Leakage of viscous bone cement into venous blood possibly resulting in pulmonary embolism may occur during percutaneous vertebroplasty. Our aim was to study if bone cement surface or cement liquid component could induce platelet aggregation or plasma coagulation in vitro. Two types of commonly used methyl-methacrylate bone cement, Palacos (Heraeus Kulzer, Germany) and Vertebroplastic (DePuy, Acro Med, England), were smeared on thin glass slides that were inserted over the bottom of cuvettes immediately or after 24 h, and platelet aggregation was recorded over 10 min. Bone cement liquid component, containing methyl-methacrylate monomer and N,N-dimethyl-p-toluidine, was tested in 2% and 4% final concentration. Partial thromboplastin time (PTT) was determined by the hook method in the presence of bone cement-smeared glass slides or 6% bone cement liquid. Both types of bone cement, either fresh or aged, did not promote platelet aggregation, whereas collagen-coated glass slides induced substantial platelet aggregation (65 +/- 37%). On the other hand, bone cement liquids reduced platelet aggregation induced by collagen solution to an average of less than 15% (p < 0.01). Bone cement, fresh or aged, had no effect on PTT, but bone cement liquids significantly prolonged PTT: median and 1st-3rd interquartile range 149 (96-171) s for Vertebroplastic and 132 (99-194) s for Palacos, p = 0.03 for both comparisons with normal pool plasma without additives that had PTT of 69 (62-71) s. We conclude that the surface of fresh or aged bone cement is not thrombogenic in vitro. The bone cement liquid inhibits platelet aggregation and plasma clotting in relatively high concentrations that cannot be expected in vivo.