The importance of selecting the correct site to apply spinal manipulation when treating spinal pain: myth or reality? A systematic review

Key points

1. 019 of 10 RCTs found no statistically significant difference between targeted and non-targeted spinal manipulation for pain or disability
2. 02The one study that favored the candidate site had high risk of bias and reported only a small pain reduction (1.2 points on an 11-point scale), below the minimal clinically important difference
3. 03Studies compared candidate sites to the same vertebral level on the contralateral side, the same spinal region, and remote spinal regions - none showed a consistent benefit of site-specificity
4. 04The clinical tools used to identify a 'candidate' vertebral level (such as motion palpation) lack demonstrated reliability and reproducibility
5. 05Results suggest SMT may work through generalized biomechanical or neurological mechanisms rather than site-specific effects

How it was conducted

Design

Systematic review of randomized controlled trials (PRISMA 2020), synthesis without meta-analysis (SWiM) due to heterogeneity

Included studies

10 RCTs published between 2003 and 2020, identified from PubMed, Embase, Index to Chiropractic Literature, and CINAHL (3,288 articles screened)

Participants

Adults with cervical or lumbar spinal pain (6 studies cervical, 4 lumbar); sample sizes ranged across studies; five studies included chronic pain patients, two included acute pain patients

Intervention comparison

SMT at a clinician-determined candidate site vs. SMT at: (i) same vertebral level contralateral side, (ii) non-specific same spinal region, or (iii) a remote spinal region

Primary outcome

Patient-reported outcomes (pain intensity, disability); secondary outcomes included pressure pain detection threshold (PPT) and range of motion

Risk of bias

Assessed using modified Cochrane RoB tool; 5 studies low RoB, 4 moderate RoB, 1 high RoB; 9 of 10 considered credible

What they found

• 9 credible studies (31 outcome comparisons) reported no statistically significant between-group differences favoring the candidate site
• 1 study (high RoB) reported a statistically significant but clinically small reduction in neck pain favoring the candidate site: mean difference 1.2 points (95% CI -1.9 to -0.5) on an 11-point scale, below the minimal clinically important difference threshold
• For low back pain: between-group differences for pain intensity (e.g., 0.5 [95% CI -0.1 to 1.1]) and PPT at the lumbar spine (e.g., -1.8 [95% CI -6.4 to 2.8]) were non-significant immediately post-treatment
• For neck pain (remote region comparisons): between-group differences for pain and disability were near zero with narrow confidence intervals across multiple time points (e.g., pain at weeks 4, 8, 12: 0.0 [-0.9 to 0.9], -0.1 [-1.0 to 0.8], -0.1 [-1.0 to 0.8])
• Side effects were either not reported or were minimal and did not differ between candidate and non-candidate SMT groups

Limitations

• Most studies examined a single or few SMT sessions in chronic pain patients, so lack of difference may partly reflect short intervention duration or chronicity effects
• Meta-analysis was not possible due to heterogeneity in study design, SMT technique, and outcome measures, limiting pooled effect estimation
• No studies reported whether participants could discern which site received SMT, leaving open the possibility of unblinded expectation bias influencing results
• The reliability and validity of clinical tests used to identify candidate sites (e.g., motion palpation) are themselves questionable, meaning the 'candidate site' concept may not be consistently operationalized

Why it matters

For patients

Patients with spinal pain can be reassured that spinal manipulation does not need to be applied at a highly specific vertebral level to produce its effects, and that choosing between targeted and non-targeted approaches is unlikely to change outcomes.

For clinicians

Clinicians who invest heavily in identifying the 'correct' vertebral level before applying SMT should be aware that the current evidence does not support site-specificity as a driver of clinical outcomes, and that SMT technique choice may matter less than other contextual and therapeutic factors.

For readers

This is the first systematic review to directly test whether SMT site selection matters, and it consistently challenges the assumption - across different spinal regions and comparison types - that targeting a specific vertebral level improves outcomes.