Although ketamine can rapidly decrease suicidal ideation (SI), its neurobiological mechanism of action remains unclear. Several areas of the cingulate cortex have been implicated in SI; therefore, we aimed to explore the neural correlates of the anti-suicidal effect of ketamine with cingulate cortex functional connectivity (FC) in depression.

Forty patients with unipolar or bipolar depression with SI underwent six infusions of ketamine over 2 weeks. Clinical symptoms and resting-state functional magnetic resonance imaging data were obtained at baseline and on day 13. Remitters were defined as those with complete remission of SI on day 13. Four pairs of cingulate cortex subregions were selected: the subgenual anterior cingulate cortex (sgACC), pregenual anterior cingulate cortex (pgACC), anterior mid-cingulate cortex (aMCC), and posterior mid-cingulate cortex (pMCC), and whole-brain FC for each seed region was calculated.

Compared with non-remitters, remitters exhibited increased FC of the right pgACC–left middle occipital gyrus (MOG) and right aMCC–bilateral postcentral gyrus at baseline. A high area under the curve (0.91) indicated good accuracy of the combination of the above between-group differential FCs as a predictor of anti-suicidal effect. Moreover, the change of SI after ketamine infusion was positively correlated with altered right pgACC–left MOG FC in remitters (r = 0.66, p = 0.001).

Our findings suggest that the FC of some cingulate cortex subregions can predict the anti-suicidal effect of ketamine and that the anti-suicidal mechanism of action of ketamine may involve alteration of FC between the right pgACC and left MOG.

Early identification of patients with bipolar disorder during their first depressive episode is beneficial to the outcome of the disorder and treatment, but traditionally this has been a great challenge to clinicians. Recently, brain-derived neurotrophic factor (BDNF) has been suggested to be involved in the pathophysiology of bipolar disorder and major depressive disorder (MDD), but it is not clear whether BDNF levels can be used to predict bipolar disorder among patients in their first major depressive episode.

To explore whether BDNF levels can differentiate between MDD and bipolar disorder in the first depressive episode.

A total of 203 patients with a first major depressive episode as well as 167 healthy controls were recruited. After 3 years of bi-annual follow-up, 164 patients with a major depressive episode completed the study, and of these, 21 were identified as having bipolar disorder and 143 patients were diagnosed as having MDD. BDNF gene expression and plasma levels at baseline were compared among the bipolar disorder, MDD and healthy control groups. Logistic regression and decision tree methods were applied to determine the best model for predicting bipolar disorder at the first depressive episode.

At baseline, patients in the bipolar disorder and MDD groups showed lower BDNF mRNA levels (P<0.001 and P = 0.02 respectively) and plasma levels (P = 0.002 and P = 0.01 respectively) compared with healthy controls. Similarly, BDNF levels in the bipolar disorder group were lower than those in the MDD group. These results showed that the best model for predicting bipolar disorder during a first depressive episode was a combination of BDNF mRNA levels with plasma BDNF levels (receiver operating characteristics (ROC) = 0.80, logistic regression; ROC = 0.84, decision tree).

Our findings suggest that BDNF levels may serve as a potential differential diagnostic biomarker for bipolar disorder in a patient's first depressive episode.