| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2006;81:770-781
© 2006 The Society of Thoracic Surgeons
a Division of Cardiac Surgery, University of Ottawa Heart Institute, Ottawa Hospital, Ottawa
b Division of Clinical Hematology, Ottawa Hospital, Ottawa
c Department of Medicine, McMaster University, Hamilton, Ontario, Canada
* Address correspondence to Dr Lam, University of Ottawa Heart Institute, 40 Ruskin Street, Suite H3404, Ottawa, Ontario, Canada K1Y 4W7 (Email: bklam{at}ottawaheart.ca).
 |
Abstract |
|---|
 Top Abstract Material and MethodsResultsCommentReferences |
|---|
|
Material and Methods |
|---|
|
TopAbstractMaterial and MethodsResultsCommentReferences |
|---|
Using the search strategy described above, it was evident that very few studies directly compared early postoperative anticoagulation strategies [5–9], and a meta-analysis was not possible because of the lack of data. In order to provide a qualitative comparison of anticoagulation strategies, an additional MEDLINE search was conducted to obtain data regarding the risk of thromboembolism and bleeding associated with each of the different approaches to early anticoagulation. This second search focused on randomized controlled trials, case series, and case-control studies reporting the results of the most common prosthetic mechanical valves currently available. The initial keywords and MESH terms were St. Jude Medical, Medtronic-Hall, Omnicarbon, Carbomedics, ATS valve, On-x valve, mechanical valve, anticoagulation, hemorrhage, bleeding, thromboembolism, and human. Manuscripts reporting the experience of mechanical valves in the aortic and/or mitral position in adult patients were included in this review. Series specifically focused on the outcomes of children or redo operations were excluded, and language restrictions were not applied. A total of 141 citations were obtained, and relevant manuscripts were reviewed by two authors (AK and BKL). The incidences of early postoperative bleeding and thromboembolism were extracted from each manuscript (absolute rates), as well as rates of perioperative mortality caused by hemorrhage or thromboembolism. Studies that did not clearly state the early postoperative anticoagulation strategy were excluded from further analysis. These data were summarized according to the three most common anticoagulation strategies to yield an approximate estimate of the risk of bleeding and embolism associated with each approach (Table 1).
|
|
Results |
|---|
|
TopAbstractMaterial and MethodsResultsCommentReferences |
|---|
Coagulation factors are also directly activated after valve implantation, leading to further clot formation as a result of the inherent thrombogenicity of the prosthetic material (suture material, Dacron sewing ring, struts, and hinge points) and sites of denuded tissue (valve excision site) [12]. Transprosthetic turbulent flow leads to regional increases in shear stress, structurally damaging the endocardium, causing a loss of local resistance to thrombosis. Recirculation areas on the outflow side of the prosthesis create flow stagnation, trapping damaged platelets and activated factors [13]. This provides an ideal milieu for thrombus formation and subsequent embolization. Thrombin can also be formed on platelet membranes after their activation, further promoting the organization and growth of platelet fibrin thrombus [14]. In sum, the particularly high thromboembolic risk that is present shortly after valve implantation involves both coagulation and platelet activation due to turbulent flow across the valve and the thrombogenicity of prosthetic material and denuded tissue.
Early thromboembolic risk after mechanical valve implantation
The insertion of a large artificial device in contact with the bloodstream exposes the patient to a continual risk of valve thrombosis and embolism. This risk is proportional to the surface area of the foreign material, which is in contact with blood, making patients with mitral valve prostheses more prone to thromboembolic complications [11]. In general, a recent mechanical valve implantation is a strong risk factor for thromboembolic complications [15, 16], especially in the first three to six months after surgery [17, 18]. The reasons for this are threefold: first, the pathologic sequelae of the patients' inherent valvular disease (atrial fibrillation, dilated left atrium, and dilated left ventricle) may predispose to areas of stasis and thrombus formation.
Second, the increased thromboembolic risk early after mechanical valve implantation may reflect incomplete endothelial proliferation on the raw intracardiac surfaces, sewing ring, and suture knots in the initial postoperative period [1, 12, 18, 19]. The presence of an endothelial lining on the valve surface effectively prevents thrombus formation, but more than one year may be required after implantation for this in-growth to fully form [20, 21]. However, in the absence of endothelial organization, the development of a mature platelet fibrin coating on the valve surface may also be favorable in terms of nonthrombogenicity and explain the absence of thrombotic deposits in explanted valves in which tissue in-growth is incomplete [20, 21]. Therefore, it is believed that the lack of host endothelial cell in-growth and mature platelet fibrin coating on the valvular surface in the postoperative period may promote early thrombus formation and contribute to the elevated early thromboembolic risk.
Finally, as pointed out by Butchart and colleagues [17, 18], the early thromboembolic risk can also be attributed to inconsistencies of oral anticoagulation management. During the first six months, the thromboembolic risk is up to seven times greater than in the after months and years. After aortic valve replacement, the risk of thromboembolic events falls from 16 per 100 patient years in the early postoperative period to 1.4 per 100 patient years at 5 years. Similarly, after mitral valve replacement, the risk falls from 21 per 100 patient years to 2.5 per 100 patient years [17, 18]. The propensity for early formation of thrombus may be related to difficulties in achieving therapeutic anticoagulation in the initial postoperative period [18, 19]. Spot - international normalized ratio (INR) tests of anticoagulated patients report as many as 50% of INR levels outside the therapeutic range [2], and a recent study of prosthetic valve anticoagulation reported subtherapeutic and variable anticoagulation as a contributing factor in the majority of thromboembolic events [22]. Variable or inadequate anticoagulation, defined as an INR 25% or more below the therapeutic range, has been demonstrated to increase the incidence of thromboembolism 2 to 6 times, especially in the postoperative period [2].
The early thromboembolic risk associated with the placement of a mechanical prosthesis depends on the complex interactions between the recipient and valve, the intrinsic thrombogenic properties of the mechanical valve components, and the diligent management of postoperative anticoagulation.
Anticoagulation therapy
The need for lifelong oral anticoagulation therapy in patients with mechanical prosthetic valves is well-recognized. In patients not receiving long-term anticoagulation therapy, the average rate of major thromboembolism is estimated to be 4 to 8 per 100 patient-years [23, 24]. This risk is reduced to 2.2 per 100 patient-years with antiplatelet therapy, and further reduced to 1 per 100 patient-years with oral anticoagulation (warfarin). Thus, the utilization of postoperative warfarin therapy reduces the incidence of major embolism by approximately 75% and has become the standard of care for all patients with mechanical prostheses [24]. In large studies conducted on anticoagulation therapy, risk factors known to modulate the risk of thromboembolism have included the number of valves replaced, the type of valve implanted, mitral valve replacement (versus aortic valve), atrial fibrillation, the presence of left atrial enlargement or reduced cardiac output, treatment with warfarin or other oral anticoagulants, adequacy of warfarin therapy, and the addition of antiplatelet drugs such as aspirin or dipyridamole [22, 24].
The optimal intensity of oral anticoagulation, defined as the level at which the incidence of both thromboembolic and bleeding complications is lowest, is still a matter of debate. Hirsh and colleagues [25] have demonstrated that increasing the INR, in mechanical valve patients, above the range of 2.5 to 3.0 also increased considerably the rate of bleeding without a reduction in the number of thromboembolic events. The American College of Cardiology/American Heart Association and the American College of Chest Physicians recommended, in their most recent guidelines, that contemporary mechanical valves in the aortic position be anticoagulated with a target INR of 2.0 to 3.0, and mechanical valves in the mitral position be anticoagulated with a target INR of 2.5 to 3.5 [26, 27]. In view of all the clinical evidence presented, the utilization of oral anticoagulation after placement of mechanical valves will remain the standard of care.
Antiplatelet therapy
In recognition of the key role of platelet activation and aggregation in thrombus formation on mechanical valve surfaces, several studies have sought to evaluate the use of antiplatelet agents for thromboembolic prophylaxis, either exclusively, or in conjunction with oral anticoagulation. Ribeiro and colleagues [28] reported on their experience with antiplatelet agents alone in patients with bileaflet aortic and mitral mechanical valves. After 22 months of follow-up, there was no thromboembolic event reported, but the valve thrombosis rate was inappropriately high at 2.1 per 100 patient years (aortic) and 10 per 100 patient years (aortic and mitral). In another report, Czer and colleagues [29] found a significant difference in the rate of valve thrombosis between patients with no anticoagulation (18.9 per 100 patient years), antiplatelet therapy exclusively (3.2 per 100 patient years) and warfarin (1.5 per 100 patient years). Similarly, a recent randomized trial comparing a combination of clopidogrel and aspirin versus warfarin in patients with mechanical aortic valves was terminated prematurely due to the increased rate of early valve thromboses in the antiplatelet arm [30]. Therefore, in the absence of anticoagulation, antiplatelet therapy seems to be insufficient to prevent the development of thrombi.
Because platelet aggregation occurs despite anticoagulation, the co-administration of antiplatelet agents with anticoagulation has been advocated to reduce the frequency of thromboembolic complications. Laffort and colleagues [31] studied the addition of aspirin on postoperative day one with mechanical mitral valve patients anticoagulated with intravenous heparin, followed by oral warfarin. Patients receiving daily aspirin had a reduction in the incidence of nonobstructive periprosthetic valve thrombi from 13% to 5% nine days after surgery, and a reduced incidence of thromboembolic events in the first 5 months, from 25% to 9% (p = 0.004). A number of studies have explored the long-term co-administration of antiplatelet agents to chronic oral anticoagulation after mechanical valve replacement, and a recent meta-analysis by Massel and colleagues [32] demonstrated that the addition of either aspirin or dipyridamole to warfarin reduced the risk of thromboembolic events by 59% (odds ratio [OR] 0.41, 95% confidence intervals [CI] 0.29% to 0.58%) and total mortality by 51% (OR 0.49, 95% CI 0.35% to 0.67%). Although the risk of major bleeding was increased with antiplatelet agents (OR 1.5, 95% CI 1.03% to 2.18%), this risk was lower with contemporary low dose (100 mg daily) aspirin (OR 1.28). A meta-analysis by Cappelleri and colleagues [33] investigating the same issue demonstrated a 67% (OR 0.33, 95% CI 0.16% to 0.69%) reduction in embolism, but an increase in major gastrointestinal hemorrhage by approximately 250% (OR 3.47, 95% CI 1.43% to 8.40%), leading to an estimate that for every 1.6 patients who had their stroke prevented by combination therapy, there was an excess of one major gastrointestinal bleed. These meta-analyses suggest that the benefits derived from the enhanced antithrombotic potential of combined therapy outweigh the risks [32, 33].
The American College of Cardiology/American Heart Association, in their most recent guidelines, recommended that the addition of aspirin (80 to 100 mg/day) to warfarin be strongly considered for all patients with mechanical valves [27]. This is particularly important for patients who have had an embolus while on warfarin therapy, those with known vascular disease, or those known to be particularly susceptible to hypercoagulability [27]. The American College of Chest Physicians, on the other hand, recommend low-dose aspirin only for mechanical valve patients with thromboembolic risk factors, such as atrial fibrillation, or as a strategy to lower the target INR from 3.0 to 2.5 in patients with bileaflet mechanical valves in the mitral position [26]. Despite the evidence and the presence of these guidelines, a survey of North American cardiac surgeons reported that only 21% of surgeons routinely use aspirin in conjunction with oral anticoagulation after mechanical valve replacement [34].
The prothrombotic milieu encountered after mechanical valve implantation is complex and multifactorial, putting patients at risk for thromboembolic complications. These undesirable outcomes are largely prevented by long-term oral anticoagulation with warfarin and antiplatelet therapy. To date, the data in the literature have predominantly focused on the late results of oral anticoagulation strategies, with a paucity of data on the effects of early postoperative anticoagulation on bleeding and thromboembolic risk.
Early Postoperative Anticoagulation
Early anticoagulation and the risk of bleeding
A clear relation exists between the intensity of anticoagulation in patients with mechanical valves and the incidence of thromboembolism, on the one hand, and of bleeding, on the other [35]. Although intravenous heparin is frequently administered after mechanical valve implantation to reduce thromboembolism, heparin has the potential to induce bleeding by inhibiting blood coagulation, impairing platelet function [36], and increasing capillary permeability [35]. Pericardial effusion and cardiac tamponade are serious complications associated with the early postoperative use of heparin and warfarin, and can present days to months after surgery. While pericardial effusion may be present in up to 64% of patients after heart surgery, cardiac tamponade is a less frequent event, occurring in 0.5% to 8.5% of patients [37, 38]. Defined as fluid in the pericardial space compromising cardiac filling, cardiac tamponade is a much more serious complication than pericardial effusion, with a mortality rate up to 16% [37, 39]. Cardiac tamponade occurs more often after valve replacement (11%) than coronary artery bypass graft surgery (CABG) (2%), and is seen almost exclusively in anticoagulated patients [38, 40]. Bleeding is another serious complication associated with anticoagulation after valve implantation. In the immediate postoperative period, bleeding requiring a return to the operating room is more common in valve patients than CABG patients [41]. However, such events usually occur in the early hours after cardiac surgery before the administration of heparin or warfarin and are therefore not usually attributable to anticoagulation. On the other hand, bleeding events, such as gastrointestinal and intracranial hemorrhage, usually present later in the postoperative period and are more likely to be associated with anticoagulation. Occurring in up to 3% of anticoagulated patients in the initial postoperative period, the incidence of all major bleeding events decreases to less than 3% annually in patients on stable anticoagulation [26, 42, 43]. These include any episodes of major internal or external bleeding that cause death, hospitalization, permanent injury, or necessitate transfusion [4]. Although upper gastrointestinal bleeding after cardiac surgery is a rare event (approximately 0.5%), it is associated with a high morbidity and a mortality of up to 30% [44, 45]. Important risk factors associated with gastrointestinal bleeding after cardiac surgery include valve replacement and anticoagulation [44, 46]. In their review of postoperative gastrointestinal bleeding, Heikkinen and colleagues [46] reported that 24% of bleeding events occurred in patients excessively anticoagulated. Bleeding associated with the central nervous system is also a major complication of anticoagulation, representing 20% to 30% of all bleeding events [2] and the most frequent cause (56%) of bleeding-related death [43]. Systemic bleeding complications and cardiac tamponade therefore represent serious complications associated with early postoperative anticoagulation. The ideal early postoperative anticoagulation strategy (with both high antithromboembolic efficacy and low bleeding risk) has yet to be determined in a systematic manner. Most reports on the topic are either retrospective or empirical.
Early postoperative anticoagulation strategies
Recommendations for long-term anticoagulation after mechanical valve replacement have been formulated by several working groups, but there is little objective evidence supporting strategies for the management of anticoagulation in the early postoperative period. The three most common early postoperative anticoagulation strategies that have been reported in the literature include the following: (1) oral warfarin starting on postoperative day one, without the use of therapeutic heparin anticoagulation; (2) intravenous unfractionated heparin beginning in the early hours after surgery, oral warfarin starting on postoperative day one, and the continuation of heparin until a therapeutic INR has been achieved; (3) low molecular weight heparin (LMWH) beginning in the early hours after surgery, oral warfarin starting on postoperative day one, and the continuation of LMWH until a therapeutic INR has been achieved. Unfractionated heparin represents a heterogeneous mixture of glycosaminoglycans that bind to antithrombin, catalyzing the inactivation of thrombin and other clotting factors. Unfractionated heparin has unpredictable pharmacokinetic and pharmacodynamic properties, and therefore requires frequent laboratory assessment of the activated partial thromboplastin time to properly monitor intravenous anticoagulation therapy. The LMWH is derived from unfractionated heparin by chemical or enzymatic depolymerization. Compared with unfractionated heparin, LMWH has reduced antifactor IIa activity relative to antifactor Xa activity, and lacks the nonspecific binding affinities of unfractionated heparin. As a result, LMWH preparations have more predictable pharmacokinetic and pharmacodynamic properties. These properties allow LMWH to be administered subcutaneously once daily without laboratory monitoring [47]. Although each regimen has been promoted, most of the published investigations lack data that would permit a firm conclusion about the optimal anticoagulation regimen.
1. Oral warfarin alone
With the early postoperative risks associated with intravenous heparin, the initiation of mechanical valve anticoagulation with oral warfarin alone has been advocated as the safest regimen [5, 48]. This is based on the increased risk of cardiac tamponade, which is up to five times greater in patients treated with therapeutic intravenous heparin (1.8%) compared with those treated with low dose heparin for deep vein thrombosis (DVT) prophylaxis (0.38%) [38, 49]. The safety of this "oral anticoagulation only" approach was highlighted in a recent randomized trial of different loading doses of warfarin after valve replacement (58% mechanical valves, 42% bioprosthetic valves), without the use of intravenous heparin. Patients randomized to receive 5 mg loading doses achieved the therapeutic range earlier than those randomized to 2.5 mg daily (1.98 vs 2.72 days, p < 0.0001). However, of those receiving 5 mg doses, 49.6% needed the dose withheld for at least one day, and the gap between the target and mean INR on day 5 was greater. There were no bleeding complications in either group, and despite the lack of intravenous heparin, there were no thromboembolic complications. Therefore, a lower loading dose of warfarin reduced the rate of excessive anticoagulation, and the absence of intravenous heparin was not associated with any thromboembolism [5]. Oral anticoagulation alone may be effective at reducing early thromboembolic and anticoagulation-related morbidity. However, oral anticoagulation requires several days of therapy before a therapeutic INR is achieved. Therefore, strategies employing oral anticoagulation alone may not be optimal in preventing the development of thrombus in patients with mechanical valves.
2. Intravenous unfractionated heparin and oral warfarin
Because the risk of thromboembolism is greatest in the early postoperative period [18], and intracardiac thrombus may develop soon after valve implantation, many centers have supported the utilization of intravenous unfractionated heparin early after surgery (within 24 hours), with oral anticoagulation starting on the first postoperative day [1, 19, 27]. However, even with the administration of intravenous heparin three hours after surgery, thrombus formation can be observed by transesophageal echocardiography in up to 18% of patients after mechanical valve implantation [50].
In a swine model evaluating anticoagulation strategies after mechanical mitral valve replacement, 69% (9/13) of pigs died from hemopericardium within 30 days when treated with concurrent intravenous heparin (125–150 units/kg q12 hours) and oral warfarin. In this study, heparin was discontinued when a target INR of greater than 2.0 was achieved [51]. Thulin and Olin [52] reported their clinical experience with anticoagulation after mechanical valve replacement in 510 patients. The morning after surgery, 10 mg of warfarin sodium was administered intravenously, followed by oral warfarin on ensuing days. Anticoagulation was supplemented with intravenous heparin until a therapeutic INR was achieved. A thromboembolism incidence of 0.2% within the first three months was reported, with only 1.3% of patients developing hemopericardium and cardiac tamponade. The authors concluded by stating that the quality of the anticoagulation approach was very important in the prevention of thromboembolic and bleeding complications.
The most recent American College of Chest Physicians Anticoagulation Consensus Guidelines encourage the use of intravenous unfractionated heparin until a therapeutic INR is achieved for mechanical valves (grade 2C recommendation), despite the lack of strong evidence [26]. This is based on the premise that starting warfarin therapy without simultaneous therapeutic heparin may enhance thrombosis in situations of high thrombotic risk. In order to avoid hypercoagulable conditions caused by varying degrees of decay in coagulation factors, heparin is continued until the INR is stable in the therapeutic range [53]. Some centers favor a more aggressive anticoagulation regimen, with the continuation of heparin therapy for two additional days after the INR is therapeutic to prevent potential thromboembolic events [53, 54]. However, little controlled data are available to ascertain the optimal use and the adequate degree of anticoagulation with intravenous heparin [6]. This lack of information and coordination has led to the establishment of recommendations that are based primarily on empirical data [26].
3. Low molecular weight heparin and oral warfarin
The LMWH has many potential advantages that may be relevant for patients with mechanical heart valves. They have a better safety profile with less thrombocytopenias [55], less bleeding, a more predictable and rapidly reached anticoagulant effect [56], the possibility of self-administration without laboratory monitoring, and shorter hospital stays and lower costs associated with outpatient administration [57]. However, unlike unfractionated heparin, LMWH is not fully reversible by protamine, and may carry an increased bleeding risk in the immediate postoperative period [47]. Nevertheless, the pharmacokinetic and biologic advantages of LMWH may be more relevant in the postoperative period of heart valve replacement because of the severe inflammation and platelet and coagulation disorders related to cardiopulmonary bypass. This may make adequate anticoagulation with unfractionated heparin more difficult [6]. Therefore, many centers have adopted the use of LMWH within 24 hours after surgery (without the use of unfractionated heparin), and oral anticoagulation starting on the first postoperative day [6, 8].
Several clinical trials have prospectively evaluated the potential use of LMWH in patients after valve replacement surgery. Bridging therapy with enoxaparin was evaluated postoperatively in a case series of 37 patients who had undergone mechanical or bioprosthetic heart valve replacements and were stable for discharge prior to achieving a therapeutic INR. Enoxaparin at 1 mg/kg every 12 hours was administered until two consecutive therapeutic INR values were reported. During the three-month study period, bleeding related to anticoagulation was reported in three patients (2 major, 1 minor). One thromboembolic event was reported the day after discontinuation of enoxaparin, and one death due to arrhythmia (possibly related to anticoagulation) was reported two weeks after discontinuation of enoxaparin. It was determined that home LMWH therapy reduced the number of hospital days by a mean 4.6 days per patient. This translated to health system savings of $168,300, based on the average cost of $1,100/day for a surgical bed [58].
Montalescot and colleagues [6] performed a comparative case-control study of subcutaneous unfractionated heparin (500 IU/kg/day, divided over three times a day dosing) versus subcutaneous enoxaparin (100 anti-Xa IU/kg every 12 hours) after heart valve replacement. Two major bleeding events occurred in each group, and one stroke occurred in the unfractionated heparin group. Another case-control study of perioperative anticoagulation after mechanical valve implantation compared 29 patients treated with enoxaparin (1 mg/kg every 12 hours) with 34 control patients treated with intravenous unfractionated heparin. There were no thromboembolic events in the LMWH group compared with two (6%) in the unfractionated heparin group (p = 0.50), and there were three (10%) bleeding episodes (pleural effusions) in the LMWH group compared with three (9%) bleeding events in the unfractionated heparin group (p = 1.0). The use of LMWH reduced the length of hospital stay after surgery (7.8 ± 4.0 to 18.1 ± 11.6, p < 0.0001) and reduced the cost per patient by an average of $6,864 [8]. Similarly, an open-label randomized trial comparing the LMWH nadroparin to intravenous unfractionated heparin after mechanical valve implantation demonstrated that patients treated with nadroparin had a shorter hospital stay, less coagulation tests, and easier dosing administration. No difference in major hemorrhagic events was reported [9]. Therefore, anticoagulation with LMWH after mechanical valve replacement appears to be a safe, feasible, and effective alternative to intravenous unfractionated heparin, and provides more predictable and rapid anticoagulation compared with unfractionated heparin. The latter is an important point when considering cost-containment issues in government funded healthcare systems.
Comparison of anticoagulation strategies
Few studies in the literature have directly compared the risks of bleeding and thromboembolism associated with different anticoagulation strategies after mechanical valve replacement. With limited data available, a systematic review was conducted, searching for controlled trials and case series involving the most common mechanical prosthetic valves and different approaches to early postoperative anticoagulation. Summarized in Table 1, a qualitative comparison of the three most common early postoperative anticoagulation strategies was performed with 20 studies describing the use of oral warfarin with subcutaneous heparin (DVT prophylactic dose) [5, 6, 59–76], 7 studies employing intravenous unfractionated heparin and oral warfarin [7, 8, 31, 52, 77–79], and 3 reports of the use of therapeutic LMWH with oral warfarin [6, 8, 58]. Through this analysis, an estimated incidence of early postoperative thromboembolism of 0.9%, 1.1%, and 0.6%, respectively, was determined for each of the three anticoagulation strategies. Furthermore, an estimated incidence of early postoperative hemorrhage of 3.3%, 7.2%, and 4.8%, respectively, was noted. The mortality rate secondary to thromboembolism or bleeding was similar between oral warfarin alone (0.3% and 0.8%) compared with intravenous unfractionated heparin with oral warfarin (0.2% and 0.8%). These data suggest that the use of early postoperative intravenous unfractionated heparin, compared with the use of oral warfarin alone, yields a substantially higher risk of bleeding without lowering the incidence of thromboembolism. This analysis, although simple in design, provides the most thorough up-to-date review of the literature on early postoperative anticoagulation. Nevertheless, these data cannot substitute for a well-designed randomized controlled trial that is clearly warranted at this time.
Through this analysis, the strategy of subcutaneous unfractionated heparin (DVT prophylaxis) with oral coumadin after mechanical valve implantation yielded a combined incidence of early postoperative thromboembolism and bleeding of 4.2%. This compared with the 8.3% combined rate for a strategy of intravenous unfractionated heparin with oral coumadin after surgery. Setting an 
level of 0.05 and ß of 0.10, approximately 300 patients per group, or 600 patients total would be required to adequately power a trial to detect a difference in these two approaches during the first month after surgery. With most surgical centers averaging 100 to 150 mechanical valve implantations per year, a multicenter trial would be best suited to answer the question at hand.
|
Comment |
|---|
|
TopAbstractMaterial and MethodsResultsCommentReferences |
|---|
In the present setting, a comprehensive approach to data collection related to anticoagulation would improve clinical decision-making. While guidelines exist regarding the reporting of prosthesis-related complications during long-term follow-up after valve surgery [4], these recommendations do not address the early postoperative complications that may be related to anticoagulation strategies, such as the development of a nonobstructive prosthesis thrombus, pericardial effusion, or cardiac tamponade. In order to facilitate this process, we propose the following staging system for early postoperative thromboembolic and bleeding events, ranked in increasing order of clinical severity and patient impact:
Thromboembolic Level (TEL)
Bleeding Level (BL)
The Need for a Trial
The initiation of an efficacious early anticoagulation protocol is important because of its potential impact on the rate of early thromboembolic complications associated with the prothrombotic state after mechanical valve implantation. The challenge in establishing a model will be to balance the risks of under-anticoagulation (thromboembolic events) against those of over-anticoagulation (bleeding complications) in an early postoperative setting when the risk of bleeding may be significant. Although a number of strategies exist, a randomized controlled trial examining the issue of early postoperative anticoagulation after mechanical valve replacement is lacking, and the paucity of knowledge in the matter is a serious impediment to adequate early care of patients with mechanical prostheses. Such a study would compare two or more early postoperative anticoagulation strategies, in addition to routine low dose antiplatelet therapy, to document the incidence of bleeding and thromboembolic complications during the first month after mechanical valve implantation. Further, it would incorporate the proposed nomenclature described above, as well as routine postoperative echocardiography to determine the incidence and clinical impact of prosthetic valve thrombi and pericardial effusions in these anticoagulated patients. Clearly, a need exists for a multicenter randomized study comparing anticoagulation regimens after mechanical valve replacement to determine the safest and most efficacious approach.
|
References |
|---|
|
TopAbstractMaterial and MethodsResultsCommentReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  N Allou, P Piednoir, C Berroeta, S Provenchere, H Ibrahim, G Baron, P Montravers, B Iung, I Philip, and N Ajzenberg Incidence and risk factors of early thromboembolic events after mechanical heart valve replacement in patients treated with intravenous unfractionated heparin Heart, October 15, 2009; 95(20): 1694 - 1700. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Charokopos, P. Antonitsis, P. Artemiou, E. Rouska, C. Foroulis, and C. Papakonstantinou Acute mechanical prosthetic valve thrombosis after initiating oral anticoagulation therapy. Is bridging anticoagulation with heparin required? Interactive CardioVascular and Thoracic Surgery, October 1, 2009; 9(4): 685 - 687. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Edwin, M. M. Tettey, E. Aniteye, L. Sereboe, M. Tamatey, K. Entsua-Mensah, D. Kotei, and K. Frimpong-Boateng Freedom from thromboembolism despite prolonged inadequate anticoagulation BMJ Case Reports, September 15, 2009; 2009(sep15_1): bcr0720092066 - bcr0720092066. [Abstract] [Full Text]
|
|
|
|
 |
|
 A. Colli, J.-P. Verhoye, R. Heijmen, J. T. Strauch, J. A.J. Hyde, D. Pagano, M. Antunes, H. Koertke, S. K. Ohri, D. H. L. Bail, et al. Corrigendum to "Antithrombotic therapy after bioprosthetic aortic valve replacement: ACTION Registry survey results" [Eur. J. Cardiothorac. Surg. 33 (2008) 531-536] Eur. J. Cardiothorac. Surg., August 1, 2009; 36(2): 429 - 429. [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Spyropoulos Outpatient-Based Primary and Secondary Thromboprophylaxis With Low-Molecular-Weight Heparin Clinical and Applied Thrombosis/Hemostasis, January 1, 2008; 14(1): 63 - 74. [Abstract] [PDF]
|
|
|
|
 |
|
 S. Talwar, A. Mathur, S. K. Choudhary, R. Singh, and A. S. Kumar Aortic Valve Replacement With Mitral Valve Repair Compared With Combined Aortic and Mitral Valve Replacement Ann. Thorac. Surg., October 1, 2007; 84(4): 1219 - 1225. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. U. Jones, J. B. Muhlestein, K. W. Jones, D. G. Renlund, T. L. Bair, T. J. Bunch, B. D. Horne, D. L. Lappe, J. L. Anderson, and D. B. Doty Reply. Ann. Thorac. Surg., October 1, 2006; 82(4): 1574 - 1575. [Full Text] [PDF]
|
|
|
|
|
|
V. Close, M. Purohit, M. Tanos, and S. Hunter Should patients post-cardiac surgery be given low molecular weight heparin for deep vein thrombosis prophylaxis? Interactive CardioVascular and Thoracic Surgery, October 1, 2006; 5(5): 624 - 629. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
