Article Text

Download PDFPDF

The new American Heart Association guidelines on the prevention of infective endocarditis: culmination of a long process of thought
  1. Xavier Duval1,2,
  2. Catherine Leport2,
  3. Nicolas Danchin3
  1. 1
    Université Paris 7, Denis Diderot, Paris, France; AP-HP Hôpital Bichat Claude Bernard, Paris, France; Inserm, CIC 007, Paris, France
  2. 2
    Université Paris 7, Denis Diderot, Paris, France; AP-HP Hôpital Bichat Claude Bernard, Paris, France; Laboratoire de pathologie Infectieuse, Paris, France
  3. 3
    Université Paris-Descartes, Faculté de Médecine; AP-HP Hôpital Européen Georges Pompidou, Paris, France
  1. Dr Xavier Duval, Service des Maladies Infectieuses et Tropicales, Hôpital Bichat Claude Bernard, Université Paris VII, 46 rue Henri Huchard, 75877 Paris Cedex 18 France; xavier.duval{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Despite recent advances in its diagnosis and treatment, infective endocarditis (IE) remains a disease with a high mortality rate, with an overall in-hospital mortality rate of 20% and a 5-year mortality rate of 40%, reported in recent large studies.1 Every effort should be made to reduce the incidence of a disease which carries such a heavy burden of morbidity and mortality. Although its efficacy has not been demonstrated in humans, antibiotic prophylaxis of IE has been recommended for subjects with predisposing cardiac conditions (PCC) since 1955.2 During the last decade, however, several factors have challenged the principles underlying these recommendations.3

The American Heart Association (AHA) has just modified its recommendations on IE prevention.4 The new recommendations represent a radical change from the previous ones: antibiotic prophylaxis using a 2 g single oral dose of amoxicillin is no longer recommended prior to dental procedure except for patients with the highest risk of adverse outcome resulting from IE (ie, previous IE, prosthetic cardiac valve, congenital heart disease with persistent risk of IE, cardiac transplantation recipients with cardiac valve diseases) and undergoing “any dental procedure that involves manipulation of the oral mucosa”. Prophylaxis is thus abandoned in patients at intermediate risk of IE, such as those with a history of aortic regurgitation or mitral regurgitation who undergo a dental procedure. It is also abandoned in all patients prior to extraoral procedures (eg, colonoscopy), regardless of their predisposing cardiac conditions.

This drastic change is supported by several considerations. Epidemiological data show that IE following dental procedures is rare, affecting fewer than 5% of patients in a large series of IE patients collected recently,5 6 reflecting a very low risk of developing IE after a dental procedure.7 Conversely, it has been suggested that transient repeated bacteraemia from everyday life activities (tooth brushing, chewing …) might more often be responsible for IE than intermittent bacteraemia following occasional procedures.8 There is no definite proof of the efficacy of antibiotic prophylaxis in the absence of randomised trials. Different case–control studies failed to identify a clear relationship between the onset of IE and preceding “at-risk” dental procedures.5 9 10 However, scaling was associated with a significant risk of viridans streptococci IE after multivariate analysis in Lacassin et al’s study. Widespread use of antibiotics before dental procedures is likely to cause iatrogenic accidents and may contribute to the selection pressure of more resistant bacteria.11 There is no validated surrogate marker of the risk of developing IE after an invasive procedure. Some procedures for which no prophylaxis is indicated appeared to be more likely to induce bacteraemia than others for which it is indicated.12 Finally, Streptococcus viridans IE, the most frequent IE due to an oral microorganism, is associated with the best prognosis when occurring on native valve as compared with other microorganisms.1

The new US guidelines represent the culmination to date of a progressive change that was initiated by the 2002 French recommendations.13 14 The French recommendations, supported by the infectious diseases, cardiologists’ and dentists’ societies, first proposed that antibiotic prophylaxis should be optional in patients at intermediate cardiac risk undergoing dental procedure. The choice left to the clinicians by the French recommendations was deliberate, as the expert group felt that there remained a true uncertainty as to the efficacy (or lack thereof) of antibiotic prophylaxis before dental procedures but that there would be a true problem of credibility and applicability for physicians who had hitherto strongly recommended antibiotic prophylaxis to their patients with valvular regurgitation, if they suddenly informed them that antibiotic prophylaxis would from now on be useless. Such a volte-face might remind the population of the time when paediatricians lost much of their credibility through a (literal!) 180° change in their recommendations on the best way to place newborns in their cradles. However, the process of a fundamental revision in the recommendations was launched, to be subsequently followed in 2006 by the British Society for Antimicrobial Chemotherapy (BSAC) recommendations and now by the AHA guidelines.4 15

In the French guidelines, the decision to use prophylaxis in optional situations was based on patients’ background characteristics (chronic immunodepression, cancer, chronic severe diseases…) associated with a severe outcome from IE. Decision on prophylaxis became based on “the risk from IE” rather than on “the risk of IE”.13

The BSAC in its guidelines recommended prophylaxis after dental procedures only in patients at high risk of IE, whereas it was recommended in patients at high and intermediate risk of IE after extradental procedures. Furthermore, the BSAC has abandoned the selection of “at-risk” dental procedures to be covered by prophylaxis based on the rate of resulting bacteraemia, given the unproven validity of this surrogate marker. Instead, it has introduced the notion of “all dental procedures involving dento-gingival manipulation or endodontics”. This recommendation and those (diametrically opposed) previously proposed by the British and European cardiologists’ societies have led to vigorous exchanges between infectious diseases specialists, cardiologists and dentists.1621

Although the AHA recommendations may appear logical and appropriate, it must not be forgotten that (as with the previous ones) they are not based on sound scientific evidence. In fact, “absence of evidence” does not mean “evidence of absence”. Therefore, these new recommendations should, like the previous ones, be considered with a touch of scepticism. More importantly, they will need scientific appraisal through new epidemiological studies, to ascertain that they do not generate a substantial increase in the number of cases of IE following dental procedures, once antibiotic prophylaxis has been abandoned in the vast majority of cardiac patients. These recommendations may also draw attention to new issues, such as the growing responsibility of Staphylococcus species in the occurrence of IE, related to the increase in healthcare-associated infections, and the role of factors other than procedures associated with the risk or severity of IE.22

In summary, the new AHA guidelines may be considered an important and courageous step—we hope, forward. Follow-up measures, however, will be necessary to document their validity. In any event, this radical change from previous policies is there to remind us that scientists and physicians should always remain modest in their assertions and the recommendations flexible in their formulation: such a dramatic change in the guidelines provides evidence that the scientific truth of today may not be the truth of tomorrow.



  • Competing interests: None.