A treatment-based classification approach to low back syndrome: identifying and staging patients for conservative treatment.

[Delitto A, Erhard RE, Bowling RW. A treatment-based classification approach to low back syndrome: identifying and staging patients for conservative treatment. Phys Ther. 1995;75:470-489.]

Key Words: Backache, Classification, Decision making, Diagnosis.

Low back syndrome (LBS), although self-limiting in most cases, leads in a small percentage of patients to chronic problems that can be very costly to manage,[1,2] and those cases that resolve are prone to recurrence at a rate of Up to 90%.[3] Contributing to the difficulty in managing low back-related disorders is the inability to identify a causative agent, thus leaving the vast majority of patients with LBS without a specific diagnosis,[4] resulting in a condition that has been described as an "illness in search of a disease."[5]

Without a specific diagnosis, the limitations of the traditional "pathology-based model," which implies symptoms should be proportional to organ pathology, become apparent.[6] Deyo[7] suggests that instead of a clinician searching for the pathological cause of LBS, three basic questions should be answered during the initial clinic visit: (1) Is there systemic or visceral disease underlying the pain? (2) Is there evidence of neurologic compromise that represents a surgical emergency? and (3) Are there findings that influence the choice of conservative therapies? Data from the history and physical examination presumably lead to answers to all three questions. Clinicians facing patients with LBS on a day-to-day basis, in our opinion, rely primarily on patient reports of signs, symptoms, and symptom-related behavior. Although appearing on the surface to be "subjective" in nature, the reliability of data related to signs and symptoms has been demonstrated in a variety of clinical situations.[8-11]

How information related to signs and symptoms is interpreted and combined to guide the clinicians in conservatively treating patients is unclear. Some have argued (without the benefit of data) that decisions regarding conservative management can be made according to dusters or patterns of data obtained during clinical testing.[12-15] When tested, however, others have shown that decisions regarding clinical clustering of data do not show a consistent pattern.[16] Such equivocal results lend credence to those who describe the decision regarding conservative intervention for LBS as "taking on the characteristics of a lottery."[17]

The purpose of this clinical perspective is to outline an examination approach for patients with acute LBS that leads to a classification that specifically directs conservative management of LBS. That is, this examination approach leads to a classification that will result in a management strategy that is detailed with regard to the precise type of treatment (ie, mobilization, extension movement, traction) to be prescribed to the patient, and not relegated to nonspecific terminology where any number of conservative strategies can be used for one classification (ie, "exercise," "active therapy").

The variables assessed in this clinical examination can be divided into the major categories of symptoms and signs. These categories are (1) historical data related to musculoskeletal as well as possible nonmusculoskeletal causes of low back symptoms; (2) movement tests where the response of the patient (in terms of behavior of signs and symptoms) is noted; and (3) signs that consist of palpatory tests, measurements, and observations that focus on alignment of body structures.

Development and Background of the Examination Procedure

Initially in developing the management scheme, we relied on input from approximately a dozen practitioners, including physical therapists, chiropractors, and physicians, to arrive at a list and description of tests and measures that would comprise a comprehensive history and physical examination for a patient with LBS.[18] Pertinent tests have subsequently been studied in a reliability investigation, and either the tests were shown to have adequate reliability (reliability coefficients [greater than or equal to].70) or judgments of reliability were suspended due to inadequate distribution of item scores.[19] Subsequent secondary analyses of the data have demonstrated adequate internal consistency rehability coefficients > .80) for the examination process.[8]

The initial sources that described the testing procedures can be traced back to writings and teachings of various "authorities" in orthopedic physical therapy, chiropractics, and medicine.[18] The descriptions of tests and measures from virtually all initial sources were lacking, however, leading our group to treat such descriptions as first approximations. We further operationalized the exact procedures for the specific tests. In addition, when data obtained from testing procedures purportedly led to treatment direction, we found a void in any clear decision rules within the writings of "authorities." Again, we have attempted to operationalize such decision rules within this article. Although some of the tests and procedures discussed in this article have been subjected to peer-reviewed investigation, we would remind the reader that much of the decision-making rules that we propose have not been tested through prospective research.

Overview of Management Approach

The structure of our approach is assessment of clinical data leading to a series of classifications that specifically direct the management of the patient with LBS. For the purpose of this text, low back syndrome is defined as a clinical entity that is characterized by the occurrence or presence of one or more of the following signs or symptoms: (1) pain in the area of the lumbosacral spine, buttock, or referred to the thigh area to the knee but thought to be of spinal origin; (2) pain, paresthesia, or other changes in cutaneous sensation located in the leg or foot area but believed to be of spinal origin radicular symptoms); or (3) alterations in reflexes or loss of motor function in the lower extremities, again from spinal origin (radicular signs).

The result of LBS is usually impaired function in daily activities. Along with others,[20,21] we recognize that the reported severity and resultant disability from LBS may often be influenced by non-movement-related factors (eg, magnified illness behavior and other psychosocial factors). The proposed management approach is designed to help clinicians recognize when these factors may be playing a critical role in the patient's symptoms and to direct the primary care clinician to refer the patient to another practitioner or to seek consultation of another professional. Only specific intervention of movement-related management strategies are included.

First-Level Classification

The first level of classification is determining whether the patient's care (1) can be managed independently and primarily by physical therapy, (2) cannot be managed by a physical therapist and instead requires referral to another health care practitioner, or (3) may be managed by a physical therapist but requires consultation with another health care practitioner (Fig. 1). In making this first-level classification, we recognize that with direct access in many states, patients with serious pathology (eg, metastatic cancer) can complain of spinal pain to a physical therapist, and the physical therapist, as a first-contact practitioner, must be able to recognize such patients and refer them to appropriate health care professionals.[22] Second, there is good evidence that in addition to physical problems, patients with LBS have psychological distress and socioeconomical concerns that can play a role in their symptoms and resultant disability.[23] We contend that the physical therapist, who is primarily trained to address physical dimensions of LBS (eg, the physical impairments), must recognize when the symptoms related to LBS have a substantial psychosocial or socioeconomic dimension and refer the patient or consult with another health care professional who has training in addressing this dimension.

In both the case of severe pathology and high psychological distress, patients may relate their symptoms in such a way that the clinician is left with an impression that the patient is experiencing a high degree of illness behavior (abnormally high response of the patient to the underlying disease process). We use three forms to assist with this level of classification: (1) a medical screening questionnaire that, in addition to screening for serious pathology, includes "nonorganic" descriptions of pain; (2) a modified Oswestry questionnaire; and (3) a pain diagram/pain scale. We do not rely on any one of the tools listed without obtaining further confirmatory data within the history and the physical examination. For example, patients may check numerous "red flags" as present, such as recent weight loss. The clinician does not immediately refer the patient to a primary care physician to rule out metastatic disease. Instead, the information checked on the questionnaire is noted by the clinician, who then further questions the patient in the history about the nature of the weight loss and makes a judgment about whether further workup is necessary.

Medical Questionnaire

The medical questionnaire is presented in Figure 2. The questionnaire is divided into two components: (1) questions that are intended to unveil medical problems that may contraindicate treatment and 2) questions that are intended to identify "nonorganic" pain behavior. The latter questions have been italicized for illustration in Figure 2. Italics are not used on the actual screening form.

If an affirmative response to any of the first group of medical screening questions is given, further information must be obtained. We believe our approach is conservative in that if we have any doubt that symptoms are from a serious pathology of musculoskeletal or nonmusculoskeletal origin, then we refer the patient to another health care professional. Affirmative responses to the latter group of questions, however, indicate the possibility of "magnified illness behavior," and, if these behaviors are confirmed through subsequent examination, we believe a multidisciplinary management regimen is needed.

Modified Oswestry Questionnaire

We use a modified version of the Oswestry Low Back Disability Questionnaire in the initial classification procedure as well as for the documentation of outcome.[24,25] The Oswestry questionnaire is a disease-specific (eg, LBS-specific) self-report. In addition to serving as a guide for initial decision making regarding classifications, we also find the information from the Oswestry questionnaire to be useful for indicating patient outcome, especially for short-term outcome studies. [26-28] The results from the Oswestry questionnaire have been reported to be highly reliable (Pearson r and intraclass correlation coefficients >.90) in its original[24] as well as modified forms.[25]

The Oswestry questionnaire is an easily administered self-report that results in an index of a patient's perceived disability based on 10 areas of limitations in performance. The areas covered are pain intensity; personal hygiene; lifting; walking; sitting; standing; sleeping; social activity; traveling; and, depending on the version used, sex life, changing status of pain, or employment/homemaking duties. Each section is scored on a six-point scale (0-5), with 0 representing no limitation and 5 representing maximal limitation. The subscales together add up to 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).

Our patients complete the Oswestry questionnaire while in the waiting room so that it is available when we take the history. The questionnaire takes less than 5 minutes to score, and we believe clinicians can use it to approximate the patients' perceived functional limitations and overall disability.

Interpretation of Scoring for First-Level Classification

Our interpretation of Oswestry scores greater than 75 is that patients may have a substantial nonmovement component to their symptoms, and this is usually manifested by emergencies that require hospitalization or by magnified illness behavior. A score of 75 or higher will include near-maximal scores in the subsections (ie, the pain is constant and severe, I cannot sit at all, I cannot stand at all I must lie down to travel), indicating the patient perceives his or her low back trouble is extremely activity limiting. If the patient is offering an accurate indication of his or her physical limitations and disability, then the clinician must be attentive to the possibility that severe underlying pathology may be causing the high degree of discomfort. Bias is a limitation of self-reports, however, so the accuracy of the patient's perception must be taken into account, and at times patients may overstate their limitations (eg, the patient whose illness behavior is amplified). We occasionally see patients early after an acute episode of back pain (eg, within a few days) who score greater than 75 and whose subsequent examination confirms the severity of the condition. In the remainder of cases where the initial Oswestry questionnaire score is greater than 75 but the patient does not demonstrate overt signs of acute distress, we begin to question whether management that totally focuses on movement-related strategies is appropriate.

At the other end of the spectrum are Oswestry questionnaire scores below 30. We believe these scores indicate patients who do not have acute conditions. The focus of further evaluation will then be directed in a direction distinct from what occurs for those with acute conditions.

Pain Scale

We use a pain scale and request that the patient rate his or her present pain (or other symptom intensity) on a scale of 0 to 10. Such visual analog scales have been shown to be repeatable in clinical settings.[29] The patient is also asked to rate his or her pain at its greatest and least intensities over the past 24 hours. in interpreting the pain scale, we first look for very high ratings (eg, [greater than or equal to]8/10) on all three judgments (present, greatest, and least in past 24 hours), a factor that for us would be indicative of serious pathology (severe, unremitting pain not affected by change in posture). Our basis for this statement is that for patients with mechanical LBS, there is usually at least one posture (eg, lying down) that will relieve pain to some degree. Second, we look at the intensity of the present pain level and make a judgment as to whether it appears to match our impression of the patient's distress level. For example, a person who fills out the pain scale indicating that present pain level is [greater than or equal to]8/10 should also, in our opinion, exhibit obvious distress, especially when the patient is asked to perform examination procedures such as forward bending or to remove his or her shoes and socks.

Pain Diagram

The patient is asked to complete a body chart diagram depicting the area of pain and other symptoms. We categorize pain diagram using the method reported by Chan et al,[30] as follows: (1) organic, (2) possibly organic, (3) possibly nonorganic, and (4) nonorganic.

Clarifying Information Gathered From Questionnaires Through the History

The questionnaires only serve to give us a first approximation of a patient's distress level and other issues related to their health status. We use the questionnaires to help us focus our further examination. Any of the data suggesting serious pathology or magnified illness behavior are further scrutinized by acquisition of the patient's history.

Serious pathology. Indicators of serious pathology are illustrated in Figure 3. First, the patient is questioned regarding his or her ability to sleep without disturbance. Our focus is on the exact manner in which sleep is disturbed, and we attempt to distinguish whether the patient may be unable to fall asleep because of the pain or may be awakened by pain during a movement, such as turning over, and fall back to sleep. The most troubling of these possibilities is awakening by pain not related to posture or position, especially if the patient also has difficulty-falling asleep. If the patient describes a sleep disturbance of this nature, we contend serious pathology must be suspected.

When a patient has reported any alteration in bladder or bowel function, particularly due to loss of sphincter control, or has reported paresthesia in the saddle area or fourth sacral dermatome, a surgical consultation should be sought without delay. These may be indicators of a central disk protrusion with compression of the fourth sacral nerve root. Failure to relieve this compression surgically may lead to permanent neurological damage, with resultant loss of control of the bowel and bladder sphincters.[31]

Questioning is required when any change in body weight has recently occurred. We focus on the extent and direction of the change, the reason for the change, and the time period over which the change has occurred. We are concerned when there is any unexplained weight loss in individuals who have probably been more sedentary than usual. We strongly believe that this change in body weight must be recognized as a warning signal of serious pathology (eg, a carcinoma). In our opinion, a loss of 4.5 kg (10 lb) of body weight for no apparent reason in a 59-kg (130-1b) patient would have more sinister implications than an 18.1-kg (40-1b) loss in a 113.4-kg (250-lb) patient who has been on a diet.

The patient is asked about any recent episodes of fever. If fever has occurred, the clinician is obligated to rule out systemic infection as a possible source of the patient's low back symptoms (eg, urinary tract infections) through additional questioning during the history and testing during the physical examination. If infection cannot be ruled out as a possible cause of the patient's symptoms, then we believe referral to another practitioner is indicated.

Finally, if the patient has had a traumatic onset of pain, we contend that an adequate radiological examination must be performed. If this has been done, as is most often the case, the examiner should at least obtain the report of these tests if the radiographs cannot be viewed. If radiological tests have not been performed, the referring physician should be consulted to determine whether radiological examination is appropriate.

Magnified illness behavior. Waddell[32] has described an illness-based clinical model for managing LBS in which the interaction of a physical problem with a patient's high degree of psychological distress produces what he has termed "magnified illness behavior." Waddell and colleagues[33,34] have described what they term "nonorganic" descriptors (symptoms) and signs of magnified illness behavior. We have included the descriptors in the medical questionnaire (Fig. 2; italicized questions). If we find that more than three of these descriptors are used by the patient, we include the tests for nonorganic signs in our physical examination. The nonorganic signs are categorized by Waddell et al[33] as follows: (1) simulation, which includes the tests of simulated rotation and axial loading; (2) distraction, which includes gathering signs of inconsistency within the physical examination such as noting a disparity between a supine straight-leg-raising and seated knee extension tests, where the ability of the patient to assume a position of a flexed hip with an extended knee is presumed to be similar; (3) tenderness, which includes exaggerated responses to such tests as very fight palpation of the spine; (4) generalized signs of weakness or sensory loss that do not fit a dermatomal or myotomal distribution; and (5) overreaction, which is a general impression of an exagerrated response from all portions of the physical examination.

In screening for serious medical conditions and magnified illness behavior, the questionnaires that we use are designed to detect the possibility of these conditions. Through interactive history taking and detailed physical examination, we further evaluate such issues. if present, referral to or consultation of another practitioner is indicated. In cases that will continue to be managed by the physical therapist, the next level of classification that we propose is to stage the patient according to the acuteness of the injury.

Second-Order Classification Staging the Patient

We have found it convenient to classify movement disorders into stages based on the acuteness of the injury. We are not alone in recognizing the usefulness of assigning patients into classifications related to acute, "subacute," and chronic.[35] Most Classifications of acuteness of injury are based on the number of days since the injury. We believe that cutoffs for categorizations in this manner are arbitrarily set and are not always useful in directing conservative care.

Although we agree that chronicity of the LBS certainly guides treatment, it has been demonstrated that most indices based on arbitrary durations are not always useful.[36] Consider, for example, a patient with LBS who complains of severe symptoms but who has had an onset of greater than 7 days, the upper limit of the category "acute low backpain" imposed by the Quebec Study guidelines.[35] We contend that most clinicians would treat this patient as an "acute" patient, a decision we propose is based on the severity of the patient's symptoms and not necessarily on the exact number of days since the onset of symptoms.

Instead of using days since onset, we base acuteness on a criterion related to the severity of disability. We have developed three different stages of LBS, with each stage based on the severity of disability. For each stage, different patient management strategies are used. An overall perspective of the staging criteria is included in Figure 4.

Although we recognize that chronicity measured in some time period in and of itself does not guide treatment management, we do note that prolonged episodes may lead to chronic pain issues, and it is important to point out that chronic pain issues are complex in a sense that they encompass more than just a physical dimension.37 Chronic pain in LBS should suggest to the physical therapist the need for comprehensive (eg, multidisciplinary) management.[37] Therefore, we find it useful to identify patients who have psychological sequelae of chronic pain so that treatment can be directed toward those health care professionals best capable of handling such problems. Rather than identifying patients as chronic by an arbitrary cutoff of symptoms of greater than 6 months' duration as per the Quebec Study guidelines,[35] we instead use accepted tests for "nonmovement" components (eg, magnified illness behavior) of LBS as a screening tool for deciding when other disciplines (eg, behavioral medicine, psychology) should be consulted.

Because this article focuses on managing "acute" injuries, we will first review staging criteria that we believe define acuteness as more related to the severity of symptoms rather than days since injury. In addition, data from other testing previously mentioned (eg, Oswestry questionnaire) will be related to staging criteria.

Stage I

A patient in stage I of the mechanical low back pain syndrome is characterized by an inability to perform the basic mechanical functions of standing, walking, or sitting. We believe that these activities can best be thought of as the foundations for other purposeful activities, and, if individuals are unable to perform these foundational activities, we cannot expect them to perform a more complex and stressful activity (eg, material handling).

We classify an individual as being in stage I if he or she is unable to stand for 15 minutes or more, to sit for 30 minutes or more, or to walk more than 0.4 km (1/4 mile) without worsening of status. This judgment is made using the self-report as a first approximation and subsequently through our interpretation of the patient's ability to sit, stand, and walk during our history taking and physical examination. This individual usually has a modified Oswestry score in excess of 40. In our experience, the majority of patients in stage I fall within the range 40 to 60. We also find that individuals who are seen shortly after an onset of LBS (eg, less than 2 weeks) will often have a modified Oswestry score greater than 60. if an individual, however, rates his or her disability between 60 and 75 after an acute episode has run its course, we suggest that he or she may still fall in stage I. In such patients, careful examination is required to distinguish those individuals who may be exhibiting signs of symptom magnification from those with a severe disability resulting from stage I LBS.

For patients in stage I, we propose that the primary focus of therapeutic intervention by the physical therapist is similar to that of other disciplines, namely, pain modulation. The primary treatment methods we propose in stage I include extension exercises, flexion exercises, lateral-shift regimens, manipulation, traction, and, occasionally, immobilization regimens. Similarly, adjunctive treatment regimens may also be prescribed by other disciplines (eg, pharmocological agents), again with the same goal of pain modulation in mind.

Stage II

An individual in stage II exceeds the requirements (time of sitting and standing, distance walked) for stage I, but he or she is unable to perform basic functional activities of daily living. Thus, the patient can sit, stand, or walk, but pain prevents the patient from performing what have been described as instrumental activities of daily living (eg, vacuuming, lifting, mowing the grass). Most individuals in stage II have modified Oswestry scores in the range of 20 to 40. Management of patient problems in stage II still includes pain modulation but broadens to include elimination of signs of physical impairment (eg, weakness, flexibility that falls outside the ideal range, poor aerobic capacity, faulty body mechanics and posture) that may predispose an individual to recurrence of acute back pain.

Stage III

Stage III is a category for the individual who is returning to an activity that places a high physical demand on the body, especially the lumbar region. Such individuals include workers whose duties include heavy material handling, athletes who participate in sports that place high demand on lumbar structures, and homemakers who must manage household chores and numerous small children simultaneously. An individual in stage III is able to perform instrumental activities of daily living and may even be able to participate in activities involving high physical demand. Sustained activities requiring high physical demands such as those that may be required by occupational duties, however, cannot be carried out. Modified Oswestry rankings are usually 20 or less. These individuals are often relatively asymptomatic, but have become generally deconditioned from a lengthy period of inactivity. The focus of our evaluation and treatment in this stage is on the ability or willingness to labor (simulated work activity) for prolonged periods of time without exacerbation of symptoms.

Each stage has different goals and treatment approaches associated with it (Fig. 4). Examination procedures are different for each phase. In the remaining portion of this article, we will focus on the clinical data and decision rules used to place patients into stage I classifications.

Third-Order Classification: Assigning Patients to State I Syndromes

The major components of the examination include (1) history taking, (2) observation of posture in standing and sitting, (3) assessment of symmetry of pelvic landmarks in standing and sitting, and (4) examination of trunk movements (forward, backward, side bending, and pelvic translocation) in the standing position. We have outlined the examination in Figure 5.

In patients who exhibit symptoms of nerve root involvement (eg, symptoms below the knee), a neurological examination is also performed that includes lower-extremity sensory testing, muscle strength assessment, reflex testing, nerve root tension tests (eg, straight leg raise and femoral nerve stretch), Babinski test, and tests for presence of clonus.[38] From a screening standpoint, the neurological examination is a traditional and important part of most evaluations of patients with LBS. In the treatment scheme that we are proposing, however, the results of such tests may not be useful in directing treatments. For example, a patient with lower-extremity symptoms consistent with an L-4 nerve root irritation (eg, weak ankle dorsiflexors, positive straight leg raise, paresthesia over medial calf and ankle) may fall into any one of four of our different proposed categories (ie, traction, extension, flexion, or lateral shift).

As with most diagnostic processes, a structured protocol for every patient is not followed.[39] Rather, key information is identified early in the examination process, and a working hypothesis (classification) is generated. Further examination proceeds until the hypothesis is confirmed or disconfirmed, the latter leading to a new hypothesis.

In addition to history data related to the first- and second-level classifications, we also ask the patient about the type and location of symptoms, mode of onset, most aggravating and relieving postures, and number and severity of previous episodes. We have outlined the pattern of responses related to each category. In many cases, the history can reveal a very compelling picture of certain categories, in which case an initial hypothesis is formulated and the examination proceeds in a fashion in which the hypothesis is systematically confirmed or disconfirmed. With other patients, however, the history is of little help in achieving an initial hypothesis, and the examination proceeds to the lower-quarter screening without an initial hypothesis in mind.

A good example of this approach includes the scenario depicted in which a patient has low back pain, leg paresthesia, and a frontal- or sagittal-plane deformity (Fig. 6). The key sign in observation of posture in standing is the presence of a movement loss so severe that it prevents the patient from assuming an upright posture. This sign may occur in the frontal plane and has been described in the literature as trunk fist,[40] lateral shift,[14] or sciatic scoliosis.[41] Alternatively, the patient may assume an acute kyphotic position of the lumbar spine from a major loss of lumbar extension.[12] The initial hypothesis is either lateral shift or an extension syndrome, depending on the deformity. The hypothesis is confirmed with movement testing; other tests may be expected to be positive. For example, patients with relevant lateral shifts will usually have asymmetrical excursions in side bending (a positive side-bending test) in the assessment of lateral bending.[42] Further confirmimg information may be gained with repeated movement testing, where either lateral pelvic translocation or extension movements impose a favorable change on the patient's status. The hypothesis, however, can be disconfirmed when a movement does not improve the patient's status. A new hypothesis or treatment category is then formulated (eg, traction).

Without an obvious hypothesis based on the patient's history, our approach is to proceed with the examination by first ruling out lower-extremity influences (eg, pelvic component, leg-length discrepancies) and then focusing on the lumbar spine, specifically movement testing. In most cases, a hypothesis (a tentative classification) should be generated from these two examinations and further examination proceeds, searching out confirmatory and disconfirmatory data. An overall structure is presented in Figure 7, and an algorithm is presented in Figure 8.

Assessment of Bony Landmarks of the Pelvis

An assessment is made of the bony landmarks of the pelvis in the standing and sitting positions. Both static (patient remaining stationary) and dynamic (patient moving) tests are performed. Most of the tests that are performed are purportedly directed toward dysfunction of the sacroiliac joints (we prefer to state that a positive composite is indicative of need for a specific manipulation technique) or a leg-length discrepancy.[43]

The actual tests used and the decision rules are outlined in Figures 7 and 8. We use four tests to make the determination of whether to intervene with manipulation to the pelvis. The four tests are (1) assessment of the symmetry of posterior superior iliac spine (PSIS) heights with the patient in a seated position, (2) the standing flexion test, (3) the prone knee flexion test, and (4) the supine to long-sitting test. A composite (three of four positive) of the four tests is used. Each test and subsequent reliability analyses have been described elsewhere.19

The first test conducted is used to assess heights of the PSISs with the patient in a sitting position. Bilateral comparisons are made, and PSISs of unequal heights constitute a positive finding. A standing flexion test is conducted next. The patient is in the standing position, and the examiner palpates the PSISs bilaterally. The patient then bends forward, with the examiner continuing to palpate the PSISs. A positive finding is present if a change in relationship is detected between the beginning and end of motion. The third test is a comparison of medial malleoli from supine to long-sitting positions. With the patient initially positioned supine, the examiner palpates the inferior aspect of the medial malleoli bilaterally and notes relative lower-extremity length. The patient then sits up, and the lengths are again compared. A change in relative lower-extremity length is a positive finding. Our fourth test is a prone knee flexion test. With the patient initially positioned prone with shoes on, the relative leg lengths are assessed visually. The patient's knees are then flexed passively to approximately 90 degrees, and the lower-extremity lengths are again observed. A change in relative lengths between the two positions is a positive finding. To place a patient in a manipulation category, three or more of these four tests must be positive. Past work has shown excellent reliability ([kappa] = .88) Of such a composite.[9]

If a positive composite is found, a specific manipulative procedure is indicated, provided there are no contraindications.[43] The procedure is illustrated in Figure 9. As an alternative to manipulation, there are also numerous muscle energy procedures that are purportedly indicated in the presence of such a composite; these procedures are described in detail elsewhere.[44]

If the composite is negative, we palpate iliac crest heights with the patient in a standing position and then assess symmetry. Any side-to-side height difference is corrected with a heel lift before movement testing begins. We are aware of the equivocal support for leg-length inequality being a cause of low back problems.[1] In our experience, however, leveling the pelvis commonly leads to a dearer picture for the movement testing that will be performed next.

Single-Movement Tests in Standing

Selected single-movement tests are examined in the standing position. First, side bending to the right and left is assessed. We prefer to have the patient slide his or her hand down the lower extremity and note the position of the fingertips along the lateral thigh and leg. This measure can be quantified by measuring the fingertips in relationship to a bony landmark (eg, fibular head or lateral malleolus) or, alternatively, to use a vertical ruler, a technique that has demonstrated excellent reliability.[19]

In either case, symmetry of side bending is evaluated. If side bending is judged to be symmetrical, then the examination will proceed by determining whether repeated movements (eg, flexion or extension) will improve the patient's status. If asymmetrical side bending is noted, the examination proceeds to determine whether pelvic translocation and extension will improve the patient's status (lateral shift) or whether the pain and restriction of motion represent a particular pattern (opening or closing pattern). The reader will recognize that the symmetrical and asymmetrical restrictions in side bending correspond with what Cyriax[12] described as "capsular" and "noncapsular" patterns, respectively. In some instances, the asymmetry is quite profound and the decision rules listed in Figures 6 and 7 can be followed. in other cases, however, the distinction between symmetrical and asymmetrical is not as clear. Part of our future work will entail further study to describe the degree of difference between left and right side bending that is meaningful and therefore can be labeled "asymmetrical" as well as other findings during the testing procedure (eg, change in pain location or elicited pain location) that may result in certain patterns.

We suggest that the examiner note the range of motion in each direction and any change in symptoms produced by the movement. Compiling information from the patient's history and from evaluation of the results of single-movement tests is instrumental in formulating the initial working hypothesis classification).

With symmetrical side bending, the examiner next assesses status change with single extension followed by flexion. At times, sagittal-plane movement can either worsen or improve a patient's status. If improvement occurs with a particular movement, the initial hypothesis is to include the patient in the category of self-treatment using the particular movement in the prescribed exercise regimen while concomitantly avoiding the opposite movement. It is more common to fall to find either movement actually improving status, in which case the clinician attempts additional movements in different positions (supine, hand-knee), all in an attempt to elicit improvement in status with movement.

Status Change

Each movement that is examined is rated according to the operational terms used to describe change in status (ie, improve, worsen, status quo). After the movement, the patient is asked to compare his or her symptoms with the baseline. Possible patient responses include (1) worsens; in which paresthesia is produced or the patient's pain or paresthesia moves distally from the lumbar spine (peripheralizes); (2) improves, in which paresthesia or pain is abolished or moves from the periphery toward the lumbar spine (centralizes); and (3) status quo, in which the patient's symptoms may increase or decrease in intensity but do not centralize or peripheralize.

Particular attention is given to those single movements (forward, backward, bilateral side bending, and pelvic translation with extension) that produce worsening or improvement in status. When a single movement worsens status, repeated testing and sustained testing using the same movement are generally contraindicated, at least in the standing position. A movement that improves status should be explored further in both the repeated movements and sustained postures to gamer further confirmation that the movement and direction actually improve the patient's status.

The major criterion in determining worsens" and "improves" in patients with back and leg symptoms is the centralization phenomenon described by Cyriax,[45] McKenzie,[14] and Donelson and colleagues,[46,47] the last of whom relate centralization to prognosis. our first interpretation of McKenzie's description of the movement tests as well as our subsequent work[18] resulted in our adopting the following decision rule: For the patient whose status improves with at least one movement in any position (eg, extension in standing), the initial hypothesis is formulated (eg, extension syndrome). Further repeated or sustained testing in the same direction either confirms or disconfirms the hypothesis. There are three potential syndromes that may exhibit improvement with active movement: (1) extension, (2) flexion, and (3) lateral shift (Figs. 7, 8).

The reader should note that these are stage I treatments and that the treatment goal is to modulate symptoms to the extent that the patient can be progressed to stage II. Not all patients experience such a favorable outcome with self-treatment, and some patients change their clinical picture once treatment is commenced. For example, some patients who are initially classified as having extension or lateral-shift syndrome may successfully use self-treatment. At a point in time, however, the status change may plateau or actually reverse direction or worsen. in our experience, it is not uncommon that patients who initially responded favorably to self-treatment (eg, extension movements) may continue to exhibit symptoms, but movements that initially improved status now do not change status or may actually worsen status. Such patients are reassessed as described later.

We acknowledge McKenzie's contribution to the body of knowledge through addition of repeated and sustained testing. We believe that McKenzie's work has refined and expanded on the descriptions provided by Cyriax with regard to movement-related symptom behavior and has improved management of individuals with LBS who demonstrate centralization of symptoms with repeated or sustained movement testing (ie, symptoms move in a proximal direction or toward the midline of the body). Studies have demonstrated that favorable outcomes accompany centralization of symptoms, allowing a clinician a powerful tool for prognosis.[46,47] Mckenzie attributes centralization and peripheralization to a phenomenon related to the disk, and was no doubt influenced by a similar explanation provided by Cyriax in an early version of his text.[45]

For those patients in whom centralization is not complete (pain remains lateral to the midline), it may be reasonable to assume that nondisk structures of the spine may be involved (eg, facet joints, sacroiliac joint). More importantly, however, repeated and sustained exercise may not be the treatment of choice in these individuals, and in some cases we believe this treatment may be contraindicated. we are unable to find a path in McKenzie's assessment algorithm[48] that adequately accounts for all patients whose status worsens or remains the same with testing or subsequent treatment. We propose an alternative treatment classification system to that of McKenzie's postural and dysfunction syndromes. Patients whose status remains the same or worsens with movement testing are candidates for this type of classification.

Status worsens. traction. One shortfall of McKenzie's assessment algorithm[48] is failure to guide the clinician in managing the patient whose symptoms worsen with movement testing (ie, do not improve with any movement and commonly worsen with all tests). These patients may or may not have deformity and may or may not exhibit asymmetrical side bending. A prime example of this type of patient is one with distal symptoms that do not improve (do not centralize) during movement testing. We propose that such patients be grouped into what we have referred to as traction syndrome.

Traction syndrome. There are three subgroups of traction syndrome: (1) capsular pattern of movement restriction that resembles flexion syndrome (ie, spinal stenosis), (2) capsular pattern of movement restriction that resembles extension syndrome (lumbar radiculopathy), and (3) noncapsular pattern of restriction that resembles lateral shift with or without distal symptoms or neurological signs (Figs. 7, 8). The goal of treatment in traction syndrome is to use mechanical (with subgroups 1 and 2) or autotraction[49] types of devices for a short period of time until on subsequent visits the patient's status may either (1) remain status quo with movement, at which time mobilization techniques may be used, or (2) improve with movement, at which time the patient will move into a self-treatment category.

Status quo: mobilization. The classification approach we propose recognizes and takes advantage of the growing body of evidence supporting the effectiveness of early, judicious use of manipulation in managing LBS.[50] We propose that single, repeated movements and sustained postures do not cause a change in status in some patients. Our approach is to classify some of these patients whose status remains unchanged as having lumbar mobilization syndrome. There are two types of patients with mobilization syndrome: those who have capsular patterns and those who have noncapsular patterns.

Patients with capsular patterns with limited extension are best managed with general mobilization techniques designed to improve extension, whereas those with limited flexion receive general mobilization techniques to restore flexion; such procedures have been described elsewhere.[51] Patients may be assigned to this category at the time of initial exanimation, or they may progress into this syndrome when status ceases to improve with self-treatments consisting of active flexion or extension.

Patients with noncapsular patterns are further divided into two subgroups: those requiring regional mobilization and those requiring specific mobilization techniques. Regional techniques are used for those patients who require passive pelvic translocation before they can be successfully managed with active (self-treatment) pelvic translocation (Figs. 7, 8). In some cases, regional techniques are not successful in improving a patient's status. In such patients with a visible list who are unable to move into lateral shift (eg, status worsens with self-correction or manual correction), the patients may require treatment with traction as outlined for traction syndrome (noncapsular pattern). Regional techniques may also be indicated in patients with a compensatory scoliosis resulting from leg-length imbalance or segmental lesions.[52]

Individuals with segmental lesions require specific manipulation or mobilization designed for closing (extending) or opening (flexing) a lumbar segment on one side. Such patients exhibit a characteristic "closing" (painful and restricted extension and ipsdateral side bending) or "opening" (painful and restricted flexion and contralateral side bending) restriction during single-movement testing. Unlike patients in the lateral-shift category, the pain of patients with segmental lesions is not as severe, is usually felt just lateral to the spinous processes in the lumbar region, and does not radiate distally. Most importantly, the pain does not improve (centralize) with repeated movement testing or regional techniques.

Immobilization. A large percentage of patients require some form of motion restriction for excessive motion in a segment. Hypermobility or segmental instability is an entity with a growing body of evidence,[53,54] and we fail to see how any of the treatment procedures described thus far effectively manage this impairment. In our experience, this is an area in which movement testing very often does not provide confirmatory information. Most common symptoms and signs of immobilization syndrome. We believe that the most useful data leading to an initial hypothesis of immobilization syndrome are generated in the patient's history. Confirmatory data include (1) frequent recurrences precipitated with minimal perturbations,[55] (2) a previous history of alternating sides of lateral shift, (3) frequent manipulation (sometimes performed by the patient) with short-lived relief, (4) trauma, (5) pregnancy or use of oral contraceptives, and (6) positive change in status with previous use of a supportive device (eg, corset). The basis for considering these data confirmatory lies in the literature[54,55] (for frequent recurrences precipitated with minimal perturbations and for a previous history of alternating sides of lateral shift) or our own experience. Confirmatory data from the physical examination are usually in the form of key signs, including (1) generalized ligamentous laxity[56]; (2) painful arc of motion or "instability catch" on forward bending of trunk, which is sometimes accompanied by a deviation to the side; (3) painful arc of motion on return from forward bending or side bending; (4) during return from forward bending, a reversal of lumbo-pelvic rhythm (the trunk is first extended and then the hips and pelvis extend to bring the body upright), which is sometimes accompanied with "thigh climbing" (using the hands pushing on the thighs to assist the trunk in attaining an upright posture); and (5) positive posterior shear test. Many of these signs have been related to segmental instability in descriptions elsewhere.[57,58]

Movement testing commonly leads to a worsening of status with sustained end-range position testing and with repeated movement when performed over a prolonged time period. The latter is commonly observed when a patient's status begins to worsen after an initial successful treatment with active flexion, extension, pelvic translocation, manipulation, or traction. Note in particular the patient with a history of recurrent lateral shifts, whose immediate acute incident may be manageable in stage I with pelvic translocation and extension but whose underlying problem may be more related to segmental instability.[55,59]

Treatment of segmental instability.

For purposes of selection of appropriate immobilization treatment strategies, we believe it is useful to grade the degree of disability resulting from the instability (Fig. 10), with the interventions corresponding to the degree of disability. On one end of the spectrum is a person with minimal disability who responds to avoidance of extreme(s) of range of motion. This may be in extension, flexion, or combined movements. At the other end of the spectrum are patients with severe disability who generally require surgical referral. In between these extremes are patients who have moderate disability and may be managed with stabilization exercise regimens or external supports. Management begins with stabilization exercise because it is the least intrusive. Corsets and braces are required when this method fails.

Summary

We propose three orders of classification to be used in the conservative management of patients with LBS (Fig. 11). We propose that the population of patients with mechanical low back pain is not homogenous, and we believe that studies oriented toward treatment regimens (eg, manipulation, stabilization techniques) that do not take this factor into account are inherently flawed. Space constraints prohibit a complete description of the treatments that accompany each syndrome, but many of these treatments are described elsewhere. Our objective in this communication is to propose that in order to improve outcome with mechanical LBS, we are obligated to classify patients into categories whereby matching treatments to classifications will result in faster, more efficient and cost-efective care.

Thus far, we have subjected a portion of this approach to research and peer review, including reliability of testing procedures and decision analyses, correlational research studying the relationship among variables necessary to classify, and predictive validation pilot testing.[8,9,18,19,22,27,28,42] We have also described various activities' processes that have led to the descriptions provided in this article. We have tested only a portion of the components of this approach, and certainly we invite others to do the same. It is not only our hope but our expectation that future peer-reviewed contributions will offer additions, deletions, and modifications to this approach or offer alternative approaches with the goal of treatment efficacy and effectiveness in the management of LBS.

RELATED ARTICLE: Invited Commentaries

Following are two invited commentaries on "A Treatment-based Certification Approach to Low Back Syndrome: Identifying and Staging Patients for Conservative Treatment."

For physical therapists to continue to play a major role in the management of patients with low back pain, cost-effective analysis (reviewing total costs for a given syndrome) and the assessment of patient perception (Did the intervention succeed in improving the quality of the patient's life?) are chemicl realities. In this regard, this article by Delitto and colleagues is timely for the profession, and they are to be congratulated for these initial efforts in suggesting a system to determine a "working classification" for patients with low back pain and an algorithm-based treatment approach to ensure consistent care based on predetermined rules rather than care influenced by examiner bias or emotion.

The authors present working classifications, which can ultimately allow for meaningful cost analysis. Physical therapists are keenly aware of the limitations in examining treatment costs for physical therapy management of low back pain using tools such as the International Classification of Diseases, ninth revision, coding system. Although these diagnoses have medical relevance, they provide little information regarding the movement dysfunction and disabiity associated with low back pain. I agree wholeheartedly with the authors' statement that the population of patients with low back pain is not homogenous and to compare treatment approaches" within a population that is not placed in appropriate diagnostic categories is illogical. Whether one agrees or disagrees with the classification system they present is not as important as first recognizing that it is imperative that physical therapists begin to think in terms of a standardized "working assessmen" or working classification."

The second major contribution is an algorithm-based treatment strategy. Predetemiined rules (signs and symptoms) govern the various steps in treatment. This strategy is different than the typical manner in which treatment is determined. Treatment decisions are often based on such factors as the amount of distress that the patient can convince the clinician that he or she is in and the clinician's own bias toward treating tissues hypothesized to be at fault. Such biases often result in "treatment approaches in search of a syndrome" (to paraphrase the wonderful way the authors referred to thid same dilemma regarding diagnosis). The authors have recognized the need for a more systematic and standardized approach, which might Ultimately ensure consistency in care.

The ongoing nature of this project is duly recognized, and this initial contribution invites some specific comments and also raises several questions for me. I agree with the utility of the first classification tier, as it should always be the first order of decision making. Once patients are identified for physical therapy management, they are further classified into categories indicating the acuteness of the problem. Both of these "classification levels" are in part determined by tools that are clinically sensible and easy to administer--the Oswestry questionnaire, a pain diagram, and a very specific intake questionnaire. These are not only important screening tools, but they also serve as tools that establish a baseline from which the clinician can evaluate progress. We also utilize the Oswestry questionnaire as part of the intake process because it provides us with a better understanding of the patients' view of their problem and how the problem affects their life. Likewise, an "exit" Oswestry questionnaire Provides the Physical therapist with information regarding the patients' perception of improvement, referred to by the authors as "short-term outcome."

By sharing their experience, the authors have contributed to the growing body of evidence, suggesting that it is necessary for clinicians to gather intake data that describe not only symptoms, but also patient perspective and current functional status. In reality, change in patient perception is the key outcome we need to assess. Measures such as strength, range of motion, sitting tolerance, walking distance, and so forth actually have lesser value because it cannot be inferred that such measures alter the patients' perception of their problem and positively influence their quality of life. Delitto and colleagues are to be congratulated not only for demonstrating the usefulness and importance of such tools, but also for providing us with their interpretive comments regarding information (scores) gained via this process and strategies that might be appropriate for given scores.

The remainder of this article focused on one of the dassifications, stage I, which they defined as the inability to perform basic mechanical functions such walking or standing and achieve relatively high Oswestry scores. My interpretation is that these patients have substantial pain, markedly limiting their activities. The authors suggest four examination components for this patient. Although I agree with the value of history taking and posture observation components, I question the clinical utility of assessing symmetry of isolated Pelvic landmarks and Pelvic movements during standing and sitting tests. The authors rightfully Point out the reliability of measurements obtained with such tests, but the correlation of such findings to the painful syndrome or disability is more difficult to determine. I am especially skeptical about the value of information gained from the standing flexion test and the supine to long-sitting test, Personally, I expect the patient with this level of discomfort to move "asymmettically." I would suggest, for example, that the individual with low back pain completes such a sit-up test using any available movement pattern to avoid the painful stimulus and that the final position of the malleoli can be due to numerous postural adjustments. Likewise, the standing flexion test seems to disregard the substantial neuromuscular influence to the movement pattern and the ability of the patient to move the lumbar spine, pelvis, and hips in various combinations the three cardinal planes. I fully realize the historical perspective of these tests as they relate to the biomechanics of iliosacral and sacrofliac motions, but such descriptions appear to fixate the examiner on "bony lesions" and do not account for the Potential of the neuromuscular system to influence a movement pattern simply to minimize the painful stimulus.

Likewise, the inteipretation of symmetry of side-bending range of motion during single movement tests begs further question. Although being a reliable test, the relationship of asymmetrical side bending to the patient's low back syndrome is questionable, unless of course it reproduces familiar pain. My observation of physical therapy students with no history of low back pain leads me to believe that asymmetry in side bending is the rule rather than the exception. Again, the historical perspective of the "capsular pattern" is not lost on me, but I wonder whether the term is as relevant today considering that the numerous anatomical constraints and neurophysiological influences to spinal movement patterns are better appreciated.

I would offer the same question to the authors in regard to their findings of "segmental lesions." Although the authors provide us with several guidelines regarding the pain pattern of such purported lesions, the assessment process to gain such information runs contrary to their original intent of standardizing diagnostic labels. The concept of "segmental lesions" might be instructive in the academic pursuit of analyzing mechanics, but the consistency of such a finding, and perhaps more importantly the correlation of any such finding (unless it specifically reproduces familiar pain) with the low back syndrome, remains questionable.

My final comment is related to the authors' brief discussion of instability. If the spine is similar to other regions of the musculoskeletal system, then age-related degenerative changes or injury affecting the various connective tissues would suggest that instability or aberrant motion might be a logical sequelae. For example, we now better appreciate how glenohumeral instability serves as a precursor to a myriad of other shoulder disorders. I am in complete agreement with the authors' assessment that there is a growing body of evidence to support the concept of instability between spinal segments. I often wonder whether muscle guarding patterns and palpable areas of increased muscle activity in the spine are, in fact, "appropriate" muscle responses occurring as result of a reflex arc whose afferent stimulus is the aberrant segmental motion placing excessive stress on spinal tissues. This area needs much further study, and the authors have appropriately raised the question.

Delitto and colleagues have provided us with a well thought out, comprehensive clinical decision-making process for the evaluation and treatment of a specific category of patients with low back pain, and they have suggested that further treatment-based classifications approaches will be forthcoming. As they note, further scientific and clinical scrutiny will refine the examination and treatment procedures suggested in this initial article, and they are to be congratulated for organizing it in such a systematic fashion to allow for peer review. This type of contribution only comes with extensive trial and error, so their efforts must be duly recognized. Their suggestions of adequately assessing patient perception and role status, standardizing classification of the syndrome into meaningful groupings, and basing treatment on predertemined rules via an algorithm are immediately relevant and absolutely essential for the clinician to heed today. I appreciate of the opportunity to provide a commentary and raise questions on this excellent work.

Carl P Derosa, PhD, PT Professor and Chairman Physical Therapy Program Northern Arizona University Box 15105 Flagstaff, AZ 86011

The opinions or assenions contained herein are the private views of the author and are not to be construed as official or as reflecting the views of the US Department of the Army or the US Department of Defense.

I appreciate the opportunity to comment on Delitto and colleagues' article "A Treatment-based Classification Approach to Low Back Syndrome: Identifying and Staging Patients for Conservative Treatment." I commend the authors for providing an examination approach for patients with acute low back syndrome (IBS) that leads to a classification system that specifically directs conservative management of LBS. The authors' approach has three levels of classification based on historical information, behavior of symptoms, and clinical signs. My comments are primarily directed to their first level of classification.

Delitto et al state the first level of classification determines whether the patient with LBS (1) can be managed independently and primarily by physical therapy, (2) may be managed by a physical therapist with consultation by another health care practitioner, or (3) cannot be managed by a physical therapist and must be referred to another health care practitioner. The authors recognize that in states where direct access to physical therapy services is allowed by law, physical therapists may serve as a "first-contact practitioner" in evaluating patients with neuromusculoskeletal (NMS) complaints. In this case, patients complaining of low back pain but not serious pathology (eg, metastatic cancer) must be identified and referred to an appropriate health care professional.

I support the first-level classification system for patients with LBS as described by Delitto et al. In direct access or prescriptive environments, physical therapists must have the skills, knowledge, and abilities to appropriately identify patients who exhibit low back symptoms but who actually have serious spinal or nonspinal pathology. Once these patients are identified, the physical therapist must refer these patients to the appropriate physician specialist.

Physical therapists in the uniformed services currently serve as nonphysician, primary health care providers in the evaluation and treatment of patients with NMS conditions.[1-3] In this role, military physical therapists use a first-level classfificaiton system, as described by Delitto et al, to evaluate and treat their patients with LBS. After taking a detailed history and performing a thorough evaluation, military physical therapists may render treatment, refer to a physician for a condition determined to be outside the scope of physical therapy practice, or treat while in consultation with another health care practitioner. Expanded privileges including referring patients to radiology for appropriate radiographic evaluation and to all specialty clients, are necessary if physical therapists are to perform NMS evaluations safely and efectively.[1,3]

As mentioned by the authors, the weightiest concern in the direct access and the prescriptive environments is that patients with serious pathology that mimicks NMS complaints might be missed. Clearly, the physical therapist's responsibility to patients is the same in either environment. The strength of the military NMS program is the efficiency with which patients with serious pathology are identified and referred to the appropriate health care practitioner. Serious pathology is identified by obtaining a detailed history and performing a thorough physical examination that clarifies the history. Military physical therapists have identified specific signs and symptoms that are "red flags" and warrant expeditious referral to the appropriate physician specialist. The "red flags" used by military physical therapists for patients with LBS are in accordance with those described by Delitto et al (Fig. 3). Nonspinal pathologies that can mimic low back symptoms (abdominal, thoracic, pelvic) and potentially serious spinal conditions (tumor, infection, spinal fracture, or cauda equina sysdrome) have been identified as "red flags" by the Agency for Health Care Policy and Research in their recent panel on acute low back problems in adults.[4]

The use of physical therapists as nonphysician health care providers in the US Army has been an overwhelming success. There is no record of any legal action being brought against Army physical therapists or the government of the United States as a result of physical therapists serving in the physician-extender role.[1] I believe this military model for evaluating and treating patients with NMS conditions supports the first-level classification system for patients with IBS as described by Delitto et al.

In addition to obtaining a patient history and performing a physical examination, the authors utilize medical, Oswestry, and Waddell questionnaires to assist in determining level 1 and 2 classifications of patients with LBS. The authors, however, remind the reader that much of the decisionmaking rules proposed in this classification system have not been tested through prospective research. Through future prospective research, perhaps the use of the medical, Oswestry, and Wadden questionnaires will be shown to provide additional, beneficial data to the physical therapy practitioner in evaluating and treating patients with LBS.

David G Greathouse, PhD, PT Colonel, US Army Chief, Army Medical Specialist Corps HQDA, DASG-DB 5109 Leesburg Pike Falls Cburch, VA 20041-3258

References

[1] Greathouse DG, Schreck RC, Benson CJ. The United States Army physical therapy experience: evaluation and treatment of patients with neuromusculoskeletal disorders. J Orthop Sports Phys Ther. 1994;19:261-266. [2] Greathouse DG, Sweeney JK, Hartwick AM. Physical therapy in a wartime environment. In: Dillingham TR, Belandres PV, eds. Rehabilitation of the Injured Soldier Textbook of Military Medicine. Washington, DC: Office of The Surgeon General, US Department of the Army. In press. [3] Benson CJ, Schreck RC, Underwood FB, Greathouse DG. The role of Army physical therapists as nonphysician health care providers who prescribe certain medications: observations and experiences. Phys Ther. in press. [4] Acute Low Back Problems in Adults: Assessment and Treatment. Rockville, Md: US Department of Health and Human Services, Agency for Health Care Policy and Research; 1994.

RELATED ARTICLE: Author Response

DeRosa questions the "clinical utility" of assessing symmetry of isolated pelvic landmarks and pelvic movements during standing and sitting tests and further questions the validity of such findings with his statement that "the correlation of such findings to the painful syndrome or disability is more difficult to determine." His "skepticism," which is primarily based on anecdotal information and personal experiences, stands in contrast with the supportive evidence per peer-reviewed publications cited in our article.[1-4] He further questions whether the tests in question are truly representative of bony alignment and iliosacral and sacroilial movement, as they are purported to be. We also question such a relationship, as evidenced by our "black box" discussion and approach to these tests. Again, our approach has been to use the tests for pelvic alignment and symmetry, and to formalize the decision rules that lead the clinician to use manipulating techniques and to evaluate the effectiveness of such an approach.

It is interesting that DeRosa chose to use a cohort of physical therapy students with no history of back symptoms and to point out that "asymmetry in side bending is the rule rather than the exception." Perhaps this discussion could be enhanced if data related to side-bending measures were available in peer-reviewed format. For example, we would be able to discuss on a more rational basis questions such as what magnitude of side-to-side difference represents an "asymmetry," an issue broached within the article. More importantly, we would like to discuss the logic of evaluating individuals without apparent back troubles and using such information in some predictive city to treat patients with low back pain. If indeed asymmetry were the rule, then we would believe from DeRosa that evaluating side bending would be useless in treating patients with low back troubles. There are a few problems with such an approach, not the least of which is finding a sample of subjects who are devoid of back troubles. Even if such individuals are found, how well can it be assured that they will not have future back troubles? For example, consider an individual who participates as a subject without back troubles who in the near future develops back pain (not an unusual occurrence considering the incidence of backache). Should such an individual's data be representative of a cohort who is devoid of back pain?

Most importantly, there are numerous tests used in everyday clinical practice that are positive in the "asymptomatic" population, yet when such tests are positive in symptomatic people, they weigh heavily in treatment decisions. For example, higher-than-normal serum uric acid levels are common in the populations of people without symptoms of gout, yet when such high levels occur with symptoms of gout, the test not only becomes important diagnostically but serves as a marker for the effectiveness of pharmacological agents. We propose that evaluating side-bending motion may, in fact, represent such a test. Although we agree with DeRosa that the validity of such an approach remains to be seen, we are not swayed by his argument nor his examples.

DeRosa uses a similar tact in presenting an argument against the validity of what we referred to as segmental lesions. Just as we avoided an argument about sacroiliac joint dysfunction, our future work would most likely steer away from spinal mechanics. Rather, our major pursuit would be focused on (1) whether such findings can be reproducibly observed in clinical situations and (2) whether the treatment approaches that match the classifications are effective.

Finally, we appreciate the comments of Dr Greathouse and fully agree with his comments with regard to the practice environment in the uniformed services.

Anthony Delitto, PhD, PT Richard E Erhard, DC, PT Richard Whowling, PT

References

[1] 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. [2] Cibulka MT, Delitto A, Koldehof R. Changes in innominate tilt after manipulation of the sacrolliac joint in patients with low back pain: an experimental study. Phys Ther. 1988;68:1359-1363. [3] Erhard RE, Delitto A, Cibulka MT. Extension program versus manipulation with flexion and extension exercises in selected patients with acute low back syndrome. Phys Ther. 1994;74: 1093-1100. [4] Delitto A, Cibulka MT, Erhard RE, et al. Evidence for an extension/mobilization category in acute low back pain: a prescriptive validity pilot study. Phys Ther. 1993;73:216-228.

[Figure 1 to 11 ILLUSTRATION OMITTED]

References

[1] Valkenburg HA, Haanen HCM. The epidemiology of low back pain. In: White AA, Gordon S, eds. American Academy of Orthopedic Surgeons Symposium on Low Back Pain. St Louis, Mo: CV Mostly Co; 1982:9-22. [2] Nachemson AL. The natural course of low back pain. In: White AA, Gordon S, eds. American Academy of Orthopedic Surgeons Symposium on Low Back Pain. St Louis, Mo: CV Mosby Co; 1982:46-51. [3] Horal J. The clinical appearance of low back disorders in the city of Gothenberg, Sweden: comparisons of incapacitated probands with matched controls. Acta Orthop Scand Suppl. 1969; 118:1-109, [4] White AA, Gorden SL. Workshop on idiopathic low-back pain. Spine. 1982;7:141-149. [5] Williams ME, Hadler NM. The illness as the focus of geriatric medicine, N Engl J Med. 1983;308:1357-1360. [6] Haldeman S. Presidential address, North American Spine Society: Failure of the pathology model to predict back pain. Spine. 1990; 15:718-724. [7] Deyo RA. Early diagnostic evaluation of low back pain. J Gen Intern Med. 1986;1:328-338. [8] 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. [9] 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. [10] Walsh K, Coggon D. Reproducability of histories of low-back pain obtained by self-administered questionnaire. Spine. 1991;16: 1075-1077. [11] McCombe PF, Fairbank JCT, Cockersole BC, et al. Reproducability of physical signs in low back pain. Spine. 1989;14:908-918. [12] Cyriax J. Textbook of Orthopaedic Medicine, Vol 1: Diagnasis of Soft Tissue Lesions. 7th ed. London, England: Bailliere Tindall; 1978:1-189, 308-621. [13] Maitland GD. Vertebral Manipulation. 5th ed. Oxford, England: Butterworth & Co (Publishers) Ltd; 1986. [14] McKenzie RA. The Lumbar Spine: Mechanical Diagnosis and Therapy. Waikanae, New Zealand: Spinal Publications Ltd; 1989:85-93. [15] Kendall HO, Kendall FP, Boynton DA. Pasture and Pain. Malabar, Fla: Krieger Publishing Co; 1985. [16] Riddle DL, Rothstein JM. Intertester reliability of McKenzie's classifications of the type of syndrome present in patients with low back pain. Spine. 1993;18:1333-1344. [17] Sikorski JM. A rationalized approach to physiotherapy for low-back pain. Spine. 1985; 6:571-579. [18] 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. [19] 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. [20] Tollison CD, Hinnant DW, Kreigel ML. Psychological concepts of pain. In: Mayer TG, Mooney V, Gatchel Rj, eds. Contemporary Conservative Care for Painful Spinal Disorders. Philadelphia, Pa: Lea & Febiger; 1991: 133-142. [21] Barnes D. Social factors affecting back pain. In: Mayer TG, Mooney V, Gatchel RJ, eds. Contemporary Conservative Care for Painful Spinal Disorders. Philadelphia, Pa: Lea & Febiger; 1991:143-148. [22] Welch WC, Erhard R, Clyde B, et al. Systemic malignancy presenting as neck and shoulder pain. Arch Phys Med Rehabil. 1994; 75:918-920. [23] Greenwood J. Socioeconomic factors in back pain and compensation systems. In: Mayer TG, Mooney V, Gatchel RJ, eds. Contemporarily Conservative Care for Painful Spinal Disorders. Philadelphia, Pa: Lea & Febiger; 1991:155-166. [24] Fairbanks JCT, Couper J, Davies JB, et al. The Oswestry Low Back Pain Disability Questionnaire. Physiotherapy. 1980;66:271-273. [25] Hudson-Cook N, Tomes-Nicholson K, Breen A. A revised Oswestry Disability Questionnaire. In: Roland MO, Jenner JR, eds. Back Pain: New Approaches to Rehabilitation and Education. New York, NY: Manchester University Press; 1989:187-204. [26] Patrick DL, Deyo RA. Generic and disease-specific measures in assessing health status and quality of life. Med Care. 1989;27:S217-S232. [27] Erhard RE, Delitto A, Cibulka MT. Relative effectiveness of an extension program and a combined program of manipulation with flexion and extension exercises in patients with acute low back syndrome. Phys Ther. 1994;74: 1093-1100. [28] Delitto A, Cibulka MT, Erhard RE, et al. Evidence for an extension/mobilization category in acute low back pain: a prescriptive validity pilot study. Phys Ther. 1993;73:216-228. [29] Downie WW, Leatham PA, Rhind VM, et al. Studies in pain rating scales. Ann Rheum Dis. 1978;37:382-384. [30] Chan CW, Goldman S, Ilstrup MS, et al. the pain drawing and Waddell's nonorganic physical signs in chronic low-back pain. Spine. 1993;18:1717-1722. [31] Kostiuk J, et al. Cauda equina syndrome and lumbar disc hemiation. J Bone Joint Surg [Am). 1986;68:386-391. [32] Waddell G. A new clinical model for the treatment of low back pain. Spine. 1987;12: 632-644. [33] Waddell G, McCulloch JA, Kummel EG, et al. Nonorganic physical signs in low back pain. Spine. 1980;5:117-125. [34] Waddell G, Pilowsky I, Bond M. Clinical assessment and interpretation of abnormal illness behavior in low back pain. Pain. 1989; 39:41-53. [35] Spitzer WO. Diagnosis of the Problem (the Problem of Diagnosis), Scientific Approach to the Assessment and Management of Activity-Related Spinal Disorders: A Monograph for Clinicians, Report of the Quebec Task Force on Spinal Disorders. Spine. 1987;12(suppl): s16-s21. [36] Von Korff M, Deyo RA, Cherkin D, Barlow W. Back pain in primary care: outcomes at 1 year. Spine. 1993;18:855-862. [37] Turck DC, Rudy TE. Toward a comprehensive assessment of chronic pain patients. Behav Res Ther. 1987;25:237-249. [38] Magee DJ. Orthopedic Physical Assessment. Philadelphia, Pa: WB Saunders Co; 1992:267-273. [39] Sackett DL, Haynes RB, Guyatt GH, et al. Clinical Epidemiology: A Basic Science for Clinical Medicine. 2nd ed. Boston, Mass: Little, Brown and Company Inc; 1991:3-19. [40] Porter RW, Miner CG. Back pain and trunk list. Spine. 1986;11:596-600. [41] Finneson BE. Low Back Pain. 2nd ed. Philadelphia, Pa: JB Lippincott Co; 1981. [42] Tenhula JA, Rose SJ, Delitto A. Association between direction of lateral lumbar shift, movement tests, and the side of symptoms in patients with low back pain. Phys Ther. 1990; 70:480-486. [43] 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. [44] Mitchell FL, Moran PS, Pruzzo NA. An Evaluation and Treatment Manual of Osteopathic Muscle Energy Procedures. Vaney Park, Mo: Mitchell, Moran and Pruzzo Associates; 1979. [45] Cyriax J. Textbook of Orthopaedic Medicine. 2nd ed. London, England: Bailliere, Tindall & Cassell Ltd; 1969:52. [46] Donelson R, Silva G, Murphy K. Centralization phenomenon: its usefulness in evaluating and treating referred pain. Spine. 1990;15: 211-213. [47] Donelson R, Grant W, Kamps C, et al. Pain response to sagittal end-range spinal motion: a prospective, randomized, multicentered trial. Spine. 1991;16:S206-S212. [48] McKenzie RA. A physical therapy perspective on acute spinal disorders. In: Mayer TG, Mooney V, Gatchel RJ, eds. Contemporary Conservative Care for Painful Spinal Disorders. Philadelphia, Pa: Lea & Febiger; 1991: 216. [49] Tesio L, Merlo A. Autotraction versus passive traction: an open controlled study in lumbar disc hemiation. Arch Phys Med Rehabil. 1993;74:871-876. [50] Shekelle PG, Adams AH, Chassin MR, et al. Spinal manipulation for low back pain. Ann Intern Med. 1992;117:590-598. [51] Mulligen BR. Manual Therapy. 2nd ed. Wellington, New Zealand: Plane View Services Ltd; 1992:40-48. [52] Mitchell FL. Structural pelvic function. Academy of Applied Osteopathy. 1965;2:178-199. [53] Frymoyer JW, Krag MH. Spinal stability and instability: definitions, classification, and general principles of management. In: Dunsker SB, Schniidek HH, Frymoyer JW, et al eds. The Unstable Spine (Thoracic, Lumbar, and Sacral Regions). Orlando, Fla: Grune & Stratton; 1986:1-16. [54] Frymoyer JW, Akeson W, Brandt K, et al. Clinical perspectives. In: Frymoyer JW, Gordon SL. New Perspectives on Low Back Pain. Park Ridge, Ill: American Academy of Orthopedic Surgeons; 1989:222-230. [55] Kirkaldy-Willis VM, Farfan HF. Instability of the lumbar spine. Clin Orthop. 1982;165: 110-123. [56] Klemp P, Stevens J, Isaacs S. A hypermobility study in ballet dancers. J Rheumatol. 1984;11:692-696. [57] Nachemson. Lumbar spine instability: a critical updata and symposium summary. Spine. 1985;10:290-291. [58] Macnab I, McCulloch J. Bacbache. 2nd ed. Baltimore, Md: Williams & Wilkins; 1990:156-159. [59] Frymoyer JW, Krag MH. Segmental instability: rationale for treatment. Spine. 1985; 10:280-286.

A Delitto, Phd, PT, is Chair, Department of Physical Therapy, School of Health and Rehabilitation Sciences, University of Pittsburgh, 101 Pennsylvania Hall, Pittsburgh, PA 15261 (USA). Address all correspondence to Dr Delitto.

RE Erhard, DC, PT, is Assistant Professor, Department of Physical Therapy, School of Health and Rehabilitation Sciences, University of Pittsburgh Medical Center, Pittsburgh, PA 15261.

RW Bowling, PT, is Assistant Professor, Department of Physical Therapy, School of Health and Rehabilitation Sciences, University of Pittsburgh Medical Center, and President and Chief Executive Officer, Centers for Outpatient Rehabilitation and Evaluation (CORE) Network, Pittsburgh, PA.

This article was submitted April 8, 1994, and was accepted January 12, 1995.