Knee Replacement or Nonsurgical Care First
A knee replacement relieves pain for most patients with end-stage arthritis, yet it adds surgical risk and leaves a sizable minority dissatisfied, so picking the right patient and the right time decides who benefits.
Key terminology
- Total knee arthroplasty (TKA)
- This operation replaces the worn ends of the femur and tibia with metal components and a plastic spacer. It is the standard operation for end-stage knee arthritis.
- Knee Injury and Osteoarthritis Outcome Score (KOOS)
- This patient questionnaire scores knee pain, symptoms, daily function and quality of life from 0 to 100. A higher score means a better knee.
- Intention-to-treat
- This analysis counts every patient in the group they were randomized to, even after a switch in treatment. It keeps randomization intact and shrinks the gap between arms when many cross over.
- Articular cartilage
- This smooth layer caps the ends of the femur and tibia inside the knee. Its gradual loss is osteoarthritis, and an x-ray shows it as a narrowed joint.
- Joint space narrowing
- The gap between femur and tibia shrinks on x-ray as cartilage wears away. It is best seen on a standing film, when body weight closes the gap.
- Kellgren-Lawrence grade
- This 0 to 4 scale grades knee arthritis on plain films by osteophytes and joint space narrowing. Grades 3 and 4 mean moderate to severe disease.
- Appropriateness criteria
- These rules rate a planned operation as appropriate, inconclusive or inappropriate from symptoms, exam and x-rays. They aim to cut variation in who gets surgery.
Takeaways
- A supervised nonsurgical program is first-line care for knee OA, even in surgical candidates. The program in the Danish trials ran 12 weeks and combined exercise, education, dietary advice, insoles and pain medication, delivered by physical therapists and dietitians. In patients not yet eligible for total knee arthroplasty (TKA), it beat written advice by 7.0 points on a 0-100 pain and function score at 2 years (Skou et al, Osteoarthritis Cartilage 2018;26:1170-1180, PMID 29723634).
- In surgical candidates, TKA plus rehab relieves pain and restores function better than rehab alone. A trial randomized 100 TKA candidates. At 12 months the Knee Injury and Osteoarthritis Outcome Score rose 32.5 points after TKA and 16.0 with the program alone (adjusted difference 15.8), and the gap was 18.3 points at 2 years. Patients could not be blinded, so a self-reported score may favor surgery (Skou et al, N Engl J Med 2015;373:1597-1606, PMID 26488691; Skou et al, Osteoarthritis Cartilage 2018;26:1170-1180, PMID 29723634).
- A knee replacement adds serious complications that a nonsurgical program avoids. In the Danish trial of 100 TKA candidates, the TKA group had 24 serious adverse events vs 6 with the program alone (P=0.005). These are counts of events, not of patients. Patients weighing surgery should hear both the larger gain and the higher risk (Skou et al, N Engl J Med 2015;373:1597-1606, PMID 26488691).
- Most surgical candidates who start with a nonsurgical program do not have a TKA within 2 years. Of those randomized to the program alone, 26% had a TKA by 12 months and 32% by 2 years, so about 2 in 3 delayed surgery. The trial used intention-to-treat analysis, so these crossovers stayed in the nonsurgical arm and shrink the measured gap (Skou et al, N Engl J Med 2015;373:1597-1606, PMID 26488691; Skou et al, Osteoarthritis Cartilage 2018;26:1170-1180, PMID 29723634).
- Knee replacement for mild radiographic arthritis predicts persistent pain and dissatisfaction. Of 49 painful TKAs with no mechanical cause, 49% had a Kellgren-Lawrence grade below 3 before surgery vs 5% of routine TKAs. In hip and knee replacements, low grade raised the odds of high pain (OR 1.56). Severe joint space narrowing (articular cartilage gone) predicts more improvement (Polkowski et al, Clin Orthop Relat Res 2013;471:162-168, PMID 22923158; Valdes et al, Semin Arthritis Rheum 2012;41:568-575, PMID 21868060; Keurentjes et al, PLoS One 2013;8:e59500, PMID 23573200).
- A sizable minority of patients remain dissatisfied after a primary TKA, mostly when expectations go unmet. In 1,703 Ontario TKAs, 19% were not satisfied with the outcome. Unmet expectations were the strongest predictor (10.7 times the risk), ahead of a poor 1-year function score, pain at rest before surgery and a complication needing readmission (Bourne et al, Clin Orthop Relat Res 2010;468:57-63, PMID 19844772).
- Experts disagree on what makes a TKA appropriate, and US surgeons share no agreed criteria. Appropriateness criteria applied to 205 US TKAs rated 34.3% inappropriate, vs about 20% in earlier non-US studies. In a follow-up, the inappropriate group gained only 2.3 function points at 1 year vs 19.8, yet 1- and 2-year recovery did not differ significantly between groups (Riddle et al, Arthritis Rheumatol 2014;66:2134-2143, PMID 24974958; Riddle et al, Arthritis Care Res 2015;67:349-357, PMID 25132662).
- NICE bases referral for joint replacement on symptoms and quality of life after nonsurgical care. Its guideline advises referral when joint symptoms substantially affect quality of life and nonsurgical management has not worked or is unsuitable. It advises against using a scoring tool to decide who gets referred (NICE guideline NG226, 2022).