This section describes efficacy outcomes from the published literature that the Committee considered as part of the evidence about this procedure. For more detailed information on the evidence, see the interventional procedure overview.
In a systematic review of 40 randomised controlled trials including 1592 patients with depression (type unspecified) treated by repetitive transcranial magnetic stimulation (rTMS, n=751) or sham stimulation (n=632), meta‑analysis of mean changes in unspecified depression rating scales showed a significant effect in favour of rTMS (Hedges' g value of 0.55, p<0.001).
In a non‑randomised comparative study of 185 patients with treatment‑resistant depression treated by conventional rTMS (n=98) or theta‑burst rTMS (n=87), Hamilton Depression Rating Scale (HDRS) scores (lower scores indicate less depression) decreased from 22.1±6.9 to 12.3±8.9 and from 21.1±5.1 to 12.7±7.9 respectively at 1‑month follow‑up (p value within groups <0.001, p value between groups not significant). In the same study, Beck Depressive Inventory scores (scores range from 0 to 63, with lower scores indicating less depression) decreased from 35.4±10.8 to 22.4±15.5 in the conventional rTMS group and from 35.9±9.9 to 20.2±13.3 in the theta‑burst rTMS group at 1‑month follow‑up (p value within groups <0.001, p value between groups not significant).
In a systematic review of 63 studies including 3236 patients treated by rTMS (n=2330), sham stimulation (n=806) or electroconvulsive therapy (ECT; n=100), percentage changes in HDRS scores (lower scores indicate less depression) were pooled and converted to Clinical Global Impression – Improvement scale (CGI‑I) scores. CGI‑I scores range from 1 to 7: a score of 4 means no change, scores of less than 4 indicate improvements in depression and scores of more than 4 indicate worsening depression. For patients with any type of depression, the mean percentage reduction in HDRS scores was 37% (CGI‑I equivalent 2.8) in the rTMS group and 22% (CGI‑I equivalent 3.4) in the sham stimulation group (p<0.05). For patients with treatment‑resistant depression, the mean percentage reduction in HDRS scores was 48% (CGI‑I equivalent 2.4) in the rTMS group and 23% (CGI‑I equivalent 3.4) in the sham stimulation group (p<0.05). When rTMS was compared against ECT in patients with any type of depression, the mean percentage reduction in HDRS scores was 34% (CGI‑I equivalent not reported) in the rTMS group and 46% (CGI‑I equivalent 2.45) in the ECT group (p<0.05).
In a systematic review of 10 randomised controlled trials including 634 patients with treatment‑resistant depression treated by bilateral rTMS, unilateral rTMS or sham simulation, clinical response data (defined as more than a 50% improvement in HDRS or Montgomery–Åsberg Depression Rating Scale scores) were compared between groups. Meta‑analysis of clinical response rates in patients treated by bilateral rTMS or sham stimulation revealed a risk ratio of 3.29 in favour of bilateral rTMS (95% confidence interval [CI] 1.69 to 6.38, p=0.0004). In the same study, meta‑analysis of remission data (classified according to predefined criteria in each included study) revealed no significant difference between patients treated by bilateral rTMS or sham stimulation (risk ratio 0.5; 95% CI 0.19 to 1.31, p=0.16).
In a systematic review of 10 randomised controlled trials including 429 patients with a primary major depressive episode treated by rTMS (n=217) or ECT (n=212), meta‑analysis of clinical response data (defined as more than a 50% improvement in HDRS scores) revealed a risk ratio of 1.52, in favour of ECT (95% CI 1.18 to 1.95, p=0.001). Meta‑analysis of remission data (classified according to predefined criteria in each included study) revealed a risk ratio of 1.42 in favour of ECT (95% CI 1.16 to 1.75, p=0.0007).
A case series evaluated 120 patients who had at least a partial response (that is, at least a 25% improvement in HDRS scores); 99 patients were recruited from the active rTMS arm of a randomised sham‑controlled trial, while 21 patients initially had sham stimulation and subsequently received active rTMS. For patients originally in the active rTMS arm of the trial, the mean HDRS score was 9.1±6.2 at the end of rTMS therapy and 9.0±7.1 at 6‑month follow‑up (p=0.537), indicating a maintained treatment effect. No pre‑treatment scores were reported. No mean HDRS scores were reported for patients who initially had sham stimulation and subsequently received active rTMS. In the same study, the relapse rate (Kaplan–Meier estimate) at 6‑month follow‑up was 13% in patients who were originally in the active rTMS arm of the trial and 16% in patients who initially had sham stimulation and subsequently received active rTMS (no p value reported).
Specialist advisers listed improvements in depressive symptoms and health‑related quality of life as efficacy outcomes.