Methods: We retrospectively reviewed 24 patients seen at the Mayo Clinic in Rochester, Minn, between 1985 and 2002 in whom PLE was suspected. Patients were identified on the basis of clinical history, and presence of cancer. Data were reviewed from magnetic resonance imaging (MRI), cerebrospinal fluid (CSF) analysis, electroencephalography (EEG), and paraneoplastic serologic studies.

Results: Common manifestations were cognitive dysfunction (92%), seizures (58%), and psychiatric symptoms (50%); 13 patients had small cell lung carcinoma; 11 had other malignancies. Paraneoplastic neuronal autoantibodies were found in 14 (64%) of 22 patients tested. Electroencephalography showed focal or generalized slowing and/ or epileptiform activity, maximal in the temporal regions, in all 22 patients tested. Magnetic resonance imaging revealed increased T2 signal involving one or both temporal lobes in 15 (83%) of 18 patients. Cerebrospinal fluid test results were abnormal in 18 (78%) of 23 patients tested. Clinical or radiographic evidence of extralimbic involvement was documented in 12 (55%) of 22 patients. No abnormality on EEG, MRI, or CSF analysis correlated with a specific cancer type or with a specific paraneoplastic autoantibody.

Conclusions: In patients with suspected PLE, EEG is invaluable for confirming cerebral dysfunction. Magnetic resonance imaging can show unequivocal involvement of temporolimbic structures and helps exclude other diagnoses. When EEG and cranial MRI are both normal, PLE is unlikely. Comprehensive testing for paraneoplastic neuronal nuclear, cytoplasmic, and ion,channel autoantibodies is an important part of the evaluation, but negative results do not rule out PLE.