OVERVIEW: What every practitioner needs to know

Are you sure your patient has prosthetic valve endocarditis? What should you expect to find?

  • Prosthetic valve endocarditis (PVE) presents at various times after the implantation of the prosthetic valve. The clinical presentations, while generally similar to those of native valve endocarditis (NVE), can be significantly impacted by close temporal relationship to cardiac surgery.

    Early onset PVE refers to disease developing within 60 days of valve placement. When infection arises prior to discharge from the hospital after surgery the typical, often subtle, manifestations of endocarditis may be obscured by findings related to surgery itself or complications thereof. Typically, patients with early onset PVE have:

    Fever

    Systemic emboli

    overt systemic emboli occur in 40 percent of patients with PVE

    approximately half of the emboli lodging in the central nervous system causing infarcts or central nervous system (CNS) hemorrhage.

  • Invasion of infection along suture lines may result in paravalvular abscess and valve dehiscence

    diagnosis is suggested by new murmurs indicative of regurgitation through paravalvular leaks

    patients may develop congestive heart failure (CHF) due to valve dysfunction, primarily regurgitation

    occasionally, functional stenosis due to large vegetations that partially obstruct the valve orifice (primarily seen with PVE at the mitral site), may cause CHF.

    abscesses involving the valve annulus and beyond, a complication that is more common with valves in the aortic position, may not only cause persisting fever in spite of appropriate antimicrobial therapy but also may disrupt the cardiac conduction system resulting in various degrees of heart block.

  • Late onset PVE (occurring more than 60 days after operation – especially that occurring a year or more after valve surgery) manifest many of the same features as early onset PVE. Additionally, some patients may present in a more subtle fashion:

    Paravalvular extension may be less common. In late onset PVE caused by less virulent pathogens.

    The presentation may be more indolent, being more typical of classic subacute NVE

    If in this setting infection persists untreated for some weeks, typical so called “peripheral manifestations of endocarditis” may be noted; these include:

    Roth spots

    conjunctival petechiae

    subungual splinter hemorrhages

    Janeway lesions, and Osler’s nodes.

    Splenomegaly

  • PVE should be considered in any patient with a prosthetic valve and fever, and high among the diagnostic considerations when the fever is otherwise unexplained. Similarly, PVE should be considered when one encounters unexplained bacteremia, especially when the organism is one that commonly causes endocarditis. The findings on physical exam include:

    Fever

    Cardiac murmurs indicating valve dysfunction

    Splenomegaly

    Evidence of systemic, especially CNS, emboli

    Peripheral manifestations of endocarditis in skin, conjunctivae, nail beds, fundi

How did the patient develop prosthetic valve endocarditis? What was the primary source from which the infection spread?

  • Early onset PVE results from direct contamination of the valve or annulus during surgery or as a consequence of hematogenous seeding from an infection which has complicated convalescence from cardiac surgery. Late onset PVE occurring 12 months or more after valve surgery, like NVE, is a consequence of hematogenous seeding from either an overt infection or more commonly due to transient bacteremia arising from a normally colonized mucosal (oral or genito-urinary surfaces) or skin.

    The microbiology of PVE developing during these time windows is reflected in the pathogenesis of infection. Cases that present between 2 and 12 months after valve surgery are a blend of delayed onset cases where infection relates to the perioperative period and the earlier onset of hematogenously seeded infection from sites unrelated to surgery. The microbiology of these cases suggest the two pathogenetic mechanisms (Table I).

Which individuals are of greater risk of developing prosthetic valve endocarditis?

  • All patients with prosthetic heart valves are at risk for developing PVE. It appears that mechanical valves carry a slightly increased risk of early onset PVE and are associated with a greater frequency of paravalvular infection in contrast with bioprostheses where infection is somewhat delayed and often limited to the valve leaflets. By the end of the third year after valve placement with frequency, risk of PVE is similar regardless of valve site or specific prosthesis (mechanical or bioprosthesis).

  • Patients with extra cardiac infection complicating convalescence after valve surgery and patients with skin or mucosal conditions associated with bacteremia, including genitourinary tract instrumentation, are at increased risk of PVE.

    Therefore, dental, skin, and genitourinary tract diseases should be resolved before valve placement and manipulation of these areas after valve placement is an indication for periprocedure endocarditis prophylaxis.

    Intravenous catheter associated bacteremia or candidemia results in PVE in 15 to 20 percent of exposed patients.

    Other than valve surgery itself, none of the potential predisposing conditions, except for illicit intravenous drug abuse (IVDA) is sufficiently common to guide diagnostic work-up or empiric therapy. IVDA carries a very high risk for NVE and PVE. Although IVDA patients more commonly have Staphylococcus aureus (methicillin susceptible or resistant) endocarditis, they can have endocarditis caused by a broad range of microorganisms from Candida species to Pseudomonas aeruginosa.

    Nevertheless, any event likely associated with bacteremia may be a clue to subsequent PVE and the causative organism.

Beware: there are other diseases that can mimic prosthetic valve endocarditis:

  • In patients with prosthetic valves, any cause of fever elicits concern that the valve might be infected. Nevertheless, few entities will present with the entire endocarditis syndrome. For patients with prosthetic heart valves and thus predisposed to endocarditis, episodes of blood stream infection are in some sense a mimic of endocarditis. However, most of these bacteremias are short-lived with scant numbers of culture sets positive in contrast with PVE where almost every blood cultures obtained is positive. Additionally, while any microorganism can cause endocarditis, some are far more likely to do so (Table I).

  • Positive blood cultures for organisms that rarely cause NVE or PVE more likely reflect extracardiac infection. In all of these instances, evaluation of the patient for extracardiac infection as well as the entire endocarditis syndrome plus anatomic assessment of the heart valves with echocardiography can help resolve questions. Prosthetic valves may be thrombogenic thus causing systemic emboli and a clinical picture suggesting PVE. Similarly, valve dysfunction resulting from technical factors at surgery, if encountered in association with fever from extracardiac conditions, could mimic PVE. Negative blood cultures argue that these two scenarios do not indicate PVE because blood culture negative PVE in the absence of confounding antibiotics is unusual.

  • There are mimics of blood culture negative endocarditis, including:

    Atrial myxoma

    Marantic endocarditis

    Hypernephroma with fever and heart murmur due to increased cardiac output

    Carcinoid syndrome,

    Systemic lupus erythematosus with Libman-Sacks endocarditis

    Statistically these conditions are not likely to coexist with a prosthetic valve.

What laboratory studies should you order and what should you expect to find?

Results consistent with the diagnosis

  • When suspecting PVE, order at least 3 blood cultures (2 bottle sets) from individual venipuncture separated in time by an hour or more. These MUST be obtained before antibiotics are started. If the illness is indolent, I prefer to obtain blood cultures and to delay empiric therapy. Then, if the initial cultures are negative after 24 hours of incubation, repeat blood cultures can be obtained without the confounding of empirically initiated antibiotic therapy.

  • Anatomic confirmation of endocarditis should be pursued using transesophageal echocardiography (TEE). Regardless of valve type or position, TEE is far more sensitive (80-93%) than transthoracic echocardiography (15-55%) when seeking anatomic evidence of PVE. Generally, cardiac magnetic resonance imaging (MRI) has not been more sensitive than TEE when seeking anatomic evidence of PVE, although on occasion such studies may provide unique information.

  • Although not diagnostic, the following tests are usually obtained to establish a baseline status or to assess for possible complications: electrocardiogram, urine analysis, complete blood count, sedimentation rate, C-reactive protein, blood urea nitrogen, and creatinine. If symptoms suggest an end-organ complication, targeted imaging or other testing of that organ system may be indicated, e.g. focal neurologic symptoms: head computed tomography (CT) or MRI and lumbar puncture for cerebrospinal fluid (CSF) analysis; heart failure: chest X-ray; belly pain: abdominal imaging.

Results that confirm the diagnosis

  • Except for findings at surgery or post mortem examination, the diagnosis of PVE is made indirectly. The Duke Criteria for endocarditis have provided a useful schema for making a clinical diagnosis of PVE or NVE. Multiple positive blood cultures, especially for organisms frequently associated with endocarditis in combination with anatomic evidence provided by a TEE clinically confirm the diagnosis. Other tests that define focal complications which are strongly associated with endocarditis – e.g. imaging a mycotic aneursym or splenic infact – can help to clinically confirm PVE, especially if the blood culture and TEE results are not definitive. Similarly, the serologic tests for
    Coxiella burnetii – elevated phase I antibody – can be confirmatory. Most other tests are to define baseline organ function as one embarks on a prolonged course of antibiotic therapy or, like the erythrocyte sedimentation rate (ESR) or C-reactive protein, provide data that are supportive but not diagnostic.

  • Positive blood cultures not only help to confirm the diagnosis but also provide the causative organisms and thus allow laboratory determination of its antimicrobial susceptibility. This information is essential for design of optimal antibiotic therapy. Occasionally microbiologic and microscopic examination of an embolic vegetation will confirm the diagnosis. Similar studies of material excised at cardiac surgery, when valve replacement is required, will be diagnostic. The latter may be a last resort when blood cultures have been negative.

What consult service or services would be helpful for making the diagnosis and assisting with treatment?

  • The diagnosis and treatment of PVE typically requires subspecialty expertise, when feasible. Promptly consult infectious disease, cardiology, and cardiac surgery. In fact, in general, if these core services are not available the patient should be transferred to a facility that can provide the necessary expertise. Extra cardiac complications occasionally require the assistance of additional consultants: neurosurgery or neuroradiology to treat a cerebral mycotic aneurysm or general surgery for splenectomy to remove a splenic abscess that cannot be drained percutaneously by interventional radiology.

If you decide the patient has prosthetic valve endocarditis, what therapies should you initiate immediately?

  • The cornerstone of optimal treatment for PVE is effective antimicrobial therapy (using bactericidal agents – ones that kill rather than inhibit the causative pathogen) and where necessary cardiac surgery to correct pathology and valve dysfunction that will not resolve with antibiotic therapy.

  • To design an optimal antimicrobial regimen one must know the causative agent and its susceptibility to specific antimicrobial agents. This information flows from laboratory assessment of the organism(s) recovered from blood cultures or occasionally from direct culture of vegetations (platelet fibrin aggregates with enmeshed microorganisms) or paravalvular abscesses. There are no randomized comparative treatment trials that define ideal therapy for specific organisms causing PVE. As a result, regimens use a single or combinations of antimicrobial agents that will be bactericidal for the causative organism and that have been shown effective in prior case series of PVE and NVE. Because the prosthetic valve is a foreign body, organisms adherent to it may be embedded in biofilm and in a stationary, non-growing state.

    Antibiotics, however, are largely most active against growing replicating organisms. Thus, prosthesis adherent non-growing stationary phase organisms, especially staphylococci, pose a unique treatment challenge. Effective therapy ideally includes a specific antimicrobial effective against these stationary phase organisms. The antibiotic regimens recommended for treatment of PVE caused by a specific organism are, in general, those used to treat NVE caused by the same organism. They are, however, administered for longer periods and, where possible, combined with other antibiotics to enhance bactericidal activity (see below). For PVE caused by staphylococci, a multi-drug regimen that includes rifampin, an antibiotic which kills these bacteria even when in stationary phase in biofilm, is recommended.

  • The intracardiac pathology associated with PVE is the other determinant of effective therapy. In 40 to 55 percent of cases, infection extends beyond the valve into paravalvular tissue resulting in valve dehiscence and paravalvular regurgitation, myocardial abscess, intracardiac fistula, or further extension disrupting the conduction system or extension into the pericardium causing pericarditis. These changes are more likely with infection occurring early rather than later after surgery, with infection of prostheses in the aortic rather than mitral position, with mechanical valves rather than bioprostheses, and with more virulent pathogens such as staphylococci.

    Alternatively, large vegetations can occlude the valve orifice causing functional stenosis and infection can perforate valve leaflets causing significant regurgitation through the valve rather than via a paravalvular route. When there is significant valve dysfunction or paravalvular infection/abscess, optimal therapy requires cardiac surgery to correct the pathology. From 40 to 50 percent of patients with PVE undergo cardiac surgery while receiving antibiotic therapy (some emergently shortly after beginning antibiotics) and in another 15 to 20 percent surgery is indicated but patients refuse or surgery is contraindicated by complications of PVE or co-morbidities.

1. Anti-infective agents

If I am not sure what pathogen is causing the infection what anti-infective should I order?

  • Understanding the microbiology of PVE is crucial to the design of effective therapy. The cause of PVE varies to some degree by the time interval from valve surgery to onset of symptomatic infection – early onset 2 months or less, intermediate onset of 2 to 12 months, and late onset of over 12 months (Table I). Infection that is health-care associated (nosocomial or non-nosocomial) or IVDA related will also impact microbiology.

  • When PVE presents as an indolent infection and the patient is hemodynamically stable, it is reasonable to obtain cultures and delay initiating antibiotic therapy pending culture results. This is particularly important if there has been recent antibiotic exposure which might render the initial cultures negative. In contrast, when PVE presents as an acute toxic infection or there is hemodynamic instability suggesting a need for early surgery, blood cultures should be obtained and empiric therapy initiated immediately thereafter. Unless there is microbiologic or epidemiologic information suggesting a specific pathogen, empiric therapy with intravenous vancomycin, gentamicin and either cefepime or a carbapenem should be started. Therapy can be revised based upon subsequent culture results.

  • Regimens for the commonly encountered organisms that cause PVE. Rifampin, as part of a three-drug regimen, is important in treating staphylococcal PVE (Table II). However, because rifampin resistance can emerge during the initial weeks of treatment, it is important to delay starting rifampin until it is confirmed that the other 2 agents in the regimen are active against the infecting strain. If the strain is resistant to gentamicin, a fluoroquinolone or another active antimicrobial should be substituted.

Table II.n

Treatment of PVE in Adults Caused by Specific Microorganisms

  • If an enterococcus exhibits high level resistance to gentamicin (grows in the face of 500 µg/ml) streptomycin should be tested for high level resistance (growth in the face of 1000 µg/ml). Streptomycin can be used if high level resistance is not present. Although standard regimens for treating enterococcal PVE and NVE call for 4 to 6 weeks of aminoglycoside therapy, some data suggest that shorter periods of aminoglycoside therapy – 2 to 3 weeks – are effective. These data support stopping the aminoglycoside component of therapy early when faced with adverse events such as deteriorating renal function or VIII nerve toxicity. If an enterococcal strain is highly resistant to both gentamicin and streptomycin, aminoglycosides cannot be used to achieve bactericidal synergy. Double beta-lactam therapy (ampicillin 12gm IV daily plus ceftriaxone 2gm IV q 12 hours) have been effective in treating E. faecalis NVE and might be considered in PVE. Treatment of PVE due to vancomycin resistant E. faecium or E. faecalis strains require extensive evaluation of the strain’s in vitro susceptibility and advice from specialist.

  • PVE due to non-HACEK gram-negative bacilli is infrequent and the
    in vitro susceptibility of the implicated organism is difficult to predict. In general, optimal therapy will require cardiac surgery plus a prolonged course of combination antibiotic therapy – high dose beta-lactam (a ureidopenicillin, third or fourth generation cephalosporin, or a carbapenem) plus an aminoglycoside.

  • Blood culture negative PVE is usually the consequence of confounding by prior antibiotic administration. In patients within 1 year of valve placement, vancomycin-gentamicin-cefepime (2gm q 8 h IV) plus rifampin at dose noted in Table I or, if occurring over 1 year after valve surgery, vancomycin-gentamicin-ciprofloxacin plus rifampin. Other causes of blood culture negative PVE are due to fastidious organisms such as Bartonella species, Coxiella burnetii, Tropheryma whippelii, Brucella species, and non-Candida fungi. Evaluation and treatment of these cases of PVE should be conducted with the assistance of an infectious disease specialist.

2. Next list other key therapeutic modalities.

  • Antibiotic therapy alone is often successful in patients with PVE caused by less virulent antibiotic susceptible organisms and who do not have heart failure, valve dysfunction, or invasive infection. In contrast, recent studies show in-hospital and short-term mortality is associated with moderate to severe heart failure associated with prosthetic valve dysfunction, S. aureus infection, persistent bacteremia, invasive infection and myocardial abscess, persistent fever for 9 days on appropriate antibiotic therapy. Mortality rates in patients with one or more of these features may be reduced by surgical intervention with debridement and valve replacement. Recent in depth analyses of the impact of surgery for PVE on mortality at hospital discharge and after 1 year suggest that who benefits for surgery is nuanced. Survival benefits primarily accrue to those patients with major valve dysfunction (aortic or mitral regurgitation), paravalvular infection, and prosthetic valve complications. In patients with S. aureus PVE it is difficult to demonstrate a survival benefit from surgery. Nevertheless, while S. aureus PVE without anatomic intracardiac complications does not routinely require surgery, patients with valvular and paravalvular complications likely require surgery for survival.

  • When surgery occurs during antibiotic therapy for PVE, a full course of antibiotic therapy is generally given after surgery unless the operation is occurring late in the planned course of antibiotic therapy and the findings at operation suggest infection was already cured.

  • The indications for urgent or emergent surgical intervention in patients with PVE include;

    Congestive heart failure (NYHA class III or IV) due to prosthesis dysfunction

    S. aureus infection with valve dysfunction

    Perivalvular abscess or invasive infection with or without resulting heart block

    Intracardiac fistula

  • Additional potential indications for surgery include:

    Relapse after optimal antimicrobial therapy

    Valve dehiscence without severe heart failure

    To prevent repeat embolization when large vegetations (>10mm) persist after an earlier embolic event

    PVE caused by organisms poorly responsive to antimicrobial therapy: fungi, Pseudomonas aeruginosa, antibiotic resistant Gram-negative bacilli, vancomycin resistant E. faecium and other enterococci for which bactericidal synergistic therapy is not available.

    Blood culture negative PVE with persistent fever after 9 days of empiric treatment

  • Anticoagulation to prevent thromboembolic complications associated with prosthetic valves is standard with mechanical valves and prostheses in the mitral position. Whether continuing this anticoagulant therapy in patients with PVE is debated. There are no prospective randomized trials to compare the risk of thromboembolic complications, stroke, or cerebral hemorrhage when anticoagulant therapy is continued versus stopped in these patients. The concern is that thromboembolic events or stroke will be increased without continued anticoagulation or alternatively that with anticoagulation strokes will convert to hemorrhagic events with increased morbidity.

    In a collation of PVE series (largely mechanical valves) among patients with no anticoagulation, 42% had strokes and 9% had cerebral hemorrhage versus those who were anticoagulated, whereby 11% had strokes and 5% experienced cerebral hemorrhage. Accordingly, continued cautious and carefully regulated anticoagulation is recommended for patients with PVE involving valves wherein anticoagulation is standard of care. If stroke or bleeding occurs, the risk of hemorrhagic transformation or additional hemorrhage must be assessed and anticoagulation reversed or reduced to decrease these risks. Anticoagulant therapy is NOT initiated in those PVE patients wherein it would not be standard of care in the absence of infection.

What complications could arise as a consequence of prosthetic valve endocarditis?

  • In addition to the intracardiac complications of PVE, arterial emboli occur in up to 50 percent of cases, with half of these going to the CNS. Emboli are most common in patients with large vegetations (>10mm diameter), mitral valve infection and PVE caused by
    S. aureus. Approximately 50 percent of emboli occur before or at PVE diagnosis; the frequency decreases rapidly by effective antimicrobial therapy. Still, in PVE embolic cerebral infarction and intracerebral hemorrhage are major causes of morbidity and mortality. Mycotic aneurysms contraindicate anticoagulation.

  • PVE may be complicated by focal extracardiac infections such as septic arthritis, osteomyelitis, renal abscess or meningitis. Focal infection may require modifications in treatment. Splenic abscess occurs in 3 to 5 percent of cases and may require percutaneous drainage or splenectomy for cure.

  • Mycotic aneurysms occur in 2 to 10 percent of patients with PVE. Intracerebral mycotic aneurysm present as localized headache or intracerebral bleeding. If greater than 5mm, they may be detected by MRI or spiral CT, whereas small ones are only reliably detected by cerebral angiography. With rupture of intracerebral aneurysms, mortality is 80%. Management requires neurosurgical evaluation. Small aneurysms may heal with antibiotic therapy and thus can be followed with serial imaging. Continued expansion with follow-up would suggest a need for surgical therapy.

  • Extracranial mycotic aneurysms may cause focal pain, a pulsatile mass, or present as rupture. These should be repaired surgically to prevent complications.

  • PVE may be complicated by significant acute renal injury due to immune complex deposition with hypocomplementemic glomerulonephritis. This typically improves with effective treatment. More commonly, acute renal injury in PVE is a result of impaired cardiac function or antimicrobial toxicity.

What should you tell the family about the patient's prognosis?

  • Although the treatment of PVE has become much more effective, PVE remains a major cause of morbidity and mortality. Among patients with PVE, almost 50 percent undergo cardiac surgery during the index hospitalization. Mortality during initial treatment ranges from 20 to 25 percent. Mortality is significantly increased with age above 65 years, healthcare-associated infection,
    S. aureus infection, persistent bacteremia, congestive heart failure, intracardiac abscess, and stroke. Among PVE wherein surgery is indicated but not performed, 75 percent die. The 5-year survival rates of PVE patients who survive surgery ranges from 54 to 82 percent. Relapse of endocarditis occurs in 6 to 15 percent and 10 percent of PVE patients treated surgically or medically, respectively.

How can prosthetic valve endocarditis be prevented?

  • Although the efficacy of periprocedure prophylaxis of endocarditis remains unproven and as such has been curtailed or abandoned by some expert statements and guidelines, the American Heart Association and Infectious Disease Society of America continue to advise antibiotic prophylaxis in patients with prosthetic valves. For patients who will have dental procedures with gingival or periapical manipulation, prophylaxis 1 hour before the procedure is advised (Table III). These antibiotics can also be used for prophylaxis in prosthetic valve patients who will undergo respiratory mucosal surgery (tonsillectomy, rigid bronchoscopy). Although not recommended, I advise prophylaxis directed at enterococci in advance of genitourinary tract manipulation. Infusion intravenously 1 hour before the procedure of ampicillin 3gm IV plus gentamicin 70mg or of vancomycin 1gm plus gentamicin 70mg is recommended.

WHAT'S THE EVIDENCE for specific management and treatment recommendations?

John, MDV, Hibberd, PL, Karchmer, AW, Sleeper, LA, Calderwood, SB. “Staphylococcus aureus prosthetic valve endocarditis: Optimal management and risk factors for death”. Clin Infect Dis. vol. 26. 1998. pp. 1302-1309. (This is a retrospective study of 33 patients with S. aureus PVE that illustrates using a multi-variable analysis that surgical therapy during early treatment reduces mortality in those patients with intracardiac complications.)

Chirouze, C, Alla, F, Fowler, VG, Sexton, D. “Impact of Early Valve Surgery on the Outcome of Staphylococcus aureus Prosthetic Valve Infective Endocarditis: Analysis in the International Collaboration of Endocarditis – Prospective Cohort Study”. Clin Infect Disease. vol. 60. 2015. pp. 741-749. (This study provides a careful assessment of the benefit of surgery in patients with S. aureus PVE and demonstrates that routine surgery in the patients may not enhance survival.

Karchmer, AW , Bayer, AS. “Surgical therapy for Staphylococcus aureus Prosthetic Valve Endocarditis: Proceed with Cautions (Caveat Emptor)”. Clin Infect Dis. vol. 60. 2015. pp. 750-752. (This is a commentary of the Chirouze et al. paper [above]. It addresses which patients with S. aureus PVE might benefit from surgery.)

Karchmer, AW, Longworth, DL. “Infections of intracardiac devices”. Infect Dis Clin N Am. vol. 16. 2002. pp. 477-505. (This paper provides an overview of the clinical features and treatment of PVE.)

Wang, A, Pappas, P, Anstrom, KJ, Abrutyn, E, Fowler, VG , Hoen, B, Miro, JM, Corey, GR, Olaison, L, Stafford, JA, Mestres, CA, Cabell, CH. “International Collaboration on Endocarditis Investigators: The use and effect of surgical therapy for prosthetic valve infective endocarditis: A propensity analysis of a multicenter, international cohort”. Am Heart J. vol. 150. 2005. pp. 1086-1091. (This paper reports a large multicenter experience with PVE. It shows in an adjusted analysis that in-hospital mortality is associated with embolic stroke and S. aureus infection. There is a trend for improvement in in-hospital mortality with surgery during active infection.)

Baddour, LM, Wilson, WR, Bayer, AS, Fowler, VG . “Infective Endocarditis in Adults: Diagnosis, antimicrobial therapy, and management of complications: A statement for healthcare professionals from the American Heart Association: Endorsed by the Infectious Diseases Society of America”. Circulation. 2015. pp. 132(This guideline statement provides a detailed approach to the treatment of endocarditis, including PVE, with discussions of antibiotic therapy and surgical intervention.)

Habib, G, Lancellotti, P, Antunes, MJ. “2015 ESC Guidelines for the Management of Infective Endocarditis”. European Heart Journal. 2015. (This is a detailed guideline statement on the treatment of endocarditis from the European Society for Cardiology. Some recommendations differ from those of the American Heart Association [Baddour et al.].)

Lalani, T, Chu , VH, Park, LP. ” In-Hospital and 1-Year Mortality in Patients Undergoing Early Surgery for Prosthetic Valve Endocarditis”. JAMA Intern Med. vol. 173. 2013. pp. 1495-1504. (This is a study of 1025 patients with PVE and provides a detailed analysis of the impact of surgery on mortality. It illustrates the nuances of which patients benefit from surgical intervention.)

Hill, EE, Herregods, MC, Vanderscheren, S, Claus, P, Peetermans, WE, Herijgers, P. ” Management of prosthetic valve infective endocarditis”. Am J Cardiol. vol. 101. 2008. pp. 1174-1178. (This series of 80 patients with PVE assesses risk factors for mortality and the role of surgery in effective treatment.)

Wang, A, Athan, E, Pappas, PA, Fowler, VG, Olaison, L, Pare, C, Almirantte, B, Munoz, P, Rizzi, M, Naber, C, Logar, M, Tattevin, P, Iarussi, DL, Selton-Suty, C, Jones, SB, Casabe, J, Morris, A, Corey, GR, Cabell, CH. ” International Collaboration on Endocarditis-Prospective Cohort Study Investigators: Contemporary clinical profile and outcome of prosthetic valve endocarditis”. JAMA. vol. 297. 2007. pp. 1354-1361. (This large multicenter experience [556 cases of PVE] characterizes the patients treated surgically as well as the risk factors for in-hospital mortality.)

Yau, JWY, Lee, P, Wilson, A, Jenkins, AJ. “Prosthetic valve endocarditis: What is the evidence for anticoagulant therapy?”. Intern Med J. vol. 41. 2011. pp. 795-797. (This paper provides a perspective on the data supporting cautious anticoagulation during treatment of PVE involving mechanical valves.)

Fernandez-Hidalgo, N, Almirante, B, Gavalda, J. “Ampicillin plus Ceftriaxone Is as Effective as Ampicillin plus Gentamicin for Treating Enterococcus faecalis Infective Endocarditis”. Clin Infect Dis. vol. 56. 2013. pp. 1261-8. (This paper compares patients treated with the ampicillin-ceftriaxone regimen for E. faecalis IE (including PVE) to a cohort receiving standard therapy and demonstrated that outcomes are comparable with reduced renal dysfunction in the non-aminoglycoside exposed patients.)