The postoperative phase of Mechanical Valve Heart Replacement (MVHR) is associated with maximal thrombo- embolic (TE) and bleeding risks and optimal early anticoagulation strategy remains controversial despite published guidelines [1] -[4] . Although the successfully use of Low Molecular Weight Heparin (LMWH) in the treatment of many cardiovascular disease: pulmonary embolism, acute coronary syndrome, atrial fibrillation and acute myocardial infarction, the LMWH still not yet routinely used for the early anticoagulation after MVHR. The three most commonly prescribed LMWH are enoxaparin (Lovenox), dalteparin (Fragmin) and tinzaparin (Innohep) of which enoxaparin is most commonly used in clinical practice. Enoxaparin sodium is obtained by alkaline depolymerization of heparin benzyl ester derived from porcine intestinal mucosa. The drug substance is the sodium salt. The average molecular weight is about 4500 daltons.

The aim of the study was to evaluate efficacy and safety of LMWH as bridge between immediate postoperative Unfractioned Heparin (UH) and effective Vitamin K Antagonist (VKA) therapy, using the rate of TE and bleeding events as primary endpoint during the early postoperative phase after Mechanical Aortic Valve Replacement (MAVR).

Methodology: It’s a prospective study approved by our institution’s Ethics Committee and conducted over 6 months in the department of cardiac surgery. It concerned 40 consecutive patients who underwent MAVR and treated by enoxaparin (Lovenox, Rhone-Poulenc-Rorer, France). All patients who underwent a MAVR were included in the study and patients with following characteristics were excluded: previous or simultaneous mitral and/or tricuspid valve surgery, renal failure (creatinin clearance < 30 ml/mn) and previous heparin-induced thrombocytopenia. Results are expressed as mean ± SD.

Study protocol: In postoperative period, intravenous UH was started on the 6th postoperative hour in the surgical intensive care unit. After transfer to the ward, UH was stopped and replaced by enoxaparin 2000, 4000 or 6000IU/12 hour respectively if the patient weight is less than 51 Kg, between 51 and 80 Kg or more than 80 Kg. The anti Xa activity was measured for all patients 4 h after the 1st and the 3rd injection of enoxaparin for a therapeutic range between 0.10 and 0.5 IU/ml. Fluindion (Previscan, Merk serono, France) was started after removal of chest drains and enoxaparin was continued until the international ratio (INR) is more than 2 (INR target = 2 - 3). Platelet aggregations inhibitors were not systematically used, but reserved to patients who had previous or concomitant coronary artery bypass graft (CABG).

Patient population: There were 40 patients (32 male and 8 female) with a mean age of 53.83 ± 16.93 years (19 - 75 years). The patients had at mean 1.29 ± 1.10 (0 - 5) cardiovascular risk factors. The average weight of patients was 74.71 ± 17.29 Kg (47 - 115 Kg). 6 patients (15%) was receiving in pre-operatory VKA therapy for atrial fibrillation in 3 cases (7.5%), previous MAVR in 2 cases (5%) and pulmonary embolism in one case (2.5%). One patient was receiving aspirin (2.5%) for previous CABG. Transthoracic echocardiography showed that 6 patients (15%) had a left atrium diameter more than 45mm and 4 patients (10%) had a left ventricular ejection fraction less than 45% (Table 1).

Statistics: Statistical analysis was performed using the statistical software package of social science (SPSS 11.5, Chicago, Illinois, USA). All data were expressed as mean ± standard deviation, median or prevalence as appropriate.

Operative details: All operations were performed by sternotomy and all patients received 3 mg/kg of heparin to achieve an activated clotting time >400 seconds. Cardio-pulmonary bypass was established by aortic and atrio- caval cannulation, the aorta clamped with cold intermittent antegrade cardioplegia and hypothermia at 32˚C (Table 2). 23 patients (57.5%) underwent simple aortic valve replacement using SJM, ATS, and Sorin prosthesis respectively in 17 (42.5%), 4 (10%), and 2 patients (5%). 17 patients (42.5%) underwent Bentall procedure using ATS valve Dacron graft for aortic root aneurysm and for type I aortic dissection respectively in 12 (30%) and 5 patients (12.5%). As associated procedures, one patient (2.5%) had epicor for surgical ablation of atrial fibrillation and one patient had resection of subaortic stenosis (Table 2). The return of patients to ward was on average of 37.2 hour (1 - 10 day) and 90% of patients returned between D1 and D2.

Anticoagulation protocol: VKA therapy (fluindion) in combination with heparin was started after removal of

MAVR: mechanical aortic valve replacement; VKA: Vitamin K Antagonist; LVEF: Left ventricular ejection fraction; LA: left atrium.

MAVR: mechanical aortic valve replacement.

chest drains and enoxaparin was continued until the target INR was superior than 2. The therapeutic range of INR was between 2 and 3. The bridging therapy with both VKA and enoxaparin lasted on average of 4.40 ± 1.69 day (3 - 10 day) and in 67.5% of cases the bridge lasted at least at the 4th day. The average of anti Xa activity was 0.32 ± 0.08 IU/ml (0.12 - 0.5 IU/ml) and all patients were in the therapeutic range (0.10 - 0.5 IU/ml) at the 1st day after the 1^st injection and the maintained in the therapeutic range after the 3rd injection. 6 patients (15%) received Platelet aggregations inhibitors for CABG.

Postoperative characteristics: As per the guidelines for reporting morbidity and mortality after valve heart operations, thrombo-embolism was defined as an embolic event that occurred in the absence of infection, after the immediate operative period, and a bleeding event was defined as an episode of major internal or external bleeding that caused death, permanent injury, or required transfusion [5] [6] .

There was no in-hospital mortality and no TE event. However, we report two major bleeding events (5%): 2 cases of reoperation for tamponade with a good outcome. The 1st patient was operated at 12th day for redo aortic stenosis with change of aortic prosthesis and CABG, he received aspirin, enoxaparin 4000 UI/12 h, fluindion, and a nonsteroidal anti-inflammatory drug for chest pain; and the 2nd patient was operated for aortic stenosis by MAVR and he received enoxaparin 4000 UI/12 h and aspirin for previous CABG. Transthoracic echocardiography control showed also a case of non-compressive and resolving pericardial effusion (2.5%) (Table 3).

Thrombo-embolic events and bleeding still account for 75% of all complications of heart valve surgery. They usually occur during the first 6 months after surgery and can affect mortality and quality of life. Several postoperative ultrasound studies demonstrated the presence of non-obstructive thromboses in about 10% of cases [7] -[9] . The propensity to early thrombus formation can be attributed to difficulties of ensuring during the initial postoperative period effective anticoagulant treatment [10] [11] . Early anticoagulation after cardiac surgery remains a controversial issue, especially in terms of intensity and timing of anticoagulation.

MAVR remains the gold standard of the surgical treatment of aortic stenosis or regurgitation especially for young patient with excellent long term results thanks to technical improvement of new generation of mechanical valve prosthesis allowing better blood flow, less stasis and made of less thrombogenic material. Despite this progress, thromboembolism and bleeding remains the most frequent complication.

LMWH has several potential advantages: better safety profile with less frequent thrombocytopenia and bleed- ing and less osteoporosis during chronic therapy, more predictable and more rapid anticoagulant effect, and the possibility of self-administration without daily laboratory monitoring. LMWH has been used successfully in the treatment and prevention of deep vein thrombosis and pulmonary embolism; and in the treatment of unstable angina [12] and acute coronary syndrome and, more recently, acute myocardial infarction [13] [14] and atrial

fibrillation [15] . The benefit-risk balance of LMWH is at least as good as that of UH. Lucas et al. concluded that enoxaparin as bridge to oral anticoagulation in patients undergoing electrical cardioversion for reduction of atrial fibrillation was safer and more effective than UH and the therapeutic zone was reached more rapidly. After successful cardioversion, no cases of intracardiac thrombosis or TE events were reported in the enoxaparin group [16] . Similar findings were reported by C. Schmidt-Lucke et al., indicating that the therapeutic range of anti-Xa activity ensuring effective anticoagulation was reached during the first 24 hours after the 1st injection in 90% of patients [17] .

The bioavailability and predictability of UH anticoagulation are poor. In the ESSENCE study [18] and TIMI 9B study [19] , in about 50% of patients receiving intravenous UH, effective anticoagulation was achieved on day 3. Hull et al. [20] , using subcutaneous UH, observed that the target Activated Partial Thromboplastin Time (APTT) was achieved on day 2 in only 37% of patients with recent deep vein thrombosis, despite an additional intravenous bolus of 5000 IU of heparin. UH has not been shown to be superior to other bridging anticoagulant regimens in patients who have recently undergone MHVR [21] .

In patients with a MHVR and requiring suspension of VKA for non-cardiac surgery, Kovacs and Douketis in two separate prospective studies including a total of 327 patients with MHVR treated by LMWH [22] [23] , concluded that a standardized periprocedural LMWH treatment is associated with a low risk of TE events or major bleeding complications.

During the early period after MHVR; when patients normally receive both heparin and VKA until a target INR was achieved, only 2 retrospective studies with a total of 131 patients receiving LMWH have been published [24] [25] and they suggested that the therapeutic zone was achieved more rapidly and more reliably with LMWH than with UH [24] and that LMWH was at least as effective and as safe as UH [24] [25] .

Montalescot [24] compared the therapeutic efficacy of LMWH and UH during the immediate postoperative period in which patients who underwent single or double MHVR received subcutaneous UH in the 1st study phase (n = 106) and LMWH in the 2nd phase (n = 102) (106 aortic valve replacement, 102 mitral valve replacement). The therapeutic zone was achieved very rapidly on the 2nd day for 87% of patients with LMWH (anti-Xa factor activity within the therapeutic range 0.5 to 1 IU/mL) versus 9% of patients with UH (APTT within the therapeutic range 1.5 to 2.5 times control) and was maintained 2 weeks for 81% of patients with LMWH versus only 21% of patients with UH. This laboratory efficacy was accompanied by a similar safety and clinical efficacy with one bleeding complication in each group and one stroke in UH group. However, follow-up was short (14 days), and the number of Mitral valve replacement was low (n = 10) [24] . Fanikos [25] compared 29 patients receiving LMWH with 34 control patients receiving UH. After 90 days follow-up, the number of events (deaths, TE and bleeding events) was slightly but not significantly lower in the LMWH group (4 vs 9 cases) and the hospital stay was shorter and the postoperative cost was lower in LMWH group [25] .

In our study; all patients were in the therapeutic range of anti Xa activity at the first day and the two major bleeding events was occurred in patients having previous cardiac surgery and receiving in addition at enoxaparin and VKA; aspirin and Non-Steroidal Anti-inflammatory Drug. In addition there was no TE event and no prosthesis thrombosis at the transthoracic echocardiography control.

Study Limitations: Our study is small and did not include a control group, we cannot comment on efficacy or safety of early LMWH after MAVR. However, as our knowledge, no randomized study evaluating LMWH early after MAVR have been conducted more than 15 years after LMWH first became available.

There are no personal conflicts of interest directly or indirectly with any commercial product or associations.

LMWH as bridging therapy between immediate postoperative UH and complete efficacy of oral anticoagulant appears to be an attractive option for the prevention of TE events in patients recently undergoing MAVR. These results need to be confirmed by a randomized study comparing LMWH and UH in this indication. This study must be completed by complementary neurological investigations to detect possible subclinical TE complications.

All the authors had participated to the patient’s management and the data collect; and had corrected the manuscript. All authors declare that there is no conflict of interest in this study.