Phenytoin toxicity: predictors of clinical course.

The records of 46 patients who were admitted to a general hospital with the diagnosis of phenytoin toxicity were retrospectively studied to identify factors present at the time of admission which correlated with severity of illness and which would therefore be of prognostic value. Length of hospital stay was used as a measure of severity of illness. Correlations were made between the length of hospital stay and 18 variables studied at the time of admission, including severity of symptoms, use of other drugs (sedative hypnotics, anticonvulsants and phenothiazines), history (seizures, cardiac arrhythmias, and alcohol abuse), laboratory evidence of liver disease or renal disease, electrolyte abnormalities, coagulopathies, prior suicide attempts, glucose levels, and white blood cell counts. Significant correlations related the length of hospital stay with the severity of symptoms, concurrent phenothiazine usage, and the presence of abnormal liver function tests on admission, but not with other factors studied. Admission phenytoin serum levels following an overdose were not a useful predictor of length of hospital stay in this series of patients.