Article Text

Download PDFPDF

What is the most appropriate treatment for patellar tendinopathy?
  1. J L Cook1,
  2. K M Khan2
  1. 1Musculoskeletal Research Centre La Trobe University, Bundoora Victoria 3086, Australia J.Cook{at}
  2. 2Department of Family Practice and School of Human Kinetics University of British Columbia, Canada kkhan{at}

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Since the demise of the concept of an inflammatory basis for tendinopathy, treatment has been based on the degenerative concept.1 However, an alternative theory of a failed healing response may be more fitting,2 and treatment that maintains or encourages the healing response in the tendon is required. Currently, treatment remains empirically based, as it is not known what interventions may best stimulate a healing response.

    The conservative treatment of patellar tendinopathy recorded in the literature includes combinations of rest,3 exercise, especially eccentric exercise,4 modalities including ultrasound, heat, and cryotherapy,5 frictions,6 biomechanical adjustment,7 and pharmaceutical treatment.8 Many of these treatments are based on “clinical experience” rather than appropriately analysed data.

    The surgical management of patellar tendinopathy includes several different operative procedures and postoperative rehabilitation protocols. The choice of surgical treatment of patellar tendinopathy appears to be based on the surgeon's preference.9

    Recent literature reviews8, 10 document a dearth of controlled trials in the treatment of patellar tendinopathy. Hence, a review of treatment of patellar tendinopathy that used a randomised allocation or were prospective in nature was undertaken.

    Ten randomised trials were reviewed. These studies investigated the effects on the patellar tendon of anti-inflammatory medication (n = 7), exercise (n = 2), and local massage (n = 1). No randomised surgical papers were identified, hence three prospective studies on the surgical treatment of this condition were included.

    Conservative treatment


    Two papers11, 12 review the effect of exercise regimens on patellar tendinopathy. These studies focus on strengthening of the muscles around the knee in subjects with jumper's knee, and measuring changes in strength, pain, and function after an eight and 12 week intervention. Outcome measures varied, but both measured changes in strength and pain; one study measured function (return to sport) as well.11

    A reduction in pain in the study period was apparent in one study, without quadriceps strength gains. Conversely, the other12 study showed improvement in quadriceps work in the study period, but it is unclear if there was a concurrent reduction in pain.


    One study13 compared treatment of patellar tendinopathy with a massage device with those treated traditionally (stretching, frictions, electrophysical modalities). The main outcome measure used in this study was a scale designed to evaluate long term changes in the patellofemoral joint,14 which may lack sensitivity to symptoms and functional changes in patellar tendinopathy. Results indicated that the main outcome measures did not differ between groups, and, although the authors concluded that the device improved clinical outcomes in patellar tendinitis, it is difficult to deduce from the data reported that the device offered a better outcome than the traditional treatment.


    Three studies investigated iontophoresis or phonophoresis using corticosteroid,15–17 two studies examined the effect of corticosteroid injection,18, 19 and two studies tested the outcome of non-steroidal anti-inflammatory drugs (one orally20 and one transcutaneously).21

    Iontophoresis with corticosteroid improved outcome in the treatment group, whereas phonophoresis with corticosteroid showed no difference in outcome when compared with phonophoresis with placebo.

    The studies that used injection indicated a good response to this intervention. The study of Capasso et al18 indicated that aprotinin offered a better outcome than either corticosteroid or placebo. The use of non-steroidal anti-inflammatory drugs for tendon pain appears to be warranted in subjects with short term symptoms.

    Surgical treatment

    There have not yet been any randomised studies of surgical treatment in patellar tendinopathy. Three studies have a clear prospective design. Khan et al22 showed that 73% of the subjects that underwent surgery for patellar tendinopathy had good results. Panni et al23 included both conservative and surgical treatment in their study. Subjects who had failed the conservative treatment were operated on, and all of these subjects were reported to have good or excellent outcome.

    Testa et al24 investigated the efficacy of percutaneous tenotomy, and reported that the technique is more effective in mid-tendon pathology than in proximal tendon pathology. This study reported that nearly 40% of subjects had poor results, and isokinetic testing showed persistent strength deficits across all outcomes.


    It is surprising that so few studies in patellar tendinopathy have a randomised or prospective design, and the few papers in this series have raised several issues in study design.

    diagnostic criteria

    Most authors used clinical assessment for diagnosis without confirmatory imaging. Pain and palpation tenderness were almost exclusively used, only two studies documented using pain on resistance of muscle contraction or other functional testing.15, 16 As tendon pain does not refer widely, the exact location of the pain would add important diagnostic accuracy to these tests.

    It can be argued that imaging is not necessary to diagnose patellar tendinopathy, as the presence of abnormal imaging does not indicate absolutely that the pain is coming from the tendon.25, 26 There is also an argument for “imaging normal” patellar tendinopathy,27 although this remains a contentious issue among tendon researchers.

    Despite these arguments, confirmation of pathology within the tendon in conjunction with clinical tests would guarantee that the best diagnostic criteria are used, and future studies should include imaging as part of the spectrum of tests needed to confirm the diagnosis.

    Palpation is used as both a diagnostic criterion and an outcome measure in this series of studies. Thus it is important that palpation is both a reliable and valid test for patellar tendinopathy. Palpation is a reliable test, but not necessarily a valid diagnostic test for this condition,28 as it does not correlate with either imaging changes or symptoms.

    Therefore, there is no specific criterion for the diagnosis of patellar tendinopathy, and the criteria used in future studies should include symptoms with objective or functional muscle testing, the pain documented on a pain map, and diagnostic imaging.

    length of symptoms

    Length of symptoms before intervention also varied widely, some studies excluding those tendons with long term symptoms (subjects with symptoms for less than five days20) whereas other studies excluded those with short term symptoms (subjects with symptoms for longer than six months23). It could be argued that these two opposing exclusion criteria are therefore investigating subjects with different conditions.

    Studies that exclude long term symptoms risk including tendons without overuse tendinopathy, and the efficacy in these studies (using anti-inflammatory intervention) may be because a true tendinopathy does not exist and because of an inflammatory component to the subjects' pain. Combined with the poorly documented and restricted diagnostic criteria, it is possible that these subjects do not have tendon mediated pain. Short term tendon pain (what this is pathologically is unknown) is not a management problem, as it appears from this series to respond well to anti-inflammatory medication. Studies of treatment efficacy should be directed at tendons that have long term symptoms, as it is these that constitute the difficult to manage group.

    study numbers

    Four studies that examined the effect of corticosteroids on tendinopathy used subjects with symptoms in several body sites. This may make the conclusions drawn about effectiveness of anti-inflammatory drugs less valid for patellar tendinopathy.

    In the remaining studies the subject numbers were relatively small: only one study had more than 100 subjects, and the remaining studies had less than 50. It is possible that these studies would have a type 2 statistical error.

    study length, treatment, and follow up

    The treatment protocol was very short in most studies (less than three weeks in six studies), except for those studies investigating exercise as an intervention (six to eight weeks). Only one study extended the follow up beyond the end of treatment,18 and evaluated subjects 12 months after treatment was completed. As patellar tendinopathy is a recurrent condition,29 these studies provide no evidence that there is any long term efficacy for any of these treatments.

    outcome measures

    Outcome measures are all subjective in patellar tendinopathy. Authors used unvalidated16 or adapted knee pain scales,12 scales not specifically designed for tendinopathy,13 or generic tendon scales30, 31 that are not necessarily appropriate or specific for the patellar tendon. The outcome scale used in both conservative and surgical papers (excellent, good, fair, poor)32 lacks specificity and sensitivity in quantifying outcome after treatment for patellar tendinopathy.

    The use of palpation as an outcome measure raises similar problems to its use as a diagnostic test for patellar tendinopathy. There is no evidence that tenderness to palpation is a valid outcome measure. The use of a gauge to standardise and quantify palpation tenderness33 may improve the validity of this outcome measure; however, normative data for tendons are required.

    concurrent treatment

    Most of the studies had either controlled11, 12 or uncontrolled16, 17 treatment or exercise during the study period that may have influenced outcome. Uncontrolled exercise was not described, as prescription was based on the individual response to treatment. In only three studies15, 20, 21 were the subjects specifically requested to refrain from other treatments and exercise.


    conservative treatment

    Studies of patellar tendinopathy have been consistently reported in the literature for 30 years. Many of these studies have suggested that it is resistant to treatment and recurrent in nature.29 Similarly, many authors suggest the need to exhaust conservative treatment options before proceeding to surgery.3, 34 Despite this, there are only three studies that have investigated conservative non-pharmaceutical intervention. Hence more studies on the conservative treatment of patellar tendinopathy are needed. From the literature reviewed for this paper, it is impossible to suggest that any one conservative treatment is more appropriate than any other to treat patellar tendinopathy.

    surgical treatment

    No studies on the surgical treatment of patellar tendinopathy fully met the criteria outlined in the methods section. When prospective studies were included, only three surgical studies were appropriate to review, all with relatively small subject numbers. None of the studies offers great insight as to the effectiveness of surgery on this tendon.

    Coleman et al9 reviewed 25 patellar tendon surgery papers and identified the methodological flaws in many of them. Their criticisms of the literature included: the nature of the studies (mainly retrospective case series); bias in subject recruitment and data collection; insensitive outcome measures (Kelly system32); and the postoperative regimens (variable).

    Similarly to conservative treatment, the literature does not offer any indication of the effectiveness or otherwise of surgical treatment of patellar tendinopathy.


    Patellar tendinopathy affects athletes in many sports and at all levels of participation, but has a particular affinity for elite, jumping athletes. These athletes can endure months of frustratingly slow rehabilitation, with treatment that appears to be based on little else other than the personal experience of the treating practitioner.

    The studies examined in this review indicate that it is impossible to recommend any treatment for this condition. Until many more studies are completed, both athletes and clinicians will remain frustrated with the limited treatment options that have been shown to have a beneficial effect on patellar tendinopathy.


    View Abstract