Acupuncture for Posttraumatic Stress Disorder: A Systematic Review of Randomized Controlled Trials and Prospective Clinical Trials

Table 1

Summary of randomized controlled trials and prospective clinical trials of acupuncture for posttraumatic stress disorder.

First author [ref]; (year); country	Population	Study design	Sample; size/𝑁, analyzed	Intervention/control group; (regime)	Treatment; session	Main outcomes	Intergroup difference	Comments
RCT (𝑛=4)
Hollifield [18]; (2007); U.S.A.	28 out of 84 identified childhood abuse/; Others, unknown trauma	3 arm parallel, open	84/73	(A) AT + AAT (𝑛=29)/; (B) CBT (𝑛=28); (C) WLC (𝑛=27)	24 sessions	(1) PTSD scale (PSS-SR); (2) Depression (HSCL-25); (3) Anxiety (HSCL-25); (4) Impairment (SDI)	(1) A versus B: 𝑃 = 0.36, MD, −0.26 [−0.83, 0.30]; A versus C: 𝑃 = 0.001, MD, −0.98 [−1.58, −0.38]; B versus C: 𝑃 = 0.004, MD, −0.85 [−1.44, −0.27]; (2) A versus B: 𝑃 = 0.92 MD, 0.03 [−0.53, 0.59]; A versus C: 𝑃 = 0.02, MD, −0.68 [−1.27, −0.10]; B versus C: 𝑃 = 0.008, MD, −0.80 [−1.38, −0.21]; (3) A versus B: 𝑃 = 0.39, MD, −0.25 [−0.81, 0.31]; A versus C: 𝑃 = 0.003, MD, −0.91 [−1.51, −0.32]; B versus C: 𝑃 = 0.008, MD, −0.79 [−1.37, −0.21]; (4) A versus B: 𝑃 = 0.98, MD, −0.01 [−0.57, 0.55]; A versus C: 𝑃 = 0.03, MD, −0.64 [−1.22, −0.06]; B versus C: 𝑃 = 0.03, MD, −0.64 [−1.22, −0.07]	The AT group had significantly better improvements in PTSD symptoms than the WLC group. But, there was no statistically significant difference between the AT group and the CBT group.
Zhang [25] (2010)China	Earthquake	4 arm parallel, open	276/256	(A) EA (𝑛=69); (B) EA + moxa (𝑛=69); (C) EA + AAT (𝑛=69)/; (D) Oral SSRI (𝑛=69)	36 sessions	(1) PTSD scale (CAPS); (2) Depression (HAMD); (3) Anxiety (HAMA)	(1) A versus D: 𝑃 = 0.43, MD, −0.13 [−0.47, 0.20]; B versus D: 𝑃 = 0.88, MD, −0.03 [−0.36, 0.31]; C versus D: 𝑃 = 0.55, MD, −0.10 [−0.44, 0.23]; (2) A versus D: 𝑃 = 0.14, MD, −0.25 [−0.59, 0.08]; B versus D: 𝑃 = 0.34, MD, −0.16 [−0.50, 0.17]; C versus D: 𝑃 = 0.23, MD, −0.21 [−0.54, 0.13]; (3) A versus D: 𝑃 = 0.34, MD, −0.16 [−0.50, 0.17]; B versus D: 𝑃 = 0.64, MD, −0.08 [−0.41, 0.25]; C versus D: 𝑃 = 0.54, MD, 0.11 [−0.23, 0.44]	The therapeutic effect of EA was not better than that of oral SSRI.
Zhang [26](2010); China	Earthquake	2 arm; parallel; open	92/81	(A) EA + moxa (𝑛=46)/; (B) Oral SSRI (𝑛=46)	36 sessions	(1) PTSD scale (CAPS); (2) Depression (HAMD); (3) Anxiety (HAMA)	(1) A versus B: 𝑃 < 0.00001, MD, −1.77 [−2.26, −1.29]; (2) A versus B: 𝑃 < 0.00001, MD, −1.96 [−2.46, −1.46]; (3) A versus B: 𝑃 < 0.00001, MD, −1.53 [−2.00, −1.07]	EA plus moxa was more effective than oral SSRI therapy.
Zhang [27]; (2011); China	Earthquake	2 arm; parallel; open	91/90	(A) Acupoint Stimulation; + CBT (𝑛=67)/; (B) CBT (𝑛=24)	3~4 sessions*	(1) PTSD scale (IES-R); (2) PTSD scale (self compiled questionnaire)	(1) A versus B: 𝑃 < 0.00001, MD, −1.56 [−2.08, −1.04]; (2) A versus B: 𝑃 = 0.01, MD, −0.59 [−1.07, −0.12]	The acupoint stimulation plus CBT showed better efficacy than CBT therapy alone.
UCT (𝑛=2)
Wang [28]; (2009); China	Earthquake	UCT	69	EA + AAT + moxa	36 sessions	(1) The number of cured/improved/non-improved	Not applicable	Treatment was effective in 65 out of 69 (94.2%).
Yuan [29]; (2009); China	Earthquake	UCT	34	AT	20 sessions	(1) The number of cured/improved/non-improved	Not applicable	AT was effective in 31 out of 34 (91.2%).

Abbreviations: RCT: randomized controlled trial; UCT: uncontrolled clinical trial; AT: classical acupuncture; EA: electro-acupuncture; moxa, moxibustion; AAT: auricular acupuncture; CBT: cognitive behavioral therapy; WLC: waitlist control; SSRI: selective serotonin reuptake inhibitors; PSS-SR: posttraumatic symptom scale-self report; HSCL-25: self-rated Hopkins symptom checklist-25; SDI: Sheehan Disability Inventory; MD: mean difference; CAPS: clinician-administered PTSD scale; HAMD: Hamilton depression rating scale; HAMA: Hamilton anxiety rating scale; IES-R: Chinese version of the incident effect scale revised; *treated a time every other day for 1 week.