Cardiopulmonary bypass (CPB) is a component of cardiac surgeries which requires full systemic heparinization before it is started^[1^]. Protamine is administered to neutralize the heparin's anticoagulant effect after CPB to return to normal hemostasis. Activated coagulation time (ACT) has been widely used to monitor the heparin's intraoperative anticoagulant effect, it needs to exceed 480 seconds before starting CPB and it is required to be equal to the preoperative value after neutralization. However, we found less bleeding in some cases whose final ACT after neutralization was a little lower than pre-heparin ACT. In these cases, the closing time was shorter than the pre-heparin one. Thereafter, a retrospective observational study was performed on patients to compare the influence of the final ACT on the outcomes of cardiac patients.

This is a retrospective observational study conducted from January 2014 to May 2017. All patients undergoing on-pump coronary artery bypass grafting (CABG) were enrolled. Exclusion criteria included patients: 1) with liver disease, 2) renal disease, 3) known bleeding diathesis, 4) who received antiplatelet (aspirin) or fibrinolytic agents within 5 days before surgery, 5) with platelet (PLT) count < 100*10⁹, 6) abnormal preoperative thromboelastography (TEG) parameters, 7) preoperative infections, 8) preoperative anemia, and 9) incomplete clinical records. Finally, 398 patients were enrolled in this study. This study was approved by the Clinical Trial Ethics Committee of the 1^st Central Hospital of Tianjin (Certificate no: E2016023L).

Patients were divided into 2 groups according to the final ACT after neutralization before they were transferred to intensive care unit (ICU). Patients whose final ACT was lower than preoperative ACT were in the Group A and those with final ACT equal to or higher than preoperative ACT were in the Group B. The following parameters were collected from our database: gender, height, weight, body surface area, CPB time, aortic cross-clamp time, minimum temperature reached during CPB, basal ACT, initial heparin bolus, post-heparinization ACT, ACT levels during CPB, final ACT, closing time, intraoperative blood loss, the first 12 hours and total bleeding from the chest drainage, need for transfusion in the ICU, and need for reoperation resulting from bleeding and other complications.

All patients' preoperative characteristics are shown in Table 1. Patients' characteristics had no statistical difference between the two groups.

All patients' perioperative characteristics are shown in Table 2. Patients' liver function, renal function, coagulation function, and homeostasis had no statistical difference between the two groups.

Protamine has been routinely administered after CPB in order to neutralize the effects of heparin for a long time. The dose of protamine used for neutralization was verified by the ACT value, which must be within normal parameters (100-140 seconds) or at the basal pre-heparin value observed in any case. Many studies aimed to evaluate the dose of protamine for neutralization^[2^,3^] which can contribute to coagulopathy if there is an excess^[4^] or deficit. However, the protamine dose for neutralizing the heparin effects was influenced by multiple factors, such as hypothermia, hemodilution, homeostasis, and so on^[5^,6^]; because of that, a dosage cannot be determined, so most centers used a dose of protamine equal to 1 or 1.5 times the initial dose of heparin and the final ACT was returned to the basal value or a little higher than the pre-heparin value.

However, we found out that after administering protamine (1:1 heparin) and adding additional protamine according to the ACT value after the first neutralization, some cases presented ACT values a little lower than pre-heparin ACT before transfer to ICU. According to the anesthesiologists' experience, because of ACT value is always higher than pre-heparin ACT after first neutralization, 10-25 mg of additional protamine would be administered. So, after 20-30 minutes, the ACT value would be lower than pre-heparin ACT in some cases. Still, the closing time was quicker than of patients with final ACT longer or equal than pre-heparin ACT. In addition, after transfer to ICU, the pericardial and mediastinal drainage in the first 12 hours was also decreasing. So, we retrospectively analyzed cases with final ACT a little lower than pre-heparin ACT and their closing time, bleeding of first 12 hours, and perioperative complications and then compared these situations with cases whose final ACT was longer than or equal to pre-heparin ACT. In our study, we aimed to evaluate the final ACT after the last neutralization, so we can skip many interference factors, such as internal environmental disorder and fluctuation of temperature on coagulation.

In recent years, TEG^[7^] has become available to measure several aspects related to coagulation. It can show several coagulation aspects, such as blood heparin levels, parameters, and the shortage of PLT. We used TEG to evaluate the PLT function and fibrinolysis situation after neutralization, which was influenced by heparin, protamine^[8^,9^], CPB, and other factors, to exclude the influence of coagulation on bleeding and closing time. In addition, we monitored other parameters, like temperature and blood gas, before measuring the last ACT to exclude other factors that could influence on the closing time and postoperative bleeding. There were no significant differences between the two groups. Also, heparin rebound, hyperfibrinolysis, and an acquired PLT defect were considered to be the main contributors to postoperative bleeding.

Heparin rebound^[10^] has been identified in many studies. This occurs because a proportion of heparin remains nonspecifically bound to plasma proteins and vascular cells, which can not be cleared by protamine, and that dissociates over time to produce an anticoagulant effect. The incidence of heparin rebound varies widely in the literature and it has been reported to be as high as 50%^[11^,12^]. Our study showed that incomplete heparin reversal and heparin rebound were a very common phenomenon after CPB, occurring in 75% of patients in Group A and 74.5% of patients in Group B. Heparin rebound would cause ACT rising directly, and according to this rising ACT, an additional small dose of protamine (25 mg) was used in Group B; but no protamine was used in Group A because the rising ACT was at the same level as the pre-heparin ACT. Although the different side effects of the administration of an additional small dose of protamine, such as anaphylactic reaction, hypotension, and pulmonary hypertension, were not found in Group B, in the Group A, these sides effects were avoided by low ACT before returning to ICU which was safer than the Group B.

Cardiopulmonary surgery is an inflammatory condition that can stimulate the synthesis of acute phase proteins that could contribute to increased protein binding. In addition, prolonged blood contact with the artificial surface of CPB will also affect PLT function which influences on the postoperative bleeding. Surgical trauma, high doses of heparin, and hypothermia all induce the activation of the inflammatory, coagulation, and fibrinolytic systems and the PLT dysfunction, leading to postoperative coagulopathy. After comparing all TEG parameters, preoperatively and after the last neutralization, no significant difference was found between the two groups and the PLT function was normal, probably due to the short CPB time. The PLT count is normal in both groups with no significant difference on the routine blood test.

The authors evaluated closing time, postoperative bleeding, and need for blood product transfusions. The time of operation was shorter in the Group A due to the quickly closing time than in the Group B. The hourly blood loss reduction between the 2 groups was modest (15-30 mL per hour on average), but there was a significant difference in the first 3 postoperative hours. There was no difference after 4 hours, because of the additional protamine in the Group B after the finding of more drainage. We just analyzed the first 12 hours drainage because after this period, drugs would be administered to some of the patients, which would interfere on coagulation factors. There were more blood product transfusions in the Group B than in Group A because of more drainage.

So, why Group A patients with lower final ACT than pre-heparin ACT had positive effect on reducing closing time and postoperative bleeding? Firstly, ACT of all patients was longer than normal preoperative ACT because low molecular weight heparin (LMWH) as anticoagulant was administered 2 times a day until the operation day. The ACT of all patients was longer than normal and always up to 200-220 before operation. Thus, even with neutralization lower than pre-heparin, the final ACT in Group A before leaving the operation room was also longer than normal. The reversion of the heparin effects after CPB referring to abnormal preoperative ACT was not useful for these patients in need of anticoagulation preoperative therapy. If the protamine has neutralized all heparin in the blood, the additional protamine was used to neutralize LMWH (60%), which made the ACT lower than the preoperative ACT. Secondly, heparin rebound was found in almost all patients, which would cause ACT longer after transfer to ICU. So, in Group A, the final ACT was lower than pre-heparin ACT after neutralization and remained equal or a little higher than pre-heparin ACT after heparin rebound, which did not require the use of additional protamine to neutralize it. But, in the Group B, ACT rose much higher than preoperative ACT after heparin rebound, so additional protamine had to be used for neutralizing it to prevent bleeding.

Some people may say that the final ACT lower than the pre-heparin ACT after neutralization may suggest the excessive use of protamine. It is also well documented that the excessive protamine leads to prolonged ACT, weakened clot structure, altered clot kinetics, and PLT dysfunction with subsequent anticoagulant action^[13^], especially if more than 1.3 mg of protamine, per each milligram of heparin, is administrated. So, we used TEG to evaluate the fibrinolytic systems and PLT dysfunction between the two groups after the final neutralization and there were no differences in all TEG parameters.

In summary, we have demonstrated that lower final ACT after neutralization is safe and has the advantage of leading to lower operative times, bleeding, and post-operative transfusions.