Static and dynamic alterations in the amplitude of low-frequency fluctuation in patients with amyotrophic lateral sclerosis

Abstract

Background Static changes in local brain activity in patients suffering from amyotrophic lateral sclerosis (ALS) have been studied. However, the dynamic characteristics of local brain activity are poorly understood. Whether dynamic alterations could differentiate patients with ALS from healthy controls (HCs) remains unclear. Methods A total of 54 patients with ALS (mean age = 48.71 years, male/female = 36/18) and 54 (mean age = 48.30 years, male/female = 36/18) HCs underwent magnetic resonance imaging scans. To depict static alterations in cortical activity, amplitude of low-frequency fluctuations (ALFF) which measures the total power of regional activity was computed. Dynamic ALFF (d-ALFF) from all subjects was calculated using a sliding-window approach. Statistical differences in ALFF and d-ALFF between both groups were used as features to explore whether they could differentiate ALS from HC through support vector machine method. Results In contrast with HCs, patients with ALS displayed increased ALFF in the right inferior temporal gyrus and bilateral frontal gyrus and decreased ALFF in the left middle occipital gyrus and left precentral gyrus. Furthermore, patients with ALS demonstrated lower d-ALFF in widespread regions, including the right lingual gyrus, left superior temporal gyrus, bilateral precentral gyrus, and left paracentral lobule by comparison with HCs. In addition, the ALFF in the left superior orbitofrontal gyrus had a tendency of correlation with ALSFRS-R score and disease progression rate. The classification performance in distinguishing ALS was higher with both features of ALFF and d-ALFF than that with a single approach. Conclusions Decreased dynamic brain activity in the precentral gyrus, paracentral gyrus, lingual gyrus, and temporal regions was found in the ALS group. The combined ALFF and d-ALFF could distinguish ALS from HCs with a higher accuracy than ALFF and d-ALFF alone. These findings may provide important evidence for understanding the neuropathology underlying ALS.

Different sensorimotor mechanism in fast and slow progression amyotrophic lateral sclerosis

“…In the current study, the decreased ALFF in the PSMA was highly consistent with the results from previous observations, including decreased glucose metabolism in 18 F‐FDG PET studies (Claassen, Josephs, & Peller, 2010; Pagani et al, 2014), decreased activity of α , β , and θ frequency bands of EEG study (Dukic et al, 2019), as well as decreased ALFF in RS‐fMRI studies (Luo et al, 2012; Ma et al, 2020; Sako et al, 2017). The consistent cross‐modality finding regarding decreased PSMA activity in ALS patients demonstrated that sensorimotor activity reduction is a robust feature.…”

“…The huge heterogeneity in disease progression is probably one of the critical factor that brings uncertainty to in vivo ALS studies, especially for resting‐state fMRI (RS‐fMRI) that detects the neuropathological spontaneous brain activity (Greicius, 2008). While some studies have found consistent reduction of RS‐fMRI local activity (Luo et al, 2012; Ma et al, 2020; Sako et al, 2017) which were highly in line with hypo‐glucose metabolism (Pagani et al, 2014) as well as decreased neural oscillation (Dukic et al, 2019) in the primary sensorimotor area (PSMA), many other studies have reported inconsistent or even contrary results of resting‐state functional connectivity (FC) in the PSMA of ALS patients. For example, a number of studies have reported decreased FC (Fekete, Zach, Mujica‐Parodi, & Turner, 2013; Jelsone‐Swain et al, 2010; Mohammadi et al, 2009; Tedeschi et al, 2012) in ALS patients, yet almost equal number of studies showed increased FC (Basaia et al, 2020; Douaud, Filippini, Knight, Talbot, & Turner, 2011; Menke, Proudfoot, Talbot, & Turner, 2018).…”

Flexibility of brain dynamics predicts clinical impairment in Amyotrophic Lateral Sclerosis

“…Hence, on the one hand, comparing our results with previous results, that were based on power-spectra or static connectivity, is not trivial. On the other hand, preliminary evidence from fMRI shows altered low-frequency brain dynamics in ALS 59 , which corroborates our results (given the comparable time-scales). However, given the different technique and type of analysis, comparing the results should be done cautiously.…”

The Predictive Value of Dynamic Intrinsic Local Metrics in Transient Ischemic Attack

“…In a previous study, excellent performance with an accuracy of 95.37% was achieved when ALFF maps were employed, followed by ReHo, fALFF, and DC. Ma X. et al (2020) chose 54 amyotrophic lateral sclerosis participants and used ALFF and d-ALFF as the SVM classification feature, and the classification accuracy was 79.63%. From what have been listed above, a reliable conclusion can be drawn that SVM has relatively high accuracy compared with traditional methods to differentiate the patients from healthy subjects.…”