Individuation of predictive parameters of posterior cranial fossa decompression in pediatric patients with Chiari I malformation and autism spectrum disorder.
Authors
Affiliations (10)
Affiliations (10)
- Department of Neurosurgery, Meyer Children's Hospital IRCCS, University of Florence, Florence, Italy. [email protected].
- Department of Neurosurgery, Meyer Children's Hospital IRCCS, University of Florence, Florence, Italy.
- College of Medicine, University of Arkansas for Medical Sciences, Little Rock, AR, USA.
- Department of Neuroscience, IRCCS Stella Maris Foundation, University of Pisa, Pisa, Italy.
- Pediatric Neuroradiology Unit, Meyer Children's Hospital IRCCS, University of Florence, Florence, Italy.
- Innovation Center, Meyer Children's Hospital IRCCS, University of Florence, Florence, Italy.
- Clinical Trial Office, Meyer Children's Hospital IRCCS, University of Florence, Florence, Italy.
- Pediatric Radiology Unit, Meyer Children's Hospital IRCCS, University of Florence, Florence, Italy.
- Mathematical Biology and Physiology, Department of Electronics and Telecommunications, Politecnico Di Torino, Turin, Italy.
- Department of Neuroscience, Psychology, Drug Research and Child Health, University of Florence, Florence, Italy.
Abstract
There are no specific guidelines for posterior cranial fossa decompression (PCFD) in asymptomatic Chiari Malformation Type I (CM-I) patients with autism spectrum disorder (ASD). However, some studies suggest that surgery for symptomatic CM-I may improve ASD symptoms. This study aims to identify skull and brain morphometric parameters that could predict surgical outcomes in symptomatic CM-I with ASD, using artificial intelligence (AI). This study included pediatric patients diagnosed with both symptomatic CM-I and ASD who underwent posterior cranial fossa (PCF) surgery. Eleven morphometric parameters were measured using computed tomography (CT) and magnetic resonance imaging (MRI) scans, including cerebellar tonsil descent, tentorium length and angle, cerebellum-to-PCF area ratio, PCF-to-cerebrum area ratio, PCF height and diameter, and various distances involving the corpus callosum, pons, fastigium, foramen magnum, and clivus length. ASD symptom changes were assessed through phone interviews and outpatient evaluations. A binary tree classifier AI model was used to identify patients who improved post-surgically. Our analysis showed that patients with a larger tentorium angle experienced some improvements in ASD symptoms after surgery, whereas those with a significantly smaller tentorium angle showed no improvement. AI identified a tentorium angle of 89.55° as a potential cut-off for distinguishing between outcome groups. No other morphometric parameters significantly influenced ASD symptom outcomes. This study evaluates the relevance of the tentorium angle width as a potentially valuable MRI-based morphometric parameter that could guide neurosurgeons in the decision-making process for this unique patient population. These findings may contribute to a more tailored approach for managing patients with CM-I and coexistent ASD.