The term tendinopathy refers to a general primary disorder of the tendon, associated with overuse in and around it, in the absence of histopathological findings. Since inflammation is absent or minimal in biopsy specimens, the descriptive term ‘tendinitis’ (implying an inflammatory process) should be used only when a histological confirmation is available [1]. More than 30% of injuries related to sports activity arise from or have an element of tendinopathy. On the other hand, tendinopathy is not restricted to athletes, affecting also sedentary population. Between 1 and 3% of the general population suffer upper extremity tendinopathies being 2 to 3.5 times more frequent in people over the age of 40, particularly if playing tennis more than 2 hours per day [2]. The risk of tendinopathy is related to occupational or sporting activity as well as the number of repetitions, strength and exposure time.

The risk increases when high strength, repetitions or exposure to vibrations during repetitive work are combined [3]. According to the current evidence regarding conservative treatment, only a few randomized controlled trials (RCTs) have been performed providing clinical evidence supporting their use; thus, the ideal treatment for tendinopathy remains unclear [4]. The management of tendinopathies should include early functional treatments, rather than rest and immobilization, but there is not enough scientific evidence to support the application of conservative treatment such as ultrasound (US), iontophoresis with NSAIDs, deep transverse friction massage (DTFM), or acupuncture for treating tendinopathy [3,5-7]. Nevertheless, in some studies these treatments show positive effects in the reduction of pain or improving the function of patients with tendinopathies (e.g., lateral epicondylitis) [3,8-10].

Other therapies, including extracorporeal shock wave therapy (ESWT), glyceril trinitrate patch, and injection of substances such as autologous blood, corticosteroid, prolotherapy, or platelet-rich plasma (PRP), may be considered if patients do not respond to the mentioned treatments and remain limited significantly in function or activity due to pain [11].

Exercise programmes incorporating eccentric muscle activity are becoming increasingly popular as they are considered to provide a more effective treatment than other forms of exercise therapy [1,12]. Some studies proposed the used of eccentric training for promoting collagen fibre cross-linkage formation within the tendon, thereby facilitating tendon remodelling [13].The objective of this systematic review was to determine the effect of eccentric training in the management of upper limb tendinopathies and to make recommendations for future research.

The study protocol followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement (PRISMA) guidelines [14]. The protocol was prospectively registered in PROSPERO, registration number CRD42014009952 [15]. A systematic, computerised search was performed using MEDLINE, SPORT Discus, CINAHL and Physiotherapy Evidence Database (PEDro) to gather information relating to the treatment of upper extremity tendinopathies with eccentric training in physically active adults. Study details were extracted and organized into (Table 1 (Included as supplementary data)) with the synopsis of the selected studies: pathology, duration of symptoms, intervention, outcome, post-treatment follow-up and PEDro score. Two authors (MOC and IMP) independently searched the databases using following key words in all relevant combinations: tendon, tendinopathy, exercise, eccentric, training. All titles and abstracts retrieved were read to assess their relevance. Studies were filtered depending on the following criteria: studies in English or Spanish; adult participants with clinical diagnosis of tendinopathy; studies design was RCT; results regarding pain or strength were assessed; and eccentric exercise was employed to treat upper extremity tendinopathies. No limits were included concerning the year of publication [1,2,5-7].

The methodological quality of each selected study was assessed independently by two reviewers (MOC, IMP) using the PEDro criteria [1]. Two authors (MOC and IMP) independently extracted data from the selected studies regarding: (i) authors; (ii) sample size; (iii) location of tendinopathy; (iv) treatment group; (v) control group; (vi) outcome measure; (vii) previous treatments; (viii) PEDro score; and (ix) the variables in the eccentric exercise protocol, such as description of the exercise, sets, repetitions, time of rest, progression of the intensity and frequency per week. Other information obtained included group sizes and socio-demographic data (age, gender), length of symptoms, dominance according to tendinopathy, previous treatments, and followup post treatment. Any discrepancies were resolved through discussion until consensus was reached.

Figure 1 reports the number of full-text studies retrieved and the number of studies excluded. Exclusion criteria included: tendinopathies were not treated (n=1); participants were cadavers (n=1); participants were non-human (n=1); eccentric work was not mentioned for the treatment of tendinopathies (n=1); RCT design was not respected (n=2).The mean PEDro score for the 12 studies was 6/10, with a range from 3 to 8.Consideringthat the intervention of eccentric exercise does not allow for blinding of the participant or therapist, these scores are relatively good. Blinding of the assessors and concealed allocation were two other criteria that were poorly reported in most of the selected studies. Thus, the highest achievable score was 8/10.

Figure 1: Flow chart of the Selection Process according to PRISMA Statement. According to: Moher D, Liberati A, Tetzlaff J, Altman DG. & PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Int J Surg Lond Engl 2010; 8: 336-41.

The two reviewers had initial agreement on 112 out of 120 criteria (k=0,861), and reached consensus on all criteria. After selection criteria were applied, 12 studies satisfied the eligibility criteria with a total of 668 participants with clinical diagnoses of epicondylar tendinopathy (n=355) or shoulder impingement (n=191). The average number of participants in each trial was 55.6 (range from 20 to 120; eccentric group mean=25.9; control group mean=28.1) and the mean duration of symptoms at baseline ranged from 1 to 156 months [1]. (Table 1) gathers the results of the 12 studies included in this systematic revision.

Functional status was assessed in 9 studies, with the TEFS (Tennis Elbow Function Scale), DASH (Disabilities of the Arm, Shoulder and Hand) questionnaire, SPADI, or SDQ (Shoulder Disability Questionnaire). Results provide us that 7 studies improved the functionality of the participants in the eccentric-group, and 2 studies had non-significant changes in that group; in terms of inter-group changes, in 2 studies were moderately and significantly better in the non-eccentric group. In 3 studies there were not significant differences, and in 4 studies they were significantly better in the eccentric-group. Pain was assessed in all of them using the VAS (Visual Analogue Scale), except in one study, in which SPADI (Shoulder Pain and Disability Index) scale was employed. In 11studies, pain decreased significantly in the group where eccentric work was employed as therapeutic modality (p<0.05); in the remaining study, no significant improvement within14 group was found in the eccentric-group (p=0.71). Regarding intergroup changes, in 2 studies 37,41 were significantly better for the noneccentric group.

In 7 studies, significant differences were not registered between groups; in 5 studies changes were significantly better in the eccentricgroup. Strength was assessed differently in 9 studies. Tyler assessed it in various forms: during wrist extension, middle-finger extension and in combination [16]. Significant within-group improvements were observed in the eccentric-group during wrist extension and combination strength (p<0.05), but not during middle-finger extension (p>0.05); in terms of inter-group changes, they were only significantly better in the eccentric-group during combined strength (p=0.01), not having differences during wrist and middle-finger extension.

The changes found by Peterson were significantly better in the eccentric-group, both within-group and intergroup (p<0.05). Svernlöv32 found also improvements significantly better in the eccentric group, both within-group and inter-group (p<0.01 and p<0.05, respectively). When did not find significant differences between groups. Other 2 studies evaluated the isometric strength. Maenhout found significant within-group changes in the eccentric-1group (p<0.05); in terms of inter-group changes, no significant differences were observed, except in isometric strength at 90º of shoulder abduction, where changes were significantly better in the eccentric-group (p=0.033) [17]. Struyf did not find any within-group changes and nor inter-group. The most common treatment protocol among these studies was established by Alfredson et al., which consisted of 3 sets of 15 repetitions performed twice daily. Most authors of the selected articles did not deviate much from Alfredson´s protocol.

There is strong evidence supporting the role of eccentric training in treating patients with epicondylar tendinopathy and shoulder impingement. Paralelly, there is strong evidence that eccentric training can produce improvements for the outcomes pain, functional status and strength in most of the studies. In terms of methodological quality, it was generally adequate.

Evaluators blinding was generally accomplished, with only 2 studies in which the blinding was not respected, and other 2 studies in which both blinding and evaluator independence were not referred. Regarding PEDRo score, only 2 studies scoring less than 5 points [18,19].

Regarding the post-treatment follow-up, 7 studies did not perform it, complicating the collection of long-term results. Nagrale developed a 4-weeks follow-up, and Martinez-Silvestrini developed a 6-weeks study with no post-treatment follow-up; despite both groups improved significantly, no inter-group changes were found, probably due to the short duration of the study period [20]. All of this, together with the characteristic “vicious cycle” of the tendon injury, may limit considerably the collection of faithful and closer information to the tendinopathy reality [21].

According to the number of included participants, Wen and Viswas did not estimatePreviously the sample size and Struyf included participants, although a minimum of 46participants was estimated as necessary to be significant [22].Other aspects, such as the supervision of home programmes or the lack of general agreement with respect to inclusion and exclusion criteria may lessen the validity of the studies. Svernlöv, Viswas, Söderberg and Nagrale set the pain at palpation of the epicondyleas an inclusion criteria; the individual participantivity in palpation skills may have lead the inclusion process to error [23].

The failure to accurately report exercise protocols and the substantial variation in exercise parameters made it difficult to assess the effectiveness of each study’s ability to isolate an eccentric exercise component and provide a progressive muscle stimulus. There is still a controversy in the literature for the ideal parameters of the eccentric training protocols. The number of sets and repetitions, as well as the number of sessions per week and the resting time between sets, vary from one study to another. In a similar vein, eccentric treatment is combined with other therapeutic modalities in most of the studies, such as13 stretching, massage, ultrasound, orthotic devices, etc. Consequently, the attempt to isolate the results and evaluate only the effect of the eccentric training becomes complicated.

We can conclude that eccentric training is effective for upper limb tendinopathies, but its superiority against other methods is not totally clear. The lack of general agreement regarding inclusion and exclusion criteria may have lessens the validity of the studies. Long term studies and longer follow-up periods become imperative, as well as more studies with supervised exercises programmes [24]. Eccentric exercise seems to be a useful treatment for upper extremity tendinopathies, but whether it is more or less effective than other forms of therapeutic exercise is unclear. The lack of general agreement regarding inclusion and exclusion criteria, adherence to home treatment or dosage exercise does not allow us to make a conclusion about the effectiveness of eccentric training.

No actual or potential conflict of interest exists and no external financial 1support was sought or received for the conduction of this review.