CHIROPRACTIC CARE FOR LOW BACK PAIN: CLINICAL PRACTICE GUIDELINE

Between March 2014 through July 2014, we conducted a systematic review to update the literature published since the previous CCGPP guideline was developed. The search included articles that were published between October 2009 through February 2014. Our question was, “What is the effectiveness of chiropractic care including spinal manipulation for nonspecific low back pain?” Table 1 summarizes the eligibility criteria for the search.

The following databases were included in the search: PubMed, Index to Chiropractic Literature, CINAHL, and MANTIS. Details of the strategy for each database are provided in Figure 1. Articles and abstracts were screened independently by 2 reviewers. Data were not further extracted.

We evaluated articles using the Scottish Intercollegiate Guideline Network checklists (www.sign.ac.uk/methodology/checklists.html) for randomized controlled trials (RCTs) and systematic reviews/meta-analyses. For guidelines, the AGREE 2013 instrument35 was used. At least 2 of the 3 investigators conducting the review (CH, SW, MK) reviewed each article. If both reviewers rated the study as either high quality or acceptable, it was included for consideration; if both reviewers rated it as unacceptable, it was removed. For AGREE, we considered “unacceptable” to be a sum of <4. If there was disagreement between reviewers, a third also reviewed the article, and the majority rating was used.

This search yielded 270 articles. Screening the articles for eligibility resulted in 18 articles included for evaluation, as detailed in Figure 2, using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart.36

Of the 18 articles included after screening, 16 were retained as acceptable/high quality12, 17, 37, 38, 39, 40, 41, 42,43, 44, 45, 46, 47, 48, 49, 50 and 251, 52 (both systematic reviews) were excluded as being of unacceptable quality according to the Scottish Intercollegiate Guideline Network checklist. Those with new relevant information were summarized and provided to the Delphi panel as background information. Table 2 lists the articles by lead author and date, and the topic addressed, if new findings were present.

Along with the literature summary, seed documents were comprised of the 3 previous CCGPP guidelines9, 10, 12; links were provided to full text versions. The original guidelines had been developed based on the evidence, including guidelines and research available at the time.16, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63 The steering committee, composed of authors on these previous guidelines, developed 51 seed statements based on the background documents, revising the previous statements if it seemed advisable based on the literature. The steering committee did not conduct a formal consensus process; however, the seed statement development was a team effort, with changes only made if all members of the steering committee were in agreement. Before conducting this project, these seed statements had gone through a local Delphi process among clinical and academic faculty at Logan University as part of their development of care pathways for their clinical faculty. This was done to assess the readability of the seed statements to a group of practicing clinicians. In the Delphi process, 7 statements were slightly modified from the original, and none of those changes were substantive, but rather for purposes of clarification. Consensus was reached for the seed document, which was then adopted by that institution for use in its teaching clinics. That document formed the seed document for the current project. For the Delphi rounds, the 51 statements were divided into 3 sections to be less onerous for the panelists to rate in a timely manner.

Panelists who served on the 3 previous consensus projects10, 11, 12 related to LBP management were invited to participate. Steering committee members made additional recommendations for experts in management of LBP who were not DCs to increase multidisciplinary input. There were 37 panelists; 33 were DCs, one of whom had dual licensure—DC and massage therapist. The 4 non-DC panelists consisted of an acupuncturist who is also a medical doctor, a medical doctor (orthopedic surgeon), a massage therapist, and a physical therapist. Thirty-three of the 37 panelists were in practice (89%); the mean number of years in practice was 27. Seventeen were also affiliated with a chiropractic institution (46%), with 2 of these associated with Logan University; 3 were affiliated with a different health care professional institution (8%); and 1 was employed with a government agency. Because this guideline focuses primarily on chiropractic practice in the United States, geographically, all panelists were from the United States, with 19 states represented. These were Arizona (1), California (4), Florida (3), Georgia (3), Hawaii (2), Iowa (2), Illinois (3), Kansas (1), Michigan (1), Minnesota (1), Missouri (3), North Carolina (1), New Jersey (2), New York (5), South Carolina (1), South Dakota (1), Texas (1), Virginia (1), and Vermont (1). Of the 33 DCs, 21 (64%) were members of the American Chiropractic Association, 2 (6%) were members of the International Chiropractors Association, and 10 (30%) did not belong to any national chiropractic professional organization.

The consensus process was conducted by e-mail. For purposes of analyzing the ratings and comments, panelists were identified by an ID number only. The Delphi panelists were not aware of other panelists’ identity during the duration of the study. As in our previous projects, we used the RAND-UCLA methodology for formal consensus.64

This methodology uses an ordinal scale of 1-9 (highly inappropriate to highly appropriate) to rate each seed statement. RAND/UCLA defines appropriateness to mean that expected patient health benefits are greater than expected negative effects by a large enough margin that the action is worthwhile, without considering costs.64

After scoring each Delphi round, the project coordinator provided the medians, percentages, and comments (as a Word table) to the steering committee. They reviewed all comments and revised any statements not reaching consensus as per these comments. The project coordinator circulated the revised statements, accompanied by the deidentified comments, to the Delphi panel for the next round.

We considered consensus on a statement’s appropriateness to have been reached if both the median rating was 7 or higher and at least 80% of panelists’ ratings for that statement were 7 or higher. Panelists were provided with space to make unlimited comments on each statement. If consensus could not be reached, it was planned that minority reports would be included.

As per recommendations for guideline development such as AGREE, we invited public comment on the draft CPG. This was accomplished by posting the consensus statement on the CCGPP Web site. Press releases and direct e-mail contacts announced a 2-week public comment period, with comments collected via an online Web survey application. Organizations and institutions who were contacted included the following: all US chiropractic colleges; members of all chiropractic state organizations; state boards of chiropractic examiners; chiropractic practice consultants; chiropractic attorneys; chiropractic media (including 1 publication sent to all US-licensed DCs); and chiropractic vendors, whose contacts also included interested laypersons. The steering committee then crafted additional or revised statements as per the comments collected through this method, and these statements were then recirculated through the Delphi panel until consensus was reached.

For scoring purposes, ratings of 1-3 were collapsed as “inappropriate,” 4-6 as “uncertain,” and 7-9 as “appropriate.” If a panelist rated a statement as “inappropriate,” he or she was instructed to articulate a specific reason and provide a citation from the peer-reviewed literature to support it, if possible. The project coordinator entered ratings into a database (SPSS v. 22.0, Armonk, NY: IBM Corp, 2013).

Most acute pain, typically the result of injury (micro- or macrotrauma), responds to a short course of conservative treatment (Table 3). If effectively treated at this stage, patients often recover with full resolution of pain and function, although recurrences are common. Delayed or inadequate early clinical management may result in increased risk of chronicity and disability. Furthermore, those responding poorly in the acute stage and those with increased risk factors for chronicity must also be identified as early as possible.

Clinicians must continually be vigilant for the appearance of clinical red flags that may arise at any point during patient care. In addition, biopsychosocial factors (also known as clinical yellow flags) should be identified and addressed as early as possible as part of a comprehensive approach to clinical management.

Chiropractic doctors are skilled in multiple approaches of functional assessment and treatment. Depending on the clinical complexity, DCs can work independently or as part of a multidisciplinary team approach to functional restoration of patients with acute and chronic LBP.

It is the ultimate goal of chiropractic care to improve patients’ functional capacity and educate them to accept independently the responsibility for their own health.

Informed consent is the process of proactive communication between a patient and physician that results in the patient’s authorization or agreement to undergo a specific medical intervention. Informed consent should be obtained from the patient and performed within the local and/or regional standards of practice. The DC should explain the diagnosis, examination, and proposed procedures clearly and simply and answer patients’ questions to ensure that they can make an informed decision about their health care choices. He or she should explain material risks* of care along with other reasonable treatment options, including the risks of no treatment. (*Note: The legal definition of material risk may vary state by state.)

Thorough history and evidence-informed examination procedures are critical components of chiropractic clinical management. These procedures provide the clinical rationale for appropriate diagnosis and subsequent treatment planning.

Assessment should include but is not limited to the following38:

• Health history (eg, pain characteristics, red flags, review of systems, risk factors for chronicity)
• Specific causes of LBP (eg, aortic aneurysm, inflammatory disorders)
• Examination (eg, reflexes, dermatomes, myotomes, orthopedic tests)
• Diagnostic testing (indications) for red flags (eg, imaging and laboratory tests)

Routine imaging or other diagnostic tests are not recommended for patients with nonspecific LBP.55

Imaging and other diagnostic tests are indicated in the presence of severe and/or progressive neurologic deficits or if the history and physical examination cause suspicion of serious underlying pathology.55

Patients with persistent LBP accompanied by signs or symptoms of radiculopathy or spinal stenosis should be evaluated, preferably, with magnetic resonance imaging or computed tomography.55

Imaging studies should be considered when patients fail to improve following a reasonable course of conservative care or when there is suspicion of an underlying anatomical anomaly, such as spondylolisthesis, moderate to severe spondylosis, posttrauma with worsening symptomatogy (consider imaging, referral, or co-management) with evidence of persistent or increasing neurological (ie, reflex, motor, and/or sensory) compromise, or other factors which might alter the treatment approach. Lateral view flexion/extension studies may be warranted to assess for mechanical instability due to excessive intervertebral translation and/or wedging. Imaging studies should be considered only after careful review and correlation of the history and examination.65

Conditions of illness and injury are typically classified by severity and/or duration. Common descriptions of the stages of illness and injuries are acute, subacute, chronic, and recurrent, and further subdivided into mild, moderate, and severe.

• Acute—symptoms persisting for less than 6 weeks.
• Subacute—symptoms persisting between 6 and 12 weeks.
• Chronic—symptoms persisting for at least 12 weeks’ duration.
• Recurrent/flare-up—return of symptoms perceived to be similar to those of the original injury at sporadic intervals or as a result of exacerbating factors.

Although most patients respond within anticipated time frames, frequency and duration of treatment may be influenced by individual patient factors or characteristics that present as barriers to recovery (eg, comorbidities, clinical yellow flags). Depending on these individualized factors, additional time and treatment may be required to observe a therapeutic response. The therapeutic effects of chiropractic care/treatment should be evaluated by subjective and/or objective assessments after each course of treatment (see “Outcome Measurement”).

Recommended therapeutic trial ranges are representative of typical care parameters. A typical initial therapeutic trial of chiropractic care consists of 6 to 12 visits over a 2- to 4-week period, with the doctor monitoring the patient’s progress with each visit to ensure that acceptable clinical gains are realized (Table 3).

For acute conditions, fewer treatments may be necessary to observe a therapeutic effect and to obtain complete recovery. Chiropractic management is also recommended for various chronic low back conditions where repeated episodes (or acute exacerbations) are experienced by the patient, particularly when a previous course of care has demonstrated clinical effectiveness and reduced the long-term use of medications.

To be consistent with an evidence-based approach, DCs should use clinical methods that generally reflect the best available evidence, combined with clinical judgment, experience, and patient preference. For example, currently, the most robust literature regarding manual therapy for LBP is based primarily on high-velocity, low-amplitude (HVLA) techniques, and mobilization (such as flexion-distraction).17, 20, 66 Therefore, in the absence of contraindications, these methods are generally recommended. However, best practices for individualized patient care, based on clinical judgment and patient preference, may require alternative clinical strategies for which the evidence of effectiveness may be less robust.

The treatment recommendations that follow, based on clinical experience combined with the best available evidence, are posited for the “typical” patient and do not include risk stratification for complicating factors. Complicating factors are discussed elsewhere in this document.

An initial course of chiropractic treatment typically includes 1 or more “passive” (ie, nonexercise) manual therapeutic procedures (ie, spinal manipulation or mobilization) and physiotherapeutic modalities for pain reduction, in addition to patient education designed to reassure and instill optimal strategies for independent management.

Although the evidence reviewed does not generally support the use of therapeutic modalities (ie, ultrasonography, electrical stimulation, etc) in isolation,67 their use as part of a passive-to-active care multimodal approach to LBP management may be warranted based on clinician judgment and patient preferences. Because of the scarcity of definitive evidence,68 lumbar supports (bracing/taping/orthoses) are not recommended for routine use, but there may be some utility in both acute and chronic conditions based upon clinician judgment, patient presentation, and preferences. Caution should be exercised as these orthopedic devices may interfere with conditioning and return to regular activities of daily living (ADLs).

The initial visits allow the doctor to explain that the clinician and the patient must