Antibiotics & Endocarditis

Cardiac Surgeons Disagree with New Guidance from NICE

Introduction to the article on nice guidance for the use of antibiotics to prevent endocarditis.

Eric Butchart, the recently-retired Senior Consultant Cardiothoracic Surgeon at the University Hospital of Wales in Cardiff, has been a medical advisor to AntiCoagulation Europe since its foundation eight years ago. As such we have greatly valued his contribution, advice and support.

The National Institute for Clinical Excellence has issued guidance on the use of antibiotics for valve patients to prevent endocarditis when having dental or other interventional procedures. The new guidance states that in most cases antibiotics are no longer necessary.

Eric Butchart, like many other Cardiothoracic Surgeons in the UK, is opposed to these new guidelines.

It is important that all valve patients read the article below .

Endocarditis is often fatal. For those that survive the physical and emotional cost cannot be measured in pounds, shillings and pence.

We would suggest that until there is more clinical evidence, that valve patients should discuss with their GP and/or cardiac surgeon whether antibiotics are advisable for them before having any dental or other interventional procedures.


If, after reading Eric Butchart's article, you wish to contact the National Institute for Health and Clinical Excellence their address is:

NICE, Midcity Place, Holborn London WC1V 6NA or

Their Chairman is Sir Michael Rawlins. NICE is the body that issues guidance on what should be available on the NHS. 'Post code lottery' must be a familiar phrase with his department and he has often stated that not everything can be funded on the NHS. One problem he faces is that of an ageing population.

NICE Guidelines on Prophylaxis against Prosthetic Valve Endocarditis

Eric G. Butchart FRCS, FETCS, FESC

Infection on an artificial heart valve, known as prosthetic valve endocarditis (PVE), is the most serious complication of valve replacement, associated with high mortality, greater than 50% in some reported series.

A recent analysis from the International Collaboration on Endocarditis, based on data from 28 countries (JAMA 2007;297:1354-61), reported an average hospital mortality of 22.8%, although in some subgroups the mortality was higher, notably the elderly (37%) and patients on dialysis (40%).

Because PVE is such a serious, life-threatening complication, antibiotic prophylaxis for dental treatment and other surgical procedures likely to introduce bacteria into the blood stream (bacteraemia) has always been considered necessary. It is therefore of great concern to both patients and cardiac surgeons that the recently published National Institute for Clinical Excellence (NICE) guidelines recommend that antibiotic prophylaxis should not be given to any prosthetic heart valve patient undergoing dental treatment or other procedures associated with transient bacteraemia, irrespective of their risk profile, in effect seeking to reverse the established practice of decades and contradicting current guidelines from the British Cardiac Society (Clin Med 2004;4:545-50), the British Society for Antimicrobial Chemotherapy (J Antimicrob Chemother 2006;57:1035-42), the European Society of Cardiology (EHJ 2004;25:267-76, 2005;26:2463-71, 2007;28:230-68), the American College of Cardiology and the American Heart Association (JACC 2006;48:e1-148, Circulation 2007;116:1736-54).

No randomised trials have ever been performed to assess the efficacy of antibiotic prophylaxis in prosthetic valve patients and, given the high mortality of PVE, it is doubtful whether such a trial would ever be considered ethical. In the absence of randomised trials, NICE base their recommendations on small case series of endocarditis in general, containing relatively few patients with PVE, claiming that there is no conclusive proof that antibiotic prophylaxis is effective and that there is no consistent association between having an interventional procedure and subsequently developing endocarditis.

However, they disregard the fact that timing the onset or cause of endocarditis is extremely difficult because the patient may experience many weeks of vague non-specific symptoms before the diagnosis is eventually made.

Furthermore, an association between endocarditis and both dental extraction and scaling has indeed been reported in some series (EHJ 1995;16:1968-74, Ann Intern Med 1998;129:761-9), although not in others. NICE also argue that, because simple tooth brushing can also cause transient bacteraemia, there is no logical basis for giving antibiotic cover for dental treatment. However, common sense dictates that procedures such as dental extraction and vigorous scaling are likely to introduce very much larger numbers of bacteria into the blood stream than simple tooth brushing.

Despite acknowledging the fact that prosthetic valve patients are at much higher risk than patients with native valve lesions, NICE accords them no separate consideration or risk stratification in making recommendations. They claim that PVE is a rare condition, that on economic grounds antibiotic prophylaxis for endocarditis in general is not cost-effective and that the number of deaths from anaphylactic reactions to antibiotics could possibly exceed those from PVE.

In fact, PVE is not rare. A large compilation of 127 series from the prosthetic valve literature shows that the incidence varies from 0%/year in some small series to 1.74%/year with an average of 0.4%/year for mechanical valves and 0.6%/year for bioprostheses (Curr Probl Cardiol 2000;25:73-156). Risk factors for PVE have been identified as previous endocarditis, diabetes, renal failure, poor functional class (i.e. very symptomatic patients or patients in heart failure), older age and double valve replacement. The risk is also higher in the first 6 months after implantation, before healing of the sewing ring of the valve has taken place, although many of these infections are probably acquired at the time of surgery. Importantly, the incidence of PVE reported in the literature is almost certainly based upon groups of patients who have been advised to take antibiotic prophylaxis for dental treatment and other interventional procedures.

Without antibiotic prophylaxis, it is likely that the incidence would be higher. NICE bases assumptions about fatal anaphylactic reactions to penicillin on 1960's general population data from the World Health Organisation. It is extremely unlikely that patients with prosthetic heart valves would never have encountered penicillin previously and be unaware of penicillin allergy. Many, perhaps most, will have taken antibiotic prophylaxis on numerous occasions already, without any adverse reaction.In making their recommendations, NICE rely heavily on a 35-page analysis, performed by health economists, based on numerous ill-supported modelling assumptions and estimations, using highly selective citations from the medical literature. For example, they determine the risk of PVE as being 93 per million (per dental procedure) based on one French article (Clin Infect Dis 2006;42:e102-7) which extrapolated risk to the whole French adult population based on a survey of 2,805 adults which included 24 patients with prosthetic valves, 4 of whom had undergone an 'at-risk dental procedure', unprotected by antibiotics in 2 cases. In their analysis model, NICE's health economists also incorrectly assume that patients will be in functional class I or II (i.e. minimally symptomatic) following valve replacement for both native and prosthetic endocarditis based on two articles, neither of which relate to endocarditis; one describes longterm results following mitral valve annuloplasty (a repair procedure, not valve replacement); the other describes longterm results with the Carpentier Edwards porcine bioprosthesis. Neither article specifically addresses the issue of outcome after valve replacement for endocarditis. The methodology of the health economists' analysis thus does not withstand scientific scrutiny and should be regarded with great scepticism.

All other current guidelines, both European and American, whilst acknowledging the lack of definite proof of the effectiveness of antibiotic prophylaxis, recommend antibiotic cover for dental and other interventional procedures in patients with prosthetic valves. Even the most recent, rather sceptical, guidelines from the American Heart Association, published in October 2007 (Circulation 2007;116:1736-54) continue to recommend antibiotic prophylaxis for prosthetic valve patients undergoing "all dental procedures which involve manipulation of gingival tissue or the periapical region of teeth or perforation of the oral mucosa". NICE are thus seriously out of step with other expert opinion. Of the 10 professional members of the NICE Guidelines Development Group, only 4 had any literature publications relating to endocarditis, according to a MEDLINE literature search. The Guideline Development Group responsible for the recommendations consisted of 4 dentists/oral surgeons (including the chairman), 2 cardiologists, 2 microbiologists, 1 pharmacist, 2 patient representatives, but only 1 cardiothoracic surgeon.

PVE has devastating consequences for the individual patient and, apart from pathologists conducting autopsies, only cardiac surgeons witness the extent of destruction of cardiac tissues it causes, with dehiscence of the prosthesis, peri-annular abscesses, spread of infection to other valves, intracardiac fistulae, damage to conducting tissue, etc., together with the complicated and prolonged postoperative course that patients often have to endure. If they escape mortality and recover from complications, they remain at risk of recurrent infection and/or paravalvular leak (leak of blood around the prosthesis) and the necessity for further high risk surgery. Few regain their previous functional status and quality of life.

Absence of definite proof of the effectiveness of antibiotic prophylaxis in preventing PVE does not constitute proof of ineffectiveness. In a situation of inconclusive evidence, patient safety should be given greater priority than economic considerations. Whilst implementation of the NICE guidelines will undoubtedly save money for the NHS, it will almost certainly result in a higher incidence of endocarditis and many more deaths from PVE. Even if only a proportion of patients avoid PVE with antibiotic prophylaxis, it must surely be justified on clinical and ethical grounds.

On the first page of their document, NICE state that their "guidance does not override the individual responsibility of healthcare professionals to make decisions appropriate to the circumstances of the individual patient, in consultation with the patient and/or guardian or carer". NICE therefore clearly distances itself from any damaging medico-legal responsibility consequent upon their published guidelines, and medico-legal accountability thus remains with individual healthcare professionals who will need to be able to defend their professional opinions and advice to patients. In a survey of members of the Society for Cardiothoracic Surgery of Great Britain and Ireland, 91% of those who responded were opposed to the NICE guidelines in respect of prosthetic valve patients. Individual patients should therefore discuss their concerns about antibiotic prophylaxis not only with their general practitioner, who may be unaware of the controversy surrounding the NICE guidelines, but also with their cardiac surgeon, in the knowledge that most cardiac surgeons in the UK are opposed to the NICE guidelines.

Eric G. Butchart FRCS, is the recently retired Senior Consultant Cardiothoracic Surgeon at the University Hospital of Wales in Cardiff. He chaired a European Society of Cardiology (ESC) committee which published recommendations for the management of patients after heart valve surgery in 2005 and was a member of an ESC Task Force creating guidelines for the management of valve disease in general, published in 2007.


Amoxicillin. NMIHI. Accessed at on December 11, 2018.

Ampicillin. NMIHI. Accessed at on September 12, 2018.

Antibiotics. NMIHI. Accessed at on September 12, 2018.

Azithromycin. NMIHI. Accessed at

Bacterial endocarditis. NMIHI. Accessed at on October 28, 2018.

Endocarditis. MedlinePlus. Accessed at on September 12, 2018.

Endocarditis. Symptoms and causes. MFMER. Accessed at on September 12, 2018.

Metronidazole. NMIHI. Accessed at on September 12, 2018.

Symptoms of bacterial endocarditis. American Academy of Family Physicians Accessed at on June 22, 2018.

What Is Infective Endocarditis? American Heart Association Accessed at on September 12, 2018.