Keratomas are an uncommon cause of lameness in equids.1 Although their pathogenesis has not been conclusively determined, keratomas are believed to develop secondary to inflammation induced by hoof wall trauma at the level of the coronary band or subsolar abscess. In most instances, keratomas manifest as an aberrant cylindrical growth of keratin that follows the horn tubules distally. Less commonly, keratomas can take the form of a solitary spherical mass within the hoof capsule. Because of their expansile nature, keratomas impinge on the distal phalanx, resulting in chronic inflammation, pressure resorption, and the characteristic semicircular lucency often observed radiographically.
Clinical signs of keratomas in equids include various degrees of lameness, recurrent foot abscesses, deformation of the hoof capsule adjacent to the underlying keratoma, and displacement of the white line toward the sole as it continues to grow distad. A diagnosis of keratoma is typically made on the basis of anamnesis, observation of the aforementioned clinical signs, radiographic findings, and, if necessary, additional imaging techniques, including CT, MRI, or, more uncommonly, nuclear scintigraphic evaluation.
Treatment to prevent recurrence requires en bloc surgical removal of the entire mass through either complete or partial hoof wall resection. Partial hoof wall resection appears to be the treatment of choice, given the lower complication rate and more rapid return to use for horses treated with this technique than for horses treated with complete resection.2–4
In a previous study,5 MRI (n = 7) or CT (3) was performed for 10 horses during surgery to more accurately define the margins of the keratoma, thereby decreasing the size of the hoof wall resection necessary to remove the keratoma, which subsequently resulted in a low complication rate (2/10; imaging modality not specified). In that study,5 barium paste had been applied to the hoof wall as a marker overlying the keratoma to guide hoof wall resection. The more common use of MRI versus CT in that study5 was reportedly attributable to the unavailability of CT at that time for most horses, and the investigators stated that they preferred the use of intraoperative CT, given the superior image quality of the hoof capsule than attainable with MRI. However, no conclusions could be drawn regarding the usefulness of CT-assisted keratoma removal specifically owing to various limitations.
Reported complication rates associated with surgical resection of keratomas in horses range from 20% to 71%.2,4,5 Such complications include excessive granulation tissue formation, hoof capsule instability, surgical site infection, and keratoma recurrence. This large degree of variability could be due to differences among studies in techniques used for keratoma removal (ie, partial vs complete hoof wall resection) as well as definitions of what constitutes a complication.
The objectives of the study reported here were describe the preoperative use of CT in the diagnosis of and surgical planning for keratoma removal in equids, describe the surgical removal techniques and postoperative patient management approaches, and determine the outcome for these patients. We surmised that preoperative CT would allow for accurate identification and delineation of keratomas, resulting in low postoperative morbidity and rapid return to previous use following surgical removal.
Materials and Methods
Case selection criteria
Medical records for equids evaluated at the University of California-Davis William R. Pritchard Veterinary Medical Teaching Hospital between October 1, 2005, and July 31, 2016, were electronically searched to identify those that underwent distal limb CT examination followed by keratoma removal during the same or a subsequent anesthetic episode. All equids meeting these criteria were included in the study.
Medical records review
Preoperative data were collected from the medical records of included equids regarding age, breed, sex, affected limb or limbs, reason for referral, lameness severity, and radiographic findings. Collected intraoperative data included total durations of anesthesia, CT examination, and subsequent hoof wall resection; CT findings and method of keratoma delineation; perioperatively administered medications; and method of keratoma removal. Collected postoperative data included duration of hospitalization, method used for hoof support, medications administered, and incidence of postoperative complications. Complications were defined as excessive granulation tissue formation at the surgical site, surgical site infection, persistent abnormal hoof growth after surgery, persistent increased lameness, and keratoma regrowth.
On initial evaluation, a complete physical examination and brief lameness evaluation had been performed for each equid. Lameness was graded on a 6-point scale, with 0 representing no detectable lameness and 5 representing non–weight-bearing lameness, on the basis of the lameness grading system of the American Association of Equine Practitioners.6
CT and surgical technique
In preparation for surgery, feed but not water was withheld for approximately 8 hours prior to anesthesia. A 14-gauge, 5.24-inch catheter with a nonstick coating was inserted into the left jugular vein by use of aseptic technique. Thirty minutes before anesthetic induction, each equid was premedicated with penicillin G procaine (22,000 U/kg [10,000 U/lb], IM) or ceftiofur sodium (2.2 mg/kg [1.0 mg/lb], IV), gentamicin sulfate (6.6 mg/kg [3.0 mg/lb], IV), phenylbutazone (4.4 mg/kg [2.0 mg/lb], IV), and tetanus toxoid (IM). Equids were sedated with xylazine hydrochloride (1.1 mg/kg [0.5 mg/lb], IV) with, in most cases, butorphanol tartrate (0.01 mg/kg [0.005 mg/lb], IV) or morphine sulfate (0.1 mg/kg [0.05 mg/lb], IV) at the discretion of the attending anesthesiologist. Anesthesia was then induced with a combination of ketamine hydrochloride (2.2 mg/kg, IV) and midazolam hydrochloride (0.05 mg/kg [0.02 mg/lb], IV) or diazepam (0.05 mg/kg, IV). After induction, orotracheal intubation was performed, equids were mechanically ventilated, and general anesthesia was maintained with isoflurane in oxygen.
Equids were positioned on a custom-built CT tablea in lateral recumbency on the side of the affected limb. The contralateral thoracic or pelvic limb was tied in a fully flexed position to provide clear access to imaging of the affected limb. At the discretion of the attending surgeon, intra-arterial contrast enhancement was used to evaluate vascularity in the region of the keratoma by means of a previously established technique.7 Contiguous slices (0.6 mm each) were obtained from the distal aspect of the proximal phalanx to the solar surface of the hoof. Computed tomographic images were obtained with 1 scannerb prior to June 2011 and then with a different scannerc afterward. The CT images were evaluated at the time of acquisition by the attending radiologist, and the height and width of the keratoma were measured with the aid of integrated diagnostic imaging softwared (Figure 1), with margins subsequently etched into the overlying hoof wall by use of electrocautery.
Representative CT images of a keratoma located at the dorsolateral aspect of the right forefoot of a 19-year-old mule gelding in transverse (A) and sagittal (B) planes showing measurements of the height and width of the keratoma for surgical planning (double-headed arrows). In the transverse plane, the keratoma is visible as a projection contiguous with the overlying hoof, causing pressure necrosis of the distal phalanx. In the sagittal plane, the height of the keratoma is apparent and a defect (arrow) is visible in the sole, representing the location of the distal aspect of the lesion.
Citation: Journal of the American Veterinary Medical Association 254, 2; 10.2460/javma.254.2.266
After CT imaging was completed, equids were typically transported to an operating suite for keratoma removal. For surgery, equids were positioned in lateral recumbency such that for dorsally or laterally located lesions, patients were positioned with the affected side on top, or for medially located lesions, patients were positioned with the affected side on the bottom. For all equids, the distal portion of the affected limb was clipped of hair from the distal metacarpophalangeal or tarsophalangeal joint to the coronary band. The hoof wall was rasped, and the sole was pared to remove excess debris, followed by aseptic preparation with 10% betadine solution and isopropyl alcohol. An abaxial sesamoid nerve block was performed by injection of 2% mepivacaine hydrochloride to desensitize the hoof, followed by tourniquet placement at the level of the proximal sesamoid bones.
In some instances, depending on surgeon preference, RLP was performed before surgery began. For RLP, a 23-gauge butterfly catheter was placed in a palmar or plantar digital vein distal to the tourniquet, and 1 g of amikacin sulfate diluted in sterile saline (0.9% NaCl) solution to a total volume of 35 mL was administered. The distal portion of the limb was draped routinely, and partial or complete hoof wall resection was performed with a pneumatic drill with burr attachmente or a reciprocating saw.f When partial hoof wall resection was chosen, a 0.75-inch Galt trephine was used to create a communication between the hoof wall resection site and the sole of the foot, leaving a rim of hoof wall at the distal extent of the hoof wall resection. Briefly, the hoof wall overlying the keratoma was first scored with the burr or reciprocating saw along the margins of the proposed site of resection, as outlined during CT. These cuts were then extended through the total thickness of the hoof wall, to the level of the sensitive lamina. Osteotomes were then placed in the lateral and medial margins of the surgically created defect in the hoof wall and used to pry the section of hoof wall free from the underlying tissues to expose the keratoma (Figure 2). In some instances, the keratoma was strongly adhered to the underlying hoof wall and was attached to the deep side of hoof wall that had been resected. After the hoof wall overlying the keratoma had been removed, sharp debridement with scalpels and Burns curettes was performed to remove all abnormal tissue, exposing the underlying distal phalanx.
Intraoperative (A) and postoperative (C) photographs of the affected foot of the equid in Figure 1 as well as lateral photograph of the resected hoof wall and underlying keratoma (B). The approximate junction between the hoof wall and keratoma formation is shown in panel B (arrows).
Citation: Journal of the American Veterinary Medical Association 254, 2; 10.2460/javma.254.2.266
For 2 equids, keratoma removal was performed at a later time with the patient standing and sedated following recovery from anesthesia for the CT examination. In these instances, equids were restrained in standing stocks and a loading dose of detomidine hydrochloride (0.01 mg/kg) combined with butorphanol tartrate (0.01 mg/kg) was administered IV. Sedation was maintained with a continuous IV infusion of detomidine hydrochloride (0.02 to 0.05 mg/kg/h [0.01 to 0.02 mg/lb/h]). An abaxial sesamoid nerve block was performed by injection of 2% mepivacaine hydrochloride to desensitize the hoof prior to keratoma removal. Hoof preparation and surgical technique were the same as described for equids in which the keratoma was removed under general anesthesia via partial hoof wall resection.
When surgery was completed and prior to tourniquet removal, the hoof wall defect was packed with antimicrobial-impregnated roll gauzeg or hypertonic saline (20% NaCl) wound dressing.h After the defect had been packed, a pressure bandage comprised of roll cottoni secured with gauzej was applied to the foot, followed by adhesive bandage material.k This initial bandage was subsequently left in place for 24 hours. Finally, a waterproof patch constructed from duct tape was applied to the solar surface of the bandage. Equids were then left to recover from anesthesia or sedation unassisted or assisted by use of head and tail ropes.
Postoperative care
After surgery, equids received antimicrobial and NSAID injections, including penicillin G procaine (22,000 U/kg, IM, q 12 h) or ceftiofur sodium (2.2 mg/kg, IV, q 12 h), gentamicin sulfate (6.6 mg/kg, IV, q 24 h), and phenylbutazone (2.2 to 4.4 mg/kg, IV, q 12 h). Most equids were transitioned to orally administered antimicrobials at the discretion of the attending surgeon prior to hospital discharge. Equids were treated with minocycline (4.0 mg/kg [1.8 mg/lb], q 12 h), sulfamethoxazole-trimethoprim (30 mg/kg [13.6 mg/lb], q 12 h), doxycycline (10 mg/kg [4.5 mg/lb], q 12 h), or chloramphenicol (50 mg/kg [22.7 mg/lb], q 8 h). Additional RLPs were provided for certain equids as previously described. In most situations, prior to hospital discharge, equids had foot support applied to the affected digit, consisting of a hospital plate with or without a metal plate bridging the hoof wall defect (Figure 2) and attached to the adjacent hoof wall with screws or a cuff shoe.l
Equids were discharged from the hospital with recommendations for confinement to a box stall and treatment with anti-inflammatory and antimicrobial medications as clinically indicated. Routine wound management continued as granulation and cornification of the hoof wall defect progressed. Continued hoof support was recommended until the hoof wall defect had completely grown out.
Outcome
Follow-up information was obtained from the medical records and via telephone conversation between equid owners or referring veterinarians and an investigator (SAK), with the aid of a questionnaire. Collected information included postoperative complications associated with healing of the surgical site during hospitalization or following hospital discharge, time elapsed between surgery and subsequent epithelialization of the surgical site, and time elapsed between surgery and regaining of preoperative activity level.
Statistical analysis
Descriptive statistics were computed. Continuous data such as age and durations are reported as mean ± SD (range).
Results
Animals
Thirty-two equids (24 geldings and 8 mares) met the inclusion criteria for the study. Mean ± SD age was 13.9 ± 5.0 years (range, 5 to 22 years). The group included 9 warmbloods, 3 Thoroughbreds, 3 Quarter Horses, 3 mules, 2 Arabians, 2 Friesians, 2 Tennessee Walking Horses, and 1 each of American Paint Horse, Belgian, Missouri Fox Trotter, Morgan, Percheron, Quarter Horse–Arabian cross, Standardbred, and Thoroughbred–Quarter Horse cross. One equid had 2 teratomas identified in the right forefoot, 1 had 2 keratomas identified in the left hind foot, and 1 had bilateral forefoot keratomas identified, resulting in the total of 35 keratomas in 33 feet of 32 equids. The right forefoot was most commonly affected (n = 15), followed by the left forefoot (7), left hind foot (5), right hind foot (4), and both forefeet (1; horse A). One equid (with 2 keratomas in the right forefoot) died during hospitalization for reasons unrelated to keratoma removal but was included in the summary data to that point.
Twenty-two equids had radiographs of the affected foot available from the referring veterinarian, and 9 had radiography performed after hospital admission. The remaining equid had been referred specifically for MRI of the affected foot, and therefore, no radiography was performed. For the 9 equids with radiography performed after admission, results indicated a characteristic semicircular defect along the solar margin of the distal phalanx of the affected limb, suggestive of keratoma. Results of lameness evaluation indicated that 23 equids had no signs of lameness while walking, 8 had lameness while walking, and 1 had no lameness while walking but had grade 3 lameness while trotting.
Reasons for referral
The most common reason for referral was recurrent foot abscesses (n = 28 [88%]). Of the 28 affected equids, 20 had radiography performed by the referring veterinarian prior to referral, revealing a semicircular radiolucent defect along the solar margin of the distal phalanx, suggestive of keratoma. The other 8 equids referred for recurrent foot abscesses had no radiography performed prior to referral. Reasons for referral for the remaining 4 equids included chronic lameness with evidence of keratoma on radiographs obtained by the referring veterinarian (n = 2); proliferative mass at the dorsomedial aspect of the digit, immediately proximal to the coronary band (1); and suspected medial collateral ligament desmitis of the distal interphalangeal joint of the affected limb (1). The equid with suspected desmitis was the one referred for MRI of the affected foot.
CT examination
Mean ± SD duration of CT scanning for all 32 equids (33 feet) was 30.2 ± 13.1 minutes (range, 16 to 65 minutes). For all 35 keratomas, CT revealed a soft tissue mass isointense to and contiguous with the overlying keratinized hoof wall. Mean height of the keratomas was 6.1 ± 1.8 cm (range, 3.0 to 11.0 cm), and mean width was 2.0 ± 0.5 cm (range, 1.0 to 3.0 cm). In all 33 affected feet, an adjacent semicircular, well-circumscribed defect was observed at the solar margin of the distal phalanx, secondary to impingement of the mass. In 2 affected feet (1 right forefoot and 1 left hind foot), 2 keratomas were identified, resulting in a total of 35 keratomas identified via CT. In one situation, the second keratoma was initially discovered on CT and had not been evident on preoperative radiographs because it was not causing pressure resorption of the adjacent distal phalanx; in the other situation, 2 keratomas had initially been suspected following radiography of the affected foot, and these 2 keratomas were confirmed via CT.
For the equid with 2 affected feet (horse A), bilateral distal extremity CT was performed owing to evidence of pressure resorption of the distal phalanx of both forefeet on referral radiographs, and CT confirmed the presence of bilateral keratomas. For the equid referred because of a proliferative mass dorsoproximomedial to the coronary band, CT revealed that the mass was continuous with a soft tissue mass underlying the hoof wall distally, which was suspected to be a keratoma.
Nine of 32 (28%) equids had intra-arterial contrast-enhanced CT performed. In all 9 horses, contrast enhancement revealed that the keratoma was devoid of blood supply and the laminar tissue deep to the keratoma had low contrast enhancement, consistent with vascular compression and loss of laminar perfusion at the site where the keratoma impinged on the distal phalanx.
Surgical technique
For the 30 equids that underwent CT followed by surgical removal during the same anesthetic episode, mean ± SD duration of anesthesia was 147.8 ± 23.7 minutes (range, 75 to 190 minutes). Mean duration of surgery for those equids was 73.7 ± 19.2 minutes (range, 35 to 120 minutes). One of the 30 equids that underwent keratoma removal during the same anesthetic episode was the case with bilateral forefoot keratomas (horse A). This equid had the left forefoot keratoma removed initially while anesthetized and the second keratoma removed from the contralateral forefoot while standing and sedated (duration of surgery, 50 minutes) 5 months later. Of the 2 remaining equids, 1 (a mule) was allowed to recover from general anesthesia after CT imaging (duration of anesthesia, 40 minutes), and the keratoma was removed 25 days later during a subsequent anesthetic episode. For keratoma removal, duration of anesthesia was 60 minutes, and that of surgery was 45 minutes. The final horse had CT performed, recovered from general anesthesia, and had a right forefoot keratoma removed while standing and sedated later the same day. For this equid, total duration of anesthesia for CT imaging was 50 minutes, and that for surgery for keratoma removal was 50 minutes.
For 30 of 32 (94%) equids, the keratomas were removed through partial hoof wall resection (total, 33 keratomas from 31 feet) and, for 2 (6%) horses, through complete hoof wall resection (total, 2 keratomas from 2 feet). Hoof wall was removed with a reciprocating saw in 23 horses and with a pneumatic drill with burr attachment in 9 equids. For the 2 equids with 2 keratomas in 1 foot, both keratomas were removed following CT imaging during the same anesthetic episode. In one of these equids, the keratomas were removed through 2 adjacent partial hoof wall resections; in the other, 1 partial hoof wall resection was adequate to remove both keratomas.
For the equid referred because of a proliferative mass that extended proximal to the coronary band, partial hoof wall resection was performed as previously described. A tunnel was then created under the remaining proximal portion of the hoof capsule beginning at the proximal aspect of the hoof wall resection and extending proximally. Remaining abnormal tissue extending proximal to the coronary band was debrided through a combination of blunt and sharp dissection, with care to avoid disrupting the coronary band or distal interphalangeal joint.
Two equids had keratoma removal performed through partial hoof wall resection while standing and sedated rather than while anesthetized for CT examination. These equids included horse A, and although the keratoma in the left forefoot was removed during the same anesthetic episode as CT image acquisition and the surgical wound was allowed to heal, the keratoma in the right forefoot was removed 5 months later with the equid sedated and standing. Duration of surgery for the second procedure in horse A was 50 minutes. The other equid underwent standing removal of a keratoma from its right forefoot performed on the same day as CT imaging, after recovery from general anesthesia. Duration of surgery in this instance was also 50 minutes.
Seventeen (53%) equids (18/33 [55%] affected limbs) received 1 g of amikacin via RLP at the time of surgery. Thirty (94%) equids had the surgically induced hoof wall defect packed with antimicrobial-impregnated roll gauze, and 2 equids first had metronidazole paste applied to the defect. In the remaining 2 (6%) equids, a hypertonic saline wound dressing and dry sterile gauze sponges had been used to pack the defect.
Postoperative care
For 31 equids (32 surgical procedures) that survived to hospital discharge, the mean ± SD hospitalization period following keratoma removal was 4.7 ± 2.1 days (range, 0 to 10 days). This included the equid (horse A) with 1 keratoma removed from each forefoot on separate occasions, with the second procedure performed on an outpatient basis. The equid that died prior to hospital discharge for reasons unrelated to keratoma removal was excluded from this calculation.
All equids received systemic antimicrobial and phenylbutazone treatments before and after surgery. All 32 equids (33 surgical procedures) received injectable antimicrobials and phenylbutazone before surgery, with 32 (100%) equids continuing to be treated with these injectable medications in the initial postoperative period. The mean ± SD treatment duration was 2.5 ± 1.2 days (range, 1 to 5 days). For removal of the contralateral forelimb keratoma on an outpatient basis (33rd procedure), horse A received 1 dose of injectable antimicrobials and phenylbutazone before surgery and was transitioned to oral antimicrobial and phenylbutazone treatment the same day after surgery. For 29 equids (30 surgical procedures, including both forefoot keratoma removals for horse A), antimicrobial administration was transitioned from injectable to oral prior to hospital discharge. The mean ± SD duration of oral medication administration was 13.7 ± 2.7 days (range, 7 to 21 days). Orally administered antimicrobials included minocycline (n = 14), sulfamethoxazole-trimethoprim (12), doxycycline (3), and chloramphenicol (1). For 3 equids, no transition to oral antimicrobial treatment was performed. This included 2 equids that received injectable antimicrobials for 5 days during hospitalization and the equid that died prior to hospital discharge.
Five equids had daily RLP performed after surgery. For 3, RLP had also been performed at the time of surgery, and afterward it was performed once daily for 2 consecutive days. The remaining 2 equids received no RLP at time of surgery, and once-daily RLP was performed after surgery for 3 or 4 consecutive days.
Initially, postoperative foot support was provided for 29 equids (30 feet) with a foot bandage alone, which was changed daily or every other day for a mean ± SD duration of 3.5 ± 1.7 days (range, 1 to 7 days). For the remaining 3 feet, 2 equids had a cuff shoe placed and 1 equid had a hospital plate with dorsal patch placed on completion of surgery, prior to recovery from anesthesia. For these 3 equids, the applied foot support was incorporated into the initial foot bandage.
For 30 feet, foot support was applied to the affected limb prior to hospital discharge (hospital plate with [n = 19] or without [5] a metal plate bridging the hoof wall defect or a cuff shoe [6]). For the 2 remaining keratomas that were removed from equids that survived to hospital discharge (2 feet), only a foot bandage was placed. This included horse A following standing removal of the keratoma from the right forefoot. In this instance, a hospital plate with dorsal patch was applied for 3 days after hospital discharge under the supervision of the referring veterinarian. The second horse was discharged in a foot bandage following removal of a keratoma from the right forefoot and had a hospital plate with dorsal patch applied 5 days following discharge, also under the supervision of the referring veterinarian.
Outcome
Follow-up information was available for all 31 equids (33 keratomas removed from 32 feet) that were initially discharged from the hospital. Data regarding the interval from keratoma removal to complete cornification of the surgical site were available for 29 equids, for a mean ± SD time of 3.5 ± 1.2 months (range, 2 to 6 months). The 2 equids for which this information was unavailable included 2 horses with suspected keratoma regrowth due to incomplete excision.
Of the 31 equids that survived to hospital discharge, 4 (13%) developed complications associated with healing of the surgical site. All 4 equids had undergone partial hoof wall resection for keratoma removal. This included 1 horse that developed excessive granulation tissue at the surgical site, 1 horse that developed excessive granulation tissue at the surgical site and transient worsening of lameness associated with subsolar abscessation, and the 2 equids with suspected keratoma recurrence.
The equid with excessive granulation tissue as a solitary complication had had a single keratoma removed from the right forefoot. Sixteen days after hospital discharge, this equid was returned for reexamination of the surgical site on an outpatient basis. The abnormal granulation tissue was debrided with the equid standing and sedated, and complete healing of the surgical site ensued without further complication.
The equid that developed excessive granulation tissue and subsolar abscessation had had 2 keratomas removed concurrently from the left hind foot through 2 adjacent partial hoof wall resections. Because of ongoing lameness, development of excessive granulation tissue, and subsolar abscessation, the equid was readmitted to the hospital 44 days following initial hospital discharge. Local wound care was initiated, and foot support was maintained in the form of a hospital plate with a dorsal metal plate attached to the hoof wall spanning both defects. Oral phenylbutazone treatment was provided, the dose and frequency of which were based on apparent comfort. The equid was discharged after an additional 34 days of hospitalization without further complication.
One equid with presumed incomplete keratoma excision and subsequent regrowth had incomplete healing of the surgical site and developed a persistent solar draining tract after surgery. One hundred fifty days following keratoma removal, additional abnormal tissue was debrided from the draining tract while the equid was standing and sedated, but no tissue sample was submitted for histologic examination.
The last of the 4 equids had delayed healing of the surgical site with suspected keratoma regrowth. This equid was the one initially referred because of a mass immediately proximal to the coronary band and had returned 27 days after initial hospital discharge, when further debridement was performed with the horse anesthetized. At surgery, the distal interphalangeal joint was inadvertently penetrated. The joint was lavaged with sterile polyionic fluids, and the horse recovered from anesthesia. Histologic examination of the excised tissues revealed no evidence of keratoma regrowth. The equid remained hospitalized for 18 days, during which it received systemic antimicrobial and NSAID treatment. Fourteen days after discharge, the equid again had signs of worsening lameness. Arthrocentesis of the distal interphalangeal joint was performed, and cytologic synovial fluid findings were consistent with septic arthritis. Arthroscopic examination of the joint was performed, revealing extensive loss of articular cartilage. The equid was euthanized 24 hours later owing to a poor prognosis for survival.
Overall, 28 of 31 (90%) equids returned to their previous activity level. Mean ± SD interval from surgery to return to previous activity was 8.9 ± 2.9 months (range, 6 to 18 months). The 3 equids that did not return to their previous activity level included the 2 with suspected keratoma regrowth and 1 with uncomplicated healing of the surgical site. This latter equid was euthanized 180 days after keratoma removal from the left forefoot because of chronic persistent lameness attributed to the distal interphalangeal joint of the same limb. The cause of lameness was believed to be a focal subchondral bone defect of the distal phalanx that had been identified on preoperative CT imaging and an adjacent full-thickness articular cartilage defect that was subsequently diagnosed on postmortem examination.
Discussion
Computed tomographic imaging of the feet in equids with a tentative diagnosis of keratoma based on clinical and radiographic findings allowed for accurate localization and measurement of the space-occupying lesion, followed by precise demarcation of the mass on the overlying hoof wall prior to removal via hoof wall resection. This allowed for keratoma removal through the smallest window possible and was associated with a low incidence of postoperative complications (13%) for the evaluated equids.
In the present study, 4 of 31 (13%) equids that survived to hospital discharge developed postoperative complications associated with healing of the surgical site, which was a lower rate than in other reports.2,5 Additionally, the 4 equids that developed complications had keratomas removed through partial hoof wall resections, whereas neither equid that underwent complete hoof wall resection developed complications. Because only 2 equids had complete hoof wall resection performed, no conclusions could be drawn regarding differences in complication rates between the 2 surgical approaches for keratoma removal.
The most common complication for equids in the present study was excessive granulation tissue formation, which has been reported as a complication encountered following keratoma removal in several other studies2,5,8 and is related to hoof wall instability.4
Both equids responded favorably to debridement of excess granulation tissue, and healthy cornification of the tissues followed. However, 1 of those 2 equids had 2 keratomas removed from 1 foot through 2 adjacent partial hoof wall resections and required prolonged hospitalization because of ongoing lameness and hoof wall instability after surgery. To avoid this complication in other equids with 2 keratomas in the same foot, it may be prudent to remove 1 keratoma, allow the surgical site to heal, and remove the second keratoma in a subsequent surgery to minimize consequent loss of laminar support to the distal phalanx.
Two equids had suspected keratoma regrowth as a complication. Although this was not confirmed via histologic examination for either equid, regrowth is a reported as a complication following keratoma removal2 and is suggested to more likely occur when no preoperative MRI or CT examination has been performed.5 In 1 equid, tissue excised from a persistent draining tract that developed at the site of previous keratoma removal was not submitted for histologic evaluation. For the second equid, excised tissues from a supracoronary site of previous keratoma removal were submitted, but no evidence of keratoma was identified in the provided samples. Efforts to debride all tissue that appeared unhealthy in this second equid led to inadvertent penetration of the distal interphalangeal joint. Despite aggressive medical and surgical management, the equid was eventually euthanized because of persistent synovial sepsis and a poor prognosis for survival.
In the study reported here, 28 of 31 (90%) equids initially discharged from the hospital returned to their previous activity level. This success rate was similar to those in previous reports1,2,5 of outcome for equids following keratoma removal. The mean interval from surgery to return to previous activity was 8.9 months for the equids of the present study. Reported intervals from keratoma removal to return to exercise differ greatly among reports,1,2,5,9–11 ranging from 1.5 months9 to 12 months.1,11 Differences in findings among studies may be related to the location and size of the keratoma as well as surgical approach necessary to facilitate removal. In the present study, equids that did not return to their previous activity level included the 2 with suspected keratoma regrowth and 1 with uncomplicated healing of the surgical site. One equid was retired to pasture, and the remaining 2 were euthanized within 6 months after surgery.
Two equids in the present study had 2 keratomas in 1 foot. Preoperative radiography revealed 2 semicircular radiolucent defects in 1 equid in the affected distal phalanx with subsequent confirmation of 2 keratomas on CT examination. In the other equid, CT revealed a second keratoma in 1 foot, which had not been recognized on preoperative radiographs because it was not yet impinging on the distal phalanx. Surgical removal of multiple keratomas from a single foot of a horse and subsequent histologic diagnosis have been reported.8 Another description12 of surgical removal of multiple keratomas from a single foot has been reported, but no histologic examination of the excised tissue was performed, so a definitive diagnosis of keratoma was not possible. These reports and the case series of the present report confirmed the possibility that multiple keratomas can occur in a single foot concurrently, and both keratomas may or may not impinge on the distal phalanx resulting in radiographic evidence of pressure resorption consistent with keratoma. Most importantly, the findings highlighted the advantages of preoperative CT not only for preoperative planning for removal of suspected keratomas identified radiographically but also to aid in identification of additional keratomas that may not be radiographically evident.
Nine of 32 (28%) equids in the present study underwent intra-arterial contrast medium administration during CT imaging for vascular enhancement. To the authors’ knowledge, this was the first reported use of contrast-enhanced CT to further define the imaging characteristics of keratomas in equids. In all 9 equids, contrast-enhanced CT revealed that the lamina deep to the keratoma had low contrast enhancement owing to compression of and loss of vascular perfusion, and keratomas appeared to be devoid of blood supply. Furthermore, contrast-enhanced CT may be useful in differentiating keratomas from other forms of neoplasia requiring hoof wall resection for removal, planning a surgical approach, or determining prognosis.
In a previous study,5 horses that underwent surgical keratoma removal returned to their previous activity level a mean of 3.6 months after surgery, whereas equids in the present study returned a mean of 8.9 months after surgery. The difference between these findings may have been related to the different methods of postoperative foot support provided but was more likely related to the size of keratomas removed. In the previous study,5 mean keratoma size was 1.6 × 3.3 cm, whereas in the present study, it was 2.0 × 6.1 cm. The larger keratoma size and size of subsequent hoof wall resections necessary for removal would result in more time required for adequate hoof regrowth to occur and inherent hoof stability to be reestablished prior to returning to full athletic activity.
Interestingly, a large proportion of equids in the present study (24/32 [75%]) were geldings, as identified in previous studies,1,2,5 although no sex predilection has been reported for keratoma development. Additionally, more forefeet than hind feet were affected in the present study, which was also consistent with previous findings.2,5 Considered together, these findings suggested that keratoma should be considered as a likely diagnosis in geldings with nonspecific clinical signs such as recurrent foot abscesses, varying degrees of lameness, and distortion of the hoof capsule affecting a forelimb.
No extensive lameness examination was deemed necessary for the equids in the present study because for most (31/32), suspicions were high at initial evaluation that they had a keratoma given the anamnesis and radiographic evidence of a characteristic radiolucent defect located along the solar margin of the affected limb. Although no equids in the present study had follow-up CT performed, a small number had follow-up radiographs obtained by referring veterinarians that revealed persistence of the characteristic semicircular lucency observed before surgery. Persistence of this lucency in the equids for which follow-up radiographs were obtained was considered an incidental finding given that none of those equids had clinical evidence of keratoma regrowth.
The present study was limited by its retrospective nature, lack of complete preoperative lameness examinations performed, and lack of follow-up imaging. The retrospective nature precluded control of preoperative, intraoperative, and postoperative variables. Nonetheless, this study represented the largest to date reporting outcome for equids following keratoma removal and the use of CT in the preoperative planning for removal of these space-occupying lesions through hoof wall resection. Findings suggested that preoperative CT imaging can be useful for this purpose and that the techniques used can provide a low complication rate and excellent prognosis for return to athletic activity for equids with keratomas. Computed tomography allowed for identification of other suspected keratomas that may not have been clinically evident or identified with other imaging modalities. Findings also suggested that contrast-enhanced CT may allow for differentiation between keratomas and other space-occupying lesions as well as identification of preexisting lesions within the feet that may influence postoperative return to function.
ABBREVIATIONS
| RLP | Regional limb perfusion |
Footnotes
Large Animal Peters Carbon Fiber CT Table, Kimsey Welding Works, Woodland, Calif.
GE HiSpeed helical scanner FX/I, GE Medical Systems, Waukesha, Wis.
16-slice GE Lightspeed scanner, GE Medical Systems, Waukesha, Wis.
eFilm workstation, Merge Healthcare, Hartland, Wis.
Hall Surgairtome, Conmed, Largo, Fla.
Ridgid JobMax 3-amp multi-tool, Ridge Tool Co, Elyria, Ohio.
Kerlix gauze bandage roll, Covidien, New Haven, Conn.
Curasalt hypertonic saline dressing, Covidien, New Haven, Conn.
Rediroll, The Franklin-Williams Co, Lexington, Ky.
Brown cling gauze, Jorgensen Laboratories Inc, Loveland, Colo.
Elastikon, Johnson & Johnson, Skillman, NJ.
Dalric special cuff shoe, Nanric, Lawrenceburg, Ky.
References
1. Lloyd KCK, Peterson PR, Wheat JD, et al. Keratomas in horses: 7 cases (1975–1986). J Am Vet Med Assoc 1988;193:967–970.
2. Boys Smith SJ, Clegg PD, Hughes I, et al. Complete and partial hoof wall resection for keratoma removal: post operative complications and final outcome in 26 horses (1994–2004). Equine Vet J 2006;38:127–133.
3. Honnas CM, Dabareiner RM, McCauley BH. Hoof wall surgery in the horse: approaches to and underlying disorders. Vet Clin North Am Equine Pract 2003;19:479–499.
4. Honnas CM. Keratomas of the equine digit. Equine Vet Educ 1997;9:203–207.
5. Getman LM, Davidson EJ, Ross MW, et al. Computed tomography or magnetic resonance imaging-assisted partial hoof wall resection for keratoma removal. Vet Surg 2011;40:708–714.
6. American Association of Equine Practitioners. Definition and classification of lameness. Guide for veterinary service and judging of equestrian events. Lexington, Ky: American Association of Equine Practitioners, 1991;19.
7. Puchalski SM, Galuppo LD, Hornof WJ, et al. Intraarterial contrast-enhanced computed tomography of the equine distal extremity. Vet Radiol Ultrasound 2007;48:21–29.
8. Rowland AL, Goodrich LR, Bass LD, et al. Diagnosis and treatment of biaxial keratomas in the left fore foot of a 15-year-old Clydesdale mare. Equine Vet Educ 2016;28:310–314.
9. O'Grady SE, Horne PA. Lameness caused by a solar keratoma: a challenging differential diagnosis. Equine Vet Educ 2001;13:87–89.
10. Seahorn TL, Sams AE, Honnas CM, et al. Ultrasonographic imaging of a keratoma in a horse. J Am Vet Med Assoc 1992;200:1973–1974.
11. Honnas CM. Supracoronary approach for keratoma removal. Equine Vet Educ 2011;23:494–495.
12. Christman C. Multiple keratomas in an equine foot. Can Vet J 2008;49:904–906.
