Brain 18FDG-PET pattern in patients with alcohol-related cognitive impairment

Abstract

Purpose Brain positron emission tomography using 18F-fluorodeoxyglucose (18FDG-PET) provides a metabolic assessment of brain function that is useful for differential diagnosis among several neurodegenerative diseases manifested by cognitive impairment (CI) [1] , [2] , [3] . The purpose of the study is to describe the pattern of 18FDG-PET abnormalities in patients with CI related to alcohol use disorder. Methods Patients admitted to the addiction medicine department of a university hospital in Paris between January 2017 and October 2018 with a confirmed diagnosis of alcohol-related cognitive impairment (ARCI) or Wernicke encephalopathy (WE) were included. Brain 18FDG-PET uptake was measured after at least one month of monitored abstinence from alcohol. Standardized uptake values (SUV) were obtained for 13 regions of interest (ROI) and normalized to the pons. Individual patients’ ROI Z-scores were calculated from healthy sex- and age-matched controls provided by Cortex ID software. A principal component analysis (PCA) on the subjects’ ROIs SUV ratios was conducted. Results Twenty-five patients were included in the analysis (20 males and 5 females; mean age 57.6 years [45 – 76 years old]). The group consisted of 19 ARCI and 6 WE cases. The mean hypometabolism was most severe in the prefrontal medial cortex (PFM) (−2.80 [± 1.30]), the prefrontal lateral cortex (−2.20 [± 1.35]), and the anterior cingulate cortex (ACC) (−2.24 [± 1.19]). Hypometabolism (Z-score Table 1 ). Other regions were also affected (with 5.32/13 hypometabolic ROIs on average [SD = 4.16, range 0 – 13]). The Z-scores in the 13 ROIs did not differ significantly between the ARCI and WE patients (P ≥ 0.05). The PCA identified two principal components. The first principal component explains 59.49% of the variance and it is formed by the clustering of six specific ROIs: sensorimotor cortex, parietal superior cortex, parietal inferior cortex, precuneus, occipital lateral cortex and primary visual cortex. The second explains 14.52% of the variance and it is formed by three ROIs: the PFM, ACC and temporal mesial cortex. Conclusions Predominant prefrontal and cingulate cortex hypometabolism was the most frequent brain 18FDG-PET pattern in our sample of patients with ARCI and WE. We believe that brain 18FDG-PET could be a potential imaging biomarker of ARCI, and predictive values for both differential diagnosis and prognosis should be prospectively tested.