The hip is a deep ball and socket articulation allowing a wide range of motion. It is inherently stable. Co-ordinated locomotion depends on the intact function of muscles around the hip and the joint itself. Orthopaedic management, beginning at birth and continuing throughout life, is focused on preserving both these features.

The femoral head, roughly two-thirds of a sphere, projects into the acetabulum from the shaft of the femur by way of the femoral neck. The neck-shaft angle averages 125⁰ in adults and is anteverted an average of 12⁰. This allows a wide range of motion and by off-setting the shaft laterally, provides a mechanical advantage for muscles to act on the hip during locomotion. The head and its corresponding surface in the acetabulum is covered with articular cartilage which provides favourable gliding properties. A thick synovial and fibrous capsule encloses the head and neck. This confines the synovial fluid and provides further stability.

The hip derives its blood supply from three sources: (1) an extracapsular ring located at the base of the femoral neck, (2) ascending cervical branches of the extracapsular ring on the surface of the femoral neck, and (3) the artery of the round ligament. The extracapsular ring is supplied primarily by the posterior femoral circumflex artery, and ascending branch of the anterior femoral circumflex, both arising from the profunda femoris artery. The ascending cervical branches pierce the capsule and ascend the neck along the osseous surface. The vessels terminate by penetrating the head at the junction of the articular cartilage. This arrangement is critical for life-long nutrition to bone and overlying cartilage. The femoral head must gain its nourishment from vessels primarily coursing from distal to proximal along a path which is prone to disruption in fracture, dislocation, or surgical approach.

Although the muscular anatomy of the hip joint is complex, it can be described functionally to allow understanding of motion, as well as the basis for various surgical approaches. The quadriceps femoris and iliopsoas are the primary hip flexors and are supplied by the femoral nerves (L2, 3, 4). On the lateral aspect of the hip, the gluteus medius and minimus along with the tensor fascia lata control hip abduction. The inferior and superior gluteal nerves (L4, 5, S1) are the primary innervation. The gluteus maximus together with the posterior hamstrings, provide hip extension. Hip adduction is controlled by the adductor longus, magnus, and brevis as well as the obtorator externus. Innervation is via the obturator nerves (L2, 3, 4).

Hip motion is extensive being greatest in the sagittal plane (15⁰ extension to 140⁰ flexion) and less in the frontal (30⁰ abduction to 25⁰ adduction) and transverse planes (external rotation 90⁰ to 70⁰ internal rotation). Common activities require at least 120⁰ of flexion and 20⁰ of external rotation. Substantial forces act on the hip, even in simple stance phase.

Mathematical analysis of the hip joint allows understanding of anatomical, pathological, and mechanical relationships. Equilibrium of a body occurs when forces acting upon it are balanced. In a pivoting joint such as the hip, this occurs when clockwise and counterclockwise movements are equal. Movements are calculated from force times distance acting on a point. In the normal hip, the centre of body gravity times the distance from the joint must be balanced by the abductor muscle force times the distance between greater trochanter and hip. The joint reaction force is the overall force generated across the ball and socket in order to maintain the pivot in space.

Pain receptors about the joint detect alterations in the resultant joint reactive force, and gait is modified by shifting the centre of gravity to the affected side so as to compensate and restore an equilibrium situation. The joint reaction force can reach three times body weight during single leg stance, but can be higher or lower depending on body position. Surprisingly, strong forces can be reached during bed rest with activities such as bedpan positioning (four times body weight) or straight leg raising.

Deterioration of hip function stems from many causes. Congenital dysplasia of the hip with varying degrees of residual subluxation predisposes to secondary deterioration of the hip later in life because of abnormal stresses on cartilage and bone. Early detection and concentric reduction of stresses is the key; the potential for remodelling often produces a well-contained hip that functions well into later adult life. Residual dysplasia is often treated with pelvic osteotomy, femoral procedures, or combined femoral and acetabular reconstruction. In these situations later reconstruction may need to be considered.

Slipped capital femoral epiphysis as a child predisposes to osteoarthritis later. Particularly in the untreated case, slipped capital femoral epiphysis results in significant morbidity. It is often not for at least 20 years that the degeneration and dysfunction results. Legg - Calvé - Perthes' disease is the other major childhood process that predisposes to later osteoarthritis. Legg - Calvé - Perthes' disease typically presents itself between the ages of 4 and 8 with a male to female ratio of 4:1. The aetiology is similar to that of osteonecrosis, with a segment of the femoral head becoming dysvascular. The key to treatment is to maintain sphericity to the head and to preserve joint motion. Progression of disease is typically associated with articular incongruity and subluxation, both of which alter the normal biomechanics of the hip and predispose to premature degeneration.

Post-traumatic arthritis is most commonly the result of articular incongruity. However, direct chondral injury often is alone responsible for residual stiffness and degeneration despite restoration of the joint surface. Infection at any age can have a devastating effect on the articular cartilage and joint (Fig. 1) 2459.

Prompt drainage of an infected hip joint gives the highest chance for preservation of joint function. Increased pressure within the joint is directly injurious to articular cartilage. Elevated white cell counts produce high concentrations of degradative enzymes and lysosomal activity highly destructive to cartilage.

Rheumatoid arthritis is the most common inflammatory arthritis necessitating joint reconstruction. The female to male ratio is approximately 2 to 3:1 with a variable age of onset.

Osteonecrosis or death of bone within the femoral head often progresses to segmental collapse and, ultimately, to degenerative arthritis of the entire hip joint. Many events can cause osteonecrosis. In the post-traumatic situation vascular interruption is the key feature. Non-traumatic causes include steroid use, alcohol abuse, haemaglobinopathies (sickle-cell), and an idiopathic group. The development of osteonecrosis is probably a combination of mechanical and biological factors leading to blockage of the osseous microcirculation.

Staging in osteonecrosis is usually by the radiographic criteria of Ficat and Arlet ( Table 1 with schema 2467 or Table 1 without schema 617). The aetiology of osteonecrosis can also have a bearing on outcome. Tooke found that patients who continued on steroids had a 50 per cent incidence of progression of osteonecrosis despite core decompression versus 22 per cent in a group in which steroid therapy was discontinued. The experience of Maistrelli with intertrochanteric osteotomy was that those with idiopathic or traumatic aetiology did better than those whose osteonecrosis was alcohol or steroid induced. At 4-year follow-up less than 4 per cent had conversion to total hip arthroplasty. Based on long-term follow-up, it was concluded that in the otherwise healthy young patient osteotomy was particularly effective in early stages of osteonecrosis. Although many patients with advanced osteonecrosis ultimately come to total hip arthroplasty, success in this group does not seem to parallel that of other aetiologies. Many factors probably contribute to this observation including underlying aetiology, bilateral disease, body weight, and significantly younger age at presentation.

The nature of the pain is important, including location, severity, and duration. The differential diagnosis for hip disorders includes simple tendinitis/bursitis, and referred pain. Buttock pain secondary to a back problem is often erroneously attributed to a hip problem by the patient. Particularly in an older population, the potential for a metastatic lesion must be considered. Pain of abdominal origin is often attributed to the hip, as are renal colic, symptomatic abdominal aortic aneurysm, inguinal or femoral hernias, or a retroperitoneal process or infection.

Hip pain is typically perceived as anterior thigh pain with associated pain deep in the groin. Buttock pain is often also a complaint as is pain distally and even including the knee. The patient may be comfortable at rest, but experience - start-up - pain when commencing activities. The pain often worsens with prolonged ambulation or weight bearing. Ultimately as the process continues, the pain may exist even at rest and may result in significant sleep disturbance.

Particularly in the degenerative hip, progression of the process results in often subtle loss of motion. Initially loss of motion may be well compensated for, but eventually the patient has increased difficulty with activities such as getting in and out of a car or bed. Activities of daily living such as dressing, placing shoes and socks, and foot care become more difficult. Stair climbing may no longer be in a reciprocating fashion and the patient may relate increased reliance on banisters.

Changes in gait often reflect both decreased strength and range of motion. Patients often walk at a slower rate and typically shorter distances as the disease process advances. Weakness of the abductors manifests as a Trendelenburg-type gait. With a painful joint an antalgic gait also can be noted with a decreased interval of weight bearing on the affected side. If a very stiff joint is present, the patient may need to circumduct the affected leg so as to be able to advance the leg during the swing phase. Patients may notice a flexion contracture when they can no longer put their leg flat in bed.

In general, as the pain and stiffness progress, increased reliance on assistive devices is required. Modification in activity can often compensate for a variable interval of time. Non-steroidal medications are often very effective in early stages but with advanced degeneration increased use of narcotic analgesics is often required.

The orthopaedic physical examination should include several important features. Observation of gait can help in the assessment of the severity of discomfort and degree of limitation in ambulation. Stiffness, balance, and weakness all can be inferred from this observation. Range of motion should be assessed. The position of the hip and lower leg often indicates the degree of stiffness and irritability of the hip. The Thomas test should be performed to assess the presence and degree of hip flexion contracture. Both knees are brought to the chest and then the unaffected leg is held in this position while the affected leg is allowed to return to maximum extension. Incomplete extension reflects a flexion contracture. Pain with internal rotation is often indicative of early hip degeneration and frank loss of internal rotation is seen as degeneration progresses.

Leg length discrepancy is often seen with joint space loss or collapse such as in avascular necrosis. Both true and apparent leg length determinations should be made. With the patient supine, measurement from anterior superior iliac spine to medial malleolus defies true leg length whereas umbilicus to medial malleolus determines apparent leg length. Precise equalization of leg lengths is desirable in total hip arthroplasty though stability should be the primary concern. Significant inequality postoperatively can be a major cause of patient dissatisfaction.

Every effort should be made to incorporate prior incisions into new scars. Although skin slough is an unusual complication in surgery about the hip, wound problems and delayed wound healing can result in superficial and/or deep infection and an adverse outcome. Motor power about the hip should also be assessed and recorded. Abductor strength should be determined (side-lying abduction against gravity and against resistance). Weakness of abductors postoperatively is associated with an abductor lurch and is also associated with a higher dislocation rate secondary to muscle imbalance. A thorough neurological examination is essential, as is a check of pulse amplitude. Involvement of other extremities must be considered and will have a bearing on the reconstruction and postoperative rehabilitation.

Radiographic deterioration of the hip joint is ultimately manifested as loss of joint space, sclerosis, and migration of the femoral head (Fig. 2) 2460.

Depending on the aetiology of the disease process, earlier signs often can be perceived. In congenital disease of the hip, subluxation of the head can be progressive, and increased sclerosis in the weight-bearing dome is noted because of the increased concentration of stress. In an inflammatory arthritis such as rheumatoid arthritis, osteopenia and symmetric joint space loss without significant osteophyte formation are common. Early stages of avascular necrosis may show subtle changes in the sphericity of the femoral head. MRI has become an increasingly valuable tool in both detection and decision making in early avascular necrosis.

Attention in terms of reconstructive efforts often centres on the acetabulum and associated deficiencies. Deficiencies can be congenital, with inadequate bone stock superiorly, or secondary to prior surgery as is often the case in revision total hip arthroplasty. Close inspection of a standard anteroposterior pelvis radiographic view can often provide valuable data. Protrusio acetabuli is a common feature in rheumatoid arthritis where the osteoporotic bone cannot withstand the stresses of weight bearing and medial protrusion of the femoral head occurs. Radiographically this is perceived as migration of the femoral head medial to the ilio-ischial line.

Radiographs are important not only for the assessment of the extent of disease and/or bony deficiency, but also, when used with templates, are important in preoperative planning. Particularly in total joint arthroplasty, the suitability of an implant in a particular case can be made well before the actual procedure.

Radiographs of the hip usually include an anteroposterior pelvis view, and anteroposterior and true lateral views of the affected hip. A - frog lateral - view is helpful in assessment for qualitative changes in the femoral head or for subchondral lucency often seen in osteonecrosis. Similar to management of the fractured acetabulum, the 45⁰ oblique views of Judet allow assessment of potential anterior and posterior column deficiencies.

Degenerative arthritis of the hip is adequately evaluated with plain radiographs. Loss of joint space, osteophyte formation, subchondral sclerosis, and cysts are the most common findings. As the disease progresses, superior and lateral migration is not uncommon in the degenerative hip. Superior and medial migration may occur with rheumatoid arthritis and osteoporosis.

After total hip replacement is performed, follow-up radiographic examinations are necessary to monitor both the implant and the bony interfaces. Radiolucent lines between the cement and bone in itself are common. Concern about the stability and longevity of the implant exists when lucent lines progress, become continuous, and are longer than 2 mm. Plain radiographs usually are insufficient in the differentiation of septic versus aseptic loosening.

Arthrography is used as an adjunct to plain radiographs in assessing a painful arthroplasty. Performed at the same time as hip aspiration (for culture and sensitivity), subtraction studies are helpful in assessing loosening of the acetabular component.

Radionucleotide imaging is also useful in the evaluation of hip pathology. Technetium scanning is commonly used for detection of metastasis, infection, and osteonecrosis, especially when plain radiographs may be normal. At the present time, indium scanning in conjunction with sulphur colloid scanning is felt to be the most accurate means of detecting infection involving an arthroplasty.

Computed tomography (CT) and magnetic resonance imaging (MRI) have expanded the ability to evaluate osteonecrosis, neoplasm, and trauma about the hip joint. CT is the imaging modality of choice for evaluating acetabular fractures and osseous abnormalities of the hip joint. MRI produces superior soft tissue imaging and is the most sensitive means of evaluating osteonecrosis of the hip.

The mainstay of medical management, regardless of aetiology, is to alleviate the acute discomfort. Often resting the joint by decreasing the patient's overall level of activity and the use of an assistive device (crutch/cane) will allow the patient to continue to function at an acceptable level. If a systemic inflammatory arthropathy is the underlying process, rest, together with an anti-inflammatory agent, is often prudent. Assistance from a physiotherapist is often required to maintain muscle strength and range of motion. Additional modalities such as heat and ultrasound are helpful when an acute flare presents.

The patient should be encouraged to pursue non-impact types of exercise such as swimming, cycling, or walking, though often at a slower rate. An overall acceptable level of activity can often be maintained for a period of time. Weight reduction is generally difficult to achieve. An overweight individual with degenerative arthritis who loses weight often reduces symptoms. However, many overweight people often are too limited by the arthritic process to pursue a high enough level of activity to achieve this goal.

Non-steroidal anti-inflammatory drugs (NSAIDs) are an integral part of medical management of arthritis. NSAIDs pharmacologically inhibit cyclo-oxygenase, which enzymatically transforms arachidonic acid to prostaglandins and their derivatives. Additionally, NSAIDs can alter the lymphocyte response to inflammation and decrease granulocyte and monocyte migration and phagocytosis. Although effective, NSAIDs do not alter the underlying disease process. NSAIDs are expensive, especially in light of the extended periods of time for which they often are taken. The most common side-effect cited for all anti-inflammatories is gastrointestinal irritation. NSAIDs should be taken with food. The use of prophylactic antiulcer-type medications is often concurrent. Other frequently cited side-effects include hepatic, renal, and anticoagulant (primarily platelet adhesiveness) abnormalities. Patients who are on NSAIDs for an extended period should have haematological monitoring, including a complete blood count and measurement of renal and hepatic function. Checking stool for occult blood loss is also reasonable. If surgery is planned, the NSAID is discontinued approximately 2 weeks before operation to allow platelet recovery and to decrease the risk of haemorrhagic complications.

Severe flares of hip arthritis occasionally require the administration of immunosuppressive agents. The administration of corticosteroids should be supervised by a physician or rheumatologist. Corticosteroids are extremely potent agents, and the therapeutic window between efficacy and toxicity can be quite narrow. All members of the health team and patient should be aware of their potential long-term toxicity including, but not limited to, drug-induced osteoporosis, avascular necrosis, fluid imbalance, and hypertension.

The actual approach used for a particular reconstructive procedure depends on several factors. These include the surgeon's experience, the anticipated procedure, and often the need to extend the dissection if further exposure is required. Many resources, including atlases and textbooks, are available which detail these approaches and discuss the relative merits and risks associated with each. What follows is a brief overview of the most common approaches utilized for the hip.

The anterior approach, often referred to as the Smith - Peterson approach, is performed with the patient supine. Besides total hip arthroplasty, this approach is also commonly used for open reduction of pelvic fractures and pelvic osteotomy. Exposure of the acetabulum is more limited than with other approaches. The superficial dissection embraces the interval between sartorious and tensor fascia lata. Deep dissection is between rectus femoris and gluteus medius. The major neurovascular structure at risk is the femoral nerve if the dissection strays medially.

The anterolateral approach is a versatile approach allowing adequate exposure of both acetabulum and femoral shaft. It is often referred to as the Watson - Jones approach. Superficially, the fascia lata is incised posterior to the tensor fascia lata. The gluteus medius is then retracted posteriorly and the tensor fascia lata anteriorly, allowing visualization of the joint capsule. Various modifications have been described for more extensile exposure. These include osteotomies of the greater trochanter or releases and detachment of the anterior portion of the gluteus medius. The femoral nerve based in its lateral position in the neurovascular bundle is at risk primarily from retractors used to facilitate exposure. The femoral vessels can be injured by incorrectly placed acetabular retractors. If a trochanteric osteotomy is used, proximal exposure of the ilium is readily accomplished and useful if extensive acetabular reconstruction is necessary. Distally, exposure of the length of the femur is achieved by splitting the fascia of the vastus lateralis and reflecting musculature subperiosteally in a posterior to anterior direction.

The transtrochanteric approach is a variation on the deep surgical dissection used in the anterolateral approach. It is typically performed in the lateral decubitus position when used for arthroplasties. With anterior and posterior delineation of the abductor musculature, a Gigli saw, an osteotome, or an oscillating saw can be used to cut the greater trochanter, beginning at the vastus tubercle. Extensive exposure is achieved without muscle denervation and, reliable bony healing is achieved. The Dall - Miles trochanteric grip and cable system seems more versatile, stronger, and allows for better control of the trochanteric segment.

Complications of any trochanteric osteotomy or fixation technique include bursitis secondary to prominent hardware, non-union and proximal migration. Migration often leads to abductor weakness with a positive Trendelenburg sign and decreased walking endurance.

In 1982 Hardinge described a modification in which the medius is split anterior to the tendinous portion of the abductor insertion and is continued distally on the trochanter and vastus lateralis. This continuous segment of tissue is then elevated subperiosteally. The minimus is also split, allowing exposures of capsule and acetabulum. The hip is dislocated anteriorly and the leg is placed in a sterile sheet when in the figure of four position to allow femoral preparation. The anterior portion of the medius is reattached directly to bone as part of an anatomical closure. The greatest danger in this approach involves splitting of the gluteus medius too proximally with resultant injury to the superior gluteal nerve and resultant abductor paralysis and weakness. A transtrochanteric approach rather than the direct lateral approach should be used in situations such as congenital disease of the hip where pre-existing superior migration of the acetabulum exists, lengthening the limb greater than 2.5 cm (1 inch) is anticipated, or significant posterior column deficiency is present.

If extensile exposure of the acetabulum is necessary for bone grafting or other revision situations, the direct lateral approach is not recommended. One advantage of the direct lateral approach is that excellent exposure is achieved on both acetabular and femoral sides. Additionally, the short external rotators and posterior capsule are left intact affording excellent stability.

The posterior approach is versatile and is used for reduction and fixation of acetabular fractures, hip dislocations, drainage of infection, and hip arthroplasty. It is often referred to as the Moore or southern approach. With the patient in the lateral decubitus position, a linear or slightly curved incision is made centred over the greater trochanter. Superficially, the fascia lata is incised in line with the skin incision. Attention is directed to the posterior edge of the gluteus medius which is retracted anteriorly. The short external rotators are then taken off their femoral insertions and reflected posteriorly protecting the sciatic nerve. The posterior capsule is incised, opening the hip joint. In total hip arthroplasty with this approach, a complete capsulectomy is typically required to gain adequate exposure. The sciatic nerve is most vulnerable during this approach or in any dissection posteriorly. Neuropraxia is most common secondary to over-vigorous retraction. In revision surgery, the nerve may be encased in scar tissue. Many authors recommend defining the nerve more distal in non-traumatized tissue and then tracing it back in a retrograde fashion. The posterior approach is popular because of the speed with which it can be performed, the need for minimal surgical assistance, and its preservation of the abductors, allowing a more rapid return to normal muscle strength and gait. The biggest criticism of this approach is stability. The posterior capsule is thin; this weakness, along with detachment of the short external rotators, can place the hip at risk for dislocation, particularly if the patient assumes a position of flexion, adduction, and internal rotation in the early postoperative period prior to reformation of a hip pseudocapsule.

The integrity of the hip joint is based on a delicate balance between the normal mechanical stress placed on the cartilage and bone and the ability of those structures to accommodate to that load. When articular incongruity exists, regardless of the aetiology, the load to the articular cartilage significantly increases. It is, in effect, carried only by a reduced contact area. When subluxation of the femoral head exists, excessive and asymmetric loading of the cartilage is produced and uneven distribution of articular pressure results. These loads in turn result in bony sclerosis, fibrocartilage degeneration, atrophy of articular cartilage, and pseudocyst formation. Experimental and clinical trials show that surgical procedures can decrease the articular pressure by increasing this weight-bearing surface. The success of the intertrochanteric osteotomy depends on the precise alteration of the neck-shaft angle so that articular congruity is maximized.

Indications for intertrochanteric osteotomy include malunions and congenital coxa vara. If avascular necrosis has resulted in only a restricted area of head involvement, the patient also may be a good candidate for osteotomy. Some change in limb length can also be produced with osteotomy though this typically is not a primary indication. The most serious complication of all osteotomies is avascular necrosis. This is secondary to injury to or indirect stretching of the posterior retinacular vessels.

Patients beyond the age of 50 are not usually perceived as osteotomy candidates. A minimum of a 70⁰ arc of motion is necessary. The underlying aetiology for hip dysfunction should be considered. Patients with early osteoarthritis are good candidates whereas those with inflammatory or rheumatoid arthritis are not.

Radiographic evaluation of the osteotomy candidate should include pelvis, anteroposterior hip in extension/internal rotation to determine the preoperative neck shaft angle, and shoot-through lateral and anteroposterior views in abduction and adduction to simulate varus or valgus osteotomies. Varus or valgus osteotomies can be combined with extension or flexion alterations depending on these radiographs. Distal advancement of the trochanter is often performed with a varus osteotomy to prevent abductor weakness and a Trendelenburg gait. A flexion osteotomy is helpful if a posterior head deformity exists such as with slipped capital femoral epiphysis or in limited avascular necrosis.

An intertrochanteric osteotomy is a very demanding operation technically. Preoperative planning with radiographs and sketches is the key to successful planning as well as execution. The surgeon must be adept in the use of the blade plate device. This technique has drastically reduced the non-union rate and facilitates rapid mobilization of the patient. Although osteotomy may not completely relieve all symptoms, it can often provide a reasonable interval during which the younger patient can continue to function at a high level.

Conversion to total hip arthroplasty after osteotomy can be undertaken when function has deteriorated. Soballe found that patients who had undergone an osteotomy an average of 6 years prior to conversion to total hip arthroplasty had comparable pain relief and range of motion. However, osteotomy patients did have a higher rate of intraoperative femur fracture and this was correlated with the degree of displacement at time of osteotomy. All these hips were converted to cemented arthroplasties. Although the mean age at conversion was 63 years, conversion to a uncemented stem in a younger individual may present problems. It is critical to have excellent stability if an ingrowth prosthesis is used and this has not been adequately studied after osteotomy.

Osteotomy can be an attractive option in a younger patient with a symptomatic hip. It is reasonable to view the osteotomy procedure as a temporizing measure with the full expectation of conversion to total hip arthroplasty at a later, more suitable, interval.

Interest in arthrodesis of the hip has increased because of concerns about the longevity of total hip arthroplasty in younger patients and the relative lack of long-term data of function and outcome with uncemented ingrowth type prosthetic designs.

Arthrodesis offers more reliable results than osteotomy or total hip arthroplasty in a younger population yet it is least accepted by this same population. Descriptions of various techniques for arthrodesis date back to the late 19th century. Early criticisms of these procedures were based on a high rate of non-union, need for spica cast immobilization, and complications related to the long period of immobilization which was required to achieve union. Many of these concerns still exist. Techniques include extra-articular arthrodesis, intra-articular arthrodesis, combination, and various modes of internal fixation. There is no doubt that arthrodesis is far more difficult to perform than total hip arthroplasty. Few surgeons have more than limited experience with arthrodesis.

The ideal candidate for arthrodesis is a young man who is physically very active and wishes to remain so. He has no evidence for contralateral hip disease and is without prior history of infection. In such an individual arthrodesis can be an effective pain relieving procedure with long-term durability. Conversion to total hip arthroplasty can be considered at a later interval. Several papers cite high success rates for such conversions. Others have noted that the altered anatomy, muscle atrophy, and other factors produce inferior outcomes. Strathy looked at the conversion of 80 hips after either spontaneous or surgical fusion. Of those that had take down of surgical fusions, a 48 per cent complication/failure rate was noted at average 10-year follow-up. Individuals younger than 50 years who had a conversion or who had more than two prior hip procedures had a significantly higher failure rate. Enthusiasm for conversion of a hip that had undergone arthrodesis to total hip arthroplasty particulary in the younger patient must be tempered given these results.

The revolution in total joint arthroplasty began in the early 1960s with the work and techniques developed by John Charnley. Although principles have not changed, refinement in techniques, improved biomaterials, and many other factors have led to increased interest by both surgeons and patients in hip arthroplasty. Surgery is indicated when adequate pain relief or function cannot be achieved by non-operative means. Most patients have failed conservative therapy, including various non-steroidal anti-inflammatory agents, use of a cane, weight loss, and modification in their overall level of activity. Patients must be well informed of the inherent risks as well as the benefits of the proposed procedure.

A careful history, physical examination, and review of radiographs is essential. The surgeon can determine the particular needs of each patient such as the optimal surgical approach, possible need for bone grafting or supplemental sources of bone (allograft) implant selection and sizing, equalization of leg lengths, and special pre- and postoperative considerations (blood donation, cell-saver, prophylaxis for deep vein thrombosis, or special monitoring devices). The choice of implants available to the surgeon is enormous. Both off-the-shelf and custom implants are readily available. Templates aid the surgeon preoperatively in deciding type and size of the most appropriate implant. Preoperative planning assures availability of appropriate materials at the time of surgery.

Acetabular components are now metal-backed shells with high density polyethylene liners. Metal backing decreases deformation of the cup thus reducing polyethylene wear, and also allowing more uniform distribution of stress to the cement mantle. Both these factors are considered to decrease the risk of loosening and increase component longevity. Many cups are of two-piece construction, composed of an outer metal shell with an inner polyethylene liner. This modularity allows the use of variable femoral head sizes. Additionally, many liners have an overhang of polyethylene up to 20⁰ that can be used to compensate for some degree of deficient bone stock or cup malposition and thus to augment stability.

Although long-term studies have shown reasonable survival rates for cemented acetabular components, concerns about late loosening has led to greater interest and use of porous coated components. The concept of ingrowth dictates a metallic outer surface with beads or fibre mesh of appropriate size so that host bone grows into this surface creating long-term biological fixation. Initial stability of the cup is based on an interference fit reaming the acetabulum precisely to the identical diameter of the cup. The cup is then impacted or press-fit into the prepared bony bed. Early stability is often augmented by the use of peripheral pegs to provide rotational stability or screws which can be placed through the cup into the superior weightbearing dome. Studies of the used prosthetic cup have questioned the extent of bony versus more common fibrous growth, yet these cups appear radiographically stable and provide excellent pain relief.

Morse taper heads increase options at surgery. This method allows adjustment of neck length and overall limb length. Heads from 22 to 32 mm in diameter are generally available. Larger diameter head sizes provide greater stability. Smaller heads have lower acetabular loosening rates and also allow for a thicker polyethylene liner. The optimal head size is probably 28 mm.

The choice of femoral component is probably the most controversial topic in total hip arthroplasty. With a cemented femoral component, two interfaces exist; the bone - cement and the cement - prosthesis. The uncemented or ingrowth component has a single bone - prosthesis interface. The amount of stress on each of these materials is directly related to its modulus of elasticity or stiffness. All femoral components, whether cobalt - chrome or titanium, have a modulus much greater than cortical bone and will result in some degree of proximal stress shielding. In cemented applications use of a smaller component with a relatively larger cement mantle transfers more stress to the cement. Similarly, a larger stem size has an increased diameter and results in a stiffer implant again with a modulus greater than native bone.

Significant improvements in cement technique have led some to believe that so-called hybrid hips (ingrowth acetabulum/cemented femur) provide the most predictable results and that non-cemented femoral applications should be used only in a physiologically younger patient or in revision situations. Success of ingrowth applications requires establishing a stable interface at the time of insertion. Most manufacturers have approached this goal with designs anatomically mimicking the internal contours of the proximal femur and/or maximizing proximal fill. With few exceptions the proximal portion of the component is porous coated because of concerns including proximal stress shielding, potential difficulty in removal, and metal ion toxicity. Reasons for the high incidence of thigh pain with ingrowth femoral components have yet to be identified and the exact aetiology is not known. Non-physiological stress concentration distally and acute cortical abutment may be factors.

Prior to dislocation of the hip, a superior retractor is placed in the outer table of the ilium, the position of the leg marked, and a leg length measurement made. The hip is dislocated and a provisional resection of the femoral head and neck made typically 2 cm above the lesser trochanter. Attention is then directed to the acetabulum, where the labrum is excised and the capsule is released subperiosteally about the acetabular rim to facilitate exposure. A large curette is used to remove soft tissue, residual cartilage, and osteophyte from the fovea so that the true floor of the acetabulum can be identified. The acetabulum is concentrically reamed and is enlarged, creating a hemisphere with punctuate bleeding bone. The component is then driven into the acetabulum with the aid of an alignment guide. Ideal cup position is approximately 40 to 45⁰ of abduction and 10 to 15⁰ forward flexion. The polyethylene liner is then snapped into place.

The femoral technique varies with the type of prosthesis being used. The medullary canal is first cannulated with a Charnley awl. Maintaining axial and valgus alignment, rasps are driven down the proximal canal in appropriate anteversion. When optimal fill is achieved a head - neck combination is snapped on to the trial rasp/component and a trial reduction performed. Leg length and stability is assessed and adjustments made as needed. Stability should be assessed in maximum extension/external rotation, throughout the mid-range and in flexion greater than 90⁰ of adduction and internal rotation. Satisfied that a stable construct has been achieved the trial femoral component is removed. For cemented application, debris is removed with a toothbrush-type device and the medullary canal further cleansed with a water pik-type device. The medullary canal is then plugged approximately 2 cm distal to the anticipated end of the prosthesis. Chilled cement is either vacuum mixed or centrifuged to eliminate voids and introduced into the medullary canal from distal to proximal with a caulking gun. A femoral pressurizer is used to improve cement intrusion, after which the prosthesis is inserted in the appropriate version, seated at the level of calcar resection and held until the cement has cured. A centralizer placed on the distal end of the prosthesis allows for a more uniform cement mantle and prevents inadvertent prosthesis positioning in varus or valgus. The hip is re-reduced, and the joint is irrigated to remove any residual debris.

The need for revision hip surgery continues to grow as patients live longer; arthroplasties are performed at earlier and earlier ages. Revision is most commonly performed for aseptic loosening, prior infection, recurrent dislocation, and or accelerated osteolysis. Whereas a hybrid index arthroplasty often has up to 95 per cent success rate at 10- to 15-year follow-up, revision surgery has not achieved such striking success. The primary contraindication to performing a revision is ongoing infection.

Revision surgery is very demanding and carries significantly higher risks than initial arthroplasties. An aspiration/arthrogram should be performed preoperatively to rule out infection. An arthrogram can help determine the degree of radiographic loosening, especially of the acetabular component. Continuous radiolucent lines, fracture of the cement mantle, and migration of the component indicate loosening. Revision is most commonly performed in the lateral decubitus position via a direct lateral or transtrochanteric approach, allowing ample exposure of the entire acetabulum and facilitating exposure for bone grafting. A complete capsulectomy is performed. Use of cement chisels and curved acetabular gouges typically frees the polyethylene from the loose cement mantle. A curette is then used to remove residual cement and membrane. The defect is assessed and bone grafting performed as needed to restore the acetabulum to its normal position. Unless a substantial surface area of acetabulum is grafted with allograft, an ingrowth acetabular component is used.

Revision of the femoral component is often more difficult than the acetabular side. The proximal femur must be cleared of all scar tissue and also the capsule so that the interfaces can be well delineated. The trochanteric osteotomy greatly facilitates exposure and visualization. Cement laterally adjacent to the trochanteric bed should be removed. An extractor can be placed about the neck of the prosthesis and a slap hammer used to facilitate removal. The use of hand cement removal instruments is recommended for residual cement. When there is significant remaining distal cement, high speed drills can be used to penetrate the cement mantle. Care should be used to avoid cortical perforation. It is important to remove all cement as residual fragments can cause eccentric reaming during preparation for the revision component and potentially result in a distal fracture.

Infection ranks as one of the most potentially devastating complications in orthopaedics in general and in total joint arthroplasty in particular. Preoperative attention should be directed particularly to skin integrity, condition of teeth, and urological function. These and any other potential sources of infection should be rectified prior to proceeding with the proposed reconstructive procedure. Infection rates of approximately 1 per cent for index arthroplasty are typically quoted. Revision surgery for documented aseptic loosening carries a 3 to 5 per cent range of infection and is attributed to the increased length of procedure, need for extensile exposure with greater soft tissue dissection, and possibly use of allograft material. Additional techniques used to enhance sterility include laminar flow, independent exhaust suits, and ultraviolet lights. These techniques, though effective, often are too costly for smaller institutions.

Salvage of an infected arthroplasty can be a formidable task. Goodman reviewed a 12-year experience with patients who had undergone total hip arthroplasty and found an infection rate of 3.2 per cent. In a large series from the Mayo Clinic, the authors noted a 13 per cent failure rate with two-stage reconstruction. The presence of retained cement at resection arthroplasty, a shorter interval between reimplantation, and a virulent organism treated with less than 28 days of intravenous antibiotics were factors associated with recurrent infection.

The diagnosis of infection in patients with a painful hip is difficult. Although an aspiration clearly is the most reliable means of detection, a negative culture does not rule out sepsis. Conflicting reports exist concerning the efficacy of technetium and/or gallium scans as aids in the diagnosis of sepsis. Initial studies show a very high level of accuracy with indium-labelled white cell studies especially with sulphur colloid scans. An accuracy of 95 per cent has been reported.

Heterotopic bone formation is a not uncommon finding associated with reconstructive surgery about the hip. The actual incidence of extensive heterotopic bone formation which impairs hip range of motion is significantly less. Prevention of heterotopic bone formation must be instituted prior to its radiographic appearance and typically within days of the operative procedure. Prophylaxis includes radiation therapy or non-steroidal anti-inflammatory medication. Ayers demonstrated 10 Gy administered over 5 days was effective. Radiotherapy is thought to interfere with the process by which pluripotential mesenchymal cells differentiate into osteoblasts. The concern with radiation therapy is the potential for radiation induced malignancy, particularly with current enthusiasm for total hip arthroplasty in a younger population. Single-dose treatment limited to 7 Gy is also effective.

Indocin has been shown to prevent or diminish heterotopic bone formation. However, non-steroidal medications may also adversely effect bone ingrowth especially in the early postoperative period. Non-steroidal medications also interfere with anticoagulation levels of coumadin commonly used as prophylaxis for deep venous thrombosis. The Hospital for Special Surgery reported a 37 per cent incidence of complications or adverse affects associated with the use of indocin in total hip arthroplasty. The surgeon needs to weigh these risks and benefits when choosing any prophylactic technique for the prevention of heterotopic bone formation.

Orthopaedic surgery in general and hip reconstructive surgery in particular is associated with a significant risk for the development of deep venous thrombosis. Factors that are associated with a high risk for the development of deep venous thrombosis include an age of more than 60 years, prior history of deep venous thrombosis, obesity, and malignant disease. Pulmonary emboli are lethal in 2 to 3 per cent of these patients. Despite the alarming prevalence of deep venous thrombosis, many orthopaedic surgeons do not employ screening techniques or provide effective prophylaxis for their patients.

Multiple techniques for the prophylaxis of deep venous thrombosis are available. The most commonly used include thigh high-gradient elastic stockings, serial compression boots, adjusted dose heparin, and coumadin. Coumadin is probably the most efficacious means of prophylaxis. At a prothrombin time of 1.3 to 1.4 times control, bleeding complications can be reduced to approximately 4 per cent without reducing efficacy. Amstutz reported a 0.5 per cent incidence of pulmonary embolism without fatality and a bleeding complication of only 1.5 per cent using this regimen in 3000 total hip arthroplasties. Contraindications to the use of coumadin include major liver disease, coagulopathy, and history of peptic ulcer disease or gastrointestinal bleeding.

Another option for high-risk patients or patients who have failed coumadin therapy is the use of vena caval filters. Filters can be readily placed under fluoroscopic guidance in a percutaneous fashion. They effectively prevent significant pulmonary embolism and have a high patency rate with a very low morbidity. They do not prevent the formation or propagation of clot in the lower extremity and thus do not lessen the potential for postphlebitic syndrome and other sequelae of deep venous thrombosis.

Serial compression boots are a very attractive means of providing prophylaxis. They are safe, easily applied, well tolerated by most patients, and are virtually devoid of complications. They reduce the incidence of deep venous thrombosis to a more acceptable range. Paiement showed a comparable incidence of deep venous thrombosis between two groups randomized to coumadin versus serial compression boots. However, venography showed that the boot group had a higher number with proximal femoral vein thrombosis, which is thought to have a higher propensity for embolization. Regardless of the prophylactic technique used, the onus is on the orthopaedic surgeon to provide patients with effective protection from venous thrombosis and its associated morbidity and mortality.

Detection of deep venous thrombosis represents an area in which vast advances have been made in recent years. Venography, though generally considered accurate, has many drawbacks. It is often painful, not readily repeatable, labour intensive, and can induce phlebitis. Indium-125 fibrinogen scanning and cuff-impedance plethysmography represent two non-invasive techniques but all are being replaced by ultrasound examination of the veins.

Aseptic loosening is an increasingly common problem confronting the reconstructive surgeon. On the acetabular side, loosening of non-metal backed cups is quite often seen, particularly 10 years after index arthroplasty. Clinically problems on the acetabular side are typically manifested by groin pain. Component undersizing and malalignment, poor cement technique with an irregular cement mantle, and deficient bone stock all contribute to higher rates of loosening. Polyethylene wear with the generation of wear debris also contributes to accelerated loosening on the acetabular side. On a microscopic level, histiocytes and macrophages predominate with a typical foreign body-type reaction. Radiographically, serial radiographs which demonstrate progression of radiolucent lines indicate loosening. Frank migration of the cup is diagnostic of loosening. Arthrography can add additional information and aspiration is important to exclude occult sepsis.

In cemented total hip arthroplasty, loosening of the femoral component is more common in the early years compared with wear on the acetabular side. Stress shielding of the proximal femur is frequently seen. Patients with loose femoral components often complain of pain or aching in the anterior thigh. Referred pain to the knee is not unusual. Pain is often exacerbated by increased level of activity. In some situations significant osteolysis is observed, although these individuals are often without significant pain.

Loosening of femoral components has been attributed to problems in their design, malposition in varus, and problems with the cement mantle. Radiographically migration or subsidence of the component is diagnostic of loosening. Failure of fixation can occur at the bone - cement interface or the component - cement interface, the former being more common. In a very loose component, it is not unusual to see an accelerated phase of osteolysis with worrisome loss of bone stock. Microscopically this is characterized by a fibrous membrane in which macrophage, histiocyte, and general inflammatory cells are seen. Again this is thought to reflect the body's reaction to wear debris.

Dislocation of the hip typically ranks second to loosening as the most troubling complication related to total hip arthroplasty. Reasons for dislocation include trauma, infection, inappropriate leg positioning, particularly in the early postoperative period, abductor muscle laxity, and or paralysis, and malposition of components.

Clearly management of the dislocation should be directed at the specific cause. Often this is not readily apparent, and a certain percentage of patients do not achieve stability either by operative or non-operative measures, but continue to dislocate.

Non-operative treatment in first-time dislocators consisting of reduction followed by traction, bracing, or spica cast for a period of time is successful in approximately two of three cases. This is typically pursued in situations where component position is felt to be appropriate. Operative intervention is necessary if closed reduction fails, component malposition is present, or for recurrent episodes of instability. Of those patients who came to surgery, success rates are highest in instances of component malposition or trochanteric revision/advancement and are typically 60 to 70 per cent.

Amstutz HC, Friscia DA, Dorey F, Carney BT. Warfarin prophylaxis to prevent mortality from pulmonary embolism after total hip replacement. J Bone Joint Surg 1989; 71A: 321 - 6.

Ayers DC, Evarts CM, Parkinson JR. The prevention of heterotopic ossification in high-risk patients by low-dose radiation therapy after total hip arthroplasty. J Bone Joint Surg 1986; 68A: 1423 - 9.

Blount WP. Don't throw away the cane. J Bone Joint Surg 1956; 30A: 695 - 708.

Coventry MB. Late dislocations in patients with Charnley total hip arthroplasty. J Bone Joint Surg 1985; 67A: 832 - 41.

Ficat P, Raymond P. Treatment of avascular necrosis of the femoral head. In: The Hip Society. The Hip. St Louis: CV Mosby Co., 1983: 280.

Hardinge K. The direct lateral approach to the hip J Bone Joint Surg 1982; 64B: 17 - 19.

Hopson CN, Siverhus SW. Ischemic necrosis of the femoral head: treatment by core depression. J Bone Joint Surg 1988; 70A: 1048 - 51.

Jensen NF, Harris WH. A system for trochanteric osteotomy and reattachment for total hip arthroplasty with a 99% union rate. Clin Orthop 1986; 208: 174 - 81.

Kakkar VV, Howe CT, Nicolaides AN, Renney JTG, Clarke MB. Deep vein thrombosis of the leg. Is there a high risk group? Am J Surg 1970; 120: 527 - 30.

Kavanagh BF, Dewitz MA, Ilstrup DM, Stauffer RN, Coventry MB. Charnley total hip arthroplasty with cement—15 year results. J Bone Joint Surg 1989; 71A: 1496 - 1503.

McDonald DJ, Fitzgerald RH, Ilstrup DM. Two-stage reconstruction of a total hip arthroplasty because of infection. J Bone Joint Surg 1989; 71A: 828 - 34.

Nordin M, Frankel V. Basic biomechanics of the musculoskeletal system. 2nd edn. Philadelphia: Lea and Febiger, 1989: 135 - 51.

Paiement GD, Wessinger SJ, Harris WH. Survey of prophylaxis against venous thromboembolism in adults undergoing hip surgery. Clin Orthop 1987; 223: 188 - 83.

Paiement G, Wessinger SJ, Waltman AC, Harris WH. Surveillance of deep venous thrombosis in asymptomatic total hip replacement patients. Am J Surg 1988; 155: 400 - 4.

Russotti GM, Coventry MB, Stauffer RN. Cemented total hip arthroplasty with contemporary techniques. Clin Orthop 1988; 235: 141 - 7.

Soballe K, Boll KL, Koford S, Severinsen B, Kristensen SS. Total hip replacement after medial displacement osteotomy of the proximal part of the femur. J Bone Joint Surg 1989; 71A: 692 - 7.

Woo RY, Morrey BF. Dislocations after total hip arthroplasty. J Bone Joint Surg 1982; 64A: 1295 - 1306.