Abstract
Patients with rheumatoid arthritis (RA) continue to undergo arthroplasty to relieve pain and restore lost function caused by joint damage, despite improvements in disease-modifying therapy. The widespread use of potent immunosuppressant medications by RA patients implies that they frequently receive these drugs at the time of surgery. In addition to the medical challenges of medication and disease management at the time of surgery, there are surgical complexities related to complex deformities, osteoporotic bone, and ligament laxity. This paper discusses the current state of the art regarding arthroplasty utilization, outcomes, and perioperative management from a medical and surgical perspective, and highlights the clinical challenges to achieve optimal outcomes.
Introduction
Patients with RA continue to undergo arthroplasty to restore lost function and relieve pain caused by damaged joints, despite the significant expansion of treatment options and the widespread use of potent disease-modifying therapies . However, outcomes may not be equivalent to patients with osteoarthritis (OA), and complications, including infection, are increased . Patients with RA pose challenges for the surgeon, due to the surgical complexity introduced by severe joint deformities as well as bone loss and tendon laxity typical of patients with RA . Patients with RA pose challenges to the rheumatologist in the perioperative period as well, as complex medication regimens need to be adjusted to balance to risk of infection with the risk of postoperative disease flare. Although the overall status of RA patients has improved with the widespread use of potent DMARDs such as methotrexate (MTX) and biologics , joint damages may prevail even in optimally managed patients . As expected, the patients at increased risk for orthopedic surgery include those with severe, seropositive, nodule-positive disease who have elevated erythromycin sedimentation rate (ESR) and C-reactive protein (CRP) levels, and frequently receive immunosuppressive DMARDs at the time of surgery . Orthopedic surgery remains an important option in patients with RA if their joint damage progresses and leads to pain and loss of function. For patients with RA, arthroplasty is indicated as the last step in a progressive approach to localized joint pain and loss of function, and surgical consensus is that arthroplasty is appropriate for patients with severe pain due to advanced arthritis . While there is regional and specialty variation in recommendations for surgery, when patients have severe, daily pain, advanced radiographic damage, and functional compromise such as the need for assistive devices, there is consensus that arthroplasty is indicated .
Utilization of orthopedic surgery in RA patients
Orthopedic surgery has typically been performed in a large number of RA patients, with up to 50% of patients undergoing an orthopedic procedure over the course of their illness . Recently, however, registry studies in Minnesota and Sweden have suggested that the disease course in RA has become less severe, and the target organ for RA, the joint, was less likely to progress to the end stage, thereby leading to less orthopedic surgery . However, others have demonstrated stable rates of hip arthroplasty (THR) and increasing rates of total knee arthroplasty (TKA) in patients with RA . Moreover, while hand and foot surgery rates may have decreased, the rates of major reconstruction of the lower extremities have remained unchanged . Some studies have demonstrated that RA patients with disease onset after age 65 are less likely to receive aggressive medical treatment, which may contribute to the observation that 2% of the RA patients enrolled in a Medicare database underwent orthopedic reconstructive surgery within 3 years of diagnosis, most commonly TKA . In a population-based study, 34% of all incident RA cases underwent orthopedic surgery over the 30-year course of the study, most commonly arthroplasty . However, the age at which RA patients undergo TKA has increased, while that of patients with degenerative arthritis has decreased, suggesting another benefit of the current treatments .
Although the large databases used for these population-based studies rarely are linked to specific information about medication use, the increased use of DMARDs and biologics during this time period may explain the stable or decreasing rates of arthroplasty. A well-characterized cohort of RA patients was recently studied and demonstrated that the concomitant use of MTX and tumor necrosis factor inhibitor (TNFi) reduced the risk of large joint replacement when compared to TNFi use alone . However, the Finnish RA registry failed to demonstrate a decrease in arthroplasty associated with the use of TNFi, when propensity score matching for on-medication disease activity was performed . Nonetheless, the proportion of arthroplasties now performed for RA has decreased from >20% of all arthroplasty cases in the early 1990s to a consistently low level of 3–4% of the current total number of arthroplasties, which has markedly increased . Regional variation also contributes to the difference in arthroplasty utilization; hand and wrist surgery varied 9fold across hospital referral regions . In addition to patient factors, regions with a high density of subspecialty rheumatologists have a significantly lower rate of hand and wrist surgery (26.2/10,000 RA patients vs. 21.5/10,000 OA patients) . It appears that treatment interventions may be contributing to the overall decrease in orthopedic surgery for patients with RA, although the use of lower-extremity arthroplasty remains high.
Utilization of orthopedic surgery in RA patients
Orthopedic surgery has typically been performed in a large number of RA patients, with up to 50% of patients undergoing an orthopedic procedure over the course of their illness . Recently, however, registry studies in Minnesota and Sweden have suggested that the disease course in RA has become less severe, and the target organ for RA, the joint, was less likely to progress to the end stage, thereby leading to less orthopedic surgery . However, others have demonstrated stable rates of hip arthroplasty (THR) and increasing rates of total knee arthroplasty (TKA) in patients with RA . Moreover, while hand and foot surgery rates may have decreased, the rates of major reconstruction of the lower extremities have remained unchanged . Some studies have demonstrated that RA patients with disease onset after age 65 are less likely to receive aggressive medical treatment, which may contribute to the observation that 2% of the RA patients enrolled in a Medicare database underwent orthopedic reconstructive surgery within 3 years of diagnosis, most commonly TKA . In a population-based study, 34% of all incident RA cases underwent orthopedic surgery over the 30-year course of the study, most commonly arthroplasty . However, the age at which RA patients undergo TKA has increased, while that of patients with degenerative arthritis has decreased, suggesting another benefit of the current treatments .
Although the large databases used for these population-based studies rarely are linked to specific information about medication use, the increased use of DMARDs and biologics during this time period may explain the stable or decreasing rates of arthroplasty. A well-characterized cohort of RA patients was recently studied and demonstrated that the concomitant use of MTX and tumor necrosis factor inhibitor (TNFi) reduced the risk of large joint replacement when compared to TNFi use alone . However, the Finnish RA registry failed to demonstrate a decrease in arthroplasty associated with the use of TNFi, when propensity score matching for on-medication disease activity was performed . Nonetheless, the proportion of arthroplasties now performed for RA has decreased from >20% of all arthroplasty cases in the early 1990s to a consistently low level of 3–4% of the current total number of arthroplasties, which has markedly increased . Regional variation also contributes to the difference in arthroplasty utilization; hand and wrist surgery varied 9fold across hospital referral regions . In addition to patient factors, regions with a high density of subspecialty rheumatologists have a significantly lower rate of hand and wrist surgery (26.2/10,000 RA patients vs. 21.5/10,000 OA patients) . It appears that treatment interventions may be contributing to the overall decrease in orthopedic surgery for patients with RA, although the use of lower-extremity arthroplasty remains high.
Optimizing outcomes of arthroplasty in RA
While the overall status for RA patients has improved, pain and functional outcomes after arthroplasty still lag behind for patients with OA . THA is one of the most reliable surgeries being performed, with improvement reported in quality-of-life measures of pain and function in most patients undergoing THA . Although patients with RA show significant improvements after THA, a higher proportion of RA patients have more pain and worse functional outcomes than those with OA ( Fig. 1 ), and RA remains a risk factor for poorer outcomes compared to patients with OA . While poorer baseline status may explain these results, involvement of multiple joints, typical for RA patients, may also play a role. In one study, RA patients without a prior arthroplasty were more likely to have worse postoperative pain than those who underwent a contralateral arthroplasty . Some studies have shown that the baseline functional status and other involved joints in patients with OA may also affect the postoperative outcomes, so patients with OA who have multiple arthritic joints are not as likely to achieve a good outcome, in particular when the lower extremity is involved .
TKA is also a remarkably successful operation, with large improvements in pain and function outcomes reported by most patients undergoing TKA , although satisfaction is typically lower for these patients when compared to THA . RA patients undergoing TKA in a high-volume orthopedic hospital have been reported to do as well as patients with OA, despite worse preoperative status, and achieved pain and function outcomes which are as good as those for OA patients . Others, however, have found that patients with OA have a significantly better functional outcome after TKA than patients with RA, while postoperative pain was no different between these groups . While worse overall outcomes have been reported for RA, in another large study that utilized an institutional surgical database, RA was associated with having less pain after TKA . While the reasons for these differences are not clear, factors such as study methodology or the temporal span of the included cases might contribute to the differences, given the changes in the status of patients with RA as well as the changes in surgical technique. Nonetheless, while RA patients lag behind patients with OA in some TKA studies, postoperative improvements in pain, function, and quality of life are clinically significant.
Surgical complexity
Coordination of care is important for the patient with RA, including not only the orthopedic surgeon and rheumatologist, but also physical and occupational therapists, social work, and anesthesiologists. While we attempt to do all extremity operations under regional anesthesia in order to protect the cervical spine, the anesthesiologist should be able to perform fiberoptic intubations if general anesthesia is required. The prevalence of cervical spine instability in patients with RA awaiting orthopedic surgical procedures is high; regional anesthesia avoids manipulation of the cervical spine . In light of this, we continue to screen all patients prior to arthroplasty with flexion/extension views of the cervical spine, which are more sensitive to instability than standard views , although this is not the practice in all high-volume centers.
Patients with RA often present with challenging joint deformities due to the disease. They frequently have poor bone quality and osteoporosis, flexion contractures, muscle atrophy, and poor skin condition, which may be due to the disease, medications including CSs, and disuse atrophy. At the hip, patients frequently have bilateral disease and may sometimes have flexion contractures. The occurrence of protrusion, or erosion with medial migration of the femoral head into the pelvis, is more frequent in RA compared to OA. At the knee, flexion contractures are common and RA patients have a higher incidence of valgus deformities compared to OA . In RA patients, the poor soft tissue envelope as well as weak muscles may explain the increased risk of hip dislocation after arthroplasty .
Surgical sequence of arthroplasty
Surgical sequence of procedures for RA needs to be individualized to the patient when multiple joints are involved. Although this has not been formally studied, there is expert consensus on some general principles regarding the sequence of surgeries. In most cases, lower extremities should be addressed before upper extremities in order to maintain ambulation and functional independence. Besides, since lower-extremity surgery typically requires the use of ambulatory aids, thus making upper extremities weight bearing, it is best to defer upper-extremity surgery until ambulatory aids are no longer needed. However, if the patient is unable to use ambulatory aids due to upper-extremity pain and dysfunction, then the upper extremities may need to be addressed first .
While surgery is performed first on the most painful joint, the interactions between joints need to be considered. In the upper extremity, our order of preference is typically wrist, hand, elbow, and then shoulder. A stable wrist is needed to provide a platform for hand functions and establish tendon lengths. Restoring hand function, especially pinch, gives the patient an incentive to use the hand which helps in postoperative therapy after elbow surgery. Elbow surgery may be more predictable than shoulder surgery as the biceps and triceps muscles are usually intact while the rotator cuff muscles of the shoulder are frequently affected. Rotator cuff attrition and tears may be irreparable in the patient with RA, so the range of motion may not be significantly improved after shoulder surgery, although pain relief is predictable. However, if the shoulder is stiff, it may be necessary to restore rotation prior to or soon after elbow surgery to prevent excessive torque on the elbow.
When addressing the lower extremity, the priority according to expert consensus is hindfoot and ankle first, followed by foot, hip, and then knee. The patient should have a pain-free, plantigrade foot before hip and knee replacements are performed in order to prevent undue stress across the implant. A patient with a fixed equinis hindfoot deformity will have difficulty ambulating without affecting the knee. Typically, the hip should be replaced before knee replacement as it is difficult to rehabilitate the knee without flexing the hip. Besides, hip pain referred to the knee will be eliminated and hip flexion contractures will be addressed when the hip replacement is performed first. When possible, patients with significant flexion contractures of both hips should have them replaced during the same surgery in order to prevent leg length discrepancy and to prevent recurrence of the contracture. This also makes their postoperative physical therapy more efficient. The same is true for patients with significant knee contractures as these are best corrected during the same operation. Overall, surgical staging in the patient with multiple joint involvement is complicated and requires careful planning and coordination of care.
Preventing complications
Adverse events
To minimize complications, RA patients undergoing arthroplasty require careful preoperative assessment and perioperative management . While certain complications such as infection and dislocation are increased in RA patients, complication risk is decreased if these patients are operated on by surgeons with RA-specific experience . RA-specific experience is defined as the number of RA surgeries performed annually, which significantly decreased the risk of complications (adjusted hazard ratio (HR): 0.81 per 10 cases; 95% CI, 0.71–0.93; p = 0.002), and was independent of the overall surgeon volume (adjusted HR: 0.84 per additional 10 RA TJA cases; 95% CI, 0.69–0.95; p = 0.009). A retrospective case–control study performed in a high-volume center with high RA-specific experience found no difference in complications including superficial surgical site infections, despite an increase in comorbid conditions and worse preoperative functional status for the patients with RA . Nonetheless, despite the recognition of volume–outcome relationships and the complexity of surgery on RA patients, 78% of RA and 79% of OA patients underwent knee replacement surgery in nonteaching hospitals. For THA, the proportion of patients undergoing THA in nonteaching hospitals is similar, 76% of RA and 75% of OA . For optimal care, elective orthopedic surgery for RA patients undergoing THA and TKA would include centralization of surgery in hospitals with high RA-specific volume and knowledge.
Deep vein thrombosis (DVT)
Deep venous thrombosis and pulmonary embolus are among the most worrisome complications after TKA and THA, with a postoperative incidence of ≥30% if preventive measures are not taken . Despite recommended prophylaxis, patients continue to develop venous thromboembolism (VTE); 1 out of 100 patients undergoing TKA and 1/200 undergoing THA will develop VTE despite receiving appropriate preventive agents . Patients with RA may have an increased overall VTE risk . For a meta-analysis of VTE risk in RA, 10 studies which included RA patients demonstrated a cumulative VTE incidence of 2.18%, and elevated odds ratio (OR) for RA patients compared to an age-, sex-, and co-morbidity-matched control population [(OR) 2.23 (95% CI: 2.02–2.47)], providing evidence for an increased VTE risk for RA. For this meta-analysis, however, studies that included the postoperative setting were specifically excluded . In RA patients undergoing arthroplasty, there was no increase in the VTE rate compared to OA, and VTE risk for RA patients is equivalent to that of OA . Although D-dimer levels increase after TKA and THA, and may be higher in RA compared to OA patients, the incidence of VTE for RA is no different than the incidence in OA undergoing TKA , even after adjusting for potential confounders such as age, sex, admission type, or type of anesthesia . In fact, although VTE risk was significantly higher for RA patients compared to the rest of the population, and hospitalization increases overall VTE risk, there was no further increase in the excess risk observed overall in the year after hospitalization in RA patients . Taken together, these studies suggest that the standard-of-care VTE prophylaxis is effective for patients with RA .
Cardiovascular events
There is a well-defined increase in the risk of cardiovascular disease associated with RA . When RA patients in their 5th and 6th decades are screened with carotid ultrasound using carotid intimal medial thickness as an indicator of atherosclerotic cardiovascular disease (ASCVD), patients with RA have more than twice the prevalence of atherosclerotic plaque as age-matched normal controls . In fact, the risk of ASCVD in RA has been estimated to be as high as the risk associated with diabetes . Cardiac morbidity and mortality associated with RA has been linked to the ongoing inflammatory burden, RA severity manifests by erosive disease, as well as disability and poor functional status . Traditional cardiac risk factors such as smoking, hypertension, diabetes, or family history of acute myocardial infarction (AMI) further increase the risk, but therapy with TNFi or MTX may decrease the risk . Cardiac events complicate 0.6% of the orthopedic surgeries, but in patients with a prior history of cardiac disease, the rate increases to 6.5% . The American College of Cardiology/American Heart Association recommend screening patients prior to surgery by careful history, and endorse the demonstration of the ability to achieve four metabolic equivalents (METS) of exertion (walking up stairs, doing housework) prior to an intermediate risk procedure such as arthroplasty. Intermediate risk procedures have a 1–5% risk of serious cardiac events including AMI . However, RA patients may be too disabled prior to arthroplasty to demonstrate an adequate functional capacity. In addition, RA patients are less likely to report angina and are more likely to have silent MIs, so preoperative cardiac risk assessment may require further testing or cardiology consultation . However, large databases have not revealed an increase in cardiac events after arthroplasty for patients with RA , and patients with diabetes have almost 3 times the risk of cardiac events (0.34% vs. 1.07%; P < 0.001) after intermediate risk procedures when compared to patients with RA . Taken together, these data suggest that the cardiac risk for patients with RA associated with undergoing intermediate risk procedures such as arthroplasty has been mitigated by current practices.
Infection
Superficial wound infection and prosthetic joint infections (PJIs) have increased along with the increase in arthroplasty utilization. Infections may occur early after surgery and are attributed to the bacteria introduced at the surgical site at the time of surgery, or late, which are thought to be blood born . The incidence rates of prosthetic knee infections have remained high, 2.05% (CI, 1.86%–2.23%) in 2001 and 2.18% in 2009 (CI, 1.99%–2.37%). Similarly, the rate of infected THA has remained high between 2001 [1.99% (CI, 1.78%–2.21%)] and 2009 [2.18% (CI, 1.97%–2.39%)] . In addition, the odds of readmission within 90 days of arthroplasty for RA patients has increased from 0.89 (95% CI 0.46–1.71) in 2009 to 1.34 (95% CI 0.69–2.61) in 2010, and to 1.74 (95% CI 1.16–2.60) in 2011, most commonly for infection .
The risk of both early and late PJI is increased in patients with RA. This was demonstrated in a study utilizing data from the Finnish Arthroplasty Registry, wherein the HR for reoperation for infection was 1.86 (95% CI 1.31–2.63) for patients with RA compared to patients with OA . A systematic review of 40 selected studies showed that RA patients had a significantly higher risk of infection were systematically reviewed; five included studies compared the PJI rate in RA versus OA, demonstrating an increased risk of infection for RA patients . Other studies have also demonstrated higher odds of infection for RA patients than OA with an OR of 10.30 (95% CI 1.31–80.2). In patients with RA, the host factors associated with infection include CS dose above 15 mg/day and being underweight ; high disease activity as well as extra-articular manifestations of RA, leukopenia, and poor functional status are some of the other factors . While use of CSs increased the risk of infection, the use of DMARDs was not considered as a risk factor in all the studies Surgical factors such as the use of prophylactic systemic antibiotics within an hour of incision, laminar flow in the operating room, and the use of antibiotic-laden cement for prosthesis fixation are generally favored for the reduction in infection. In a meta-analysis reporting on the outcome in 36,033 THA, a 50% reduction in infection was reported with the prophylactic use of antibiotic-laden cement . However, another meta-analysis including both THA and TKA showed that the use of antibiotic-laden cement had only a protective effect in the prevention of deep infection (RR = 0.41; 95% CI, 0.17–0.97; P = 0.04) . Other studies using large administrative databases have not consistently demonstrated a decrease in infection risk using antibiotic-laden cement, however .
The presence of an orthopedic implant increases the pathogenic potential of a small number of bacteria to infect an implant by forming a bacterial biofilm, which adheres to the prosthetic material, rapidly forming microcolonies . This makes both diagnosis and treatment of infection more difficult, because biofilm maturation via aggregation and formation of a polysaccharide matrix make the bacteria inaccessible to normal host defenses . Cytokines including tumor necrosis factor-alpha (TNFa) play an important role in the initial host defense, and may explain the somewhat increased risk of PJI observed in patients receiving TNFa inhibitors . Perioperative DMARD and biologic management may be an area where infection risk modification is feasible.
Medication management
Although a high proportion of patients with RA are treated with DMARDs and biologics at the time of undergoing THA and TKA, there are no definitive answers to questions regarding optimal medication management . As the standardized infection ratio (SIR) was 0.46 for deep infection in one high-volume orthopedic hospital, formal study of infection risk associated with all DMARDs and biologics would be extremely difficult given the number of patients ( N > 50,000) needed to perform an adequately powered study. The study is further hampered by the varying stop and start dates for medications in the perioperative period, which make comparisons between centers difficult . Nonetheless, certain information is available for consideration.
TNFis and other biologics
The utilization of TNFis has clearly brought benefits to RA patients including improved quality of life and function, even for those with advanced or refractory RA who may be more likely to undergo joint replacement surgery . Over 40% of 159 RA patients undergoing TKA in a high-volume center were treated with biologics, and 67% were on nonbiological DMARDs, with no increase in either superficial or deep infection compared to matched controls . However, there are recognized general infection risks associated with TNFi use, recently confirmed in a meta-analysis that additionally demonstrated the risk to be dose dependent . Infection risk associated with elective orthopedic surgery and the recent use of TNFi was studied using pooled data representing 3681 patients with recent exposure to TNFis and 4310 with no recent exposure to TNFis at the time of surgery for a systematic review and meta-analysis ( Fig. 2 ). This study demonstrated a significant increased risk of infection in the patients with recent exposure to TNFis, with an OR of 2.47 (95% CI = 1.66–3.68; P < 0.0001). Although this study supports the practice of discontinuing TNFi prior to elective orthopedic surgery, there was no consistent adjustment for other potential confounders such as steroid use and disease activity, and importantly, not all included studies corrected for confounding by indication with propensity scores. Taken together, these data indicate that infection risk, both superficial and deep, may be increased between 2% and 4% in patients on biologics at the time of surgery; in order to prevent a single infection, administration of biologics would need to be stopped in 25–50 patients. Given the severe consequences of PJI, and until definitive studies are available, the guidelines issued by the American Society of Rheumatology can be followed (Class C evidence). These advise that biologic agents should not be used during the perioperative period, for at least 1 week prior to and 1 week after surgery if risk of infection is for all but minor surgeries such as cataract explant .
Related posts:
Treating to target in established rheumatoid arthritis: Challenges and opportunities in an era of novel targeted therapies and biosimilars
How does established rheumatoid arthritis develop, and are there possibilities for prevention?
Monitoring in established RA: Role of imaging and soluble biomarkers
Evaluating and mitigating fracture risk in established rheumatoid arthritis
Patient-centred care in established rheumatoid arthritis
Quality in rheumatoid arthritis care
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
