CORR Insights®: Do Patients Live Longer After THA and Is the Relative Survival Diagnosis-specific?

Where Are We Now?

According to the World Health Organization, life expectancy around the world has increased, and is currently above 80 years in 29 countries [6]. As people live longer, they will need more orthopaedic care. The rising number of THAs performed annually, for example, has the potential to cause an economic burden on healthcare systems around the world.

Additionally, advances in medical knowledge and technology have led to a relaxation of some surgical indications. For example, the introduction of crosslinked polyethylene resulted in increased longevity of THA implants [1], and perhaps as a result of improvements like this, younger patients are being offered the procedure. Because of the young age of these patients, healthcare economics questions arise not just about quality of life after these procedures or their cost-effectiveness, but about the survival of the patients themselves after THA.

In this context, the paper by Cnudde and colleagues is important because it compares the life expectancy in patients undergoing THA to that of the general population. Based on the Swedish Hip Arthroplasty Register, the study shows that THA for primary osteoarthritis is associated with slightly greater life expectancy at least during the 10-year period following implantation. However, the reason for that association remains unclear. Other groups assessed mortality and complication rates in subgroups of patients after THA compared to different cohorts of patients after THA. For example, by analyzing a large national database, Boniello and colleagues [2] found that octogenarians had a higher mortality and more complications than younger patients within the first 30 days. But despite the unique advantages associated with national joint registries, an inherent weakness of these tools is that, generally speaking, the cohorts are insufficiently characterized in terms of patient-level diagnostic information such as physical activity, overall health, comorbidities, or disease severity [3].

Where Do We Need To Go?

As economic pressure on all healthcare systems continues to increase, quantitative data regarding the cost and outcomes of THA will become more relevant. Policymakers need to know the value of a given treatment to allocate resources accordingly. And as surgeons, we need more data in order to advise a patient about what one can and cannot expect from an intervention and how it will affect them for the rest of their lives. We need to identify subgroups of patients who do not improve as expected, as well as those who express dissatisfaction with their surgical results. To quantify the benefit of an operation, many variables should be considered including subjective parameters such as quality of life, activity levels, pain, and ability to work.

The value of a particular surgical procedure can only be measured if a treated cohort is compared against a control. As various subjective factors play into the process of indicating a THA, selection bias will cloud a thorough analysis of its efficiency. In order to have the best-possible controls, detailed descriptions of large populations are necessary including the severity of disease, comorbidities, activity levels, and socioecomonic status.

Looking at patient survival data is a crude but essential first step. Other endpoints of interest such as quality-adjusted life-years (the product of quality of life and length of life [4]) and patient-reported outcome measures (PROM) are of equal importance. One issue with institutional and national registries is that different PROMs are utilized for different treatments and patient groups, complicating comparisons among various treatments and across populations. Is it possible to unify outcome measures in national registries?

How Do We Get There?

While randomized controlled trials do minimize selection bias, their ability to detect rare events or to quantify resource utilization in normal practice is poor [3], and their results may not be necessarily applicable to the entire healthcare systems. Registry studies and big-data analysis from other large databases will continue to give us important information regarding the value and safety of interventions and implants. And because we can find excellent long-term performances of orthopaedic implants with survival rates of over 93% at 10 years for the most popular combinations of stems and cups [1], large-scale clinical use without very solid data on revision rates will no longer be acceptable. The FDA’s clinical research Phase I-IV studies identify ineffective or dangerous treatments as they are established in the pharmaceutical industry [5]; similar studies are long overdue in orthopaedics. For example, a new implant should first be tested in a small cohort of selected patients. Only if its performance is sufficient at mid-term, should the implant be tested in subsequent studies in larger cohorts. Such a scheme will slow down development, but the innovations will become more sustainable.