Levodopa combined with a peripheral dopa-decarboxylase inhibitor (DCI) has been considered the therapy of choice for Parkinson's disease (PD). Levodopa is nearly always effective, but has a high incidence of adverse effects with long term use, including response fluctuations (on/off phenomena) and dyskinesias. Dopaminergic agonists, acting directly at the receptor level, would be able to decrease the incidence of these motor complications.In progressive neurodegenerative diseases, such as PD, modification of the rate of disease progression (often referred to as neuroprotection) is currently a highly debated topic. Increased oxidative stress is thought to be involved in nigral cell death, that is characteristic of PD. This oxidative stress may be further exacerbated by levodopa therapy. These mechanisms have been proven in vitro and animal models, but it's relevance in humans remains speculative.Based on the considerations above, the emerging therapeutic strategies for PD advocate early use of dopamine agonists in the treatment of PD. A number of recent well-controlled studies have proven the efficacy of dopamine agonists used as monotherapy. Moreover, as predicted by animal studies, on the long term, dopaminergic agonists induce significantly less motor complications than levodopa.In the last 2years, three new dopamine agonists have been launched, including ropinirole, pramipexole and cabergoline. These new agonists have been added, as therapeutical options to well-established drugs, like pergolide, bromocriptine or talipexole. The recently launched compounds have proven efficacy in monotherapy and as adjunctive therapy to levodopa. Unfortunately, only a very limited amount of comparative data among the different agonists is available. Pergolide has proven to be a superior drug to bromocriptine as adjunctive therapy to levodopa in a significant number of studies and is considered the gold standard dopamine agonist. Nevertheless, none of the recently launched compounds has compared itself against pergolide.A comparison of monotherapy trials is difficult, because of differences in design and populations. In a recently completed trial pergolide was statistically significantly better than placebo in all the efficacy parameters tested, with 57% of pergolide treated patients improving over 30% in the motor section of the UPDRS, as compared to 17% in the placebo arm. Interestingly, these results were obtained in the absence of any other antiparkinsonian drug during the trial. Recent monotherapy trials done with ropinirole and pramipexole achieved also significant improvements as monotherapy, but in these cases selegeline, a drug that causes a symptomatic improvement in PD, was allowed as co-medications during the trial. Not all trials used the same efficacy measures, i.e. monotherapy trials with pergolide and ropinirole used a "responder" based analysis (responder were all patients that improved 30% or more on the motor section of UPDRS), as well as a baseline to endpoint improvement in motor scores. Pramipexole monotherapy trials used only the latter approach, which is clinically less powerful than a responder analysis.Even with the difficulties mentioned above, all the recent trials with dopamine agonists have proven that these drugs are a useful symptomatic long term treatment for PD with or without levodopa and that the early use of dopamine agonists reduces the incidence of motor complications as compared to levodopa.