Computational analysis of tongue reconstruction surgery: The impact of donor site and flap volume on post-operative anatomy and biomechanics.

Abstract

Tongue reconstruction requires a series of decisions tailored to patient needs to restore anatomy and preserve speech and swallowing. The impact of these interdependent choices is difficult to evaluate, as clinical outcomes depend on case-specific factors. However, computational analysis offers a method for analyzing these interdependencies in a controlled fashion. The present study systematically quantifies the impact of key surgical decisions, namely donor site selection (radial forearm vs. anterolateral thigh) and flap volume on the final anatomical and biomechanical outcomes. To achieve this, we developed an automated framework that simulates free flap tongue reconstruction. The framework leverages biomechanically optimized flap design to generate a multi-component virtual flap, which is then computationally sutured to the resection site, and its long-term tissue atrophy is simulated to predict the final neotongue state. Using this platform, we conducted a 120-simulation factorial study for the systematic analysis. Four clinically plausible tongue reconstruction scenarios, six levels of flap stiffness, and five levels of flap overbulking (intentional excess volume) were simulated. The results reveal a fundamental, physics-based trade-off: while increasing flap overbulking was the dominant factor in restoring pre-operative anatomy, it came at the cost of a predictable increase in biomechanical strain imposed on the native tongue. Furthermore, stiffer flaps induced significantly higher baseline strain. The anatomical benefit of overbulking was significantly modulated by tissue properties. These findings provide a biomechanical rationale for clinically observed functional trade-offs. This work presents an open-source, physics-based, and robust computational testbed for systematically evaluating interdependent surgical variables. Ultimately, the framework's automation and scalability offer a pathway toward personalized, simulation-informed surgical planning for tongue reconstruction.