Platelet factor 4 (PF4) is a potent antiheparin in vitro. In view of the large amount of PF4 secreted from platelet alpha-granules during routine blood collection and processing techniques, the potential significance of this release was investigated using three measurements of heparin activity: the activated partial thromboplastin time (aPTT), the thrombin time, and factor Xa inactivation using the chromogenic substrate S2222 for assay of factor Xa. The results demonstrate that purified PF4 neutralizes heparin activity when added in increasing amounts to normal platelet-poor plasma containing a fixed concentration of commercial porcine gut mucosal heparin. This effect was seen when assaying heparin activity by all three methods. In addition, when heparin was added in increasing concentrations to pooled plasma samples that were collected from normal volunteers, there was neutralization of heparin activity in blood samples collected by routine citrate anticoagulation (CIT60) in comparison to blood samples collected simultaneously with platelet secretion inhibiting agents added to the anticoagulant (CIT+). This effect was seen when assaying heparin by the aPTT and thrombin time. These data confirm that both purified and secreted PF4 have significant antiheparin activity when heparin is added in vitro to normal plasma. Neutralization of circulating heparin by PF4 secreted during blood collection from anticoagulated patients could result in underestimation of the actual in vivo heparin concentration. In order to evaluate the significance of this effect, purified PF4 was added to plasma collected from heparinized patients and again PF4 neutralized heparin activity. This was seen, however, only when heparin activity was measured by the thrombin time or Xa inactivation assays. There was minimal shortening of the aPTT when PF4 was added in final concentrations up to 1000 ng/ml. When blood samples were simultaneously collected from anticoagulated patients by both routine and special collection methods, these results were confirmed. There was a significant difference between heparin activities measured in the CIT+ (secreted PF4 58 ng/ml) and CIT60 (secreted PF4 1074 ng/ml) plasma samples by both thrombin time and Xa inactivation. There was no difference, however, in the aPT when both types of plasma samples were simultaneously collected and assayed for each anticoagulated patient. This suggests that there may be circulating heparin fractions which can prolong the aPTT but which do not interact with PF4.(ABSTRACT TRUNCATED AT 400 WORDS)