Abstract and Introduction
Abstract
Objective
This prospective investigation is designed to determine the prognostic factors associated with the response to conservative therapy of subacromial impingement syndrome.
Materials and Methods
We treated 102 patients with subacromial impingement syndrome, diagnosed by physical examination and a subacromial lidocaine injection test, with a standardized conservative protocol. We followed the patients for a period of 12 months. Outcome was evaluated with Constant score and effects of 8 variables: Age, sex, pretreatment symptom duration, dominant shoulder, initial Constant score, active range of motion, acromion morphology, and acromial spur on patient outcomes were evaluated with univariate and multivariate analyses.
Results
Among 89 patients (44 men; mean age, 56.4 years) who finished the study, the mean difference between initial and final scores was 15.9 (95% confidence interval [CI]: 13.9–17.8). Three variables, the initial Constant score (b = .52, 95% CI: .28–.76), the duration of disease before treatment (b = −4.4, 95% CI: from −7.2 to −1.6), and acromial morphology (b = −5.3, 95% CI: from −9.8 to −.8) were found to be independent predictors of outcome (model R2 = .68).
Conclusion
Patients with more severe disease, a long duration of symptoms, and type II or III acromion may require more invasive therapeutic options as the first intervention.
Introduction
Subacromial impingement and rotator cuff tears are common causes of shoulder disability. For the first time, Neer[1] popularized the concept of this syndrome, and described it as impingement of the rotator cuff tendon below the anteroinferior part of the acromion. This impingement caused tendonitis and eventual tendon tear. He described a clinical sign in which pain was aggravated when the affected shoulder was placed in forward-flexed and internally rotated position. He also introduced an impingement test that evaluated pain relief in response to subacromial injection of local anesthetics.
Three current accepted approaches to treatment of impingement syndrome include conservative management, arthroscopic surgery, and open surgical interventions. All of these therapeutic modalities have been found to be effective, even though there is controversy in the medical literature[2–6] about their success rates. Based on the effectiveness of noninvasive therapies, many investigators recommend a period of conservative management for at least 6–12 months.[6–10] Nevertheless, with the advent of arthroscopic surgical techniques, which are remarkably less invasive and require shorter rehabilitation periods than open surgical procedures, some surgeons suggest earlier arthroscopic intervention in rotator cuff disorders.[11]
A significant amount of our knowledge about the prognostic factors associated with the response to conservative therapy is from retrospective studies. The aim of our study is to determine these prognostic factors on the basis of a prospective investigation.
Materials and Methods
From March 2001 to February 2002, a total of 102 patients who referred to our outpatient orthopaedic clinic, and who consented to participate in the study, were enrolled. Participants eligible for this study were those with a newly confirmed diagnosis of subacromial impingement syndrome. Diagnosis was performed on the basis of positive Neer's[1] and Hawkins and Abrams'[12] signs as well as Neer's impingement test (injection of 10 cc of 1% lidocaine lateral to the anterior edge of the acromion and looking for reduction in pain severity). For each participant, a series of radiographs was done and included anteroposterior, true lateral, supraspinatous outlet, and axillary views of the affected shoulder.
This study was approved by the ethics committee of our institutional review board. Patients with a history of any of the following were excluded: fracture in the proximal part of the humerus, frozen shoulder, cervical radiculopathy, and long-lasting diabetes mellitus. Exclusion criteria also included the presence of radiographic signs of glenohumeral or acromioclavicular arthritis. A total of 177 consecutive patients were evaluated for inclusion of whom 49 did not meet the inclusion criteria. Thirty-five patients did not agree to sign the informed consent. This yielded a participation rate of 79%. There was no significant difference between the Constant scores of these patients and the participants (results not shown).
For each participant, a detailed questionnaire about his/her demographic factors and medical history was completed by a trained interviewer. A thorough physical examination was performed by 1 physician. Participants then entered the 12-month treatment program under a particular treatment protocol, which included an oral nonsteroidal anti-inflammatory drug (indomethacin, 75 mg 3 times daily) and (if needed) local corticosteroid injections (in which a maximum of 2 injections was allowed). After the pain was under control and the participants were able to tolerate the exercise, a detailed physical therapy program was initiated under the supervision of a specialist and continued for the rest of the follow-up period.
Follow-up visits were performed bimonthly. At each visit, a questionnaire was completed about any adverse events, participants' adherence to drug therapy, and any subjective report of pain alleviation or aggravation. Patients were given a daily physical treatment program and were asked to log their activities and drug usage. Participants were reminded of visits by telephone calls. The Constant scores[13] were calculated for the first and the last follow-up visits.
Correlations of 8 baseline variables (age, sex, pretreatment Constant score, duration of symptoms, active range of motion, involvement of dominant or nondominant side, acromial morphology, and the presence of acromial spur) to the final Constant score were evaluated. A multivariate linear regression model was used to determine the independent effects of variables on clinical outcome. With the use of a linear model, we avoided categorizing treatment results on the basis of arbitrary cut-off values. To evaluate the fitness of the multivariate model, each significant variable was entered into a univariate model; coefficients of determination (adjusted R-square) and the standard error of the estimates were compared.
Results
Of the 102 participants who entered the study, 13 (12.7%) were excluded from the final analysis because of short follow-up (7 cases), discontinuation of drugs due to gastrointestinal adverse effects (2 cases), and noncompliance to the follow-up program (4 cases). Baseline characteristics of 89 participants who completed the follow-up program are shown in Table 1.
Table 1.
Baseline Characteristics of the Study Participants
| Variable | Total (n = 89) |
|---|---|
| Age (years) | |
| Mean | 56.4 |
| SD | 10.6 |
| Range | 34–80 |
| Sex | |
| Men | 45 |
| Women | 44 |
| Symptom duration (days) | |
| Mean | 133 |
| Range | 14–730 |
| Dominant shoulder | |
| Involved | 73 |
| Spared | 16 |
| Acromial morphology | |
| Type I | 8 |
| Type II | 69 |
| Type III | 12 |
| Active range of motion* | |
| Normal | 53 |
| Mildly impaired | 18 |
| Moderately impaired | 12 |
| Severely impaired | 6 |
| Acromial spur | |
| Present | 8 |
| Absent | 81 |
| Initial constant score | |
| Mean | 66.7 |
| SD | 11.1 |
| Range | 40–90 |
Range of motion was measured in active abduction and forward flexion and categorized as normal (>/= 140°), mildly impaired (120°-139°), moderately impaired (90°-119°), and severely impaired (<90°).
Mean constant scores of the participants at the initial and last follow-up visits were 66.7 (SD = 11.1) and 82.6 (SD = 13.5), respectively. Participants had an average increase of 15.9 (95% confidence interval: 13.9–17.8) in their Constant scores. At the last follow-up, 69% (61 cases) of participants had an improvement of more than or equal to 10 points in their Constant scores; 27% (24 cases) had minor or no improvement (less than 10 points); and 4.5% (4 cases) experienced deterioration of the disease.
The association of 8 baseline variables to the final outcome was evaluated, in which 4 variables (pretreatment Constant score, pretreatment duration of symptoms, active range of motion, and acromial morphology) were found to be correlated to the final constant score (Table 2).
Table 2.
Results of Univariate and Multivariate Analyses of Association of Baseline Variables With Final Outcome
| 95% Confidence Interval | ||||||
|---|---|---|---|---|---|---|
| Variable | Pearson Correlation Coefficient | P Value | Multivariate Regression Coefficient | Lower | Upper | P Value |
| Age | −.12 | .26 | .002 | −.15 | .19 | .81 |
| Sex | .02 | .85 | −1.49 | −5.03 | 2.04 | .40 |
| Symptom duration* | −.73 | < .001†† | −4.40 | −7.24 | −1.56 | .003†† |
| Dominant shoulder involvement | .05 | .63 | .65 | −3.83 | 5.13 | .77 |
| Acromial morphology | −.60 | < .001†† | −5.30 | −9.85 | −.75 | .023†† |
| Active range of motion | −.52 | < .001†† | .54 | −2.01 | 3.09 | .67 |
| Acromial spur | −.12 | .27 | −3.69 | −9.01 | 1.63 | .17 |
| Initial Constant score | .73 | < .001†† | .52 | .28 | .76 | < .001†† |
| Regression constant | – | – | 77.52 | 43.76 | 111.27 | < .001†† |
Log transformation was used to achieve a normal distribution.
P < .05 was considered significant.
In a multivariate linear model, 3 variables, including the pretreatment Constant score, the duration of involvement, and acromial morphology, were found to be independent predictors of the treatment outcome (Table 2). Analysis of residuals revealed a normal distribution (1-sample Kolmogorov-Smirnov P value = .3). The 3-variable model had an R-square of .68 with a standard error of the estimate of 7.6. This model had a better fitness than each of the univariate models on the basis of 3 mentioned variables (Table 3).
Table 3.
Fitness of Multivariate and Univariate Regression Models
| Models (Predictor Variables) | Adjusted R2 | Standard Error of the Estimate |
|---|---|---|
| Model 1 (acromion morphology) | .36 | 10.9 |
| Model 2 (logarithm of symptom duration) | .50 | 9.6 |
| Model 3 (initial Constant score) | .52 | 9.8 |
| Model 4 (3-variable model) | .68 | 7.6 |
Discussion
We found that the predictive value of the pretreatment Constant score could be empowered by taking into account the effects of acromion morphology and pretreatment symptom duration. This is quantitatively shown by better fitness of the 3-variable model than the univariate models.
Diagnosis of subacromial impingement syndrome in our study was based on the classic Neer's[1] sign and impingement test. Preferred reference tests for assessment of the shoulder in soft-tissue disorders are arthroscopy and surgery,[14] which are not applicable in a conservative management setting as was used in this study. Neer's test has been estimated to have a high sensitivity and a relatively low specificity.[14] However, our exclusion criteria seemed to refine this specificity, because it is likely to exclude patients with other diagnostic labels, such as adhesive capsulitis and glenohumeral osteoarthritis. Other diagnostic tests, such as ultrasonography and magnetic resonance imaging (MRI), are unlikely to be superior to Neer's test in diagnosis of the impingement syndrome.[14]
Conservative therapy used in different studies consists of several modalities, including oral nonsteroidal anti-inflammatory drugs, intra-articular and subacromial glucocorticoid injections, oral glucocorticoid treatment, physiotherapy, and hydrodilation.[4,15–18] Many investigators who have reported on nonoperative treatment of subacromial impingement syndrome have documented successful results, especially when patients were managed in the early stages of the disease.[19–21] However, systematic reviews have revealed inconclusive evidence for the real efficacy of such therapies.[3,5,6]
In a retrospective study assessing the outcome of conservative treatment on 616 patients,[22] a favorable outcome was observed for patients between 41 and 60 years old, whereas Bartolozzi and associates[19] in another retrospective study found no difference in outcome among the different age groups. We also did not find any remarkable correlation between age and patients' outcome. In another prospective study,[21] type I (flat) acromion was associated with better outcome with conservative management, whereas patients with type III (hooked) acromion had the worst response. Our findings are in accordance with this study. In one study,[19] the duration of symptoms for more than 1 year was found to be related with unfavorable outcome, whereas in another study,[23] a cut-off of 6 months for the duration of burden showed no significant difference between outcomes of conservative treatment in the 2 groups. Our study, however, showed a significant predictive value for this variable.
The prospective design of this study allowed us to minimize several potential biases and apply a unique diagnostic approach and therapeutic program to all patients. However, our study had some limitations. First, we did not evaluate the correlation between the extension of rotator cuff pathology and patient outcome because we didn't perform MRI or ultrasonography tests for all participants. Rotator cuff pathology has been found to have a predictive value on final outcome,[19] although in practical settings, it is not feasible to perform these tests on all patients, particularly those with good initial conditions. Second, although the Constant scoring system is widely used and its method is clearly described,[13] it has been criticized for its relatively low reliability and for being highly dependent on range of motion and muscle strength, while ignoring important factors, such as patient quality of life.[24,25]
This study did not attempt to compare the results of conservative treatment with other treatment modalities. Furthermore, considering current opinion on the usefulness of conservative therapies as the first line of treatment[4,16,18] and regarding ethical issues on performing invasive procedures on this population, we did not recruit a matched control group treated with surgical procedures; therefore, these results cannot be weighed against surgical-intervention reports.
Conservative therapy could be a valuable option in patients with subacromial impingement syndrome. Severity of pretreatment conditions does not necessitate a more invasive therapy. Predictive power of initial functional impairment could be refined if one considers the duration of symptoms and acromial morphology. In patients with such poor prognostic factors, it seems better to choose aggressive therapeutic methods as the first-line intervention. However, further investigations with larger sample sizes, longer follow-up periods, and with the use of modern, validated-function assessment tools are needed to develop a precise decision guideline for choosing the appropriate treatment method for patients.
Acknowledgments
The authors thank Dr. Mohammad Reza Guity (Assistant Professor of Orthopedics, Tehran University of Medical Sciences, Tehran, Iran) for his considerable assistance in the design and supervision of this study.
Contributor Information
Afshin TaheriAzam, Department of Orthopedic Surgery, Khatam Hospital, Zahedan University of Medical Sciences, Zahedan, Iran.
Mohsen Sadatsafavi, Research Development Center, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran.
Alireza Moayyeri, Research Development Center, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran.
References
- 1.Neer CS. Anterior acromioplasty for the chronic impingement syndrome in the shoulder: a preliminary report. J Bone Joint Surg Am. 1972;54:41–50. [PubMed] [Google Scholar]
- 2.Bolukbasi S, Guzel VB, Simsek A, et al. Modified and classic acromioplasty for impingement of the shoulder. Int Orthop. 2002;26:10–12. doi: 10.1007/s00264-001-0300-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.van der Heijden GJ, van der Windt DA, Kleijnen J, et al. Steroid injections for shoulder disorders: a systematic review of randomized clinical trials. Br J Gen Pract. 1996;46:309–316. [PMC free article] [PubMed] [Google Scholar]
- 4.Plafki C, Steffen R, Willburger RE, et al. Local anaesthetic injection with and without corticosteroids for subacromial impingement syndrome. Int Orthop. 2000;24:40–42. doi: 10.1007/s002640050010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.van der Heijden GJ, van der Windt DA, de Winter AF. Physiotherapy for patients with soft tissue shoulder disorders: a systematic review of randomized clinical trials. BMJ. 1997;315:25–30. doi: 10.1136/bmj.315.7099.25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Green S, Buchbinder R, Glazier R, et al. Systematic review of randomized controlled trials of interventions for painful shoulder: selection criteria, outcome assessment, and efficacy. BMJ. 1998;316:354–360. doi: 10.1136/bmj.316.7128.354. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ellman H. Arthroscopic subacromial decompression: analysis of one- to three-year results. Arthroscopy. 1987;3:173–181. doi: 10.1016/s0749-8063(87)80061-0. [DOI] [PubMed] [Google Scholar]
- 8.Mantone JK, Burkhead WZ, Noonan J. Nonoperative treatment of rotator cuff tears. Orthop Clin North Am. 2000;31:295–311. doi: 10.1016/s0030-5898(05)70149-8. [DOI] [PubMed] [Google Scholar]
- 9.Speer KP, Lohnes J, Garrett WE. Arthroscopic subacromial decompression: results in advanced impingement syndrome. Arthroscopy. 1991;7:291–296. doi: 10.1016/0749-8063(91)90130-p. [DOI] [PubMed] [Google Scholar]
- 10.Wirth MA, Basamania C, Rockwood CA. Nonoperative management of full-thickness tears of the rotator cuff. Orthop Clin North Am. 1997;28:59–67. doi: 10.1016/s0030-5898(05)70264-9. [DOI] [PubMed] [Google Scholar]
- 11.Itoi E, Tabata S. Incomplete rotator cuff tears. Results of operative treatment. Clin Orthop. 1992;284:128–135. [PubMed] [Google Scholar]
- 12.Hawkins RJ, Abrams JS. Impingement syndrome in the absence of rotator cuff tear. Orthop Clin North Am. 1987;18:373–382. [PubMed] [Google Scholar]
- 13.Constant CR, Murley AH. A clinical method of functional assessment of the shoulder. Clin Orthop. 1987;214:160–164. [PubMed] [Google Scholar]
- 14.Dinnes J, Loveman E, McIntyre L, et al. The effectiveness of diagnostic tests for the assessment of shoulder pain due to soft tissue disorders: a systematic review. Health Technol Assess. 2003;7:iii, 1–166. doi: 10.3310/hta7290. [DOI] [PubMed] [Google Scholar]
- 15.Berry H, Fernandes L, Bloom B, et al. Clinical study comparing acupuncture, physiotherapy, injection and oral anti-inflammatory therapy in shoulder-cuff lesions. Curr Med Res Opin. 1980;7:121–126. doi: 10.1185/03007998009112038. [DOI] [PubMed] [Google Scholar]
- 16.Ellman H. Diagnosis and treatment of incomplete rotator cuff tears. Clin Orthop. 1990;254:64–74. [PubMed] [Google Scholar]
- 17.Neer CS, II, Welsh RP. The shoulder in sports. Orthop Clin North Am. 1977;8:583–591. [PubMed] [Google Scholar]
- 18.Scheib JS. Diagnosis and rehabilitation of the shoulder impingement syndrome in the overhand and throwing athlete. Rheum Dis Clin North Am. 1990;16:971–988. [PubMed] [Google Scholar]
- 19.Bartolozzi A, Andreychik D, Ahmad S. Determinants of outcome in the treatment of rotator cuff disease. Clin Orthop. 1994;308:90–97. [PubMed] [Google Scholar]
- 20.Itoi E, Tabata S. Conservative treatment of rotator cuff tears. Clin Orthop. 1992;275:165–173. [PubMed] [Google Scholar]
- 21.Wang JC, Horner G, Brown ED, et al. The relationship between acromial morphology and conservative treatment of patients with impingement syndrome. Orthopedics. 2000;23:557–559. doi: 10.3928/0147-7447-20000601-12. [DOI] [PubMed] [Google Scholar]
- 22.Morrison DS, Frogameni AD, Woodworth P. Non-operative treatment of subacromial impingement syndrome. J Bone Joint Surg. 1997;79:1–5. doi: 10.2106/00004623-199705000-00013. [DOI] [PubMed] [Google Scholar]
- 23.Bokor DJ, Hawkins RJ, Huckell GH, et al. Results of nonoperative management of full-thickness tears of the rotator cuff. Clin Orthop. 1993;294:103–110. [PubMed] [Google Scholar]
- 24.Conboy VB, Morris RW, Kiss J, et al. An evaluation of the Constant-Murley Shoulder Assessment. J Bone joint Surg Br. 1996;78:229–232. [PubMed] [Google Scholar]
- 25.Kirkley A, Griffin S, Dainty K. Scoring systems for the functional assessment of the shoulder. Arthroscopy. 2003;19:1109–1120. doi: 10.1016/j.arthro.2003.10.030. [DOI] [PubMed] [Google Scholar]