Background: This study aims to identify a metabolomic signature that facilitates the classification of syncope and the categorization of the unexplained syncope (US) to aid in its management. Methods: We compared a control group (CTRL, n = 10) with a transient loss of consciousness (TLC) group divided into the OH group (n = 23) for orthostatic syncope, the NMS group (n = 26) for neuromediated syncope, the CS group (n = 9) for cardiological syncope, and the US group (n = 27) for US defined as syncope without a precise categorization after first- and second-level diagnostic approaches. Results: The CTRL and the TLC groups significantly differed in metabolic profile. A new logistic regression model has been developed to predict how the US will be clustered. Using differences in lysophosphatidylcholine with 22 carbon atom (C22:0-LPC) levels, 96% of the US belongs to the NMS and 4% to the CS subgroup. Differences in glutamine and lysine (GLN/LYS) levels clustered 95% of the US in the NMS and 5% in the CS subgroup. Conclusions: We hypothesize a possible role of C22:0 LPC and GLN/LYS in re-classifying US and differentiating it from cardiological syncope.
