High incidence of antibodies to protamine and protamine/heparin complexes in patients undergoing cardiopulmonary bypass

Abstract

Protamine is routinely used to reverse heparin anticoagulation during cardiopulmonary bypass (CPB). Heparin interacts with protamine to form ultralarge complexes that are immunogenic in mice. We hypothesized that patients exposed to protamine and heparin during CPB will develop antibodies (Abs) to protamine/heparin (PRT/H) complexes that are capable of platelet activation. Specimens from a recently completed prospective clinical trial (HIT [for heparin-induced thrombocytopenia] 5801 study; n = 500) of CPB patients were examined for PRT/H Abs at baseline, at time of hospital discharge (between days 3 through 7), and 30 days after CPB. PRT/H antibody features were characterized and correlated with adverse cardiovascular outcomes. We found a high incidence of PRT/H antibody formation (29%) in patients undergoing cardiac surgery. PRT/H Abs were of high titer (mean titer 1:14,744), showed heparin-dependent binding, and activated platelets in the presence of protamine. PRT/H Abs showed no cross-reactivity to platelet factor 4/heparin complexes, but were cross-reactive with protamine-containing insulin preparations. In the absence of circulating antigen at day 30, there were no complications of thrombocytopenia, thrombotic events, or long-term cardiovascular events. These studies show that Abs to PRT/H occur commonly after cardiac bypass surgery, share a number of serologic features with HIT Abs, including platelet activation, and may pose health risks to patients requiring drug reexposure.

Figure 1

Incidence of PRT/H Abs in healthy patients and patients undergoing CPB. (A) PRT/H antibody reactivity was measured by ELISA in healthy patients (Controls, n = 101) and patients after CPB (n = 500) at baseline, days 3 through 7, and day 30 after CPB. (B) Time course of seroconversion in all seropositive patients (n = 154; A_450nm >1.2). Each line represents the seroconversion profile of a patient after CPB.

Figure 2

Serologic characteristics of PRT/H Abs. (A) Representative titration curves of high-titer PRT/H Abs in 19 patients demonstrating antibody titer range of 1:1800 to 1:175 000. (B) Specificity of high-titer PRT/H Abs with PRT/H antibody levels of A_450nm >3 (n = 32). Each symbol represents an individual CPB patient. (C) Heparin-dependent binding of PRT/H Abs of patients depicted in B.

Figure 3

PRT/H Abs activate platelets in a protamine-dependent manner. Washed ¹⁴C-serotonin–labeled platelets from healthy donors were incubated with or without protamine (PRT; 1.25 ug/mL) in the presence of 3 representative patient samples obtained at day 30 (CPB patients or control patients [CTRL]) with increasing amounts of unfractionated heparin (0, 0.25, 1, 100 U or 1U+ IV.3). ¹⁴C releasate was measured by liquid scintillation counting. These results are representative of 4 independent experiments.

Figure 4

Correlation of PF4/H and PRT/H seropositivity. (A) PRT/H results at day 30 plotted as a function of PF4/H positivity (A_405nm >0.4; R² = 0.001454). (B) PF4/H results at day 30 plotted as a function of PRT/H positivity (A_405nm >1.2; R² = 0.0008695).

Figure 5

PRT/H antibody cross-reactivity with NPH insulin. Plasma from 32 CPB patients with high PRT/H antibody reactivity (A_450nm >3.0) was analyzed for reactivity by ELISA on microtiter plates coated with NPH insulin, which contains protamine (diluted 1:160 in PBS), and with regular insulin, which does not contain protamine (diluted 1:160 in PBS; P = < .0001). Each symbol represents an individual CPB patient.