Statistics from Altmetric.com
Despite major advances in treating valvular heart disease, the in-hospital mortality (15–20%) and 1-year mortality (≃40%) for infective endocarditis (IE) has not improved even with modern antibiotics and surgical therapy. Further, stroke (17%), embolisation other than stroke (23%), heart failure (HF) (32%) and other complications remain common; therefore, all precautions to help prevent IE should be employed where indicated. In underdeveloped countries, IE is most often associated with rheumatic heart disease. In developed countries, IE is increasingly associated with prosthetic valves and intracardiac devices, with the risk of IE being 50 times higher in patients with a prosthetic valve compared with the general population. Other populations that are at higher risk of IE who might benefit from specific precautions or improved systems of care to prevent infection are intravenous drug users, the elderly and those with diabetes mellitus or with other forms of immunosuppression.
While the need to prevent bacteraemia in these high-risk populations is universally accepted, the efficacy of antibiotic prophylaxis when bacteraemia is likely to occur has been controversial over the years. Most of this controversy stems from the dogma of historical prophylaxis management, the knowledge that endocarditis has very high morbidity/mortality rates, as well as the misperception that antibiotic prophylaxis is harmless. The new AHA/ACC guidelines1 reflect the scarce evidence base for IE prophylaxis. There has been a paucity of data to support historically aggressive prophylaxis measures as evidenced by a Cochrane Review2 concluding that there was no evidence for benefit using penicillin to prevent endocarditis, as well as other data indicating lack of benefit from prophylaxis in non-dental procedures such as endoscopy/cystoscopy. Guidelines from the European Society of Cardiology (ESC) and from the Working Party of the British Society for Antimicrobial Chemotherapy harmonise with the new AHA/ACC recommendations and have been in place for a number of years. Multiple epidemiological studies have now been performed, which demonstrate that there has been no increase in the rate of IE since the adoption of the ESC guidelines recommending more restrictive use of IE prophylaxis. The National Institute for Health and Care Excellence (NICE, UK) guidelines have taken an even more radical departure from the previous prophylaxis standards in not recommending antibiotic prophylaxis for dental or non-dental procedures. Similarly, subsequent epidemiological studies performed in the wake of the NICE guideline revisions have demonstrated no increase in clinical cases or deaths from IE.
For patients with IE, the new ACC/AHA guidelines place a great deal of emphasis on the role of the heart valve team (HVT)3 in decision making regarding antibiotic therapy and the type and timing of surgical intervention. The IE valve team should include infectious disease specialists, cardiologists and a cardiac surgical team. The concept of a HVT is especially important in IE as making the diagnosis can be difficult and frequently is delayed, which places the patient at risk of progressive and potentially irreparable structural damage to the heart and other organ systems secondary to vascular-embolic and immunologically mediated events. Guidelines recommend an integrated diagnostic approach based on blood culture results, echocardiographic findings and use of validated diagnostic criteria; however, early diagnosis requires that primary care physicians be alert to the possibility of IE in febrile patients with risk factors. Appropriate antibiotic therapy is essential for effective treatment of IE with the choice and duration of specific medications determined by infectious disease specialists based on the antibiotic sensitivities of the causative organism.
In addition to appropriate antibiotic therapy, surgical intervention is needed in about 50% of patients with IE to eradicate infection or treat complications due to valve destruction (table 1).4 However, decisions about the type and timing of surgery in these patients can be quite challenging, requiring a multispecialty approach to ensure the best outcomes. A risk-scoring system using the Society for Thoracic Surgery database has been developed to predict risk of surgery in patients with IE to help better counsel patients and more objectively define risks of surgery.5 In the 2014 AHA/ACC guidelines, we have been diligent to adhere to the current body of evidence available to us in terms of indications for surgery and timing of operative intervention. Similarities and differences in recommendations for surgical intervention between the ACC/AHA and European Society of Cardiology (ESC) guidelines6 published in 2009 will be compared and contrasted in the remainder of this editorial. In general, both the ACC/AHA and the ESC recommendations apply to both native and prosthetic left-sided valve endocarditis with a few additional recommendations that apply specifically to prosthetic valves or intracardiac devices.
Patients with IE in which there is no debate as to the need for early surgical management are those with valve dysfunction resulting in HF or uncontrolled infection resulting in heart block, abscess or a destructive penetrating lesion. The International Collaboration on Endocarditis-Prospective Cohort Study (ICE-PCS)7 reported an in-hospital mortality rate in 21% patients with IE with HF treated with surgery versus 45% in those who were medically treated. In complicated left-heart native valve endocarditis, baseline features independently associated with 6-month mortality were (1) abnormal mental status, (2) moderate-to-severe HF, (3) bacterial aetiology other than Viridans streptococci and (4) medical therapy without valve surgery. This risk stratification has been validated and reproduced in both retrospective propensity studies and prospective studies.
Despite the uniformity in recommendations between the AHA/ACC guidelines and the ESC guidelines in these patient situations, the definitions in timing of surgery vary slightly. In the ACC/AHA guidelines, ‘early surgery’ is defined as occurring during the initial hospitalisation before the completion of a full therapeutic course of antibiotics, whereas the ESC guideline breaks surgical timing down into more specific categories of ‘emergency’ (performed within a few hours), ‘urgent’ (within a few days) or ‘elective’ (after at least 1 or 2 weeks of antibiotic therapy). (table 1) While there are patients who can be triaged into these specific time frames, the definitions used in the majority of the published surgical studies were less precise. Future studies focused upon the exact timing of surgical intervention will be required to address this issue more definitively. The new ACC/AHA guidelines also propose a more liberal interpretation of the degree of IE-induced HF required to trigger an ‘early’ surgical management approach based upon definitions used in the ICE-PCS study and other more recent studies. Specifically, the definition in these studies was based upon the mere presence of cardiac valvular or paravalvular-related pulmonary oedema as opposed to the requirement that a patient have medically ‘refractory or persistent’ oedema or shock. This is a subtle but very important clinical distinction; clear evidence of HF, regardless of severity or response to medical therapy, is an indication for surgical intervention.
The other minor, but potentially clinically significant, difference in surgical recommendations between the AHA/ACC and ESC documents relates to the emphasis placed upon surgical management in patients with large (>10 cm) vegetations. While both documents highlight the increased risk of embolic events, the ESC guideline recommends ‘urgent’ surgery (class I indication if one or more embolic episodes occur despite appropriate antibiotic therapy and class IIb if the vegetation is >15 mm), whereas the AHA/ACC considers that ‘early’ surgery is only a class IIb (‘may be considered’) recommendation in these patients. While Kang et al8 have reported a small randomised controlled clinical trial in this population showing a statistically significant reduction in embolic events (0% in early surgery group vs 21% in conventionally treated group), there was no significant difference in all-cause mortality. However, the conventionally treated group in this study did eventually require surgery for HF or other complications in 77% of cases so the recommendation for early intervention may be strengthened depending on the results of forthcoming larger trials.
Compared with native valve endocarditis, prosthetic valve endocarditis (PVE) is associated with a higher mortality as well as higher post-treatment HF-related disability. Staphylococcal PVE has been associated with a mortality rate as high as 70%. Given the difficulty in eradicating Staphylococcus spp. when prosthetic or avascular material is involved in the infection, surgical debridement with removal of the infected valve is associated with higher survival rates than medical therapy alone. Thus, in addition to all of the recommendations that apply to both native and prosthetic IE, early surgery is recommended in the patient with a prosthetic valve and relapsing infection—defined as recurrence of bacteraemia after a complete course of appropriate antibiotics with initially negative blood cultures. The AHA/ACC and ESC are equally aggressive in recommendations for the management of PVE and for patients with cardiac device-related IE. Pacing system removal is a very important management consideration in these patients. Among patients with cardiac device IE, the rates of both concomitant valve infection and mortality are high.9
Overall, the ACC/AHA and ESC IE guidelines are concordant and exhibit only subtly nuanced differences. While IE antibiotic prophylaxis remains controversial as to its efficacy, early surgical management of the IE patient with valve dysfunction resulting in HF is clearly lifesaving. Emphasis on a coordinated HVT will be able to fill in for the knowledge gaps that our current literature has not yet bridged.
Contributors Both authors contributed to this editorial.
Competing interests None.
Provenance and peer review Commissioned; internally peer reviewed.
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.