|Year : 2018 | Volume
| Issue : 2 | Page : 80-97
Sacroiliac joint denervation, the joint and ligaments: aretrospective study and algorithm
Mohammad-Hazem Ahmad-Sabry1, Gholamreza Shareghi2
1 Department of Anesthesiology, Alexandria University, Alexandria, Egypt
2 Advanced Pain Therapeutics of Knoxville, Knoxville, Tennessee, USA
|Date of Submission||13-Sep-2017|
|Date of Acceptance||19-Nov-2017|
|Date of Web Publication||28-Jun-2018|
Department of Anesthesiology, Alexandria University, Alexandria, 21111
Source of Support: None, Conflict of Interest: None
Introduction We included patients who had sacroiliac joint complex pain (SIJCP) and failing all conservative approaches to alleviate SIJCP, all had SI joint procedure with fluoroscopic guidance. Some had more than 3 months of better than 50% with SI steroid injection (62/144 patients), others became candidates for sacroiliac joint complex sensory radiofrequency (RF) denervation for long-term pain relief (24 patients), with facet denervation (22 patients) or with piriformis injection (19 patients).
Materials and methods In all, 170 patient charts were reviewed, and 17 were excluded for unclear block results and 26 for discogenic pathology. Fluoroscopy and 10 or 15-mm active tip RF cannula was used. An imaginary stripline was drawn about 3–5 cm medial and parallel to the joint line. Posterior primary sensory denervation was then carried out through 10–15 mm segments in series. Denervation was repeated 9–12 months later if greater than 50% of original pain had returned (first repeat and second repeat). The denervation process was then expanded to other sets of patients having (II) SIJCP plus ipsilateral lumbar facet pain and (III) SIJCP plus ipsilateral piriformis syndrome exclusively.
Results All groups showed an increase in activities of daily living (through Oswestry Scores), better pain score, and at least 45 weeks improvement of pain score after RF denervation.
Conclusion SIJCP can be effectively (50% or better) reduced for 4.5 months with steroid injection, or 10–12 months with denervation. The procedures were repeated once or twice with the same effectiveness. The method of pain management through our approach is effective, safe, simple, and reproducible.
Keywords: back pain, denervation, radiofrequency, sacroiliac joint
|How to cite this article:|
Ahmad-Sabry MH, Shareghi G. Sacroiliac joint denervation, the joint and ligaments: aretrospective study and algorithm. Res Opin Anesth Intensive Care 2018;5:80-97
|How to cite this URL:|
Ahmad-Sabry MH, Shareghi G. Sacroiliac joint denervation, the joint and ligaments: aretrospective study and algorithm. Res Opin Anesth Intensive Care [serial online] 2018 [cited 2018 Jul 19];5:80-97. Available from: http://www.roaic.eg.net/text.asp?2018/5/2/80/235485
| Introduction|| |
Sacroiliac (SI) joint is a large diarthrodial joint in which the anterior one-third (between sacrum and ilium) is a true synovial joint. The posterior two-thirds are full of ligamentous connections . The main function of this joint is stability, with little rotation about the three axes. Cadaver studies show that the total range of motion during flexion or extension rarely exceeds 2° ,.
Most of the problems of pain attributed to the SI joint have been greatly discussed in orthopedic, physical medicine, and pain medicine literature ,,,,. They arise from (a) the anatomical location of the joint, the load it must bear, and (b) the limits of mobility of this joint, which is held in place by ligaments both anteriorly and posteriorly that cover the caudal two-thirds of the joint.
There may still be a debate as to the exact number and nature of all the sensory nerves to the SI joint, but a fair estimate would be that the joint line and the ligaments are mostly innervated from sensory branches of L4 through S4 ,,,,,.
Anatomical dissections in humans have revealed that the innervations to the SI joint complex are provided exclusively by spinal nerves from dorsal rami of S1–S4 .
Similar to other major joints in the human body, there is reason to suspect that the ligaments surrounding this joint may have their own pain innervation(s). Suspicion that more than the SI joint itself is involved with pain production from this region arises from what physiologically makes sense, as well as a number of comments made in the literature ,,,.
Diagnostic SI joint area pain is started on the basis of a good history, a full clinical examination with a strong suspicion of the involvement of the SI joint as a pain progenitor ,,,, followed by diagnostic SI area joint injections ,.
The premise of the injection is that as stressing the joint by examinations produces pain, and pain is generally present in the SI area, then injection of a local anesthetic to the pain-producing region should relieve the pain ,,,.
We postulated that SI area pain generation was from more than the nerves that go inside the joint. Our hypothesis was that (a) all the nociceptive sensory pain nerves that supply the posterior ligaments of the SI joint and the SI joint itself came off the spinal cord posteriorly, and run in the posterior half of the body to innervate their designated targets. (b) If we could interrupt all of these sensory painful signals, in a safe location, then the pain from this complex should be much reduced as long as these nerves’ interruption continues.
| Materials and methods|| |
- We obtained informed consent and followed all guidelines of the Declaration of Helsinki (https://www.wma.net/policy/archived-policies/).
- After obtaining proper consents, we proceeded with the procedures. At each step of each procedure if any patient wanted to opt out of the protocol, or was not 100% satisfied with the protocol, that patient was given, other pain management options to the best of our abilities to his/her satisfaction.
- All patients had pain complaints related to the SI joint complex for at least 6 months. Most had undergone several diagnostic tests by their referring physicians (e.g. computer tomography scan, or MRI of lumbar spine; lower back and/or hip radiographies; electromyograph of lower extremities; laboratory studies regarding infection, rheumatoid arthritis, and so on). Many had done physical therapy, and were pharmacologically being treated with a possible combination of NSAIDs, muscle relaxants, membrane stabilizers, and opioids.
- A detailed full history and clinical examination (including full neurological and musculoskeletal exam) was carried out on each patient upon admission and changes noted after each different diagnostic blocks.
After patients had been ruled out for facet/piriformis/discogenic/nerve root pain problems, they underwent the first SI area injection. Those who originally had greater than 50% relief but after 4–6 weeks needed a second injection [i.e. if >50% of their pain had returned, 10 ml total (5 ml of local anesthetic+4 ml of Omnipaque-240+40 mg of depomedrol)] and those in whom the first injection was good enough (i.e. >50% of their pain had stayed gone) were evaluated after 3 months from the first or second (their last injection). Therefore, these patients either had 40 mg of steroid in one injection, or 80 mg of steroid in two injections by 3 months from their last injection. If satisfied with the pain relief (at least >50%), they were excluded from denervation study. They were offered a ‘booster’ shot of 10 ml of local anesthetic and 40 mg of depomedrol in the SI area, every 3–12 months, as their relief approached 50% of original.
A total of 17 patients out of 144 ([Table 1]) actually began the study, but during the 10 ml injection, part of the local anesthetic/dye/steroid went inferoanterior to the most caudal point of the SI joint, bathed the sciatic nerve, and caused transient conduction block. These patients were excluded from our study.
For denervation, we chose a line medial and parallel to the SI joint, about 3–5 cm away. Along this line, we hoped to interrupt all nociceptor pain progenitors (efferents) coming from the SI joint and its posterior ligaments. Full denervation along the above strip denervation would render the patient free of pain if all the pain sensory efferents were posteriorly located, and if they were silenced during the denervation the pain relief would continue for the duration of the nerves’ signal interruption. The stripline denervation technique used almost exclusively (>90%) a G18, 10 mm active tip unipolar radiofrequency (RF) electrode, with multiple overlapping ‘burns’ in series, which hopefully not only denervates along the bare metal but up to 2–4 mm surrounding it. The bent oblique superficial-to-deep orientation of active tip in the posterior ligaments adjacent to sacral bone presumably denervated any posterior nociceptive nerves intended for the ligaments, or the SI joint, crossing the stripline.
- Our above hypothesis was tested on a group of our patients (designated group A) who had to fit a certain strict exclusion criteria, described in section MM II E.
- Patients with lower back area and/or buttock pain radiating to the hip or groin, and/or with radicular pain complex, usually have pain originating from (a) SI area, (b) lumbar facet area, (c) lumbar discogenic pain (with or without radiculopathy), and (d) piriformis muscle syndrome. They rarely may have other diagnoses listed in [Table 2] that must also be detected and ruled out. Patients commonly either have one of the above four more common diagnoses or a combination of two or more. Patients who were suspected of having diagnoses number (b), (c), or (d), or a combination thereof, first underwent standard diagnostic blocks used in the interventional pain practice (for (b) see ref  and for (d) see group C) to diagnose the problems(b) or (d). If (c) was then strongly suspected, those patients underwent discography.
|Table 2 Differential diagnostic list of lower back/hip pain with/or without radiculopathy|
Click here to view
- Those responding positively to these blocks were excluded from group A study. Here we were interested in patients with clinical presentation of only SI joint area pain. If patients had lumbar facet pain, or piriformis syndrome pain from the above studies, we took note and later evaluated whether they would belong to group B or group C, respectively. If there was any suspicion of discogenic pain with/without radiculopathy, they were removed from our study altogether.
- Patients with suspected SI area pain as their only diagnosis underwent our standard diagnostic criteria for sacroiliac joint complex pain (SIJCP), and if the diagnostic SI area blocks were successful, they underwent denervation ([Table 1]).
- If a patient had a combination of SIJCP and ipsilateral lumbar facet pain, these patients first underwent diagnostic facet block of the medial branch nerves of the suspected facets . Within 20 min after completion of this block, the patient was ambulated and asked to flex and extend their torso. We did the standing, flexion of torso, and rotation test, and we pressed hard over the blocked facet joints. If the patient had little or no pain now to the above maneuvers [i.e. the facets were now minimally painful (≥75% of their pain gone) or not at all painful], but provocative tests of SI joint were still very painful ,, then we suspected possible SI area pain involvement. These patients were then placed prone on an radiographic table and a diagnostic SI joint complex area block with contrast and local anesthetic was carried out. If the SI injection took greater than or equal to 80% of their remaining pain away in a few minutes, after full clinical examination, we strongly suspected SI area pain involvement. These patients then underwent the standard two sets of diagnostic/confirmatory facet medial branch blocks, and if properly qualified it was followed by RF medial branch denervation of these nerves.
Three to four weeks after facet denervation, these patients were re-examined for facet joint pain and SI joint pain. If the facet pain areas were now reasonably not painful (≥75% of their facet-related pain was gone in comparison to before), but the SI joint pain areas were (primary and referred areas ), then these patients underwent SI joint area diagnostic protocol (MM II F).
When these patients became candidates for SI area denervation, they had a successful denervation (originally most patients had >70–80% decrease of their SI area pain) that was confirmed during the initial postprocedure physical examination follow-up. What if within a few (e.g. 5–6) months 50% of a pain came back? Was it facet related? Or SI related? From group A results, we had learned that most patients still have greater than or equal to 50% pain relief at about 9 months. From literature, we were aware that 5–6 months after facet medial branch denervation, it is likely that more of these patients will have greater than or equal to 50% of their facet pain returned. After a full clinical examination of the patient was done and compared with their previous exams before/after each diagnostic block, we often had our answer. Earlier pain returns (within 5–6 months) were usually facet pain in nature. If there was any doubt about this pain, then a diagnostic facet medial branch block or a diagnostic SI joint area block (without steroids) removed the doubt.
If a patient was suspected of a combination of piriformis pain syndrome and SIJCP on the same side, diagnosis of piriformis syndrome starts with a clinical suspicion. Owing to the nature of this syndrome presenting often with bizarre symptoms , history is not very specific in helping the diagnosis. Generally during clinical examination with the patient prone on an exam table, if a straight line is drawn between ipsilateral tip of greater trochanter of the hip and the posterior superior iliac spine, applying sustained thumb pressure at midpoint of this line compresses the piriformis muscle over sciatic nerve onto the bony sciatic notch. Within a few seconds to minutes, patients with this syndrome will usually start complaining of pain there, and then it may radiate elsewhere (see bizarre symptoms above). To substantiate the diagnosis, a piriformis muscle block may be indicated (below):
These patients first underwent diagnostic piriformis muscle injection as follows:
The patient was placed prone on the fluoroscopy table. Skin projection of the most caudal point of the SI joint on the PA radiographic view was anesthetized with local anesthetic. Then, a G22 Spinocan spinal needle (B. Braun Medical Inc., Bethlehem, PA, USA) was advanced under fluoroscopy toward a point 1.5–2.0 cm caudad and 1.5–2.0 cm medial to the caudal-most point of the SI joint. The tip of the needle is then contacting the sacral bone. The tip was then slowly walked off laterally until the tip is kissing the edge of the sacral bone going deeper.
On cross-table lateral radiographic view, the tip was then advanced 4–5 mm anterior to the ventral surface of the sacral bone [tip is invariably in the original portion or piriformis muscle, and because of only 4–5 mm ventral depth even if aberrant anatomy (divided sciatic nerve in the piriformis muscle, or divide piriformis muscle with the sciatic nerve intersecting it), the needle tip is not ventral or lateral enough to impale the sciatic nerve (for a description of the anatomy see Benzon et al. .
The injection administered into the piriformis muscle comprised 10 ml of 0.25% bupivacaine, 10 ml of Omnipaque-240, and 10 mg of depomedrol.
The patient then got off the procedure table. Thirty minutes later, a full examination was performed. If the piriformis-related symptoms were greatly reduced (≥75%), they were noted in detail; we then preceded to our standard selection criteria for SI joint complex.
When these patients became candidates for SI joint complex denervation, they had a successful denervation (originally most patients had 70–80% decrease in their SI area pain). This was confirmed during the postprocedure examination and interview. What if within a few weeks to a few months the pain came back? Was it piriformis related? Or SI related? From group A results and literature, we focused on piriformis muscle blocks lasting from a few days to a few weeks, and the SI complex denervation lasting 9–10 months. Then, a full clinical examination of the patient was done and compared with their previous examinations done before and after each diagnostic/therapeutic block. Most often we had our answer. Pain returning after a few days to weeks was normally piriformis muscle tensing up again. If there was any doubt, a diagnostic SI joint complex block (without steroid) or a piriformis muscle block (without steroid) removed the doubt.
MM I: injection technique
The injection technique was similar to that of Dussatt et al. , except that we used the lower one-third to one-fourth of the joint for initial entry and from a perpendicular fluoroscopy we then moved the beam about 5° cephalocaudad and about 5° oblique ipsilaterally to see the caudal SI joint. In this view, the lower fourth of the joint (both the anterior and posterior portions) usually looks like two sets of parallel lines.
A gauge 22 Spinocan spinal needle was then advanced toward the posterior entry to the joint. When maneuvering the tip through the ligaments and capsule of the joint, there is usually a pop. Some small angling up/down/left/right was sometimes necessary. Intra-articular position was confirmed by injection of 0.5–1.5 ml of Isovue M-200 (or Omnipaque-240) through the needle. In 94/122 (77%) nondenervation patients (group D), and in 37/52 (71%) patients of our (group A) denervation cases, the needle was within the joint. As we wanted to cover all the pain generators, from the joint and the posterior surrounding ligaments, we injected a total of 10 ml of a mixture containing 5 ml of Isovue M-200 (or Omnipaque-240), 4 ml of 0.25% bupivacaine, and 1 ml=40 mg of depomedrol. In patients in whom depomedrol was not indicated, we added 1.0 ml of normal saline instead. If resistance was encountered during the injection, the needle was withdrawn 2–4 mm posteriorly and injection was attempted again, under fluoroscopic view. The contrast within the joint and extravasated from the joint could be seen on the anteroposterior view, along the tissues and ligaments, mostly posterior to the joint. We then waited a few minutes, asked the patient for quantification of pain reduction, and performed the same physical examination of this area done before the injection.
MM II: assessments and statistics
MM II A: outcome assessment of pain:
- 100-point visual analog pain scale (VAS).
- Subjective improvement in activities of daily living (ADL) secondary to SIJCP reduction [the revised Oswestry disability index (for lower back dysfunction, in the revised form the sex question has been changed to changes in degree of pain)]. This questionnaire was given before denervation and 3 and 9 months after denervation (see [Table 3] for results).
|Table 3 Revised Oswestry disability index scores for evaluation of activities of daily living|
Click here to view
- Objective reduction in the amount of opioid used on a daily basis for pain reduction 2 weeks before denervation versus 2–3 months after denervation. To calculate this dose of all opioids taken per 24 h orally or transdermally or transmucosally, they were converted to an equipotent equivalent dose of oral morphine sulfate first, and compared with each other.
- A simple question of (YES/NO) if the patient is satisfied with the overall pain relief obtained as a result of our procedure(s), asked at specific time intervals during the study ([Table 4]).
MM II B: postprocedure assessment
- Telephone interview within 1 week and 2 weeks, with specific questions regarding outcome assessment of pain (above), and then once-a-month during the study.
- Formal face-to-face interview 2–3 weeks after the first block and before the second block. Specific questions regarding estimation of duration of pain relief after each block.
- The once-a-month pain assessment continued until almost all of (80–100%) the original pain had returned. If patients could not be reached to be assessed after 2 months, they were removed from this study.
MM II C: time to recurrence assessment
- This is defined as return of 50% or more of the patient’s original SIJCP.
MM II D: statistics
- χ2-test, t-test, and analysis of variance when applicable.
MM II E: exclusion criteria from SI joint complex diagnostic block
- Lumbar facet-related pain.
- Lumbar discogenic pain (with and/or without radiculopathy).
- Piriformis pain syndrome.
- Clear evidence of other diagnoses besides 1, 2, and 3 above that are listed in [Table 2].
- Active (litigation or workman’s compensation claim) cases.
MM II F: selection of patients for SI complex sensory denervation
- A full history and clinical examination is obtained from each patient.
- Each patient received two valid SI joint complex injections. Each injection contained 40 mg of depomedrol (1 ml), 4 ml of 0.25% bupivacaine, and 5 ml of Isovue M-200 (or Omnipaque-240) mixture and was performed under fluoroscopy. We allowed 2–4 weeks between subsequent injections.
- If patients reported greater than 80% relief for a few hours to a few weeks, and greater than 50% overall pain relief lasting greater than 3 months, and if they were satisfied with the pain relief from steroids, we then gave them intermittent booster SI area joint injections of local anesthetic and 40 mg of depomedrol every 3–12 months, as needed. These patients were excluded from denervation (group D).
- If patients reported greater than 80% relief for a few hours only, with return of their original pain in less than a few days, and repeated second injection to SI joint complex yielded a similar short term result, then these patients became candidates for SI joint complex sensory denervation.
MM II G: SI complex sensory pain generator denervation with RF
- The patient was in prone position, and upper buttock area was prepped with alcohol and draped.
- Fluoroscopy in posteroanterior (PA) to 5–10° contralateral oblique view to see the SI joint en-face, trying to look at the whole length of the joint cephalocaudad, as well as the ipsilateral sacral nerve root origins (all in the same fluoroscopy image).
- A linestrip denervation is drawn in the physician’s mind about 2.5–5.0 cm medial and parallel to the joint ([Figure 1],[Figure 2],[Figure 3]).
|Figure 2 Posterior sensory nerves sacroiliac joint and posterior ligaments complex.|
Click here to view
- Skin projection over the middle point on the SI joint is anesthetized; a G20 or G18 RF cannula with 10 or 15 mm bent active tip is advanced to the most caudad part of the strip denervation. A measure of 40–50 ml of 0.15% bupivacaine and 1 mg/ml depomedrol is injected along the length of the linestrip. Then RF denervation is carried out at 80°C for 90 s ([Figure 4]).
- We made an attempt to denervate medial and deep (to purposefully encounter long and short posterior SI ligaments) in our segmental serial 10–15 mm denervations.
- Denervation is repeated after repositioning the RF needle tip 10 or 15 mm more cephalad along the strip denervation, and so on. It usually takes at least 10–12 ‘burns’ of the 10 or 15 mm active tip cannula for an effective strip denervation ([Figure 4]).
Creating a strip of denervation lesion, parallel and at a safe distance away from the SI joint, and away from the nerve roots, encompasses both the pain generators in this joint and outside (posterior) to the joint, which we postulate may exist there; see [Figure 1], [Figure 3], and [Figure 4]. In this method, the L5 dorsal ramus is automatically denervated along the strip denervation, as well as all sensory contributions from presumably L4, L5, S1, S2, S3, and S4.
The problem of pain in the lower back, with or without radiculopathy, or with hip/buttock-related pain, falls into a broad category ([Table 2]). By the time the patient is seen in the pain clinic, it is hoped that the list of differential diagnoses in [Table 2] have already been narrowed to half a dozen or less!
As seen in [Table 5] of the total number of our patients, 65 underwent 227 primary and repeat (twice) RF SI joint complex area denervation. Both RF primary first and second denervations yielded statistically similar outcomes of duration and percentage of pain relief, with indistinguishable satisfaction surveys obtained during the period of pain relief. The ADL scores show statistically significant and maintainable improvement after primary, first, and second repeat denervations in groups A, B, and C.
We are therefore advocating the use of the strip denervation method for the SI area pain: that is, to perform denervation 2.5–5.0 cm parallel and medial to the SI joint, by the physician’s comfort level with RF (e.g. group A, B, or RF denervation).
There are other published articles in the literature with valiant methods of RF destruction of SI joint area pain generators. Some of these studies are listed as follows.
Yin et al.  performed a sensory stereotaxy-guided approach to SI joint area complex. Their functional mapping revealed 100% symptomatic nerves from L5 and S1, 90% from S2, and 40% from S3. A successful neurotomy response was defined as patient perceived improvement greater than 60% with a concomitant greater than 50% decrease in VAS 6 months after RF neurotomy process. Four of their patients had previously undergone one/more laminectomy/fusion. None had undergone provocative discography. In the discussion section, these physicians point to the fact that ‘… further study of the deep interosseous (IO) SIJ ligament injection in diagnosing not only intra-articular, but ligamentous pain, is suggested. The lack of a ‘gold standard’ for the diagnosis of SIJC ligamentous pain confounds this diagnostic dilemma.’
Kim et al.  presented selective neurotomy of sacral lateral branches for pain of SI joint dysfunction. They excluded patients with objective evidence of lumbosacral radiculopathy (through physical exam). Then performed at least two diagnostic blocks with dye plus saline, dye plus 1 ml of 0.5% bupivacaine, and dye plus 1 ml of 4% lidocaine. A better than 50% decrease in pain for the duration of local anesthetic was considered positive. RF neurotomies (unipolar) were performed after localizing the ‘pathological branches’ and adopting a sensory stimulation method. The number of RF lesioned nerve branches were six patients (L5, upper S2, middle S1, lower S1, lower S2, upper S1). At 6 months, five of 16 original patients (31%) reported 100% pain relief, and another five reported moderate relief.
Ferrante et al.  reported RF denervation through bipolar electrode lesioning of 50 consecutive SI joint denervations in 33 patients who had failed conservative management, and had undergone diagnostic SI local anesthetic blocks before denervation. They evaluated VAS, physical exam, SI joint provocation tests, lumbar spine range of motion, and opioid use before and after denervation. Successful denervation was at least greater than or equal to 50% decrease in VAS for 6 months. In all, 12 of 23 (36.4%) met this criteria. The average duration of pain relief was 12.0±1.2 months in responders versus 0.2±0.7 in nonresponders (P≤0.0001)
In the study of Burnham et al. , nine patients with SIJ pain confirmed by local anesthetic block underwent bipolar RF denervation through strip lesions made adjacent to the lateral edge of S1–S3 dorsal foraminal apertures. The L5 dorsal ramus received a conventional unipolar lesioning. There was significant improvement in pain intensity (8 to 4.1), a decrease in disability scores, lower back pain, leg pain, and opioid use. Satisfaction survey was very satisfied (78%: 1 month; 67%: 3 months; 67%: 6 months; 89%: 9 months; 67%: 12 months).
Cohen et al.  in a randomized placebo-controlled study of 14 patients, RF denervated L4 and L5 primary dorsal rami and cool-probe denervated ipsilateral S1, S2, and S3 after a local anesthetic block. One, 3, 6, and 12 months pain relief of treated patients was 79, 64, 57, and 14%, respectively, compared with 14% of placebo group at 1 month. Once again a full (stripline like) denervation (from L4 to S3) as depicted from their data provided intermediate-term (6 months >50%) pain relief of their patients. It is conceivable that as RF denervations of nerves and ligaments usually are longer-lasting than cool-probe denervations of the same, our denervation data and theirs would be very similar regarding duration of pain relief outcomes!
Dreyfuss et al.  used multi-site multi-depth blocks on IO and dorsal SI ligaments (DSL). After keeping IO and DSL insensate, they injected the inferior SI joint with dye (capsular distention). Their study showed different sources of pain coming from intra-articular portion and posterior ligaments, a conclusion in general agreement with our findings.
Our study not only tried to take into account all the nociceptive pain progenitors involved with SI joint complex, but follows this denervation concept successfully through two cycles of denervation-reinnervation if/when they occur. The outcome of our denervation method is statistically similar whether primary, first, or second set of denervations are performed, or whether denervations are done on a ‘pure’ SI joint complex problem, or on patients in whom other problems (lumbar facet pain, piriformis syndrome) are also involved. Please note that this study is not exclusively the proof of distinct sensory innervations to the posterior ligament attachments of SI joint . Our data are, however, very consistent with the presence of such innervations.
Therefore, the SI joint area complex as a progenitor of pain can be managed either with local anesthetic or anti-inflammatory steroid to the joint plus the posterior ligaments for a substantial (>50% pain reduction) of 19-week duration. If the steroids are not sustaining the pain relief, then the denervation procedure can successfully satisfy a substantial percentage of the patients (lowest 45% to highest 73%) even after 9 months postdenervation.
The longer time of 9-12 months to get greater than equal to 50% of the original pain returned presumably indicates the duration of time that the pain generator nerves are growing back toward their original targets.
In all, 29/227 (12.8%) of our procedures developed a transient cutaneous dysesthesia, or hypoesthesia, and/or itching along the surface of the ipsilateral buttock mirroring the SI joint’s skin projection. These abnormal sensations generally disappeared within 2-10 weeks after denervation in almost 93% of these patients. It is possible that some cutaneous nerve branches may have been denervated during the ‘deep denervation procedure’ resulting in the above transient symptoms!
Study patients with lumbar facet pain, or piriformis pain syndrome, as well as SI area joint pain, were able to be distinctly addressed regarding each pain problem. Similar concepts may be extrapolated to patients with lumbar discogenic pain and/or nerve root pain plus SI joint area pain. Patients may also have two or more of the above conditions plus SI joint area pain.
This review encountered a number of patients with more than one of the above three pain problems (i.e. piriformis syndrome and/or facet pain and/or discogenic pain) plus their SI joint area problem. By diligent application of facet medial branch denervation and/or piriformis block and/or lumbar disc-related therapies and SI joint area denervation, we have been mostly able to manage our patients (Flow Chart 1-4)
| Results|| |
A total of 514 new patients were referred to the pain clinic with low back pain (LBP); their chart was reviewed to be included in the study. The diagnosis of main pathology was facet in 185 (36%) patients, SI in 170 (33.1%) patients, piriformis in 84 (16.3%) patients, and discogenic in 75 (14.6%) patients ([Table 6]). The patients’ main diagnosis was SI pain at the beginning of the study, but in 17 patients during the 10 ml injection, part of the local anesthetic/dye/steroid went inferoanterior to the most caudal point of SI joint, bathed the sciatic nerve, and caused transient conduction block. These patients were excluded from our study, as well as 26 patients who showed radiculopathic symptoms and discogenic pain.
|Table 6 Mix of diagnoses of new patients referred to our pain clinic with chronic low back pain with or without radiculopathy|
Click here to view
A total of 65 patients had denervation techniques with a total of 227 procedures, and 62 patients had steroid injections ([Table 1]). Group A primary denervation includes 24 patients, 14 female and 10 male, in whom a total of 67 denervation procedures were done. In group A primary denervation, the duration of pain relief in weeks was 48.9±15.2 (P≤0.02) with percentage of pain relief 85.6±11.8 (P≤0.01) and had percentage decrease of opioid dose used by 53±22 (P≤0.05). Revised Oswestry disability index scores for evaluation of ADL was done within 2 months before denervation and 3 months and 9 months after denervation. In group A primary denervation, 18 out of 24 patients completed the revised Oswestry disability index; the score was 37.8±6.1 (P≤0.001) before denervation, 23.5±5.4 (P≤0.01) 3 months after denervation, and 26.0±4.7 (P≤0.01) 9 months after denervation. Satisfaction survey showed decreased rate by time from 91% at 0 month (1–2 weeks from procedure), 87% at 3 months, 79% at 6 months, 62% at 9 months, and 46% at 12 months.
Group A: first repeat denervation includes 15 patients with a total of 19 denervation procedures
In group A first repeat denervation, the duration of pain relief in weeks was 54.7±13.2 (P≤0.01), with a percentage of pain relief of 84.9±12.5 (P≤0.01). In all, 12 out of 15 patients completed the revised Oswestry disability index; the score was 21.1±6.2 (P≤0.02) 3 months after denervation and 24.9±6.6 (P≤0.05) 9 months after denervation. Satisfaction survey showed decreased rate by time from 93% at 0 month (1–2 weeks from procedure), 80% at 3 months, 80% at 6 months, 73% at 9 months, and 53% at 12 months.
Group A: second repeat denervation includes 11 patients with a total of 13 denervation procedures
In group A second repeat denervation, the duration of pain relief in weeks was 50.3±14.4 (P≤0.02), with a percentage of pain relief of 81.1±13.4 (P≤0.01). In all, nine out of 11 patients completed the revised Oswestry disability index; the score was 19.6±5.2 (P≤0.05) 3 months after denervation and 23.7±4.8 (P≤0.05) 9 months after denervation. Satisfaction survey showed decreased rate by time from 100% at 0 month (1–2 weeks from procedure), 90% at 3 months, 72% at 6 months, 45% at 9 months, and 36% at 12 months.
Group B: primary denervation includes 22 patients, 13 female and nine male, with a total of 94 denervation procedures
In group B primary denervation, the duration of pain relief in weeks was 43.4±16.3 (P≤0.01), with a percentage of pain relief of 86.8±9.5 (P≤0.05), and the percentage of opioid dose used decreased by 55±13 (P≤0.05). Revised Oswestry disability index scores for evaluation of ADL was performed within 2 months before denervation and 3 months and 9 months after denervation. In group B primary denervation, 17 from 22 patients completed the revised Oswestry disability index; the score was 53.6±7.5 (P≤0.005) before denervation, 35.9±7.1 (P≤0.005) 3 months after denervation, and 38.2±6.0 (P≤0.001) 9 months after denervation. Satisfaction survey showed decreased rate by time from 91% at 0 month (1–2 weeks from procedure), 87% at 3 months, 79% at 6 months, 62% at 9 months, and 46% at 12 months.
Group B: first repeat denervation includes 16 patients with a total of 23 denervation procedures
In group B first repeat denervation, the duration of pain relief in weeks was 45.7±11.7 (P≤0.01), with a percentage of pain relief of 81.7±12.9 (P≤0.01). In all, 14 out of 16 patients completed the revised Oswestry disability index; the score was 32.2±6.5 (P≤0.01) 3 months after denervation and 37.5±5.3 (P≤0.01) 9 months after denervation. Satisfaction survey showed decreased rate by time from 94% at 0 month (1–2 weeks from procedure), 94% at 3 months, 81% at 6 months, 69% at 9 months, and 50% at 12 months.
Group B: second repeat denervation includes 13 patients with a total of 17 denervation procedures
In group B second repeat denervation, the duration of pain relief in weeks was 49.1±14.0 (P≤0.05), with a percentage of pain relief of 86.5±11.6 (P≤0.02). In all, 10 out of 13 patients completed the revised Oswestry disability index; the score was 40.4±5.9 (P≤0.05) 3 months after denervation and 42.3±7.2 (P≤0.02) 9 months after denervation. Satisfaction survey showed decreased rate by time from 84% at 0 month (1–2 weeks from procedure), 77% at 3 months, 69% at 6 months, 54% at 9 months, and 46% at 12 months.
Group C: primary denervation includes 19 patients, 10 female and nine male, with a total of 66 denervation procedures
In group C primary denervation, the duration of pain relief in weeks was 47.6±11.6 (P≤0.02) with a percentage of pain relief of 77.4±10.2 (P≤0.01) and had percentage decrease of opioid dose used by 41±16 (P≤0.05). Revised Oswestry disability index scores for evaluation of ADL was done within 2 months before denervation and 3 months and 9 months after denervation. In group C primary denervation, 17 out of 19 patients completed the revised Oswestry disability index; the score was 46.1±5.8 (P≤0.005) before denervation, 27.4±6.4 (P≤0.005) 3 months after denervation, and 29.0±4.9 (P≤0.005) 9 months after denervation. Satisfaction survey showed decreased rate by time from 95% at 0 month (1–2 weeks from procedure), 84% at 3 months, 84% at 6 months, 68% at 9 months, and 52% at 12 months.
Group C: first repeat denervation includes 15 patients with a total of 19 denervation procedures
In group C first repeat denervation, the duration of pain relief in weeks was 49.8±9.9 (P≤0.01), with a percentage of pain relief of 68.6±13.6 (P≤0.01). In all, 11 out of 14 patients completed the revised Oswestry disability index; the score was 30.1±6.6 (P≤0.03) 3 months after denervation and 34.8±5.1 (P≤0.01) 9 months after denervation. Satisfaction survey showed decreased rate by time from 86% at 0 month (1–2 weeks from procedure), 78% at 3 months, 64% at 6 months, 57% at 9 months, and 43% at 12 months.
Group C: second repeat denervation includes 10 patients with a total of 13 denervation procedures
In group C, second repeat denervation duration of pain relief in weeks was 57.7±14.3 (P≤0.01), with a percentage of pain relief of 77.1±12.9 (P≤0.005). In all, seven out of 10 patients completed the revised Oswestry disability index; the score was 28.7±5.5 (P≤0.05) 3 months after denervation and 33.7±7.0 (P≤0.05) 9 months after denervation. Satisfaction survey showed decreased rate by time from 100% at 0 month (1–2 weeks from procedure), 90% at 3 months, 80% at 6 months, 60% at 9 months, and 50% at 12 months.
Total patients with denervation average of pain relief in weeks was 48.7±5.1 (P≤0.001) with a percentage of pain relief of 81.4±4.9 (P≤0.0005) and had a percentage decrease of opioid dose use by 50±9.7 (P≤0.001). From the 65 patients who had denervation procedures, 53 patients showed improvement in ADL.
Group D (steroid injection) had a total of 62 patients, 47 female and 15 male; of this, 30 patients had SIJCP, 19 patients had SIJCP with facet pain, and 13 patients had SI joint pain complex with piriformis. In group D, the duration of pain relief in weeks was 19.4±7.5 (P≤0.02), with a percentage of pain relief of 76.1±12.3 (P≤0.01). The number of patients with repeated procedures was 15 patients with 32 procedures; of them, five patients needed second repeat. Patient satisfaction and ADL was not done for the steroid injection group.
| Discussion|| |
Although many pain specialists point to the SI joint itself as the center and source of SI pain, it is possible that a portion of the pathology and the pain is generated from the SI complex − that is, the tissues surrounding the joint ,,,,.
Our 7-year retrospective study of patients with SIJCP resulted in the following:
- A method of trying to isolate patients with ‘pure’ SI area joint pain is offered.
- A simple method of denervation of posterior sensory pain nerves to this SIJCP is offered.
- The denervation method is straightforward. It results in over 10–12 months of greater than 50% pain relief.
- The stripline denervation involves posterior nociceptive nerves directed toward (a) the SI joint itself, (b) the ligaments posterior and medial to the SI joint.
- After the pain returns (>50% of the original pain), the denervation can be repeated with the same safety, quickness (strip denervation in 25–30 min), efficacy, and duration of pain relief.
- Some patients’ pain can originate from specific DSLs. Denervation of such ligaments also can result in long-term pain relief.
- The SI joint itself may have another kind of sensory innervation (sympathetic/mechanoreceptor?) not related to 1–3 above (see below).
- Clinically for about 43% (62/144) of the patients, the local anesthetic and steroid was good enough to obtain substantial pain relief for an average of 19 weeks if we took into account bathing the posterior ligaments, as well as the joint, with the local anesthetic and steroid. This finding is not surprising as it anatomically and physiologically makes sense.
Following strict clinical guidelines, with frequent questioning of any changes in clinical symptoms, and trying to substantiate these changes with ‘diagnostic blocks’ when possible, that is, proper use of selective piriformis muscle injection versus SI joint injection versus lumbar facet diagnostic block versus selective nerve root block, it was often possible to distinguish pain of ‘only’ SI joint complex origin. Following strict clinical criteria from our Flow Charts, we hoped to construct a percentage table ([Table 6]) as to what are the main diagnostic problems of chronic low back pain population referral to our clinic? In addition, in the SI area’s case, we tried to further subdivide that population ([Table 1]).
170/514=33.1% for the patients distribution referred had SIJPC involvement ([Table 6]), 63/170=37% of the SIJCP patients=12.2% of the 514 had ‘pure’ SI joint complex (inside the joint and its posterior ligaments) pain.
This study revealed that about 43% (62/144) patients referred to our pain clinic with SIJCP obtained satisfactory relief from local anesthetic and steroid injection of the joint and the posterior ligaments, for an average of 19.4±7.5 (P≤0.02) weeks. Both primary or repeat injections were carried out, and these patients continued to get similar satisfactory results from the injections (group D). Although each injection was done under fluoroscopy, the spread of the dye in the SI joint and about the ligaments was observed and documented.
In an attempt to evaluate other (ligament) possible sources of pain generation, we denervated caudad to cephalad the posterior spinal sensory branches of presumably L4 to S4 that travel from the spinal cord over the sacrum to innervate the SI joint only about 1–2 cm away from the edge of the SI joint. We now attempted to evaluate other possible sources of pain production by provocation of SI joint! On denervation day, after full denervation was completed, the following procedures ensued.
A G22 Quincke spinal needle was advanced from anesthetized skin toward certain designated points, as described on group F, and then 1.0 ml and then a total of 5.0 ml of Omnipaque-240 was injected into the SI joint at a constant rate of 1.0 ml/20 s. Each patient was asked to rate the quantity (VAS) and the quality (type) of possible pain experienced during injection.
Each patient in group F had been chosen as per standard criteria discussed previously (failure of group D patients, etc.). As depicted, 19/39 (48%) patients had no pain elicited by the needle at the rim, in the joint, or after 1.0 or 5.0 ml of dye injection. In all, 10/39 (26%) patients had pain similar to their SI area pain upon needle in the joint, and 1.0 or 5.0 ml of dye injection, whereas another 26% had pain dissimilar to their SI area pain upon a similar provocation!
These data suggest that the sensory information coming from the SI joint itself is more than the input of dorsal nerves traveling from presumably L4 to S4 over the posterior sacrum, crossing the stripline denervation on their way to the SI joint. Whether sympathetic innervation and/or other visceral innervation ventrally are involved is yet to be defined. In addition, this other innervation is not uniform in all the individuals tested.
The presence of nociceptive neural tissue in ligaments posterior to the SI joint has been much studied: for example, SI Joint ,,, sacrospinous ligament , sacrotuberous ligament ,, iliolumbar ligament , and the long and short posterior SI ligaments ,,. For the study patients, after stripline denervation, a few patients were noted to have pain remaining specific to certain posterior SI ligaments (see below). To evaluate and pinpoint the contribution(s) of these ligaments, we proceeded as follows.
For contribution(s) of long and short posterior SI ligaments to pain production, as of 1/2007, we moved the parallel stripline 1–1.5 cm medial to SI joint, mostly lateral to short and definitely away from the long posterior SI ligament. We encountered some of these patients (α and β below) and already had some others (γ and δ below) in whom after 2 weeks of stripline denervation of SI joint area they returned with specific pain complaints in that general area, not related to the joint itself, to be evaluated. Spinal needle provocation of the stripline, or lateral to it toward the joint, or in the joint itself, with/without dye injection if needed, did not produce any pain. However, in different groups of patients designated α (short posterior SI ligament), β (long posterior SI ligament), γ (sacrotuberous ligament), and δ (iliolumbar ligament), the injection of 1–2 ml of Omnipaque-240, into that ligament, produced the exact specific pain pattern(s) consistent with the patient’s pain complaint(s). Each injection for each ligament in question was repeated twice, at least 1 week apart, and both had to produce patient’s exact (specifically related) pain complaint(s).
On a subsequent visit for each patient before each denervation of a ligament segment, after placement of radio-opaque dye, sensory stimulation and motor stimulation was carried out ruling out any dangerous proximity to any somatic nerve. Then a small amount (1–2 mls of 0.25% bupivacaine±1 mg/ml depomedrol) bathed that ligament. Then the ligament was denervated with (a) RF cannula (G18, 10 mm bent active tip lesioning − unipolar) or (b) with a G20, 10 mm active tip Stryker-SGS (self-grounding series) (bipolar) RF lesioning.
Over time, some of the innervation of the posterior SI ligaments apparently returned causing appearance of similar pain patterns in a few patients. As per the above method, after excluding SI joint itself, and ruling in the pain-producing ligament, we proceeded with repeat RF denervation of such a ligament.
If at any time the local anesthetic plus steroid protocol failed to give any of group D patients greater than or equal to 50% pain relief lasting less than 6–8 weeks, a repeat injection was performed. If its outcome also was now the same (not satisfactory=not getting ≥3 months of >50% pain relief), then the patient became a candidate for denervation procedure. This happened to 11 patients during the study. The data of these 11 patients is used in [Table 3]; although their post SI denervation data are in full agreement with other patients in group A, they were not included in any group A statistics.
The use of a larger volume of diagnostic/therapeutic mixture of local anesthetic plus dye plus anti-inflammatory steroid solution here did not breech anyone’s previous ‘set gold standard’.
As a part of our original hypothesis, we supposed that all the sensory progenitors of pain to the SI joint complex would be in the joint, and posteriorly located to the joint. In this light, we intentionally tried to cover both the sensory innervations within the joint, and posterior to the joint (see Injection Technique and addendum B). In addendum B, if the first injection did not possibly ‘cover’ the posterior ligaments, then a second injection was given within the first 3 months. We made intentionally sure that some solution bathed the posterior ligaments, and if the patient’s response was greater than equal to 80% pain relief within minutes of this injection, then this was considered a successful initial block. We then waited to see how long this would last (see addendum B). If less than or equal to 80% pain relief was obtained on each of two successive blocks of (the SI joint and the posterior ligaments), then the possibility of the pain originating from SIJC was much reduced.
The injection of local anesthetic plus steroid into the SI joint is not new, nor is the injection of the same into ligaments near the SI joint . In this study, we combined both of these methods as we believed our pain progenitor hypothesis makes sense. Especially validating the hypothesis when through proven interventional pain management methods other differential diagnoses were able to be ruled out (e.g. lumbar facet pain, piriformis pain, selective nerve root pain, lumbar discogenic pain, etc.). Borowsky et al.  presented retrospective reviews of two large case series that concluded that significant posterior sources of SI region exist, and intra-articular diagnostic blocks underestimate the prevalence of SI region pain.
This retrospective study shows a simple, reproducible method for long-term sensory denervation of all the posterior nociceptive pain generators involved with the SI joint area.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Bowen V, Cassidy JD. Macroscopic and microscopic anatomy of sacroiliac joint from embryologic life until the eight decade. Spine 1981; 6:620–628.
Vleeming A, vanWingarden JP, Snijder CJ, Stoeckart R, Stijnen T. Load application to the sacrotuberous ligament: influence of sacroiliac joint mechanics. Clin Biomech 1989; 4:204–209.
Vleeming A, vanWingarden JP, Dijkstra PF, Stoeckart R, Snijder CJ, Stijnen T. Mobility in the sacroiliac joint in the elderly: a kinematic and radiologic study. Clin Biomech 1992; 7:170–176.
Cohen SP. Sacroiliac joint pain: a comprehensive review of anatomy, diagnosis and treatment. Anesth Analg 2005; 101:1440–1453.
Bernard TN, Cassidy JD. The sacroiliac syndrome. In: Frymoyer JW, editor. Pathophysiology, diagnosis and management. New York, Raven: The Adult Spine: Principle and Practice; 1991. 2107–2130.
Schwarzer AC, Aprill GN, Bagduk N. The sacroiliac joint in chronic low back pain. Spine 1995; 20:31–37.
Paradise LA, Raj P. Sacroiliac joint blocks. In: Prithvi Raj P, editors. Interventional pain management image-guided procedures. 2nd ed. Philadelphia, USA: Elsevier (Saunders); 2008. 429–442.
Slipman CW, Jackson HB, Lipetz JS, Chan KT, Lenrow D, Vresilovic EJ. Sacroiliac joint pain referral zones. Arch Phys Med Rehabil 2000; 81:334–338.
Grob KR, Neuhuber WL, Kissling RO. Innervations of the sacroiliac joint of the human. Z Rheumatol 1995; 54:117–122.
Yin W, Willard F, Carreiro J, Dreyfuss P. Sensory stimulation-guided sacroiliac joint radiofrequency neurotomy: technique based on neuroanatomy of the dorsal sacral plexus. Spine 2003; 28:2419–2425.
Dreyfuss P, Henning T, Malladi N, Goldstein B, Bogduk N. The ability of multi-site multi-depth sacral lateral branch block to anesthetize the sacroiliac joint complex. Pain Med 2009; 10:679–688.
Szadek KM, Hoogland JM, Piet V, Zuurmond WWA, DeLang JJ, Perez RSGM. Possible nociceptive structures in the sacroiliac joint cartilage: an immunohistochemical study. Clin Anat 2010; 23:192–198.
Fortin JD, Kissling RO, O’Conner BL, Vilensky JA. Sacroiliac joint innervation and pain. Am J Orthoped 1999; 28:687–690.
Berthelot JM, Labatt JJ, LeGoff B, Gouin F, Maugavs Y. Provacative sacroiliac joint maneuvers and sacroiliac joint blocks are unreliable for diagnosingsacroiliac joint pain. Joint Bone Spine 2006; 73:17–23.
Luukkainen RK, Nissila M, Asikainen EL, Sanila MT, Lehtinen K, Alanaauts A, Kautiainen HH. Periarticular corticosteroid treatment of sacroiliac joint in patients with Seronegative spondylo-arthropathy. Clin Exp Rheumatol 1999; 17:88–90.
Luukkainen RK, Wennerstrand PV, Kautiainen H, Sanila MT, Asikainen EL. Efficacy of corticosteroid treatment of the sacroiliac joint in non-spondylarthropathic patients with chronic pain in the region of sacroiliac joint. Clin Exp Rheumatol 2002; 20:52–54.
Maigne JY, Aivaliklis A, Pfefer F. Results of sacroiliac joint double block and value of sacroiliac pain provocation test in 54 patients with low back pain. Spine 1996; 21:1889–1892.
Vleeming A, DeVries HJ, Mens JMA, vanWingarden JP. Possible role for the dorsal sacral ligament in women with peripartum pelvic pain. Acta Obstet Gynecol Scand 2002; 81:430–436.
Fortin JD, Washington WJ, Falco FJE. Three pathways between the sacroiliac joint and neural structures. Am J Neurorad 1999; 20:1429–1434.
Haufe SM, Mork SA. Sacroiliac joint debridement: a novel technique for treatment of sacroiliac joint pain. Photomed Laser Surg 2005; 23:596–598.
Ikeda R. Innervation of sacroiliac joint. Macroscopic and histologic studies. J Nippon Med School 1991; 58:587–596.
Dreyfuss P, Michaelsen M, Pauza K, Mclarty J, Bogduk N. The value of medical history and physical examination in diagnosing sacroiliac joint pain. Spine 1996; 21:2594–2606.
Van der Wurff P, Buijs EJ, Groen GJ. A multitest regimen of pain provocation tests as an aid to reduce unnecessary minimally invasive sacroiliac joint procedures. Arch Phys Med Rehabil 2006; 87:10–14.
Laslett M, Aprill CN, McDonald B, Young SB. Diagnosis of sacroiliac joint pain: a validity of individual provocation tests and composites of tests. Man Ther 2005; 10:207–218.
Hansen HC, McKenzie-Brown AM, Cohen SP, Swicegood JR, Colson JD, Manchikanti L. Sacroiliac joint interventions: a systemic review. Pain Physician 2007; 10:165–184.
Fortin JD, Pier J, Falco F. Sacroiliac joint injection: pain referral mapping and arthrographic findings. In: Fleeming A, Mooney V, Darman T, Snijder C, Stoeckart R, editors. Movement, stability and low back pain, the essential role of the pelvis. New York: Churchill Livingstone; 1997. 271–285
Fortin JD, Dryer AP, West S. Sacroiliac joint: pain referral maps upon applying a new injection/arthrography technique − part I: asymptomatic volunteers. Spine 1994; 19:475–482.
Fortin JD, Tolchin R. Sacroiliac arthrograms and post-arthrography CT. Arch Phys Med Rehab 1993; 74:1259–1264.
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.
Boswell MV, Singh V, Staats PS, Hirsch JA. Accuracy of precision diagnostic blocks In the diagnosis of chronic spinal pain of facet or zygapophysial joint origin: a systemic review. Pain Physician 2003; 6:449–456.
Raj P, Lou L, Erdine S, Staats P, Waldman S. Piriformis muscle injection. In: PrithviRaj P. Radiographic imaging for regional anesthesia & pain management. Churchill Livingstone (London, England): Elsevier Publisher; 2003. 250–253.
Benzon HT, Katz JA, Benzon HA, Iqbal MS. Piriformis syndrome: anatomical considerations, a new injection technique, and a review of the literature. Anesthesiol 2003; 98:1442–1448.
Dussatt RG, Kaplan PA, Anderson MW. Fluoroscopy guided sacroiliac joint injection. Radiology 2000; 214:273–277.
Kim H-J, Shin D-G, Kim HL, Shin D-A. Selective neurotomy of sacral lateral branches for pain of sacroiliac joint dysfunction. J Korean Neurosurg Soc 2005; 38:338–343.
Ferrante FM, King LF, Roche EA, Kim PS, Aranda M, DeLaney LR et al.
Radiofrequency sacroiliac joint denervation for sacroiliac syndrome. Reg Anesth Pain Med 2001; 26:137–142.
Burnham RS, Yasui Y. An alternative method of radiofrequency neurotomy of the sacroiliac joint: a pilot study of the effect of pain, function, and satisfaction. Reg Anesth Pain Med 2007; 32:12–19.
Cohen SP, Hurley RW, Buckenmaier CC 3rd, Kurihara C, Morlando B, Dragovich A. Randomized placebo-controlled study evaluating lateral branch radiofrequency denervation for sacroiliac joint pain.Anesthesiology 2008; 109:279–288.
Vilensky JA, O’Conner BL, Fortin JD, Merkel GJ, Jimenez AM, Scofield BA, Kleiner JB. Histological analysis of neural elements in the human sacroiliac joint. Spine 2002; 27:1202–1207.
Barksdale PA, Gasser RF, Gauthier CM, Elkins TE, Wall LL. Intraligamentous nerves as potential sources of pain after sacrospinous fixation of the vaginal apex. Int Urogynecol J Pelvic Floor Dysfunct 1997; 8:121–125.
Roshanravan SM, Weislander CK, Schaffer JI, Corton MM. Neurovascular anatomy of the sacrospinous ligament region in female cadavers: implications in sacrospinous ligament fixation. Am J Obstet Gynec 2007; 197:660 e1–660 e6.
Woodly SJ, Kennedy E, Mercer SR. Anatomy in practice: the sacrotuberous ligament, New Zeeland J Physiol 2005; 33:91–94.
Kiter E, Karaboyun T, Tufan AC, Acar K. Immunohistochemical demonstration of nerve endings in iliolumbar ligament. Spine 2010; 35:E 101–E 104.
Keel JC. Fluoroscopically guided vs. blind steroid injections for chronic iliolumbar syndrome: case control study. Arch Phys Med Rehab 2007; 88:E70.
Varga E, Dudas B, Tile M. Putative propioceptive function of the pelvic ligaments: biochemical and histological studies. Injury 2008; 39:858–864.
Borowsky CD, Fagen G. Sources of sacroiliac region pain: insights gained from a study comparing standard intra-articular injection with a technique combining intra- and peri-articular injection. Arch Phys Med Rehab 2008; 89:2048–2056.
McGrath C, Nicholson H, Hurst P. The long posterior sacroiliac ligament: a histological study of morphological relation in the posterior sacroiliac region. Joint Bone Spine 2009; 76: 57–62.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]