A 45-year-old man presented with bilateral hip pain (greater on the right side than the left) that had been worsening over the past 5 years. The pain was located in the groin and was exacerbated by walking. His hip motion had deteriorated over time. Radiographs demonstrated severe osteoarthritis, and total hip arthroplasty (THA) was recommended ( Fig. 38.1 ).
THA was performed by a less-invasive direct lateral approach without complication. At regular postoperative assessments during the first 6 months, he complained of low-grade right hip pain, but he was able to return to work.
He presented 2.5 years after surgery with continued and mildly worsening right hip pain that he said had never resolved ( Fig. 38.2 ). Radiographs were obtained, and lucency was observed in Gruen zones 1 and 7. We assumed that ingrowth of the femoral stem had failed and that it had become loose. Laboratory studies, including erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP), were ordered to rule out infection. Because the ESR and CRP levels were mildly elevated, aspiration was ordered. It yielded minimal fluid and a culture that was positive for Staphylococcus epidermidis . Repeat aspiration was performed, and a white blood cell count of 57,500 cells/µL was obtained with 93% polymorphonuclear leukocytes (PMNs). The culture result was again positive for S. epidermidis . The patient was treated for infection in stages with radical débridement, placement of an antibiotic-laden polymethylmethacrylate articulating spacer, and a 6-week course of intravenous antibiotics followed by reimplantation.
Total hip arthroplasty (THA) is increasingly common and very successful, but a small percentage of patients continues to have hip pain postoperatively.
A detailed history and physical examination are necessary to accurately diagnose the source of the failed THA.
The differential diagnosis is broad, and an algorithmic approach must consider intrinsic and extrinsic causes of pain.
Even in the presence of obvious mechanical failure or instability, infection must be ruled out, because the treatment is fundamentally different from that for an aseptic revision.
Metal-on-metal bearings are associated with unique complications, but because data to guide the workup are sparse, skillful and attentive communication with the patient is required.
Total hip arthroplasty (THA) is one of the most common orthopedic procedures. An estimated 280,000 procedures are performed annually at a cost of $12 billion in the United States. The number is projected to increase to approximately 572,000 procedures performed annually by the year 2030. Analyses have demonstrated that THA is one of the most cost-effective interventions in health care.
Follow-up assessments have demonstrated high clinical success rates, defined by patient satisfaction, reduction in pain, and improvement in function. Despite these successes, approximately 25% of patients report some level of pain at the 6-month follow-up. Britton and colleagues reported that the pain tended to improve slightly from 6 months to 2 years postoperatively but began to gradually deteriorate after 4 years. They also reported that pain correlated with patient opinion and was predictive of revision.
The estimated probability of THA revision at 5 years is 4.1%, with revisions accounting for approximately 18% of the THA procedures performed annually in the United States. This revision burden is expected to grow over the next 2 decades. Revision THA results in increased institutional costs, longer lengths of hospital stay, longer operative times, and increased risks for patients with less predictable clinical benefits. These are most commonly performed because of instability, mechanical loosening, and infection. The need for hip arthroplasty surgeons to systematically and accurately evaluate the patient with a painful, potentially failed THA will become increasingly important as patient volume grows in the setting of diminishing resources.
The medical history and physical examination require minimal equipment. A goniometer is helpful to accurately quantify motion of the hip and knee. Graduated blocks may be required to accurately assess a limb length discrepancy (LLD) in the standing patient. A reflex hammer may assist in a thorough neurologic examination that includes assessing deep tendon reflexes. Access is needed to plain radiography viewing equipment, such as a picture archiving and communication system (PACS) or a light box.
Laboratory evaluation is necessary in most cases. It is facilitated by convenient access to a laboratory with a full range of services, including synovial fluid analysis and ESR and CRP determinations. Testing of serum levels of metal ions, including chromium and cobalt, use specific protocols that, for example, require collection of blood samples in certified metal-free vacutainers. Compliance with published protocols should be verified before basing clinical decisions on reported levels.
When a patient presents with a painful THA, a thorough investigation must follow to ensure that important information is not overlooked. An algorithmic approach is recommended. The algorithm may be completed in a single office visit, but return appointments and follow-up testing often are required. Communication with the patient regarding imaging and laboratory testing results cannot be overemphasized, because pain after THA may be a tremendous source of anxiety for patients. Application of this algorithm requires an understanding of the broad differential diagnosis of pain after THA ( Table 38.1 ).
Step 1. History
The first step in the algorithm is to obtain the patient’s medical history. The history should focus on the nature of the pain, including its temporal onset and precipitating factors, duration and exacerbating factors, location, character, and severity. These characteristics provide insight into the cause of pain. For example, if the patient’s pain has persisted since surgery, it should be established whether the nature of pain has changed from its preoperative status. If not, a wrong diagnosis resulting in THA might have been made. Preoperative radiographs should be obtained, and attention should focus on extrinsic causes of hip pain that might have been the source of pain before surgery. Most commonly, however, the pain has a different nature postoperatively, necessitating a focus on intrinsic and extrinsic causes.
When pain has persisted without relief since surgery and the nature is different from that experienced preoperatively, it may indicate infection, poor implant stability, periprosthetic fracture, or a poor indication for surgery in the first place. However, if the patient experienced relief of symptoms after surgery with later onset of pain, the issue may be soft tissue irritation (e.g., bursitis, tendinitis), infection, aseptic loosening, instability, fracture, or osteolysis. Pain with initiation of movement or during weight bearing that is relieved with rest may be consistent with prosthetic loosening. Vascular or neurogenic claudication may be the issue if pain begins after a consistent degree of activity. Certain patients, especially those with lumbar spine and hip disease, may experience a worsening of neurogenic claudication symptoms after THA and resulting increased activity. Sepsis or malignancy may cause pain that is constant, including at night or with rest, and it may be associated with symptoms such as weight loss, loss of appetite, and general malaise. The temporal onset of limb length inequality in relation to surgery must be adequately established. Limb lengthening does occur during THA and may be a source of pain. Conversely, progressive limb shortening may indicate mechanical failure in the THA.
Location of the pain is informative. Pain localized to the trochanteric region suggests bursitis or fracture. Groin pain or pain in the deep buttock is most often associated with acetabular component loosening, iliopsoas tendinitis, and periprosthetic infection, or it may suggest vascular or neurogenic claudication. Although less common, groin pain may also be associated with obturator, femoral, or inguinal hernia; inguinal lymphadenopathy; psoas abscess; and genitourinary or gynecologic conditions. Radiating pain to the buttock is often associated with spinal stenosis, but if the pain radiates beyond the knee, it is more likely radicular.
Thigh pain is most often related to femoral component loosening or a mismatch in the modulus of elasticity between the implant and the host bone, but it may also result from lateral femoral cutaneous nerve injury. Characterization of the thigh pain helps to differentiate its potential sources. Burning pain or dysesthetic pain is more likely related to nerve pathology, whereas deep, achy pain is more frequently associated with loosening or infection. Sharp pain is related to fracture, bursitis, or tendinitis.
Many conditions may result in a range of pain severity, and precipitating causes should be considered. Pain out of proportion to the surgical history may be caused by complex regional pain syndrome (CRPS). Pain after a traumatic fall may be caused by component loosening or fracture. A history of systemic illness or a recent dental, gastrointestinal, or genitourinary procedure with subsequent onset of pain increases the likelihood of an acute hematogenous infection. Pain associated with a history of wound drainage or hematoma and with prior treatment with antibiotics suggests infection. Immunosuppression, rheumatoid arthritis, diabetes, and obesity also increase the risk of infection in patients undergoing THA.
Step 2. Physical Examination
A well-performed physical examination should include the painful hip, the contralateral hip, the knee and spine, and the neurovascular systems. The goal is to determine the source of pain by performing maneuvers that reproduce or exacerbate the patient’s symptoms.
Examination begins with observation of the patient’s gait pattern, noticing any LLD, antalgia, and signs of muscle weakness, including abductor deficiency. Abductor deficiency is the cause of a Trendelenburg gait pattern and is more common in cases using the direct lateral approach to THA. Femoral or sciatic nerve injury results in distal motor weakness, which may be apparent during gait. If LLD is suspected, limb length should be measured objectively with the patient in the standing and supine positions. Graduated blocks are helpful in determining LLD in the standing position. Fixed pelvic obliquity due to adductor contracture or spinal disease may cause a discrepancy between real and apparent limb lengths and should be considered when assessing LLD.
The skin in the area of prior incisions should be inspected for warmth, erythema, fluctuance, or drainage indicating infection. Muscle wasting may be evident on inspection of the extremity and may corroborate suspicion of peripheral nerve injury or deconditioning. The location of pain should be identified and palpated. This is most easily done with the patient in the standing position and may include the greater trochanter, buttock, thigh, or lumbar spine. Less commonly, conditions, such as femoral or inguinal hernia and lymphadenopathy may be identified on palpation of the painful region.
The patient is then examined in the supine position. Limb lengths are again assessed and degree of motion of the hips and knees established, observing which positions reproduce the patient’s pain. Reproducible pain at extremes of motion indicates instability or impingement. Pain with active motion should be differentiated from pain with passive motion, which correlate with component loosening and occult infection, respectively. Pain with active or resisted hip flexion is indicative of iliopsoas tendinitis. A passive straight leg raise should be performed to determine whether sciatic nerve tension reproduces the patient’s symptoms. Stressing the pelvis with the patient in the supine position may exacerbate pain related to pelvic stress or insufficiency fracture.
Step 3. Plain Radiography
Plain radiography should be the first step in the imaging evaluation of a painful THA. Images should include an anteroposterior view of the pelvis, an anteroposterior view of the affected hip that includes the tip of the stem, and a lateral view of the hip. We routinely use a lateral frog view for follow-up hip evaluation but occasionally obtain a cross-table lateral view to evaluate acetabular component version. Serial plain radiographs should be inspected for signs of migration, osteolysis, or loosening of the implant. Radiographs should be taken in a consistent manner in regard to positioning, penetration, and rotation for optimal comparison.
A cementless technique is used for most THAs in the United States. Engh and colleagues and Dorr and co-workers described criteria for the evaluation of femoral and acetabular component fixation, respectively. Engh and associates identified major and minor radiographic criteria for implant fixation. The absence of reactive lines around the porous-coated portion of the implant and the presence of endosteal spot welds are major signs of osseointegration of the implant. Calcar atrophy, the absence of a pedestal, and a stable distal stem are considered minor signs of osseointegration. Extensive reactive lines around the porous-coated region of the implant are considered to be a major radiographic sign of failure to osseointegrate. The lack of endosteal spot welds constitutes a minor sign of failed osseointegration.
Dorr and colleagues found that progression of radiolucent lines more than 2 years after the operation and any new radiolucent line greater than 1 mm that appeared more than 2 years postoperatively were most predictive of cementless acetabular component loosening at an average of 90 months’ follow-up. Their criteria had 100% positive predictive value and specificity. O’Neill and Harris have a similar system for evaluation of cemented components, but it did not prove to be reliable on the acetabular side, and radiolucencies at the bone–cement interface were subsequently shown to represent internal remodeling rather than deterioration of the interface. Radiolucencies at the prosthesis–cement interface are likely signs of debonding and loosening of the implant. We evaluate radiolucent lines on follow-up radiographs, the involved zones, and the thickness of the line. Progression over time is correlated with patient symptoms.
Plain radiographs should not be relied on for the diagnosis of joint sepsis. However, infection is suggested by certain radiographic findings such as periosteal new bone formation, endosteal scalloping, osteopenia, and extensive lysis. Infection should be suspected if rapid osteolysis or bone disintegration is observed radiographically, especially in the first year postoperatively.
Polyethylene wear can be identified with close scrutiny of plain radiographs and may prompt review for associated osteolytic lesions. Areas of osteolysis may become symptomatic in severe cases and lead to fracture.
Step 4. Evaluate for Infection
After obtaining a history, performing the physical examination, and evaluating the plain radiographs, it is sometimes clear that the patient has an infection. For example, there may be persistent wound drainage, an abscess, or a sinus tract, or the patient may present with aspirate results from an outside physician. Radiographs may show component loosening or early postoperative osteolysis ( Fig. 38.3 ). In this situation, laboratory evaluation, including ESR and CRP values, should be obtained with or without aspiration of the hip joint. If all data are consistent, the surgeon should proceed with staged treatment of the infection, including radical débridement and placement of an antibiotic-laden polymethylmethacrylate spacer ( Fig. 38.4 ), an interim course of intravenous antibiotics, and prosthesis reimplantation ( Fig. 38.5 ). There is no reason to consider other causes of hip pain at this point because treatment of a periprosthetic joint infection takes precedence.