An 8-year-old warmblood mare was referred to the Ghent University Faculty of Veterinary Medicine for evaluation after an acute onset of severe neurologic abnormalities. Two days prior to referral, the mare developed ataxia, circling, and compulsive behavior. The mare was head-pressing and, when turned out in the paddock, would run into the fence. The owner also reported that the mare appeared to have difficulties in passing feces. The following day, the mare was lethargic and had a poor appetite.
At the time of initial examination at the referral hospital, the mare appeared to have normal mentation, although it appeared irritated and excited when approached and examined. Results of a physical examination and abdominal ultrasonography were unremarkable. Rectal temperature was 38.3°C (100.9°F), and rectal palpation did not reveal any abnormalities other than dry feces in the distal portion of the colon. Results of a CBC and serum biochemical profile, including serum ammonia concentration, were within reference limits.
Neurologic examination revealed a diminished menace response in the right eye; however, no visual deficits or other cranial nerve deficits were detected. Mild ataxia was evident at a walk, with normal resistance to sideways traction of the tail. Anal reflex and tail tone were normal. To rule out a spinal cause of the ataxia, transcranial magnetic stimulation of the motor cortex of the brain was performed with a magnetic stimulator.a Magnetic motor-evoked potentials were normal, indicating normal spinal cord conduction. Neurologic abnormalities reported by the owner were suggestive of a supratentorial lesion, and CT of the head was recommended. Treatment with prednisolone (1 mg/kg [0.45 mg/lb], PO, q 24 h) was initiated.
Computed tomography was performed with a third-generation scanner.b The horse was premedicated with romifidine (0.08 mg/kg [0.036 mg/lb], IV), and anesthesia was induced with midazolam (0.6 mg/kg [0.27 mg/lb], IV) and ketamine (10 mg/kg [4.54 mg/lb], IV) and maintained with isoflurane.
Transverse, 5-mm-thick, contiguous images of the brain were obtained with a standard algorithm (120 kVp, 130 mA, and 3-second scanning time), and examination of CT images revealed slightly hyperdense masses involving both lateral ventricles. Examination of additional images obtained after IV administration of iopromidec (800 mL of a solution containing 370 mg of iodine/mL) revealed heterogeneous contrast enhancement of the intraventricular masses (Figure 1). The masses had a bilobate appearance and filled the rostral two thirds of the lateral ventricles. Maximum width, length, and height were 5.7, 6, and 4.8 cm, respectively. The most caudal portions of the lateral ventricles were dilated with CSF. Given the location and CT appearance of the masses, the most likely diagnosis was considered to be cholesterinic granuloma, although choroid plexus papilloma1 and papillary ependymoma2 could not be ruled out.
Transverse, contrast-enhanced CT image of the brain of a horse with cholesterinic granulomas. The image was obtained at the level of the hypophysis; the left side of the brain is on the right side of the image. Notice the large, contrast-enhanced masses in both lateral ventricles (black arrow). Several hypodense areas representing necrosis are visible (white arrow).
Citation: Journal of the American Veterinary Medical Association 233, 6; 10.2460/javma.233.6.950
Transverse, contrast-enhanced CT image of the brain of a horse with cholesterinic granulomas. The image was obtained at the level of the hypophysis; the left side of the brain is on the right side of the image. Notice the large, contrast-enhanced masses in both lateral ventricles (black arrow). Several hypodense areas representing necrosis are visible (white arrow).
Citation: Journal of the American Veterinary Medical Association 233, 6; 10.2460/javma.233.6.950
Transverse, contrast-enhanced CT image of the brain of a horse with cholesterinic granulomas. The image was obtained at the level of the hypophysis; the left side of the brain is on the right side of the image. Notice the large, contrast-enhanced masses in both lateral ventricles (black arrow). Several hypodense areas representing necrosis are visible (white arrow).
Citation: Journal of the American Veterinary Medical Association 233, 6; 10.2460/javma.233.6.950
The horse recovered from anesthesia without complications. Over the next few days, the neurologic abnormalities resolved, and no additional signs developed. However, biopsy of the masses was recommended to further refine the diagnosis and to provide more precise information regarding prognosis and treatment.
Two days prior to biopsy of the masses, the dosage of prednisolone was increased (2 mg/kg [0.9 mg/lb], PO, q 24 h) in an attempt to decrease the risk that cerebral edema or inflammation would develop following the procedure. Treatment with ceftiofur (1 mg/kg, IM, q 12 h) was begun 1 day before the biopsy procedure.
Buprenorphine (0.006 mg/kg [0.003 mg/lb], IV) was administered immediately prior to anesthetic induction. The horse was premedicated with detomidine hydrochloride (10 μg/kg, IV), and anesthesia was induced with thiopental (2 g, IV) and guaifenesin (100 mg/kg [45.5 mg/lb], IV) and maintained with isoflurane. To avoid an increase in intracranial pressure, hypocapnia was created by means of controlled hyperventilation.3,4
The biopsy procedure required that the horse be positioned in lateral recumbency. The horse was positioned in left lateral recumbency, as it was thought that this provided the most direct access to the greatest amount of abnormal tissue. Computed tomography of the regions of interest was performed following IV administration of iopromide, and images were examined to determine the most appropriate transverse plane for the biopsy procedure on the basis of accessibility of the lesion and the shortest path between the skull and the lesion that provided the lowest risk for vital structures. The site for insertion of the needle was chosen in this transverse plane, and the distance between the skull and the lesion was measured. The CT table was then moved out of the gantry to allow for preparation of the operative field.
Hair was clipped from an area in the parietal and temporal regions bordered by the right supraorbital process, the zygomatic arch, the nuchal crest, and the midline, and the skin was surgically prepared. A skin incision was made at the chosen entry point 3 cm to the right of the crista sagittalis externa at the level of the highest point of the zygomatic arch. The interscutularis and temporalis muscles were elevated and retracted laterally, and a 1.5-cm-diameter burr hole was created in the underlying parietal bone with a trephine. An arthroscopic sleeved (2.4 mm diameter) with stylet was inserted through the burr hole into the brain parenchyma to the depth of the previous measurement. The CT table was then returned to the gantry, and a laser light in the gantry was used to ensure that it was positioned as before. Computed tomographic images were obtained to verify the position of the arthroscopic sleeve at the proximal margin of the lesion (Figure 2). Once positioning of the arthroscopic sleeve was considered to be adequate, the stylet was removed, and a brain biopsy needlee was advanced through the arthroscopic sleeve to the depth of the lesion. A difference in consistency between normal brain tissue and tumor tissue was perceptible while inserting the needle. A CT image was obtained to verify correct placement of the biopsy needle, and brain biopsy was then performed. First, an aspiration biopsy specimen was obtained by applying suction to the needle with a 20-mL syringe, and direct smears were prepared. Subsequently, 2 tissue specimens were obtained. Specimens were immediately fixed in neutral-buffered 10% formalin and submitted for histologic evaluation.
Transverse, contrast-enhanced CT image of the brain of the horse in Figure 1 obtained during biopsy of the intracranial masses. An arthroscopic sleeve (right arrow) has been advanced in the transverse plane through a trephine hole in the skull to the proximal margin of the mass (left arrow).
Citation: Journal of the American Veterinary Medical Association 233, 6; 10.2460/javma.233.6.950
Transverse, contrast-enhanced CT image of the brain of the horse in Figure 1 obtained during biopsy of the intracranial masses. An arthroscopic sleeve (right arrow) has been advanced in the transverse plane through a trephine hole in the skull to the proximal margin of the mass (left arrow).
Citation: Journal of the American Veterinary Medical Association 233, 6; 10.2460/javma.233.6.950
Transverse, contrast-enhanced CT image of the brain of the horse in Figure 1 obtained during biopsy of the intracranial masses. An arthroscopic sleeve (right arrow) has been advanced in the transverse plane through a trephine hole in the skull to the proximal margin of the mass (left arrow).
Citation: Journal of the American Veterinary Medical Association 233, 6; 10.2460/javma.233.6.950
The horse recovered from anesthesia without complications, and no neurologic deficits were apparent following the biopsy procedure. A 20% mannitol solution (1 g/kg, IV, q 8 h) was administered for 1 day, flunixin meglumine (1.1 mg/kg [0.5 mg/lb], IV, q 24 h) was administered for 2 days, and ceftiofur (1 mg/kg, IV, q 12 h) was administered for 7 days after the procedure. Prednisolone was administered at a dosage of 2 mg/kg, PO, every 24 hours for the first day after the procedure. The dosage was then decreased to 1 mg/kg, PO, every 24 hours for 1 day and then to 0.5 mg/kg (0.23 mg/lb), PO, every 24 hours for 3 days.
For the owner's convenience, the horse was hospitalized for 21 days after the biopsy procedure. Mentation and cranial nerve function remained normal throughout the period of hospitalization.
Histologic examination of the brain biopsy specimens revealed multiple cholesterol clefts surrounded by giant cells and hemosiderin-laden macrophages (Figure 3). Some macrophages contained iron pigment, providing evidence of bleeding. The diagnosis of a cholesterinic granuloma was confirmed. Cytologic examination of smears of the brain aspirates revealed only aggregates of neutrophils and a few macrophages, necrotic debris, and RBCs.
Photomicrograph of a portion of the intracranial mass in the horse in Figure 1. Cholesterol clefts (arrows) are seen embedded in a paucicellular stroma with few inflammatory cells. H&E stain; bar = 40 μm.
Citation: Journal of the American Veterinary Medical Association 233, 6; 10.2460/javma.233.6.950
Photomicrograph of a portion of the intracranial mass in the horse in Figure 1. Cholesterol clefts (arrows) are seen embedded in a paucicellular stroma with few inflammatory cells. H&E stain; bar = 40 μm.
Citation: Journal of the American Veterinary Medical Association 233, 6; 10.2460/javma.233.6.950
Photomicrograph of a portion of the intracranial mass in the horse in Figure 1. Cholesterol clefts (arrows) are seen embedded in a paucicellular stroma with few inflammatory cells. H&E stain; bar = 40 μm.
Citation: Journal of the American Veterinary Medical Association 233, 6; 10.2460/javma.233.6.950
Discussion
The choroid plexus is a highly vascularized tissue in the ventricular system of the brain that produces CSF. Cholesterinic granulomas are thought to develop as a result of chronic or intermittent congestion and edema with secondary hemorrhage.5,6 Cholesterol and other breakdown products of extravasated RBCs are deposited in the tissues and elicit granulomatous inflammation.5,6 They appear more frequently in the plexus of the fourth ventricle, where many are clinically silent, but can cause signs of a cerebral disorder when located in the lateral ventricles in horses.5,6 Signs most commonly seen in horses with cholesterinic granulomas are lethargy,5–8 reluctance to move,5,8 compulsive behavior,5–7 circling,5–7,9 ataxia,6 blindness,5 and seizures.5,6,9 Less commonly, vague signs of colic may be seen.6,7 Often, signs are episodic, with affected horses appearing normal between episodes.5–9 Neurologic signs in horses with large cholesterinic granulomas are thought to be a result of direct compression of neuronal tissue or obstruction to CSF drainage resulting in secondary hydrocephalus.5,6 The intermittency of clinical signs is thought to be a result of variable CSF pressure, with high CSF pressure alternating with drainage of CSF.7 Intermittent exacerbations of clinical signs might also be attributable to periodic release of irritating substances, such as cholesterol, which causes aseptic, chemical meningitis.7 In affected horses, therefore, treatment with anti-inflammatory agents could decrease edema and stabilize inflammation, temporarily alleviating neurologic signs.
Previously reported cases5–9 of cholesterinic granulomas involved horses with neurologic abnormalities attributable to large, primarily bilateral granulomas located in the lateral ventricles. Whereas these horses had recurrent episodes of progressive neurologic signs or only minimal signs of improvement leading eventually to euthanasia, the horse described in the present report had only a single episode of neurologic abnormalities. Long-term (ie, 5 years) follow-up did not reveal any recurrence of neurologic deficits, and the horse returned to its previous training level and competed successfully. Cholesterinic granulomas are often identified as an incidental postmortem finding, having caused no abnormalities during life,5–8 and the uncommon clinical course in the horse described in the present report could suggest that the cholesterinic granulomas were an incidental finding and the clinical signs were actually a result of some concurrent cerebral disease.
Possible causes of cerebral dysfunction in adult horses include head trauma, hemorrhage, abscess, neoplasia, metabolic disease (eg, hepatic encephalopathy and hyperammonemia), encephalomyelitis (ie, viral, bacterial, protozoal, fungal, or parasitic infection), and CNS toxicosis (eg, lead and fumonisin B1). Head trauma with hemorrhage and abscess formation could be ruled out on the basis of CT findings in the horse described in the present report, and the history and disease course were atypical for lead poisoning or leukoencephalomalacia. Protozoa commonly associated with protozoal encephalomyelitis were not endemic in the region of origin of the horse and neither were eastern and western equine encephalitis viruses, Borna virus, and West Nile virus. Equine herpesvirus is ubiquitous in most European countries, and herpesvirus infection could not be completely ruled out, but equine herpesvirus infection typically results in lesions in the spinal cord and cauda equina and generally causes a different clinical course. Hepatic dysfunction can produce clinical signs similar to those seen in this horse, but was ruled out on the basis of results of serum biochemical analyses. The possibility of transient hyperammonemia in the absence of liver disease was considered, but was less likely in this horse because serum ammonia concentration was within reference limits and no signs of gastrointestinal tract dysfunction were detected.10
Cholesterinic granulomas with comparable dimensions have been found to cause pronounced neurologic deficits in other horses, and we believe that masses seen in the horse described in the present report were the cause of the transient cerebral dysfunction. Presumably, neurologic signs were a result of partial, temporary obstruction of the interventricular foramina with high CSF pressure and dilatation of the ventricular system, with signs remitting once CSF pressure returned to normal. However, the exact explanation of the singular appearance of neurologic signs in this horse remains speculative, and the natural course of this condition in horses is not really known. To our knowledge, this is the first description of a horse with large, bilateral cholesterinic granulomas that was capable of resuming its previous training level. Considering the disease course in other reported cases, one may presume that neurologic abnormalities could indeed recur in this horse, and the owners were warned at the time of discharge about the risks intrinsic to managing and riding such horses.
The introduction of CT and magnetic resonance imaging in equine medicine has proven to be an indispensable asset in identifying and localizing intracranial lesions, including cholesterinic granulomas.6 However, both CT and magnetic resonance imaging have limited specificity, and only histologic examination of a tissue sample can provide a definitive diagnosis. Surgical and chemotherapeutic treatments of intracranial masses in horses have not been described. Therefore, unlike the situation in human and small animal medicine, in which brain biopsy is performed to obtain a definitive diagnosis because treatment and prognosis is dependent on the diagnosis,11 few indications exist for performing brain biopsy in horses. Although CT findings for the horse described in the present report were highly suggestive of cholesterinic granuloma, biopsy was recommended to exclude less common tumors (eg, choroid plexus papilloma1 and papillary ependymoma2) that have been identified in horses. Image-guided brain biopsy has become the standard method for establishing a diagnosis in humans and dogs with intracranial lesions, and several reports3,12–20 have described the safety and accuracy of CT-guided brain biopsy performed freehand or with the aid of a stereotactic device.
Major complications that can occur during brain biopsy include hemorrhage and cerebral edema leading to intracranial hypertension and, potentially, an exacerbation of clinical signs, seizures, coma, or death.3,13,16,21 Minor complications that have been described include wound infection, headache, and nausea. In the horse described in the present report, no clinically detectable adverse effects associated with the biopsy procedure were encountered. This lack of adverse effects was believed to be attributable to certain characteristics of the tumor itself and of the trajectory chosen for the biopsy needle in conjunction with medical treatment to prevent cerebral edema. In addition, in 1 study,13 life-threatening complications were common in dogs that had severe neurologic abnormalities prior to brain biopsy, whereas the horse described in the present report did not have any severe neurologic abnormalities at the time of brain biopsy.
Fatal complications of brain biopsy are primarily related to the location and vascularity of the tumor.17,22 Although cholesterinic granulomas are deep-seated tumors, we were able to choose, on the basis of contrastenhanced CT images, a biopsy trajectory that avoided vital brain areas. The biopsy trajectory crossed the parietal lobe of the telencephalon and, more specifically, the cerebral cortex, the limbic lobe of the limbic system, and the corpus callosum, and these areas are less developed in horses than in people. Cholesterinic granulomas are well-circumscribed lesions, and these tend to yield more diagnostic samples than do heterogeneous,13,17 poorly circumscribed,13 or nonneoplastic lesions.13
Cytologic examination of aspirate smears obtained from the horse in the present report did not provide any useful information. Reasons for this poor result could be the small sample size and the firm structure of cholesterinic granulomas, which can make aspiration of tumor tissue more difficult. For this reason, tissue samples should be obtained in addition to aspirates. Collection of at least 2 tissue samples has been recommended.13,16
To date, treatment of cholesterinic granulomas in horses has been nonspecific and aimed at decreasing CNS inflammation, cerebral edema, and intracranial pressure. Although there have been several reports4,23,24 of successful aspiration and surgical debulking of intracranial abscesses in horses, to our knowledge, surgical removal of an intracranial tumor in a horse has not been reported. Initially, we had hoped to be able to surgically remove the tumor from this horse, depending on the outcome of brain biopsy. Ultimately, we elected to not attempt surgical removal because the horse's neurologic abnormalities had resolved and surgical removal could have caused serious, potentially fatal complications given the location and size of the masses.
On the basis of our experience with this case, we believe that CT-guided brain biopsy of deep-seated lesions is feasible in horses and that this technique can provide histologically representative specimens, allowing for in vivo diagnosis of intracranial lesions. In addition, this technique may have the potential to assist with surgical removal of intracranial lesions or guided placement of medications.
ABBREVIATION
| CT | Computed tomography |
Magstim 200, Novametrix, Whitland, Wales.
Pace Plus, GE Medical Systems, Milwaukee, Wis.
Ultravist, Schering, Berlin, Germany.
Richard Wolf GmbH, Knittlingen, Germany.
Nashold biopsy needle, 2-mm diameter, Integra Radionics, Burlington, Mass.
References
- 1.↑
Pirie RS, Mayhew IG & Clarke CJ, et al. Ultrasonographic confirmation of a space-occupying lesion in the brain of a horse: choroid plexus papilloma. Equine Vet J 1998;30:445–448.
- 2.↑
Huxtable CR, de Lahunta A & Summers BA, et al. Marginal siderosis and degenerative myelopathy: a manifestation of chronic subarachnoid hemorrhage in a horse with a myxopapillary ependymoma. Vet Pathol 2000;37:483–485.
- 3.
Moissonnier P, Blot S & Devauchelle P, et al. Stereotactic CT-guided brain biopsy in the dog. J Small Anim Pract 2002;43:115–123.
- 4.
Cornelisse CJ, Schott HC & Lowrie CT, et al. Successful treatment of intracranial abscesses in 2 horses. J Vet Intern Med 2001;15:494–500.
- 5.↑
Jackson CA, Delahunta A & Dykes NL, et al. Neurological manifestation of cholesterinic granulomas in three horses. Vet Rec 1994;135:228–230.
- 6.↑
Vink-Nooteboom M, Junker K & Van Den Ingh T, et al. Computed tomography of cholesterinic granulomas in the choroid plexus of horses. Vet Radiol Ultrasound 1998;39:512–516.
- 7.↑
Johnson PJ, Lin TL, Jennings DP. Diffuse cerebral encephalopathy associated with hydrocephalus and cholesterinic granulomas in a horse. J Am Vet Med Assoc 1993;203:694–697.
- 8.
Divers TJ. Structural and functional disease of the equine brain. Clin Tech Equine Pract 2006;5:67–73.
- 9.
Cobb S. Studies in mammalia neuropathology: 1. Cerebral cholesteatoma in the horse. J Mammal 1923;4:221–223.
- 10.↑
Peek SF, Divers TJ, Jackson CJ. Hyperammonaemia associated with encephalopathy and abdominal pain without evidence of liver disease in four mature horses. Equine Vet J 1997;29:70–74.
- 11.↑
Arbit E, Galicih JH. Importance of image-guided stereotactic biopsy to confirm diagnosis in an oncological setting. Ann Surg Oncol 1994;1:368–372.
- 12.
Harari J, Moore MM & Leathers CW, et al. Computed tomographic-guided, freehand needle biopsy of brain tumours in dogs. Prog Vet Neurol 1993;4:41–44.
- 13.↑
Koblik PD, Lecouteur RA & Higgins RJ, et al. CT-guided brain biopsy using a modified Pelorus Mark III stereotactic system: experience with 50 dogs. Vet Radiol Ultrasound 1999;40:434–440.
- 14.
Moissonnier P, Bordeau W & Delisle F, et al. Accuracy testing of a new stereotactic CT-guided brain biopsy device in the dog. Res Vet Sci 2000;68:243–247.
- 15.
Flegel T, Podell M & March PA, et al. Use of a disposable real-time CT ctereotactic navigator device for minimally invasive dog brain biopsy through a mini-burr hole. Am J Neuroradiol 2002;23:1160–1163.
- 16.
Savitz MH. CT-guided needle procedures for brain lesions: 20 years' experience. Mount Sinai J Med 2000;67:318–321.
- 17.
Hall WA. The safety and efficacy of stereotactic biopsy for intracranial lesions. Cancer 1998;82:1749–1755.
- 18.
Seliem RM, Assaad MW & Gorombey SJ, et al. Fine-needle aspiration biopsy of the central nervous system performed freehand under computed tomography guidance without stereotactic instrumentation. Cancer 2003;99:277–284.
- 19.
Goldstein S, Gumerlock MK, Neuwelt EA. Comparison of CT-guided and stereotaxic cranial diagnostic needle biopsies. J Neurosurg 1987;67:341–348.
- 20.
Wen DY, Hall WA & Miller DA, et al. Targeted brain biopsy: a comparison of freehand computed tomography-guided and stereotactic techniques. Neurosurgery 1993;32:407–412.
- 21.
Bernstein M, Parrent AG. Complications of CT-guided stereotactic biopsy of intra-axial brain lesions. J Neurosurg 1994;81:165–168.
- 22.
McGirt MJ, Woodworth GF & Coon AL, et al. Independent predictors of morbidity after image-guided stereotactic brain biopsy: a risk assessment of 270 cases. J Neurosurg 2005;102:897–901.
- 23.
Janicek JC, Kramer J & Coates JR, et al. Intracranial abscess caused by Rhodococcus equi infection in a foal. J Am Vet Med Assoc 2006;228:251–253.
- 24.
Allen JR, Barbee DD & Boulton CR, et al. Brain abscess in a horse: diagnosis by computed tomography and successful treatment. Equine Vet J 1987;19:552–555.
