Marked, disabling fluctuations in motor performance (on-off phenomena) appear after chronic L-Dopa therapy in Parkinson's disease (PD). Intravenous infusion of L-Dopa greatly reduces these motor fluctuations, but it is not reliable as a chronic treatment. Therefore, infusion of the potent, water-soluble dopaminergic agonist lisuride has been tested. However, many patients did not respond to infusion of lisuride alone, and no clinical parameter is known to correlate with the lacking response. In order to study this problem, we performed the TRH test (200 micrograms i.v.) in 8 PD patients with severe motor fluctuations; before and during lisuride subcutaneous infusion, we measured PRL and TSH responses to TRH. Both PRL and TSH receive an inhibitory control from dopaminergic receptors on pituitary cells, whereas they are stimulated by TRH. The TRH test, given during lisuride infusion, allows an indirect evaluation of the 'brake function' of the dopaminergic system on anterior pituitary, i.e. of dopaminergic receptor sensitivity in vivo. In our study, TRH induced a significant TSH rise in all PD patients, before and during lisuride infusion. Moreover, the lisuride responders (i.e. patients showing constant 'on' period during lisuride infusion, 4 patients) showed a significant lower TSH response as compared to nonresponders. PRL levels followed the same trend without reaching statistical significance. These data are compatible with the presence, in the two groups, of a different pituitary dopaminergic sensitivity which would suggest the presence of pharmacodynamic factors associated with the lacking response to intravenous lisuride infusion.