Background: Interest in unicompartmental knee arthroplasty has
recently increased in the United States, making a firm understanding of the
indications for this procedure important. The purpose of this study was to
examine the cost-effectiveness of unicompartmental knee arthroplasty compared
with total knee arthroplasty in elderly low-demand patients.
Methods: A Markov decision model was used to evaluate the
cost-effectiveness of unicompartmental knee arthroplasty as compared with
total knee arthroplasty in the elderly population. Transition probabilities
were estimated from the Norwegian Arthroplasty Register and the arthroplasty
literature, and costs were based on the average Medicare reimbursement for
unicompartmental, tricompartmental, and revision knee arthroplasties. Outcomes
were measured in quality-adjusted life-years.
Results: Our model showed unicompartmental knee arthroplasty to be a
cost-effective strategy for this population as long as the annual probability
of revision is <4%. The cost of unicompartmental knee arthroplasty must be
greater than $13,500 or the cost of total knee arthroplasty must be less than
$8500 before total knee arthroplasty becomes more cost-effective.
Conclusions: Our model suggests that, on the basis of currently
available cost and outcomes data, unicompartmental knee arthroplasty and total
knee arthroplasty have similar cost-effectiveness profiles in the elderly
low-demand patient population. However, several important parameters that
could alter the cost-effectiveness analysis were identified; these included
implant survival rates, costs, perioperative mortality and infection rates,
and utility values achieved with each procedure. The thresholds identified in
this study may help decision-makers to evaluate the cost-effectiveness of each
strategy as further research characterizes the variables associate with
unicompartmental and total knee arthroplasties and may be helpful for
designing future appropriate clinical trials.
Level of Evidence: Economic and decision analysis, Level
II. See Instructions to Authors for a complete description of levels of