| Objective To explore the differences of Event-related potential(ERP)components P1,vertex positive potential(VPP),and N170 in emotional face processing of two subtypes of Major depressive disorder(MDD).The correlation between disease severity and the P1,N170,VPP were analyzed.To provide a basis for individualized clinical research and treatment.Methods According to inclusion and exclusion criteria,61 cases of Melancholic depression(MD),and 48 cases of Anxious-somatizing depression(ASD),and 22 healthy controls matched with age,sex,and education were included.The general demographic data of all subjects,and the 17 items of Hamilton Depression rating scale were collected,and 30 items of Inventory of Depressive Symptomatology(IDS-30)in the case group were collected as the subtype scale.The latency and amplitude of ERP components P1,VPP,and N170 of all subjects with neutral facial stimulation were collected.All data were analyzed by SAS 9.4 statistical software.Chi-square test was used to compare the qualitative variables among the three groups.The variance analysis was used to compare the fixed variables among the three groups,and the Bonferroni correction method was used to compare the two groups,and make box charts,then the generalized linear model was used for regression analysis.Correlation analysis uses Person correlation analysis and make scatter plot,and then uses multivariate linear model for regression analysis.P < 0.05 was considered statistically significant.Results1.Comparison of general demography and clinical scale among MD group,ASD group and control group There were significant differences in the total score of HAMD and the scores of5 factors among the three groups(P < 0.05),and there were significant differences in anxiety/somatization and delay between MD group and ASD group(P < 0.05).Others are not statistically significant.2.Study on latency and amplitude of P1,N170 and VPP in MD group,ASD group and control group The difference of the amplitude of P1 at the P7 electrode was statistically significant(P < 0.05).Compared with the control group,the amplitude of P1 in the ASD group was increased.The latency of N170 at the P7,PO7 and PO8 electrodes of the three groups was statistically significant(P < 0.05).Compared with the control group,the latency of N170 at the P7,PO7,and PO8 electrode sites was prolonged in the MD group,and the latency of N170 at the PO7 and PO8 electrode sites was prolonged in the ASD group.The latency of VPP at the CZ electrode of the three groups was statistically significant(P < 0.05).Compared with the control group,the latency of VPP at CZ electrode was prolonged in the MD group.After excluding the influence of age,gender and educational background,the latency of N170 at the P7,PO7 and PO8 electrodes in the MD group and the ASD group were statistically significant compared with the control group(P < 0.05).Others are not statistically significant.3.Study on the correlation between HAMD-17 total score,factor score with index in case group In the case group,the HAMD-17 total score was negatively correlated with the latency of P1 at the electrode point of P7(r=-0.2,P=0.04),and the block factor score was negatively correlated with the latency of P1 at the electrode points of P7(r=-0.22,P=0.02).There was no significant correlation between the HAMD-17 total score,factor score with the latency and amplitude of N170.In the case group,the block factor score was positively correlated with the latency of VPP at the electrode point of FCZ(r=0.21,P=0.03)and the electrode point of CZ(r=0.23,P=0.02).The HAMD-17 total score was positively correlated with the amplitude of VPP at the electrode point of CZ(r=0.2,P=0.04).The sleep disturbance factor was positively correlated with the amplitude of VPP at FCZ(r=0.23,P=0.02)and CZ(r=0.22,P=0.02)electrodes.Others are not statistically significant.After excluding the influence of age,gender and educational background on the results,for every 1 point increase in the HAMD total score,the latency of P1 at the P7 electrode was predicted to be extended by 0.0005 ms,and the amplitude of VPP at the CZ electrode was increased by 0.359μv.For each 1 point increase of the blocking factor,the latency of P1 at the P7 electrode was predicted to decrease by 0.002 ms,that the latency of P1 at the PO7 electrode decrease by 0.002 ms,that the latency of VPP at the FCZ electrode extended by 0.002 ms,and that the latency of VPP at the CZ electrode extended by 0.003 ms.For each 1 point increase in sleep factor,the amplitude of VPP at the FCZ electrode was predicted to increase by 1.043 μV,and the amplitude of VPP at the CZ electrode increase by 0.939 μV.4.Study on the correlation between HAMD-17 total score,factor score with index in MD group The N170 latencies of P8(r=-0.31,P=0.02)and P08(r=-0.29,P=0.03)electrodes were negatively correlated with the scores of sleep disorders in MD group.The VPP amplitude of CZ(r=0.26,P=0.04)electrode was positively correlated with the factor score of sleep disorder.After excluding the influence of age,gender and educational background on the results,for every 1 point increase in sleep factor,the latency of N170 at the P8 and PO8 electrodes were predicted to be decreased by 0.001-0.002 ms,and the amplitude of VPP at the CZ electrode increase by 0.939 μV.Others are not statistically significant.5.Study on the correlation between HAMD-17 total score,factor score with index in ASD group The P1 latencies P7(r=-0.37,P=0.01)and P07(r=-0.29,P=0.047)electrodes were negatively correlated with the scores of block factor in ASD group,and the P8(r=-0.29,P=0.047)electrode points of the P1 latencies was negatively correlated with the anxiety/somatization factor.Excluding the influence of age,sex and educational background on the results,for every 1 point increase in the blocking factor score,the latency of P1 at the P7 electrode was predicted to be decreased by 0.003 ms,every 1 point increase in the anxiety/somatization factor score,the latency of P1 at the P8 electrode was predicted to be decreased by 0.003 ms.There is no correlation between PO7 electrodes in multivariate analysis.Others are not statistically significant.Conclusion1.The brain has a lateralization effect on the electrophysiological characteristics reflected by facial information,and the overall effect is more prominent in the left hemisphere.2.The facial processing ERP components of P1,N170,and VPP can be used to identify the difference between depressive disorder and normal controls,but there is no evidence that they could be an effective biomarker to distinguish between the two depression subtypes.3.The abnormal of the P1 and the VPP components may be a susceptible electrophysiological index to predict the severity of MDD.The correlation between P8 and PO8 electrodes of N170 latency,CZ electrodes of VPP amplitudes and the severity of sleep disorders can be used as biomarkers to predict and distinguish patients with MD subtypes in the study of subtype homogeneity. |
PDF Full Text Request