How frequently is ultrasound used for guiding rheumatology procedures?

The first comprehensive report on such application comes from a general survey performed in 1999 . In this survey, focussed on the practice of ultrasound by rheumatologists, 30% of respondents found this technique a useful guiding tool especially for hip, shoulder and knee joint injections. Despite the possible bias related to such a questionnaire, distributed during an international meeting and for which only people interested in the field are likely to respond, guiding procedures seemed to be an essential part of the rheumatology ultrasound practice. Later surveys confirmed these findings . Because the number of rheumatologists performing ultrasound is growing over time, as well as the number of publications into the field, a concomitant increase in the number of rheumatologists performing ultrasound-guided interventions would have been expected. However a recent survey conducted in Europe among members of the European League Against Rheumatism (EULAR), specifically focussed on the practice of ultrasound-guided injections, showed that only 10% of the responding sonographers used routinely ultrasound for performing local procedures . The main barrier seemed to be the lack of specific training and of clear procedures on when and how to use ultrasound to guide injections. In particular the absence of clear guidelines concerning the mode of sterilisation/disinfection was a major concern. On the contrary sonographers having answered positively reported an increased facility of the act as well as a major confidence in their accuracy. The conclusion of the authors was the need of a clear standardisation and of a specific training. Specific training is actually missed because ultrasound-guided injections represent just a minor aspect during the teaching courses.

How accurate is the ultrasound-guided modality?

Despite the lack of a specific training there are several publications on the use of ultrasound to guide injections. Most of these studies have addressed the success rate of ultrasound-guided versus conventional arthrocentesis as well as the accuracy of ultrasound-guided versus conventional joint injection, while only a relatively small number of studies have addressed outcomes . The results of these studies are sometimes contradictory. A randomised comparative study evaluating short-term response to blind injection versus ultrasound-guided glucocorticoid injection in patients with painful shoulder suggested that ultrasound-guided injections should be indicated in patients with poor response to previous blind injection to ensure accurate medication placement in order to improve therapeutic effectiveness . The same gain in efficacy was demonstrated in a recent meta-analysis of an individual patient showing a greater short-term improvement in wrist pain with ultrasound-guided glucocorticoid injections as compared to clinical-guided injections . Another systematic review evaluating the efficacy of ultrasound-guided injections compared with injections using palpation/anatomic landmarks showed greater accuracy in the ultrasound-guided group, independent of anatomic site, and greater short-term (<6 weeks) improvement. Long-term outcomes were not proven to be different between ultrasound- or anatomic-guided groups. Also foot/ankle and wrist/hand, small joint space injections showed a greater accuracy with ultrasound-guided injections contrasted to larger joint spaces. In large joints, such as knee and shoulder, the improvement in function, pain, intensity and range of motion seemed independent of the injection technique. This observation was also confirmed by a Cochrane review on randomised or semi-randomised control trials . The objective was to assess whether in patients with shoulder pain ultrasound-guided glucocorticoid injections improve patient-relevant outcomes as compared to landmark-guided or systemic intramuscular injections. The conclusion of the authors was that despite moderate efficacy evidence from five trials and an evident increase in the accuracy of the ultrasound-guided act, no clear improvement was demonstrated by using guided injections with respect to blinded injections. Additional studies are needed to evaluate possible superiority of the efficacy.

How accurate is the ultrasound-guided modality?

Despite the lack of a specific training there are several publications on the use of ultrasound to guide injections. Most of these studies have addressed the success rate of ultrasound-guided versus conventional arthrocentesis as well as the accuracy of ultrasound-guided versus conventional joint injection, while only a relatively small number of studies have addressed outcomes . The results of these studies are sometimes contradictory. A randomised comparative study evaluating short-term response to blind injection versus ultrasound-guided glucocorticoid injection in patients with painful shoulder suggested that ultrasound-guided injections should be indicated in patients with poor response to previous blind injection to ensure accurate medication placement in order to improve therapeutic effectiveness . The same gain in efficacy was demonstrated in a recent meta-analysis of an individual patient showing a greater short-term improvement in wrist pain with ultrasound-guided glucocorticoid injections as compared to clinical-guided injections . Another systematic review evaluating the efficacy of ultrasound-guided injections compared with injections using palpation/anatomic landmarks showed greater accuracy in the ultrasound-guided group, independent of anatomic site, and greater short-term (<6 weeks) improvement. Long-term outcomes were not proven to be different between ultrasound- or anatomic-guided groups. Also foot/ankle and wrist/hand, small joint space injections showed a greater accuracy with ultrasound-guided injections contrasted to larger joint spaces. In large joints, such as knee and shoulder, the improvement in function, pain, intensity and range of motion seemed independent of the injection technique. This observation was also confirmed by a Cochrane review on randomised or semi-randomised control trials . The objective was to assess whether in patients with shoulder pain ultrasound-guided glucocorticoid injections improve patient-relevant outcomes as compared to landmark-guided or systemic intramuscular injections. The conclusion of the authors was that despite moderate efficacy evidence from five trials and an evident increase in the accuracy of the ultrasound-guided act, no clear improvement was demonstrated by using guided injections with respect to blinded injections. Additional studies are needed to evaluate possible superiority of the efficacy.

How to perform an ultrasound-guided injection?

Generally, all ultrasound-guided interventions are performed by using one of two methods: indirect and direct visualisation. Indirect visualisation involves the performance of a pre-intervention ultrasound examination, which helps to identify the putative site and planning the best clinical approach. The direct visualisation method involves the performance of ultrasound-examination simultaneously with the local intervention, which allows the real-time visualisation of the target lesion along with the inserted instrument.

Larger amounts of fluids that localise superficially may be aspirated or injected without ultrasound guidance in clinically clear situations. This is the case, for example, in large knee joint effusions. Often, ultrasound may be performed before an injection for localising the area of interest. The area where the needle needs to be introduced can be marked. A paper clip that shows reverberation artefacts on the ultrasound image can help identifying the correct area for the needle to be introduced . Finally, the needle will be introduced without ultrasound guidance.

Other situations require ultrasound guidance for introducing and advancing the needle towards the area that shall be injected or aspirated. This applies particularly to deeper structures that are not palpable.

Approaches

Often the needle can be introduced some distance away from the probe. Thus it will not be necessary to use a sterile cover of sterile gel. In the case of superficially localised anatomic structure it may be necessary to introduce the needle closely to the ultrasound probe. Then the probe should be covered with a sterile probe cover or a sterile glove; sterile ultrasound gel needs to be put into the cover between the probe and the cover as well as on the skin.

The needle may be introduced either parallel to the probe or perpendicular to the probe. When the needle runs parallel to the probe it will be seen in full length. If it is advanced perpendicular to the probe the needle is only represented by a hyperechoic spot in the ultrasound image. The sonographer needs to either shift the probe or push the needle slightly forward or backward in order to make sure that the tip of the needle is directly in the area of interest.

Fig. 1 shows several approaches for the needle in relation to the probe . The first graph (1A) shows the approach with the best view on the needle. The needle runs parallel to the probe. The tip of the needle can be easily localised. Often anatomy does not allow introducing the needle parallel to the probe. Then, the needle is advanced angled to the probe (1B). The needle becomes less visible with increasing angles between the probe and the needle. Some ultrasound machines allow angling the direction of the image towards the needle for increasing its visibility. Usually the needle and the probe are held separately by the same examiner. Alternatively, one person can perform the ultrasound and the other person introduces the needle. Probes are available with a device to introduce the needle right next to the middle of the probe (1C). This can be also done freehand. Ultrasound beam direction and direction of the needle are similar in this case. Thus, only a spot is seen on the ultrasound image that represents the needle. This is also the case if the needle is introduced further away from the probe but perpendicular to the probe as described in the previous paragraph. In a few anatomical regions, as between the toes, the needle can be advanced towards the probe as shown in Fig. 1 D.

Approaches of ultrasound guided injections with possible positions of needles and probes.

The following text describes the approach to the most relevant anatomical structures for ultrasound-guided injections in rheumatology.

Shoulder

Ultrasound is particularly important with regard to injections in the shoulder region.

The following abnormalities most commonly occur in rheumatological practice in relation to ultrasound-guided injections:

• •
  

  subdeltoid bursitis,

  
  
• •
  

  tenosynovitis of the biceps tendon,

  
  
• •
  

  synovitis of the glenohumeroid joint with or without rotator cuff tear,

  
  
• •
  

  synovitis of the acromioclavicular joint,

  
  
• •
  

  synovitis of the sternoclavicular joint and

  
  
• •
  

  entrapment of the suprascapular nerve.

Each of these entities requires a different approach for ultrasound-guided injections. Ultrasound helps determine the region to be injected. Only large palpable bursitis that extends closely to the skin surface can be injected directly without ultrasound guidance with a high probability of reaching the target. The other pathologies occur more deeply. Thus they cannot be localised by palpation.

The subdeltoid bursa localises between the deltoid muscle and the rotator cuff. Glucocorticoid injections into that bursa are particularly helpful in calcifying tendinitis of the rotator cuff as the injected material will spread along the surface of the rotator cuff. In other conditions, the bursa is directly inflamed displaying an increased amount of fluid. The anterior transverse scan of the rotator cuff or of the proximal long biceps tendon is used to guide the injection. External rotation of the shoulder may increase the amount of fluid in the bursa. The injection needle is inserted just lateral to the probe in the antero-lateral aspect of the shoulder, about 1–2 cm below the needle ( Fig. 2 ). The needle is then advanced parallel to the probe into the bursa. The needle displays well in the ultrasound image because it is parallel to the probe. Even a bursa that is not extended may be injected successfully with ultrasound guidance as it directly localises anteriorly to the rotator cuff.

Transverse anterior scan of the shoulder. Injection of the subdeltoid bursa from lateral, parallel to the probe. The needle as seen as a hyperechoic line coming from the left.

Tenosynovitis or fluid collections of the tendon of the long head of the biceps muscle are common in shoulder disease. The compartments of the glenohumeral joint and the tendon sheath communicate with each other. Otherwise, tenosynovitis can exist independently of glenohumeral joint inflammation. The tendon sheath can be injected with the same approach that has been described above for the subdeltoid bursa.

Ultrasound can delineate the glenohumeral joint from the dorsal and from the axillary region. The joint can be accessed from the anterior side only in case of a complete subscapularis tendon tear. The axillary recess may not be injected because of many nerves, the axillary artery and vein that localise in this region, and because of hygienic issues. Therefore, in most patients, the only possible way of injecting the glenohumeral joint is from its posterior aspect. The posterior transverse scan provides a direct view on the joint space together with the posterior aspects of the glenoid labrum. In fact the probe is not positioned strictly dorsally but about 45° at the postero-lateral aspect of the shoulder. The maximum amount of effusion will be detected in external rotation with abduction of the arm . The needle can be introduced cranially to the probe in the level of the middle of the probe with an angle of about 10–20° heading caudally. With this approach the needle is not well visible. The length of the needle needs to be 2.5–4 cm. Alternatively, the glenohumeral joint can be accessed from the medial or lateral side. The medial approach is easier as the needle can be introduced with a 40–50° angle in relation to the skin surface and the ultrasound probe. The needle is better visible compared to the previous approach. However, it is again not perfectly visible as it is not parallel to the probe. The lateral approach is more difficult than the medial approach as the needle runs closely to the rotator cuff and the humeral head. Thus, only larger posterior effusions are accessible with the lateral approach. The needles need to be long, usually 5 cm and, in large patients, the needles need to be even longer in order to access the glenohumeral joint. In the case of a complete rotator cuff tear the glenohumeral joint can be accessed also anteriorly applying the anterior transverse scan of the shoulder applying the same approaches as previously described for the posterior approaches.

The suprascapular nerve and an adjacent small artery can be found more medially in a groove. The probe is moved medially from the scan that shows the glenohumeral joint and slightly cranially. Now the probe is positioned nearly completely dorsally. The approaches for the needle are similar to those described for the posterior aspect of the glenohumeral joint. The lateral approach is easier as the needle passes anatomical areas that localise more dorsally to the rotator cuff and the humeral head.

The acromioclavicular and sternoclavicular joints can be easily palpated. They may be either injected blindly after the ultrasound examination has confirmed synovitis or injected after marking the spot on the skin where the needle should be introduced. Alternatively, the joints may be reached with ultrasound-guided procedures by placing the probe longitudinally to the axis of the joint. This is done by the standard scan of the acromioclavicular joint that is transverse with respect to the body axis. The sternoclavicular joint is examined by ultrasound with a 45° oblique scan with the cranial aspect of the probe pointing laterally. When both bones that form the joints are visible on the ultrasound image the needle is introduced perpendicular to the probe. When the needle is under the probe, a hyperechoic spot is seen. Ultrasound-guided procedures of the acromioclavicular or sternoclavicular joints are particularly useful if smaller or larger amounts of fluid need to be aspirated.