Machine Learning Prediction of Spinal Anaesthesia Block Duration
The duration of sensory and motor block after spinal anaesthesia varies widely between patients given the same dose of local anaesthetic. Much of this variability is explained by differences in lumbosacral cerebrospinal fluid volume, which cannot be measured routinely in clinical practice but is related to simple body measurements such as abdominal circumference and vertebral column length. This prospective observational study will develop and internally validate prediction models for the duration of sensory and motor block following spinal anaesthesia in adults undergoing elective surgery. Preoperative clinical and anthropometric variables will be recorded, and block regression will be assessed serially after intrathecal injection. Machine learning methods (lasso regression, ridge regression, random forest, extreme gradient boosting, and support vector regression) will be developed and compared against multivariable linear regression as the reference model. The aim is a practical tool that helps anaesthetists anticipate how long a spinal block will last in an individual patient, supporting decisions about case scheduling, supplementation, and discharge planning.
Adults scheduled for elective surgery under single-shot spinal anaesthesia [+2]
Contraindication to neuraxial anaesthesia (coagulopathy, local or systemic infec... [+6]