SummaryTrans-oesophageal echocardiography and continuous monitoring of the partial pressure of carbon dioxide at end expiration (PETCO2) were performed in 63 dogs during total hip replacement and in three dogs during femoral stem revision. The PET CO2 usually did not change during insertion of the acetabular prosthesis, but in 17.5% of the cases, it decreased by more than 10 mmHg after insertion of the femoral prosthesis. In all of the cases, transoesophageal echocardiography revealed echogenic structures in the right side of the heart, immediately after insertion of the acetabular and of the femoral prostheses. These appeared as either numerous pinpoint echoes, which passed through the right side of the heart in a “snow flurry” like fashion, or more echogenic particles. The embolaemia, seen ultrasonographically, was categorized into one of six grades of severity. Statistical analysis showed significant positive correlations between the severity of the embolaemia and the extent of the decrease in PETCO2 (rs = 0.405; P = 0.001; n = 60), and between the severity of the embolaemia and the duration of the decrease in PETCO2 before it returned to baseline values (rs = 0.345; P = 0.007; n = 60).Trans-oesophageal echocardiography demonstrated that embolaemia was induced during cemented prosthesis insertion. It was proposed that an increase in intramedullary pressure, during insertion, forces bone marrow into the venous circulation. Bone marrow itself can constitute an embolus, but also it may activate intravascular coagulation, thus leading to thromboembolism, in addition to bone marrow embolism. Elongated particulate emboli, up to several centimetres in length, were seen on ultrasonogrammes and were thought to be, at least in part, products of intravascular coagulation. A significant decrease in the platelet concentration in peripheral venous blood, after insertion of the femoral prosthesis, was also indicative of intravascular coagulation. The extent of the impairment of gas exchange was documented by means of capnography. The significant positive correlation between the embolaemia, seen via ultrasonography, and the decrease in PETCO2 strongly suggest a causal relationship between the two variables. The rapid recovery, that occurred in most patients, is attributed to the enormous lytic ability of pulmonary tissue. Prophylactic measures include the maintenance of normovolaemia during the operations, thorough lavage of the intramedullary cavity before prosthesis insertion and steps to prevent an increase in intramedullary pressure during implantation.Using trans-oesophageal echocardiography, embolaemia in the right side of the heart was consistently detected immediately after insertion of cemented acetabular as well as cemented femoral prostheses in dogs. There was a significant correlation between the severity of the embolaemia and the decrease in carbon dioxide at end expiration (PETCO2) after femoral prosthesis insertion.