AUTHOR=Li Changjian , Zhang Yong , Liao Ying , Han Lu , Zhang Qingyou , Fu Jia , Zhou Dan , Long Shuai , Tian Hong , Jin Hongfang , Du Junbao TITLE=Differential Diagnosis Between Psychogenic Pseudosyncope and Vasovagal Syncope in Children: A Quantitative Scoring Model Based on Clinical Manifestations JOURNAL=Frontiers in Cardiovascular Medicine VOLUME=9 YEAR=2022 URL=https://www.frontiersin.org/journals/cardiovascular-medicine/articles/10.3389/fcvm.2022.839183 DOI=10.3389/fcvm.2022.839183 ISSN=2297-055X ABSTRACT=

The study was designed to explore a clinical manifestation-based quantitative scoring model to assist the differentiation between psychogenic pseudosyncope (PPS) and vasovagal syncope (VVS) in children. In this retrospective case-control study, the training set included 233 pediatric patients aged 5–17 years (183 children with VVS and 50 with PPS) and the validation set consisted of another 138 patients aged 5–15 years (100 children with VVS and 38 with PPS). In the training set study, the demographic characteristics and clinical presentation of patients were compared between PPS and VVS. The independent variables were analyzed by binary logistic regression, and the score for each variable was given according to the approximate values of odds ratio (OR) to develop a scoring model for distinguishing PPS and VVS. The cut-off scores and area under the curve (AUC) for differentiating PPS and VVS cases were calculated using receiver operating characteristic (ROC) curve. Then, the ability of the scoring model to differentiate PPS from VVS was validated by the true clinical diagnosis of PPS and VVS in the validation set. In the training set, there were 7 variables with significant differences between the PPS and VVS groups, including duration of loss of consciousness (DLOC) (p < 0.01), daily frequency of attacks (p < 0.01), BMI (p < 0.01), 24-h average HR (p < 0.01), upright posture (p < 0.01), family history of syncope (p < 0.05) and precursors (p < 0.01). The binary regression analysis showed that upright posture, DLOC, daily frequency of attacks, and BMI were independent variables to distinguish between PPS and VVS. Based on the OR values of each independent variable, a score of 5 as the cut-off point for differentiating PPS from VVS yielded the sensitivity and specificity of 92.0% and 90.7%, respectively, and the AUC value was 0.965 (95% confidence interval: 0.945–0.986, p < 0.01). The sensitivity, specificity, and accuracy of this scoring model in the external validation set to distinguish PPS from VVS were 73.7%, 93.0%, and 87.7%, respectively. Therefore, the clinical manifestation-based scoring model is a simple and efficient measure to distinguish between PPS and VVS.