Association between direction of lateral lumbar shift, movement tests and side of symptoms in patients with low back pain syndrome.

J Tenhula, MHS/PT, OCS, is Orthopedic Clinical Specialist, Irene Walter Johnson Rehabilitation Institute, 509 S Euclid Ave, St Louis, MO 63110, and Instructor, Program in Physical Therapy, Washington UniversitY School of Medicine, 660 S Euclid Ave, PO Box 8083, St Louis, MO 63110. Address all correspondence to Ms Tenhula at Program in Physical Therapy, Washington University School of Medicine, 660 S Euclid Ave, PO Box 8083, St Louis, MO 63110 (USA).

S Rose, PhD, PT, FAPTA, was Associate Professor and Associate Director for Research, Program in Physical Therapy, Department of Orthopedics and Rehabilitation, University of Miami, Coral Gables, FL 33143, at the time of his death. He was Director and Associate Professor, Program in Physical Therapy, WashingtOn University School of Medicine, and Director, Department of Physical Therapy, Irene Walter Johnson Rehabilitation Institute, when this study was performed.

A Delitto, MHS, PT, is Instructor, Program in Physical Therapy, Washington University School of Medicine.

This study was approved bY an internal review board.

This article was submitted July 13, 1989, and was accepted February 28, 1990.

Patients with low back pain syndrome (LBS) often have a clinical presentation of postural deviation of a lateral lumbar shift (LIS), which is also known as a sciatic scoliosis. The determination of the presence and direction of this lateral postural deviation is necessary to guide treatment in certain physical therapy management approaches (eg, autotraction, McKenzie shift correction). In some cases, the frontal-plane deviations of LLS are very difficult to observe, even by an experienced clinician. The shift may even be confused with a longstanding postural deviation of pelvic obliquity or a pre-existing LLS or scoliosis that is not related to the current symptoms.

Literature Review

Several authors [1-7] have reported finding pain or restricted movement with lateral bending (side bending) in patients with a lateral shift, This finding on movement testing has been used to confirm the presence of a lateral shift or to infer the presence of a lateral shift that is difficult to discern.[7]

McKenzie[8] describes a lateral shift as occurring when a vertebra has rotated and laterally flexed in relationship to the vertebra below. This combined movement carries the upper trunk with it. Thus, the top half of the patient's body has moved to the right or left in relationship to the bottom half He states that it must be determined whether the lateral shift is relevant to the present symptoms or merely a congenital or developmental abnormality. Waddell et al[9] described a sciatic list as a condition in which the shoulders are offset from the pelvis. He distinguished this condition from a true structural scoliosis, which has an element of rotation and compensatory curves above and below. A review of the literature reveals agreement that pain can be provoked if the patient bends the lumbar spine toward the convexity of the sciatic scoliosis (ie, in the direction opposite to the lateral shift).[1,3-5,7] McKenzie[8] combines the two movements of side bending and rotation into one movement of side gliding. He states that, in the presence of a lateral shift, there is always some unilateral loss of side gliding. When this loss occurs, movement of the trunk is restricted or completely blocked in the direction opposite to the lateral shift.[8] This loss of movement correlates with restricted side bending to the opposite (contralateral) side of the direction of the lateral shift.

Jayson[2] states that certain patterns of deformity or loss of movement are diagnostic and that one of the most sensitive signs of acute disk prolapse is sciatic scoliosis or spasm of the sacrospinalis muscles on the side of the prolapse. Lateral flexion is often selectively limited according to whether the disk protrusion is medial or lateral to the nerve root that is being irritated. Jayson[2] and other authors[6,8] attribute these acute deformities or loss of movement to a disk protrusion, and they assert that the position of the nerve root in relationship to the protrusion determines the direction of the lateral shift or lumbar scoliosis. These postural deviations are described as an involuntary protective measure that provides some degree of pain relief by reducing nerve-root pressure related to stretch or tension. [1,3,4,6,7,10-13]

Grieve[1] does not believe that a lateral lumbar shift is due to disk protrusion. He argues that nerve root and disk relationships in many cases may have little to do with inducing these postural deviations. Porter and Miller's study[14] of 100 patients with low back pain and associated "trunk list" provides evidence that the etiology of LLS is not related to the disk. Twenty of the 100 patients studied required surgery for a lumbar disk lesion. Findings at the time of surgical disk excision revealed that the direction of the list was related neither to the side of disk protrusion nor to the topographic position of the disk.

Several reasons other than disk protrusion have also been cited for production of a nonstructural lumbar scoliosis and LLS. In addition to apparent or real lower limb-length discrepancy, other causes described are 1) reflex splinting or reflex protective posture [14,15]; 2) habitual poor posture and muscle imbalance or developmental asymmetry[16]; 3) pain and muscle "spasm" secondary to a painful lesion of a spinal nerve root or to inflammation, a neoplasm, or an abdominal lesion[16]; 4) facet-joint impingement or sprain[17]; 5) sacroiliac joint dysfunction or abnormalities with a rotation of an innominate[1,7,18,19]; 6) encroachment of foraminal territory by bony articular facets, edema, or synovial effusion[1]; and 7) residual lateral deviation after recovery from a previous episode of back or sciatic pain.[1,8]

Three different management approaches for patients with LLS have been described in the literature. One of those treatment approaches is the autotraction system, as developed by Lind and Natchev.[7] When a lumbar scoliosis (lateral shift) is present, a "treatment diagnosis" is established as either a "lateral LBS" or a "medial LBS," according to whether 1) the curve is convex (lateral) or concave (medial) on the pathological side, 2) the pain increases during lateral bending to the same (lateral) or contralateral (medial) side of symptoms according to an expected pattern of behavior, and 3) a Lasegue test's results are positive. The assumption is made that pain will increase with side bending toward the convexity of the curve (ie, in the opposite direction of the LLS). If a lumbar scoliosis is not present, a treatment diagnosis of "sub-LBS" is made, and, according to Natchev,[7] the "pain behavior" is said to exhibit medial or lateral behavior. This treatment diagnosis is based primarily on the pain response during lateral bending, and treatment is for medial or lateral LBS.

Another treatment that is based on a finding of a protective scoliosis is unilateral lumbar traction, as described by Saunders.[20,21] He recommends maintaining the scoliosis by using a unilateral traction technique for those patients who seem to have an increase in pain when conventional bilateral traction is used.

The third treatment approach includes McKenzie's techniques for manual and self-correction of the lateral shift.[8,22] He states that the significance of derangement with deformity of scoliosis is much greater than is generally recognized, and this lateral shift must be corrected first before beginning extension movements. He states that the incidence of a lateral shift in patients with low back pain is as high as 52% and that all patients with low back pain must be tested for evidence of a lateral shift that is relevant to the present symptoms. An undetected minor or barely discernible lateral shift may be worsened by treatment with extension exercises, mobilization, or manipulation. Grieve[1] states that the incidence of slight lateral listing may be higher than generally supposed because minor degrees of deviation may be undetected or ignored. Maitland[23] agrees that lateral displacement of the thorax on the pelvis is a very important test movement (using McKenzie's side-gliding movement examination), and he emphasizes the importance of assessing the degree of pain reproduced by countering the patient's deformity.

Thus, establishing the presence or absence of a clinically meaningful LLS is an important step in the management of patients with LBS. Because an LLS may be readily observable but not related to current symptoms, or not readily observable but related to the current LBS, it is important to establish other confirming or disconfirming clinical signs that would indicate the need for a particular treatment. Natchev[7] has suggested that certain tests can be used to confirm the presence of a clinically meaningful LLS (eg, the production of symptoms during side bending), but his data are anecdotal. The purposes of this study were to determine in patients with LBS 1) whether a contralateral side-bending movement test is always positive in the presence of a lateral lumbar shift and 2) whether a relationship exists between the side of symptoms and the direction of the lateral shift. Our research hypotheses were 1) that there would be a significant association between the presence of a lateral lumbar shift and a positive contralateral side-bending movement test and 2) that there would be a significant association between the side of symptoms and the direction of the shift.

Method

Subjects

The subjects in this study were 24 patients with LBS (17 male, 7 female) who were found on examination to have an observable LLS. The subjects' ages ranged from 19 through 50 years (X = 37.1, SD = 12.0). The onset of LBS occurred less than 3 weeks prior to examination in 20 patients (83%), from 3 to 6 weeks prior to examination in 2 patients (8%), and longer than 6 weeks prior to examination in 2 patients (8%). Seventeen patients (71%) reported a history of one or more episodes of LBS. Excluded from the study were patients with a pre-existing scoliosis, established by patient history or by the presence of a rotational deformity as determined by the presence of a rib hump or prominence of one side in the lumbar area, and patients with an observable lateral pelvic tilt indicating an apparent leg-length difference. Each subject signed an institutionally approved consent form to participate in the study.

Equipment

Equipment used in the patients' routine physical examination consisted of a posture grid of dark vertical and horizontal lines spaced 6 in* apart on a light-colored wall a Plexiglas** footplate. a plumb line, a marking pen, a Nikon (model EM B7593749) 35-mm camera*** with a Nikon (model L2509151) 50/1.8 lens,*** and a slide projector.

* in = 2.54 cm.

** Rohm & Haas Co, Independence Mall W, Philadelphia, PA 19105.

***Nippon Kogaku KK, Fuji Bldg 2-3, Marunouchi 3-chrome, Chiyoda-ku,

Tokyo 100, Japan.

Procedure

The data are collected from each patient's routine physical examination arid medical history, which were administered by the principal investigator (JAT). Information about hand dominance, side and duration Of symptoms, previous episodes of LBS, and the patient's perceived cause was obtained from each patient's history of LBS. Although these four factors were not included in the data analysis, they are presented to better describe the patient sample. During the physical examination, the patient's back was exposed to the level of the S2-S3 spinal segments. Visual appraisal of the patient's posture was performed to determine the presence or absence of an observable LLS. We defined lateral lumbar shift as occuring when the upper trunk and shoulders appeared to be trans-located in the frontal plane in relationship to the pelvic girdle or when there was a lateral deviation or curve of the lumbar spine from the midline. For a patient to be classified as having a lateral shift, the deviation had to have been observable by the examiner.

Side-bending movement tests were performed by asking each patient to stand as erect as possible with feet comfortably apart and arms hanging down at the sides with all fingers kept straight. Right side bending was tested first. The patient was instructed to bend directly to the right side, letting the right hand slide down the lateral thigh toward the knee as far as possible. The patient was instructed to report the effect of this movement on his or her symptoms while the examiner observed the range of motion and estimated the distance of tile fingertips to the lateral joint line of the knee. The patient then performed the side-bending movement to the left side, repeating the same procedure. Patients were instructed to try to avoid trunk flexion or extension during the side-bending movements and to avoid lifting the opposite heel from the floor. After completion of the side-bending tests, the patient was asked to report in which direction the movement produced a greater increase in symptoms. The side-bending test was considered positive if it produced or increased the patient's symptoms greater in one direction than in the other. In the absence of a change in symptoms, the test was considered positive if the side-bending movement was limited more in one direction by a difference greater than 1 in.

Following the physical examination, photographs of both anterior and posterior views of the patient were taken while he or she stood in front of the posture grid. Instructions were given to each patient to stand with his or her feet centered on the footplate and with his or her weight evenly distributed over both feet. The plumb line was aligned with the center vertical line of the grid and footplate through the camera viewfinder. The distance between the camera and the posture grid was kept consistent (15 ft*). The principal investigator marked the following bony landmarks for visualization in the photograph: sternal notch; umbilicus; and spinous processes of S2, L1, and T1 or T2 (depending on visibility as related to head and neck posture). (* ft = 0.3048 m.)

The presence and direction of LLS was determined by the principal investigator during the physical examination of each patient. To confirm the principal investigator's observation, a second investigator (SJR) also determined the presence and direction of LLS by viewing projected slide images of each patient. Interrater agreement between each investigator's determination of the direction of LLS was assessed using Cohen's Kappa statistic (K).[24] Perfect agreement (K = 1.0) was found.

Data Analysis

The degree of relationship between the following pairs of variables was tested using a 2 x 2 chi-square statistic with the Yates correction for continuity[25]: 1) direction of shift and symptoms produced on side bending and 2) side of symptoms and direction of shift. The level of significance was established at .05.

Results

Of the 24 patients studied, 10 (42%) demonstrated a right LLS and 14 (58%) demonstrated a left LLS. Positive side-bending tests were determined using the criteria for change in symptoms. The use of the limited-motion criteria, therefore, was not necessary. The side-bending movement tests were positive to the contralateral side in 17 patients (71%), positive to the ipsilateral side in 5 patients (21%), and negative in both directions in 2 patients (8%). These two patients' data were excluded from the statistical analysis. Chi-square analysis revealed a statistically significant relationship between the direction of the lateral shift and the direction of the positive side-bending test (X[sup.2] = 5.37, P [less than] .05).

Thirteen patients (54%) complained of right-sided symptoms, 9 (38%) of left-sided symptoms, and 2 (8%) of central symptoms. Seven (32%) of those analyzed (n = 22) were shifted toward the side of symptoms, and 15 (68%) were shifted away from the side of symptoms. Chi-square analysis revealed no significant relationship between the side of pain and the direction of shift (X[sup.2] = 1.49, P [greater than] .05)

Discussion

Low back pain syndrome has long been a difficult management problem in the health care delivery system. A major reason for this difficulty has been the inability of practitioners to specifically diagnose this condition, resulting in the majority of patients suffering from backache being placed into the category "unclassified low back pain." This problem has prompted some practitioners to suggest redirecting the emphasis in LBS away from the use of medical diagnoses, where ascribing an etiology is paramount, and toward the use of classifications. These classifications can then be used to establish a prognosis and a rational treatment plan.[26-30]

Some clinicians[8,20] have attempted to subclassify patients previously diagnosed with unclassified low back pain according to the patients' signs and symptoms. One of these subclassifications is sciatic scoliosis, also known as protective scoliosis or lateral listing, or what we refer to as a "relevant LLS." The presence of a relevant LLS, according to some authorities,[7,8,20,21] guides the clinician in treating a patient in a certain way. These practitioners may not agree with what treatment to use, but they all suggest specific treatments for patients with LLS that are different from treatments suggested for other patients with LBS.

Our finding of the association between a positive side-bending movement test and the direction of LLS indicates that patients with LBS who demonstrate a lateral shift are more likely than those who do not demonstrate a lateral shift to have a positive side-bending test to the opposite direction of the lateral shift (contralateral side). This finding partially confirms (71% association) Natchev's[7] use of side-bending tests to establish guidelines for treatment using autotraction. However, the occurrence of negative results in 8% of the patients studied and positive side-bending tests to the ipsilateral side in 21% of the patients studied seems to preclude the use of this test alone to establish a treatment diagnosis of LLS.

The only other movement test described in the literature that establishes that a lateral shift is relevant to present symptoms is the side-gliding test.[8,22] This test confirms the presence of LLS if symptoms improve wit repeated side gliding, and

McKenzie[8,22] suggests performing repeated side-gliding movements in the treatment of LLS. McKenzie uses this test alone to confirm an LLS and reports that this test is useful for treating patients who demonstrate "lateral shift behavior" but who do not have a readily observable LLS. He also uses the side-gliding test to rule out a shift that is not related to current symptoms. If the patient's symptoms do not change with the side-gliding test, then, according to McKenzie, a clinically meaningful LLS is not present, even if there is an observable LLS.

Our data suggest that it is not plausible for a clinician to base an entire treatment on the results of the side-bending test. Although significant, the magnitude of the association between direction of LLS and positive side-bending tests in our study was low. In our study, we chose only patients with an obvious, observable LLS, thus eliminating any patient with a subtle or nonobservable LLS. in patients without an observable LLS, the auto-traction diagnostic system must depend solely on the side-bending test, which is not justified by the low magnitude of the association found in this study. We believe that adding other tests that can establish the presence or absence of a clinically meaningful US will increase the specificity of a classification of LLS.

Practitioners appear to use three different tests to establish the presence of LLS: 1) observation, 2) the positive side-bending test, and 3) the positive side-gliding test. We believe that the diagnosis of LLS should be based on a combination of findings (or clusters of findings) from these three tests and possibly on the use of other tests. Further research is needed to identify the exact make-up of the cluster (eg, weighting of each test, decision rules). Once established, the clusters should be more specific to the diagnosis of LLS than any one individual test. Further research is also needed to determine whether a combination of these three tests can be used to confirm that LLS is a diagnostic category of LBS. The focus of our future research will be directed toward examining the validity of an LLS treatment category.

The lack of correlation between direction of the lateral shift and side of symptoms in this study agrees with the results of Porter and Miller.[14] Our results of 15 of 22 patients (68%) who deviated away from the painful side differ from McKenzie's[22] findings of 479 of 500 patients (96%) with sciatic scoliosis who deviated away from the painful side. Grieve[1] also states that the patient usually deviates away from the painful side and less frequently toward the painful side, although he does not present any data to support his observation. We must consider, however, the sample differences among these studies, including the small number of patients in our study, and possible differences in demographics.

We believe that our literature review of the causes of an LLS indicates that trying to associate this clinical sign with pathology is not useful in directing physical therapy management. We believe that it would be more meaningful to be able to establish and confirm the presence of LLS with simple clinical tests that a physical therapist can perform. If an LLS can be determined with confidence, the clinician can use that information to help make a diagnosis that can be used to direct treatment.

Conclusion

The results of this study show a significant relationship between a positive contralateral side-bending movement test and an LLS, indicating that the positive side-bending test is a useful clinical test to help confirm or to establish the presence of a lateral shift in patients with LBS. The results, however, do not show a significant relationship between the side of symptoms and the direction of the LLS in the patients studied.

Table 1. Association Between Clinical Data and Direction of Lateral Lumbar Shift in Patients with Low Back Pain Syndrome. (LBS) (N = 24)

(* OMITTED)

  Table 2. Association of Direction of
Lateral Shift to Positive Side-Bending Test[sup.a]
                            Positive                   Positive
                            Side                       Side
Direction of                Bending                    Bending
Shift                       to Right                   to Left
Right                        2                         7
Left                        10                         3
[sup.a]x[sup.2] = 5.37, P[less than].05.
  Table 3.  Association of Direction of
Lateral Shift to Side of Pain[sup.a]
Direction of                  Side of Pain
Shift                         Right     Left
Right                         4         6
Left                          9         3
[sup.a]x[sup.2] = 1.49, P[greater than].05

References

1. Grieve GP. Common Vertebral Joint Problems New York, NY: Churchill Livingstone Inc; 1981.

2. Jayson MIV, ed. The Lumbar Spine and Back Pain. 2nd ed. London, England: Pitman Publishing Ltd; 1981.

3. McNab I. Backache. Baltimore, Md: Williams & Wilkins; 1977.

4. Cyriax JA. Textbook of Orthopaedic Medicine: Soft Tissue Lesions. 7th ed. London, England; Bailliere Tindall; 1978:1.

5. Herlin L. Sciatic and Pelvic Pain Due to Lumbosacral Nerve Root Compression. Springfield, Ill: Charles C Thomas, Publisher; 1966.

6. Finneson BE. Low Back Pain. 2nd ed. Philadelphia, Pa: JB Lippincott Co; 1980.

7. Natchev E. A Manual on Autotraction. Stockholm, Sweden: Folksam Scientific Council; 1984.

8. McKenzie RA. The Lumbar Spine: Mechanical Diagnosis and Therapy Waikanae, New Zealand: Spinal Publications; 1981.

9. Waddell G, Main CJ, Morris EJ, et al. Normality and reliability in clinical assessment of backache. Br Med J 1982;284:1519-1523.

10. Kendall HO, Kendall FP, Boynton DA. Posture and Pain. Melbourne, Fla: RE Krieger Publishing Co Inc; 1967.

11. Hardy RW, ed. Lumbar Disc Disease. New York, NY: Raven Press; 1982.

12. White AA, Gordon SL. American Academy of Orthopedic Surgeons Symposium on Idiopathic Low Back Pain. St Louis, Mo: CV Mosby Co; 1980.

13. White AA, Panjabi MM. Clinical Biomechanics of the Spine. Philadelphia, Pa: JB Lippincott Co; 1978.

14. Porter RW, Miller CG. Back pain and trunk list. Spine. 1986;11:596-600.

15. Cailliet R. Low Back Pain Syndrome. 3rd ed. Philadelphia, Pa: FA Davis Co; 1981.

16. Seiman LP. Low Back Pain: Clinical Diagnosis and Management. Norwalk, Conn: Appleton-Century-Crofts; 1983.

17. Salter RB. Textbook on Disorders and Injuries of the Musculoskeletal System. 2nd ed. Baltimore, Md: Williams & Wilkins; 1983.

18. Saunders HD. Classification of musculoskeletal spinal conditions. Journal of Orthopaedic and Sports Physical Therapy. 1979;1:3-15.

19. Erhard R, Bowling R. The recognition and management of the pelvic component of low back and sciatic pain. Bulletin of the Orthopaedic Section, American Physical Therapy Association. 1977;2:4-15.

20. Saunders HD. Use of spinal traction in the treatment of neck and back conditions. Clin Orthop 1983; 179:31-38.

21. Saunders HD. Unilateral lumbar traction: a clinical report. Phys Ther. 1981;61:221-225.

22. McKenzie RA. Manual correction of sciatic scoliosis. N Z Med J 1972;76:194-199.

23. Maitland GD. Vertebral Manipulation. 5th ed. London, England: Butterworth & Co (Publishers) Ltd; 1986.

24. Cohen J: A coefficient of agreement for nominal scales. Educational and Psychological Measurement. 1960;20:27-36.

25. Hays WL. Statistics. 3rd ed. New York, NY: Holt, Rinehart and Winston Inc; 1981:551-552.

26. Spitzer WO. Diagnosis of the problem (the problem of diagnosis): scientific approach to the assessment and management of spinal disorders-a monograph for clinicians. Spine. 1987; 12 (suppl); 16-21.

27. Sahrmann SA. Diagnosis by the physical therapist-a prerequisite for treatment: a special communication. Phys Ther. 1988;68:1703-1706.

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

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

30. Mooney V. The syndromes of low back disease. Orthop Clin North Am. 1983; 14:505-515.