| Objective:High misdiagnosis rate of bipolar depression can result in inappropriate treatment with antidepressants without using mood stabilizers as the foundation,which increases the risk of affective switching to hypomania or mania while not preventing new depressive episodes.Therefore,searching for the biomarker to distinguish BD(bipolar disorder)and UD(unipolar depression):is very important.Unfortunately,to date,no laboratory or imaging marker has been identified that allows for distinguishing BD and UD.This study aimed to evaluate whether the neurocognitive function and neurophysiological characteristics could be a biomarker to distinguish BD and UD by exploring the differences of neurocognitive function and neurophysiological characteristics among UD,bipolar depression,and DWB(depression with bipolarity)groups,furtherly,to confirm that DWB is closer to BD or UD.Methods:Subjects with major depressive episodes,aged between 18 and 45 years,based on DSM-5,were enrolled in this study.HAMD-17 and YMRS were applied to assess the severity of illness(HAMD-17≥17 points and YMRS<6 points).Patients with bipolar disorder I and II entered the BD group,and then patients with UD accepted the second interview.According to the criteria of depression with bipolarity,patients met the criteria entered the D WB group,remaining patients entered the UD group.Each group included 30 patients.30 healthy individuals were enrolled as healthy control(HC)group.THINC-it(THINC-integrated tool),WCST(Wisconsin Card Sorting Test),CPT(continue performance test),and ERP(event-related potentials)were applied to assess neurocognitive function and neurophysiological characteristics.Standardized Z-scores were calculated to compare performance on both objective and subjective cognitive assessments on the THINC-it,using scores of HC group as reference([participantx score on testy-mean of HC on testy]÷standard deviation of HC on testy).All data were entered and analyzed using SPSS 24.0.One-way analysis of variance was used on measurement data among groups.The chi-square test was used on count data,Pearson correlation analysis was used to analyze the link between cognitive function and demographic data or clinical data.Results:(1)There were no significant differences on demographic data(gender,age,years of education and BMI)among four groups,as well as on some clinical data(family history,whether with psychotic symptom and total duration)and clinical scale scores(CGI,HAMD-17 and HAMA)among three patient groups.There were no significant differences in the age of onset between BD and DWB groups,while the ages of onset in BD(P=0.019)and DWB(P=0.008)groups were younger than the UD group.There were no significant differences in attacks between BD and DWB grouos,while the attacks of BD(P<0.001)and DWB(P=0.041)groups were more than the UD group.There were no significant differences in the course of this episode between BD and DWB groups,while the course of this episode in BD(P<0.001)and DWB(P=0.014)groups were shorter than UD group.The difference in scores of MDQ between BD and DWB groups was not significant(P=0,216),while there were significant differences between BD and UD group(P<0.001)as well as DWB and UD group(P<0.001).(2)Thinc-itTHINC-it objective cognition(Z-score)The difference of the Z-score betweenDWB and BD was not significant(P=0.098),Z-scores of BD group(P<0.001)and DWB group(P=0.005)were lower than UD group,while the Z-scores of three patient groups were lower than HC group.THINC-it subjective cognition(Z-score)The difference of the Z-score betweenDWB and BD was not significant(P=0.360),Z-scores of BD group(P=0.001)and DWB group(P=0.013)were lower than UD group,while the Z-scores of three patient groups were lower than HC group.(3)WCSTNumber of trails completedThere was no significant difference in the number of trials completed between BD and DWB,the number of trails and categories completed of BD(P<0.001)and DWB(P<0.001)were higher than UD,the number of trails and categories in three patient groups were higher than HC group.Number of categories completedThere was no significant difference in the number of categories completed between BD and DWB,the number of categories completed of BD(P<0.001)and DWB(P=0.006)were lower than UD,the number of trails and categories in three patient groups were lower than HC group.Total number of correctA total number of correct in BD group were less than DWB(P=0.003)and UD(P<0.001),and which in DWB were less than UD(P=0.043),while the total number of correct in three patient groups were less than HC group.Total number of errorsThe total number of errors in the BD group was more than DWB(P=0.001)and UD(P<0.001),and which in DWB were more than UD(P<0.001),while the total number of errors in three patient groups were more than HC group.Trails to completed first categoryDifferences on trails to completed the first category among three patient groups were not significant,while the trails to completed the first category of three patient groups were more than the HC group.Perseverative responsesPerseverative responses of the BD group were more than DWB(P=0.009)and UD(P<0.001)groups,and that of DWB was more than UD(P=0.011),while perseverative responses of three patient groups were more than HC group.Perseverative errorsThere was no significant difference in perseverative errors between BD and DWB groups(P=0.581),perseverative errors of BD(P<0.001)and DWB(P=0.002)groups were more than UD,and that of three patient groups were higher HC group.Nonperseverative errorsNonperseverative errors of the BD group were more than DWB(P<0.001)and UD(P<0.001)groups,and that of the DWB group was more than the UD group(P=0.011),while perseverative errors of three patient groups were more than HC group.Percent conceptual level responsesPercent conceptual level responses of the BD group was lower than DWB(P<0.001)and UD(P<0.001),and that of DWB was lower than UD(P=0.008),while percent conceptual level responses of three patient groups were lower than HC group.(4)CPT2-digit numbers and 3-digit numbersThere were no significant differences in leakage responses,false responses,and mean reaction time among three patient groups,while leakage responses and false responses of three patient groups were more than the HC group,and mean reaction time of three patient groups were longer than HC group.4-digit numbersThere were no significant differences in leakage responses,false responses and mean reaction time between BD and DWB group,as well as mean reaction time among three patient groups.Leakage responses and false responses of BD and DWB groups were more than the UD group,and that of 3 patient groups was more than the HC group.(5)ERPThere were no significant differences in N2 and N3 latency among three patient groups,as well as on P2 latency among four groups,while N2 and N3 latency of three patient groups were longer than the HC group.(6)Correlation between cognition and demographic data,clinical data or clinical scales scoreThere was no significant correlation between cognition and demographic data or clinical data.Standardized Z-scores of THINC-it(whether objective part or subjective part)and ERP latencies were not correlated with clinical scales score(HAMD-17,HAMA,CGI-S and MDQ).Number of trails completed,Total number of errors,Trails to completed first category,Perseverative responses,Perseverative errors,leakage responses and false responses of 2-digit numbers CPT,leakage responses and mean reaction time of 3-digit numbers CPT,leakage responses,false responses and mean reaction time of 4-digit numbers CPT were positively correlated with CGI-S scores.Number of categories completed was negatively correlated with CGI-S scores.There was no significant correlation between Total number of correct,Nonperseverative errors,Percent conceptual level responses,mean reaction time of 2-digit numbers CPT or false responses of 3-digit numbers CPT and CGI-S scores.Number of trails completed,Total number of errors,Trails to completed first category,Perseverative responses,Perseverative errors,leakage responses and mean reaction time of CPT were positively correlated with HAMD-17 scores.Number of categories completed,Total number of correct and Percent conceptual level responses were negatively correlated with HAMD-17 scores.There was no significant correlation between Nonperseverative errors or false responses of CPT and HAMD-17 scores.Number of trails completed,Total number of errors,Perseverative responses,Perseverative errors,mean reaction time of 2-digit numbers CPT,leakage responses,false responses and mean reaction time of 3-digit numbers CPT,leakage responses and mean reaction time of 4-digit numbers CPT were positively correlated with HAMA scores.Number of categories completed and Percent conceptual level responses were negatively correlated with HAMA scores.There was no significant correlation between Total number of correct,Trails to completed first category,Nonperseverative errors,leakage responses and false responses of 2-digit numbers CPT or false responses of 4-digit numbers CPT and HAMA scores.Number of trails completed,Total number of errors,Nonperseverative errors,leakage responses of 2-digit numbers CPT,mean reaction time of 3-digit numbers CPT,leakage responses of 4-digit numbers CPT were positively correlated with MDQ scores.Percent conceptual level responses were negatively correlated with MDQ scores.There was no significant correlation between Number of categories completed,Total number of correct,Trails to completed first category,Perseverative responses,Perseverative errors,false responses and mean reaction time of 2-digit numbers CPT,leakage responses and false responses of 3-digit numbers CPT or false responses and mean reaction time of 4-digit numbers CPT and MDQ scores.Conclusion:(1)Patients with BD,UD and DWB suffered general cognitive dysfunction.(2)Clinical characteristics of D WB might be closer to BD other than UD.(3)Cognitive dysfunction of BD was severer than UD,and that of DWB was closer to BD other than UD.(4)The severity of the disease was related to cognitive dysfunction.The greater possibility of bipolar disorder which patients with is,the worse cognitive dysfunction will be.(5)The neurocognitive function may be a biomarker to distinguish BD and UD. |
PDF Full Text Request