Clinical Observations
Musculoskeletal Imaging
October 2005

Diagnostic and Therapeutic Use of Sonography-Guided Iliopsoas Peritendinous Injections

Abstract

OBJECTIVE. Our objective was to review our experience performing sonography-guided iliopsoas bursal/peritendinous injections as a diagnostic and therapeutic tool in the workup and treatment of patients with hip pain.
CONCLUSION. Sonography-guided iliopsoas bursal/peritendinous injections are useful in determining the cause of hip pain. They can provide relief to most patients with iliopsoas tendinosis/bursitis after hip replacement. The results of injection alone are not as successful in cases of idiopathic iliopsoas tendinosis/bursitis, but the technique can help determine which patients may benefit from a surgical tendon release.

Introduction

Groin pain can present a difficult therapeutic and diagnostic dilemma. The source of pain in this region can range from abdominal wall hernias to muscle abnormalities, primary hip disorders (i.e., osteoarthritis), and secondary mechanical wear or impingement from indwelling orthopedic hardware (i.e., total hip replacements). A commonly recognized cause of hip and/or groin pain is iliopsoas tendinosis, which is often seen in the setting of rheumatoid arthritis, acute trauma, overuse injury, or after total hip arthroplasty [16].
In cases where iliopsoas tendinosis is suspected, an anesthetic–corticosteroid injection into the tendon's adjacent bursa can confirm the diagnosis or help determine an alternate cause. Complete or near complete relief of the symptoms confirms the diagnosis. This injection may provide temporary or permanent symptomatic relief and may postpone or avoid surgical intervention [1]. Failure of the injection to provide relief may indicate an alternate cause for hip pain, such as labral degeneration or tear. Current imaging techniques, particularly MRI, can be helpful in delineating anatomic abnormalities. However, the abnormalities seen on imaging are multifactorial and cannot be readily sorted out during physical examination alone.
Interventional musculoskeletal procedures have traditionally been guided by fluoroscopy or CT. However, these imaging techniques are not without limitations [7]. Fluoroscopy does not provide direct visualization of soft-tissue interfaces and requires needle manipulation with test injections of contrast material to confirm needle placement. CT allows for direct visualization of soft tissue; however, it traditionally has not been performed in real time, provides poor tissue contrast, and requires intermittent needle manipulation. Real-time CT fluoroscopy is an option on many new generation scanners but it does not provide the same degree of soft-tissue contrast present on sonography. Real-time observation during needle placement under and during injection would be possible, however. Both fluoroscopy and CT use ionizing radiation.
In this article, we will describe our experience using sonography guidance to perform therapeutic injections of the iliopsoas bursa. The real-time nature of sonography allows one to continuously monitor needle placement and the distribution of the therapeutic agent.

Materials and Methods

This study, which was approved by our institutional review board, involved a retrospective review of the results of sonography-guided injections into the iliopsoas bursa in 39 patients from April 2001 to May 2003. The 39 patients fell into one of two categories: Group A consisted of 11 patients with total hip replacements (seven women and four men; age range, 50–81 years; mean age, 67 years). The average age of the hip replacements was 18.9 months (range, 3–44 months). Group B consisted of 28 patients (18 women and 10 men; age range, 16–67 years; mean age, 35 years) who had not undergone prior total hip arthroplasty. Both groups of patients presented with similar symptoms of hip and/or groin pain that were aggravated by hip flexion or straight-leg raises. Patients in Group B experienced a wider range of symptoms, including tenderness, snapping, clicking, grinding, and catching. Patients were referred for sonography-guided injection for the presumed diagnosis of iliopsoas tendinosis that was refractory to conservative treatment (e.g., nonsteroidal antiinflammatory drugs) and to help establish the diagnosis when clinical and other imaging studies were inconclusive.
Fig. 1 Axial image of iliopsoas tendon at level of joint line in 19-year-old man. Fast spin-echo proton density image (left) and corresponding sonogram (right) show anatomic relationship of iliopsoas tendon (arrow), femoral head (F), and iliopectineal eminence of acetabulum (A). Tendon (arrow) is seen as elliptic structure situated medial and superficial to joint capsule and along lateral margin of iliopectineal eminence.
All injections were performed using a freehand technique with the patient supine. The tendon was localized using a 3.5- to 7.5-MHz transducer in sector or curved linear geometry, with transducer selection depending on factors such as body habitus and geometry. Better visualization of the tendon was generally possible in thinner patients, in whom a linear 7.5-MHz transducer can be used. Larger patients required use of a curved linear transducer. All scans were performed on a Sonoline Elegra (Siemens Medical Solutions) scanner. Injections were performed using a lateral approach at the level of the joint line (iliopectineal eminence) and a 22- or 20-gauge spinal needle (Fig. 1). A standardized therapeutic mixture consisting of 5 mL 1% lidocaine, 2 mL bupivacaine, and 40–80 mg of triamcinolone was administered under sterile conditions. The techniques used have been previously reported for a variety of similar sonography-guided interventions [7]. The end point of the injection was always direct distention of the iliopsoas bursa and/or the presence of microbubbles in a peritendinous distribution (Figs. 2A and 2B). If there was preexisting fluid distention of the bursa, then the needle was directed into the bursa. When no preexisting bursal distention had occurred, the needle was positioned between the tendon and the hip capsule at the level of the iliopectineal eminence. Test injections were performed using 1% lidocaine while observing in real time to ensure bursal distention.
The only potential immediate complication was a transient femoral nerve palsy. This results in diminished sensation over the anterior thigh and inability to flex the hip or put weight on the extremity. Patients were observed in the waiting area immediately after the procedure for onset of symptoms. If patients exhibited any symptoms, they were observed in the ultrasound laboratory until their resolution. When patients were able to ambulate approximately 50 feet (15 m) without a sensation of the leg “giving way,” they were released. In our experience, this seemed to be the effect of lidocaine infiltrating the deep tissue during needle placement.
Patient charts were reviewed and immediate follow-up for 17 of the 39 patients (all from Group B) was documented in their medical charts. This information was recorded after receiving telephone calls from these patients within 1 to 5 days after the injection. For the remaining 22 patients, immediate pain assessment was based on their recollection of pain relief, which was obtained during the initial follow-up telephone call.
All 39 patients were sent a standardized pain assessment questionnaire and then contacted by telephone to discuss the survey. Patients were asked to rate their pain relief in quartile percentages (i.e., 0–25%, 25–50%, etc.). If immediate follow-up data were not available in their charts, patients were asked to recall their percentage of pain relief within the first 24 hr after the procedure. They also used the percentage scale to assess their present pain relief compared with their baseline (i.e., before the injection). Inquiries were made about the use of antiinflammatory drugs at the time of the injection and the time of the follow-up telephone call. In addition, the patients were asked if they would consider undergoing the procedure again should the problem recur.
Those patients who were contacted at less than 12 months and who indicated a positive response were contacted again via telephone a year after the injection. For purposes of this study, a positive outcome was defined as pain relief subjectively greater than 50%.

Results

A total of 55 injections were made in 39 patients, with a mean time to follow-up of 13.5 months (range, 1 to 26 months) for each injection. Eight patients were injected twice and four patients were injected three times (mean time between injections, 6.2 months). A short learning curve in optimally performing these injections became apparent. There were two separate issues: needle placement and the volume of injected material. In the Group B patients, the anatomy was generally well defined. Test injection with 1% lidocaine served as a method to ensure filling of the bursa. During the course of treating several patients, it became clear that needle placement deep in relation to the tendon at the level of the iliopectineal eminence would more readily fill the bursa, thereby reducing the amount of needle manipulation. In Group A patients, the iliopsoas tendon was invariably abnormal but also anatomically distorted by the indwelling hardware. The tendon was typically displaced anteriorly and medially by the acetabular cup. It took imaging several patients before recognizing this as a consistent pattern. Alterations in the volume of injected material were changed somewhat based on discussions between the radiologist performing the procedure and the referring surgeons. A large volume of anesthetic was preferred to achieve better overall coverage of the bursa and tendon. The amount of steroid injected remained constant.
Fig. 2A Sonography-guided iliopsoas bursal injection in 27-year-old woman. Axial sonogram of iliopsoas tendon (t) with 22-gauge spinal needle (arrow) in position for sonography-guided injection. Lateral approach is used with transducer operating in sector format. Bony eminence of acetabulum (a) provides target for needle placement. Femoral head (fh) is labeled.
Fig. 2B Sonography-guided iliopsoas bursal injection in 27-year-old woman. Injection (long arrow) is performed while observing in real time. Fluid distention of bursa is illustrated with presence of microbubbles, producing echogenic foci along nondependent surface of bursa (short arrow, arrowhead).
Twenty-three patients received more than one telephone call, provided that the first telephone call occurred less than a year after the injection and the patient indicated an initial positive response. For Group A patients who received more than one call, the initial call occurred a mean of 3.9 months after the injection and the second call, a mean of 13 months after injection. For Group B patients, the initial call was made a mean of 3.5 months after the first injection and the second call, a mean of 18 months after the injection.
Thirty-seven of the 39 patients had not been taking analgesics for a minimum of 2 weeks at the time of the injection. The remaining two patients (from Group B), who only experienced relief for the first few hours, were taking oral analgesics at the time of the injection and were also taking them at the time of the follow-up telephone call. Neither of these patients was considered a responder to the injection and both had other etiologic factors implicated as their source of pain.
Twenty-seven of the 39 patients were willing to repeat the procedure. Those patients who were most willing to undergo the procedure again were those who experienced relief extending beyond the first 24 hr.
The results varied in the two groups. In Group A, one patient did not obtain immediate or delayed relief despite two iliopsoas injections. It was thought that the hip pain was not related to the iliopsoas tendon. Of the remaining 10 patients, all experienced immediate relief, with nine of 10 (90%) maintaining at least 50% pain relief at 1 year. The one patient without sustained success had relief for 2 weeks, while a second injection did not provide improvement. One patient had three injections, while five patients received two injections, including the patient not achieving sustained relief (mean time for pain relief, 9.1 months; range, 2 weeks to 17 months). One of the injections was repeated due to technical problems.
In Group B, 10 of the 28 patients did not experience immediate or delayed relief. Thus, it was thought that the pain was not related to the iliopsoas tendon. Of the 10, one underwent varus osteotomy of the proximal femur with resolution of pain. One experienced relief after arthroscopic débridement of a labral tear. One had continuing problems from untreated hip dysplasia. Two had refractory pain after hip arthroscopy. One patient had relief after a subsequent intraarticular injection and was considered a candidate for arthroscopic débridement. In four patients, the underlying cause remains unclear.
Of the remaining 18 patients in Group B, all sustained immediate pain relief after iliopsoas injection. Even fewer achieved sustained relief; eight of 18 (44%) of these patients, including one who required two injections, had continued relief at 1 year. One patient has had three injections, each providing short-term relief. The patient was not willing to consider surgical treatment.
Four Group B patients underwent surgical release of the iliopsoas for continued symptoms after temporary relief from iliopsoas injections. One procedure was a complete release done open in a patient with a previous varus osteotomy. The patient obtained complete pain relief. Three releases were partial and done arthroscopically. Two of the three obtained relief from the procedure; the third is under consideration for a repeat tendon release. Of the remaining patients who had temporary relief, four had improvement after a course of physical therapy. One patient had relief after an additional cortisone injection to the hamstring origin.
No major complications occurred from the injections. Two patients experienced a temporary femoral nerve palsy, which cleared in 30 min despite visualization of the femoral neurovascular bundle. There were no infections or vascular injuries.

Discussion

The cause of hip and groin pain is multifactorial. Although advanced imaging techniques are helpful to limit the number of possible etiologic factors, the final arbiter is clinical response to a therapeutic intervention. Once primary hip disorders have been eliminated, the hip flexors and abductors are among the next most common sources of pain. It can be very difficult to ascertain whether the pain is from an intraarticular or extraarticular source. Imaging evidence of joint-related pathology may be present, and this can help to establish the source. In the absence of an obvious cause, or if more than one potential cause is present, it is often helpful for the patient to undergo a diagnostic injection, with documentation to assess the degree of pain relief. The injection can be intra- or extraarticular, depending on clinical and imaging-based findings. Clinical features such as restricted motion, catching, or snapping may be present with a variety of conditions [1, 2, 8].
The iliopsoas muscle, the most powerful flexor of the thigh, is a compound muscle composed of the psoas and iliacus muscles [9]. The psoas, originating from the lumbar spine, and the iliacus, arising from the pelvis, converge to form the iliopsoas muscle and insert anteromedially onto the lesser trochanter of the femur as the iliopsoas tendon. Problems may occur when inflammation of the tendon is present or if bursitis exists between the tendon and the pelvis or anterior hip capsule [1, 2]. Iliopsoas irritation may also occur after total hip arthroplasty, especially if a prominent socket impinges on the tendon [3, 5, 6, 10].
Using well-defined landmarks, the tendon is sonographically apparent at the level of the iliopectineal eminence (Fig. 1). Sonography-guided needle placement in this location will generally allow distention of the iliopsoas bursa. As with other sonography-guided interventions, we have found a test injection with local anesthetic is helpful to confirm needle position. The real-time nature and lack of ionizing radiation allow simple continuous needle adjustments. Sonography also has the advantage of direct visualization of the neurovascular bundle. Application of deep local anesthesia may result in temporary femoral nerve palsy, as was the case in two of our patients (it cleared within 30 min). In no patient was direct impingement of the neurovascular bundle an issue.
Some limitations of our study include its retrospective nature, which does not allow a systematically controlled and identical follow-up for each patient. Follow-up interviews varied with respect to the timing of the initial telephone contact after the injection. The use of a visual pain scale and asking the patient to recall level of pain are subjective. Alternatively, it is their subjective assessment that brings them to seek medical attention. Performance of a well-defined prospective study may be of value in the future, in which patients are randomized to different therapy regimens. Large body habitus occasionally posed problems in visualizing the tendon and tendon anisotropy. In patients with total hip replacements, there was anatomic distortion, with the tendon often more anterior and medial. In cases of severe tendinosis, the tendon was not as readily apparent as in the population with de novo problems or in those patients referred after hip arthroscopy. There was a short learning curve in the total amount of volume of anesthetic–cortisone mixture to inject and in optimally positioning the needle to ensure effective bursal distention.
Our initial experience with sonography-guided iliopsoas bursal/peritendinous injections has proven useful in determining the cause of hip and/or groin pain. In our experience, anesthetic–corticosteroid injection performed under sonography guidance provides a safe and effective method to inject the bursa and to exclude the iliopsoas tendon as a possible cause for pain. In patients with temporary relief, the technique is of value to determine whether patients would be candidates for surgical tendon release. These injections can provide long-term relief in arthroplasty patients and in some nonarthroplasty patients as well. Ninety percent of patients with iliopsoas-related symptoms after hip arthroplasty achieved significant relief without the need for surgical release or revision arthroplasty. The lower success rate in the nonarthroplasty patients probably reflects other causative factors involved in their groin pain, such as labral degeneration or anterior impingement, which are difficult to rectify by injection alone. In our series, a substantial number of nonarthroplasty patients (10) were ultimately shown to have other etiologic factors (e.g., labral tears) as the source of their hip pain. Eighteen of 28 (64%) in this group of patients responded, helping to confirm the diagnosis of iliopsoas tendinosis.
Based on our initial experience, sonography-guided iliopsoas peritendinous injections have become a standard part of the clinical algorithm in evaluating these patients at our institution. In patients with indwelling hip replacements, in whom aseptic loosening or infection are excluded, sonography-guided injection has become the next step in the assessment of groin pain. In those patients with de novo hip problems, the rationale to try a sonography-guided injection remains exclusionary—that is, to help sort out possibilities suggested by clinical examination and imaging assessments. In patients who have long-term relief, sonography-guided injection provides a useful therapeutic option to surgical release. The relative ease of the procedure coupled with a very low morbidity makes it an excellent choice in the treatment of refactory iliopsoas tendinitis bursitis.

Footnote

Address correspondence to R. S. Adler ([email protected]).

References

1.
Johnston CA, Wiley JP, Lindsay DM, Wiseman DA. Iliopsoas bursitis and tendinitis: a review. Sports Med 1998; 25:271-283
2.
Allen WC, Cope R. Coxa saltans: the snapping hip revisited. J Am Acad Orthop Surg 1995; 3:303-308
3.
Trousdale RT, Cabanela ME, Berry DJ. Anterior iliopsoas impingement after total hip arthroplasty. J Arthroplasty 1995; 10:546-549
4.
Taylor GR, Clarke NM. Surgical release of the snapping iliopsoas tendon. J Bone Joint Surg Br 1995; 77:881-883
5.
Heaton K, Dorr LD. Surgical release of iliopsoas tendon for groin pain after total hip arthroplasty. J Arthroplasty 2002; 17:779-781
6.
Taher RT, Power RA. Iliopsoas tendon dysfunction as a cause of pain after total hip arthroplasty relieved by surgical release. J Arthroplasty 2003; 18:387-388
7.
Sofka CM, Collins AJ, Adler RS. Use of ultrasonographic guidance in interventional musculoskeletal procedures: a review from a single institution. J Ultrasound Med 2001; 20:21-26
8.
McCarthy JC, Day B, Busconi B. Hip arthroscopy: applications and technique. J Am Acad Orthop Surg 1995; 3:115-122
9.
Tatu L, Parratte B, Vuillier F, Diop M, Monnier G. Descriptive anatomy of the femoral portion of the iliopsoas muscle: anatomical basis of anterior snapping of the hip. Surg Radiol Anat 2001; 23:371 —374
10.
Jasani V, Richards P, Wynn-Jones C. Pain related to the psoas muscle after total hip replacement. J Bone Joint Surg Br 2001; 84:991-993

Information & Authors

Information

Published In

American Journal of Roentgenology
Pages: 940 - 943
PubMed: 16177412

History

Submitted: July 30, 2004
Accepted: October 25, 2004
First published: November 23, 2012

Authors

Affiliations

Ronald S. Adler
Department of Radiology and Imaging, Hospital for Special Surgery, 535 East 70th St., New York, NY 10021.
Robert Buly
Department of Orthopaedic Surgery, Hospital for Special Surgery, New York, NY 10021.
Regina Ambrose
Department of Radiology and Imaging, Hospital for Special Surgery, 535 East 70th St., New York, NY 10021.
Thomas Sculco
Department of Orthopaedic Surgery, Hospital for Special Surgery, New York, NY 10021.

Metrics & Citations

Metrics

Citations

Export Citations

To download the citation to this article, select your reference manager software.

Articles citing this article

Media

Figures

Other

Tables

Share

Share

Copy the content Link

Share on social media