Coagulation and neonatal bypass: an assessment of changes in coagulation factor concentration during cardiopulmonary bypass in neonates, with modified ultrafiltration

Bleeding is of major concern in neonates after cardiopulmonary bypass. More severe coagulopathy can be expected in children who weigh under 8 kg. The effect of cardiopulmonary bypass on factor concentration has been described previously using a prime of fresh whole blood. Fresh whole blood is not generally available in the UK. In addition we describe the effect of modified haemofiltration and changes in factor V concentration, which has not previously been described.

Ethical approval was gained and consent was obtained from the patient's parents. An observational design was adopted for the present study. Cardiopulmonary bypass was conducted as is routine at the Freeman Hospital. The pump prime had an average volume of 900 ml. This consisted of one unit of fresh frozen plasma, packed cell to target haematocrit of 25%, sodium bicarbonate, heparin, and a balanced crystalloid solution. All patients were subjected to extreme hypothermia (17–21°C), received heparin (3 mg/kg) and protamine reversal, and received modified ultrafiltration after separation from bypass. Blood samples were taken before bypass, on full flow, at minimum temperature, after rewarming, after modified ultrafiltration and on return to intensive care. The samples were analyzed for haemoglobin, platelets, standard coagulation tests and colloid osmotic pressure. The samples were stored for analysis of factors II, V, VII, VIII, IX and X, antithrombin III, and markers of activation (TAT, PTf 1 + 2, D-dimers).

Fourteen patients were recruited (age 3–21 days). Thirteen patients had arterial switch procedures, one patient had repair of an interrupted aortic arch. Weights were 2.2–4.4kg.

Before bypass most levels were at the low end of the expected range for neonates of this age. After initiation of bypass, factor concentrations fell to extremely low levels. With the exception of factors V and VII, the change was predictable from the degree of haemodilution with the pump prime. Factor V levels fell to less than 50% of predicted activity. During hypothermic bypass there was little change in factor activity or evidence of blood activation.

After discontinuation of bypass and modified ultrafiltration, factor concentrations rose to an unpredictable degree. Although factor V activity increased by almost 200%, platelet counts rose by only 4%. In addition there was evidence of marked activation at this stage with increases in TAT, PTf 1 + 2, and D-dimers levels. On return to intensive care factors V and VIII levels were below the expected range in the majority of patients (11 and 7 patients, respectively). Fibrinogen levels were below age normal in eight patients, but were greater than 0.5 g/dl in all but one.

The low levels of factor V on bypass are of interest. Further work has demonstrated that loss of factor V occurs as soon as the FFP is added to the circuit (without connection to the patient). This suggests absorption into the circuit plastics.

The factor deficiencies during bypass are profound but predictable. The increase after ultrafiltration points to a further advantage of this technique, but evidence of platelet consumption is a concern. Continuing deficiencies of factors VIII and V may add to bleeding postoperatively. Factor V concentrate is not available.