Concerning Decline in Antibiotic Prophylaxis for Infective Endocarditis
IE incidence had been declining prior to updated guidelines published in 2007. Now some eligible patients are not getting the drugs they need.
After the American Heart Association (AHA) released more restrictive recommendations regarding antibiotic prophylaxis for infective endocarditis (IE) in 2007, there was a concerning decline in prescriptions for high-risk patients undergoing dental procedures, who remained eligible under the new guidance, administrative data show.
At the same time, a trend toward declining IE incidence slowed in this group, researchers led by Martin Thornhill, MBBS, BDS, PhD (University of Sheffield School of Clinical Dentistry, England), report in a study published in the Journal of the American College of Cardiology. The findings also were presented last week at the AHA 2018 Scientific Sessions in Chicago, IL.
In moderate-risk patients, there was a more appropriate marked drop in prescription of antibiotic prophylaxis after the new recommendations were released, accompanied by a more modest slowing of the decline in IE incidence.
Although the observational study does not establish a causal relationship between prescribing patterns and incidence trends, Thornhill told TCTMD in an email that the findings are supportive of the AHA’s guidance, which recommends using antibiotic prophylaxis in people with a high risk of IE and its complications who are undergoing invasive dental procedures, but not in those with a moderate risk.
The findings suggest, however, “that further measures may be desirable to ensure that all those at high risk of infective endocarditis who are undergoing invasive dental procedures are offered antibiotic prophylaxis,” Thornhill said. “The small increase in endocarditis in those at moderate risk [compared with the expected decline] suggests that the benefits of antibiotic prophylaxis may be insufficient to warrant recommending it or that some of those at moderate risk might benefit whilst others would not. More research on those at moderate risk may be needed to determine which is the case.”
Commenting for TCTMD, Robert Bonow, MD (Northwestern University Feinberg School of Medicine, Chicago, IL), said “the conclusion that this requires further investigation is probably the best way to put it,” pointing out there’s no true evidence of a causal relationship between prescribing practices and IE incidence.
What this study adds is more information on what types of patients are getting antibiotic prophylaxis, but the data are not detailed enough to make firm conclusions about its effectiveness in preventing IE, said Bonow, who was one of the authors of the 2007 AHA guidance.
In particular, he said, there’s no information on what organisms were causing the cases of endocarditis. Prophylaxis is administered before dental procedures to prevent endocarditis caused by oral bacteria, so if endocarditis is caused by something else, prophylaxis would not be of benefit. Many people placed in the high-risk group had prosthetic heart valves, Bonow noted, and most cases of endocarditis that develop in this population are not caused by oral bacteria.
Trying to link together endocarditis and an invasive dental procedure in high-risk patients is made difficult, too, by the fact that many of these patients are older and have other diseases—like diabetes or kidney disease—that make them susceptible to infections. It’s unclear whether receiving one dose of oral antibiotics at the dentist would make an impact, Bonow said.
“You’re giving antibiotics for the 30 minutes you’re in a dental chair and you’re not giving antibiotics for the hundreds of hours when you get bacteremia brushing your teeth and flossing,” he said. “So even if you had an oral organism it’s very difficult to tie that to the single visit to the dentist and whether or not you took an antibiotic.”
Asked whether the 2007 recommendations are still sound, Bonow said they’re “about as good as you can make them. And the reason for recommending antibiotics in the high-risk group isn’t so much to prevent the infection per se, although you want to do that, it’s because you’re identifying the high-risk group where the consequences of having endocarditis are a problem.”
Prescriptions Fell Across Risk Groups
The AHA first recommended antibiotic prophylaxis for IE before invasive medical and dental procedures in 1955, but the effectiveness of the approach has never been proven. That lack of evidence, along with concerns about adverse effects and the development of antibiotic resistance, led to the more restrictive 2007 AHA recommendations. European bodies followed with similar guidance shortly thereafter.
Prior studies have examined the impact of the 2007 recommendations on IE incidence, with mixed results, but none have included data on prophylaxis prescribing or stratification by risk group at the same time, Thornhill said.
To overcome those limitations, the investigators dug into administrative data on people with Medicare or commercial health insurance included in Truven Health MarketScan databases. The study covered May 2003 to August 2015, with a prerecommendation period (May 2003 to April 2007), a transition period (May 2007 to October 2008), and a postrecommendation period (November 2008 to August 2015).
The analysis included 198.5 million enrollee-years of data, with 0.64%, 5.91%, and 93.45% in people considered high, moderate, and unknown/low risk, respectively.
IE incidence and antibiotic prophylaxis prescriptions were on the decline before 2007, possibly because of increasing diagnostic specificity for IE and because of a growing awareness of views that many patients previously recommended for prophylaxis did not need it, Thornhill et al say. The dearth of evidence supporting prophylaxis and concerns about the downsides of using it also may have been playing a role, they add.
Even taking those pre-existing trends into account, the researchers estimated that prophylaxis prescriptions dropped in each risk group following the 2007 recommendations: by 52% in people with unknown/low risk, by 64% in those with moderate risk, and by 20% in those with high risk.
IE incidence continued to decline across risk groups, but the rate of decline changed after the recommendations were released. Compared with what was expected based on pre-existing trends, there was no significant increase in IE incidence in the group with unknown/low risk, but 75% and 177% increases in incidence in the moderate- and high-risk groups.
“What we found was that, despite the recommendation to continue prophylaxis for those at high risk, it fell significantly in this group, and in parallel there was a highly significant increase in endocarditis incidence,” Thornhill said. “While in those at moderate risk, despite a much larger fall in antibiotic prophylaxis prescribing, the increase in endocarditis incidence was small and barely reached significance. Together, these findings support the AHA recommendation to give antibiotic prophylaxis to those at high risk.”
Difficulty Distinguishing Risk
The results also suggest that determining risk of IE may be difficult for clinicians, especially dentists, Thornhill added.
“This could be the result of a number of educational or service delivery issues, and it is important that the guidance distinguishing these groups is as clear as possible and that clinicians are properly educated about these issues,” Thornhill said.
“However, it also needs to be borne in mind that cardiological factors distinguish these groups, yet it is dentists who ultimately decide when antibiotic prophylaxis is required, and they are not specialists in heart disease,” he continued. “Obviously, good communication between cardiologists and dentists is important, therefore, in determining which individuals should or should not be considered for antibiotic prophylaxis. Perhaps those at high risk should be educated about their risk status and be provided with a card that they can use to identify that they are at high risk when attending a dentist.”
Bonow also underscored the importance of identifying higher-risk individuals. “They should be receiving not only antibiotics but also very good oral hygiene and dental care and all the other things we do to prevent infections in hospitals,” he said.
In an accompanying editorial, Ann Bolger, MD (University of California, San Francisco), says if “taken at face value, these results suggest that this failure to comply with the antibiotic prophylaxis guidelines may have contributed to potentially avoidable cases of endocarditis in the patients most at risk.
“If these two trends in antibiotic prophylaxis and IE were proven to be linked in a cause-and-effect manner,” she continues, “these results would strike both fear and hope into the hearts of dedicated IE worriers—fear that some patients at high risk have been inadequately protected from IE due to lack of compliance with antibiotic prophylaxis recommendations, and hope that there might ﬁnally be some indication that antibiotic prophylaxis is actually effective in avoiding some cases of endocarditis.”
However, Bolger points out—like Bonow—that the lack of information on causative organisms limits the establishment of a causal relationships.
So for now, “there is no respite from worrying about preventing IE,” she says.
“For our patients’ sakes, thoughtful risk assessment, clear communication of that risk, and commonsense guidance on oral health and avoiding preventable bacteremia should be the core of our prevention strategy,” she says. “Even when antibiotic prophylaxis is recommended, counseling about oral hygiene, responding to unexplained fever by presenting for evaluation, consideration of blood cultures before initiating antibiotic use, and vigilance regarding skin breakdown or vascular entry are critically important.”
Todd Neale is the Associate News Editor for TCTMD and a Senior Medical Journalist. He got his start in journalism at …
Thornhill MH, Gibson TB, Cutler E, et al. Antibiotic prophylaxis and incidence of endocarditis before and after the 2007 AHA recommendations. J Am Coll Cardiol. 2018;72:2543-2554.
Bolger AF. Preventing endocarditis: no rest for the worrier. J Am Coll Cardiol. 2018;72:2555-2556.
- The study was funded by a research grant from Delta Dental of Michigan and its Research and Data Institute.
- Thornhill reports receiving support from the Delta Dental Research and Data Institute.
- Bolger reports no relevant conflicts of interest.