Tendinopathies around the Elbow Part 1: Lateral Elbow Tendinopathy
- * Department of Orthopaedics, Royal Gwent Hospital, Newport, UK
- † Department of Orthopaedics, St Elisabeth Hospital, Tilburg, The Netherlands
- Correspondence
Oliver Donaldson, Gordon House, 10 Bridge Street, Usk, Monmouthshire NP15 1BG, UK. Tel.: 0797 3663340. Fax: 0151 2826619. E-mail: ollydonaldson@hotmail.com
Abstract
Tendinopathies of the elbow and in particular of the common extensor origin are a common cause of elbow pain. Part one of this two part review of tendinopathies of the elbow will focus on the pathophysiology and management of lateral elbow tendinopathy, frequently referred to as tennis elbow. Lateral elbow tendinopathy is a common condition with an incidence of 1–2%. The pathology arises from the origin of extensor carpi radialis brevis where changes, consistent with all tendinopathies, of angiofibroblastic hyperplasia occur secondary to repetitive micro trauma. It is not an inflammatory condition. Clinical history and examination is usually sufficient for diagnosis although MRI and ultrasound can be used. The many treatment options that have been proposed have a mixed quality of supporting evidence. Thus management protocols are difficult to define. Treatment depends on the length of symptoms. Acute presentation is managed through conservative measures including activity modification, topical NSAIDs and physiotherapy. For patients with recalcitrant symptoms, injection therapy with, for example, platelet rich plasma can be used. Alternatively surgical excision of the diseased tissue can be performed. This review article will consider the available evidence in order to identify both treatments that are effective and those that are not.
INTRODUCTION
Elbow pain can commonly be caused by tendinopathies. The most common tendinopathy around the elbow involves the common extensor origin, frequently known as tennis elbow, which has a prevalence of 1% to 3% [1]. However, tendinopathies to the flexor origin (‘Golfer's elbow’), distal biceps and distal triceps also occur. This two-part review article will looks at each of these four tendinopathies individually and provides a thorough overview of the current management principles. Before doing this, it is necessary to define the term tendinopathy.
Tendinopathies are evident at many sites, including the rotator cuff, patellar tendon and Achilles tendon. There is a common misconception that a tendinopathy is an inflammatory condition and this has been further exacerbated by the term tendinitis used to describe many of these conditions. Tendinopathy is a non-inflammatory condition characterized by ‘angiofibroblastic hyperplasia” [2,3], a combination of immature fibroblastic accumulation and neovascularization. This causes disorganized collagen matrix [4] and mucoid degeneration, producing the characteristic macroscopic appearance of a grey, amorphous tendon (Fig. 1). Inflammation is not a histological feature and so the term tendinopathy or tendinosis should be the preferred method of describing these conditions.
Fig. 1
Lateral elbow release demonstrating characteristic tendinopathy.
LATERAL ELBOW TENDINOPATHY (TENNIS ELBOW)
The first description of lateral elbow tendinopathy was by Runge, in 1873, where he described pain arising from the lateral aspect of the elbow. He called this ‘Schreibkrampf’, translated to ‘writer's cramp” [5]. It took another decade before Morris described the condition in the English literature using the phrase ‘rider's sprain” [6]. The following year, in 1883, Major [7] linked this to lateral elbow pain in lawn tennis players at which time the term Tennis Elbow was first introduced.
EPIDEMIOLOGY
Although tennis elbow is a common condition, the data on the epidemiology are relatively limited and out-dated. The incidence in the general population is 1% to 2% [8], which increases to 9% in tennis players [9], with novice tennis players being affected more than professional players. The peak incidence occurs in the fifth decade, with women of that age having a prevalence as high as 10% [8]. However, the overall prevalence remains equal in males and females. The incidence is notably higher in those who are manual workers [10,11].
ANATOMY AND PATHOGENESIS
The common extensor origin consists of the conjoined tendons of extensor carpi radialis brevis (ECRB), extensor digitorum communis (EDC) and, to a lesser degree, extensor carpi ulnaris. These attach along the anterior border of the lateral epicondyle. The extensor carpi radialis longus (ECRL) and brachoradialis originate more proximally. ECRB lies deep within the common extensor origin, adjacent to the lateral aspect of the capitellum and attaches to the base of the third metacarpal distally.
The articular side of the origin of ECRB is the most commonly cited region for degeneration to occur with the characteristic histology previously described seen here. Repeated loading of the common extensor origin with eccentric and concentric exercises causes microtrauma. An impaired or incomplete reparative response follows, leading to tendinopathy. Studies have demonstrated the presence of free nerve endings [12], increased glutamate [13] and substance P receptors [14] within the insertion of ECRB in patients with lateral (and medial) elbow tendinopathy. This may explain the pain experienced with this condition and why corticosteroid injections give short term relief. However, it is most likely that the pain is multifactorial and numerous studies have implicated intra-articular pathology in the form of plicae or synovitis as a source of pain [15,16].
CLINICAL FEATURES
Activity related lateral elbow pain of insidious onset is the most common history. The pain can radiate down the forearm, and patients often describe poor grip strength or difficulty carrying objects in the affected hand. Patients who develop symptoms from playing tennis usually complain of pain when performing a backhand. Biomechanical studies have shown that professional and novice players hit the backhand stroke differently. Professionals have their wrists in a position of extension with a likelihood of further extension in follow-through compared to novices hitting the ball with their wrist in flexion with a likelihood of further flexion [17]. This may be the reason why novice tennis players are more pre-disposed to tennis elbow because they eccentrically load their common extensor origin. It is important in the history to rule out other causes for lateral elbow pain. A history of paraesthesia or other sensory symptoms would suggest a peripheral nerve entrapment such as radial tunnel syndrome. Mechanical symptoms of locking or instability may point towards an intra-articular cause for pain.
Examination reveals localized tenderness at the origin of ECRB, just distal and anterior to the lateral epicondyle. A number of provocative tests have been described (Table 1) [18] based on resistance testing of ECRB. This is usually assessed by resisted extension of the middle finger (Maudsley's test) or resisted wrist extension with the elbow extended and wrist pronated (Cozen's test). Grip strength with the wrist in extension has been shown to be as much as 50% less than the non-affected side [19] and so is a useful adjunct in clinical examination.
Table 1
Provocative tests for lateral elbow tendinopathy
Differential diagnoses
Table 2 gives a list of possible causes of lateral elbow pain. Radial tunnel syndrome, in particular, can present with similar symptoms to lateral elbow tendinopathy. Provocative tests, as described above, can be positive in both conditions. However, the area of maximal tenderness is usually 3 cm to 4 cm distal to the lateral epicondyle. Sensory disturbance in the distribution of the superficial branch of the radial nerve may also be present.
Table 2
Differential diagnosis of lateral elbow pain
INVESTIGATIONS
A good clinical history and examination is usually all that is required to make a diagnosis of lateral elbow tendinopathy. In patients with refractory symptoms, further investigations can be useful in order to exclude alternative pathology. Magnetic resonance imaging (MRI) (Fig. 2) demonstrates a thickened and oedematous common extensor origin in approximately 90% of symptomatic patients [20⇓–22]. However, the severity of radiological findings does not correlate with the severity of symptoms [23]. MRI can also evaluate the lateral collateral ligament complex, the chondral surfaces of the radiocapitellar joint and the synovium to rule out coexisting pathology.
Fig. 2
Magnetic resonance imaging showing high signal intensity at common extensor origin consistent with lateral elbow tendinopathy.
Ultrasound can also be used in evaluating lateral elbow tendinopathy. The sensitivity has been reported to be approximately 80% with a specificity of 50% [24⇓–26]. The findings seen include tendon thickening, intrasubstance calcification and focal hypoechoic areas.
Plain radiographs are the least helpful in the diagnosis of lateral elbow tendinopathy. Subtle calcification in the common extensor tendon can be seen but is present in less than 10% of cases [27]. However, they have some use in ruling out bony pathology such as osteochondritis dissecans of the capitellum in younger patients or osteoarthritis in the older patient.
TREATMENT
Treatment choices for patients should be made with the aid of good quality, robust evidence. There is certainly no shortage of research in the treatment of lateral elbow tendinopathy, although the quality of this research has been questioned [28,29]. Cowan et al. considered the quality of 54,000 randomized control trials assessing the treatment of lateral elbow tendinopathy [29]. They found less than 10% of these trials could be considered as level 1 evidence, with > 87% rated as fair or poor on the Coleman Methodology Score. The deficiencies included poor descriptions of recruitment (> 90%), lack of power calculations (73%), poor/no randomization (58%) and inadequate blinding (90%). Therefore, there is no robust evidence to support or refute the majority of the treatment modalities presented here.
Lateral elbow tendinopathy management encompasses a large spectrum of both non-operative and operative modalities. It is important to remember that the natural history of lateral elbow tendinopathy is improvement of symptoms over a period of 12 months in approximately 80% of patients [30,31]. The reasons for this are unclear but most likely represent activity modification either in the workplace or recreational. Patient education is therefore an important part of treating lateral elbow tendinopathy.
NON-OPERATIVE MANAGEMENT
There is an abundance of evidence, many of it conflicting, for the potential non-operative treatment modalities. These include corticosteroid injections, physiotherapy, bracing, nonsteroidal anti-inflammatories (topical or oral), acupuncture, ultrasound, extracorporeal shockwave therapy (ESWT), autologous blood injection, botulinum toxin injection, platelet-rich plasma (PRP) injections, etc. The current evidence available for each is discussed below.
Nonsteroidal anti-inflammatory therapy
Nonsteroidal anti-inflammatory drugs (NSAIDs) have a role in the symptomatic treatment of most painful and inflammatory conditions including tendinopathies around the elbow. There have been a number of studies assessing the efficacy of NSAIDs in the treatment of lateral elbow tendinopathy, including over 10 randomized control trials and a Cochrane review [32]. Only one of these studies followed up patients for longer than 1 month.
Topical NSAIDs (diclofenac, diflam) have been shown to have benefit over placebo in reducing pain up to 4 weeks after the onset of symptoms [33⇓–35]. The results for oral NSAIDs, however, have been less convincing, with only one study showing reduced pain compared to placebo [36]. This may be explained by lateral elbow tendinopathy being a non-inflammatory condition and so improvement may be a result of the analgesic effects of the NSAIDs and not the anti-inflammatory effect.
Studies comparing NSAIDs with corticosteroid injection have demonstrated reduced pain in patients treated with injection. Hay et al. performed a randomized controlled trial assessing a single corticosteroid injection (20 mg of methylprednisolone and 0.5 ml of 1% lignocaine), naproxen (500 mg b.d. for 2 weeks) and placebo (vitamin C for 2 weeks) [37]. Outcome measures of patient's perception of benefit, pain and function were assessed at 4 weeks, 6 months and 12 months. Most patients had improved by 12 months regardless of which treatment group they were in, reinforcing the apparent natural history of lateral elbow tendinopathy. The naproxen group had no benefit over placebo at 4 weeks. The corticosteroid group showed significant benefit over NSAIDs and placebo at 4 weeks, although this effect did not continue to 6 months or 12 months with a small number of patients in the corticosteroid group relapsing compared to the other groups.
In summary, the evidence would support the use of topical NSAIDs in the symptomatic treatment of lateral elbow tendinopathy during the initial presentation. The use of oral NSAIDs, however, is not indicated and should be avoided, particularly considering the potential side effects they can cause.
Corticosteroid injections
All tendons can experience tendinopathy, with the Achilles, patellar, rotator cuff and common extensor origin being the most common. These are all characterized by non-inflammatory angiofibroblastic hyperplasia but, despite this, corticosteroid injections remain a common method of treatment.
A systematic review of the use of corticosteroid injection in tendinopathies showed improvement in pain in the short term but the effects were reversed at intermediate and long term [38]. This review included high-quality randomized controlled trials and so, of the 3824 trials that were identified, only 41 were included. There were 12 trials looking specifically at lateral elbow tendinopathy assessing slightly different interventiolns and controls. All studies showed a significant improvement in pain at 4 weeks to 6 weeks with corticosteroid injection regardless of the dose or volume of steroid, or whether local anaesthetic was given concurrently or not. However, the results by 26 weeks were worse for corticosteroid compared to no intervention, NSAIDs, physiotherapy or PRP injections.
Bisset et al. compared a single injection of triamcinolone and lignocaine with physiotherapy or a wait and see approach [39]. The corticosteroid group demonstrated significant improvement by 6 weeks but their symptoms relapsed after 6 weeks, with significantly poorer results compared to wait and see or physiotherapy at 52 weeks. The physiotherapy group showed some short term benefit over a wait and see approach at 6 weeks but no difference at 52 weeks. Smidt et al. compared three similar groups and found success rates of 92% for corticosteroid injection, 47% for physiotherapy and 32% for wait and see policy at 6 weeks [30]. However, the success rates at 52 weeks were 69%, 91% and 83%, respectively, showing a significant detrimental effect of corticosteroid injections at 1 year.
The type or dose of corticosteroid does not affect the results [40]. Patients having multiple injections versus a single injection demonstrate a worse outcome at 18 months with significant increased pain and a worse Verhaar elbow score [41].
In addition, a recent randomized controlled trial of corticosteroid injection versus placebo injection demonstrated detrimental clinical outcomes in the corticosteroid group at 1 year [42]. The same study assessed physiotherapy as a treatment and found no significant benefit.
In summary, there is very little evidence to support the use of corticosteroid injections in the treatment of lateral elbow tendinopathy as a result of the relapse rates seen at 6 months and 12 months.
PRP injections
The use of PRP injections has escalated significantly over recent years for treating many ligament and tendon conditions. Whole blood is centrifuged or filtered forming plasma that has a high concentration of platelets and endogenous growth factors. It is assumed (but not proven) that these growth factors, when infiltrated at sites of tendinous or ligamentous injury, facilitate healing. There is a slowly increasing body of evidence to support their use but studies are still relatively sparse. One of the advantages that PRP has is very few side effects and this factor has probably aided in its increasing use.
The ‘precursor” to injection of PRP was the injection of autologous blood. Autologous blood was shown to be more effective at 8 weeks than corticosteroid in a randomized controlled trial of 60 patients [43]. In addition, two randomized controlled trials have compared autologous blood with PRP. Creaney demonstrated successful treatment (an improvement in patient-related tennis elbow evaluation score of 25 points) in 66% and 72% of patients undergoing two ultrasound-guided injections of PRP and autologous blood, respectively, at 6 months [44]. The difference between the groups did not achieve statistical significance, despite a power analysis being performed, implying equivalent results with both modalities. In a smaller study of 28 patients [45], autologous blood was also shown to be equivalent to PRP at 6-month follow-up. There was a statistically significant better visual analogue score at 6 weeks for PRP but all other times demonstrated equivalence.
Two prospective case series [46,47] both showed significant improvement with injection of autologous blood at 6-month and 9.5-month follow-up, respectively. Connell injected autologous blood under sonographic guidance in 35 patients with refractory lateral elbow tendinopathy (symptoms greater than 12 months) [46]. Nirschl scores improved from a median of 6 to 0 at 6 months, with visual analogue scores (VAS) improving from a median of 9 to 0. Edwards demonstrated a similar reduction in pain and improvement in Nirschl scores in 28 patients injected with 2 ml of autologous blood [47]. Seven patients required two injections and two patients required three. Those receiving multiple injections for refractory symptoms made a full recovery with VAS and Nirschl scores of 0.
Peerbooms et al. performed a randomized controlled trial comparing PRP and corticosteroid injections [48]. One hundred patients were enrolled in the study having had symptoms of lateral elbow tendinopathy for greater than 6 months and were followed up for 1 year. There was a quicker improvement in the corticosteroid group but, by 8 weeks, both groups had similar improvement. The corticosteroid group reached their peak improvement by 8 weeks and their symptoms progressively deteriorated, whereas the PRP group continued to improve throughout the follow-up time period of 1 year. A successful outcome was considered as a 25% reduction in visual analogue score or Disabilities of the Arm, Shoulder and Hand (DASH) score without reintervention at 1 year. Forty-nine percent of the corticosteroid group and 73% of the PRP group were considered a success according to the visual analogue scores (p< 0.001), with 51% and 73% success, respectively, with regards the DASH score (p = 0.005).
Krogh et al. performed a randomized controlled trial comparing saline, glucocorticoid and PRP with a proposed follow-up of 12 months [49]. Only 16 out of 60 patients completed the full 12-month follow-up and so meaningful data was published for follow-up to 3 months alone. There was no difference in pain scores or patient related tennis elbow evaluation score at 3 months between the groups. A significant reduction in colour Doppler and tendon thickness in the glucocorticoid group was evident at 1 months and 3 months. Interestingly, of those who dropped out of the study, 26 patients chose to be treated with PRP, 12 with glucocorticoids, two with surgery, two with sclerosing agents and two declined further treatment.
Mishra treated 140 patients with PRP or a local anaesthetic injection [50]. Sixty percent improvement was seen with PRP compared to 16% in the control group at 8 weeks. As a result of the poor outcomes in the control group, the majority sought further treatment and so could not be followed up long term. However, the PRP group were followed up for approximately 2 years with 81% improvement in pain at 6 months increasing to 93% at final follow-up (p < 0.0001).
Taylor published a systematic review on the use of PRP injections in tendon and ligament injuries [51]. The review highlighted only three randomized controlled trials on the subject including the study performed by Peerbooms [48]. One trial assessed chronic Achilles tendinopathy, in which PRP was found not to affect outcome with regards pain and function compared to an eccentric exercise programme [52]. The other trial assessed the use of PRP in anterior cruciate ligament reconstruction which showed a mature (hamstring) graft signal on MRI at 1 year in 100% of the PRP group compared to 78% in the control group [53]. The review concluded that the evidence to support the use of PRP injections was, at present, limited to lateral elbow tendinopathy. The benefit of PRP over autologous blood, however, has not been fully established and presently demonstrates equivalent results.
Other injection modalities
A variety of other injection treatments have been tried for lateral elbow tendinopathy. These include botulinum toxin, prolotherapy (hypertonic glucose and local anaesthetic), sodium hyaluronate and polidocanol (sclerosing solution).
Botulinum toxin type A (60 Dysport units) injection appears to be beneficial in reducing pain up to 18 weeks compared to placebo injection [54,55]. Weakness of finger extension, particularly the middle finger, was noted in a significant proportion of patients treated with botulinum toxin, as would be expected, although this appeared to resolve by finalfollow-up. Keizer compared botulinum toxin injection to surgical release in a randomized controlled trial in 40 patients [56]. The results at 1 year demonstrated 65% and 75% success rates in the botulinum toxin and surgical release groups, respectively.
Prolotherapy involves an injection of dextrose with an extract of cod liver oil called sodium morrhuate. The mechanism of action is poorly understood, although animal studies have shown that prolotherapy may strengthen ligament and tendon insertions. Two randomized controlled trials have assessed prolotherapy versus a placebo injection [57] and corticosteroid injection [58]. Both studies had small study samples of 24 patients and poor follow-up. Scarpone et al. were unable to follow up the control group to 1 year but the prolotherapy group demonstrated 60% resolution [57]. Bearing in mind the natural history of lateral elbow tendinopathy is improvement in 80% of patients at 1 year, the use of prolotherapy as a treatment is not recommended.
Hyaluronic acid has been used effectively in a number of joints. Peri-articular use is less common but there is some evidence of positive effect in acute ankle sprain [59]. Petrella et al. performed a randomized controlled trial on 331 consecutive patients with symptoms for a minimum of 3 months comparing 1.2 ml of 1% sodium hyaluronate with 1.2 ml of saline placebo [60]. A significant improvement in pain and grip strength was seen at 30 days and this persisted to final follow-up at 1 year. Moreover, those in the sodium hyaluronate group were able to return to ‘pain-free and disability-free sport” by 18 (± 11) days, whereas none of the patients in the control group were able to do so. There have been no further studies on the use of hyaluronic acid despite the convincing evidence supplied by this study.
Polidocanol is a sclerosing agent that is injected into the part of the common extensor tendon showing increased blood flow. Zeisig assessed the use of polidocanol injection using either ultrasound or colour doppler versus a control group of lignocaine and adrenaline [61]. There was no difference in outcome between the groups at 1 year, with improvement seen in approximately half of patients. Again, the natural history of the condition brings these results into question.
Orthotics, acupuncture and physiotherapy
A number of trials have included counter-force braces as a treatment or as an adjunct to treatment for lateral elbow tendinopathy. A Cochrane review [62] concluded there is no clear evidence for the use of orthotic devices either as a treatment or an adjunct. Few studies showed long-term follow-up. In the short term (6 weeks), corticosteroid [63] and physiotherapy [64] were more effective at providing pain relief.
Acupuncture has been the subject of another Cochrane review [65] and it is likely to have some short-term benefit in pain relief but no difference at 3 months or 6 months. Further systematic reviews have agreed that there is some evidence to support its use in the short term for pain relief [66,67].
A variety of physiotherapy modalities have been used. These include exercise, stretching, eccentric loading exercises, friction massage, ultrasound and manipulation. Pienimaki et al. compared an 8-week exercise and stretching programme with pulsed ultrasound over a similar time period [68]. Although the numbers in the study were small (39 patients), there was a statistical significant reduction in pain, sleep disturbance and improvement in grip strength in the exercise group at 8 weeks. The intermediate or long-term effects were not assessed.
Eccentric loading exercises have been used as the mainstay of physiotherapy management of tendinosis, particularly patellar and Achilles, for a number of years [69]. A recent systematic review has, however, questioned whether isolating the eccentric component has any advantage over eccentric-concentric exercises [70]. Croisier performed a nonrandomized trial comparing nonstrengthening standard physiotherapy with isokinetic eccentric training in 92 patients [71]. Eccentric strengthening was better at reducing pain, maintaining strength and normalizing the tendon appearance on ultrasound. The physiotherapy regime was prescribed three times a week for a mean duration of 9 weeks. Long-term follow-up was not assessed. A further study compared a 6-week programme of stretching, concentric exercises or eccentric exercises, with no difference being found between the three groups [72].
A combination of exercise and manipulation has been shown also to be effective. The concept, ‘mobilization-with-movement” was first described by Mulligan in 1995 [73] and has been used to treat a number of conditions including lateral elbow tendinopathy. The technique involves applying and maintaining a manual force, in the form of a joint glide, to the lateral aspect of the elbow at the same time that a previously impaired action is performed, such as a grip. A number of studies have demonstrated immediate improved pain free grip strength using mobilization-with-movement [74,75], although these studies have not assessed whether this improvement is maintained. Kochar and Dogra, however, have shown improved grip strength and reduced pain at 12 weeks compared to placebo [76].
Pulsed, low-intensity ultrasound administered on a daily basis demonstrated no benefit over sham ultrasound at 12 weeks, with improvement in pain seen in 64% and 57%, respectively [77]. A systematic review also failed to demonstrate the efficacy of ultrasound as a treatment modality [66]. In the same review, phonophoresis (ultrasound with hydrocortisone gel) was shown to be of no additional benefit over ultrasound alone.
In summary, the efficacies of physiotherapy modalities have weak evidence and are unlikely to be the ‘solution” in isolation. Combined therapy of mobilization with movement appears to have the most encouraging evidence, although the intermediate and long-term results are lacking to fully support its use at present.
Electrotherapeutic interventions (laser therapy, extracorporeal shockwave therapy, pulsed electromagnetic field and iontophoresis)
Laser therapy has been used in musculoskeletal disorders for over two decades with mixed results. Laboratory trials have shown that laser therapy has both anti-inflammatory effects and biostimulatory effects on collagen synthesis [78], which likely explains the persistence in assessing its efficacy for musculoskeletal disorders. A number of systematic reviews have reviewed its use as a treatment for lateral elbow tendinopathy and concluded that there is no benefit from low-level laser therapy in the short or long term compared to placebo [66,79,80]. However, a more recent review looked at the different doses of laser therapy used and demonstrated that a wavelength of 904 nm targeted at the common extensor insertion (as opposed to trigger or acupuncture points) was effective at reducing pain and improving grip strength up to 8 weeks, particularly when used in conjunction with an exercise regimen [81]. It is interesting that systematic reviews using the same contemporary evidence can produce contradictory conclusions. For this reason, the use of laser therapy in the treatment of lateral elbow tendinopathy should be considered with caution.
It is not clear how extracorporeal shockwave therapy exerts its effects on soft tissues, although there has been a surge in interest in its use for conditions such as lateral elbow tendinopathy, plantar fasciitis, Achilles tendinopathy and rotator cuff tendinopathy. Following a number of trials and systematic reviews, including a Cochrane review, it has become apparent that its use in lateral elbow tendinopathy is not indicated [66,67,82⇓–84] as a result of no benefit over placebo in the short or long term. In addition, more side effects were seen with the use of ESWT including erythema and haematomas to the areas treated.
Pulsed electromagnetic field (PEMF) therapy has been used in the treatment of non-unions and, to a lesser degree, tendinopathies. Magnetic pulses are directed at the site of tendinopathy, which causes electrical signals that induce cellular repair. The evidence to support its use is sparse. Devereaux et al. performed a randomized controlled trial comparing PEMF with placebo showing no difference in pain reduction at 8 weeks [85]. Uzunca et al. performed a quasi-randomized controlled trial in 60 patients comparing PEMF, sham PEMF and corticosteroid injection [86]. The PEMF group had reduced pain at rest and on activity compared to the other groups at 3 months. However, the p-values have not been reported and so it is unclear whether the results they achieved are statistically significant.
Iontophoresis is a method of administering transdermal drugs through the use of electrical charges. NSAIDs and corticosteroids have been administered using this method for lateral elbow tendinopathy, although systematic reviews have been unable to endorse or refute its use [66]. Nirschl et al. were unable to demonstrate a significant difference in outcome between iontophoresis with corticosteroid solution and placebo at 4 weeks [87]. This was supported by Runeson and Haker [88], who found no difference up to and including final follow-up at 6 months compared to placebo. Grossi et al. were able to show an improvement in pain with high and low dose NSAID iontophoresis at 2 weeks compared to saline or sham iontophoresis [89]. However, the results beyond 2 weeks are unknown. The use of NSAIDs iontophoresis, on this level of evidence, can therefore not be supported, particularly given that there is reasonable evidence to support the use of topical NSAIDs.
Glyceryl trinitrate (GTN)
Animal studies have shown that inhibition of nitric oxide synthase results in poorer tendon healing, suggesting that nitric oxide is important for tendon healing and collagen synthesis [90]. Consequently, topical GTN patches have been used in the treatment of lateral elbow tendinopathy. Paoloni performed a randomized-controlled trial on 86 patients comparing a GTN patch (1.25 mg/24 hour) with a placebo patch [91]. Both groups also included a ‘tendon rehabilitation programme’. The results were encouraging with reduced pain at 2 weeks, 6 weeks and 12 weeks. By 6 months, 81% of patients had an excellent outcome compared to 60% in the placebo group (p = 0.005). The main concern to its use was the complication rate which caused the GTN patch to be discontinued in 12% (five of 43) of patients as a result of severe headaches or local dermatitis. McCallum et al. performed a 5-year follow-up study of these patients showing that there was no difference between the two groups at this time [92].
As a result of the relative success of the previous study, Paoloni et al. performed a further study assessing different doses of GTN patches (0.72 mg/24 hour, 1.44 mg/24 hour, 3.6 mg/24 hour) [93]. This study did not include a physiotherapy programme. The study was abandoned at 8 weeks because the only clinically significant result found was an improvement in pain using the 0.72 mg/24 hour patch compared to placebo, although the differences were marginal. They concluded that GTN patch can improve pain in the short term but its efficacy was likely to be related to a concurrent exercise programme. The side effects of headaches appeared also to be dose related.
There have been no further studies assessing GTN and, with the disappointing results of the last trial and the notable risk of side effects with GTN patches, it is likely that this treatment will not become more widespread.
Radiofrequency microtenotomy
Radiofrequency is used in many orthopaedic arthroscopic procedures to ablate connective tissues. It can also be used in a non-ablative mode to heat collagen-based connective tissue to promote a healing response. By causing a ‘thermal injury’, an inflammatory response follows with influx of inflammatory mediators including macrophages. The technique is commonly carried out as an open procedure under appropriate anaesthetic, although there have been reports of percutaneous and even non-invasive administration [94].
Radiofrequency has, understandably, been considered as a promising treatment for tendinopathies. Unfortunately, there is little evidence at present to support its use in lateral elbow tendinopathy, with only one randomized controlled trial published and two case series. Meknas et al. compared radiofrequency microtenotomy with surgical release and repair in 24 patients, demonstrating an improvement in symptoms and functional scores in both groups at 3 weeks, 6 weeks and 12 weeks and at approximately 1 year [95]. Grip strength was significantly better at 12 weeks in the radiofrequency group compared to the surgical group. They concluded that radiofrequency offers an alternative to surgery. Tasto et al. performed open radiofrequency microtenotomy on 13 patients showing significant pain reduction and improved functional scores throughout the 2-year follow-up [96]. Lin et al. demonstrated that radiofrequency could be performed safely using a percutaneous technique with good results in 34 patients with 6-month follow-up [97]. Both these studies represent case series with small numbers. Further trials are required to evaluate whether radiofrequency provides the ‘answer” to treating lateral elbow tendinopathy.
OPERATIVE MANAGEMENT
Surgical techniques for treating lateral elbow tendinopathy have been used since the 1920s [98]. There have been a multitude of different techniques described with many modifications applied to these techniques in an attempt to improve the clinical results achieved. The literature, in general, is littered with case series demonstrating good clinical outcomes for each technique used and, consequently, it is difficult to form meaningful conclusions regarding whether surgery is beneficial and, if so, which technique is the best.
A Cochrane review in 2002 [99] was unable to find any randomized-controlled trials assessing surgery for lateral elbow tendinopathy. In 2011, this review was updated [100] having found five trials with a total of 191 patients. It is encouraging to see that more randomized-controlled trials are now being performed, although also of concern that this only amounts to five, despite the procedure being around for almost a century.
Surgical options can be performed through open, arthroscopic or percutaneous techniques and are directed at the insertion of ECRB. Techniques described include debridement, release, repair, denervation and lengthening. Karkhanis et al. performed a quantitative review on operative management, showing that surgical techniques differ, to a certain extent, on the geographical location of the surgeon [101]. European surgeons favour a release, where as North American surgeons favour debridement as originally described by Nirschl [102].
Open surgery
Debridement
The most established open technique for surgical treatment is based upon the seminal paper by Nirschl and Pettrone, published in 1979 [3]. They were one of the first to recognize that the pathology appeared to lie within the ECRB tendon, describing the characteristic amorphous, greyish appearance of the diseased tendon. They retrospectively reviewed 1213 patients diagnosed with tennis elbow. Only 82 patients (88 elbows) required operative treatment with almost 93% responding to conservative measures. The operative technique involves a curvilinear incision over the lateral epicondyle. An incision is made between ECRL and the common extensor aponeurosis with ECRL reflected anteriorly to expose ECRB. ECRB is then incised and inspected with abnormal looking tendon being fully excised, including, if necessary, parts of EDC and ECRL (Fig. 3). In their series, they found it was generally necessary to remove approximately 75% of ECRB. A small arthrotomy is also performed to rule out intra-articular pathology. The anterior aspect of the lateral epicondyle is decorticated using an osteotome or multiple drill holes to improve blood supply. The interface between ECRL and the extensor aponeurosis is then closed.
Fig. 3
Lateral elbow release with debridement of extensor carpi radialis brevis and decortication of insertion.
Sixty-six (75%) patients were reported to have an excellent result, nine were good, 11 were fair and two were considered to have failed. Of note, only 51 patients underwent the technique described, with the other patients undergoing release of the extensor digitorum communis, as described by Hohmann [98], or a combination of EDC and ECRB.
A number of studies, only one of which was a randomized-controlled trial, have assessed the outcome following the Nirschl technique. The results varied marginally between studies with a successful outcome reported in a mean of 82% of patients [101]. Dunkow et al. performed a randomized-controlled trial comparing the Nirschl technique with percutaneous release in 45 patients with follow-up of 1 year [102]. Those undergoing percutaneous release were more satisfied, returned to work quicker and had better function than those in the open surgical technique group. Dunn et al. reported the long-term results of the Nirschl technique at a minimum of 10 years follow-up in 139 patients [103]. Three patients required revision surgery. Patient satisfaction was a mean of 8.9/10, with 93% returning to sporting activities. Meknas et al., as discussed earlier, performed a randomized-controlled trial comparing surgical release with radiofrequency microtenotomy [95]. The surgical technique was modified from the original description by Nirschl as they repaired the ECRB following debridement. The results of the study showed equivalent good results in both groups.
The Nirschl technique has been modified in a variety of ways to try and improve clinical results. The common modifications have been repair of the ECRB or common extensor origin [103⇓–105] and not decorticating the lateral epicondyle [106,107]. It is not possible to determine whether repairing the tendon improves outcome because there have been no randomized controlled trials assessing this. Khashaba, however, has performed a double-blinded randomized controlled trial on drilling of the lateral epicondyle when performing the Nirschl technique [107]. More pain and stiffness were seen in the drilled group. The trial only included 23 patients and p-values were not given despite a power analysis being performed, and so the results must be interpreted with caution. Das and Maffulli did not show a difference in outcome at 2 years between the two groups [106]. It is, therefore, also not possible to conclude whether decortication of the lateral epicondyle improves outcome.
Release with or without repair
Release with or without repair was first described by Hohmann [98]. A number of studies have advocated this technique with good long term results. For example, Verhaar et al. performed a lateral extensor release under local anaesthetic as an outpatient procedure in 63 patients [108]. The operative technique included infiltration with lidocaine and epinephrine (1 : 10,000). The common extensor origin was exposed and incised transversely and allowed to retract distally. Adequate release was confirmed by getting the patient to extend the wrist. Seventy-six percent of patients had no pain at 1 year, with 91% being pain free at 5 years. Karkhanis et al. evaluated the studies using the Hohmann technique, showing a wider variation in good outcomes (50% to 92%) than was seen with the Nirschl technique (79% to 85%) [101]. There is no clear consensus regarding whether repair of ECRB confers a better outcome as no study has compared this.
Other techniques
The pathology of tennis elbow has only really been established since 1979 [3] but surgical treatments have been present for many years prior to this. It is understandable that many of these surgical treatments have been founded on the pathology perceived to be the cause and so, consequently, they show mixed results. In 1955, Bosworth described four different approaches for the surgical treatment of tennis elbow in 27 patients [109]. These included division of the common extensor origin with or without excision of a synovial fringe between the radial head and capitellum or partial excision of the annular ligament with or without division of the common extensor origin. Although only four patients underwent partial division of the annular ligament alone, they all improved markedly and were considered asymptomatic. Bosworth concluded that the annular ligament plays an important role in the pain experienced by lateral elbow tendinopathy. Further studies have reproduced good results [110,111], although the surgical technique is rarely performed currently as a result of the change in attitude of the pathology of lateral elbow tendinopathy.
Radial tunnel syndrome is the most likely differential diagnosis for lateral elbow pain and so decompression of the posterior interosseus nerve (PIN) has been performed as a surgical treatment. Leppilahti et al. performed a randomized-controlled trial comparing decompression of the PIN with Z-lengthening of the ECRB in 28 patients [112]. The operation was considered successful in 50% of the PIN group and 43% of the ECRB group with an overall 25% reoperation rate. These results do not compare favourably with many non-operative treatments and other surgical techniques.
Percutaneous surgery
The results of percutaneous release of the common extensor origin were first published in 1982 with a retrospective case series of 35 patients with lateral elbow tendinopathy and six patients with medial elbow tendinopathy [113]. The results were considered excellent in 91% and unsatisfactory in 9% at a mean follow-up of almost 3 years. A number of retrospective case series have since been published with similar results [114⇓–116].
The only randomized controlled trial on percutaneous release was published in 2004 [102], comparing it with open surgery. Forty-five patients were randomized to open surgery or percutaneous release. Both procedures were performed under a general anaesthetic. The percutaneous technique involved a 1-cm incision being made at the lateral epicondyle. Artery forceps were placed behind the common extensor origin and this was then divided. The wrist was flexed to ensure a gap of approximately 1 cm was made. Statistically significant improvement was seen in patient satisfaction, time to return to work, DASH score and sporting activities in the percutaneous group. The percutaneous group improved faster and returned to work on average 3 weeks earlier than the open surgery group.
Arthroscopic surgery
The advantage of arthroscopic surgery in the treatment of lateral elbow tendinopathy is that it allows for intra-articular pathology to be evaluated and addressed. The pathology is present on the undersurface of the ECRB tendon and so this can be directly visualized following partial capsulectomy of the lateral margin of the radiocapitellar joint. Arthroscopic surgery, however, has a steep learning curve and its use in the management of lateral elbow tendinopathy is a relatively recent advance.
Baker and Jones classified arthroscopic changes seen in lateral elbow tendinopathy into three distinct patterns (Table 3) [117]. He also published, in a separate paper, the results of 42 consecutive arthroscopic tennis elbow releases with a minimum 2-year follow-up [15]. Some 95% of patients considered themselves to be ‘better” or ‘much better” following treatments, with a mean return to work of less than 3 weeks. Long-term follow-up of these patients reported continuing success of arthroscopic treatment at 10 years [118]. Interestingly, they reported intra-articular pathology in 69% of patients: synovitis/thickening (55%), bone spurs (12%), valgus extension overload (5%), loose bodies (7%) and degenerative joint disease (2%).
Table 3
Baker classification of arthroscopic findings of lateral elbow tendinopathy
Intra-articular pathology has been noted in most studies assessing arthroscopic intervention although the incidence varies significantly. Szabo et al. reported 44% of patients had intraarticular pathology [94], Latterman et al. reported 31% [119], and Owens et al. reported only 18.8% [120]. All these studies showed significant success in patients treated with arthroscopic surgery.
There have been no randomized controlled trials assessing arthroscopic surgery as an intervention but a few retrospective comparative studies are present in the literature. Szabo et al. compared open, arthroscopic and percutaneous surgery [104]. They reported good outcomes regardless of surgical techniques but felt that arthroscopic surgery had the benefit of addressing intra-articular pathology. Peart et al. performed a retrospective review comparing the results of arthroscopic treatment to open surgery with a minimum of 6 months of follow-up [121]. There was no significant difference in outcomes between the open and arthroscopic groups with good or excellent outcomes in 69% and 72%, respectively.
CONCLUSIONS AND PREFERRED MANAGEMENT
As clinicians, we strive to use evidence-based medicine to direct our medical and surgical management. There is certainly no shortage of evidence published on lateral elbow tendinopathy but the optimal management remains somewhat of an enigma. To quote a famous, and clichéd saying: ‘there is more than one way to skin a cat’. The treatment of lateral elbow tendinopathy can undoubtedly have this phrase ascribed to it. Further research will improve our understanding of which treatments are beneficial and, more likely, which treatments are not. The recent increase in randomized controlled trials has already started to provide some answers.
It is important to differentiate patients suffering from acute lateral elbow tendinopathy and those with recalcitrant symptoms. Those with acute symptoms should pursue a conservative management. Topical NSAIDs, activity modification and physiotherapy, using mobilization with movement, are likely to provide the best initial symptomatic treatment. If symptoms fail to respond to these simple measures then an injection should be considered prior to considering operative intervention. The evidence for corticosteroid injections clearly demonstrates that it is not helpful in recalcitrant lateral elbow tendinopathy and may in fact be detrimental. We would, therefore, advocate avoiding corticosteroid injections. Injection of PRP or sodium hyaluronate would appear to be the most reasonable choices given the good results seen in randomized controlled trials.
If symptoms persist despite injection therapy, then operative intervention or radiofrequency ablation should be considered. Further studies are required to assess radiofrequency ablation but the limited evidence, to date, has been encouraging. It is impossible to differentiate between surgical techniques because patient outcome is similar with open, percutaneous and arthroscopic surgery. It appears logical to perform arthroscopic surgery because the pathology is on the articular side of ECRB. It is also possible, with this technique, to evaluate and address intra-articular pathology. These advantages have to be balanced against the learning curve and additional operative time usually seen with arthroscopic intervention.
- Received March 4, 2013.
- Accepted May 14, 2013.
Nenhum comentário:
Postar um comentário