Meta-analysis of Salt Valproate Prevent Switch Associated with Antidepressants in Chinese Depressive Patients

Objective: to study the ecacy of valproate in preventing switch rate related to antidepressant in depressive patients. Methods: The related literature were searched in Chinese Biomedical Database(CBM), China National Knowledge , Chinese Social from 1 January 2005 to 1 January 2020 The rate of switch between groups was synthesized and discussed.


Background
Valproate may be applied to depression for the following reasons: First, in the study, as an important intensive treatment, it is used in the treatment of depression [1]. Second, valproate can improve the symptoms of refractory depression [2,3]. Third, valproate is bene cial to the improvement of irritability, impulse and anxiety symptoms of irritable depression [4,5]. Fourth, valproate is widely used in post-stroke depression in China [6,7].
In the consensus of experts on the psychiatric application of valproate, the prevention of manic conversion is also one of the important roles of valproate [8]. During the treatment ,manic symptoms may appear, which are related to antidepressant ,patients were diagnosed as bipolar disorder [9,10,11]. In fact, one of the principles is to avoid turning manic as the treatment of bipolar depression [12]. However, since most bipolar disorder often starts with depression, which often treated with antidepressant and switch [13,14].The lithium can prevent the switch and decrease the more 50% switch rate [15].As valproate can improve the irritable,compulsive and mixed symptom in depression,so the avoidance of switch.This study just is a metaanalysis about prevention of switch by valproate. Methods 1. Literature retrieval methods:

2.Comparison of switch rate between experimental group and control group.
A total of 549 subjects were included in 7 studies. The results showed that the switch rate of salt valproate was 0.11%(3/279),switch rate of antidepressant was 11.11%( 30/270), which was very different in switch rate(OR=0.13 ,95% CI: 0.05-0.35) .The switch rate in experimental group was signi cantly lower than in that in control group (Z=4.11,P<0.0001). And it also indicated that salt valproate reduced switch rate was 99%( 11.11%-0.11%/11.11%).See gure2.
The funnel plot analysis of study about switch rate show a gap,which indicate there maybe a bias of publication.see gure3.

Discussion
The salt valproate primarily was used for bipolar disorder. Just according to type of depressive episode,salt valproate more was used in treatment bipolar depression [24,25],In a study, lurasidone adjunctive with lithium or valproate demonstrated signi cant improvement in depressive symptoms based on the MADRS from weeks 2-5 but not at the primary week 6 endpoint [25].Meta-analysis showed a signi cant difference in favour of valproate for reduction in depressive symptoms, both on depression symptom scales (standardized mean difference (SMD) -0.35 (95% con dence interval, -0.69, -0.02)), and participants with at least 50% improvement in symptoms -relative risk (RR) 2.00 (1.13, 3.53) [24].
But valproate also was used for other type depression,such as TRD [2,3],depression with mixed future,with agitated or anxiety symptoms [26,27].It also hint valproate maybe effective in prevention of switch associated with antidepressant.In fact,the risk of switch associated with antidepressant in patients with depression was monotherapy by antidepressant and without use of mood stabilizer [12].
In this study,valproate can decrease the possibility of switch associated with antidepressant. The switch rate in experimental group was signi cantly lower than in that in control group (Z = 2.18 ~ 3.47,P = 0.0033 ~ 0.0005,OR = 0.11-0.18). In total,the results showed that the switch rate of salt valproate was 0.11% (3/279),switch rate of antidepressant was 11.11%( 30/270), which was very different in switch rate(OR = 0.13 ,95% CI: 0.05-0.35,Z = 4.11,P < 0.0001). And it also indicated that salt valproate reduced switch rate was 99%( 11.11%-0.11%/11.11%).So we need cautiously regard this results that salt valproate decrease the switch rate induced by antidepressant during treatment for patients with depression.
The certain number of depressive patients switch to mania or exciting status during treatment by antidepressant,of which could be diagnosed by criteria of bipolar disorder in DSM-5 [28 ].But this is not successful therapeutic plan for patient duo to switch,because it induce the mania ahead [29].So avoiding switch to mania is important part of therapeutic plan,whereas the patients is unipolar or bipolar depression.
The switch-inducing potential of antidepressants is unclear, which can trigger mood episode switches in patients with bipolar disorder or soft bipolar disorder [30].
This study had several limitations. Firstly, the sample size of this meta-analysis was relatively small. Only 7 studies and 549 subjects were involved. Secondly, collecting data style may in uence the result of investigation, for example,different criteria of switch can get different detection rate of switch. so it was very import to establish a diagnostic criteria for switch associated with antidepressant.Thirdly, not all the studies had blind observation. These factors are partly responsible for the source of pool rate of switch associated with antidepressant, also affect us to see the real signi cance and risk of switch.

Conclusion
As mood stabilizer,the both sodium valproate and magnesium valproate can reduced switch rate related to antidepressant in depressive patients.  Comparison of switch rate between experimental group and control group. The xed random model was used. The results showed that the switch rate of salt valproate was 0.11%(3/279), switch rate of antidepressant was 11.11%(30/270), which was very different in switch rate(OR=0.13, 95% CI: 0.05-0.35). The switch rate in experimental group was signi cantly lower than in that in control group (Z=4.11,P<0.0001).

Figure 3
The funnel plot analysis of study about switch rate. switch rate show a gap, which indicate there maybe a bias of publication.

Figure 4
Subgroup comparison of switch rate according to salt valproate. The xed random model was used. The switch rate of sodium valproate group was 0.00%(0/145), switch rate of antidepressant group was 5.7%% (8/140), which was very different in switch rate(OR=0.18, 95% CI=0.04-0.84). The switch rate in experimental group was signi cantly lower than in that in control group (Z=2.18, P=0.0033). he switch rate of magnesium valproate group was 2.2%(3/134), switch rate of antidepressant group was 16.92%( 22/130), which was very different in switch rate(OR=0.11, 95% CI=0.03-0.39). The switch rate in experimental group was signi cantly lower than in that in control group (Z=3.47, P=0.0005)