The practice of routinely prescribing antibiotic prophylaxis prior to dentistry for the prevention of serious infections has undergone a shift in recent years due to lack of convincing evidence of benefit, balanced with the known risks of antibiotic overuse. The transition away from prophylaxis for patients with orthopedic implants in the United States is a good example. In 2008 and 2012 respectively, the United Kingdom and Sweden adopted policy recommendations for cessation of prescribing routine dental antibiotic prophylaxis for patients at high risk of infectious endocarditis. Professional organizations in the U.S., including the American Heart Association, still recommend prophylaxis before certain dental procedures for all patients considered to be at high risk of infectious endocarditis.

A recent before and after study utilized the Swedish national registry to compare incidence of infectious endocarditis in nearly 77,000 adults at high risk (based on diagnosis codes of prior endocarditis, prosthetic heart valve, or cyanotic congenital heart disease) with almost 400,000 adults at low risk of infectious endocarditis (but who all had diabetes) from 2008-2018, with 2012 marking the beginning of the recommendation to stop prescribing routine dental prophylaxis. The high-risk group was stratified into prevalent and incident cohorts to avoid survivorship bias. The decision to stratify was based on data suggesting relatively high mortality in the first few years after hospitalization for infectious endocarditis or after heart valve replacement, which could lead to lower incidence of infectious endocarditis in those at high risk who had survived until the start of the study period in 2008. Those who were already considered at high risk in 2008 were assigned to the prevalent cohort; those who became high risk after 2008 were assigned to the incident cohort. Patients were followed for no more than two years due to concern about falsely diluting the incidence after that period if people were to succumb to the illness in the first two years, which is statistically probable.

The results demonstrated no significant increase in incidence of infectious endocarditis among high-risk patients in the six years studied after the recommended cessation of routine dental prophylaxis in Sweden. There was also no difference found when prevalent and incident cohorts were analyzed separately, with adjusted hazard ratios for the prevalent cohort of 13.6 (95% CI 11.2-16.4) before and 11 (95% CI 9-13.5) after and for the incident cohort of 20.9 (95% CI 17-25.6) before and 23.9 (95% CI 19.4-29.3) after. Amoxicillin prescribing by dentists (the main prescribers of prophylaxis in Sweden) fell by 40 percent in the six years following the recommendation to stop.

The obvious assumption to be drawn here is that routine dental prophylaxis in people at high risk of infectious endocarditis doesn’t make a difference; either it’s not necessary or prophylaxis is ineffective. If this assumption is true, this would represent a clear victory for antibiotic stewardship. However, there are uncertainties in this study, such as who did and did not get antibiotics, as the distribution of the 40 percent reduction in prescribing is unknown, which may (or may not) have affected the incidence in some subpopulations. Additionally, the choice to use patients with diabetes as a low-risk control is curious and invites a level of potential confounding, considering that patients with diabetes are probably at higher risk of infections such as endocarditis than a similar group without diabetes. Are these flaws enough to say Sweden should go back to recommending prophylaxis? No. The Swedish position favors inaction (versus action) when data is lacking, which falls very much in line with the principles of reducing overuse, less is more, first do no harm, etc. But, given these persistent uncertainties about a practice that affects a large number of people across the globe, a high-quality randomized trial may well be justified to answer this question definitively.

For more information, see the topic Endocarditis Prophylaxis in DynaMed®.