As technology continues to advance, so too does the possibility of better outcomes in spine surgery.
As more physicians lean into AI, software, mapping capabilities and robotics, patients are expected to continue to see more efficient procedures with fewer complications and revision rates.
Theresa Pazionis, MD, assistant professor of orthopedic surgery and sports medicine at the Lewis Katz School of Medicine at Temple University in Philadelphia, spoke with Becker's about her work in 3D-printed spinal devices and how AI can be used for preoperative planning and predictive modeling.
Question: How has the field of AI and robotics changed in spine surgery since you began practicing?
Dr. Theresa Pazionis: When I began practicing in 2014, spine surgery relied heavily on freehand screw placement with minimal navigation. We were templating spine surgeries on paper or early versions of surgimap, which gave us great outcomes but did not have the level of integration and consistency we see with robotic planning platforms today. Over time, we’ve integrated AI and robotics into preoperative planning, using systems like the Globus Robot, which integrates with surgimap, Medtronic AiBLE, Mazor, UNiD adaptive spine intelligence and patient specific rod and Aprevo’s personalized interbody implants. These technologies enable us to map surgical goals with high precision, optimize screw placement and design patient-specific implants, resulting in more consistent outcomes, fewer complications and improved correction accuracy compared to traditional methods. We are also able to track patient and surgeon specific outcomes using radiomics (data extraction from serial radiographs) to use predictive analytics to better plan and personalize surgical plans for our patients.
Q: What does the future of 3D-printed spinal implants look like?
TP: It is important to emphasize that robotics and customized implants are not a substitute for excellent surgical technique, medical optimization including bone health and sarcopenia management, and excellent and efficient intraoperative carpentry are and always will be the gold standard of care. These implants are just the icing on the cake. The future is trending towards personalized surgical care- Using these implants is just like using any other implant a spine surgeon typically would but the enhanced emphasis on surgical planning does give globally more reproducible outcomes. Whether or not we are using personalized or off the shelf implants, planning using predictive analytics and robotics platforms allows us to give the patient the best surgery for their anatomy and physiology. Companies like Carlsmed are producing patient-specific cages and implants tailored to individual anatomies. These allow for better load distribution across the endplate as opposed to point loading, more precise correction and reduced revision rates. With systems like Medtronic’s UNiD patient specific rods, we now leverage predictive analytics and digital twins to simulate outcomes, ensuring the best surgical strategy. This not only benefits the patient with improved quality of life but also reduces costs for healthcare systems by minimizing reoperations.
Q: What are some of the challenges you have run into being one of the first people to pilot these implants?
TP: The biggest challenge is cost. Custom implants, like those from Aprevo, are more expensive upfront than off-the-shelf options and have a longer lead time- it takes 5 days to 2 weeks to obtain custom implants from the planning to production phase. However, when you factor in lower revision rates and great long-term outcomes, they save money overall. Another challenge is adoption—healthcare systems are often slow to embrace these advancements despite the data supporting their efficacy. Educating providers and payers about the long-term benefits is crucial for wider implementation.
Q: What other payer challenges are you facing in terms of implant approval?
TP: Implant approval challenges largely depend on insurance coverage criteria, particularly for custom implants required for specific deformities. A standardized office process using tailored language and data helps justify implant use, which makes it easier to advocate for coverage and ensure patients receive the implants they need.
Q: What other technology or advancements are you keeping an eye on?
TP: Thus far, I’ve been enthusiastic in using predictive analytics in the spinal deformity space, and am looking forward to working to help develop spine oncology predictive analytics algorithms.