For decades, antibiotic prophylaxis was recommended for patients at high risk of IE undergoing specific IPs. Due to the lack of clear evidence supporting its use, recommendations for antibiotic prophylaxis ceased in the mid-2000s, except for high-risk patients undergoing invasive dental procedures (IDPs). This study aims to quantify the association between IPs and IE incidence.
All IE admissions in England from April 2010 to March 2016 were identified through the national admissions register (n = 14731). All IPs performed within the 15 months prior to admission were then identified. To determine whether the odds of developing IE increased within 3 months post certain IPs, the incidence of IPs in the 3 months immediately before IE admission (case period) was compared with the incidence in the preceding 12 months (control period). The odds of IE increased following permanent pacemaker and defibrillator implantation (OR = 1.54; p < .001), dental extraction (OR = 2.14; p = .047), upper (OR = 1.58; p < .001) and lower (OR = 1.66; p < .001) gastrointestinal endoscopy, bone marrow biopsy (OR = 1.76; p = .039), bronchoscopy (OR = 1.33; p = .049), and blood product transfusions (OR = 1.2; p = .012).
The study’s conclusion indicates a clear association between specific IPs (permanent pacemaker and defibrillator implantation, dental extraction, gastrointestinal endoscopy, and bronchoscopy) and subsequent development of IE. These findings should prompt a necessary reevaluation of current IE prophylaxis recommendations for high-risk individuals.
COMMENTARY:
In developed countries, several factors are contributing to the rising IE incidence observed in the past decade. These include an aging population, increased use of implantable cardiac devices (ICDs), the advent of transcatheter prostheses, more patients on hemodialysis, improved IE detection due to advancements in diagnostic techniques (especially echocardiography), and the guideline change in the early 2000s that discontinued antibiotic prophylaxis for IE in diagnostic-therapeutic invasive procedures that compromise epithelial or mucosal integrity or pose a risk of bacteremia. Pre-2007 and 2009 AHA and ESC guidelines recommended antibiotic prophylaxis for patients at moderate to high risk of IE undergoing invasive procedures. However, due to the lack of evidence at that time linking these procedures with IE (except for dental procedures affecting the gingiva or the pulp and root), and concerns about adverse effects and antibiotic resistance, the AHA and ESC restricted prophylaxis to high-risk patients undergoing IDPs (e.g., those with previous IE, valve prostheses, including transcatheter valves, or unrepaired cyanotic congenital heart defects). Despite studies suggesting a possible rise in IE incidence, the 2015 ESC/EACTS guideline recommendations remained unchanged due to conflicting study results. The 2021 ESC/EACTS guideline on valve disease diagnosis and treatment also did not modify the IE prophylaxis recommendations.
To date, the strongest evidence for an association between IE and IPs comes from a Swedish study conducted in 2018 by Janszky et al. Interestingly, the findings of this study are nearly consistent with those in the current study led by Thornhill MH et al. Both studies found a clear relationship between IE development and certain IPs, such as recent ICD implantation, upper and lower GI endoscopy, and bronchoscopy. Notably, the current study also demonstrates a clear association between IE and IDPs performed within three months, a factor not analyzed in the Janszky et al. study.
Most healthcare systems in industrialized countries ensure specific antibiotic coverage for IDPs, ICD implantation, and bronchoscopy to prevent infections. High-risk patients undergo dental extractions per IE prophylaxis guidelines. Patients undergoing ICD implantation or bronchoscopy also receive antibiotic coverage according to procedure-specific guidelines, such as the 2015 British guideline on ICD infection prevention and the 2013 guideline on flexible bronchoscopy. However, these guidelines do not recommend antibiotic prophylaxis specifically to prevent IE, but rather to prevent procedural infectious complications (e.g., generator infection or pneumonia). Despite these three invasive procedures typically receiving antibiotic prophylaxis, this study showed a significant association with IE within three months post-procedure, suggesting possible incomplete prophylaxis, limited guideline adherence in clinical practice, or ineffective prevention protocols.
Furthermore, there was a strong association between upper and lower GI endoscopy and IE, as demonstrated in Janszky et al.’s Swedish study. However, this was not observed with endoscopic retrograde cholangiopancreatography, likely because these patients already receive intensive antibiotic prophylaxis to prevent cholangitis. It is concerning that routine GI endoscopy, unlike other invasive procedures studied, has no general or IE prophylactic antibiotic recommendation, making it a critical point for future guideline updates.
Regarding hematologic procedures, an association was found between bone marrow biopsy and blood product transfusions (including red blood cells and plasma), though these results must be interpreted cautiously, as these procedures are often performed under a misdiagnosis of hematologic malignancy during the initial diagnostic phase of IE. No association was observed between otorhinolaryngology, dermatology, obstetric procedures, or IE, though the number of these procedures was minimal. Similarly, no association was found with cystoscopy and prostate endoscopic procedures, potentially explained by the frequent use of antibiotics for postprocedural urinary infection prevention.
In conclusion, although this study was conducted using administrative databases with inherent limitations, its results are highly relevant. As in Janszky et al.’s study, a clear association was demonstrated between IPs and subsequent IE development within a short period. The accumulated evidence from these studies necessitates reconsideration of antibiotic prophylaxis for high-risk patients undergoing IPs such as ICD implantation, upper and lower GI endoscopy (notably, the only one lacking routine antibiotic coverage), bronchoscopy, and potentially any other procedure that could induce bacteremia.
REFERENCES:
Thornhill MH, Crum A, Campbell R, Stone T, Lee EC, Bradburn M, Fibisan V, Dayer M, Prendergast BD, Lockhart P, Baddour L, Nicoll J. Temporal association between invasive procedures and infective endocarditis. Heart. 2023 Jan 11;109(3):223-231. doi:10.1136/heartjnl-2022-321519
Janszky I, Gémes K, Ahnve S, Asgeirsson H, Möller J. Invasive procedures associated with the development of infective endocarditis. J Am Coll Cardiol. 2018 Jun 19;71(24):2744-2752. doi:10.1016/j.jacc.2018.03.532
