ABSTRACT
Introduction Paraplegia is a serious complication after thoracic and thoracoabdominal aortic aneurysm surgery. The aortic cross-clamp blocks blood flow to the intercostal artery as a feeding blood vessel, so the spinal cord is at risk of being exposed to ischemia. Hypothermia with systemic cooling is a useful means of avoiding spinal ischemia caused by aortic blockade but has various side effects. Theoretically, local cooling by epidural cooling catheter is an effective method to reduce the side effects. However, the use of needle sensors to measure the temperature of the human spinal cord is not ethically applicable in the real clinical field. The purpose of the study is to build computer modeling of human-sized spinal cords and a basic platform for simulating spinal cord cooling. This is being done to prove that local cooling can cool the human spinal cord in the same way, even in human spinal cords larger than laboratory animals. Methods We tried to model a horizontal cross-section of tissue near the spinal cord at a size equivalent to that of an adult human. The tissue around the spinal cord was decomposed into many small matrices for analysis using the finite element method. Next, the analysis was performed using a high-speed computer on the assumption that the matrix exchanges heat with the adjacent matrix over time according to Pennes' bio-heat equation. Repeated calculations were performed on a high-speed computer to calculate temperature changes in the central part of the spinal cord. Result By setting the temperature of the cooling catheter to 20°C, temperatures at the center of the spinal cord after 5, 10, 15, 20 and 25 minutes were 34.08°C, 33.64°C, 33.48°C, 33.40°C, and 33.36°C, respectively. After stopping the cooling, the temperature at the center of the spinal cord recovered to baseline temperature within 10 minutes. Conclusion Results were similar to those of previous animal studies using our local cooling system, suggesting that evaluation of cooling catheter's performance by computational simulation (CS) is effective.
