A 19-year-old Thoroughbred gelding (horse 1) with a 3-year history of chronic nasal discharge that was unresponsive to systemic treatment with antimicrobials was referred for evaluation and treatment. Prior to referral, guttural pouch mycosis was diagnosed by the referring veterinarian by means of endoscopic examination and cytologic observation of fungal hyphae in samples of plaques in the guttural pouch (auditory tube diverticulum).
Initial endoscopic examination at our hospital confirmed the presence of a fungal plaque on the dorsomedial aspect of the medial compartment of the right guttural pouch involving the right ICA with suspected secondary bacterial infection on the basis of endoscopic appearance, results of gram staining of aspirates, and presence of purulent nasal discharge. Endoscopic images were suggestive but not conclusive of branching of the ICA at the distal to mid level of the guttural pouch where the ICA was not obscured by scar tissue or fungal plaque (Figure 1). No signs of hemorrhage or neurologic deficits commonly associated with guttural pouch mycosis, such as dysphagia, Horner syndrome, or laryngeal neuropathy were present at admission.
Treatment by means of transarterial coil embolization was planned. A jugular catheter was placed, and perioperative medications consisted of tetanus toxoid (IM, once), potassium penicillin (22,000 IU/kg [10,000 U/lb], IV, q 6 h), gentamicin sulfate (6.6 mg/kg [3.0 mg/lb], IV, q 24 h), and flunixin meglumine (1.1 mg/kg [0.45 mg/lb], IV, q 12 h) for 24 hours, then phenylbutazone (2.2 mg/kg [1 mg/lb], PO, q 12 h) for the remainder of the hospitalization. The horse was premedicated with xylazine (1.1 mg/kg, IV). General anesthesia was induced with midazolam (0.05 mg/kg [0.023 mg/lb], IV) and ketamine (2.2 mg/kg, IV), an endotracheal tube was placed, and anesthesia was maintained with isoflurane in oxygen. The horse was then positioned in lateral recumbency with the affected right side up, and standard anesthesia monitoring was performed. The head and neck were positioned on a plexiglass table to allow fluoroscopya during the procedure. The surgical approach and transarterial coil embolization technique for the affected ICA were performed as previously described.1 Digital subtraction angiography by means of injection of iohexol confirmed the presence of a separate site of origin for the right occipital artery and right ICA (Online Supplement Video 1 available at http://avmajournals.avma.org/toc/javma/247/12). Selective digital subtraction angiography of the horse's vasculature confirmed the presence of a bifurcation of the right ICA (Figure 2).
On angiography, the right ICA was noted to have an aberrant branch or bifurcation at the caudoventral aspect of the guttural pouch, with both branches being of comparable diameter. After a single ICA origin from the common carotid artery, the ICA bifurcated with the aberrant branch originating more caudally than the typical ICA with both vessels overlapping one another in the lateral imaging plane. On correlating the endoscopic fungal plaque location and suggestive presence of branching of the ICA inside the pouch to the angiographic vascular anatomy, we were suspicious of fungal involvement of the aberrant or caudal branch of the ICA. The decision was made to embolize both ICA branches. Stainless steel embolization coilsb covered with polyethylene terephthalate fibers were deployed in the distal aspects of both ICA branches and proximal to the bifurcation (Figure 3; Online Supplement Video 2 available at http://avmajournals.avma.org/toc/javma/247/12). For this report, the term distal refers to a location in the vessel farther from the heart, whereas proximal refers to a location closer to the heart. Throughout the procedure, the delivery catheter was flushed frequently with heparinized saline (0.9% NaCl) solution, and iodinated contrast was administered as necessary to assure correct coil position and to confirm cessation of bloodflow in both ICA branches. Preoperative endoscopy indicated that the ECA and maxillary artery were unaffected by the fungal plaque, and there were no abnormalities in these vessels on angiography; therefore, they were not embolized.
Coil sizes selected and placed included an 8-cm × 8-mm and a 5-cm × 10-mm coil in the distal aspect of the rostral or true ICA, a 5-cm × 10-mm coil in the distal aspect of the caudal or aberrant branch, and a 5-cm × 10-mm coil in the proximal aspect of the ICA just prior to the branching or bifurcation. The distal coils were placed at the level of the basisphenoid bone matching the dorsal aspect of the guttural pouch, whereas the proximal coils were placed at the caudal aspect of the pouch after the ICA branched off the common carotid artery.
No complications were observed during the procedure. The horse had a mild episode of paddling and nystagmus during recovery that resolved spontaneously; further recovery was uneventful, and these signs were assumed to be associated with normal recovery behavior.
Postoperatively, the affected right guttural pouch was treated topically with enilconazole (5 to 8 mL diluted in 50 mL of saline solution), penicillin G procaine (20 mL), and transendoscopic gentle debridement of the fungal plaque as an adjuvant treatment for 2 days. The horse was discharged 3 days after admission without complications other than a small corneal ulcer of the left eye associated with recovery from anesthesia that was treated with a triple antibiotic ointment administered topically and had resolved at the time of discharge.
At telephone follow-up 10 months after surgery, the patient had returned to its intended use as a trail riding horse. The owner reported that at recheck endoscopy with the referring veterinarian 2 months after surgery, the fungal plaque was completely resolved, and no nasal discharge had been observed.
A 15-year-old Thoroughbred gelding (horse 2) with a 3-month history of nasal discharge unresponsive to antimicrobials was referred for treatment. Prior to referral, left guttural pouch mycosis was diagnosed by the referring veterinarian by means of endoscopic examination and cytologic observation of fungal hyphae after sampling the plaques in the guttural pouch. No signs of hemorrhage or neurologic deficits were present. Endoscopic examination at our institution confirmed the presence of a fungal plaque on the dorsomedial aspect of the medial compartment of the left guttural pouch involving the ICA only; secondary bacterial infection was also suspected on the basis of results of endoscopy, results of gram staining of aspirates, and presence of purulent nasal discharge. As for horse 1, endoscopic images were suggestive but not conclusive of branching of the ICA at the distal to mid level of the medial compartment of the guttural pouch.
Treatment with transarterial coil embolization was planned, and the same perioperative protocol was used as for horse 1, with the exception that flunixin meglumine (1.1 mg/kg, IV, q 12 h) was administered for 72 hours versus 24 hours. The surgical approach and procedure were exactly as described for horse 1.1 Angiography of the ICA confirmed the presence of a separate site of origin for the left occipital and ICA arteries and the presence of a bifurcation of the left ICA (Figure 2; Online Supplement Video 3 available at http://avmajournals.avma.org/toc/javma/247/12).
In this patient, the angiographic appearance of the bifurcated ICA was similar to that described for horse 1. Stainless steel embolization coilsb covered with polyethylene terephthalate fibers were deployed in the distal aspects of both ICA branches and proximal to the ICA bifurcation (Figure 3; Online Supplement Video 4 available at http://avmajournals.avma.org/toc/javma/247/12). The distal aspect of the rostral or true ICA was embolized with an 8-cm × 8-mm and a 6-cm × 8-mm coil, an 8-cm × 8-mm coil was placed in the distal aspect of the caudal or aberrant branch, and 2 coils (8 cm × 10 mm and 6 cm × 8 mm) were placed in the ICA just prior to the branching or bifurcation.
No complications were observed during the procedure. However, the horse showed pronounced paddling and incoordination during recovery from anesthesia and required several attempts to stand. On standing, the horse was markedly ataxic with a dropped left ear and muzzle deviation to the right. The horse was placed in a padded stall overnight and treated with dexamethasone sodium phosphate (0.1 mg/kg [0.045 mg/lb], IV) and hypertonic saline (7.2% NaCl) solution (2 mL/kg [0.9 mg/lb], IV). An additional dose of dexamethasone was administered 24 hours later.
The next morning, neurologic examination revealed severe vestibular ataxia; conscious and unconscious proprioceptive deficits in all 4 limbs; decreased left-sided facial and nasal septum sensation; hemiparesis of the left ear, eyelids, and muzzle; base-wide stance on the hind limbs; and left-sided limb spasticity with right-sided limb weakness.2 These signs were compatible with inadvertent coil embolization of the left caudal cerebellar artery and secondary effects to the nuclei of cranial nerves V, VII, and VIII (Figures 3 and 4; Online Supplement Videos 3 and 4 available at http://avmajournals.avma.org/toc/javma/247/12). A lateral skull radiograph obtained 2 days after surgery ruled out coil migration when compared with fluoroscopic images obtained at the completion of the surgical procedure. Postoperatively, the affected guttural pouch was treated topically for 3 days as described for horse 1.
Over the course of hospitalization, the neurologic signs improved progressively, and the horse was discharged 1 week after admission. At that time, the vestibular ataxia was assessed as moderate, facial and trigeminal nerve deficits were moderately improved, and the right-sided weakness was markedly improved but still present.2 After 4 weeks of restricted exercise, the horse was readmitted for a follow-up evaluation. Endoscopic examination of the affected guttural pouch revealed complete resolution of the mycosis and normal laryngeal function. Neurologic examination revealed mild vestibular ataxia, absence of trigeminal nerve deficits, and marked improvement of facial nerve deficits.2
At telephone follow-up 7 months after the procedure, there were no cranial nerve V, VII, or VIII deficits, and for the previous 5 months, the horse had been ridden at a low level of exercise (trail riding and low-level jumping). The owner reported that the horse behaved quieter than before surgery and showed mild exercise intolerance.
Discussion
Guttural pouch mycosis is a disease without age, sex, or breed predilection that predominantly affects the ICA and less commonly the ECA or maxillary artery of horses.3 The predominant clinical signs include epistaxis, nasal discharge, dysphagia, Horner syndrome, and laryngeal dysfunction.3 Several fungal species have been isolated from affected horses, but Aspergillus spp appear to have a special predilection for the guttural pouch environment.4 After endoscopic diagnosis, most clinicians recommend surgery rather than medical treatment to prevent fatal complications. Several surgical techniques to eliminate blood flow in the affected vessels have been reported with variable success.1,5–9 Reported techniques include ligation of the common carotid artery, occlusion with balloon-tipped catheters, and transarterial delivery of intravascular stainless steel embolization coils or nitinol plugs.1,5–7 Embolization techniques prevent retrograde and normograde flow in the affected arteries via a minimally invasive technique aided by fluoroscopic guidance1,7 and are currently used in many referral institutions and specialty hospitals, depending on the availability of trained personnel and equipment. As these procedures become more commonplace, potential complications and variant anatomy need to be recognized and reported. The main reported complications associated with transarterial ICA embolization for guttural pouch mycosis are air or clot embolization and implant migration.10
Whereas the vascular anatomy of the equine head and neck is well described, variants in vascular anatomy have been reported, including variable origins of the common occipital artery and ICA, aberrant branching of the ICA or ECA, and aberrant ICA path.1,11–16 Aneurysm formation has also been described for the ICA and the ECA.12 Although mentioned in some reports, there are few angiographic descriptions of aberrant branching or bifurcation of the ICA in the veterinary literature.12–15 In the present report, we provided angiographic findings for this anatomic variant in 2 horses with guttural pouch mycosis. In addition, we described neurologic complications associated with inadvertent coil embolization of the caudal or aberrant ICA branch, with the coil placed too distally (noncardiac side), leading to caudal cerebellar ischemia in one patient (horse 2).
In prior reports,12–15 the caudal or aberrant branch of the ICA was identified as unilateral with the potential to be identified during endoscopy or angiography in horses with or without guttural pouch mycosis, but this was not described in detail. On the basis of the 2 horses described in this report, we suggest that given its size and anatomic pathway, it may be clearer to define this aberrant branch as a bifurcation of the ICA, with a rostral branch traversing toward the cavernous sinus with an S-shaped curve and a caudal branch leading to the caudal brainstem. Once the caudal branch is identified, it has been recommended to occlude it at its origin to prevent retrograde flow to the rostral ICA or mycotic erosion in case of direct contact with the guttural pouch lining.12,14,16
Freeman et al14 reported inadvertent balloon occlusion of an aberrant branch of the ICA during guttural pouch mycosis treatment in a mare. This branch was found to communicate with the basilar artery during necropsy, and the authors suggested that it represents an unusual manifestation of the caroticobasilar artery.14 Bacon et al16 reported a similar case, although the occluded artery was the ICA, which followed an abnormal path before penetrating the cranium rather than an aberrant branch. At necropsy, the ICA was found to join the junction between the basilar and caudal cerebellar artery rather than traverse toward the circle of Willis, and the catheter was inadvertently placed at that aforementioned junction.16
Despite being mentioned in several publications, the anatomic description of these aberrant branches of the ICA is somewhat confusing, and angiographic imaging is lacking. The overlying bony structures of the equine head limit the utility of standard angiography. The use of digital subtraction angiography improves characterization of real-time arterial anatomy and was therefore used in these cases.17 In the patients of the present report, the results of digital subtraction angiography suggested that the caudal or aberrant branch of the ICA originated at the level of the guttural pouch, had a similar pathway as the ICA, and communicated extracranially with the basilar artery via the caudal cerebellar artery.
In horses, the caudal cerebellar artery perfuses portions of the medulla oblongata, pons, and cerebellum and receives a limited number of distal anastamoses.18 The human equivalent of the equine caudal cerebellar artery is the inferior cerebellar artery, which perfuses nuclei of cranial nerves VII and VIII and parts of V (pons) as well as portions of the cerebellum and the spinothalamic tract.19 Human patients suffering a stroke involving this vessel commonly have symptoms including ipsilateral ataxia, contralateral numbness, hearing loss, and ipsilateral facial numbness and weakness. These neurologic symptoms are compatible with the clinical signs observed in horse 2 and reinforce our hypothesis of a communication between the caudal or aberrant branch of the ICA and the caudal cerebellar artery. Neurologic signs associated with the brainstem and cerebellar ischemia were also seen during anesthesia or recovery in the 2 prior reports of horses in which balloon-tipped catheters were inadvertently placed in aberrant arterial branches.14,16
Angiographic images in the 2 horses of this report revealed the presence of the caudal cerebellar artery anastomosing with the caudal or aberrant branch of the ICA after presumptively branching off the basilar artery. In horse 1, review of the angiographic imaging suggested that the distal coil in the aberrant branch was placed just proximal to the caudal cerebellar artery. It is unclear whether the neurologic signs evident during recovery from anesthesia were related to anesthesia or to a transient ischemic event involving the caudal cerebellar artery and related to coil placement. In horse 2, the coil was unintentionally deployed farther distally than for horse 1 and was thought to block blood flow from the caudal cerebellar artery, resulting in neurologic signs compatible with ischemia of the brainstem and cerebellum. Placement of the coil at that level was a surgical error and should be avoided. Histologic alterations compatible with caudal cerebellar artery occlusion were also confirmed in the case reported by Bacon et al.16 We believe that the notable neurologic improvement seen in horse 2 may have been a result of collateral flow or clot dissolution as well as recanalization of the caudal cerebellar artery. Recanalization of coilembolized arteries has been observed histologically as soon as 1 month after surgery.1
It has been suggested that if an aberrant ICA branch is identified, it should be ligated at its origin (proximally).12,14,16 Also, in a prior report of 1 horse, Cheramie et al13 elected to prevent retrograde flow from the aberrant branch to the ICA by occluding the ICA just distal to the branch origin rather than occluding the branch itself. Whereas this was successful for that case, in our 2 patients, we elected to embolize the caudal or aberrant branch farther distally for several reasons. First, results of endoscopic examination suggested the presence of both branches beneath the guttural pouch lining at the distal to mid aspect of the pouch, although examination of the proximal aspect was hindered by the presence of scar tissue and fungal plaque (Figure 1). Second, angiography revealed that both branches were intertwined at the proximal aspect of the guttural pouch where the fungal plaque was present in the pouch, making it difficult to determine confidently whether none, one, or both of the branches were affected by the fungal plaque. Third, in horse 1, the caudal or aberrant branch originated caudally but had a more rostral course, potentially being the only branch affected by the mycotic plaque. Finally, if the caudal or aberrant branch were affected by the mycotic plaque at the dorsomedial aspect of the guttural pouch (distal side of the vessel), epistaxis may not have been prevented with the technique reported by Cheramie et al13 or by embolization of the branch only at its origin. Therefore, we recommend embolization of the caudal branch just proximal to the junction with the caudal cerebellar artery to prevent mycotic erosion of the vessel in cases where the dorsomedial aspect of the guttural pouch is covered by fungal plaque and involvement of both branches is a concern. As in the prior reports,13–16 we also recommend distal and proximal embolization of the rostral or true ICA branch in the standard manner. An important question to be answered is how to determine more accurately whether the caudal or aberrant branch is affected by mycotic invasion and whether the risk for potential neurologic complications with distal coiling has to be assumed. Although in our cases we did not use direct endoscopic observation of the guttural pouch during transarterial coil embolization, we believe that, in addition to preoperative endoscopy and intraoperative angiography, direct endoscopic visualization during treatment can help to better elucidate the presence or absence of each branch within the pouch during treatment.
Lastly, the caudal or aberrant branch of the ICA has been described as an unusual form of the caroticobasilar artery or a direct communication with the basilar artery.14,16 We question this anatomic characterization for several reasons. The caroticobasilar artery anastomoses the ICA and the basilar artery intracranially rather than extracranially as seen in the 2 horses of the present report.12,18 Furthermore, the neurologic signs seen in horse 2 and the necropsy results of the case reported by Bacon et al16 are specific to caudal cerebellar artery ischemia.
To our knowledge, the occasional presence of an aberrant branch of the ICA has been reported previously12–15; however, there is confusing information regarding the specific vascular anatomy, the regions of the nervous system perfused by this branch, and the implications of occluding blood flow in this branch when treating a horse with guttural pouch mycosis. The aim of the present report was to gather information from the literature and our 2 patients to provide clinicians with further knowledge of the anatomy of this vessel and to highlight a possible complication that may be encountered if the vessel is occluded too close to the junction with the caudal cerebellar artery. We acknowledge that further anatomic variations may exist among horses, although the caudal or aberrant branch of the ICA seems to follow a comparable pathway in all reported cases.
ABBREVIATIONS
ECA | External carotid artery |
ICA | Internal carotid artery |
OEC 9900 Elite Cardiac C-Arm, GE Healthcare, Salt Lake City, Utah.
Tornado embolization coil, Cook Medical Inc, Bloomington, Ind.
References
1. Lévéille R, Hardy J, Robertson JT, et al. Transarterial coil embolization of the internal and external carotid and maxillary arteries for prevention of hemorrhage from guttural pouch mycosis in horses. Vet Surg 2000; 29: 389–397.
2. Reed SM. Neurologic exam. J Equine Vet Sci 2003; 23: 484–492.
3. Freeman DE, Hardy J. Guttural pouch. In: Auer JA, Stick JA, eds. Equine surgery. 4th ed. St Louis: Elsevier Saunders, 2012; 623–642.
4. Ludwig A, Gatineau S, Reynaud MC, et al. Fungal isolation and identification in 21 cases of guttural pouch mycosis in horses (1998–2002). Vet J 2005; 169: 457–461.
5. Greet TRC. Outcome of treatment in 35 cases of guttural pouch mycosis. Equine Vet J 1987; 19: 483–487.
6. Freeman DE, Donawick WJ. Occlusion of internal carotid artery in the horse by means of a balloon-tipped catheter: clinical use of a method to prevent epistaxis caused by guttural pouch mycosis. J Am Vet Med Assoc 1980; 176: 236–240.
7. Delfs KC, Hawkins JF, Hogan DF. Treatment of acute epistaxis secondary to guttural pouch mycosis with transarterial nitinol vascular occlusion plugs in three equids. J Am Vet Med Assoc 2009; 235: 189–193.
8. Matsuda Y, Nakanishi Y, Mizuno Y. Occlusion of the internal carotid artery by means of microcoils for preventing epistaxis caused by guttural pouch mycosis in horses. J Vet Med Sci 1999; 61: 221–225.
9. Cheramie HS, Pleasant RS, Robertson JL, et al. Evaluation of a technique to occlude the internal carotid artery of horses. Vet Surg 1999; 28: 83–90.
10. Freeman DE. Complications of surgery for diseases of the guttural pouch. Vet Clin North Am Equine Pract 2008; 24: 485–497.
11. Lepage OM. Hemorrhagie dans les poches gutturales. 1. Anatomie, diagnostic et etiologie. Prat Vet Equine 1994; 24: 255–261.
12. Colles CM, Cook WR. Carotid and cerebral angiography in the horse. Vet Rec 1983; 113: 483–489.
13. Cheramie HS, Pleasant RS, Dabareiner RM, et al. Detachable latex balloon occlusion of an internal carotid artery with an aberrant branch in a horse with guttural pouch (auditory tube diverticulum) mycosis. J Am Vet Med Assoc 2000; 216: 888–891.
14. Freeman DE, Staller GS, Maxson AD, et al. Unusual internal carotid artery branching that prevented arterial occlusion with a balloon-tipped catheter in a horse. Vet Surg 1993; 22: 531–534.
15. Lepage OM, Piccot-Crézollet C. Transarterial coil embolisation in 31 horses (1999–2002) with guttural pouch mycosis: a 2-year follow-up. Equine Vet J 2005; 37: 430–434.
16. Miller BC, Wilson DA, Martin DD, et al. Complications of balloon catheterization associated with aberrant cerebral arterial anatomy in a horse with guttural pouch mycosis. Vet Surg 1998; 27: 450–453.
17. Macdonald DG, Fretz PB, Baptiste KE, et al. Anatomic, radiographic and physiologic comparisons of the internal carotid and maxillary artery in the horse. Vet J 1999; 158: 182–189.
18. Nanda BS. Blood supply to the brain. In: Getty R, ed. Sisson and Grossman's the anatomy of the domestic animals. 5th ed. Philadelphia: WB Saunders, 1975; 578–584.
19. Burger KM, Tuhrim S, Naidich TP. Brainstem vascular stroke anatomy. Neuroimaging Clin N Am 2005; 15: 297–324.