An ever growing literature is providing evidence that levodopa, the major therapeutic agent for Parkinson's disease, and its metabolites (dopamine and quinone derivatives) are toxic to neurons, particularly in tissue culture. Neurons do not survive well in the presence of levodopa and undergo apoptosis. Toxicity has also been seen in somein vivomodels. The most commonly suggested mechanism by which levodopa and dopamine induce toxicity is by promoting oxidative stress. In support of this theory are findings that certain anti-oxidants, including sulphydryl compounds and ascorbic acid, decrease the neurotoxic effect. On the other hand, there are also a few studies that have failed to show that levodopa and dopamine are neurotoxic in animal models.There is no evidence that levodopa is toxic in humans, but the findings of neurotoxicityin vitroand in somein vivomodels raise concern about the safety of levodopa in humans. There is a need for clinical research to clarify this issue. A clinical trial that evaluates whether there is hastening of Parkinson's disease in patients treated with levodopa might be able to answer this question, and, fortunately, such a clinical trial is planned. The results should provide clinicians with information on how best to proceed in treating patients who have Parkinson's disease. In the meantime, until this information is available, the use of levodopasparing strategies, particularly for young patients who would be taking levodopa for a great many years, may be appropriate. Such a strategy calls for utilising non-levodopa medications that have symptomatic effects in the place of levodopa or as adjuvants to keep the dosage of levodopa as low as possible. Another approach is to develop agents that block the neurotoxicity of levodopa and dopamine, and which can be given in combination with levodopa.