Preliminary Results of Orthotic Treatment for Chronic Low Back Pain

Thomas M. Gavin, CO
James B. Boscardin, MD
Avinash G. Patwardhan, PhD
Wilton H. Bunch, MD, PhD
Michael R. Zindrick, MD
Mark A. Lorenz, MD
Lori A. Vrbos, RN, MS

ABSTRACT

The use of spinal orthoses for chronic low back pain has traditionally been random and empirical. Frequently, a lumbosacral corset is prescribed and, upon failure of the corset to reduce pain, orthotic treatment is abandoned. In this prospective study, the test instrument of Willner, consisting of a rigid aluminum thoracic lumbosacral orthosis (TLSO) with a sagittal lumbar pad that is adjustable in lumbar flexion or extension, is used to test the hypothesis that a five-day trial in a test orthosis will be a good predictor of the outcome of orthotic treatment for chronic low back pain.

Introduction

Eighty percent of our population experiences low back pain at some time of their lives (1-3). Andersson et al. reported the total cost for low back pain to industry in 1988 was estimated between $26.8 and $56 billion (4). Herron and Turner reported approximately 200,000 lumbar laminectomies for back and/or leg pain are performed yearly in the United States (5). Failure rate of lumbar laminectomies for back and/or leg pain as reported in various studies ranges from I to 48 percent with a 10 to 20 percent poor outcome most frequently noted (6).

Orthoses have been used as a conservative treatment modality for low back pain. Perry summarized the results of a questionnaire answered by 3,460 orthopedic surgeons who reported that when prescribing a lumbosacral orthosis, the primary expectation is reduction of lumbosacral motion and the secondary expectation is to provide some degree of abdominal support (7).

Some clinical studies of orthotic treatment for low back and/or leg pain secondary to spondylolisthesis have been reported. Micheli et al. reported on low back and leg pain secondary to adolescent and young adult spondylolysis and spondylolisthesis in which 100 percent of patients had good to excellent pain relief in Boston orthoses fitted in lumbar flexion (8). The authors also reported a 16 percent nonoperative healing rate of the isthmic fractures in adolescents treated in the study. Willner reported on orthotic treatment of adults with stenosis and symptomatic isthmic spondylolisthesis with documented slippage of 25 to 50 percent of L5 on S1(9). All patients were fitted with rigid TLSOs in lumbar flexion. Eighty-seven percent of patients in this study reported complete pain relief in the orthosis while all patients reported some pain relief.

Both of these studies show significant improvement of symptoms from spondylolysis, spondylolisthesis and stenosis when patients were fitted with rigid orthoses in lumbar flexion. Willner reported that using rigid orthoses to treat low back pain, when prescribed randomly, independent of diagnosis, was effective only about 50 percent of the time (9).

Willner, in a later study, described a test instrument for predicting the success of rigid orthoses for low back pain (10). This test instrument consists of an adjustable lumbosacral orthosis that can either flex or extend the lumbar spine. Willner advocated a 30-minute trial wearing of the test instrument while the patient performed tasks of normal daily activity. Upon terminating the trial, it was then determined if the patient was a candidate for a rigid orthosis, fabricated in the flexion/extension value determined by the test instrument. The test device was designed to test patients in an orthosis before proceeding to a definitive rigid TLSO and to determine if the patients' symptoms improved in flexion, neutral or extension of the lumbar spine.

A prospective study was designed to test the hypothesis that a five-day trial wearing the test instrument of Willner would be a good predictor of the outcome of orthotic treatment for chronic low back pain, would enhance patient compliance and identify optimal lumbar sagittal geometry to reduce symptoms (see Figure 1 ).

Methods

The criteria for patient selection for a five-day trial in the test orthosis was as follows:

Patients had back and/or leg pain for six months or longer.
All had diagnoses documented with either radiographs, M.R. imaging and, in some cases, C.T. scans, myelograms, discograms and/or any combination of the aforementioned diagnostic methods.
All known conventional forms of treatment short of spinal fusion, including bed rest, physical therapy, lumbosacral corset, epidural injection, and in some cases, chiropractic adjustment, acupuncture, traction and laminectomy/discectomy, were exhausted without resolution of pain.
Pain severity was subjectively assessed on a 0-10 scale, 0 being pain-free and 10 being complete incapacitating pain. Patients had to rate between 5-10 on the scale to be considered a candidate for the trial wear.
Patients did not have muscle spasms in the paraspinal musculature.
Patients reported pain relief in either prone, supine or lateral recumbent positions.

Thirty-six patients met the criteria and were selected for trial wear of the test instrument. Twenty-six patients had discogenic pathologies, including disc herniation, multiple level disc bulging, degenerative disc disease and combinations of the previously mentioned. Three patients had post-laminectomy lumbar instability. All discogenic and lumbar instability patients were fitted in maximum lumbar extension in the test instrument (see Figure 2 ).

Four spondylolisthesis patients with slippage of 25 to 50 percent, two with lumbar stenosis and one with facet syndrome were fitted with the test instrument for five days in maximum lumbar flexion (see Figure 3 ).

Patients were asked to carefully monitor any symptomatic changes during the trial wear. Trial wear included all waking hours, excluding driving time for those who felt incapable of operating their automobiles while in the test instrument. All were instructed to keep their activity level equal to or greater than normal during the trial. Patients were instructed to remove the test instrument immediately and return if pain increased at any time. If patients were unsure whether the test instrument was reducing pain, they were asked to remove it for four hours on the fourth day and walk, monitoring any increase of symptoms while out of the test instrument.

After five days in the test instrument all patients were reassessed. Criteria for successful trial wear was: (1) reduction of pain score to 0-5 on the 0-10 scale, (2) an increase of pain to pretrial value within five hours of doffing the test instrument while vertical and (3) patient desired to proceed into a custom-molded TLSO for three to six months. All were given the option to either proceed into a rigid TLSO or to discontinue orthotic treatment. If the decision was made to proceed with the rigid orthosis, a custom-molded low-profile TLSO was fabricated from a plaster impression taken in either lumbar flexion or extension as determined optimal in the trial. The TLSO was then worn for up to six months at which point the orthosis was discontinued, and patients received physical therapy. Patients were then reassessed one to three months after discontinuing use of the orthosis to determine whether they maintained pain relief.

Results

During an 18-month period, 36 patients entered the test instrument for a five-day trial wear. The mean age in this study was 43.4 years (range 17.1 - 80.8 years). The patient pain score average preceding the trial was 7.7 (range 5-9). Upon completing the five-day trial, the average pain score reduced to 3.3 (range 0-8). The relative pain reduction average of this group was 4.3 (range 0-9).

Twenty-nine patients chose to proceed to the rigid TLSO, and all 29 patients fit the criteria of success in the test instrument (see Figure 4 ). The average pain score of the 29 successful patients was 2.8 (range 0-5). The relative pain reduction average of this group was 5.5 (range 3-9). All 29 patients proceeded to rigid TLSOs, fabricated in the geometry deemed optimal in the test instrument for three to six months. Lumbar extension was consistently the optimal method for pain reduction in the discogenic patients, whereas lumbar flexion relieved pain for the spondylolisthesis, stenosis and facet syndrome patients. All 29 patients reported not only a reduction of pain severity but a decrease in the frequency of stabbing pains and an increase in walking endurance while in the test instrument. All patients who proceeded to rigid TLSOs reproduced pain reduction values from the test orthosis to the TLSO.

Of the patients who chose not to proceed from the test instrument to the rigid orthosis after the trial wear, none of the seven fit the criteria of success in the test instrument. Five of these patients reported increased back and leg pain while in the test instrument and doffed the orthosis immediately. Although lumbar extension/flexion angles were altered upon initial report of failure in the test instrument, the desired result was not achieved. All increased symptoms returned to pre-trial levels within five hours of doffing the test instrument. The five patients with the increased symptoms were all diagnosed with discogenic pathologies at the L5-S1 level. Three of these patients were fitted in baycast spicas for one week with near-complete resolution of symptoms.

To date, 25 of the 29 TLSO patients have completed the orthotic treatment course of three to six months and have discontinued their TLSOs. These patients were assessed with an average post-orthosis pain score of 3.3 (range 08). Six patients, after completing orthotic treatment, demonstrated a significant return of symptoms and underwent posterior spinal fusion with instrumentation. Of these six patients, three returned to their normal employment, two were recommended for vocational rehabilitation, and one was retired prior to onset of pain. Of the patients who underwent spinal fusion, all used their TLSOs for postoperative immobilization after adjustments were made to relieve extension/flexion.

Nineteen of the 25 patients maintained a nonoperative status after completing treatment. Of the 19, 11 returned to pre-treatment employment, four were retired prior to treatment, one was a full-time student and returned to school, one was a housewife who returned to normal daily activity, one was recommended for vocational rehabilitation, and one is still pending decision. These results are short-term, ranging from three to 12 months post-treatment.

Statistical Analysis

Due to the experimental nature of this study, all data were analyzed blindly. All descriptive and categorical data were measured on an ordinal scale. A chi-square distribution was applied to all cross-tabulation and contingency tables. Data were summarized according to observed vs. expected frequencies in an attempt to ascertain whether a true experimental effect existed as opposed to results due to chance fluctuations. The Mann-Whitney U test was selected as an alternative to the t-test because it is appropriate in experiments dealing with small samples. In addition, if any ties in the ranks occur, only a minimal effect is exerted. A two-tailed test was performed since no prediction was made by the authors as to the direction of the scores. The significance level was set at .01 (p<.01).

The test statistics were compared to theoretical chi-square tables. Tabled values for the following pairings exceeded the .01 significance level, thus establishing statistical significance: (1) in-test instrument vs. in-TLSO score, (2) in-test instrument vs. post-TLSO score and (3) in-TLSO vs. post-TLSO score.

Discussion

Five mechanisms of action for lumbosacral orthoses have been hypothesized. The most relevant to our work is the concept of limitation of segmental motion of the lumbosacral spine.

Several authors have reported on the effectiveness of an orthosis in limiting segmental motion of the spine (11-13). Generally, lumbosacral orthoses are more effective in limiting motion at the upper lumbar levels than the lower and may actually increase motion at the L5-S1 level. Only the baycast spica was effective in limiting motion at this level.

Other proposed mechanisms of action include limiting gross trunk motion, increasing intra-abdominal cavitary pressure, reducing muscle activity and reducing intradiscal pressure (1421). The results of these works are generally inconsistent and do not seem pertinent here.

Not all orthoses produce the same degree of immobilization. More than half of the patients who were successful in the test instrument previously wore well-fitted lumbosacral corsets without experiencing any reduction in pain. Immobilization, as the treatment of choice, should not be discarded until it is certain that state has been achieved.

Although our preliminary study is small, we have attempted to formulate an hypothesis that may be tested on others as well as in our future work. These ideas are based on specific patients.

The L5-S1 discogenic patients did not do well in the test instrument as they all increased symptoms in extension, neutral and flexion. However, three of the same patients reported near-complete pain relief in the baycast spica. We postulate that immobilization is the mechanism of action that reduces L5-S1 discogenic pain.

Three of our discogenic (not L5-S1) patients were initially fitted in lumbar flexion. All three developed intense sciatic pain within 15 minutes. They were immediately refitted in lumbar extension and had successful results in their trial periods.

A patient with spinal stenosis was fitted in extension, which led to an increase in leg pain. The orthosis was changed into flexion, and the result was a successful trial period. All of the spondylolisthesis patients in our study used their TLSOs for postoperative had slippage of LS on 51 and had significant symptomatic relief in flexion without a thigh extension.

For these patients we propose the sagittal plane alignment is the most likely critical factor and achieving an maintaining optimal lumbar geometry is the primary mechanism of action Lumbar immobilization is a necessary but not sufficient condition for this group of patients. Thus, we add an ad additional proposed mechanism for the action of orthoses in the treatment o low back pain.

These results are short-term, and back complaints tend to reoccur. We make no predictions about long-tern effectiveness. However, our preliminary findings suggest if a patient with chronic low back pain has been treated with all conventional methods-short of spinal fusion-without pain relief. orthotic treatment will provide acceptable results for most patients when prescribed after analysis of five-day trial wear with a test instrument.

THOMAS M. GAVIN, CO, is director of clinical services at BioConcepts inc., Willowbrook, Ill. Gavin is also a research orthotist in the orthopedic biomechanics laboratory at the Rehab R&D Center, DVA hospital, Hines, Ill.

JAMES B. BOSCARDIN, MD, is an orthopedist in private practice with the Parkview Orthopaedic Group in Palos Heights, Ill. Boscardin is also a faculty member of Loyola University Medical Center, Stritch School of Medicine, Maywood, Ill.

AVINASH G. PATWARDHAN, PhD, is professor of orthopedics at Loyola University Medical Center, Stritch School of Medicine, Maywood, Ill. Patwardhan is also the director of the orthopedic biomechanics laboratory in the Rehab R&D Center, DVA hospital, Hines, Ill.

WILTON H. BUNCH, MD, PhD, is presently the medical director at Lakeshore Hospital, Birmingham, Ala. Bunch was the chairman editor of The Atlas of Orthotics, second edition.

MICHAEL R. ZINDRICK, MD, is a private practice orthopedist with Hinsdale Orthopaedic Associates, Hinsdale, Ill. Zindrick is also a faculty member of Loyola University Medical Center, Stritch School of Medicine, Maywood, Ill.

MARK A. LORENZ, MD, is a private practice orthopedist with Hinsdale Orthopaedic Associates, Hinsdale, Ill. Lorenz is also a faculty member of Loyola University Medical Center, Stritch School of Medicine, Maywood, Ill.

LORI A. VRBOS, RN, MS, is a researcher and statistician in the department of orthopedic surgery, Loyola University Medical Center, Maywood, Ill.

References:

Anderson JAD. Back pain and occupation. In: Jayson MIV, ed. The Lumbar Spine and Back Pain. New York: Churchill Livingstone, 1987:16.
Schultz AB. Loads on the lumbar spine. In: Jayson MIV, ed. The Lumbar Spine and Back Pain. New York: Churchill Livingstone, 1987:204.
Wood PHN, Badley EM. Epidemiology of back pain. In: Jayson MIV, ed. The Lumbar Spine and Back Pain. New York: Churchill Livingstone, 1987:1.
Andersson GBJ, Pope MH, Frymoyer JW, Snook S. Epidemiology and cost. In: Occupational Low Back Pain: Treatment and Prevention. Mosby Year Book, 991:110.
Herron LD, Turner J. Patient selection or lumbar laminectomy and discectomy vith a revised objective rating system. Clin Rel Res 1985;199:145. Goel VK, Weinstein JN. Biomechanics of the Spine: Clinical and Surgical Perspecye. Boca Raton, Fla.: CRC Press, 1990:2.
Perry J. The use of external support in the treatment of low back pain. J Bone Dint Surg 1970;52A:1440-2.
Micheli U, Hall JE, Miller ME. Use of the modified Boston brace for back injuries in athletes. AmerJ Sports Med 1980; 8:3516.
Willner S. Effect of a rigid brace on back pain. Acta Orthop Scand 1985;56:40-2.
Willner S. Test instrument for predicting the effect of rigid braces. Prosth Orthot Intl 1990;14:1:22-6.
Norton PL, Brown T. The immobilizing efficiency of the back braces; their effect on the posture and motion of the lumbosacral spine. J Bone Joint Surg 1957; 39A:11l-39.
Lumsden RM, Morris JM. An in vivo study of axial rotation and immobilization at the lumbosacral joint. J Bone Joint Surg 1968;50A: 1591.
Fidler MW, Plasmans CMT. The effect of four types of support on the segmental mobility of the lumbosacral spine. J Bone Joint Surg 1983;65A:943-7.
Waters RL, Morris JM. Effect of spinal supports on the electrical activity of muscles of the trunk. J Bone Joint Surg 1970;52A:51-60.
Nachemson A, Schultz AB, Andersson GBJ. Mechanical effectiveness of lumbar spine orthoses. Scand J Rehab Med (Suppl.) 1983;9:139-49.
Nachemson A, Morris JM. In vivo measurements of intradiscal pressure. J Bone Joint Surg 1964;46A:1077-92.
Morris JM, Lucas DB. Physiological considerations in bracing of the spine. Orthop Prosth Appl 1963;37:44.
Morris JM et al. The role of the trunk in stability of the spine. J Bone Joint Surg 1961 ;43A:327.
Lantz SA, Schultz AB. Lumbar spine orthosis wearing-II. Effect on trunk muscle myoelectric activity. Spine 1986;11:83842.
Lantz SA, Schultz AB. Lumbar spine orthosis wearing-I. Restriction of gross body motions. Spine 1986; 11:834-7.
Krag MH, Byrne KB, Pope MH, Bayliss D. The effect of back braces on the relationship between intra-abdominal pressure and spinal loads. Advances in Bioengineering 1986;2:12:22-3.