| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 The Russell H. Morgan Department of Radiology and Radiological Science, Johns
Hopkins University School of Medicine, 600 N Wolfe St., Baltimore, MD
21287.
2 Department of Diagnostic Radiology, Eberhard-Karls-University Tübingen,
Tübingen, Germany.
3 Department of Internal Medicine, Division of Rheumatology,
Eberhard-Karls-University Tübingen, Tübingen, Germany.
Received April 16, 2008;
accepted after revision June 28, 2008.
J. S. Lewin is a cofounder of and a shareholder in Interventional Imaging,
Inc.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionAPPENDIX 1: Sacroiliac STIR...References |
|---|
SUBJECTS AND METHODS. Over a 12-month period, 60 patients (32 women and 28 men; median age, 28 years; age range, 18–49 years) with chronic lower back pain suspected to originate from the sacroiliac joints were enrolled in the study. Based on diagnostic MRI findings, MR fluoroscopy–guided sacroiliac joint injections were performed in 57 (95%) patients. Diagnostic injections (35, 58.3%) were performed if nonspecific or degenerative MRI findings were present. Therapeutic injections (22, 36.7%) were performed in patients with inflammatory arthropathy. In three (5%) patients, no injections were performed. Technical effectiveness was assessed by analyzing, first, the rate of intraarticular injection; second, the time required for the procedure; third, image quality; and, fourth, occurrence of complications and clinical outcome by analyzing pain intensity changes and volume and signal intensity of sacroiliac inflammatory changes.
RESULTS. The rate of intraarticular injection was 90.4% (103/114).
The mean length of time for the procedure was 50 minutes (range, 34–103
minutes), with exponential shortening over time (p 
0.001). The
contrast-to-noise ratios of the needle and tissues were sufficiently different
for excellent delineation of the needle. No complications occurred. Diagnostic
injections identified the sacroiliac joints as generating significant pain in
46.9% (15/32) of the patients. Three months after therapeutic injections, pain
intensity had decreased by 62.5% (p 0.001) and the volume and
relative signal intensity of inflammatory changes had decreased by 37.5%
(p = 0.003) and 47.6% (p 0.001), respectively.
CONCLUSION. We accept the hypothesis that combined diagnostic and interventional MRI of the sacroiliac joints can be performed efficiently and effectively for comprehensive diagnosis and therapy of lower back pain originating from the sacroiliac joints.
Keywords: arthrography • diagnostic MRI • interventional MRI • lower back pain • MR guidance • sacroiliac joints
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionAPPENDIX 1: Sacroiliac STIR...References |
|---|
Among the available imaging techniques, MRI plays an important role in the diagnostic workup of sacroiliac disorders and can readily depict stress fracture, neoplasia, infection, and inflammatory arthropathy. However, MRI may not provide specific findings for the identification of sacroiliac joint pain related to degenerative joint disease, deformity, shear injuries, social or psychologic distress, pain related to prior trauma that may or may not involve the sacroiliac joints, or idiopathic sacroiliac joint pain [6]. In such cases, diagnostic sacroiliac joint injections are used to identity symptomatic sacroiliac joints [3, 7–9]. In addition, intraarticular steroid injection of the sacroiliac joint is a potent therapeutic option in patients with inflammatory arthropathy [3, 10–12].
Therefore, diagnostic or therapeutic sacroiliac joint injections may be required after diagnostic MRI. Procedures are typically performed under CT or fluoroscopic guidance. Because this procedure requires an additional clinical appointment for the patient and because CT and fluoroscopy are associated with exposure to ionizing radiation, combined diagnostic MRI and subsequent MR-guided injection of the sacroiliac joints is desirable. This approach became possible with the recent introduction of clinical openbore 1.5-T MR systems.
Therefore, to test the hypothesis that this combination can be performed efficiently and effectively, we prospectively assessed the procedure and clinical outcomes of combined diagnostic and interventional MRI of the sacroiliac joints in patients with chronic low back pain suspected to originate from the sacroiliac joints.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionAPPENDIX 1: Sacroiliac STIR...References |
|---|
Inclusion criteria were lower back pain clinically suspected to originate from the sacroiliac joints that had been present for at least 12 weeks despite conservative management (nonsteroidal antiinflammatory drugs, physical therapy, or both); the ability to provide voluntary, written informed consent to participate in this study; and no history of surgical procedures during the prior year.
Women who were pregnant or lactating were excluded. Additional exclusion criteria were heavy narcotic use; uncontrolled psychiatric disorders; inability to be positioned prone; spinal implants near the sacroiliac joints; and a history of adverse reaction to local anesthetics, steroids, or gadolinium-based contrast agents.
The study design is outlined in Figure 1. Clinical history, physical examination, and laboratory data were used to rule out specific causes of chronic lower back pain such as an underlying malignant, visceral, or metabolic disease; infection; or neurologic deficit requiring surgical intervention [13]. Patients with a clinical diagnosis of nonspecific lower back pain suspected to originate from the sacroiliac joints were referred for combined diagnostic and interventional MRI of the sacroiliac joints (Figs. 2A, 2B, 2C, 2D, and 2E).
|
|
|
|
|
|
For diagnostic MRI evaluation of the sacroiliac joints, a coronal oblique STIR MR sequence (TR/TE, 7,000/75; slice thickness, 4 mm; field of view [FOV], 22 cm; FOV phase, 100%; base resolution, 256; phase resolution, 70%; acquisition time, 6 minutes 32 seconds) and an axial T1-weighted turbo spin-echo MR sequence (400/17; slice thickness, 4 mm; FOV, 22 cm; FOV phase, 100%; base resolution, 320; phase resolution, 75%; acquisition time, 3 minutes 45 seconds) were used (Fig. 2B). MR images were evaluated immediately after the diagnostic phase at the workstation. MR-guided sacroiliac joint injections were performed according to the results of diagnostic MRI (Fig. 1).
Dorsolateral needle paths for direct access to the posteroinferior aspect of the synovial compartment of the sacroiliac joints were planned bilaterally on axial T1-weighted turbo spin-echo MR images (Fig. 2B) that were acquired during the diagnostic phase. Subsequently, the puncture sites were prepared and draped in a sterile fashion and the superficial tissue was anesthetized using 1–2 mL of lidocaine 1% (Fig. 2C).
A continuously acquired and displayed single-slice T1- and T2*-weighted FLASH 2D MR sequence (9.3/3.5; slice thickness, 5 mm; band width, 170 Hz) was used for near-real-time MRI guidance (MR fluoroscopy). The slice was prescribed to the slice position of the T1-weighted image of the diagnostic phase chosen for planning. Determination of both skin entry points was achieved using a syringe filled with gadolinium-enhanced saline moved over the skin during MR fluoroscopy with the FLASH 2D sequence (Fig. 2D). After antiseptic preparation of the skin, draping, and administration of local anesthetic, two small skin incisions were created and the puncture needle was navigated into the sacroiliac joint, again using the FLASH 2D MR sequence (Fig. 2D). The patient was then moved out of the bore and the injection was performed. Subsequently, the contralateral sacroiliac joint was punctured and injected in a similar manner. Figure S2F, a cine view available online at www.ajronline.org, shows a puncture.
For diagnostic sacroiliac joint injections, 1 mL of bupivacaine 1% and gadopentetate dimeglumine (mixed at a ratio of 200 parts bupivacaine to 1 part gadopentetate dimeglumine) was injected. For therapeutic sacroiliac joint injections, 1.2 mL of triamcinolone acetonide, ropivacaine, and gadopentetate dimeglumine (mixed at a ratio of 200 parts triamcinolone acetonide, 40 parts ropivacaine, and 1 part gadopentetate dimeglumine) was injected.
After bilateral sacroiliac joint injections had been performed, distribution of the injectant was assessed using a fat-saturated T1-weighted spin-echo MR sequence (480/17; slice thickness, 4 mm; FOV, 22 cm; FOV phase, 100%; base resolution, 320; phase resolution, 70%; acquisition time, 4 minutes 52 seconds) (Fig. 2E).
After diagnostic and therapeutic MRI, patients were monitored for 1 hour in the prone position and subsequently were discharged from the outpatient facility.
Assessment of Technical Effectiveness
The technical success of joint punctures and subsequent distribution of
injectant (intraarticular vs extraarticular) were qualitatively assessed on
postinterventional fat-saturated T1-weighted spin-echo MR images by two
readers in consensus. Minimal dorsal reflux of injectant without spread to
adjacent spinal structures in cases with sufficient intraarticular fluid
accumulation was categorized as intraarticular injection. Evaluation was
performed immediately after the procedure.
The total time for each procedure was divided into three phases. The diagnostic phase was defined as the period from the patient entering the MRI room to the end of diagnostic MR image evaluation. The interventional phase was defined as the period from the end of the diagnostic phase to withdrawal of the needle after injection and dressing of injection sites. The postinterventional phase consisted of performing the fat-saturated T1-weighted spin-echo MR sequence. Time points for each phase were documented during the procedure.
The image quality and ability to distinguish structures on FLASH MR fluoroscopic images were quantified by calculating contrast-to-noise ratios (CNRs) (Table 1).
|
Potential complications were documented during the procedure and were assessed during follow-up appointments.
Assessment of Clinical Outcome
Changes in pain intensity were assessed using an 11-point visual analog
scale (VAS) ranging from 0 (no pain) to 10 (worst pain). Data were obtained
before the procedure and 30 minutes after the procedure for diagnostic
injections and before the procedure and at 3-month follow-up for therapeutic
injections. In diagnostic injections, pain relief of 50% at 30 minutes after
the proce dure (positive test result) was regarded as sufficient to
characterize the sacroiliac joints as a significant source of the patient's
chronic lower back pain.
In patients with inflammatory arthropathy, therapeutic success was assessed by comparing the volume (volumehyper) and average signal intensity (SIRelHyper) of subchondral sacroiliac STIR hyperintensity on MR images acquired on the day of the intervention and at 3-month follow-up.
Quantitative Evaluation
Values are given as the median, with the minimum and maximum values in
parentheses, or as the arithmetic mean with first SDs
[14]. Measurements on MR
images were performed by a single person. The arithmetic mean of three
measurements was used. Intrarater variability was assessed using the
coefficient of variation (CV), which was calculated as follows:
 |
is the first SD and µ is the arithmetic mean
[15]. Box-and-whisker diagrams
were used to summarize values of VAS scores, volumes, and signal intensities,
whereas the boxes resemble the second and third quartile and the whisker-caps
show the 10th and 90th percentiles. A p value of 0.05 or less was
considered significant. All statistical analysis was performed using SPSS
software (version 11.5, SPSS). Time requirements were evaluated with regression statistics over the 12-month period. Curve fitting based on the Marquardt-Levenberg algorithm was used for approximation of the type of regression and calculation of regression coefficients and significance levels.
Signal intensity values were obtained using the measurement tool of the MRI
workstation (Syngo 5, Siemens Medical Solutions). CNR was defined as follows:
 |
 |
Significant differences between baseline and 3-month follow-up data of VAS scores were assessed using Wilcoxon's signed rank test [16].
Volumes and average signal intensities of the subchondral STIR
hyperintensity were measured (Matlab 6.5, MathWorks). First, the sacroiliac
joints were defined by manually outlining the cortex of both iliac bones and
the sacrum. Vascular and neural structures were excluded. For each set of STIR
MR images, the average signal intensity of the physiologic sacral bone marrow
(SIsacrum) was measured by placing an ROI (
200
pixels) at the center of the first sacral vertebra. The second positive SD of
SIsacrum was used as the threshold selection criterion for
a hyperintense pixel.
The average signal intensity of sacroiliac STIR hyperintensity (SIRelHyper) was calculated as shown in Appendix 1. The total volume of the subchondral sacroiliac STIR hyperintensity (volumehyper), given in cubic centimeters (cm3), was obtained by the addition of hyperintense voxels.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionAPPENDIX 1: Sacroiliac STIR...References |
|---|
Diagnostic MRI of the sacroiliac joints showed acute inflammatory arthropathy (22/60, 36.7%), nonspecific degenerative changes (21/60, 35%), stress fracture (2/60, 3.3%), and osteoid osteoma (1/60, 1.7%) and was normal in 14 of 60 (23.3%) patients.
Bilateral MR-guided sacroiliac joint injections were performed in 57 patients: 35 (61.4%) diagnostic injections in patients with normal MR findings and degenerative changes and 22 (38.6%) therapeutic injections in patients with inflammatory arthropathy. No complications occurred.
Technical Effectiveness
Puncture was successful in 108 of 114 (94.7%) sacroiliac joints, whereas in
the six (5.3%) remaining joints, the joint could not be accessed. In those
cases, 1 mL of bupivacaine 1% was injected into the posterior ligamentous
apparatus of the sacroiliac joints. In successfully punctured joints, 103 of
108 (95.4%) injections resulted in intraarticular accumulation of injectant.
Of the 11 of 114 (9.6%) sacroiliac joints with paraarticular injections, two
bilateral sacroiliac joints were of the inflammatory arthropathy group and
three bilateral and one single sacroiliac joint were of the group that showed
nonspecific degenerative changes.
SNRs were calculated as SNRfat = 174.4 ± 69.9 (CV = 4.1%), SNRneedle = 27.2 ± 10.9 (CV = 12.0%), SNRmuscle = 105 ± 155.3 (CV = 7.8%), and SNRbone = 31.6 ± 12.9 (CV = 15.5%). All SNRs were statistically significantly different (Table 1).
The mean time needed for the diagnostic phase was 22.5 minutes (range,
20.0–30.0 minutes). The mean time for the interventional phase was 22.5
minutes (5.0–67.5 minutes). The mean time for the postinterventional
phase was 6.0 minutes (5.0–7.5 minutes). The mean total time of the
procedure was 50 minutes (34–103 minutes). Regression analysis
identified a significant overall exponential decay of the total time of the
procedure (p 0.001), indicating significant shortening over time
(Figs. 3A,
3B,
3C, and
3D).
|
|
|
|
0.001) from 8 (range of VAS scores, 6–10)
before the intervention to 3 (0–4) 30 minutes after the intervention.
Ten (66.7%) patients had normal MR findings and five (33.3%) patients had
nonspecific degenerative changes. Patients with paraarticular diagnostic
injections (n = 3) were not included in this evaluation. The patient
who received one intraarticular and one paraarticular injection and had a
negative test result was included.
|
0.001). Median change of volume of subchondral sacroiliac STIR
hyperintensity (volumehyper) was –5.4 cm3
(range, –17.9 to 5 cm3; –37.5%) from 11.3
cm3 (2.9–23.1 cm3; CV = 7.0%) at baseline to 7.6
cm3 (0–17.9 cm3; CV = 5.4%) at 3 months' follow-up
(p = 0.003). Median change of signal intensity of sacroiliac STIR
hyperintensity (SIRelHyper) was –0.90 (range,
–2.11 to 0.55; –47.6%) from 1.91 (1.20–2.53; CV = 1.34%) at
baseline to 0.98 (–0.10 to 2.67; CV = 1.1%) at 3 months' follow-up
(p 0.001). The two patients with paraarticular injections of an
anesthetic were not included in this follow-up evaluation.
|
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionAPPENDIX 1: Sacroiliac STIR...References |
|---|
Several imaging techniques can be used for evaluation of the sacroiliac joints [23]. MRI is frequently used because it detects stress fracture, neoplasia, infection, and inflammatory arthropathy involving the sacroiliac joints with high sensitivity [24]. However, identification of symptomatic sacroiliac joints based on morphologic evaluation may not be possible in conditions such as degenerative joint disease, deformity, shear injuries, social or psychologic distress, pain related to prior trauma that may or may not involve the sacroiliac joints, and idiopathic sacroiliac joint pain [1, 4, 5, 22].
In the presence of equivocal MRI findings, diagnostic intraarticular injections can be applied to functionally test the hypothesis that the sacroiliac joints are symptomatic [3, 7–9]. Diagnostic intraarticular sacroiliac joint injections are considered the most sensitive test for the identification of symptomatic sacroiliac joints [3]. In our study, diagnostic injections identified symptomatic sacroiliac joints in 46.9% of patients with equivocal or nonspecific MRI findings, whereas in 53.1%, the sacroiliac joints were shown not to be involved in pain generation.
Therapeutic sacroiliac joint injections have been shown to decrease lower
back pain and decrease inflammatory activity in patients with inflammatory
arthropathy of the sacroiliac joints
[10–12].
In accordance with those findings, our results show a significant overall
decrease in both parameters at 3 months' follow-up (Figs.
6A,
6B,
6C, and
6D). Therapeutic sacroiliac
joint injections are a valid therapeutic option because widely available
systemic long-term treatment is rarely effective for the sacroiliac joints
[25] and recently introduced
more potent systemic drugs, such as tumor necrosis factor-
-blocking
agents, carry the risk of opportunistic infections and possibly of malignancy
[26]. MR-guided local drug
delivery to the sacroiliac joints has a very low risk profile
[27–29];
likewise, no complications occurred in our patient population.
|
|
|
|
Intraarticular drug delivery is imperative in diagnostic sacroiliac joint injections to ensure exclusive testing of the targeted sacroiliac joints by preventing injection into different compartments (i.e., false-negative test results) or spread to adjacent structures (i.e., false-positive test results). Therapeutic injections may also be performed intraarticularly to increase drug concentration at the site of inflammation and to avoid fat necrosis, which can occur in paraarticular steroid injections.
Because of the complex 3D anatomy of the sacroiliac joints and their close relationship to critical structures, the achievement of an acceptable rate of intraarticular sacroiliac joint injections requires imaging guidance [30]. "Blind" injections have been shown to result in intraarticular injection in only 22% of injections, with inadvertent epidural injections in 24% of injections [31]. Several techniques have been described for x-ray fluoroscopy, CT, and MRI guidance using open low-field MR systems [6, 28, 29, 32, 33].
In this study, MR guidance resulted in intraarticular injection in approximately 90% of the cases. This rate may be minimally lower compared with x-ray fluoroscopy and CT if performed by an experienced operator because small osteophytes of the posterior sacroiliac joints that may impede entering the joint space are less well visualized on MRI than on x-ray fluoroscopy and CT.
Real-time needle guidance is helpful to perform sacroiliac joint injections efficiently. Freehand MR-guided sacroiliac joint injections using high-field MR systems became available through the recent introduction of newly designed clinical wide-bore MR systems. With a distance of approximately 60 cm from the outer edge to the isocenter of the magnet, this design allows sufficient access to the sacroiliac joints, similar to that provided for CT-guided interventions. The use of an MR fluoroscopy mode allows real-time guided puncture similar to CT fluoroscopy and x-ray fluoroscopy [7, 34].
MR fluoroscopic needle placement contributed to an average time for the interventional phase of 22.5 minutes. Different concepts for fast MRI for interventional applications exist [35]. In general, gradient-echo MR sequences are well suited to achieve an MR fluoroscopic MRI mode because of the use of low flip angles and lack of 180° refocusing pulse, as opposed to spin-echo sequences [36]. The FLASH 2D sequence used in this study was characterized by a temporal resolution of approximately 1 frame per second and produced reliable and well-demarcated needle visualization as well as high and sufficiently different SNR from the SNRs of adjacent structures for good anatomic delineation.
In this pilot study, we used a high-resolution T1-weighted MR sequence with spectral fat saturation to detect the distribution of injectant with highest sensitivity to assess the rate of MR-guided intraarticular injections. In further studies, this more time-consuming MR sequence (4 minutes 52 seconds) may be replaced by a faster T1-weighted gradient-echo sequence with fat saturation to further decrease the total time needed for the procedure.
MR fluoroscopic guidance in this study was characterized by a remarkably constant and reliable needle artifact, thereby allowing excellent needle visualization during the interventions. The average overestimation of the true needle length on MR images is 1–2 mm for gradient-echo sequences [37], which provides sufficient accuracy for sacroiliac joint punctures.
Combined diagnostic and interventional MRI of the sacroiliac joints was found to be beneficial for several reasons: First, it allowed comprehensive evaluation of the sacroiliac joints by morphologic and functional assessments or therapy in one setting. This strategy obviates a second appointment and may result in fewer lost workdays for patients. Second, MRI is not associated with exposure to ionizing radiation as opposed to CT and x-ray fluoroscopy. Exposure to ionizing radiation at the level of the sacroiliac joints is especially concerning because of the vicinity of the sacroiliac joints to the reproductive tract. In addition, with MR guidance, exposure of the operator to ionizing radiation can be avoided as compared with x-ray fluoroscopy and CT fluoroscopy guidance [34]. MRI guidance can also be beneficial in performing serial injections to eliminate the risk of repetitive exposure to ionizing radiation and seems to be ideal for the treatment of young patients. Inflammatory arthropathy of the sacroiliac joints typically presents in patients between the ages of 20 and 30 years [38]. Third, monitoring the distribution of injectant is a common practice to increase the validity of clinical response [6, 32, 39]. Interventional MRI can be used in patients with a hypersensitivity to iodine, in whom iodine-based contrast agents such as those used in CT and x-ray fluoroscopy are contraindicated.
We did not include a cost analysis because such an analysis is not universally applicable and cost and availability of MRI examinations differ among institutions and may vary greatly among different health care systems. However, after a learning phase, the time required for the interventional phase gradually approached that of the diagnostic phase allowing individual cost estimation.
Currently the availability of interventional MR systems may be limited. However, we speculate that with the increasing acceptance of clinical open-configuration high-field MR systems, the availability of suitable MR systems will increase in the near future.
In conclusion, on the basis of the results of this investigation, we accept the hypothesis that combined diagnostic and interventional MRI of the sacroiliac joints can be performed efficiently and effectively for comprehensive diagnosis and therapy of patients with chronic lower back pain originating from the sacroiliac joints.
APPENDIX 1: Sacroiliac STIR Hyperintensity
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionAPPENDIX 1: Sacroiliac STIR...References |
|---|
 |
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionAPPENDIX 1: Sacroiliac STIR...References |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Fritz, C. Thomas, N. Tzaribachev, M. S. Horger, C. D. Claussen, J. S. Lewin, and P. L. Pereira MRI-Guided Injection Procedures of the Temporomandibular Joints in Children and Adults: Technique, Accuracy, and Safety Am. J. Roentgenol., October 1, 2009; 193(4): 1148 - 1154. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Fritz, C. Thomas, S. Clasen, C. D. Claussen, J. S. Lewin, and P. L. Pereira Freehand Real-Time MRI-Guided Lumbar Spinal Injection Procedures at 1.5 T: Feasibility, Accuracy, and Safety Am. J. Roentgenol., April 1, 2009; 192(4): W161 - W167. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
