A specific biomarker for amyotrophic lateral sclerosis: Quantitative susceptibility mapping. Academic Article uri icon

Overview

abstract

  • OBJECTIVE: Accurate and timely diagnosis of amyotrophic lateral sclerosis (ALS) is a diagnostic challenge given the lack of specific diagnostic and imaging biomarkers as well as the significant clinic overlap with mimic syndromes. We hypothesize that MR quantitative susceptibility mapping (QSM) can help differentiate ALS from mimic diagnoses. METHODS: In a blinded retrospective study of MRIs with QSM from 2015 to 2018, we compared motor cortex susceptibility along the hand and face homunculi in ALS patients and patients with similar clinical presentations. Inclusion required a confirmed ALS or a mimic diagnosis. Comparative groups included age-matched patients with MRIs performed for non-motor neuron symptoms that were reported as normal or demonstrated leukoaraiosis. Quantitative susceptibility values were compared with ANOVA and Tukey-Kramer (post-hoc). ROC analysis and Youden's index were used to identify optimal cutoff values. RESULTS: Fifty ALS, 35 mimic, and 70 non-motor neuron symptom patients (35 normal, 35 leukoaraiosis) were included. Hand and face homunculus mean susceptibility values were significantly higher in the ALS group compared to the mimic (p=0.001, p=0.004), leukoaraiosis (p<0.001, p=0.003), and normal (p<0.001, p<0.001) groups. ROC curve analysis comparing ALS to mimics resulted in an area under the curve of 0.71 and 0.67 for the hand and face homunculus measurements, respectively. In differentiating ALS from mimics, Youden's index showed 100% specificity and 36% sensitivity for hand homunculus measurements. CONCLUSIONS: QSM has diagnostic potential in the assessment of suspected ALS patients, demonstrating very high specificity in differentiating ALS from mimic diagnoses.

publication date

  • January 4, 2021

Research

keywords

  • Amyotrophic Lateral Sclerosis
  • Motor Cortex

Identity

Scopus Document Identifier

  • 85100390325

Digital Object Identifier (DOI)

  • 10.1016/j.clinimag.2020.12.018

PubMed ID

  • 33548870

Additional Document Info

volume

  • 75