Patient selection and reported occupational demands have an important role in determining outcomes. More work is required to identify factors predicting outcome in this difficult subgroup. (J Hand Surg 2009;34A:849–857. Copyright © 2009 by the American Society for Surgery of the Hand. All rights reserved.) Type of study/level of evidence Therapeutic IV. Key words Arthroscopy, lateral epicondylitis, tennis elbow, chronic pain, workers’ compensation.
FromtheDivisionofOrthopedicSurgery,UniversityofWesternOntario,HandandUpperLimbCenter,St Joseph’sHealthCare,London,Ontario,Canada;andtheFractureandOrthopaedicHospital,Pali,Ahmedabad, India.
Corresponding author: Ruby Grewal, MD, Division of Orthopedic Surgery, University of Western Ontario,HandandUpperLimbCenter,StJoseph’sHealthCare,268GrosvenorStreet,LondonON,N6A 4L6, Canada; e-mail: [email protected].
Received for publication October 15, 2008; accepted in revised form February 5, 2009.
0363-5023/09/34A05-0007$36.00/0 doi:10.1016/j.jhsa.2009.02.006
No benefits in any form have been received or will be received related directly or indirectly to the subject of this article.
© ASSH 䉬 Published by Elsevier, Inc. All rights reserved. 䉬 849
850
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
tennis elbow, is a common cause of elbow pain and dysfunction. Although this disorder can occur in tennis players and other athletes, the most common clinical scenario involves a worker who performs heavy or repetitive tasks.1 Cyriax wrote that tennis elbow is a self-resolving condition whose symptoms usually last for 8 to 12 months; however, he also stated that there is a “chronic occupational type” of this disorder that is less likely to resolve.2 A recent literature review of the surgical treatment of lateral epicondylitis found that most patients have a fairly similar profile.3 They are in their fifth decade of life (range, 30 – 63 years), their dominant arm is most commonly affected (74%), they have an average duration of symptoms of 19 months, and they have failed nonsurgical treatment.3 Published case series4 –10 generally report favorable clinical outcomes with open, percutaneous, or arthroscopic surgery but there are few reports that use standardized, validated measures to report outcomes.3 The purpose of this study was to evaluate outcomes of arthroscopic tennis elbow release in patients with chronic recalcitrant symptoms, many with workers’ compensation claims, and high occupational demands, using standardized outcome measures, including a detailed objective assessment of workplace demands.
L
ATERAL EPICONDYLITIS, OR
MATERIALS AND METHODS Arthroscopic debridement and release of the extensor carpi radialis brevis (ECRB) tendon has been the preferred form of surgical intervention for lateral epicondylitis by the senior author since 1999. Surgical indications included failed conservative treatment (physiotherapy, occupational therapy, tennis elbow bracing, and cortisone injections) and persistence of symptoms for a minimum of 1 year. The senior author performed arthroscopic ECRB releases in 48 patients between 1999 and 2006. Patients with revision surgery were excluded. A total of 36 patients were able to return for reassessment. Twelve patients could not be reached for follow-up: 1 had a recent stroke, 2 could not be located, 7 failed to return our phone calls or were unable to attend their follow-up appointment, and 2 refused to return for follow-up because of their geographical distance from our clinic. An examiner who was not involved with their care evaluated all 36 patients. Patients completed the Patient-Rated Tennis Elbow evaluation (PRTEE), which has a visual analog scale for pain and function,11 the American Shoulder and Elbow Surgeons Elbow (ASES-e) patient self-evaluation form for pain and
function,12 and the Short Form-12 (SF-12), version 2, health survey for general health and well-being.13 Patients who were working (35 of 36) also completed a 26-item version of the Work Limitations Questionnaire (WL-26).14,15 The WL-26 score consists of 5 domains: scheduling, mental, physical, social, and output demands (ability to work quickly enough and finish work on time). Each domain is scored separately (best score ⫽ 0, worst score ⫽ 100); a score of 50 indicates the average patient is having difficulty meeting the demands of that domain for half of each workday (score 33 ⫽ difficulty for one third of the workday). Patients were also asked to assess their results as being completely cured, much better, better, the same, or worse. Objective assessment included a detailed clinical examination and a standardized protocol for measuring grip16 and motion.17 Strength testing was conducted by the LIDO WorkSET isokinetic dynamometer (Loredan Biomedical, West Sacramento, CA).18 Isometric grip strength was measured as maximum and pain-free grip strength with elbow extended. We also measured the isometric wrist extension and flexion strength. All measurements were recorded thrice after an interval of 30 seconds between each measurement. Statistical softare (SPSS v15.0; Chicago, IL) was used to calculate descriptive statistics and correlations and test for differences in subgroups using t-tests and side-to-side differences (paired t-tests). Surgical technique Under general anesthesia, patients were positioned in the lateral decubitus position. The elbow was injected with 20 mL saline to distend the joint. Our standard visualization portal for tennis elbow release was the proximal anteromedial portal. The anterior elbow joint was inspected, and the capsule overlying the common extensor origin was visualized. Any capsular or tendon abnormality was noted. Patients were classified as having type 1, 2, or 3 lesions.19 Type 1 lesions have inflammation and fraying deep to the extensor carpi radialis brevis (ECRB) tendon without evidence of a frank tear. Type 2 lesions are linear tears at the undersurface of the ECRB tendon, and type 3 lesions are retracted with partial or completed avulsions of the tendon. The proximal anterolateral working portal was made using an outside-in method. The capsule and the ECRB tendon were resected using a motorized shaver. Care was taken to avoid injury to the lateral ulnar collateral ligament by limiting the resection to the portion anterior to the mid-axis of the radial head. We did not decorticate the lateral epicondyle. We used only 2 arthroscopic
JHS 䉬 Vol A, May–June
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
portals and did not perform a posterior elbow arthroscopy in this cohort. Portals were sutured with 3– 0 nylon sutures and a soft bulky dressing was applied. Postoperative care Active motion was encouraged postoperatively. Patients were instructed to remove the dressing 48 hours postoperatively and apply a bandage over each portal site. Sutures were removed at 10 days. Patients did not routinely attend physiotherapy but were encouraged to return gradually to their normal activities as their symptoms allowed. Heavy or repetitive work was not permitted for 6 weeks postoperatively. Demographic data There were 20 males and 16 females in the study group. The average age of our subjects was 45.3 ⫾ 7.0 years (range, 29 – 61 years). Thirty patients had their dominant arm affected. A total of 23 patients were involved in a workers’ compensation claim. Ten patients recalled a specific event leading to pain, whereas 26 of them described the onset of pain as insidious. Thirteen patients described their work as heavy, 12 described their work as repetitive, 10 had clerical duties or other physically light jobs, and 1 was retired. Before developing lateral epicondylitis, 25 patients were involved in recreational sports. The mean duration of symptoms before surgery was 30 ⫾ 20.5 months (range, 1–10 years); 12 patients had symptoms for 2 to 3 years, and 9 patients had symptoms for ⬎3 years. The average number of previous cortisone injections was 2.5 ⫾ 1.9 (range, 0 –10 injections), with 17 patients receiving 3 or more injections and 4 patients receiving no injections. Although all patients ultimately failed nonsurgical measures, 23 reported some transient symptomatic relief with the cortisone injections. There were 12 patients with type 1 lesions, 4 with type 2 lesions, 2 with type 3 lesions, and 18 with no lesion of the ECRB tendon identified intraoperatively.19 Additional intraoperative arthroscopy findings were as follows: 13 patients had synovitis, 4 had osteophytes, and 4 had chondromalacia of the radial head. One patient had a localized synovectomy at the time of arthroscopic release of the ECRB and one had an open carpal tunnel release under the same anesthetic. One patient developed mild erythema around the portals that required oral antibiotics; no other complications were observed. RESULTS Pain and disability The average length of follow-up was 42 months (range, 19 –74 months). Thirty subjects felt that they had im-
851
proved with surgery; 14 patients described themselves as completely asymptomatic, 16 were better or much better, 5 remained the same, and 1 felt worse. On a visual analog score, the average satisfaction was 8 out of 10. A final PRTEE was calculated, with 100 representing the worst possible score (maximal pain and functional disability). The average PRTEE for pain was 14.6 ⫾ 12.9 out of 50 (range, 0 –37). The average PRTEE for functional disability was 11.3 ⫾ 11.9 out of 50 (range, 0 –37). The average total PRTEE was 26.2 ⫾ 24.3 out of 100. The average ASES-e pain score was 16.1 ⫾ 15.0 (range, 0 –50, where 0 ⫽ no pain); the average ASES-e function score was 27.9 ⫾ 8.8 (range, 0 –36, where 36 ⫽ full function). The mean Mayo Elbow Performance Index (MEPI) score20 was 78.6 ⫾ 16.5; 22 patients had a good to excellent result, 9 had a fair result, and 5 had a poor result. The mean SF-12v2 physical component summary score was 44.6 and the mental component summary score was 49.8 (US population norms ⫽ 50). At final follow-up, there was a mean improvement of 27.3% in grip strength (affected grip strength was expressed as a percentage of the unaffected grip strength). This improvement was statistically significant (p ⫽ .001). There were slight side-to-side differences (affected/unaffected arm) demonstrated in elbow flexion (138.0°/141.4°, p ⫽ .014), elbow extension (1.9°/0.9°, p ⫽ .054), and grip strength (36.0/40.2 kg, p ⫽ .009), although the clinical importance of these differences is questionable. No side-to-side differences were noted in wrist extension (9.2/10.0 Nm, p ⫽ .17) or wrist flexion strength (10.7/11.4 Nm, p ⫽ .19). We found no associations among patient age, duration of symptoms, presence of additional intraoperative findings (ie, cartilage damage, synovitis), type of lesion, presence or absence of ECRB lesion at time of arthroscopy, presence of bilateral symptoms, and final outcomes. Patients who reported temporary relief with cortisone injections had less pain and disability at final follow-up compared with patients who received no benefit from these injections (mean PRTEE, 18.5 vs 39.2, p ⫽ .04). Before developing tennis elbow, 25 patients were involved in recreational sports, 16 of whom were able to return to sports (11 unrestricted, 5 restricted). Effect of work history on outcomes Of 36 patients in this cohort, 23 were involved in workers’ compensations claims. Fifteen of the 23 patients were able to return to work after surgery. Two patients involved in heavy labor and 2 involved in
JHS 䉬 Vol A, May–June
852
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
repetitive tasks made a full return to work. The remaining 11 patients returned to work but required restrictions on their physical demands (2 of these patients were heavy laborers, 6 were involved in repetitive tasks, 3 were involved in clerical or other light duties at work). Seven patients with workers’ compensations claims were unable to return to work, and 1 patient retired. Four patients changed their jobs as a result of their elbow pain: 3 of these patients were previously involved in repetitive work and 1 was involved in heavy work. The remaining 3 patients did not return to work owing to persistent arm pain: 1 had work that involved repetitive tasks and 2 were employed in heavy labor. Of the 13 patients who had no compensation claims, 1 was retired, another retired from heavy labor as a result of his elbow pain, and the remaining 11 were all able to return to work. Five patients returned to jobs involving heavy labor, 5 returned to a job involving clerical work, and 1 returned to his job as a janitor but required activity restrictions. None of the patients were required to change their occupations. The time taken off work after surgery averaged 18.5 ⫾ 27.5 weeks for the whole group (range, 0 –104 weeks). The compensation patients took an average of 24.5 ⫾ 32.6 weeks off work (range, 1–104 weeks), whereas the non-compensation patients took an average of 10.3 ⫾ 16.6 weeks off work (range, 0 –52 weeks) (p ⫽ .2). Patients performing heavy or repetitive work (n ⫽ 18) took an average of 23.9 ⫾ 31.4 weeks off work, and those performing light work (n ⫽ 8) took an average of 6.4 ⫾ 8.0 weeks before returning to work (p ⫽ .04). Patients with workers’ compensation claims had significantly worse outcome scores (MEPI, ASES-e pain, function, satisfaction, and PRTEE scores) than those without claims (Table 1). Patients involved in heavy or repetitive occupations also had significantly worse ASES-e pain, function, satisfaction, and PRTEE scores than patients performing clerical jobs, but no significant difference occurred in the MEPI scores (Table 2). The Workplace Limitations Questionnaire14,15 identified difficulties in all 5 work domains. The mean score for difficulty handling scheduling demands was 18.0 ⫾ 25.9, social demands was 19.9 ⫾ 28.8, physical demands was 22.9 ⫾ 22.9, mental demands was 26.1 ⫾ 33.2, and output demands was 27.1 ⫾ 33.8. As expected, physical demands correlated significantly with all outcome measures (p ⬍ .05); however, we found that mental, social, and scheduling demands also correlated significantly (p ⬍ .05) with self-report outcome measures (Table 3). Patients with workers’ compensation claims had greater difficulties in all occupational
TABLE 1. Relationship Between Outcomes and History of Workers’ Compensation Claim Workers’ Compensation Case Yes
No
p Value
71.8
90.0
.001
Pain
21.0
6.8
Function
24.1
34.3
⬍.001
7.0
9.8
⬍.001
36.7
7.6
⬍.001
Mental
47.7
54.3
.16
Physical
41.4
51.3
.04
MEPI ASES
Satisfaction PRTEE
.003
SF-12
TABLE 2. Relationship Between Outcomes and Type of Work Heavy or Repetitive Work Yes
No
p Value
75.1
85.7
.08
Pain
18.6
8.1
.05
Function
25.9
33.0
.02
7.7
9.9
.001
31.2
13.1
.02
Mental
48.6
55.1
.23
Physical
42.2
54.0
.02
MEPI ASES
Satisfaction PREE SF-12
domains (not only physical demands) compared with those who were not involved with compensation claims. These differences were statistically significant in the domains of mental demands (36.0 vs 9.0, p ⫽ .05), output demands (40.8 vs 3.1, p ⫽ .002), physical demands (36.8 vs 3.2, p ⬍ .001), and social demands (27.7 vs 6.3, p ⫽ .05). Although the results were not statistically significant, workers’ compensation patients also showed greater difficulties with scheduling demands (25.1 vs 5.5, p ⫽ .09) (Fig. 1). Patient selfassessment statements also correlated with high workplace limitation scores. Patients who felt they were worse, unchanged, or only slightly better after surgery had remarkably greater workplace limitations than
JHS 䉬 Vol A, May–June
853
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
TABLE 3.
Pearson Correlation Coefficients for Occupational Demands and Outcome Scores
Outcome Scores
Scheduling Demands
Mental Demands
Output Demands
Social Demands
Physical Demands
–0.375
–0.412
–0.324
–0.195
–0.436*
MEPI ASES Function
0.516*
–0.607†
–0.532*
–0.448*
–0.696†
Pain
0.320
0.598†
0.287
0.431*
0.484*
SF-12 Physical
–0.530*
–0.660†
–0.412
–0.599*
–0.706†
Mental
–0.724†
–0.914†
–0.436
–0.760†
–0.741†
0.530*
0.638†
0.401
0.410
0.589†
PRTEE Total Pain
0.440*
0.558†
0.390
0.326
0.560†
Function
0.560†
0.646†
0.405
0.436*
0.612†
*p ⬍ .05. †p ⬍ .001.
45 40 35 30 25 20 15 10 5 0
Scheduling Demands
Mental Demands*
Physical Demands*
Workers Compensation
Social Demands*
Output Demands*
Nonworkers Compensation
FIGURE 1: Workplace Limitation scores in workers’ compensation and noncompensation patients (*p ⬍ .05).
those who felt they were much better or completely cured. The limitations were evident not only in the physical and output demands of their occupation, but also in the mental demands experienced in the workplace (Fig. 2). When examining the relationship between workplace limitation scores and job type, heavy laborers and those with repetitive jobs experienced more limitations than patients involved in light work or clerical jobs; however, only the domains of scheduling, output, and physical demands achieved statistical significance (Fig. 3). We conducted a chart review for the 12 patients who were unable to return for follow-up. At the time of last assessment, 2 patients were completely asymptomatic,
4 were markedly better, 2 were only slightly better, and 2 had no improvement at all. The outcomes of 2 patients who were not seen past their immediate postoperative visit could not be assessed. DISCUSSION There is a small subgroup of patients with lateral epicondylitis whose symptoms do not respond to conservative measures.2 These are the patients who usually present to surgeons. A variety of surgical treatments have been described (open, percutaneous, and arthroscopic), and all report good but not definitive results.7,8,21,22 In recent years, the arthroscopic technique has gained in popularity. There are many benefits to an
JHS 䉬 Vol A, May–June
854
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
70 60 50 40 30 20 10 0
Scheduling Demands
Mental Demands*
Physical Demands*
slightly better, same or worse
Social Demands
Output Demands*
much better or completely cured
FIGURE 2: Workplace Limitation scores and self-assessment (*p ⬍ .05).
40 35 30 25 20 15 10 5 0
Scheduling Demands*
Mental Demands
Physical Demands*
heavy or repetitive
Social Demands
Output Demands*
light or desk
FIGURE 3: Workplace Limitation scores and type of work (*p ⬍ .05).
arthroscopic approach to the treatment of lateral epicondylitis. The arthroscopic technique allows excellent visualization of the joint and an accurate localization of the ECRB tendon.9 There is also a lower morbidity and earlier return to work and activities than seen with open surgery9; however, the arthroscopic method is more technically demanding. This study demonstrates a positive change in health and work status after arthroscopic release in a difficult subgroup of patients; the study also reports a remarkable level of residual disability despite positive overall outcomes. Few reports in the literature describe the results of arthroscopic treatment for lateral epicondylitis.9,10 Owen et al. published a case series outlining the results of 16 patients observed for an average of 24.1 months
after arthroscopic treatment for lateral epicondylitis.9 In that cohort, there were no workers’ compensation cases and only one patient was employed as a manual laborer. The average duration of conservative treatment in that cohort was 31.7 months. In that series, there were no reported complications: 10 of 12 patients reported feeling “much better,” 2 were “better,” and none considered themselves to be the same or worse after surgery. Using a pain analog scale (0 –10, with 10 being the worst pain), the average pain at rest was 0.58, pain with activities of daily living was 1.58, and the average pain with sports and work was 3.25.9 Baker et al. also reported a series of patients treated arthroscopically.10 They observed 40 patients (42 elbows) for 2 years. The average duration of symptoms
JHS 䉬 Vol A, May–June
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
before surgery was 14 months. Ten patients were involved in a workers’ compensation claim and 22 patients performed heavy labor. The results in that cohort were also favorable. Of the 39 elbows in the 37 patients who were available for follow-up, 37 were rated “better” or “much better.” The average pain at rest was 0.9, pain with activities of daily living was 1.4, and average pain with sports and work was 1.9. Unfortunately, only 13 of their patients returned for objective follow-up. In those 13, the average MEPI score was 93.6 of 100, and grip strength averaged 96% of the strength of the unaffected limb.10 Although the literature demonstrates favorable results for the treatment of lateral epicondylitis, a recent Cochrane review23 concluded there was insufficient evidence to support any specific surgical intervention at this time. There are even fewer reports describing the outcomes of surgical treatment in patients with chronic recalcitrant symptoms, in a cohort of primarily workers’ compensation patients. The overall results in this cohort are comparable to those of Owens et al.9 and Baker et al.,10 although our patients had a longer duration of symptoms before surgery (mean, 29.7 months) and there was a larger number of patients involved in workers’ compensation claims (23) and in physically demanding occupations (25). Despite the number of potential predictors of poorer outcomes, 30 subjects felt that they had improved with surgery and patient satisfaction was high (8 of 10). Despite these good subjective results, only 22 had good or excellent results based on the MEPI score. Most patients had some residual symptoms that affected their measured and self-reported function and ability to participate in work. Although the overall results in this cohort were good, the outcome among patients involved in workers’ compensation claims or in heavy or repetitive labor was not as favorable. Patients with less demanding occupations, without workers’ compensation claims, were most likely to show favorable results, as were those who responded favorably to a previous steroid injection. Patients with workers’ compensation claims had worse outcomes based on all self-report outcome scores (ASES-e pain, function, satisfaction, and PRTEE) (Tables 1 and 2). Occupational demands also play a considerable role in determining outcomes in repetitive strain injuries such as lateral epicondylitis; however, no published reports exist in the literature that assess workplace demands in an objective, standardized fashion. Because of the high proportion of work-related injuries in our cohort, we included outcome measures that addressed
855
work performance in more detail than has been addressed in previous studies. The WL-26 score reflects the amount of time an individual is experiencing difficulties while at work. Work limitations reported using a slightly different version of the WL questionnaire in a previous study were 20 to 28 for patients with osteoarthritis (type of osteoarthritis in cohort: hand, 30.3%; back, 13.9% knee, 22.3%; neck, 13.9%; hip, 13.9%; other, 5.5%) and 4 to 10 for controls.15 The average scores in the current study were 18.0 ⫾ 25.9 for scheduling demands, 19.9 ⫾ 28.8 for social demands, 22.9 ⫾ 22.9 for physical demands, 26.1 ⫾ 33.2 for mental demands, and 27.1 ⫾ 33.8 for output demands, which suggests that patients who successfully return to work after arthroscopic release continue to experience a moderate amount of difficulty similar to that experienced by patients with osteoarthritis. We found that patients employed in heavy labor or repetitive jobs had significantly greater difficulty in the domains of physical, output, and scheduling demands, as one might expect. Our data also demonstrated that patients with workers’ compensation claims had greater difficulties in all occupational domains compared with those who were not involved with compensation claims. The differences were statistically significant not only in the domains of physical (p ⬍ .001) and output (p ⫽ .002) demands, but also in the domains of mental demands (p ⫽ .05) and social demands (p ⫽ .05), and showed a trend toward statistical significance in scheduling demands (p ⫽ .09). We found that patients’ selfreported global ratings of function also correlated with self-reported workplace limitation scores in all domains, but most significantly in the domains of physical, mental, output, and social demands. Whereas the physical aspects of job demands are typically of primary concern for surgeons, mental and social demands are often overlooked. The results of this cohort show that all domains of occupational demands are closely linked to outcomes and patient self-assessment and should be more closely explored in future studies, particularly in workers’ compensation cases. A recent series published by Kay in 2003 reviewed the results of a series of 108 consecutive patients with tennis elbow who were also seeking compensation.24 In that series, 42 of 108 patients received surgery. Of those patients, 74% denied subjective improvement in overall symptoms, 26% had some improvement, and none felt they were cured. Only 12% were able to return to their previous jobs, 50% were unable to return to work, 29% changed jobs, and 9% formally retired. These figures demonstrate that epicondylitis in patients seeking com-
JHS 䉬 Vol A, May–June
856
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
pensation can be a severe, persistent, disabling condition, with outcomes that are significantly worse than seen in the general population. The role of workers’ compensation was also examined in a cohort of patients treated with an open tennis elbow release. Balk et al. found that both workers’ compensation and non-workers’ compensation groups achieved similar levels of pain relief, although there was a trend toward more complete pain relief in the non-compensation group.25 The return to work rate was also similar between groups, although a greater percentage of patients with workers’ compensation claims required a job change (24% vs 4%, p ⫽ .007). Although the workers’ compensation patients in our cohort demonstrated inferior results compared with those not seeking compensation, their outcomes were much better than those reported by Kay24 and comparable to those of Balk et al.25 The overall satisfaction in our cohort of patients seeking compensation was 7 out of 10; 5 felt they were cured, 12 had some improvement (better or much better), 5 were unchanged, and only 1 felt that results were worse after surgery. Out of 23 patients, 15 were able to return to their previous jobs (4 returned to full duties and 11 had some physical restrictions) and only 8 were unable to return to their previous work (4 changed jobs, 3 are not working, and 1 retired). Lateral epicondylitis can result in significant disability for individuals, affect workplace productivity, and possibly lead to the development of chronic pain. The cohort represented in this study includes a large number of patients seeking compensation and reflects a subgroup of patients with chronic, recalcitrant symptoms. Although these factors did adversely affect outcomes, the general results are still considered favorable, much better than results of litigants previously cited in some reports.24 The lack of detailed preoperative subjective and objective evaluation is a significant limitation of this retrospective study. However, the high percentage of patients who returned for detailed outcomes assessment, including evaluations specifically tailored to determine functioning in the workplace, is a significant strength. Another limitation of this study is that we included no control groups; as such, it provides only level IV evidence. We cannot conclude that our outcomes are better or worse than might have been achieved with other surgeries, but our results indicate that arthroscopic release of the ECRB is safe and provides symptomatic improvement in most patients with lateral epicondylitis, even when treating a
population of heavy manual laborers with recalcitrant symptoms. Patient selection and perceived occupational demands likely have an important role in determining outcomes, and more work is required to identify which factors can predict who will and will not respond to surgical intervention. Based on these preliminary findings, a multifactorial model will need to be considered, incorporating all domains of occupational demands. To determine whether this technique is more efficacious than either the percutaneous or open methods, a randomized controlled trial is required. REFERENCES 1. Coonrad RW, Hooper WR. Tennis elbow: its course, natural history, conservative and surgical management. J Bone Joint Surg 1973;55A: 1177–1182. 2. Cyriax JH. The pathology and treatment of tennis elbow. J Bone Joint Surg 1936;18:921–940. 3. Lo MY, Safran MR. Surgical treatment of lateral epicondylitis: a systematic review. Clin Orthop Relate Res 2007;463:98 –106. 4. Kumar VS, Shetty AA, Ravikumar KJ, Fordyce MJ. Tennis elbow: outcome following the Garden procedure: a retrospective study. J Orthop Surg (Hong Kong) 2004;12:226 –229. 5. Verhaar J, Walenkamp G, Kester A, van Mameren H, Van der Linden T. Lateral extensor release for tennis elbow: a prospective long-term follow-up study. J Bone Joint Surg 1993;75A:1034 – 1043. 6. Yerger B, Turner T. Percutaneous extensor tenotomy for chronic tennis elbow: an office procedure. Orthopedics 1985;8:1261–1263. 7. Nirschl RP, Pettrone FA. Tennis elbow: the surgical treatment of lateral epicondylitis. J Bone Joint Surg 1979;61A:832– 839. 8. Baumgard SH, Schwartz DR. Percutaneous release of the epicondylar muscles for humeral epicondylitis. Am J Sports Med 1982;10: 223–226. 9. Owens BD, Murphy KP, Kuklo TR. Arthroscopic release for lateral epicondylitis. Arthroscopy 2001;17:582–587. 10. Baker CL, Murphy KP, Gottlob CA, Curd DT. Arthroscopic classification and treatment of lateral epicondylitis: two-year clinical results. J Shoulder Elbow Surg 2000;9:475– 482. 11. MacDermid JC. Outcome evaluation in patients with elbow pathology: issues in instrument development and evaluation. J Hand Ther 2001;14:105–114. 12. King GJW, Richards RR, Zuckerman JD, Blasier R, Dillman C, Friedman RJ, et al. A standardized method for assessment of elbow function. J Shoulder Elbow Surg 1999;8:351–354. 13. Ware JE, Kosinski M, Keller SD. A 12-item short-form health survey construction of scales and preliminary tests of reliability and validity. Med Care 1996;34:220 –233. 14. Lerner D, Amick BC III, Rogers WH, Malspeis S, Bungay K, Cynn D. The Work Limitations questionnaire. Med Care 2001;39:72– 85. 15. Lerner D, Reed JI, Massarotti E, Wester LM, Burke TA. The Work Limitations questionnaire’s validity and reliability among patients with osteoarthritis. J Clin Epidemiol 2002;55:197–208. 16. MacDermid JC, Alyafi T, Richards RS. Test–retest reliability of static and endurance grip strength tests performed on the Jamar and NK. Physiother Can 2001;53:48 –54. 17. Armstrong AD, MacDermid JC, Chinchalkar S, Stevens RS, King GJ. Reliability of range-of-motion measurement in the elbow and forearm. J Shoulder Elbow Surg 1998;7:573–580. 18. Patterson LA, Spivey WE. Validity and reliability of the LIDO active isokinetic system. J Orthop Sports Phys Ther 1992;15:32–36. 19. Baker CL, Jones GL. Current concepts: arthroscopy of the elbow. Am J Sports Med 1999;27:251–264.
JHS 䉬 Vol A, May–June
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
20. Morrey BF, Adams RA. Semiconstrained arthroplasty for the treatment of rheumatoid arthritis of the elbow. J Bone Joint Surg 1992; 74A:479 – 490. 21. Grundberg AB, Dobson JF. Percutaneous release of the common extensor origin for tennis elbow. Clin Orthop 2000;376:137–140. 22. Dunkow PD, Jatti M, Muddu BN. A comparison of open and percutaneous techniques in the surgical treatment of tennis elbow. J Bone Joint Surg 2004;86B:701–704.
857
23. Buchbinder R, Green SE, Youd JM, Assendelft WJ, Barnsley L, Smidt N. Shock wave therapy for lateral elbow pain. Cochrane Database Syst Rev 2005, CD003524, 1–51. 24. Kay NRM. Litigants’ epicondylitis. J Hand Surg 2003;28B:460 – 464. 25. Balk ML, Hagberg WC, Buterbaugh GA, Imbriglia JE. Outcome of surgery for lateral epicondylitits (tennis elbow): effect of workers’ compensation. Am J Orthop 2005;34:122–126.
JHS 䉬 Vol A, May–June
Functional Outcome of Arthroscopic Extensor Carpi Radialis Brevis Tendon Release in Chronic Lateral Epicondylitis Ruby Grewal, MD, Joy C. MacDermid, PhD, Parag Shah, MD, Graham J.W. King, MD
Purpose To evaluate outcomes of arthroscopic tennis elbow release in a population of patients with chronic, recalcitrant symptoms, a large number of workers’ compensation claims, and high occupational demands using standardized outcome measures, including a detailed objective assessment of workplace demands. Methods We treated 36 patients with chronic lateral epicondylitis with an arthroscopic release. A standardized protocol was used to measure strength, motion, and outcomes (American Shoulder and Elbow Surgeons Elbow [ASES-e] score, Short Form-12, PatientRated Tennis Elbow evaluation [PRTEE], and Work Limitations Questionnaire-26). Results The mean duration of symptoms before surgery was 30 months. A total of 25 of 36 patients were employed in heavy or repetitive occupations and 23 of 36 were involved in a workers’ compensation claim. The final overall results were favorable, with 30 of 36 subjects reporting improvement with surgery. The final mean Mayo Elbow Performance Index score was 78.6 ⫾ 16.5 (22 ⫽ good to excellent, 9 ⫽ fair, and 5 ⫽ poor). The average total PRTEE was 26.2 ⫾ 24.3 out of 100. The average ASES-e pain score was 16.1 ⫾ 15.0 and the average ASES-e function score was 27.9 ⫾ 8.8. Patients in heavy or repetitive occupations and those with workers’ compensation claims had significantly worse outcome scores (Mayo Elbow Performance Index, ASES, and PRTEE). Based on Work Limitations Questionnaire-26 scores, patients with workers’ compensation claims had significantly greater difficulties with physical (36.8 vs 3.2, p ⬍ .001), output (40.8 vs 3.1, p ⫽ .002), mental (36.0 vs 9.0, p ⫽ .05), and social (27.7 vs 6.3, p ⫽ .05) workplace demands. Conclusions Arthroscopic tennis elbow release provides symptomatic improvement in most patients with lateral epicondylitis. Patient selection and reported occupational demands have an important role in determining outcomes. More work is required to identify factors predicting outcome in this difficult subgroup. (J Hand Surg 2009;34A:849–857. Copyright © 2009 by the American Society for Surgery of the Hand. All rights reserved.) Type of study/level of evidence Therapeutic IV. Key words Arthroscopy, lateral epicondylitis, tennis elbow, chronic pain, workers’ compensation.
FromtheDivisionofOrthopedicSurgery,UniversityofWesternOntario,HandandUpperLimbCenter,St Joseph’sHealthCare,London,Ontario,Canada;andtheFractureandOrthopaedicHospital,Pali,Ahmedabad, India.
Corresponding author: Ruby Grewal, MD, Division of Orthopedic Surgery, University of Western Ontario,HandandUpperLimbCenter,StJoseph’sHealthCare,268GrosvenorStreet,LondonON,N6A 4L6, Canada; e-mail: [email protected].
Received for publication October 15, 2008; accepted in revised form February 5, 2009.
0363-5023/09/34A05-0007$36.00/0 doi:10.1016/j.jhsa.2009.02.006
No benefits in any form have been received or will be received related directly or indirectly to the subject of this article.
© ASSH 䉬 Published by Elsevier, Inc. All rights reserved. 䉬 849
850
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
tennis elbow, is a common cause of elbow pain and dysfunction. Although this disorder can occur in tennis players and other athletes, the most common clinical scenario involves a worker who performs heavy or repetitive tasks.1 Cyriax wrote that tennis elbow is a self-resolving condition whose symptoms usually last for 8 to 12 months; however, he also stated that there is a “chronic occupational type” of this disorder that is less likely to resolve.2 A recent literature review of the surgical treatment of lateral epicondylitis found that most patients have a fairly similar profile.3 They are in their fifth decade of life (range, 30 – 63 years), their dominant arm is most commonly affected (74%), they have an average duration of symptoms of 19 months, and they have failed nonsurgical treatment.3 Published case series4 –10 generally report favorable clinical outcomes with open, percutaneous, or arthroscopic surgery but there are few reports that use standardized, validated measures to report outcomes.3 The purpose of this study was to evaluate outcomes of arthroscopic tennis elbow release in patients with chronic recalcitrant symptoms, many with workers’ compensation claims, and high occupational demands, using standardized outcome measures, including a detailed objective assessment of workplace demands.
L
ATERAL EPICONDYLITIS, OR
MATERIALS AND METHODS Arthroscopic debridement and release of the extensor carpi radialis brevis (ECRB) tendon has been the preferred form of surgical intervention for lateral epicondylitis by the senior author since 1999. Surgical indications included failed conservative treatment (physiotherapy, occupational therapy, tennis elbow bracing, and cortisone injections) and persistence of symptoms for a minimum of 1 year. The senior author performed arthroscopic ECRB releases in 48 patients between 1999 and 2006. Patients with revision surgery were excluded. A total of 36 patients were able to return for reassessment. Twelve patients could not be reached for follow-up: 1 had a recent stroke, 2 could not be located, 7 failed to return our phone calls or were unable to attend their follow-up appointment, and 2 refused to return for follow-up because of their geographical distance from our clinic. An examiner who was not involved with their care evaluated all 36 patients. Patients completed the Patient-Rated Tennis Elbow evaluation (PRTEE), which has a visual analog scale for pain and function,11 the American Shoulder and Elbow Surgeons Elbow (ASES-e) patient self-evaluation form for pain and
function,12 and the Short Form-12 (SF-12), version 2, health survey for general health and well-being.13 Patients who were working (35 of 36) also completed a 26-item version of the Work Limitations Questionnaire (WL-26).14,15 The WL-26 score consists of 5 domains: scheduling, mental, physical, social, and output demands (ability to work quickly enough and finish work on time). Each domain is scored separately (best score ⫽ 0, worst score ⫽ 100); a score of 50 indicates the average patient is having difficulty meeting the demands of that domain for half of each workday (score 33 ⫽ difficulty for one third of the workday). Patients were also asked to assess their results as being completely cured, much better, better, the same, or worse. Objective assessment included a detailed clinical examination and a standardized protocol for measuring grip16 and motion.17 Strength testing was conducted by the LIDO WorkSET isokinetic dynamometer (Loredan Biomedical, West Sacramento, CA).18 Isometric grip strength was measured as maximum and pain-free grip strength with elbow extended. We also measured the isometric wrist extension and flexion strength. All measurements were recorded thrice after an interval of 30 seconds between each measurement. Statistical softare (SPSS v15.0; Chicago, IL) was used to calculate descriptive statistics and correlations and test for differences in subgroups using t-tests and side-to-side differences (paired t-tests). Surgical technique Under general anesthesia, patients were positioned in the lateral decubitus position. The elbow was injected with 20 mL saline to distend the joint. Our standard visualization portal for tennis elbow release was the proximal anteromedial portal. The anterior elbow joint was inspected, and the capsule overlying the common extensor origin was visualized. Any capsular or tendon abnormality was noted. Patients were classified as having type 1, 2, or 3 lesions.19 Type 1 lesions have inflammation and fraying deep to the extensor carpi radialis brevis (ECRB) tendon without evidence of a frank tear. Type 2 lesions are linear tears at the undersurface of the ECRB tendon, and type 3 lesions are retracted with partial or completed avulsions of the tendon. The proximal anterolateral working portal was made using an outside-in method. The capsule and the ECRB tendon were resected using a motorized shaver. Care was taken to avoid injury to the lateral ulnar collateral ligament by limiting the resection to the portion anterior to the mid-axis of the radial head. We did not decorticate the lateral epicondyle. We used only 2 arthroscopic
JHS 䉬 Vol A, May–June
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
portals and did not perform a posterior elbow arthroscopy in this cohort. Portals were sutured with 3– 0 nylon sutures and a soft bulky dressing was applied. Postoperative care Active motion was encouraged postoperatively. Patients were instructed to remove the dressing 48 hours postoperatively and apply a bandage over each portal site. Sutures were removed at 10 days. Patients did not routinely attend physiotherapy but were encouraged to return gradually to their normal activities as their symptoms allowed. Heavy or repetitive work was not permitted for 6 weeks postoperatively. Demographic data There were 20 males and 16 females in the study group. The average age of our subjects was 45.3 ⫾ 7.0 years (range, 29 – 61 years). Thirty patients had their dominant arm affected. A total of 23 patients were involved in a workers’ compensation claim. Ten patients recalled a specific event leading to pain, whereas 26 of them described the onset of pain as insidious. Thirteen patients described their work as heavy, 12 described their work as repetitive, 10 had clerical duties or other physically light jobs, and 1 was retired. Before developing lateral epicondylitis, 25 patients were involved in recreational sports. The mean duration of symptoms before surgery was 30 ⫾ 20.5 months (range, 1–10 years); 12 patients had symptoms for 2 to 3 years, and 9 patients had symptoms for ⬎3 years. The average number of previous cortisone injections was 2.5 ⫾ 1.9 (range, 0 –10 injections), with 17 patients receiving 3 or more injections and 4 patients receiving no injections. Although all patients ultimately failed nonsurgical measures, 23 reported some transient symptomatic relief with the cortisone injections. There were 12 patients with type 1 lesions, 4 with type 2 lesions, 2 with type 3 lesions, and 18 with no lesion of the ECRB tendon identified intraoperatively.19 Additional intraoperative arthroscopy findings were as follows: 13 patients had synovitis, 4 had osteophytes, and 4 had chondromalacia of the radial head. One patient had a localized synovectomy at the time of arthroscopic release of the ECRB and one had an open carpal tunnel release under the same anesthetic. One patient developed mild erythema around the portals that required oral antibiotics; no other complications were observed. RESULTS Pain and disability The average length of follow-up was 42 months (range, 19 –74 months). Thirty subjects felt that they had im-
851
proved with surgery; 14 patients described themselves as completely asymptomatic, 16 were better or much better, 5 remained the same, and 1 felt worse. On a visual analog score, the average satisfaction was 8 out of 10. A final PRTEE was calculated, with 100 representing the worst possible score (maximal pain and functional disability). The average PRTEE for pain was 14.6 ⫾ 12.9 out of 50 (range, 0 –37). The average PRTEE for functional disability was 11.3 ⫾ 11.9 out of 50 (range, 0 –37). The average total PRTEE was 26.2 ⫾ 24.3 out of 100. The average ASES-e pain score was 16.1 ⫾ 15.0 (range, 0 –50, where 0 ⫽ no pain); the average ASES-e function score was 27.9 ⫾ 8.8 (range, 0 –36, where 36 ⫽ full function). The mean Mayo Elbow Performance Index (MEPI) score20 was 78.6 ⫾ 16.5; 22 patients had a good to excellent result, 9 had a fair result, and 5 had a poor result. The mean SF-12v2 physical component summary score was 44.6 and the mental component summary score was 49.8 (US population norms ⫽ 50). At final follow-up, there was a mean improvement of 27.3% in grip strength (affected grip strength was expressed as a percentage of the unaffected grip strength). This improvement was statistically significant (p ⫽ .001). There were slight side-to-side differences (affected/unaffected arm) demonstrated in elbow flexion (138.0°/141.4°, p ⫽ .014), elbow extension (1.9°/0.9°, p ⫽ .054), and grip strength (36.0/40.2 kg, p ⫽ .009), although the clinical importance of these differences is questionable. No side-to-side differences were noted in wrist extension (9.2/10.0 Nm, p ⫽ .17) or wrist flexion strength (10.7/11.4 Nm, p ⫽ .19). We found no associations among patient age, duration of symptoms, presence of additional intraoperative findings (ie, cartilage damage, synovitis), type of lesion, presence or absence of ECRB lesion at time of arthroscopy, presence of bilateral symptoms, and final outcomes. Patients who reported temporary relief with cortisone injections had less pain and disability at final follow-up compared with patients who received no benefit from these injections (mean PRTEE, 18.5 vs 39.2, p ⫽ .04). Before developing tennis elbow, 25 patients were involved in recreational sports, 16 of whom were able to return to sports (11 unrestricted, 5 restricted). Effect of work history on outcomes Of 36 patients in this cohort, 23 were involved in workers’ compensations claims. Fifteen of the 23 patients were able to return to work after surgery. Two patients involved in heavy labor and 2 involved in
JHS 䉬 Vol A, May–June
852
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
repetitive tasks made a full return to work. The remaining 11 patients returned to work but required restrictions on their physical demands (2 of these patients were heavy laborers, 6 were involved in repetitive tasks, 3 were involved in clerical or other light duties at work). Seven patients with workers’ compensations claims were unable to return to work, and 1 patient retired. Four patients changed their jobs as a result of their elbow pain: 3 of these patients were previously involved in repetitive work and 1 was involved in heavy work. The remaining 3 patients did not return to work owing to persistent arm pain: 1 had work that involved repetitive tasks and 2 were employed in heavy labor. Of the 13 patients who had no compensation claims, 1 was retired, another retired from heavy labor as a result of his elbow pain, and the remaining 11 were all able to return to work. Five patients returned to jobs involving heavy labor, 5 returned to a job involving clerical work, and 1 returned to his job as a janitor but required activity restrictions. None of the patients were required to change their occupations. The time taken off work after surgery averaged 18.5 ⫾ 27.5 weeks for the whole group (range, 0 –104 weeks). The compensation patients took an average of 24.5 ⫾ 32.6 weeks off work (range, 1–104 weeks), whereas the non-compensation patients took an average of 10.3 ⫾ 16.6 weeks off work (range, 0 –52 weeks) (p ⫽ .2). Patients performing heavy or repetitive work (n ⫽ 18) took an average of 23.9 ⫾ 31.4 weeks off work, and those performing light work (n ⫽ 8) took an average of 6.4 ⫾ 8.0 weeks before returning to work (p ⫽ .04). Patients with workers’ compensation claims had significantly worse outcome scores (MEPI, ASES-e pain, function, satisfaction, and PRTEE scores) than those without claims (Table 1). Patients involved in heavy or repetitive occupations also had significantly worse ASES-e pain, function, satisfaction, and PRTEE scores than patients performing clerical jobs, but no significant difference occurred in the MEPI scores (Table 2). The Workplace Limitations Questionnaire14,15 identified difficulties in all 5 work domains. The mean score for difficulty handling scheduling demands was 18.0 ⫾ 25.9, social demands was 19.9 ⫾ 28.8, physical demands was 22.9 ⫾ 22.9, mental demands was 26.1 ⫾ 33.2, and output demands was 27.1 ⫾ 33.8. As expected, physical demands correlated significantly with all outcome measures (p ⬍ .05); however, we found that mental, social, and scheduling demands also correlated significantly (p ⬍ .05) with self-report outcome measures (Table 3). Patients with workers’ compensation claims had greater difficulties in all occupational
TABLE 1. Relationship Between Outcomes and History of Workers’ Compensation Claim Workers’ Compensation Case Yes
No
p Value
71.8
90.0
.001
Pain
21.0
6.8
Function
24.1
34.3
⬍.001
7.0
9.8
⬍.001
36.7
7.6
⬍.001
Mental
47.7
54.3
.16
Physical
41.4
51.3
.04
MEPI ASES
Satisfaction PRTEE
.003
SF-12
TABLE 2. Relationship Between Outcomes and Type of Work Heavy or Repetitive Work Yes
No
p Value
75.1
85.7
.08
Pain
18.6
8.1
.05
Function
25.9
33.0
.02
7.7
9.9
.001
31.2
13.1
.02
Mental
48.6
55.1
.23
Physical
42.2
54.0
.02
MEPI ASES
Satisfaction PREE SF-12
domains (not only physical demands) compared with those who were not involved with compensation claims. These differences were statistically significant in the domains of mental demands (36.0 vs 9.0, p ⫽ .05), output demands (40.8 vs 3.1, p ⫽ .002), physical demands (36.8 vs 3.2, p ⬍ .001), and social demands (27.7 vs 6.3, p ⫽ .05). Although the results were not statistically significant, workers’ compensation patients also showed greater difficulties with scheduling demands (25.1 vs 5.5, p ⫽ .09) (Fig. 1). Patient selfassessment statements also correlated with high workplace limitation scores. Patients who felt they were worse, unchanged, or only slightly better after surgery had remarkably greater workplace limitations than
JHS 䉬 Vol A, May–June
853
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
TABLE 3.
Pearson Correlation Coefficients for Occupational Demands and Outcome Scores
Outcome Scores
Scheduling Demands
Mental Demands
Output Demands
Social Demands
Physical Demands
–0.375
–0.412
–0.324
–0.195
–0.436*
MEPI ASES Function
0.516*
–0.607†
–0.532*
–0.448*
–0.696†
Pain
0.320
0.598†
0.287
0.431*
0.484*
SF-12 Physical
–0.530*
–0.660†
–0.412
–0.599*
–0.706†
Mental
–0.724†
–0.914†
–0.436
–0.760†
–0.741†
0.530*
0.638†
0.401
0.410
0.589†
PRTEE Total Pain
0.440*
0.558†
0.390
0.326
0.560†
Function
0.560†
0.646†
0.405
0.436*
0.612†
*p ⬍ .05. †p ⬍ .001.
45 40 35 30 25 20 15 10 5 0
Scheduling Demands
Mental Demands*
Physical Demands*
Workers Compensation
Social Demands*
Output Demands*
Nonworkers Compensation
FIGURE 1: Workplace Limitation scores in workers’ compensation and noncompensation patients (*p ⬍ .05).
those who felt they were much better or completely cured. The limitations were evident not only in the physical and output demands of their occupation, but also in the mental demands experienced in the workplace (Fig. 2). When examining the relationship between workplace limitation scores and job type, heavy laborers and those with repetitive jobs experienced more limitations than patients involved in light work or clerical jobs; however, only the domains of scheduling, output, and physical demands achieved statistical significance (Fig. 3). We conducted a chart review for the 12 patients who were unable to return for follow-up. At the time of last assessment, 2 patients were completely asymptomatic,
4 were markedly better, 2 were only slightly better, and 2 had no improvement at all. The outcomes of 2 patients who were not seen past their immediate postoperative visit could not be assessed. DISCUSSION There is a small subgroup of patients with lateral epicondylitis whose symptoms do not respond to conservative measures.2 These are the patients who usually present to surgeons. A variety of surgical treatments have been described (open, percutaneous, and arthroscopic), and all report good but not definitive results.7,8,21,22 In recent years, the arthroscopic technique has gained in popularity. There are many benefits to an
JHS 䉬 Vol A, May–June
854
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
70 60 50 40 30 20 10 0
Scheduling Demands
Mental Demands*
Physical Demands*
slightly better, same or worse
Social Demands
Output Demands*
much better or completely cured
FIGURE 2: Workplace Limitation scores and self-assessment (*p ⬍ .05).
40 35 30 25 20 15 10 5 0
Scheduling Demands*
Mental Demands
Physical Demands*
heavy or repetitive
Social Demands
Output Demands*
light or desk
FIGURE 3: Workplace Limitation scores and type of work (*p ⬍ .05).
arthroscopic approach to the treatment of lateral epicondylitis. The arthroscopic technique allows excellent visualization of the joint and an accurate localization of the ECRB tendon.9 There is also a lower morbidity and earlier return to work and activities than seen with open surgery9; however, the arthroscopic method is more technically demanding. This study demonstrates a positive change in health and work status after arthroscopic release in a difficult subgroup of patients; the study also reports a remarkable level of residual disability despite positive overall outcomes. Few reports in the literature describe the results of arthroscopic treatment for lateral epicondylitis.9,10 Owen et al. published a case series outlining the results of 16 patients observed for an average of 24.1 months
after arthroscopic treatment for lateral epicondylitis.9 In that cohort, there were no workers’ compensation cases and only one patient was employed as a manual laborer. The average duration of conservative treatment in that cohort was 31.7 months. In that series, there were no reported complications: 10 of 12 patients reported feeling “much better,” 2 were “better,” and none considered themselves to be the same or worse after surgery. Using a pain analog scale (0 –10, with 10 being the worst pain), the average pain at rest was 0.58, pain with activities of daily living was 1.58, and the average pain with sports and work was 3.25.9 Baker et al. also reported a series of patients treated arthroscopically.10 They observed 40 patients (42 elbows) for 2 years. The average duration of symptoms
JHS 䉬 Vol A, May–June
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
before surgery was 14 months. Ten patients were involved in a workers’ compensation claim and 22 patients performed heavy labor. The results in that cohort were also favorable. Of the 39 elbows in the 37 patients who were available for follow-up, 37 were rated “better” or “much better.” The average pain at rest was 0.9, pain with activities of daily living was 1.4, and average pain with sports and work was 1.9. Unfortunately, only 13 of their patients returned for objective follow-up. In those 13, the average MEPI score was 93.6 of 100, and grip strength averaged 96% of the strength of the unaffected limb.10 Although the literature demonstrates favorable results for the treatment of lateral epicondylitis, a recent Cochrane review23 concluded there was insufficient evidence to support any specific surgical intervention at this time. There are even fewer reports describing the outcomes of surgical treatment in patients with chronic recalcitrant symptoms, in a cohort of primarily workers’ compensation patients. The overall results in this cohort are comparable to those of Owens et al.9 and Baker et al.,10 although our patients had a longer duration of symptoms before surgery (mean, 29.7 months) and there was a larger number of patients involved in workers’ compensation claims (23) and in physically demanding occupations (25). Despite the number of potential predictors of poorer outcomes, 30 subjects felt that they had improved with surgery and patient satisfaction was high (8 of 10). Despite these good subjective results, only 22 had good or excellent results based on the MEPI score. Most patients had some residual symptoms that affected their measured and self-reported function and ability to participate in work. Although the overall results in this cohort were good, the outcome among patients involved in workers’ compensation claims or in heavy or repetitive labor was not as favorable. Patients with less demanding occupations, without workers’ compensation claims, were most likely to show favorable results, as were those who responded favorably to a previous steroid injection. Patients with workers’ compensation claims had worse outcomes based on all self-report outcome scores (ASES-e pain, function, satisfaction, and PRTEE) (Tables 1 and 2). Occupational demands also play a considerable role in determining outcomes in repetitive strain injuries such as lateral epicondylitis; however, no published reports exist in the literature that assess workplace demands in an objective, standardized fashion. Because of the high proportion of work-related injuries in our cohort, we included outcome measures that addressed
855
work performance in more detail than has been addressed in previous studies. The WL-26 score reflects the amount of time an individual is experiencing difficulties while at work. Work limitations reported using a slightly different version of the WL questionnaire in a previous study were 20 to 28 for patients with osteoarthritis (type of osteoarthritis in cohort: hand, 30.3%; back, 13.9% knee, 22.3%; neck, 13.9%; hip, 13.9%; other, 5.5%) and 4 to 10 for controls.15 The average scores in the current study were 18.0 ⫾ 25.9 for scheduling demands, 19.9 ⫾ 28.8 for social demands, 22.9 ⫾ 22.9 for physical demands, 26.1 ⫾ 33.2 for mental demands, and 27.1 ⫾ 33.8 for output demands, which suggests that patients who successfully return to work after arthroscopic release continue to experience a moderate amount of difficulty similar to that experienced by patients with osteoarthritis. We found that patients employed in heavy labor or repetitive jobs had significantly greater difficulty in the domains of physical, output, and scheduling demands, as one might expect. Our data also demonstrated that patients with workers’ compensation claims had greater difficulties in all occupational domains compared with those who were not involved with compensation claims. The differences were statistically significant not only in the domains of physical (p ⬍ .001) and output (p ⫽ .002) demands, but also in the domains of mental demands (p ⫽ .05) and social demands (p ⫽ .05), and showed a trend toward statistical significance in scheduling demands (p ⫽ .09). We found that patients’ selfreported global ratings of function also correlated with self-reported workplace limitation scores in all domains, but most significantly in the domains of physical, mental, output, and social demands. Whereas the physical aspects of job demands are typically of primary concern for surgeons, mental and social demands are often overlooked. The results of this cohort show that all domains of occupational demands are closely linked to outcomes and patient self-assessment and should be more closely explored in future studies, particularly in workers’ compensation cases. A recent series published by Kay in 2003 reviewed the results of a series of 108 consecutive patients with tennis elbow who were also seeking compensation.24 In that series, 42 of 108 patients received surgery. Of those patients, 74% denied subjective improvement in overall symptoms, 26% had some improvement, and none felt they were cured. Only 12% were able to return to their previous jobs, 50% were unable to return to work, 29% changed jobs, and 9% formally retired. These figures demonstrate that epicondylitis in patients seeking com-
JHS 䉬 Vol A, May–June
856
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
pensation can be a severe, persistent, disabling condition, with outcomes that are significantly worse than seen in the general population. The role of workers’ compensation was also examined in a cohort of patients treated with an open tennis elbow release. Balk et al. found that both workers’ compensation and non-workers’ compensation groups achieved similar levels of pain relief, although there was a trend toward more complete pain relief in the non-compensation group.25 The return to work rate was also similar between groups, although a greater percentage of patients with workers’ compensation claims required a job change (24% vs 4%, p ⫽ .007). Although the workers’ compensation patients in our cohort demonstrated inferior results compared with those not seeking compensation, their outcomes were much better than those reported by Kay24 and comparable to those of Balk et al.25 The overall satisfaction in our cohort of patients seeking compensation was 7 out of 10; 5 felt they were cured, 12 had some improvement (better or much better), 5 were unchanged, and only 1 felt that results were worse after surgery. Out of 23 patients, 15 were able to return to their previous jobs (4 returned to full duties and 11 had some physical restrictions) and only 8 were unable to return to their previous work (4 changed jobs, 3 are not working, and 1 retired). Lateral epicondylitis can result in significant disability for individuals, affect workplace productivity, and possibly lead to the development of chronic pain. The cohort represented in this study includes a large number of patients seeking compensation and reflects a subgroup of patients with chronic, recalcitrant symptoms. Although these factors did adversely affect outcomes, the general results are still considered favorable, much better than results of litigants previously cited in some reports.24 The lack of detailed preoperative subjective and objective evaluation is a significant limitation of this retrospective study. However, the high percentage of patients who returned for detailed outcomes assessment, including evaluations specifically tailored to determine functioning in the workplace, is a significant strength. Another limitation of this study is that we included no control groups; as such, it provides only level IV evidence. We cannot conclude that our outcomes are better or worse than might have been achieved with other surgeries, but our results indicate that arthroscopic release of the ECRB is safe and provides symptomatic improvement in most patients with lateral epicondylitis, even when treating a
population of heavy manual laborers with recalcitrant symptoms. Patient selection and perceived occupational demands likely have an important role in determining outcomes, and more work is required to identify which factors can predict who will and will not respond to surgical intervention. Based on these preliminary findings, a multifactorial model will need to be considered, incorporating all domains of occupational demands. To determine whether this technique is more efficacious than either the percutaneous or open methods, a randomized controlled trial is required. REFERENCES 1. Coonrad RW, Hooper WR. Tennis elbow: its course, natural history, conservative and surgical management. J Bone Joint Surg 1973;55A: 1177–1182. 2. Cyriax JH. The pathology and treatment of tennis elbow. J Bone Joint Surg 1936;18:921–940. 3. Lo MY, Safran MR. Surgical treatment of lateral epicondylitis: a systematic review. Clin Orthop Relate Res 2007;463:98 –106. 4. Kumar VS, Shetty AA, Ravikumar KJ, Fordyce MJ. Tennis elbow: outcome following the Garden procedure: a retrospective study. J Orthop Surg (Hong Kong) 2004;12:226 –229. 5. Verhaar J, Walenkamp G, Kester A, van Mameren H, Van der Linden T. Lateral extensor release for tennis elbow: a prospective long-term follow-up study. J Bone Joint Surg 1993;75A:1034 – 1043. 6. Yerger B, Turner T. Percutaneous extensor tenotomy for chronic tennis elbow: an office procedure. Orthopedics 1985;8:1261–1263. 7. Nirschl RP, Pettrone FA. Tennis elbow: the surgical treatment of lateral epicondylitis. J Bone Joint Surg 1979;61A:832– 839. 8. Baumgard SH, Schwartz DR. Percutaneous release of the epicondylar muscles for humeral epicondylitis. Am J Sports Med 1982;10: 223–226. 9. Owens BD, Murphy KP, Kuklo TR. Arthroscopic release for lateral epicondylitis. Arthroscopy 2001;17:582–587. 10. Baker CL, Murphy KP, Gottlob CA, Curd DT. Arthroscopic classification and treatment of lateral epicondylitis: two-year clinical results. J Shoulder Elbow Surg 2000;9:475– 482. 11. MacDermid JC. Outcome evaluation in patients with elbow pathology: issues in instrument development and evaluation. J Hand Ther 2001;14:105–114. 12. King GJW, Richards RR, Zuckerman JD, Blasier R, Dillman C, Friedman RJ, et al. A standardized method for assessment of elbow function. J Shoulder Elbow Surg 1999;8:351–354. 13. Ware JE, Kosinski M, Keller SD. A 12-item short-form health survey construction of scales and preliminary tests of reliability and validity. Med Care 1996;34:220 –233. 14. Lerner D, Amick BC III, Rogers WH, Malspeis S, Bungay K, Cynn D. The Work Limitations questionnaire. Med Care 2001;39:72– 85. 15. Lerner D, Reed JI, Massarotti E, Wester LM, Burke TA. The Work Limitations questionnaire’s validity and reliability among patients with osteoarthritis. J Clin Epidemiol 2002;55:197–208. 16. MacDermid JC, Alyafi T, Richards RS. Test–retest reliability of static and endurance grip strength tests performed on the Jamar and NK. Physiother Can 2001;53:48 –54. 17. Armstrong AD, MacDermid JC, Chinchalkar S, Stevens RS, King GJ. Reliability of range-of-motion measurement in the elbow and forearm. J Shoulder Elbow Surg 1998;7:573–580. 18. Patterson LA, Spivey WE. Validity and reliability of the LIDO active isokinetic system. J Orthop Sports Phys Ther 1992;15:32–36. 19. Baker CL, Jones GL. Current concepts: arthroscopy of the elbow. Am J Sports Med 1999;27:251–264.
JHS 䉬 Vol A, May–June
ARTHROSCOPIC LATERAL EPICONDYLITIS RELEASE
20. Morrey BF, Adams RA. Semiconstrained arthroplasty for the treatment of rheumatoid arthritis of the elbow. J Bone Joint Surg 1992; 74A:479 – 490. 21. Grundberg AB, Dobson JF. Percutaneous release of the common extensor origin for tennis elbow. Clin Orthop 2000;376:137–140. 22. Dunkow PD, Jatti M, Muddu BN. A comparison of open and percutaneous techniques in the surgical treatment of tennis elbow. J Bone Joint Surg 2004;86B:701–704.
857
23. Buchbinder R, Green SE, Youd JM, Assendelft WJ, Barnsley L, Smidt N. Shock wave therapy for lateral elbow pain. Cochrane Database Syst Rev 2005, CD003524, 1–51. 24. Kay NRM. Litigants’ epicondylitis. J Hand Surg 2003;28B:460 – 464. 25. Balk ML, Hagberg WC, Buterbaugh GA, Imbriglia JE. Outcome of surgery for lateral epicondylitits (tennis elbow): effect of workers’ compensation. Am J Orthop 2005;34:122–126.
JHS 䉬 Vol A, May–June