Objectives: To assess the efficacy and safety of acupuncture therapy (monotherapy or adjuvant therapy), compared with placebo, conventional interventions, or no treatment in treating patients with idiopathic Parkinson's disease (IPD).

Data sources: International electronic database: (1) The Cochrane Controlled Trials Register, (2) Academic Search Premier, (3) ACP Medicine, Alternative Medicine, (4) CINAHL, (5) EBM Reviews, (6) EMBASE, (7) MEDLINE, (8) OLD MEDLINE, (9) ProQuest Medical Library. Chinese electronic databases searched included: (1) VIP, (2) CJN, (3) CBM disk, (4) China Medical Academic Conference. Hand searching was conducted on all appropriate journals. Reference lists of relevant trials and reviews were also searched to identify additional studies.

Selection criteria: All randomized controlled trials (RCTs) of any duration comparing monotherapy and adjuvant acupuncture therapy with placebo or no intervention were included.

Data collection and analysis: Data were abstracted independently by Y. C. Lam and S. C. Man onto standardized forms, and disagreements were resolved by discussion.

Main results: Ten (10) trials were included, each using a different set of acupoints and manipulation of needles. None of them reported the concealment of allocation. Only two mentioned the number of dropouts. Two (2) used a nonblind method while others did not mention their blinding methods. Nine (9) studies claimed a statistically significant positive effect from acupuncture as compared with their control; only one indicated that there were no statistically significant differences for all variables measured. Only 2 studies described details about adverse events.

Conclusions: There is evidence indicating the potential effectiveness of acupuncture for treating IPD. The results were limited by the methodological flaws, unknowns in concealment of allocation, number of dropouts, and blinding methods in the studies. Large, well-designed, placebo-controlled RCTs with rigorous methods of randomization and adequately concealed allocation, as well as intention-to-treat data analysis are needed.