Use of a classification approach to the treatment of 3 patients with low back syndrome.

Key Words: Classification, Low back syndrome, Rehabilitation.

In a pathology-based model, disease can be fully accounted for and defined by deviations from structural or physiological norms.[1] Treatment is directed toward identifying the underlying structural abnormality and applying the appropriate corrective measures. Signs and symptoms are assumed to occur in direct relation to the structural abnormality, and they are expected to disappear once the abnormality is corrected.[2,3]

Despite advances in diagnostic procedures and surgical techniques, low back syndrome (LBS), which includes the pain, disability, and economic costs resulting from injury of the lumbar spine, continues to grow both in prevalence and cost to society.[4,5] Continued reliance on pathology-based models of disease has been credited by some authors[2,6,7] as contributing to the continued epidemic of LBS. With pathology-based models, identification of structural abnormalities guides treatment decisions. Patients with severe symptoms of LBS, however, frequently have no identifiable structural pathology.8 Furthermore, structural abnormalities on imaging studies are known to lack sensitivity in predicting the presence of symptoms. Radiographs, magnetic resonance imaging (MRI), and computed tomography (CT) all have high false-positive rates in individuals without low back symptoms.[9-11] According to pathology-based models, identification of the causative pathology is necessary before treatment decisions can be made. Given the disparity between pathology and symptoms, it is not surprising that the choice of conservative treatment measures for people with LBS was described by Sikorski as "taking on the characteristics of a lottery."[12]

Several authors[4,7,13] have proposed classification systems as an alternative to diagnosis via a pathology-based model. A classification system is a method of categorizing disorders that might otherwise be considered homogenous entities. The process of classification requires an organizational theme about which the larger entity can be organized.[14] Decisions regarding the conservative treatment of patients with LBS can be organized into subgroups, or classifications, on the basis of signs and symptoms identified during the examination and the patient interview.[7,12,15(pp61-98),16] The placement of a patient into a diagnostic classification due to a cluster of signs and symptoms can direct clinicians to specific treatment approaches. Decisions made using a classification model of LBS do not rely on identifying or inferring underlying pathology, but are based on the theory that specific treatments have a higher likelihood of succeeding in patients with specific clusters of signs and symptoms.

Rose[13] described 2 primary purposes of a classification approach: (1) to form a foundation for clinical diagnosis and provide a basis for making treatment decisions with the greatest likelihood of success for a given patient and (2) to improve research by establishing relevant subgroups of patients for treatment outcome studies. Although several classification systems for patients with LBS have been proposed,[4,7,12,16] little research evidence exists regarding the ability of any system to achieve these purposes.

Delitto and colleagues[7,17] proposed a classification model for the treatment of people with LBS. Classification is based on historical information, the behavior of symptoms, and clinical signs. Patients are placed into a treatment-based classification that, in turn, guides conservative treatment (Tab. 1).[7] This article presents 3 patients who had clinical and imaging findings consistent with a pathology-based diagnosis of herniated intervertebral disks and similar anatomical distributions of pain, yet were treated with 3 different approaches based on the specific classification given to each patient using the approach developed by Delitto and colleagues. Case reports cannot provide any evidence for a cause-and-effect relationship between treatment and outcome. The purpose of this report is to illustrate the use of a classification approach in patients who may appear, based on pathology, to belong to a homogenous group, yet responded positively to substantially different treatments.

Table 1. Classification Categories, Key Examination Findings, and Treatment Approaches for Patients With Acute Low Back Symptoms According to the Treatment-Based Classification System Described by Delitto et al(7)

Classification        Key Signs and Symptoms

Extension             Flexion activities worsened symptoms (sitting,
                       bending)
                      Symptoms improved with extension movement
                       testing
                      Symptoms worsened with flexion movement
                       testing
Flexion               Extension activities worsened symptoms
                       (standing, walking)
                      Symptoms improved with flexion movement
                       testing
                      Symptoms worsened with extension movement
                       testing
Mobilization          Sacroiliac: positive sacroiliac tests
(lumbar/sacroiliac)   Lumbar: opening or closing pattern with
                       movement testing
Immobilization        Frequent prior episodes due to minimal spinal
                       perturbations
                      Prolonged static postures worsened symptoms
                      Symptoms worsened with sustained movement
                       testing
                      Symptoms improved with repeated movement
                       testing
Lateral shift         Visible lateral shift (lateral translation of
                       the trunk relative to the pelvis)
                      Asymmetrical side-bending range of motion
                      Symptoms improved with pelvic translocation
                       movement testing, worsened with opposite
                       translocation
Traction              Radicular symptoms
                      Symptoms did not improve with any movement
                       tests
                      Symptoms worsened with most movement tests

Classification        Treatment Approach

Extension             Extension exercises
                      Restriction of flexion (education/bracing)
Flexion               Flexion exercises
                      Restriction of extension (education/bracing)
                      Partially unloaded walking
Mobilization          Mobilization/manipulation
(lumbar/sacroiliac)
Immobilization        Avoidance of end-range or sustained postures
                       (bracing/education)
                      Trunk strengthening and stabilization
                        exercises
Lateral shift         Pelvic translocation exercise in standing or
                       prone position
                      Pelvic translocation exercise combined with
                       extension
                      Autotraction
Traction              Mechanical traction
                      Autotraction

Overview of Examination and Classification Procedures

The classification approach used for the patients in this report has been described by Delitto et al.[7,17] In this approach, the clinician is not searching to identify underlying pathologies, but seeks to make 2 primary determinations: (1) whether the patient is appropriate for primary management by physical therapy and (2) based on signs and symptoms, which treatment approach has the greatest likelihood of success.

Appropriateness for Physical Therapy Management

The first decision was to determine whether the patients in this report could be managed primarily by physical therapy or whether consultation with another medical specialist would be required. The patients were screened for evidence of serious pathology of a nonmusculoskeletal origin through a medical screening questionnaire. Certain questions have been shown to be useful in screening patients for evidence of a serious underlying condition. These questions focus on determining whether there is a history of cancer, unexplained weight loss, immunosuppression, and pain that is worst at rest.[18] In addition, descriptions of recent alterations in bowel or bladder function or of paresthesias in the saddle region would indicate compression of the cauda equina and would have necessitated immediate surgical consultation.19 Psychological consultation would have been considered for any patient demonstrating signs of abnormal illness behavior, defined as inappropriate or mal-adaptive modes of perceiving and responding to one's own state of health.[20] Waddell and colleagues[21, 22] have described both physical examination findings and questions regarding symptoms designed to identify abnormal illness behavior. The 8 questions described by Waddell et al[21] were included on the medical screening questionnaire. If several positive responses were found, examination procedures to detect Waddell's physical signs[22] were included.

Additional screening tools also were used to determine the patients' appropriateness for physical therapy management. A pain drawing can be used to help identify abnormal illness behaviors. Chan et al[23] have shown a correlation between Waddell's signs and pain drawings with a high degree of nonorganic symptoms, including widespread, nondermatomal pain or paresthesia distributions. A pain scale was used to screen for nonmusculoskeletal pathology or abnormal illness perceptions. The patients rated their present pain level on a scale of 0 to 10. They also rated their pain at its lowest and highest levels within the previous 24 hours. Very high scores (8-10) on all 3 ratings would have indicated serious pathology (unremitting pain unaffected by positions) or abnormal illness behavior. Because comparison with findings on other screening tools is essential, a Modified Oswestry Low Back Disability Questionnaire[24] also was administered. The Oswestry questionnaire is a disease-specific, self-report measure of functional disability with documented reliability and construct validity.[25] The modified Oswestry questionnaire asks the patient to rate his or her level of function in 10 areas, with each area scored from 0 to 5. The total score is expressed as a percentage, with 100% representing total disability and 0% representing no disability. The Oswestry score provides information regarding the patient's perception of his or her degree of disability and, for the patients I am describing, was used as a treatment outcome measure. Each patient in this report was deemed appropriate for physical therapy management (Tab. 2).

Table 2. Data Used in Determining Appropriateness for Physical Therapy Management

                       Patient 1             Patient 2

Age (y)                35                    39
Sex                    Male                  Female
Chief complaint        Low back pain, left   Low back pain, right
                       lower-extremity       lower-extremity pain
                       pain/numbness
Diagnostic imaging     MRI:(a) herniated     No imaging studies
  results              L5-S1 disk with       performed
                       compression of the
                       left S1 nerve root
Waddell's nonorganic
  signs score          2/8                   0/8
Pain drawing           No nonorganic         No nonorganic symptoms
                       symptoms
Modified Oswestry
  questionnaire score  60%                   36%
Pain scale score(b)
  At present           6/10                  3/10
  At best              2/10                  1/10
  At worst             8/10                  6/10

                                   Patient 3

Age (y)                            50
Sex                                Female
Chief complaint                    Low back pain, left
                                   lower-extremity pain
Diagnostic imaging results         MRI: herniated L5-S1 disk with
                                   compression of the left S1
                                   nerve root
                                   Radiography: degenerative
                                   disk disease L4-5 and L5-S1
Waddell's nonorganic signs score   1/8
Pain drawing                       No nonorganic symptoms
Modified Oswestry questionnaire    46%
  score
Pain scale score(b)
  At present                       4/10
  At best                          2/10
  At worst                         8/10

(a) MRI=magnetic resonance imaging.

(b) Patients were asked to rate their level of pain, on a scale of 0 to 10, at the present time, at its best (least pain) in past 24 hours, and at its worst (most pain) in the past 24 hours.

Classification of the Patients

For the patients in this report, the second determination was the choice of the treatment method that was believed to have the greatest likelihood of success. The physical therapist made this determination by classifying the patients, which, in turn, guided treatment choices (Tab. 1). Classification was based on signs and symptoms identified during the therapist's physical examination. The major components of the physical examination were history taking, observation of posture, neurological examination, assessment of the pelvis, and assessment of changes in symptoms with movement testing.

Some historical data were gathered prior to the physical examination via the medical history questionnaire and pain diagram. Further information regarding the mode of onset, prior episodes of LBS and treatment received, and aggravating or relieving postures was sought by direct interview of the patient prior to beginning the physical examination. Underlying segmental instability, which has been defined as a loss of spinal stiffness such that normally tolerated loads will result in pain or deformity,[26] may be indicated by numerous prior episodes of LBS precipitated by minimal perturbations, increasing frequency and severity of episodes, frequent manipulations, or a prior positive response to bracing.[7,27] Information regarding postures that relieve or exacerbate symptoms can be obtained by asking the patient to rank order sitting, standing, and walking in terms of symptom provocation. These activities place different demands on the spine. Sitting places the lumbar spine into a more flexed position, whereas standing and walking require more extension. Information about posture is helpful, in my view, for identifying patients whose symptoms are posture-dependent and who may be treated by having them perform specified movements (ie, flexion and extension classifications).[28,29]

The patients were observed to identify deformities that necessitate specific treatment interventions. Based on the system used with patients with LBS, 2 deformities are of primary concern: (1) acute kyphotic postures, which are sagittal-plane deformities,[15(p382)] and (2) lateral shifts, which occur in the frontal plane. A lateral shift is a visible shift of the trunk and shoulders to the left or right in relation to the pelvis.[30] Several authors[7,14,31] have proposed that a lateral shift represents a unique classification of patients with LBS, suggesting the need for a specific treatment.

With the system used for these patients, a neurological examination is not required for every patient with LBS, but is reserved for those patients demonstrating symptoms of potential nerve root compression (eg, symptoms below the knee). The neurological examination includes the measurement of lower-extremity reflexes, sensation and manual muscle testing, and assessment of dural tension signs (straight-leg-raising test, femoral nerve test).[32]

Assessment of the pelvis began by palpating the patients' iliac crests and the posterior and anterior superior iliac spines with the patients in a standing position. The relative heights of the right and left sides were compared. Consistent increases in height on one side were considered evidence of a leg-length discrepancy, whereas an inconsistent pattern with some landmarks high on one side and other landmarks high on the other may suggest a sacroiliac (SI) problem.[33] Confirmation of an SI region problem is sought through additional tests. Four tests were used when needed (a test of symmetry of the posterior superior iliac spines in a sitting position, a standing flexion test, a supine-to-long-sitting test, and a prone knee flexion test). These tests are described in detail in previous publications.[7,34] Three or more positive findings indicated a classification that called for mobilization of the SI joint.[34]

Movement testing was performed initially with the patients in a standing position. The patients were asked to side bend to the left and right and to flex and extend as far as possible. Pelvic translocations, in which the pelvis is shifted relative to the shoulders in the frontal plane, also were assessed on patients with a visible lateral shift. Movement tests may be judged by the examiner to have 1 of 3 effects on a patient's symptoms:

1. Symptoms worsen: paresthesia is produced or the pain moves distally from the spine (peripheralizes).

2. Symptoms improve: paresthesia or pain is abolished or moves toward the spine (centralizes).

3. Status quo: symptoms may increase or decrease in intensity, but do not centralize or peripheralize.

Movement testing may include observation of other postures (sitting, prone, quadruped) or of repeated movements or sustained postures. A judgment of change in symptoms was made for each test performed. Any movement found to worsen symptoms was not tested further, whereas movements that improved symptoms were tested further in other postures or with repeated movements or sustained postures, and were likely to be used in treatment. Patients improving with movement were classified into flexion, extension, or lateral shift categories (Tab. 1).

Throughout the history taking and physical examination, hypotheses are formulated to guide in the subsequent selection of evaluation procedures.[35] This method of clinical diagnostic decision making has been labeled by Sackett et al[36] as the "hypothetico-deductive strategy" and defined as "the formulation, from the earliest clues, of a `short list' of potential diagnoses or actions, followed by the performance of those clinical and paraclinical maneuvers that will best reduce the length of the list."[36(p15)] For example, a patient reporting production of symptoms with standing or walking and relief with sitting may result in an initial hypothesis that flexion-oriented treatment is indicated (ie, flexion syndrome). This initial hypothesis is confirmed or rejected based on further evidence from the physical examination. A worsening of symptoms with extension movements and improvement with flexion movements would be further evidence for a flexion classification. If single flexion movements in a standing position did not cause an improvement in symptoms, repeated or sustained flexion in other postures (sitting, supine, quadruped) would be indicated. Improvement in symptoms on one of the subsequent tests would confirm the initial hypothesis, whereas an inability to improve symptoms with flexion may lead to another classification hypothesis (eg, traction).

Overview of Treatment Strategies

The choice of a treatment approach in these cases was based on the classification assigned to each patient. The classification was arrived at through the hypothetico-deductive process of hypothesis formation and evidence gathering during the evaluation. Even though each patient had low back and unilateral radicular pain and evidence of nerve root compression (based on imaging studies or the physical examination), 3 different classifications were assigned: lateral shift, extension, and flexion. The treatment progression for each patient will be described. In this classification system, the treatment of the patients is divided into stages, based on the severity of symptoms and the degree of disability.[7] Patients with severe disability (Oswestry score generally [is greater than] 40) who have difficulty performing the basic mechanical tasks of sitting, standing, and walking are considered to be in stage I, and the goals of treatment are symptom modulation, decreasing disability, and progression into subsequent stages. Because the patients in this case series had Oswestry scores near or above 40 and reported difficulty with sitting, standing, or walking, they were considered initially to be in stage I.

Patients with less disability who can tolerate the basic tasks of sitting, standing, and walking, but cannot perform more demanding activities such as lifting tasks and more strenuous employment or homemaking duties, are considered to be in stage II. Management focused on reducing functional limitations and disability. Further evaluation was required to identify force, flexibility, and conditioning deficits that need to be addressed. The focus of this report is on stage I management. A complete discussion of stage II treatment is beyond the scope of this report.

Three case reports will now be presented as illustrations of a classification approach to the examination and treatment of 3 patients. Four different therapists, each of whom was familiar with the classification approach just outlined, participated in the care of these patients. The first and third patients presented were evaluated by the same therapist; the second patient was evaluated by a different therapist. Two additional therapists assisted in administering treatments for these patients. The evaluating physical therapist coordinated the care of the patients. No data currently exist regarding the reliability of the classification approach used with these patients. The classification and treatment procedures, therefore, may have differed had another therapist been making the decision for each patient.

Patient 1

The first patient was a 35-year-old man with a complaint of left leg pain and numbness and, to a lesser extent, low back pain. Onset occurred 6 days prior to his being seen for physical therapy after he had been roller-blading for about 10 hours. The patient was employed as an instructor at a roller rink, but he was unable to work after his injury. He had been evaluated by a physiatrist and referred for physical therapy. Magnetic resonance imaging had been performed and showed a left paracentral disk herniation at the L5-S1 level with compression of the left S1 nerve root.

History Taking (Tab. 3)

Table 3.
Patient History

                                  Patient 1
Onset of symptoms                 Developed low back pain
                                  after roller-blading 10
                                  hours, progressed to left leg
                                  pain the next day
Time between onset of symptoms    6 days
 and physical therapy evaluation
Categorization of postures
 (sit/stand/walk)
  Best                            Standing
  Worst                           Sitting/walking
Prior history of back pain        Several prior episodes of back
                                  pain only in past 8 years,
                                  becoming more frequent
                                  and severe; first episode
                                  followed an automobile
                                  accident
Treatment received for prior      Responded favorably to rest,
  episodes                        anti-inflammatory drugs,
                                  and use of an elastic lumbar
                                  support

                            Patient 2            Patient 3
Onset of symptoms           Gradual onset of     Sudden onset of low
                            low back and right   back pain and
                            leg pain over a      shooting pain in
                            couple of days       the left leg while
                                                 walking
Time between onset of       2 weeks              3 weeks
 symptoms and physical
 therapy evaluation
Categorization of
 postures (sit/stand/walk)
  Best                      Standing/walking     Sitting
  Worst                     Sitting              Walking/standing
Prior history of back       Two prior episodes   Numerous prior
 pain                       in the past 10       episodes of back
                            years, both          pain only, no
                            associated with      identifiable
                            lifting injuries     precipitating
                                                 factors
Treatment received for      Responded            Responded favorably
 prior episodes             favorably to rest    to modified
                            and activity         activity and
                            modification         manipulation

The onset of LBS was gradual, with the patient noting pain that increased at night and left leg pain the following morning. Sitting and walking increased symptoms, and standing was his least painful posture. Several previous episodes with back pain only were reported. The initial episode followed a motor vehicle accident 8 years prior to the current episode. He reported that episodes during the past year had been more severe and frequent. Prior episodes had resolved with rest for a few days and occasional use of an elastic lumbar support.

Physical Examination (Tab. 4)

Table 4.
Physical Examination Findings

                                 Patient 1
Observation                      Right lateral shift
                                 Antalgic gait, with decreased
                                  weight bearing on left lower
                                  extremity
Neurological examination         Diminished sensation in left
                                  lateral thigh and calf
                                 Absent left ankle reflex
                                 Diminished force with left great
                                  toe extension and dorsiflexion
                                 Positive straight leg raise on the
                                  left
Palpation of pelvic landmarks    All landmarks elevated on the left
                                  side
Changes in status with single-
 movement testing
  Right side bending             Status quo with full motion
  Left side bending              Symptoms worsened with substantial
                                  motion restriction
Flexion                          Symptoms worsened with minimal
                                  motion restriction
Extension                        Symptoms worsened with substantial
                                  motion restriction
Right pelvic translation         Symptoms improved with minimal
                                  motion restriction
Left pelvic translation          Not assessed
Repeated movement testing        Status quo with repeated right
                                  pelvic translation

                            Patient 2            Patient 3
Observation                 No deformity noted   No Deformity noted
                            Normal gait          Normal gait
Neurological examination    Diminished           No deficits in
                             sensation in right   sensation
                             anterior/lateral
                             calf
                            Diminished right     No deficits in
                             ankle reflex         lower-extremity
                                                  reflexes
                            Diminished force     No deficits in
                             in right plantar     lower-extremity
                             flexion              force
                            Negative straight    Negative straight
                             leg raise            leg raise
Palpation of pelvic         Pelvic landmarks     Pelvic landmarks
 landmarks                   even                 even
Changes in status with
 single-movement testing
Right side bending          Status quo with      Status quo with
                             full motion          full motion
Left side bending           Status quo with      Status quo with
                             full motion          full motion
Flexion                     Symptoms worsened    Status quo with
                             with minimal         minimal motion
                             motion restriction   restriction
Extension                   Status quo with      Symptoms worsened
                             full motion          with moderate
                                                  motion restriction
Right pelvic translation    Not assessed(a)      Not assessed
Left pelvic translation     Not assessed         Not assessed
Repeated movement testing   Symptoms improved    Symptoms improved
                             with repeated        with repeated
                             extension in         flexion in
                             quadruped position   quadruped position

(a) Pelvic translocations are not assessed in patients without a visible lateral shift.

The first physical therapist performed all of the physical examination procedures. Observation of the patient revealed an antalgic gait apparently due to decreased weight bearing on the left lower extremity. He was also noted to be in a right lateral shift, with his shoulders transposed to the right relative to the pelvis in the frontal plane. The neurological examination demonstrated sensory, reflex, and force deficits and a positive straight-leg-raising test on the left. Assessment of the pelvis indicated a consistent increase in the heights of the left landmarks. The validity of this assessment may have been questionable because of the patient's difficulty with full weight bearing on the left lower extremity.

With movement testing, there was a worsening of symptoms with flexion, extension, and left side bending, with substantially restricted range of motion (ie, [is greater than] 50%) with extension and left side bending. Asymmetrical range of motion in side bending (a positive side-bending test) is anticipated in a patient with a lateral shift[7,31] and was found in this patient, with side bending to the left showing much greater restriction than side bending to the right. Because the patient was observed to have a lateral shift and a positive side-bending test, pelvic translocations were performed. The therapist stood at the side to which the patient's shoulders were shifted with his shoulder against the patient's shoulder and his arms around the patient's pelvis. The movement of translocation was produced by pushing the patient's shoulders away while bringing the pelvis toward the midline of the patient's body.[16] This movement produced an improvement in the patient's status of centralization of symptoms. Repeated right pelvic translocations, therefore, were performed. The initial improvement in symptoms observed with single-movement testing was not found during repeated testing, and the patient's symptoms were judged to be status quo.

Hypothesis Formation and Testing

The physical therapist performing the physical examination formed the initial hypothesis from the observation of a lateral shift. The reliability of observational judgments of the presence of a lateral shift has been questioned,[37] but other authors[30] have suggested that a readily observable shift can be reliably determined. Further evidence for the presence of a lateral shift came from the judgment of asymmetrical side-bending motions (positive side-bending test)[31] and from the effect of correction of the deformity (pelvic translocation) on the patient's symptoms.[16] Judgments of change in symptoms during pelvic translocations have been shown to be reliable between therapists.[37] In this case, a positive side-bending test and an initial improvement in symptoms with single-movement pelvic translocation testing confirmed the physical therapist's initial hypothesis that the patient belonged to a lateral shift classification group.

Stage I Treatment (Tab. 5)

Table 5.
Stage I Treatment Procedures and Outcomes

                                Patient 1
Classification                  Lateral shift
Stage I treatment duration      7 weeks
No. of sessions of              12
 physical therapy
Treatment progression           Autotraction to correct deformity
                                Use of flexion-limiting orthosis
                                Repeated right pelvic translocation
                                 with extension in standing
                                Autotraction for extension
                                Repeated extension in prone
                                 exercises

Stage I treatment outcome(a)
Modified Oswestry               22%
 questionnaire score
Pain scale score (present)      2/10

Stage II considerations         Stabilization, general strengthening
                                 and conditioning exercises
                                Address residual deficits of left
                                 dorsiflexor strength and
                                 hamstring muscle flexibility

                                Patient 2
Classification                  Extension Category
Stage I treatment duration      4 weeks
No. of sessions of              4
 physical therapy
Treatment progression           Repeated extension in quadruped
                                 exercises
                                Education to avoid excessive
                                 flexion in sitting or lifting
                                Repeated extension in standing
                                 exercises
                                Progressive walking program

Stage I treatment outcome(a)
Modified Oswestry               0%
 questionnaire score
Pain scale score (present)      1/10

Stage II considerations         Continue walking program
                                Address residual deficit of
                                 right plantar-flexor force
                                Continued monitoring of sitting
                                 posture

                                Patient 3
Classification                  Flexion category
Stage I treatment duration      5 weeks
No. of sessions of              9
 physical therapy
Treatment progression           Repeated flexion in quadruped
                                 exercises
                                Partially unloaded treadmill
                                 ambulation
                                Repeated flexion in supine
                                 exercises
                                Progressive walking program

Stage I treatment outcome(a)
Modified Oswestry               8%
 questionnaire score
Pain scale score (present)      0/10

Stage II considerations         Continue walking program
                                Stabilization exercises

(a) Outcome measures assessed at completion of stage I.

The patient was placed in a lateral shift classification. His symptoms did not improve with repeated translocation of the pelvis in a standing position. When movements are unable to improve symptoms, traction becomes the treatment of choice, according to Delitto et al.[7] In the case of a lateral shift, the autotraction protocol described by Natchev[31] may be useful. An autotraction table permits adjustments to patient positioning in all 3 planes of movement, allowing the deformity of the patient to be accommodated. The patient then provides a traction force through the use of his or her arms and legs, while the therapist corrects the deformity. For this patient, his right lateral shift was initially accommodated, and his pelvis was gradually moved toward a neutral position. After each autotraction treatment, a flexion-limiting orthosis(*) was applied to prevent the patient from flexing his spine. The patient wore the orthosis during all waking hours during the initial phase of treatment. Four autotraction treatments, administered over a 10-day period by the physical therapist who performed the initial evaluation and by another physical therapist, were required to correct his deformity.

The patient was then able to improve his status with repeated pelvic translocations. McKenzie[16] has recommended the use of extension exercises following correction of a lateral shift deformity.[16] Extension was first performed in a standing position. The patient stood in a doorframe, using his arms to brace his shoulders. The patient actively performed a right pelvic translocation, followed by active lumbar extension. He was progressed to performing extension exercises in the prone position by placing his hands on the support surface and then actively extending his elbows while keeping the pelvis in contact with the surface, creating extension of the lumbar spine. The autotraction treatment was continued, with the emphasis switched from correction of the lateral shift deformity to improving extension.

After 7 weeks of stage I treatment administered by the 2 different physical therapists, including autotraction, pelvic translocation and extension exercises, and the use of a flexion-limiting orthosis, the patient's modified Oswestry questionnaire score was reduced from 60% to 22%, his pain scale score was reduced from 6/10 to 2/10, and he reported improvement in functional status, including improved tolerance to sitting and walking. At this time, the physical therapist who had initially evaluated the patient judged his symptoms to be status quo on all movement tests, and he had only intermittent radicular symptoms during the day. The flexion-limiting orthosis was discontinued because flexion no longer worsened the patient's symptoms. Stage II treatment was directed at stabilization exercises[38] and residual deficits in force production. Stage II treatment was deemed appropriate due to the patient's residual disability (ie, 22% Oswestry score). Stabilization exercises, including lumbar extensor and oblique abdominal muscle strengthening exercises,[39,40] were considered appropriate due to his history of frequent episodes of LBS, with increasing frequency and severity. This history suggested the possibility, of an underlying segmental instability.[27] He continued in stage II treatment for 4 more weeks, after which he returned to full-time employment. He resumed use of his elastic lumbar support for work activities.

Patient 2

Patient 2 was a 39-year-old woman employed as a research assistant. She reported a gradual onset of low back pain without a known precipitating event, which progressed to the right lower extremity 2 weeks prior to being seen for physical therapy. Her complaint was right-sided low back pain of greater intensity than right leg pain. She had been off work for 2 days, but had subsequently been able to return to work. The patient had been examined by a physiatrist and referred for physical therapy. No diagnostic imaging studies had been performed.

History Taking (Tab. 3)

The onset of symptoms had been gradual, occurring over a couple of days, beginning with back pain and progressing to right leg pain. Her most aggravating posture was sitting. Standing and walking helped reduce the symptoms. Forward bending and lifting activities were reported to be difficult. She noted 2 prior episodes of back pain in the past 10 years, both related to lifting incidents. Both episodes had responded to self-imposed activity modification, and she had not previously sought medical attention.

Physical Examination (Tab. 4)

The physical examination procedures were performed by a physical therapist who did not participate in the care of the first patient. Postural observation did not reveal any deformity. The neurological examination showed diminished sensation to light touch of the anterior and lateral calf on the right and diminished right plantar-flexor strength and Achilles tendon reflex. Palpation of the pelvic landmarks revealed no asymmetry. Single-movement testing in a standing position revealed full lumbar range of motion, with no change in symptoms during side bending to the left and right or during extension. Flexion in a standing position was found to be minimally restricted and produced a worsening of symptoms (peripheralization). Because flexion was found to worsen the patient's symptoms, repeated extension was performed first with the patient in a standing position. No change in symptoms was found. Repeated extension was further assessed with the patient in the quadruped position. The quadruped position was chosen because it reduces the weight-bearing load on the spine and may permit the patient to move the spine more comfortably. This movement test produced an improvement in symptoms.

Hypothesis Formation and Testing

Information from the patient's history, taking (increased symptoms in sitting, improvement with standing and walking) led the therapist who performed the initial evaluation to the hypothesis that the patient's symptoms would worsen with flexion movements and improve with extension movements (extension classification). Worsening of symptoms with the single-motion testing of flexion helped to confirm this hypothesis. Single-motion testing of extension did not improve her symptoms. Because the hypothesis remained extension classification, repeated extension was performed, first in a standing position and then in a quadruped position. Repeated extension in a quadruped position improved the patient's symptoms and further confirmed the extension classification.

Stage I Treatment (Tab. 5)

Patient 2 was classified into the extension category. Stage I treatment consisted of repeated extension exercises, initially in the quadruped position and later in the standing position. Extension in the quadruped position began with the patient in the hands-and-knees position. She was then instructed to rock forward, bringing her pelvis toward the support surface and producing extension of the spine. Extension in a standing position was performed by having the patient place her hands on her hips and extend her spine without allowing the knees to bend. Education on how to avoid prolonged flexion with sitting was important due to the patient's occupation, which required her to sit for extended periods of time. She was encouraged to maintain a lordotic posture while sitting and to take frequent standing breaks. A progressive walking program was initiated for general conditioning and because the patient benefited from the repetitive extension created by walking.

The patient was seen for physical therapy once a week for 4 weeks. The same physical therapist who performed the initial evaluation administered each of the treatment sessions. After the 4 weeks of physical therapy, her modified Oswestry questionnaire score was reduced from 36% to 0%, her pain scale score was reduced from 3/10 to 1/10, and she had no radicular symptoms. She was discharged from physical therapy and encouraged to continue her walking program and to maintain a lordotic posture during sitting at work. She also was instructed in how to perform strengthening exercises in order to correct a slight residual deficit in plantarflexion force on the right.

Patient 3

Patient 3 was a 50-year-old woman employed as a medical technician. She reported a sudden onset of shooting pain in the left lower extremity approximately 3 weeks prior to being seen for physical therapy. She had subsequently developed left lower-extremity pain and, to a lesser extent, low back pain. She continued to work. The patient was evaluated and referred by her family physician for physical therapy. Magnetic resonance imaging had been performed and showed a disk herniation at the L5-S1 level with compression of the left S1 nerve root. Radiographs had revealed degenerative disk disease at both L4-5 and L5-S1.

History Taking (Tab. 3)

The onset of symptoms was sudden and occurred while walking. The patient ranked sitting as the posture that least aggravated her symptoms, with standing and walking causing an increase in symptoms. She reported numerous prior episodes of back pain without any leg pain or difficulty walking. She noted no pattern regarding frequency and intensity of episodes. Prior episodes responded favorably to reduced activity or chiropractic manipulation.

Physical Examination (Tab. 4)

The same physical therapist who performed the initial examination of the first patient also performed the physical examination of the third patient. The pastural observation and neurological examination were normal for this patient. Assessment of the pelvic landmarks did not reveal any asymmetry. Single-movement testing was status qua, with full range of motion in right and left side bending. Flexion also was status qua with a minimal restriction of motion. Extension caused a worsening of symptoms, with some restriction of motion. Repeated flexion was tested with the patient in a quadruped position to reduce weight-bearing stress on the lumbar spine. This test produced an improvement of symptoms.

Hypothesis Formulation and Testing

The patient's sudden onset of symptoms during walking and her reports of worsening symptoms with standing or walking and of improvement in a sitting position led the therapist who performed the initial evaluation to hypothesize initially that her symptoms would worsen with extension movements and improve with flexion movements (flexion classification). Worsening of symptoms with single-movement testing of extension provided further evidence for flexion classification. Her symptoms were judged to be status qua during single flexion movements. Repeated flexion in the quadruped position, however, produced an improvement of symptoms, further confirming the flexion classification.

Stage I Treatment (Tab. 5)

The third patient was placed into the flexion classification. The treatment program was overseen by the physical therapist who performed the initial evaluation and by another therapist who had no involvement with the previous 2 patients. Treatment consisted of repeated flexion exercises in the quadruped and supine positions and a progression to harness-supported treadmill ambulation, in which a traction force is used in an effort to reduce compressive loading on the spine during ambulation. This treatment combination has been reported to be useful for patients in the flexion classification.[41] Patients in the flexion classification typically have limited walking tolerance. They also frequently report claudication-like pain brought on by walking and relieved by sitting. These symptoms have been related to the compressive loading and spinal extension that occur with ambulation.[42,43] The reduced compressive loading afforded by harness-supported treadmill ambulation may allow the patient to ambulate with reduced or alleviated symptoms. The traction force is gradually reduced until the patient is able to ambulate independently without symptoms.[41]

Flexion exercises also were performed in both the quadruped and supine positions. From the quadruped position, the patient was instructed to arch her spine to produce flexion and then to rock backward, approximating her buttocks to her heels to produce further flexion of the hips and lumbar spine. In the supine position with the knees flexed, the patient was instructed to bring one knee to her chest and hold the position for 10 to 20 seconds, and then repeat the exercise with the opposite leg.

The patient participated in 9 physical therapy sessions over a 5-week period (ie, about 2 times a week), after which her modified Oswestry questionnaire score was reduced from 46% to 8%, her pain scale score was reduced from 4/10 to 0/10, and she reported being able to tolerate walking for 30 minutes during daily activities. Two additional physical therapy sessions were conducted with an emphasis on stage II treatments, which consisted of lumbar stabilization and continuation of walking progression to increase duration of walking. Stabilization exercises were appropriate, in my view, due to the patient's history of frequent prior episodes and multiple manipulation treatments.

Discussion

The 3 patients described in this report had similar complaints of low back and unilateral radicular pain. Each patient demonstrated evidence of compressive nerve root pathology obtained from imaging tests (patients 1 and 3) and physical examination findings (patients 1 and 2). These findings could suggest similar conservative treatment approaches, particularly for clinicians using pathology-based treatments. Each patient, however, was treated with a different approach based on a cluster of signs and symptoms identified during evaluation. Had a pathology-based classification approach been used, the patients would have received similar treatments based on the apparent similarities among them in these areas. The signs and symptoms identified during the examination, however, revealed differences among the 3 patients. One patient had a postural deformity, one patient's symptoms worsened with flexion activities, and one patient's symptoms worsened with extension activities. Utilizing a classification approach based on signs and symptoms led to 3 different treatment approaches, each of which appeared to successfully accomplish the goal of symptom modification and improved disability (ie, as demonstrated by modified Oswestry questionnaire scores). In the absence of control, I cannot state with certainty that the approach led to the outcome or that a pathology-based approach would not have worked.

Research on treatment of individuals with LBS with interventions such as physical agents, shoe lifts, lumbar corsets and braces, traction, and exercise routines (eg, flexion or extension exercises) has not been able to provide evidence in randomized clinical trials for the effectiveness of these commonly used interventions.[18] One reason for the inability to demonstrate the therapeutic effect of these and other interventions may be that researchers have not accounted for subgroup differences within the larger entity of low back pain.[44] Rose[13] termed this the "washout effect," in which the positive effect of a therapeutic intervention in a smaller subgroup of patients is washed out by the large number of patients belonging to other subgroups for whom the intervention is not effective. The 3 patients described in this case series provide an example of how a washout effect may occur. Had they been considered a homogenous group of patients with acute low back pain and had all 3 patients been treated with the same approach (eg, extension exercises), it could be argued that perhaps only one of the patients (patient 2) would have had as rapid a recovery as was achieved using the classification approach. A classification system that identifies clinically relevant subgroups of patients with LBS may be a prerequisite to conducting randomized clinical trials that can determine the efficacy of many of the treatment approaches used by physical therapists, including the categorization scheme used for these 3 patients.

Further research is needed to identify the most reliable and valid classification scheme for patients with LBS. Evidence of improved treatment outcomes through the use of a classification approach is ultimately needed before any system can be advocated for widespread use. Case reports cannot provide evidence of improvement in treatment outcomes. Obtaining such evidence can only be accomplished through clinical trials, preferably randomized trials, comparing treatments matched to a patient's classification with unmatched treatments. The inherent shortcomings of case studies must be considered, including the tendency for symptoms of LBS to improve with time regardless of the treatment administered.

Conclusion

The identification of effective conservative treatment approaches for patients with LBS has been elusive. Yet, as the costs to society due to LBS continue to escalate, the importance of the undertaking becomes more apparent. Reliance on the nature of the underlying pathology or distribution of symptoms to guide treatment decisions does not appear to be adequate. Signs and symptoms identified during the evaluation were used to classify 3 patients and to guide their treatments. This report illustrates the use of a classification scheme in practice, and a case was made as to why this classification scheme may provide a more effective approach than pathology-based classification approaches and treatments. Further research is needed to modify, refine, and provide evidence for the utility of existing classification schemes.

(*) CASH Anterior Hyperextension Orthosis, Ralph Storrs Inc, 197 South West Ave, Kankakee, IL 60901.

References

[1] Engel GL. The need for a new medical model: a challenge for biomedicine. Science. 1977;196:129-136.

[2] Waddell G. A new clinical model for the treatment of low-back pain. Spine. 1987;12:632-644.

[3] Taylor FK. The medical model of the disease concept. Br J Psychiatry. 1976; 128:588 -594.

[4] Spitzer WO. 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):S1-S59.

[5] Andersson GBJ. The epidemiology of spinal disorders. In: Frymoyer JW, ed. The Adult Spine: Principles and Practices. New York, NY: Raven Press: 1991:107-146.

[6] Haldeman S. Presidential address, North American Spine Society: Failure of the pathology model to predict back pain. Spine. 1900;15: 718-724.

[7] 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;75:470-485.

[8] Nachemson AL. Advances in low back pain. Clin Orthop. 1985;200: 266-278.

[9] Frymoyer JW, Newberg A, Pope MH, et al. Spine radiographs in patients with low back pain: an epidemiological study in men. J Bone Joint Surg Am. 1984;66:1048-1055.

[10] Wiesel S, Tsourmas N, Feffer HL, et al. A study of computer-assisted tomography, I: the incidence of positive GAT scans in an asymptomatic group of patients. Spree. 1984;9:549-551.

[11] Boden SD, Davis DO, Dina TS, et al. Abnormal magnetic-resonance scans of the lumbar spine in asymptomatic individuals: a prospective investigation. J Bone Joint Surg Am. 1990;72:403-408.

[12] Sikorski JM. A rationalized approach to physiotherapy for low-back pain. Spine. 1985;10:571-579.

[13] Rose SJ. Physical therapy diagnosis: role and function. Phys Ther. 1989;69:535-537.

[14] Katz JN, Buchbinder R. Soft tissue syndromes. Baillieres Clin Rheumatol. 1995;9:585-598.

[15] Cyriax J. Textbook of Orthopaedic Medicine, Volume I: Diagnosis and Management of Soft Tissue Lesions. 6th ed. London, England: Bailliere Tindall; 1976:61-98, 382.

[16] McKenzie RA. The Lumbar Spine: Mechanical Diagnosis and Therapy. Waikanae, New Zealand: Spinal Publications Ltd; 1989.

[17] Bowling RW, Truschel DW, Delitto A, Erhard RE. Conservative management of low back pain with physical therapy. In: Erdil M, Dickerson OB, eds. Cumulative Trauma Disorders: Prevention, Evaluation, and Treatment. New York, NY: Van Norstrand Reinhold; 1997:499-593.

[18] Bigos S, Bowyer O, Braen G, et al. Acute Low Back Problems in Adults. Rockville, Md: Agency for Health Care Policy and Research, Public Health Service, US Department of Health and Human Services; December 1994. Clinical Practice Guideline No. 14, AHCPR Publication No. 95-0642.

[19] Kostvik JP, Harrington I, Alexander D, et al. Gauda equina syndrome and lumbar disc herniation. J Bone Joint Surg Am. 1986;68:386-391.

[20] Pilowsky I. Low back pain and illness behavior (inappropriate, maladaptive, or abnormal). Spine. 1995;20:1522-1524.

[21] Waddell G, McCulloch JA, Kummel EG, Venner RM. Nonorganic physical signs in low-back pain. Spine. 1980;5:117-125.

[22] Waddell G, Pilowsky I, Bond M. Clinical assessment and interpretation of abnormal illness behaviour in low back pain. Pain. 1989;39:41-53.

[23] Chan CW, Goldman S, Ilstrup DM, et al. The pain drawing and Waddell's nonorganic physical signs in chronic low-back pain. Spine. 1993;18:1717-1722.

[24] Hudson-Cook N, Tomes-Nicholson K, Breen A. A revised Oswestry Disability Questionnaire. In: Roland RO, Jenner JR, eds. Back Pain: New Approaches to Rehabilitation and Education. New York, NY: Manchester University Press; 1989:187-204.

[25] Kopec JA, Esdaile JM. Functional disability scales for back pain. Spine. 1995;20:1943-1949.

[26] Frymoyer JW, Akeson W, Brandt IL et al. Clinical perspectives. In: Frymoyer JW, Gordon SL, eds. New Perspectives on Low Back Pain. Park Ridge, Ill: American Academy of Orthopaedic Surgeons; 1989:225.

[27] Kirkaldy-Willis WH, Farfan HF. Instability of the lumbar spine. Clin Orthop. 1992;165:110-123.

[28] Katz JN, Dalgas M, Stucki G, et al. Degenerative lumbar spinal stenosis: diagnostic value of history and physical examination. Arthritis Rheum. 1995;98:1236-1241.

[29] Fritz JM, Erhard RE, Delitto A, et al. Preliminary results of the use of a two-stage treadmill test as a clinical diagnostic tool in the differential diagnosis of lumbar spinal stenosis. J Spinal Disord. 1997;10:.410-416.

[30] Tenhula JA, Rose SJ, Delitto A. Association between direction of lateral lumbar shift; movement tests, and side of symptoms in patients with low back pain syndrome. Phys Ther. 1990;70:480-486.

[31] Natchev E. A Manual on Autotraction. Stockholm, Sweden: Folsom Scientific Counsel; 1984.

[32] Hopperfield S. Physical Examination of the Spine and Extremities. East Norwalk, Conn: Appleton-Century-Crofts; 1976:237-263.

[33] Greenman PE. Principles of Manual Medicine. 2nd ed. Baltimore, Md: Williams & Wilkins; 1996:310-311.

[34] Cibulka MT, Delitto A, Koldehoff RM. 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.

[35] Delitto A, Snyder-Madder L. The diagnostic process: examples in orthopedic physical therapy. Phys Ther. 1995;75:203-211.

[36] Sackett DL, Haynes RB, Guyatt GH, et al. Clinical Epidemiology: A Bas Science for Clinical Medicine. 2nd ed. Boston, Mass: Little, Brown and Company Inc; 1991:3-19.

[37] Donahue MS, Riddle DL, Sullivan MS. Intertester reliability of a modified version of McKenzie's lateral shift assessments obtained on patients with low back pain. Phys Ther. 1996;76:706-716.

[38] Norris CM. Spinal stabilisation, 5: an exercise programme to enhance lumbar stabilisation. Physiotherapy. 1995;81:139-146.

[39] McGill SM. The mechanics of torso flexion: sit-ups and standing dynamic flexion manoeuvres. Clinical Biomechanics. 1995;10:184-192.

[40] Callaghan JP, Gunning JL, McGill SM. The relationship between lumbar spine load and muscle activity during extensor exercises. Phys Ther. 1998;78:8-18.

[41] Fritz JM, Erhard RE, Vignovic M. Nonsurgical treatment approach for patients with lumbar spinal stenosis. Phys Ther. 1997;77:962-973.

[42] Penning L. Functional pathology of lumbar spinal stenosis. Clinical Biomechanics. 1992;7:3-17.

[43] Schonstrom N, Lindahl S, Willen J, et al. Dynamic changes in the dimensions of the lumbar spinal canal: an experimental study in vitro. J Orthop Res. 1989;7:115-121.

[44] Leboeuf-Yde C, Lauristen JM, Lauristen T. Why has the search for causes of low back pain largely been nonconclusive? Spine. 1997;22:877-881.

JM Fritz, PT, ATC, is Assistant Professor, Department of Physical Therapy, School of Health and Rehabilitation Sciences, University of Pittsburgh, 6035 Forbes Tower, Pittsburgh, PA 15260 (USA) ([e-mail]).