Cardiac surgery is a major consumer of blood products worldwide. The nature of the surgery when the heart-lung machine is used, is such that the patient's blood is significantly diluted by intravenous fluids. This dilution further compounds the blood's ability to form blood clots to seal the wound, thus increasing bleeding after surgery. The problem of this dilution is particularly severe in patients with low body weight, of whom a large percentage are women undergoing heart surgery. As a result, in this population there is an increased risk of needing a blood transfusion during and after surgery. The effects of this dilution can be potentially reversed by filtering the blood and removing the fluid after the principal part of the procedure has been completed through a process referred to as modified ultrafiltration (MUF). This procedure has been shown to be effective in several small clinical studies, however the interpretation of the results has been complicated by the fact that the studies were not "blinded". In other words, because the physicians and surgeons making the decisions about blood transfusions knew what treatment the patients received (i.e. MUF or not) it may have biased their judgement and affected the validity of the findings of the studies. The investigators believe that MUF may be a useful procedure to limit blood transfusions, particularly in patients of low body size, and weight undergoing heart surgery. This can only be demonstrated in a large clinical trial, and in the best case scenario, in a trial in which total blinding has been undertaken. However, blinding in this manner is quite difficult to achieve and it is necessary first to demonstrate that this is possible. Further, as the trial will involve several heart centers, it is essential to demonstrate that the relatively complex study interventions can be carried out in each of the centers in a reproducible manner. With this data in hand, the investigators will be able to submit for funding at a later date for a very large trial to determine if MUF decreases the need for blood transfusions in heart surgery patients. The investigators will also be able to determine the effect of this in terms of helping conserve blood as a benefit to the blood-banking agency (the Canadian Blood Service).

The demographic of the cardiac surgery population has progressively changed such that there is a greater proportion of older patients and females who by virtue of their co-morbidities and smaller average size have an increased risk of receiving blood products during and after surgery. Cardiopulmonary bypass (CPB) is the major contributing factor to this adverse event, as the technique is not only associated with coagulation and platelet abnormalities, but also by necessity, significant hemodilution as a consequence of the pump prime and the cardioplegia solution used to induce cardiac arrest. The hemodilution and the resultant decreased hematocrit (Hct) may trigger transfusion by the surgical team and there is evidence that platelet and coagulation function may be further compromised, resulting in worsening of postoperative bleeding. Modified ultrafiltration (MUF), a technique commonly utilized in pediatric cardiac surgery to reverse the effects of hemodilution after CPB, utilizes a circuit with an inflow attached to the aortic cannula, an ultrafiltration device, and outflow of the hemoconcentrated product directed back into the right atrium. Although MUF has been demonstrated to be effective at increasing the Hct, this technique has not received general acceptance as a blood conservation technology in the adult cardiac surgery population for several reasons. First, the necessary prolongation of the operative time with MUF may be perceived to be inconvenient by the operating staff and this may dissuade their use of this technique. Second, the results from the limited adult trials utilizing MUF, though positive in terms of blood conservation, have not shown changes of a magnitude that would routinely impact standard of care. It is possible that the perceived benefits of MUF in these trials have been less than impressive primarily due to the inclusion of patients at low risk of bleeding and the lack of blinding, thus contributing to the concern of treatment bias. Finally, none of these trials have been blinded and thus treatment bias may have contributed to the results. We believe that ultimately the use of MUF in this target population must be tested in a double-blind randomized controlled trial. However, the logistics of carrying out a trial of this magnitude must be established beforehand in a pilot trial that convincingly demonstrates the feasibility of blinding of this operative intervention, as well as the reproducibility of the model in several clinical centers. We are proposing to carry out a two phase multi-centre (University of Ottawa Heart Institute [OHI], Kingston General Hospital, Royal Victoria Hospital, University Hospital London) randomized controlled trial to evaluate the feasibility of an experimental model testing MUF versus a "sham" circuit in low-body weight (= 65 kg) patients at high risk to receive transfusion. In Phase I (n = 16 patients - 4 patients/center), the model will be optimized, and in Phase II (n = 64 patients - 16 patients/center) the project will be expanded and the blinding tested. The primary outcome will involve the success of blinding of the treating physicians (surgeon, anesthetist, and intensivist) in Phase II as measured by a novel Blinding Index. The information derived on the study model will be used to support a proposal in the subsequent definitive multi-center trial for which we will seek funding in the following year (n = 278/group). The primary outcome of the definitive trial will address the mean number of units of blood transfused in patients treated with or without MUF. There is also evidence in an animal model of the benefit of MUF to limit neurologic injury related to cardiac surgery. This aspect has not been assessed in a clinical model and therefore we intend to address this by evaluating the impact of MUF on early and late neurocognitive dysfunction related to CPB as a secondary outcome in the definitive trial. The proposed pilot and the definitive trial have the potential to address the role of MUF in adult cardiac surgery as a blood conservation technology. The potential application has significant implications as the procedure is inexpensive, easily teachable and physiologically sound. Further, it has very considerable relevance to women undergoing heart surgery due to their relative small body weight and increased current transfusion risk. As such, we believe that the magnitude of the findings from these trials will greatly impact current care in cardiac surgery worldwide.

Inclusion Criteria:

• Adult cardiac surgery patients with weight ≤ 65 kg or body surface area (BSA) < 1.7 m2

Exclusion Criteria:

• Emergency surgery
• Abciximab (Reopro™) use < 7 days
• Inability to obtain consent
• Age < 18 years
• Patients undergoing off-pump surgery
• Preoperative anemia