Characteristics of the included trials and participants
| Included studies | Diagnostic criteria | No. of participants | Mean age (years) | Female No (%) | Duration (weeks) | Intervention | Outcome measures | Sponsor type |
| NSAID | ||||||||
| Müller 200610 | DSM-IV | 40 | 44.4 | 20 (50) | 6 | Reboxetine 4–10 mg/d+Ccelecoxib 400 mg/d vs Seboxetine 4–10 mg/d+placebo | Scores at baseline and endpoint (HAMD-17) Remission (HAMD ≤7 in endpoint) | CI |
| Akhondzadeh 200911 | DSM-IV HAMD-17≥18 with item 1≥2 | 40 | 34.43 | 25 (62.5) | 6 | Fluoxetine 20–40 mg/d+Celecoxib 400 mg/d vs Fluoxetine 20–40 mg/d+Placebo | Score changes from baseline to endpoint (HAMD-17) Remission (HAMD ≤7 in endpoint) | NPO |
| Abbasi 201212 | DSM-IV-TR HAMD-17≥18 with item 1≥2 | 40 | 34.65 | 13 (32.5) | 6 | Sertraline 200 mg/d+Celecoxib 400 mg /d vs Sertraline 200 mg/d+Placebo | Score changes from baseline to endpoint (HAMD-17) Response (≥50% reduction in HAMD) Remission (HAMD ≤7 in endpoint) | NPO |
| Majd 201513 | DSM-IV-TR HAMD-17 from 18 to 36 | 32 | 35.29 | 32 (100) | 8 | Sertraline 200 mg/d+Celecoxib 400 mg/d vs Sertraline 200 mg/d+Placebo | Score changes from baseline to endpoint (HAMD-17) Response (≥50% reduction in HAMD) Remission (HAMD ≤7 in endpoint) | NC |
| Omega-3 FA | ||||||||
| Nemets 200214 | DSM-IV HAMD-24≥18 | 19 | 53.15 | 16 (84.2) | 4 | TAU +EPA 2 g/d vs TAU +Placebo | Score changes from baseline to endpoint (HAMD-24) Response (≥50% reduction in HAMD) Remission (HAMD ≤7 in endpoint) | NC |
| Marangell 200315 | DSM-IV MADRS≥12 HAMD-28 ≥17 | 35 | 47.33 | 28 (80) | 6 | DHA 2 g/d vs Placebo | Score changes from baseline to endpoint (HAMD-28, MADRS) Response (≥50% reduction in MADRS) | CI |
| Su 200316 | DSM-IV HAMD-21 ＞18 | 22 | 38.43 | 18 (82) | 8 | TAU+DPA 4.4 g/d+DHA 2.2 g/d vs TAU+Placebo | Scores at baseline and endpoint (HAMD-21) | CI |
| Freeman 200817 | DSM-IV EPDS ≥9 | 59 | 30.38 | 59 (100) | 8 | EPA 1.1 g/d+DHA 0.8 g/d vs Placebo | Scores at baseline and endpoint (EPDS, HAMD) | NPO |
| Rees 200818 | DSM-IV EPDS ≥13, HAMD-17＞14 or MADRS＞25 | 26 | 32.85 | 26 (100) | 6 | DHA 1.6 g/d+EPA 0.4 g/d vs Placebo | Scores at baseline and endpoint (EPDS, HAMD-17, MADRS) | CI |
| Su 200819 | DSM-IV HAMD-21≥18 | 36 | 31.1 | 36 (100) | 8 | EPA 2.2 g/d+DHA 1.2 g/d vs Placebo | Scores at baseline and endpoint (HAMD, EPDS, BDI) Response (≥50% reduction in HAMD) Remission (HAMD ≤7 in endpoint) | NPO |
| Jazayeri 200820 | DSM-IV HAMD-17≥15 | 32 | 34.8 | 13 (40.6) | 8 | Fluoxetine 20mg+EPA 2 g vs Fluoxetine 20mg+Placebo | Response (≥50% decrease in baseline HAMD) | NPO |
| da Silva 200821 | DSM-IV | 29 | 64.4 | 17 (58.6) | 12 | EPA 0.72 g/d+DHA 0.48 g/d vs Placebo TAU+EPA 0.72 g/d+DHA 0.48 vs TAU+Placebo | Response (≥50% reduction in MADRS scores from baseline to endpoint). | NC |
| Carney 200922 | DSM-IV BDI-II≥16 | 122 | 58.35 | 41 (33.6) | 10 | Sertraline 50 mg/d+Omega-3 2 g/d vs Sertraline 50 mg/d+Placebo | Scores at baseline and endpoint (BDI-II, HAMD-17) Response (≥50% reduction in BDI-II) Remission (BDI-II ≤8 in endpoint) | CI |
| Mischoulon 200923 | DSM-IV HAMD-17≥18 | 24 | 45 | 15 (62.5) | 8 | TAU +EPA 1 g/d vs TAU +Placebo | Scores at baseline and endpoint (HAMD-17) Response (≥50% reduction in HAMD) Remission (HAMD ≤7 in endpoint) | CI |
| Rondanelli 201024 | DSM-IV-TR GDS ＞10 | 46 | 83.91 | 46 (100) | 8 | EPA 1.67 g/d+DHA 0.83 g/d vs Placebo | Scores at baseline and endpoint (GDS) Remission (GDS＜11 in endpoint) | CI |
| Bot 201025 | DSM-IV | 25 | 54.0 | 13 (52) | 12 | TAU +EPA 0.9 g/d vs TAU +Placebo | Scores at baseline and endpoint (MADRS) | CI |
| Gertsik 201226 | DSM-IV HAMD-21 ＞17 | — | 40.5 | — | 9 | Citalopram 20–40 mg/d+EPA 1.8 g/d+DHA 0.4 g/d vs Citalopram 20–40 mg/d+Placebo | Full remission (HAMD score ＜7 in endpoint) | CI |
| Meyer 201327 | DSM-IV HAMD＞16 | 83 | 45.27 | 51 (61.4) | 16 | TUA+EPA 0.6 g/d+DHA 2.2 g/d vs TUA+Placebo | Data not available | CI |
| Mischoulon 201528 | DSM-IV HAMD≥15 CGI-S ≥3 | 177 | 45.8 | 105 (59.3) | 8 | EPA 1.06 g/d+DHA 0.274 g/d vs EPA 0.18 g/d+DHA 0.9 g/d vs Placebo | Score changes from baseline to endpoint (HAMD) Response (≥50% reduction in HAMD) Remission (HAMD ≤7 in endpoint) Quality of life (Q-LES-Q) | CI |
| Rapaport 201629 | DSM-IV GIS≥3 HAMD-17≥15 | 155 | 46.1 | 91 (58.7) | 8 | EPA 1.06 g/d+DHA 0.26 g/d vs EPA 0.18 g/d+DHA 0.9 vs Placebo | Score changes from baseline to endpoint (HAMD-17) Remission (HAMD ≤7 in endpoint) | CI |
| Shinto 201630 | DSM-IV | 39 | 51.25 | 36 (92.3) | 12 | EPA 1.95 g/d+DHA 1.35 g/d vs Placebo | Response rates (≥50% improvement in MADRS) Quality of life (SF-36) | NPO |
| Statin | ||||||||
| Ghanizadeh 201331 | DSM-IV HAMD-17 ＞17 | 60 | 32.1 | 22 (36.7) | 6 | Fluoxetine 40 mg/d+Lovastatin 30 mg/d vs Fluoxetine 40 mg/d+Placebo | Score changes from baseline to endpoint (HAMD-17) | NC |
| Haghighi 201432 | DSM-5 HAMD-17≥25 | 60 | 32.25 | 28 (46.7) | 12 | Citalopram 40 mg/d+Atorvastatin 20 mg/d vs Citalopram 40 mg/d+Placebo | Scores at baseline and endpoint (HAMD-17) Response (＞50% reduction in HAMD) Remission (HAMD ≤7 in endpoint) | NPO |
| Gougol 201533 | DSM-IV HAMD≥22 | 44 | 35.3 | 29 (65.9) | 6 | Fluoxetine 20 mg/d+Simvastatin 20 mg/d vs Fluoxetine 20 mg/d+Placebo | Score changes from baseline to endpoint (HAMD-17) Response (≥50% reduction in HAMD) Remission (HAMD ≤7 in endpoint) | NPO |
| Minocycline | ||||||||
| Emadi-Kouchak 201634 | DSM-IV-TR HAMD up to 18 | 46 | 35.53 | 16 (34.8) | 6 | Minocycline 200 mg/d vs Placebo | Score changes from baseline to endpoint (HAMD-17) | NPO |
| Dean 201735 | DSM-IV MADRS≥25 | 71 | 49.42 | 47 (66.2) | 12 | TAU+Minocycline 200 mg/d vs TAU+Placebo | Scores at baseline and endpoint (MADRS) Response (≥50% reduction in HAMD) Quality of life (Q-LES-Q-SF) | NPO |
| Husain 201736 | DSM-5 | 40 | 35* | 20 (50) | 12 | TAU+Minocycline 200 mg/d vs TAU+Placebo | Score changes from baseline to endpoint (HAM-D −17) Response (≥50% reduction in HAMD) Quality of life (EQ-5D) | NPO |
| Pioglitazone | ||||||||
| Sepanjnia 201237 | DSM-IV-TR HAMD-17≥22 with item 1≥2 | 40 | 32.1 | 29 (72.5) | 6 | Citalopram 20–30 mg/d+Pioglitazone 30 mg/d vs Citalopram 20–30 mg/d+Placebo | Score changes from baseline to endpoint (HAMD-17) Response (≥50% reduction in HAMD) Remission (HAMD ≤7 in endpoint) | NPO |
| Modafinil | ||||||||
| DeBattista 200338 | DSM-IV HAMD-21 14–28 | 136 | 45 | 95 (69.9) | 6 | TAU+Modafinil 100–400 g vs TAU+Placebo | Score changes from baseline to endpoint (HAMD-17) | CI |
| Abolfazli 201139 | DSM-IV-TR HAMD≥18 with item 1≥2 | 44 | 33.2 | 23 (52.3) | 6 | Fluoxetine 40 mg/d+Modafinil 400 mg/d vs Fluoxetine 40 mg/d+Placebo | Score changes from baseline to endpoint (HAMD-17) Response (≥50% reduction in HAMD) Remission (HAMD ≤7 in endpoint) | NPO |
| NAC | ||||||||
| Berk 201440 | DSM-IV-TR MADRS≥18 | 127 125 | 50.2 | 159 (63.1) | 12 | TAU+NAC 2 g/d vs TAU +Placebo | Score changes from baseline to endpoint (MADRS) Response (≥50% reduction in MADRS) Remission (MADR ≤7 in endpoint) Quality of life (Q-LES-Q) | CI |
*The study used a median to describe age.
BDI, Beck’s Depression Inventory;CGI-S, Clinical Global Impression-Severity Scale; CI, commercial industry; DHA, docosahexaenoic acid; DSM, Diagnostic and Statistical Manual for Mental Disorders;EPA, eicosapentanoic acid; EPDS, Edinburgh Postpartum Depression Scale; FA, fatty acid; GDS, Geriatric Depression Scale; HAMD, Hamilton Rating Scale for Depression; MADRS, Montgomery–Asberg Depression Rating Scale; NAC, N-acetylcysteine; NC, not clear; NPO, non-profit organisation; NSAID, non-steroidal anti-inflammatory drug; SF-36, Short Form 36; TAU, treatment as usual.