Total knee replacement: an evidence-based analysis
Medical Advisory Secretariat
Abstract:Objective: The aim of this review was to assess the effectiveness, in terms of pain reduction and functional improvement, and costing of total knee replacement (TKR) for people with osteoarthritis for whom less invasive treatments (such as physiotherapy, analgesics, anti-inflammatory drugs, intra-articular steroids, hyaluronic acids, and arthroscopic surgery) have failed. Clinical need: Osteoarthritis affects an estimated 10% to 12% of Canadian adults. The therapeutic goals of osteoarthritis treatment are to improve joint mobility and reduce pain. Stepwise treatment options include exercise, weight loss, physiotherapy, analgesics, anti-inflammatory drugs, intra-articular steroids and hyaluronic acids, arthroscopic surgery, and, in severe cases, total joint replacement with follow-up rehabilitation. These treatments are delivered by a range of health care professionals, including physiotherapists, occupational therapists, family physicians, internists, rheumatologists, and orthopedic surgeons. TKR is an end-of-line treatment for patients with severe pain and functional limitations. More women than men undergo knee replacement, and most patients are between 55 and 84 years old. The technology: TKR is a surgical procedure in which an artificial joint or prosthesis replaces a damaged knee joint. The primary indication for TKR is pain, followed by functional limitation. Usually, a person's daily activities must be substantially affected by pain and functional limitations for him or her to be considered a candidate for TKR. There are 3 different types of knee replacement prostheses. Non-constrained prostheses use the patient's ligaments and muscles to provide the stability for the prosthesis. Semi-constrained prostheses provide some stability for the knee and do not rely entirely on the patient's ligaments and muscles to provide the stability. Constrained prostheses are for patients whose ligaments and muscles are not able to provide stability for the knee prosthesis. The most common risks and complications associated with TKR are deep venous thrombosis, infection, stiffness, loosening, and osteolysis. To prevent deep venous thrombosis, patients are treated with heparin prophylactically and/or given support stockings to wear. Patients are also given antibiotics for 24 hours after surgery to minimize the risk of infection. Stiffness is another associated complication. In most patients, it can be avoided by keeping the knee moving in the days and weeks following surgery. THE NATIONAL INSTITUTES OF HEALTH IN THE UNITED STATES CONCLUDED THAT THE INDICATIONS FOR TKR SHOULD INCLUDE THE FOLLOWING: radiological evidence of joint damage, moderate to severe persistent pain that is not adequately relieved by nonsurgical management, and clinically significant functional limitation resulting in diminished quality of life. Review strategy: In March 2005, the following databases were searched: Cochrane Library International Agency for Health Technology Assessment (first quarter 2005), Cochrane Database of Systematic Reviews (first quarter 2005), Cochrane Central Register of Controlled Trials (first quarter 2005), MEDLINE (1966 to March 2005), MEDLINE In-Process and Other Non-indexed Citations (1966 to March 14, 2005), and EMBASE (1980 to 2005 week 9). The Medical Advisory Secretariat also searched Medscape on the Internet for recent reports on trials that were unpublished but that were presented at international conferences. In addition, the Web site Current Controlled Trials (www.controlled-trials.com) was searched for ongoing trials investigating TKR or unicompartmental knee replacement. No studies were identified that compared TKR to an alternative treatment. Several studies have been reported that compare preoperative measurement scores on targeted measures of functioning and pain to postoperative measurement scores in patients undergoing various TKR procedures. In order for the Medical Advisory Secretariat to measure the effectiveness of TKR and to compare the effectiveness of TKR across studies, effect sizes were calculated in studies that reported the standard deviations of the preoperative and postoperative measurement scores. Percent change was also calculated. For this review, a 20% improvement in outcome score was defined as the minimal clinically important difference. Summary of findings: Overall, patients who undergo TKR surgery for osteoarthritis have substantial improvements in terms of reduction of pain and improvement of function. A comparison of the mean effect score and the percent change in 19 studies that reported preoperative and postoperative outcome scores for patients who had TKR showed that the procedure is effective. The 19 studies included patients of various ages and used a variety of prostheses and techniques to implant the device. TKR was effective in all of the studies. The revision rates ranged from 0% to 13% in the studies that reported at least 5 years of follow-up. As for the factors that predict TKR outcomes, a variety of factors have been evaluated, including obesity, age, gender, prosthesis design, and surgical techniques; however, none of these have been shown to predict outcomes (pain or function) consistently across studies. However, the regression analyses identified accounted for only 12% to 27% of the variance, indicating that over 70% of the variance in the outcomes of TKR is unexplained. In terms of the timing of TKR surgery, 2 studies found that the severity of osteoarthritis does not predict outcome, but 1 study was found that higher functioning patients had significantly less pain and better function up to 2 years after surgery compared with lower functioning patients. It is important to note that the patients in the low and high function groups were evenly matched on comorbid conditions. Unicompartmental knee replacement surgery seems to be as effective as TKR surgery for people who meet the indications for it. This is a subset of people who have osteoarthritis of the knee, because for unicompartmental knee replacement to be indicated, only 1 (usually the medial) compartment of the knee can be affected. Patients who undergo this kind of surgery seem to have shorter hospital stays and faster recovery times than do patients who have TKR surgery. Conclusion: There is substantial evidence to indicate that TKR effectively reduces pain and improves function.