Are Age-based Cutoffs for THA Logically Consistent?

Although THA is routinely offered to patients age 85 years and older, surgeons might hesitate to operate on patients deemed "too young." This study tests whether such hesitancy reflects a consistent standard.

(1) How much utility (net benefit) can be expected from THA, as a function of patient age and operation-related clinical variables? (2) What relative cost must be attributed to revision of a THA such that the imposition of this penalty renders the expected utility of the operation too low to justify it in younger people?

A calculator was programmed using a state-transition Markov model to estimate the expected lifetime benefit of THA as a function of patient age and other user-specified parameters. A novel unit of benefit, the joint-adequate life year (JALY), was defined. One JALY is earned in each postoperative year that the patient has an intact prosthesis and adequate clinical outcome. JALYs can also be forfeited when a patient enters a revision state. Thus, previously accumulated JALYs may be partially or completely offset, resulting in a net lifetime total that is lower than the total years of an adequate outcome. Users of the calculator specify patient sex and age at the time of THA (we modeled sex rather than gender in order to be able to derive life expectancy from sex-specific life tables published by the Social Security Administration), the anticipated fraction of patients expected to have a clinically adequate outcome (termed here the "satisfaction fraction"), prosthetic life expectancy, the expected annual risk of revision attributed to catastrophic events such as fracture or infection, and a discount rate, which is used to convert future gains and losses to their net present value. The prosthetic life expectancy parameter was used to calculate an annual risk of needing revision for wear. Users also assign a negative value to the revision state, representing the number of JALYs that a patient would be willing to forgo to avoid that state. This penalty was applied in the first instance the failure state was encountered, and a value 1.5 times this penalty was applied for any subsequent encounter. The model was then interrogated to provide the expected JALYs earned as a function of failure-state penalties ranging from 0 to 40 JALYs. For this calculation, prosthetic life expectancy was set to 15 years, the anticipated satisfaction fraction was set to 90%, the annual rate of catastrophic failure was set to 0.5%, and no temporal discounting was applied. Sensitivity analyses were then performed for different values of prosthetic life expectancy, satisfaction fraction, catastrophic failure rate, second-revision penalty ratio, and with 2% and 10% temporal discounting applied. Based on the common use of THA in 85-year-old patients (a group with a life expectancy of about 6 years after THA), a benchmark of six JALYs was established, as logically no patient can have more "adequate joint years" than total years of life. We then determined the penalty that must be assigned to the revision state, such that a 45-year-old male's expected benefit would fall below this six-JALY benchmark.

A cohort of 45-year-old males can be expected to earn 25 lifetime JALYs under the base case assumptions and a revision penalty of five JALYs. Lifetime JALYs for this cohort do not drop below the threshold of six until the revision penalty reaches 25 JALYs. Sensitivity analysis demonstrated that this finding was robust across wide variation in clinically plausible values for prosthetic life expectancy, satisfaction, failure rates, and discounting.

Hesitation to offer THA to 45-year-old patients with impairing hip disease is logically inconsistent with the procedure's broad acceptance in 85-year-old patients. This is because under plausible assumptions of the clinical parameters, 45-year-old patients stand to accrue considerably more lifetime benefit. Indeed, even accounting for the inevitable failures younger patients may encounter (and the penalties these failures impose), such patients can expect to accrue more than six JALYs across their lifetimes unless implausibly large failure penalties are assumed.

Level II, Economic and Decision Analyses.