Research Review by Dr. Shawn Thistle©


Oct. 2006

Study Title:

Traction for low back pain with or without sciatica: An updated systematic review within the framework of the Cochrane Collaboration


Clarke J et al.

Public Information:

Spine 2006; 31(14): 1591-1599.


A common treatment for low back pain and sciatica is traction. For many years traction has been employed as a single treatment, or in combination with other modalities for lumbar spine complaints, most commonly involving disc pathology with radiculopathy.

As many of us in practice know, there are many companies who are advertising and selling very elaborate traction machines with equally elaborate claims of clinical success. One just needs to check the office mail to find a new flyer or brochure for such a product.

The purpose of this systematic review (within the framework of the Cochrane Collaboration) was to assess the efficacy of traction for low back pain with or without sciatica/radiculopathy, compared to reference treatments, placebo/sham traction, or no treatment.

The most frequently used traction techniques (all included in this review) are:
  1. mechanical or motorized traction (via machine)
  2. manual traction (via therapist)
  3. auto-traction (patient controlled by grasping/pulling on bars attached to a traction table)
Most traction devices utilize a harness around the rib cage and another around the iliac crest area. The force exerted can be intermittent or continuous, but only in motorized units can it be standardized. Factors contributing to the actual force transmitted through the spine include muscular counterforces, skin stretch and subcutaneous tissue girth, abdominal pressure, and the frictional interface between table and patient.

The following have been proposed as mechanisms of action for traction (or spinal elongation):
  • decreasing lordosis and increasing intervertebral space
  • inhibition of nociceptive impulses
  • improved mobility
  • decreased mechanical stress
  • reduced muscle spasm or spinal nerve compression
  • release of luxation of a disc or capsule from the Z-joints
  • release of adhesion from the Z-joints
It is worth noting here that none of the above-listed mechanisms have been satisfactorily supported by empirical data.

There is also controversy in the literature regarding the most effective amount of load required for effective traction. Case reports exist citing adverse reactions to traction forces in excess of 50% body weight while others question the utility of forces less than 20% body weight (even referring to it as "placebo traction").

The objective of this review was to update a previous systematic review that was done up to 1992. Randomized controlled trials including men or women (over 18 yoa) treated for non-specific LBP with some type of traction were included. Other treatments could be provided, as long as traction was the main treatment component. LBP patients in these trials included acute, subacute, and chronic cases.

Outcome measures investigated were: pain (measured with VAS or NRS), a global measure (overall improvement etc.), back-pain specific functional status (ex. Roland Morris or Oswestry), and return to work.

In total, 24 studies involving 2177 patients were included in the review. All 24 studies included at least some patients with sciatica.

Relevant observations and recommendations from this review include:
  • for mixed groups of patients with LBP (with and without sciatica), traction is no more effective than comparison treatments (including placebo and sham traction)
  • for patients with sciatica, there is contradictory evidence regarding traction compared to sham, placebo, or no treatment
  • for patients with sciatica, there is contradictory evidence regarding different types of traction (i.e. which type of traction is superior - mechanical, manual, or auto)
  • there is limited evidence that there is no difference between light and normal force traction for patients with sciatica
  • several new trials have been published since the last review (3 of them being of high quality), but the results have not changed substantially

Conclusions & Practical Application:

Conclusions and Practical Application: The main conclusion of this review is that traction as a single treatment for LBP with or without sciatica is no more effective than placebo, sham, no treatment, or other treatments.

It should be noted that the lack of strong, consistent evidence regarding the use of traction results from the lack of well designed, high quality studies, heterogeneity of study populations, and lack of statistical power in existing studies.

The reviewers found NO high quality studies supporting possible positive effects of ANY of the types of traction included in this review. They also state that some of the earlier published positive evidence came from low quality studies that investigated auto-traction.

From a clinical perspective, the reviewers state that no studies were found that investigated traction as part of a pragmatic multidisciplinary approach to LBP - perhaps illuminating a useful avenue for future studies. As I have said before, it often doesn't surprise me that individual treatments (ex. SMT, massage, etc.) are not supported as single treatments for LBP. Most practicing clinicians utilize a number or treatment modalities…a remaining disconnect between clinical practice and scientific research.

At this time, traction cannot be recommended as a single treatment for LBP with or without sciatica.