This article is mentioned in:

Introduction

Depression is a common mood disorder, which is characterized by mood swings, chaos at work, difficulties in learning, eating disorders, lack of interest in daily activities and entertainment, insomnia or excessive sleep, restlessness, excitement, fatigue, feeling of worthlessness, difficulties in thinking or concentrating and suicidal thoughts or behaviors. It has been reported that depression can increase the incidence and mortality from somatic diseases (1-3). In 2010, there were ~298 million patients suffering from depression worldwide, while the annual incidence of depression is estimated to be ~50%. Therefore, the rising incidence of depression has become a serious challenge in medical research (4,5).

Currently, first-line antidepressant drugs mainly include serotonin re-uptake inhibitors and selective 5-HT-norepinephrine re-uptake inhibitors (6). However, the underlying mechanism of action of vortioxetine differs from other drugs, since it generally acts in a mixed manner via modulating receptor activity and inhibiting the re-uptake of neurotransmitters. Vortioxetine exerts its pharmacological activity in vivo via 5-hydroxytryptamine type 3 (5-HT3) receptor, 5-HT7 receptor and 5-hydroxytryptamine (serotonin) receptor 1D (5-HT1D) antagonism, 5-HT1B receptor partial agonism, 5-HT1A receptor agonism and 5-HT transporter inhibition (7,8). Interestingly, a previous study demonstrated that vortioxetine had no effect on norepinephrine and dopaminergic neurons virtually (9). Additionally, another study showed that vortioxetine could effectively treat patients with acute-phase depression, while it could also effectively prevent recurrence during the maintenance phase (10).

On September 30th, 2013, vortioxetine was approved by the US Food and Drug Administration (FDA) for the treatment of adults with depression. The drug is provided in doses of 5, 10, 15 and 20 mg (11). Previous randomized controlled trials (RCTs) and systematic reviews in different databases revealed that vortioxetine displayed improved efficacy and safety compared with the positive drugs paroxetine and venlafaxine, or placebo (12-14). However, the results of the aforementioned studies were considered insufficient, since the different doses of vortioxetine and outcome measures were limited. The two other meta-analysis about vortioxetine also provided few outcomes and safety evaluation. Sufficient outcome evaluation could provide more suggestions to physicians. With the advancement of clinical trials and the increased demand for patient medication, it is necessary to re-evaluate the efficacy and safety of vortioxetine as an increasingly popular first-line drug for the treatment of depression in adults. The re-evaluation of its clinical efficacy and safety has also gained increasing attention from psychiatrists and clinical pharmacists in several countries. Therefore, in terms of systematic reviews, collecting more detailed data from clinical trials on vortioxetine via evaluating more factors associated with the efficacy and safety of different doses of vortioxetine could provide the necessary evidence for decision-making on medical treatments and clinical applications.

Materials and methods

The present study was performed and reported according to the Cochrane Handbook for Systematic Reviews of Interventions, which is used for conducting systematic reviews and meta-analyses for observational studies (15,16), as well as the Preferred Reporting Items for Systematic reviews and Meta-Analyses statement (17). To account for any data that could be missing from the final analysis, the results were assessed using the Last Observation Carried Forward method.

Study eligibility criteria

The clinical trials that met the following criteria were included in the present study meta-analysis: i) Double-blind, parallel-controlled and randomized clinical trials; ii) Adult patients suffering from major depressive disorder (MDD), dysthymic disorder and other psychotic disorders according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) (18), DSM-IV Text Revision (19) and/or ICD-10 Classification of Mental and Behavioral Disorders (20-22); iii) Patients treated with 5, 10, 15 or 20 mg/once per day (QD) vortioxetine and placebo; iv) The efficacy outcome was determined based on the changes in the total scores of the 24-Items Hamilton Rating Scale for Depression (HADRS-24) (23), Sheehan Disability Scale (SDS) (24), Montgomery-Asberg Depression Rating Scale (MADRS) (25) and Clinical Global Impression Scale-Improvement (CGI-I) (26), and changes in HADRS-24 response rate, and MADRS response (≥50%) and remission (≤10%) rates, from baseline; and v) The safety outcome was defined as the rate of discontinuation due to adverse effects (>5%).

The exclusion criteria were as follows: i) Systematic reviews; ii) Review articles; iii) Case-control studies; iv) Animal studies; v) Comments; vi) Studies with incomplete data; vii) Case reports; viii) Studies where inappropriate statistical methods were used; ix) Duplicate publications; and x) studies that the diagnostic criteria were not reported.

Data sources and searching strategy

Literature search was performed on PubMed\Medline, EBSCO, Embase, Cochrane library, OVID and Web of Science from database inception to November 2022. There were no limits in terms of language, race, sex and nationality. Potentially relevant unpublished data were searched on ClinicalTrials.gov, the FDA web site (Drugs@FDA; https://www.accessdata.fda.gov/), Chinese Clinical Trial Registry (http://www.chictr.org.cn/), European Union Drug Regulating Authorities Clinical Trials (https://eudract.ema.europa.eu/index.html), World Health Organization and International Clinical Trials Registry Platform (http://www.who.int/ictrp/en/). All studies were hand-searched for randomized clinical trials that met the inclusion criteria. The search terms were as follows: ‘vortioxetine’, ‘brintellix’, ‘Lu AA21004’, ‘placebo’, ‘MDD’, ‘dysthymic disorder’, ‘adult patients’, ‘efficacy’, ‘safety ‘, ‘tolerability’, ‘clinical trial’, ‘randomized controlled trial’, ‘RCT’, ‘double-blind’ and ‘parallel-controlled’. The PubMed search string used was as follows: ‘(vortioxetine or brintellix or Lu AA21004 or trintellix) and (placebo) and (MDD or dysthymic disorder) and (adult patients) and (efficacy) and (safety or tolerability) and (clinical trial or randomized controlled trial or RCT) and (double-blind) and (parallel-controlled) and (human or humans)’. In addition, the Embase search string used was the following: ‘(vortioxetine*.ti or brintellix*.ti or Lu AA21004*.ti or Trintellix*.ti) and (MDD*.ti or dysthymic disorder*.ti) and (adult patients*.ti) and (efficacy*.ti) and (safety*.ti or tolerability*.ti) and (clinical trial*.ti or randomized controlled trial*.ti or RCT*.ti) and (double-blind*.ti) and (parallel-controlled*.ti) and (human*.ti or humans*.ti)’.

Study selection

Each search was performed separately, and each study was downloaded as a separate file using Endnote X6. To minimize selection bias, two researchers (SG and XX) independently screened the titles, abstracts and full texts of each article and data were extracted based on the pre-defined eligibility criteria. The above two researchers evaluated the quality of the literature. In case of disagreement, a third researcher (LF) was involved to reach consensus.

Data extraction

Two researchers (SG and XX) extracted the study characteristics, baseline characteristics of patients, including age, body mass index (BMI), race, HADRS-24 total score, MADRS total score, CGI-I total score and race, interventions and outcome measures, including efficacy [HADRS-24 total score change, SDS total score change, MADRS total score change, CGI-I total score change, HDRS-24 response rate, MADRS response rate (≥50%) and MADRS remission rate (≤10%)], and safety (adverse effects) outcomes.

Quality assessment

The quality of literature was evaluated using the Cochrane Handbook for Systematic Reviews of Interventions (version 5.1.0) for assessing risk of bias in RCTs (27). The evaluation components included random sequence generation, allocation concealment, blinding, analysis of incomplete outcome data and intention-to-treat analysis, while there was no selective reporting or other bias. Each item was defined as ‘yes’ (low risk of bias), ‘no’ (high risk of bias) or ‘unclear’.

Statistical analysis

All outcomes were evaluated using Revman 5.3 software (http://www.cochrane.org/). Risk ratios (RR) with 95% confidence intervals (CIs) were calculated for dichotomous outcomes, such as response and remission rates. Continuous outcomes, such as scale score, are expressed as the mean difference (MD). The I2 statistic was calculated to estimate heterogeneity using Review Manager. I2≤50% was considered to indicate that the studies were homogeneous and a fixed effect model with the Mantel-Haenszel (M-H) method was performed. Otherwise, the random-effect model (REM) was adopted (28). Publication bias was assessed by visually inspecting funnel plots (29).

Results

Literature search and study characteristics

After duplicates were removed, a total of 262 studies were screened. Among them, 248 were excluded, since they did not meet the inclusion criteria based on in vitro studies, animal studies, trials with healthy volunteers and review articles. The remaining 14 studies were assessed according to the pre-determined inclusion criteria. Finally, 11 RCTs were included in the meta-analysis. The studies evaluated the efficacy and safety of vortioxetine via randomizing adult patients into different study arms with different doses of vortioxetine (Fig. 1) (30-39) with the exception of one study (40). As shown in Table I, in nine studies patients were treated with vortioxetine for eight weeks (30,31,33-39), while in the remaining two studies for six weeks (32,40). The main characteristics of patients included in the 11 studies (30-40), such as age, race, BMI, CGI-I and MADRS basic scores were well described. The demographic or clinical characteristics between trials were equivalent to the baseline. In the selected studies, 4,098 adult patients with MDD were treated with vortioxetine and 1,852 with placebo. The bias risk assessment demonstrated that there was a low risk of bias in randomization and blinding. However, the presence of other biases, such as recruitment bias and clinical settings-related bias could not be ruled out (Fig. 2).

Figure 1 PRISMA flowchart of study selection.

Figure 2 Risk of bias summary: Judgements of review authors about each risk of bias item for each included study.

Table I Basic characteristics of literatures (Baseline/mean ± SD).

Table I

Basic characteristics of literatures (Baseline/mean ± SD).

Study	Interventions	Patients (n)	Age, years	Body mass, index kg/m2	CGI-S score	HDRS-24 total score	MADRS total score	Treatment duration (weeks)	Race	Outcome measure
Henigsberg et al (30) 2012	Vorti 5-mg	140	47.3±12.0	26.4±5.1	Not	32.1±5.04	30.6±2.83	8	White, Black,	①②③④
	Vorti 10-mg	140	46.4±12.3	26.2±4.6	reported	33.1±4.77	31.6±3.83		Asian, Other	⑤⑥⑦⑧
	Placebo	140	46.4±12.3	26.4±4.6	32.7±4.40	30.6±2.89
Nishimura et al (31) 2018	Vorti 5-mg	144	44.2±11.89	25.06±5.43	4.7±0.65	Not	31.6±3.67	8	Asian	⑤⑥⑦
	Vorti 10-mg	150	45.7±10.9	25.93±5.46	4.7±0.66	reported	31.8±4.02
	Vorti 20-mg	154	44.0±11.79	24.82±5.21	4.7±0.65		31.7±3.73
	Placebo	152	43.6±11.57	24.82±5.13			31.6±3.56
Jain et al (32) 2013	Vorti 5-mg	300	42.5±13.0	30.5±8.2	Not	32.7±5.4	42.5±13.0	6	White, Black,	①②③④
	Placebo	300	42.4±12.7	30.8±7.7	reported	32.2±5.5	42.4±12.7		Asian, American	⑤⑥⑧
									Indian/Alaskan,
									Pacific Inlander
McIntyre et al (33) 2014	Vorti 10-mg	195	45.4±12.2	Not	4.60±0.62	Not	45.4±12.2	8	Not reported	②④⑥⑦
	Vorti 20-mg	207	46.1±11.8	reported	4.62±0.58	reported	46.1±11.8
	Placebo	196	45.6±12.1		4.55±0.63		45.6±12.1
Mahablesh warkar, et al (34) 2013	Vorti 2.5-mg	153	42.6±12.9	29.5±7.5	4.6±0.62	29.8±5.4	42.6±12.9	8	White, Black,	①③④⑤
	Vorti 5-mg	153	43.1±13.9	31.4±8.8	4.6±0.65	29.0±5.6	43.1±13.9		Asian, American	⑥⑧
	Placebo	153	42.6±13.8	29.6±7.3	4.5±0.62	29.5±6.1	42.6±13.8		Indian/Alaskan,
									Pacific Inlander
Baldwin et al (35) 2012	Voeti 2.5-mg	155	46.0±12.5	Not	4.8±0.7	29.6±5.8	46.0±12.5	8	Asian	②⑧
	Vorti 5-mg	157	44.7±13.1	reported	4.8±0.7	31.3±5.8	44.7±13.1
	Vorti 10-mg	151	45.2±13.1		4.8±0.7	30.4±5.4	45.2±13.1
	Placebo	148	43.4±12.5		4.8±0.7	29.8±5.1	43.4±12.5
Katona et al (36) 2012	Vorti 5-mg	156	70.5±4.8	Not	4.8±0.7	29.2±5.0	70.5±4.8	8	Not Reported	①⑤⑥⑦
	Placebo	145	70.3±4.4	reported	4.7±0.7	29.4±5.1	70.3±4.4			⑧
Mahablesh warkar et al (37) 2015	Vorti 20-mg	198	44.2±12.2	Not	4.6±0.6	Not	44.2±12.2	8	Black, Asian,	②④⑧
	Placebo	194	45.0±12.1	reported	4.6±0.6	reported	45.0±12.1		Other
Boulenger et al (38) 2014	Vorti 15-mg	151	47.0±14.6	Not	4.9±0.6	Not	47.0±14.6	8	Not Reported	⑥⑦⑧
	Vorti 20-mg	151	46.2±13.4	reported	4.8±0.7	reported	46.2±13.4
	Placebo	158	48.1±13.1		4.9±0.7		48.1±13.1
Mahablesh warkar et al (39) 2015	Vorti 15-mg	147	43.1±12.28	31.3±7.48	4.5±0.55	Not	43.1±12.28	8	White, Black,	⑥⑦⑧
	Vorti 20-mg	154	42.8±12.40	30.9±7.63	4.5±0.60	reported	42.8±12.40		Asian, Native
	Placebo	161	42.4±12.55	31.1±7.88	4.6±0.58		42.4±12.55		American/
									Alaskan native
Alvarez et al (40) 2012	Vorti 5-mg	108	43.8±11.6	Not	5.2±0.7	29.9±5.4	43.8±11.6	6	Not reported	⑧
	Vorti 10-mg	100	42.3±13.1	reported	5.1±0.7	29.3±5.6	42.3±13.1
	Placebo	105	42.0±10.9		5.1±0.7	29.7±5.0	42.0±10.9

[i] LOCF, last observation carried forward; HDRS-24, 24-item Hamilton Depression Rating Scale; MADRS, Montgomery-Asberg Depression Rating Scale; SDS, Sheehan Disability Scale; CGI-I, Clinical Global Impression Scale-Improvement; Vorti, Vortioxetine; ①HDRS-24(LOCF), 24-item Hamilton Depression Rating Scale; ②MADRS(LOCF), Montgomery-Asberg Depression Rating Scale; ③SDS(LOCF), Sheehan Disability Scale; ④CGI-I(LOCF), Clinical Global Impression Scale-Improvement; ⑤, HDRS-24 response rate; ⑥, MADRS remission rate; ⑦, MADRS response rate; ⑧, Safety.

Efficacy outcomes

To evaluate the efficacy of vortioxetine on MDD, the changes in the total scores of HADRS-24, MADRS, SDS and CGI-I, and the HADRS-24 and MADRS response rates (≥50%) and those in MADRS remission rate (≤10%) were retrieved from the included studies. The HADRS-24 response rate, MADRS response rate (≥50%) and MADRS remission rate (≤10%) were considered as the primary efficacy outcomes. In addition, the HADRS-24, MADRS, SDS and CGI-I total score changes were considered as the secondary efficacy outcomes. The effect of vortioxetine (5, 10, 15 and 20 mg) compared with placebo on the response of patients with MDD was assessed via measuring the changes in the total scores of HADRS-24, MADRS, SDS and CGI-I, and in HADRS-24 response rate, MADRS response rate (≥50%) and MADRS remission rate (≤10%) at 8 or 6 weeks. Patients were considered to be the measure index responders when reduced total scores from baseline at the end of the study were obtained. The present study analysis revealed the following results: i) In the 5-mg vortioxetine group vs. the placebo group, no statistically significant difference was observed in terms of HADRS-24, MADRS, SDS and CGI-I total score change, and HADRS-24 response rate (P>0.05). However, the MADRS response (≥50%; M-H RR=1.57; 95% CI=1.23-1.99) and remission (≤10%; M-H RR=1.28; 95% CI=1.10-1.49; I2=49%) rates were significantly enhanced in the 5-mg vortioxetine dose group compared with the placebo group (P<0.05). ii) In the 10-mg vortioxetine group vs. the placebo group, a statistically significant difference was observed in the total scores of HADRS-24 [MD=-4.93; 95% CI=-5.12-(-4.76)], MADRS [MD=-3.65; 95% CI=-5.61-(-1.69)], SDS [MD=-1.54; 95% CI=-1.76-(-1.32)] and CGI-I [MD=-0.58; 95% CI=-0.64-(-0.52)], and in HDRS-24 response rate (M-H RR=2.16; 95% CI=1.52-3.05), MADRS response rate (≥50%; M-H RR=1.44; 95% CI=1.15-1.81; I2=39%) and MADRS remission rate (≤10%; M-H RR=1.61; 95% CI=1.38-1.89; I2=34%; P<0.05). iii) The HADRS-24, MADRS, SDS, CGI-I total scores and HDRS-24 response rate were not reported for the 15-mg vortioxetine group. However, there was no statistically significant difference in the MADRS response (≥50%) and remission rates (≤10%) between the two groups (P>0.05) (4). Consistently, in the 20-mg vortioxetine group vs. placebo, the HADRS-24 and SDS total score changes, and the HADRS-24 response rate were not reported. A statistically significant difference was obtained between the above groups in terms of the total scores of MADRS [MD=-2.30; 95% CI=-2.45-(-2.15)] and CGI-I [MD=-0.58; 95% CI=-1.13-(-0.02)], MADRS response rate (≥50%; M-H RR=1.55; 95% CI=1.07-2.23) and MADRS remission rate (≤10%; M-H RR=1.54; 95% CI=1.17-2.03; all P<0.05). Details of efficacy and heterogeneity assessment are presented in Table II and Fig. S1, Fig. S2, Fig. S3 and Fig. S4.

Table II Comparison of therapeutic effect analysis in each trial group.

Table II

Comparison of therapeutic effect analysis in each trial group.

	HADRS-24 total score change (MD)	MADRS total score change (MD)	SDS total score change (MD)	CGI-I total score change (MD)	HADRS-24 response rate (RR)	MADRS response rate (≥50%) (RR)	MADRS remission rate (≤10%) (RR)
Vortioxetine 5-mg vs. Placebo	-7.46 (-16.69 to 1.77) Z=1.58; (P=0.11) 4 trials	-2.07 (-4.37 to 0.23) Z=1.76 (P=0.08); 3 trials	-0.27 (0.94 to 0.40) Z=0.79 (P=0.43); 3 trials	-0.22 (0.48 to 0.03) Z=1.71 (P=0.09); 3 trials	1.34 (1.00 to 1.80) Z=1.96 (P=0.05); 4 trials	1.57 (1.23 to 1.99) Z=3.66 (P<0.05); 3 trials	1.28 (1.10 to 1.49)a Z=3.13 (P<0.05); 5 trials
Vortioxetine 10-mg vs. Placebo	-4.93 (-5.12 to -4.74) Z=52.18 (P<0.05); 1 trial	-3.65 (-5.61 to -1.69) Z=3.65 (P<0.05); 3 trials	-1.54 (-1.76 to -1.32) Z=13.86 (P<0.05); 1 trial	-0.58 (-0.64 to -0.52) Z=19.35 (P<0.05); 2 trials	2.16 (1.52 to 3.05) Z=4.34 (P<0.05); 1 trial	1.44 (1.15 to 1.81)a Z=3.13 (P<0.05); 3 trials	1.61 (1.38 to 1.89)a Z=5.88 (P<0.05); 3 trials
Vortioxetine 15-mg vs. Placebo	Not reported	Not reported	Not reported	Not reported	Not reported	1.36 (0.75 to 2.47) Z=1.00 (P=0.32); 2 trials	1.41 (0.91 to 2.19) Z=1.53 (P=0.13); 2 trials
Vortioxetine 20-mg vs. Placebo	Not reported	-2.30 (-2.45 to -2.15) Z=30.37 (P<0.05); 1 trial	Not reported	-0.58 (-1.13 to -0.02) Z=2.02 (P<0.05); 2 trials	Not reported	1.55 (1.07 to 2.23) Z=2.32 (P<0.05); 4 trials	1.54 (1.17 to 2.03) Z=3.06 (P<0.05); 4 trials

[i] HDRS-24, 24-item Hamilton Depression Rating Scale; MADRS, Montgomery-Asberg Depression Rating Scale; SDS, Sheehan Disability Scale; CGI-I, Clinical Global Impression Scale-Improvement;

[ii] ^aI²≤5; MD, Mean Difference; R, Risk Ratio.

Safety outcomes

Subsequently, a parallel, independent meta-analysis was performed. The meta-analysis of the 11 articles included 16 adverse reactions, including nausea, headache, nasopharyngitis, dizziness, diarrhea, constipation, dry mouth, insomnia, adverse events (AEs) leading to discontinuation, serious AEs (SAEs), fatigue, hyperhidrosis, decreased appetite, somnolence, vomiting and upper respiratory tract infection, with a total adverse drug reaction rate of 5%. Therefore, the analysis revealed that compared with the placebo group: i) A statistically significant difference was observed in the onset of nausea (response rate, M-H RR=2.48; 95% CI=1.99-3.10; I2=0%) in the 5-mg vortioxetine group. However, no difference was observed in headache, nasopharyngitis, dizziness, diarrhea, constipation, dry mouth, insomnia, AEs leading to discontinuation, SAEs, fatigue, hyperhidrosis, decreased appetite, somnolence and vomiting, between the two groups (P>0.05). ii) A statistically significant difference was obtained in nausea (response rate, M-H RR 3.02; 95% CI, 2.16=4.23; I2=0%) and hyperhidrosis (response rate, M-H RR=4.65; 95% CI=1.36-15.95; I2=0%) between the 10-mg vortioxetine group compared with the placebo group. No differences in the remaining AEs were recorded (P>0.05). iii) In the 15-mg vortioxetine group, a statistically significant difference was only obtained in nausea (response rate, M-H RR=3.12; 95% CI=2.18-4.46; I2=0%), compared with the placebo group, and not for the other AEs (P>0.05). iv) In the 20-mg vortioxetine group, a statistically significant difference in nausea (response rate, M-H RR=3.39; 95% CI=2.54-4.51; I2=0%), insomnia (response rate, M-H RR=2.25; 95% CI=1.09-4.68; I2=22%) and vomiting (response rate, M-H RR=13.42; 95% CI=1.78-101.37) was observed. There were no statistically significant differences in headache, nasopharyngitis, dizziness, diarrhea, constipation, dry mouth, AEs leading to discontinuation, SAEs, fatigue, decreased appetite, vomiting and upper respiratory tract infection between the two groups (P>0.05). The details of AE assessment are demonstrated in Table III and Fig. S5, Fig. S6, Fig. S7 and Fig. S8. Furthermore, the visual inspection of funnel plots revealed low obvious publication bias (Fig. 3).

Figure 3 Publication bias for vortioxetine in all included studies.

Table III Comparison of safety analysis in each trial group.

Table III

Comparison of safety analysis in each trial group.

	Vortioxetine 5-mg	Vortioxetine 10-mg	Vortioxetine 15-mg	Vortioxetine 20-mg
Nausea	2.48% (1.99 to 3.10%)a I2=0%; 7 trials	3.02% (2.16 to 4.23%)a I2=0%; 4 trials	3.12% (2.18 to 4.46%)a I2=0%; 2 trials	3.39% (2.54 to 4.51%)a I2=0%; 4 trials
Headache	1.01% (0.82 to 1.25%) I2=4%; 6 trials	0.89% (0.67 to 1.20%) I2=0%; 4 trials	1.36% (0.89 to 2.08%) I2=0%; 2 trials	1.19% (0.88 to 1.60%) I2=0%; 4 trials
Nasopharyngitis	1.24% (0.84 to 1.84%) I2=0%; 4 trials	0.81% (0.52 to 1.27%) I2=0%; 4 trials	0.65% (0.24 to 1.74%) 1 trial	0.95% (0.62 to 1.46%) I2=0%; 3 trials
Dizziness	1.00% (0.72 to 1.40%) I2=0%; 7 trials	1.20% (0.72 to 2.01%) I2=35%; 4 trials	1.53% (0.35 to 6.60%) I2=78%; 2 trials	1.70% (0.82 to 3.52%) I2=52%; 4 trials
Diarrhea	0.95% (0.52 to 1.73%) I2=61%; 5 trials	0.75% (0.43 to 1.29%) I2=34%; 3 trials	1.87% (1.04 to 3.38%) I2=33%; 2 trials	1.13% (0.54 to 2.39%) I2=59%; 4 trials
Constipation	1.00% (0.60 to 1.68%) I2=0%; 5 trials	1.12% (0.49 to 2.60%) I2=25%; 3 trials	0.87% (0.35 to 2.13%) 1 trial	1.73% (0.89 to 3.37%) I2=0%; 2 trials
Dry mouth	1.20% (0.86 to 1.68%) I2=0%; 6 trials	0.99% (0.55 to 1.80%) I2=31%; 3 trials	0.97% (0.54 to 1.76%) I2=0%; 2 trials	1.42% (0.92 to 2.18%) I2=23%; 4 trials
Insomnia	1.46% (0.83 to 2.58%) I2=0%; 4 trials	1.05% (0.50 to 2.23%) I2=0%; 3 trials	0.68% (0.23 to 2.02%) 1 trial	2.25% (1.09 to 4.68%)a I2=22%; 2 trials
AE leading to discontinuation	0.35% (0.07 to 1.70%) 1 trial	2.38% (0.94 to 6.03%) 1 trial	Not reported	1.72% (0.60 to 4.92%) I2=58%; 2 trials
Any SAE	2.10% (0.19 to 22.88%) 1 trial	3.06% (0.32% to 29.09%); 1 trial	Not reported	5.03% (0.60 to 42.57%); 1 trial
Fatigue	1.07% (0.58 to 1.97%) I2=0%; 4 trials	1.08% (0.44 to 2.66%) I2=0%; 2 trials	1.57% (0.45 to 5.45%) 1 trial	1.31% (0.36 to 4.78%) 1 trial
Hyperhidrosis	1.58% (0.75 to 3.31%) I2=0%; 4 trials	4.65% (1.36 to 15.95%)a I2=0%; 2 trials	0.87% (0.27 to 2.80%) 1 trial	0.08% (0.00 to 1.42%) 1 trial
Decreased appetite	2.07% (0.73 to 5.92%) I2=0%; 3 trials	0.47% (0.04 to 5.14%) 1 trial	Not reported	2.92% (0.31 to 27.86%) 1 trial
Somnolence	1.16% (0.58 to 2.32%) I2=0%; 3 trials	0.94% (0.28 to 3.19%) 1 trial	Not reported	Not reported
Vomiting	1.28% (0.45 to 3.63%) I2=0%; 2 trials	1.32% (0.43 to 4.07%) 1 trial	7.57% (0.94 to 60.80%); 1 trial	13.42% (1.78 to 101.37%)a 1 trial
Upper respiratory tract infection	Not reported	Not reported	0.49% (0.17 to 1.38%) 1 trial	0.75% (0.31 to 1.82%) 1 trial

[i] ^aP<0.05. AE, adverse effect; SAE, serious AE.

Discussion

Summary of findings. It has been reported that serious mental illness (SMI), such as depression, bipolar disorder and schizophrenia, not only seriously affects patient's interpersonal relationships, work and independent living capabilities, but also is a significant risk factor for increasing the prevalence and mortality of somatic diseases (41). Previous studies verified that patients with SMI were at a higher risk of developing acute organ dysfunction (42) and their life expectancy was lower by an average of 20 years compared with the general population (2). Among the aforementioned conditions, cardiovascular diseases, cancer and diseases of the endocrine system, such as diabetes and obesity, and respiratory system have become the main cause of SMI-related death (43). Numerous studies suggested that cardiovascular diseases were the most common diseases among patients with SMI, characterized by relatively high mortality rate (44,45). Among them, the lifetime prevalence rate of myocardial infarction, angina pectoris and stroke were ~30% (46). Cognitive dysfunction is common in patients with MDD, while it has been reported that cognitive impairment may persist after the relief of depressive symptoms (47). Since cognitive impairment is the most significant residual symptom and can reduce the quality of life of patients with MDDs, persistent cognitive impairment may prevent full recovery from depressive episodes. Several scholars have advocated ‘remission of cognitive function’ as a novel target for the treatment of MDD. Chen et al (48) confirmed that the changes in the number of dendritic spines, dendritic spine density, dendritic length and branching in the hippocampus could promote the maturation of hippocampal dendritic spines and improve cognitive function in animals, thus indicating that they could also potentially affect cognitive function in humans. Another study suggested that vortioxetine could increase histamine levels in the cerebral cortex and hippocampus, and it could indirectly improve cognitive function via regulating serotonin in the vegetarian system (49).

In the present systematic review, 11 randomized, double-blind, placebo-controlled trials were included based on the inclusion and exclusion criteria (30-40). Subsequently, an evidence-based medical evaluation of the efficacy and safety of the four vortioxetine preparation doses (5, 10, 15 and 20 mg) was carried out. All 11 studies showed that the methodological assessments performed were of high quality, with a low risk of bias, thus verifying that this was a systematic review with high data credibility. Additionally, all included studies displayed favorable general-data consistency and baseline balance, while they were comparable. Furthermore, the heterogeneity of several indicators in the reference material was assessed by Q test. In the efficacy assessment, there were varying degrees of heterogeneity in each group (5, 10, 15 and 20 mg). Due to the heterogeneity (I2>50%) in the reference material and since all the clinical research data were valid, the weight ratio difference analysis of the relevant effect indicators could not accurately exclude heterogeneous sources. Therefore, to evaluate the efficacy of each indicator in the four groups (5-, 10-, 15- and 20-mg vortioxetine groups), a meta-analysis using a fixed effect model was performed for only a few indicators (MADRS remission rate in the 5- and 10-mg vortioxetine groups and MADRS response rate in the 10-mg vortioxetine group), while the remaining indicators were assessed by a REM. For patients with depression, who were initially treated with 5 mg vortioxetine QD, the results revealed that there was a significant difference in the MADRS response and remission rates at the end of the treatment period. In addition, the clinical symptom scores of patients in the 10- and 20-mg QD vortioxetine groups were markedly changed with increasing dosage. Compared with the placebo group, there was a significant difference in symptom scores and response rates (P<0.05). However, data processing revealed that the data from five indicators, namely HADRS-24 total score, MADRS total score, SDS total score, CGI-I total score and HADRS-24 response rate, were missing for the 15-mg QD vortioxetine group. Therefore, only the data for two indicators, namely MADRS response rate and MADRS remission rate, were used for the combined analysis in this group. There was no statistically significant difference between the aforementioned two indicators in the 15-mg QD vortioxetine group compared with the placebo group (P>0.05), possibly due to the effect of reference source, selection and publication bias. Therefore, a descriptive analysis was performed, and a single double-blind RCT showed that patient treatment with 15 mg vortioxetine QD could significantly improve symptom score (39). The aforementioned results were consistent with those reported by Meeker et al (50), which verified that patient treatment with 5, 10, 15 or 20 mg vortioxetine could significantly affect the response and remission rates. In the present study, the results also showed that administration of 5, 10, 15 or 20 mg vortioxetine could notably alter HADRS-24, MADRS, SDS and CGI-I total scores, and the HADRS-24 response rate compared with placebo. However, the results also demonstrated that treatment with 5 mg vortioxetine had no effect on HADRS-24, MADRS, SDS and CGI-I total scores, and HADRS-24 response rate, compared with placebo. There was an effort to use the subgroup analysis due to the heterogeneity, but the included research in every efficacy outcome measure is no more than four, therefore the subgroup analysis cannot solve heterogeneity (I2>50%), Finally, it was decided not to use the subgroup analysis. Nevertheless, more high-quality clinical studies are urgently needed to further investigate and confirm the aforementioned findings.

In terms of safety evaluation, an independent merged meta-analysis on 16 adverse reactions, namely nausea, headache, nasopharyngitis, dizziness, diarrhea, constipation, dry mouth, insomnia, AEs leading to discontinuation, SAEs, fatigue, hyperhidrosis, decreased appetite, somnolence, vomiting and upper respiratory tract infection was carried out. The total incidence rate of AEs in the 11 included studies was >5% (30-40). The results revealed that nausea, hyperhidrosis, insomnia and vomiting were more common in all trial vortioxetine groups (5, 10, 15 and 20 mg) compared with placebo. Regarding the association between AEs and vortioxetine dosage, no absolute linear association was found in this evaluation system. However, this result could be greatly affected by bias and the completeness of the data. Therefore, further high-quality clinical studies are required to verify the aforementioned results.

Moreover, there have been two systematic reviews about vortioxetine (51,52); the study conducted by Zhang et al (51) only the response rate, remission rate and tolerability were analyzed, which is insufficient. In the study conducted by Thase et al (52), the safety outcomes were not analyzed. In addition, in the present study more and new outcome measures were analyzed. The efficiency outcome measures not only contained MADRS response rate, MADRS remission rate, but also HADRS-24, MADRS, SDS, CGI-I and HADRS-24 response rate. The safety outcome measure was also substantial. Although the conclusion was different, Zhang et al (51) considered that vortioxetine is more advantageous over placebo in treating MDD among adults, while the present study revealed that there was no statistical difference between vortioxetine 5-mg and placebo in HADRS-24, MADRS, SDS, CGI-I total score change and HADRS-24 response rate. Overall, the present study provided sufficient outcome measure, new conclusions compared with other meta-analysis of vortioxetine and could offer significant information to clinicians who prescribe vortioxetine to patients suffering from SMI.

Limitations

Even though the included studies in the present meta-analysis were strictly screened according to the inclusion and exclusion criteria, a rigorous standardized quality evaluation was conducted for the grouped 11 references. There were nevertheless certain limitations. The following deficiencies still exist in the meta-analysis of the system evaluation and effect indicators: i) In multiple databases, there could be omissions in the collection of references due to the retrieval strategy applied; ii), Since two researchers reviewed the references in an independent and parallel way according to the inclusion and exclusion criteria, the risk of reference selection bias during the screening of reference material should not be excluded; iii) Screening results from relevant clinical trial registration websites were unsatisfactory. For several grey references and evidence from non-traditional sources, the query results may not be a strong supplement for the required entry data due to the effects of confidentiality agreements. Therefore, a particular degree of publication bias risk could be included in the analysis. iv) Since the 11 studies included into the analysis were all in English, the risk of language bias could not be avoided. However, the aforementioned risks could be present in any systematic review and meta-analysis. The strict and careful reference screening and quality assessment, and the appropriate methodological handling can guarantee scientific, complete and accurate results with high clinical reference value.

The present systematic review demonstrated that 5-20 mg was significantly effective compared with placebo in the treatment of MDD. However, treatment of patients with 5 mg vortioxetine displayed no difference in the HADRS-24, MADRS, SDS and CGI-I total scores, and HADRS-24 response rate. Interestingly enough, increased vortioxetine doses were associated with improved tolerance and high safety. Considering the potential bias and confounding of the studies included in this meta-analysis, more well-conducted and large-scale RCTs are needed to confirm the findings of the present study.

Supplementary Material

Forest plot of comparison of (A) HDRS-24, (B) MADRS, (C) Sheehan Disability Scale and (D) Clinical Global Impression Scale-Improvement total score changes, and (E) HDRS-24 response rate, (F) MADRS response rate (≥50%) and (G) MADRS remission rate (≤10%) when administered 5 mg vortioxetine vs. placebo. HDRS-24, 24-Items Hamilton Rating Scale for Depression; MADRS, Montgomery-Asberg Depression Rating Scale.

Forest plot of comparison of (A) HDRS-24, (B) MADRS, (C) Sheehan Disability Scale and (D) Clinical Global Impression Scale-Improvement total score changes, and (E) HDRS-24 response rate, (F) MADRS response rate (≥50%) and (G) MADRS remission rate (≤10%) when administered 10 mg vortioxetine vs. placebo. HDRS-24, 24-Items Hamilton Rating Scale for Depression; MADRS, Montgomery-Asberg Depression Rating Scale.

Forest plot of comparison of (A) MADRS response rate (≥50%) and (B) MADRS remission rate (≤10%) when administered 15 mg vortioxetine vs. placebo. HADRS-24 total score change, MADRS total score change, Sheehan Disability Scale total score change, Clinical Global Impression Scale-Improvement total score change and HDRS-24 response rate not reported. HDRS-24, 24-Items Hamilton Rating Scale for Depression; MADRS, Montgomery-Asberg Depression Rating Scale.

Forest plot of comparison of (A) MADRS and (B) Clinical Global Impression Scale-Improvement total score changes, and (C) MADRS response rate (≥50%) and (D) MADRS remission rate (≤10%) when administered 20 mg vortioxetine vs. placebo. HADRS-24 and Sheehan Disability Scale total score change, and HDRS-24 response rate not reported. HDRS-24, 24-Items Hamilton Rating Scale for Depression; MADRS, Montgomery-Asberg Depression Rating Scale.

Forest plot of comparison of (A) nausea, (B) headache, (C) nasopharyngitis, (D) dizziness, (E) diarrhea, (F) constipa-tion, (G) dry mouth, (H) insomnia, (I) AE leading to discontinuation, (J) any serious AE leading to discontinuation, (K) fatigue, (L) hyperhidrosis, (M) decreased appetite, (N) somnolence and (O) vomiting when administered 5 mg vortioxetine vs. placebo. Upper respiratory tract infection was not reported. AE, adverse event.

Forest plot of comparison of (A) nausea, (B) headache, (C) nasopharyngitis, (D) dizziness, (E) diarrhea, (F) constipa-tion, (G) dry mouth, (H) insomnia, (I) AE leading to discontinuation, (J) any serious AE leading to discontinuation, (K) fatigue, (L) hyperhidrosis, (M) decreased appetite, (N) somnolence and (O) vomiting when administered 10 mg vortioxetine vs. placebo. Upper respiratory tract infection was not reported. AE, adverse event.

Forest plot of comparison of (A) nausea, (B) headache, (C) nasopharyngitis, (D) dizziness, (E) diarrhea, (F) constipa-tion, (G) dry mouth, (H) insomnia, (I) fatigue, (J) hyperhidrosis, (K) vomiting and (L) upper respiratory tract infection when administered 15 mg vortioxetine vs. placebo. AEs, serious AEs and decreased appetite were not reported. AE, adverse event.

Forest plot of comparison of (A) nausea, (B) headache, (C) nasopharyngitis, (D) dizziness, (E) diarrhea, (F) constipa-tion, (G) dry mouth, (H) insomnia, (I) AE leading to discontinuation, (J) any serious AE leading to discontinuation, (K) fatigue, (L) hyperhidrosis, (M) decreased appetite, (N) vomiting and (O) upper respiratory tract infection when administered 20 mg vortioxetine vs. placebo. Somnolence was not reported. AE, adverse event.

Acknowledgements

Not applicable.

Funding

Funding: The present study was supported by the Science and Technology Plan Fund from Yaan Science and Technology Bureau (grant no. 22KJJH0039).

Availability of data and materials

All data generated or analyzed during this study are included in this published article.

Authors' contributions

SG, LF, XX and DZ designed the study, collected the data, undertook the statistical analyses, drafted the manuscript and contributed equally to this work. SG and XX participated in the review of the study, performed the statistical analyses, helped to interpret data and drafted the manuscript. All authors read and approved the final version of the manuscript. Data authentication is not applicable.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References