Valve disease is treated increasingly by interventional catheter techniques or surgical valve repair. The recommendations given for treatment of native valve disease would seem most applicable in such patients; that is, antithrombotic treatment depends on associated stroke risk factors. Data on thrombotic complications after the Ross procedure are limited, but the risk seems low, and the recommendations for a biologic valve seem reasonable.39
Following percutaneous edge-to-edge repair for mitral regurgitation using the Mitra-Clip device, aspirin (325 mg/d) is recommended for 6 months along with clopidogrel (75 mg/d) for 30 days after the procedure.40 After percutaneous balloon mitral/aortic valvuloplasty, no specific antithrombotic therapy is recommended unless it is needed for another indication (eg, left atrial thrombus, atrial fibrillation).
The risk of bleeding during surgery for a patient while on antithrombotic therapy has to be weighed against the thromboembolic risk caused by stopping the therapy. This should take into account the type of surgical procedure; risk factors for thromboembolism; and the type, location, and number of mechanical valves.2 In a worst-case scenario (mechanical valve with previous thromboembolism), the risk of embolus is 0.08-0.16% if VKA therapy is stopped for 3 days,41 Although antithrombotic therapy must be individualized, some generalizations apply. For procedures where bleeding is unlikely or would be inconsequential if it occurred, antithrombotic therapy should not be stopped. This can apply to surgery on the skin, dental prophylaxis, or simple treatment for dental caries.2 Eye surgery, in particular surgery for cataracts or glaucoma, is usually associated with very little bleeding; many ophthalmologists do not alter antithrombotic therapy.
For patients with bileaflet aortic valve and no risk factors for thromboembolism, the risk of stopping VKA for a few days is small and these patients do not need bridging therapy. VKA is stopped 2-4 days before surgery, to allow the INR to fall < 1.5 when surgery can be safely performed. Once bleeding risk allows (usually 12-24 hours after surgery) VKA is restarted.2
For patients with mechanical MVR, older-generation AVR (ball-in-cage valve, eg Starr Edwards valve), and mechanical AVR with other risk factors for thromboembolism (AF, previous thromboembolism, hypercoagulable condition, LV systolic dysfunction (LV ejection fraction < 30%), or > 1 mechanical valve), bridging anticoagulation with intravenous UFH or subcutaneous LMWH is recommended to reduce the risk of adverse effects.2 VKA is stopped 2-4 days before surgery. When the INR falls to < 2.0, intravenous heparin infusion or weight-adjusted LMWH (twice daily) is initiated. This is stopped 4-6 hours (for intravenous UFH) or 12 hours (for LMWH) before surgery. Once bleeding risk allows (usually 12-24 hours after surgery), a VKA is restarted. Bridging heparin after surgery needs to be individualized depending on the bleeding risk and risk of thrombosis. The use of LMWH allows outpatient bridging but has the disadvantage of partial anticoagulation reversal with protamine, a long half-life, and is contraindicated in those with severe renal failure.
If a patient is taking aspirin, it should be discontinued 7 days before the procedure and restarted as soon as it is considered safe by the surgeon. Clopidogrel and ticagrelor should be stopped 5 days before the procedure, whereas prasugrel should be stopped 7 days in advance.
Cardiac Catheterization and Angiography
Urgent cardiac catheterization procedures can be performed in patients taking warfarin. For nonurgent cases, warfarin should be stopped 72 hours before cardiac catheterization so that the INR is < 1.8 (for femoral access) or < 2.2 (for radial access) at the time of the procedure,42 and restarted the same day after the procedure for patients who are at low risk for thromboembolism. For patients at high thromboembolic risk, heparin should be started 48 hours before the procedure and continued until warfarin is resumed and the desired INR is achieved. Of note, the European guidelines advocate avoidance of interruption of VKAs (but not non-VKA oral anticoagulants) in elective percutaneous coronary intervention (PCI) (with radial access as the preferred option) in order to avoid the bleeding or ischemic complications associated with bridging therapy.43 A post hoc analysis of the WOEST trial44 found that uninterrupted oral anticoagulation (periprocedural INR 2.5) was not associated with excess bleeding complications and tended to have reduced major adverse cardiac events as compared to bridging therapy. In patients who are scheduled to undergo transseptal puncture, all antithrombotic therapy should be discontinued, and the INR should be < 1.2 prior to the procedure.41
Coronary Artery Stents in Patients With Valve Disease
Dual antiplatelet therapy (DAPT) with aspirin and a thienopyridine (eg, clopidogrel, prasugrel) is superior to oral anticoagulation for preventing stent thrombosis in patients following PCI with stent implantation.45 On the other hand, oral anticoagulation is superior to DAPT for the prevention of thromboembolic events in patients with AF and mechanical valves.46,47 Hence, patients with a compelling indication for anticoagulation (eg, mechanical valves, MS with AF, CHA2DS2-VASc score ≥ 2), undergoing PCI require triple therapy with aspirin, clopidogrel, and an oral anticoagulant (VKA in case of mechanical valve and MS with AF).43,48 Triple antithrombotic therapy with aspirin, clopidogrel, and warfarin increases the bleeding risk with increasing duration.49 In order to balance the ischemic and bleeding complications, certain recommendations with general consensus have been made43,48: (1) the duration of triple therapy must be kept as short as possible; (2) low-dose aspirin (< 100 mg) should be used; (3) the use of prasugrel or ticagrelor as part of triple therapy should be avoided; (4) prophylactic proton pump inhibitor use is recommended to reduce gastrointestinal bleeding; (5) a lower target INR of 2-2.5 is recommended; and (6) for patients with a low risk of bleeding (HAS BLED score ≤ 2), the newer generation drug-eluting stents (DES) are recommended in preference to bare metal stent (BMS), given that the former are associated a lower risk of restenosis and stent thrombosis. For patients with a high risk of bleeding (HAS BLED score > 2), the choice of stent between a BMS and DES should be individualized. A BMS is preferred if the risk of bleeding is very high and the risk of restenosis low.
The optimal duration of triple therapy depends on the clinical setting (stable coronary artery disease or acute coronary syndrome), the risk of stent thrombosis, and the risk of bleeding.38,43 For stable coronary artery disease, patients undergoing PCI with a high bleeding risk (HAS BLED score > 2), triple therapy (oral anticoagulants, aspirin 75-100 mg daily, and clopidogrel 75 mg daily) is recommended for 4 weeks followed by dual therapy with an oral anticoagulant and clopidogrel 75 mg/d (or alternatively, aspirin 75-100 mg/d) up to 12 months.50 For patients at low bleeding risk (HAS BLED score ≤ 2), the minimum duration of triple therapy is 4 weeks (not to exceed 6 months depending on risk of stent thrombosis with 6 months recommended for acute coronary syndrome)51 followed by dual therapy with oral anticoagulant and clopidogrel 75 mg/d (or alternatively, aspirin 75-100 mg/d) up to 12 months. The WOEST trial50 compared dual therapy (VKA plus clopidogrel alone) to triple therapy (VKA plus aspirin and clopidogrel) in 573 patients taking long-term oral anticoagulation who received a coronary stent. The primary outcome of any bleeding was significantly reduced in the dual therapy arm without any increase in the rate of thrombotic events (stent thrombosis). Further, there was a significant reduction in the all-cause mortality (secondary end point) at 12 months in the dual therapy arm. On the other hand, the ISAR TRIPLE trial,51 which randomized 614 patients with an indication for oral anticoagulation to either 6 weeks or 6 months of triple therapy with aspirin, clopidogrel, and warfarin, followed by double therapy with aspirin and warfarin thereafter, found no difference in the primary end points of death, myocardial infarction, definite stent thrombosis, stroke, or thrombolysis in myocardial infarction (TIMI) major bleeding at 9 months. There was no difference in the individual end points either. However, Bleeding Academic Research Consortium (BARC) type ≥ 2 bleeding was reduced in the dual therapy arm.51
After 12 months, stable coronary artery disease patients may be managed with oral anticoagulants alone. But in the case of mechanical valves, the addition of low-dose aspirin to anticoagulant therapy has been shown to be beneficial in the long term and is recommended. In select cases such as left main stenting, proximal bifurcation stenting, or proximal left anterior descending coronary artery stenting, clopidogrel 75 mg/d may be preferred over low-dose aspirin for the long-term treatment.
Pregnancy with a mechanical heart valve constitutes the World Health Organization (WHO) risk class III (significantly increased risk of maternal mortality or severe morbidity).52 Pregnancy is associated with alterations in hemostasis and coagulability that increase the risk of thromboembolic events among women with mechanical PHVs.53 Estimates of maternal mortality range from 1%53 in developed countries to 3% in emerging countries,54 with most deaths attributable to thrombotic complications. The possibility of fetal toxicity and changes in dose requirements of antithrombotic agents during pregnancy further complicate management decisions in this high-risk population. The primary objective should be to prevent the occurrence of valve thrombosis and its lethal consequences for both the mother and the fetus.52 There is no ideal anticoagulation regimen. Anticoagulation options are (1) use of VKA (warfarin) throughout pregnancy with switch to UFH at 36 weeks; (2) use of UFH/LMWH restricted to the first 6-12 weeks, followed by warfarin up to 36 weeks with a switch to heparin; and (3) use of UFH/LMWH throughout pregnancy.2,52,55,56 The last regimen is usually not favored, as continuous use of UFH/LMWH throughout pregnancy markedly increases the risk of prosthetic heart valve thrombosis (PHVT).
Warfarin crosses the placenta and is associated with teratogenicity, increased incidence of spontaneous abortion, prematurity, stillbirth, and fetal bleeding (including cerebral hemorrhage).57 Embryopathy resulting from in utero exposure to warfarin appears to be greatest during the first trimester, particularly between weeks 6 and 12 of gestation. The risk of warfarin-associated embryopathy appears to be dose-related. The incidence is < 3% with a daily dose of ≤ 5 mg but increases to > 8% with a warfarin dose of > 5 mg/d.58,59 Some centers have used acenocoumarol (a coumarin derivative) with a very low incidence of warfarin-related embryopathy.60 With respect to valve thrombosis and thromboembolism, warfarin given throughout pregnancy is the safest option.61 UFH does not cross the placenta and is considered a safer alternative to warfarin in terms of fetal adverse effects (no teratogenicity or fetal bleeding), but it may be less efficacious in reducing maternal thromboembolic risk compared with warfarin.61 Heparin is preferably given as an intravenous infusion rather than subcutaneous injections. Prolonged exposure to heparin is associated with thrombocytopenia and osteoporosis. Given the short half-life of heparin and the change in its volume of distribution during pregnancy, frequent dose adjustments and higher doses may be needed. There is also the added risk of intravenous line infection, which may rarely lead to prosthetic valve infective endocarditis.
In the absence of controlled clinical trials, current recommendations are based on limited, observational data. In a systematic review comprising 24 studies of over 900 pregnant women with mechanical PHVs, Chan and colleagues61 evaluated maternal and fetal outcomes according to the type of antithrombotic regimen used during pregnancy: oral anticoagulants alone, oral anticoagulants substituted with heparin during the first trimester, heparin throughout pregnancy, or antiplatelet agents alone. Results from this analysis (Table 53–3) suggested that warfarin is more efficacious than UFH for thromboprophylaxis of pregnant women with mechanical PHVs but is associated with an increased risk of embryopathy. Warfarin therapy throughout pregnancy is recommended for patients with a daily dose requirement of < 5 mg/d and for those who at are at very high risk of thromboembolism (older generation mitral valve prosthesis or those with previous history of thromboembolism).2
TABLE 53–3.Frequency of Fetal and Maternal Complications by Anticoagulation Regimen During Pregnancy ||Download (.pdf) TABLE 53–3. Frequency of Fetal and Maternal Complications by Anticoagulation Regimen During Pregnancy
|Anticoagulation Regimen ||Congenital Fetal Anomalies ||Thromboembolic Complications |
|Regimen 1 (oral anticoagulants alone)a ||35/549 (6.4%) ||31/788 (3.9%) |
|Regimen 2 (heparin during first trimester/oral anticoagulants)a ||6/174 (3.4%) ||21/229 (9.2%) |
|Regimen 3 (heparin alone)a ||0/17 (0.0%) ||7/21 (33.3%) |
|Regimen 4 (no anticoagulation)a ||3/92 (3.3%) ||26/107 (24.3%) |
|LMWH throughout pregnancy66 ||0/92 ||9/92 (9.8%) |
LMWHs offer several advantages over UFH with longer half-life, greater bioavailability, and more predictable anticoagulant response. During pregnancy, LWMHs do not cross the placenta and have a lower incidence of osteopenia and thrombocytopenia compared with UFH.62 Previously, several authors had reported fatal valve thrombosis in pregnant women receiving weight-adjusted LMWH alone.63,64 Of late, there have been a number of reports of successful pregnancies using dose-adjusted LMWH (twice daily) according to anti-Xa levels (see Table 53–3).65,66 Using this approach, the rates of valve thrombosis and spontaneous abortions are lower than with UFH.2 The dosage requirement of LMWH may increase by as much as 50% over the course of pregnancy. Hence frequent dose adjustments are needed. There remain some unresolved questions—the optimal anti-Xa levels, the importance of peak vs trough levels, the best time interval for monitoring anti-Xa level, and the optimal timing of dosage (twice a day or three times a day). In the absence of monitoring of anti-Xa levels, LMWH should not be used.2,62
The decisions on the choice of anticoagulation should be made by both the physicians and patient, who needs to be fully informed of the potential risks and benefits associated with the various therapeutic options.52
Current guidelines recommend frequent monitoring of anticoagulation therapy during pregnancy irrespective of the antithrombotic regimen chosen.2,52,56 INR should be measured weekly. For mechanical MVR, older-generation AVR, and those with risk factors for thromboembolism, the target INR should be 3 (range 2.5-3.5). For patients with a bileaflet aortic valve and no risk factors for thromboembolism, the recommended target INR is 2.5 (range 2-3). Low-dose aspirin (75-100) is recommended for pregnant patients with either mechanical or bioprosthetic valves in the second and third trimester. The dose of unfractionated heparin should prolong the 6-hour postinjection aPTT to at least twice the control. Recommendations on LMWH suggest twice-daily administration to achieve an anti-factor Xa level of 0.8 to 1.2 U/mL 4 to 6 hours2 after injection performed every 1-2 weeks.52,62
Delivery should be planned so that the mother does not enter labor in a fully anticoagulated state, which would increase the risk of severe maternal hemorrhage and place the neonate at risk of intracranial hemorrhage (in case of VKA).52 A switch from VKA to UFH/LMWH is done at 36 weeks. LMWH is switched to UFH 36 hours before planned delivery. UFH is discontinued 4-6 hours before planned delivery and restarted 4-6 hours after delivery if there are no bleeding complications. Vaginal delivery is preferred. Catheter placement for epidural anesthesia is not advisable within 12 hours of the last LMWH dose because of the risk of spinal or epidural hematoma. VKA can be restarted on same day of an uneventful vaginal delivery and after 1-2 days in those with caesarean section. A complete switchover to VKA should be done in-hospital. VKA is safe during breast feeding, as warfarin is highly protein bound and not readily detectable in breast milk.
In case of an emergency delivery with the patient on UFH/LMWH, protamine is used to reverse the effect of heparin although it only partially reverses the anticoagulant effect of LMWH (60%-80%). If emergency delivery occurs with the patient on VKA, the neonate is at risk of intracranial hemorrhage during vaginal delivery. Hence caesarean section is recommended.52 Prothrombin complex concentrates or fresh frozen plasm is administered to achieve an INR < 2. A management algorithm for pregnant women with PHV is outlined in Fig. 53–2.67
Management of a woman with a prosthetic heart valve (PHV) at time of pregnancy. IV, intravenous; SVD structural valve deterioration; VHD, valvular heart disease.