Abstract
Aspiration and injection of joints and soft tissues is an indispensable skill used in everyday practice by the clinical rheumatologist. Most rheumatologists recognise that performing these procedures using anatomical landmarks is not always successful, particularly in the case of small or infrequently injected joints, bursae or tendon sheaths. Musculoskeletal ultrasound confirms the local pathological – anatomical diagnosis and is the most applicable and feasible imaging method that can be applied in clinical practice in guiding musculoskeletal interventional procedures. From 1993, there has been substantial examination of the accuracy of landmark- and imaging-guided procedures. We have searched the literature and ascertained whether imaging techniques improve the accuracy of musculoskeletal procedures and whether the accuracy of needle placement can be translated into improved clinical outcome (efficacy).
Introduction
Despite major advances in systemic therapy for the treatment of inflammatory rheumatic diseases, corticosteroid (steroid) injection into the joints, bursae, tendon sheaths or other musculoskeletal soft tissues remains an important tool used in daily clinical practice. These injections are usually guided by the palpation of anatomical landmarks and are termed landmark-guided injection (LMGI) (also known as palpation-guided, clinical examination-guided or blind injection techniques). Jones et al., in 1993 were the first to report the success rates of needle placement using a palpation-guided approach . The authors proposed that accurate placement of the needle could diminish the discomfort experienced by the patient and reduce incidental damage to adjacent structures and tissue atrophy within the extra-articular soft tissues.
Ultrasound (US) guidance of aspiration was first reported by Gompels and Darlington in 1981 . The first step of an US-guided procedure is always to confirm and localise the joint or soft tissue pathology. In clinical practice, US has been shown to frequently change the anatomical and pathological diagnosis made on clinical grounds, which alters the decision of whether or not to inject and where to place the needle . The superior diagnostic accuracy of US scanning compared with clinical examination is a major factor that should be examined in future studies that compare the outcomes of LMGIs and US-guided injections (USGIs). In addition to aspiration and steroid injections, it is now possible to perform US-guided percutaneous biopsies of the joints, bursae, tendon sheaths, major salivary glands and muscles; to assist insertion of drainage catheters; to detect soft tissue foreign bodies; and to perform nerve blocks. There are two common methods for USGIs: semi-guided or indirect method (skin surface marking) and needle guidance under direct sonographic vision (direct method) . Direct visualisation of the needle is considered to be the superior technique because it confirms the correct position of the needle. The needle is observed in real time while it is passing through the tissue. There are no studies that directly compare the performance of these two USGI techniques.
US is the optimal imaging modality for guiding musculoskeletal interventional procedures because it is extensively available at bed side, provides high definition imaging of both needle and joint tissues, incurs relatively low costs and does not use radiation . In this article, we address whether USGIs improve the accuracy of therapeutic musculoskeletal injections compared with LMGIs and, further, the accuracy of needle placement can be translated into improved clinical outcome (efficacy) in the different anatomical regions. We searched all English language articles published in the last 35 years using PubMed database and collected clinical trials comparing the accuracy and/or efficacy of LMGIs with USGIs. We focused on human rheumatic and musculoskeletal conditions and excluded oncology, traumatology and spinal procedures.
Shoulder
Glenohumeral joint
Accuracy
In a study by Cunnington et al., injections of steroid and contrast agent were given in 184 different joints randomised for LMGIs and USGIs. The accuracy was assessed by radiography of the injected joint, which localised the contrast agent in the injectate. The accuracy of the glenohumeral joint injections reported for USGIs was 63% and for LMGIs was 40% (p = NS) . It should be noted that in this study, USGI was performed using the posterior approach by a relatively inexperienced ultrasonographer with one year of practical US experience, and 40% accuracy was achieved by rheumatologists with a median of 14 years of clinical experience in joint injection. It would be expected that the accuracy of a skilled ultrasonographer interventionist would be much higher, and the sample size in this study was small making statistical analysis difficult.
There are eight clinical studies that investigated the use of LMGIs. Three of these used magnetic resonance (MR) arthrography for accurate injection verification, four used contrast medium and fluoroscopy and one used arthroscopy. According to these reports, the mean accuracy of LMGIs to the glenohumeral joint was 67% (variation 10–99%) . In a paper by Tobola et al. , three different injection approaches were investigated (anterior, posterior or supraclavicular), and statistically significant differences were absent between the three approaches. However, the anterior route was the most accurate and was independent of the experience level of the provider.
There are five other studies that have investigated the accuracy of USGIs before MR arthrography . The mean reported accuracy of USGIs to the glenohumeral joint were 96% (variation 92–100%). One of these studies compared US- and fluoroscopy-guided techniques and noted that the first attempt was successful in 72% of the injections in the fluoroscopy-guided group and 94% in the US-guided group .
In a cadaveric study with 80 shoulder specimens, the accuracy of USGIs and LMGIs were 92.5% and 72.5% (p = 0.025), respectively . Two operators injected radiopaque contrast through a posterior approach. After the injections, radiographs of the specimens were obtained to assess the accuracy of the injections.
There are three cadaveric studies that investigated LMGIs to the glenohumeral joint . The mean accuracy was 94% (variation 91–96%). One study used superior injection approach and noted that caution for the likelihood of penetrating the long head of biceps tendon should be considered .
These findings provide evidence that USGIs to the glenohumeral joint are more accurate than LMGIs ( Table 1 ).
| LMGI (%) | USGI (%) | |
|---|---|---|
| Comparative clinical study (1) | 40 | 63 |
| Clinical LMGIs studies (8) | 67 | – |
| Clinical USGIs studies (5) | – | 96 |
| Comparative cadaveric study (1) | 72.5 | 92.5 |
| Cadaveric LMGIs studies (3) | 94 | – |
Efficacy
There is a single comparative study of LMGIs and USGIs that investigated the efficacy of glenohumeral joint injections using these techniques. In this trial for patients with adhesive capsulitis, the US group had a lower reduction in pain VAS; however, no significant difference was observed between groups for function at 6 weeks post injection . In Cunnington’s study, there was no significant difference for outcome variables between USGIs and LMGIs when all the joints were grouped together. The results for single joints were not reported. However, there was a greater improvement in the VAS score for function in the accurate injection group . A recent Cochrane review reported no significant improvement in efficacy with USGIs to the shoulder . However, the authors did not perform the analysis on the basis of the exact injection location. In addition, there were fewer side effects in the USGI groups.
Presently, there is not enough data to conclude that USGIs are more efficacious than LMGIs in the treatment of glenohumeral joint diseases.
Subacromial space
Accuracy
In assessing the studies on injection into the subacromial space, it should be noted that different studies have used different volumes of injectate, and a larger volume may aid the accuracy but leads to more tissue damage. Two clinical studies have compared LMGIs and USGIs using MR arthrography for verifying the success of the intervention. Rutten et al. reported that both methods were 100% accurate , and Dogu et al. found that accurate injections were performed in 65% in the USGI group and 70% in the LMGI group .
Four clinical studies have investigated the accuracy of LMGIs into the subacromial space . One used MR arthrography and the others used fluoroscopy plus contrast media for verification. The mean accuracy was 61% (variation 29–91%).
Three cadaveric studies have examined the accuracy of LMGIs into the subacromial space . The mean accuracy was 81% (variation 70–91%). Thus, the accuracy of LMGIs into the subacromial space is sub-optimal and depends on the operator.
There is not enough data to confirm that USGIs are more accurate than LMGIs, and further research is required to clarify this ( Table 2 ).
| LMGI (%) | USGI (%) | |
|---|---|---|
| Comparative clinical studies (2) | 85 | 82.5 |
| Clinical LMGIs studies (4) | 61 | – |
| Cadaveric LMGIs studies (3) | 95 | – |
Efficacy
Five studies have evaluated the efficacy of USGIs and LMGIs into the subacromial space, and they all showed better outcomes for the USGI procedures in the short term . However, two additional studies found that there is no difference in efficacy between accurate and inaccurate injections .
According to the literature, USGIs into the subacromial space produce a significantly greater improvement in terms of pain reduction and functional gain than LMGIs in chronic shoulder pain, at least in short term.
Biceps tendon sheath, acromioclavicular joint and sternoclavicular joint
Accuracy and efficacy
Injection into the tendon sheath of the long head of the biceps brachii can be more accurately performed by USGI than that by LMGI according to a single study . USGIs showed 87% accuracy, whereas LMGIs showed only 27% accuracy. Computed tomography (CT) with contrast agent was used for the verification of the injection; however, the efficacy of this was not reported.
In one clinical study, the accuracy of LMGIs to the acromioclavicular joint was 43%. This was verified using radiographic contrast material . There are no papers reporting the efficacy of acromioclavicular injections.
The exact position of USGI and LMGI into the acromioclavicular joint has been examined in three cadaver studies . The mean accuracies reported were 96% (variation 90–100%) for USGIs and 62% (variation 40–75%) for LMGIs. In a cadaveric study, Partington and Broome found that LMGIs were successful in 67% of the acromioclavicular injections .
It can be concluded that LMGIs to the biceps tendon sheath and acromioclavicular joint are sub-optimal and USGIs are more accurate; however, efficacy studies for this are missing.
The only study evaluating injection accuracy into the sternoclavicular joint used LMGI approach with a reported accuracy of 78% . This was a cadaveric study ( Table 3 ).
| LMGI (%) | USGI (%) | |
|---|---|---|
| Biceps tendon sheath: comparative clinical study (1) | 27 | 87 |
| Acromioclavicular joint: clinical LMGIs study (1) | 43 | – |
| Acromioclavicular joint: comparative cadaveric studies (3) | 62 | 96 |
| Acromioclavicular joint: cadaveric LMGIs study (1) | 67 | – |
| Sternoclavicular joint: cadaveric LMGIs study (1) | 78 | – |
Advanced procedures of the shoulder
Some procedures in the shoulder area require exact imaging guidance. Percutaneous US-guided fenestration and aspiration of calcific tendinosis is an example . It was originally done as a fluoroscopic procedure; however, US has largely replaced fluoroscopy because of its excellent safety profile and clinical efficacy. In the shoulder region, nerve blocks can be performed using US guidance (e.g. suprascapular and dorsal scapular nerve blocks). US may be used to diagnose and aspirate a ganglion cyst compressing the nerve in the suprascapular or spinoglenoid notch. Other injectable therapies such as platelet-rich plasma are now available for rotator cuff tendinopathy; however, further studies are required to clarify their efficacy, and there is no comparison of USGI and LMGI for these therapies.
Shoulder
Glenohumeral joint
Accuracy
In a study by Cunnington et al., injections of steroid and contrast agent were given in 184 different joints randomised for LMGIs and USGIs. The accuracy was assessed by radiography of the injected joint, which localised the contrast agent in the injectate. The accuracy of the glenohumeral joint injections reported for USGIs was 63% and for LMGIs was 40% (p = NS) . It should be noted that in this study, USGI was performed using the posterior approach by a relatively inexperienced ultrasonographer with one year of practical US experience, and 40% accuracy was achieved by rheumatologists with a median of 14 years of clinical experience in joint injection. It would be expected that the accuracy of a skilled ultrasonographer interventionist would be much higher, and the sample size in this study was small making statistical analysis difficult.
There are eight clinical studies that investigated the use of LMGIs. Three of these used magnetic resonance (MR) arthrography for accurate injection verification, four used contrast medium and fluoroscopy and one used arthroscopy. According to these reports, the mean accuracy of LMGIs to the glenohumeral joint was 67% (variation 10–99%) . In a paper by Tobola et al. , three different injection approaches were investigated (anterior, posterior or supraclavicular), and statistically significant differences were absent between the three approaches. However, the anterior route was the most accurate and was independent of the experience level of the provider.
There are five other studies that have investigated the accuracy of USGIs before MR arthrography . The mean reported accuracy of USGIs to the glenohumeral joint were 96% (variation 92–100%). One of these studies compared US- and fluoroscopy-guided techniques and noted that the first attempt was successful in 72% of the injections in the fluoroscopy-guided group and 94% in the US-guided group .
In a cadaveric study with 80 shoulder specimens, the accuracy of USGIs and LMGIs were 92.5% and 72.5% (p = 0.025), respectively . Two operators injected radiopaque contrast through a posterior approach. After the injections, radiographs of the specimens were obtained to assess the accuracy of the injections.
There are three cadaveric studies that investigated LMGIs to the glenohumeral joint . The mean accuracy was 94% (variation 91–96%). One study used superior injection approach and noted that caution for the likelihood of penetrating the long head of biceps tendon should be considered .
These findings provide evidence that USGIs to the glenohumeral joint are more accurate than LMGIs ( Table 1 ).
| LMGI (%) | USGI (%) | |
|---|---|---|
| Comparative clinical study (1) | 40 | 63 |
| Clinical LMGIs studies (8) | 67 | – |
| Clinical USGIs studies (5) | – | 96 |
| Comparative cadaveric study (1) | 72.5 | 92.5 |
| Cadaveric LMGIs studies (3) | 94 | – |
Efficacy
There is a single comparative study of LMGIs and USGIs that investigated the efficacy of glenohumeral joint injections using these techniques. In this trial for patients with adhesive capsulitis, the US group had a lower reduction in pain VAS; however, no significant difference was observed between groups for function at 6 weeks post injection . In Cunnington’s study, there was no significant difference for outcome variables between USGIs and LMGIs when all the joints were grouped together. The results for single joints were not reported. However, there was a greater improvement in the VAS score for function in the accurate injection group . A recent Cochrane review reported no significant improvement in efficacy with USGIs to the shoulder . However, the authors did not perform the analysis on the basis of the exact injection location. In addition, there were fewer side effects in the USGI groups.
Presently, there is not enough data to conclude that USGIs are more efficacious than LMGIs in the treatment of glenohumeral joint diseases.
Subacromial space
Accuracy
In assessing the studies on injection into the subacromial space, it should be noted that different studies have used different volumes of injectate, and a larger volume may aid the accuracy but leads to more tissue damage. Two clinical studies have compared LMGIs and USGIs using MR arthrography for verifying the success of the intervention. Rutten et al. reported that both methods were 100% accurate , and Dogu et al. found that accurate injections were performed in 65% in the USGI group and 70% in the LMGI group .
Four clinical studies have investigated the accuracy of LMGIs into the subacromial space . One used MR arthrography and the others used fluoroscopy plus contrast media for verification. The mean accuracy was 61% (variation 29–91%).
Three cadaveric studies have examined the accuracy of LMGIs into the subacromial space . The mean accuracy was 81% (variation 70–91%). Thus, the accuracy of LMGIs into the subacromial space is sub-optimal and depends on the operator.
There is not enough data to confirm that USGIs are more accurate than LMGIs, and further research is required to clarify this ( Table 2 ).
| LMGI (%) | USGI (%) | |
|---|---|---|
| Comparative clinical studies (2) | 85 | 82.5 |
| Clinical LMGIs studies (4) | 61 | – |
| Cadaveric LMGIs studies (3) | 95 | – |
Efficacy
Five studies have evaluated the efficacy of USGIs and LMGIs into the subacromial space, and they all showed better outcomes for the USGI procedures in the short term . However, two additional studies found that there is no difference in efficacy between accurate and inaccurate injections .
According to the literature, USGIs into the subacromial space produce a significantly greater improvement in terms of pain reduction and functional gain than LMGIs in chronic shoulder pain, at least in short term.
Biceps tendon sheath, acromioclavicular joint and sternoclavicular joint
Accuracy and efficacy
Injection into the tendon sheath of the long head of the biceps brachii can be more accurately performed by USGI than that by LMGI according to a single study . USGIs showed 87% accuracy, whereas LMGIs showed only 27% accuracy. Computed tomography (CT) with contrast agent was used for the verification of the injection; however, the efficacy of this was not reported.
In one clinical study, the accuracy of LMGIs to the acromioclavicular joint was 43%. This was verified using radiographic contrast material . There are no papers reporting the efficacy of acromioclavicular injections.
The exact position of USGI and LMGI into the acromioclavicular joint has been examined in three cadaver studies . The mean accuracies reported were 96% (variation 90–100%) for USGIs and 62% (variation 40–75%) for LMGIs. In a cadaveric study, Partington and Broome found that LMGIs were successful in 67% of the acromioclavicular injections .
It can be concluded that LMGIs to the biceps tendon sheath and acromioclavicular joint are sub-optimal and USGIs are more accurate; however, efficacy studies for this are missing.
The only study evaluating injection accuracy into the sternoclavicular joint used LMGI approach with a reported accuracy of 78% . This was a cadaveric study ( Table 3 ).
| LMGI (%) | USGI (%) | |
|---|---|---|
| Biceps tendon sheath: comparative clinical study (1) | 27 | 87 |
| Acromioclavicular joint: clinical LMGIs study (1) | 43 | – |
| Acromioclavicular joint: comparative cadaveric studies (3) | 62 | 96 |
| Acromioclavicular joint: cadaveric LMGIs study (1) | 67 | – |
| Sternoclavicular joint: cadaveric LMGIs study (1) | 78 | – |
Advanced procedures of the shoulder
Some procedures in the shoulder area require exact imaging guidance. Percutaneous US-guided fenestration and aspiration of calcific tendinosis is an example . It was originally done as a fluoroscopic procedure; however, US has largely replaced fluoroscopy because of its excellent safety profile and clinical efficacy. In the shoulder region, nerve blocks can be performed using US guidance (e.g. suprascapular and dorsal scapular nerve blocks). US may be used to diagnose and aspirate a ganglion cyst compressing the nerve in the suprascapular or spinoglenoid notch. Other injectable therapies such as platelet-rich plasma are now available for rotator cuff tendinopathy; however, further studies are required to clarify their efficacy, and there is no comparison of USGI and LMGI for these therapies.
Elbow
Elbow joint: accuracy and efficacy
Two studies have compared the accuracies of LMGIs and USGIs to the elbow joint. The study by Cunnington reported accuracies of 64% and 91% for LMGI and USGI, respectively (p = 0.1). Kim reported an accuracy of 100% using USGI and 77.5% using LMGI by posterior (olecranon) approach. They verified the injection accuracy using the fluoroscopic contrast method in 80 patients .
Two studies have evaluated LMGI accuracy to the elbow joint, and the mean accuracy was 92% (variation 84–100%) . Both the studies used radiographic contrast media to verify the success of the intra-articular injection. Of these, only Lopes et al. reported the efficacy and noted significant improvement in the VAS for pain at rest and during movement, VAS for oedema and morning stiffness .
It appears that LMGIs and USGIs to the elbow joint do not differ much in accuracy, although the number of studies is small ( Table 4 ).
| LMGI (%) | USGI (%) | |
|---|---|---|
| Clinical comparative studies (2) | 71 | 96 |
| Clinical LMGIs studies (2) | 92 | – |
Lateral epicondylitis and advanced procedures around the elbow
Lateral epicondylitis is a common condition in which different types of treatments have been investigated with varying results. There are no studies of the accuracy of LMGIs and USGIs in this area. The definition of accuracy is also complex given that some injection approaches target the surface of the common extensor origin, whereas others aim to penetrate and disrupt the tendon structure. US scanning plays an important role in the differential diagnosis of lateral elbow pain given that the CEO, lateral collateral ligament and humeroradial joint are so closely apposed.
Different types of US-guided procedures such as needle tendon fenestration and platelet-rich plasma injections have been and are to be investigated in the treatment of lateral epicondylitis.
US guidance has an important role in performing nerve blocks at the elbow level such as blocking of the posterior interosseous nerve and radial nerve at the spiral groove. Ganglion cysts compressing nerves can be diagnosed using US and also aspirated under US guidance.
Wrist and hand
Wrist and hand joints: accuracy
Cunnington et al. achieved 79% and 75% accuracy using USGI and LMGI, respectively, in the wrist joint (p = 0.8) . In two studies, LMGIs to the wrist joint resulted in accuracies of 97% and 50% . In these three papers, the radiographic contrast medium method was used for verifying the success of intra-articular injection. Choudur et al. used US guidance in MR arthrography and found that 99% were successfully injected into the wrist .
Smith et al. studied the accuracy of USGIs and LMGIs to the scaphotrapeziotrapezoid joint . This cadaver study showed that USGIs were 100% accurate, whereas LMGIs were only 80% accurate. There are two additional cadaver studies examining the accuracies of USGIs to the trapeziometacarpal and distal radioulnar joints, and the accuracies were 94% and 100%, respectively .
A single study compared the accuracy of LMGIs and USGIs to PIP and MCP joints . The needle position was intra-articular in 50% of PIP joints and 80% of MCP joints using LMGI, and the corresponding figures for USGIs were 92% and 100%. In this study, visualisation of the needle tip within the joint space and distension of the joint capsule following steroid injection defined intra-articular positioning of the needle.
The accuracies of USGIs to the wrist and hand are overall higher than LMGIs, but because of the paucity of studies, further research is required to make definite conclusions ( Table 5 ).
| LMGI (%) | USGI (%) | |
|---|---|---|
| Wrist joint: clinical comparative study (1) | 75 | 79 |
| PIP and MCP joints: clinical comparative study (1) | 65 | 96 |
| Wrist joint: clinical LMGIs studies (2) | 74 | – |
| Wrist joint: clinical USGIs study (1) | – | 99 |
| Scaphotrapeziotrapezoid joint: comparative cadaveric study (1) | 80 | 100 |
| Trapeziometacarpal joint: cadaveric USGIs study (1) | – | 94 |
| Distal radioulnar joint: cadaveric USGIs study (1) | – | 100 |
Related posts:
Imaging in rheumatic and musculoskeletal conditions
Imaging in the diagnosis and management of axial spondyloarthritis
Imaging of osteoarthritis (OA): What is new?
Imaging of vasculitis: State of the art
Imaging of regional pain syndromes; from syndromes to conditions using imaging?
Imaging in paediatric rheumatology: Is it time for imaging?
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
