When to use the manipulation? (Letters to the Editor).

To the Editor:

In "Evaluation of the Presence of Sacroiliac Joint Region Dysfunction Using A Combination of Tests: A Multicenter Intertester Reliability Study" (August 2002), Riddle et al seem to understand the original purpose of the 4 tests they examined, yet they go on to misinterpret and mischaracterize the utilization of these tests.

In their abstract, the authors state that they have examined intertester reliability of measurements, but in the conclusion of the abstract they state that they "suspect it is likely that either the proper treatment technique will not be chosen based on the test results or the intervention will be applied to the wrong side." The standing flexion, asymmetry of posterior superior iliac spines in sitting, long sitting, and prone knee bend tests were selected from a myriad of "sacroiliac joint [SIJ] signs" by Cibulka et al (1) to operationally define SIJ dysfunction in order to conduct clinical trials. Indeed, the study by Cibulka et al, which Riddle et al state a purpose of their study was to replicate, contained 2 steps: the first to establish the reliability for the tests in question, and the second to obtain pretest-posttest measurements following manipulation or sham manipulation. A device combining a caliper, a protractor, and a plumb line was used to determine changes in sagittal-plane angles. At the University of Pittsburgh, we are pursuing a similar question with current technology.

Researchers in 2 subsequent intervention studies (2,3) also used these 4 criteria and the same manipulative technique. In these projects, it was necessary to operationally define "patients" as those who would benefit by extension exercises, because the existing guidelines were not adequate to assign patients to the category. The 4 SIJ tests were used to assign patients to the category where it was expected that the manipulation procedure would be of benefit, and the tests were not intended to determine the technique or the side of application. The technique and the side to be treated were determined a priori to answer the research question: When patients with acute low back pain are identified as having 3 or more of the signs, how do they respond to a mobilization technique purported to affect the SIJ in such patients? According to Delitto et al, "Briefly, the technique involved placing the patient supine with the spine laterally flexed to the symptomatic side." (3)(P218)

The study allegedly replicated by Riddle and colleagues is 14 years old. Thus, the stated second purpose, which was "to examine the degree of agreement between therapists by taking into account the side of the presumed dysfunction and the type of asymmetry present," harkens to the past when non-weight-bearing tests were interpreted to describe "malposition" of stable weight-bearing joints. I contend that the problem with this approach is that it fails to recognize the effect of gravity on anatomical landmarks, which succumb to the relative sufficiency of the muscles once gravity is neutralized. The 2 non-weight-bearing tests help, I believe, to identify the muscles placed on stretch (ie, hamstring and rectus femoris muscles with the long sitting and prone knee bending tests, respectively). I argue that these insufficiencies can be confirmed by hamstring muscle length tests and the Thomas test.

The authors, in their first composite analysis, "collapsed all positive ratings (independent of the side and type of asymmetry determined to be present) and determined the extent of agreement when paired therapists rated 3 or more tests as positive or negative." But consider the patient selection bias: the mean duration of pain was 45.2 weeks (SD=80.4). More than one quarter (27%) of their patients reported not having pain that interfered with jobs or housekeeping, and some apparently scored zero on their 10-cm visual analog scale (see authors' Tab. 2). Clearly, this sample had chronic pain by any definition, and they apparently had low-grade pain. In contrast, the subjects in the studies by Delitto et al (3) and Erhard et al (2) had mean durations of pain of 8 (SD=6) and 21 (SD=20) days, respectively. The chronicity and intensity of symptoms differ markedly between the samples and may account for discrepancies in findings. Additionally, early investigations of iliac crest heights using a pelvic level revealed the necessity of locating a level section of the floor each time a measurement was taken. (4) It is unclear whether similar precautions were used in the study by Riddle et al.

The authors' general conclusion is to recommend provocation tests in lieu of tests presumed to measure SIJ alignment or movement. Readers should be aware that the authors suggest abandoning all alignment and movement tests, not just those that they investigated. Provocation was not examined in the study, but several articles were referenced. A review of these articles, however, reveals mixed results at best. Dreyfuss et al, (5) Maigne et al, (6) and Slipman et al (7) all reported an inability to support provocation. Fortin et al (8) relied on injections to people with no complaints of low back pain to establish pain patterns.

Laslett and Williams (9) concluded that 5 of 7 tests yielded reliable measurements, but fell to the same criticism as Cibulka et al, (1) in that the pair of testers having the highest agreement examined 43% of the subjects. Interestingly, Laslett and Williams explained the poor agreement between the results of their study and those of McComb et al (10) by emphasizing their use of multiple training sessions. Broadhurst and Bond, (11) referenced in the "Conclusion" section, examined 6 provocation tests and found 3 of them to be predictive of SIJ pain. Just one of these tests was also found to be of value by Laslett and Williams, and all 6 provocation tests were found to be useless by Dreyfuss, Maigne, and Slipman.

Even if these studies had supported provocation unanimously, it would beg the same question posed by Riddle and colleagues: How does provocation determine choice of the technique or the side to be treated? Injections for everyone?

My mentor once pointed out that if the wrong question is asked, the study is flawed, even if performed perfectly. The pertinent question is: When to use the manipulation?

Richard E Erhard, PT, DC
Assistant Professor
Department of Physical Therapy
School of Health and Rehabilitation Sciences
University of Pittsburgh
6035 Forbes Tower
Pittsburgh, PA 15260

References

(1) 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.

(2) Erhard RE, Delitto A, Cibulka MT. Relative effectiveness of an extension program and a combined program of manipulation and flexion and extension exercises in patients with acute low back syndrome. Phys Ther. 1994;74:1093-1100.

(3) Delitto A, Cibulka MT, Erhard RE, et al. Evidence for use of an extension-mobilization category in acute low back syndrome a prescriptive validation pilot study. Phys Ther. 1993;73:216-228.

(4) NIOSH Low Back Atlas of Standardized Tests And Measurements. Washington, DC: US Dept of Health and Human Services, Public Health Service, Centers for Disease Control, National Institute for Occupational Safety and Health; December 1988.

(5) Dreyfuss P, Michaelsen M, Pauza K, et al. The value of medical history and physical examination in diagnosing sacroiliac joint pain. Spine. 1996;21:2594-2602.

(6) Maigne JY, Aivaliklis M, Pfefer E Results of sacroiliac joint double block and value of sacroiliac pain provocation tests in 54 patients with low back pain. Spine. 1996;21:1889-1892.

(7) Slipman CW, Sterenfeld EB, Chou LH, et al. The predictive value of provocative sacroiliac joint stress maneuvers in the diagnosis of sacroiliac joint syndrome. Arch Phys Med Rehabil. 1998;79:288-292.

(8) Fortin JD, Aprill CN, Ponthieux B, Pier J. Sacroiliac joint: pain referral maps upon applying a new injection/arthrography technique, part II: clinical evaluation. Spine. 1994; 19:1483-1489.

(9) Laslett M, Williams M. The reliability of selected pain provocation tests for sacroiliac joint pathology. Spine. 1994; 19:1243-1249.

(10) McComb PF, Fairbank JCT, Cockersole BC, Pynsent PB. Reproducibility of physical signs in low-back pain. Spine. 1989;14: 908-918.

(11) Broadhurst NA, Bond MJ. Pain provocation tests for the assessment of sacroiliac joint dysfunction. J Spinal Disord. 1998;11: 341-345.

Author Response:

Erhard claims that we "misinterpret and mischaracterize the utilization" of the 4 tests we examined in our study. More specifically, he claims that misinterpretation and misuse of the tests occurred because we concluded that incorrect interventions would likely be selected based on the unreliable measurements obtained from the 4 tests. We are not sure why Erhard believes our conclusion is incorrect. Whether the results of these 4 tests are used to choose an intervention or to assign a patient to a category, any clinical decision based on unreliable measurements is likely to be unsound. We do not recommend using unreliable measurements in clinical practice.

We chose to examine the series of 4 tests originally recommended by Cibulka et al (1) for one simple reason: it was the only study we found that suggested measurements of sacroiliac joint (SIJ) movement or alignment had an acceptable level of reliability. The bulk of the evidence suggested to us that physical therapists should reconsider the use of these types of tests. (2) By replicating the study of Cibulka et al, we hoped to provide additional evidence, one way or the other, to make a recommendation about the utility of these types of tests for clinical practice.

Given our study and evidence from the literature, we believe that when most physical therapists conduct tests of SIJ movement and alignment, the measurements they obtain are unreliable. We indicated, based on our earlier article, (2) that there is evidence to suggest that measurements obtained from tests of pain provocation from the SIJ region are at least moderately reliable. There is also some evidence, albeit weak as pointed out by Erhard, for the validity of measurements obtained from these tests. We believe that the available evidence indicates that, for the large majority of physical therapists, tests of SIJ alignment or movement are not helpful and that data obtained from pain provocation tests are more likely to provide useful information.

Erhard also contends that the duration of our patients' symptoms was excessive. Our patients had a mean symptom duration of 45.2 weeks (SD-80.4, range=1-330). We chose this sample because Cibulka et al (1) did not specify symptom duration for their sample and we had no data-based or theoretical argument as to why we should restrict symptom duration in our sample. In our experience, physical therapists examine for the presence of SIJ region dysfunction no matter the duration of symptoms. Erhard apparently believes that measurements of SIJ dysfunction are more reliable and more useful when obtained on patients with acute low back pain. We re-examined our data and found that we had 32 patients who had symptoms for 8 weeks or less (mean=3.6 weeks, SD=2.4, range= 1-8). The results of this analysis are provided in the Table. The reliability is only slightly higher for patients with acute low back pain than for the entire sample and is still too low for patient use, in our opinion.

Given the lack of evidence for traditional measures of SIJ alignment or mobility, we are encouraged to hear that Erhard and colleagues are working on developing new methods for identifying patients with SIJ region dysfunction. We believe that new methods that have sound psychometric properties are clearly needed to replace the more traditional methods of visual estimation of SIJ alignment and movement.

Erhard concludes by asking how tests of pain provocation determine the choice of intervention or the side to be treated. Based on available evidence, tests of pain provocation do not indicate that a specific intervention should be used. Pain provocation tests, however, do provide therapists with moderately reliable data that could be used in clinical decision making.

We agree that Erhard's last question of when to use manipulation (or any intervention, for that matter) is an important one. The work of Erhard and colleagues (3,4) provides some data to suggest that a manipulative procedure applied to the SIJ region is effective, at least in the short term, for patients with acute LBP and at least 3 positive SIJ tests (out of the 4 SIJ tests examined by Cibulka et al(1)). What the studies of Erhard and colleagues do not tell us is whether the 4 SIJ tests provided the critical diagnostic information. For example, the patients in their studies may have improved following manipulation because they had acute low back pain and, had leg pain only rarely, and not because they had 3 positive tests for SIJ region dysfunction. A recent presentation (5) and an upcoming publication (6) provide initial evidence to indicate whether the measurements obtained using the 4 tests studied by Cibulka et al (1) are valid for predicting patient response to the manipulation studied by Erhard and colleagues. (3,4) The study by Flynn and colleagues (6) suggests to us that the SIJ movement and alignment tests recommended by Cibulka et al (1) do not predict whether patients respond favorably to the manipulation. At least for now, it appears that SIJ movement or alignment tests are not the answer for predicting who will respond favorably to a manipulation.

Table.
Intertester Reliability for Composite Results of the Four Tests for
Patients With Acute Low Back Pain (n=32) (a)

                             %       [kappa]     [P.sub  [P.sub
Finding                   Agreement   (SE)        .pos]   .neg]

Composite analysis 1:
  3 of 4 tests (+ or -)     68.8     .31 (.18)   54.5    76.2

Composite analysis 2:
  3 of 4 tests (+ right,
  + left, negative)         65.6     .25 (.18)   31.0    66.0

Composite analysis 3:
  3 of 4 tests (anterior
  right, anterior left,
  negative, posterior
  right, posterior left)    67.9     .36 (.16)  38.7    67.9

(a) [kappa]=kappa coefficient, [P.sub.pos]=observed proportion
of positive agreement, [P.sub.neg]=observed proportion of negative
agreement.

Daniel L Riddle, PT, PhD
Associate Professor
Department of Physical Therapy
Medical College of Virginia Campus
Virginia Commonwealth University
1200 E Broad St
Richmond, VA 23299-0223
([e-mail])

Janet K Freburger, PT, PhD
NRSA Postdoctoral Research Fellow
Cecil G Sheps Center for Health Services
Research
Assistant Professor
Division of Physical Therapy
University of North Carolina at Chapel
Hill
Chapel Hill, NC

References

(1) 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.

(2) Freburger JK, Riddle DL. Using published evidence to guide the examination of the sacroiliac joint region. Phys Ther. 2001; 81:1135-1143.

(3) Erhard RE, Delitto A, Cibulka MT. Relative effectiveness of an extension program and a combined program Of manipulation and flexion and extension exercises in patients with acute low back syndrome. Phys Ther. 1994;74:1093-1100.

(4) Delitto A, Cibulka MT, Erhard RE, et al. Evidence for use of an extension-mobilization category in acute low back syndrome: a prescriptive validation pilot study. Phys Ther. 1993;73:216-228.

(5) Flynn T, Fritz J, Allison S, et al. Characteristics of patients with low back pain who respond best to sacroiliac region manipulation. J Orthop Sports Phys Ther. 2001;31: A9-A10.

(6) Flynn T, Fritz JM, Whitman J, et al. A clinical prediction rule for classifying patients with low back pain who demonstrate short term improvement with spinal manipulation. Spine. In press.