Persistent buttock pain caused by sacroiliac dysfunction has been documented by Fortín et al.1 Although some would argue that this is not a real clinical entity, precise injection studies have demonstrated it is a cause of chronic low back pain.2 In the Fortin et al1 study, of 100 consecutive patients who had chronic back pain that was undefinable by standard physical examination and radiographic procedures, 43 had pain in the buttock overlying a sacroiliac joint and 30% had appropriate pain relief by specific injections (13% of the entire group). Thus, given this strict criteria of reproduction and relief by two separate injections, sacroiliac incompetence can be symptomatic in a significant number of people.
There is question as to the appropriate treatment for these patients. Although cortisone injections have been documented as effective,3 the antiinflammatory effect is not permanent. In addition, the injections do not offer an opportunity to stabilize an incompetent joint.
Manipulation also has been demonstrated to be effective in sacroiliac pain.4 However, the exact mechanism of pain relief and how manipulation could create long-term stabilization in an incompetent joint is unclear. Moreover, what is being manipulated and how one measures the "dose" of manipulation remains puzzling.
Other treatment options for stabilization of an incompetent joint include a sacroiliac belt or fusion. A sacroiliac belt can be effective in certain circumstances,5 while fusion of the unstable joint may be successful in most patients.6 However, orthotic devices often are poorly tolerated, and fusion is an extreme form of stabilization with an unavoidable rate of failure and complications.
Stabilization of the sacroiliac joint using the basis of dynamic muscle activity via an exercise program may offer an additional treatment option. A theoretical rationale for sacroiliac stabilization by muscle activity was presented by Vleeming et al.7 In their study, anatomic dissections revealed a strong connection between the gluteus on one side and the latissimus dorsi on the other by means of the thoracolumbar fascia. In cadaver studies, loads were directly transferred from one muscle group to the other. This is a reciprocal relationship that has been suggested in the past8 but not documented in living patients.
This study sought to document the reciprocal relationship between the contralateral gluteus and the latissimus dorsi, and evaluate me efficacy of physical training of these stabilizing muscles for treatment of sacroiliac pain.
Figure 1 : MedX torso machine (MedX, Ocala, FIa).
MATERIALS AND METHODS
To test this stabilization hypothesis, it is necessary to rotate the trunk without allowing any pelvic motion, as well as evaluate natural rotation during gait. If the actions of the gluteus and latissimus dorsi during rotation as described above are confirmed, this would confirm the theory developed from the dissections in the cadaver studies.
Fifteen healthy individuals (12 women and 3 men) without any history of back or leg pain complaints were recruited to be evaluated for surface myoelectric activity of their latissimus and gluteus musculature. Electromyographic (EMG) results were recorded on the ME3000p (Mega Electronics Ltd, Finland).
Electrodes were placed in the same locations on all study participants; the interelectrode distance was 1.5*. The electrodes were copper and 3/8" in diameter (Mediocotest M-OO-S, 01stykke, Denmark).
Electrodes were placed on the right and left gluteus and the latissimus on both sides. Electrodes on me gluteus were located 10 cm medial and parallel at the level of the greater trochanter. Latissimus electrodes were placed 1 cm medial and inferior to the interior lateral comer of the scapula.
Electrodes came packed with their own gel, and the ME3000p system is designed to have high input impedance; therefore, skin preparation was not necessary. Two test performances were carried out. The EMG bandwidth was 15500 Hz. Data were gathered by a digital recorder at a sampling rate of 1000 Hz. Data were collected and stored, and subsequently downloaded on a personal computer by way of a fiber-optic connection. Data were analyzed with programs from Mega Electronics as well as our own program.
In the first test, the surface myoelectric activity was tested during gait using a tteadmill at constant speed. Study participants also were placed in a MedX torso rotation machine (MedX, Ocala, FIa). This device evaluates strength during torso rotation while the patient is sitting with hip and leg restraints that stabilize the pelvis but allow rotation in a neutral sitting position. Participants performed dynamic eccentric and concentric exercises on the torso rotation machine9 (Figure 1). Initial resistance was set at 50% of the maximum isometric torque noted at initial testing.
Based on the stabilization hypothesis, mis same equipment was used to treat patients with sacroiliac pain. Five women who had a clinical history consistent with sacroiliac pain were treated. They all had unilateral buttock pain with a history of sudden onset. Pain duration averaged 8 months. In all of the patients, pain was relieved by steroid and local anesthesia injections into their sacroiliac joint under fluoroscopy.
For testing, surface electrodes were placed in the same location as on the asymptomatic participants. Patients underwent strength testing and then began the same twice-weekly exercise protocol as asymptomatic participants, starting at 50% of initial isometric torque. Once patients could achieve 20 repetitions at a given resistance, the amount of resistance was increased approximately 5% at the next exercise session. Training lasted for approximately 2.5 months with an average of 20 sessions.
All healthy participants walking on a treadmill at a constant rate confirmed the reciprocal relationship between me latissimus on one side versus the gluteus maximus on the contralateral side (Figure 2). Raw EMGs demonstrated no significant difference between men and women. When the root mean square of all 15 healthy participants was displayed graphically, the signal amplitude of the latissimus was far less than that of the gluteus. The right gluteus usually had a lower root mean square than the left. Twelve of the 15 healthy participants were right handed (Figure 3). This reciprocal relationship of muscles correlates with normal reverse rotation of shoulders versus pelvis in normal gait.
When the healthy participants exercised on the torso rotation machine, the reciprocal relationship between the latissimus on one side and gluteus on the contralateral side again was evident (Figure 4). With right rotation of the trunk, the right latissimus dorsi is significantly more active than the left, but the left gluteus is more active than the right gluteus, although less than the latissimus. The torso goes through an arc of 720° from the right to left rotation and reverse.
There was no significant difference in the EMG amplitude between rotation to the left or right (Figure 5). When the root mean square of all healthy participants was displayed, the findings of the raw EMG patterns were reflected by all participants. In the sitting pelvisfixed posture, the latissimus was more active than the gluteus, but the gluteus contralateral from the active latissimus was more active than the homolateral gluteus maximus (Figures 6 and 7).
Sacroiliac patients demonstrated a strikingly different pattern. On initial evaluation, the gluteus on the symptomatic side was more active than demonstrated by the healthy participants (Figure 8). However, the reciprocal relationship between latissimus and gluteus was still present.
Sacroiliac patients remained in the training program until strength gains plateaued. At completion of training, there was a significant increase in isometric strength. The raw data from a typical patient are presented in Figure 8. Myoelectric activity changed with training. The reason for improvement in strength is demonstrated by increased latissimus activity after training. There also was diminished activity of the gluteus on the involved side (Figure 8). Gluteus EMG did not return to the low levels typical of the asymptomatic participants, but did return to a more physiologic status. The average change in strength is depicted in Figure 9. The range was from an 18% to 29% increase. The other four sacroiliac patients demonstrated similar myoelectric patterns.
Figure 2: EMG pattern of a woman walking at constant speed on a treadmill. During ambulation, there is a reciprocal function between the latissimus dorsi on one side and the gluteus on the contralateral side. Figure 3: Summary of myoelectric activity depicted as an average root mean square (amplitude) of the gluteus versus the latissimus in healthy individuals. The right gluteus consistently used less amplitude than the left. The amplitude of the left and right latissimus while walking was minimal but equal. Figure 4: Raw EMG pattern of a healthy woman carrying out rotation on the torso rotation machine shows a greater degree of right latissimus activity contrasted to the reciprocal left gluteus activity. Figure 5: In the torso rotation machine, the mean amplitude (all healthy participants) of left and right latissimus dorsi was approximately the same. Figure 6: Normal activity of the latissimus dors» (all healthy participants) in the torso rotation machine was strongest in the direction of rotation; the contralateral gluteus was more active than the homolateral gluteus. Figure 7: Similar characteristics of function were noted in right rotation with the right latissimus more active than the left; the contralateral gluteus (left) was more active than the homolateral. Figure 8: Raw EMG recording of a patient with a painful right sacroiliac joint going through left torso rotation demonstrates hyperactivity of the gluteus on the painful side (A). Raw EMG from the same patient documents that with improvement in strength and reduction of pain, EMG during left torso rotation returned to the normal status of large latissimus activity contrasted to small gluteus activity (B). Channel 4 (left gluteus) has now become more active to substitute for the right gluteus. (Abbreviations: LL=left latissimus [channel 1 ], RL=right latissimus [channel 2], RG=right gluteus [channel 3], and LG=left gluteus [channel 4]). Figure 9: Average increase for all patients in strength of rotation from the beginning to the conclusion of the exercise program (approximately 2.5 months). Strength diminished compared to normal, especially at full external rotation.
Symptoms improved markedly in all patients. None of the patients required continued pain medication and all returned to normal physical activity.
Although root mean square myoelectric activity cannot be truly quantified, these qualitative myoelectric findings confirm the interrelationship between the latissimus on one side and the gluteus maximus on the contralateral side as demonstrated by the anatomic findings of Vleeming et al.7 This was demonstrated both in walking by the healthy participants and torso rotation in both sacroiliac patients and healthy participants.
We cannot explain the finding that the right gluteus maximus consistently had less amplitude than the left during gait. It should be noted, however, that in a large series of patients diagnosed as having primary sacroiliac pain, 45% had pain on the right, 35% had pain on the left, and 20% had bilateral pain.10 The small number of patients and lack of knowledge of leg dominance are insufficient data on which to speculate.
The reciprocal gluteus-latissimus relationship was suggested by Gracovetsky8 from mathematical analysis of the biomechanics of gait. This relationship, however, has seldom been described in articles on gait analysis."
From a therapeutic sense, the importance of these findings is that they offer an opportunity for a rational treatment program for dysfunction to the sacroiliac joint. Stabilization of a symptomatic and apparently incompetent joint is certainly theoretically possible by simultaneously activating the two major muscles whose fascial connection traverses the sacroiliac joint.
An incompetent sacroiliac joint is a difficult diagnosis to define and thus difficult to treat rationally. There are few clinical tests believed to be reliable.2 In a study by Schwarzer et al,2 in which the sacroiliac joint was carefully injected to evaluate pain production and relief, no specific physical examination test correlated with all of the positive responders. It also has been determined that there is no difference in the amount of motion at the sacroiliac joint in symptomatic versus nonsymptomatic individuals.12 It also has been demonstrated that abdominal musculature can carry out a self-bracing effect on the sacroiliac joint, both during standing and sitting.13
In healthy, asymptomatic individuals, we found that during sitting, the latissimus dorsi in the torso rotation movement had significantly greater EMG activity than the contralateral gluteus. This would be expected based on the biomechanics of the situation. In the sitting position, the hip is not moving. However, our finding that the gluteus maximus EMG activity in this test situation was significantly greater in patients with sacroiliac painful dysfunction than in healthy individuals may indicate the gluteus maximus tries to stabilize the apparently incompetent sacroiliac joint. All patients demonstrated this hyperactivity. The more normal EMG activity of the gluteus after treatment, associated with increased strength, supports the view that stabilization by the musculature is possible and probably associated with less inhibition from the more stabilized sacroiliac joint.
The comparison of myoelectric patterns in normal individuals to symptomatic patients forms the basis of this study. Because individuals vary as to their strategies of myoelectric activity, no statistical statements can be made. However, the differences in patterns are apparent from the various depictions noted above. A surprising finding was significant hyperactivity of the gluteus on the painful side during sitting when stress was applied to the sacroiliac joint during torso rotation. The assisting role of the contralateral latissimus also is important and thus forms a rational goal for rehabilitative exercise programs.
1 . Fortin JD, Aprili CN, Ponthieux B. Pier J. Sacroiliac joint: pain referral maps upon applying a new injection/arthrography technique, II: clinical evaluation. Spine. 1994; 19:1483-1489.
2. Schwarzer AC, Aprili CN, Bogduk N. The sacroiliac joint in chronic low back pain. Spine. 1995; 20:31-37.
3. Bernard TN, Kirkaldy- Willis WH. Recognizing specific characteristics of nonspecific low back pain. Clin Orthop. 1987: 217:266-280.
4. Cassidy JB, Kirkaldy-WiUis WH, MacGregor M. Spinal manipulation for the treatment of chronic low back and leg pain: an observational study. In: Berger AA, Greenman PE, eds. Empirical Approaches to the Validation of Spinal Manipulation. Springfield, IU: Charles C. Thomas Publisher; 1985:119-148.
5. Vleeming A, Buyurk M, Stoekart RA. A study of the biomechanical effects of pelvic belts. Am J Obstet Gynecol. 1992; 162:535-543.
6. Moore MR. Diagnosis and surgical treatment of chronic painful sacroiliac dysfunction in the integrated function of the lumbar spine and sacroiliac joint. In: Vleeming A, Mooney V, Dormán T, Snijders C, eds. The Integrated Function of Lumbar Spine and Sacroiliac Joint. New York. NY: Churchill Livingstone; 1995:341-354.
7. Vleeming A, Pool-Goudzwaard AL. Stoekart R, The posterior layer of the thoracolumbar fascia: its function in load transfer from spine to legs. Spine. 1995; 20:753-758.
8. Gracovetsky SA. Locomotion - linking the spinal engine with the leg. In: Vleeming A, Mooney V, Dormán T, Snijders C, eds. The Integrated Function of Lumbar Spine and Sacroiliac Joint, New York, NY: Churchill Livingstone; 1995:171-174.
9. Carpenter DJ, Graves M, Pollock S, et al. Quantitative assessment of isometric torso rotation net muscular torque. Arch Phys Med Rehab. 1991;72:804-808.
10. Bernard TN Jr. Cassidy JD. The sacroiliac joint syndrome - pathophysiology, diagnosis and management. In: Frymoyer JW, ed. The Adult Spine: Principles and Practice. New York. NY: Raven Press; 1991:2107-2130.
11. Gauge JR, Deluca PA, Rinshaw TS. Gait analysis: principles and applications. J Bone Joint Surg Am. 1995; 77: 1607- 1 623.
12. Sturesson BT, Selvick G. Uden A. Movements of the sacroiliac joints. A roentgen stereophotogrammetric. Spine. 1989; 14:162165.
13. Soijáers CJ, Bakker MP, Vleeming A. Stoeckart R. Oblique abdominal muscle activity on standing and sitting on hard and soft seats. ClinBiomech. 1995; 10:1073-1078.