INFLAMMATORY REACTION AND CAPILLARY LEAK SYNDROME-RELATED TO CARDIOPULMONARY BYPASS IN NEONATES UNDERGOING CARDIAC OPERATIONS

We studied the inflammatory reaction related to cardiopulmonary bypass in 24 neonates (median age 6 days) undergoing the arterial switch ope ration for simple transposition of the great arteries, with respect to the development of postoperative capillary leak syndrome. Complement proteins, leukocyte count, tumor necrosis factor-alpha, and histamine levels were determined before, during, and after cardiopulmonary bypas s. Additionally, protein movement from the intravascular into the extr avascular space during cardiopulmonary bypass was assessed by the meas urement of plasma concentrations of proteins with molecular weights ra nging from 21,200 to 718,000. Capillary leak syndrome developed in 13 of the 24 neonates. Patients with capillary leak syndrome, as compared with those without, had preoperatively higher C5a levels (C5a, 3.0 +/ - 0.6 mu g/L vs 0.9 +/- 0.2 mu g/L) (mean +/- standard error of the me an) (p < 0.05) and higher leukocyte counts (leukocytes, 17.9 +/- 2.1 x 10(3) cells/ml versus 11.7 +/- 0.8 x 10(3) cells/ml) (p < 0.05), sugg esting in these neonates a preoperative inflammatory state. Preoperati ve clinical and operative data were identical in both patient groups. Before cardiopulmonary bypass, serum protein concentrations were simil ar in all patients. Ten minutes after institution of cardiopulmonary b ypass, protein concentrations fell to significantly lower values in pa tients with capillary leak syndrome than in those without: albumin (19 % +/- 1.5% vs 30% +/- 6% of the prebypass value, p < 0.05), immunoglob ulin G (17% +/- 1.5% vs 29% +/- 5.5%, p < 0.001), and alpha(2)-macrogl obulin (15% +/- 1.2% vs 25% +/- 4%, p < 0.02). During cardiopulmonary bypass, albumin concentrations remained significantly lower in patient s with capillary leak syndrome than in those without, whereas hematocr it values were similar in both groups. During cardiopulmonary bypass, patients with capillary leak syndrome also had lower concentrations of complement proteins C3 and C4 but not C1 inhibitor. C3d/C3 ratio and C5a levels were similar in both patient groups. In contrast, histamine liberation during cardiopulmonary bypass was significantly more prono unced in patients with capillary leak syndrome than in those without ( 725.2 +/- 396.7 pg/ml vs -54.1 +/- 58.4 pg/ml, p < 0.05). Tumor necros is factor-alpha levels after protamine administration were also signif icantly higher in patients with capillary leak syndrome (38.1 +/- 10.0 pg/ml vs 15.3 +/- 3.4 pg/ml, p < 0.05). Leukocyte count during and af ter cardiopulmonary bypass was similar in both patient groups. This st udy demonstrates increased protein leakage as early as 10 minutes afte r initiation of cardiopulmonary bypass in patients having clinical sig ns of postoperative capillary leak syndrome. Patients with capillary l eak syndrome displayed more pronounced histamine liberation and tumor necrosis factor-alpha liberation than patients without capillary leak syndrome, suggesting a relationship between cardiopulmonary bypass-rel ated inflammatory reaction and perioperative capillary damage. A preop erative inflammatory state in patients with capillary leak syndrome co uld have enhanced the inflammatory response to cardiopulmonary bypass.