Table 1 Studies included in the analysis
Study | Design | Baseline characteristics | Comparison # med trials (treatment resistance) | Outcome | CEBM/GRADE | |
|---|---|---|---|---|---|---|
Fregni F, et al. Inter Jrl Neuropsychophar. 2006 [24] | Pooled data from 6 clinical trials – retrospective analysis Countries: Canada (N = 25 patients), US (N = 60), Austria (N = 29), Brazil (N = 21) | 153 patients; Age = 51.1 ± 15.1; M/F = 63/90 | ≥2 medication trials defined as refractory (medication resistant). | Treatment refractoriness was a significant predictor of clinical response to rTMS (P < 0.0001); use of Model 3 excluded Tel Aviv patients (N = 42) due to only one failed medication trial. Therefore 195–42 = 153 patients included in the analysis. | 1c/B | |
Brakemeier E-L, et al. Jrl. Psych Res. 2007 [30] | Prospective case series Country: Germany | 70 patients; Age = 49.5 ± 12.5; M/F = 44/26 | Comparison medication resistant (≥2 trials) (N = 51) to non-medication resistant (1 trial) (N = 19) | Patients with a shorter duration of depressive episode and lower level of medication resistance showed a greater response to rTMS. | 4/C | |
O’Reardon JP, et al. Biol Psych. 2007 [25]; Lisanby SH, et al. Neuropsychopharm. 2009 [18] | Double blind multisite (23 centers) RCT | 301 patients; Age = 48.3 ± 10.8; M/F = 142/159 | Comparison medication resistant (≥2 trials) (N = 147) to non-medication resistant (1 trial) (N = 164) | The likelihood of responding to rTMS was 4 times higher if patients had received one unsuccessful medication trial before rTMS in comparison with patients having received 2 or more unsuccessful trials (P = 0.021). Effect size in patients with one failed therapy was 0.83. Post hoc analysis performed by Lisanby. | 1b/B | |
Countries: 20 sites US; 2 Austria; 1 Canada | ||||||
Brakemeier E-L, et al. Jrl Affect Disord. 2008 [31] | Prospective and retrospective case series Country: Germany | 79 patients; Age = 49.1 ± 14.3; M/F = 35/43 | Comparison within rTMS treatment arm medication resistant (≥2 trials) to non-medication resistant (1 trial) | Non-treatment resistant patients with a short duration of episode were more likely to respond to rTMS than medication resistant (43% vs. 18%) (P = 0.023) | 4/C | |
Cohen RB, et al. Jr. Nerv Ment Dis. 2010 [32] | Single center observation study Country: Brazil | 56 patients; Age = 48 ± 15; M/F = 26/30 | Comparison low treatment resistance [1 trial] (n = 34) to high treatment resistance [≥2 trials] (N = 22) | Low treatment resistance has a statistically significant effect (P < 0.01) on treatment outcome as measured by HDRS. | 4/C | |
Carpenter LL, et al. Depress Anxiety. 2012 [33] | Multicenter observational study Country: US | 307 patients; Age = 48.6 ± 14.2; M/F = 102/205 | Comparison low treatment resistance [≤1 trial] (n = 140) to high treatment resistance [≥2 trials] (N = 167) | Low treatment resistance had a modest influence on treatment outcome as measured by CGI-S and PHQ-9 outcomes. No statistical difference between groups on response and remission but a higher percentage of patients having response (59.4% vs. 56.8%; CGI-S; 57.2% vs. 55.6%; PHQ-9) or remission (39.9% vs. 34.9%; CGI-S; 31.9% vs. 26%; PHQ-9)with low treatment resistance | 4/C | |
Huang M-L, et al. Aust & NZ Jrl Psych. 2012 [26] (Note: Huang L et al., Zhejiang Da Xue Bio Yi Xue Ban. 2011 is a duplicate study [43]. However it is in Chinese so Huang M-L et al. Aust & NZ Jrl Psych 2012 used) | Single center RCT Country: China | Active = 28; Age = 32.8 ± 7.3; M/F = 9/19 | Sham = 28; Age = 31.6 ± 7.4; M/F = 8/20 | Comparison of rTMS plus citalopram (N = 28) vs. rTMS sham plus citalopram (N = 28) in first episode major depressive disorder on response and remission after 4 weeks. First 2 weeks use of rTMS (active or sham). Second two week citalopram only in both groups. | Significantly greater number of early improvers (using HAMD-17) at 2 weeks with rTMS vs. sham/citalopram (P = 0.031). No difference in response (46% vs. 36%; P = 0.586) or remission (39% vs. 29%; P = 0.572) at 4 weeks. | 1b/B |
Wang H-N, et al. Translational Psych. 2017 [27] | Single center RCT Country: China | rTMS+med = 82; Age = 42.3 ± 11.4;M/F = 22/60 | Med =108; Age = 40 ± 11.5; M/F = 23/85 | Comparison in first episode depressed patients: rTMS (N = 91) vs. antidepressant (N = 108) vs. rTMS plus antidepressant (N = 82) over 12 months. Examination of relapse/recurrence. | Relapse/recurrence at 12 months significantly lower in rTMS plus antidepressant group (20%) vs. antidepressant group (44.4%) (P = 0.033). | 1b/B |
Wang Y-M, et al. Psych Res. 2017 [28] | Single center RCT Country: China | Active = 22; Age = 28.8 ± 8.5; M/F = 12/10 | Sham = 23; Age = 30.1 ± 9.5; M/F = 13/10 | Comparison in treatment naïve patients rTMS plus paroxetine (N = 22) [active] to rTMS sham plus paroxetine (N = 21) [sham] | Response and remission rate of [active vs. sham] 95.5% vs. 71.4 and 68.2% and 38.1% respectively. (P < 0.05) | 1b/B |
Yang H, et al. Jrl Psych Brain Sci. 2017 [29] | Single center RCT Country: China | Active =41 patients; Age = 35.5 ± 12; M/F = 17/24 | Sham = 41 patients; Age = 35.4 ± 12.1; M/F = 15/25 | Comparison in treatment naïve patients rTMS plus escitalopram (N = 41) [active] to rTMS sham plus escitalopram (N = 41) [sham] | Active rTMS plus escitalopram significantly more effective (≥50% reduction in HAMD-17 score) (N = 36) than sham (N = 17) at 4 weeks (P < 0.05) | 1b/B |