Impact of hospital volume on the economic value of computer navigation for total knee replacement

Abstract

Background: An aim of the use of computer navigation is to reduce rates of revisions of total knee replacements by improving the alignment achieved at the surgery. However, the decision to adopt this technology may be difficult for some centers, especially low-volume centers, where the cost of purchasing this equipment may be high. The purpose of this study was to examine the impact of hospital volume on the cost-effectiveness of this new technology in order to determine its feasibility and the level of evidence that should be sought prior to its adoption.

Methods: A Markov decision model was used to evaluate the impact of hospital volume on the cost-effectiveness of computer-assisted knee arthroplasty in a theoretical cohort of sixty-five-year-old patients with end-stage arthritis of the knee to coincide with the peak incidence of knee arthroplasty in the United States.

Results: Computer-assisted surgery becomes less cost-effective as the annual hospital volume decreases, as the cost of the navigation increases, and as the impact on revision rates decreases. Centers at which 250, 150, and twenty-five computer-navigated total knee arthroplasties are performed per year will require a reduction of the annual revision rate of 2%, 2.5%, and 13%, respectively, per year over a twenty-year period for computer navigation to be cost-effective.

Conclusions: Computer navigation is less likely to be a cost-effective investment in health-care improvement in centers with a low volume of joint replacements, where its benefit is most likely to be realized. However, it may be a cost-effective technology for centers with a higher volume of joint replacements, where the decrease in the rate of knee revision needed to make the investment cost-effective is modest, if improvements in revision rates with the use of this technology can be realized.

Fig. 1

The Markov model used to model patients with end-stage arthritis of the knee. Each patient is treated with a total knee arthroplasty, either with computer-assisted surgery (CAS) or without computer-assisted surgery (TKA). If a patient survives the operation, that patient is considered to stay well until the time of death from other causes or the need for a revision. If a patient has a revision and survives, he or she is assumed to stay well with that revision until the time of death from other causes. The model continues until all patients reach the death health state.

Fig. 2

Results of the sensitivity analysis demonstrating the relationship between hospital volume and the reduction in the revision rate following primary total knee arthroplasty with computer-assisted surgery (CAS). The blue area demonstrates the profiles for which total knee arthroplasty with computer-assisted surgery is more cost-effective (the cost is less than $50,000 per quality-adjusted life-year [QALY]). The green area demonstrates the profiles for which total knee arthroplasty without computer-assisted surgery is more cost-effective. The reduction in the revision rate is expressed as a percentage of the revision rate following total knee arthroplasty performed without navigation (TKA). For example, for computer navigation to be cost-effective at a center at which one hundred and sixty arthroplasties are performed per year, the reduction in revisions with the use of navigation would have to be just under 3%, or the revision rate would have to be just under 97% of the revision rate for total knee arthroplasties performed without navigation.

Fig. 3-A

Figs. 3-A, 3-B, and 3-C Results of the sensitivity analyses demonstrating the relationship between the annual cost of computer navigation (CAS) and the reduction in the revision rates at centers at which joint replacements are performed. The blue area demonstrates the profiles for which total knee arthroplasty with computer-assisted surgery is more cost-effective (the cost is less than $50,000 per quality-adjusted life-year [QALY]). The green area demonstrates the profiles for which total knee arthroplasty without computer-assisted surgery is more cost-effective. The reduction in the revision rate is expressed as a percentage of the revision rate following total knee arthroplasty performed without navigation (TKA). Fig. 3-A Results for centers at which 250 joint replacements are performed per year. For example, for computer navigation costing $48,000 annually to be cost-effective, the center would need to have just under a 2% reduction in revisions with the use of navigation, or have a revision rate of just under 98% of the revision rate for total knee arthroplasties performed without navigation.

Fig. 3-B

Results for centers at which 150 joint replacements are performed per year. For example, for computer navigation costing $48,000 annually to be cost-effective, the center would need to have a 2.5% reduction in revisions with the use of navigation, or have a revision rate of 97.5% of the revision rate for total knee arthroplasties performed without navigation.

Fig. 3-C

Results for centers at which twenty-five joint replacements are performed per year. For example, for computer navigation costing $48,000 annually to be cost-effective, the center would need to have a 13% reduction in revisions with the use of navigation, or have a revision rate of 87% of the revision rate for total knee arthroplasties performed without navigation.