The United States' current spend on healthcare equates to 18% of GDP or USD 3.81 trillion and it is predicted rise to 19.7% by 2028 equating to USD 6.2 trillion. The trend is similarly reflected by most other countries and is simply unsustainable. Healthcare costs must be reduced in both the short- and long-term.

In the short-term, I envision a more optimized health delivery system, one where AI and machine learning enable physicians to incorporate the power of large data sets into personalized patient care. We know every patient is unique, and as is eloquently pointed out by Dr. Keller, the importance of intuition in medical decision-making should also not be underestimated. The synergy of this data driven personalized care will ultimately result in the most appropriate treatment plan within the most appropriate timescale.

Optimization also extends to reducing current resource burdens on the healthcare system, such as periprosthetic joint infection. Both the American and Australian registry annual reports 2022 highlight infection as the most common reason for revision in both hip and knee arthroplasties. The challenging diagnosis and treatment mean that any modifiable risk factors, such as metal bearings, should be avoided. Optimized healthcare delivery is a great value proposition which allows us to provide improved and more cost-effective care to more people in the immediate future.

When considering the long-term, we must be aware that the investment needed in technology is required today, not tomorrow. Robotics, navigation, and more durable devices will all assist in the reduction of future costs by reducing the revision burden for arthroplasty patients. The challenge, however, is predicting the long-term benefit in the short-term. Randomized Controlled Trials, as Prof. Stengel also recognizes, are not the be all and end all as ideal predictors for long-term benefits, despite being frequently required by state agencies or healthcare systems for acceptance.

Better methodologies to assess big data are growing quickly in our field and include the evolution of registry data using machine learning algorithms as well as predictive analytics and AI models which can be used to determine the potentials benefits of new technology. It stands to reason that a tool which is more precise, such as robotics or navigation and a bearing surface that reduces not only wear, but the risk of revision for infection and eliminates metal sensitivity should always be the default and not the exception. In this I can only agree with Prof. Trebše, prevention of infection is an absolute priority.

Much of the research in the field of human longevity and lifespan is based on optimization and personalization to improve health and extend life. David Sinclair, author of ‘Lifespan’ hypothesizes that in the next ten years we will have the knowledge of how to extend life significantly. If these hypotheses become a reality a hip replacement would have to be durable enough to potentially last 50 years. While investment in new technology is costly, the current trends in spending are ultimately unsustainable going forward. The question should not be "Can we afford to use these new technologies?", the question should be "Can we afford NOT to use these new technologies?".

As a final note, if the patients were to be objectively presented with all the current evidence of new technologies, it is my belief that most of them would choose to receive a knee replacement performed with robotics, a hip replacement optimized with navigation, and consolidated bearing surfaces that have the potential to last a lifetime, such as ceramic-on-ceramic. So why the hesitant uptake of these technologies? I’ll let you answer that question.

Stefan Kreuzer MD, MSc