Evaluation of eccentric exercise in treatment of patellar tendinitis.

Evaluation of Eccentric Exercise in Treatment of Patellar Tendinitis Patellar Tendinitis (jumper's knee) is an insertional tendinopathy most commonly affecting the patellar tendon's origin on the inferior pole of the patella. [1-4] The clinical manifestations of patellar tendinitis are characterized by an aching pain localized to the bone-tendon interface without swelling, locking, or giving way. [5] Definitive diagnosis of patellar tendinitis is based on the presence of local tenderness with palpation of the inferior (or superior) patellar pole. Radiographic examination may show radiolucency or elongation of the involved pole. [2,5] In addition, histological evidence has shown the presence of pseudocystic cavities at the border line between mineralized fibrocartilage and bone. [2,6] The occurrence of this finding varies, [2] but it has been suggested that tendon ruptures may occur as a result of repetitive microtrauma to the tendon-bone interface. [7]

The etiology of patellar tendinitis is attributed to an overuse of the musculoskeletal unit. Blazina et al coined the term "jumper's knee" to reflect what was thought to be the primary factor causing this injury. [5] Subsequently, it was found that athletes involved in activities requiring repetitive jumping, kicking, or running were most often affected. Individuals beginning these types of sport activities for the first time have also been diagnosed with patellar tendinitis. [8] In a review of 407 volleyball players, Ferretti found 23% had a history of patellar tendinitis, with a similar incidence for male and female athletes. [9] In addition, surgical exploration showed no derangements of the articular cartilage. [3] Malalignment of the patellofemoral articulation is generally considered as a separate and distinct diagnostic category and is not necessarily a significant cause of symptoms in patellar tendinitis. [3,8,10]

Surgical treatment for patellar tendinitis has had marginal success. [5,6] Several authors have described conservative treatment approaches for patellar tendinitis including muscle stretching exercises, rest, heat, and ice. [2,5,11-15] In general, the success of conservative treatment has been acceptable during the early stages of the tendinitis and poor for the later stages. [1,2,5]

The use of eccentric exercise in the treatment of patellar tendinitis has been advocated but has not been tested systematically. [12-14] The high tension levels developed during eccentric contraction of the quadriceps femoris muscle have been implicated as a possible factor in the microscopic failure of the patellar tendon fibers. Curwin and Stanish have described a conservative treatment approach that emphasizes the use of gradually incremented eccentric exercises to strengthen the quadriceps femoris muscle-patellar tendon unit. [12] They have suggested that such a rehabilitation program would closely replicate and better prepare the patellar tendon for the stresses it must withstand during sport activities.

Our study was designed to assess the effectiveness of a supplemental eccentric exercise program in the treatment of patellar tendinitis. The effect of this treatment approach was determined by measuring changes in the quadriceps femos muscle's ability to generate work during eccentric contractions. This study addressed the following question with respect to patellar tendinitis: Do patients receiving supplemental eccentric training show differences in quadriceps femoris muscle work capacity compared with patients and healthy subjects participating in only a muscle stretching exercise program?

Method

Subjects

A total of 31 subjects (15 female, 16 male; aged 21-45 years) completed the eight-week study. Fifteen subjects had a diagnosis of patellar tendinitis, and 16 subjects were "normal" (healthy) controls. The subjects with tendinitis were recreational athletes referred to the study by orthopedic surgeons at the University of Wisconsin Hospital and Clinics Sports Medicine Center. Informed consent was obtained from all of the participants. Demographic characteristics of the subjects are shown in Table 1.

For those subjects with patellar tendinitis, the average length of time of involvement was 19 months, with a range from 3 months to 4 years. Diagnosis of patellar tendinits was made by an orthopedic surgeon. The following criteria were used for inclusion in this study:

1. Subject's reports of tenderness with palpation of the inferior pole of the patella.

2. No history of trauma to the knee.

3. Onset of symptoms a minimum of six weeks prior to participating in the study.

4. No anti-inflammatory oral medication or injection for at least two weeks prior to starting the study.

5. No other current knee or lower extremity problems including chondromalacia, muscle strains, and hip or ankle injuries.

Procedure

The subjects were randomly divided into one of four groups: patellar tendinitis-home exercise only (T-H) (n=7), patellar tendinitis-additional eccentric exercise (T-A) (n=8), normal-home exercise only (N-H) (n=11), and normal-additional eccentric exercise (N-A) (n=5). Subjects in all groups were instructed in two quadriceps femoris and two hamstring muscle stretching exercises that served as the home exercise program (Figure). The stretching exercises were performed two times per day, seven days per week, on the involved side only (or right side for normal subjects). Compliance was confirmed through a written log completed by each subject.

Subjects in the T-A and N-A Groups also made three visits per week to the physical theraphy department to participate in an eccentric quadriceps femoris muscle strengthening program using the KIN/COM [TM] dynamometer. (*1) Subjects initially performed isokinetic eccentric quadriceps femoris muscle contractions with the involved (or right) knee at a speed of 30 [degrees]/sec. The intensity of the exercise was progressed over eight weeks by gradually increasing the speed of the eccentric contraction to 70 [degrees]/sec. The angular velocity progression is described in Table 2. This approach was chosen because the force-velocity curve for eccentric muscle contractions shows muscle tension increases as contraction velocity increases. n16 This method of exercise progression has been advocated by Curwin and colleagues [12,13,17] and by Renstrom and Johnson. [14]

All subjects were given a bilateral quadriceps femoris muscle strength test using the KIN/COM [TM] at the initial visit (Test 1) and at four weeks (Test 2) and eight weeks (Test 3) after starting the study. Testing was performed in the velocity mode (isokinetic resistance) with a minimal force threshold of 20 N. We used the KIN/COM [TM] software program, (*1) version 2.16, for data collection. Prior to each strength test, the subjects performed 10 submaximal concentric-eccentric knee extensions (50 [degrees]/sec isokinetic speed) to familiarize them with the testing setup. Each subject then performed five maximal isokinetic concentric-eccentric contractions of the quadriceps femoris muscle at a speed of 50 [degrees]/sec. The arc of movement was from 90 degrees to 0 degrees of knee flexion. Eccentric work values for the five trials were computed and summed by the KIN/COM [TM] software program. This procedure gave each subject on eccentric work measurement per test.

All subjects with patellar tendinitis were asked to evaluate their knee pain at the beginning and end of the eight-week program. Two pain scales were used. The first scale assessed the subjective intensity level of the pain (Appendix 1). The second scale measured the occurrence of pain during activity (Appendix 2). The latter scale is a modified version of the pain ratings proposed by Blazina et al [5] and Curwin and Stanish. [12]

Reliability of Repeated Measures

Tredinnick and Duncan tested the reliability of peak torque and work measurements obtained from the KIN/COM [TM] during both eccentric and concentric contractions. [18] The results of this study may not be generalizable to our investigation because Tredinnick and Duncan [18] used healthy subjects (rather than patients with patellar tendinitis) and their analysis averaged data from three trials as compared with our summation of five repetitions.

Amore critical index of reliability is found with the performance of the hardware. Assuming a uniform manufacturing process, identical protocols should be comparable on different KIN/COM [TM] dynamometers. Farrell and Richards showed that the KIN/COM [TM] is reliable in both static and dynamic modes for measuring joint position and force production and for controlling lever arm speed. [19] The reliability of this system, therefore, has been verified.

Reliability was not evaluated separately for each group and for each condition. In our study, an intraclass correlation coefficient (ICC[3,1]) provided a conservative estimate of overall reliability of repeated measurements obtained on the KIN/COM [TM]. [20] The ICC was calculated from the mean squares of the diagnosis factor and the error term (Tab. 4) and was significantly different from zero (ICC = .37, p [is less than] .05). A significant ICC in this case is an indication of the reliability of the testing protocol.

Research Design

A 2 X 2 factorial design with three repeated measures was used to analyze quadriceps femoris muscle eccentric work ratios. The first and second factors each had two levels: 1) diagnosis (normal vs patellar tendinitis) and 2) method (home program vs additional eccentric training).

Data Analysis

An analysis of variance for repeated measures was used to determine the statistical main effects and interactions at each level of the research design. Eccentric quadriceps femoris muscle work was expressed as a percentage with respect to body side (right-left for normal subjects or involved-uninvolved for subjects with patellar tendinities). This normalization procedure was necessary to provide a basis of comparison between normal subjects and a patient population with a unilateral pathological condition. A significant F-ratio test result was studied further using Tukey's post hoc analysis (studentized range test). Because group size was not uniform, a harmonic N was calculated (N = 7.16) to serve as the divisor for the error term. All post hoc comparisons were made at the .05 level of significance and were corrected for all possible pair-wise comparisons. Ratings on each pain scale were correlated with quadriceps femoris muscle work ratios using a Spearman rank-order correlation coefficient because of the ordinal nature of the pain scales.

Results

MEans and standard deviations for the ratios of quadriceps femoris muscle work are displayed in Table 3. Normal subjects demonstrated significantly larger ratios for quadriceps femoris muscle eccentric work compared with subjects with patellar tendinitis (Tab. 4). Post hoc and descriptive analyses were organized as follows:

1. Comparison of the two groups receiving additional eccentric training (N-A and T-A Groups).

2. Comparison of the difference between mean eccentric work ratios in Test 1 with Test 3 for each group.

3. Comparison of the home muscle stretching program with the additional eccentric exercise program (N-H vs N-A Groups and T-H vs T-A Groups).

4. Comparison of the two groups participating in the home muscle stretching program only (N-H vs T-H Groups).

The analysis comparing additional eccentric training for normal subjects and for subjects with patellar tendinitis reveals that the largest significant difference for eccentric ratios occurred during the last test (Test 3). Mean quadriceps femoris muscle eccentric work ratio was 140% for the N-A Group but only 106% for the T-A Group (Tab. 3). This difference means that the trained limb for normal subjects produced a greater work rating with respect to the untrained limb and that this relative increase in strength was not matched by subjects with patellar tendinitis.

Some important trends, however, are found when work capacity was analyzed to compare the beginning and end of the eight-week training session. The normal subjects' trained limb showed a 23% improvement with additional eccentric exercise. The involved (trained) limb in the T-A Group also demonstrated increased work capacity (mean work ratios were 78% and 106% for Tests 1 and 3, respectively). This change in relative work capacity represented a 26% increase for the patient groups over the eight-week training course. Changes in mean eccentric work ratios for the normal and patient groups participating in only the muscle stretching exercise program did not show increases of this magnitude. The N-H Group had a decrease in the mean work ratio of 4%, and the T-H Group showed only a 13% increase between Test 1 and Test 3 (Tab. 3).

Quadriceps femoris muscle eccentric work capacity in normal subjects receiving only the home muscle stretching protocol did not differ statistically from the normal subjects with additional eccentric training. Although no statiscally significant differences were found between the patients participating in only the home muscle stretching program and the patients receiving additional eccentric training, the T-A Group showed greater improvement with involved limb work capacity, as noted previously.

The home muscle stretching program produced only one significant difference between the normal subjects and the subjects with patellar tendinitis (Test 1, Tab. 3). All other comparisons between the N-H and T-H Groups were not statistically significant. Correlations between the final pain ratings and quadriceps femoris muscle work ratios for Test 3 were determined using Spearman rank-order correlation coefficients. We have focused on pain ratings recorded at the end of the eight-week training session. The negative relationship of eccentric work ratios to pain frequency (p = -.80) and pain intensity (p = -.78) indicates that as pain increased in the T-A Group, the work ratios decreased. The T-A Group's pain questionnaire average frequency rating was 1.68 [+ or -] 0.86, and the average intensity rating was 0.88 [+ or -] 0.60.

Discussion

Conservative management of patellar tendinitis must be carefully planned because physical therapy is usually attempted before any surgical procedure. The use of eccentric exercise training has been advocated as a primary feature of tendinitis rehabilitation. [12-14,17] The purpose of our study was to determine the effects of an eight-week eccentric exercise program on quadriceps femoris muscle work capacity in a group of patients with patellar tendinitis compared with three groups of control subjects. We found that additional eccentric training had a significant effect on increasing the strength difference between right and left extremities in normal subjects. We also found that subjects with patellar tendinitis increased their quadriceps femoris muscle eccentric work capacity in the trained (involved) limb, but this trend was not statistically significant.

There are several possible explanations for the limited gains in quadriceps femoris muscle eccentric work capacity in the T-A and T-H Groups. Pain and discomfort in the knee were like to account for some of the difference between the N-A and T-A Groups. Loading the quadriceps femoris tendon eccentrically produces tension levels exceeding concentric or maximal isometric contractions. [16] A pain or protective response could limit the levels of eccentric work when resistance is applied to the joint. The negative correlations between work ratios and pain rating show that even mild to moderate discomfort will affect muscle performance, particularly under maximum tendon-loading conditions. It should be noted that an eccentric work ratio of 100% indicated equal work capacity for both legs. Ratios less than 100% showed that the involved limb (or right limb for normal subjects) generated less work than the opposite limb. In contrast, ratios greater than 100% indicated that more work was produced from the right or involved quadriceps femoris muscle with respect to the controlled untrained limb. The last test in the series (Test 3) showed the T-A Group's mean work ratios at or near 100% (Tab. 3). The involved limb of the subjects with tendinitis slightly exceeded work output of the less involved or noninvolved side. In contrast, the N-A Group showed dramatic and statistically significant work capacity increases in the trained (right) limb. The muscle stretching protocol produced no consistent differences between the normal subject and patient groups or within groups (across method). Therefore, supplemental eccentric exercise training, as described in this article, resulted in limited improvement in quadriceps femoris muscle strength for the subjects with patellar tendinitis.

It could be argued that the composition of the sample in terms of the proportion of male and female subjects may have skewed the results. The N-A Group, however, showed the largest gains in quadriceps femoris muscle eccentric work. The N-A Group was exclusively female (Tab. 1), and we would expect normative strength values for this group to be lower because of a relatively smaller mass of the quadriceps femoris muscle in women than in men. [21] Sex bias, therefore, was not a prominent confounding variable.

The stage or severity of tendinitis is largely determined by the report of pain associated with activity. [1,2,5,12,13,17] During the early stages, the pain is reported to be intermittent and relieved with activity. As the tendinitis worsens, the pain becomes more persistent. In extreme cases, the pain is present before, during, and after activity.

The failure of eccentric exercise to significantly increase quadriceps femoris muscle work capacity may be related to the stage of injury. Martens et al reported that conservative treatment failed in 16 out of 38 cases when the tendinitis was in an advanced stage. [2] In contrast, surgical procedures yielded favorable results in a higher percentage of patients. [2,6,7] It is important to note that conservative therapy was effective during early stages of the tendinitis. [2] In our study, the small number of subjects in the T-A and T-H Groups precluded a further subgrouping based on the stage of tendinitis.

Clinical Implications

Optimizing the intensity level of eccentric exercise may improve quadriceps femoris muscle strength in patients with jumper's knee. The protocol described in this article was effective for training normal subjects but only showed a tendency for increasing work output in subjects with patellar tendinitis. Can patients with patellar tendinitis develop large increases in eccentric quadriceps femoris muscle work capacity? How dependent is the outcome of conservative therapy on the stage of tendinitis? Future research should focus on intensity variations of eccentric exercise that will facilitate gains in quadriceps femoris muscle work capacity for treating overload injuries of the knee. Studies should also be designed to evaluate the dependency of therapeutic outcomes on the stage of injury and level of discomfort during exercise.

Conclusion

Normal controls receiving eccentric exercise training showed greater increases in quadriceps femoris muscle work output (with respect to the untrained limb) compared with subjects with patellar tendinitis. The subjects with tendinitis showed a tendency for improving eccentric quadriceps femoris muscle work ratios over the eight-week training course. Knee pain in the T-H and T-A Groups was correlated with a decrease in eccentric quadriceps femoris muscle work ratios. We concluded that patients with tendon overload injuries can make gains in eccentric work capacity but that these patients may not be able to generate the eccentric loads that will result in optimal strength increases.

(*1) Chattecx Corp, 101 Memorial Dr, PO Box 4287, Chattanooga, TN 37405.

References

[1] Roels J, Martens M, Burssens A: Patellar tendinitis (jumper's knee). Am J Sports Med 6: 362-368, 1978

[2] Martens M, Wouters P, Burssens A, et al: Patellar tendinitis: Pathology and results of treatment. Acta Orthop Scand 53:445-450, 1982

[3] Ferretti A: Epidemiology of jumper's knee. Sports Med 3:289-295, 1986

[4] Ferretti A, Puddu G, Mariani PP, et al: Jumper's knee: An epidemiological study of volleyball players. The Physician and Sportmedicine 12(10):97-106, 1984

[5] Blazina ME, Kerlan RK, Jobe FW, et al: Jumper's knee. Orthop Clin North Am 4:665-678, 1973

[6] Ferretti A, Ippolito E, Mariani P, et al: Jumper's knee. Am J Sports Med 11:58-62, 1983

[7] Kelly DW, Carter VS, Jobe FW, et al: Patellar and quadriceps tendon ruptures: Jumper's knee. Am J Sports Med 12:375-379, 1984

[8] Hunter HC: Patellofemoral arthralgia. Journal of the American Osteopathic Association 85:580-585, 1985

[9] Ferretti A, referred to in Ferretti A, Puddu G, Mariani PP, et al: The natural history of jumper's knee. Int Orthop 8:239-242, 1985

[10] Percy EC, Strother RT: Patellalgia. The Physician and Sportsmedicine 13(7):43-59, 1985

[11] Black JE, Alten SR: How I manage infrapatellar tendinitis. The Physician and Sportsmedicine 12(10):86-92, 1984

[12] Curwin S, Stanish WD (eds): Tendinitis: Its Etiology and Treatment. Toronto, Ontario, Canada, D C Health Canada Ltd, 1984

[13] Stanish WD, Curwin S, Rubinovich MR: Tendinitis: The analysis and treatment for running. Clin Sports Med 4:593-609, 1985

[14] Renstrom P, Johnson RJ: Overuse injuries in sports. Sports Med 2:316-333, 1985

[15] Ferretti A, Puddu G, Mariani PP, et al: The natural history of jumper's knee. Int Orthop 8:239-242, 1985

[16] Komi PV: Measurement of the force-velocity relationship in human muscle under concentric and eccentric contractions. In Cerquiglini S, et al: Proceedings of Third International Seminar on Biomechanics, Rome, Italy. Basel, Switzerland, S Karger AG, Medical and Scientific Publishers, 1973, pp 224-229

[17] Stanish WD, Rubinovich MR, Curwin S: Eccentric exercise in chronic tendinitis. Clin Orthop 208:65-68, 1986

[18] Tredinnick TJ, Duncan PW: Reliability of measurements of concentric eccentric isokinetic loading. Phys Ther 68:656-659, 1988

[19] Farrell M, Richards JG: Analysis of the reliability and validity of the kinetic communication exercise device. Med Sci Sports Exerc 18:44-49, 1986

[20] Shrout PE, Fleiss JL: Intraclass correlations: Uses in assessing rater reliability. Psychol Bull 86:420-428, 1979

[21] Wells C: Women, Sports and Performance: A Physiological Perspective. Champaign, IL, Human Kinetics Publishers Inc, 1985, p 35

K Jensen, BS, PT, is Supervisor, Sports Medicine Physical Theraphy Unit, University of Wisconsin Hospital and Clinics, Center for Health Sciences, 600 Highland Ave, Madison, WI 53792.

R Di Fabio, PhD, PT, is Director of Physical Theraphy, University of Wisconsin Hospital and Clinics, Center for Health Sciences, 600 Highland Ave, Madison, WI 53792 (USA).

This article was submitted January 27, 1988; was with the authors for revision for 22 weeks; and was accepted October 26, 1988.