Assessment of intrasynovial injection in horses by contrast‐enhanced ultrasonography using air bubbles created by agitation of solution

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2020 The Authors. Equine Veterinary Journal published by John Wiley & Sons Ltd on behalf of EVJ Ltd *The abstract is available in German in the Supporting Information section of the online version of this article Faculty of Veterinary Science, Leahurst Equine Hospital, University of Liverpool, Neston, UK


| INTRODUC TI ON
Intrasynovial injections are performed commonly in horses to diagnose and treat lameness. 1 The accuracy of intrasynovial injections ranges widely (32%-96%) and has only been reported for a limited number of joints, including the proximal and distal interphalangeal joints, metacarpophalangeal joint and distal tarsal joint. [1][2][3][4][5] Intrasynovial injections of smaller joints, particularly where effusion is not readily palpable, can be technically challenging and result in inaccurate injection. [1][2][3][4] Verification of accurate injection has traditionally relied on synovial fluid retrieval and ease of injection. 6 However, the validity of these methods has been questioned. 2 Increased accuracy of injection has been shown for a number of ultrasonography-guided injection techniques, where accuracy of injection was assessed by evaluating radiographic or computed tomographic images following injection of iodinated contrast, 7-11 but these techniques do not allow the operator to establish the accuracy of any given injection. In a study investigating the accuracy of injection techniques targeting the bicipital bursa (n = 12) using large volumes of air (15 mL) with radiopaque contrast (10 mL), ultrasonography predicted the location of the injectate as determined by radiography following all 12 injections. 12 Radiographic examination following injection of iodinated contrast is the only practical imaging modality verifying accuracy of intrasynovial injection in a clinical setting but is rarely undertaken. Therefore, the accuracy of a particular injection in a clinical setting often remains unknown. Confirming the accurate deposition of local anaesthetic or therapeutic agents enables the veterinary surgeon to interpret results of diagnostic analgesia appropriately and offers reassurance that any failure of treatment did not arise from extrasynovial injection. A clinically applicable technique for the determination of the location of injectate following an intended intrasynovial injection, as intrasynovial or extrasynovial would permit this reassurance. Such a technique has been described in humans by using ultrasonography imaging following injection with agitated air-glucocorticoid-saline admixture. [13][14][15] A high sensitivity of such a test would be desirable as this would allow for identification of unsuccessful intrasynovial injections, and thus would permit repeat injection if appropriate. A high specificity would be useful for verifying successful intrasynovial injections with positive results providing the clinician with confidence that the injection was intrasynovial.
Contrast-enhanced ultrasonography (CEUS) is an ultrasonography imaging technique where traditional diagnostic imaging ultrasonography is used with a contrast agent, administered either locally or systemically, to improve the interpretation of the ultrasonography. 16 Contrast echocardiography was first described in human medicine in 1968 after accidental intravenous injection of agitated saline. The resulting air bubble-saline mixture enhanced contrast of the aorta and cardiac chambers. 17 Since then, commercially available second-generation contrast agents with smaller and more stable gas bubbles have been developed. 16,[18][19][20] However, first-generation contrast (air bubbles) remains in clinical use in both human and veterinary medicine. A number of CEUS techniques using various contrast agents have been reported in experimental studies in horses 18-20 but have not been reported clinically. Part 1 of the study aimed to establish a CEUS technique, using agitated injectate (air bubbles), for attempted intrasynovial injections and determine the diagnostic sensitivity and specificity of this technique in cadavers. The objectives for Part 2 were to describe the application of CEUS in horses over a variety of clinical cases; to determine whether bubbles generated by agitation would be visible ultrasonographically when clinically applicable injectates were used; and to report our experience with the effect of walking the horse on bubble visibility. We hypothesised that the CEUS technique, described herein, would classify with both high sensitivity and specificity the outcome of attempted intrasynovial injection in cadavers and that the technique would be transferable to horses in a clinical setting.

| Part 1: Cadaver study
Synovial structures from three freshly subjected to euthanasia horses were used. The details of each horse were recorded including age, breed, sex and reason for euthanasia. All horses were subjected to euthanasia using intravenous Cinchocaine Hydrochloride, Quinalbarbitone Sodium (Somulose; Dechra Veterinary Products).
CEUS was performed on injected and adjacent synovial structures.
The aim was to inject the following synovial structures in all horses: distal interphalangeal joints, proximal interphalangeal joints, metacarpophalangeal/metatarsophalangeal joints, digital flexor tendon sheaths, cubital joints, scapulohumeral joints, carpal sheaths, tarsometatarsal joints and tarsocrural joints. In addition, the aim was to inject either the antebrachiocarpal or middle carpal joint for each carpus, and either medial femorotibial or femoropatellar compartment for each stifle. For the carpus and stifle joints, the target joint was randomly determined using a dice roll. Ninety synovial structures were included in the study, with 39 synovial structures excluded due to either rigor mortis preventing flexion and extension of the limb (n = 36), the presence of intrasynovial echoes (gas bubbles from decomposition) (n = 2) identified using ultrasonography prior to injection or the region being unable to be effectively assessed ultrasonographically (n = 1). A total of 51 synovial structures were injected including 13 different anatomical synovial structures (Table 1).
horse, air, bubble, synovial injection, joint injection, contrast medium Following euthanasia, the skin over the synovial structures was clipped, cleaned and ultrasonography gel applied. Synovial structures were assessed using GE Vivid 7 ultrasonography © (GE Medical Systems Ltd) in two of the cadavers and using GE Vivid E ultrasonography © (GE Medical Systems Ltd) in one cadaver. In all cadavers, a linear transducer (ML6-15) was used.
Ultrasonography was performed before injection to confirm the normal appearance of synovial structures.
The contrast for CEUS was created by drawing up room air (equating to 20% of the volume of the injectate) into a syringe containing 10mg/ml Methylene Blue (MB) solution. The syringe was shaken vigorously for 10 seconds before expelling any remaining air from the syringe. Injections were performed either ultrasonography guided or by palpation by one of two authors (N.K.O. and J.D.S) (Item S1). The stifle compartments, scapulohumeral and cubital joints were injected using ultrasonography guidance. Both veterinarians used the same injection techniques for all cadavers (Item S1). The volume of injectate was 5-20 mL depending on the size of the synovial structures ( Table 1). One of the cadavers was a pony (168 kg) and the volume of injectate was reduced by a proportional volume accordingly. The needle was removed following injection.
Contrast-enhanced ultrasonography was then performed of both injected and adjacent synovial structures immediately following injection, and repeated following flexion and extension of the limb. The injected synovial structures were imaged via two acoustic windows: at the injection site and at a site remote from the injection to assess distribution of injectate within the synovial cavity. For example, the proximal interphalangeal joint was injected using a palmar approach; following injection CEUS was performed at the injection site (palmar) and on the dorsal aspect of the joint (remote from the injection site). A positive result for CEUS was recorded when contrast was considered to be visualised within the synovial structure, and a negative CEUS when no contrast was visualised or when contrast was considered to be outside the target synovial structure ( Table 2). Intersynovial communication was considered to be present on CEUS if contrast was identified within an adjacent synovial structure. The contrast appeared on CEUS as characteristic multifocal, hyperechoic dots, patches or lines identified within the synovial structure (   F I G U R E 3 Ultrasonography images of the lateral digital extensor tendon sheath from horse 14 before (A) and after injection (B). In the superficial aspect of the tendon sheath, the contrast can be seen as a hyperechoic line. The ultrasonography probe is placed horizontally over the tendon proximal to the wound just proximal to the tarsus volume of injectate was drawn into the syringe and, with a sterile gloved finger sealing the syringe, it was shaken vigorously for 10 seconds before expelling remaining air from the syringe. Injection was then performed promptly (within 10 seconds) using aseptic technique. If this time elapsed, then the agitation process was repeated.
Contrast-enhanced ultrasonography was then performed immediately following injection and was recorded as positive if contrast was visualised within the injected synovial structure(s) and negative when no contrast was visualised or when contrast was visualised outside the target synovial structure (Table 2). Adjacent synovial structures were also examined with CEUS and the results recorded similarly. In instances with a negative CEUS result, the horse was walked for three to four strides and CEUS repeated. When contrast was visualised but the operator was unable to determine if the contrast was intrasynovial or extrasynovial, the results were recorded as inconclusive ( Table 2). In cases with subcutaneous fluid accumulations, contrast was injected into nearby synovial structure(s) to assess for communications and both the injected structure and the fluid accumulation were assessed with CEUS.

| Data analysis
Confidence intervals (CI) were calculated for proportions using Exact    joint, and both the long and lateral digital extensor tendons sheaths.

| Part 2: Clinical cases
The remainder of the injections were performed based on palpation.  In a clinical setting identification of an unsuccessful intrasynovial injection has obvious benefit, namely, permitting repeat injection.

| D ISCUSS I ON
Additionally, clinicians and owners alike can be reassured following intrasynovial injection with therapeutic agents that any failure of treatment did not result from incorrectly located injection. CEUS was negative following three attempted intrasynovial injections in clinical cases (horses 2 and 6). Based on the results from part 1, we concluded that these injections were probably extrasynovial (incorrectly located), and we repeated them. Traditionally, veterinary practitioners rely on the retrieval of synovial fluid prior to injection, ease of injection and/or distension of the synovial structure following injection for verification of intrasynovial injection. 6 However, these techniques can be unreliable.
For example, synovial fluid was not retrieved prior to injection for many intrasynovial (correctly located) injections structures in Part 1 of this study. Furthermore, movement of the needle during injection, particularly when attaching the syringe to the needle, may result in advancement or withdrawal of the tip of the needle leading to extrasynovial injection despite retrieval of synovial fluid initially. Low resistance to injection has also been shown to be unreliable and was not associated with accurate intra-articular injection of the shoulder joint in equine cadavers. 9 Noticeable distension of large joints is unlikely to be appreciable following injection of small volumes of injectate. 23 Contrast radiography has also been described for evaluating acquired or anatomical communications between synovial structures. 2 Vigorous shaking of the injectate prior to instillation generated the ultrasonography contrast used in both parts of this study. Air bubbles, dispersed within the injectate by shaking, interact with ultrasonography waves creating characteristic hyperechoic speckles within the fluid on the image generated. The incorporation of room air into the syringe to form the bubbles is considered safe for a number of reasons. Firstly, room air is commonly detected on radiography following intrasynovial injection and air is often aspirated when performing centesis of the carpal joints, both without reported consequence. 26 Secondly, human physicians inject relatively large volumes of intra-articular air when performing certain types of contrast arthrography, again without reported increased risk of synovial sepsis. [13][14][15][16]27 Thirdly, CEUS with agitated injectate and room air has been used by human rheumatologists without complication. 13,14 And finally, although only a limited number of horses underwent CEUS in Part 2 of this study, no adverse effects were seen (14 horses; 23 synovial structures). In dusty environments, it would be considered prudent to draw air into the syringe through an appropriate filter.
The use of agitated saline as an ultrasonography contrast agent has been described in dogs and humans to diagnose congenital heart defects and portosystemic shunts. 28,29 Bubbles generated in this fashion are relatively large and are unstable over time resulting in limitations of transit within the small pulmonary circulation and the bubbles dissipating quickly within blood. 16 Commercial ultrasonography contrast agents have been developed to improve the longevity and reduce the size of the microbubbles, and are used routinely in human medicine. 16 In the equine literature, the safety and feasibility of commercial second-generation contrast agents administered systemically have been described experimentally in horses for monitoring and quantifying perfusion of the uterus in pregnant mares 20 as well as for imaging of the distal limb. 19 In an ex-vivo study, CEUS using a commercially available contrast medium with stabilised microbubbles was shown to improve the ability of the operator to detect longitudinal tears of the deep digital flexor tendon. 18 As a tool for verification of intrasynovial injection in the horse the contrast agent is not required to last for long periods, neither are the bubbles required to be small enough to pass through small vessels, making an agitated injectate a suitable contrast agent. However, ultrasonography examinations of the target structure should be performed as quickly as possible following injection. Contrast from an air-steroid mixture used in human joints has been reported to be visible for over 5 minutes. 14 9). While the reason for this has yet to be elucidated, the relatively large bubbles generated by agitation may obstruct their movement between adjacent synovial structures where the communication has a small aperture. Interestingly, in humans, intersynovial communications between tibiotalar joint and tibialis posterior tendon sheath, and between the metatarsophalangeal joint and the intermetatarsal burse have been described using agitated injectate CEUS. 14

| CON CLUS ION
Agitation of the injectate resulted in air-bubble formation within the injectate creating a reliable, economical and safe ultrasonography contrast agent, suitable for intrasynovial injection. The CEUS technique described has high diagnostic sensitivity. The technique was also used to diagnose intersynovial communications following trauma.

E TH I C A L A N I M A L R E S E A RCH
Ethical approval was granted by the University of Liverpool Research Ethic Committee (VREC790 and VREC731).

CO N FLI C T O F I NTE R E S T S
No competing interests to declare.

I N FO R M E D CO N S E NT
All owners gave informed written consent for their horse's inclusion in the study.

AUTH O R CO NTR I B UTI O N S
All authors were involved in the study design. N. Ogden and D. Stack were involved in acquiring images for part 1. All authors were involved with acquiring images for part 2. N. Ogden was involved in the data analysis and all authors were involved in the manuscript preparation and final approval.

DATA ACCE SS I B I LIT Y S TATE M E NT
The data that support the findings of this study are available from the corresponding author upon reasonable request.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/evj.13388.