A Systematic Review of the physiotherapy management of lower limb tendonopathies

Tendonitis is a condition which is comparatively commonly seen in various clinics. The largest cohort of patients tend to have developed their condition as a result of various sports-related activities but it is acknowledged that there is a substantial cohort of RSI sufferers and occupation-related forms of tendonitis. (Kader et al 2002) In this piece we aim to review the various treatment modalities and to concentrate primarily on the eccentric muscle strengthening modalities of treatment, the rationale behind them and any evidence that they actually work.

Before we can consider the direct question of eccentric loading as treatment for tendonopathies we must examine the rationale for its uses well as the basic science and theory behind the actual practice. We will do this largely by the mechanism of a literature review.

Methodology

In this review we shall be examining the literature for not only the methods that are currently employed in treating the various lower limb tendonopathies but also for justification for these methods and the quality of the science behind them. We shall therefore critically review the literature available and present it in a rational form.

In addition to this we intend to present an overview of various factors in a wider picture that are relevant to our considerations. We shall consider the current views on the pathophysiology of tendonitis and the experimental evidence on the response of the tendon to exercise in general terms.

Although it is accepted that the majority of patients currently seen in clinical practice with various forms of lower limb tendonitis are suffering from a sports related injury, we shall also look at the effects of ageing on tendon physiology as it is acknowledged that the elderly are another highly represented group with tendonitis.

We conclude the preamble with a number of clinical considerations, most prominently the difficulties posed by the differences in nomenclature and terminology which renders both assessments and comparisons between clinical trials difficult.

We conclude the dissertation with a review of various currently employed treatment modalities and the rationale behind them. We focus specifically on the use and place of eccentric muscle strengthening exercises in the spectrum of rational treatments..

Pathophysiology of tendonitis

At the macro-anatomical level, the tendon is usually easily defined as a semi-rigid white or grey structure, generally found in close proximity to synovial joints. One of its prime functions is to transmit forces generated by muscles to the skeletal system, often inducing movement. (Huxley HE 1979).

At the micro-anatomical level, it’s structure is very much more complex and requires a detailed examination before we can realistically and meaningfully consider the issues relating to the therapy of tendonitis.

Tendons form part of the anatomical structures that are functionally grouped together as the extracellular matrix (ECM). The rate of turnover – both synthesis and degradation – is influenced by a number of different factors including metabolic and disease related factors, but the strongest influence on the turnover rate is mechanical stress, usually as a result of various degrees of physical activity. (Agar Pet al 2000)

Tendon (and intramuscular) collagen, turns over at a rate which is about half as fast as myofibrillar protein turnover. The main physiological stimulus to turnover appears to be the multiple stimuli arising from mechanical or contractile activity.(Cuthbertson D et al2005)

At the cellular level, degradation of collagen is mediated largely byte metalloprotease group of enzymes and synthesis is most strongly influenced by a number of different trophic factors which are released at the cellular level. (Algren MS. 1999)

These growth factors are mainly responsible for both the transcriptional changes as well as the post-translational modifications that take place as a result of either physiological changes or disease processes. (Sand Meier et al 1997)

Until comparatively recently, tendon tissue was thought to be fairly inert. Recent research work has given good supportive evidence that the internal metabolic processes, the internal vascular responses (Alstom et al 1994) and the actual catabolic turnover of the collagen protein in response to physical activity, is considerably greater than originally thought. The converse is also true, as inactivity appears to have the same inhibitory effect on tendon tissue as the better known effect of wasting in muscle tissue. (Abrahamson SO et al 1996). This effect is of particular importance in our considerations (later) when we consider that some authorities suggest that outright rest is inappropriate initial treatment for tendonitis.

Collagen is a large polymer-type protein made up of many repeating subunits, (triple helices of polypeptides with a high proportion of proline and hydroxyproline). It is made by fibroblasts. In the muscle, it forms a basket-like network around the muscle fibres but then forms progressively more coherent and solid structure as it forms discrete tendon. In this way it allows the efficient transmission of forces generated by the myofibrils to the tendon – and hence to the bone. (Kjaer M 2004).

Training, in the form of physical work, exercise or repetitive movements, will have a trophic effect on the tendon as a whole. Collagen turnover can be increased and there can be an overall increase in the amount of collagen protein in the tendon. (Herzog W et al 2002)

Collagen, in the form in which it is found in a tendon, has enormous on-elastic tensile strength and a modest degree of ability to bend under lateral stress. As the amount of collagen in a tendon increases, the tendon’s mechanical (or more accurately, viscoelastic,) properties change. It decreases it’s stress levels for a given load, and thereby renders it more load resistant.(Fowls JL et al. 2000). Again this facts of great relevance to our clinical considerations later in this piece.

The stiffness, or resistance to lateral stress, is a function of the cross-linking of sulphur bonds across the parallel bands of protein. In general terms, the more cross-links, the stiffer the tendon. The degree of cross-linking is a result of a complex interaction between a number of enzyme systems in the matrix of the tendon. (Hamill OP et al.2001)

Polyglycans are an important feature of this enzyme cascade and become an increasingly important functional component as age increases. Older or ageing collagen will tend to exhibit glycolated cross links in addition to the sulphur links of youth. This is part of the reason why older tendons are less flexible (and possibly more prone to injury). (Inglemark BE 1948).

The functional significance of these links is that they render the tendon even stiffer and less able to bend.(Davidson PF 1989).Understanding these processes is fundamental to the prescribing of a rational treatment regime for tendon injuries and other pathologies.

It is also important to have a complete understanding of both the vascular and neurologically mediated adaptation processes that are present in the my-tendon complex. These work on a far more rapid and immediate time frame than the processes that we have just described, and are primarily responses to rapid changes in the mechanical loading stresses.

As muscle tissue develops physiologically, there is a symbiotic relationship between the muscle and the extracellular matrix. The various physiological mechanisms that stimulate muscle growth and hypertrophy appear to have a similar effect on the extracellular matrix. (MacLean et al 1991) But in the latter case, they are less well understood.

We know that that significant and repeated mechanical loading will trigger off, or initiate a process, which starts with the activation of trophic gene in a cellular nucleus, (Banes AJ et al.1999), it progresses through the complex processes of protein synthesis and functionally ends with the deposition of collagen in the tendon tissue.(Yasuda et al 2000)

Responses of the tendon to exercise

There would appear to be some form of integration between the muscular and the extracellular matrix signalling pathways, which optimises the co-ordinated activity of the trophic processes in response to the stimuli (which can be both loading and tensile in nature), which produce the response in the first place. (Viidik A.1993). This co-ordination mechanism must exist, as it is a well-recognised phenomenon that a tendon hypertrophies to accommodate the increased mechanical stress that its associated hypertrophied muscle produces. (Derwin et al 1999)

Considerable research effort has been expended in trying to delineate the mechanism, but to date, the results have not increased our understanding of the situation significantly. (Vierck J et al 2000)

Specific studies in this area have been able to show a clear correlation between collagen response and an increase in physical training. (Langberg et al 2001). The response was detectable after a 4week training programme and was maximal at 11 weeks.

When we consider the pathophysiology of RSI (repetitive strain injury) or even chronic overload syndrome, the stimuli that can produce muscle hypertrophy or increase muscle fibrosis can also produce fundamental changes in the tendon structure. (Birk DE et al 1990)

These changes can include changes in both the chemistry and the functionality of cross bonding of the collagen fibres, (Barnard K et al1987), changes in the size of the collagen fibrils, areas of locally increased blood flow (known as hyper vascularisation zones), and an increase in the catabolic processes which can result in either (or both) collagen being synthesised and laid down, or increase in fibroblastic activity which increases the fibrous component of the tendon.
(Greenfield EM et al 1999)

It is a fundamental recognition of the fact that these processes require “adjusted loading” rather than an enforced absence of loading(immobilisation) to reverse the physiological processes, that underpins most of the thrust of this review.( Howell JN et al 1993), (JärvinenTAH et al 2002)

The experimental evidence to support this view comes from the classic set of investigations by Gibson (et al 1987) who compared the rate of collagen synthesis and turnover in an immobilising long-cast leg with the rate of turnover in the unaffected leg. The rate of collagen synthesis dropped by half over a seven week period in the immobilised leg. The investigators also found an adaptive (and compensatory)reduction in the rate of collagen degradation which had the overall effect of reducing the protein loss in the tendons.

In the overall context of our investigation it is also important to note that the authors also found that minimal electrical stimulation of the muscle (5% of maximum voluntary contraction for 1 hr. per day),increased protein synthesis to such an extent that there was no net protein loss over the same seven week period of the trial. (Gibson etal 1989)

In a study that was remarkable for its invasiveness (the authors took repeated biopsies of human patella tendon after periods of exercise), Miller (et al 2004) demonstrated that tendon collagen synthesis showed a 30% rise within 6hrs of exercise and up to a 50%rise within a 24 hr. period. This was found to exactly follow the pattern of protein synthesis in skeletal muscle. This finding is strongly supportive of the assertions made earlier in this essay, that there would appear to be a mechanical or humeral mechanism that links the trophic effects that are apparent in both tendon and skeletal muscle.

Various authors have postulated different mechanisms (it has to be said – with scant evidence), including integrin’s, (Levenhagen et al2002), growth factors including transforming growth factor beta (TGFB) (Moore et al.2005), or mechano growth factor (MGF) (Rennie et al 2004),which they suggest may be responsible for the co-ordination of the trophic effects of perimysium collagen, tendon collagen and the myofibrils.

More concrete evidence exists (and is arguably of greater relevance to our investigation here), for the fact that dietary protein alone can produce a trophic stimulus for tendon collagen. (Jefferson &Kimball 2001). It is postulated that there is some form of amino acid sensor that is responsive to the availability of amino acids. This haste effect of changing the availability of various protein kinases in the extracellular matrix generally and a subsequent enzymatic cascade which results in an increase in various anabolic signalling molecules which are, in turn, responsible for the activation of mRNA. This is then responsible for the increased synthesis of collagen (and other related proteins), in tendon and other extracellular matrix tissues. This series of very elegant experiments was done in carefully controlled conditions which removed the possibility of other anabolic factors being relevant as the only variable was the availability of amino acids. (Cuthbertson et al 2005)

There is further evidence of the effect of exercise on tendon structure in the form of the set of experiments by Rennie and disco-workers. Looking specifically at the metabolism of collagen Rennie found that after strenuous exercise, the rate of incorporation of a marker into tendon collagen followed a specific pattern (Rennie &Tipton 2000). There was a latent period of about 90 mines after exercise where there was no change in metabolic rate. It was then noticed that there was a dramatic increase to about 5 times normal rates of synthesis, which peaked at about 12 hrs., was maintained for about 12hrs, and then declined over the next 48 hrs.

In line with the findings of Cuthbertson (above) the investigators noted that the rise in levels of synthesis is greatest if associated with an amino acid load just pre- or post-exercise, and this effect can be further enhanced by the administration of insulin secretagogues(such as glucose). There is therefore little doubt that feeding helps the post exercise response. (Atherton P et al 2005)

The effects of ageing on tendon pathophysiology

We have already commented, in passing, on the physiological effects of ageing in relation to the polyglycan cross bonding in tendons. There are a number of other changes which will naturally occur in relation to advancing years, which are of direct relevance to our considerations here. It is clearly a matter of observation that muscles, bones and tendons deteriorate as age increases. This deterioration leads to physical symptoms such as loss of strength, mobility and suppleness together with an increase in fatigability and a general reduction in proprioception. This condition is sometimes called “sarcopenia”.(Forbes 1987)

Epidemiological studies (Dorrens et al 2003), provide good evidence to support the popularly held view that an active lifestyle into old age is more likely to support a higher level of bone density, muscle bulk and tendon flexibility, than a sedentary one. One can postulate that the trophic mechanisms referred to above, stay active for longer when constantly stimulated by mechanical activity. One effect of ageing that has been experimentally demonstrated, is that the trophic effects of available amino acids in the bloodstream are not as great in the elderly as in the young. The elderly appear to have an ability to develop resistance to the trophic effects of amino acids, which was not present when they were younger. (Cuthbertson et al 2005)

Another physiological change that can be demonstrated in the elderly, is a reduced RNA : DNA ratio in tendon tissue, which is a marker of a reduced ability to manufacture protein. This, together with reduction in the amount of detectable anabolic signalling proteins, seems to be central in the failure of the muscle and tendon synthesising mechanisms. (Smack et al.2001).

If we add these findings to other work of Smack (et al 2001) and Leverhagen (et al 2002) which shows that the elderly can show responsiveness in terms of trophic changes in the collagen content of tendons by manipulation of the diet. Both studies showed that maximising the protein : energy ratio of ingested food is a reasonable strategy. It should also be noted that they also demonstrated that one has to be careful to keep the energy content of the food low in order to minimise unwanted weight gain.

The elderly could reasonably be assisted to maximise the benefit they get from training (resistance training in these particular studies), by integrating it with feeding concentrated in the immediate pre- or post-exercise period. This appears to have the effect of increasing the positive synergistic relationship between exercise and amino acid delivery.( Williams et al. 2002)

Clinical considerations

Differential diagnosis

The first and possibly most fundamental issue that we have to consider when looking at the issues of the treatment of tendonitis, is the issue of correct diagnosis. This, sadly, is compounded by the fact that there appear to be several different terminology vocabularies in common clinical use. It therefore can be difficult to directly compare treatment studies of “tendonitis “ unless one has direct and clear diagnostic criteria. (Saxena 1995)

Tendonitis may be taken in some medical circles to include all those conditions which come under the broad heading of “painful overuse tendon conditions” (Khan et al 1999). This is generally accepted by the uncritical, as meaning that this equates with a painful inflammatory reaction in the tendon tissue. Histological investigation of the typical chronically painful tendon, generally shows an absence of the polymorphonuclear and other associated inflammatory cells. In some literature we can see the emergence and replacement of the term tendonitis with tendinitis. This latter term tends to be defined as pertaining to areas of collagen degeneration, increased ground substance and neo-vascularisation. (Purdue et al 1996)

To both illustrate and clarify the point, let us consider thevarious clinical entities that may either present like, or may be diagnosed as, “tendonitis”.

For ease of classification and clarity, in this section we shall consider the term “tendonitis” in specific relation to the Achilles tendon.

Williams (1986) produced the (arguably) most commonly currently accepted definitions of Achilles tendon pathologies. He classified them into:-

Rupture,
Focal degeneration,
Tendinitis,
Per tendonitis (peritendonosis),
Mixed lesions,
Origin/insertion lesions,
Other cases such as metabolic/rheumatic causes.

In common clinical parlance, any of them can be referred to, with reasonable accuracy, as “tendonitis”. (Galloway et al 1999)

The aetiologies can vary (and this may well have a bearing on treatment), from trauma, reduced flexibility, abnormal or changed biomechanical considerations (such as excessive pronation, supination or limb length inequalities) to name but a few. (Saxena, A 1998)

It should be noted that the anatomy of the Achilles tendon is unusual and certainly different from any other in the lower limb. It does not have a true synovial sheath but a petition which extends from its origin in the muscle to its insertion in the calcaneus. Peritendonosisis therefore a commonly misdiagnosed as Achilles tendonitis. It is also clinically significant that there is a region of decreased vascularity in the tendon, which is typically about 6 comes above its insertion (Hume 1994).

The clinical difference between these two conditions is that true Achilles tendonitis may, if chronic, be characterised by fucoid, or fatty focal degenerative, changes in the tendon itself, where asperitendonitis will not involve the Achilles tendon at all. (Kvist1994).

These degenerative changes may be extremely resistant to non-surgical forms of treatment. In practice, the two conditions may well be presenting the same individual. (Killer et al 1998)

The differentiating signs are, however, fairly easy to detect and the two conditions can be separately distinguished in most cases. Per tendonitis is the inflammation of the petition and can usually be clinically distinguished by the presence of clinical crepitus as the Achilles tendon tries to glide back and forth along the inflamed petition. This sign together with pain, generally tends to increase with activity and the tenderness is normally felt along the whole length of the tendon. Achilles tendonitis on the other hand classically gets better with movement and is at its worst after a period of rest. The discomfort tends to be more localised into discrete areas and is more commonly found in cases where there has been either a partial or even a complete rupture in the past. (Clement et al 1994)

Other pathologies can arise associated with the Achilles tendon, and for the sake of completeness we should briefly consider them as they could be potentially confounding factors in any trial which aims to consider tendonitis.

Tendocalcinosis is an inflammatory process which involves the Achilles tendon but only at the point of insertion to the calcaneal bone. It typically will result in calcification and therefore should be considered a different entity to Achilles tendonitis as such. It is characterised by localised pain, and prominence of the calcaneal insertion of the tendon which may well be associated with a retro-tendon bursitis. (Williams 1986)

If we apply the same rationale to the patella tendon, we are again faced with a bewildering array of terminology and conditions which tend to get lumped together as “tendonitis” and may also therefore be confounding factors in any study. We shall therefore spend a few paragraphs delineating them.

Some authors point to the fact that conditions that had been previously referred to as tendonitis, when examined at a histological level, are found to be the result of collagen breakdown rather than inflammation (Khan et al 1996), and therefore suggest the title oftendinosis is more appropriate. (Cook et al 2000) (I)

The whole issue of the role of the inflammatory process in the tendonopathies appears to be far from clear. An examination of the literature can point to work (such as that by Khan – above), who demonstrated that the prime histological changes were non-inflammatory and were more typical of fucoid, hyaline or fibrous degeneration with occasional calcific processes being identified. Other investigators however, point to the clinical picture which commonly includes the classic inflammatory triad of dolour, rub our and tumour (pain, redness and swelling)(Almekinders et al 1998). This, associated with the evidence of the relieving effect of NSAIA’s or corticosteroids(Friedberg 1997) leads to an ambiguous picture.

The pathophysiology of this condition is most commonly thought tube related to jumping and landing activity which is the mechanism which appears to cause the rupture of the collagen filaments and hence the histological appearances. The characteristics of this type of condition are that it tends to be focal, and often in the region of the lower pole of the patella. Initially it tends to be self healing but as the chronicity increases, the pain levels can increase to the point where pain is experienced even at rest (Cook et al 2000) (II)

This type of condition must clearly be differentiated from there-patella bursitis (Housemaid’s knee) which is often mistakenly diagnosed as a patella tendonitis.
(Halaby et al 1999)

Factors which appear to predispose to tendonopathy

Many authors identify chronic overuse as being one of the major factors in tendonopathy generally. (Kist 1994) (King et al 2000). This applies equally to the occupational tendonopathy as much as the sports-related conditions. (Jon stone 2000) (Kraushaar et al 1999). We should acknowledge that the term overuse can refer equally to overuse in terms of repetitive action just as much as it can refer to overloading. The two factors being independent (but often related).

Some of the current literature points to the fact that there can be differentiation in the spectrum of overuse injuries between those conditions that arise from some form of biochemical change in the structure of the tendon itself (Joss et al 1997), those that are associated with biomechanical changes (such as change in function or previous injury) (Alstom 1998) and those that arise as a result of ageing or other degenerative changes (Alstom et al 1995).

These factors can arise as a result of, or independently from, other factors such as the fact that the anatomical path of a tendon can take it over (or in close proximity to) friction-inducing structures such as a bony prominence – as in the case of the tibias posterior tendon, (Benjamin et al 1998) or factors relating to the site of insertion of the tendon into the bone – as in the case of theAchilles-calcaneum interface.(Benjamin et al 1995)

We can point to evidence that extraneous factors can also predispose to tendonopathy. There are genetic factors (Singer et al 1986), and a relationship to blood type (Joss et al 1989). The presence of certain concomitant chronic or debilitating illnesses can certainly be associated with tendonopathies (Kannur et al 1991) as can the chronic use of certain medications – most notably the fluoroquinolone group.(Huston 1994)(Ribard et al 1992). The mechanism in the latter case appears to be associated with an increase in the amount of MMP and its associated activity which seems to be associated with an increase in the rate of degradation of protein (especially collagen) in certain tissues. (Williams et al 2000).

Other authors have identified biomechanical factors as being significant (rather than necessarily causal), in the development oftendonopathies, but we shall discuss this in specific relation to treatment, and so will not discuss it further here

The spectrum of currently available treatment

Before beginning any rational consideration of the various forms of treatment available, one must appreciate a common truth in medicine, and that is that different treatments and different patients will respond differently to a specific treatment modality, and one of the factors that will influence this phenomenon is the skill and experience of the practitioner concerned. For example, a surgeon may well find that he gets good results from tenotomise but poor results from eccentric exercises and therefore will recommend surgery. Physiotherapist may find the converse. It is therefore important to be critical of such factors in any appreciation and appraisal of different techniques for the treatment of the lower-limb tendonopathies.

In this section we shall examine the available literature to try to obtain an overview of the various treatment modalities that are currently being prescribed and examine the rationale behind their use and efficacy

Most authors seem to agree that, before considering the specific conditions, a general approach of conservative measures (such as load reduction, strengthening exercises, and massage) should be tried before other modalities such as medication and physical interventions(ultrasound etc.), and that surgery should only realistically be considered as a last resort. The only obvious exception to that approach would be when complete (or sometimes perhaps partial ) rupture of the tendon has occurred, and then surgery may well be considered the prime intervention. (Cook et al 2000) (I)

Let us consider the various options in turn.

In this section we will begin (again, for the sake of clarity), by specifically considering the options available for patella tendonitis. We accept that there will, of course, be overlap between the treatments for the various tendonopathies, but it makes for a rational approach to consider each in turn.

The first comment that we must make is that, after examination of the literature it is noticeable that there are only a comparatively few well constructed, placebo controlled randomised trials in this area.(Almekinders et al 1998). Those that we can examine appear to suggest that the traditional treatments aimed at minimising the inflammatory processes in the condition are largely ineffective. The authors (Cooked al 2000) (II) suggest that this may well be because of the findings we have quoted earlier (Khan et al 1996) that histologically, the prime pathology is not inflammatory.

Relative Rest

Cook (et al 2000) (I) points to the fact that many strategies can rationally involve load reduction and the (now outmoded) instruction to “Stop everything and rest” is positively contraindicated. The rationale for this relates to the mechanisms that we have examined earlier in this piece. Immobilisation of a tendon is actually harmful as we can point to evidence (above) that shows that tensile stress and mechanical action not only stimulates collagen production, it also is vital in tendon to ensure it’s optimal fibre alignment. Rational treatment suggests that a programme of “Relative rest” may be beneficial. By that, the authors (Cook et al 2000)(I) suggest that activity should continue as long as the prime traumas of jumping, landing or sprinting can be avoided and reintroduced in a carefully graded fashion.

Biomechanical Correction

Because patella tendonitis is primarily related to jumping and sprinting sports ( in numbers that present clinically), we will consider treatment in relation to them. The forces that are generated in the patella tendon on landing after a jump are considerably greater than those that produced the jump in the first place. (Richards et al1996). It logically follows that if biomechanical methods can be employed to more efficiently minimise the forces, they would be best employed on landing strategies than jumping ones.

One should appreciate that the energy-absorbing capacity of the limbs dependant, not only on the patella tendon, but factors at the hip and ankle as well.
Studies show that the ankle and calf are the prime sites of absorbing the initial landing load (Richards et al 1996) and, if these structures are not biomechanically sound, then this will increase the forces transmitted to the knee.

Prilutskii and his co-workers (et al 1993) completed a series of studies which showed that up to 40% of the energy absorbed on landing is transmitted proximally from the ankle/calf mechanism. It follows that it must be biomechanically sound if it is to absorb the 60% bulk of the load which otherwise would be transmitted upwards to the knee mechanism.

Another set of studies (Prapavessis et al 1999) concluded that when flat-foot and fore-foot landings were compared, the latter generated less forces throughout the lower limb and that the forces could be reduced further (up to another 25%) by increasing the range of both hip and knee flexion on landing.

There are a number of other potential biomechanical deficiencies that can be amenable to correction and should therefore be sought outspans planes may be an obvious anatomical problem detectable at an initial examination (Kaufman et al 1999), but there are other types of functional abnormality (such as excessively rapid pronation on landing) (McCrery et al 1999), that may require far more sophisticated evaluation. Outhouses inside shoes may go a long way to help these problems

Some authors, (McCrery et al 1999), regard a reduced range of movement in the sub-taller joints as an aggravating factor which places and undue stress on the Achilles tendon and that manual mobilisation of the joint is indicated in these cases.

Cry therapy

In the light of the histological findings mentioned earlier,cryotherapy has a rational place in treatment. It is thought th