Superiority of centrifugal pump over roller pump in paediatric cardiac surgery: prospective randomised trial

The merits of centrifugal pump in adult cardiopulmonary bypass are well established. This study compares the effects of the Medtronic Biomedicus centrifugal pump with conventionally used roller pump in routine cardiopulmonary bypass in infants and children.

Between June 1996 and March 1997, 42 children (aged 2 days-13 years) undergoing elective cardiac surgery were assigned to either centrifugal or roller pump bypass. The following variables were studied: haemolysis (haematocrit, free plasma haemoglobin, haptoglobins), platelet activity (platelet counts, Beta-thromboglobulin), leukocyte count, cytokine release (IL-2, IL-6, IL-8), complement activation (C3a and C5a), blood and blood product requirements, urine output on bypass, post-operative blood urea, duration of ventilation, intensive care and hospital stay.

Age, weight, disease complexity, duration of bypass, and a number of other variables were comparable in the two groups. The centrifugal pump resulted in lower plasma free haemoglobin (mean +/- SD, 50 +/- 23 vs. 72 +/- 35 mg/dl, P < 0.01), higher platelet count (133.1 +/- 34.8 vs. 63.5 +/- 29.6 x 10(9)/l, P < 0.01), less platelet activation (beta-TG 1253 +/- 633 vs. 1657 +/- 677 ng/ml; P < 0.05), less cytokine release (IL-6 329 +/- 57 vs. 392 +/- 59 pg/ml; P < 0.05), and reduced levels of C3a (4822 +/- 274 vs. 5933 +/- 393 ng/ml, P < 0.01). Differences were detected in favour of the centrifugal pump in urine output on bypass (4.1 +/- 0.5 vs. 2.3 +/- 1.9 ml/kg per h, P < 0.01), post-operative maximal urea (6.5 +/- 3.1 vs. 10.2 +/- 6.7 mmol/l, P < 0.02), ventilation time (18.9 + 6.5 vs. 56.5 + 51.7 h, P < 0.01), duration of intensive care (1.4 +/- 0.79 vs. 3.33 +/- 2.8 days, P < 0.05) and hospital stay (5.7 +/- 1.4 vs. 15.75 +/- 23.9 days, P < 0.01), but not in blood and blood product requirements (RCC: 11.26 +/- 4.6 vs. 10.77 +/- 4.2 ml/kg per 24 h, P > 0.05).

The centrifugal pump as compared to roller pump results in less blood trauma, reduced platelet activation and less pronounced inflammatory response. There is also an improved renal response during and after bypass. This is translated clinically into reduced requirement for ventilation, shorter intensive care and hospital stays. These results strongly favour the use of centrifugal pump in routine paediatric cardiac surgery.