Causative factors and rehabilitation of patellar tendinopathy: A systematic review

Background Patellar tendinopathy (PT) is a common chronic pathology of the knee, with a high prevalence in athletes and the general population. Objectives The objectives of this article were to systematically investigate all the evidence applicable to the intrinsic and extrinsic causative factors and rehabilitation of PT, and then integrate and link rehabilitation with the main causative factors identified. Method The Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines were followed. Various tools were used to evaluate the methodological quality of the eligible articles. Data were interpreted descriptively, and the causative factors and rehabilitation of PT were analysed. Results Twenty studies were included in the review. The distinctive factor responsible for PT is the mechanical theory. Seven intrinsic and four extrinsic risk factors were identified, with the main intrinsic risk factors being muscle flexibility and strength, and extrinsic risk factors being acquisition and level of skills. PT can be treated with numerous different therapeutic modalities, although eccentric muscle training showed exceptional results. The intrinsic and extrinsic risk factors can only be transformed and reduced by rehabilitation, which is inevitable to improve PT pain and function. Conclusion The essence of an integrated management protocol for PT is to identify the dominant contributing factors, whether intrinsic or extrinsic, and to reduce the load on the patellar tendon by modifying these factors by either rehabilitation intervention or direct modification of the equipment or environment to obtain a positive outcome towards pain management and function.


Introduction
Patellar tendinopathy (PT) is a common chronic pathology of the knee, with a high prevalence in both athletes (Frizziero et al. 2014) and the general population (Toppi et al. 2015). It is characterised by microscopic tears and tissue degeneration because of excessive and repetitive mechanical loading of the patellar tendon. Epidemiological studies indicate that tendon injuries account for possibly up to 50% of injuries sustained during sporting activities, with tendon overuse because of running-associated sports accounting for nearly 30%. Athletes with PT can experience uncomfortable symptoms and decreased function for up to 3 years (Saggini et al. 2012), which have a negative effect on quality of life (Toppi et al. 2015). The risk factors for and treatment of PT can be challenging and unsatisfying (Silva et al. 2015). A number of risk factors for the development of PT have been identified, such as age, gender, heavy physical work, type of training surface, high training volume and level of participation (De Vries et al. 2015a). A better understanding of the aetiology of PT will facilitate the identification of modifiable risk factors and make a valuable contribution to planning of preventative measures and interventions (Van der Worp et al. 2012). It is unclear how these risk factors specifically relate to the available rehabilitation options for PT.
Only one published systematic review article (Van der Worp et al. 2011b) indicated the risk factors for PT, whereas three others described different rehabilitation options for PT (Frizziero et al. 2014;Malliaras et al. 2013;Mani-Babu et al. 2015). However, none of these reviews integrated the risk factors and rehabilitation. One of the objectives of this review was therefore to integrate the risk factors and rehabilitation of PT in a systematic review. This followed a methodical investigation of all evidence applicable to the intrinsic and extrinsic causative factors as well as rehabilitation of PT, as another objective of this review. Addressing the intrinsic and extrinsic factors with rehabilitation is considered to be a good measure in the successful rehabilitation of PT, and therefore, this systematic review provides a unique The search strategy included the following keywords in order to identify all the relevant articles for inclusion in this review: ("patella* tendinopath*" or (patella* and tendinit*)) AND ("intrinsic factor*" or age or gender or "body composition*" or "fat mass" or "body weight" or "body mass index" or injur* or "joint instability*" or "musc* strength" or "musc* power" or "range of motion" or "range of movement" or "anatomic* alignment*" or "postural stability*" or "sport* specific technique*" or "level of skill*" or "skill* level*" or "extrinsic factor*" or strapping or bracing or "foot wear" or footwear* or shoe* or "training surface*" or "eccentric decline squat*" or "skill* acquisition" or proprioception* or flexib* or "muscle activat*" or etiolog* or aetiolog*) AND (rehab* or "return to sport" or "return to play" or "motor reeducat*") and (exercise* or train* or sport*) This search was conducted twice during the study period (approximately 3 months apart), in order to cross-reference the results and ensure that all possible articles eligible for inclusion were identified. The combination of results from the two searches yielded 120 possible articles for inclusion before the elimination of duplicate studies (Figure 1).

Study selection
The study selection process from the eligible 120 articles was independently conducted by all members of the research team, against the inclusion and exclusion criteria (Box 1). When there was vagueness concerning the eligibility, a conversation was held between all the members of the research team to resolve any disagreement. Figure 1 shows the process of the search strategy to determine the final sampling of the articles for the review.

Methodology quality appraisal
Different methodology quality appraisal tools were selected to evaluate the eligible articles because of the different research designs of these articles. Twenty-five articles were assessed for methodology quality scoring after which 20 articles were included in this systematic review (see Figure 1). Table 1 lists more details of the methodological quality appraisal for the final 20 articles included in this systematic review. The independent researcher was responsible for verifying the data obtained from the scoring done by the authors. The average total methodological quality score for all the articles included (n = 20) was 72%.

Data extraction
A custom-made Excel spreadsheet was developed to extract applicable information from each article and a summary is displayed in Tables 2, 3 and 4. Information was documented with regard to the age, gender, body composition, muscle strength and flexibility, anatomical alignment, medical history, strapping or orthosis, training surfaces, and acquisition and level of skill of the participants described in each article. The researcher conducted the data extraction autonomously. The information recorded was confirmed by the rest of the research team who checked it jointly for mistakes after completion of the process.

Data analysis
Pooling of the data for the purpose of framing a metaanalysis was not an aim of the systematic review because of dissimilarities in the outcomes. Data were summarised descriptively by means of gathering information about the characteristics of the causative factors and rehabilitation of PT to form a data analysis.

Results
The search strategy yielded 20 articles that met the described inclusion criteria shown in Box 1. Results from these 20 articles are included and discussed in the next section.

Demographic information
The age of the study participants varied for each of the 15 studies in this review that revealed the age of the participants (see Tables 2, 3 and 4). All the study participants were older than 18 years, with Toppi et al. (2015) having the oldest study participants (with a mean age of 67 years). The majority of the study participants were aged between 18 and 35 years. According to Van der Worp et al. (2012), the probability of sustaining PT decreased with age, but they argued that the risk for PT increased for individuals older than 30 years of age because of changes in the tendon structure and its mechanical properties. However, the correlation between PT and age is still uncertain (Van der Worp et al. 2012).
Fourteen of the studies included in this review (Tables 2, 3 and 4) described gender with 13 indicating that the majority of participants were men. Three of these studies determined gender to be a risk factor, with men more likely to develop Eligible arƟcles for inclusion in final systemaƟc review (n = 20)

Eligible arƟcles (n = 25)
Conducted a methodological quality check Exclusion of a further 5 arƟcles due to poor methodological quality (Appendix 1)

Eligible arƟcles (n = 42)
Obtained and screened full-text of 60 eligible arƟcles Exclusion of a further 17 arƟcles due to irrelevancy based on full-text screening (Appendix 1)

Eligible arƟcles (n = 54)
Obtained and screened abstracts of 75 eligible arƟcles Exclusion of a further 12 arƟcles due to irrelevancy based on abstract screening (Appendix 1)

Eligible arƟcles (n = 93)
Title screening conducted according to inclusion and exclusion criteria (Box 1) Exclusion of 39 arƟcles due to irrelevancy based on Ɵtle screening (Appendix 1)

Eligible arƟcles (n = 120)
DuplicaƟon check conducted Exclusion of 27 arƟcles due to duplicaƟon PT than women (De Vries et al. 2015a;Van der Worp et al. 2011a, 2012. This hypothesis could be justified in that the quadriceps muscle generates more force during contraction in male than in female athletes, as well as the fact that women's oestrogen plays a role in influencing the tendon structures positively ( Van der Worp et al. 2012). Nevertheless, PT is a multifactorial pathology and it is challenging to obtain direct evidence with regard to gender differences and the part played by oestrogen (Van der Worp et al. 2011a).

Muscle flexibility
According to the three systematic reviews, impaired lower limb muscle flexibility was noted as being a risk factor for PT. The muscles described in these reviews are the iliotibial band (

Anatomical alignment
Four studies referred to anatomical alignment in athletes with PT. Saggini et al. (2012) reported a definite correlation between an inferior patellar pole alignment and the probability to develop PT. Secondly, leg-length discrepancy, where the longer leg is the preferred take-off leg in jumping, can also be associated with PT. However, there is limited evidence to confirm this (Van der Worp et al. 2011b). Another hypothesis is that lower foot arch heights might cause knee and soft tissue injuries (Silva et al. 2015).

Body composition
Three studies mentioned body mass index (BMI), with two of them not finding any relationship between BMI and PT (De Vries et al. 2015a;Toppi et al. 2015). The third study indicated that a higher BMI can contribute to PT because of a theoretically greater loading of the patellar tendon (Van der Worp et al. 2011b). Samukawa (2011) andSilva et al. (2015) mentioned in both their case reports that decreased dorsiflexion was a risk factor for PT. This must, however, be interpreted with caution because of the methodology used (i.e. only one study participant per case report).

Type of sport
Only one article referred to the type of sport as an extrinsic causative risk factor for PT. Volleyball players are more likely to develop PT than basketball players, possibly because of the difference in the number of jumps performed and the players' jumping technique (Van der Worp et al. 2012). Saggini et al. (2010) were the only authors who reported that training on hard surfaces is a risk for PT because of the load

Authors (year) Study design Participants Causative intrinsic factors Causative extrinsic factors Rehabilitation
Silva et al. Case

Strapping or orthosis
According to De Vries et al. (2015b), a patella strap or sports tape decreases pain in an experimental group compared with a control group in the short term, although none of them is more effective than placebo taping. The long-term effect remains unclear, with an amplified effectiveness of taping in female participants (De Vries et al. 2015b).

Rehabilitation
Thirteen of the 20 studies included in the systematic review had one or more component that described parts  . Malliaras et al. (2013) suggested that an eccentric-concentric loading programme for individuals with PT is more beneficial than EE on its own, particularly in athletes with noticeable concentric muscle weakness that may not recover with isolated EE because of the muscle type contraction.

PT and other treatment modalities
Non-specified drug therapy (Dimitrios et al. 2012) and deep transverse friction massage also improve the symptoms of PT (Samukawa 2011).

Hip and core strengthening
The gluteus maximus muscle is considered to be an important hip extensor muscle to strengthen during PT rehabilitation (Silva et al. 2015).

Stretches
Improved flexibility of the muscles of the lower limb, especially muscles surrounding the hip and knee, forms an essential component of rehabilitation and the resolution of chronic symptoms of PT (Samukawa 2011).

Return to sport
There are contradicting views regarding return to sport during the rehabilitation intervention. Silva et al. (2015) advised that sport participation can be continued, but the pain experienced during the activity may not exceed 3/10 on the Visual Analogue Scale. On the contrary, it is recommended that activity participation must only resume after tendon healing is complete (Dimitrios et al. 2012), which is suggested to take approximately 12 weeks (Frizziero et al. 2014). Sportspecific activities that cause pain in the tendon must be avoided until the tendon has healed completely (Dimitrios et al. 2012).

Sport-specific technique
Jumping with appropriate kinematics is an important factor to consider in PT. The jumps in PT athletes must be evaluated and modified accordingly to produce a 'soft landing' to reduce the load on the tendon (Silva et al. 2015).

Discussion
PT is a diverse and complex pathology with numerous challenges regarding risk factors for developing PT and its rehabilitation, and it can be more complicated than assumed to identify these aspects. A limited number of articles (n = 20) were included in the systematic review because of the inclusion and exclusion criteria for this review, as well as poor methodological quality of a number of studies initially identified. This, in addition to the heterogeneity of the studies when considering specific aspects related to the causative factors and rehabilitation, prevented a meta-analysis to be performed.

Intrinsic and extrinsic causative risk factors for patellar tendinopathy
The initial aim of this systematic review was to review the literature concerning the intrinsic and extrinsic causative risk factors for PT. The distinctive factor responsible for the development of PT is the mechanical theory. It can be defined as a failed healing process with micro-injuries to the patellar tendon because of overloading, which is responsible for matrix and cell changes and altered mechanical properties of the tendon (Van der Worp et al. 2011b). The data varied and seven intrinsic and four extrinsic risk factors were identified. The main intrinsic risk factors identified were muscle flexibility and strength. Regarding the extrinsic risk factors, acquisition and level of skills were the prominent factors. Evidence for all the other risk factors was indecisive.
Although several risk factors for PT were recognised, the systematic review did not demonstrate robust evidence for numerous intrinsic or extrinsic causative risk factors, which http://www.sajp.co.za Open Access has possibly been limited by the inclusion and exclusion criteria for this review.
If the mechanical pathophysiological theory discussed previously ( Van der Worp et al. 2011b) is taken in consideration, it clearly indicates that all intrinsic and extrinsic causative risk factors for PT mentioned in the results are responsible for loading the patellar tendon in altered ways (see intrinsic causative risk factors and Table 2). If the intrinsic and extrinsic causative factors could be addressed in rehabilitation and the load on the tendon could be minimised, improvement in pain and function will be noticeable.

Rehabilitation for patellar tendinopathy
Regarding the rehabilitation, mainly descriptive articles with a widespread range of research designs were included. PT can be treated with numerous different therapeutic modalities; EE training showed exceptional results, with 10 of the 13 articles on rehabilitation for PT reporting positive results. The perception is still that EE is the gold standard conservative treatment modality and superior to other treatment options based on its excellent and sturdy outcomes over the years. This happens when EE lengthens the tendon, which is responsible for 'squeezing out' the flow of neovessels (McCreesh et al. 2013). Doppler ultrasound imaging after a period of EE indicated a minimal improvement in tendon echogenicity and a substantial decline in tendon neovascularity, with a dramatic improvement in the symptoms and pain in the tendon (Da Cunha et al. 2012;Samukawa 2011). This explains why EE is so successful and beneficial in the treatment of PT. EE not only assists with diminishing of neovascularity, but also facilitates improvements in neuromuscular activation, improved muscle strength (Samukawa 2011) and muscle endurance (Saggini et al. 2012).
EE is performed as a single leg squat on a decline board with an angle of 25° at a slow speed to assist with tissue healing (Dimitrios et al. 2012) and has superior results over a squat that is performed on a flat-step (Frizziero et al. 2010;Goldman & Lentz 2010). EE can be performed in an aggressive manner causing pain (Da Cunha et al. 2012) or without pain, because evidence reveals that both establish improvements in pain and function (Da Cunha et al. 2012). The frequency and repetitions of performing EE differed in the reviewed articles, but generally included three sets of 15, one or twice daily and 5-7 days a week (Goldman & Lentz 2010;McCreesh et al. 2013;Van der Worp et al. 2011c).
EE can be performed at home without full-time supervision, although the athlete's compliance may influence the outcomes (Dimitrios et al. 2012). The high intensity of performance of EE seems to be decisive for favourable therapeutic results (Da Cunha et al. 2012). More than one of the articles included in this review combined EE with an additional treatment modality. Although this systematic review did not specifically aim to address the possible different treatment methods in combination with EE, it has been noticed that EE has positive outcomes when used in combination with other treatment modalities.
Multiple treatment options are available for PT, and the collective aim is to improve pain and function in PT. One of these additional modalities is ESWT, of which the therapeutic outcome is based on its bio-stimulating effects (Saggini et al. 2012). The authors suggested different dosage prescriptions which vary from one (Saggini et al. 2012) to three treatment sessions per week for 3 weeks (Van der Worp et al. 2011c). Secondly, static stretches contribute to the reversal of PT (Dimitrios et al. 2012), and deep transverse friction massage contributes to reduce the adhesions and enable realignment of collagen fibres (Samukawa 2011).
An important factor in managing symptoms of tendinopathy during the rehabilitation intervention is regulating or reducing the tendon load for good execution of the exercise programme. The benefits of a conservative rehabilitation programme can include an improvement in knee range of movement, quadriceps muscle strength, reduction in PT pain symptoms and general improvement in knee function (Goldman & Lentz 2010).

The association between intrinsic and extrinsic causative risk factors with rehabilitation
Intrinsic and extrinsic causative risk factors can only be addressed by rehabilitation which is essential to the improvement of PT pain and function. Therefore, it is essential to address each intrinsic and extrinsic causative risk factor individually during the rehabilitation period and make adaptations to the environment if necessary. Research studies describing the rehabilitation of intrinsic and extrinsic causative factors were found in the results of this systematic review (see Tables 2, 3 and 4).
A positive observation from this systematic review is that inadequate lower limb flexibility and muscle strength, especially around the hip and knee, has an adverse effect on knee kinematics (Samukawa 2011) and can be addressed by a stretching and strengthening programme. It is advised to start with non-weight bearing strengthening exercises and then progress to more advanced exercises that will contribute to improve lower limb biomechanics during landing kinematics (Silva et al. 2015). This indicates a direct link http://www.sajp.co.za Open Access between addressing intrinsic and extrinsic causative factors in rehabilitation and the difference in the pain and function associated with PT.
Acquisition and level of skills are extrinsic risk factors for PT that can be modified during rehabilitation to correct the execution technique of activities and lower the demand on the tendon. For example, in a task requiring lower limb effort, such as hopping, it has been found that athletes with PT perform the task differently to 'off-load' the knee and decrease the effort on the affected joint (Souza et al. 2010 Some recommendations regarding the causative factors and rehabilitation are relatively broad rather than explicit guidelines. Therefore, there is a need for continuous highquality research studies in order to identify better evidence regarding the intrinsic and extrinsic risk factors associated with PT and to investigate the diverse rehabilitation interventions for tailoring a rehabilitation programme to manage this challenging pathology.
A limitation of this systematic review was the inclusion of a limited number of articles because of the specific research focus, inclusion and exclusion criteria set for this review, and the poor methodological quality of a number of studies initially identified. In an attempt to address this limitation, the researchers applied the search strategy twice during the study period and also examined the reference lists of all the included studies in an effort to identify possible additional studies. This additional measure, however, yielded no further articles for inclusion.
Strengths of this systematic review include the perspectives presented, which give insight into the causative factors and rehabilitation for PT. This will enhance the knowledge of health care professionals involved in the management of PT. Another strength is the inclusion of articles with a good methodological quality (i.e. with an average of 72%) following the rigorous methodology quality assessment. The results, however, indicate a high incidence of similar authors included in the systematic review, which may direct publication bias. PT is a troublesome pathology which requires intensive research and rehabilitation and the high methodological quality of articles included in this review contribute to reliable outcomes in the reporting of the findings.

Conclusion
This systematic review presents a unique perspective on the integration of intrinsic and eccentric causative factors with the rehabilitation of PT. This insight into the latest evidence highlights the essence of an integrated management protocol for PT to obtain a positive outcome with regard to pain management and function in athletes with PT. Appendix starts on the next page →