JBJS | The Efficacy of an Injection of Steroids for Medial Epicondylitis. A Prospective Study of Sixty Elbows*

The Journal of Bone & Joint Surgery, Volume 79, Issue 11

Articles | November 01, 1997

The Efficacy of an Injection of Steroids for Medial Epicondylitis. A Prospective Study of Sixty Elbows*

SHALOM STAHL, M.D.†; TEDDY KAUFMAN, M.D.†, HAIFA, ISRAEL

Investigation performed at the Hand Surgery Unit and the Department of Plastic and Reconstructive Surgery, Rambam Medical Center, Haifa

The Journal of Bone & Joint Surgery. 1997; 79:1648-52

5 Recommendations (Recommend) | 3 Comments | Saved by 3 Users Save Case

text A A A

Abstract

We report the details of a prospective, randomized, double-blind study that was undertaken to analyze the short-term and long-term effects of the local injection of methylprednisolone to treat medial epicondylitis. Fifty-eight patients (sixty elbows) were assigned to receive a single injection of 1 per cent lidocaine with either forty milligrams of methylprednisolone (experimental group) or saline solution (control group); both groups were also managed with physical therapy and the use of non-steroidal anti-inflammatory drugs. The two groups were not significantly different with regard to the age and gender of the patients, the duration of the symptoms, the degree of pain before the injection, or the number of dominant upper limbs involved. Six weeks after the injection, the experimental group had significantly less pain than the control group (p < 0.03), as determined with a modification of the grading system of Nirschl and Pettrone. However, the groups did not differ with regard to pain at three months and at one year. The intensity of pain, as measured on a visual-analog scale, did not differ between the two groups six weeks and one year after the injection. We believe that the improvement observed in both groups primarily reflects the natural history of the disorder, and we conclude that the local injection of steroids provides only short-term benefits in the treatment of medial epicondylitis.

Figures in this Article

Introduction

Introduction | Materials and Methods | Results | Discussion | References

Medial epicondylitis is a clinical entity characterized by pain in the medial aspect of the elbow. The accumulated pathological evidence suggests that the process is associated with fibrillary degeneration of collagen and angiofibroblastic hyperplasia at the origin of the flexor-pronator muscle mass²¹, microfragmentation or tears of the tendon^2,25, accumulation of vascular granulation tissue^19,25, and tendinous necrosis²⁵, all of which are also accompanied by a secondary inflammatory reaction.

The use of non-operative measures^{8,10,11,14,20}, consisting of various modalities such as rest, physical therapy, and the administration of non-steroidal anti-inflammatory medications, is an accepted approach to the treatment of medial epicondylitis^2,18,20; however, the use of a local injection of steroids is controversial. Labelle et al. stated that the treatment of epicondylitis is "based on either the personal experience of the different authors, or on a few even earlier reports rather than on scientific evidence." It has been suggested that injection of methylprednisolone is the treatment of choice for epicondylitis, as thirty-three of thirty-six patients who had had such treatment had faster relief of symptoms than patients who had been managed with an injection of lidocaine or saline solution⁹. In another study, 278 of 290 patients had a satisfactory result after local injections of hydrocortisone acetate⁸, and similar results were found in a study in which patients were managed with injections of methylprednisolone or hydrocortisone⁷. Some authors have advocated that cortisone injections should be used cautiously as multiple intratendinous injections have been reported to result in atrophy or actual dissolution of the tendon²⁰. However, when cortisone is injected below the origin of the muscle, immediate and total relief of pain has been observed²⁰. Furthermore, it has been shown that patients who were managed operatively had previously had an unsatisfactory response to multiple injections^12,20,25. In contrast, previous studies have suggested that the result after injection of steroids is not superior to that after oral administration of indomethacin¹⁴ or local application of steroid hypospray¹³.

According to the system of Chalmers et al. for the evaluation of the quality of randomized, controlled trials, only eighteen of 185 studies on tennis elbow were found to include adequately selected test groups¹⁵. Only five of these studies involved treatment with steroids^4,7,9,13,14, and the scores for the quality of those trials (expressed as a percentage of the maximum possible score) ranged from 6 to 26 per cent. We were unable to find any reports in which the system of Chalmers et al. had been used to evaluate studies of the treatment of medial epicondylitis. Although temporary relief of pain after the injection of steroids has been noted by several investigators^{9,10,12,17,19,21,23,25,26}, the long-term benefits of this treatment for medial epicondylitis are doubtful and complications have been reported^1-3,24. The present prospective, randomized, double-blind study was undertaken to analyze the short-term and long-term effects of the local injection of steroids to treat medial epicondylitis.

*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.

†Hand Surgery Unit (S. S.) and Department of Plastic and Reconstructive Surgery (T. K.), Rambam Medical Center, P.O.B. 9602, Haifa 31096, Israel.

*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.

†Hand Surgery Unit (S. S.) and Department of Plastic and Reconstructive Surgery (T. K.), Rambam Medical Center, P.O.B. 9602, Haifa 31096, Israel.

View Larger

TABLE I CHARACTERISTICS OF THE PATIENTS

*NS = not significant.†The values are given as the mean and the standard error of the mean.

	Experimental Group	Control Group	Difference between Groups*
Total no. of elbows	30	30	NS
In male patients	24	22	NS
In female patients	6	8	NS
No. of dominant limbs involved	24	26	NS
Duration of symptoms†(mos.)	4.33 ± 0.25	4.72 ± 0.25	NS

Add to My JBJS

TABLE I CHARACTERISTICS OF THE PATIENTS

*NS = not significant.†The values are given as the mean and the standard error of the mean.

	Experimental Group	Control Group	Difference between Groups*
Total no. of elbows	30	30	NS
In male patients	24	22	NS
In female patients	6	8	NS
No. of dominant limbs involved	24	26	NS
Duration of symptoms†(mos.)	4.33 ± 0.25	4.72 ± 0.25	NS

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1 Bar graph showing the mean pain-phase score. The score was significantly lower (p < 0.03) in the experimental group (steroid) than in the control group (non-steroid) six weeks after the injection. No significant differences were detected between the groups at the other time-points studied.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2 Bar graph showing the mean scores on the visual-analog scale for the intensity of pain. The scores were similar in the two groups before the injection (p > 0.5) and six weeks (p > 0.1) and one year (p > 0.9) after the injection.

Materials and Methods

Introduction | Materials and Methods | Results | Discussion | References

Patients: Fifty-eight patients (sixty elbows) who had medial epicondylitis were assigned to the present study between December 1989 and July 1994. Twenty-seven patients had a white-collar job, fourteen had a job that seldom involved the performance of strenuous duties, and seventeen worked as manual laborers. Eleven patients were non-professional athletes. As our purpose was to evaluate patients who had isolated medial epicondylitis, we excluded patients who had associated conditions involving the upper limb, such as an injury of the elbow, carpal tunnel syndrome, lateral epicondylitis, or ulnar neuropathy.

Clinical evaluation: Physical examination included palpation to localize the tenderness around the elbow, measurement of the range of motion, determination of the stability of the elbow joint, and application of resistance to flexion of the wrist and to pronation of the forearm to elicit pain. Radiographs were made for all patients. Medial epicondylitis was clinically diagnosed according to the following criteria: pain in the medial epicondyle during daily activities at work or participation in sports, tenderness over the origin of the flexor-pronator muscle mass, tenderness over the medial epicondyle of the humerus, and increased pain with pronation of the forearm and flexion of the wrist against resistance. In patients in whom ulnar neuropathy was suspected, the clinical evaluation also included the elbow flexion test, testing for the Tinel sign, measurement of two-point discrimination in the fingers, and nerve-conduction studies.

Assessment of pain phase²¹ We used a modification of the grading system of Nirschl and Pettrone to evaluate the pain phase²¹ before the injection and at six weeks, three months, and one year after the injection. According to our modification of the pain-phase scale, a score of 0 points is given if the patient has full activity and no pain; a score of 1 point, if the patient has no pain during normal daily activities and mild pain during sports or occupational activity; a score of 2 points, if the patient has occasional pain during normal daily activities and moderate pain during sports or occupational activity; a score of 3 points, if the patient has mild-to-moderate pain during normal daily activities and severe pain during sports or occupational activity; and a score of 4 points, if the patient has pain at rest. Our modification reflects our intention to differentiate between pain evoked during normal daily activities and that elicited by strenuous activities, which is difficult to do with the system of Nirschl and Pettrone. Scores of 0 and 4 points on the pain-phase scale are compatible with the grades of excellent and failure according to the system of Nirschl and Pettrone. We modified the system and added another score to facilitate the differentiation among varying intensities of pain induced during normal and strenuous activities. We believe that our modified pain-phase scale allows a more accurate and informative evaluation of the patient.

Assessment of intensity of pain: Pain is considered to be the major indication for operative treatment whenever non-operative treatment has failed¹⁸. Hence, the evaluation of the intensity of pain was the only additional criterion used to estimate the beneficial effects of the treatment modalities in the present study. The patients were asked to use a ten-centimeter visual-analog scale to indicate the most severe pain that they had ever had during activity or rest, with zero indicating no pain and ten centimeters representing the maximum intensity. A numerical value (1 to 10 points) was given to correspond with the mark that the patient had made on the scale. All patients evaluated pain with this scale before the injection and six weeks and one year afterward.

Treatment groups: All fifty-eight patients (sixty elbows) were treated according to a standard protocol consisting of the use of non-steroidal anti-inflammatory medications, the elimination of activities that caused pain, and physical therapy. A number of different non-steroidal anti-inflammatory drugs were used, but all patients received an equivalent amount of medication. Only patients who had not previously been managed with injections of steroids were included in the study. The additional treatment for each patient was determined by the draw of a card in a sealed envelope. The study was designed to include sixty elbows; therefore, there were sixty sealed envelopes that each contained an assignment card for either the experimental or the control group. If a patient was removed from the study for any reason, his or her assignment card was placed in another envelope for reallocation to another patient. After informed consent had been obtained, the patients were assigned to groups according to the predetermined randomized schedule. The experimental group consisted of thirty elbows that were treated with one injection of forty milligrams of methylprednisolone (one milliliter) with one milliliter of 1 per cent lidocaine. The control group consisted of thirty elbows that were treated with one injection of one milliliter of 1 per cent lidocaine with one milliliter of saline solution (0.9 per cent NaCl). Thus, the total volume of solution that was injected was identical for the two groups, eliminating differences due to distention of tissues. The solutions to be injected were prepared and coded by a member of the medical staff who was not involved in the study. The injection was given, after the sealed envelope had been drawn and the coded card had been removed, by another member, who was unaware of the solution to be given. Similarly, none of the patients knew which solution had been injected. The pain phase²¹ and the intensity of the pain were evaluated and recorded by one of us (S. S.), who also was blinded with regard to treatment. Only after the results had been assessed were the treatment groups uncoded and the data correlated with the respective treatments.

Assessment of complications of local injection of steroids: Follow-up consisted of interviews and physical examination for possible local complications, such as fat atrophy, depigmentation of skin, and disruption of the muscle origin², as well as for post-injection flare, facial flushing, and iatrogenic infection³.

Statistical analysis: The scores on the pain-phase scale and those on the visual-analog scale for intensity of pain were expressed as the mean (and standard error of the mean), median, and frequency. The t test was used to evaluate the differences between the mean values for the two groups, and the Mann-Whitney test was used to evaluate the differences between the median values. Linear multiple-regression analysis was used to describe the results for the two groups as a function of time. Values of p < 0.05 and of r² > 0.7 were considered to be significant.

Results

Introduction | Materials and Methods | Results | Discussion | References

None of the patients dropped out of the study during the one-year period of follow-up. The mean age (and standard error of the mean) was 43 ± 1.22 years for the patients in the experimental group (injection of the steroid) and 41.3 ± 1.48 years for the patients in the control group; this difference was not significant. The two groups also were similar with regard to variables such as gender, the duration of symptoms, and the number of dominant limbs involved (Table I). The range of motion of all of the elbows except one was found to be normal; the exceptional elbow had a flexion contracture of 7 degrees. The radiographic findings included calcifications or osteophytes in the region of the medial epicondyle in three elbows in the experimental group and in one elbow in the control group. Because of the small number of non-professional athletes in each group, statistical analysis could not be performed on the results for this subgroup.

Pain phase: The mean score on the pain-phase scale was similar for the two groups before the injection (2.4 ± 0.15 points for the experimental group compared with 2.3 ± 0.15 points for the control group; p > 0.05) (Fig. 1). Six weeks after the injection, the mean score was significantly lower for the experimental group than for the control group (1.2 ± 0.21 compared with 1.9 ± 0.19 points; p < 0.03). In contrast, the mean scores for the two groups were not found to differ significantly at three months (1.2 ± 0.19 compared with 1.3 ± 0.19 points; p > 0.7) or at one year (0.5 ± 0.14 compared with 0.6 ± 0.22 points; p > 0.5). With regard to the rate of improvement, the experimental group had a significantly greater decrease in the mean pain-phase score, compared with the control group, six weeks after the injection (p < 0.001); however, with the numbers available for study, we could not detect a difference at three months or at one year (p > 0.27 for both).

Intensity of pain: The mean values on the visual-analog scale were similar (p > 0.5) for the two groups before the injection (3.7 ± 0.26 for the experimental group compared with 3.5 ± 0.24 for the control group). With the numbers available for study, we could not detect a significant difference in the values at six weeks (1.5 ± 0.30 compared with 2.2 ± 0.29; p > 0.1) or at one year (0.5 ± 0.18 compared with 0.6 ± 0.17; p > 0.9) (Fig. 2). Linear multiple-regression analysis demonstrated a similar trend in both groups, and the trends did not differ between the groups (r² = 0.45).

Complications: No local complications could be attributed to the injection of methylprednisolone; however, facial flushing was detected within twenty-four hours after the injection of the steroid in one female patient.

Discussion

Introduction | Materials and Methods | Results | Discussion | References

Medial epicondylitis is characterized by pain in the medial aspect of the elbow and dysfunction induced by degenerative changes in the origin of the flexor-pronator muscle mass²; the entity also may be associated with other disorders of the elbow region^12,19,25. In the present series, medial epicondylitis was diagnosed clinically according to the typical history and the physical findings. All patients were referred to us after a relatively short duration of symptoms and were managed with common modalities of physical therapy in addition to the randomly assigned treatment. The two groups were similar in terms of the parameters before the injection; thus, the influence of biological variables was minimized.

Failure to alleviate persistent pain, the major symptom of medial epicondylitis, is the usual indication for operative treatment¹⁹. Hence, a visual-analog scale was used to quantify the intensity of pain and a modification of the system of Nirschl and Pettrone was used to determine dysfunction due to pain.

Although the hand dynamometer is a useful means with which to estimate grip strength¹⁶, the device was not employed in the present study. In our patients, grip strength would more likely reflect pain-related variations in exerted effort rather than other neuromuscular dysfunctions. In addition, the validity of measurements made with a hand dynamometer is questionable, as non-standardized verbal instructions may substantially affect the results²². The measurement of maximum grip strength is not reliable, as pain may inhibit maximum performance.

As all patients were managed with non-steroidal anti-inflammatory medications and physical therapy in addition to the injection of the steroid, the precise effectiveness of the injection in both groups might be questionable. However, all patients were managed in the same physical therapy department according to very similar protocols. In addition to these therapeutic measures, only one local injection was given in each group (one milliliter of methylprednisolone with one milliliter of 1 per cent lidocaine in the experimental group and one milliliter of 1 per cent lidocaine with one milliliter of saline solution in the control group); thus, any observed differences should be attributed only to the pharmacological effects of the steroid⁹.

At six weeks, the patients who had been managed with an injection of methylprednisolone had a significantly better pain-phase score and had had a significantly faster decrease in the intensity of pain compared with the patients in the control group. However, with the numbers available, we could detect no significant differences between the two groups at three months or at one year. We attribute the temporary improvement at six weeks to the beneficial effects of the steroid on the secondary inflammatory reaction that may be associated with medial epicondylitis. Our findings are in accordance with other observations^12,21,25. Both of our groups had improvement of the pain-phase score and a decrease in the intensity of pain during the follow-up period, with similar results at one year. We believe that this improvement primarily reflects the natural course of medial epicondylitis and the therapeutic effects of physical therapy and non-steroidal anti-inflammatory medications. Although the injection of steroids may alleviate pain temporarily, it does not appear to influence the final result.

We conclude that the local injection of a steroid has only short-term beneficial effects in the treatment of medial epicondylitis. Early diagnosis and the institution of other well known therapeutic modalities, such as rest, physical therapy, and the use of non-steroidal anti-inflammatory medications, are probably the best ways to treat this clinical entity.

NOTE: The authors thank S. Linn, M.D., Ph.D., Department of Clinical Epidemiology, Rambam Medical Center and Faculty of Medicine, Technion, Israel Institute of Technology, for analysis of the data.

References

Introduction | Materials and Methods | Results | Discussion | References

Balasubramaniam, P., and Prathap, K.: The effect of injection of hydrocortisone into rabbit calcaneal tendons. J. Bone and Joint Surg.,54-B(4): 729-734, 1972.54-B(4)729 1972

Bennett, J. B.: Lateral and medial epicondylitis. Hand Clin.,10: 157-163, 1994.10157 1994 [PubMed]

Brand, C.: Intra-articular and soft tissue injections. Australian Fam. Phys.,19: 671-675,678,680-682, 1990.19671 1990

Brattberg, G.: Acupuncture therapy for tennis elbow. Pain,16: 285-288, 1983.16285 1983 [PubMed]

Burgess, R. C.: Tennis elbow. J. Kentucky Med. Assn.,88: 349-354, 1990.88349 1990

Chalmers, T. C.; Smith, H., Jr.; Blackburn, B.; Silverman, B.; Schroeder, B.; Reitman, D.; and Ambroz, A.: A method for assessing the quality of a randomized control trial. Controlled Clin. Trials,2: 31-49, 1981.231 1981 [PubMed]

Clarke, A. K., and Woodland, J.: Comparison of two steroid preparations used to treat tennis elbow, using the hypospray. Rheumatol. and Rehab.,14: 47-49, 1975.1447 1975

Coonrad, R. W., and Hooper, W. R.: Tennis elbow: its course, natural history, conservative and surgical management. J. Bone and Joint Surg.,55-A: 1177-1182, Sept. 1973.55-A1177 1973

Day, B. H.; Govindasamy, N.; and Patnaik, R.: Corticosteroid injections in the treatment of tennis elbow. Practitioner,220: 459-462, 1978.220459 1978 [PubMed]

Dijs, H.; Mortier, G.; Driessens, M.; De Ridder, A.; Willems, J.; and De Vroey, T.: A retrospective study of the conservative treatment of tennis elbow. Acta Belgica Med. Phys.,13: 73-77, 1990.1373 1990

Ernst, E.: Conservative therapy for tennis elbow. British J. Clin. Pract.,46: 55-57, 1992.4655 1992

Gabel, G. T., and Morrey, B. F.: Operative treatment of medial epicondylitis. Influence of concomitant ulnar neuropathy at the elbow. J. Bone and Joint Surg.,77-A: 1065-1069, July 1995.77-A1065 1995

Hughes, G. R., and Currey, H. L.: Hypospray treatment of tennis elbow. Ann. Rheumat. Dis.,28: 58-62, 1969.2858 1969 [PubMed]

Kivi, P.: The etiology and conservative treatment of humeral epicondylitis. Scandinavian J. Rehab. Med.,15: 37-41, 1983.1537 1983

Labelle, H.; Guibert, R.; Joncas, J.; Newman, N.; Fallaha, M.; and Rivard, C.-H.: Lack of scientific evidence for treatment of lateral epicondylitis of the elbow. An attempted meta-analysis. J. Bone and Joint Surg.,74-B(5): 646-651, 1992.74-B(5)646 1992

Mathiowetz, V.; Weber, K.; Volland, G.; and Kashman, N.: Reliability and validity of grip and pinch strength evaluations. J. Hand Surg.,9A: 222-226, 1984.9A222 1984

Murley, A. H. G.: Tennis-elbow treated with hydrocortisone acetate. Lancet,2: 223-225, 1954.2223 1954

Nirschl, R. P.: Lateral and medial epicondylitis. In The Elbow, pp. 129-148. Edited by B. F. Morrey. New York, Raven Press, 1994.

Nirschl, R. P., and Pettrone, F. A.: Tennis elbow. The surgical treatment of lateral epicondylitis. J. Bone and Joint Surg.,61-A: 832-839, Sept. 1979.61-A832 1979

Nirschl, R. P., and Sobel, J.: Conservative treatment of tennis elbow. Phys. and Sports Med.,9: 43-54, 1981.943 1981

Ollivierre, C. O.; Nirschl, R. P.; and Pettrone, F. A.: Resection and repair for medial tennis elbow. A prospective analysis. Am. J. Sports Med.,23: 214-221, 1995.23214 1995 [PubMed]

Price, R.; Sinclair, H.; Heinrich, I.; and Gibson, T.: Local injection treatment of tennis elbow—hydrocortisone, triamcinolone and lidocaine compared. British J. Rheumatol.,30: 39-44, 1991.3039 1991

Sölveborn, S.-A.; Buch, F.; Mallmin, H.; and Adalberth, G.: Cortisone injection with anesthetic additives for radial epicondyalgia (tennis elbow). Clin. Orthop.,316: 99-105, 1995.31699 1995 [PubMed]

Unverferth, L. J., and Olix, M. L.: The effect of local steroid injections on tendon. J. Sports Med.,1: 31-37, 1973.131 1973

Vangsness, C. T., and Jobe, F. W.: Surgical treatment of medial epicondylitis. Results in 35 elbows. J. Bone and Joint Surg.,73-B(3): 409-411, 1991.73-B(3)409 1991

Verhaar, J. A. N.; Walenkamp, G. H. I. M.; van Mameren, H.; Kester, A. D. M.; and van der Linden, A. J.: Local corticosteroid injection versus Cyriax-type physiotherapy for tennis elbow. J. Bone and Joint Surg.,78-B(1): 128-132, 1996.78-B(1)128 1996

Topics

View Large | Download Slide(.ppt)
Add to My JBJS

View Larger

TABLE I CHARACTERISTICS OF THE PATIENTS

*NS = not significant.†The values are given as the mean and the standard error of the mean.

	Experimental Group	Control Group	Difference between Groups*
Total no. of elbows	30	30	NS
In male patients	24	22	NS
In female patients	6	8	NS
No. of dominant limbs involved	24	26	NS
Duration of symptoms†(mos.)	4.33 ± 0.25	4.72 ± 0.25	NS

Add to My JBJS

TABLE I CHARACTERISTICS OF THE PATIENTS

*NS = not significant.†The values are given as the mean and the standard error of the mean.

	Experimental Group	Control Group	Difference between Groups*
Total no. of elbows	30	30	NS
In male patients	24	22	NS
In female patients	6	8	NS
No. of dominant limbs involved	24	26	NS
Duration of symptoms†(mos.)	4.33 ± 0.25	4.72 ± 0.25	NS