In the Wake of Guidelines, Antibiotic Prophylaxis for Infective Endocarditis Declined for Moderate-Risk Patients

Following a 2007 guideline revision, prophylaxis is mostly—but not entirely—confined to high-risk groups.

In the Wake of Guidelines, Antibiotic Prophylaxis for Infective Endocarditis Declined for Moderate Risk Patients

Revised guidelines from the American Heart Association (AHA) that advised against antibiotic prophylaxis in all but patients at highest risk for infective endocarditis were swiftly implemented into practice, a new study suggests. But its authors assert that the same guidance cannot be blamed for a more recent rise in endocarditis-associated hospitalizations.

After 50 years of recommending prophylaxis for a variety of patients scheduled to undergo dental or respiratory procedures, the 2007 revision by the AHA concluded that this broad-sweep approach, which never had any hard evidence to support it, is only reasonable for those with underlying cardiac conditions who have the highest risk of adverse outcomes from infective endocarditis.

“There are no randomized clinical trials at all in this area,” noted Pallav Garg, MBBS, MSc (Western University, London, Canada), lead author of the new study, which was published online ahead of print in Circulation. “In fact, there is very little data to support the use of antibiotic prophylaxis for the prevention of infective endocarditis” regardless of patient baseline risk, he added to TCTMD.

Jawahar L. Mehta, MD, PhD (University of Arkansas for Medical Sciences, Little Rock), who commented on the study for TCTMD, agreed, adding that many physicians and dentists likely think they are erring on the side of caution in prescribing prophylaxis, even in the absence of elevated risk.

“People say it’s just one pill, what the big deal? It seems easy enough to do,” Mehta said. “But I think we need a trial in high-risk patients to at least have some data.”

Revision Had Intended Effect

To measure the impact of the guideline revision on the incidence of infective endocarditis, Garg and colleagues looked at rates of prophylaxis prescription and endocarditis-related hospitalizations in Canada over a 13-year span that included the periods before and after the revision was implemented.

“What we found was that the 2007 American Heart Association guidelines revision had their intended effect,” Garg said. “The overall rates of antibiotic prophylaxis dropped quite dramatically and quite early after the guideline was put out, particularly in the population at moderate risk, which was the target of the guideline revision.”

Among moderate-risk patients, which included those with acquired valvular heart disease, hypertrophic cardiomyopathy and other congenital cardiac malformations, there was a decrease in the mean quarterly rate of prophylaxis prescriptions from 30,680 to 17,954 per million population in the periods before and after the guideline release (P = 0.0004). In contrast, only a minimal change in prescribing was seen in the high-risk group, which included those with previous infectious endocarditis, prosthetic valve replacement or prosthetic material used in valve repair and certain forms of congenital heart disease.

Of note, however, across all risk groups, an increase was observed in the rate of new infective endocarditis hospitalizations beginning in 2010. The incidence ranged from 336 to 1,915 per million population in the highest-risk group and from 180 to 440 per million population in the moderate-risk group.

Although it can take years from the time a guideline change is introduced until it becomes ingrained in practice, Garg said the existing evidence suggests that the recommendation to limit prophylaxis happened swiftly, with a quick drop in the rate of prescriptions for moderate-risk patients occurring in the first two quarters after the guideline revision was released. That evidence supports Garg and colleagues’ contention that the increase in hospitalizations for infective endocarditis that happened in subsequent years was unrelated to the discontinuation of prophylaxis in moderate-risk patients that began almost immediately after the 2007 guideline revision.

As for what is behind the increase in those hospitalizations, Garg and colleagues say it remains unclear but may be partially explained by documented rises in comorbidities and intravenous drug use.

Conflicting Results

Previous efforts to evaluate the impact of the guideline revision have yielded conflicting results. Data from the Nationwide Inpatient Sample demonstrated an increase in streptococcal, but not staphylococcal, infectious endocarditis and no change in hospitalization rates. But Garg and colleagues saw no increase in streptococcal infections, which they say provides further support for the guideline change because the primary purpose of the prophylaxis is to prevent strep infections. Staphylococcus aureus and streptococcal species were seen in 30.3% and 26.4% of patients, respectively. Another 10.5% of the cases involved some other staph species.

“If prophylaxis did work and we gave less of it, then we should see an increase of strep infections over time,” Garg observed. “Our study showed there was no change in the proportion of strep infections over time in the older age group and in the younger age group, there was a decrease. So both of these provide further evidence that there is no link between the decrease in prophylaxis [beginning in 2007] and this late increase in endocarditis after 2010.”

Garg also pointed out that while the drop in prophylaxis prescribing in the moderate-risk group is encouraging, there has not been a full stop as would be expected if everyone was in compliance with the guideline revision.

“The AHA is saying don’t prescribe it in the moderate-risk group, but there were still 20,000 prescriptions per million population being given between 2008 and 2014,” he noted.

Mehta said ongoing education on this subject at meetings, in journals, and across specialties is probably needed. “This message has been around for a long time, but it has to be continued so that that [physicians and dentists] understand the individual level of risk of the patient,” he said.

  • Garg and Mehta report no relevant conflicts of interest.