• 
• 

Three Part Question

In patients who have undergone [mitral valve repair] does [warfarin] have any benefit in terms of [survival or reducing thromboembolic complications]?

Clinical Scenario

You are looking after a 55-year old man with mitral regurgitation who underwent a mitral valve repair with a quadrangular P2 resection and an annuloplasty ring. He is in sinus rhythm and well. You had talked to him about warfarin in the context of a mitral replacement in case you couldn't repair the valve. However he is now keen to eliminate any additional risk of stroke as his mother had a severe stroke 5 years ago. He says that he would like to take warfarin if there was any additional chance of preventing a stroke. You decide to review the literature to answer his question.

Search Strategy

Medline1966 to July 2006, Embase 1980 to July 2006 using the OVID interface.
This search was repeated in the Cochrane database of systematic reviews. The American College of Cardiology, AHA/ACCP, NICE, SIGN, European society of cardiology and the British Society for Haematology guidelines were also hand searched.
[mitral valve repair.mp OR mitral repair.mp OR annuloplasty.mp OR mitral ring.mp] AND [exp warfarin/ OR warfarin.mp OR exp.anticoagulants/ OR anticoagulation.mp OR exp coumarins/ OR coumarin.mp].

Search Outcome

The search identified 63 papers on Medline and 64 papers on Embase in addition to the hand searched guidelines. 12 papers gave the best evidence on this topic

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Braunberger et al, 2001, France	162 patients with non rheumatic valve insufficiency undergoing repair between 1970 and 1984. All patients received warfarin for 2 months and this was stopped as long as sinus rhythm was restored.	Retrospective cohort study (level 2b)	Survival	3/162 died in first month 20 years survival 48% (95% CI 40% to 57%),	Mean age 56 years 45%were in Atrial Fibrillation All patients had an annuloplasty ring
			Thromboembolic events	4 patients had a thromboembolic strokes (of whom 3 were in atrial fibrillation),
Chauvaud et al, 2001, France	951 patients with rheumatic mitral valve insufficiency, operated on from 1970 to 1994 at a single institution All patients received warfarin for 2 months and this was stopped as long as sinus rhythm was restored. Mean follow up 12 years, maximum 29 years.	Retrospective Cohort Study (level 2b)	Survival	89±19% at 10 years 82±18% at 20 years	Mean age 25.8 years 95% of patients received a rigid Carpentier Edwards ring. 47% of patients were in AF
			Rates of thrombosis	0.4% patients per year (33 events in 35 patients), with 3 deaths Three events occurred in the first 3 months of postoperative course (2 hemiplegia and 1 gastrointestinal hemorrhage). After this there were 29 CVAs 4 limb emboli All thromboemboli associated with Atrial fibrillation
			Freedom from reoperation	82±19% at 10 years 55±25% at 20 years
Aramendi et al, 1998, Spain	235 patients between 1990 and 1995 underwent either mitral valve repair (MVr) or replacement (MVR): 122 patients were given Ticlopidine for 90 days. Number of mitral repair patients: (n= 20). 40 patients were randomised to have warfarin. Number of mitral valve repairs: (n= 11).	Prospective cohort study (level 2b)	Incidence of thromboembolic complications in both groups	6 episodes of thrombo-embolism in 209 survivors. All episodes of thrombo-embolism occurred in the first post-operative year. 2 patients were in the ticlopidine group. Linearised incidence 0.5 %per pt-yr 4 episodes of thrombo-embolism in the warfarin group. Linearised incidence of 3 % per pt-yr. Freedom from thromoembolism was 98±1 % for the ticlopidine group compared to 86±5 % for the warfarin group. (p= 0.002) Freedom from thrombo-embolism, there was no significant differences between the types of prosthesis used, aortic or mitral position, repair or bioprosthesis, or cardiac rhythm (sinus or atrial fibrillation) 4 episodes of haemorrhage for the entire series. All occurred in the first 3 months of follow-up. 3 episodes in the ticlopidine group with the linearised incidence of 0.75 % per pt-yr 1 episode of haemorrhage in the warfarin group. The linearised incidence of 0.75 % per pt-yr. (p= NS)	Non-randomised study with the proportion of patients with atrial fibrillation more in warfarin group, compared to ticlopidine group (50 % versus 30 %; p< 0.05).
Galloway et al, 1988, USA	148 patients underwent Carpentier reconstruction between 1980 and 1986. All patients were placed on warfarin on the third post-operative day. Warfarin was discontinued at the end of 3 months and patients were switched to antiplatelet therapy. High risk patients with predisposing factors were continued on warfarin long term basis	Retrospective cohort study (level 2b)	Long-term thrombo-embolic rates following mitral valve reconstruction.	Of the 125 survivors without re-operation, (73 of 125) 58.4 % were on an anticoagulant or antiplatelet therapy.(34 of 125) 27.2 % were on warfarin and (39 of 125) 31.2 % were receiving aspirin with or without dipyridamole Incidence of anti-coagulation related complications was 0.33 per 100 patient-years 6 late thromboembolic complications in 5 patients (2 strokes and 4 TIAs) 2 patients were on warfarin when they had TIAs at 2.5 months and 13.2 months post surgery 4 patients were on aspirin; 2 had a stroke, 6.5 and 16.3 months post surgery; 1 patient died and the other recovered from the stroke, 2 episodes of TIAs, at Ĕ 1month and 15.6 months post surgery Total freedom from late thrombo-embolism was 97.6 % at 1 year and remained 95.2 % for 2- 7 years.	This is a retrospective follow-up study and therefore no causal or prognostic inferences can be made.
Butchart et al, 2005, UK	Guideline from the Working Groups on Valvular Heart Disease, Thrombosis, and Cardiac Rehabilitation and Exercise Physiology, European Society of Cardiology	Guideline (level 1a)	Recommendation for patients undergoing a mitral repair	Warfarin for the first 3 months, in all patients with bioprostheses or mitral valve repair involving the use of a prosthetic annuloplasty ring. Although there is widespread use of aspirin as an alternative to anticoagulation for the first 3 months in patients with no other indications for anticoagulation, there are no randomized studies to support the safety of this strategy.	INR 2.5 recommended for people without risk factors or 3.0 for those with factors such as AF , LA>50mm, EF<35%.
			Follow up	Patients with bioprostheses or mitral valve repair who are not on anticoagulation require close follow-up not only to detect evidence of structural degeneration or recurrence of mitral regurgitation, but also to detect the onset of AF.
Vaughan et al, 2005, UK	A Survey of anticoagulation practice among UK cardiothoracic surgeons 52% (97/185) replies	Survey (level 3b)	Warfarin after Mitral repair	64% (52/81) use warfarin after mitral valve repair, when an annuloplasty ring is inserted. Aspirin is used long term by 54% (44/82) of consultants after mitral valve repair.	Only a 52 % response rate. Only UK surgeons
Deloche et al, 1990, USA	206 patients underwent mitral valve repair with a prosthetic ring between 1972 and 1979. 195 patients survived the operation. They followed the Carpentier protocol post mitral valve repair of warfarin on the third post-operative day. Warfarin was discontinued at the end of 3 months and patients were switched to antiplatelet therapy. High risk patients with predisposing factors were continued on long term basis	Retrospective Cohort Study (Level 2b)	Long-term freedom from thrombo-embolism and anti-coagulation related haemorrhage.	Of the 189 patients available for follow-up. 10 patients had a thromboembolic event. Seven of these were TIAs without any major sequelae. Two patients had permanent neurologic deficit and one died Actuarial rate of freedom from thromboembolism at 15 years of 93.97 % ±2.3 % Linearised rate of thromboembolism of 0.4 % /pt -yr 88 patients (49 %) received long-term anticoagulation with warfarin because of AF. 6 haemorrhagic complications. Actuarial rate of freedom from haemorrhagic complications 95.6 %±1.97% at 15 years for the total series and a rate of 91 % on the 88 patients on warfarin. Linearised rate of 0.5 %/pt-yr.
Jovin et al, 2005, USA	245 patients who underwent mitral valve repair (MVR). 73 patients had a history of AF or were in AF on admission. 172 patients had no previous arrhythmias and were in sinus rhythm	Retrospective Cohort study (level 2b)	Frequency of AF at discharge from hospital	65 (27 %) patients had AF at discharge. 56 (86 %) were in AF prior to surgery and 9 (14 %) were in sinus rhythm. 180 patients were discharged in sinus rhythm and of these 64 (36 %) had an episode of AF during the post-operative period 106 patients never had AF before or after surgery. Risk factors for AF at discharge: - AF on admission OR 59.66 Age > 55 OR 3.51 LA >5 cm OR 4.32 Ejection fraction > 40 % OR 3.29. Of 65 patients who were in AF at discharge 61 were discharged on warfarin and 2 were discharged on warfarin and aspirin. Of 180 in sinus rhythm at discharge 98 were discharged on warfarin	This is a retrospective study and therefore no causal or prognostic inferences can be made.

Comment(s)

The ACC/AHA 1998 guidelines for the management of patients with valvular heart disease do not provide recommendations for patients who have undergone a mitral valve repair and neither do the ACCP guidelines of 2004. The European Society of Cardiology do provide guidelines for these patients, stating that there are no randomized controlled trials to support the safety of omitting warfarin after mitral repair. They recommend 3 months of warfarin at a target INR of 2.5 or 3.0 if there are additional risk factors. They acknowledge that this is based on expert consensus and acknowledge that many surgeons do not follow this guideline. Vaughan et al in a survey of UK surgeons found that 64% of consultants used warfarin post mitral repair, thus demonstrating that there is much variation in the anticoagulation management of patients post mitral repair in the UK. Of the large series' of patients with mitral valve repair, Carpentier et al have provided the longest follow up. They reported their long term results of 928 patients with rheumatic regurgitation and 162 patients with non-rheumatic regurgitation, with a follow up of up to 29 years. This group used 2 months of warfarin for all patients, and only 3 patients had a stroke in the first 3 months. There were 37 thromboembolic events in these patients and there was a very high association with atrial fibrillation. Jovin A et al reviewed 245 patients who underwent mitral repair. 181 patients were discharged in sinus rhythm; 64 (36 %) of these had an episode of atrial fibrillation(AF) during the post operative period. Of the 180 patients who were in sinus rhythm at discharge, 98 (54 %) were discharged on warfarin, 78 (43 %) were discharged on aspirin and 3 (2%) received no anticoagulation or antithrombotic therapy at discharge. The authors recommended oral anticoagulation for 3 months, until the endothelialization of the newly implanted prosthesis and suture material takes place. Aramendi JL et al reported on the thromboembolic risk of 235 patients having replacements or repairs from 1990-1995. In total six episodes of thrombo-embolism were reported. All occurred in the first post-operative year, four reported during the first three months follow up. 2 patients were on ticlopidine and four episodes were in patients having warfarin. The risk of thromboembolism in the first three months of follow up analyzed by hazard function showed the highest risk in the first month, but this rapidly declined thereafter. Galloway et al reported on 148 patients after mitral valve reconstruction surgery. All patients were started on warfarin on the third post-operative day for 3 months. Six late thromboembolic complications were reported in five patients. Two patients were on warfarin during the event at 2.5 months and 13.2 months post surgery. Four episodes occurred when the patient was on aspirin at 1, 6.5, 15.6 and 16.3 months post surgery. One patient died from the stroke. Total freedom from late thromboemboli was 97.6 % at 1 year and remained at 95.2 % for years 2-7. Deloche A et al followed up 195 patients who underwent mitral repair. All patients were started on warfarin on the third post-operative day and discontinued after 3 months. Ten patients had a thromboembolic event, for an actuarial rate of 94 % patients free of thromboembolism at 15 years (linearised rate of 0.4 %/pt-yr). Of the ten thromboembolic events, seven were TIAs without major sequelae two patients had permanent deficit and one patient died. 88 patients remained on long-term anticoagulation because of AF. There were six reported haemorrhagic complications, for an actuarial rate of 95.6 %±1.97 %patients free of anticoagulation related haemorrhage at 15 years for the total series and 91 % for the 88 patients receiving anticoagulation.

Clinical Bottom Line

The current European Society of Cardiology guidelines support the use of warfarin for 3 months post mitral repair, citing an absence of studies supporting the safety of omitting warfarin. They acknowledge that this is based on expert consensus and that many surgeons do not follow this guideline. The longest follow up studies of patients post-mitral repair report excellent results using short term warfarin, and they also show that a third of patients discharged in sinus rhythm will have an episode of atrial fibrillation shortly after. In addition the highest risk of thromboembolism occurs in the early months post surgery. Therefore until studies demonstrate the safety of omitting warfarin for patients undergoing mitral valve repair 3 months of anticoagulation should remain the standard of care.

References

1. Bonow RO, Carabello B, de Leon AC et al. ACC/AHA Guidelines for the Management of Patients With Valvular Heart Disease. Executive Summary. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guide (Committee on Management of Patients With Valvular Heart Disease). J Heart Valve Dis 1998;7(6):672-707.
2. Bonow RO, Carabello B, de LA, Jr et al. Guidelines for the management of patients with valvular heart disease: executive summary. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Management of Patients with Valvular Heart Disease). Circulation 1998;98(18):1949-84.
3. Salem DN, Stein PD, Al-Ahmad A, Bussey HI, Horstkotte D, Miller N, Pauker SG. Antithrombotic Therapy in Valvular Heart Disease - Native and Prosthetic. The seventh ACCP conference on Antithrombotic and thrombolytic therapy. Chest 2004;126:457S-82S.
4. Vaughan P, Waterworth PD, Vaughan P, Waterworth PD. An audit of anticoagulation practice among UK cardiothoracic consultant surgeons following valve replacement/repair. J Heart Valve Dis 2005;14(5):576-82.
5. Braunberger E, Deloche A, Berrebi A et al. Very Long-Term Results (More Than 20 Years) of Valve Repair With Carpentier's Techniques in Nonrheumatic Mitral Valve Insufficiency. Circulation 2001;104(90001):8I-11.
6. Chauvaud S, Fuzellier JF, Berrebi A, Deloche A, Fabiani JN, Carpentier A. Long-Term (29 Years) Results of Reconstructive Surgery in Rheumatic Mitral Valve Insufficiency. Circulation 2001;104(90001):12I-15.
7. Aramendi JL, Agredo J, Llorente A, Larrarte C, Pijoan J, Aramendi JL, Agredo J, Llorente A, Larrarte C, Pijoan J. Prevention of thromboembolism with ticlopidine shortly after valve repair or replacement with a bioprosthesis. J Heart Valve Dis 1998;7(6):610-4.
8. Galloway AC, Colvin SB, Baumann FG et al. Long-term results of mitral valve reconstruction with Carpentier techniques in 148 patients with mitral insufficiency. Circulation 1988;78(3 Pt 2):I97-105.
9. Butchart EG, Gohlke-Barwolf C, Antunes MJ, Tornos P, De Caterina R, Cormier B, Prendergast B, Lung B, Bjornstad H, Leport C, Hall RJ, Vahanian A. Working Groups on Valvular Heart Disease, Thrombosis, and Cardiac Rehabilitation and Exercise Physiology, European Society of Cardiology. Recommendations for the management of patients after heart valve surgery. European Heart Journal 2005;26:2463-71.
10. Vaughan P, Waterworth PD. An audit of anticoagulation practice among UK cardiothoracic consultant surgeons following valve replacement/repair. Journal of Heart Valve Disease 2005:14(5):576-82.
11. Deloche A, Jebara VA, Relland JY, Chauvaud S, Fabiani JN, Perier P, Dreyfus G, Mihaileanu S, Carpentier A. Valve repair with Carpentier techniques. The second decade. J Thorac Cardiovasc Surg 1990;99(6):990-1001.
12. Jovin A, Hashim S, Jovin IS, Clancy JF, Klovekorn WP, Muller-Berghaus G. Atrial fibrillation at discharge from the hospital in patients undergoing mitral valve repair. Thorac Cardiovasc Surg 2005;53(1):41-5.