|Home | About | Journals | Submit | Contact Us | Français|
The significance of collateral ligament desmitis of the tarsocrural joint is often clinically underestimated, because it is an uncommon injury with a guarded prognosis for athletic soundness. The objective of this study was to describe the clinical presentation, treatment, and outcome of 12 horses with collateral ligament desmitis, along with tarsocrural joint synovitis secondary to hemarthrosis. Criteria for inclusion in this study included clinical signs of tarsocrural joint synovitis and sonographic evidence of collateral ligament desmitis. This retrospective study evaluated horses over an 8-year period. Median follow-up after treatment was 5.5 years. Four horses in the study returned to their previous level of performance, 6 horses remained lame due to pain in the tarsus, 1 was euthanized, and 1 is in convalescence. This study highlights the importance of collateral ligament desmitis, and emphasizes the need for early, aggressive treatment to prevent the development of osteoarthritis.
Résultats cliniques des lésions du ligament collatéral du tarse. L’importance du ligament collatéral desmitis de l’articulation tarsocrural est souvent cliniquement sous-estimé, parce que c’est une blessure rare avec un pronostic gardée pour athletic solidité. L’objectif de cette étude était de décrire la présentation clinique, le traitement et le résultat de 12 chevaux avec ligament collatéral desmitis, avec tarsocrural mixte synovite secondaire à hémarthrose. Critères d’inclusion dans la présente étude comprenaient des signes cliniques de synovite conjointe de tarsocrural et de preuve échographie du ligament collatéral desmitis. Cette étude rétrospective a évalué chevaux sur une période de 8 ans. Médiane de suivi après que traitement était de 5,5 ans. Quatre chevaux dans la présente étude revient à leur niveau précédent de performances, six chevaux est resté lame en raison de la douleur dans le tarse, 1 a été euthanasiés et 1 est en convalescence. Cette étude met en évidence l’importance du ligament collatéral desmitis et met l’accent sur la nécessité d’un traitement précoce, agressif empêcher le développement de l’arthrose.
(Traduit par les auteurs)
Tarsocrural collateral ligament desmitis has been reported as a cause of severe hind limb lameness in horses (1). Clinical signs of tarsocrural collateral ligament injury include: edema over the injured collateral ligament, pain on palpation over the injured ligament, and/or effusion within the tarsocrural joint (1). Lameness varies from mild to non-weight bearing with joint flexion exacerbating a pain response (1). Chronic pain and osteoarthritis may develop secondary to joint instability leading to cartilage damage in horses with collateral ligament injuries of the tarsocrural joint (1).
Tarsocrural joint collateral ligament desmitis has also been documented to cause hemarthrosis (2). There have been multiple reports of pain associated with hemarthrosis in horses and humans (3–5). Joint capsule distention and/or inflammation within the joint were thought to be the cause of lameness in a study investigating a group of horses administered an intra-articular dose of autologus blood into the metacarpophalangeal joint (4).
The purpose of this retrospective study was to describe the clinical presentation, treatment, and outcome of horses with tarsocrural joint with collateral ligament desmitis. We hypothesized that tarsocrural joint synovitis due to hemarthrosis can result from concomitant collateral ligament desmitis of the tarsocrural joint.
Medical records at Oakridge Equine Hospital were searched for horses having had a tarsocrural joint ultrasound between December 2003 and February 2011 (Table 1). Horses were included in the study if they had evidence of tarsocrural joint collateral ligament desmitis based on ultrasonographic examination. Information obtained from the medical records included signalment; history (duration of lameness, how injury occurred); physical and lameness examination (6) findings (grade of lameness, response to flexion, amount of edema, and effusion within the tarsocrural joint); diagnostic imaging results; arthroscopic surgical findings; culture and sensitivity results, and treatment. Diagnostic analgesia was not performed because clinical signs localized the lameness to the tarsocrural region.
Follow-up information was collected from the owners by telephone questionnaire in 10 of 12 cases (1 horse was euthanized prior to the conclusion of the study and a second horse was convalescing). The questionnaire included the following: is the horse sound (yes/no), how long until the horse returned to soundness (number of days), and is the horse performing at the pre-injury level of performance (yes/no). The median duration of follow-up was 5.5 y with a range of 9 mo to 8 y.
All horses in the study were examined radiographically (EDR3; Sound-Eklin, Carlsbad, California, USA). During the initial examination, 4 standard radiographic views of the affected tarsus (lateromedial, dorsoplantar, dorsolateral-plantaromedial oblique, and dorsomedial-plantarolateral oblique) were obtained.
Each horse had an ultrasound examination of the affected and contralateral tarsus at the time of the initial examination (MyLab 50 with 7.5 to 12 mHz linear array transducer probe; Biosound Esaote, Indianapolis, Indiana, USA). The contralateral limb was considered normal for each case. Echogenicity, fiber pattern, and cross-sectional area were evaluated in each collateral ligament. The synovial fluid was evaluated for signs of hemarthrosis (increase in echogenicity and/or a swirling echogenic pattern) (7). The synovial lining was evaluated for thickening and fibrinous loculations in the tarsocrural joint (7). The medial and lateral long and short collateral ligaments of the tarsus were examined in longitudinal and transverse planes, from proximal to distal.
A combination of treatments was used for this group of horses, depending on the case and the surgeons’ discretion. While hospitalized, most horses were medicated with one or more of the following antimicrobials: procaine penicillin G (Aquacillin; Vedco, St. Joseph, Missouri, USA), 22 000 IU/kg body weight (BW), IM, BID; gentamicin sulfate (Gentavid 100; Vedco), 6.6 mg/kg BW, IM or IV, SID; metronidazole (Metronidazole; Krakow Pharmaceutical Company, Krakow, Poland),15 mg/kg BW, PO, BID; and/or enrofloxacin (Baytril 100; Bayer Health Care, Shawnee Mission, Kansas, USA), 7.5 mg/kg BW, IM or IV SID, along with phenylbutazone (Equi-Bute; Schering-Plough, Summit, New Jersey, USA), 4.4 mg/kg BW, PO, SID. Each horse also had a dimethyl sulfoxide (DMSO 90% Gel; Butler Animal Health, Dublin, Ohio, USA)/nitrofurazone (Nitrotop; Butler Animal Health) sweat wrap placed on the injured tarsocrural joint. The sweat wraps were removed and/or replaced every 24 h. Treatment also consisted of extracorporeal shock wave therapy (Versatron; High Medical Technologies, Lengwil, Switzerland) and joint lavage with Lactated Ringers solution (Lactated Ringers Injection; Baxter Animal Healthcare, Deerfield, Illinois, USA) in a through and through fashion with 14-gauge needles. Arthroscopic examination, lavage, and debridement of the tarsocrural joint were performed through standard approach to the dorsal and plantar pouches of the tarsocrural joint (2).
Local antimicrobials were administered via intra-articular injections of amikacin sulfate (Equiglide, Vedco) 0.5 to 1.0 g per joint or by intravenous regional limb perfusion with amikacin sulfate 1 g per perfusion, cefotaxime sodium (Cefotaxime; West-Ward Pharmaceutical Corp, Eatontown, New Jersey, USA), 2 g per injection; or imipenem and cilastatin (Primaxin; Merck and Co. Inc., Whitehouse Station, New Jersey, USA), 250 mg per injection. Intra-articular injections or regional limb perfusions were performed once every 24 h. Antimicrobials used in each case were based on each surgeon’s preference. Intra-articular injection with hyaluronate sodium (Hyvisc; Boehringer Ingelheim Vetmedica, St. Joseph, Missouri, USA), 22 mg/joint was also given in select cases.
All horses were discharged with instructions to the owners to ensure stall rest for 15 to 90 d. Some horses were also discharged with systemic antimicrobials to be administered for 6 to 14 d following the day of discharge. A successful outcome was defined as a horse returning to its previous level of performance.
Twelve horses met the criteria to be included in the study, (6 geldings, 4 mares, 2 stallions) with a median age of 5.5 y (mean 6.3, range: 3 to 12 y). Of these 12 horses; 8 were Quarter horses, 2 Appaloosas, 1 Paint, and 1 Thoroughbred. Six horses were used for barrel racing, 5 for roping and/or reining, and 1 was used for racing (Table 1). The hospital breed distribution is 50% Quarter horses and hospital use distribution is 30% western performance events. Three of the horses sustained an injury to the collateral ligament while performing, and 9 were injured while they were in a stall or turned out to pasture/paddock.
The median lameness score was 4.0/5 (mean 3.5, range: 2–4/5). The median duration of clinical signs prior to presentation was 5 d (mean 8.8 d, range: 1 to 28 d). On clinical examination, 9 horses were lame in the right hind limb and 3 were lame in the left hind limb. The ligaments that were damaged included the long lateral collateral ligament (7 cases), long medial collateral ligament (3 cases), and both medial and lateral long collateral ligaments (2 cases). Specific information on lameness and ligaments injured are outlined in Table 1. Damage to the short oblique collateral ligaments was not found in any of the cases.
Each horse withdrew the limb and held the tarsus in flexion when minimal digital pressure was applied over the damaged collateral ligament. All horses resisted manipulation of the tarsus (flexion, medial, and lateral bending force) and had some degree of periarticular edema with marked edema focused around the damaged collateral ligament. Subjectively, 8 horses had severe effusion and 4 horses had moderate effusion within the tarsocrural joint.
Bacterial culture and sensitivity performed were performed for 2 cases, one of which was negative (case 6) and the other was positive for Streptococcus (case 8). Cytology was not performed in any of the cases.
All cases had radiographic evidence of soft tissue thickening, with marked thickening focused around the injured collateral ligament. Ten of the 12 horses had no significant articular abnormalities within the tarsocrural joint during the radiographic examination performed at the time of presentation. Radiographic findings within each tarsocrural joint are described in Table 2. Three horses had evidence of osteoarthritis within the tarsometatarsal and distal intertarsal joint.
Follow-up radiographs were taken between 30 d and 240 d following the initial examination. Four of the 6 horses with follow-up radiographs had signs of progressive osteoarthritis at the time of follow-up examination (Table 2).
All horses had evidence of thickened synovium, fibrinous loculations, and/or coagulated blood and hyperechoic swirling synovial fluid within the tarsocrural joint (Figure 1). The damaged collateral ligament in each case had moderate diffuse fiber pattern disruption that extended the length of the ligament, along with increased cross-sectional area compared with the contralateral limb (Figure 2). None of the injured collateral ligaments had complete disruption of the fibers within the ligament. Cartilage irregularity was seen on the medial trochlear ridge in case 7. During arthroscopic surgery, a full-thickness cartilage erosion was seen on this area of the medial trochlear ridge.
Of the 7 horses that had follow-up ultrasonography, horses of cases 4, 3, 9, and 11 had a normal collateral ligament fiber pattern at 80, 90, 120, and 120 d, respectively, following the initial examination. Three of the 7 horses with follow-up examinations had collateral ligaments that remained larger than the tarsocrural collateral ligament on the contralateral limb. At 90 d following the initial examination, case 2 had a 50% improvement in fiber pattern and 75% improvement in cross-sectional area. The case 7 horse had 50% improvement in fiber pattern at 190 d, along with roughening of the medial trochlear ridge of the talus. The case 12 horse had 80% improvement in fiber pattern at 120 d. The other 6 horses were lost to ultrasound follow-up due to the owner declining further evaluation of the horse.
Twelve horses were hospitalized for a median of 6.5 d (mean 5.6 d, range: 4 to 17 d) during the study period (Table 3). Nine of 12 horses were medicated with procaine penicillin G and gentamicin sulfate while they were hospitalized. After 4 d of treatment with procaine penicillin G and gentamicin sulfate, the case 8 horse was switched to enrofloxacin because clinical signs became more severe. The remaining horse (case 3) received only gentamicin sulfate. Table 3 outlines the treatment details per case. The horses of cases 11 and 12 did not receive any systemic antibiotics.
Each horse received systemic phenylbutazone and a DMSO/nitrofurazone sweat wrap over the affected tarsocrural joint for 3 to 10 d, starting the day of presentation. Three horses (cases 2, 3, 8) had a single joint lavage of the affected tarsocrural joint with lactated Ringer’s solution. One horse (case 2) had the joint lavage 2 wk prior to arthroscopic surgery and 1 horse (case 8) had the joint lavage 3 d prior to arthroscopic surgery.
Local antimicrobial therapy is summarized in Table 3. Six horses received intra-articular injections of amikacin sulfate. Following minimal improvement with intra-articular injections of amikacin sulfate, the horses of cases 6 and 8 received regional limb perfusion with cefotaxime initially, followed by imipenem and cilastatin.
Extracorporeal shock wave therapy over the damaged collateral ligaments was performed in 3 cases (cases 4, 7, and 12). The horse of case 4 had 1 treatment 14 d following the initial injury. The horse of case 7 had 2 treatments at 30 d and 60 d following the initial injury. The horse of case 12 had 1 treatment 60 d following the initial injury.
Arthroscopic lavage and debridement of the tarsocrural joint was performed on 8 of 12 horses. Median duration of time from injury to surgery was 14 d (mean 14.4 d, range: 1 to 28 d) (Table 4). Serosanguinous synovial fluid, fibrin, and blood clots were flushed and debrided from the tarsocrural joint in all 8 horses that had arthroscopic surgery (Figure 3). There was widespread cartilage thinning and fibrillation within the tarsocrural joint in all 8 horses. Six of 8 horses had focal full thickness cartilage erosions and/or bone fragments removed during the initial arthroscopic surgery (Table 4; Figure 4). Synovial proliferation along with a tear in the joint capsule over the collateral ligament and fraying of the collateral ligament were seen during arthroscopic surgery in each case. The horses of cases 2 and 6 had a second arthroscopic surgery to remove additional fibrin and fibrillated cartilage (Table 4). Following the second arthroscopic surgery, the horse of case 6 was hospitalized for 17 d and received phenylbutazone for 10 d. The horse of case 2 was hospitalized for 2 d and received phenylbutazone for 4 d. Systemic and local antimicrobial therapy for these 2 cases is outlined in Table 3.
In total, 12 horses were discharged from the hospital with instructions to have strict stall rest for 30 to 120 d, along with ultrasound examinations every 30 to 60 d. Thirty days of stall rest with 15 min of daily hand walking began once the horses had an 80% improvement in lameness, along with significant sonographic improvement in the fiber pattern of the damaged collateral ligament. Following stall rest with hand walking, horses were turned out in a 30′ × 30′ paddock for 30 d and then 30 d of increasing exercise under saddle. The horses of cases 4 and 5 were discharged with instructions to administer doxycycline hylate (Doxycylcine; West-Ward) 10 mg/kg BW, PO, BID, for 4 and 10 d, respectively. The horses of cases 6 and 8 were discharged with instructions to administer chloramphenicol (Viceton Biomeda, Le Sueur, Minnesota, USA), 50 mg/kg BW, PO, TID for 7 and 14 d, respectively.
At the time of follow-up 4 horses were athletically sound (Table 1). The horse of case 3 went back to its previous level of performance 3 mo following the initial presentation; it was retired after 6 mo of performance due to a carpal injury. Seven horses with follow-up for more than 3 y were retired due to tarsocrural joint lameness despite strict stall rest, and remained pasture sound at the time of follow-up (Table 1). The horse of case 10 was still in convalescence 60 d after injury, and the horse of case 8 was euthanized due to a persistent severe lameness in the affected tarsus 30 d following the initial presentation.
Three of the 4 horses that returned to their previous level of performance were grade 3/5 lame and had moderate effusion at presentation; all 4 horses had been seen within 24 h of the initial injury. Six of 7 horses that did not have a positive outcome were grade 4/5 lame and all 7 horses had severe tarsocrural joint effusion.
Our hypothesis that tarsocrural joint synovitis due to hemarthrosis can result from concomitant collateral ligament desmitis of the tarsocrural joint was confirmed in each case in this study. All 12 horses had clinical signs consistent with a tarsocrural joint synovitis. Eight of the twelve horses had evidence of hemarthrosis and synovitis confirmed during arthroscopic surgery, and the remaining cases were confirmed via ultrasonographic examination. Moderate to severe inflammation and pain has been documented in horses with hemarthrosis (2,3), which is a common finding in acute cases of collateral ligament tears of the tarsocrural joint (1). Due to the intimate association between the collateral ligaments and joint capsule (1,8), we suspected that a lateral to medial bending force, or vice versa, can result in damage to the collateral ligament and joint capsule, causing a hemarthrosis. The hemarthrosis is secondary to bleeding from the joint capsule (3).
Collateral ligament injury generally involves diffuse damage throughout the ligament (7). Sonographically there was an increase in the cross-sectional area, decreased echogenicity, and diffuse fiber pattern disruption within each injured collateral ligament. None of the damaged collateral ligaments were completely ruptured. In this study, the lateral collateral ligament was injured more frequently than the medial collateral ligament.
All of the horses in this study had desmitis in the long portion of the lateral collateral ligament. Research has shown that the long collateral ligaments are under more tension when the tarsus is in extension (9). Because most of the horses were injured while turned out, we speculate that the injury occurred from a bending force while the limb was in extension and the long collateral ligaments were under tension.
Tarsocrural joint osteoarthritis developed in 4 of the 6 horses with follow-up radiographs. It was not possible to determine if the osteoarthritis in the 4 cases with follow-up radiographs was secondary to severe chronic inflammation, cartilage defects, subtle changes in joint stability (1), or a combination of all three. Chronic inflammation within a joint can cause cartilage degradation (10,11). In addition, cartilage defects and cartilage surrounding the defect have been shown to be degenerative and to progress to osteoarthritis (12). It is possible that shear forces applied to the articular cartilage within the tarsocrural joint at the time the collateral ligaments were injured, caused a cartilage flap or full thickness cartilage erosion seen arthroscopically in 6 cases. The horse of case 6 had no changes on radiographs taken at 30 d after injury, which may have been too soon after the initial injury to see progressive signs of osteoarthritis. This horse was later retired due to a chronic lameness in the affected tarsocrural joint. Each horse with signs of progressive osteoarthritis also had some degree of cartilage damage noted during arthroscopic surgery; none of these horses regained full athletic potential. The horse of case 9 was the only horse to have cartilage damage noted during arthroscopic surgery and still return to athletic soundness. All 4 horses that returned to soundness had no evidence of osteoarthritis on radiographs at the time of presentation and at the time of follow-up.
Re-injury of the tarsocrural collateral ligament has been documented in the literature (13,14) and may be a cause for persistent lameness in the horses in the present study. Unfortunately, we had follow-up ultrasound examinations on only 7 horses. Three of 7 horses that remained lame or were euthanized had follow-up ultrasound examinations, and only the horse of case 4 had a normal collateral ligament at 80 d following the initial injury. All 4 horses that regained soundness had a follow-up ultrasound examination, 3 out of 4 had a normal appearing collateral ligament at the time of follow-up (90 to 120 d). Future studies need to be performed to determine if additional therapy [such as platelet rich plasma, stem cells, extracorporeal shock wave therapy (ESWT)], directed toward the collateral ligament would improve the clinical outcome.
Bacterial culture from the synovial fluid, fibrin, or synovium identifies the etiologic agent in 32% to 87% of septic arthritis cases (15–19). Administration of antimicrobials prior to synovial fluid aspiration decreases the chances for a positive synovial fluid culture (20). Because of the wide-ranging results for obtaining a positive bacterial culture and most horses receiving antimicrobials prior to a sample for culture being taken, only 2 horses in the current study had a culture and sensitivity performed on the synovial fluid, synovium, and/or fibrin collected from the tarsocrural joint.
Despite the low number of bacterial cultures in this study, systemic and local antimicrobials were used aggressively in cases with severe clinical signs associated with a septic arthritis (15,20) due to the likelihood of developing septic arthritis as a result of hemarthrosis (15) or idiopathic tarsocrural septic arthritis (1). The clinical signs and lameness became worse in all horses in this study that received stall rest and nonsteriodial anti-inflammatory medication prior to presentation. The surgeons treating theses cases felt that there was a synovitis and possible septic arthritis because the clinical signs were getting worse with time and anti-inflammatory medication. Each horse showed considerable improvement in clinical signs 1 to 2 d following the initiation of systemic and/or local antimicrobials. The literature states that horses with heat, swelling, and pain localized to the joint should be diagnosed and treated for septic arthritis even if the etiologic agent cannot be identified (15). Healthy animals can have bacteria enter the blood stream from the gastrointestinal tract or other sources (21). These bacteria are normally eliminated by the immune system (21), but in rare cases it appears that the bacteria gain access to the joint and establish an infection prior to being eliminated from the blood stream (15). The surgeons chose broad-spectrum antimicrobials for systemic administration and the best antimicrobial for the most likely potential pathogen (15) for administration via regional limb perfusion and/or intra-articular injections. Comparison of antimicrobial therapy was not performed because of the wide range of treatment regimens used.
The decision to perform arthroscopic surgery on the injured tarsocrural joint was based on the surgeons’ impression of the severity and duration of clinical signs and on financial constraints. The more severe and longer the duration of clinical signs, the more likely arthroscopic surgery was chosen. This may have negatively biased horses receiving arthroscopic surgery, as these cases were more advanced in the disease process leading to a worse prognosis for a full recovery. The horse of case 9 was the only horse that had arthroscopic surgery and returned to soundness. It is unknown if arthroscopic surgery earlier in the disease process would have yielded a more favorable outcome.
Twelve horses with tarsocrural joint collateral ligament damage along with synovitis of the tarsocrural joint were identified in an 8-year period, indicating that this condition is relatively rare in our population of horses. Nine horses sustained the injury while in a stall or in uncontrolled exercise, suggesting that the injury may not be related to any specific activity. Western performance Quarter horses, which are allowed pasture turnout during periods throughout the day, were over-represented in this study. A study investigating collateral ligament rupture in the metacarpo/metatarsophalangeal joint found that horses housed in pastures may have increased risk of collateral ligament injury because they are at risk of stepping in a hole or getting caught in a fence (13).
Only 4/12 horses returned to their previous level of performance, 6 remained lame, 1 was euthanized due to severe lameness in the affected tarsus, and 1 was in convalescence. These results show a guarded prognosis for horses to return to athletic soundness, which is in agreement with prior research (14). Three of the 4 horses that returned to soundness had been admitted to the hospital within 24 h of the initial injury. Of the horses that remained lame only 1 was seen within 24 h of the injury and the rest were seen between 5 and 28 d. This suggests that early aggressive treatment may improve the prognosis for athletic soundness.
In this study, horses with less severe clinical signs and/or a shorter duration of time between injury and onset of clinical signs may have a better prognosis to return to athletic soundness compared with horses that remained lame. The retrospective nature of the study limited our ability to make comparisons between treatment regimens. The perception that horses that underwent arthroscopic surgery had a worse prognosis for returning to soundness was skewed, because the cases with the worst clinical signs received arthroscopic surgery. The authors stress the importance of identifying cases with tarsocrural collateral ligament desmitis and possible synovitis secondary to hemarthrosis within 24 h of injury, and initiating aggressive anti-inflammatory and antimicrobial (systemic/local) therapy prior to onset of prominent clinical signs of severe synovitis and/or septic arthritis. CVJ
Use of this article is limited to a single copy for personal study. Anyone interested in obtaining reprints should contact the CVMA office (gro.vmca-amvc@nothguorbh) for additional copies or permission to use this material elsewhere.