| Objective To evaluate the predictive factors of antidepressant effect of high-frequency repetitive transcranial magnetic stimulation(HF-rTMS)over left dorsolateral prefrontal cortex(DLPFC)and establish a predictive model by comparing the clinical data and serum thyroid hormone levels of patients with different antidepressant effects before and after treatment,aiming to provide an objective basis for early clinical screening of patients with depression more suitable for HF-rTMS treatment and thereby improve the overall response rate and reduce the waste of medical resources.Methods From January 2020 to December 2020,a total of 95 inpatients were recruited in Hangzhou Seventh People’s Hospital,who met the DSM-V diagnostic criteria for major depressive disorder,with a HAMD(24-items)Score≥21,and ages 18-70 yeas old.All patients underwent 20 sessions(once daily,five days per week)of 10-Hz rTMS(100%MT)over the left DLPFC.The severity of depressive symptoms was assessed within 24 h before and after the rTMS treatment using the depression assessment instruments,including HAMD,HAMA,BDI;in the meanwhile,blood samples were collected to measure the serum levels of thyroid hormones.According to their HAMD score reduction rates at the end of the rTMS treatment,the patients were divided into a response group(score reduction rate ≥50%)and a no-response group(score reduction rate<50%).The clinical data and serum thyroid hormone levels of the two groups before and after treatment were analyzed by univariate analysis.With the response rate being the dependent variable,a binary logistic regression model was built to interpret the statistically significant results of the univariate analysis and identify independent predictors of response to rTMS in patients with depression.Finally,the prediction model is established,and the receiver operating characteristic(ROC)curve is used to evaluate the prediction ability of different indicators.Results 1.Out of the 95 participants,84 had completed the the entire study,among whom 37 were assigned to the response group and the rest 47 to the no-response group,suggesting a response rate of 44.0%.After the rTMS treatment,the HAMD,HAMA,and BDI scores in both groups were remarkably reduced when compared with the pre-treatment levels,and the differences were statistically significant(p<0.05,respectively).2.By comparing the results of the univariate analysis of clinical data,it was found that there were significant differences between the two groups in duration of current episode and scores for psychomotor retardation and sleep disorders(p<0.05,respectively).The binary logistic regression model was applied to these predictors,and the results suggested significant influences of the duration of current episode(OR=0.976,p=0.020)and the score for sleep disorders(OR=1.856,p=0.014)on the response rate.3.The univariate analysis of serum thyroid hormone levels between the two groups before and after treatment,it was found that the serum TSH level of the response group pre-treatment was lower than that of the no-response group,and the difference was statistically significant(p<0.05).Intra-group comparison found that the serum TSH level of the response group post-treatment was higher than that pre-treatment,the difference was statistically significant(p<0.05).According to the binary logistic regression analysis,it was revealed that the pre-treatment serum level of TSH was significant in predicting the response to rTMS(OR=0.408,p=0.013).4.A logistic regression equation was developed using the three predictors,including duration of current episode,sleep disorders score,and pre-treatment serum TSH,and then transformed to obtain the calculation formula of the joint predictor:L=X1-0.045X2-1.257X3.The area under curve(AUC)of the joint predictor(0.800,95%CI 0.704-0.897,p<0.001)was larger than the other three single indicators,and had a certain predictive value.The best cut-off value of the joint predictor was 0.421 when the Youden’s index was the largest,sensitivity was 78.4%,and specificity was 74.5%.Conclusions 1.Duration of current episode,sleep disorders score,and pre-treatment serum TSH are three potential predictors of response to left DLPFC HF-rTMS in patients with depression.2.In those showing good response to left DLPFC HF-rTMS,the serum TSH appears to increase dramatically compared with the pre-treatment level,suggesting possible correlations between the hypothalamic-pituitary-thyroid(HPT)axis and the anti-depression effects of rTMS.3.AUC of the joint predictor is the largest,and its predictive efficiency is higher than that of a single indicator.It has a certain value for early screening of patients with depressive disorder who are more suitable for left DLPFC HF-rTMS treatment. |
PDF Full Text Request