| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
Pain Management Divisions, Departments of Anesthesiology and Critical Care Medicine, Johns Hopkins Medical Institutions, Baltimore, MD and Walter Reed Army Medical Center, Washington, DC
Address correspondence and reprint requests to Steven P. Cohen, MD, Johns Hopkins Hospital Pain Management Center 550 North Broadway, Suite 301 Baltimore, MD 21205. Address electronic mail to scohen40{at}jhmi.edu.
 |
Abstract |
|---|
 Top Abstract AnatomyFunction and BiomechanicsPrevalenceMechanism of InjuryDiagnosis and PresentationTreatmentConclusionsReferences |
|---|
|
Anatomy |
|---|
|
TopAbstractAnatomyFunction and BiomechanicsPrevalenceMechanism of InjuryDiagnosis and PresentationTreatmentConclusionsReferences |
|---|
|
|
Age-related changes in the SI joint begin in puberty and continue throughout life. During adolescence, the iliac surface becomes rougher, duller, and coated in some areas with fibrous plaques. These senescent changes accelerate during the third and fourth decades of life and are manifested by surface irregularities, crevice formation, fibrillation and the clumping of chondrocytes. Degenerative changes on the sacral side generally lag 10–20 yr behind those affecting the iliac surface. In the sixth decade, motion at the joint may become markedly restricted as the capsule becomes increasingly collagenous and fibrous ankylosis occurs. By the eighth decade of life, erosions and plaque formation are inevitable and ubiquitous (4).
Innervation
The innervation of the SI joint remains a subject of much debate. The lateral branches of the L4-S3 dorsal rami are cited by some experts as composing the major innervation to the posterior SI joint (1). Other investigators claim that L3 and S4 contribute to the posterior nerve supply (6,7). The innervation of the anterior joint is similarly ambiguous. Early 20th century German literature asserts the anterior SI joint is supplied by the obturator nerve, superior gluteal nerve and the lumbosacral trunk (8). More recent literature suggests the anterior joint is innervated by L2-S2 (1), L4-S2 (9), and the L5-S2 ventral rami (10). Some authors have even suggested that the anterior SI joint is devoid of nervous tissue (7,11). In a study testing the ability of L5 dorsal ramus and S1-4 lateral branch blocks to protect the SI joint from an experimental stimulus, 6 of 10 subjects retained the ability to perceive ligamentous probing (12).
A neurophysiologic study conducted in cats identified 29 mechanosensitive afferent units, 26 of which were found in the joint capsule and 3 in adjacent muscles (13). Twenty-eight of these units were classified as nociceptive and 1 as proprioceptive. Among these 29 receptive fields, 16 were located in the proximal third of the posterior SI joint and 11 in the middle third. The average mechanical threshold of an SI joint nociceptive unit was 70 g, as compared to the 6 g mean mechanical threshold for lumbar facet joint nociceptive units and the 241 g threshold for units residing in the anterior lumbar disk (14–16). This indicates that the pain sensitivity of the SI joints may be lower than that of the lumbar facet joints but higher than the anterior portions of lumbar discs. As all animals underwent posterior midline incisions, somatosensory units in the anterior SI joint were not stimulated.
|
Function and Biomechanics |
|---|
|
TopAbstractAnatomyFunction and BiomechanicsPrevalenceMechanism of InjuryDiagnosis and PresentationTreatmentConclusionsReferences |
|---|
There have been numerous attempts to discern the biomechanics of the SI joint. These motion studies can be summarized as follows: the SI joint rotates about all 3 axes, although the movements are very small and difficult to measure (19,20). Miller et al. (21) studied the load-displacement behavior of single and paired SI joints in 8 elderly cadavers. Various static test loads were applied in the superior, lateral, anterior and posterior directions, and rotations about all 3 axes were measured. These tests were conducted with one and both ilia fixed. The authors found that with 1 leg immobile, movements in all planes ranged from between 2 to 7.8 times more than that measured with both legs fixed. In a series of cadaveric studies, Vleeming et al. (22,23) found that the total range of motion during flexion and extension at the SI joint rarely exceeded 2 degrees, with 4 degrees being the upper limit during sagittal rotation. In another cadaver study, Brunner et al. (24) found that the main motion in male specimens tended to be translation, whereas in female specimens it was rotational. The maximum range of motion in this study was 1.2 degrees in men and 2.8 degrees in women. In a study by Egund et al. (25) examining SI joint movements in 4 volunteers using radiographic stereophotogrammetry, the authors found the maximal rotations and translations to be 2.0 degrees and 2.0 mm, respectively. A larger study (n = 24) by Jacob and Kissling (26) conducted in healthy young and middle-aged men and women found similarly small limits on rotation (1.7 degrees) and translation (0.7 mm). However, one individual with a history of SI joint pain exhibited more than 6 degrees of rotation about the y-axis. Finally, Sturesson et al. (27) measured multiple SI joint movements in 25 patients diagnosed with SI joint pain. Movements in all planes were found to be small, with translations never exceeding 1.6 mm, and rotations being limited to 3 degrees. No differences were found between symptomatic and asymptomatic joints, leading the authors to conclude that 3-dimensional motion analysis was not useful for identifying painful SI joints in most patients. Possible exceptions to the finding that hypermobility is not a typical cause of SI joint pain include traumatic instability, multiparity, muscular atrophy, and lower motor neuron disease (28).
|
Prevalence |
|---|
|
TopAbstractAnatomyFunction and BiomechanicsPrevalenceMechanism of InjuryDiagnosis and PresentationTreatmentConclusionsReferences |
|---|
Schwarzer et al. (30) conducted a prevalence study involving 43 consecutive patients with chronic LBP principally below L5-S1 using fluoroscopically guided SI joint injections. Fifty-seven other patients with LBP were excluded on the basis of more rostral symptoms. Three criteria were used to diagnose SI joint pain: analgesic response to local anesthetic (LA), abnormalities on post-arthrography computed tomography (CT) scanning, and concordant pain provocation during joint distension. Using significant pain relief after LA injection as the sole criterion for diagnosis, the prevalence of SI joint pain in the 43 subjects was determined to be 30% (95% confidence interval [CI], 16%–44%), with 4 patients obtaining complete pain relief. Using analgesic response combined with a ventral capsular tear (the most common radiologic finding) as the criteria, the prevalence decreased to 21% (95% CI, 9%–33%). Only 7 patients satisfied all 3 diagnostic criteria, for a lower limit prevalence rate of 16% (95% CI, 5%–27%). The presence of groin pain was the only referral pattern found to distinguish patients with SI joint pain from those with LBP of non-SI joint origin.
Maigne et al. (31) conducted a prevalence study in 54 patients with unilateral LBP using a series of blocks done with different LA based on International Spinal Injection Society guidelines (32). Nineteen patients had a positive response (
75% pain relief) to the lidocaine screening block. Among these patients, 10 (18.5%) responded with >2 h pain relief after the confirmatory block with bupivacaine and were considered to have true SI joint pain (95% CI, 9%–29%). Based on these studies, the prevalence of SI joint pain in carefully screened LBP patients appears to be in the 15%–25% range.
|
Mechanism of Injury |
|---|
|
TopAbstractAnatomyFunction and BiomechanicsPrevalenceMechanism of InjuryDiagnosis and PresentationTreatmentConclusionsReferences |
|---|
In addition to etiologic sources, there are numerous factors that can predispose a person to gradually develop SI joint pain. Risk factors that operate by increasing the stress borne by the SI joints include true and apparent leg length discrepancy (37), gait abnormalities (38), prolonged vigorous exercise (39), scoliosis (40), and spinal fusion to the sacrum (41). Whereas increased SI joint uptake using scintigraphy has been demonstrated after lumbar spine fusion (42), at least one study examining the long-term effects of spinal fusion on SI joint function concluded that neither biomechanical nor anatomical changes were more common in fusion patients than in those who underwent decompression procedures (43). Lumbar spine surgery has also been purported to trigger SI joint pain for reasons unrelated to increased force transmission. These factors include SI ligament weakening and/or surgical violation of the joint cavity during iliac graft bone harvest (44) and postsurgical hypermobility (45).
Pregnancy predisposes women to SI joint pain via the combination of increased weight gain, exaggerated lordotic posture, the mechanical trauma of parturition, and hormone-induced ligamental laxity (46,47). The laxity associated with pregnancy is attributable to increased levels of estrogen and relaxin, and it predisposes parturients to sprains of the SI joint ligaments. SI subluxation has also been reported to cause back pain in pregnancy (48).
Inflammation of one or both SI joints is considered to be an early and prominent symptom in all seronegative and HLA-B27-associated spondylarthropathies (49). Although the precise etiology of spondylarthropathy remains unknown in most patients, the strong association with HLA-B27 supports the view that these conditions are attributable to a genetically determined immune response to environmental factors in susceptible individuals. In a subset of patients with Reiter's syndrome/reactive arthritis, the disease is clearly induced by infection (50) (Table 1).
|
The specific etiologies that can result in SI joint pain are widespread and protean. Potential causes range from rare events such as pyogenic infection (51) and malignancy (52), to more mundane occurrences such as bracing one's legs in a motor vehicle accident (53), falls (53), athletic injuries (54), prolonged lifting and bending (55), and torsional strain (55). In a retrospective study by Chou et al. (56) assessing the inciting events in 54 patients with injection-confirmed SI joint pain, the authors found trauma was the cause in 44% of patients, 35% were idiopathic, and 21% were attributed to the cumulative effects of repeated stress. In the 24 patients who cited trauma as the source of their pain, the most common events were motor vehicle accidents (n = 13), falls onto the buttock (n = 6), and childbirth (n = 3).
|
Diagnosis and Presentation |
|---|
|
TopAbstractAnatomyFunction and BiomechanicsPrevalenceMechanism of InjuryDiagnosis and PresentationTreatmentConclusionsReferences |
|---|
|
The reliability of provocative SI joint maneuvers and alignment/mobility tests has also been questioned, with most studies conducted by chiropractors and physical therapists. Whereas some of these studies have found moderate to high inter-examiner reliability (61–63), most have not (64–68). Generally, reproducibility has been found to be greater for provocative tests than for mobility and alignment assessments. In the Dreyfuss et al. study (59) conducted in 85 patients with injection-confirmed SI joint pain, there was moderate agreement (kappa = 0.6) between chiropractors and medical doctors with regard to provocative maneuvers of painful joints. Even when agreement was perfect, the maneuvers were still found to lack diagnostic utility.
Radiological Studies
Results of studies examining radiologic findings in patients with SI joint pain have been similarly disappointing. In studies by Maigne et al. (69) and Slipman et al. (70), the investigators found sensitivities of 46% and 13%, respectively, for the use of radionuclide bone scanning in the identification of SI joint pain. Despite the high specificites in these studies (89.5% for Maigne et al. and 100% for Slipman et al.), the low sensitivies indicate bone scanning is a poor screening test for SI joint pain. Poor correlation with diagnostic injections and symptoms have also been found for CT and radiographic stereophotogrammetry (71,27). In a retrospective analysis by Elgafy et al. (71), CT imaging was found to be 57.5% sensitive and 69% specific in diagnosing SI joint pain.
Pain Referral Patterns
There have been several attempts to identify pain referral patterns from SI joints. In one of the earliest studies conducted in 10 asymptomatic volunteers, Fortin et al. (72) performed provocative SI joint injections using contrast and lidocaine. Sensory changes were localized to the ipsilateral medial buttock inferior to the posterior superior iliac spine in 6 of the 10 subjects. In 2 subjects, the area of hyperesthesia extended to the superior aspect of the greater trochanter. The last 2 subjects experienced sensory changes radiating into the upper thigh. In a follow-up study, independent examiners selected 16 individuals among 54 with chronic LBP whose pain diagrams most closely resembled the pain referral patterns obtained in the first study (73). These 16 patients proceeded to undergo provocative SI joint injections with contrast and LA. All 16 experienced concordant pain during the injection, with 14 obtaining pain relief after deposition of LA. Ten patients reported 50% pain reduction. Six of the 16 patients had ventral capsular tears revealed during arthrography. After the SI joint injections, provocative discography and lumbar facet joint injections were performed in 9 patients each. In none of the patients was either test positive.
Slipman et al. (74) conducted a retrospective study to determine the pain referral patterns in 50 patients with injection-confirmed SI joint pain. In contrast to the findings by Fortin et al. (72) and Schwarzer et al. (30), the authors found the most common referral patterns for SI joint pain to be radiation into the buttock (94%), lower lumbar region (72%), lower extremity (50%), groin area (14%), upper lumbar region (6%), and abdomen (2%). Twenty-eight percent of patients experienced pain radiating below their knee, with 12% reporting foot pain. Based on the existing data, the most consistent factor for identifying patients with SI joint pain is unilateral pain (unless both joints are affected) localized predominantly below the L5 spinous process (30,59,72–74).
Diagnostic Blocks
It is often assumed that an analgesic response to a properly performed diagnostic block is the most reliable method to diagnose SI joint pain. Although this may seem to be self-evident, the validity of intraarticular SI joint blocks remains unproven. There are many factors that can impact on the sensitivity and specificity of diagnostic blocks. These include the placebo effect, convergence and referred pain, neuroplasticity and central sensitization, expectation bias, unintentional sympathetic blockade, systemic absorption of LA, and psychosocial issues (75). In addition, SI joint block can be one of the most challenging spinal injection procedures. Extravasation of LA to surrounding pain-generating structures such as muscles, ligaments, and lumbosacral nerve roots can lead to false-positive blocks. Conversely, failure to obtain adequate LA spread to the anterior and cephalad portions of the SI joint can result in false-negative blocks. In a classic study by North et al. (76) examining the specificity and sensitivity of a battery of lumbosacral LA blocks in 33 patients with a chief complaint of sciatica, the authors found the specificity of all blocks to be exceedingly low. SI joint blocks were not performed in this study.
In a pilot study by Fortin et al. (72) mapping SI joint referral patterns in asymptomatic volunteers, extravasation of contrast (mean 1.6 mL injected) occurred in 9 of 10 subjects during SI joint injection, with half having at least moderate spread outside the joint. After the injection of LA, 40% of subjects noted lower extremity numbness, indicating inadvertent anesthetization of the lumbosacral nerve roots. In the Maigne et al. (31) study, 3 of the initial 67 patients were excluded because of "sciatic palsy" after the screening block and another 7 were excluded because penetration of the SI joint was impossible. Other investigators have reported much less frequent (<5%) failure rates with fluoroscopically guided SI joint injections (30,59,77). Technical difficulties may be more frequently encountered in elderly patients and those with spondylarthropathies, in whom degenerative changes are more pronounced. If persistent difficulties entering the SI joint are encountered using fluoroscopy, the 3-dimensional imaging capabilities of CT may facilitate entry into the joint (34, 78) (Fig. 3).
|
Regardless of the imaging modality used to confirm intraarticular injection, SI joint injections should never be performed blindly. Rosenberg et al. (79) performed a double-blind study in 37 patients (39 joints) to determine the accuracy of clinically guided SI joint injections using CT imaging as the standard. The authors found that intraarticular injection was accomplished in only 22% of patients, whereas sacral foraminal spread occurred 44% of the time. In 3 patients, no contrast was seen on CT scanning, indicating probable vascular uptake. In 24% of injections, contrast extended into the epidural space.
As for facet blocks, some experts have advocated using a series of SI joint blocks to reduce the incidence of false-positives. In a prospective study involving 67 patients with unilateral LBP, SI joint-compatible referral patterns and joint tenderness, Maigne et al. (31) sought to determine the prevalence of SI joint pain using a series of blocks with 2 different LA. In the 54 patients who completed the study, 19 obtained 75% pain relief with the lidocaine screening block. After the bupivacaine confirmatory block, only 10 of the 19 patients achieved 75% pain relief lasting 2 or more hours, for a prevalence rate of 18.5%. The false-positive rate of 17% in this study is less than that previously reported for lumbar facet blocks (80). Yet without a diagnostic "gold standard," there is no way of determining how many true positives were false positives and how many false positives were actually true positives. In clinical practice confirmatory SI joint blocks are almost never performed because a) the block itself is considered to be definitive treatment; b) double-blocks are not cost-effective (81); and c) the negative consequences of obtaining a false false-positive block (misdiagnosing true SI joint pain) outweigh the ramifications of overdiagnosing the condition. In summary, there is no infallible, universally accepted method for diagnosing pain originating in the SI joint(s).
|
Treatment |
|---|
|
TopAbstractAnatomyFunction and BiomechanicsPrevalenceMechanism of InjuryDiagnosis and PresentationTreatmentConclusionsReferences |
|---|
Psychosocial Issues
Recent studies have provided incontrovertible evidence that psychopathology and other psychosocial factors can influence both the development of chronic pain conditions and the response to treatment. In a study by Polatin et al. (82) conducted in 200 chronic LBP patients, the authors found that 77% met lifetime criteria and 59% demonstrated current symptoms for at least one psychiatric diagnosis, with the most common being depression, substance abuse, and anxiety disorders. Notably, more than 50% of those with depression and more than 90% of patients with substance abuse or an anxiety disorder experienced symptoms before the onset of LBP. Most, but not all, studies have shown untreated psychopathology to negatively affect LBP treatment outcomes (83).
In addition to psychiatric illness, social factors have been demonstrated to impact the prognosis of LBP. These include return-to-work issues, secondary gain, catastrophizing, poor role models, codependency behavior, inadequate coping mechanisms, and attitudes, beliefs, and expectations (84). To optimize outcomes, the identification and treatment of concomitant psychosocial issues is of paramount importance. This is best accomplished via a multidisciplinary approach.
Conservative Management
The non-interventional management of SI joint pain should ideally address the underlining pathology. In patients with true or apparent leg length discrepancy, this might include the use of shoe inserts to more equitably distribute the load borne by the SI joints. Because leg length discrepancies are frequently found in asymptomatic individuals (37) and many patients already compensate for their lower extremity length difference by altering their gait or posture, most experts recommend starting out cautiously with inserts that correct only half the incongruity. For SI joint pain resulting from altered gait mechanics and spine malalignment, physical therapy and osteopathic or chiropractic manipulation have been reported to reduce pain and improve mobility (85,86). However, there are no prospective, controlled studies supporting these modalities.
Nonsurgical stabilization programs have been advocated for SI joint pain. These range from the application of pelvic belts that reduce the sagittal rotation of incompetent SI joints in pregnant women (87,88) to exercise-induced pelvic stabilization programs (89). In a study by Mooney et al. (89), the authors found that 5 women with injection-confirmed SI joint pain had electromyographic-documented hyperactivity of the ipsilateral gluteus muscles and contralateral latissimus muscle compared with 15 asymptomatic control patients. After a 2-1/2 month exercise program, all 5 patients achieved a significant reduction in pain and a return of myoelectric activity to normal patterns.
Ankylosing spondylitis (AS), the most well known and studied spondylarthropathy, is an inflammatory rheumatic disease characterized by spine and SI joint involvement that manifests as spondylitis and sacroiliitis. There are numerous studies assessing the efficacy of various pharmacologic treatment modalities in AS and other spondylarthropathies, but the conclusions that can be drawn from these studies are limited by several factors. These include the protean nature of rheumatic involvement, the systemic manifestations accompanying the disorders, and the fact that although the outcome measures generally include changes in pain complaints and mobility, most do not specifically address SI joint pain. Table 3 lists the evidence supporting various pharmacotherapies for AS and other spondylarthropathies.
|
Intraarticular Injections
Intraarticular injections with steroid and LA often serve the dual function of being therapeutic and aiding in diagnosis. To summarize these studies, most but not all investigators have found radiologically guided SI joint injections to provide good to excellent pain relief lasting from 6 mo to 1 yr (Table 4). Along with a multitude of studies demonstrating prolonged pain relief after intraarticular SI joint steroid injections, double-blind studies have shown a beneficial effect for periarticular corticosteroid treatment as well (35,36).
|
Radiofrequency Denervation Procedures
Several investigators have performed radiofrequency (RF) denervation procedures in an attempt to provide prolonged pain relief to patients suffering SI joint pain. The techniques used have ranged from denervating the nerves supplying the SI joint (90–92) to cre-ating lesions in the joint itself (93), with one study using a combination of the two (94). The success rates of studies targeting the nerve supply are higher than those focusing on the joint itself, with approximately two thirds of patients reporting significant pain relief. The major drawback to percutaneous RF denervation procedures is that they should not be expected to alleviate pain emanating from the ventral SI joint. In the study by Schwarzer et al. (30), ventral capsular pathology was shown to account for 69% of all CT pathology in the 13 patients with a positive response to diagnostic SI joint blocks. Complicating matters further are that the nerves lesioned during RF procedures innervate other pain-generating structures besides the SI joint, and the SI joint is likely innervated by other nerves inaccessible for denervation (Table 5).
|
Surgical and Other Invasive Interventions
In 1999, Srejic et al. (95) reported 12–16 mo of significant pain relief in 4 patients with SI joint pain who received a series of 3 intraarticular injections with hyaluronic acid. Three of these patients had postsurgical SI joint pain and one suffered from severe osteoarthritis of the spine. The rationale for this treatment stems from studies demonstrating long-term pain relief with hyaluronic acid injections in degenerative joint disease of the knee (96). However, one meta-analysis found only scant evidence for the use of viscosupplementation in knee osteoarthritis, especially when lower molecular weight hyaluronate formulations are used (97). In addition to questionable efficacy, another factor that mitigates against intraarticular hyaluronic acid injections for SI pain is that degenerative joint disease accounts for only a small percentage of cases.
Proliferative therapy (a.k.a. "prolotherapy") has been advocated as a treatment for nonspecific LBP and SI joint pain (98). The rationale behind the use of "prolotherapy" is that the ligaments and other soft tissue structures are of primary importance in the development of LBP. Thus, the injection of a drug promoting fibroblast hyperplasia should theoretically increase the strength and reduce sensitization of these structures. In a double-blind study, Ongley et al. (99) found that LBP patients who received 6 wk of proliferant therapy had lower pain scores and disability indices at their 6-mo follow-up than "control" patients who received saline injections. Despite these findings, the lack of specific diagnoses, the numerous other treatment differences between groups, and the targeting of pain generators outside the SI joints limit the relevance of this study.
In patients with SI joint pain unresponsive to more conservative measures, several investigators have advocated surgical stabilization. Unfortunately, all published reports on SI joint fusion have been small case series or retrospective studies. Whereas the primary indications for SI joint fixation are either joint instability or fractures (100,101), successful arthrodesis has also been reported for degenerative joint disease (102). In some patients, successful stabilization can be done percutaneously using CT guidance (103). Regardless of the underlying etiology, based on the existing studies the long-term success rate for SI joint fusion appears to be in the range of 70% (100–102). Neuroaugmentation of the third sacral nerve root has also been reported to provide adequate pain relief in 2 patients with severe SI joint pain unresponsive to conventional therapy (104).
|
Conclusions |
|---|
|
TopAbstractAnatomyFunction and BiomechanicsPrevalenceMechanism of InjuryDiagnosis and PresentationTreatmentConclusionsReferences |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractAnatomyFunction and BiomechanicsPrevalenceMechanism of InjuryDiagnosis and PresentationTreatmentConclusionsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. S. Williams and S. P. Cohen Greater Trochanteric Pain Syndrome: A Review of Anatomy, Diagnosis and Treatment Anesth. Analg., May 1, 2009; 108(5): 1662 - 1670. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. P Cohen, C. E Argoff, and E. J Carragee Management of low back pain BMJ, December 22, 2008; 337(dec22_1): a2718 - a2718. [Full Text]
|
|
|
|
|
|
S. P. Cohen and R. W. Hurley The Ability of Diagnostic Spinal Injections to Predict Surgical Outcomes Anesth. Analg., December 1, 2007; 105(6): 1756 - 1775. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Buijs, L. Visser, and G. Groen Sciatica and the sacroiliac joint: a forgotten concept Br. J. Anaesth., November 1, 2007; 99(5): 713 - 716. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
