Chiropractic Management of Low Back Pain and Low Back-Related Leg Complaints: A Literature Synthesis

Chiropractic Management of Low Back Pain and Low Back-Related Leg Complaints: A Literature Synthesis

Dana J. Lawrence, DC, MMedEd, William Meeker, DC, MPH, Richard Branson, DC, Gert Bronfort, DC, PhD, Jeff R. Cates, DC, MS, Mitch Haas, DC, MA, Michael Haneline, DC, MPH, Marc Micozzi, MD, PhD, William Updyke, DC, Robert Mootz, DC, John J. Triano, DC, PhD, Cheryl Hawk, DC, PhD

Dana J. Lawrence, DC, MMedEd,
Senior Director, Center for Teaching and Learning,
Palmer College of Chiropractic, Davenport, IA.

OBJECTIVES:   The purpose of this project was to review the literature for the use of spinal manipulation for low back pain (LBP).

METHODS:   A search strategy modified from the Cochrane Collaboration review for LBP was conducted through the following databases: PubMed, Mantis, and the Cochrane Database. Invitations to submit relevant articles were extended to the profession via widely distributed professional news and association media. The Scientific Commission of the Council on Chiropractic Guidelines and Practice Parameters (CCGPP) was charged with developing literature syntheses, organized by anatomical region, to evaluate and report on the evidence base for chiropractic care. This article is the outcome of this charge. As part of the CCGPP process, preliminary drafts of these articles were posted on the CCGPP Web site www.ccgpp.org (2006-8) to allow for an open process and the broadest possible mechanism for stakeholder input.

RESULTS:   A total of 887 source documents were obtained. Search results were sorted into related topic groups as follows: randomized controlled trials (RCTs) of LBP and manipulation; randomized trials of other interventions for LBP; guidelines; systematic reviews and meta-analyses; basic science; diagnostic-related articles, methodology; cognitive therapy and psychosocial issues; cohort and outcome studies; and others. Each group was subdivided by topic so that team members received approximately equal numbers of articles from each group, chosen randomly for distribution. The team elected to limit consideration in this first iteration to guidelines, systematic reviews, meta-analyses, RCTs, and coh ort studies. This yielded a total of 12 guidelines, 64 RCTs, 13 systematic reviews/meta-analyses, and 11 cohort studies.

CONCLUSIONS:   As much or more evidence exists for the use of spinal manipulation to reduce symptoms and improve function in patients with chronic LBP as for use in acute and subacute LBP. Use of exercise in conjunction with manipulation is likely to speed and improve outcomes as well as minimize episodic recurrence. There was less evidence for the use of manipulation for patients with LBP and radiating leg pain, sciatica, or radiculopathy.

From the FULL TEXT Article

Discussion

A total of 887 source documents were initially obtained. This included a total of 12 guidelines, 64 RCTs, 20 systematic reviews/meta-analyses, and 12 cohort studies. Table 1 provides an overall summary of the number of studies evaluated.

Acute LBP

There was fair evidence that HVLA has better short-term efficacy than mobilization or diathermy and limited evidence of better short-term efficacy than diathermy, exercise, and ergonomic modifications.

Chronic LBP

The HVLA procedure combined with strengthening exercise was as effective for pain relief as nonsteroidal antiinflammatory dugs with exercise. Fair evidence indicated that manipulation is better than physical therapy and home exercise for reducing disability. Fair evidence shows that manipulation improves outcomes more than general medical care or placebo in the short-term and to physical therapy in the long-term. The HVLA procedure had better outcomes than home exercise, transcutaneous electrical nerve stimulation, traction, exercise, placebo and sham manipulation, or chemonucleolysis for disk herniation.

Mixed (Acute and Chronic) LBP

Hurwitz [23] found that HVLA was the same as medical care for pain and disability; adding physical therapy to manipulation did not improve outcomes. Hsieh [24] found no significant value for HVLA over back school or myofascial therapy. A short-term value of manipulation over a pamphlet and no difference between manipulation and McKenzie technique were reported by Cherkin et al. [25] Meade [26, 27] contrasted manipulation and hospital care, finding greater benefit for manipulation over both short-term and long-term. Doran and Newell [28] found that SMT resulted in greater improvement than physical therapy or corsets.

Sham and Alternate Manual Method Comparisons

The study of Hadler [34] balanced for effects of provider attention and physical contact with a first effort at a manipulation sham procedure. Patients in the group that entered the trial with greater prolonged illness at the outset were reported to have benefited from the manipulation. Similarly, they improved faster and to a greater degree (QS, 62.5). Hadler [34, 35] demonstrated that there was a benefit for a single session of manipulation compared to a session of mobilization (QS, 69). Erhard [36] reported that the rate of positive response to manual treatment with a hand-heel rocking motion was greater than with extension exercises (QS, 25). Von Buerger [37] examined the use of manipulation for acute LBP, comparing rotational manipulation to soft tissue massage. He found that the manipulation group responded better than the soft tissue group, although the effects occurred mainly in the short-term. The results were also hampered by the nature of the forced multiple choice selections on the data forms (QS, 31). Gemmell [38] compared 2 forms of manipulation for LBP of less than 6 weeks of duration as follows: Meric adjusting (a form of HVLA) and Activator technique (a form of mechanically assisted HVLA). No difference was observed, and both helped to reduce pain intensity (QS, 37.5). MacDonald [39] reported a short-term benefit in disability measures within the first 1 to 2 weeks of starting therapy for the manipulation group that disappeared by 4 weeks in a control group (QS, 38). The work of Hoehler, [40] although containing mixed data for patients with acute and chronic LBP, is included here because a larger proportion of patients with acute LBP were involved in the study. Manipulation patients reported immediate relief more often, but there were no differences between groups at discharge (QS, 25).

Sham and Alternate Manual Methods

Triano [58] found that SMT produced significantly better results for pain and disability relief for the short-term, than did sham manipulation (QS, 31). Cote [68] found no difference over time or for comparisons within or between the manipulation and mobilization groups (QS, 37.5). The authors posed that failure to observe differences may have been due to low responsiveness to change in the instruments used for algometry, coupled with a small sample size. Hsieh [24] found no significant value for HVLA over back school or myofascial therapy (QS, 63). In the study by Licciardone, [69] a comparison was made between osteopathic manipulation (which includes mobilization and soft tissue procedures as well as HVLA), sham manipulation, and a no-intervention control for patients with chronic LBP. All groups showed improvement. Sham and osteopathic manipulation were associated with greater improvements than seen in the no-manipulation group, but no difference was observed between the sham and manipulation groups (QS, 62.5). Both subjective and objective measures showed greater improvements in the manipulation group compared to a sham control, in a report by Waagen [70] (QS, 44). In the work of Kinalski, [71] manual therapy reduced the time of treatment of patients with LBP and concomitant intervertebral disk lesions. When disk lesions were not advanced, a decreased muscular hypertonia and increased mobility was noted. This article, however, was limited by a poor description of patients and methods (QS, 0).

Harrison et al [72] reported a nonrandomized cohort controlled trial of treatment of chronic LBP consisting of 3-point bending traction designed to increase curvature of the lumbar spine. The experimental group received HVLA for pain control during the first 3 weeks (9 treatments). The control group received no treatment. Follow-up at a mean of 11 weeks showed no change in pain or curvature status for controls but a significant increase in curvature and reduction of pain in the experimental group. Average number of treatments to achieve this result was 36. Long-term follow-up at 17 months showed retention of benefits. No report of relationship between clinical changes and structural change was given.

Haas and colleagues [73] examined the dose-response patterns of manipulation for chronic LBP. Patients were randomly allocated to groups receiving 1, 2, 3, or 4 visits per week for 3 weeks, with outcomes recorded for pain intensity and functional disability. A positive and clinically important effect of the number of chiropractic treatments on pain intensity and disability at 4 weeks was associated with the groups receiving the higher rates of care (QS, 62.5). Descarreaux et al [74] extended this work, treating 2 small groups for 4 weeks (3 times per week) after 2 baseline evaluations separated by 4 weeks. One group was then treated every 3 weeks; the other did not. Although both groups had lower Oswestry scores at 12 weeks, at 10 months, the improvement only persisted for the extended SMT group.

What Are the Relevant Outcome Measures?

The Clinical Practice Guidelines formulated by the Canadian Chiropractic Association and the Canadian Federation of Chiropractic Regulatory Boards [103] note that there are a number of outcomes that may be used to demonstrate change as a result of treatment. These should be both reliable and valid. According to the Canadian guidelines, appropriate standards are useful in chiropractic practice because they are able to perform the following:

• consistently evaluate the effects of care over time;

• help indicate the point of maximum therapeutic improvement;

• uncover problems related to care such as noncompliance;

• document improvement to the patient, doctor, and third parties;

• suggest modifications of the goals of treatment if necessary;

• quantify the clinical experience of the doctor;

• justify the type, dose, and duration of care;

• help provide a database for research; and

• assist in establishing standards of treatment of specific conditions.

The broad general classes of outcomes include functional outcomes, patient perception outcomes, physiologic outcomes, general health assessments, and subluxation syndrome outcomes. This chapter addresses only functional and patient perception outcomes assessed by questionnaires and functional outcomes assessed by manual procedures.

Functional Outcomes

These are outcomes that measure the patient’s limitations in going about his or her normal daily activities. What is being looked at is the effect of a condition or disorder on the patient (ie, LBP, for which a specific diagnosis may not be present or possible) and its outcome of care. Many such outcome tools exist. Some of the better known include the following:

• Roland Morris Disability Questionnaire,

• Oswestry Disability Questionnaire,

• Pain Disability Index,

• Neck Disability Index,

• Waddell Disability Index, and

• Million Disability Questionnaire.

These are only some of the existing tools for assessing function.

In the existing RCT literature for LBP, functional outcomes have been shown to be the outcome that demonstrates the greatest change and improvement with SMT. Activities of daily living, along with patient self-reporting of pain, were the 2 most notable outcomes to show such improvement. Other outcomes fared less well, including trunk range of motion (ROM) and straight leg raise.

In the chiropractic literature, the outcome inventories used most frequently for LBP are the Roland Morris Disability Questionnaire and the Oswestry Questionnaire. In a study in 1992, Hsieh [104] found that both tools provided consistent results over the course of his trial, although the results from the 2 questionnaires differed.

Conclusion

Existing research evidence regarding the usefulness of spinal adjusting/manipulation/mobilization indicates the following:

1. As much or more evidence exists for the use of SMT to reduce symptoms and improve function in patients with chronic LBP as for use in acute and subacute LBP.

2. Use of exercise in conjunction with manipulation is likely to speed and improve outcomes as well as minimize episodic recurrence.

3. There was less evidence for the use of manipulation for patients with LBP and radiating leg pain, sciatica, or radiculopathy.

4. Cases with high severity of symptoms may benefit by referral for comanagement of symptoms with medication.

5. There was little evidence for the use of manipulation for other conditions affecting the low back and very few articles to support a higher rating.

Exercise and reassurance have been shown to be of value primarily in chronic LBP and low back problems associated with radicular symptoms. A number of standardized, validated tools are available to help capture meaningful clinical improvement over the course of low back care. Typically, functional improvement (as opposed to simple reported reduction in pain levels) may be clinically meaningful for monitoring responses to care. The literature reviewed remains relatively limited in predicting responses to care, tailoring specific combinations of intervention regimens (although the combination of manipulation and exercise may be better than exercise alone), or formulating condition-specific recommendations for frequency and duration of interventions. Table 2 summarizes the recommendations of the team, based on the review of the evidence.