*Corresponding author: Daisuke Hasegawa, DVM, PhD; Department of Veterinary Radiology, Nippon Veterinary and Animal Science University, Musashinoshi, Tokyo, Japan.

Abstract: Excessive recurrent seizures, such as status epilepticus (SE) and/or cluster seizures, have been shown to induce epileptic (secondary) brain damage (EBD) in experimental studies and human epileptic patients. EBD results in various behavioral (psychological) and/or neurological dysfunctions. In this report, an epileptic dog that had been diagnosed and treated with antiepileptic drugs showed behavioral changes similar to Klüver-Bucy syndrome, which results from bilateral limbic lesion in primates after severe SE. Previous cases of EBD in small animals have shown rage reaction/aggression or were euthanized before recovering from SE. We report here a dog that showed distinct clinical and pathological similarities to Klüver-Bucy syndrome of primates, including human patients.

An 8-year and 9 months old, ovariohysterectomized female Welsh corgi weighing 10.6 kg was referred to the Animal Medical Center of Nippon Veterinary and Animal Science University (AMC-NVAU) for behavioral problems, which began after status epilepticus (SE). The dog had an initial seizure at 2 years of age, and a subsequent seizure frequency of 1-3 times per month. At 1 month after the initial seizure, the dog was subjected to physical, hematological (CBC and serum chemistry), and neurological examinations, as well as computed tomography and serum canine distemper virus (CDV) antibody titer another veterinary hospital. These examinations showed normal findings, and a tentative diagnosis of idiopathic epilepsy was made. Antiepileptic drug (AED) therapy with phenobarbital (PB: 3-8 mg/kg, PO, q12h) and potassium bromide (KBr: 30 mg/kg, PO, q24h) was initiated. However, the seizure frequency did not improve significantly (1-3 times per month). The seizures of this dog usually began as partial, consisting of salivation, gazing, immobilization and head tremor, followed by a secondary generalized tonic-clonic seizure. During the interictal period, the dog was obedient to the owner's family, but had a cowardly character and usually barked at or threatened strangers.

Two months before being referred to AMC-NVAU, the owner found the dog in SE; SE lasted at least 3 hours until she was taken to another veterinary hospital. The SE continued during hospitalization for about 1 hour after intravenous injection of diazepam, followed by pentobarbital.

After recovering from SE, the dog did not recognize her owners, did not vocalize, and changed her personality to being obedient and friendly to everyone. The dog did not react to calling or any noise, and was not startled by any loud sounds. In addition, the dog ate anything around her such as towels, shoes, stool, etc. (pica).

The behavioral signs persisted until referral to AMC-NVAU. On admission there, the history revealed that routine vaccinations (including CDV and rabies) and filarial prophylaxes had been performed every year. AED therapy with PB (6 mg/kg, PO, q12h) and KBr (30 mg/kg, PO, q24h) had been continued until admission, and no seizures had been observed after the episode of SE. The dog's general appearance was normal and no abnormalities were found on physical examination. Neurological examinations revealed the behavioral signs described above plus mild bilateral proprioceptive deficits in the pelvic limbs and failure to react to any sounds. CBC, serum chemistry, blood gases, electrolytes, thoracic radiographs, and electrocardiogram were normal; notably, there was no increase in ALP as a result of PB therapy.

Brainstem auditory evoked potentials (BAEP), recorded using an evoked potential examination system^a indicated the brainstem auditory pathways were normal bilaterally.

Electroencephalography (EEG) was performed with both referential and bipolar derivations.^b During recording the dog was very quiet and drowsy without sedation. Sporadic and/or synchronous paroxysmal discharges such as spikes, polyspikes, and sharp waves were numerous bilaterally in the temporal and parietal regions dorsally (Figure 1).

Click on the image to see a larger view Figure 1. The electroencephalogram (EEG) was recorded 2 months after the episode of status epilepticus (just before the MRI shown in Figure 2). Sporadic (arrow heads) and/or synchronized (arrows) paroxysmal discharges (spikes and sharp waves) were recorded frequently. The EEG epoch shown here was recorded using a referential derivation (reference electrode on the nasion). Time constant; 0.1 ms, sampling frequency; 250 Hz, high-cut filter; 60 Hz. L/RF; left/right frontal, L/RP; left/right parietal, L/RT; left/right temporal, L/RO; left/right occipital.

Magnetic resonance imaging (MRI) using a 1.5 Tesla MRI system^c was performed under general anesthesia. Symmetric and cavitating necrotic lesions extending from the amygdalae to the lateral temporal lobes were detected as hyperintensities on T2-weighted images (T2WI) and hypointensities on fluid-attenuated inversion recovery (FLAIR) and T1-weighted images (T1WI). Around these lesions, i.e. bilaterally in the pyriform, temporal and parietal cortices and bilaterally in the hippocampi, hyperintensities were detected on T2WI and FLAIR, and hypointensities were detected on T1WI (Figure 2). These lesions were not enhanced by gadolinium.

Click on the image to see a larger view Figure 2. Transverse planes of fluid-attenuated inversion recovery (FLAIR: TR/TE/TI = 8000/120/2000 ms) imaging at the level of the amygdala (A) and hippocampus (B). Symmetric cavitary and laminar hypointense lesions on FLAIR extend from the amygdalae to the lateral temporal lobe cortices (arrows). Hyperintensity is detected around these lesions and the hippocampi (arrowheads) on FLAIR.

At the time of the MRI, cerebrospinal fluid (CSF) was collected by cisternal puncture. No specific abnormalities were found, except a mild increase of neuron-specific enolase (NSE: 25 ng/mL, reference range < 15 ng/mL). CDV-antibody and antigen were not detected in the CSF.

After the MRI examination, euthanasia was carried out as requested by the owner. Necropsy and histopathological examinations were performed. On gross examination of the brain, the surfaces of the pyriform lobes and the ventral parts of the temporal lobes were found to be depressed bilaterally. Histopathologically, the amygdalae, parahippocampal gyri and extratemporal cortices showed extensive, bilaterally symmetrical necrosis and formed the cavitary lesions visible on MRI. (Figure 3). Neuronal loss, pyknosis, and neuropile rarefaction with fat-laden macrophages and astrocytic proliferation were observed in the remaining cortex of the pyriform, parahippocampal and temporal lobes. In the hippocampi, there was severe neuronal loss and pyknosis, with astrocytosis in CA3, 4, and 1 of the pyramidal cell layer (Figure 4). CA 2 and the dentate gyrus showed slight damage. There were no abnormalities in other brain regions, and inclusion bodies and/or perivascular cuffing suggesting virus and/or non-suppurative encephalitis were not detected. Furthermore, immunohistochemistry using CDV monoclonal antibody ^d was performed and showed a negative result.

Click on the image to see a larger view Figure 3. Macroscopic hematoxylin-eosin section at the same level as in Figure 1A. Symmetric necrotic, cavitary lesions are visible in the amygdalae (arrows).

	Click on an image to see a larger view Figure 4. Histopathological sections of the hippocampi from the patient and from a normal dog. A: Patient. Magnification: x 40, H E stain. Severe neuronal loss and gliosis are seen in CA3, 4 and 1 of the pyramidal cell layer. CA2 and the dentate gyrus (DG) also show slight damage. Magnification: x 40, H E stain. B: Normal dog.

Excessive recurrent seizures have been shown to produce epileptic brain damage (EBD), which has been demonstrated by many cases of intractable human epilepsy and the studies of various experimental animal models. Hippocampal sclerosis (HS) or mesial temporal sclerosis (MTS) are typical EBDs and are seen in kainic acid (KA) induced models^1-5 and in intractable human temporal lobe epilepsy.^6,7 On the other hand, the occurrence of EBD in the dog has been controversial because different reports suggest either no specific pathological findings in several cases of idiopathic epilepsy^8-11 or similar findings as HS/MST^12-15.

Previously, we studied the canine model of KA-induced complex partial status epilepticus^4,5. In those studies, although KA was injected into the unilateral amygdala and the partial seizures of early stage started at the injected amygdala, secondary generalized seizures and contralateral partial seizures with secondary generalized seizures were observed frequently during SE. Bilateral extended necrosis in the amygdala forming a cavitary lesion and typical HS were observed. Those results suggested that canine partial seizures, especially limbic seizures, are easy to generalize and that the canine limbic system has a higher sensitivity to seizure activity. MRI and pathological findings in the present case are exactly similar to those observed in the canine KA model. At present, the excitotoxicity theory is believed to be a pathophysiological mechanism of EBD^1-3. Recently, Mellena et al. reported 4 similar cases of MRI abnormalities following recurrent seizures¹⁹, and Mariani et al. reported a similar case of polioencephalomalacia using MRI and pathologic findings²⁰. They also found symmetric and/or asymmetric limbic lesions and suggested that excitotoxicity from severe seizures might be the cause of those lesions.

Along with the excitotoxicity theory, the pathophysiological mechanism of HS has been investigated in KA-models using rodents and cats^2,3. Recently, Buckmaster et al. investigated HS, including mossy fiber sprouting in intractable epilepsy in dogs using Timm stain and, based on their findings, suggested that HS and temporal lobe epilepsy are not common causes of canine epilepsy¹¹. Although we did not investigate the mossy fibers, the histological findings in the hippocampi of the present case are consistent with those of HS in experimental models and human epilepsy. Therefore we suggest that HS is induced in dogs depending on the seizure type and/or SE.

Other studies suggested polioencephalomalacia in the limbic system, however, many cases were associated with CDV encephalitis^16-19. The present case was distinguished from CDV clinically, pathologically and immunohistochemically. Although CDV encephalitis is the most common cause of canine cortical necrosis, other causes exist. Cerebral infarction, hypoxia, chemical toxicity and other types of encephalitis have been suggested as possible causes^{17, 20,21}. Cerebral infarction usually occurs unilaterally, and evidence of infarction was not observed in the present case. In addition, hypoxia and chemical toxicities are not obvious from her history and pathology. Other types of encephalitis include necrotizing meningoencephalitis and unknown slowly-progressive viral or immunogenic encephalitis. Necrotizing meningoencephalitis (NME), also known as Pug dog encephalitis, is characterized by acute to chronic progressive cerebral necrosis, which may be caused by autoimmune mechanism²², and occurs in some breeds other than the Pug dog^23,24. Chronic necrotic lesions of NME are similar to the observations in this case. However, NME shows progressive neurological dysfunctions, and granulomatous lesions including perivascular cuffs are typically observed. Therefore, the possibility that our case had NME is extremely low. Montgomery and Lee¹⁰ and Gredal²⁵ reported intraneuronal inclusions similar to Lafora's bodies in epileptic beagle dogs. Lafora's bodies are observed in human Lafora disease showing myoclonic epilepsy. A reported beagle dog also showed progressive myoclonus and did not show cortical necrosis²⁵. The seizures of the present dog were not myoclonic, and inclusion bodies were not found. Recently, in humans, non-herpetic and non-paraneoplastic limbic encephalitis have attracted a great deal of attention^26-28. This limbic encephalitis is also thought to be an autoimmune disease and shows good response to steroid therapy. However, the cause of this disease has not been established and there is no report of this disease in veterinary medicine.

Interestingly, in the present case, the clinical signs of brain dysfunction were similar to "Klüver-Bucy syndrome" (KBS), and were consistent with bilateral injury to the amygdalae and temporal lobes. KBS was reported to result from bilateral temporal lobectomy in monkeys by Klüver and Bucy in 1930s and is a clinical sign of dysfunction in the amygdalae and temporal lobes in animals and humans.^29,30 KBS is characterized by the following behavioral changes: 1) 'psychic blindness', which is the inability to recognize and to evaluate objects; 2) 'hyperorality', which is the ingestion and eating of inappropriate objects (pica); 3) 'hypersexuality' toward both genders and/or different species; 4) 'altered emotional behavior', which is characterized by placidity and loss of fear of an enemy and/or surroundings. Few reports have documented behavioral changes after damage to the amygdalae in dogs and cats. Tanaka et al. reported a rage reaction when the amygdala was excited in a feline hippocampal KA model³¹. Caldwell and Little³² and Mariani et al.²⁰ also reported aggression in dogs with amygdaloid and hippocampal lesions. These aggressive behaviors are thought to be paradoxical symptoms caused by moderate damage or stimulation of the amygdala and other limbic system structures. Although we did not evaluate cases in detail, our previous KA-treated dogs also changed their behavior to be obedient after CPSE⁹. The clinical signs in the present case, which included loss of fear reaction to loud sounds in spite of normal auditory function, pica and non-recognition of the owner, are similar with KBS reported in monkeys and human. However, hypersexuality was not observed. Behavioral changes and polyphagia are well known side effect of AEDs³³. We cannot completely exclude pharmacologically induced KBS-like behavioral change in the present case because the serum levels of AEDs were not measured before and after SE. However, AED therapy after SE was not altered from previous doses, and the dog never showed altered behaviors prior to SE. A latent possibility exists that the dog's sensitivity to AEDs was enhanced by the pathological changes.

To the authors' knowledge, this is the first reported case of KBS in the dog. Depending on more detailed observations of postictal or interictal behavior, KBS in dogs may be recognized in chronic encephalitis, such as NME and CDV, in addition to EBD. Further study of emotion and behavior in dogs and cats, clinically, neuroanatomically and neuropathologically, is necessary.

Footnotes

a Neuropack, Nihon Koden, Tokyo, Japan
b Neurofax, Nihon Koden, Tokyo, Japan
c Visart, Toshiba medical system, Tokyo, Japan
d Canine distemper virus monoclonal antibody, Veterinary Medical Research & Development, Pullman, U.S.A.

References

1.  Wasterlain CG, Fujikawa DG, Penix L, et al. Pathophysiological mechanisms of brain damage from status epilepticus. Epilepsia 1993;34:S37-S53.

2.  Ben-Ari Y. Limbic seizure and brain damage produced by kainic acid: mechanisms and relevance to human temporal lobe epilepsy. Neuroscience 1985;14: 375-403.

3.  Tanaka T, Tanaka S, Fujita T, et al. Experimental complex partial seizures induced by a microinjection of kainic acid into limbic structures. Prog Neurobiol 1992;38:317-334.

4.  Hasegawa D, Orima H, Fujita M, et al. Complex partial status epilepticus induced by a microinjection of kainic acid into unilateral amygdala in dogs and its brain damage. Brain Res 2002;955:174-182.

5.  Hasegawa D, Orima H, Fujita M, et al. Diffusion-weighted imaging in kainic acid-induced complex partial status epilepticus in dogs. Brain Res 2003: 983: 115-127.

6.  Mathern GW, Babb TL, Armstrong DL. Hippocampal sclerosis. In: Engel J Jr, Pedley TA, eds. Epilepsy: A comprehensive textbook. Philadelphia. PA: Lippincott-Raven, 1997;133-155.

7.  Salmenpera T, Kalviainen R, Partanen K, et al. MRI volumetry of the hippocampus, amygdala, entorhinal cortex, and perirhinal cortex after status epilepticus. Epilepsy Res 2000;40:155-170.

8.  Palmer AC. Pathologic changes in the brain associated with fits in dogs. Vet Rec 1972;90:167-173.

9.  Koestner A. Neuropathology of canine epilepsy. Prob Vet Med 1989;1:516-534.

10. Montgomery DL, Lee AC. Brain damage in the epileptic beagle dog. Vet Pathol 1983;20:160-169.

11. Buckmaster PS, Smith MO, Buckmaster CL, et al. Absence of temporal lobe epilepsy pathology in dogs with medically intractable epilepsy. J Vet Intern Med 2002;16:95-99.

12. Andersson B, Olsson SE. Epilepsy in a dog with extensive bilateral damage to the hippocampus. Acta Vet Scand 1959;1:98-104.

13. Yamasaki H, Furuoka H, Takechi M, et al. Neuronal loss and gliosis in limbic system in an epileptic dog. Vet Pathol 1991;28:540-2.

14. Morita T, Shimada A, Takeuchi T, et al. Cliniconeuropathologic findings of familial frontal lobe epilepsy in Shetland sheepdogs. Can J Vet Res 2002;66:35-41.

15. Hasegawa D, Fujita M, Nakamura S, et al. Electrocorticographic and histological findings in a Shetland sheepdog with intractable epilepsy. J Vet Med Sci 2002;64:277-279.

16. Braund KG, Vandeveld M. Polioencephalomalacia in the dog. Vet Pathol 1979;16:661-672.

17. Hartley WJ. Polioencephalomalacia in dogs. Acta Neuropathol 1963;2:271-281.

18. Fischer K. Herdformige symmetrische Hirngewebsnekrosen bei Hunden mit epileptiformen Krampfen. Pathol Vet 1964;1:133-160.

19. Mellema LM, Koblik PD, Kortz GD, et al. Reversible magnetic resonance imaging abnormalities in dogs following seizures. Vet Rad Ultrasound 1999:40:588-595

20. Mariani CL, Platt SR, Newell SM, et al. Magnetic resonance imaging of cerebral cortical necrosis (polioencephalomalacia) in a dog. Vet Rad Ultrasound 2001;42:524-531.

21. Summer BA, Cummings JF, De Lahunta A. Degenerative disease of the central nervous system. In: Summer BA, Cummings JF, De Lahunta A, eds. Veterinary Pathology. St Louis. Mosby, 1995; 244-246.

22. Uchida K, Hasegawa T, Ikeda M, et al. Detection of autoantibody from Pug dogs with necrotizing encephalitis (Pug dog encephalitis). Vet Pathol 1999;36:301-307.

23. Stalis IH, Chadwick B, Dayrell-Hart B, et al. Necrotizing meningoencephalitis of Maltese dogs. Vet Pathol 1995;32:230-235.

24. Lotti D, Capucchio MT, Gaidolfi E, et al. Necrotizing encephalitis in a Yorkshire terrier: clinical, imaging, and pathologic findings. Vet Rad Ultrasound 1999;40:622-626.

25. Gredal H, Berendt M, Leifsson PS. Progressive myoclonus epilepsy in a beagle. J Small Anim Pract 2003;44:511-514.

26. Shoji H, Asaoka K, Ayabe M, et al. Non-herpetic acute limbic encephalitis. A new subgroup of limbic encephalitis? Intern Med 2004;43:348.

27. Asaoka K, Shoji H, Nishizaka S, et al. Non-herpetic acute limbic encephalitis. MRI findings and CSF cytokines. Intern Med 2004;43:42-48.

28. Bien CG, Schulze-Bonhage A, Deckert M, et al. Limbic encephalitis not associated with neoplasm as a cause of temporal lobe epilepsy. Neurology 2000;55:1823-1828.

29. Klüver H, Bucy P. Preliminary analysis of functions of the temporal lobe in monkeys. Arch Neurol Psychiatry 1939;42:979-1000.

30. Hayman LA, Rexer JL, Pavol MA, et al. Klüver-Bucy syndrome after bilateral selective damage of amygdala and its cortical connections. J Neuropsychiatry Clin Neurosci 1998;10:354-358.

31. Tanaka T, Kaijima M, Daita G, et al. Electroclinical features of kainic acid-induced status epilepticus in freely moving cats. Microinjection into the dorsal hippocampus. Electroencephalogr Clin Neurophysiol 1982;54:288-300.

32. Caldwell DS, Little PB. Aggression in dogs and associated neuropathology. Can Vet J 1980;21:152-154.

33. Vernau KM, LeCouteur RA, Maddison JE. Anticonvulsant drugs. In: Maddison M, Page S, Church D, eds. Small Animal Clinical Pharmacology. Philadelphia. PA: W.B.Saunders, 2002;327-341.