Patient-specific rod planning is often too optimistic in hypokyphotic adolescent idiopathic scoliosis patients

The human spine is not a straight pole; it is a graceful, S-shaped curve that allows us to stand upright while absorbing the shocks of walking and running. In the upper back, the spine naturally bends outward, creating a gentle arch known as thoracic kyphosis. This curve is essential for balance and for protecting the spinal cord. However, in a condition called adolescent idiopathic scoliosis, the spine twists and bends sideways, often flattening out this natural outward curve. When the upper back becomes too straight, it can lead to long-term problems with posture and pain. Surgeons have long sought a way to restore this natural shape during corrective surgery, but doing so is incredibly difficult because the spine is stiff and the body's response to surgery is hard to predict.

For years, surgeons have relied on patient-specific rods to solve this puzzle. Instead of bending a straight metal rod by hand during the operation—a process that varies from surgeon to surgeon—engineers now use computer models to design custom rods that are pre-shaped to the exact curve the surgeon hopes to achieve. The idea is that if the rod is perfectly shaped before it ever enters the operating room, the spine will naturally follow that shape once the surgeon attaches it. This technology promises a more precise and harmonious result, potentially preventing the spine from becoming too flat or too curved after the operation. But does the computer model tell the whole story, or does it sometimes promise more than the human body can deliver?

A team of researchers at hospitals in Paris set out to test the reliability of these custom rods, specifically looking at a group of patients who presented a unique challenge: those whose spines were already dangerously flat before surgery. The study involved 210 young patients, all under the age of 19, who underwent spinal fusion surgery to correct their scoliosis. The researchers divided these patients into two groups based on the shape of their upper backs before the operation. One group had a "hypokyphotic" spine, meaning their natural outward curve was less than 20 degrees, making their upper back nearly flat. The other group had a "normokyphotic" spine, with a natural curve between 20 and 40 degrees, which is considered normal and healthy.

The surgeons used a sophisticated software system to plan the surgery for every single patient. They uploaded X-rays of the patients' spines and used a predictive model to design a custom rod that would correct the sideways curve while simultaneously restoring the natural outward curve of the upper back. The goal was to bring the flat spines up to a healthy 20 to 40-degree curve. The computer models were optimistic, predicting that the custom rods would successfully reshape the spines of the flat-backed patients, adding roughly 20 degrees of curve to their upper backs. The plan was validated by the treating physicians weeks before the surgery, and the custom rods were manufactured and shipped to the operating room.

When the surgeries were completed and the patients were examined afterward, the results revealed a significant gap between the plan and reality, particularly for the patients with the flattest spines. While the custom rods worked well for the group that started with a normal curve, the results were different for the hypokyphotic group. In this group, the surgery did improve the shape of the spine, but not as much as the computer had promised. On average, the surgeons managed to add only 10 degrees of curve to the upper back, falling short of the 20-degree target that had been planned. In fact, for one in five of these patients, the difference between what was planned and what was achieved was so large that the result was considered insufficient. The computer model had assumed the spine would be flexible enough to bend into the new shape, but in reality, the stiffness of these specific spines resisted the full correction.

For the patients who started with a normal curve, the custom rods performed as expected, with the final shape of the spine staying very close to the surgical plan. However, the study highlighted that the technology is not a magic bullet that works the same way for every body type. The researchers found that the computer simulations consistently overestimated how much the flat spines could be corrected. The custom rods were effective at restoring a harmonious shape to the spine in the majority of cases, and the vast majority of patients ended up with a healthy, natural curve after surgery. Yet, the specific subgroup with the flattest preoperative spines proved to be much harder to correct than the models suggested.

The study concludes that while patient-specific rods are a powerful tool for spinal surgery, the planning process needs to be more cautious when dealing with patients who have very flat upper backs. The current computer models tend to be too optimistic, assuming a level of flexibility that these patients often do not possess. To get the best results, surgeons need to adjust their expectations and their plans to account for the specific stiffness of the patient's spine and their own personal surgical techniques. The technology works, but it requires a human touch to temper the computer's predictions with the reality of the patient's unique anatomy.