Evidence for use of an extension-mobilization category in acute low back syndrome: prescriptive validation pilot study.

Up to 90% of patients treated by conservative management of low back syndrome (LBS) remain undiagnosed.[1,2] Traditional diagnostic testing, some of which has improved dramatically in recent times (eg, radiographic imaging tests), has enhanced the clinician's ability to detect abnormalities in the human body. These technological advances are associated with high costs, however, and the questionable sensitivity and specificity of implicating abnormal low back structures as causative to LBS has caused some to question their usefulness and cost-effectiveness.(3-6)

The lack of any identifiable pathologic entities in the majority of patients with LBS has led some authorities to suggest expanding the traditional "pathology model" of health and disease with LBS.[7] Many researchers and practitioners encourage classifying patients with LBS according to signs and symptoms as opposed to diagnosing patients, where a specific mechanistic etiology must be implicated.[8-12] Spitzer[12] argues that, because of the lack of uniformity in pathology-based low back diagnoses, classification of spinal disorders should be based on "clinical entities encountered in practice." A summary of classification schemas is presented in Table, 1.

[TABULAR DATA OMITTED]

With surgical intervention becoming less of an option,[7,9] all of the classification systems guide the clinician toward conservative management for the majority of patients with LBS. Exactly what constitutes conservative management, however, remains elusive. With the exception of Sikorski[10] and McKenzie,[11] conservative care is not specifically addressed in the classification systems in Table 1. Although Sikorski and McKenzie offer specific direction for conservative care, they report successes in descriptive terms and their approaches have not been subjected to rigorous evaluation and peer review.

A classification system can only be justified if (1) a clinician can examine a Patient and then reliably assign the patient to a classification that directs care and (2) implementing the assigned treatment results in more efficacious management of patients than comparative, nonspecific management strategies (prescriptive validity). Previous work has indicated that assignment of patients to treatment-oriented categories is possible by using measurements obtained during a Physical examination.[13] The purpose of our study was to assess whether a priori classification followed by implementation of directed and specific conservative management results in more effective treatment of patients who have LBS than does use of a nonspecific conservative treatment regimen.

Method

Subjects

Patients referred to physical therapy for management of LBS were included in the study. All patients were classified as having an acute episode of LBP (<7 days since onset) or a subacute episode (7 days to 7 weeks) according to the Quebec Task Force guidelines.[12](pS17) Demographic information about the patients is given in Table 2.

Table 2. Demographic and Other Information

                        Experimental       Comparison
                          Group              Group
                         (n = 14)            (n=10)
Age (y)(b)                    37 (12)         27 (10)

Sex
 Male                          7               7
  Female                       7               3

Days since onset of
  low back
  syndrome(c)                  6 (5)          11 (6)
Back pain only                11               8
Leg symptoms(d)                3               2

(a) Standard deviations shown in parentheses
(b) [tau] = 2.269, P<.05.
(c) [tau] = 1.962, P>.05.
(d) Pain on paresthesia below the buttock area.

Procedure

The experimental procedure is outlined in Appendix 1. The details of the experiment were explained to each patient, and those patients consenting to participate signed a written consent form. The patient was then examined by a physical therapist (MTC) and categorized as belonging to either an extension-mobilization group or to some other group. Subjects who were not included in the extension-mobilization group were dismissed from the study. Although the classification system we are developing groups patients with LBS into six different categories, for the purposes of this report, we describe only the extension-mobilization criteria because we used only those criteria in this study.

The variables we used to classify the patients can be divided into two major categories: (1) movement tests in which the response of the patient is noted and (2) physical signs that focus on pelvic alignment. Previous work[13] has documented internal consistency of theta [greater than or equal to] .80 using an examination protocol that included tests used in this study.

Movement Testing

Variables used to classify a patient's response to movement testing are described in Appendix 2. For these Variables, the patient was asked to Concentrate on the amplitude and location of his or her pain and the location of any paresthesia (baseline level). The patient was then asked to perform a specified movement (see Appendix 2 for the sequence) and return to the original starting position (ie, standing, supine, or prone). After the movement, the patient was asked to compare his or her pain or paresthesia with the baseline. Possible responses of the patient were (1) worsens, in which paresthesia was produced or the patient's pain or paresthesia moved distally from the lumbar spine (peripheralized), or, in the case of no paresthesia, the intensity of the pain was at a greater level than baseline; (2) improves, in which paresthesia was abolished or the patient's pain moved from the periphery toward the lumbar spine (centralized), or, in the absence of paresthesia, reported pain levels were lower after the movement; and (3) status quo, in which the patient's pain and paresthesia were not affected by the movements.

A major criterion in determining "worsens" and "improves" in patients with back and leg symptoms is the centralization phenomenon described by McKenzie[11 pp85-93] and Donelson and colleagues,14,15 who relate centralization to prognosis. Our interpretation of McKenzie's description of the movement tests[ll pp85-93] as well as our previous work[16] resulted in our adopting the following decision rule: For patients without referred pain or root symptoms below the buttock, judgments would be made according to the change in lumbar/buttock symptom intensity after the movement. A patient whose symptoms improved (eg, intensity decreased with movement) with at least two variables related to extension movements and worsened with at least one flexion movement was placed in the extension category (eg, patient reported centralization of symptoms during extension in standing and repeated extension in the prone position while reporting peripheralization with flexion in the supine position). For patients with referred pain or root symptoms (pain or paresthesia) below the buttock, we used the centralization phenomenon to judge status. A subject had to improve (achieve centralization) in at least two extension movements and worsen with at least one flexion movement to be placed in the extension category.

Physical Signs

The second category of variables used for classification involves signs that have been related to sacroiliac joint region pain.[17,18] Manipulative techniques, also purportedly directed to the sacroiliac joint, are indicated when such signs are positive.[19] The etiology of pain arising from the sacroiliac joint is an extremely controversial topic.[20] The presence of pain in the sacroiliac region (eg, dull pain over the posterior superior iliac spine [PSIS]), however, is described by many patients and may be explained by a variety of non - joint-related causes.[21] Rather than implicate the sacroiliac joint in the etiology of such pain, we prefer to label the signs and symptoms as indicative of sacroiliac regional pain and, if a composite (a group) of such signs is positive, indicative of a specific manipulative technique.

For the assessment of sacroiliac regional pain, four tests were used and are described in detail elsewhere.[17,18,22] where.[17,18,22] The first test conducted is used to assess heights of the PSISs with the subject in a sitting position. Bilateral comparisons are made, and PSISs of equal heights constitute a negative finding. A high right or left side constitutes a positive finding. A sitting flexion test is conducted next. The subject is initially in a sitting position, and the examiner palpates the PSIS bilaterally. The subject then bends forward, with the examiner continuing to palpate the PSISs. A positive test occurs if an asymmetry in motion is detected between the PSISs during forward flexion. The third test is a supine to long-sitting test. With the subject initially positioned supine, the examiner palpates the medial malleoli bilaterally and notes the relative lower-extremity lengths. The subject then sits up, and the lengths are again compared. The observation of a change in relative lower-extremity length between positions is a positive finding. Our fourth test is a prone knee flexion test. With the patient initially positioned prone and with shoes on, the relative limb lengths are assessed visually. The patient's knees are then passively flexed to approximately 90 degrees, and the lower-extremity lengths are again observed. A change in relative length between the two positions is a positive finding. To place a patient in a manipulation category, we use a composite of three of four positive tests. Past work[22] has shown excellent reliability of such a composite.

Once the patients were placed in the extension-mobilization group, they were randomly assigned to either an experimental group (n=14, matched and specifically directed) or a comparison group (n=10, unmatched and nonspecific). Random assignment was accomplished by flip of a coin.

Treatment

The experimental group was treated first with a mobilization technique purported to affect the sacroiliac join and described previously.[22] Briefly, the technique involved placing the patient supine with the spine laterally flexed to the symptomatic side. The therapist stood on the opposite side of the patient. The patient's hands were clasped behind the neck. The therapist threaded one arm through the patient's clasped hands, rotating the upper trunk toward the therapist. The therapist then placed the free hand on the patient's anterior superior iliac spine (ASIS) that was farthest away from the therapist. The therapist applied a posterior force to the ASIS while the patient maintained full upper per trunk rotation (Fig. 1). This manipulation treatment was followed by an extension-oriented treatment regimen proposed by McKenzie.* Prone press-ups were prescribed and the use of flexed postures discouraged (eg, use of a lumbar roll to discourage lumbar kyphotic postures while sitting).

We used a flexion-oriented exercise regimen, as described by Williams,23 for treatment of the comparison group. The flexion program was used in an attempt to mimic an exercise regimen often prescribed by physicians. In our geographical area of practice, a flexion-oriented handout, usually printed by pharmaceutical companies, is commonly given to almost all patients with LBS seeking health care. This exercise program, therefore, constitutes a generalized, commonly used, nonspecific exercise regimen for the treatment of LBS.

All patient treatments were supervised within the physical therapy setting on a three-times-per-week schedule. At time.1/2, the treatment period continued over a weekend, and in those instances the 3- or 5-day point would actually be a 4- or 6-day period, respectively. In addition, patients were given a handout illustrating the prescribed exercises and encouraged to carry out the exercise regimen at home on at least three occasions per day.

On return visits, the major focus was to assess how the patient was performing the prescribed exercise regimen. First, proficiency in the exercise regimen was checked by a physical therapist by asking the patient to demonstrate the exercises. If the patient performed the exercises appropriately, then minimal supervision was offered. If the patient needed further guidance, this guidance was provided by a physical therapist or a physical therapy aide. Both groups were closely monitored, and by the end of the study, all patients were performing the exercises as instructed.

Outcome Measure

In diseases in which etiology is obscure (eg, LBS), management is almost totally directed at preservation of function and limitation of symptoms.[24] We decided to use a self-report questionnaire as the measure of outcome because of the following advantages: (1) ease of administration, (2) nonparticipation of the treating clinician and therefore not subject to examiner bias, (3) better reflection of various functional activities that would be logistically impossible to replicate in the clinic or to predict from indexes based on physical examination data, and (4) documented reliability.[25] We recognized the limitations of self-reports, including their susceptibility to patient bias. There is, however, good support for self-report measures of health status in clinical investigations involving patients with LBS.[26] Thus, we decided that the advantages of a self-report far outweighed the disadvantages.

We decided to use a disease-specific (ie, LBS-specific) self-report instrument. Patrick and Deyo[27] suggest that disease-specific self-reports are most sensitive to change, especially for short-term outcome studies. The Oswestry Low Back Pain Questionnaire is an easily administered, disease-specific self-report instrument that provides an index of a patient's perceived disability based on 10 areas of limitations in performance.[28] These areas are pain intensity, changes in the status of pain, personal hygiene, lifting, walking, sitting, standing, sleeping, social activity, and traveling. Each section is scored on a six-point scale (0-5), with 0 representing no limitation and 5 representing a maximal limitation. The subscales added together yield a total maximum score of 50. The score is then doubled and interpreted as a percentage of the patient-perceived disability (the higher the score, the greater the disability). Past work,[25] which included the Oswestry questionnaire and other self-reports such as the Million Scale,[29] Roland and Morris' Disability Scale,[30] and the Waddell and Main Disability Index,[31] concluded that a great deal of redundancy (high inter-correlations) characterizes these self-reports. We therefore felt that one index would be sufficient.

Data Analysis

Oswestry questionnaire data were analyzed with a 2 x 3 (treatment group x treatrnent period) analysis of variance (ANOVA), with treatment group a between-group factor and treatment period a within-group factor.[32] As a check on the randomization process, we performed an a posteriori comparison using between-group [tau] tests of the following data: age, onset of LBS, and the initial Oswestry questionnaire scores.

Results

Twenty-four of the 39 patients who initially took part in the study were classified as having signs and symptoms indicating treatment with an extension-mobilization approach. The randomization process resulted in 14 patients being in the experimental group and 10 patients being in the comparison group. Between-group comparisons of the initial Oswestry questionnaire scores and onset of LBS were not significant (t = 1.922 and t = 1.962, respectively; P [greater than or equal to] .05). The comparison group, however, was significantly younger than the experimental group (t = 2.234, P [greater than or equal to] .05). Although these results support both groups as being equivalent at the start of the study, we recognize that because of the small number of subjects in each group, the power (ie, likelihood of finding a difference) for such tests is extremely low.

The ANOVA results are summarized in Table 3. These results indicate main effects of treatment group as well as treatment period. The significant main effects were precluded by a significant treatment group x treatment period interaction (F[1,22] = 4.91, P<.05). Thus, the rate of improvement, as indicated by the Oswestry questionnaire scores, was dependent on the treatment group to which the patient was assigned. This result is illustrated in Figure 2.

[TABULAR DATA OMITTED]

Discussion

The results of this study support our hypothesis that a priori classification of certain patients with LBS followed by a matched, specifically defined conservative management strategy may result in a more effective outcome when compared with an unmatched, nonspecific treatment. The magnitude of change in the Oswestry questionnaire scores was substantial. Fairbanks et al[28] described ranges of Oswestry questionnaire scores and compared the scores with disability ratings (Tab. 4). The average score of the experimental group at the completion of our study was 11.2%, corresponding to a minimal disability rating. The average score of the comparison group was 30%, corresponding to a moderate disability rating. We contend that the results of this study represent not only statistically significant findings, but also that the magnitude of the findings is clinically meaningful, especially in the relatively short time frame of the intervention.

Table 4. Ranges of Oswestry
Questionnaire Scores and Relationship to
Levels of Disability[28]

Score           Rating of Disability

 0 - 20           Minimal disability
21 - 40           Moderate disability
41 - 60           Severe disability
61 - 80           Crippled
                  Bed-bound or exaggerating
81 - 100          symptoms

With such a small sample, there is always the possibility that the groups were not equivalent at the start of the study. We believe that the best assurance for equivalence of groups is the type of random assignment we used. We further assessed the equivalence of groups, however, by conducting a posteriori t tests on the variables of time of onset, initial Oswestry questionnaire scores, and age of the groups. As we have noted, the results suggest that there was no difference between the groups in time of onset as well as initial Oswestry questionnaire scores, but there was a significant difference in age between groups (Tab. 2). We do not, however, see how the age difference between groups could affect our conclusion that the changes in the dependent measure were related to the different treatment approaches.

The classification systems described in Table 1 have different approaches to the patient with LBS. Not surprisingly, the authors have different expectations of their respective classification systems. For example, Mooney[8] takes an approach in which clinical data lead to a decision of whether to operate on a patient, whereas Waddell and Hamblen[9] take a decision-making approach concerning what ancillary testing is needed. In many settings, physicians and surgeons are only the initial contacts for patients with LBS, and once the decisions concerning surgery and ancillary tests are determined, the vast majority of patients are directed toward conservative treatment in which treatment decisions are often independent of physician intervention.

For the clinician who must administer triage to patients with LBS, the classification systems offered by Mooney[8] and Waddell and Hamblen[9] have obvious value. Once the determination to proceed with conservative care is made, however, such classifications are of little help in selecting specific treatments. Clinicians involved in the day-to-day conservative management of patients with LBS presumably rely on signs and symptoms to guide treatment. The results of our study suggest that classification based on signs, symptoms, and response to movement can be used to assist the clinician in deciding which of the many LBS interventions to use.

Numerous researchers and practitioners[33-39] have espoused in the general sense the use of classification and diagnosis in everyday physical therapy Practice. We would argue that there are few clinical entities for which classification has greater importance than for LBS. Diagnosis is virtually impossible for LBS, and conservative treatment is so poorly defined that the selection of treatment strategies has been likened to "taking on the characteristics of a lottery."[10] The results of our study illustrate the usefulness of classification specifically guiding treatment. We believe that refining classification systems will help clarify the role of conservative management in patients with LBS. This could result in the most efficient use of health care dollars as well as maximal ethical benefit to the patient.

Limitations

Although we have attempted to describe our sample in detail, we recognize the sample size was small; thus, generalizability may be limited. There was no control group, and there also was no follow-up. Our purpose was to assess the short-term effectiveness of classification and treatment, and we therefore did not include a follow-up in our research design. In addition, our research design does not permit us to rule out the possibility that the experimental treatment manipulation followed by extension-oriented exercises) may be beneficial to all patients, regardless of their classification. our design also did not allow us to differentiate between the relative effectiveness of the two elements of treatment we used (ie, manipulation and the extension-oriented exercise regimen). We will attempt to address these limitations in our current and future work.

Appendix 1. Outline of Experimental Procedure

1. Patient with low back syndrome referred to physical therapy. 2. Patient agrees to participate in study; completes Oswestry Low Back Pain Questionnaire. 3. Patient undergoes physical examination. 4. Patient is classified as belonging to an extension-mobilization group or to another group. 5. Extension-mobilization group subjects are randomly assigned to an experimental group

(matched, specific treatment) or a comparison group (unmatched, nonspecific); others are

dismissed from study. 6. Assigned treatment is carried out on all patients during their initial visit; each patient is

scheduled for follow-up on a three-times-per-week basis. 7. Patient returns for follow-up on day 3(a) and completes Oswestry questionnaire prior to

treatment; treatment is monitored and assistance is given as needed. 8. Patient returns for follow-up on day 5(a) and completes Oswestry questionnaire prior to

treatment; treatment is monitored and assistance is given as needed.

Appendix 2. Variables Used to Classify Patient Response to Movement Testing(a)

Single Movements

1. Flexion in standing position 2. Flexion in supine position 3. Extension in standing position 4. Extension in prone position

Repeated Movements

1. Pelvic translocation left (standing) 2. Pelvic translocation right (standing) 3. Extension in prone position 4. Flexion is supine position

Sustained Movement

1. Flexion in supine position

(a) Order of testing was as follows: standing, prone, supine. Judgments of worsen, improve, or status quo were made.

Conclusion

We described a procedure for categorizing patients with LBS a priori by use of a treatment-oriented classification scheme that specifically directed management toward mobilization and extension exercises. We compared the relative effectiveness of a specifically directed management strategy with a nonspecific treatment. We conclude that a priori classification of patients with LBS into a treatment category of extension and mobilization and subsequently treating the patient accordingly with specified interventions can be an effective approach to conservative management of patients with LBS.

References

[1] Valkenburg HA, Haanen HCM. The epidemiology of low back pain. In: White AA, Gordon S, eds. Proceedings of the American Academy of Orthopedic Surgeons Symposium on Low Back Pain. St Louis, Mo: CV Mosby Co; 1982: 9-22. [2] Nachemson AL. The natural course of low back pain. in: White AA, Gordon S, eds. Proceedings of the American Academy of Orthopedic Surgeons Symposium on Low Back Pain. St Louis, Mo: CV Mosby Co; 1982:46-51. [3] Fullenlove TM, Williams AJ. Comparative roentgen findings in symptomatic and asymptomatic backs. JAMA. 1957;168:572-574. [4]Magora A, Schwartz A. Relation between the low back pain syndrome and X-ray findings. Scand J Rehabil Med. 1976;8:115-125. [5]Hitselberger WE, Wittin RM. Abnormal myelograms in asymptomatic patients. J Neurosurg. 1984;28:204-206. [6] Wiesel SW, Tsourmas N, Feffer HL, et al. A study of computer assisted tomography: the incidence of positive CAT scans in an asymptomatic group of patients. Spine. 1984;9:549-551. [7] Haldeman S. Presidential Address, North American Spine Society: Failure of the pathology model to predict low back pain. Spine. 1990;15:718-724. [8] Mooney V. The syndromes of low back disease. Orthop Clin North Am. 1983; 14:505-515. [9] Waddell G, Hamblen DL. The differential diagnosis of backache. Procedures in Practice. 1983;227:1167-1175. [10] Sikorski JM. A rationalized approach to physiotherapy for low-back pain. Spine. 1985; 10:571-579. [11] McKenzie RA. The Lumbar Spine: Mechanical Diagnosis and Therapy. Waikanae, New Zealand: Spinal Publications Ltd; 1981. [12] Spitzer WO. Diagnosis of the problem (the problem of diagnosis). Spine. 1987;12(suppl): S16-S21. [13] Delitto A, Shulman AD, Rose SJ, et al. Reliability of a physical examination to classify patients with low back syndrome. Physical Therapy Practice. 1992;1:1-9. [14] Donelson R, Silva G, Murphy K. Centralization phenomenon: its usefulness in evaluating and treating referred pain, Spine, 1990;15:211-213. [15] Donelson R, Grant W, Kamps C, et al. Pain response to sagittal end-range spinal motion: a prospective, randomized, multicentered trial. Spine. 1991;16(suppl):S206-S212. [16] Delitto A, Shulman AD, Rose SJ. On developing expert-based decision support systems in physical therapy: the NIOSH Low Back Atlas. Phys Ther. 1989;69:554-558. [17] NIOSH Low Back Atlas of Standardized Tests and Measurements. Washington, DC, US Dept of Health and Human Services, Public Health Service, Center for Disease Control, National Institute for Occupational Safety and Health, December 1988. [18] MaGee Dj. Orthopedic Physical Assessment. 2nd ed. Philadelphia, Pa: WB Saunders Co; 1992;309-310, 320-321. [19] Porterfield JA, DeRosa C. The sacroiliac joint. In: Gould JA, ed. Orthopedic and Sports Physical Therapy. 2nd ed. St Louis, Mo: CV Mosby Co; 1990:564-573. [20] Frymoyer JW, Gorden SL. New Perspectives on Low Back Pain. Park Ridge, Ill: American Academy of Orthopedic Surgeons; 1989:240-242. [21] McGill SM. A biomechanical perspective of sacro-iliac pain. Clin Biomech. 1987;2:145-151. [22] Cibulka MT, Delitto A, Koldehoff R. Changes in innominate tilt after manipulation of the sacroiliac joint in patients with low back pain: an experimental study. Phys Ther. 1988; 68:1359-1363. [23] Williams PC. Low Back and Neck Pain: Causes and Conservative Treatment. Springfield, Ill: Charles C Thomas, Publisher; 1974; 35-43. [24] Deyo FA Measuring functional outcomes in therapeutic trials for chronic disease. Controlled Clin Trials. 1984;5:223-240. 25 Rose SJ, Shulman AD, Strube MJ. Functional assessment of patients with low back syndrome. Topics in Geriatric Rehabilitation. 1986;1:9-30. [26] Deyo RA, Patrick DL. Barriers to the use of health status measures in clinical investigation, patient care, and policy research. Med Care. 1989;27(suppl):S254-s268. [27] Patrick DL, Deyo RA. Generic and disease-specific measures in assessing health status and quality of life. Med Care. 1989;27(suppl): S217-S232. [28] Fairbanks JCT, Couper J, Davies JB, et al. The Oswestry Low Back Pain Disability Questionnaire. Physiotherapy. 1980;66:271-273. [29] Million R, Hall W, Haavik RD, et al. Assessment of the progress of the back pain patient. Spine. 1982;7:204-211. [30] Roland M, Morris R. A study of the natural history of back pain, 1: development of a reliable and sensitive measure of disability in low-back pain. Spine. 1983;8;141-144. [31] Waddell G, Main CJ. Assessment of the severity in low-back disorders. Spine 1984;9: 204-208. [32] Wilkinson L. SYSTAT: The System for Statistics. Evanston, Ill: SYSTAT Inc; 1988:405-418. [33] Rose SJ. Description and classification: the cornerstones of pathokinesiological research. Phys Ther 1986;66:379-381. [34] Rose SJ. Physical therapy diagnosis: role and function. Phys Ther. 1989;69:535-537. [35] Rose SJ. Musing on diagnosis. Phys Ther. 1988;68:1665. Editorial. [36] Guccione AA. Physical therapy diagnosis and the relationship between impairment and function. Phys Ther. 1991;71:499-503. [37] Jette AM. Commentary on "Physical Therapy Diagnosis and the Relationship Between Impairments and Function." Phys Ther. 1991; 71:503-504. [38] Jette AM, Diagnosis and classification by physical therapists: a special communication. Phys Ther. 1989;69:967-969. [39] Sahrmann SA. Diagnosis by the physical therapist - a prerequisite for treatment: a special communication. Phys Ther. 1988;68:1703-1706.