Total wrist arthroplasty and total wrist arthrodesis in rheumatoid arthritis: a decision analysis from the hand surgeons' perspective

Abstract

Purpose: Treatment of severely destroyed rheumatoid wrists with total wrist arthroplasty or total wrist arthrodesis relies on patient and surgeon preferences rather than rigorous prospective outcomes data. The purpose of this study is to develop a decision analytic model of arthroplasty and arthrodesis in rheumatoid arthritis (RA) using utility values obtained from a random sample of hand surgeons.

Methods: A utility survey using a time trade-off design was administered to 175 members of the American Society for Surgery of the Hand. Based on the results of the survey, the utility values that surgeons assign to health states associated with arthroplasty and arthrodesis and their complications were calculated. By combining utility values with complication rates in the published literature, we developed a decision tree to calculate the expected quality-adjusted life years (QALYs) for each procedure compared to living with a painful RA wrist.

Results: Based on surgeon preferences, living for 30 years with a painful, poorly functioning RA wrist (utility = 0.54) is associated with 16.2 QALYs. Treatment with arthroplasty (utility = 0.85) is associated with 25.5 QALYs, a gain of 9.3 QALYs over nonsurgical management. Arthrodesis (utility = 0.82) is associated with 24.6 QALYs, a gain of 8.4 QALYs over nonsurgical management. Arthroplasty is associated with a small incremental increase in QALYs (0.9) compared to arthrodesis.

Conclusions: Based on utility scores, hand surgeons feel that living with a painful, poorly functioning RA wrist for 30 years is worth approximately half as many years with a painless, well-functioning wrist. The outcomes for arthroplasty and arthrodesis are valued more than nonsurgical management. On the basis of its higher expected gain in QALYs, arthroplasty should be the preferred treatment. The minimal increase in utility for arthroplasty over arthrodesis suggests however, that surgeons do not view arthroplasty as superior to arthrodesis.

Type of study/level of evidence: Economic and decision analysis IV.

Figure 1

Outline of decision tree for TWA, TWF, and non-operative management. Square indicates decision node, circle indicates chance node, and triangle indicates terminal node.

Figure 2

Decision tree with calculations. Probabilities are shown for each category of possible complications. Expected QALYs for each management strategy are shown on the left.

Figure 3

Equation used for calculation of QALYs.

Figure 4

Sensitivity analysis evaluating change in expected QALYS with range of 15 to 35 healthy remaining years of life.