Historically the treatment for painful elbow ankylosis was a resection arthroplasty; resection of the painful joint was achieved by excising the entire olecranon, radial head and distal humerus above the condyles (Fig. 40.1A). Whilst in most cases this gave good pain relief, instability was common as there were no soft tissue or bony stabilizers remaining after surgery. Stability relied purely on the fibrous scar tissue between the cut bone surfaces. Despite this, good results were reported in posttraumatic, tuberculosis and rheumatoid arthrirtis.²^–⁴ Defontaine introduced the interposition as a refinement to the resection arthroplasty in 1887.⁵ By removing less bone and leaving a fulcrum between the olecranon and the distal humerus, he addressed the difficulty of instability experienced in resection arthroplasty (Fig. 40.1B). Tissue was positioned in the joint to act as a spacer, keeping the collateral ligaments functional and providing a pain-free articulation. The material used was varied and included muscle transposition, skin, fat or even pig’s bladder. Despite the use of an interposition graft, however, because less bone was excised, stiffness and even reankylosis could still be a problem. Hass cut the distal humerus into a wedge with the idea of reducing bone contact while still providing a fulcrum, however, pain, stiffness and wear of the articulation remained a persistent problem.⁶ Schüller first published the use of interposition in rheumatoid elbows in 1893. Hurri et al reported on 76 rheumatoid elbows and found that interposition arthroplasty provided better stability compared with the group of patients who had had a resection arthroplasty alone.³ However, their results also showed that only 40% were pain-free compared with 64% in the resection group. These were also the findings of Buzby, who concluded that patients were happier with a pain-free, flail elbow than they were with the more stable but painful interposition arthroplasty, the surgery often providing only a few degrees of extra motion.² Even in these early reports it became clear that movement was always achieved at the expense of stability, while movement with stability resulted in less predictable pain relief (Fig. 40.2).⁴

Figure 40.1 Resection arthroplasty of the elbow. (A) Extraperiosteal resection arthroplasty had high rates of instability. (B) The extent of the resection arthroplasty with the facial interposition. The ‘V’ cut maintains a fulcrum and is resurfaced with facia.

Figure 40.2 Relationship of range of motion, pain and instability during surgical release and debridement. As the range of motion is increased the elbow is subject to increasing instability (A). Gaining movement while maintaining stability can result in continuing pain (B).

In 1971, Peterson and Janes published the experience from the Mayo clinic, which identified a spectrum of pathology in the rheumatoid elbow; those with deranged mechanics and those patients whose primary complaint was pain.¹ In patients whose primary complaint was pain, a limited open debridement, synovectomy with radial head excision produced better results than interposition. The use of interposition arthroplasty has declined with the increasing success of total elbow arthroplasty, which gives better pain-free range of motion and stability in the low-demand patient.

It was not until the 1970s that Minami ⁷ and Kashiwagi⁸ published the first papers giving a detailed description and recommended treatment option for less severe elbow osteoarthritis. A procedure first described by Outerbridge and then published by Kashiwagi became known as the Outerbridge–Kashiwagi procedure. In 1992 this was modified by Morrey, who used a trephine to remove osteophytes encroaching on the olecranon and coronoid fossae, with elevation rather than splitting of the triceps, the so-called ulnohumeral arthroplasty.⁹ The column procedure approached the elbow via a lateral incision, debriding osteophytes and releasing the capsule through both anterior and posterior intervals. A more extensive debridement arthroplasty was described by Tsuge and Mizusek, which involved a formal disarticulation of the joint, with or without release of the collateral ligaments, excision of the capsule and reshaping of the radial head.¹⁰

As each refinement in surgical technique has evolved, so has our understanding of their limitations and the patients they will benefit. Each development has removed patients from the list that previously would have received a fusion or resection arthroplasty, leaving only a few for whom there is no alternative. Less invasive techniques with more predictable results now benefit patients who traditionally would not have been offered surgery at all. This chapter looks at open debridement and interposition arthroplasty and identifies those patients who will benefit from them. The preferred technique, with tips and tricks, is illustrated and the published results are reviewed.

Aetiology of pain stiffness around the elbow

Primary osteoarthritis

This accounts for only 2–3% of patients presenting with elbow arthritis.¹¹ It seems to almost exclusively affect men who are engaged in repetitive heavy manual labour and is presumably as a result of a genetic predisposition, followed by environmental stimulus for wear. The pathological changes which occur as the disease process progresses have been described.^12,¹³ The elbow forms osteophytes which occur on the coronoid, olecranon and fill their respective fossae on the humerus. Due to the high congruency of the joint, there is an early decrease in the range of movement. Pain can occur where the extra bone growth impinges on neighbouring normal bone or due to the development of a ‘kissing lesion’. The joint space, however, is initially maintained and the ulnohumeral articular cartilage is not worn.^10,^14,¹⁵ It is this characteristic which allows for the successful treatment with debridement of the extrinsic bone and soft tissue, leaving the relatively preserved ulnohumeral articulation alone. Osteophytes can break off and form loose bodies which cause locking as they interpose themselves within the joint. The ulnar nerve can also be irritated by the degenerative processes within the elbow and can often be a leading source of pain. Finally, ulnohumeral articular wear develops; this tends to cause pain which persists throughout the entire range of motion. Removal of osteophytes and soft tissue release will be less successful in these patients, as the ulnohumeral joint has intrinsic wear.

Posttraumatic osteoarthritis

This can be as a result of malunion after fracture of the articular surface, leading to incongruence and wear, or may be due to an osteochondral injury, which can also result in early degenerative changes. If there is significant intrinsic wear to the intra-articular cartilage of the ulnohumeral joint, open debridement may not be successful. Trauma can also lead to injury or attenuation of the medial collateral ligaments of the elbow. With time the resulting valgus laxity will cause posteromedial impingement with osteophyte formation on the posteromedial olecranon and humerus. This wear pattern will have identical radiological appearances to primary osteoarthritis and can be managed in a similar way with open debridement. In our experience, lorry drivers who repeatedly exert a valgus strain when tying down their load, and some athletes, such as boxers, weightlifters and throwers, can develop this pattern of osteoarthritis.

Heterotopic ossification

This is new bone formation within non-osseous tissues typically after elbow dislocation, with an incidence varying from 25% to 75%.^12,¹⁶ This high incidence is thought to be as a result of brachialis, which is mainly muscular as it crosses the joint, being torn at the time of dislocation. It also occurs after surgery and its incidence is increased in the presence of a concomitant head injury.

Inflammatory arthritis

This is a less common indication for open debridement. The soft tissues are not prone to contracture and osteophytes are not a common feature of the arthropathy. The wear pattern is more intra-articular in nature and therefore less suitable for joint preserving debridement surgery. Only 5% of rheumatoid patients have isolated elbow arthritis.

Clinical Pearl 40.2

Open debridement is a joint preserving operation and is contraindicated in the presence of significant ulnohumeral articular cartilage wear.

Presentation, investigation and treatment options

The decision to operate and which procedure to perform is reached following a careful history, examination and investigations. Some specific questions should be answered to help in the decision-making process.

History

The patient’s age and occupation is important. Find out why the patient has consented to treatment. While most patients will complain of both pain and loss of movement, one is often more of a problem than the other. Is the loss to range of movement functionally significant? Does it stop them doing their job? If so, document whether it is flexion or extension which is the main limiting factor. This will guide you into which soft tissues or bone needs to be debrided to give the elbow a functional range of movement (Fig. 40.3). Can the patient reduce the demands on the elbow or even change jobs?

Figure 40.3 The anatomy which blocks flexion/extension. (A) Bone and soft tissue which block flexion. (B) Bone and soft tissue which block extension. (C) Lateral margins of osteophytes.

Are there symptoms of locking, clicking or instability? Locking and clicking suggest loose bodies which can be easily removed without the need to proceed to a more extensive debridement. Has the patient got any distal neurological symptoms, such as sensory disturbance or muscle weakness? In this case the ulnar nerve will need to be addressed. Even in the presence of marked radiographic degenerative changes, when neurological symptoms are the main complaint, ulnar nerve release alone can adequately relieve symptoms.

If trauma is involved, the exact mechanism, the fracture pattern, and orientation and management is useful information. If it is an inflammatory arthropathy, other joints which are affected must be noted, as they can have an impact on the function of the elbow, along with the medications and current status of the disease process. Aetiology of the arthritis will give important information as to the condition of the bone and soft tissues prior to surgery. I specifically enquire about any history of previous infection.

It is also important to ascertain patient expectations. They must be realistic as the elbow is not going to be returned to normal. The literature suggests an increase in the arc of motion of between 22° and 35° can be expected and is dependent on patient compliance with postoperative physiotherapy. They must be made aware that pain relief after open debridement can be modest and the range of motion will deteriorate with time. If there is any doubt as to their commitment to the process then surgery should not be offered.

Summary Box 40.2

• Discuss ways in which the patient might manage at home or work without having an operation.

• Really get to the bottom of what is causing the main problem. Even in the presence of gross radiological degenerative changes, can you get away with reducing the level of pain with less radical surgery? That is, management of loose bodies, ulnar nerve or radial head alone.

• Realistic patient expectations and commitment to the rehabilitation are vital.

Clinical examination

There are several specific points that must be looked at in the examination prior to surgery. On inspection look for previous surgical incisions and the quality of the skin, consider whether your surgery can be approached through the old scar.

Is pain limited to the endpoint of movement, suggesting impingement due to osteophytes which can be removed? Or, is it throughout the arc of movement, suggesting intra-articular wear, which is less likely to benefit from joint preserving open debridement (Fig. 40.4). Does the endpoint have a solid, bony block to it, or is it soft and springy, which would suggest a soft tissue contracture.

Figure 40.4 (A,B) Radiographic appearances of intrinsic wear.

Investigation

Anteroposterior and lateral radiographs will help locate the osteophytes that need to be removed. Evidence of joint space narrowing will suggest excessive articular wear. Adequacy and the quality of the bone are evaluated. Loose bodies can often be seen along with calcification of the ulnar collateral ligament, the latter can be associated with ulnar nerve symptoms. The role of computed tomography (CT) has not been established, but is certainly helpful for identifying the position and extent of heterotopic ossification and loose bodies. I find a three-dimensional CT helpful in planning surgery. Magnetic resonance imaging can be used to evaluate ligament anatomy and stability. Finally, nerve conduction studies and electromyographic (EMG) sampling can be useful to exclude radiculopathy or pain of neurological origin.

Ultimately the decision to perform debridement is based on the disease process, anatomical culprit of the pain and stiffness, patient and surgeon factors. Table 40.1 summarizes the factors which the surgeon should consider in the management of elbow arthritis.

Table 40.1 Factors to consider prior to surgery

Patient factors	Age
	Activity level
	Willingness to modify activity level
	Poor rehabilitation potential
Anatomical factors	Articular wear
	Impingement spurs
	Loose bodies
	Ulna nerve symptoms
	Soft tissue contracture
	Stability
Disease factors	Primary osteoarthritis
	Inflammatory arthritis
	Trauma
	OCD in young patient
	Athletes
Surgeon factors	Training
	Experience
	Technical expertise

Summary Box 40.3

At the end of your assessment you should be able to answer:

• Is there any significant ulnohumeral wear which would be a contraindication to open debridement?

• What bone and soft tissue is causing the pain and stiffness, do you need to debride the anterior/posterior compartment or both?

• Does the radial head need to be excised?

• Does the ulnar nerve need transposing?

Treatment options

Next we must consider the approach and extent of the surgery which will be required to achieve the goals identified. There is now an accepted treatment algorithm for the management of elbow arthropathy which starts with simple measures such as analgesia together with modification of daily activities. Unfortunately, the patient group that typically experiences the degenerative changes is that of middle-aged manual workers who are usually the least able to modify their occupation. If non-steroidal antiinflammatory analgesia is insufficient to control the symptoms then intra-articular steroid injection can be helpful in reducing pain and swelling. As with all joints, however, the benefits are often short term and subsequent injections provide diminishing relief of symptoms. Physiotherapy can help maintain range of motion and occupational therapy can provide splints for pain relief or devices to help modify activities of daily living.

While total joint replacement has been adopted as the treatment of choice in most other joints of the body, the role of total elbow replacement in degenerative conditions at the elbow remains limited. Kozak reported complications in four out of five elbow replacements for primary osteoarthritis at 3 years.¹⁸ In addition, elbow arthritis again largely affects middle-aged men in manual work who wish to maintain their high demands on the elbow joint; they are obviously not good candidates for arthroplasty.⁹

Surgical debridement serves to remove painful osteophytes which impinge causing pain and loss of movement, along with a release of soft tissues to improve movement. This is only effective in patients with relatively well-preserved ulnohumeral cartilage. If there is excessive wear and joint-space narrowing the elbow will continue to be painful and total elbow arthroplasty or interposition arthroplasty would be the better option. Various techniques to debride the elbow can now be undertaken arthroscopically. Recent advances in arthroscopic surgery to the elbow have meant many patients with less severe arthritis can be treated in this way. However, it is contraindicated in the presence of extensive osteophytes, multiple pathology, previous surgery and uncertain anatomy. While there may have been some excellent reported outcomes, arthroscopic debridement is regarded by many surgeons as having an increased risk of complications,¹⁹^–²³ with no clear clinical benefit over open surgery.^24,²⁵ Open debridement of the elbow remains the mainstay of treatment for moderate to severe elbow arthritis.

No consensus exists among orthopaedic surgeons as to the most effective method of an open debridement and on which patients they should be performed. One operation does not fit all; each described procedure provides us with techniques which can be deployed on a patient-to-patient basis. The techniques are illustrated later in this chapter.

The column procedure is best suited to treat soft tissue contractures without bony impingement. It can be performed either through a medial or lateral incision. It makes sense to use the medial approach when the ulnar nerve needs to be decompressed; it also allows the surgeon to address any osteophytes in the medial gutter, while protecting the nerve and the anterior oblique bundle of the medial collateral ligament. The disadvantage is the limited access to the olecranon fossa and as a consequence sometimes inadequate removal of bone.

The Outerbridge–Kashiwagi (OK) procedure gives excellent exposure to the central posterior osteophytes and is therefore best suited to patients with bony impingement. It has the advantage of giving access to both the front and back of the joint through a posterior incision, without excessive soft tissue disruption. This is achieved by making a window through the olecranon fossa and reaching through this fenestration into the anterior joint. Reducing surgical soft tissue trauma in this way means there is less swelling which is important as bleeding and swelling after surgery turns into scar tissue which ultimately results in stiffness. Good results depend on early postoperative mobilization.

The Morrey ulnohumeral arthroplasty (UHA) is a modification of the OK procedure, which does not involve splitting the triceps, but includes performing a clean-cut trephine through the olecranon fossa. This has the benefit of less postoperative pain with less bone dust seeding into the muscle. Both the OK and UHA techniques are, however, limited by the failure to address any radial head pathology; furthermore adequate release of the anterior capsule is difficult to perform safely through the 15 mm window. By extending the incision and dissecting medial or laterally, the anterior capsule can be approached around the side, thus combining the UHA and column procedures.

An extensile debridement can be performed when more extensive hypertrophic changes are present. The joint is disarticulated by releasing the lateral collateral ligament complex and the common extensor origin, similar to the technique that can be used in total elbow arthroplasty. This gives the very best exposure through 360°, including the radial head, but at the expense of considerable soft tissue dissection. There have been no reports of postoperative instability in patients who have had their collateral ligaments detached in this way.

Contraindications for open debridement

The most frequent contraindication is in patients with poor rehabilitation potential. Some patients are easily identified, such as those with a history of drug and alcohol abuse. These patients are rarely compliant and, even after counselling, they rarely engage fully in their postoperative rehabilitation protocols and have a poor outcome as a result. Most others will comply to a varying degree; however, it must be made absolutely clear before surgery that rehabilitation will be painful and arduous. In addition spasticity or brain injury with limited voluntary control is a relative contraindication for surgical release; however, it may be required for hygiene purposes. Poor soft tissue coverage can make postoperative splintage hazardous, as wound breakdown will slow rehabilitation, and recurrent soft tissue contracture is sure to follow.

Patients with minimal osteophyte formation, ulnohumeral joint space narrowing on X-ray and pain throughout the full range of movement, have symptoms from ulnohumeral cartilage wear. Patients with this intrinsic wear pattern will not benefit from a joint-preserving operation and will be better served with an interposition arthroplasty.

The presence of metabolically active heterotopic ossification is a relative contraindication to early open debridement. The optimal timing for excision of the heterotopic ossification has not been established. A normal alkaline phosphatase is useful in patients who have large amounts of heterotopic ossification, typically around the hip. However, a relatively small amount of heterotopic ossification around the elbow results in only a small rise in alkaline phosphatase levels, which makes it an unreliable indicator of osteoblast activity. A three-phase technetium scan has been reported as being 90% sensitive for metabolically active heterotopic ossification and is therefore the preferred method.²⁶ Heterotopic ossification can absorb over time, especially in children who have a recovering head injury. There is some evidence that earlier release is preferable (6 months) as it gives better long-term results.²⁷^–²⁹ The use of preoperative radiotherapy and non-steroidal antiinflammatory drugs also decreases the risk of recurrence after excision.²⁹

Interposition arthroplasty is an option if the ulnohumeral joint has intra-articular wear and the patient is too young or active for a total elbow replacement. Interposition arthroplasty does not carry the same weight-lifting restriction as total elbow arthroplasty (5 kg lifting or 2 kg repetitive restriction) and has been shown to be durable in the active patient.^30,³¹ Alternatives are limited in these patients; arthrodesis severely limits the functional tasks that can be performed and is poorly tolerated. Resection arthroplasty creates severe disabling instability in most patients and is only indicated in those with active infection. Satisfactory results have been published for interposition arthroplasty in healthy, active patients with severe posttraumatic or rheumatoid elbow arthrosis.^30,³²^–³⁵

This salvage procedure, however, should only be considered when the patient is disabled by pain with loss of function and non-operative measures have been exhausted.³⁵ There is no age limit as a guide, but interposition arthroplasty is typically performed in patients with severe inflammatory arthritis (<30 years old) or posttraumatic arthritis (<60 years old). It is important that the elbow is stable as there is a significant association between poor outcomes and preoperative instability.^30,³⁵ For the interposition arthroplasty to have the best results there needs to be both bony congruence and soft tissue stability. The most common contraindication is insufficient bone stock which will increase postoperative instability due to lack of congruency. This can be restored by grafting the distal humerus as a staged procedure. An alternative is to use the calcaneal bone graft attached to the Achilles tendon at the time of surgery. Deformity which alters the mechanical axis of the arm will need to be corrected with an osteotomy if possible, as the pull of the flexors and extensors must be centred over the arthroplasty for stability. Ten degrees of varus or valgus is considered to be the limit. The interposition arthroplasty will not, however, provide prolonged pain-free stability if the patient is involved in very heavy manual labour which requires lifting with abduction of the shoulder. In these patients either a new occupation must be found or rarely an arthrodesis may be a better option. The sustained use of crutches or having to transfer from bed to a chair is also a relative contraindication.

As with any joint reconstruction there has to be active movement across the articulation; in this case the absence of the elbow flexors is an absolute contraindication. The use of dead interposition graft material means any active infection must be eradicated, typically free of infection for 6 months after the cessation of antibiotics. Preoperative assessment alone is not a sufficient indication to proceed to interposition; the articular cartilage must be assessed at the time of surgery after dislocation of the joint. It is important not to confuse the areas of thin or absent cartilage in the mid portion of the olecranon as being arthritic in origin. If there is insufficient wear to warrant interposition then an extensile debridement alone should be performed.

The principle of the surgery is to place scar tissue between the distal humerus and olecranon in place of the articular cartilage. In 1990 Morrey developed three additional features to the technique: distraction using an external fixator, early postoperative motion and a larger interposition graft.³⁶ Various interposition materials have been used in the ulnohumeral joint, including fascia lata,³⁵ cutis graft,³⁷ Achilles tendon allograft,³⁰ Gelfoam³⁸ and silicone.³⁹ The Wrightington experience has been with Achilles tendon allograft, which has the benefit of no donor site morbidity and an adequate thickness without having to suture several layers together. There is some evidence that it may survive longer than fascial interposition graft, leading to fewer patients requiring revision surgery in the early postoperative period.³⁰ The elbow is then held with a dynamic external fixator which is used as a means of initiating early motion while neutralizing forces and thus protecting the soft tissues, the interposed graft, and any ligament repair or reconstruction.^31,³⁶

Recent advances: the lateral compartment resurfacing elbow arthroplasty

Several studies have identified that the initial degenerative changes occur within the lateral compartment of the elbow, with the medial compartment often being well preserved. The lateral compartment resurfacing elbow arthroplasty (LRE) was designed for patients with primary osteoarthritis in whom symptoms had recurred following debridement and in whom the medial compartment joint surfaces were seen to be well preserved. Its indications were then broadened to patients with more advanced degenerative change secondary to rheumatoid disease and trauma. The principle is that resurfacing the articular surfaces of the lateral compartment will remove the painful articulation and off-load the ulnohumeral joint.

Initial results suggest that the LRE may lead to better outcomes than alternative procedures such as the Outerbridge–Kashiwagi procedure. Although long-term results are not yet known, it would appear that LRE may be an alternative to open debridement and has the potential to delay or even avoid total elbow arthroplasty.

Summary Box 40.4

• The treatment options discussed above allow for increasing access to the joint for open debridement.

• Debridement must be complete but aim to keep soft tissue dissection to access the joint to a minimum.

• Figure 40.5 provides a suggested algorithm for the management of degenerative changes in the elbow.

Figure 40.5 Treatment algorithm for the surgical management of elbow arthritis. OA, osteoarthritis; RhA, rheumatoid arthritis.

Surgical techniques and rehabilitation

Goals of management

Ideally, the arc of movement should be between 30° and 130° of flexion, with 50° of supination and 50° of pronation. This range of motion has been demonstrated to be sufficient for most activities of daily living.⁹ There is no doubt, however, that the loss of terminal flexion, which is required for bringing the hand towards the face, for feeding and grooming, etc., is a more significant disability than the loss of terminal extension. The second goal should be stability. Stability is more important for patients who do heavy lifting and manual activities, especially when the arm is internally or externally rotated. In this position maximum valgus and varus stresses are applied across the elbow and any instability will be apparent. The third goal is to extinguish pain. If surgery increases the range of motion significantly but the joint is too painful to move, the patient will not regard it as a successful operation.

Surgical technique – open debridement

Our currently preferred technique combines an ulnohumeral arthroplasty with a lateral column procedure. The reason for this being the difficulty of totally excising the anterior capsule through the 15 mm fenestration alone. The patient is placed in the lateral decubitus position, with the arm over a short post with sufficient room for the elbow to flex beyond 90°. The tourniquet must be positioned as high as possible to give maximum access to the elbow and inflated to 240 mmHg. A posterior incision is employed; it must be full thickness down to the facia overlying the triceps tendon. This decreases the risk of wound edge necrosis in patients with poor skin. The posterior incision is also preferred as it enables access to both medial and lateral sides. Curving the incision slightly to the lateral side allows even better access, moves the wound away from the point of the elbow and avoids the medial side where any scar tendons may cause problems when resting the elbow on a hard surface.

Ideally the dissection should be kept to a minimum; however, there should be a low threshold for decompressing the ulnar nerve. If the patient has had previous surgery it may have been transposed. However, a full exploration should take place before any dissection over the medial side of the elbow takes place. As a rule it is safer to look for the nerve in the more proximal virgin tissue, as it passes through the intermuscular septum, and trace it distally. Even in documented cases of previous transposition the nerve might have slipped back into its groove. A vascular loop protects the nerve by allowing retraction out of the operative field.

Using sharp dissection the triceps can be reflected off the side of the olecranon in order to aid exposure. This can be either lateral or medial to the triceps, depending on what else needs to be done, i.e. if the ulnar nerve requires release and the radial head is to be left alone then the medial approach is used. Conversely, if a radial head excision is necessary the lateral approach will be more appropriate. If more than 50% of the triceps tendon is reflected off the bone, a formal intraosseous repair of the tendon will be required prior to closure. Osteophytes around the olecranon can now be excised using an osteotome and the edges smoothed with rongeurs.

Sufficient exposure is required to accurately place the trephine in the centre of the olecranon fossa. Too distal and the trochlea will be thinned and insufficient bone will be removed to allow the olecranon to engage, too lateral and the trephine will cut into the capitellum as it breaks through on the volar surface and too medial will weaken the thin medial column. Extensive osteophytes and previous debridement can mean that accurate placement is difficult. In this instance it is best to perform a thorough debridement of the olecranon first and then extend the elbow which will then indicate the position of the fossa. The distal humerus flexes as it approaches the elbow, so the trephine should be angled 15–20° proximally front to back to avoid damaging the anterior articular surfaces (Fig. 40.6). Once the volar cortex is cut through the dowel is removed. The fenestration can be widened and the edges smoothed using Cloward up-biting rongeurs, which have the benefit of protecting the volar structures. They can also be used to remove excess bone from the anterior compartment.

Figure 40.6 Positioning of the trephine. Sufficient exposure is required to accurately position the trap line. The distal humerus flexes, so the trephine must be angled at 15–20° to avoid cutting into the volar articular surfaces.

A volar capsule contracture either causes loss of extension or forms as a result of it; either way the volar capsule needs to be released. The anterior compartment is safely entered through an incision made along a line extending from the lateral epicondyle to the equator of the radial head. Dissection anterior to this line will protect the lateral ulnar collateral ligaments. Sticking close to bone and using blunt dissection, the muscle and soft tissues are stripped off the lateral column and then the capsule, looking out for the posterior interosseous nerve which can run close to the radial volar capsule. In cases of revision surgery, the safest approach is to find the radial nerve in the virgin tissue proximal to the old scar as it passes through the intermuscular septum, then carefully dissect distally.

A Homan retractor can be used to lift the brachialis muscle off the capsule giving access across to the medial side, allowing a clear view to excise the entire capsule (Fig. 40.7). Full flexion and extension should now be achieved with gentle short lever arm manipulation. Any bone impingement or soft tissue contracture which may be blocking range of motion can be palpated by placing a finger front and back during flexion and extension. Bone wax can be used to seal the bleeding cancellous bone which is necessarily exposed during the operation. If more than 50% of the triceps has been lifted off the olecranon it must be formally repaired with osseous sutures. The use of a drain will reduce postoperative swelling, particularly if there is a large potential space under the posterior skin flaps.

Figure 40.7 (A–D) Exposure of volar capsule. A retractor is placed on the medial condyle lifting the extensor carpi radialis brevis and brachialis off the volar capsule.

Surgical technique – interposition arthroplasty

An indwelling catheter is sited to give adequate postoperative pain relief to allow continuous passive range of motion for 24–48 hours postoperatively. The patient is positioned and the incision is made in the same way as that described for the open debridement. If patients have had previous surgery, any scars should be incorporated into the incision to reduce the potential risk of skin necrosis. Once the full thickness skin flaps have been raised both medially and laterally the bony ridge of the ulna is palpated through the fascia. The soft tissue plane between anconeus and extensor carpi ulnaris (Kocher’s interval) can be utilized, although I prefer the interval between the ulna and anconeus. An incision is made 5 mm or so lateral to the ulnar ridge, exposing the anconeus muscle bellow. This 5 mm cuff of tissue allows easy repair of the fascia at the end of the operation. The anconeus is then stripped off the lateral face of the proximal ulna down to the supinator crest where the lateral ligament complex of the elbow inserts. The annular ligament is lifted off the supinator crest using subperiosteal dissection (Fig. 40.8A). If the radial head is still present and requires excision it can be delivered into the wound and removed. At this stage it should be possible to supinate the forearm, lifting the ulna off the trochlea. The articular cartilage can then be clearly visualized to confirm that the ulnohumeral articulation requires an interposition arthroplasty. If more than 50% of the surface is worn then interposition is warranted. Any worn areas should not be confused with the normal absent or thin articular cartilage in the mid-portion of the olecranon. Before proceeding to interposition arthroplasty the lateral side of the distal humerus must be exposed to allow access to the anterior compartment of the elbow. To achieve this, using sharp dissection and sticking close to bone, the anconeus, lateral ulnar collateral ligament and common extensor origin are lifted off as one big sleeve of tissue until it can be folded over the epicondyle. It should now be possible to see the volar capsule, which will need to be released to gain useful elbow motion. Careful dissection of the capsule away from the overlying tissue before excision will protect the posterior interosseous nerve from injury. If there is any doubt as to the safety of this, the capsule can be stripped off the humerus more proximally using a periosteal elevator. This will then allow extension without removing the capsule. However, it is likely that without excision of the capsule the anterior release will be less effective.

Figure 40.8 (A–C) Interposition arthroplasty. (D) Suturing the graft to the prepared distal humerus: (i) horizontal mattress sutures passed through the graft and the drill hole; (ii) transosseous sutures are used to hold the graft out to the edge of the prepared distal humerus. (E,F) Interposition arthroplasty continued.

Moving to the medial side, the ulnar nerve is identified proximally in normal tissue and dissected out distally. It is then lifted out of its bed so that the ulnar gutter can be approached with safety. If the anterior bundle of the ulnar collateral ligament has not been destroyed by the disease process, it can be clearly demarcated and protected as it is vital for postoperative stability. While leaving it attached makes the exposure for the rest of the operation more difficult, I believe it should be protected because instability is a leading cause of failure. The remainder of the medial ligament and scar tissue is removed from the ulnar gutter along with any excess bone in the olecranon fossa. The triceps can be lifted off the distal humerus with a retractor. There are usually large osteophytes here, partly due to the posteromedial impingement found in valgus laxity, and also because they are often left over after previous debridement, as they lie under the nerve and near the anterior ligament. These osteophytes along with the tip of the olecranon should be excised; this will aid in dislocation of the joint (Fig. 40.8B). The articular surfaces are then fully exposed.

The next step is to prepare the bone of the trochlea and olecranon. The aim is to have a smooth articulation of subchondral bone between the trochlea and the olecranon, leaving a 3 mm gap between the two to accommodate the graft. If too much bone is removed, this will result in increased wear. Alternatively, if there are any ridges there will be poor congruence and if sclerotic bone is left the graft will not adhere onto the trochlea (Fig. 40.8C). Using the technique described by Cheung and Morrey,³⁵ three or four drill holes are made from posterior to anterior across the distal humerus at the proximal edge of the articulation. Horizontal mattress sutures are passed through the Achilles graft and then the drill holes, with further sutures passed through the bone at the edges (Figs 40.8D and E). The joint is reduced and checked for range of motion and congruence. The lateral soft tissues are sutured to the lateral condyle and the annular ligament fixed back to the supinator crest with transosseous sutures. The range of motion is checked once more. Some elbows are stable and in these cases an external fixator is not essential, however, the fixator also distracts the joint, maintaining the gap while the graft integrates with the bone. For this reason we prefer to apply an external fixator in all cases (Fig. 40.8F).

Rehabilitation

Recurrent stiffness is the most common problem after open debridement or interposition arthroplasty. Patients must be made aware that the efforts of the surgeon are futile if they do not engage fully with the often painful and arduous postoperative rehabilitation. Muscle function around the elbow can also contribute to recurrent stiffness; factors such as length, tone, guarding and co-contraction can work against therapy. Therapy will also not achieve a greater range of motion than that obtained at the time of surgery; the idea that a small amount of contracture can be left and it will ‘stretch out with therapy’ is erroneous.

The formation of postoperative scar tissue (which can develop into contracture if left static) has three phases and an understanding of this will help with its management. In the inflammatory phase scar tissue forms from postoperative bleeding and oedema, which must be minimized. Postoperative bleeding is reduced by meticulous haemostasis prior to closure and the use of drains. Bone wax can reduce osseous bleeding, however, it introduces a foreign material which will create its own inflammatory reaction and increase the risk of infection.⁴⁰ Bleeding can be further reduced in the immediate postoperative period by applying a light compression bandage and elevating the arm with the elbow extended. The wounds should be dressed with a sterile non-adherent dressing and bulk must be kept to a minimum. The use of cryotherapy to reduce oedema also appears to help. However, it can be cumbersome and care must be taken to ensure that it does not interfere with early range of motion. The use of continuous passive mobilization (CPM) has not been established. The role of CPM is not solely to passively maintain the range of motion. Indeed, when used properly it has been shown to reduce swelling.⁴¹ The motion should be full so the elbow alternately flexes and extends to tighten the tissues, augment lymphatic drainage and prevent oedema. The CPM machine should be set at the range of motion that was achieved at the time of surgery and should not be stopped in the first 24–36 hours. The total end time has been shown to be more important than the number of repetitions, so the slowest speed setting should be used. It can be paused for 30-minute intervals at end range positions only.

Analgesia is usually achieved with continuous axillary block again for 24–48 hours. The expertise of the anaesthetist is related to the success of the block, however pain is a warning sign of tissue tearing rather than elongation. In my practice the CPM is maintained for 24–36 hours. It is not unusual, however, for patients to insist it is turned off in order to sleep the second night after surgery. This allows swelling and oedema to develop, and restarting the following day is often not tolerated. Patients are discharged with a customized static splinting programme. It is usual for the patient to have limitation of movement in one direction; either flexion or extension and splinting should be tailored to this. The night or resting splint should be employed in the direction of maximizing the end range, which is usually extension. During the day the elbow is splinted 10–15 minutes at the other extreme every 2–3 hours, and in between exercises and activities of daily living are encouraged. In the first 2 weeks the splint should be placed at the limit of active range of motion as passive stretch in this early period will increase swelling which will be counterproductive.

Fibroblastic healing is the second phase of maturation of scar tissue; collagen is laid down between 2 and 6 weeks after surgery. This collagen is not organized and can still be lengthened and any loss of motion since the CPM was stopped can now be regained. The tissues can now be stretched with passive customized splinting to the point of resistance or discomfort but not pain; to do this earlier (in the inflammatory phase) would result in oedema. Heat can be useful when applied at full tissue stretch, as it influences tissue extensibility and plastic deformity.

The third phase (6 weeks to 6 months) is the scar maturation or remodelling phase. Now the priority is to achieve active range of motion throughout the same arc as passive. The muscles and tendons should now be out to length, but may not exhibit power at the extremes of their new range. If the patient cannot power out to full extension or flexion, contracture will reoccur. Strength and endurance exercises assist in maintaining an equal balance between active and passive motion. The splinting programme should be continued, gradually weaning off day splints but maintaining night splints for 6 months or even longer. The duration of splintage is proportional to the duration of contracture.

Dynamic distraction with a hinged external fixator is used in patients who have had an interposition arthroplasty or those in whom there is some concern regarding postoperative stability. It permits active and passive motion, guiding joint range of motion with enough distraction to allow for healing of the interposition graft onto the new articular surfaces. There are two main types: the Universal Compass Hinge has a worm gear, which can be turned to aid in passive stretch after 2 weeks. The Mayo Dynamic Joint Distracter has no worm gear and relies on CPM, therapists and patient compliance for passive motion. While the worm gear has an obvious advantage, the Universal Hinge is bulky and as a consequence sometimes awkward to use.

Outcome including literature review

Open debridement

On review of the literature, most authors report modest improvement in pain relief and range of motion, with patients being generally satisfied with the procedure.^9,^10,^15,⁴²^–⁴⁶ Table 40.2 summarizes these data.

Table 40.2 Summary of the literature review

Survivorship

It does appear, however, that the range of motion gained at surgery deteriorates with time; this has been demonstrated in several studies. Minami et al reported a 10% deterioration in range of motion when they compared the same group of patients at 5 years after surgery and then again at 12 years.⁴⁹ Phillips et al extrapolated the best and worst scenarios for patient-perceived benefit over time.⁴² They predicted that 50% of patients would still be satisfied at 8 years after surgery. Wada et al measured a 24° increase in the arc of movement 1 year after surgery in their 33 patients.⁴⁷ However, at the latest follow-up 70% had had a deterioration, a mean of 7° loss between these two time points. The authors concluded that the disease process was progressive, and that patient satisfaction would deteriorate with time.⁴⁷ When range of motion at follow-up is compared across all the published data, there is a trend that would suggest that studies with the longest follow-up have the poorest range of motion. At our institution 59 patients were followed up for 7.6 years. We looked at the arc of movement at the end of surgery, at 3 months and at a mean follow-up of 7.8 years. Figure 40.9 shows our data along with all the other outcome studies for open debridement. It again demonstrates that the range of motion gained at surgery deteriorates with time.⁴⁰

Figure 40.9 Deterioration of range of motion with time.

Prognostic factors

Forster et al found that patients in whom duration of symptoms was less than 2 years, with considerable pain and one or more loose bodies benefited most from the open debridement.⁴⁸ Phillips et al⁴² and Antuna et al¹⁵ found no correlation between the functional assessment and reappearance of the bone with the fenestration of the olecranon fossa.

Recurrence of osteophytes

Oka et al ⁴⁴ studied the recurrence of osteophytes radiographically and found mild arthrosis in 75% and moderate recurrence of osteophytes in the remaining 25%. The spurs were less severe, however, than the preoperative arthrosis in all cases and none had required revision surgery at a mean of 5 years. Phillips et al⁴² found no correlation between deteriorating function and recurrence of bone with the fenestration of the olecranon fossa. However, Wada et al⁴⁷ reoperated on two patients who had lost motion as a result of pain and osteophyte recurrence.

Return to work

Oka et al ⁴⁴ included two populations: the 26 athletes returned to their sport and all 24 manual labourers returned to their occupation with only one having some limitation. Phillips et al⁴² reported 75% return to their previous jobs, 6% changed to a lighter work, 2% become unemployed and 6% retired. In the Wada series⁴⁷ all of the office workers returned to their work, while 76% of heavy manual labourers returned to the same job, 12% retired and 12% transferred to working in an office.

Interposition arthroplasty

Only a few papers have reported the results of interposition arthroplasty; it is not a common operation and the numbers are small. The extent of bone resection, the type of graft used, the role of ligament reconstruction and the use of a dynamic external fixation have not been optimized. The Mayo group has the largest series and are currently pleased with the use of Achilles allograft. However, the results of interposition arthroplasty are inconsistent with 24–30% of patients having a poor result, confirming that it is salvage surgery and should be considered only as a last resort.^30,^31,³⁶^–³⁸ ^,50 Eight to 15% required revision to total elbow replacement as a result of pain or instability; fortunately total elbow replacement after a failed interposition arthroplasty has good results. Instability is the leading cause for failure. Patients who have preoperative instability have a higher failure rate and as a consequence it should be considered a contraindication. Most studies report a reduction in pain after the interposition surgery but rarely achieve an asymptomatic joint; therefore patients with a painless loss of motion should not be offered this surgery.

Complications of open debridement

The literature demonstrates that there is a significant complication rate associated with open elbow debridement; between 12% and 19% of patients experienced an adverse incident or required further surgery. Forster et al ⁴⁸ reported a 17% complication rate (three ulnar nerve and one wound infection) with 8% requiring revision surgery; two patients had revision OK procedures and a third patient underwent an ulnar nerve release. Tashjian et al⁴⁵ reported 12%; one infection and one patient required subcutaneous transposition of the ulnar nerve. Antuna and Morrey recorded two acute neurological complications, two patients returned to theatre for manipulation under anaesthesia to maintain the range of motion within the first 8 weeks and a further two patients required revision debridement. In addition 29% complained of some degree of ulnar nerve symptoms postoperatively; they also found that untreated ulnar nerve symptoms had contributed to a poorer outcome at follow-up.¹⁵ As a consequence they concluded that the nerve should be aggressively evaluated prior to surgery. Phillips et al reported no infection or ulnar nerve damage.⁴²

In our series, 24% of patients had an adverse incident or required further surgery. One patient developed an anterior interosseous nerve palsy, which recovered over a period of 3 months. One patient developed myositis ossificans and another developed a superficial wound infection that responded to a course of intravenous antibiotics. Two of four patients who had bone wax used on the trephined bone developed deep wound infections, although both made a full recovery. As a result of this, we no longer use bone wax. Two patients had triceps tendon ruptures that involved exploration and repair of the torn aponeurosis. In one patient the rupture developed spontaneously while in the other it occurred during a manipulation under general anaesthesia. Two patients required revision ulnohumeral arthroplasty at 2 years and 3 years after the index procedure. Both patients had an unsatisfactory result for the second procedure. Due to persistent pain five patients underwent total elbow replacement surgery with satisfactory outcomes.