Psychotic depression (PD) is described in the International Classification of Diseases, 10th edition (ICD-10) and the American Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV), a division with clear therapeutic implications.PD in the classical classification refers specifically to depressive episodes with psychotic symptoms (e.g., delusional depression with hallucinations, delusions, or depressive lucidity with xylophobia), which is a subtype of depression. Probably due to different diagnostic criteria, differences in the definition of psychotic symptoms and ethnic differences, the ratio of depressed patients with psychotic symptoms to the overall depressed patients is reported to be 5%-38% in China and 20%-33% in foreign countries, but because many of these patients do not show cognitive difficulties and delusions, or even deny suicidal tendencies, it makes the diagnosis of PD difficult and often not effectively However, because many of these patients do not show cognitive difficulties and delusions, or even deny suicidal tendencies, PD is difficult to diagnose and often not effectively treated. Even if the correct diagnosis is made, many patients do not receive timely and effective treatment. PD is also a difficult psychiatric disorder, and different PD patients may show different responses to standard treatments, which are summarized by the author as follows. 1. Treatment with antipsychotics or antidepressants alone The 2nd generation antipsychotics such as risperidone and olanzapine are effective in PD. In a 6-week multicenter, double-blind, parallel-group trial of 123 patients, the efficacy of risperidone was compared with the combination of haloperidol and amitriptyline. Although the combination showed a slightly higher efficacy rate, the Risperidone treatment group also showed a 37% and 51% reduction in Brief Psychiatric Rating Scale (BPRS) and Mania Rating Scale (BRMS) scores, respectively. In another report, after the combination of fluoxetine and trifloxystrobin was ineffective in one PD patient, the combination of fluoxetine, trifluoperazine and electroconvulsive therapy was also ineffective, and finally, after 1 week of risperidone alone, there was a significant improvement in psychotic symptoms and mood disorders without adverse effects. It is suggested that risperidone may be effective when other treatments are ineffective. Olanzapine alone has also been successful in PD patients, who showed significant improvement in symptoms after treatment with olanzapine alone. With the exception of amoxapine, tricyclic antidepressants alone are usually ineffective in patients with PD. Amoxapine, a tricyclic drug containing the structure of the antipsychotic ketepin and the antidepressant clopidogrel, has the function of inhibiting 5-HT and DA, so much so that it is considered to be an atypical antipsychotic. Antipsychotics are at risk for extrapyramidal adverse effects in the treatment of patients with PD, and certain clinicians and patients try to avoid these drugs. Although the exact mechanism of selective 5-hydroxytryptamine reuptake inhibitors (SSRIs) for the treatment of PD has not been elucidated, numerous studies have reported the efficacy of SSRIs alone for both depressive and psychotic symptoms. For example, fluvoxamine treatment in PD patients has a similar efficacy to the combination and electroconvulsive therapy. In a double-blind trial of sertraline and paroxetine in patients with delusional depression, the efficacy of sertraline and paroxetine was 75% and 46%, respectively. However, in a Japanese patient with PD treated with fluvoxamine and sertraline, psychotic symptoms improved with fluvoxamine, worsened when switching to sertraline, and improved again when fluvoxamine was reintroduced. The analysis may be due to their different mechanisms of action with σ-1 receptors or to the strong inhibitory effect of sertraline on dopamine transporter proteins. Reports of different treatment outcomes with the same SSRIs are not isolated cases, which require us psychiatrists to be cautious when using SSRIs alone for PD. Combined antipsychotic and antidepressant treatment Combined antipsychotic and antidepressant treatment for PD patients is one of the most effective treatment methods available. Traditionally, the combination of tricyclic antidepressants and typical antipsychotics has been the main treatment option, but many double-blind trials have now shown that the combination of SSRIs and antipsychotics is a more optimal combination. For example, in 30 patients with PD treated with fluoxetine and endorphin, 22 had a reduction rate of more than 50% on the Hamilton depression scale (HAMD) after 5 weeks, which was the same as the efficacy of amoxapine, electroconvulsive therapy and the combination of tricyclics and antipsychotics. The results of Meyers et al. Dube et al. compared the efficacy of the combination with that of single drug and placebo, and the mean HAMD reduction rate of the combination of olanzapine and fluoxetine was significantly greater than that of the olanzapine and placebo control group. Also the significant rate was 56% with the combination, 36% with olanzapine and 30% with placebo. However, it has also been reported that the combination of fluoxetine or sertraline induces or worsens psychotic symptoms in PD patients taking antipsychotics. 3. Hypothalamic-pituitary-adrenal (HPA) axis inhibitor therapy Some studies suggest that psychotic symptoms and cognitive impairment in PD patients may be due to significantly abnormal HPA axis activity and excess cortisol, and that the use of HPA axis inhibitors may be a new treatment. There are two receptors for cortisol in humans, a salt corticosteroid receptor (MR) and a low-affinity receptor for glucocorticoid receptor (GR), and HPA axis activity is mainly maintained by the balance of these two receptors. Mifepristone is a potent GR inhibitor, which on the one hand improves clinical symptoms and cognitive performance by directly inhibiting GR, and on the other hand leads to a large amount of cortisol that can only interact with MR, indirectly downregulating or resetting HPA axis activity. Numerous studies have shown that mifepristone rapidly improves psychiatric symptoms in the treatment of PD patients. belanoff et al. randomized 30 patients to receive 50 mg, 600 mg and 1200 mg of mifepristone daily in 3 groups for 7d. The results showed significant improvement in psychotic symptoms in the two high-dose groups, with more than 40% of patients having more than 50% reduction in HAMD scores and more than 30% reduction in BPRS scores. In another study, a controlled study of 207 patients on mifepristone and placebo, after 1 week of mifepristone treatment and 3 weeks of follow-up, showed significant improvement in psychotic symptoms in the mifepristone group (41% and 27% in the mifepristone and placebo groups, respectively; P < 0.05), but no significant difference was found for depressive symptoms. Recently, Blasey et al. did a similar study in 433 PD patients and found a significant and rapid improvement in psychotic symptoms at mifepristone blood levels above 1660 ng/ml, as well as a good linear relationship between mifepristone blood levels and the responsiveness of psychotic symptoms. These studies show that GR and MR, as important pathological markers of HPA axis dysfunction, provide new avenues for the treatment of PD by acting on GR. Mifepristone has been suggested to be used in PD patients during exacerbations, followed by a switch to antidepressants alone or a combination of antipsychotics and antidepressants. However, such drugs inhibit GR along with progesterone, especially in young women, and the problems associated with long-term treatment include delayed menstruation and potential estrogen breakthrough bleeding, which should be guarded against.