Richard A. Lecouteur, BVSc, PhD, DACVIM (Neurology)
Although it is possible to localize a problem to the brain, and sometimes to localize a problem within the brain, it must be remembered that clinical signs may be the same regardless of the underlying cause. Clinical signs reflect the location of a problem. Brain tumors, infections, congenital disorders, trauma, vascular disorders, degeneration, immunologic and metabolic disorders, toxicities, and idiopathic disorders may result in similar clinical signs. For this reason it is essential to follow a logical diagnostic plan for a cat or dog with signs of brain dysfunction.
Following a complete history and physical and neurological examinations, a minimum data base should be obtained. This should include a hemogram, serum chemistry panel, and urinalysis. Survey thoracic radiographs and abdominal ultrasound help to rule out problems elsewhere. The major objective in doing these tests is to exclude disease outside the brain as a cause of the signs of cerebral dysfunction.
Plain skull radiographs are useful for detecting problems of the skull or nasal cavity that may have extended to the brain. Occasionally, lysis or hyperostosis of the skull may accompany a primary brain tumor (e.g., meningioma of cats) or there may be mineralization within a neoplasm. Skull radiographs are of little value in detecting dysfunction within the brain.
Analysis of cerebrospinal fluid (CSF) is recommended as an aid in the diagnosis of a brain disorder. Results of CSF analysis may help to identify inflammatory causes of cerebral dysfunction, and in some cases may support diagnosis of a brain tumor. CSF bathes the entire CNS, both internally (the ventricles and central canal) and externally (the subarachnoid space). CSF composition may be affected by many nervous system diseases and the ease with which this fluid may be collected has made it a useful diagnostic tool in the diagnosis of CNS disease. Unfortunately, for cells to be shed into the CSF a disease must involve the ventricular system or the subarachnoid space. Disorders involving deeper brain structures (e.g., neoplasms) may not result in shedding of cells into CSF. Frequently these diseases disrupt the blood-brain barrier allowing protein to leak into the CSF and resulting in an increased protein level. CSF must be evaluated keeping in mind history and clinical signs. Neoplasms and some other non-inflammatory diseases may result in inflammatory changes in CSF composition. CSF composition also may change as a disease becomes more chronic. Also, following various therapies (e.g., glucocorticoids) CSF no longer may accurately reflect an etiology.
Care should be used in the collection of CSF, because frequently an increased intracranial pressure (ICP) may be present in association with a brain tumor, and pressure alterations associated with CSF removal may cause brain herniation. Because CSF pressure measurements are of limited usefulness, it often is desirable to utilize techniques such as hyperventilation to decrease intracranial pressure prior to CSF collection.
CSF may be collected at either the cerebellomedullary cistern or by lumbar puncture. In general collection from the cerebellomedullary cistern is easier to perform, allows collection of a larger volume, and generally results in less blood contamination. All patients undergoing CSF collection should be anesthetized appropriately. If it is suspected that intracranial pressure is elevated the patient should be hyperventilated for several minutes prior to collection as well as during and after collection in order to decrease arterial CO2 and intracranial pressure. Complications of CSF collection include needle injury to the brain stem and herniation of the brain, usually due to high intracranial pressure. Both these complication may be fatal if appropriate steps to reduce intracranial pressure (hyperventilation and mannitol administration) are not instituted immediately.
Always note from where the CSF was collected. CSF from the cisternal space has a lower protein than fluid from the lumbar space. Analysis should include gross visual examination, cytologic analysis, protein measurement and possibly fluid culture. Normal CSF should appear clear and colorless. Turbidity may indicate increased white cell numbers and/or protein elevation. A pink color to the fluid may be due to blood contamination. Persistence of this color change after centrifugation of a small amount of sample may indicate the presence of free hemoglobin, suggesting previous subarachnoid hemorrhage, rather than sample contamination. Xanthochromia develops if more than 48 hours have elapsed following hemorrhage.
Cytological preparation should be completed as soon as possible (within 30 minutes of collection), since some CSF cells undergo degeneration rapidly. WBCs degrade faster than RBCs in the CSF, making interpretation difficult. Refrigeration does help to slow WBC lysis. The cytologic evaluation should begin with a total cell count on the unconcentrated fluid. A slide should then be prepared of concentrated fluid for evaluation of cell morphology and differential cell counts. The two most common slide preparation techniques are cytocentrifugation and sedimentation. The sedimentation technique is easy and reliable and may preserve cell architecture better than centrifugation methods.
Normal CSF should contain less than 5 WBC/ul. Cell numbers may increase most commonly with inflammatory disease but also with tumors, necrosis, trauma, and vascular injury. The type and number of cells may reflect the cause of the inflammation. CSF is most accurate in acute, untreated illnesses. Correcting the number of WBCs in the CSF for blood contamination is possible only with very mild contamination. The WBC numbers in the CSF increase approximately 1 WBC/ul for each 500 to 700 RBCs/ul. A predominance of polymorphonuclear leukocytes in CSF may indicate a suppurative meningitis probably due to bacterial infection or to severe viral encephalitis. The presence of multiple cell types in CSF, including macrophages, lymphocytes, neutrophils and sometimes plasma cells is generally the result of a granulomatous inflammation, such as occurs with fungal, protozoal, and some idiopathic diseases. Mixed cell populations may also be seen with inadequately treated chronic bacterial infections and in response to foreign bodies. A nonsuppurative inflammation is diagnosed if the increased cell numbers are primarily mononuclear cells, especially lymphocytes. It is most characteristic of viral and rickettsial infections, although it can also be seen with neoplasms.
CSF should be submitted for both aerobic and anaerobic bacterial culture and antibiotic sensitivity testing whenever abnormalities consistent with meningitis are found. Negative cultures are common even when bacterial or fungal organisms can be seen. Culturing the sedimented or centrifuged CSF may increase the likelihood of a positive culture. Causative organisms may also be isolated from the blood of animals with systemic infections. It is recommended that a large volume of CSF, preferably 2 or 3 ml, be collected for bacterial and/or fungal culture. Serology also may be useful in diagnosis of CNS fungal infections.
In normal cerebellomedullary CSF, protein is less than 25 mg/dl and less than 35 mg/dl with lumbar puncture. Since storage of CSF does not affect protein it can be sent to an outside laboratory for protein evaluation. CSF protein is elevated with many diseases including encephalitis, meningitis, neoplasms, trauma and infarcts. With noninflammatory disruption of the blood-brain-barrier, (e.g., neoplasm and infarcts), most (75% of the CSF protein will be albumin, whereas with increased intrathecal production (e.g., encephalitis) it is predominantly globulin. Where both albumin and globulin are elevated, an inflammatory process affecting both the meninges and the CNS (e.g., FIP) should be suspected.
Although increased CSF protein content and a normal to increased CSF white blood cell count have been considered "typical" of a brain neoplasm, in one study, only 39.6% of dogs with a primary brain tumor exhibited "typical" CSF alterations. CSF analysis was normal in 10% of the dogs in this study, while the remaining 50.4% of dogs had a variety of nonspecific CSF changes. The CSF from dogs with a meningioma often may have an elevated white blood cell count (> 50/ul), with more than 50% of these cells being polymorphonuclear leukocytes. Neoplastic cells may be present in CSF, particularly when sedimentation techniques are used for analysis. The use of CSF protein electrophoresis, and IgG index of CSF, may aid in the determination of the presence of a brain neoplasm. Little information is available regarding CSF alterations seen in association with feline brain tumors; however, changes appear to be similar to those described for dogs.
Skull radiographs are useful in patients with suspected bony or cartilaginous changes (e.g., head trauma, bony tumors), but in general plain skull films are of limited value in patients with brain disease. Angiography and venography were used in the past to try to diagnose brain disorders. These techniques have severe limitations and in most instances fail to define the exact extent of a neoplasm and its precise relationship to surrounding structures. These techniques have now been replaced with computed tomography (CT) and magnetic resonance imaging (MRI).
CT and MRI
CT and MRI allow imaging of brain tissue rather than just the surrounding bony skull. Both can distinguish lesions which have only slightly different densities than the surrounding tissues and this can be further enhanced by contrast agents allowing the identification of masses and other abnormal tissues within the brain. Images obtained by means of MRI may be superior to those of CT, especially in certain areas such as the brain stem, although CT usually is better for bony lesions (e.g., middle ear studies). While the major tumor types are reported to have characteristic CT or MRI appearances, nonneoplastic lesions may mimic the CT or MRI appearance of a neoplasm, and occasionally a metastasis may resemble a primary brain tumor on CT or MRI images. Patients for either CT or MRI must be anesthetized and intubated. Proper patient positioning is extremely important. The animals should be placed in sternal recumbency with the head extended. The entire calvaria should be examined in the non-contrast series of images. This should be followed by a post contrast series of images.
At the present time, biopsy is the sole method available for the definitive diagnosis of brain lesions (including tumor type) in cats or dogs. Biopsy methods include ultrasound-guided biopsy, and CT-guided biopsy. Rapid cytological evaluation of a brain lesion from a biopsy sample can provide crucial information on operative management, medical management, chemotherapy, or radiation therapy. Smear preparations are generally wet fixed in 95% alcohol and stained with hematoxylin and eosin although toluidine blue, Giemsa, or Papanicolaou's stain also may be used.
In a recent study in which the results from the smear technique were compared with those from sections of paraffin embedded tissue, the overall diagnostic was about 80%. The main advantages of this method of intraoperative diagnosis are speed, ease of preparation, technical simplicity, need for minimal equipment, high degree of cytological resolution compared to frozen preparations, low cost and small sample size required. A limitation of this system is that it is difficult to prepare adequate smear preparations in certain tough and coherent tumors (e.g., schwannomas, fibrillary astrocytomas, and some meningiomas). Smear preparations provide excellent cytologic detail, however these differ from the conventional histologic appearance of HE stained paraffin-embedded tissue. Experience is required in the correct interpretation of smear preparations.
1. LeCouteur RA: Tumors of the nervous system. In Withrow SJ, MacEwen EG (ed): Small Animal Clinical Oncology, 3rd ed. Philadelphia, WB Saunders, 2001, pp. 500-531.
2. Vernau KM, Higgins RJ, Bollen AW, et al.: Primary canine and feline nervous system tumors: Intraoperative Diagnosis using the smear technique. Vet Pathol 38: 47-57, 2001.
3. Koblik PD, RA LeCouteur, RJ Higgins, et al.: CT-guided brain biopsy using a modified Pelorus Mark III stereotactic system: Experience with 50 dogs. Veterinary Radiology and Ultrasound 40:434-440, 1999.
4. LeCouteur RA: Cerebral meningiomas: Diagnostic and therapeutic considerations. In August JR (ed): Consultations in Feline Internal Medicine, Volume 4, WB Saunders Company, Philadelphia, pp 385-292, 2001.