Should We Use Traction to Treat Lumbar Radiculopathy? | Article of The Week #34

Traction is an ‘old school’ treatment used to treat low back pain but does it work?

Low back pain in the most common condition a physiotherapist will see and it affects some 60-80% of people during thier lifetime. A small percent of these people will develop lumbar radiculopathy, a pain syndrome caused by compression and/or irritation to the lumbar nerve roots.

Deep Dive into Back Pain on Physioplus

Typical first line treatment for lumbar radiculopathy is conservative management and this can take many forms. Traction is something which has been used for decades and is still being used to treat radiculopathy even though there is a lack of evidence to support it’s use. Despite this it is still being used in clinical practice in place of treatments which are proven to be more effective.

A new systematic review has been published to again explore whether or not traction is effective for treating lumbar radiculopathy.


This systematic review was registered on PROSPERO, follows PRISMA recommendations and used GRADE and RoB Tool to assess for bias and research quality. The databases used for this systematic review were Cochrane, PubMed, CINAHL, Scopus, WoS and PEDro. The addition of the latter is good to see as many reviews investigating physiotherapy intervention choose not to include PEDro in favor of more well known databases but this can lead to specific evidence being missed from the data analysis.

The search terms used for the review we adapted for each databse and each search string is available to download. The string for PubMed is below so you can see an example of one of the searches. The searches appear thorough and systematic.

(“traction”[MeSH Terms] OR “physical therapy modalities”[MeSH Terms]) AND (“low back pain”[MeSH Terms] OR “sciatica”[MeSH Terms] OR “radiculopathy”[MeSH Terms] OR “radicular syndrome”[All Fields] OR “nerve root pain”[All Fields] OR “leg pain”[All Fields]) AND ((randomized controlled trials as topic[MeSH Terms]) OR (randomized controlled trial [All Fields] OR controlled clinical trial[All Fields] OR random*))

Databases were searched from any date up to March 31st 2019. Additional records were gathered via reference list mining. Two independent blinded reviewers conducted the study selection elements and then the authors systematically compared their results with a third person available for consensus if disagreement occurred.

Studies were included within the review if they were RCTs published in English, had participants 18 years of age or older with lumbar radiculopathy confirmed by the presence of at least 2 of the following:

  1. radicular symptoms (LBP with pain and/or numbness radiating below the knee
  2. 1 or more radicular sign: (sensory loss / parasthesia in any L4-S1 dermatome, diminished patellar or achillies reflex or L4-S1 weakness)
  3. Positive CT/MRI

Studies involving any form of traction (if described in detail) were included when applied alone or in conbination with other conservative or drug treatments.

Primary outcome measures were perceived pain in lumbar and/or sciatic region neasured using a numerical rating scale. Secondary outcomes include Oswestry Disability Index, Roland & Morris Disability Questionnaire, mobility, psychological parameters. QoL measures and treatment adherence.

This is a lot of secondary outcome measures which is often found in systematic reviews however if you cast your net wide enough you will catch something. The research question the authors are trying to answer are what are the effects of traction on lumbar radiculopathy so perhaps this number of secondary measures is ok

Results & Clinical Take Home

In total three studies were included within the analysis, this was from an initial total of 3,673 identified studies. In total these 3 studies included a pooled total of 85 patients. Two different types of traction were involved (vertical and inversion) with two using intermittent traction and one continuous duration.

The duration of the course of treatment ranged from 1 week to 2 months and the length of tractions ranged from 10 – 45 minutes. The force used ranged from upper half of patients bodyweight to the full patients bodyweight. Traction was compared to physiotherapy, medications and bed rest (this isn’t even a control!).

Overall the studies included had moderate to high risk of bias largely due to an inability to blind participants and therapists however there were also concerns about selection, attrition and performance biases.

Quick Summary of Systematic Review of Traction of Lumbar Radiculopathy

  • There is no support for the use of lumbar traction in the management of lumbar radiculopathy
  • Lumbar traction is more effective than bed rest alone
  • Lumbar traction is not more effective than any other physiotherapy treatment method for radiculopathy and makes no additional benefit when added in combination treatments

The results of the systematic review show that intermittent traction was ineffective when compared to usual physiotherapy even when combined with another treatment. Additionally there was no positive effect on increasing physical activity. Even when added to usual medical management traction provided no additional effects to treatment. The only thing traction was more effective than was bed rest but that’s clearly obvious and not relevant to clinical practice. 

In summary traction does not appear to provide any beneficial effects when used to treat proven lumbar radiculopathy. Furthermore, the existing evidence is of poor quality and until more high quality larger scale trials are created  traction should not form part of treatment plans for lumbar radiculopathy. Even if these studies take place is this the direction the profession should go in? Probably not. this systematic review is another reason traction should be confined to the treatment modality graveyard.