Pulmonary embolism in reamed femoral nailing has been reported and discussed over recent years. Does an unreamed nailing technique with a solid nail prevent this rare but serious complication of intramedullary fixation? In an animal model in rabbits, we studied the pathophysiologic impact on pulmonary function and the impact on hemostasis of reamed and unreamed nailing of intact femora and tibiae, and of femoral fracture in relation to intramedullary pressure. No statistical difference of PaO2, PaCO2, and PCO2et was found in the femur whether a reamed or unreamed procedure was performed. Two of six animals with unreamed femoral nailing, one of six animal with reamed femoral nailing, and one of five animals with a femoral fracture fulfilled four of four or three of four criteria for embolization (increase of the difference of PaCO2 and PCO2et, decrease of PaO2, increase of blast cells in central-venous blood and bone marrow/fat in histologic section of the lungs and bone). Tibial nailing did not alter pulmonary function in either group. Intramedullary pressure was increased in all animals with perioperative impairment of pulmonary function (375 to 676 mbar). Analysis of the hemostatic results showed a significant difference of platelet activation in reamed versus unreamed nailing of the femur 1 hour after nailing (p < 0.01) and a significant decrease of fibrinogen and antithrombin III (p < 0.001/p < 0.01) in reamed femoral nailing. We conclude that unreamed nailing of the femur with a solid rod may also cause bone marrow embolization with alteration of pulmonary function as long as an important increase of the intramedullary pressure is generated during the nailing procedure.(ABSTRACT TRUNCATED AT 250 WORDS)