Beyond operative time: multidimensional CUSUM analysis of institutional learning with case-complexity adjustment in endoscopic endonasal transsphenoidal surgery for pituitary neuroendocrine tumours: a retrospective single-centre longitudinal cohort study
This retrospective study of 510 endoscopic endonasal transsphenoidal surgeries demonstrates that while operative efficiency improved across three distinct phases after adjusting for case complexity, safety, resection extent, and cerebrospinal fluid leak rates remained stable throughout, suggesting that institutional learning is multidimensional and outcome-specific rather than defined by a single proficiency threshold.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Surgery is often imagined as a linear journey: a surgeon performs a procedure, makes mistakes, learns from them, and eventually reaches a state of perfect mastery. In the world of pituitary surgery, where delicate tumors are removed from the base of the skull through the nose, this journey is traditionally measured by a single number: how long the operation takes. The assumption is that as a team gains experience, the clock ticks down, signaling that they have learned the necessary skills. However, this simple metric ignores a crucial reality: not all tumors are created equal. Some are small and simple, while others are large, invasive, and technically demanding. If a surgical team starts with easy cases and gradually moves to harder ones, a shorter operation time might not mean they are getting faster; it might just mean they are getting better at handling complexity without losing their cool. The question of how to truly measure when a surgical program has matured remains a complex puzzle, especially when the cases themselves are changing as the team gains experience.
A team of neurosurgeons in Mardan, Pakistan, set out to solve this puzzle by looking at 510 consecutive surgeries performed between May 2022 and April 2026. They did not just count the minutes spent in the operating room; they built a detailed picture of the learning curve by tracking four different things at once: how long the surgery took, whether the tumor was completely removed, if there were any serious complications, and if cerebrospinal fluid leaked after the operation. Crucially, they adjusted their analysis to account for the difficulty of each specific tumor. They considered factors like the size of the tumor, how deeply it had invaded nearby structures, whether it was a repeat surgery, and how high it had grown above the brain. By doing this, they could separate the surgeon's growing skill from the natural increase in case difficulty that often happens as a program matures.
The researchers found that the story of learning was not a single straight line but a series of distinct chapters, and these chapters looked different depending on which outcome they were watching. When they looked at operative efficiency—essentially how quickly the team could perform the surgery relative to the difficulty of the case—they saw two clear turning points. The first transition occurred after the 40th surgery, where the team began to operate significantly faster than expected for the types of tumors they were facing. A second, more subtle shift happened around the 342nd surgery. At this later stage, the team was facing increasingly difficult tumors, including those that were more invasive and required more complex surgical approaches. Despite this rising difficulty, the team managed to keep their surgery times even shorter than before, suggesting they had developed a remarkable ability to absorb complexity without slowing down.
However, the story changed completely when the researchers looked at safety and the completeness of the tumor removal. For major complications, such as severe bleeding, infection, or new nerve damage, the data showed no turning points at all. The rate of these serious events remained steady and low throughout the entire 510-surgery sequence, regardless of how many procedures had been performed. Similarly, the success rate of completely removing the tumor did not show a sudden jump or a distinct phase of improvement. Instead, the ability to remove the tumor entirely was consistently tied to the anatomy of the tumor itself; harder tumors were harder to remove, no matter how much experience the team had. The same was true for postoperative leaks of cerebrospinal fluid, which occurred in a small number of cases but showed no pattern of improvement or decline over time.
This distinction is vital because it challenges the common belief that a single number of surgeries defines when a surgeon or a team has "arrived." The study suggests that learning is specific to the task at hand. A team can become highly efficient at managing the clock and the workflow while facing harder cases, yet the safety record and the ability to fully remove difficult tumors may not follow the same timeline of improvement. In fact, the data showed that as the team gained experience, they were actually taking on more difficult cases, yet they maintained their safety standards. This indicates that the program had matured into a state where it could handle greater challenges without compromising patient safety, even if the traditional markers of "mastery" like a sudden drop in complication rates were not visible.
The researchers concluded that judging a surgical program by a single threshold of experience is misleading. Instead, a multidimensional view is required, one that recognizes that efficiency, safety, and resection success are different skills that may develop at different rates. The team in Mardan demonstrated that a surgical program can evolve into a highly capable unit that manages increasing complexity with steady safety, even if the traditional "learning curve" looks different for every outcome. This approach offers a more faithful way to understand how surgical teams grow, moving beyond simple stopwatch metrics to a deeper appreciation of how experience interacts with the inherent difficulty of the work.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.