Valvular Heart Disease: Recommendations for Investigation and Management29 April 2002
Valvular Heart Disease: Recommendations for Investigation and Management
Summary of guidelines produced by a working group of the British Cardiac Society and the Research Unit of the Royal College of Physicians
This summary is taken from the Journal of the Royal College of Physicians, Vol30, No4, 1996 and is reproduced by permission.
B D Prendergast, MRCP, British Heart Foundation Research Fellow
A P Banning, MRCP, British Heart Foundation Research Fellow
R J C Hall, MD, FRCP, Consultant Cardiologist, Department of Cardiology, University Hospital of Wales, Cardiff
Abstract - Despite the decline of rheumatic fever, valvular heart disease remains a common and important problem. The investigation and treatment of patients have been revolutionised by the widespread availability of echocardiography, which provides information about disease aetiology, accurate and non-invasive assessment of severity and a means of monitoring progression. These advances have facilitated progress in reconstructive valve surgery and the application of balloon valvuloplasty. Improved valve prostheses and reduction in peri-operative mortality have also broadened the indications for valve replacement surgery. A working party convened by the British Cardiac Society and the Royal College of Physicians of London have reviewed current practice and provided guidelines, summarised in this article, for the investigation and management of adults with valvular heart disease in primary care, district general hospitals and tertiary referral centres.
In 1995 the Joint Audit Committee of the British Cardiac Society and the Royal College of Physicians of London convened a working group to recommend guidelines for the investigation and management of adults with valvular heart disease. The group included representatives from general practice, the nursing profession, community medicine, district general hospital cardiology, tertiary centre cardiology and cardiac surgery, the legal profession and regional health authority purchasing bodies. This paper is a summary of the full guidelines , to which reference should be made for clarification or further information.
the general practitioner
1.\tPatients in whom valvular heart disease is life-threatening.
2.\tPatients with symptomatic valvular heart disease, who will benefit from exact diagnosis and planned management.
3.\tPatients in whom valvular heart disease may underlie the presenting condition, e.g. heart failure caused by aortic stenosis, atrial fibrillation caused by mitral valve disease.
4. \tPatients with valvular heart disease in special situations, e.g. pregnancy.
A background of rheumatic fever, ischaemic heart disease, hypertension or intravenous drug abuse should be considered and potential symptoms of angina, syncope or dyspnoea evaluated. Examination should determine the heart rhythm and blood pressure and the presence of any cardiac murmurs or signs of heart failure.
The finding of a systolic murmur in conjunction with any of the above symptoms should not be dismissed as aortic "sclerosis" (a defunct term). Given the poor prognosis of symptomatic aortic stenosis  and the difficulty of precise diagnosis, such patients should generally be referred urgently for a specialist opinion.
Routine investigations are unnecessary. Simple blood tests, e.g. haemoglobin, thyroid function tests, should be requested according to clinical circumstances. An electrocardiogram or chest x-ray are unlikely to influence the referral decision and are unnecessary.
If a patient has suspected valvular heart disease it would be usual to refer unless the patient is unwilling or other conditions are judged to contra-indicate referral. The degree of urgency is indicated below: in uncertain cases telephone advice should be sought from an experienced specialist, ideally of registrar grade or above.
·\tEmergency. Valvular heart disease with uncontrolled symptoms or suspected infective endocarditis. Hospital admission will usually be required.
·\tUrgent. Symptomatic valvular heart disease. The patient should attend a specialist clinic as soon as possible.
·\tRoutine. Diagnostic or long-term management issues. The patient should be seen in accordance with routine clinic procedure.
Electrocardiogram All patients should have an electrocardiogram as part of the initial consultation.
Chest x-ray A chest x-ray is not indicated routinely at the first assessment . Baseline examination is useful if serial assessment is likely. The detection of coincident pulmonary disease may be important, particularly when invasive investigation is being considered.
Echocardiography This is the non-invasive method of choice for the diagnosis and evaluation of valvular heart disease. Trans-oesophageal echocardiography is a specialised technique with specific indications in the assessment and management of valvular heart disease (Table 1). Referral to a centre with this facility may be necessary.
Assessment and advice The general practitioner should be promptly informed of the specialist assessment, with emphasis on diagnosis and management, and any other important issues, e.g. endocarditis prophylaxis, anticoagulation. The patient should receive advice on the following issues, where appropriate:
·\tcontraception or the possibility/advisability of pregnancy in women of childbearing age
·\tendocarditis awareness and prophylaxis
·\tsafety to drive.
Nurse specialists can play an important role in patient education and should be available if possible. Their advice can be supplemented by appropriate literature.
Follow-up arrangements should be made clear, with emphasis on any indications for early re-referral.
Many patients do not require hospital follow-up. Asymptomatic patients with a good long-term prognosis (e.g. "innocent" murmurs, trivial mitral regurgitation, mild aortic regurgitation in a hypertensive patient) can be discharged after suitable advice and reassurance, with clear guidelines for re-referral.
The frequency of follow-up for patients with significant native valve disease should be determined according to the diagnosis and its severity. Most patients should attend a consultant-led clinic annually. Those with significant aortic stenosis, evidence of ventricular impairment/dilatation or symptoms suggestive of disease progression should be seen more often
(Fig 1a, b).
The patient should be seen by an experienced cardiovascular physician, i.e. a consultant or staff grade cardiologist (or physician experienced in the care of cardiovascular disease), a senior registrar in cardiology or an experienced cardiology registrar or clinical assistant. Senior house officers should be supervised and have access to experienced colleagues for advice. Ideally, the patient should see a consultant every 2-3 years.
Facilities for an electrocardiogram, chest x-ray and echocardiogram should be available on the day of clinic attendance. Anticoagulant control should be reviewed. Patients with new or deteriorating symptoms, an embolic event or unexplained febrile illness require detailed investigation. If there is a definite suspicion of severe haemolysis or endocarditis the patient should be admitted to hospital.
Patients should be intermittently reviewed by the general practitioner, possibly in a formalised "shared care" arrangement. Those with new signs or symptoms should be re-referred without delay. Any febrile illness of uncertain origin should be considered as infective endocarditis and referred urgently prior to any antibiotic administration.
Vasodilators palliate valvular regurgitation  and left ventricular impairment and should be considered for all patients with significant aortic or mitral regurgitation. They are contra-indicated in aortic stenosis. Digoxin is the conventional treatment for atrial fibrillation and may have a role in patients with heart failure, even in sinus rhythm, in addition to diuretics and angiotensin converting enzyme inhibitors. Anticoagulant therapy will be discussed later.
Poor prognostic features and referral for intervention
Patients with the following features have a poor prognosis and should receive close attention: atrial fibrillation, left ventricular dysfunction or dilatation, pulmonary hypertension, co-existent coronary artery disease and disease affecting more than one valve.
Patients with increasing symptoms, echocardiographic or clinical evidence of disease progression, or who develop new poor prognostic features should be considered for referral with a view to intervention. Ideally, elective investigations should be undertaken in the tertiary centre within three months of referral. In urgent circumstances the waiting time should not exceed one month; occasionally, immediate admission may be necessary (Table 2).
the role of the tertiary centre
All patients with symptomatic mitral stenosis should be actively considered for percutaneous balloon mitral valvuloplasty. Contra-indications include the presence of left atrial thrombus, severe subvalvar involvement or valve calcification, and significant mitral regurgitation. It is mandatory to detect these beforehand, using trans-oesophageal echocardiography .
The surgical alternatives are open mitral commissurotomy or mitral valve replacement. Recent studies comparing balloon valvuloplasty (in suitable patients) with surgery for mitral stenosis have shown no difference in outcome or complication rates [6,7].
Timing of intervention should be guided by the severity of symptoms and evidence of progressively increasing left ventricular size. Echocardiography is the investigation of choice; an end-systolic dimension greater than 5.5cm suggests the need for surgery .
The benefits of mitral valve repair  may favour earlier intervention if repair is thought feasible. Trans-oesophageal echocardiography is essential for pre- and peri-operative assessment .
Symptomatic severe aortic stenosis has a very poor prognosis . Unless there are exceptional circumstances, urgent surgery is always indicated. The presence of left ventricular impairment solely due to aortic stenosis is not a contra-indication to surgery. The management of patients with asymptomatic severe aortic stenosis is controversial: surgery is often indicated, especially if the trans-valve gradient exceeds 100mmHg.
Initial enthusiasm for balloon aortic valvuloplasty as an alternative to valve replacement in the elderly  has been tempered by poor long-term results . The procedure should be restricted to occasional patients with severe aortic stenosis in the following circumstances:
·\twhen other major surgery is proposed for underlying disease with an uncertain prognosis
·\tas a bridge to aortic valve replacement if immediate surgery is inadvisable.
Echocardiographic evidence of progressive left ventricular dilatation (end-systolic dimension >5.5cm), even in the absence of major symptoms, indicates the need for surgery . Aortic valve replacement is usually necessary and some patients with dilatation of the ascending aorta (>5.5cm) may require root replacement.
Mixed valve disease
Assessment and treatment comprise a summation of the factors for each individual valve lesion. The mortality and morbidity of double valve replacement are significantly higher and this may influence the decision to proceed.
Special surgical considerations
With appropriate patient selection, the results of valve replacement in the elderly can approach those in younger patients. Mortality after aortic valve replacement is consistently lower than after mitral valve replacement [13,14]. The risk of both procedures rises with advancing age or if there is co-incident coronary artery disease. A simple, quick procedure is important in the elderly; additional bypass grafting should only be undertaken for severe coronary artery disease.
Pre-operative coronary angiography
Echocardiographic techniques have negated the need for routine invasive assessment of valve disease. Most patients undergoing valve surgery require pre-operative coronary angiography to detect co-existent coronary artery disease. In some, e.g. the young with no history of chest pain or risk factors for coronary artery disease, this may be unnecessary. Equally, coronary angiography should not delay a potentially life-saving operation, e.g. emergency aortic valve replacement.
The choice of prosthesis
Mechanical prostheses These offer durability and longevity. Despite the problems associated with anticoagulation and small risks of thrombosis-in-situ they are the first choice for the majority of patients.
Tissue prostheses - Heterografts (or xenografts) offer a low thrombo-embolic rate and avoid the need for anticoagulation. The high incidence of long-term valve failure restricts their use to elderly patients with aortic valve disease whose life expectancy is less than that of the valve (10-12 years), situations where anticoagulants are contra-indicated and some patients with infective endocarditis. They are seldom used in the mitral position due to a high risk of complications [15,16] and also because anticoagulation is usually indicated for other reasons (left atrial enlargement and/or atrial fibrillation). Homografts are also associated with a low thrombo-embolic rate. Relative indications include aortic valve replacement in active endocarditis, aortic root replacement and aortic valve replacement in the young when a tissue valve is necessary.
Urgency of surgical treatment
The urgency should be defined for each patient and guided by clinical judgement. Circumstances will vary in individual centres; in particular, the quantity of emergency work undertaken can restrict elective waiting list admissions.
The patient's condition and the presence of any cardiac murmurs at the time of discharge should be documented. The general practitioner should be provided with essential information (Table 3), supplemented by a fuller summary in due course. Patients also need specific advice on recovery, exercise, sex, work, sport, driving, flying, dentistry, drug therapy (particularly anticoagulants) and the reporting of new symptoms, especially those suggesting valve failure or endocarditis. This should be accompanied by an information booklet.
The broad principles are the same as those in native valve disease. Intermittent follow-up by a physician with a cardiovascular interest or cardiac surgeon is satisfactory, providing there are no significant complications. Duplication of follow-up in the district general hospital and the interventional centre should be avoided.
The patient and general practitioner should be aware of the need for early review if problems arise. When medical advice is needed the general practitioner should contact a senior member of the team involved in the patient's hospital care.
This subject has been reviewed by a European working party which has produced comprehensive guidelines  including indications for anticoagulation and International Normalised Ratio (INR) recommendations for patients with prosthetic valves or native valve disease.
Background - In native valve disease the risk of thrombosis is related to the following factors: atrial fibrillation, left atrial enlargement, mitral obstruction and impaired left ventricular function .
General surgery in anticoagulated patients - The peri-operative anticoagulation regime should be discussed in advance with a cardiovascular physician or cardiac surgeon. Patients with native valve disease can often stop or reduce their anticoagulants. Patients with prosthetic heart valves should not discontinue oral anticoagulation: if the patient cannot take oral medication, intravenous warfarin or heparin can be used. Heparin is less effective in preventing thrombosis: if it is used, then warfarin should be re-instituted as soon as possible. Most surgical procedures can be safely performed with an INR of around 2.0. In procedures resulting in large raw surfaces the risk of bleeding must be weighed against the risk of valve thrombosis if the INR falls below 2.0.
2. \tInfective endocarditis
Prevention - Working parties in the United Kingdom and America have produced guidelines [19,20] which are summarised in the full guidelines. Patient information cards are available from The British Heart Foundation.
Diagnosis - Diagnosis can be difficult. Minimum investigations include a full blood count, biochemical screen, erythrocyte sedimentation rate (and/or C-reactive protein) and at least three sets of blood cultures from two separate sites before starting antibiotic therapy. Trans-thoracic echocardiography should be performed though a negative study does not exclude the diagnosis. Trans-oesophageal imaging has a high negative predictive value  and is particularly sensitive in the detection of vegetations  or abscesses . Echocardiographic features have been incorporated in the new "Duke" diagnostic criteria  which are superior to the old von Reyn criteria.
Treatment - Optimal management requires close liaison between the cardiologist, cardiac surgeon and microbiologist. Antibiotic therapy should be started promptly after blood cultures have been taken. Recommendations for the treatment of streptococcal and staphylococcal endocarditis have been published . Indications for surgery include heart failure due to valvular regurgitation, abscess formation, failure to respond to medical therapy and an unstable infected prosthetic valve. Urgent surgery should not be delayed by an arbitrary period of pre-operative antibiotic therapy.
Prosthetic valve endocarditis - Rigorous prophylaxis is vital and intravenous cannulae, urinary catheters and indwelling lines should be avoided whenever possible. Prosthetic valve endocarditis has a mortality of approximately 50%, and further surgery is usually required, often as an emergency. It should be managed in a cardiothoracic surgical centre unless close communication and rapid transfer are possible. Trans-thoracic echocardiography is inadequate and prosthetic valve patients with unexplained pyrexia should be referred for trans-oesophageal imaging.
3. \tValvular heart disease in pregnancy and young women
This group of patients should be managed by a specialist with experience in this area.
Planning of family - The oral contraceptive pill is safe in valvular heart disease, though anticoagulated patients should be aware of the potential drug interaction. Patients in New York Heart Association class I or II before conception will usually tolerate pregnancy without difficulty. If intervention seems likely this should be carried out before pregnancy if at all possible.
The choice of prosthetic valve for a young woman anticipating pregnancy is difficult. Bioprostheses do not require anticoagulation but degenerate more quickly in pregnancy . Mechanical prostheses are durable but carry the need for anticoagulation. The choice should be made in consultation with the patient.
The pregnant patient with a murmur - These patients should be seen by an obstetrician and cardiologist as early as possible in case pregnancy is contra-indicated. Minimum investigations are a haemoglobin level and electrocardiogram; echocardiography should be performed if the diagnosis is uncertain.
Follow-up - The patient should be followed by an obstetrician and cardiologist. Deterioration is most common in mitral and aortic stenosis and these patients should receive particular attention.
Medical treatment - Beta-blockers may prevent deterioration in pregnant women with mitral stenosis . Angiotensin converting enzyme inhibitors are contra-indicated in pregnancy.
Intervention - Balloon dilatation of mitral or aortic stenosis can be performed in early pregnancy. If valve replacement is required this should be deferred until pregnancy is sufficiently advanced to allow delivery by Caesarean section immediately before cardiac surgery .
Endocarditis - Prophylaxis is not indicated to cover normal delivery in women with native valve disease. It should be used in patients with prosthetic valves or a history of previous endocarditis. Advice should be sought from a microbiologist concerning the choice of antibiotics.
Delivery - The mode of delivery and anaesthesia should be discussed in advance. Epidural anaesthesia should be avoided because of the risk of hypotension.
Anticoagulation & prosthetic valves in pregnancy - Patients with tissue valves who are not chronically anticoagulated generally do well but need careful follow-up because of the risk of valve degeneration. The risk of pregnancy for patients with mechanical prostheses is small. The main hazard is valve thrombosis and embolism, and meticulous anticoagulation is required. Most patients, particularly those with prosthetic valves, should continue oral anticoagulants with close INR monitoring. Warfarin carries a 3-5% risk of embryopathy  but the risk of fetal bleeding is very low at doses less than 5mg/day . It is not significantly excreted in breast milk. Heparin is not teratogenic but has a narrow therapeutic index: its use should be confined to the first trimester of pregnancy in women whose warfarin requirement is high (ie greater than 5mg/day). The convention has been to admit patients with mechanical prostheses for peri-partum anticoagulant control with heparin. Anticoagulant control in this situation can be extremely difficult, and for this reason we recommend elective caesarean section at 38 weeks of pregnancy with only a brief 3-day interruption of warfarin therapy (two days preoperative and one day postoperative) .
These guidelines represent the consensus views of the working group. The aims have been to provide a balanced set of recommendations and to guide practice without substituting for individual clinical judgement. These standards should stimulate audit of the care of valvular heart disease and highlight areas where practice can be improved.
Members of the Working Group
R J C Hall, Consultant Cardiologist, University Hospital of Wales, Cardiff;
A P Banning, Senior Registrar in Cardiology, John Radcliffe Hospital, Oxford;
N Boon, Consultant Cardiologist, The Royal Infirmary of Edinburgh;
R Boyle, Consultant Cardiologist, York District Hospital;
E G Butchart, Consultant Cardiothoracic Surgeon, University Hospital of Wales, Cardiff;
K D Dawkins, Consultant Cardiologist, Southampton General Hospital;
G J Green, Consultant Cardiologist, Glan Clwyd District General Hospital, Rhyl;
R H King, General Practitioner, Cambridge;
S Livesey, Consultant Cardiothoracic Surgeon, Southampton General Hospital;
H Marsh, General Practitioner, Cardiff;
C Oakley, Professor of Clinical Cardiology, The Hammersmith Hospital, London;
J Parker, President of The British Cardiac Society and Consultant Cardiothoracic Surgeon, St. George's Hospital, London;
B D Prendergast, BHF Junior Research Fellow, University of Wales College of Medicine, Cardiff;
S Saltissi, Consultant Physician & Cardiologist, Royal Liverpool University Hospital;
Sally Smith, Barrister, 1, Crown Office Row, The Temple, London;
D Spence, Regional Audit Co-ordinator, Trent Regional Health Authority;
H Swanton, Consultant Cardiologist, The Middlesex Hospital, London;
A Townsend, Coronary Care Unit Sister, Freeman Hospital, Newcastle-upon-Tyne;
T Treasure, Professor of Cardiothoracic Surgery, St. George's Hospital, London;
C F Weston, Senior Lecturer in Cardiology, University of Leeds;
D Wheatley, Professor of Cardiothoracic Surgery, Glasgow Royal Infirmary;
P Wilkinson, Consultant Cardiologist, Ashford & Harefield Hospitals, Middlesex.
Dr Prendergast and Dr Banning were supported by the British Heart Foundation Junior Research Fellowships.
1\tPrendergast BD, Banning AP, Hall RJC. Valvular heart disease: recommendations for \t\tinvestigation and management. Guidelines produced by a working group of the British Cardiac \tSociety and the Research Unit of the Royal College of Physicians. London: Royal College of \tPhysicians, 1996.
2\tRapaport E. Natural history of aortic and mitral valve disease. Am J Cardiol 1975;35:221-7.
3\tRoberts G, Saxton H, Field S. Making the best use of a department of clinical radiology: \t\tguidelines for doctors (2nd edn). London: Royal College of Radiologists, 1993.
4\tScognamiglio R, Rahimtoola SH, Fasoli G, Nistri S, Dalla Volta S. Nifedipine in asymptomatic \tpatients with severe aortic regurgitation and normal left ventricular function. N Engl J Med \t1994;331:689-94.
5\tGoldstein SA, Campbell AN. Mitral stenosis: evaluation and guidance of valvuloplasty by \ttransesophageal echocardiography. Cardiol Clin 1993;11:409-25.
6\tCohen JM, Glower DD, Harrison JK, Bashore TM et al. Comparison of balloon valvuloplasty \twith operative treatment for mitral stenosis. Ann Thorac Surg 1993;56:1254-62.
7\tReyes VP, Soma Raju B, Wynne J, Stephenson LW et al. Percutaneous balloon valvuloplasty \tcompared with open surgical commissurotomy for mitral stenosis. N Engl J Med 1994;331:961-7.
8\tZile MR. Chronic aortic and mitral regurgitation: choosing the optimal time for surgical correction. \tCardiol Clin 1991;9:239-53.
9\tSand ME, Naftel DC, Blackstone EH, Kirklin JW, Karp RB. A comparison of repair and \t\treplacement for mitral valve incompetence. J Thorac Cardiovasc Surg 1987;94:208-19.
10\tFreeman WK, Schaff HV, Khandheria BK, Oh JK et al. Intraoperative evaluation of mitral valve \tregurgitation and repair by transesophageal echocardiography: incidence and significance of \tsystolic anterior motion. J Am Coll Cardiol 1992;20:599-609.
11\tCribier A, Saoudi N, Berland J, Savin T et al. Percutaneous transluminal valvuloplasty of \tacquired aortic stenosis in elderly patients: an alternative to aortic valve replacement? Lancet \t1986;i:63-7.
12\tBernard Y, Etievent J, Mourand J-L, Anguenot T et al. Long-term results of percutaneous aortic \tvalvuloplasty compared with aortic valve replacement in patients more than 75 years old. J Am \tColl Cardiol 1992;20:796-801.
13\tFreeman WK, Schaff HV, O'Brien PC, Orszulak TA et al. Cardiac surgery in the octogenarian: \tperioperative outcome and clinical follow-up. J Am Coll Cardiol 1991;18:29-35.
14\tDavis EA, Gardner TJ, Gillinov AM, Baumgartner WA et al. Valvular disease in the elderly: \tinfluence on surgical results. Ann Thorac Surg 1993;55:333-8.
15\tBloomfield P, Wheatley DJ, Prescott RJ, Miller HC. Twelve-year comparison of a Bjork-Shiley \tmechanical heart valve with porcine bioprostheses. N Engl J Med 1991;324:573-9.
16\tHammermeister KE, Sethi GK, Henderson WG, Oprian C et al. A comparison of outcomes in men \t11 years after heart-valve replacement with a mechanical valve or bioprosthesis. N Engl J Med \t1993;328:1289-96.
17\tGohlke-Barwolf C, Acar J, Oakley C, Butchart E et al. Guidelines for prevention of \t\tthromboembolic events in valvular heart disease. J Heart Valve Dis 1993;2:398-410.
18\tStroke Prevention in Atrial Fibrillation Investigators. Predictors of thromboembolism in atrial \tfibrillation: II. Echocardiographic features of patients at risk. Ann Intern Med 1992;116:6-12.
19\tAntibiotic prophylaxis of infective endocarditis. Recommendations from the Endocarditis \tWorking Party of the British Society for Antimicrobial Chemotherapy. Lancet 1990;335:88-9.
20\tDajani AS, Bisno AL, Chung KJ, Durack DT et al. Prevention of bacterial endocarditis. \t\tRecommendations by the American Heart Association. JAMA 1990;264:2919-22.
21\tLowry RW, Zoghbi WA, Baker WB, Wray RA, Qui-ones MA. Clinical impact of \t\ttransoesophageal echocardiography in the diagnosis and management of infective endocarditis. \tAm J Cardiol 1994;73:1089-91.
22\tErbel R, Rohmann S, Drexler M, Mohr-Kahaly S et al. Improved diagnostic value of \t\techocardiography in patients with infective endocarditis by transoesophageal approach. A \tprospective study. Eur Heart J 1988;9:43-53.
23\tDaniel WG, MYgge A, Martin RP, Lindert O et al. Improvement in the diagnosis of abscesses \tassociated with endocarditis by transoesophageal echocardiography. N Engl J Med \t\t1991;324:795-800.
24\tDurack DT, Lukes AS, Bright DK. New criteria for diagnosis of infective endocarditis: utilization \tof specific echocardiographic findings. Am J Med 1994;96:200-9.
25\tAntibiotic treatment of streptococcal and staphylococcal endocarditis. Report of a Working Party \tof the British Society for Antimicrobial Chemotherapy. Lancet 1985;ii:815-7.
26\tBadduke ER, Jamieson RE, Miyashima RT, Munro AI et al. Pregnancy and childbearing in a \tpopulation with biologic valvular prostheses. J Thorac Cardiovasc Surg 1991;102:179-86.
27\tOakley CM. Pregnancy in heart disease. In: Best A (ed), Cardiovascular physiology in women. \tCardiovasc Clin 1988;19:57-78.
28 \tOakley CM. Pregnancy in heart disease. In: Jackson G (ed), Difficult cardiology. Practical \tmanagement and decision making. Martin Dunitz Ltd., London 1990:1-20.
29\tde Swiet M. Prescribing in Pregnancy. Anticoagulants. BMJ 1987;294:428-30.
30\tCotrufo M, de Luca TSL, Calabro R, Mastrogiovanni G, Lama D. Coumarin anticoagulation \tduring pregnancy in patients with mechanical valve prostheses. Eur J Cardiothorac Surg \t1991;5:300-5.
Address for correspondence: Professor R J C Hall, Department of Cardiology, Hammersmith Hospital, Du Cane Road, London W12 0HS.
The full version of these guidelines and recommendations, published in booklet form, will be circulated to members of the British Cardiac Society.
They will also be available for sale from the Royal College of Physicians, price £7.00 (overseas £9.00) - address 11 St. Andrews Place, London, NW1 4LE