Background
Acupuncture is frequently used to treat the side effects of cancer treatment, but the safety of this intervention remains uncertain. The current meta‐analysis was conducted to assess the safety of acupuncture in oncological patients.
Methods
The PubMed, Cochrane Central Register of Controlled Trials, and Scopus databases were searched from their inception to August 7, 2020. Randomized controlled trials in oncological patients comparing invasive acupuncture with sham acupuncture, treatment as usual (TAU), or any other active control were eligible. Two reviewers independently extracted data on study characteristics and adverse events (AEs). Risk of bias was assessed using the Cochrane Risk of Bias Tool.
Results
Of 4590 screened articles, 65 were included in the analyses. The authors observed that acupuncture was not associated an with increased risk of intervention‐related AEs, nonserious AEs, serious AEs, or dropout because of AEs compared with sham acupuncture and an active control. Compared with TAU, acupuncture was not associated with an increased risk of intervention‐related AEs, serious AEs, or drop out because of AEs but was associated with an increased risk for nonserious AEs (odds ratio, 3.94; 95% confidence interval, 1.16‐13.35; P = .03). However, the increased risk of nonserious AEs compared with TAU was not robust against selection bias. The meta‐analyses may have been biased because of the insufficient reporting of AEs in the original randomized controlled trials.
Conclusions
The current review indicates that acupuncture is as safe as sham acupuncture and active controls in oncological patients. The authors recommend researchers heed the CONSORT (Consolidated Standards of Reporting Trials) safety and harm extension for reporting to capture the side effects and better investigate the risk profile of acupuncture in oncology.
Lay Summary
According to this analysis, acupuncture is a safe therapy for the treatment of patients with cancer.
Acupuncture seems to be safe compared with sham acupuncture and active controls.