Clinical Effectiveness of TMS for Adults with PTSD
Both systematic reviews had very broad search criteria, identifying studies that used TMS for treating PTSD.20,21 Wahbeh et al.21 was more explicit in describing their inclusion criteria, process of study selection and method of quality assessment compared to Karsen et al20 (i.e. the data extraction process was described a priori and study screening and extraction were done by two independent reviewers). Karsen and colleagues20 did not detail any inclusion/exclusion criteria, their process of study selection, or test for publication bias. Publication bias was mentioned as a possible limitation by Wahbeh et al.,21 due to the inclusion of 17 positive trials and five negative trials, however it was not explicitly tested. The methods used in the meta-analysis20 are also concerning because two studies contributed two sets of data to the pooled effect size. This may have led to an inflation of results due to the overrepresentation of two of the three studies.
The RCT22 ensured blinding of patients and assessors, provided a detailed description of the intervention, but did have some limitations. The time frame of recruitment and method of allocation were not described, study power was a concern, and the results may not be generalizable to all patients with PTSD. The traumatic events experienced by patients included in this study were non-military in nature, including patients having experienced a motor vehicle accident, domestic violence or physical assaults.
Clinical Effectiveness of TMS for Adults with GAD
The review conducted by Zwanger et al.23 described itself as a systematic review; however it did not follow the protocol of a properly conducted systematic review. Inclusion and exclusion criteria were not listed; there was no detail of the search strategy, study selection and data extraction process, and had no reported assessment of study quality. The definition of anxiety, which included PTSD and panic disorder, was unclear and not what is typically seen in the literature.
Clinical Effectiveness of TMS for Adults with Depression
The quality of the 2014 Canadian HTA report24 was high. The scope and context of the report are well described; the methods for searching the literature, extracting data, and critically appraising the studies are well documented and conducted. Multiple databases were accessed, screening and study selection were done in duplicate, and standardized forms were used to extract study data. Furthermore, an economic analysis was completed, and the social implications and implementation concerns were discussed. The HTAs from the United States2,26 were also well conducted but variable. The Agency for Healthcare Research and Quality accessed multiple literature databases and study screening and selection was done by two independent reviewers, whereas the Blue Cross and Blue Shield Association only searched PubMed and did not report having completed study screening in duplicate. The scope of the report was also limited in that an economic analysis was not completed and patient and family perspectives were not considered in drawing conclusions regarding the use of TMS for treating depression in adults.
The two identified systematic reviews of meta-analyses12,27 provided a list of included studies, the characteristics of each, and a qualitative review of their findings. The comprehensiveness of the search was limited because only PubMed was accessed, and it is unclear if the study screening, selection and data extraction were done in duplicate. Finally, the reviews discussed the quality of the individual studies, however it was unclear if the level of quality was appropriately considered when the authors stated their conclusions.
The included systematic reviews33,34 and meta-analyses28–32,35,36 were generally of high quality. The objectives of the study and the literature search strategy were all presented a priori. In more recent reviews,28–30,35,36 the study selection and data extraction was done in duplicate and reasons for the exclusion of specific studies were provided. In most studies, where it was applicable, study heterogeneity and publication biases were assessed,29–32,35,36 as was the quality of included studies.30,31,35 Heterogeneity between studies was found in some cases.30,31 The small sample sizes of the studies included in two reviews30,36 were highlighted as limitations to the cited benefits of rTMS. The subgroup analysis performed by Xie et al.36 (according to stimulation parameters frequency, number of stimuli, motor threshold and treatment duration) may be of concern given the limited sample size included for each parameter. Finally, the generalizability of findings by Sarkar et al.30 may be limited due to its focus on studies published in the Indian context.
The included RCTs37–41 were generally of high quality based on the reporting by study authors. All authors stated that blinding and randomization took place, however the authors of one study mentioned the inherent limitations of using sham coils in TMS38 and details of the methods of randomization were not stated in two studies.38,40 The objectives, interventions, patient characteristics, and outcome measures in all studies were well described. In some cases, the underlying population from which the sample was recruited from was not well defined,37,40,41 specific P-values were not reported,38 methods of allocation were unclear,40,41 and drop-out rate was a concern.39 Where drop-out rate was a concern in one study,39 the authors accounted for missing values using the last observation carried forward method.
Guidelines Associated with the use of Transcranial Magnetic Stimulation for Adults with PTSD, GAD or Depression
All evidence-based guideline documents were based on a systematic search of the literature. A clear link between the evidence and recommendations was provided in some documents,42,45 but was less explicit in others.43,44,46 Generally, all guidelines provided recommendations that were easily identifiable in their respective documents.42–46 Most guidelines appropriately described the competing interests of their working group members, but there was no discussion of how these conflicts were, if at all, addressed. The level of specificity and ambiguity of the guidelines varied across guidelines, with Canadian guidelines42 offering a reasonable synopsis of where rTMS fits into therapy for depression and recommendations for specific rTMS stimulation parameters. The American guidelines43,44 generally had less specific recommendations while the European guidelines45 offered a similar level of specificity as the Canadian guidelines.42 The European guidelines45 were unique in that they provided a discussion of the resource implications for implementing TMS into practice.
