Review Article

A Systematic Review and Meta-Analysis of Efficacy, Cost-Effectiveness, and Safety of Selected Complementary and Alternative Medicine for Neck and Low-Back Pain

Table 3

Summary of findings of acupuncture for neck pain (only pain and functional outcomes).

Duration and cause of painOutcomesGRADE*Findings

Acupuncture versus no treatment

Acute/subacute, chronic, and mixed, (specific, nonspecific)NAInsufficient
No trial
NA

Unknown specificPain intensity score (SF-MPQ)Low
Design: RCT
ROB: Medium
Consistency: NA (only 1 trial)
Directness: yes
In one trial [75], acupuncture was significantly better than no treatment in reducing pain intensity short-term after the end of treatment (mean change: versus , ).

Unknown nonspecificNAInsufficient
No trial
NA

Acupuncture versus placebo

Acute/subacute specific, nonspecificNAInsufficient
No trial
NA

Chronic specificPain intensity score (VAS)Moderate
Design: RCT
ROB: Medium
Consistency: yes
Directness: yes
Precision: yes
In three trials, acupuncture [77, 209] or dry needling [78] was similar to sham acupuncture [77] or laser acupuncture [78, 209] immediately or at short term after the treatment. In one of these trials [78], posttreatment mean VAS values in dry needling and sham laser acupuncture groups were 29.2 (±21.9) and 28.0 (±19.4), respectively.
The meta-analysis of two trials indicated no significant difference between acupuncture and placebo immediately after the end of treatment (pooled mean difference: 0.27, 95% CI: −0.60, 1.13) [79]. See Figure 6.
NDI scoreLow
Design: RCT
ROB: Medium
Consistency: NA (only 1 trial)
Directness: yes
In one trial [77], the mean disability score was not significantly different between acupuncture and sham-acupuncture groups immediately posttreatment ( versus , ).

Chronic nonspecificPain intensity score (VAS)Low
Design: RCT
ROB: Medium
Consistency: no
Directness: yes
Precision: yes
The meta-analysis of three trials showed no significant difference between acupuncture and sham-acupuncture immediately posttreatment (pooled mean difference: −0.24, 95% CI: −1.20, 0.73) [8082] (See Figure 6). Trials comparing acupuncture to other types of placebos (e.g., TENS, drug) [83, 8587, 210] could not be pooled due to heterogeneity across outcomes, followup periods, or missing data.
NDI scoreModerate
Design: RCT
ROB: Low
Consistency: NA (only 1 trial)
Directness: yes
In one trial [83, 210], intermediate posttreatment mean disability was significantly reduced in acupuncture compared to placebo group ( versus , ).

Mixed specificPain intensity score (VAS)Low
Design: RCT
ROB: High
Consistency: NA (only 1 trial)
Directness: yes
In one trial [88], there was no significant difference between acupuncture and placebo (laser pen) at intermediate-term posttreatment followup ( versus , ).

Mixed nonspecificNAInsufficient
No trial
NA

Unknown specificNo pain or disability outcome reportedNAOne trial [27] reporting % subjects without symptoms.

Unknown nonspecificNAInsufficient
No trial
NA

Acupuncture versus pain medication

Acute/subacute, mixed (specific, nonspecific)NAInsufficient
No trial
NA

Chronic specificPain intensity score (VAS, SF-MPQ)Low
Design: RCT
ROB: High
Consistency: no
Directness: yes
Of the three trials [8991] comparing acupuncture to medications, in two [89, 90] there was no significant difference between acupuncture and injection of lidocaine [89, 90], lidocaine plus corticoid [90], or botulinum toxin [90] at short-term posttreatment followup. In one of the trials [89], two-week posttreatment mean VAS values were for acupuncture and for lidocaine ( ). In another trial [91], acupuncture was better than NSAIDs immediately after treatment (mean VAS score: versus , ).

Chronic nonspecificPain intensity score (VAS)Low
Design: RCT
ROB: High
Consistency: yes
Directness: yes
None of three trials comparing acupuncture to medication (e.g., NSAIDs, analgesics) demonstrated significant between-group differences [69, 70, 87]. In one of the trials [69], acupuncture had a better mean score versus pain medication group at immediate (mean VAS score: versus ) or at intermediate-term followup (mean VAS score: 2.5 versus 4.7) [69, 72].

Unknown specificPain intensity score (VAS, SF-MPQ)Low
Design: RCT
ROB: High
Consistency: yes
Directness: yes
In two trials [28, 92], acupuncture was significantly more effective than injection of lidocaine in the short-term. In one trial [28], the mean pain scores were versus ( ).

Unknown nonspecificNAInsufficient
No trial
NA

Acupuncture versus physiotherapy

Acute/subacute, chronic, mixed, or unknown (specific, nonspecific)NAInsufficient
No trial
NA

Acupuncture versus mobilization

Acute/subacute, mixed, or unknown (specific, nonspecific)NAInsufficient
No trial
NA

Chronic specificNAInsufficient
No trial
NA

Chronic nonspecificPain intensity score (VAS)Low
Design: RCT
ROB: Medium
Consistency: NA (only 1 trial)
Directness: yes
In one trial [93], there was no significant difference between acupuncture and standard localized mobilization techniques at short- or intermediate-term posttreatment followup (no numerical data on mean scores were reported).
Disability (NPQ score)Low
Design: RCT
ROB: Medium
Consistency: NA (only 1 trial)
Directness: yes
In one trial [93], there was no significant difference between acupuncture and standard localized mobilization techniques at short- or intermediate-term posttreatment followup (no numerical data on mean scores were reported).

Acupuncture versus usual care

Acute/subacute, mixed, or unknown (specific, nonspecific)NAInsufficient
No trial
NA

Chronic specificNAInsufficient
No trial
NA

Chronic nonspecificDisability (NPQ score)Low
Design: RCT
ROB: Medium
Consistency: NA (only 1 trial)
Directness: yes
In one trial [94], acupuncture was added to general practice care and showed no difference in disability (NPQ) compared to general practice care alone immediately posttreatment ( versus , ).

Acupuncture versus manipulation

Acute/subacute, mixed, or unknown (specific, nonspecific)NAInsufficient
No trial
NA

Chronic specificPain intensity score (VAS)Low
Design: RCT
ROB: High
Consistency: NA (only 1 trial)
Directness: yes
In one trial [24], there was no significant difference between acupuncture and spinal manipulation at short-term followup (mean VAS: versus ).

Chronic nonspecificPain intensity score (mean % VAS)Low
Design: RCT
ROB: High
Consistency: no
Directness: yes
In one trial [69], acupuncture was better than manipulation in reducing pain intensity at short-term followup (50.0% versus 42.0%).
In another trial [70], immediate posttreatment reduction in pain intensity was significantly greater in manipulation (VAS: 33.0%) versus acupuncture (VAS score % reduction not reported).
Pain intensity score (VAS)Low
Design: RCT
ROB: High
Consistency: NA (only 1 trial)
Directness: yes
In one trial [69, 72], median pain intensity scores in the acupuncture and manipulation groups did not differ at intermediate-term followup (VAS median scores: 2.5 versus 2.8, ).
Disability score (NDI)Low
Design: RCT
ROB: High
Consistency: yes
Directness: yes
Two trials demonstrated significant superiority of manipulation over acupuncture in improving neck disability.
In the first trial [70], median NDI score reduction in neck disability immediately posttreatment was significantly greater in manipulation (−10.0, 95% CI: −14.0, −4.0) than acupuncture group (−6.0, 95% CI: −16.0, 2.0).
In the second trial [69], the posttreatment NDI values were significantly more improved in manipulation (median: 22; range: 2–44) than acupuncture group (median: 30; range: 16–47); value not reported.

Acupuncture versus massage

Acute/subacute, mixed, or unknown (specific, nonspecific)NAInsufficient
No trial
NA

Chronic specificPain intensity score (VAS)Low
Design: RCT
ROB: High
Consistency: NA (only 1 trial)
Directness: yes
In one trial [209], acupuncture was significantly better (VAS score scale: 0–100) compared to massage in a short-term posttreatment followup (mean VAS score change from baseline: 24.22 versus 7.89, ).

Chronic nonspecificNAInsufficient
No trial
NA

*Precision in formal grading was applied only to pooled results.
VAS: visual analog scale; RMDQ: Roland-Morris disability scale; NHP: Nottingham health profile; MPQ: McGill pain questionnaire; PDI: pain disability index; SF: short form; NPQ: neck pain questionnaire; SF-PQ: short form pain questionnaire; PRI: pain rating index; PPI: present pain intensity; NA: not applicable; NDI: neck disability index.