Materials and Methods

Medical records of 63 equids incurring puncture wounds through the frog or collateral sulci of a single foot and admitted to the University of California-Davis William R. Pritchard Veterinary Medical Teaching Hospital between 1998 and 2008 were retrospectively reviewed. The essential criterion for inclusion in the study was a penetrating injury with a foreign body in a region confined to the frog, lateral collateral sulcus, or medial collateral sulcus. This diagnosis was based on histories that reported the presence of a penetrating injury involving the frog or collateral sulci or positive physical examination findings including a foreign body or puncture tract in the frog or collateral sulci. Diagnostic radiography, including plain and contrast studies, was also used to verify the location and depth of the puncture tract. When possible, a sterile blunt probe and injection of contrast agent were used to identify the depth and direction of the tract. In some affected equids, contrast radiography of the distal interphalangeal joint, navicular bursa, or digital flexor tendon sheath was used to determine communication between these synovial structures and the tract. Lameness at the time of hospital admission was graded for each animal on a scale of 0 (sound) to 5 (non–weight bearing).⁴ Results of cytologic analysis of synovial samples, where available, were also examined.

Each equid was assigned a grade from 1 to 4 on the basis of severity of penetration confirmed at the time of admission by use of radiographic evidence or synoviocentesis. A grade 1 injury involved the superficial corium only (< 1 inch), grade 2 injury involved the deeper corium (> 1 inch) or digital cushion but not the distal phalanx, grade 3 injury involved the distal phalanx but not the synovial structures (Figure 1), and grade 4 injury involved a synovial structure (distal interphalangeal joint, navicular bursa, or digital flexor tendon sheath; Figure 2).

View Full Size

Lateromedial radiographic view of the left hind foot of a horse with a non–weight-bearing lameness of the left hind limb. As an example of a grade 3 injury, notice that a long radiopaque foreign body (ie, nail) has penetrated the foot to the solar surface of the distal phalanx but does not involve a synovial structure.

Citation: Journal of the American Veterinary Medical Association 239, 8; 10.2460/javma.239.8.1104

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Lateromedial radiographic views of the right hind foot of a horse with a non–weight-bearing lameness of the right hind limb before (A) and after (B) injection of contrast material through a sterile canula. As an example of a grade 4 injury, notice that the penetrating tract communicates with the navicular bursa.

Citation: Journal of the American Veterinary Medical Association 239, 8; 10.2460/javma.239.8.1104

• Download Figure
• Download figure as PowerPoint slide

Factors retrieved from the medical record and evaluated included breed, age, sex, the degree of lameness, foot affected, type of foreign body, duration of time from injury to hospital admission, location of penetration, severity of injury in terms of penetration, and treatment. Equids were further placed into 3 groups: equids that were euthanized prior to treatment (n = 13), equids that were treated conservatively or with a standing surgical procedure (35), and equids that were treated surgically under general anesthesia (15).

Follow-up on all equids was achieved through use of either medical records or telephone conversation with the owners to determine outcome. Outcome was classified as whether the animal was sound and able to return to intended use, lame and unable to return to intended use, or euthanized because of poor prognosis, poor response to treatment, or financial constraints of the owner.

Statistical analysis—Factors that appeared to influence the eventual outcome on the basis of univariate analyses were examined with a multivariate logistic regression model and proportional hazards regression models by use of a commercially available statistical program.^a Assumptions of linearity in the log hazard and log odds for continuous variables were analytically verified. Level of significance was P ≤ 0.05. Statistical results are presented as odds ratios, hazard ratios, and their respective 95% CIs. The distributions of breeds of the study population and hospital population were compared by use of a χ² test of homogeneity.

Results

Although most breeds were represented, Quarter Horses were the most frequently seen, accounting for 35% (22/63) of the equids. Thoroughbreds were highly represented, accounting for 27% (17/63) of the equids, followed by Arabians (6.3% [4/63]), Morgans (4.8% [3/63]), Hanoverians (3% [2/63]), Trakehners (3%), warmbloods (3%), and 1 each of Andalusian, Selle Francias, Haflinger, Pinto, Tennessee Walking Horse, Oldenburg, Rocky Mountain, Welsh Pony, Westphalian, Paint, and mule (1.6% [1/63]). The distribution of breeds (Thoroughbreds, Quarter Horses, and others) in the study population was not significantly (P = 0.16) different from the distribution of the hospital population during the study period. Thirty-four of the equids were geldings, 25 were mares, and 4 were stallions.

The most common penetrating foreign body was a nail (54/63 [85.7%]). Others included a screw (3/63 [4.8%]), metal wire (2/63 [3%]), glass (1/63 [1.6%]), and a piece of wood (1/63 [1.6%]). The type of foreign body was unknown in 2 equids. The right hind foot was the most frequently affected in 23 of 63 (36.5%) equids, the left hind foot in 14 (22.2%), the left forefoot in 13 (20.6%), and the right forefoot in 13 (20.6%). One of the 63 (1.6%) equids had a grade 1 lameness (on a scale from 1 to 5), 11 of 63 (17.5%) had a grade 3 lameness, 33 of 63 (52.4%) had a grade 4 lameness, and 18 of 63 (28.6%) were non–weight bearing (ie, grade 5 lameness) on the affected limb. The time between injury and the hospital admission varied from 1 hour to 90 days (mean, 7.8 days).

As defined by the inclusion criteria, the puncture wounds were located within a region confined to the frog, lateral collateral sulcus, and medial collateral sulcus. For the purposes of the study, this region was divided into thirds: cranial aspect of the frog and collateral sulci, middle aspect of the frog and collateral sulci, and caudal aspect of the frog and collateral sulci. Thirteen equids had a penetration injury of the cranial third, 19 had a penetration injury of the middle aspect of the frog or sulci region, and 31 equids were affected at the caudal third. Penetration of the frog occurred in 29 equids, the medial collateral sulcus in 20, and lateral collateral sulcus in 14 equids. Thirty-four equids had penetration of 1 or more synovial structures: 4 (11.8%) had penetration of the cranial third of the frog or sulci, 11 (32.4%) had penetration of the middle third, and 19 (55.9%) had penetration of the caudal third.

On the basis of the injury severity grading system, 2 of 63 (3.2%) equids had a grade 1 injury, 18 (28.6%) had a grade 2 injury, 9 (14.3%) had a grade 3 injury, and 34 (54.0%) had a grade 4 injury (results of diagnostics in 2 of the equids were inconclusive, but because of a high degree of suspicion of synovial penetration, they were classified as having a grade 4 injury). Of the 34 equids with a grade 4 injury, 22 had penetration of the navicular bursa, 3 had penetration of the distal interphalangeal joint, 2 had penetration of the digital flexor tendon sheath, 4 had penetration of the navicular bursa and the distal interphalangeal joint, and 1 had penetration of the digital flexor tendon sheath, distal interphalangeal joint, and navicular bursa. Two equids with a grade 4 injury had inconclusive results on radiographic examination and synoviocentesis.

Thirteen of 63 (20.6%) equids were euthanized at the time of admission because of poor prognosis based on radiographic findings and results of cytologic analysis of synovial fluid. All these equids had a grade 4 injury because of the involvement of a synovial structure.

Thirty-five of 63 (55.6%) equids underwent conservative nonsurgical treatment (n = 33) or conservative treatment and a surgical procedure with the horse standing (2). This treatment group included 2 equids with a grade 1 injury, 18 with a grade 2 injury, 9 with a grade 3 injury, and 6 with a grade 4 injury. In all cases, treatment included superficial paring out of the tract, aseptic preparation of the foot, bandaging with a sterile dressing, and parenteral administration of broad-spectrum antimicrobials. Sixteen of 35 conservatively treated equids received antimicrobial regional limb perfusions IV in the affected leg. Horseshoe treatment plates were used in 7 of 35 conservatively treated equids where deeper debridement of the tract was undertaken (2 equids with a grade 2 injury, 4 with a grade 3 injury, and 1 with a grade 4 injury).

Nine of 35 conservatively treated equids had a grade 3 injury. Six of the 9 equids received antimicrobial regional limb perfusions IV. One of the equids with suspected septic osteitis underwent curettage of the distal phalanx at the region of the nail penetration with the horse standing. Of the 9 equids with a grade 3 injury, 8 returned to full soundness; the remaining equid was euthanized because of poor response to treatment as a result of osteomyelitis and sequestrum formation at the site of penetration at the distal phalanx. Of the 8 equids that recovered, 1 developed osteomyelitis and a sequestrum but recovered following a second curettage, IV antimicrobial regional limb perfusion, and systemic treatment with antimicrobials.

Six of 35 conservatively treated equids had a grade 4 injury. Four of the 6 equids had confirmed penetration of the navicular bursa, whereas for 2 equids, there was a high degree of suspicion of penetration of a synovial structure. All 6 equids with a grade 4 injury received systemic treatment with antimicrobials, debridement, and bandaging. Four of the 6 equids received antimicrobial regional limb perfusions IV. One equid underwent a deep digital flexor tendon fenestration and navicular bursotomy with the horse standing (commonly called a street-nail procedure) and regained soundness. Of the 6 equids with a grade 4 injury, 4 regained soundness, 1 remained chronically lame, and 1 was euthanized because of failure to respond to treatment and progressive radiolucency observed in the flexor cortex of the navicular bone on radiographic examination.

In summary, conservative treatment, including debridement and other surgical procedures with the horse standing, resulted in a return to soundness in 32 of 35 (91.4%) equids, of which 4 had had a grade 4 injury. Two of 35 (5.7%) conservatively treated equids (1 with a grade 4 injury and 1 with a grade 3 injury) were euthanized because of poor response to treatment and an unfavorable prognosis. One of 35 (2.9%) conservatively treated equids (with a grade 4 injury) remained chronically lame and was unable to be used for intended purposes. The length of time from treatment to return to normal activity ranged from 1 to 7 months (mean, 3 months). The length of hospitalization for 31 of 35 conservatively treated equids ranged from 1 to 18 days (mean, 5 days). The remaining 4 equids were treated exclusively in the field without hospital referral.

Fifteen of 63 (23.8%) equids underwent surgical treatment involving general anesthesia. All 15 surgically treated equids had a grade 4 injury. Of the 15 equids, 9 had penetration of the navicular bursa alone, 2 had penetration of the digital flexor tendon sheath alone, 3 had penetration of the navicular bursa and distal interphalangeal joint, and 1 had penetration of the navicular bursa, distal interphalangeal joint, and digital flexor tendon sheath. Of 15 surgically treated equids, 3 underwent navicular bursoscopy and lavage with debridement of the nail tract to facilitate drainage, 2 with penetration of the tendon sheath underwent endoscopy and lavage of the digital flexor tendon sheath, and 10 underwent a deep digital flexor tenectomy (ie, street-nail procedure).² Six of 15 surgically treated equids underwent at least 2 surgeries involving general anesthesia. Ten of the 15 equids were treated with antimicrobial regional limb perfusions IV. Of the 15 equids, 7 were euthanized because of poor response to treatment or financial constraints of the owner, 6 became sound and fit for intended use, and 2 remained chronically lame. Two of the 3 horses that underwent navicular bursoscopy returned to soundness, compared with 4 of 10 equids that underwent a deep digital flexor tenectomy.

Overall, 26 of the 50 treated equids received an antimicrobial regional limb perfusion of amikacin sulphate IV in the affected leg. The amount of amikacin sulphate administered and the number of treatments varied from 500 mg to 3 g and from 1 to 5 treatments at 24-hour intervals.

Equids that were treated later after injury were associated with having a worse prognosis (odds ratio, 0.81 for each additional day; 95% CI, 0.674 to 0.97; P = 0.02). Of the 38 equids that were treated within 48 hours of injury, 31 (81.6%) returned to soundness, 1 (2.6%) became chronically lame, and 6 (15.8%) were euthanized. Of the 10 equids that were treated between 2 and 7 days of injury, 6 returned to soundness, 1 was lame, and 3 were euthanized. Of the 15 equids that were treated at > 1 week after injury, 1 returned to soundness, 1 remained chronically lame, and 13 were euthanized.

Of the 26 equids with forefoot injuries, 11 regained soundness, 2 remained chronically lame, and 13 were euthanized. The overall return to soundness for a forefoot injury was 42% (11/26). Of the 37 equids with hind foot injuries, 27 regained full soundness, 1 was chronically lame, and 9 were euthanized. The overall return to soundness was 73% (27/37). Equids with a hind foot injury were more likely to have a better outcome, compared with equids with a forefoot injury (odds ratio, 7.49; 95% CI, 0.98 to 57.3; P = 0.053).

All equids with an injury severity grade of 1 (2/63) and 2 (18/63) returned to soundness. Eight of 9 equids with a grade 3 injury returned to soundness, and the remaining equid was euthanized. Of the 21 equids with a grade 4 injury that were treated, 10 (47.6%) returned to soundness, 3 (14%) remained chronically lame, and 8 (38%) were euthanized. Each additional increase in grade of injury severity by 1 unit was associated with having a poorer prognosis (odds ratio, 0.011; 95% CI, 0.0002 to 0.75; P = 0.036). With each additional unit increase in injury severity grade, equids took longer to return to soundness (hazard ratio, 0.72; 95% CI, 0.53 to 0.99; P = 0.044).

Overall, 38 of 63 (60%) equids returned to soundness. Excluding the 13 equids that were euthanized prior to being treated because of poor prognosis or financial constraints of the owner, 38 of 50 (76%) equids receiving treatment returned to soundness. Including the 13 equids that were euthanized following diagnosis but prior to treatment, a total of 21 of 34 (62%) equids with a grade 4 injury were euthanized and 10 of 34 (29%) regained soundness. Excluding the 13 equids that were euthanized following a diagnosis but prior to treatment, 10 of 21 (47.6%) equids with a grade 4 injury returned to soundness.

Discussion

Foreign body penetration of the foot is a common cause of lameness in the horse and emergency complaints in equine practice. All puncture wounds of the foot should be considered as being serious; however, those affecting the central region of the sole, including the frog and collateral sulci, require particular attention because of the potentially serious consequences associated with infection of the underlying structures (the deep digital flexor tendon, navicular bursa, navicular bone, deep digital flexor tendon sheath, distal interphalangeal joint, and pedal bone). If aggressive treatment is not promptly initiated, infection of these structures is potentially career ending and life threatening.⁵

Of the 63 equids included in the study, 54 (85.7%) had penetration with a nail. This finding would advocate the need for increased client education in regard to appropriate husbandry measures to prevent this, such as the use of magnets and regular checks of stalls and pastures for nails and other potentially dangerous objects.

The solar surface of the foot can be divided into 2 regions: the central region and the sole. The central region of the foot consists of the frog and the collateral sulci. Steckel et al³ found that 95% of horses with a puncture through the sole returned to soundness, whereas only 50% of the horses receiving a puncture wound through the frog or collateral sulci regained soundness, reflecting a poorer prognosis. For the purposes of the present study, only equids incurring puncture wounds to the frog itself or the collateral sulci were included to ascertain a prognosis for an area that, if affected, is considered potentially more serious. Equids with a puncture wound in the caudal third and middle third of the foot were twice as likely to have penetration of 1 or more synovial structures as were equids with a puncture wound in the cranial region. The location of the puncture wound may be considered one of the first prognostic indicators in the initial presentation and management of these cases.

The time since injury is also an important variable in determining a prognosis early in the evaluation and deciding the best course of treatment. Identification of a tract in the frog and sulci can be difficult if the penetrating object is no longer in place, as the elastic tissue of the horn tends to seal over once the object is no longer in situ. Any time lapse between the introduction of bacteria and treatment can lead to establishment of a serious infection with marked tissue destruction and extension of infection into synovial cavities such as the navicular bursa, digital flexor tendon sheath, or distal interphalangeal joint.⁶ In cases where more than a week has passed before commencement of treatment, the prognosis is grave. Horses in which the injury is < 1 week old tend to have a better prognosis.⁶ This was evident in the present study in which equids that were treated within 48 hours of injury had a much better prognosis for return to soundness in comparison with equids that were treated > 7 days after injury (81.6% [31/38] vs 6.7% [1/15] return to soundness). The longer the duration of infection, the higher the incidence of permanent damage, especially where synovial structures are concerned. The prolonged inflammatory response seen in chronic infections may result in synovial hyperplasia and hypertrophy, thrombosis of the synovial vessels, and pannus formation leading to fibrosis of the joint capsule. Chronic infections may also contribute to deterioration in the articular cartilage.⁷ Irreversible cartilage damage is the end stage of infectious arthritis and contributes to impaired joint function, permanent lameness, and ultimately a poor prognosis for future use and life.⁸

In the present study, equids that had a hind foot affected had a better prognosis than did those that were affected in the forefoot, which is consistent with findings in other studies.^2,3,9 One suggested reason for this may be the fact that more weight is carried on the forelimbs.² Another factor that should be considered is that hind limb lameness is often more subtle than forelimb lameness and may not be easily detected by owners.

Fifteen equids underwent surgical treatment involving general anesthesia, and 7 of the 15 were euthanized. The most common reasons for euthanasia included poor response to treatment, findings indicative of osteomyelitis of the flexor cortex of the navicular bone, osteitis of the third phalanx, evidence of sepsis of the deep digital flexor tendon, and financial constraints of the owner. These findings are similar to those of another study,¹⁰ in which the most common reason for euthanasia of horses with deep puncture injuries of the frog region was septic navicular bursitis with septic deep digital flexor tendinitis and osteomyelitis of the navicular bone. In the present study, 2 of 3 horses that underwent navicular bursoscopy returned to soundness, compared with 4 of 10 equids that underwent a deep digital flexor tendon fenestration and navicular bursotomy. One study⁹ in horses reported a success rate of 75% (12/16) for the treatment of contaminated and septic bursae by use of endoscopy of the navicular bursa. However, only 10 of the 16 horses returned to athletic soundness. A second study² found that 12 of 38 (31.6%) horses treated for septic navicular bursitis by use of the so-called street-nail procedure (ie, deep digital flexor tendon fenestration and navicular bursotomy) had a satisfactory outcome, but only 4 of 38 (10.5%) horses returned to athletic function, 5 (13.2%) were used as broodmares, and 3 (7.9%) had no long-term follow-up. Steckel et al³ found 6 of 19 horses treated with a deep digital flexor tenectomy were sound and back in athletic work. All of these studies discuss the complications and long convalescence associated with the so-called street-nail procedure, which involves removing the central one-half to two-thirds of the frog and underlying digital cushion to expose the deep digital flexor tendon. The navicular bursa is then entered by resecting a portion of the deep digital flexor tendon, thus facilitating drainage of the bursa.² Wright et al⁹ reported a better success rate using endoscopy of the navicular bursa, along with fewer postoperative complications and shorter periods of hospitalization. This is probably the result of a less invasive procedure without a large open wound requiring longer periods of aftercare and further debridement. In the study by Richardson et al,² 15 of 34 (44%) horses that underwent the deep digital flexor tenectomy also required further debridement of the wound following the initial surgery. Further studies are warranted to compare these techniques and ascertain the reported benefits of the navicular bursoscopy as a treatment for penetrating injuries to the bursa.

Overall, 60% (38/63) of the total number of equids returned to soundness, and 76% (38/50) of the equids that received treatment returned to soundness in the present study. Equids that responded to conservative treatment had comparatively less severe injuries than did those that were treated with more invasive surgical management, all of which included penetration of a synovial structure. Of the 35 conservatively treated equids, 91.4% (32/35) returned to a previous level of soundness. Most of these equids had lower-grade injuries (ie, 2 had a grade 1 injury, 18 had a grade 2 injury, and 9 had a grade 3 injury) in comparison with those undergoing more invasive surgical treatment involving general anesthesia, all of which had a grade 4 injury. However, of the 35 conservatively treated equids, 6 had a grade 4 injury. Of these 6 equids, 4 regained soundness, 1 remained chronically lame, and 1 was euthanized. Of the 4 equids treated conservatively that returned to soundness, 3 were treated within 24 hours of injury and 1 was treated within 3 days of injury. The equid that remained chronically lame began treatment at 12 days after injury, and the equid that was euthanized began treatment at 7 days after injury. This would indicate that aggressive early treatment in equids with synovial penetration may be of benefit, excluding the need for surgical treatment involving general anesthesia. These results differ from those of the study by Steckel et al,³ in which only 2 of 27 horses responded to medical treatment alone and none of the horses with septic synovitis responded to medical treatment alone.