Procoagulant activity after off-pump coronary operation: is the current anticoagulation adequate?

Background . Hemostasis is preserved after off-pump coronary operations compared with conventional coronary procedures. However, this preserved hemostasis may result in a procoagulant activity.

Methods . We prospectively studied coagulation in 22 patients who underwent off-pump coronary operation either through a midline sternotomy (n = 14) or with minimally invasive anterolateral thoracotomy (n = 8).

Results. Procoagulant activity, represented by prothrombin factor 1 and 2, remained at baseline levels during operation but increased significantly on postoperative day 1. Factor VII remained at baseline levels during the operation but decreased significantly on postoperative day 1. Fibrinolysis was increased as indicated by the fibrin degradation products on postoperative day 1. A promoted hemostasis attributable to endothelial activation was indicated by the increase in von Willebrand factor on postoperative day 1. Platelets counts and platelet activation (<Рисунок: beta>-thromboglobulin) remained at baseline levels after the operation. No adverse clinical events occurred.

Conclusions . Patients undergoing off-pump coronary operation show an increased procoagulant activity in the first postoperative 24 hours regardless of the surgical approach (midline sternotomy or anterolateral thoracotomy). This procoagulant activity is not mediated by platelet-related factors. Therefore, a specific perioperative prophylactic pharmacologic regimen is advisable.

 

 

 

Recently, several minimally invasive techniques for off-pump myocardial revascularization have been reported [1] [2] [3] [4] [5] . In these minimally invasive techniques different accesses have been used to perform coronary operations without using cardiopulmonary bypass. These off-pump techniques for surgical myocardial revascularization have several advantages, either through a midline sternotomy [1] [2] (off-pump coronary artery bypass, OPCAB) or with minimal accesses [3] [4] [5] (minimally invasive coronary surgery, MICS). The main advantages of these techniques are the opportunity to avoid the use of cardiopulmonary bypass with its damaging effects [6] [7] [8] [9] [10] and to improve and expedite patient recovery. These advantages can ultimately optimize the use of health care resources [11] [12] . Furthermore, after off-pump coronary operation the hemostasis is preserved and postoperative bleeding is significantly reduced compared with procedures using cardiopulmonary bypass [1] [2] [3] [4] [5] [11] [12] . Consequently, the use of donor blood and derivates is significantly reduced as an additional effect.

However, the preserved hemostasis may result in a procoagulant activity after off-pump coronary operation. This procoagulant activity is a well-known phenomenon in major general surgery and it is also a phenomenon to be expected after major surgical procedures such as coronary operation. The procoagulant activity increases the risk of venous thrombosis and potentially endangers the patency of coronary anastomoses. In addition, operating on the beating heart and the use of minimal surgical access may increase the technical difficulty of coronary operation and some investigators have expressed concern for an excessive rate of graft failure [13] [14] [15] [16] [17] . Some of these failures and complications could be explained by the occurrence of graft thrombosis.

To investigate the procoagulant activity after off-pump coronary operation we prospectively studied 22 patients with single-vessel disease involving the left anterior descending artery (LAD) who underwent off-pump coronary operation. Because the extent of tissue damage may influence the procoagulant activity, the patients were randomly assigned to OPCAB or MICS. The main coagulation and clinical variables were measured intraoperatively and postoperatively.

This prospective study was designed to investigate whether a procoagulation activity can occur after off-pump coronary operation. Twenty-two patients with documented myocardial ischemia and single-vessel disease involving the LAD were assigned to two methods of off-pump myocardial revascularization: OPCAB through a midline sternotomy or MICS with anterolateral small thoracotomy. The main coagulation variables, prothrombin factor 1 and 2 (F1+2), factor VII, and fibrin degradation products, were recorded at scheduled intervals: just before the start of the operation (baseline), after heparin administration (T1), at the end of the operation (T2) and 24 hours later (T3). In addition, platelet counts, <Рисунок: beta>-thromboglobulin (a marker of platelet activation), and von Willebrand factor were recorded at baseline and T3. Clinical variables, such as activated clotting time (ACT), perioperative myocardial infarction, blood loss, blood transfusion, and hospital stay, were also prospectively recorded. Blood loss was measured at T3.

From May to October 1997, 22 consecutive patients with single-vessel disease entered the study and were randomly assigned to two treatments: 14 underwent OPCAB through a midline sternotomy and 8 underwent MICS with a small anterolateral thoracotomy. Mean age was 61.5 ± 11 years (range, 37 to 83 years). There were no differences in age and preoperative risk factors between the two groups. Inclusion criteria were the presence of single-vessel disease involving the LAD (type B2 or C stenosis [18] ) with documented myocardial ischemia and a normal or moderately depressed ventricular function (ejection fraction, > 35%). All patients underwent elective and first operations. Exclusion criteria were the presence of any associated cardiac disease requiring surgical treatment with the use of cardiopulmonary bypass (eg, left ventricular aneurysm or valvular disease), myocardial infarction either evolving or acute (within 48 hours) or recent (within 2 weeks). The use of intravenous heparin before operation was also considered as an exclusion criterion.

A panel of cardiologists and cardiac surgeons gave general consensus on the feasibility of both OPCAB and MICS. Then OPCAB and MICS were presented to the patients who were informed of MICS as an alternative treatment to OPCAB, which was presented as the standard treatment. The final choice of treatment was left to the patient, who gave informed consent.

Patients were prepared and draped as for a conventional coronary operation down to the transverse umbilical line. Down from this line the patients were covered with a warm air-cushion rewarming blanket to avoid heat dispersion and the consequent hypothermia during the procedure. The skin incision was begun 2 to 3 cm caudal to the sternal notch and was extended to a point 2 to 3 cm cranially to the xiphoid. A full midline sternotomy was performed. The left internal mammary artery was harvested as a pedicle with electrocautery from its origin to its distal bifurcation. Then heparin was given (100 IU/kg) and the mammary pedicle was divided. The ACT was checked 3 minutes later. A sternal spreader carrying a coronary stabilizer (MV Access Platform, CardioThoracic System, Inc, Portola Valley, CA) was positioned in the sternal wound. The pericardial sac was opened with a conventional reversed T incision and suspended with stay sutures. The coronary stabilizer was positioned and locked in place. The LAD was surrounded by two looping 5−0 polypropylene sutures, proximally and distally to the chosen site for the anastomosis. The LAD was opened and the mammary-to-coronary anastomosis was performed with a running 7−0 or 8−0 polypropylene suture. No protamine was given at the end of the procedure. The pericardium was closed after the procedure, the midline sternotomy was closed in layers, and three drains (one left pleural, one pericardial, one mediastinal) were left in place.

Patients were prepared as previously described [4] . A double-lumen endotracheal tube was used. A skin incision of approximately 8 to 10 cm was made in the fifth intercostal space. Once the pleural cavity had been opened, the left lung was deflated and the left internal mammary artery was identified by palpation.

The wound spreader (IMA Retractor, CardioThoracic System) was secured in place and gently opened to avoid rib fractures. The left internal mammary artery was harvested as a pedicle from the first rib down to the seventh intercostal space. Then heparin was given (100 IU/kg) and the mammary pedicle was divided. The ACT was checked 3 minutes later. The LAD was surrounded by two looping 5−0 polypropylene sutures, proximally and distally to the chosen site for the anastomosis. The coronary stabilizer (Stabilizer, CardioThoracic System) was positioned and locked into place. The LAD was opened and the mammary-to-coronary anastomosis was performed with a running 7−0 or 8−0 polypropylene suture. No protamine was given at the end of the procedure. The small thoracotomy wound was closed in layers and one pleural drain was left in place.

In both groups the patients were transferred to the intensive care unit still intubated. Extubation was scheduled in the intensive care unit after stabilization of hemodynamic and respiratory states.

All patients received 80 mg of oral aspirin and 7,500 IU Fraxiparine subcutaneously beginning on the afternoon of postoperative day 1. Subcutaneous Fraxiparine was continued until discharge and then suspended. Oral aspirin was continued after discharge. No additional anticoagulation drug was administered either preoperatively or on the day of the operation.

Blood samples were taken before surgical incision (baseline), 3 minutes after heparinization (T1), at the end of operation (T2), and 24 hours after operation (T3) from the indwelling radial artery catheter. Platelet count and hematocrit were measured from blood samples anticoagulated with 0.1 mmol/L ethylenediaminetetraacetic acid and counted by a cell counter (Cell-Dyn 610, Sequoia-Turner Corp, CA). Blood samples anticoagulated with 3.06% sodium citrate and hirudin were centrifuged at 1,000 g for 10 minutes. After centrifugation, the platelet-poor plasma was divided in separate aliquots and stored at -80°C for further determination of coagulation and fibrinolysis. Prothrombin F1+2 was determined by an enzyme-linked immunosorbent assay (Behringwerke AG, Marburg, Germany; normal values, 0.4 to 1.1 nmol/L). Factor VII was determined by a chromogenic assay (Chromogenix AB, Molndal, Sweden). Fibrin degradation was determined by an enzyme-linked immunosorbent assay (Organon Teknika, Boxtel, the Netherlands; normal upper limit, 650 ng/mL).

Activation of platelets was indicated by the release of <Рисунок: beta>-thromboglobulin as determined by a radioimmunoassay (Amersham International, Inc, Amersham, UK). The von Willebrand factor was determined by an enzyme-linked immunosorbent assay (Gradipore Ltd, North Ryde, Australia).

Statistical analysis

All clinical and laboratory data were collected prospectively in a customized database. All data were processed using the SPSS 7.5 statistical package (SPSS, Inc, Chicago, IL). All continuous variables are expressed as mean ± standard deviation. All tests of significance were two-tailed: <Рисунок: chi> 2 and Fisher's exact test were used for discrete variables and analysis of variance and Mann-Whitney for continuous variables. The Student's t test was used to compare the values of the variables between the two treatments and within each treatment at different times. We considered a statistical probability of less than 0.05 as indicative of significance.

All operations were uneventful. The ACT after heparinization was not different between the two groups (284 ± 26 seconds in the OPCAB group and 277 ± 6 seconds in the MICS group). No perioperative myocardial infarction occurred. Mean postoperative blood loss was significantly higher in the OPCAB group (771 ± 284 mL for OPCAB and 244 ± 171 mL for MICS; p = 0.03). Hematocrit on postoperative day 1 was not significantly different between the two groups (33.4 ± 0.4 for OPCAB and 36.8 ± 0.5 for MICS). Two patients in the OPCAB group required donor blood transfusion, whereas no patient in the MICS group received any donor blood. Mean postoperative intensive care unit stay was 1 day in both groups. Mean postoperative hospital stay was significantly shorter in the MICS group (6.2 days for OPCAB and 4.2 ± 0.5 for MICT; p < 0.01).

Procoagulant activity, represented by prothrombin F1+2, was low and stable during operation . [Table 1] There was no significant difference between the OPCAB and the MICS groups. However, prothrombin F1+2 increased significantly in both groups on postoperative day 1 compared with baseline as well as with the concentration at the end of the operation (T2; ). [Figure 1] Factor VII was stable during the operation but decreased significantly on postoperative day 1 in both groups , [Figure 2] indicating the consumption of coagulation factors through the extrinsic pathway. Fibrinolysis was also activated in both groups as indicated by the significant increase in fibrin degradation products on postoperative day 1 . [Figure 3] A promoted hemostasis as a result of endothelial activation was also present on postoperative day 1 as indicated by the significant increase in von Willebrand factor . [Table 2]

Platelets were not significantly different between the two groups on postoperative day 1 (179 ± 42 Ч 10 3 for OPCAB and 184 ± 46 Ч 10 3 for MICS). A slight decrease in <Рисунок: beta>-thromboglobulin [Table 3] was observed on postoperative day 1, which can be related to a slight degree of hemodilution. This indicates that no platelet activation occurred.

Cardiopulmonary bypass induces clotting disorders and platelet dysfunction [6] [7] [19] [20] [21] . However, such an effect on hemostasis, which increases postoperative bleeding and is usually considered as adverse, also has a desirable effect in protecting anastomosis patency. An impaired hemostasis prevents thrombosis of the coronary grafts, presumably until a full postoperative regimen of antiplatelet therapy is established. In addition, years of experience in peripheral vascular surgery have indicated the importance of an aggressive pharmacologic prophylactic regimen to prevent postoperative graft occlusion, in particular for small vascular anastomoses [22] . A crucial difference between a peripheral vascular anastomosis (ie, infrainguinal) and a mammary-to-coronary anastomosis is the impact of sudden occlusion. The occlusion of an anastomosis involving the LAD leads to catastrophic consequences. In our institute, during a 3−year experience in off-pump coronary operations, two sudden cardiac deaths and two nonfatal pulmonary embolisms (2 OPCAB) occurred in a total of 287 patients. These two sudden cardiac deaths (1 MICS and 1 OPCAB) were related to graft thrombosis in the presence of a patent anastomosis at autopsy. The two cases of nonfatal pulmonary embolism occurred in 2 patients with a further uncomplicated postoperative course.

Although we have not encountered clinical evidence of this problem in this series (all 22 patients were free from myocardial ischemia or pulmonary embolism postoperatively), the previous clinical experience and the result of this study may indicate a possible danger for the early graft patency as well as an increased risk of venous thrombosis.

In fact, the results of this study clearly show that patients undergoing both OPCAB and MICS have a procoagulant activity that is strongly evident 24 hours after operation. This is indicated by the increase in F1+2 and fibrin degradation products and by the decrease in Factor VII ( and ). The significant increase in von Willebrand factor on postoperative day 1 is further supporting this evidence. In fact, the increase in von Willebrand factor is an indirect sign of promoted hemostasis as a result of endothelial activation. Moreover, the procoagulant activity observed in this study is not influenced by the contribution of platelet-related factors. In fact, the total number of platelets remains unchanged and no platelet degranulation is detectable.

Current protocols of perioperative anticoagulation are substantially empirical [1] [2] [3] [4] [5] and derived from the protocols applied to patients who undergo conventional coronary operations with cardiopulmonary bypass. However, cardiopulmonary bypass is known to affect hemostasis and platelet function for at least 24 hours. Conversely, patients undergoing off-pump coronary operation are lacking the impairing effect of cardiopulmonary bypass on hemostasis.

In addition, Oltrona and colleagues [23] have reported that heparin administration cannot suppress thrombin activity in an undetected percentage of patients showing a heparin-resistant thrombin activity. Heparin-resistant thrombin activity is associated with the evidence of intravascular thrombosis, which can play a role in early occlusive complications as well as in promoting late restenosis [23] [24] . This intravascular thrombosis is most likely to occur when the ACT does not increase to more than 300 seconds after the administration of heparin [23] . In our series the average ACT was less than 300 seconds after the administration of 100 IU/kg heparin. Therefore, this dose of heparin, conventionally (and empirically) accepted for off-pump coronary operation, is not enough to track and subsequently to treat patients with heparin-resistant thrombin activity.

In conclusion, patients undergoing off-pump coronary operation face an increased procoagulant activity regardless of what surgical access is used (midline sternotomy or anterolateral thoracotomy). This increased procoagulant activity is independent of platelet activation. As a result of this increased procoagulant activity patients undergoing off-pump coronary operation should be considered at increased risk of thrombotic graft occlusion and pulmonary embolism in the first postoperative 24 hours. Therefore, we believe that the perioperative anticoagulation policy for patients undergoing off-pump coronary operation should be more aggressive than that for patients undergoing conventional coronary procedures with cardiopulmonary bypass. Therefore, a careful anticoagulation prophylactic regimen should be begun the day before operation and should be continued immediately after the operation up to the first postoperative 24 hours. In addition the dose of heparin administered intraoperatively should be enough to keep the ACT more than 300 seconds. We do not advise to antagonize the heparin with protamine at the end of the procedure, unless an uncontrollable diffuse bleeding should be present.

Prospective randomized trials comparing conventional postoperative treatment for coronary patients and a combined preoperative and postoperative treatment are advisable to clarify further this issue.

 

 

 

1.Benetti F.J., Naselli G., Wood M., Geffner L.. Direct myocardial revascularization without extracorporeal circulation. Experience in 700 patients. Chest 1991;100:312−316.

2.Buffolo E., Gomes W.J., Andrade J.C.. Myocardial revascularization without extracorporeal circulation. Surgical results in 1090 patients. Arq Bras Cardiol 1994;62:149−153.

3.Benetti F.J., Mariani M.A., Sani G.. Video-assisted mini-invasive coronary surgery without cardiopulmonary bypass. A multicenter study. J Thorac Cardiovasc Surg 1996;112:1478−1484.

4.Boonstra P.W., Grandjean J.G., Mariani M.A.. An improved method for direct coronary surgery without cardiopulmonary bypass via antero-lateral small thoracotomy. Ann Thorac Surg 1997;63:567−569.

5.Jatene F.B., Pego-Fernandes P.M., Hayata A.L.. VATS for complete dissection of the LIMA in minimally invasive coronary artery bypass grafting. Ann Thorac Surg 1997;63:110−113.

6.Butler J., Rocker G.M., Westaby S.. Inflammatory response to cardiopulmonary bypass. Ann Thorac Surg 1993;55:552−559.

7.Edmunds L.H.. Why cardiopulmonary bypass makes patients sick. Adv Card Surg 1995;6:131−167.

8.Gu Y.J., Mariani M.A., van Oeveren W M.A., Grandjean J.G., Boonstra P.W.. Reduction of systemic inflammatory response in patients undergoing minimally invasive coronary artery bypass grafting. Ann Thorac Surg 1998;65:420−424.

9.Kurusz M., Butler B.D.. Embolic events and cardiopulmonary bypass. Cardiopulmonary bypass. Williams and Wilkins,Baltimore, 1993, 267−290.

10.Roach G.W., Kanchuger M., Mangano C.M.. Adverse cerebral outcomes after coronary bypass surgery. N Engl J Med 1996;335:1857−1863.

11.King R.C., Reece T.B., Hurst J.L.. Minimally invasive coronary artery bypass grafting decreases hospital stay and cost. Ann Thorac Surg 1997;225:805−809.

12.Zenati M., Domit T.M., Saul M.. Resources utilization for minimally invasive direct and standard coronary artery bypass grafting. Ann Thorac Surg 1997;63:S84−S87.

13.Lytle B.W.. Minimally invasive cardiac surgery. J Thorac Cardiovasc Surg 1996;111:554−555.

14.Machiraju V.R.. A word of caution regarding minimally invasive coronary artery bypass procedures. Ann Thorac Surg 1997;64:294- .

15.Izzat M.B., Yim A.P.. Trouble-shooting in minimally invasive direct coronary artery bypass. Lancet 1997;350:665−666.

16.Pagni S., Qaqish N.K., Senior D.G., Spence P.A.. Anastomotic complications in minimally invasive coronary artery bypass grafting. Ann Thorac Surg 1997;63:S64−S67.

17.Alessandrini F., Gaudino M., Glieca F.. Lesion of the target vessel during minimally invasive myocardial revascularization. Ann Thorac Surg 1997;64:1349−1353.

18.Ryan T.J., Bauman W.B., Kennedy J.W.. Guidelines for percutaneous transluminal coronary angioplasty. Circulation 1988;88:2987−3003.

19.Tabuchi N., de Haan J., van Oeveren W.. Rapid recovery of platelet function after cardiopulmonary bypass. Blood 1993;82:2930−2931.

20.Tabuchi N., de Haan J., Boonstra P.W., van Oeveren W.. Activation of fibrinolysis in the pericardial cavity during cardiopulmonary bypass. J Thorac Cardiovasc Surg 1993;106:828−833.

21.Boonstra P.W., van Imhoff G.W., Eysman L.. Reduced platelet activation and improved hemostasis after controlled cardiotomy suction during clinical membrane oxygenator perfusions. J Thorac Cardiovasc Surg 1985;89:900−906.

22.Lindblad B., Wakefield T.W., Stanley T.J.. Pharmacological prophylaxis against postoperative graft occlusion after peripheral vascular surgery. Eur J Vasc Endovasc Surg 1995;9:267−271.

23.Oltrona L., Eisemberg P.R., Masala J.M., Sewall D.J., Shelton M.E., Winters K.J.. Association of heparin-resistant thrombin activity with acute ischemic complications of coronary interventions. Circulation 1996;94:2064−2071.

24.Violaris A.G., Melkert R., Hermann J.R., Serruys P.W.. Role of angiographically identifiable thrombus on long-term renarrowing after coronary angioplasty. Circulation 1996;93:889−897.