Introduction

Osteoarthritis (OA) is a leading cause of lameness and morbidity and presents significant treatment challenges in animals and humans.¹ Although the pathophysiology of OA is still incompletely understood, it is now well recognized that the OA disease process typically starts with disease of the synovial membrane itself and is driven primarily by macrophages.² These synovial tissue macrophages appear to exist in a hybrid state of activation that overall displays regulatory responses that are crucial for joint homeostasis but are unable to control inflammation when regulatory functions become impaired or overwhelmed.

A 2.5% synthetic cross-linked injectable polyacrylamide hydrogel (2.5 iPAAG) has been used for years in the bulking of soft tissues such as skin and the bladder neck.³ Intra-articular injection of 2.5 iPAAG is now used to treat OA in horses^4–6 and humans,⁷ with significant effects on joint pain, joint effusion, lameness, and capsular stiffness. The 2.5 iPAAG is biocompatible, nonabsorbable, nonresorbable, and nondegradable, and since the material is chemically inert the mode of action is considered purely mechanical.^8,9 Nevertheless, histologic studies of mice, rats, rabbits, pigs, goats, horses, and humans have shown that 2.5 iPAAG administration supports cellular growth and integration and possesses a permanent and stable augmentation effect due to constant water exchange with its host tissue.^3,10 Data from clinical studies support that the clinical benefits remain long-term, up to 2 years in horses⁴ and up to 3 years in humans.⁷

This observational in vivo study was performed to characterize the nature of cellular changes in the synovium in response to 2.5 iPAAG injected IA into normal equine joints.

Methods

The study was conducted under animal ethics committee approval, and informed owner consent was obtained. A total of 10 Thoroughbred horses were included in the study, ranging in age from 3 to 5 years (median, 3.8 years), with 5 geldings and 5 mares. These animals were normal, healthy animals not suffering from OA or signs of joint disease (as determined by clinical lameness examination, including joint health and mobility and flexion tests) and chosen from available horses being retired from racing (for reasons unrelated to lameness such as age, poor racing performance, or financial constraints of the owners), horses rested from racing at a nearby training facility, or resident horses at the study facility.

Horses were kept in small 100 X 100-m grass paddocks consistent with local conditions, except for the first 48 hours after treatment when they were restricted in a 4 X 4-m stall. All 10 horses were administered either 2 mL (50 mg) or 4 mL (100 mg) 2.5 iPAAG (ArthramidVet; Contura Vet) injected into a specific joint(s): the metacarpo-phalangeal (MCP) or middle carpal joint (MCJ), chosen at random in sequence as the horses were enrolled by the treating veterinarian. Doses were predetermined, as the purpose of the study was to demonstrate target animal safety at 1 or 2 times the recommended label dose in a single joint, or up to or 3 times the label dose in the target animal for regulatory purposes. Therefore, selected horses had multiple joints injected with varying doses but not all joints were sampled for synoviocentesis or histology.

Timing of sample collection was predetermined to address specific questions about what was happening at a cellular level in the joint tissues between 14 and 42 days when tissue integration and a clinical benefit are known to occur and to assess any longer-term effects of up to 90 days postinjection. Overall, synovial fluid was aspirated from a total of 10 horses/13 joints (Table 1). Synovial fluid was aspirated prior to injection of 2.5 iPAAG (at day 0) in 7 of the 10 horses, and these joints acted as controls.

Five horses had synovial fluid aspirated at day 14 postinjection from the same treated joint (3 MCP and 2 MCJ). Four horses had synovial fluid aspirated at day 42 postinjection, all from MCP joints. One horse had fluid aspirated at day 90 postinjection only but from a total of 4 treated joints (2 MCP and 2 MCJ).

On each of days 14 and 42, 2 horses (a total of 4) owned by the study sponsor were humanely euthanized (sedation with xylazine, 1.5 mg/kg, IV, followed by a pentobarbital, 100 mg/kg, IV, bolus) and the fetlock or carpal joints were removed for gross and histologic examination of the joint capsule. Untreated joints on the contralateral limbs were used as controls.

The numbered slides used in the final analysis (Table 2) were those that had distinct cellular and architectural detail evident and that enabled the pathologists to precisely characterize the nature of the synovial reaction to the gel. All horses and limbs were observed daily for 2 days (48 hours) following treatment for swelling, effusion, and lameness and at weekly intervals throughout the study by an experienced equine veterinarian blinded to which joint had been treated.

Results

The data in this in vivo study demonstrated that an IA injection of the 2.5 iPAAG induces a simple macrophage-driven phagocytic response to foreign material.¹²

Histology

Histology results from 2 independent observers are collated and described below. No analysis for interobserver variation took place. Synovial membrane histology scores were analyzed by t test at 95% CI (Table 2) with controls set at fixed effect. Overall scores for histologic assessment of the synovium for synovial hyperplasia (mean, 1.3; SE, 0.18; P < .001), cellular infiltration (mean, 1.8; SE, 0.18; P < .001), and vascularity (mean, 1.6; SE, 0.21; P < .001) were significantly higher in the 2.5 iPAAG–injected group compared to controls. No differences between the MCP and MCJ were assessed.

In injected joints, basophilic blue gel was not noted overlying the synovial membrane (ie, in the joint space) but was present in the superficial subintima in mostly small amounts in most sections (Figure 1). In the superficial subintima, the gel had a relatively intense fibrillar/beaded appearance and there were further deposits of pale-staining, finely fibrillar basophilic material in superficial and/or deeper layers of the synovium and external connective tissues, particularly in perivascular locations.

View Full Size

Amount and distribution of the gel. Photomicrographs show sections from equine fetlock joints 14 (A) and 42 (B) days after injection with 50 mg of 2.5% synthetic cross-linked injectable polyacrylamide hydrogel (2.5 iPAAG) showing small amounts of gel (blue) in the superficial subintima (arrows). The gel has a relatively intense fibrillar/beaded appearance, and there are further deposits of pale-staining finely fibrillar basophilic material in superficial and/or deeper layers of the synovium. Photomicrographs show sections harvested from the middle carpal joint (MCJ) 14 (C) and 42 (D) days after treatment with 50 mg 2.5 iPAAG showing intermediate amounts of gel in the subintima and thickened villous projections. H&E stain; bar = 100 µm.

Citation: Journal of the American Veterinary Medical Association 262, 5; 10.2460/javma.23.10.0553

• Download Figure
• Download figure as PowerPoint slide

An inflammatory response was noted, consisting of variable but, in some sections, significant numbers of macrophages and surface synoviocytes (type A cells, which are also macrophages) infiltrating the gel material (Figure 2). Nodular aggregates of macrophages were not noted (ie, there were no organized granulomas). There were occasional binucleate histiocytic (macrophage) cells with no evidence of neutrophilic inflammation, fibrin deposition, cell death, or other inflammatory cell types but rather a simple phagocytic response to foreign material. There was no evidence of mineralization.

View Full Size

Inflammatory reaction in response to the gel. H&E stain. An inflammatory response is noted, consisting of variable but, in some sections, large numbers of macrophages and surface synoviocytes infiltrating the gel material. A—Fourteen days after treatment, depicting significant amounts of (blue) gel material with numerous macrophages infiltrating it in an attempt to phagocytose the material; bar = 50 µm. B—A higher-magnification view (X40) 42 days after treatment showing the cytoplasm of macrophages (arrows) filled with basophilic gel material, and the gel material between them is granular with vacuoles (holes); bar = 20 µm. In addition to the macrophage infiltration, in many sections (C) harvested 42 days after treatment the numbers of synoviocyte layers were increased (hyperplasia) and those cells swollen (hypertrophy); bar = 100 µm. A higher-magnification view (D; X20) showing again the synoviocyte hyperplasia and hypertrophy (same horse) and the unique demonstration of a reactive synovial lining layer (arrows) with no evidence of neutrophil infiltration, fibrin deposition, cell death, or other inflammatory cell types; bar = 50 µm.

Citation: Journal of the American Veterinary Medical Association 262, 5; 10.2460/javma.23.10.0553

• Download Figure
• Download figure as PowerPoint slide

In contrast to the control joints, villous hyperplasia (Figure 3) was seen in many of the injected joints (to various degrees), with long, thick, and/or branching villous projections with mild to moderate hypertrophy 4 to 6 cells thick. Hypervascularity was also noted in most sections and in some involved all layers, including external tendon/ligament tissue. Fibroblasts were more prominent around blood vessels in inflamed synovium, but overall differences between injected and noninjected joints were considered to be subtle. In control joints, the number of intimal blood vessels were variable and relatively numerous in some areas, indicating this could have been preexisting (presumably) subclinical disease.

View Full Size

Villous hyperplasia is further demonstrated in these images comparing an untreated MCJ with a single layer of synoviocytes (A) with an MCJ 42 days after treatment (B) and depicting long, thick, and branching villous projections (core of subintima) lined on each side by synoviocytes and in which significant amounts of subintimal (blue) gel are noted. H&E stain; bar = 100 µm.

Citation: Journal of the American Veterinary Medical Association 262, 5; 10.2460/javma.23.10.0553

• Download Figure
• Download figure as PowerPoint slide

Scanning electron microscope

Scanning electron microscope (Figure 4) confirmed that fresh 2.5 iPAAG had a physical and uniform 3-D scaffolding structure with intact cross-linked strands. In injected joints, tissue integration with the surrounding host tissue (synovial membrane) was extensive, as evidenced by gel traversed by thin strands of connective tissue (collagen) fibers. The gel structure appeared unchanged by this process, at least to 42 days in this study.

View Full Size

Scanning electron microscope images of the fresh gel (A) X 7,500 µm showing typical uniform 3-D cross-linking structure. B, C, and D—Images X 3,500 µm of joint tissue 42 days after injection with 50 mg 2.5 iPAAG showing integration of the gel with the surrounding host tissue. Cells and thin strands of connective tissue (collagen) fibers can be seen inside the gel, and the gel structure appears unaltered in appearance.

Citation: Journal of the American Veterinary Medical Association 262, 5; 10.2460/javma.23.10.0553

• Download Figure
• Download figure as PowerPoint slide

Discussion

This observational study investigating the synovial reaction, localization, and retention of 2.5 iPAAG in normal equine joints demonstrates that the 2.5 iPAAG causes a minimal transient cellular response in the synovial fluid before localizing in the subintima of the synovial lining, where it induces a reactive sublining layer that is predominately monocytic in nature.

Our findings are consistent with those of previously published studies of horses,¹⁰ with the notable exception that no free gel was discernible at gross postmortem in the joint cavity or adhering to the inner joint capsule as described at 4 months in an induced OA goat model by Tnibar et al.⁸ This was the first time some of the changes in synovial fluid cellular parameters have been characterized with the 2.5 iPAAG gel. This was also the first time that the uniformity of the 2.5 iPAAG molecule has been clearly demonstrated in situ with SEM imaging (Figure 4), an observation that in part may explain why the 2.5 iPAAG has such a unique biocompatibility and safety profile across multiple published studies of both animals⁹ and humans.⁷

The dose of 2.5 iPAAG recommended on the product label is 1 to 4 mL/joint (2 mL for the equine fetlock joint). The product has a nonpharmacological mode of action, so these recommended dose rates are based on what has been shown to be safe and effective in several independent studies,^4–6,9 and up to 4 mL may be used at the discretion of the veterinarian. The data presented in this study were used for a margin of safety submission in the test country (at 1 and 2 times the recommended dose in a single joint and up to 3 times the dose in the target animal); therefore, up to a 4-mL (100-mg) dose (2 times the recommended dose) was used in an individual MCP or MCJ. No adverse events were reported. Interestingly, in human orthopedics a single injection of 6 mL of 2.5 iPAAG is now being used to alleviate symptoms in participants with moderate to severe knee OA.⁷ This may suggest that veterinary practitioners might also consider administering higher doses in some patients depending on the severity and duration of disease.

Christensen¹⁰ reported that tissue integration of the gel was evident as early as 7 to 14 days after injection in the horse, and as a result 14 days was the first time point used in the current study. Injection with 2.5 iPAAG did not cause any statistically significant changes in the percentage of mononuclear cells, lymphocytes, TP content, or RBCs in synovial fluid. However, a statistically significant increase was observed for TNC following injection. For all parameters, however, it should be noted that case numbers were limited, and findings should be viewed in terms of clinical relevance. All parameters assessed in the synoviocentesis sampling remained within normal laboratory reference ranges throughout the study (between baseline day 0 and day 14, 42, or 90 postinjection). This minimal, transient cellular reaction in response to the gel, which is predominately macrophage driven, further supports the low level of irritation, antigenicity, and high degree of safety seen with administration of 2.5 iPAAG.

In consideration of these findings, the authors now recommend that patients may benefit from a period of reduced exercise (walking, swimming, treadmill, or light cantering exercise) while this transient inflammatory response and the integration process is taking place, at least for the first 10 to 14 days following treatment. It may also be important for veterinarians to understand this phenomenon when managing any post-treatment joint flare. On the basis of available published studies, the complication rate for IA injection of 2.5 iPAAG is estimated at 0.04%.⁹ In the authors’ own combined experiences this is similar with only occasional cases (< 1:2,000 or 0.05%) exhibiting a mild transient joint edema that responds well to conservative therapy..

Gross postmortem examination showed no observable amounts of gel material overlying the synovial surface, and no free gel was observed in the joint space at any time point. Three of the authors performed the examinations concurrently, and findings could have differed if examinations had been performed at other time points or in diseased joints. Christensen et al¹⁰ reported macroscopic findings from 13 horse joints (7 horses aged 5 to 13 years; median, 10 years) presented with veterinarian-diagnosed OA that had been injected with 2.5 iPAAG between 7 days and 2 years previously. They reported that free 2.5 iPAAG inside the cavity could not be discerned from joint fluid on naked-eye inspection but also described the PAAG appearing “as a thick, smooth, glistening, yellow substance or in the coffin joint as small, clear deposits along the inner part of the longitudinal tendon facing the cavity.” Tnibar et al⁸ in an induced OA model in goats at 4 months reported gel was seen “in various amounts adhering to the inner side of the joint capsule in all the treated OA goats.” Diseased (OA) joints will often show signs of gross inflammation, such as synovitis and villous hyperplasia,¹³ which may affect the gross appearance and possibly even the uptake of 2.5 iPAAG in situ. However, 2.5 iPAAG is transparent and histologic findings in horse, goat, and rabbit joints all show similar histologic findings to those in our study; the gel is clearly demonstrated to be fully integrated into the subintima of the joint capsule between 7 and 14 days after injection.^8,10 In rabbits, free gel was still discernible in the cavity at 10 days but not at 30 days.¹⁰ Therefore, in the authors’ opinion, it is important not to interpret these statements as free gel residing in the joint cavity, but rather that the transparent gel is integrated within a thin layer of transparent synoviocytes with induced synovial hypertrophy and hyperplasia.

The gel acts as a bioscaffold, enabling cell migration and vessel integration.^9,10 In our study, the gel was seen as a blue granular (basophilic) material with a local tissue response consisting predominately of macrophages and surface synoviocytes that appear to be attempting, unsuccessfully, to phagocytose the gel material. This finding contrasts with an earlier study¹⁴ of a 4% PAAG that showed synovial macrophages were able to phagocytose that gel. This property presumably therefore relates to variations in chemistry and manufacturing that imparts differences in the molecular size, stability, and stiffness of different PAAG products.¹⁵ This may also explain the superior and long-term efficacy seen with 2.5 iPAAG in contrast to simple visco-supplement treatments like hyaluronic acid that are metabolized quickly after injection,¹³ triamcinolone acetonide, and hyaluronic acid combinations,^6,9 as well as 4% PAAG that shows only a reduction in lameness grade in the single efficacy study published so far.¹⁶ In comparison, efficacy studies with 2.5 iPAAG have used complete resolution of lameness as their successful outcome measure.^4–6,9 A direct comparison between 2.5 iPAAG and 4% PAAG could be a focus of future studies.

The synovial hypertrophy and villus hyperplasia observed in this study was described as a low-level phagocytic response to foreign material that was inflammation by definition. This type of local tissue response to a foreign body is typical in approved medical devices in humans¹² and importantly does not imply a high level of irritation or antigenicity. In our study, the inflammation showed no evidence of neutrophilic inflammation, fibrin deposition, mineralization, cell death, or other inflammatory cell types. Our findings support that the 2.5 iPAAG does not appear to act as a nidus for infection, nor does it cause fibrosis. Less is known about the performance or safety of other PAAG products, and the widespread and largely indiscriminate use of these products in some countries has caused serious long-term complications, mainly infection and granulomatous reactions.^9,15 It is therefore important for veterinarians to consider this when selecting a particular PAAG product for clinical use.

Our findings clearly demonstrate that by 14 days 2.5 iPAAG is fully integrated into the subintima with the formation of a de novo intimal cell layer and does not remain in the joint space. It is not known whether integration of gel into the subintima affects joint pharmacokinetics, and specific comparisons between 2.5% iPAAG injected and noninjected joints should be a focus of future studies.

It has been suggested that the notable macrophage-driven cellular reaction in the synovial intima may act to improve the nature of the synovial fluid.^10,14 Other studies, as well as the authors’ own experiences, have recognized the limitations in the detection of inflammatory cytokines in synovial fluid, even in samples from patients experiencing marked inflammation.^2,4,16,17 However, synovial fluid quality could be assessed with a simple viscometer, rheological analysis, tribology, or similar objective measures.¹⁸ Recent advances in transcriptomic RNA analysis may also provide new opportunities in the future.¹⁹ We used nonclinical joints, and we did not attempt to evaluate the effects of the 2.5 iPAAG on any measures of synovial fluid quality or markers of inflammation, but this should be a focus for future studies.

Scanning electron microscopy was undertaken to gain a better understanding of the gel’s physical response to integration. This enabled us to observe the 3-D structure of the gel. It is in the manufacturing process that the manufacturer utilizes its unique and proprietary (IL-X cross-linking) technology and where polymerization is progressing under tight temperature control to provide a minimum of variation in formed chain length and distribution of cross-links. It is this stable and homogenous gel structure that is accredited for the 2.5 iPAAG’s stability, biocompatibility, and safety in comparison to other hydrogels.^9,15 This uniformity is clearly demonstrated in the SEM images (Figure 4) and further illustrates the gel acts as a bioscaffold, undergoing extensive tissue integration. There was no evidence of any untoward effect on the gel structure itself (ie, the gel structure appeared stable and remained unchanged in appearance, at least by 42 days).

The 2.5 iPAAG does however induce significant macrophage-driven villous hypertrophy and hyperplasia as compared to noninjected joints and significant hypervascularity, although the differences between injected and noninjected joints for this variable were considered subtle. Notwithstanding, these findings may be inconsistent with what is normally considered desirable for joint health.¹³ All of the control joints were set at fixed effect for statistical analysis, which may have influenced the outcome of the analysis, but subjective observations from 2 independent and experienced equine pathologists also supported that there were significant histologic changes induced by the 2.5 iPAAG. However, multiple clinical trials have now demonstrated superior long-term (up to 2 years) clinical efficacy with complete resolution of lameness in 65.3% to 83.3% of horses treated with 2.5 iPAAG across a range of equine disciplines.^4–6,9 Previous studies have suggested either a biomechanical mechanism of action through an increase in joint capsule elastance⁸ or some means of improvement in synovial membrane function.^8–10 What is increasingly recognized, however, is that macrophages play a central role in both the synovial inflammation leading to OA and the restoration of the joint to homeostasis.^2,20

Our study had several limitations: the population of horses used in this study all had a history of being used previously for athletic pursuits (including racing) and so findings could indicate preexisting subclinical disease; a limited number of experimental animals were included, none with clinical evidence of OA; and subjective clinical measures were used, including reaction to flexion and effusion, different types of joints, variation in 2.5 iPAAG doses, and different time periods for follow-up. Some of our findings also highlight the importance in future studies of knowing the preexisting level of pathology when trying to estimate the effect of the gel, and even though our experimental design was aimed at minimizing variability, synovial histologic parameters could vary with joint and site within the joint, and this could also have contributed to a degree of variability among samples. Also, no attempt was made in these normal joints to stain for nerve endings or perform joint capsule elastance studies of equine samples to compare with previous studies performed of goats with OA.⁶

In conclusion, injection of 2.5 iPAAG induces a low-level foreign body response, which is predominately macrophage driven, and with no evidence of fibrosis or mineralization. By 14 days, the gel is fully integrated into the subintima, with no free gel remaining in the joint cavity. These findings support previous studies that have found 2.5 iPAAG is safe, and our findings can be used to guide future directions for research and evidence-based treatment decisions with the likes of 2.5 iPAAG.

Disclosures

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. This study received funding from Innovative Medical Solutions (IMS) Ltd, which is the Market Authorization Holder for 2.5 iPAAG (ArthramidVet) in Australia and New Zealand. The funder had the following involvement with the study: provision of investigatory product and payment of costs associated with laboratory analysis. The authors have provided the following statement: I/We, [Jason Lowe and Leigh de Clifford], hereby disclose potential conflicts of interest related to the research presented in this article. I/We are employed by/shareholders in [IMS Vet], the organization that sponsored and funded this research project. My/Our employment/shareholding with [IMS Vet] may be perceived as a conflict of interest, as it could potentially influence the interpretation and reporting of the study results. To address this potential bias, I would like to clarify my/our primary obligation as a scientist is to uphold the highest standards of research integrity and to contribute to the advancement of knowledge in a fair and unbiased manner. If there are any questions or concerns about potential conflicts of interest or the study’s integrity, please feel free to contact me directly at [jason@imsvet.com]. All other authors declare no competing interests.

No AI-assisted technologies were used in the generation of this manuscript.

Funding

Funding for the 2.5% iPAAG product and laboratory testing costs was provided by Innovative Medical Solutions Ltd.