Vortioxetina para la depresión en adultos
Resumen
Antecedentes
El trastorno depresivo mayor es un trastorno mental frecuente que afecta la mente, la conducta y el cuerpo de la persona. Se expresa a través de diversos síntomas y se asocia con deficiencia significativa. A pesar de la variedad de opciones de tratamiento farmacológicas y no farmacológicas, queda mucho por hacer para mejorar el tratamiento farmacológico de la depresión en cuanto a la eficacia y la tolerancia. El antidepresivo disponible más reciente es la vortioxetina. Se supone que la acción antidepresiva de la vortioxetina está relacionada con una modulación directa de la actividad de los receptores serotoninérgicos y la inhibición del transportador de la serotonina. El mecanismo de acción no se comprende completamente, pero se ha señalado que es novedoso. La vortioxetina se clasificó en la categoría de "Otros" antidepresivos y, por lo tanto, puede proporcionar una opción a los fármacos antidepresivos existentes.
Objetivos
Evaluar la eficacia y la aceptabilidad de la vortioxetina en comparación con placebo y otros fármacos antidepresivos en el tratamiento de la depresión aguda en adultos.
Métodos de búsqueda
Se hicieron búsquedas en el Registro Especializado del Grupo Cochrane de Revisión de Depresión, Ansiedad y Neurosis (Cochrane's Depression, Anxiety and Neurosis Review Group's Specialised Register) hasta mayo 2016, sin aplicar ninguna restricción de fecha, idioma o estado de publicación. Se verificaron las listas de referencias de los estudios y las revisiones relevantes, los informes del organismo regulador y las bases de datos de ensayos.
Criterios de selección
Se incluyeron los ensayos controlados aleatorizados que compararon la eficacia y la tolerancia, o ambas, de la vortioxetina versus placebo u otro agente antidepresivo en el tratamiento de la depresión aguda en adultos.
Obtención y análisis de los datos
Dos autores de la revisión, de forma independiente, seleccionaron los estudios y extrajeron los datos. Se extrajeron los datos sobre las características de los estudios, las características de los participantes, los detalles de la intervención y las medidas de resultados en cuanto a la eficacia, la aceptabilidad y la tolerancia. Se analizaron solamente los datos por intención de tratar y se utilizaron los riesgo relativo (RR) como tamaños del efecto para los datos dicotómicos y las diferencias medias (DM) para los datos continuos, con los intervalos de confianza (IC) del 95%. Los metanálisis utilizaron modelos de efectos aleatorios.
Resultados principales
Se incluyeron 15 estudios (7746 participantes) en esta revisión. Siete estudios fueron controlados con placebo; ocho estudios compararon vortioxetina con inhibidores de la recaptación de serotonina‐norepinefrina (IRSN). No fue posible identificar estudios que compararan la vortioxetina con fármacos antidepresivos de otras clases como los inhibidores selectivos de la recaptación de serotonina (ISRS).
La vortioxetina puede ser más efectiva que placebo en los tres resultados de eficacia: respuesta (RR de Mantel‐Haenszel 1,35; IC del 95%: 1,22 a 1,49; 14 estudios, 6220 participantes), remisión (RR 1,32; IC del 95%: 1,15 a 1,53; 14 estudios, 6220 participantes) y síntomas depresivos medidos mediante la Montgomery‐Åsberg Depression Scale (MADRS) (rango de puntuación: 0 a 34; la puntuación mayor significa un resultado peor: DM ‐2,94; IC del 95%: ‐4,07 a ‐1,80; 14 estudios, 5566 participantes). La calidad de la evidencia fue baja para la respuesta y la remisión y muy baja para los síntomas depresivos. No se encontró evidencia de una diferencia en la tasa total de abandonos (RR 1,05; IC del 95%: 0,93 a 1,19; 14 estudios, 6220 participantes). Más participantes interrumpieron la vortioxetina en comparación con placebo debido a los efectos adversos (RR 1,41; IC del 95%: 1,09 a 1,81; 14 estudios, 6220 participantes) pero menos la interrumpieron debido a la falta de efectividad (RR 0,56; IC del 95%: 0,34 a 0,90; P = 0,02; 14 estudios, 6220 participantes). La calidad de la evidencia para los abandonos fue moderada. Los análisis de subgrupos y de sensibilidad no mostraron factores que influyeran significativamente en los resultados.
En comparación con otros antidepresivos, evidencia de muy baja calidad de ocho estudios no mostró diferencias clínicamente significativas entre la vortioxetina y los IRSN como una clase en la respuesta (RR 0,91; IC del 95%: 0,82 a 1,00; 3159 participantes) ni en la remisión (RR 0,89; IC del 95%: 0,77 a 1,03; 3155 participantes). Hubo una diferencia pequeña que favoreció a los IRSN en las puntuaciones de los síntomas depresivos en la MADRS (DM 1,52; IC del 95%: 0,50 a 2,53; ocho estudios, 2807 participantes). Evidencia de calidad muy baja de ocho estudios (3159 participantes) no mostró diferencias significativas entre la vortioxetina y los IRSN como una clase en las tasas totales de abandonos (RR 0,89; IC del 95%: 0,73 a 1,08), los abandonos debidos a eventos adversos (RR 0,74; IC del 95%: 0,51 a 1,08) ni en los abandonos debidos a la falta de efectividad (RR 1,52; IC del 95%: 0,70 a 3,30).
Versus antidepresivos individuales, los análisis indicaron que la vortioxetina puede ser menos efectiva que la duloxetina en cuanto a las tasas de respuesta (RR 0,86; IC del 95%: 0,79 a 0,94; seis estudios, 2392 participantes) y las puntuaciones de los síntomas depresivos en la MADRS (DM 1,99; IC del 95%: 1,15 a 2,83; seis estudios, 2106 participantes). Versus la venlafaxina, el metanálisis de dos estudios no encontró diferencias estadísticamente significativas (respuesta: RR 1,03; IC del 95%: 0,85 a 1,25; 767 participantes; puntuaciones de los síntomas depresivos: DM 0,02; IC del 95%: ‐2,49 a 2,54; 701 participantes). En cuanto al número de participantes que informaron al menos un efecto adverso (tolerancia), la vortioxetina fue mejor que los IRSN como una clase (RR 0,90; IC del 95%: 0,86 a 0,94; ocho estudios, 3134 participantes) y que la duloxetina (RR 0,89; IC del 95%: 0,84 a 0,95; seis estudios, 2376 participantes). Sin embargo, el análisis de sensibilidad suscita algunas dudas sobre este resultado, ya que solo dos estudios administraron una dosis comparable.
Se consideró que ninguno de los estudios tuvo alto riesgo de sesgo en ningún dominio, pero se consideró que todos los estudios tuvieron un riesgo incierto de sesgo de informe selectivo y de otros sesgos.
Conclusiones de los autores
El lugar de la vortioxetina en el tratamiento de la depresión aguda no está claro. Los análisis mostraron que la vortioxetina puede ser más efectiva que placebo en cuanto a la respuesta, la remisión y los síntomas depresivos, pero la relevancia clínica de estos efectos no está clara. Además, en general la calidad de la evidencia para apoyar estos resultados fue baja. En comparación con los IRSN, no se encontró una ventaja de la vortioxetina. La vortioxetina fue menos efectiva que la duloxetina, pero menos pacientes informaron efectos adversos cuando se trataron con vortioxetina en comparación con duloxetina. Sin embargo, estos resultados no están claros y no están bien apoyados por la evidencia. Una limitación principal de la evidencia actual es la falta de comparaciones con los ISRS, que por lo general se recomiendan como tratamientos de primera línea para la depresión aguda. Se necesitan estudios con comparaciones directas con los ISRS para analizar esta brecha y se pueden complementar mediante metanálisis en red para definir la función de la vortioxetina en el tratamiento de la depresión.
PICO
Resumen en términos sencillos
Vortioxetina para el tratamiento de la depresión en adultos
¿Por qué es importante esta revisión?
Muchas personas presentan depresión mayor. La depresión mayor es una enfermedad grave que provoca una angustia significativa en los pacientes y sus familias. La depresión mayor afecta el trabajo y las relaciones de los pacientes, pero también puede afectar a los pacientes físicamente, por ejemplo, mediante cambios en la concentración o el apetito. Los fármacos antidepresivos disponibles no siempre son efectivos para tratar la depresión mayor y también pueden tener efectos secundarios desagradables. Esta revisión compara un antidepresivo nuevo, la vortioxetina, con placebo (un tratamiento simulado, p.ej., comprimido de azúcar) y con otros antidepresivos. Se supone que la vortioxetina funciona de un modo diferente al de otros antidepresivos disponibles y es importante saber si es un tratamiento efectivo y una alternativa posible a los tratamientos ya disponibles.
¿Quién estará interesado en esta revisión?
A los pacientes afectados por depresión mayor y a sus familias, a los médicos generales (MG), los psiquiatras, los farmacéuticos y a otros profesionales que trabajan en servicios de salud mental para adultos.
¿Qué preguntas pretende contestar esta revisión?
¿La vortioxetina es más efectiva que el placebo para tratar a los pacientes con un episodio de depresión mayor?
¿La vortioxetina es más o menos efectiva que otros tratamientos antidepresivos disponibles?
¿Más o menos pacientes continúan con el tratamiento cuando se tratan con vortioxetina en comparación con placebo u otros antidepresivos?
¿Más o menos pacientes tienen efectos secundarios cuando se tratan con vortioxetina en comparación con otros antidepresivos?
¿Qué estudios se incluyeron en la revisión?
En mayo de 2016, se efectuaron búsquedas en bases de datos médicas electrónicas para encontrar ensayos que compararan la vortioxetina con placebo u otros antidepresivos. Se incluyeron solo los estudios que utilizaron un diseño controlado aleatorizado (donde los pacientes se asignan al azar a uno de dos o más grupos de tratamiento) e incluyeron adultos (con una edad mayor de 18 años) con un diagnóstico de depresión mayor. Se incluyeron 15 ensayos con 7746 participantes en la revisión.
¿Qué nos dice la evidencia de la revisión?
La calidad de la evidencia varió de muy baja a moderada según el resultado (el síntoma o efecto que se midió) y la comparación. La vortioxetina fue más efectiva que placebo, pero no fue más efectiva que otros antidepresivos utilizados habitualmente. Los estudios no encontraron diferencias en los pacientes que interrumpieron el tratamiento en comparación con placebo u otros antidepresivos. La vortioxetina solo se comparó con un tipo de fármaco (los llamados IRSN) y no se comparó con los antidepresivos prescritos con mayor frecuencia. Los resultados variaron notablemente entre los estudios.
¿Qué debe suceder a continuación?
No se pueden establecer conclusiones firmes sobre la vortioxetina. La vortioxetina fue efectiva para tratar la depresión mayor aguda, pero no mostró una ventaja clara en comparación con algunos tratamientos que ya están disponibles. Las conclusiones también son difíciles de establecer porque faltan comparaciones con los antidepresivos prescritos con mayor frecuencia (los llamados ISRS). Además, no está claro si la vortioxetina tiene una ventaja con respecto a los efectos secundarios específicos asociados con los antidepresivos habitualmente prescritos, por ejemplo, los problemas sexuales. Estas cuestiones deben abordarse en estudios futuros.
Authors' conclusions
Summary of findings
| Vortioxetine compared to Placebo for adults with Major Depressive Disorder | ||||||
| Patient or population: adults with Major Depressive Disorder | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Quality of the evidence | Comments | |
| Risk with Placebo | Risk with Vortioxetine | |||||
| Response | Study population | RR 1.35 | 6220 | ⊕⊕⊝⊝ | All studies were sponsored by the pharmaceutical companies that manufacture vortioxetine. Small difference favouring vortioxetine. | |
| 356 per 1,000 | 480 per 1,000 | |||||
| Total number of drop‐outs | Study population | RR 1.05 | 6220 | ⊕⊕⊕⊝ | No difference between vortioxetine and placebo. | |
| 160 per 1,000 | 168 per 1,000 | |||||
| Remission | Study population | RR 1.33 | 6217 | ⊕⊕⊝⊝ | Small difference favouring vortioxetine. | |
| 224 per 1,000 | 299 per 1,000 | |||||
| Depressive Symptoms | The change in depressive symptoms score ranged from 10.8 to 15.9 points | The change was 2.94 points higher | ‐ | 5566 | ⊕⊝⊝⊝ | Small difference favouring vortioxetine. |
| Drop‐out due to adverse events | Study population | RR 1.41 | 6220 | ⊕⊕⊕⊝ | Small difference favouring placebo. | |
| 38 per 1,000 | 53 per 1,000 | |||||
| Drop‐out due to inefficacy | Study population | RR 0.56 | 6220 | ⊕⊕⊕⊝ | Small difference favouring vortioxetine. | |
| 31 per 1,000 | 18 per 1,000 | |||||
| Tolerability | Study population | RR 1.12 | 6182 | ⊕⊕⊕⊝ | Small difference favouring placebo | |
| 564 per 1,000 | 632 per 1,000 | |||||
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1 A serious risk of bias is present, as about 30% of the studies showed an overall dropout rate above 20%, evidence was downgraded by one level 2 A moderate degree of heterogeneity (I‐squared 30‐60%) is present, evidence was downgraded by one level 3 A substantial degree of heterogeneity (I‐squared 60‐90%) is present, evidence was downgraded by two levels | ||||||
| Vortioxetine compared to SNRIs for adults with Major Depressive Disorder | ||||||
| Patient or population: adults with Major Depressive Disorder | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Quality of the evidence | Comments | |
| Risk with SNRIs | Risk with Vortioxetine | |||||
| Response | Study population | RR 0.91 | 3159 | ⊕⊝⊝⊝ | All studies were sponsored by the pharmaceutical companies that manufacture vortioxetine. No difference between vortioxetine and SNRIs. | |
| 577 per 1,000 | 525 per 1,000 | |||||
| Total number of drop‐outs | Study population | RR 0.89 | 3159 | ⊕⊝⊝⊝ | No difference between vortioxetine and SNRIs. | |
| 212 per 1,000 | 189 per 1,000 | |||||
| Remission | Study population | RR 0.89 | 3155 | ⊕⊝⊝⊝ | No difference between vortioxetine and SNRIs. | |
| 370 per 1,000 | 329 per 1,000 | |||||
| Depressive Symptoms | The change in depressive symptoms score ranged from 14.1 to 23.4 points | The change was 1.52 points lower | ‐ | 2807 | ⊕⊝⊝⊝ | Small difference favouring SNRIs. |
| Drop‐out due to adverse events | Study population | RR 0.74 | 3159 | ⊕⊝⊝⊝ | No difference between vortioxetine and SNRIs. | |
| 97 per 1,000 | 72 per 1,000 | |||||
| Drop‐out due to inefficacy | Study population | RR 1.52 | 3159 | ⊕⊝⊝⊝ | No difference between vortioxetine and SNRIs. | |
| 14 per 1,000 | 21 per 1,000 | |||||
| Tolerability | Study population | RR 0.90 | 3134 | ⊕⊕⊝⊝ | Small difference favouring vortioxetine | |
| 690 per 1,000 | 621 per 1,000 | |||||
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1 A very serious risk of bias is present as 60% of the studies had more than 20% dropouts overall, evidence was downgraded by two levels 2 A moderate degree of heterogeneity (I‐squared 30‐60%) is present, evidence was downgraded by one level 3 The 95% CI crossed both 1 (no differences) and 0.75 (appreciable benefit for vortioxetine), evidence was downgraded by one level 4 The 95% CI crossed 1 (no differences), 0.75 (appreciable benefit for vortioxetine) and 1.25 (appreciable benefit for SNRIs). Outcome is very imprecise: evidence was downgraded by two levels | ||||||
Background
Description of the condition
Major depressive disorder (MDD) affects a person's mind, behaviour and body and is expressed in a variety of symptoms. The core features are depressed mood and loss of interest or pleasure. Other diagnostic criteria include significant changes in bodyweight, decreased or increased appetite, sleep disturbances, psychomotor agitation or retardation, fatigue, feelings of worthlessness or guilt, reduced concentration and suicidal ideation (APA 2013). According to the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM‐IV), a person must have symptoms for at least two weeks (APA 2013). The core features of major depression have not been changed from DSM‐IV to DSM‐5. Although there are some differences in the diagnosis of depression between DSM and International Classification of Diseases (ICD), diagnoses of major depression according to DSM‐IV seems to be congruent with severe or moderate depressive episodes according to ICD‐10 (Saito 2010).
MDD is a common mental disorder, but prevalence rates vary markedly across countries. Lifetime prevalence was estimated at 14.6% on average in high‐income countries and 11.1% in low‐income countries, with a female:male ratio of about 2:1 (Bromet 2011). MDD is associated with substantial impairment. According to the Global Burden of Disease study, MDD is the second leading cause of disability worldwide (Vos 2012).
Description of the intervention
A variety of pharmacological and non‐pharmacological treatment options is available for the treatment of MDD. Pharmacological treatment options comprise monoamine oxidase inhibitors (MAOIs), tricyclic antidepressants (TCAs) or heterocyclic antidepressants, selective serotonin re‐uptake inhibitors (SSRIs), serotonin‐norepinephrine (‐noradrenaline) reuptake inhibitors (SNRIs), and other antidepressant agents (e.g. mirtazapine, bupropion, reboxetine, agomelatine), as well herbal products (e.g. hypericum). Current guidelines for depression recommend the use of an antidepressant or psychological treatment for people with moderate depression (APA 2010; NICE 2009) as first‐line treatment. According to these guidelines, the effectiveness between classes of antidepressants is similar. However, SSRIs are recommended over TCAs and MAOIs due to their favourable adverse effect profile (APA 2010; NICE 2009). SSRIs have now become the most prescribed antidepressant class in most parts of the world (Bauer 2008; Grover 2013; Zhang 2013).
Vortioxetine was licensed for the treatment of depression by the Food and Drugs Administration (FDA) in September 2013 in the USA (FDA 2014) and by the European Medicines Agency (EMA) in December 2013 for the EU (EMA 2014). Despite the similarities to SSRIs, the mechanism of action of vortioxetine is claimed to be novel (see How the intervention might work). According to the ATC classification of the World Health Organization (WHO), vortioxetine is placed in the category of "Other" antidepressants (WHO 2016). Due to the recent marketing authorisation, clinical experience and data on clinical use of vortioxetine is very limited at this time.
How the intervention might work
The mechanism of action of vortioxetine is not fully understood, but it is assumed to be related to a direct modulation of serotonergic receptor activity and inhibition of the serotonin transporter. Vortioxetine is an antagonist to 5‐HT3, 5‐HT1D and 5‐HT7 receptors, a partial agonist to the 5‐HT1B receptor and a 5‐HT1A receptor agonist (EMA 2014a). However, it is unclear if and how these mechanism contribute to an antidepressant effect. It is hypothesised that serotonin transporter inhibition, combined with the several other actions of vortioxetine at 5‐HT receptors, mainly at 5‐HT3 receptors, enhances the release of serotonin and modulates the release of other neurotransmitters within various brain circuits (enhanced release of norepinephrine, dopamine, histamine, acetylcholine and glutamate; reduced gamma‐aminobutyric acid (GABA) signalling). These actions could improve the efficiency of information processing in malfunctioning brain circuits by facilitating long‐term potentiation, neuroplasticity and increased firing of pyramidal neurons (Du Jardin 2016; Pehrson 2016; Stahl 2015).
Why it is important to do this review
Despite the common use of antidepressants and the variety of available treatment options, there is still an ongoing debate about the use of antidepressants in general as some studies show modest effects compared to placebo (Kirsch 2008). A significant proportion of people do not achieve remission with current treatments (Pigott 2011). Furthermore, although the adverse effect profile of SSRIs is in general judged favourable over TCAs or MAOIs, many people are dropping out of antidepressant treatment (Pigott 2011), mainly due to adverse effects (Bull 2002). Thus, there is still room for improvement of the pharmacological treatment of depression.
Another subject of debate is the comparative effects of modern antidepressants. One multiple‐treatment meta‐analysis of 12 new‐generation antidepressant drugs concluded that sertraline and escitalopram may be more favourable in terms of efficacy and acceptability compared to the other included antidepressants (Cipriani 2009). Notably, this finding could not be replicated in another comprehensive review (Gartlehner 2011). The latter review concluded that there are no substantial differences in efficacy between the antidepressants, but that antidepressants differ in onset of action and adverse events (Gartlehner 2011).
Vortioxetine was approved in late 2013 for the USA and EU and is currently the latest available antidepressant. Randomised controlled trials (RCTs) comparing vortioxetine to placebo or other antidepressants have been published. Several systematic reviews and meta‐analyses have been published on the efficacy and tolerability on vortioxetine. The first meta‐analysis included seven studies comparing vortioxetine to placebo (Berhan 2014). One systematic narrative review gave an overview of the vortioxetine studies and reported results of 10 RCTs in adults with major depression without pooling the results (Citrome 2014). Two meta‐analyses did not conduct a systematic review of the data. The first pooled analysis selected 11 short‐term placebo‐controlled trials and five long‐term open‐label studies to evaluate the safety and tolerability of vortioxetine (Baldwin 2016). The second pooled analysis analysed a selected subset of five studies to examine the effect of vortioxetine on quality of life (Florea 2015). Furthermore, 11 studies were included in an indirect comparison of vortioxetine, duloxetine, sertraline, vilazodone, levomilnacipram and escitalopram (Citrome 2016). Two other meta‐analyses focused on a subset of a specific dose of vortioxetine 5 mg (Fu 2015) or 10 mg (Li 2016) compared to placebo. However, another meta‐analysis, based on an analysis of 11 studies, found no effect differences related to dosing (Meeker 2015). This meta‐analysis included comparisons of vortioxetine and placebo as well as vortioxetine and active comparators, but it included subtherapeutic doses below 5 mg. In sum, these reviews agree that vortioxetine has a significant advantage compared to placebo in terms of efficacy, but found no advantage compared to other available antidepressants.
A summary of the FDA review of vortioxetine is also available (Zhang 2015). The FDA review included 10 short‐term placebo‐controlled trials and concluded that vortioxetine demonstrated efficacy in six of the included trials, but reported that only vortioxetine 20 mg/day showed superiority over placebo in US trials and showed smaller effects in general in US populations. The most recent meta‐analysis, which was co‐authored by employees of the manufacturers of vortioxetine, analysed an almost identical dataset and included data from treatment arms in the approved dose range of 11 short‐term studies which had results published on ClinicalTrials.gov (Thase 2016). Most analyses were conducted with aggregated data, but additional individual participant data were used. Apparently, the analyses did not follow an a priori defined protocol. However, the analyses are in line with the findings from previous reviews, showing a statistically significant advantage of vortioxetine compared to placebo, but a clear dose‐response relationship could not be established. It also confirms the findings of smaller effects in US studies. Thus, despite several attempts to synthesise available literature on vortioxetine, a practical interpretation of data from current systematic reviews may be limited by comprehensiveness issues (lack of thorough search for unpublished data, selective inclusion of trials in the analyses), and possible conflicts of interest. Our review provides an independent, comprehensive and up‐to‐date summary of the available evidence of the efficacy and acceptability of vortioxetine compared to placebo and other active pharmacological treatment options.
Objectives
To assess the efficacy and acceptability of vortioxetine compared with placebo and other antidepressant drugs in the treatment of acute depression in adults.
Methods
Criteria for considering studies for this review
Types of studies
We included RCTs but excluded quasi‐RCTs. For trials with a cross‐over design, we considered only the results from the first randomisation period. We included cluster‐RCTs if sufficient information was available to account for the clustering (see Unit of analysis issues).
Types of participants
Characteristics
Participants of both sexes, of any ethnicity, and aged 18 years and older.
Diagnosis
Participants with a primary diagnosis of unipolar major depression according to DSM‐III (APA 1980), DSM‐ III‐R (APA 1987), DSM‐IV (APA 1994), DSM‐IV‐TR (APA 2000), DSM‐5 (APA 2013), ICD‐10 (WHO 1992), Feighner (Feighner 1972) or Research Diagnostic Criteria (Spitzer 1978), and Chinese Classification of Mental Disorders (CCMD‐3, Chinese Society of Psychiatry 2001). We excluded studies of people with treatment‐resistant depression, defined as inadequate treatment response to at least four weeks of adequate antidepressant treatment.
Comorbidities
We included studies with people with comorbid psychiatric disorders. We excluded antidepressant trials in people with depression with a serious concomitant physical illness (e.g. myocardial infarction, diabetes, cancer, etc.).
Setting
Any setting.
Subset data
Based on National Institute for Health and Care Excellence (NICE) guidelines for the treatment of depression in adults (NICE 2009), we included studies in which less than 20% of the included participants had bipolar depression.
We included studies with relevant subsets of data (e.g. some participants aged below 18 years) if the data were available for the relevant subset and the randomisation was stratified by the criterion in question. Inclusion of these studies was examined in sensitivity analyses.
Types of interventions
Experimental intervention
-
Vortioxetine monotherapy. To increase the clinical applicability of the review, we excluded treatment arms employing dosages below the lowest effective dose of 5 mg/day (EMA 2014b). We included treatment arms with fixed and flexible dosing schemes. Fixed doses are set a priori and are independent from participant criteria, while in flexible dosing schemes the dose is adapted according predefined criteria, for example, insufficient response.
Comparator intervention
-
Placebo.
-
Another antidepressant as monotherapy, including:
-
conventional TCA or heterocyclic antidepressants (amitriptyline, amoxapine, clomipramine, desipramine, dosulepin/dothiepin, doxepin, imipramine, lofepramine, maprotiline, nortriptyline, protriptyline, trimipramine);
-
SSRIs (fluoxetine, fluvoxamine, citalopram, paroxetine, escitalopram);
-
SNRIs (venlafaxine, duloxetine, milnacipran);
-
MAOIs (phenelzine, isocarboxazide, tranylcypromine, moclobemide, brofaromine);
-
other antidepressant agents (mirtazapine, bupropion, reboxetine, agomelatine) or non‐conventional antidepressive agents (herbal products such as hypericum).
-
We applied no restrictions on dosage of the comparators, but we conducted sensitivity analyses and excluded studies with unequal dosing.
Types of outcome measures
We included studies that meet the above inclusion criteria regardless of whether they report on the following outcomes.
Primary outcomes
-
Response to treatment: the primary efficacy outcome was the number of participants who responded to acute treatment, as defined by a reduction of at least 50% on the Hamilton Depression Rating Scale (HAM‐D) scale (Hamilton 1960) or Montgomery‐Åsberg Depression Scale (MADRS; Montgomery 1979), or any other depression scale, or "much or very much improved" (score 1 or 2) on Clinical Global Impression ‐ Improvement (CGI‐I) (Guy 1970). We did not consider other definitions of response in this Cochrane Review.
Where more than one scale was provided, we gave preference as listed. We used response rate instead of a continuous symptom score for the primary efficacy analysis to make the interpretation of results easier (Guyatt 1998).
-
Total number of dropouts: primary outcome measuring acceptability was the total number of participants dropping out during the trial as a proportion of the total number of randomised participants.
Secondary outcomes
-
Remission: number of participants who achieved remission. We defined remission a priori as:
-
-
7 points or less on the 17‐item HAM‐D and as 8 points or less for all the other longer versions of HAM‐D;
-
10 or less points on the MADRS;
-
"not ill or borderline mentally ill" (score 1 or 2) on Clinical Global Impression ‐ Severity (CGI‐S) (Guy 1970) at endpoint.
-
We did not consider other definitions of remissions in this Cochrane Review.
-
Depressive symptoms: endpoint mean scores, or mean change scores at endpoint on HAM‐D, MADRS, or any other depression rating scale score.
-
Dropouts due to adverse events: number of participants who dropped out due to adverse events during the trial as a proportion of the total number of randomised participants.
-
Dropouts due to inefficacy: number of participants who dropped out due to inefficacy during the trial as a proportion of the total number of randomised participants.
-
Tolerability: evaluated using the total number of participants experiencing at least one adverse event.
We collected any data on specific adverse effects and reported these data in tables. However, due to the poor reporting in RCTs (Zorzela 2014), we did not summarise these data in meta‐analyses.
Although the effect on cognition is a relevant outcome for people treated with psychopharmacological treatments, we decided to exclude cognition as an outcome. According to the EMA, cognition was not systematically assessed (EMA 2014a). The EMA report on vortioxetine indicated that only three studies reported this outcome and that a meta‐analysis by the manufacturer of these studies had outcome reporting bias (EMA 2014a). Furthermore, the EMA report pointed out that it was not possible to distinguish between an effect on cognition and a relief of depressive symptoms, because no active comparator was included.
Timing of outcome assessment
Our primary outcomes were the acute phase treatment response (between four and 12 weeks). When studies reported efficacy data at different time points, we gave preference to the time point closest to eight weeks.
Hierarchy of outcome measures
We included efficacy data measured by HAM‐D (Hamilton 1960), MADRS (Montgomery 1979), or any other depression scale. Response or remission rates may also be based on CGI‐I scores (Guy 1970) or a combination of these outcomes. Where more than one criterion was provided, we planned to use the data according to the following hierarchy: HAM‐D, MADRS, other depression scales and CGI and combination of these in cases of remission and response rates. However, due to the reporting of the MADRS (see Description of studies) we gave preference to MADRS outcomes (see Differences between protocol and review). We did not include CGI scores as a continuous outcome. The HAM‐D scale is available in various versions which differ in the number of included items. The versions with 17, 21 or 24 items are the most common and we gave preference to these in this order. Studies may also use different criteria for response or remission. We gave preference according to the list described in the corresponding outcome section (see Primary outcomes; Secondary outcomes).
Search methods for identification of studies
Cochrane Collaboration Depression, Anxiety and Neurosis Review Group's Specialized Register (CCDANCTR)
The Cochrane Collaboration Depression, Anxiety and Neurosis Group (CCDAN) maintain two clinical trials registers at their editorial base in Bristol, UK: a References Register and a Studies Register. The CCDANCTR‐References Register contains over 37,000 reports of RCTs in depression, anxiety and neurosis. Approximately 60% of these references have been tagged to individual, coded trials. The coded trials are held in the CCDANCTR‐Studies Register and records are linked between the two registers using unique Study ID tags. Coding of trials is based on the EU‐PSI coding manual, using a controlled vocabulary (see Cochrane Collaboration Depression, Anxiety & Neurosis Group for further details). Reports of trials for inclusion in the Group's registers are collated from routine (weekly), generic searches of MEDLINE (from 1950), Embase (from 1974) and PsycINFO (from 1967); quarterly searches of the Cochrane Central Register of Controlled Trials (CENTRAL) and review‐specific searches of additional databases. Reports of trials are also sourced from international trials registers via the WHO trials portal (the International Clinical Trials Registry Platform (ICTRP)), pharmaceutical companies, handsearching of key journals, conference proceedings, and other (non‐Cochrane) systematic reviews and meta‐analyses.
Details of CCDAN's generic search strategies (used to identify RCTs) can be found on the Group's website.
Electronic searches
We performed the following electronic searches with no restrictions on date, language or publication status.
-
CCDANCTR‐Studies Register using the following controlled search terms:
Condition = depress*
AND
Intervention = Vortioxetine or "Lu AA21004"
-
CCDANCTR‐References Register using a more sensitive set of free‐text terms to identify additional untagged/uncoded reports of RCTs:
Free‐text = (depress* or dysthymi* or "mood disorder*" or "affective disorder*" or "affective symptom*") and (Vortioxetine or "Lu AA21004" or LuAA21004 or Brintellix)
-
International trial registries via the WHO trials portal (ICTRP) and ClinicalTrials.gov to identify unpublished or ongoing studies, together with the trial registries of relevant pharmaceutical companies:
Searching other resources
Reference lists
We checked the reference lists of all included studies and relevant systematic reviews to identify additional studies missed from the original electronic searches (e.g. unpublished or in‐press citations). Also, we conducted a cited reference search on the Web of Science.
Correspondence
We contacted trialists and subject experts for information on unpublished or ongoing studies or to request additional trial data.
Data collection and analysis
Selection of studies
Two review authors (GO, MK) independently screened titles and abstracts for inclusion of all studies identified by the search and coded them as 'potentially eligible.' We retrieved the full‐texts of study reports/publications rated as 'potentially eligible' by one or both review authors. Two review authors (GO, MK) independently screened the full‐text articles and identified studies for inclusion, and identified and recorded reasons for exclusion of the ineligible studies (see Characteristics of excluded studies table). We resolved any disagreements through discussion or, if required, by consulting a third review author (CB). We collated multiple reports of the same study and included them a single study.
Data extraction and management
We used a data collection form which had been piloted on at least one study in the review to extract study characteristics and outcome data. Two review authors (CB, MK) independently extracted study characteristics and outcome data from included studies. We extracted the following study characteristics.
-
Methods: blinding, total duration of study, details of 'run in' periods, number of study centres and location, study setting, withdrawals and date of study.
-
Participants: sample size, mean age, age range, gender, severity of condition, diagnostic criteria, inclusion criteria and exclusion criteria.
-
Interventions: intervention, comparison, concomitant medications, excluded medications, dose and dosing scheme (fixed versus flexible).
-
Outcomes: primary and secondary outcomes specified and collected, and time points reported.
-
Notes: funding for trial and notable conflicts of interest of trial authors.
We noted in the Characteristics of included studies table if outcome data were not reported in a usable way. We resolved disagreements by consensus or by involving a third review author (CB). One review author (MK) transferred data into Review Manager 5 (RevMan 2014). Two review authors (GO, JB) double checked the data entered for correctness by comparing the data presented in the systematic review with the study reports. A third review author (CB) spot‐checked study characteristics for accuracy against the trial report.
Main planned comparisons
We combined the comparators (see Types of interventions) into classes in the meta‐analyses. Therefore, the main planned comparisons were:
-
vortioxetine versus placebo;
-
vortioxetine versus TCAs/heterocyclics;
-
vortioxetine versus SSRIs;
-
vortioxetine versus SNRIs;
-
vortioxetine versus MAOIs;
-
vortioxetine versus other antidepressant agents.
Wherever suitable, we presented data with substances as subgroups within each class.
Assessment of risk of bias in included studies
Two review authors (MK, GO) independently assessed risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We resolved any disagreements by discussion or by involving a third review author (CB). We assessed the risk of bias in the included studies according to the following domains.
-
Random sequence generation.
-
Allocation concealment.
-
Blinding of participants and personnel.
-
Blinding of outcome assessment.
-
Incomplete outcome data.
-
Selective outcome reporting.
-
Other bias.
We judged potential sources of bias as 'high,' 'low' or 'unclear' and provided supporting quotation a from the study report together with a justification for our judgement in the 'Risk of bias' table. The risk of bias judgements were summarised across different studies for each of the domains listed. We considered blinding separately for different key outcomes where necessary (e.g. for unblinded outcome assessment, risk of bias for treatment discontinuation may be different than for a participant‐reported scale). Where information on risk of bias relates to unpublished data or correspondence with a trial author, we noted this in the 'Risk of bias' table.
When considering treatment effects, we took into account the risk of bias for the studies that contributed to that outcome.
Measures of treatment effect
Dichotomous data
We analysed dichotomous data as risk ratios (RRs), because RRs are more intuitive in their interpretation than odds ratios (Grant 2014), with 95% confidence intervals (CI). Where reported intention‐to‐treat (ITT) analysis was based on a 'modified' ITT not including all dropouts (e.g. leaving out dropouts without postbaseline assessment), we applied a conservative approach and considered these dropouts as non‐responders or non‐remitters (i.e. assumed they would have experienced the negative outcome by the end of the trial, e.g. failure to respond to treatment).
Continuous data
We analysed continuous data as mean difference (MD) with 95% CI, if all studies reported the necessary data from HAM‐D‐scales as an outcome. In cases that data from different scales were combined, we used standardised mean difference (SMD) with 95% CI. We entered and presented data with a consistent direction of effect. Data were analysed as endpoint data. We combined endpoint and change scores only if data were analysed as MD. Analyses were conducted with ITT data as reported (e.g. data from last observation carried forward (LOCF) or mixed model for repeated measurements (MMRM) methods).
We conducted meta‐analyses only where this was meaningful (i.e. if the treatments, participants and underlying clinical question were similar enough for pooling to make sense).
Unit of analysis issues
Cluster‐randomised controlled trials
Cluster‐RCTs were eligible for inclusion if sufficient information was available to account for the clustering (see Differences between protocol and review).
Cross‐over trials
For trials with a cross‐over design, we considered only the results from a first randomisation period (see Differences between protocol and review).
Studies with multiple treatment groups
If more than one treatment arm was reported in a single trial, we only included the relevant treatment arms. In case of multiple relevant treatment arms (e.g. different dosages), we combined these treatment arms into a single group. For dichotomous outcomes, we summarised data across groups; while for continuous outcomes, we combined means and standard deviations according to Chapter 7 of the Cochrane Handbook for Systematic Reviews of Interventions (Section 7.7.3.8, Higgins 2011).
In case of relevant treatment arms that could not be combined (e.g. different SSRIs as comparators), we divided the sample size of the shared group so that the two arms were treated as independent comparisons.
Dealing with missing data
We contacted study authors or study sponsors to verify key study characteristics and missing outcome data. Attempts to contact authors and sponsors were documented, as well as additional data we received in these correspondences.
In the absence of supplemental data from the authors, we planned to calculate the SDs of the HAM‐D (or any other depression scale) and response/remission rates according to validated imputation methods (Furukawa 2005; Furukawa 2006). We planned to examine the validity of these imputations in sensitivity analyses.
Assessment of heterogeneity
Variations in participants and interventions lead to clinical heterogeneity. We extracted basic study characteristics (see Data collection and analysis) from the studies and described these in the Characteristics of included studies tables. A decision was then made as to whether studies were similar enough to combine in meta‐analyses.
We quantified statistical heterogeneity of the studies using the I2 statistic and Chi2 test. As the Chi2 test is known to have low power, we used P = 0.10 as a threshold for statistical significance. Our interpretation of the I2 statistic followed the recommendation of the Cochrane Handbook for Systematic Reviews of Interventions and we considered the I2 statistic as:
-
0% to 40%: might not be important;
-
30% to 60%: may represent moderate heterogeneity;
-
50% to 90%: may represent substantial heterogeneity;
-
75% to 100%: considerable heterogeneity.
We also assessed heterogeneity by visual inspection of forest plots.
Assessment of reporting biases
Publication bias was scrutinised by visual inspection of funnel plots if we include more than 10 studies in the analysis of the outcome in question.
Data synthesis
We used a random‐effects model in our primary analysis. We expected some heterogeneity in the studies included and the random‐effects model incorporates the variance between studies in the model. As a result, CIs are wider. The random‐effects model has the highest generalisability in an empirical examination of summary effect measures for meta‐analyses (Furukawa 2002). We routinely examined the robustness of this summary measure by checking the results under a fixed‐effect model. Material differences between the models were reported.
Subgroup analysis and investigation of heterogeneity
We only conducted subgroup and sensitivity analysis for the primary outcomes.
A priori, we planned to perform the following subgroup analyses.
-
Vortioxetine dosing: fixed versus flexible dosing schemes.
-
Treatment setting: primary care versus inpatient care versus outpatient care.
-
Older people (aged more than 65 years): included versus excluded.
Sensitivity analysis
We planned to perform the following sensitivity analyses to examine the robustness of the effect size.
-
Exclusion of trials with unequal dosing. We defined comparability of doses by comparing the percentage of the maximum licensed daily dose in both groups (e.g. vortioxetine 10 mg/day (50% of 20 mg/day) equals fluoxetine 40 mg/day (50% of 80 mg/day)).
-
Exclusion of studies that did not employ a double‐blind approach.
-
Exclusion of studies with subsets of people with bipolar disorders.
-
Exclusion of trials with dropout rates of more than 20% in one of the treatment arms included.
-
Exclusion of studies with imputed data.
-
Exclusion of studied sponsored by the manufacturer of vortioxetine.
'Summary of findings' table
We employed the GRADE approach to interpret findings (Langendam 2013) and used GRADEpro to import data from Review Manager 5 (RevMan 2014) to create 'Summary of findings' tables. These tables provide outcome‐specific information concerning the overall quality of evidence from each included study in the comparison, the magnitude of effect of the interventions examined and the sum of available data on the outcomes.
For each comparison, we reported the primary outcomes (response and total number of dropouts) and secondary outcomes (remission, depressive symptoms, dropouts due to adverse events, dropouts due to inefficacy and participants experience at least one adverse event).
Results
Description of studies
Results of the search
The literature search identified 113 records including eight duplicates. We excluded 45 of the remaining 105 records based on the abstracts. We retrieved 60 full‐text articles for detailed examination, which led to the exclusion of 45 records. The majority (35 trials) were secondary publications of already included or excluded trials. Two studies did not meet our inclusion criterion for acute treatment (Jacobsen 2015a; Jacobsen 2015b); two were relapse prevention studies (Boulenger 2012; NCT02371980); one trial did not meet our criterion for depression because it randomised remitted participants and healthy controls only (Browning 2014); and one study randomised participants to vortioxetine or agomelatine after an inadequate response to at least six weeks of SSRI or SNRI treatment (Montgomery 2014). We identified two ongoing studies (NCT02294305; NCT02389816) which may fulfil the inclusion criteria of this review and two studies are awaiting assessment as there are no published results (NCT02272517; NCT02279966). Request of additional information by the manufactures or the authors was not necessary. The literature search was last updated in May 2016 (see Figure 1; Characteristics of excluded studies table).
Study flow diagram.
Included studies
Fifteen studies were included in this systematic review (Alvarez 2012; Baldwin 2012; Boulenger 2014; Henigsberg 2012; Jacobsen 2015; Jain 2013; Katona 2012; Mahableshwarkar 2013; Mahableshwarkar 2015a; Mahableshwarkar 2015b; Mahableshwarkar 2015c; McIntyre 2014; NCT01255787; Takeda 2011; Wang 2015). Two of these were unpublished trials carried out by a pharmaceutical company (Takeda) (NCT01255787; Takeda 2011). (See Characteristics of included studies table).
Design
All included studies were randomised trials and applied double‐blind methodology. Seven studies were three‐armed with vortioxetine, an active comparator and placebo (Alvarez 2012; Baldwin 2012; Boulenger 2014; Katona 2012; Mahableshwarkar 2013; Mahableshwarkar 2015a; Mahableshwarkar 2015b). Eight studies were two‐armed, among these seven studies were placebo‐controlled (Henigsberg 2012; Jacobsen 2015; Jain 2013; Mahableshwarkar 2015c; McIntyre 2014; NCT01255787; Takeda 2011), and one study was active controlled only (Wang 2015).
Trial duration
Two studies lasted six weeks (Alvarez 2012; Jain 2013), and 13 studies lasted eight weeks (Baldwin 2012; Boulenger 2014; Henigsberg 2012; Jacobsen 2015; Katona 2012; Mahableshwarkar 2013; Mahableshwarkar 2015a; Mahableshwarkar 2015b; Mahableshwarkar 2015c; McIntyre 2014; NCT01255787; Takeda 2011; Wang 2015).
Sample sizes
Overall, the studies included 7746 participants. Of these, 4134 were randomised to vortioxetine. Of the remaining 3612 participants, 2299 were randomised to placebo, 1313 to SNRIs (344 to venlafaxine and 969 to duloxetine). The mean sample size per arm was 209 participants (range 105 to 448).
Setting
All studies were multicentre trials. Six studies were conducted in a single nation: the USA (Jacobsen 2015; Jain 2013; Mahableshwarkar 2013; Mahableshwarkar 2015a; Mahableshwarkar 2015c) and Japan (Takeda 2011). One multinational study recruited Asian participants only (Wang 2015). The other studies were multinational across continents. An overview of countries where participants were recruited is given in Figure 2.
Countries participating in trials. The categories represent the number of studies randomising participants within a country.
Four studies enrolled both inpatients and outpatients (Baldwin 2012; Boulenger 2014; McIntyre 2014; Wang 2015). Three studies recruited exclusively outpatients (Alvarez 2012; Jacobsen 2015; Jain 2013). Eight studies did not explicitly report the setting (Henigsberg 2012; Katona 2012; Mahableshwarkar 2013; Mahableshwarkar 2015a; Mahableshwarkar 2015b; Mahableshwarkar 2015c; NCT01255787; Takeda 2011).
Participants
All studies included participants with a diagnosis of MDD. No trial enrolled people with comorbid psychiatric disorders.
Thirteen studies randomised participants from 18 years of age: eight studies recruited participants aged between 18 and 75 years (Baldwin 2012; Boulenger 2014; Henigsberg 2012; Jacobsen 2015; Jain 2013; Mahableshwarkar 2013; Mahableshwarkar 2015a; Mahableshwarkar 2015c), and four studies recruited between the ages of 18 and 65 years (Alvarez 2012; Mahableshwarkar 2015b; McIntyre 2014; Wang 2015). Two studies recruited participants from 20 years of age: one study between the ages of 20 and 64 years (NCT01255787), and one study between the ages of 20 and 75 years (Takeda 2011). One study included only older participants (Katona 2012; aged 65 to 88 years).
Interventions and comparators
Eight studies compared vortioxetine to SNRIs: two compared vortioxetine to venlafaxine (Alvarez 2012; Wang 2015), and six compared vortioxetine to duloxetine (Baldwin 2012; Boulenger 2014; Katona 2012; Mahableshwarkar 2013; Mahableshwarkar 2015a; Mahableshwarkar 2015b). Seven studies compared vortioxetine to placebo only (Henigsberg 2012; Jacobsen 2015; Jain 2013; McIntyre 2014; Mahableshwarkar 2015c; NCT01255787; Takeda 2011). We found no studies comparing vortioxetine to TCAs/heterocyclics, SSRIs, MAOIs or other antidepressant agents.
One study used a flexible vortioxetine dose scheme (range from 10 mg/day to 20 mg/day) (Mahableshwarkar 2015b). The other 14 trials used a fixed vortioxetine doses scheme (5 mg/day, 10 mg/day, 15 mg/day or 20 mg/day). Two studies applied subtherapeutic dosages of vortioxetine below 5 mg/day (1 mg/day (Henigsberg 2012); 2.5 mg/day (Mahableshwarkar 2013)). These treatment arms were excluded.
Outcomes
All 15 studies provided efficacy data (either as dichotomous or as continuous outcome) and tolerability/acceptability data and could be entered into a meta‐analysis.
Nine studies used the MADRS for their primary outcome measures (Alvarez 2012; Baldwin 2012; Boulenger 2014; Jacobsen 2015; Mahableshwarkar 2015a; Mahableshwarkar 2015c; NCT01255787; Takeda 2011; Wang 2015). Four studies used the HAM‐D‐24 (Henigsberg 2012; Jain 2013; Katona 2012; Mahableshwarkar 2013), two studies used the Digit Symbol Substitution Test (DSST) (Mahableshwarkar 2015b; McIntyre 2014), and one study additionally the Rey Auditory Verbal Learning Test (RAVLT) (McIntyre 2014) for primary outcome measures. For secondary outcomes, the studies used mainly MADRS, CGI‐I, Sheehan Disability Scale (SDS), CGI‐S, HAM‐D‐24, and Hamilton Anxiety Rating Scale (HAM‐A). Two studies also used the HAM‐D‐17 (Henigsberg 2012; Takeda 2011), and two studies used additional cognitive tests (DSST, RAVLT, Trail Making Test ‐ A (TMT‐A), Trail Making Test ‐ B (TMT‐B), Stroop, Perceived Deficits Questionnaire (PDQ), Simple Reaction Time (SRT), Cognitive Reflection Test (CRT), Groton Maze Learning Test (GMLT), Detection Task (DT), Identification Task (IT), and One‐Back Task (Mahableshwarkar 2015b; McIntyre 2014)). All studies reported response rates and remission rates. Thirteen studies defined the response rate as 50% or greater decrease from baseline in MADRS total score and two studies as 50% or greater decrease from baseline in HAM‐D‐24 total score (Jain 2013; Mahableshwarkar 2013). All studies defined the remission rate as MADRS total score of 10 or less.
All studies reported dropouts due to any reason and dropouts due to adverse effects. All but one study (McIntyre 2014) reported dropouts due to inefficacy and the total number of participants who experienced adverse effects.
As expected, the reporting of the individual adverse effects varied markedly. Seven studies reported adverse events if the incidence was at least 5% per arm (Alvarez 2012; Baldwin 2012; Boulenger 2014; Katona 2012; Mahableshwarkar 2015b; McIntyre 2014; Wang 2015). Another seven studies set the threshold at 2% (Henigsberg 2012; Jacobsen 2015; Jain 2013; Mahableshwarkar 2013; Mahableshwarkar 2015a; Mahableshwarkar 2015c; NCT01255787), and one study set the threshold at 0% (Takeda 2011).
Excluded studies
Overall, we excluded six studies (14 references) from the systematic review because they did not meet the inclusion criteria. They were designed as relapse prevention studies (Boulenger 2012; NCT02371980), were not conducted in acute therapy (Jacobsen 2015a; Jacobsen 2015b), recruited randomised remitted participants or healthy controls (Browning 2014), or included participants with a treatment‐resistant depression (Montgomery 2014) (see Figure 1 and Characteristics of excluded studies table).
Studies awaiting classification
Two studies have recently been completed, but have not yet published results (NCT02272517; NCT02279966). Both studies are short‐term randomised, double‐blind trials of eight weeks' duration, which examine the effects of vortioxetine on cognitive functions in people with depression in comparison to an SSRI (see Characteristics of studies awaiting classification table).
Ongoing studies
We identified two ongoing studies (see Characteristics of ongoing studies table). The ongoing studies are short‐term randomised, double‐blind trials of eight weeks' duration (NCT02389816) or 12 weeks' duration (NCT02294305). One study examines the efficacy of vortioxetine for the treatment of depression in people with comorbid social anxiety disorder (NCT02294305). The other study is comparing the efficacy of vortioxetine for the treatment of depression in Japanese people (NCT02389816). One study is ongoing, but not recruiting (NCT02294305), the other is currently recruiting participants (NCT02389816).
Risk of bias in included studies
For graphical representations of the judgements of risk of bias, refer to Figure 3 and Figure 4. Full details of judgements for every included study are presented in the 'Risk of bias' tables within the Characteristics of included studies table.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
We rated none of the studies as having a high risk of bias in any domain, but we rated all studies at unclear risk of bias in at least two domains (see Figure 3 and Figure 4 for summary graphs). All studies were sponsored by the pharmaceutical companies that manufactures vortioxetine (Lundbeck, Takeda), and two of them were unpublished.
Allocation
Eight studies did not report details on sequence generation and were judged at unclear risk of bias (Henigsberg 2012; Jacobsen 2015; Mahableshwarkar 2013; Mahableshwarkar 2015a; Mahableshwarkar 2015b; Mahableshwarkar 2015c; NCT01255787; Takeda 2011). In addition, five studies did not adequately describe allocation concealment (Henigsberg 2012; Jacobsen 2015; Mahableshwarkar 2015b; NCT01255787; Takeda 2011).
Blinding
All RCTs were reported as double‐blind and so were at low risk of bias. All studies used at least identically appearing capsules for blinding.
Incomplete outcome data
Nine studies had a dropout rate below 20% in all treatment arms and so were at low risk of attrition bias (Alvarez 2012; Henigsberg 2012; Jacobsen 2015; Katona 2012; Mahableshwarkar 2015b; Mahableshwarkar 2015c; McIntyre 2014; NCT01255787; Takeda 2011). Of the six remaining studies, two studies had a dropout rate above 20% in the vortioxetine arm (Baldwin 2012; Boulenger 2014), one study in the active control arm (Wang 2015), one study in the placebo arm (Jain 2013), one study in the vortioxetine and in the active control arm (Mahableshwarkar 2015a), and one study in all arms (Mahableshwarkar 2013). The range in these six studies was from 20.1% to 27.4%. These studies were at unclear risk of attrition bias.
Selective reporting
We rated all included studies at unclear risk of selective reporting bias, because published protocols were unavailable. However, publications and entries in clinical trial registers did not reveal discrepancies.
Other potential sources of bias
All studies were sponsored by the pharmaceutical companies that manufactures vortioxetine (Lundbeck, Takeda) and were, therefore, assessed as having an unclear risk of bias.
Effects of interventions
See: Summary of findings for the main comparison Vortioxetine compared to Placebo for adults with Major Depressive Disorder; Summary of findings 2 Vortioxetine compared to SNRIs for adults with Major Depressive Disorder
We have reported the results of the present systematic review by grouping the comparators into two classes: placebo and SNRIs. Specific comparators are presented in subgroups where possible. We could not identify relevant studies comparing vortioxetine with TCAs, heterocyclics, SSRIs, MAOIs or other antidepressants.
Comparison 1. Vortioxetine versus placebo
Fourteen studies including 6220 participants contributed data to the comparison of vortioxetine versus placebo (see summary of findings Table for the main comparison). The quality of evidence contributing to all outcomes was rated as moderate to very low, because of high dropout rates and statistical heterogeneity.
Primary outcomes
1.1. Response to treatment
There was evidence that vortioxetine was more effective than placebo (Mantel‐Haenszel RR 1.35, 95% CI 1.22 to 1.49; P < 0.001; 14 studies, 6220 participants). Statistical heterogeneity was substantial between studies (I2 = 60%) (Analysis 1.1; Figure 5).
Forest plot of comparison: 1 Vortioxetine versus placebo, outcome: 1.1 Response.
1.2. Total number of dropouts
There was no evidence that vortioxetine was associated with a lower or higher total dropout rate than placebo (RR 1.05, 95% CI 0.93 to 1.19; P = 0.40; 14 studies, 6220 participants). There was no heterogeneity (I2 = 0%) (Analysis 1.2; Figure 6).
Forest plot of comparison: 1 Vortioxetine versus placebo, outcome: 1.2 Total number of dropouts.
Secondary outcomes
1.3. Achieved remission
There was evidence that more participants achieved remission with vortioxetine than with placebo (RR 1.32, 95% CI 1.15 to 1.53; P < 0.001; 14 studies, 6220 participants). Heterogeneity was substantial between studies (I2 = 58%) (Analysis 1.3).
1.4. Depressive symptoms
There was evidence that vortioxetine was significantly more effective in lowering MADRS score compared to placebo (MD ‐2.94, 95% CI ‐4.07 to ‐1.80, P < 0.001; 14 studies, 5566 participants). Heterogeneity was high between studies (I2 = 79%) (Analysis 1.4).
1.5. Dropout due to adverse events
There was evidence that vortioxetine was associated with a higher dropout rate due to adverse events compared to placebo (RR 1.41, 95% CI 1.09 to 1.81; P = 0.008; 14 studies, 6220 participants). There was no heterogeneity (I2 = 0%) (Analysis 1.5).
1.6. Dropout due to inefficacy
There was evidence that vortioxetine was associated with a lower dropout rate due to inefficacy compared to placebo (RR 0.56, 95% CI 0.34 to 0.90; P = 0.02; 14 studies, 6220 participants). Heterogeneity between studies was moderate (I2 = 41%) (see Analysis 1.6).
1.7. Tolerability
There was evidence that more participants experienced adverse effects when treated with vortioxetine than when treated with placebo (RR 1.12, 95% CI 1.07 to 1.16; P < 0.001; 14 studies, 6182 participants). Heterogeneity between studies was low (I2 = 8%) (see Analysis 1.7).
Specific adverse effects compared to placebo are reported descriptively in Analysis 1.14. One study reported all adverse effects mentioned (Takeda 2011). Due to the limits of graphs in Review Manager 5 and in line with the majority of studies, we only reported adverse effects with an incidence of 2% or greater in one of the treatment arms for this study. Serious adverse events are reported in Analysis 1.15.
This analysis was not conducted with ITT data according to our conservative approach (see Measures of treatment effect), but with ITT data as reported in the trials.
Comparison 2. Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors
Eight studies including 3159 participants contributed data to the comparison of vortioxetine versus SNRIs (see summary of findings Table 2). The quality of evidence contributing to all outcomes was very low because of high dropout rates and substantial statistical heterogeneity.
Primary outcomes
2.1. Response to treatment
There was no evidence that vortioxetine was less or more effective than SNRIs as a whole (RR 0.91, 95% CI 0.82 to 1.00; P = 0.06; 8 studies, 3159 participants). Heterogeneity was substantial between studies (I2 = 61%) (Analysis 2.1; Figure 7).
Forest plot of comparison: 3. Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors, outcome: 3.1 Response.
Although there was no statistically significant difference between the specific comparators (Chi2 = 2.84, degrees of freedom (df) = 1, P = 0.09), response rates were significantly lower for vortioxetine compared to duloxetine (RR 0.86, 95% CI 0.79 to 0.94; P = 0.001; 6 studies, 2392 participants; I2 = 28%) while there was no difference in response rates compared to venlafaxine (RR 1.03, 95% CI 0.85 to 1.25; P = 0.73; 2 studies, 767 participants; I2 = 69%).
2.2. Total number of dropouts
There was no evidence that vortioxetine was associated with a lower or higher total dropout rate than SNRIs as a whole (RR 0.89, 95% CI 0.73 to 1.08; P = 0.25; 8 studies, 3159 participants). Heterogeneity between studies was moderate (I2 = 44%) (Analysis 2.2; Figure 8).
Forest plot of comparison: 3. Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors, outcome: 3.2 Total number of dropouts.
There was no significant difference between trials comparing vortioxetine to duloxetine or venlafaxine (Chi2 = 2.87, df = 1, P = 0.09), but total dropout rates were significantly lower for vortioxetine compared to venlafaxine (RR 0.70, 95% CI 0.52 to 0.93; P = 0.02; 2 studies, 767 participants; I2 = 0%). There was no statistically significant difference between vortioxetine and duloxetine for total dropouts (RR 0.96, 95% CI 0.76 to 1.21; P = 0.74; 6 studies, 2392 participants; I2 = 45%).
Secondary outcomes
2.3. Achieved remission
There was no significant difference in the number of participants who achieved remission between vortioxetine and SNRIs as a whole (RR 0.89, 95% CI 0.77 to 1.03; P = 0.11; 8 studies, 3155 participants). Heterogeneity between studies was substantial (I2 = 57%) (Analysis 2.3).
There was no statistically significant difference between the specific comparators (Chi2 = 1.15, df = 1, P = 0.28) and there were no statistically significant differences in remission rates between vortioxetine and venlafaxine (RR 0.99, 95% CI 0.81 to 1.20, P = 0.88; 2 studies, 767 participants; I2 = 37%) or vortioxetine and duloxetine (RR 0.85, 95% CI 0.70 to 1.02; P = 0.09; 6 studies, 2388 participants; I2 = 58%).
2.4. Depressive symptoms
There was evidence that vortioxetine was less effective in lowering depression scores compared to SNRIs as a whole (MD 1.52, 95% CI 0.50 to 2.53; P = 0.003; 8 studies, 2807 participants). Heterogeneity between studies was moderate (I2 = 50%) (Analysis 2.4).
There was no significant difference between trials comparing vortioxetine to duloxetine or venlafaxine (Chi2 = 2.11, df = 1, P = 0.15). Comparing vortioxetine to duloxetine, the depression scores were significantly more reduced by duloxetine (MD 1.99, 95% CI 1.15 to 2.83; P < 0.001; 6 studies, 2106 participants; I2 = 6%). There was no significant difference for this outcome between vortioxetine and venlafaxine (MD 0.02, 95% CI ‐2.49 to 2.54; P = 0.99; 2 studies, 701 participants; I2 = 65%).
2.5. Dropout due to adverse events
There was no evidence that vortioxetine was associated with a lower or higher dropout rate due to adverse events compared to SNRIs as a whole (RR 0.74, 95% CI 0.51 to 1.08; P = 0.12; 8 studies, 3159 participants). Heterogeneity between studies was moderate (I2 = 55%) (Analysis 2.5).
There was a statistically significant difference between the comparators (Chi2 = 7.07, df = 1, P = 0.008). Dropout rates due to adverse events were significantly lower for vortioxetine compared to venlafaxine (RR 0.42, 95% CI 0.26 to 0.67; P < 0.001; 2 studies; 767 participants; I2 = 0%). There was no statistically significant difference between vortioxetine and duloxetine (RR 0.92, 95% CI 0.65 to 1.31; P = 0.65; 6 studies, 2392 participants; I2 = 30%).
2.6. Dropout due to inefficacy
There was no evidence that vortioxetine was associated with a lower or higher dropout rate due to inefficacy compared to SNRIs as a whole (RR 1.52, 95% CI 0.70 to 3.30; P = 0.29; 8 studies, 3159 participants). Heterogeneity between studies was moderate (I2 = 30%) (Analysis 2.6).
There was no significant difference between trials comparing vortioxetine to duloxetine or venlafaxine (Chi2 = 1.29, df = 1, P = 0.26). Furthermore, there were no significant differences in dropout rates due to inefficacy between vortioxetine and venlafaxine (RR 2.68, 95% CI 0.99 to 7.24; P = 0.05; 2 studies, 767 participants; I2 = 0%) or vortioxetine and duloxetine (RR 1.16, 95% CI 0.41 to 3.31; P = 0.78; 6 studies, 2392 participants; I2 = 34%).
2.7. Tolerability
There was evidence that fewer participants experienced adverse effects when treated with vortioxetine than when treated with SNRIs as a whole (RR 0.90, 95% CI 0.86 to 0.94; P < 0.001; 8 studies, 3139 participants). There was no heterogeneity (I2 = 0%) (Analysis 2.7).
There was no statistically significant difference between the specific comparators (Chi2 = 0.09, df = 1, P = 0.76). The comparison between vortioxetine and duloxetine showed that fewer participants experienced adverse effects when treated with vortioxetine (RR 0.89, 95% CI 0.84 to 0.95; P < 0.001; 6 studies, 2376 participants; I2 = 18%). There was no significant difference between vortioxetine and venlafaxine (RR 0.91, 95% CI 0.82 to 1.00; P = 0.06; 2 studies, 758 participants; I2 = 0%).
Specific adverse effects compared to SNRIs are reported descriptively in Analysis 2.16. Specific serious adverse events are reported in Analysis 2.17.
This analysis used ITT data as reported in the trials.
Subgroup analyses
Comparison 1. Vortioxetine versus placebo
Fixed versus flexible dosing schemes
Two studies compared placebo to a flexible dose of vortioxetine (Alvarez 2012; Mahableshwarkar 2015b). There were no significant differences between the subgroups in terms of treatment response (test for subgroup differences: Chi2 = 0.05, df = 1, P = 0.82; Analysis 1.8) and total number of dropouts (Chi2 = 0.70, df = 1, P = 0.40; Analysis 1.9).
Treatment setting: primary care versus inpatient care versus outpatient care
We found no studies in primary care settings and all studies including inpatients also included outpatients, so it was impossible to conduct this subgroup analysis.
Older people (aged greater than 65 years): included versus excluded
We excluded four studies from this subgroup analysis, because it was unclear if older participants were included (Henigsberg 2012; Jain 2013; Mahableshwarkar 2013; Takeda 2011). Four studies excluded older participants (Alvarez 2012; Mahableshwarkar 2015b; McIntyre 2014; NCT01255787). One study recruited only older participants (Katona 2012). There were no differences in response rates between the subgroups (Chi2 = 0.52, df = 1, P = 0.47; Analysis 1.10), but dropout rates differed significantly (Chi2 = 5.02, df = 1, P = 0.02; Analysis 1.11). In the studies including older participants, the dropout rates were significantly lower in the placebo groups (RR 1.25, 95% CI 1.05 to 1.49; P = 0.01; I2 = 0%). The number of total dropouts was not significantly different compared to placebo in the studies excluding older participants (RR 0.90, 95% CI 0.71 to 1.13; P = 0.36; I2 = 0%).
Comparison 2. Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors
Fixed versus flexible dosing schemes
One study compared a flexible dose of vortioxetine 10 mg to 20 mg versus a fixed dose of duloxetine 60 mg/day (Mahableshwarkar 2015b). All other studies used a fixed dose scheme. The study with a flexible dose found no significant differences in terms of response rates (Chi2 = 0.03, df = 1, P = 0.86; Analysis 2.8) and total dropouts (Chi2 = 0.04, df = 1, P = 0.84; Analysis 2.9), as compared to the fixed‐dose studies.
Treatment setting: primary care versus inpatient care versus outpatient care
We found no studies in primary care settings and all studies including inpatients also included outpatients, so it was impossible to conduct this subgroup analysis.
Older participants (aged greater than 65 years): included versus excluded
We excluded one study from this analysis, because it was unclear if older participants were included in the study population (Mahableshwarkar 2013). Three studies comparing vortioxetine to an SNRI excluded older participants (Alvarez 2012; Mahableshwarkar 2015b; Wang 2015). There was no significant difference between the subgroups in response rates (test for subgroup differences: Chi2 = 2.52, df = 1, P = 0.11; Analysis 2.10) and total dropout rates (test for subgroup differences: Chi2 = 2.38, df = 1, P = 0.12; Analysis 2.11).
Sensitivity analyses
The following sensitivity analyses were defined a priori.
Exclusion of trials with unequal dosing
This sensitivity analysis is only meaningful for the comparisons of vortioxetine with active comparators.
In six studies, vortioxetine and control antidepressants were compared using unequal doses. Two studies compared vortioxetine 5 mg/day (25% of the maximum dose) to duloxetine 60 mg/day (50% of the maximum dose) (Katona 2012; Mahableshwarkar 2013), and one compared vortioxetine 5 mg/day or 10 mg/day (50% of the maximum dose) to venlafaxine 225 mg/day (100% of the maximum dose for moderately depressed outpatients) (Alvarez 2012). Two studies used higher vortioxetine doses and compared vortioxetine 15 mg/day or 20 mg/day (100% of the maximum dose) to duloxetine 60 mg/day (50% of the maximum dose) (Boulenger 2014; Mahableshwarkar 2015a). One study compared flexible dosing of vortioxetine to a fixed dose of duloxetine (Mahableshwarkar 2015b). To increase the usability of the analysis, the analysis was conducted by grouping the comparisons into subgroups according to the direction of the imbalance in dosing.
The analysis of response rates revealed statistically significant differences between the groups (Chi² = 14.99, df = 3, P = 0.002; Analysis 2.12). The two trials with fair (or comparable) dosing found no differences between vortioxetine and SNRIs (RR 1.09, 95% CI 0.97 to 1.22; P = 0.13; 912 participants; I2 = 0%), but the studies using higher and lower vortioxetine doses than comparator showed significantly higher response rates in participants treated with SNRIs (higher: RR 0.80, 95% CI 0.72 to 0.90; P < 0.001; 2 studies, 903 participants; I2 = 0%; lower: RR 0.87, 95% CI 0.78 to 0.98; P = 0.02; 3 studies, 936 participants; I2 = 9%).
The analysis of the total number of dropouts found no statistically significant differences between the subgroups (Chi2 = 3.68, df = 3, P = 0.30; Analysis 2.13). In the trials with fair (or comparable) dosing, vortioxetine showed statistically significant lower total dropout rate than SNRIs as a whole (RR 0.75, 95% CI 0.59 to 0.96; P = 0.02; 2 studies, 912 participants; I2 = 0%). The studies using higher vortioxetine doses, lower vortioxetine doses and flexible versus fixed dosing found no statistically significant dropout rates between vortioxetine and SNRIs (higher dose: RR 1.33, 95% CI 0.74 to 2.39; P = 0.33; 2 studies, 903 participants; I2 = 72%; lower vortioxetine dose: RR 0.77, 95% CI 0.59 to 1.02; P = 0.06; 3 studies, 936 participants; I2 = 0%; flexible versus fixed dosing: RR 0.94, 95% CI 0.60 to 1.47; P = 0.77; 1 study, 408 participants).
Exclusion of studies that did not employ a double‐blind approach
All our included studies were double‐blinded.
Exclusion of studies with subsets of participants with bipolar disorders
None of the included studies randomised people with bipolar disorder.
Exclusion of trials with dropout rates of more than 20% in one of the treatment arm included
This sensitivity analyses were performed for the comparisons of vortioxetine with placebo and vortioxetine with SNRIs.
Six studies reported overall dropout rates of more than 20% and were excluded (Baldwin 2012; Boulenger 2014; Jain 2013; Mahableshwarkar 2013; Mahableshwarkar 2015a; Wang 2015).
The analyses of response rates showed no significant differences between dropout rates below and above 20% for the comparison between vortioxetine and placebo (Chi2 = 1.00, df = 1, P = 0.32; Analysis 1.12) and the comparison between vortioxetine and SNRIs (Chi2 = 0, df = 1, P = 0.95; Analysis 2.14).
The analyses of the total number of dropouts found no statistically significant differences between the subgroups for the comparison between vortioxetine and placebo (Chi2 = 0.21, df = 1, P = 0.64; Analysis 1.13) and the comparison between vortioxetine and SNRIs (Chi2 = 0.11, df = 1, P = 0.74; Analysis 2.15).
Exclusion of studies with imputed data
There is no need for this sensitivity analysis, because the analysis did not include imputed data.
Exclusion of studied sponsored by the manufacturer of vortioxetine
All studies were conducted by the manufacturers of vortioxetine, therefore this sensitivity analysis is not meaningful.
Reporting bias
We examined the funnel plot for the primary outcomes for the comparison of vortioxetine and placebo only, because the comparison with SNRIs did not contain more than 10 studies. Both funnel plots were inconclusive. We were able to identify two unpublished trials (NCT01255787; Takeda 2011).
Discussion
Summary of main results
Our review included 15 studies (7746 participants). Statistically, vortioxetine was more effective than placebo. This advantage was consistent across the three efficacy outcomes of response, remission and depressive symptoms, although the quality of the evidence was low for response and remission and very low for depressive symptoms (see summary of findings Table for the main comparison). Clinically, there is uncertainty on the relevance of these differences. According to some authors, statistically significant differences in response and remission rates should be accepted as clinically relevant (Montgomery 2009), but in the present review we estimated a RR of 1.35 for response and 1.32 for remission, which correspond to a number needed to treat for an additional beneficial outcome (NNTB) of 8 (95% CI 5 to 12) for response and 13 (95% CI 8 to 29) for remission (assuming a baseline response and remission rate of 356/1000 for response and 224/1000 for remission; see summary of findings Table for the main comparison). Thus, doctors would need to treat 8 (95% CI 5 to 12) people with vortioxetine, rather than placebo, for one additional responding patient and 13 people (95% CI 8 to 29) for one additional remitting patient. Furthermore, the difference in change scores between vortioxetine and placebo was estimated to be fewer than 3 points on the MADRS at study endpoint. Some authors suggested that a difference of 2 points on HAM‐D or MADRS is already clinically relevant (Montgomery 2009), but others have shown that an improvement of up to 3 points on the HAM‐D‐17 corresponded to "no change" on CGI (Leucht 2013). Clearly, this casts uncertainty on the real added value of vortioxetine. We found no statistically significant difference in terms of total dropout rates, although more participants discontinued vortioxetine because of adverse effects, while significantly more participants discontinued placebo because of inefficacy. The subgroup or sensitivity analysis revealed no factors that significantly influenced the results.
In comparison with other antidepressants, very low quality of evidence found no clinically significant advantage in efficacy of vortioxetine over the SNRIs as a class (see summary of findings Table 2). Against individual antidepressants, the analyses of response rates and change in depressive symptoms scores suggested that vortioxetine may be less effective than duloxetine, although in terms of remission rates, there was no difference. Against venlafaxine, meta‐analysis of two studies found no statistically significant differences. In terms of tolerability, our analyses of total dropout rates and dropouts due to adverse events or inefficacy found no significant differences between vortioxetine and the SNRIs as a class. In terms of number of participants reporting at least one adverse effect, vortioxetine was better than the SNRIs as a class and duloxetine. However, the sensitivity analysis casts some doubts on this result, as only two studies used comparable dosing.
Overall completeness and applicability of evidence
All studies included were conducted in highly selected populations, excluding, for example, people with psychiatric comorbidities and suicidal ideation, which are commonly seen in everyday practice. Thus, the external validity of our results may be limited. Furthermore, we were unable to identify studies that compared vortioxetine to pharmacological classes other than SNRI, which is a major limitation in terms of applicability of the evidence in routine care, as the SSRIs are the most commonly prescribed antidepressants as first‐line treatment.
In accordance with the protocol, we did not conduct meta‐analysis on adverse events and did not include scales assessing accompanying symptoms, although these aspects may be of paramount relevance in clinical practice, often guiding the selection of an antidepressant (Gartlehner 2012). Our list of adverse events shows that there is considerable variance in naming, grouping and reporting adverse events; similarly, functional outcomes or accompanying symptoms were erratically assessed and reported. Thus, meta‐analyses of these outcomes cannot be informative and may rather be misleading in informing clinical practice.
Quality of the evidence
We chose seven outcomes for assessing the quality of evidence and to construct the 'Summary of findings' tables (the chosen outcomes being: response, total number of dropouts, remission, depressive symptoms, dropouts due to adverse events, dropouts due to inefficacy, tolerability expressed as patient experiencing as least one adverse event). We constructed two separate 'Summary of findings' tables: one for vortioxetine compared to placebo (summary of findings Table for the main comparison), and one for vortioxetine compared to SNRIs (summary of findings Table 2).
The quality of the evidence for the comparisons of vortioxetine and placebo was low for remission and response rates and very low for the analysis of depressive symptoms, showing uncertainty in the magnitude of effect. The quality was downgraded because of high dropout rates in some of the trials or because of statistical heterogeneity. We rated the quality of the evidence with respect to treatment discontinuation as moderate for all three outcomes in studies of vortioxetine versus placebo (summary of findings Table for the main comparison).
For studies comparing vortioxetine versus the SNRIs, the quality of the evidence was very low for all outcomes (summary of findings Table 2). The quality was downgraded because of high dropout rates and substantial statistical heterogeneity.
Study limitations (risk of bias)
All studies employed proper double‐blind procedures and adequate masking of the outcome assessment. However, we judged that all outcomes in studies comparing vortioxetine versus placebo had a serious risk of bias, as about 30% of studies showed an overall dropout rate above 20%. We downgraded the quality of evidence by one level accordingly. In studies comparing vortioxetine versus the SNRIs, the risk of bias was very serious for all outcomes, as about 60% of studies had a dropout rate above 20%.
Consistency of effect
In studies comparing vortioxetine versus placebo, there was a moderate degree of heterogeneity (I2 = 30% to 60%) (Schünemann 2013) for the outcomes 'response' and 'remission.' The quality of evidence was downgraded by one level. There was a substantial degree of heterogeneity (I2 = 60% to 90%) for the outcome 'depressive symptoms.' For this reason, we downgraded the quality of evidence by two levels for these outcomes. In the comparison vortioxetine versus SNRIs, there was a moderate degree of heterogeneity for four outcomes, namely 'response,' 'remission,' 'depressive symptoms' and 'dropouts due to adverse events.' We downgraded the quality of evidence by one level. Reasons underpinning this statistical heterogeneity are uncertain. In general, included studies were homogeneous in terms of clinical features of participants, as the vast majority enrolled participants with the same diagnosis and within the same range of age. Also trial duration was very similar between studies, while the characteristics of the setting differed between studies (see Included studies).
Indirectness
We judged that there was no serious risk of indirectness, as the diagnostic criteria, the characteristics of participants, type of interventions and outcomes largely matched those of interest. Most studies were conducted in high‐ and middle‐income countries, especially in the US. Evidence for people from large parts of the world, for example, Africa, South America and the Middle East was lacking (see Figure 2).
Imprecision
For studies comparing vortioxetine versus placebo, there was no serious risk of imprecision, because of the relatively large number of participants and events. However, for studies comparing vortioxetine versus SNRIs, we downgraded the outcomes 'total dropouts' and 'dropouts due to adverse events' by one level for serious risk of imprecision, as the 95% CI crossed both 1 (no differences) and 0.75 (appreciable benefit for vortioxetine) (according to Guyatt 2011). The outcome 'dropouts due to inefficacy' was downgraded by two levels for very serious risk of imprecision, as the 95% CI crossed 1 (no differences), 0.75 (appreciable benefit for vortioxetine) and 1.25 (appreciable benefit for SNRIs).
Publication bias
Visual inspection of the funnel plot did not suggest a relevant risk of publication bias. Our review included some unpublished studies and this may have reduced the impact of publication bias. As experts in the field were contacted during the search phase, it is unlikely that relevant studies were overlooked. However, although the search was thorough, there is still the possibility of not having identified unpublished studies, considering that standardised procedures to perform this type of search are lacking (Chan 2012), and that all studies were sponsored by the pharmaceutical company.
Potential biases in the review process
Although we judged our funnel plots to be inconclusive and we cannot rule out that we missed studies, we consider that our study sample is rather comprehensive and less prone to publication bias in comparison to, for example, reviews of agomelatine (Guaiana 2013; Koesters 2013). We included two unpublished trials only, but, for example, the failed trial and the three negative trials listed in the FDA medical review (CDER 2013) have been published and are included in our review (Baldwin 2012; Jain 2013; Mahableshwarkar 2013; Mahableshwarkar 2015c). However, there is some evidence for a sponsorship bias in antidepressant research favouring the experimental drug (Barbui 2004; Weinmann 2008). Our analysis contained only trials conducted by the manufacturers, so we cannot rule out a sponsorship bias in favour of vortioxetine in our data. Furthermore, our analysis combined different doses of vortioxetine, which may have increased heterogeneity of the effects. However, considering that previous reviews found no significant dose‐response effects (Meeker 2015; Thase 2016), this approach may be justified. Furthermore, doses may also vary considerably across participants within study arms and it therefore may be necessary to review these effects based on individual participant data.
Agreements and disagreements with other studies or reviews
Our review is in line with previous reviews of vortioxetine (e.g. Meeker 2015; Thase 2016; Zhang 2015) showing that vortioxetine has a statistically significant advantage compared to placebo in terms of efficacy, but did not show an advantage compared to SNRIs. However, considering the quality of the evidence and other potential sources of bias, we suggest caution in the interpretation of efficacy data. This cautious interpretation is in line with findings from previous reviews, for example by not finding a clear dose‐response relationship (Meeker 2015; Thase 2016), and the noteworthy finding that only high doses (20 mg/day) of vortioxetine were significantly superior to placebo in US populations (Thase 2016; Zhang 2015). Furthermore, the findings that duloxetine was more effective than vortioxetine, and that there was no difference in terms of efficacy between vortioxetine and venlafaxine, are inconsistent with the evidence showing the superior efficacy of venlafaxine over duloxetine (Cipriani 2009; Schueler 2011). We are aware of the limitations of such an indirect comparison, but these inconsistencies raise some questions about the validity of the findings.
Study flow diagram.
Countries participating in trials. The categories represent the number of studies randomising participants within a country.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Forest plot of comparison: 1 Vortioxetine versus placebo, outcome: 1.1 Response.
Forest plot of comparison: 1 Vortioxetine versus placebo, outcome: 1.2 Total number of dropouts.
Forest plot of comparison: 3. Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors, outcome: 3.1 Response.
Forest plot of comparison: 3. Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors, outcome: 3.2 Total number of dropouts.
Comparison 1 Vortioxetine versus placebo, Outcome 1 Response.
Comparison 1 Vortioxetine versus placebo, Outcome 2 Total number of dropouts.
Comparison 1 Vortioxetine versus placebo, Outcome 3 Remission.
Comparison 1 Vortioxetine versus placebo, Outcome 4 Depressive symptoms.
Comparison 1 Vortioxetine versus placebo, Outcome 5 Dropout due to adverse events.
Comparison 1 Vortioxetine versus placebo, Outcome 6 Dropout due to inefficacy.
Comparison 1 Vortioxetine versus placebo, Outcome 7 Tolerability.
Comparison 1 Vortioxetine versus placebo, Outcome 8 Subgroup analysis: fixed vs flexible dosing ‐ response.
Comparison 1 Vortioxetine versus placebo, Outcome 9 Subgroup analysis: fixed vs flexible dosing ‐ total number of dropouts.
Comparison 1 Vortioxetine versus placebo, Outcome 10 Subgroup analysis: inclusion of older (aged > 65 years) participants ‐ response.
Comparison 1 Vortioxetine versus placebo, Outcome 11 Subgroup analysis: inclusion of older (aged > 65 years) participants ‐ total number of dropouts.
Comparison 1 Vortioxetine versus placebo, Outcome 12 Sensitivity analysis ‐ exclusion > 20% dropouts ‐ response.
Comparison 1 Vortioxetine versus placebo, Outcome 13 Sensitivity analysis ‐ exclusion > 20% dropouts ‐ total number of dropouts.
Comparison 1 Vortioxetine versus placebo, Outcome 14 Adverse events.
Comparison 1 Vortioxetine versus placebo, Outcome 15 Serious adverse events.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 1 Response.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 2 Total number of dropouts.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 3 Remission.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 4 Depressive symptoms.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 5 Dropout due to adverse events.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 6 Dropout due to inefficacy.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 7 Tolerability.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 8 Subgroup analysis: fixed vs flexible dosing ‐ response.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 9 Subgroup analysis: fixed vs flexible dosing ‐ total number of dropouts.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 10 Subgroup analysis: inclusion of older (aged > 65 years) participants ‐ response.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 11 Subgroup analysis: inclusion of older (aged > 65 years) participants ‐ total number of dropouts.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 12 Sensitivity analysis ‐ unequal dosing ‐ response.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 13 Sensitivity analysis ‐ unequal dosing ‐ total number of dropouts.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 14 Sensitivity analysis ‐ exclusion > 20% dropouts ‐ response.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 15 Sensitivity analysis ‐ exclusion > 20% dropouts ‐ total number of dropouts.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 16 Adverse events.
Comparison 2 Vortioxetine versus serotonin‐norepinephrine reuptake inhibitors (SNRIs), Outcome 17 Serious adverse events.
| Vortioxetine compared to Placebo for adults with Major Depressive Disorder | ||||||
| Patient or population: adults with Major Depressive Disorder | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Quality of the evidence | Comments | |
| Risk with Placebo | Risk with Vortioxetine | |||||
| Response | Study population | RR 1.35 | 6220 | ⊕⊕⊝⊝ | All studies were sponsored by the pharmaceutical companies that manufacture vortioxetine. Small difference favouring vortioxetine. | |
| 356 per 1,000 | 480 per 1,000 | |||||
| Total number of drop‐outs | Study population | RR 1.05 | 6220 | ⊕⊕⊕⊝ | No difference between vortioxetine and placebo. | |
| 160 per 1,000 | 168 per 1,000 | |||||
| Remission | Study population | RR 1.33 | 6217 | ⊕⊕⊝⊝ | Small difference favouring vortioxetine. | |
| 224 per 1,000 | 299 per 1,000 | |||||
| Depressive Symptoms | The change in depressive symptoms score ranged from 10.8 to 15.9 points | The change was 2.94 points higher | ‐ | 5566 | ⊕⊝⊝⊝ | Small difference favouring vortioxetine. |
| Drop‐out due to adverse events | Study population | RR 1.41 | 6220 | ⊕⊕⊕⊝ | Small difference favouring placebo. | |
| 38 per 1,000 | 53 per 1,000 | |||||
| Drop‐out due to inefficacy | Study population | RR 0.56 | 6220 | ⊕⊕⊕⊝ | Small difference favouring vortioxetine. | |
| 31 per 1,000 | 18 per 1,000 | |||||
| Tolerability | Study population | RR 1.12 | 6182 | ⊕⊕⊕⊝ | Small difference favouring placebo | |
| 564 per 1,000 | 632 per 1,000 | |||||
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1 A serious risk of bias is present, as about 30% of the studies showed an overall dropout rate above 20%, evidence was downgraded by one level 2 A moderate degree of heterogeneity (I‐squared 30‐60%) is present, evidence was downgraded by one level 3 A substantial degree of heterogeneity (I‐squared 60‐90%) is present, evidence was downgraded by two levels | ||||||
| Vortioxetine compared to SNRIs for adults with Major Depressive Disorder | ||||||
| Patient or population: adults with Major Depressive Disorder | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Quality of the evidence | Comments | |
| Risk with SNRIs | Risk with Vortioxetine | |||||
| Response | Study population | RR 0.91 | 3159 | ⊕⊝⊝⊝ | All studies were sponsored by the pharmaceutical companies that manufacture vortioxetine. No difference between vortioxetine and SNRIs. | |
| 577 per 1,000 | 525 per 1,000 | |||||
| Total number of drop‐outs | Study population | RR 0.89 | 3159 | ⊕⊝⊝⊝ | No difference between vortioxetine and SNRIs. | |
| 212 per 1,000 | 189 per 1,000 | |||||
| Remission | Study population | RR 0.89 | 3155 | ⊕⊝⊝⊝ | No difference between vortioxetine and SNRIs. | |
| 370 per 1,000 | 329 per 1,000 | |||||
| Depressive Symptoms | The change in depressive symptoms score ranged from 14.1 to 23.4 points | The change was 1.52 points lower | ‐ | 2807 | ⊕⊝⊝⊝ | Small difference favouring SNRIs. |
| Drop‐out due to adverse events | Study population | RR 0.74 | 3159 | ⊕⊝⊝⊝ | No difference between vortioxetine and SNRIs. | |
| 97 per 1,000 | 72 per 1,000 | |||||
| Drop‐out due to inefficacy | Study population | RR 1.52 | 3159 | ⊕⊝⊝⊝ | No difference between vortioxetine and SNRIs. | |
| 14 per 1,000 | 21 per 1,000 | |||||
| Tolerability | Study population | RR 0.90 | 3134 | ⊕⊕⊝⊝ | Small difference favouring vortioxetine | |
| 690 per 1,000 | 621 per 1,000 | |||||
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1 A very serious risk of bias is present as 60% of the studies had more than 20% dropouts overall, evidence was downgraded by two levels 2 A moderate degree of heterogeneity (I‐squared 30‐60%) is present, evidence was downgraded by one level 3 The 95% CI crossed both 1 (no differences) and 0.75 (appreciable benefit for vortioxetine), evidence was downgraded by one level 4 The 95% CI crossed 1 (no differences), 0.75 (appreciable benefit for vortioxetine) and 1.25 (appreciable benefit for SNRIs). Outcome is very imprecise: evidence was downgraded by two levels | ||||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Response Show forest plot | 14 | 6220 | Risk Ratio (M‐H, Random, 95% CI) | 1.35 [1.22, 1.49] |
| 2 Total number of dropouts Show forest plot | 14 | 6220 | Risk Ratio (M‐H, Random, 95% CI) | 1.05 [0.93, 1.19] |
| 3 Remission Show forest plot | 14 | 6220 | Risk Ratio (M‐H, Random, 95% CI) | 1.32 [1.15, 1.53] |
| 4 Depressive symptoms Show forest plot | 14 | 5566 | Mean Difference (IV, Random, 95% CI) | ‐2.94 [‐4.07, ‐1.80] |
| 5 Dropout due to adverse events Show forest plot | 14 | 6220 | Risk Ratio (M‐H, Random, 95% CI) | 1.41 [1.09, 1.81] |
| 6 Dropout due to inefficacy Show forest plot | 14 | 6220 | Risk Ratio (M‐H, Random, 95% CI) | 0.56 [0.34, 0.90] |
| 7 Tolerability Show forest plot | 14 | 6182 | Risk Ratio (M‐H, Random, 95% CI) | 1.12 [1.07, 1.16] |
| 8 Subgroup analysis: fixed vs flexible dosing ‐ response Show forest plot | 14 | 6220 | Risk Ratio (M‐H, Random, 95% CI) | 1.35 [1.22, 1.49] |
| 8.1 Fixed dose | 12 | 5513 | Risk Ratio (M‐H, Random, 95% CI) | 1.35 [1.19, 1.52] |
| 8.2 Flexible dose | 2 | 707 | Risk Ratio (M‐H, Random, 95% CI) | 1.38 [1.16, 1.63] |
| 9 Subgroup analysis: fixed vs flexible dosing ‐ total number of dropouts Show forest plot | 14 | 6220 | Risk Ratio (M‐H, Random, 95% CI) | 1.05 [0.93, 1.19] |
| 9.1 Fixed dose | 12 | 5513 | Risk Ratio (M‐H, Random, 95% CI) | 1.07 [0.94, 1.22] |
| 9.2 Flexible dose | 2 | 707 | Risk Ratio (M‐H, Random, 95% CI) | 0.91 [0.64, 1.30] |
| 10 Subgroup analysis: inclusion of older (aged > 65 years) participants ‐ response Show forest plot | 10 | 4525 | Risk Ratio (M‐H, Random, 95% CI) | 1.38 [1.24, 1.54] |
| 10.1 Older participants included | 6 | 2616 | Risk Ratio (M‐H, Random, 95% CI) | 1.34 [1.15, 1.55] |
| 10.2 Older participants excluded | 4 | 1909 | Risk Ratio (M‐H, Random, 95% CI) | 1.45 [1.23, 1.71] |
| 11 Subgroup analysis: inclusion of older (aged > 65 years) participants ‐ total number of dropouts Show forest plot | 10 | 4528 | Risk Ratio (M‐H, Random, 95% CI) | 1.11 [0.96, 1.28] |
| 11.1 Older participants included | 6 | 2619 | Risk Ratio (M‐H, Random, 95% CI) | 1.25 [1.05, 1.49] |
| 11.2 Older participants excluded | 4 | 1909 | Risk Ratio (M‐H, Random, 95% CI) | 0.90 [0.71, 1.13] |
| 12 Sensitivity analysis ‐ exclusion > 20% dropouts ‐ response Show forest plot | 14 | 6220 | Risk Ratio (M‐H, Random, 95% CI) | 1.35 [1.22, 1.49] |
| 12.1 < 20% dropouts | 9 | 3927 | Risk Ratio (M‐H, Random, 95% CI) | 1.41 [1.26, 1.57] |
| 12.2 > 20% dropouts | 5 | 2293 | Risk Ratio (M‐H, Random, 95% CI) | 1.26 [1.03, 1.52] |
| 13 Sensitivity analysis ‐ exclusion > 20% dropouts ‐ total number of dropouts Show forest plot | 14 | 6220 | Risk Ratio (M‐H, Random, 95% CI) | 1.05 [0.93, 1.19] |
| 13.1 < 20% dropout | 9 | 3927 | Risk Ratio (M‐H, Random, 95% CI) | 1.02 [0.86, 1.21] |
| 13.2 > 20% dropout | 5 | 2293 | Risk Ratio (M‐H, Random, 95% CI) | 1.08 [0.91, 1.28] |
| 14 Adverse events Show forest plot | 14 | Risk Ratio (M‐H, Random, 95% CI) | Totals not selected | |
| 14.1 Constipation | 11 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.2 Nausea | 14 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.3 Diarrhoea | 13 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.4 Dry mouth | 12 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.5 Vomiting | 9 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.6 Abdominal pain upper | 6 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.7 Dyspepsia | 7 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.8 Gastro‐oesophageal reflux disease | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.9 Abdominal discomfort | 4 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.10 Abdominal pain | 3 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.11 Stomach discomfort | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.12 Flatulence | 5 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.13 Fatigue | 12 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.14 Pain | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.15 Thirst | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.16 Irritability | 5 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.17 Decreased appetite | 6 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.18 Increased appetite | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.19 Anorexia | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.20 Headache | 14 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.21 Dizziness | 13 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.22 Somnolence | 10 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.23 Tremor | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.24 Sedation | 4 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.25 Dysgeusia | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.26 Hypoaesthesia | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.27 Poor‐quality sleep | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.28 Ejaculation delayed (men) | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.29 Erectile dysfunction (men) | 3 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.30 Dysmenorrhoea | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.31 Hyperhidrosis | 9 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.32 Pruritus generalised | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.33 Pruritus | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.34 Vision blurred | 5 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.35 Nasopharyngitis | 11 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.36 Influenza | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.37 Respiratory tract infection | 6 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.38 Gastroenteritis viral | 3 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.39 Sinusitis | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.40 Urinary tract infection | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.41 Bronchitis | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.42 Anorgasmia | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.43 Insomnia | 9 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.44 Restlessness | 3 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.45 Abnormal dreams | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.46 Anxiety | 4 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.47 Depression | 3 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.48 Libido decreased | 3 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.49 Orgasm abnormal | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.50 Nightmare | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.51 Middle insomnia | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.52 Suicidal ideation | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.53 Tachycardia | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.54 Palpitations | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.55 Tinnitus | 3 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.56 Back pain | 7 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.57 Arthralgia | 4 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.58 Myalgia | 3 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.59 Musculoskeletal pain | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.60 Muscle spasms | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.61 Pain in extremity | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.62 Hypertension | 4 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.63 Hot flush | 3 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.64 Fall | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.65 Ligament sprain | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.66 Muscle strain | 3 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.67 Accidental overdose | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.68 Road traffic accident | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.69 Nasal congestion | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.70 Yawning | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.71 Rhinorrhoea | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.72 Oropharyngeal pain | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.73 Weight decreased | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.74 Blood pressure increased | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 14.75 Hepatic function abnormal | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15 Serious adverse events Show forest plot | 14 | Risk Ratio (M‐H, Random, 95% CI) | Totals not selected | |
| 15.1 Varicella zoster infection | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.2 Kidney infection | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.3 Herpes zoster infection | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.4 Puncture site infection | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.5 Gastroenteritis | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.6 Pyelonephritis | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.7 Brain tumour | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.8 Gallbladder cancer | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.9 Colon cancer | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.10 Laryngeal cancer | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.11 Renal cell carcinoma | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.12 Bile duct cancer | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.13 Prostate cancer | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.14 Breast cancer recurrent | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.15 Worsening of major depressive disorder | 6 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.16 Suicidal ideation | 3 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.17 Suicide attempt | 5 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.18 Intentional self‐injury | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.19 Self‐injurious behaviour | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.20 Panic attack | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.21 Suicidal behaviour | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.22 Middle ear effusion | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.23 Jaundice cholestatic | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.24 Cholecystitis | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.25 Pelvic fracture | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.26 Intentional overdose | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.27 Lumbar vertebral fracture | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.28 Injury | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.29 Hip fracture | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.30 Head injury | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.31 Stress fracture | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.32 Craniocerebral injury | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.33 Subdural haematoma | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.34 Brain contusion | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.35 Road traffic accident | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.36 Serotonin syndrome | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.37 Dizziness | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.38 Cerebrovascular accident | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.39 Convulsion | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.40 Transient ischaemic attack | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.41 Lumbar radiculopathy | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.42 Syncope | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.43 Cerebral haematoma | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.44 Subarachnoid haemorrhage | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.45 Adenomyosis | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.46 Vaginal haemorrhage | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.47 Pulmonary embolism | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.48 Blood pressure decreased | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.49 Tachycardia | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.50 Acute myocardial infarction | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.51 Atrial fibrillation | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.52 Coronary artery disease | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.53 Pancreatitis | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.54 Hiatus hernia | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.55 Drug hypersensitivity | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.56 Abortion spontaneous | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.57 Ectopic pregnancy | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.58 Abortion missed | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.59 Abortion induced | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.60 Type 1 diabetes mellitus | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.61 Hypertension | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 15.62 Renal colic | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Response Show forest plot | 8 | 3159 | Risk Ratio (M‐H, Random, 95% CI) | 0.91 [0.82, 1.00] |
| 1.1 Vortioxetine vs venlafaxine | 2 | 767 | Risk Ratio (M‐H, Random, 95% CI) | 1.03 [0.85, 1.25] |
| 1.2 Vortioxetine vs duloxetine | 6 | 2392 | Risk Ratio (M‐H, Random, 95% CI) | 0.86 [0.79, 0.94] |
| 2 Total number of dropouts Show forest plot | 8 | 3159 | Risk Ratio (M‐H, Random, 95% CI) | 0.89 [0.73, 1.08] |
| 2.1 Vortioxetine vs venlafaxine | 2 | 767 | Risk Ratio (M‐H, Random, 95% CI) | 0.70 [0.52, 0.93] |
| 2.2 Vortioxetine vs duloxetine | 6 | 2392 | Risk Ratio (M‐H, Random, 95% CI) | 0.96 [0.76, 1.21] |
| 3 Remission Show forest plot | 8 | 3155 | Risk Ratio (M‐H, Random, 95% CI) | 0.89 [0.77, 1.03] |
| 3.1 Vortioxetine vs venlafaxine | 2 | 767 | Risk Ratio (M‐H, Random, 95% CI) | 0.99 [0.81, 1.20] |
| 3.2 Vortioxetine vs duloxetine | 6 | 2388 | Risk Ratio (M‐H, Random, 95% CI) | 0.85 [0.70, 1.02] |
| 4 Depressive symptoms Show forest plot | 8 | 2807 | Mean Difference (IV, Random, 95% CI) | 1.52 [0.50, 2.53] |
| 4.1 Vortioxetine vs venlafaxine | 2 | 701 | Mean Difference (IV, Random, 95% CI) | 0.02 [‐2.49, 2.54] |
| 4.2 Vortioxetine vs duloxetine | 6 | 2106 | Mean Difference (IV, Random, 95% CI) | 1.99 [1.15, 2.83] |
| 5 Dropout due to adverse events Show forest plot | 8 | 3159 | Risk Ratio (M‐H, Random, 95% CI) | 0.74 [0.51, 1.08] |
| 5.1 Vortioxetine vs venlafaxine | 2 | 767 | Risk Ratio (M‐H, Random, 95% CI) | 0.42 [0.26, 0.67] |
| 5.2 Vortioxetine vs duloxetine | 6 | 2392 | Risk Ratio (M‐H, Random, 95% CI) | 0.92 [0.65, 1.31] |
| 6 Dropout due to inefficacy Show forest plot | 8 | 3159 | Risk Ratio (M‐H, Random, 95% CI) | 1.52 [0.70, 3.30] |
| 6.1 Vortioxetine vs venlafaxine | 2 | 767 | Risk Ratio (M‐H, Random, 95% CI) | 2.68 [0.99, 7.24] |
| 6.2 Vortioxetine vs duloxetine | 6 | 2392 | Risk Ratio (M‐H, Random, 95% CI) | 1.16 [0.41, 3.31] |
| 7 Tolerability Show forest plot | 8 | 3134 | Risk Ratio (M‐H, Random, 95% CI) | 0.90 [0.86, 0.94] |
| 7.1 Vortioxetine vs venlafaxine | 2 | 758 | Risk Ratio (M‐H, Random, 95% CI) | 0.91 [0.82, 1.00] |
| 7.2 Vortioxetine vs duloxetine | 6 | 2376 | Risk Ratio (M‐H, Random, 95% CI) | 0.89 [0.84, 0.95] |
| 8 Subgroup analysis: fixed vs flexible dosing ‐ response Show forest plot | 8 | 3159 | Risk Ratio (M‐H, Random, 95% CI) | 0.91 [0.82, 1.00] |
| 8.1 Fixed dosing | 7 | 2751 | Risk Ratio (M‐H, Random, 95% CI) | 0.91 [0.81, 1.01] |
| 8.2 Flexible dosing | 1 | 408 | Risk Ratio (M‐H, Random, 95% CI) | 0.93 [0.75, 1.14] |
| 9 Subgroup analysis: fixed vs flexible dosing ‐ total number of dropouts Show forest plot | 8 | 3159 | Risk Ratio (M‐H, Random, 95% CI) | 0.89 [0.73, 1.08] |
| 9.1 Fixed dosing | 7 | 2751 | Risk Ratio (M‐H, Random, 95% CI) | 0.89 [0.71, 1.11] |
| 9.2 Flexible dosing | 1 | 408 | Risk Ratio (M‐H, Random, 95% CI) | 0.94 [0.60, 1.47] |
| 10 Subgroup analysis: inclusion of older (aged > 65 years) participants ‐ response Show forest plot | 7 | 2850 | Risk Ratio (M‐H, Random, 95% CI) | 0.92 [0.83, 1.03] |
| 10.1 Older participants included | 4 | 1675 | Risk Ratio (M‐H, Random, 95% CI) | 0.87 [0.77, 0.98] |
| 10.2 Older participants excluded | 3 | 1175 | Risk Ratio (M‐H, Random, 95% CI) | 1.00 [0.88, 1.15] |
| 11 Subgroup analysis: inclusion of older (aged > 65 years) participants ‐ total number of dropouts Show forest plot | 7 | 2854 | Risk Ratio (M‐H, Random, 95% CI) | 0.92 [0.74, 1.15] |
| 11.1 Older participants included | 4 | 1679 | Risk Ratio (M‐H, Random, 95% CI) | 1.05 [0.76, 1.45] |
| 11.2 Older participants excluded | 3 | 1175 | Risk Ratio (M‐H, Random, 95% CI) | 0.76 [0.60, 0.97] |
| 12 Sensitivity analysis ‐ unequal dosing ‐ response Show forest plot | 8 | 3159 | Risk Ratio (M‐H, Random, 95% CI) | 0.91 [0.82, 1.00] |
| 12.1 Equal dosing | 2 | 912 | Risk Ratio (M‐H, Random, 95% CI) | 1.09 [0.97, 1.22] |
| 12.2 Vortioxetine dose higher | 2 | 903 | Risk Ratio (M‐H, Random, 95% CI) | 0.80 [0.72, 0.90] |
| 12.3 Vortioxetine dose lower | 3 | 936 | Risk Ratio (M‐H, Random, 95% CI) | 0.87 [0.78, 0.98] |
| 12.4 Flexible vs fixed dosing | 1 | 408 | Risk Ratio (M‐H, Random, 95% CI) | 0.93 [0.75, 1.14] |
| 13 Sensitivity analysis ‐ unequal dosing ‐ total number of dropouts Show forest plot | 8 | 3159 | Risk Ratio (M‐H, Random, 95% CI) | 0.89 [0.73, 1.08] |
| 13.1 Equal dosing | 2 | 912 | Risk Ratio (M‐H, Random, 95% CI) | 0.75 [0.59, 0.96] |
| 13.2 Vortioxetine dose higher | 2 | 903 | Risk Ratio (M‐H, Random, 95% CI) | 1.33 [0.74, 2.39] |
| 13.3 Vortioxetine dose lower | 3 | 936 | Risk Ratio (M‐H, Random, 95% CI) | 0.77 [0.59, 1.02] |
| 13.4 Flexible vs fixed dosing | 1 | 408 | Risk Ratio (M‐H, Random, 95% CI) | 0.94 [0.60, 1.47] |
| 14 Sensitivity analysis ‐ exclusion > 20% dropouts ‐ response Show forest plot | 8 | 3159 | Risk Ratio (M‐H, Random, 95% CI) | 0.91 [0.82, 1.00] |
| 14.1 < 20% dropouts | 3 | 1039 | Risk Ratio (M‐H, Random, 95% CI) | 0.91 [0.82, 1.01] |
| 14.2 > 20% dropouts | 5 | 2120 | Risk Ratio (M‐H, Random, 95% CI) | 0.90 [0.77, 1.07] |
| 15 Sensitivity analysis ‐ exclusion > 20% dropouts ‐ total number of dropouts Show forest plot | 8 | 3159 | Risk Ratio (M‐H, Random, 95% CI) | 0.89 [0.73, 1.08] |
| 15.1 < 20% dropouts | 3 | 1039 | Risk Ratio (M‐H, Random, 95% CI) | 0.86 [0.64, 1.14] |
| 15.2 > 20% dropouts | 5 | 2120 | Risk Ratio (M‐H, Random, 95% CI) | 0.92 [0.68, 1.24] |
| 16 Adverse events Show forest plot | 8 | Risk Ratio (M‐H, Random, 95% CI) | Totals not selected | |
| 16.1 Constipation | 6 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.2 Diarrhoea | 7 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.3 Dry mouth | 8 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.4 Nausea | 8 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.5 Vomiting | 4 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.6 Abdominal pain upper | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.7 Dyspepsia | 3 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.8 Abdominal discomfort | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.9 Abdominal pain | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.10 Stomach discomfort | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.11 Flatulence | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.12 Fatigue | 6 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.13 Irritability | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.14 Decreased appetite | 6 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.15 Anorexia | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.16 Dizziness | 8 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.17 Headache | 8 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.18 Somnolence | 4 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.19 Tremor | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.20 Sedation | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.21 Dysgeusia | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.22 Poor quality sleep | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.23 Ejaculation delayed (men) | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.24 Erectile dysfunction (men) | 3 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.25 Hyperhidrosis | 6 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.26 Vision blurred | 3 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.27 Nasopharyngitis | 4 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.28 Upper respiratory tract infection | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.29 Gastroenteritis viral | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.30 Urinary tract infection bacterial | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.31 Anorgasmia | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.32 Insomnia | 5 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.33 Restlessness | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.34 Abnormal dreams | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.35 Anxiety | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.36 Depression | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.37 Libido decreased | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.38 Orgasm abnormal | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.39 Nightmare | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.40 Middle insomnia | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.41 Palpitations | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.42 Tinnitus | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.43 Back pain | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.44 Arthralgia | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.45 Myalgia | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.46 Musculoskeletal pain | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.47 Muscle spasms | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.48 Pain in extremity | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.49 Hot flush | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.50 Fall | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.51 Muscle strain | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.52 Accidental overdose | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.53 Yawning | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.54 Rhinorrhoea | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.55 Weight decreased | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.56 Blood pressure increased | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 16.57 Heart rate increased | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17 Serious adverse events Show forest plot | 8 | Risk Ratio (M‐H, Random, 95% CI) | Totals not selected | |
| 17.1 Varicella zoster infection | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.2 Brain tumour | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.3 Gallbladder cancer | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.4 Bile duct cancer | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.5 Prostate cancer | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.6 Worsening of major depressive disorder | 6 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.7 Suicidal ideation | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.8 Suicide attempt | 3 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.9 Intentional self‐injury | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.10 Self‐injurious behaviour | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.11 Panic attack | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.12 Anxiety | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.13 Middle ear effusion | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.14 Vertigo positional | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.15 Jaundice cholestatic | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.16 Pelvic fracture | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.17 Intentional overdose | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.18 Lumbar vertebral fracture | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.19 Hip fracture | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.20 Head injury | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.21 Stress fracture | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.22 Craniocerebral injury | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.23 Subdural haematoma | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.24 Asthenia | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.25 Serotonin syndrome | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.26 Dizziness | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.27 Convulsion | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.28 Transient ischaemic attack | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.29 Adenomyosis | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.30 Vaginal haemorrhage | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.31 Varicocele | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.32 Pulmonary embolism | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.33 Blood pressure decreased | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.34 Acute myocardial infarction | 2 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.35 Atrial fibrillation | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.36 Coronary artery disease | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.37 Abortion induced | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 17.38 Ligament sprain | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
