INTRODUCTION — Rheumatoid arthritis (RA) is a chronic, progressive disorder in many patients. Despite therapy with disease-modifying antirheumatic drugs (DMARDS), including biologic agents, joint erosion and destruction can develop over time. (See "General principles and overview of management of rheumatoid arthritis in adults".)
The major symptom of joint destruction is pain. With further progression of the arthritis, there are signs of loss of motion, diminution in motor strength, and ultimately decline in function. (See "Evaluation and medical management of end-stage rheumatoid arthritis".)
This topic will review the surgical indications, procedures, management approach, and complications of total joint replacement (TJR) in patients with severe RA. Overviews of total hip and knee arthroplasty and their complications, as well as the perioperative management of patients with rheumatic diseases, are presented elsewhere. (See "Total hip arthroplasty" and "Total knee arthroplasty" and "Complications of total hip arthroplasty" and "Complications of total knee arthroplasty" and "Preoperative evaluation and perioperative management of patients with rheumatic diseases".)
FREQUENCY OF JOINT REPLACEMENT — The advent of total joint replacement (TJR) has been considered to be a major breakthrough and an important therapeutic option in the management of rheumatoid arthritis (RA). A 23-year, prospective, longitudinal, observational study (from a single referral center) of TJR and its predictors in 1600 patients with RA followed between 1974 and 1997 from the time of diagnosis revealed the following [1]:
●Approximately 25 percent underwent TJR.
●For patients with one TJR, 25 percent required an additional arthroplasty within one year, and 50 percent required an additional arthroplasty within seven years.
●Ten years after TJR, approximately 6 percent of implanted knees and 4 percent of implanted hips had been replaced with a second arthroplasty.
●The risk of requiring a TJR increased by three-fold to six-fold, with highly abnormal values obtained with any of the following: the Health Assessment Questionnaire Disability Scale, global severity of disease assessment, and measurement of the erythrocyte sedimentation rate (ESR).
Subsequent studies revealed similar results [2,3]. A 2010 study showed that the rates of joint surgery in patients with RA have declined since reaching the peak in the 1990s [4]. Among younger patients with RA, aged 40 to 59, the rates of knee and hip surgery decreased by 19 and 40 percent, respectively. However these rates increased in patients older than 60 years. These results may reflect the beneficial effect of changes in antirheumatic treatments, with an expected lag in older patients.
INDICATIONS AND SURGICAL OPTIONS — When medical therapy fails to control rheumatoid arthritis (RA) in a specific joint or joints, one should consider reconstructive surgery, depending upon the joint in question. The main indications for surgery are intractable joint pain (with activity or at rest) and the presence of an unacceptable functional decline. Pain or functional declines are ascribed to joint destruction. This must be accompanied by failure of all of the nonsurgical approaches to the management of the arthritis including a change in disease-modifying, immunosuppressive, or biologic drug regimens, the use of more potent analgesics, and rehabilitation therapies. Plain radiographs should be obtained to confirm the end-stage nature of the arthritis. In addition, bursitis, tendinitis, and nerve entrapment syndromes must be excluded as causes of joint pain. (See "Overview of soft tissue musculoskeletal disorders".)
The goal of surgery in the management of RA is the relief of pain and the restoration of joint function. When clinically indicated, surgery is recommended to take place in a timely manner because a prolonged delay can lead to fixed deformities, soft tissue contractures, or excessive muscle atrophy, which prevents a good postoperative recovery [5]. Various surgical procedures, including total joint arthroplasty, may be indicated for the damaged joint.
Surgical options for the patient with RA include the following procedures:
●Tenosynovectomy to excise inflamed tendon sheaths or to repair a recent tendon rupture (eg, hand tendons).
●Arthroscopic repair of tendon rupture (eg, rotator cuff).
●Arthroscopic or open synovectomy to excise inflamed synovium in an attempt to reduce pain, temporarily prevent cartilage destruction, or remove joint debris that impinges upon normal joint range of motion (any large joint). These operations have become much less common with the development of more effective medical treatment approaches since the 1990s.
●Osteotomy to realign weightbearing bones to correct valgus or varus deformities of the knee.
●Joint fusion to stabilize destroyed joints that are not easily replaced (eg, ankle, wrist, thumb, and cervical spine).
●Soft tissue release to correct severe contractures that develop around joints with attendant marked loss of motion. This was more commonly seen in the past when there were many severely debilitated and bedridden RA patients.
●Small joint implant arthroplasty (eg, metacarpophalangeal joints) to reduce pain and to improve hand function.
●Metatarsal head excision arthroplasties to alleviate severe forefoot pain and to improve gait.
●Total joint replacement (TJR) (eg, shoulder, elbow, wrist, hip, knee, ankle), which will be the focus of the remainder of this topic review.
Contraindications — The major contraindication to joint replacement is active systemic or articular infection. Improvements in the perioperative management of patients have markedly reduced the risk associated with comorbid medical conditions, such as hypertension, cardiovascular disease, diabetes, obesity, or bleeding disorders.
TOTAL JOINT ARTHROPLASTY
Preoperative evaluation — Preoperative evaluation for patients with rheumatic diseases is presented in more detail elsewhere. (See "Preoperative evaluation and perioperative management of patients with rheumatic diseases".)
●Cervical spine – In addition to the usual preoperative screening for surgical clearance, the patient with rheumatoid arthritis (RA) needs to be evaluated for cervical spine instability. Flexion and extension radiographs of cervical spine showing an increased anterior atlantodental interval (ADI) of greater than 3 mm or a posterior atlantodental interval of less than 14 mm suggest severe cervical spine subluxation and instability [6]. These patients are at risk for cord impingement and neurologic compromise as a result of the manipulations of the cervical spine that may occur with intubation and surgical positioning; as a result, general anesthesia and intubation should be approached with caution. (See "Cervical subluxation in rheumatoid arthritis".)
●Minimizing risk of prosthetic joint infection – Infectious complications associated with joint implant surgery remain a serious concern in the patient with RA. Potential sources for infection, such as dental caries, skin lesions, and urinary tract infections, should be screened for and treated to prevent future hematogenous spread of these infections to prosthetic joints [7,8]. In a prospective study of all total joint replacement (TJR) surgeries from the Norwegian arthroplasty register, a 1.6-times higher risk of total knee replacement (TKR) revision surgery due to infection was found in RA patients compared with osteoarthritis (OA) patients. No increase in the risk of total hip replacement (THR) revision surgery due to infection was found in RA patients compared with OA patients in the first five postoperative years. A higher incidence of late infection after six years was noted in both THR and TKR patients with RA [9].
●Deep venous thrombosis (DVT) prophylaxis – DVT prophylaxis for orthopedic surgical patients is described in detail separately. (See "Prevention of venous thromboembolism in adults undergoing hip fracture repair or hip or knee replacement".)
Management of antirheumatic drug therapy — The management of antirheumatic drug therapy in the perioperative period, including the use of disease-modifying antirheumatic drugs (DMARDs) and antiinflammatory drugs, and the risk of suppression of the hypothalamic-pituitary-adrenal axis due to glucocorticoid use, is discussed separately. (See "Preoperative evaluation and perioperative management of patients with rheumatic diseases", section on 'Medication management' and "Preoperative evaluation and perioperative management of patients with rheumatic diseases", section on 'Rheumatoid arthritis'.)
Surgical priorities — When the patient with RA has multiple joints in need of surgical attention, establishing the proper surgical sequence is an important issue. Lower extremity surgery should generally predate upper extremity surgery, when possible [10]. In the upper extremity, wrist deformities should be corrected before hand surgery in order to maintain distal stability and alignment. Whether elbow or shoulder replacement should precede wrist and hand surgery is controversial and should be decided on an individual basis [11,12].
In the lower extremity, foot and ankle surgery should be performed before hip and knee replacement in order to provide stability for lower extremity rehabilitation [11]. However, if knee malalignment is particularly severe, this must be addressed first. In general, ankle and hindfoot procedures should be performed before forefoot reconstruction [11]. Unless there is severe pain, instability, or contracture of the knee joint, arthroplasty of the ipsilateral destroyed hip should be performed prior to knee reconstruction, because hip rehabilitation generally requires less functional range of motion of the ipsilateral knee than knee rehabilitation requires of the hip.
Total hip arthroplasty — THR has become one of the most frequently performed reconstructive surgical procedures in orthopedic surgery. (See "Total hip arthroplasty".)
The success rate has improved markedly since John Charnley popularized the technique in the late 1950s [13]. Factors influencing the surgical outcome include patient age, general health, and preoperative functional status. At 10 to 20 years of follow-up, a 90 percent success rate can be predicted for the average patient undergoing a cemented total hip arthroplasty [14,15]. Sixty to 90 percent of THRs in the United States utilize cementless designs and, at 10-year follow-up, have a 95 percent rate of prosthesis survival, with some maintaining this success rate at 20-year follow-up [16]. Since the advent of more effective medical treatments for RA, the incidence of THR for young adults with RA has diminished substantially; less than 10 percent of THR surgeries in young adults are performed for inflammatory arthropathies. The majority of THR surgeries in patients under 30 years of age are for osteonecrosis or secondary OA [17]. Outcome studies have indicated that the failure rate of cementless implants is primarily in the acetabular component and was less than 7 percent with a follow-up of 4 to 16 years, with the exception of one study with a failure rate of 49 percent attributed to an unsuccessful acetabular hydroxyapatite coating.
An important operative decision is whether to implant the hip with or without cement. Cemented fixation techniques have been used since Charnley introduced them, but their use became widespread in the 1970s. Cementless components were introduced in the late 1980s. A cementless component is generally advised for the young patient and for anyone with good bone stock, and a cemented implant is generally preferred for the older, low-demand patient with thinner bone stock (see "Total hip arthroplasty"). Patients can apply weight on a cemented hip prosthesis within a few days following surgery, while they may have to wait for weeks before applying weight to a cementless prosthesis. A disadvantage of cemented hip prostheses is the higher risk of prosthesis loosening, particularly in patients with RA [18].
The incidence of infection in primary total hip arthroplasty varies from 0.4 to 1.5 percent. In a prospective study of all THRs from the Norwegian arthroplasty registry, no increase in the risk of THR revision surgery due to infection was found in RA patients compared with OA patients in the first five postoperative years, but a higher rate of late infection after six years was noted [9]. The pathogenesis, clinical manifestations, treatment, and prevention of prosthetic joint infections are presented separately. (See "Prosthetic joint infection: Epidemiology, microbiology, clinical manifestations, and diagnosis" and "Prosthetic joint infection: Treatment" and "Prevention of prosthetic joint and other types of orthopedic hardware infection".)
Total knee arthroplasty — The reported success rate for total knee arthroplasty (TKA, knee replacement) is in excess of 85 percent at 10 years in patients with RA [19-21]. The indications, preoperative evaluation, selection of components and of fixation technique, perioperative management, and outcomes of TKA are presented separately. (See "Total knee arthroplasty".)
The relative benefits of TKA were described in a 2015 study that compared the impact of TKR in patients with RA and with OA who were participants in the National Data Bank for Rheumatic Diseases and underwent such surgery between 1999 and 2012 [22]. A smaller proportion of patients with RA than OA underwent TKA (5.3 percent of 15,818 versus 10.2 percent of 3079 patients). The patients with RA had less preoperative pain but more involvement of other joints, underwent TKA at a younger age, and showed less improvement in most pain- and health-related quality of life measures, although both of the latter groups exhibited benefit.
Revision rates due to septic or aseptic loosening are low. Problems with the patellofemoral articulation, including patellar instability, loosening, fracture, and component failure, are a leading cause of the need for TKA reoperation. (See "Complications of total knee arthroplasty", section on 'Patellofemoral disorders'.)
In a prospective study of all TKRs from the Norwegian arthroplasty register, a 1.6-fold higher risk of TKR revision surgery due to infection was found in RA patients compared with OA patients [9]. Removal and delayed replacement of the knee prosthesis may result in the best functional results. (See "Complications of total knee arthroplasty", section on 'Surgical site infection'.)
It is unclear whether the use of continuous passive motion devices in the postoperative management of TKA results in enough clinical benefit to justify the inconvenience and expense of the procedure. (See "Total knee arthroplasty", section on 'Postoperative course and follow-up'.)
Total shoulder arthroplasty — Prosthetic replacement of the glenohumeral joint was first introduced by Neer in 1951; since that time, improvements in prosthesis design have led to better outcomes with respect to pain reduction, function, and longevity. Biomechanical studies revealed that the prosthetic head thickness and glenohumeral alignment should replicate presurgical measurements to maximize function and promote implant durability. Anatomic studies led to the development of modular designs with different stem and head sizes that could be mixed and matched, allowing for better fit to accommodate individual differences. Most systems in use are modular with stems made of cobalt-chrome alloy, with proximal porous ingrowth coating and cemented all-polyethylene glenoid components [23,24].
Radiographs in patients with RA do not show the subchondral sclerosis, inferior humeral head osteophytes, and posterior subluxation of the humeral head on the glenoid seen in advanced OA, but rather a more symmetric pattern of joint space narrowing with juxtaarticular osteopenia and central glenoid wearing. Magnetic resonance imaging (MRI) can be used to evaluate the extent of rotator cuff disease and will show advanced cartilage degeneration. Most patients with RA have some degree of rotator cuff tear, and 20 to 35 percent have full thickness tears [23]. Computed tomography (CT) is valuable to determine the degree of glenoid wear and amount of residual bone stock, since glenoid recontouring procedures or bone grafting may be necessary to recenter the humeral head if it is subluxed.
Total shoulder arthroplasty (TSA) is considered superior to shoulder hemiarthroplasty (HA) with respect to pain relief, activity level and function, longevity, and necessity for revision in patients with RA and an intact rotator cuff. However, in patients with RA and severe rotator cuff tears, the evidence for whether TSA or HA, or the newer reverse shoulder arthroplasty, should be performed is less clear.
In a 2007 study of 195 TSAs and 95 HAs, involving 247 patients with RA who underwent surgery between 1976 and 1991, with at least two years of follow-up (mean 11.6 years) or follow-up until revision, there was comparable pain relief in patients with or without an intact rotator cuff [25]. However, in the patients with an intact rotator cuff who received a TSA the risk of undergoing a revision procedure was significantly less, and the range of motion and degree of pain relief were significantly greater. There was a high incidence of glenoid component periprosthetic lucency in the TSA group (72 percent) and of glenoid wear in the HA group (98 percent).
A 2014 follow-up study of patients from the same group identified 165 TSAs and 90 HAs available for clinical analysis with at least five years of follow-up (mean 13.8 years), finding reasonable longevity for both TSA and HA and that none of the variables analyzed, including age, gender, and rotator cuff status, could differentiate the risk of revision surgery in TSA versus HA patients [26]. Both TSA and HA continued to provide pain relief and improved motion, although both pain and motion showed greater improvement in patients with an intact rotator cuff who underwent TSA. In patients with a thin or torn rotator cuff, there was no difference in pain relief between TSA and HA. Survivorship free of revision at 10 years following TSA with and without intact rotator cuff was 97 and 91 percent, respectively, and following HA was 76 and 93 percent, respectively. Among the HA patients, there was a very high rate of glenoid wear (98 percent), and eight patients had revision to TSA for symptomatic wear of the glenoid, with 11 revisions overall. Among the TSA patients, although there was a high incidence of periprosthetic glenoid lucency on radiography (73 percent), there was a low rate of revision for glenoid loosening (nine patients, four of which were revised to HA), and an additional eight TSA patients required revision surgery for other reasons.
Despite these findings, irreparable rotator cuff tears are still considered as a contraindication to TSA because deltoid muscle contraction allows the humerus to migrate upward, resulting in early glenoid component loosening. Originally developed in the 1980s in Europe, reverse shoulder arthroplasty has been available since 2003 as a surgical option in the United States for these patients. In this procedure, the socket component is attached to the humerus and the ball component is attached to the glenoid fossa. By reversing the ball and socket components, a fulcrum is created around which the deltoid can pull to restore forward elevation. This surgery is considered advantageous compared with HA because it restores motion in addition to relieving pain. Follow-up studies of reverse shoulder arthroplasty in RA show improvement in outcomes and a complication rate of 14 percent but are limited in number of patients and duration of outcomes [27,28]. Complication rates vary with the surgeon's experience, and some experts avoid this surgery in patients under 70 years of age because of higher demands and need for longer implant survivorship. However, as implant design and surgical techniques have improved since the early 2000s, orthopedic surgeons are cautiously considering this procedure for a wider spectrum of patients, including younger patients with RA.
Total elbow arthroplasty — Pain is the main indication for total elbow arthroplasty. Pain relief following total elbow replacement is excellent in most patients with RA. In some younger patients, excision of the radial head and synovectomy of the elbow have offered a conservative and effective method of treatment for refractory elbow pain. However, subsequent conversion of this procedure to a standard capitellocondylar total elbow replacement is more difficult. In one series, the degree of improvement in such patients was lower than in those who underwent primary elbow arthroplasty [29]. In RA patients with radiographic evidence of advanced joint destruction and with pain, stiffness, and instability that limits activities, implant arthroplasty is preferred to radial head excision with synovectomy. Since 1975, more than 20 elbow protheses have been developed, but comparison of results is difficult because of the small numbers of patients reported and because of the lack of standardized assessment. Rating systems have been established that are enabling more standardized comparison of prosthesis design [30-32]. Two types of total elbow arthroplasties are frequently chosen by orthopedists. One is a semiconstrained metal-to-polyethylene hinge; the other is an unconstrained capitellocondylar design [32].
Overall, pain relief is successful in more than 90 percent of patients, and long-term complication rates are approximately 11 percent [33]. Though infection has historically been the main complication in elbow replacements, further development in surgical techniques has reduced the incidence of infection to less than 3 percent [33]. Loosening of the implant remains a concern [34]. Patients who have had previous elbow infections or a history of heterotopic ossification should be advised to avoid total elbow arthroplasty, if possible [30].
Total wrist arthroplasty — The wrist joint is the key to stability of the hand that leads to normal function. Recurrence of ulnar deviation at the metacarpophalangeal joint almost always occurs unless wrist stabilization is first obtained [35]. Until the 1970s, wrist arthrodesis (fusion) was the only commonly employed surgical procedure to relieve pain and to provide stability to this joint. Limited or total wrist arthrodesis continues to be performed depending upon the degree and/or location of joint damage.
Total wrist arthroplasty (TWA) is another surgical option. However, its use has been limited due to the high incidence of complications, including loosening, infection, subsidence, and dislocations. Complications have been reported in 40 percent of patients with the Voltz implant [36-38]. A 2008 systematic review of 18 TWA studies that included 500 procedures compared the outcomes of TWA with total wrist arthrodesis for RA [39]. Outcomes for total wrist arthrodesis were at least comparable and possibly better than those for TWA. Fusion provided pain relief in 98 percent of the cases compared with 90 percent for arthroplasty, and only 3 out of 14 studies with appropriate data showed an average active arc of wrist motion within the functional range for patients undergoing arthroplasty. Patient satisfaction was comparable and was reported as more than 90 percent in each group. There were more major complications in the arthroplasty group, and 21 percent underwent revision surgery, including prosthesis removal without replacement or salvage arthrodesis in 5 percent [39].
Total ankle replacement — Because the ankle is a hinged joint, and normal gait requires only 10 to 12 degrees of ankle extension and 20 degrees of ankle flexion, loss of motion is not critical to function, and ankle arthrodesis (AA) can result in a stable, pain-free ankle and improve quality of life. It has remained the treatment of choice for relief of severe ankle and hindfoot pain in patients with RA [40]. The main drawback is the later development of arthrosis in the adjacent joints, particularly fusion of the subtalar joint. Total ankle replacement (TAR) has the potential to preserve range of motion, restore normal gait, and protect adjacent joints.
TAR was introduced in the 1970s, but initially high failure and complication rates were noted. Since that time, three generations of implants have been developed, with several total ankle arthroplasty systems approved for use in the United States by the US Food and Drug Administration (FDA) since 2005. Third-generation systems feature a metallic baseplate fixed to the tibia and a domed component resurfacing the talus, with a polyethylene bearing surface interposed between the two.
Although there is an impression that the two procedures have similar outcomes, there is no conclusive evidence. A 2014 systematic review of data comparing TAR with arthrodesis found three retrospective studies and only one prospective study, which was non-randomized [41]. There were multiple other methodological problems noted with the studies. Two of the studies showed statistically significant improvements in the TAR group, but the other two studies showed no differences between the two procedures. Careful selection of patients for either procedure is important. Younger, heavy, physically active males might be better candidates for arthrodesis, and patients with preexisting subtalar arthritis or hip or knee impairment that would be worsened by loss of ankle joint motion might be better candidates for TAR, but randomized trials are needed to confirm these preferences.
Total metacarpophalangeal arthroplasty — A 2005 review of the studies reporting results of hand surgery in RA concluded that the available studies do not allow for a clear definition of surgical indications or criteria [42]. This view was shared by authors of another review of hand surgery in RA patients across the United States [43]; these investigators found large variations in the surgical management of the rheumatoid hand that depended upon geographic location. Rheumatologists and hand surgeons often disagree about the indications for RA hand procedures [12,44].
Outcomes involving surgery of the metacarpal phalangeal (MCP) joints have traditionally been measured by range of motion restored, patient satisfaction, subjective functional outcome, and complication rate. The MCP joints, which are critical for the positioning and functioning of the digits, are condylar joints. Thus, they are less stable than the proximal interphalangeal (PIP) joints.
The final outcome with implant arthroplasty is never a normal functioning articulation; rather, the result is a painless arthroplasty with a useful arc of motion [45]. Benefits usually include an arc of motion in the 40 to 50 degree range, improvement of extensor lag, and correction of ulnar deviation to 0 to 20 degrees [46,47].
A prospective, multicenter, cohort study of 46 RA patients followed two years after reconstructive surgery found objective improvement in MCP arc of motion, ulnar drift, and extension lag that correlated well with patient satisfaction scores as measured by the Michigan Hand Questionnaire (MHQ), despite the fact that there were minimal improvements in grip and pinch strength [48]. A 2000 systematic review concluded that silicone implant arthroplasty was beneficial [49].
Patients with excellent PIP joints tend to gain less flexion of their MCP joints with surgery; in this setting, special attention is, therefore, undertaken during rehabilitation to improve MCP function [47].
Possible complications with implant arthroplasty include recurrence of ulnar deviation, loss of the MCP's arc of flexion, infection, and implant breakage [50]. A broken prosthesis does not necessarily require surgical replacement [51].
COMPLICATIONS — The safety and reliability of total joint arthroplasty have improved markedly since the initial use of these procedures. However, there are a number of potential complications, the most important of which are infection, thromboembolic disease, and implant loosening. (See "Complications of total hip arthroplasty".)
Although there have been rare case reports of sarcomas of bone or soft tissue adjacent to prosthetic joints, a large and long-term epidemiologic study of cancer rates among nearly 117,000 such patients, followed for a mean of seven years, found no increase in the overall risk of cancer or sarcoma [52].
Infection — Periprosthetic joint infection is a serious complication of hip and knee replacement surgery with a prevalence of 1 to 3 percent and is a major reason for hip and knee revision surgery [53-56]. Patients with rheumatoid arthritis (RA) appear to be at higher risk for infection after total knee arthroplasty (TKA) and, possibly, after total hip arthroplasty, compared with patients with osteoarthritis (OA) [57]. As an example, in a large population-based study, there was an approximately 50 percent greater risk of infection during the two years following primary knee arthroplasty in patients with RA compared with those with OA, with an absolute risk of 1.2 percent in the RA patients [58]. It is most likely that the largest shares of this risk are related to the underlying disease (RA), as well as its associated comorbidities and use of immunomodulatory therapies. Medication management in rheumatic disease patients undergoing joint replacement surgery is discussed in detail separately. (See "Preoperative evaluation and perioperative management of patients with rheumatic diseases".)
Several subsequent studies have also documented the infectious risk in patients with RA, the association between relatively low doses of prednisone therapy and increased one-year risk of prosthetic joint infection in patients with RA undergoing total hip arthroplasty/total knee arthroplasty, and the dose-response relationship between prednisone dose and these adverse outcomes [59,60]:
●In a Danish nationwide register-based cohort study, the one-year risks of prosthetic joint infection and death following total hip arthroplasty/total knee arthroplasty were greater in patients with RA than OA [59]. Patients treated with biologic disease-modifying antirheumatic drug (DMARD) therapies did not have a higher risk of prosthetic joint infection compared with a comparable biologic-naïve group; however, patients treated with glucocorticoids had an increase in the risk of prosthetic joint infection and death (prednisone >7.5 mg/day with greater risk than prednisone ≤7.5 mg/day).
●Another study, using Medicare Claims and MarketScan administrative data for patients with RA undergoing total hip arthroplasty/total knee arthroplasty, found a numerically increased one-year risk of prosthetic joint infection with receipt of prednisone 5 to 10 mg daily and a greater and statistically significantly increased one-year risk of prosthetic joint infection among patients receiving more than 10 mg/day [60]. In this same study, the one-year risk of prosthetic joint infection was similar among the various biologic disease-modifying therapies.
The epidemiology, clinical manifestations, microbiology, and diagnosis of prosthetic joint infections are reviewed in detail elsewhere. (See "Prosthetic joint infection: Epidemiology, microbiology, clinical manifestations, and diagnosis".)
The erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP), which may be helpful in identifying postoperative infections in patients with noninflammatory causes of joint replacement, may not be helpful in patients with active RA. All culture specimens obtained should be sent for aerobic, anaerobic, fungal, and acid fast organism cultures, since patients with RA may be immunocompromised from RA as well as from their preoperative medications.
Thromboembolism — The incidence of venous thromboembolism (VTE), which includes deep venous thrombosis (DVT) and/or pulmonary embolism (PE), has steadily declined since 1990, with the adoption of less invasive surgical techniques and of earlier postoperative mobilization and discharge. In one study, the risk of VTE within 90 days after total hip arthroplasty was lower among patients with RA compared with OA (0.4 versus 1.4 percent), while the risk after TKA was comparable [58]. VTE still remains a significant cause of morbidity and potential mortality, although VTE-related deaths in patients undergoing orthopedic surgical procedures were infrequently reported in trials since 2003 [61].
The 2012 American College of Chest Physicians guideline estimated that the risk of nonfatal, symptomatic VTE rate was reduced from 4.3 to 1.8 percent in the month after major orthopedic surgery when low molecular weight heparin (LMWH) was utilized perioperatively [61]. A number of other alternative agents may also be utilized. The prevention of venous thromboembolism in orthopedic surgical patients, including those undergoing joint arthroplasty, is discussed in detail separately. (See "Prevention of venous thromboembolism in adults undergoing hip fracture repair or hip or knee replacement".)
Implant loosening and failure and cobalt toxicity — Implant loosening, which typically presents with pain, is the major cause of implant failure. Loosening may occur due to infection, mechanical failure related to surgical technique, or osteolysis [62]. (See 'Infection' above.)
Osteolysis results when particulate polyethylene debris released from worn implants causes a cellular immune response that forms a synovial-like membrane around the prosthesis [63], and this has been reported in several designs of cemented and cementless implants [64]. Increased cellular proliferation to both acrylic and cobalt-chromium has been observed in patients with aseptically loosened metal-on-metal and ceramic-on-ceramic prostheses [65]. Metal-on-metal implants have been shown to fail at a significantly higher rate than ceramic-on-ceramic or metal-on-polyethylene implants, and a number of concerns have arisen regarding the use of these prostheses. These include observations described in case reports that both metal-on-metal and metal-on-polyethylene implant wear can release cobalt into the bloodstream, resulting in systemic manifestations, including cardiomyopathy; hypothyroidism; neuropathy; smell, hearing, and taste loss; rashes; muscle weakness; and fatigue. In these cases, elevated serum cobalt levels were noted, and with surgery to remove the prostheses, cobalt levels decreased and clinical symptoms improved [66,67]. (See "Complications of total hip arthroplasty", section on 'Sequelae from metal-on-metal wear debris'.)
Early detection of implant loosening is required for successful revision and prevention of catastrophic implant failure. The diagnosis of hip implant loosening is often confirmed if there is a radiolucent zone of 2 mm or more in width in a patient who has pain on weightbearing that is relieved by rest. Conventional radiography is the mainstay for evaluating these joints, with computed tomography (CT) scanning and scintigraphy playing smaller roles [68]. Another technique is digital subtraction arthrography, which may be useful if conventional x-rays are normal. When used to evaluate the femoral component in 78 painful hip prostheses, this technique had 96 percent sensitivity and 100 percent specificity for the diagnosis of component loosening [69].
Hip dislocation — Patients with RA may be at higher risk of hip dislocation as a complication of total hip arthroplasty compared with patients with OA. A systematic review of 40 studies included a meta-analysis of four studies that analyzed the results of surgery in 1637 patients with RA and in 61,810 patients with OA, which found a significantly higher risk of hip dislocation in the RA patients (OR 2.16, 95% CI 1.52-3.07) [57]. A subsequent population-based study, which addressed some of the methodologic limitations of prior studies, also found a greater risk of dislocation within two years following total hip arthroplasty among RA patients compared with OA (2.6 versus 1.2 percent) [58]. Different surgical approaches and techniques may have an impact on the degree of risk; a careful discussion about such particular risks should take place between surgeon and patient. (See "Complications of total hip arthroplasty", section on 'Dislocation'.)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Knee replacement (The Basics)" and "Patient education: Hip replacement (The Basics)")
●Beyond the Basics topics (see "Patient education: Total knee replacement (Beyond the Basics)" and "Patient education: Total hip replacement (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●The main indication for joint replacement surgery in patients with rheumatoid arthritis (RA) is joint destruction, despite medical and rehabilitative therapy, which has resulted in intractable joint pain (with activity or at rest) or in an unacceptable functional decline. Other causes of articular and periarticular pain should be excluded. The goal of surgery in RA is the relief of pain and the restoration of joint function. Once a determination has been made that surgery is clinically indicated, it should take place in a timely manner to avoid fixed deformities, soft tissue contractures, or excessive muscle atrophy, which prevent a good postoperative recovery. (See 'Indications and surgical options' above.)
●The only major contraindication to joint replacement is active systemic or articular infection. (See 'Contraindications' above.)
●Particular attention should be given to the preoperative evaluation, including identification of patients with cervical spine instability and who are at increased risk of neurologic complications, and to the identification of potential sources for infection, such as dental caries, skin lesions, and urinary tract infections, for which the patient should be screened and treated to prevent future hematogenous spread of these infections to prosthetic joints. (See 'Preoperative evaluation' above and "Preoperative evaluation and perioperative management of patients with rheumatic diseases".)
●The dosing of certain drugs used in the treatment of RA may need to be adjusted to minimize perioperative and postoperative complications. (See 'Management of antirheumatic drug therapy' above.)
●When the patient with RA has multiple joints in need of surgical attention, establishing the proper surgical sequence is an important issue, which depends upon the specific joints involved and upon the need to optimize stability, alignment, and rehabilitative potential. (See 'Surgical priorities' above.)
●The specific evaluation and outcomes can differ between joints. Hip and knee arthroplasties generally have a high level of success. (See 'Total hip arthroplasty' above and 'Total knee arthroplasty' above and 'Total shoulder arthroplasty' above and 'Total elbow arthroplasty' above and 'Total wrist arthroplasty' above and 'Total ankle replacement' above and 'Total metacarpophalangeal arthroplasty' above.)
●The most important of the potential complications of joint replacement in RA include infection, thromboembolic disease, and implant loosening. (See 'Complications' above and 'Infection' above and 'Thromboembolism' above and 'Implant loosening and failure and cobalt toxicity' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Michael H Weisman, MD, who contributed to an earlier version of this topic review.
