Meta-analysis of transcranial magnetic stimulation to treat post-stroke dysfunction

摘要

OBJECTIVE: To evaluate the effects of transcranial magnetic stimulation (TMS) on post-stroke dysfunction. DATA SOURCES: A computer-based online search of the VIP and PubMed databases was performed to identify TMS studies published between January 1989 and December 2010 using the key words "transcranial magnetic stimulation, stroke" in Chinese and English. A total of 61 articles were collected. STUDY SELECTION: Repetitive articles were excluded. Articles published recently in the authoritative journals of the same domain were selected. The full-text of selected articles was searched. A total of seven articles were randomized controlled experiments regarding the application of TMS in the treatment of post-stroke dysfunction. The methodology quality of included articles was evaluated according to standards of Cochrane Reviewer’s Handbook system and analyzed using RevMand4.2 software. MAIN OUTCOME MEASURES: Therapeutic effects and difference evaluation indices were represented by odds ratios, weighted mean difference (WMD) and 95% confidence interval (CI). Potential publication bias was described using a funnel plot. RESULTS: A total of seven randomized, controlled studies were included involving 281 patients. Following TMS treatment, meta-analysis results revealed that scores in the Mini-Mental State Examination were higher in the TMS group compared with the control group [WMD = 3.96, 95% CI (2.44, 5.49), P = 0.08]; scores in the Hamilton Rating Scale for Depression were significantly lower in the TMS group compared with the control group [WMD = -6.21, 95% CI (-7.55, -4.87), P = 0.92]; scores in National Institutes of Health Stroke Scale were lower following TMS treatment compared with the control group [WMD = -0.89, 95% CI (-1.98, 0.19), P = 0.04]. Performance of patients undergoing TMS treatment was better than the controls in the line bisection test [WMD = -0.78, 95% CI (-1.43, -0.12), P = 0.005] and line cancellation test [WMD = -0.47, 95% CI (-0.78, -0.15), P < 0.000 01]. Sensitivity analysis produced identical results after eliminating articles with unknown diagnostic standards or statistical methods. CONCLUSION: TMS effectively improved post-stroke dysfunction, manifested by improved cognitive function and memory performance compared with controls.