In recent years, serum S100B has been used as a secondary endpoint in some clinical trials, in which serum S100B has successfully indicated the benefits or harm done by the tested agents. Compared to clinical stroke studies, few experimental stroke studies report using serum S100B as a surrogate marker for estimating the long-term effects of neuroprotectants. This study sought to observe serum S100B kinetics in PIT stroke models and to clarify the association between serum S100B and both final infarct volumes and long-term neurological outcomes. Furthermore, to demonstrate that early elevations in serum S100B reflect successful neuroprotective treatment, a pharmacological study was performed with a non-competitive NMDA glutamate receptor antagonist, MK-801. Serum S100B levels were significantly elevated after PIT stroke, reaching peak values 48 h after the onset and declining thereafter. Single measurements of serum S100B as early as 48 h after PIT stroke correlated significantly with final infarct volumes and long-term neurological outcomes. Elevated serum S100B was significantly attenuated by MK-801, correlating significantly with long-term beneficial effects of MK-801 on infarct volumes and neurological outcomes. Our results showed that single measurements of serum S100B 48 h after PIT stroke would serve as an early and simple surrogate marker for long-term evaluation of histological and neurological outcomes in PIT stroke rat models.