Tips and tricks in ultrasound-guided musculoskeletal interventional procedures

The clinical indications for US-guided interventions are many and varied, and the specific clinical context of each patient should be taken in account prior to embarking on any interventional procedure. The most common injections are performed using a combination of local anesthetic and either a particulate or water-soluble steroid preparation to address symptoms of pain or swelling, usually localized to a joint (arthritis), bursa (bursitis), tendon sheath (tenosynovitis or tendinosis), or nerve (neuritis/neuroma). If more conservative measures exist for the management of patient symptoms, these should be attempted prior to entertaining an invasive procedure^(3–5). In many cases, the targeted approach of an US-guided injection can be beneficial to confirm accuracy and may improve patient experience due to increased ease of accurate needle placement. Sonographic guidance should also be considered if a prior blind injection has not achieved the desired clinical effect, as it may indicate an inaccurate or suboptimal injection. Evidence for the effectiveness of the most common injections in the MSK system has been well documented in the literature^(3,6–8).

There are also potential absolute and relative contraindications to US-guided injections that must be addressed. Absolute contraindications are few, but active septic arthritis, bursitis, or tenosynovitis contraindicates steroid injection into the relevant space. Additionally, overlying superficial soft tissue infection should not be traversed with a needle en route to an injection target due to the risk of seeding that compartment. If an overlying cellulitis is encountered in the setting of a proposed intra-articular intervention, an alternate route should be sought that avoids the contaminated soft tissues. In cases of severe allergy to a particular injectate, a suitable alternative should be utilized⁽⁷⁾.

Relative contraindications will differ based on the management context of individual patients as well as the complement of treatment options available in each situation. Acute subchondral insufficiency fractures, most common in the femoral head, have been considered a relative contraindication due to the theoretical risk of worsening fracture and ischemic effects to the subchondral bone promoted by local steroid effects, increasing the likelihood of osteonecrosis⁽⁹⁾. Indeed, a causal relationship between steroid injection and subchondral insufficiency fracture and osteonecrosis has been proposed but never definitively proven^(10,11). Non-severe allergic reactions should be managed on a case-by-case basis, and appropriate pharmacologic modifications or pretreatment regimens can be utilized should the procedure be absolutely clinically necessary.

A common clinical context for patients receiving intra-articular steroid injections is pain management of arthritis, to avoid or at least delay definitive surgical management, namely arthroplasty. However, if a surgery is planned following steroid injection, the surgeon’s preference should be discussed, as evidence suggests that intra-articular steroid injection increases risk of prosthetic joint infection if performed within three months of surgery⁽¹²⁾.

An exhaustive list of reported complications is not presented here, but informed consent should be tailored to each patient, including both common and uncommon risks, as well as any risks relevant to the clinical context and pharmaceuticals used, for example skin depigmentation or soft tissue atrophy in the setting of a superficial particulate steroid injections⁽¹³⁾. Systemic effects of corticosteroids are well established and relatively common, including hyperglycemia, especially in those with diabetes mellitus, and an immediate or delayed flushing reaction that is possible within the first 24–48 hours of the injection⁽¹⁴⁾. Non-target injection and the subsequent potential complications associated with them should be mentioned, but it should also be explained that these are much less likely to occur in US-guided procedures compared to blind or landmark-guided procedures, as targeted injection will be confirmed in real-time^(8,15). An additional benefit of US’s precision compared to blind injections is that lower volumes and doses of an anesthetic are required to produce the same clinical effect, thus minimizing potential local soft tissue complications related to high doses of anesthetics⁽¹⁶⁾.

The most important preventable complication for any US-guided percutaneous MSK intervention is iatrogenic infection, including septic arthritis, which can be devastating if not promptly recognized and treated. Strict adherence to universal precautions, aseptic technique, and precise needle placement are critical to minimizing infection risk⁽¹⁷⁾.

Many US imaging systems are equipped with a range of interchangeable transducers, and appropriate selection is critical to a successful US-guided intervention. The optimal transducer for a given injection is mainly determined by the footprint size and frequency for the given size and depth of the desired target. A small footprint size is ideal for small injection targets with small or otherwise challenging surface anatomy (for example, a lesser toe interphalangeal joint or the peroneal tendons as they curve behind the lateral malleolus). The smaller footprint size will allow the relevant surface anatomy to accommodate both the transducer and the needle. Higher frequency transducers will optimally resolve more superficial structures in greater soft tissue detail but are suboptimal for deep or large injection targets. The latter are more optimally visualized by low-frequency transducers, which usually have a larger footprint and display a broader region of deeper anatomy, such as the hip joint (Fig. 1). One potential pitfall is that the high-frequency transducer may not penetrate thick skin or subcutaneous tissues, as can be seen in the sole of the foot or the palm of the hand, and a lower frequency transducer may be necessary even if the target tissue depth is relatively superficial.

Fig. 1.

The universal precautions exercised in accordance with individual institutional policies and common clinical practice guidelines for all interventional procedures should be no different for US-guided interventions. Indeed, US-guided interventions are not known to be associated with an increased risk of infectious complications⁽¹⁷⁾. However, the transducer utilized for real-time guidance is neither disposable nor single-use as is the case for all other instruments used in the procedure. Commonly, the transducer is covered with a single-use sterile plastic barrier, most often a condom-type cover. An alternative approach is to leave the transducer footprint uncovered and drape the sides of the transducer and its cord with a sterile adhesive drape. The footprint can be sterilized with a chlorhexidine, Betadine (Povidone-iodine), or a similar product. The advantage of this approach is that the transducer footprint is unencumbered by gel and plastic covering, significantly improving the guidance image quality. This approach to US transducer preparation has been shown to be safe from an infection control standpoint, with one study identifying no confirmed infectious complications in over 6,000 injections performed at a single center⁽¹⁸⁾.

Geometry is important to the success of an US-guided injection, defining the correct needle trajectory by taking into consideration the depth of the target, distance of the target from the needle entry site, and distance of the needle entry site from the edge of the transducer footprint⁽¹⁴⁾.

In the so-called “long axis” or “in-plane” injection approach, the transducer is orientated such that the needle will be visualized along its length, with the length and tip of the needle kept in view at all times (Fig. 2 A). The so-called “short axis” or “out-of-plane” approach is achieved with the transducer positioned perpendicular to the trajectory of the needle, such that the needle will appear as a single point of reflection in short axis, often with a reverberation artifact deep to it, when it passes below the transducer (Fig. 2 B). While either method is acceptable, the surface anatomy or geometry of the injection site may necessitate favoring one approach over another⁽¹⁹⁾. Generally, small and narrow superficial joints, such as the interphalangeal joints of the finger or toe, are best accessed by the short axis approach, so the needle can be guided steeply into the center of the joint in contrast to a large, wide and deep joint that lends itself well to the more shallow approach of the long axis or in-plane technique (e.g. the hip or shoulder joints).

Fig. 2.

Fortuitously, needles are excellent specular reflectors^(20,21), making them conspicuous on US images, but the more acute the angle of insonation from the transducer upon the needle, the less signal is detected and rendered on the image. Therefore, the best visualization is afforded by in-plane views, where the sound beam reflects off the needle at a ninety-degree angle (Fig. 3). Heel-toe maneuvering of the transducer or use of sonographic systems with beam steering capabilities may be useful to maximize the reflected signal from the needle. Specialized reflective needles with surface alterations or echogenic tip coatings designed to accentuate the reflection of the needle may be useful but are generally not necessary for most US-guided interventions^(20–22).

Fig. 3.

Autologous PRP has been touted as a useful biotherapeutic in countless clinical contexts, with ever-expanding therapeutic uses. PRP is rich in growth factors, released upon platelet activation, that have been shown to be vital in stimulating and propagating tissue regeneration, such as healing of tendinopathy, tendon or ligament tears, and bone healing in the context of chondromalacia and osteoarthritis. In general, the goal in PRP preparation is to maximize platelet concentration in the plasma layer above the physiologic concentration by 3–5 times, ideally around 1,000,000 platelets per microliter of injectate. Refrigerated centrifuges, if available, should be used to maximize platelet concentration; this technique has been shown to double the concentration obtained from autologous blood samples⁽³⁵⁾. Many local tissue factors are hypothesized to affect the efficacy of PRP injections. Intratendinous PRP injection is often accompanied by dry needling to promote bleeding and angiogenesis⁽³⁶⁾. If avoidable, local anesthetic, such as lidocaine, should not be injected directly at the site of PRP delivery, as it has been shown to decrease platelet aggregation response and theoretically may diminish the growth factor and cytokine release thought to initiate a therapeutic benefit⁽³⁷⁾. No consensus exists on the ideal needle caliber for the delivery of PRP, and may be dictated by the volume of PRP injectate and the site of injection. Specifically, thin needle gauge has not been shown to disrupt platelet viability or activity.

US-guided injection or arthrocentesis of both large joints, such as the shoulder (glenohumeral) and hip (femoroacetabular) joints as well as the small joints, such as tarsometatarsal or metatarsophalangeal joints, are commonplace. Although many intra-articular injections are feasible without direct visualization, using surface landmarks or palpation, US offers excellent real-time visualization to ensure accurate positioning of the needle tip in the joint throughout an injection, while minimizing trauma to adjacent structures. Confirmation of target injections and increased confidence of the needle position on the part of the practitioner additionally benefit patient comfort and potentially improve clinical outcomes.Each joint presents a slightly different scenario for needle approach and patient positioning, which will not be reviewed exhaustively here. But the principles guiding successful sonographic injections in each of these joints are similar and will be discussed here in the context of the glenohumeral joint. We will provide several practical tips for successful US-guided intra-articular injections.

The glenohumeral joint is easily visualized and injected with the assistance of US (Fig. 4, Fig. 5 A). If desired, diagnostic images of the joint, rotator cuff muscles and tendons, and subacromial/subdeltoid bursa can be obtained at the time of US-guided intervention. Findings such as a rotator cuff tear, osteoarthritis, bursitis, and calcific tendinitis, to name a few, may be revelatory regarding the etiology of a patient’s symptoms, and may assist in appropriate management of the patient, even informing the type of percutaneous intervention that may be required.

Fig. 4.

Fig. 5.

Most commonly, US-guided glenohumeral joint injections are performed with the patient lying in an oblique prone, lateral decubitus position, with the posterior aspect of the targeted shoulder exposed for injection. This so-called “posterior approach” to the glenohumeral joint is safe, with no major nerves or vasculature along the planned needle trajectory⁽³⁸⁾. A major advantage of US guidance, over other imaging modalities and blind injections, is the ability to visualize neurovascular structures accurately in real-time. Patients may also prefer the posterior approach, as the practitioner and any instruments are not in the patient’s direct sightline during the procedure.

With the transducer footprint placed longitudinally with respect to the glenohumeral joint, the joint will be imaged as depicted (Fig. 5 A). Because the needle trajectory traverses the infraspinatus tendon and posterior joint capsule, visualization can often be improved with further adduction and internal rotation of the ipsilateral arm. We prefer orienting the injection such that the interventionalist faces the console screen to easily view the real-time guidance images as they proceed. To facilitate this, an approach can be planned either from medial to lateral or from lateral to medial (Fig. 4, Fig. 5 B), both of which are reasonable trajectories.

Infiltration of the subcutaneous soft tissues with a thin (for example, 25 gauge) needle is followed by advancement of a longer, lower gauge needle depending on the required depth of the injection. Many joints can be accessed with a single 1.5-inch thin-gauge needle. Care should be taken to anesthetize at the puncture site in the immediate superficial subcutaneous adipose tissue, and deeper anesthesia should also be provided, noting that the joint capsule is richly innervated, so pericapsular anesthesia may significantly improve patient comfort. In the case of glenohumeral joint injections, additional anesthesia instillation within the subacromial/subdeltoid bursa, which is innervated and can be an independent pain generator, may also enhance comfort.

Once the needle reaches an intra-articular location, the desired medication can be administered. Real-time visualization is important to ensure an accurately targeted injection. The particulate nature of many steroid formulations can be used to the practitioner’s advantage in that the particles are small reflectors, which may create a “contrast effect” in the joint (Video 1)⁽³⁹⁾. In smaller joints, especially using an out-of-plane approach, the “waterfall” appearance may be seen (Video 2), confirming an intra-articular injection. Otherwise, progressive distension of the joint capsule is a reliable sign (Video 3). Although injection of gas bubbles should be avoided in many injections (e.g. intra-articular contrast injections for magnetic resonance arthrograms) occasionally, small gas bubbles accompanying a test injection may either be intentionally or accidentally introduced, collecting anti-dependently, deep to the joint capsule, producing a typical ringdown artifact that can be recognized and confirm adequate subcapsular positioning of the needle tip (Fig. 6 A). Use of color or power Doppler has also been described as a reliable method for imaging the jet of injectate administered deep to the joint capsule (Fig. 6 B and C)⁽⁴⁰⁾.

Fig. 6.

Bursal injections are commonly performed juxta-articular interventions to address bursitis, a painful inflammatory condition that can be confused for intra-articular pathology such as arthritis or other structural pathology around the joint. Commonly injected bursae include the subacromial/subdeltoid bursa and the scapulothoracic bursa at the shoulder, and the greater trochanteric and iliopsoas bursae at the hip. Pre-injection sonography of these potential spaces can confirm evidence of bursitis, usually indicated by bursal thickening, fluid distension, proliferative synovium, and hypervascularity (Fig. 7 A). US guidance ensures accurate injection of the bursal space, which can be a very narrow target, difficult to accurately inject blindly^41,42. Nevertheless, these are safe injections that can be visualized well by US, given the generally superficial location of the bursae. Intra-bursal injection is confirmed by real-time visualization of bursal distension by the injectate (Fig. 7 B, Video 4).

Fig. 7.

US-guided aspiration of calcium deposits in the context of calcific tendinopathy, bursitis, and periarthritis is known as barbotage⁽⁴³⁾. Most commonly occurring along the bursal-sided fibers of the rotator cuff tendons, hydroxyapatite calcium deposits can occur throughout the MSK system, also appearing regularly in the peritrochanteric region of the proximal femur, involving the gluteus minimus/medius tendon insertions, and about the joint capsules of the knee or in the hands and feet (so-called calcific periarthritis). Both single-needle and dual-needle lavage techniques have been described^(43–46). At our institution, we find the single-needle technique suffices in the majority of situations. Barbotage may not be possible when the calcium is small or fragmented; in such cases, needle fenestration and mechanical disruption followed by therapeutic bursal injection may suffice^(47,48). For single-needle barbotage, the deposit can be accessed with an 18–20 gauge needle, utilizing a delicate single pass to avoid rupturing the pericalcific pseudocapsule containing the calcification⁽⁴⁹⁾ (Fig. 8). Pulsed lavage with lidocaine and saline disrupts the organized deposit, allowing aspiration of calcium into the lavage syringe, and creating a characteristic “fish-mouth” appearance on dynamic US imaging (Video 5). If intraosseous migration of the calcium has occurred, barbotage will be a less effective therapy⁽⁵⁰⁾. It is essential that barbotage procedures of calcium occurring near or in a bursa (most often the subacromial/subdeltoid or trochanteric bursae) be completed with an intrabursal steroid injection as the final step in the procedure. The injection will offset and alleviate any symptoms associated with reactive chemical bursitis caused by the mobilized calcium during the procedure^(43,44).

Fig. 8.

Ischiofemoral impingement is a commonly encountered cause of buttock and posterior thigh pain that can be difficult to diagnose and for which many treatments exist, although no single therapeutic approach has been shown to be the most effective⁽⁵¹⁾. US-guided ischiofemoral interval injections are among the non-surgical therapeutic options for patients suffering from buttock pain or sciatica related to this entity. US guidance for injection into this deep subcutaneous space is vital to prevent iatrogenic neurovascular injury or non-target injections. Recognition of the anatomic landmarks on US is vital (Fig. 9 A–C) and will aid in accurately targeting the injection. Potential therapeutic maneuvers include steroid and anesthetic injections into the quadratus femoris muscle or the perisciatic fat⁽⁵²⁾. Additionally, ischiofemoral interval prolotherapy or intramuscular botulinum toxin injection of the quadratus femoris or piriformis muscles (Fig. 10 A and B) have been reported with positive, though temporary, therapeutic outcomes^(51,53,54).

Fig. 9.

Fig. 10.

Ultrasound-guided dry needling (tenotomy) of tendons has been utilized to treat chronic painful tendinopathy and chronic partial tendon tears at various anatomic sites. The technique involves serial fenestration of the affected tendon with ultrasound guidance. Needle gauges ranging from 22- to 25-gauge have been reported to have good outcomes in the literature, with serial fenestration performed by taking multiple (usually 20 or more) passes through the most abnormal appearing portion of the tendon, orienting the needle parallel to the tendon fibers. Dry needling may be used in isolation to promote local angiogenesis, fibroblastic proliferation, and collagenization within the fenestrated tissue^(55–57).

US affords excellent soft tissue resolution, especially for superficial structures such as peripheral nerves in the extremities. Therefore, the peripheral nerves are optimal targets for US-guided diagnostic and therapeutic maneuvers. Indeed, US has long been an important guidance technique for performing peripheral nerve blocks in many locations throughout the body across a wide variety of clinical specialties^(58,59).

The diagnostic utility of US should not be ignored when performing an US-guided perineural injection for diagnostic or interventional purposes. The nerve should be scanned in its entirety to recognize and localize any imaging abnormality of the nerve such as extrinsic compression, overt neuritis, or a mass like a neuroma or peripheral nerve sheath tumor^(23,60,61).

To confirm a particular nerve or imaging abnormality of a nerve as a symptom generator, diagnostic perineural injections can be performed utilizing a long-acting local anesthetic agent such as bupivacaine, which will produce a nerve block of the desired nerve distribution. Patients can then record their symptoms and compare them to baseline, revealing the extent of response to the nerve block, often informing the need or utility of subsequent intervention upon the nerve in question.

Neuritis is commonly treated with a combination of steroid and local anesthetic injected in the perineural fat at the site of maximal visual abnormality of the nerve or at the site of presumed nerve compression or irritation^(23,60). When approaching the nerve, care should be taken to avoid direct injury with the needle or intraneural injection of steroid or anesthetic. In our experience, an in-plane approach with the needle, while visualizing the nerve in short axis, allows optimal visualization of the outer epineurium and nerve fascicles. Following a low-volume test injection with 1% lidocaine to achieve initial hydrodissection of the nerve, perineural spread of the injectate can be visualized in both short and long axis (Video 6 and Video 7, Fig. 11) after which the desired medication(s) can be administered into this potential space. For therapeutic perineural injections, we prefer to use a long-acting anesthetic, such as 0.5% ropivacaine or 0.75% bupivacaine, and a rapidly absorbed corticosteroid, such as betamethasone. The rapid systemic absorption of betamethasone is preferred to avoid long dwell time of the steroid in the local soft tissue, which may impair the healing reaction if a surgical intervention is planned. Hydrodissection with local anesthetic in the setting of nerve impingement or perineural scar entrapment has been described in association with positive clinical outcomes in such clinical scenarios as piriformis syndrome and carpal tunnel syndrome^(52,62). For large-volume perineural injections, for example along the sciatic nerve, we utilize sterile saline or a dilute anesthetic-saline solution to avoid a strong sciatic nerve block.

Fig. 11.

Large-volume hydrodissection of peripheral nerves is frequently utilized to address impingement or entrapment neuropathy related to soft tissue entrapment of a peripheral nerve by a scar or other anatomic features such as muscle or fascia⁽⁶³⁾. Release of the nerve by fluid dissection is thought to have both a mechanical consequence in the setting of entrapment and vascular consequences improving perineural venolymphatic flow by increasing potential perineural space. Hydrodissection is best performed utilizing ultrasound guidance for precise, careful placement of injectate about the nerve, ideally allowing circumferential separation of the nerve from surrounding structures. The hydrodissection should be performed at the site of maximal impingement or entrapment, if it is visible on ultrasound and percutaneously accessible. A 25-gauge or 27-gauge needle is best suited for small, superficial peripheral nerves, whereas a 22-gauge needle may be necessary for larger and deeper nerves such as the sciatic nerve. Many injectates have been utilized and reported to show benefit, the most popular being a large-volume injection of normal saline or local anesthetic diluted by normal saline. Non-dilute anesthetic such as lidocaine will produce a dense nerve block, which may be undesirable. A water-5% dextrose solution has shown superiority to normal saline alone for median nerve hydrodissection in patients with carpal tunnel syndrome⁽⁶⁴⁾.

These minimally invasive procedures can be useful alternatives to potentially morbid surgical interventions, or at least serve to confirm a nerve or specific site along the nerve as the symptom generator to be addressed by subsequent definitive therapy^(52,23).