Infective Endocarditis

Authors: Matthew E. Levison, M.D.

The characteristic pathology of infective endocarditis is the vegetation, a lesion that is the result of successive deposition of platelets and fibrin on the endothelial surface of the heart. Infection is the most common cause, and the usual pathogen is one of a variety of bacterial species, although fungi may be the cause in some patients. Microscopic colonies of the pathogen are buried beneath the surface of fibrin, usually in the absence of an inflammatory reaction. Most commonly the heart valve is the site of the vegetation, but in certain instances vegetations may occur on other parts of the endocardium. Involvement of extracardiac endothelial sites, which can produce an illness clinically similar to endocarditis, is properly termed endarteritis.


Non-Cardiac Risk Factors

In population-based studies, the age and sex-adjusted incidence of infective endocarditis is about 5 per 100,000 person-years (119). However the incidence rate of infective endocarditis may be influenced by the prevalence of risk factors in the local population. For example, the incidence rate of infective endocarditis was found to be approximately 12 per 100,000 person-years the Philadelphia region, the excess incidence attributed almost entirely to the increased frequency of intravenous drug use in this region, where infective endocarditis in intravenous drug users represented 46% of all cases (10). Other significant non-cardiac risk factors risk factors (Table 1) for infective endocarditis are: 1) advancing age (in part due to the increased prevalence of predisposing cardiac lesions, e.g. , prior infective endocarditis, degenerative cardiac lesions and prosthetic cardiac valves; and 2) male gender (in part due to the increased prevalence of certain cardiac lesions, such as bicuspid aortic valve, in males). In the pre-antibiotic era the median age of patients with infective endocarditis was between 30-40 years (60); now over 50% of patients are older than 50 years of age, although intravenous drug users with infective endocarditis tend to be younger (1093). The incidence rate ratio for those 65 year of age or older is almost 9 times that of those under 65 years, and for males about 1.5 to 2.5 times that of females (10). Lower risk for infective endocarditis caused by oral flora has been found among edentulous cases than patients who had teeth but did not floss; also reduced risk was found among those who floss daily, which suggests that poor dental hygiene is a risk factor, especially among those with cardiac risk factors (125); but prior dental procedures have not been found to be risk factor for infective endocarditis caused by dental flora (126,66). Other non-cardiac risk factors for infective endocarditis that have been reported include uremia (probably related to dialysis catheter sepsis) and diabetes mellitus (125), and surgery and skin infections and “infectious episodes” (66).

back to top

Cardiac Risk Factors

Predisposing cardiac lesions are found in about 3/4 of patients with infective endocarditis (126). Patients without a predisposing cardiac risk factor more likely have nosocomial infective endocarditis, have infective endocarditis caused by more virulent organisms, such as S. aureus, or are intravenous drug users.

Conditions that have been identified as cardiac risk factors (Table 1) in patients with infective endocarditis include various degenerative valvular lesions, congenital heart disease, bicuspid aortic valves, mitral valve prolapse with mitral insufficiency or thickened mitral leaflets, rheumatic valvular heart disease, prosthetic cardiac valves, and previous infective endocarditis (126). However, inferring risk from the relative frequency of various lesions in case series of infective endocarditis can be problematic, since those lesions found more frequently in the general population are also most likely to be common in case series of patients with infective endocarditis. For example, although rheumatic valvular heart disease is still common both in the general population and the most common cardiac risk for infective endocarditis in developing countries (21), the frequency of rheumatic valvular heart disease in developed countries both in the general population and in patients with infective endocarditis has markedly declined (93). Mitral valve prolapse is now the most frequent cardiac lesions found in patients with infective endocarditis in developed countries, which reflects the high prevalence of mitral valve prolapse in the general population, estimated to be between 2-21% (93).

The true degree of risk for infective endocarditis of a specific cardiac lesion can be determined only by measuring the incidence rate of infective endocarditis among those who have a particular cardiac abnormality (119): The highest incidence rates (over 2000/100,000 patient years) occur in patients who undergo valve replacement of an infected prosthetic cardiac valve. Patients who either undergo prosthetic valve replacement for native-valve endocarditis, have had previous native-valve endocarditis, or have a prosthetic cardiac valve in place also have high incidence rates (300-740/100,000 patient-years). For prosthetic valve endocarditis, risk is greatest during the first few post-operative months as a result of intra-operative and peri-operative contamination, and probably does not vary by site of placement or type of prosthetic valve material. Similar incidence rates are found in patients with valvular rheumatic valvular heart disease.

Surgically uncorrected congenital heart disease at risk for endocarditis include patent ductus arteriosus, ventricular septal defects, tetralogy of Fallot, and coarctation of the aorta. Surgical repair can eliminate the risk of infective endocarditis 6 months after surgery if no residual shunt or valvular lesion remains. However, corrective surgery is not always protective: prosthetic material and devices remain at risk for the first 6 months after placement because of incomplete endothelialization; also persistent risk remains for palliative shunts and conduits and sites of turbulent blood flow that remain despite placement of prosthetic material in an attempt to surgically correct congenital heart disease (267589).

Transvenous permanent pacemakers have been reported to have an incidence rate for infective endocarditis of about 50/100,000 patient-years (33).

Similarly, the incidence rate for mitral valve prolapse with murmur is also about 50/100,000 patient-years (only about 10-fold higher than that of the general population) (119), but mitral valve prolapse may be especially a problem in elderly males with mitral valve thickening (93). However, valvular rheumatic valvular heart disease, congenital heart disease and mitral valve prolapse represent groups of patients with diverse types and severity valvular abnormalities that have different levels of risk for acquisition of infective endocarditis, which complicates an accurate assessment of the true risk of acquisition of infective endocarditis in individual patients with a specific underlying cardiac condition.

Patients with cardiac lesions at no greater risk than the general population include those with secundum atrial septal defects, atherosclerosis, previous coronary artery bypass graft surgery, mitral valve prolapse without murmur, and previous rheumatic valvular heart disease without valvular dysfunction.

back to top


The following sequence of events is thought to result in infective endocarditis: The normal endothelium is non-thrombogenic; but when damaged the endothelium is a potent inducer of blood coagulation. Turbulent blood flow produced by certain types of congenital or acquired heart disease, e.g. , flow from a high to low pressure chamber or across a narrowed orifice, traumatizes the downstream endothelium and predisposes for deposition of platelets and fibrin (104), the so-called “nonbacterial thrombotic endocarditis” lesion (NBTE), on the surface of the traumatized endothelium (4). Subsequent episodes of bacteremia with species capable of survival in the blood stream, adherence to the nonbacterial thrombotic endocarditis, and proliferation at this site can then result in infective endocarditis (28).

The adherence of microorganisms in the blood stream to nonbacterial thrombotic endocarditis involves a complex interaction between multiple types of microbial adhesions, matrix molecules, and platelets on the surface of the damaged endothelium. Dextran and fimA serve as two of the adhesins for oral streptococci. The avidity of adherence in vitro to a fibrin-platelet matrix, as well as the ability to produce experimental infective endocarditis in rabbits with traumatized cardiac valves has been found to be related to the amount of dextran produced by these streptococci (99,117). Dextran-producing streptococci are also more likely to be recovered from blood of patients with viridans streptococcal infective endocarditis than are non-dextran producing strains (96). Viridans streptococci also contain fimA protein, a major adhesion to the fibrin/platelet matrix of the nonbacterial thrombotic endocarditis (13). In contrast, S. aureus can adhere directly to fibronectin that covers the surface of uninjured endothelial cells; uninjured endothelial cells in tissue culture can phagocytose adherent S. aureus (77), which multiply intracellularly, kill the endothelial cell, and thereby cause fibrin-platelet deposition; this may explain the propensity of S. aureus, unlike oral streptococci, to initiate infective endocarditis on normal heart valves (88,137).

Further deposition of platelets and fibrin occurs on the surface of adherent bacteria. Bacteremia is sustained by subsequent fragmentation of the vegetation, which exposes the underlying microbial colonies. Scanning electron microscopy of vegetations, however, rarely shows bacteria on the luminal surface of the vegetation, as the exposed bacteria are rapidly covered by deposition of fibrin and platelets. The vegetation enlarges as circulating bacteria redeposit on the surface of the vegetation, which in turn stimulates further deposition of fibrin and platelets. The resultant vegetation then is composed of successive layers of fibrin and clusters of buried bacteria, with rare red blood cells and leukocytes (30). Sustained bacteremia with a constant number of bacteria per ml of blood is characteristic of infective endocarditis, and results from an equilibrium between the rate of release of bacteria from the vegetation and clearance of circulating bacteria by the reticuloendothelial system in liver, spleen, and bone marrow (9). On the left side of the heart, microorganisms become buried in the depths of the vegetation and secluded from host defenses, such as leukocytes and humoral factors; there the microorganisms, multiply initially as rapidly as do bacteria in broth cultures to reach maximal microbial densities of 108-11 colony forming units per gram of vegetation within a short time (28). Right-sided vegetations have lower bacterial densities, which may be the consequence of host defense mechanisms, such as polymorphonuclear leukocytic activity, that are active at this site (45138). In the mature vegetations of streptococcal infective endocarditis, over 90% of the microorganisms are metabolically inactive and non-growing (29). Microorganisms in this phase are least responsive to the bactericidal effects of antibiotics that inhibit bacterial cell wall synthesis, such as the beta-lactams, as a consequence of lack of expression of penicillin binding proteins, the target of beta-lactam antibiotics (124).

When healing occurs, either spontaneously, e.g. , right-sided endocarditis, or under the influence of antibiotic therapy, the surface layer of fibrin is invaded by fibroblasts and the rough surface of the vegetation is progressively covered by smooth endothelium, which reduces the risk of reseeding of the vegetation (30).

back to top


Compression of infected foci release large numbers of bacteria into the blood stream (71). Bacteremia can also occur spontaneously in uninfected individuals, perhaps related to unsuspected minor infected foci such as periodontitis (94), or following trauma to uninfected skin or mucosal surfaces that are populated by a dense endogenous micro flora. These mucosal sites include the gingival crevice, oropharynx, terminal ileum, colon, distal urethra, and vagina. The intensity of the resulting bacteremia is related directly to the magnitude of the trauma, the density of the microbial flora, and the presence of inflammation or infection at the site. For example, spontaneous and procedure-induced bacteremia occurs more frequently in the presence of periodontal infections (3694), which is a likely consequence of hyperemia and a more abundant microflora in the infected periodontal tissues surrounding the teeth.

Trauma to skin or mucosal site releases many different microbial species into the blood stream, the types of which depend on the unique endogenous microflora that colonizes the particular traumatized site (71). However, only a few of the many bacterial species that thus gain entry into the bloodstream, namely viridans streptococci, staphylococci, and enterococci, are commonly capable of causing infective endocarditis; these microorganisms account for over 80 % of cases of infective endocarditis (93). Bacteremia caused by viridans streptococci and other oral streptococci occurs in 18-85% of dental extractions and periodontal procedures. Bacteremia caused by oral streptococci also follows esophageal dilatation and sclerotherapy for esophageal varices. Enterococcal bacteremia occurs less frequently following genitourinary and gastrointestinal invasive procedures such as prostatic surgery, endoscopic retrograde cholaniopancreatography for biliary obstruction, biliary tract surgery, surgery on lower intestinal mucosa, and urethral dilatation.

A history of such procedures within the preceding 2 months has been found in 25% of patients with viridans streptococcal infective endocarditis and 40% of patients with enterococcal infective endocarditis. However, it is now believed that the period between the inciting bacteremia and the clinical onset of infective endocarditis is more likely to be less than 2 weeks (116). Indeed, these medical procedures (particularly dental procedures) mentioned above are common in the general population, which makes assessment of the risk of the procedure for production of infective endocarditis difficult; the mere temporal association of a particularly common procedure, such as a dental procedure, with a rare disease like infective endocarditis does not necessarily infer causation. Minor mucosal trauma as routine as brushing teeth, flossing, chewing hard candy or other everyday experiences commonly causes asymptomatic bacteremia that is characterized by small numbers (usually <10 colony forming units/ml of blood), and short duration (15-30 minutes). Although transient bacteremia is a common, everyday event, and each event may be associated with only a very small risk for infective endocarditis, the cumulative risk of these transient episodes of low grade bacteremia may be sufficient to account in large part for the 75% of patients with viridans streptococcal infective endocarditis or 60% of patients with enterococcal infective endocarditis who fail to recall a medical or dental procedure that preceded the onset of their endocarditis (103). Even transient bacteremia from everyday events may be additionally responsible for some cases of infective endocarditis in patients who give a history of a recent preceding procedure. Two recent studies of patients with infective endocarditis that used age- and sex-matched controls failed to demonstrate a relationship between recent dental procedures and infective endocarditis due to oral microorganisms (66126).

back to top


Viridans streptococci (i.e. , alpha-hemolytic streptococci), such as Streptococcus sanguis, S. mitis, S. mutans, S. salivarius, and S. bovis (a non-enterococcal Group D streptococcus), are responsible for 35-40% of cases. In addition, Abiotrophia and Granulicatella species, which are viridans streptococci (previously referred to as nutritionally variant streptococci or NVS) that require pyridoxal HCl or thiol-compounds for growth and satellite around staphylococcus when cultured on blood agar plates, cause about 5% of cases of infective endocarditis (11). Infective endocarditis caused by NVS and Gemella morbellorum is associated with greater mortality and morbidity (clinical and bacteriologic relapse, CHF, prosthetic valve placement, and embolization) (102123). S. bovis infective endocarditis occurs more frequently in the elderly and is associated with preexisting colonic lesions (e.g. , colonic malignancies) (91). Enterococcus species are responsible for up to 10% of cases. Occasional cases of infective endocarditis are due to beta-hemolytic streptococci or rarely Streptococcus pneumoniae. Infective endocarditis caused by the Streptococcus anginosus group (also known as S. milleri group that includes S. anginosus, intermedius and constellatus) is frequently complicated by destructive valvular lesions and purulent metastatic foci similar to that caused by S. aureus. S. pneumoniae, S. pyogenes, and Groups BC, and G streptococci are relatively uncommon causes of infective endocarditis, but have also been associated with greater morbidity and mortality.

In recent series, staphylococci have become the most common cause of infective endocarditis, perhaps due to the increasing frequency of nosocomial acquisition and prevalence of prosthetic valves and intravenous drug users (43); however infective endocarditis in the subgroup of patients who are adult non-intravenous drug users and acquire infective endocarditis in the community is reported to be still most commonly due to viridans streptococci (93). Coagulase-negative staphylococci, usually S. epidermidis, occasionally cause native valve endocarditis in patients with underlying valvular abnormalities and commonly cause of early-onset prosthetic valve endocarditis (PVE); in the case of PVE, they are almost always resistant to methicillin or oxacillin, which is consistent with their hospital-acquisition. S. aureus also causes PVE and nosocomial infective endocarditis, and infective endocarditis in intravenous drug users. Staphylococcus lugdunensis, a coagulase-negative staphylococcus, can cause a highly destructive native valve endocarditis, similar to infective endocarditis caused by S. aureus (97).

The so-called HACEK group of fastidious gram-negative bacilli (GNB) (Haemophilus aphrophilusActinobacillus actinomycetemocomitansCardiobacterium hominisEikenella corrodensKingella kingae) cause up to 5% of cases. GNB other than the HACEK group rarely cause infective endocarditis. However, GNB, including Enterobacteriaceae and Pseudomonas aeruginosa may cause infective endocarditis in intravenous drug users and in patients with a prosthetic cardiac valve (22, 69). Strains of GNB that cause infective endocarditis exhibit resistance to complement-mediated lysis by normal human sera that perhaps allows their survival in the blood stream and on the endothelial surface of the cardiac valve (31139).

Most cases of fungal endocarditis also occur in intravenous drug users or in patients with a prosthetic cardiac valve; such patients often have central vascular catheters, and may be immunocompromised. The most common fungal species is Candida albicans, followed by Candida parapsilosis. Polymicrobial endocarditis is unusual, but is seen in intravenous drug users. Salmonella species are common causes of endarteritis on an atherosclerotic abdominal aortic plaque (39). Blood cultures may be negative in up to one third of patients with infective endocarditis; although in one half of these patients the negative blood cultures can be attributed to prior antibiotic therapy that has suppressed the infection (12). Other causes of culture-negative infective endocarditis (Table 2 and 3) include fastidious organisms such as Coxiella burnetiiBartonella quintanaTropheryma whipplei, and Brucella species. The frequency of culture-negative endocarditis and its etiologic agents is likely to be strongly correlated to epidemiology of the microbial agent in a particular geographic region, e.g. , Brucella infective endocarditis is related to exposure to contaminated milk, cheese or meat and Coxiella burnetii infective endocarditis to infectious aerosols or milk from infected farm animals in certain regions of the world. Bartonella quintana infective endocarditis is seen in the homeless, body lice-infested, and alcoholic population (101).

back to top


In the pre-antibiotic era, when infective endocarditis was uniformly fatal, a short duration of illness of less than 6 weeks prior to death was used to characterize acute endocarditis; in contrast, subacute and chronic infective endocarditis had a more indolent course until death in six weeks to two years (5759). The distinction based on chronicity has continued to prove useful in the antibiotic era. Chronicity is now used in reference to the duration of illness prior to presentation. Acute infective endocarditis is usually (50-70%) caused by S. aureus, especially when accompanied by marked signs of general infection and suppurative embolic phenomena and has a rapidly fatal course if treatment is delayed. Acute infection may develop on a previously normal valve: In the non-intravenous drug users, the aortic valve is commonly involved; in the intravenous drug users, the tricuspid valve. Acute tricuspid valve lesions may be associated with multiple septic pulmonary emboli. Therefore, clinically acute infective endocarditis, especially in the presence of a pre-existing intravascular devise, e.g. , intravenous catheter, can serve a diagnosis of presumptive S. aureus endocarditis, which will require use of empiric anti-staphylococcal antibiotic therapy that can be modified when results of initial blood cultures become available. Subacute infective endocarditis, commonly caused by streptococci and enterococci, in contrast, often develops on previously damaged endocardium, has a less dramatic clinical course with an onset within 14 days of the inciting bacteremia (116), and the clinical manifestations are characteristically fever, progressive wasting and debility, and non-suppurative peripheral vascular phenomena.

Clinical manifestations result from 1) the valvular infection itself; 2) embolization of fragments of the vegetation; 3) suppurative complications that result from hematogenous spread of infection; or 4) immunologic response to the infection in the form of immune complex vasculitis. Systemic manifestations of infective endocarditis include most commonly fever and other symptoms that may accompany fever, such as drenching night sweats, arthralgias, myalgias, pain in the low back and thighs, and weight loss. Fever is usually low grade, the temperature peaks rarely exceeding 39.4oC. However, fever may be high and spiking in patients with acute infective endocarditis or absent in a few patients, e.g. , those who are very elderly or severely debilitated, have significant renal or heart failure, or are taking antipyretics or antibiotics (55).

Murmurs of cardiac valvular insufficiency due to destruction or distortion of the infected valve and its supporting structures are commonly present; less commonly murmurs of valvular stenosis occur due to large vegetations. Murmurs are likely to be absent in tricuspid infective endocarditis or may be absent when a patient with acute infective endocarditis is first seen. Prosthetic valve endocarditis (PVE) may result in regurgitant systolic or diastolic murmurs as a result of dehiscence of the valve at the annulus, the suture lines being usual site of infection in early PVE, or muffling of the usual crisp prosthetic mechanical valve clicks.

In native valve endocarditis, valve ring abscess due to local extension of infection occurs in the weakest portion of the annulus, which is near the membranous interventricular septum and AV node. Usually the noncoronary cusp of the aortic valve is involved. In post-operative PVE, valve ring abscess occurs commonly because the annulus, rather than the leaflet is the site of infection. Valve ring abscess can lead to persistent fever despite appropriate antimicrobial therapy, recurrent emboli, heart block as a result of destruction of conduction pathways in the area of the AV node and bundle of His in the upper interventricular septum, pericarditis or hemopericardium as a result of burrowing abscesses into the pericardium, or shunts between cardiac chambers or between the heart and aorta. Myocardial infarction may occur as a result of coronary artery embolization, and myocardial abscess can occur as a result of bacteremia. Diffuse myocarditis may occur as a consequence of immune-complex vasculitis.

Congestive heart failure (CHF) is a common complication of infective endocarditis and carries a grave prognosis; it develops in patients with infective endocarditis as a consequence of valvular or myocardial involvement during antimicrobial therapy (2/3 of patients will develop CHF within the first month of therapy), or may precede the onset of infective endocarditis as a consequence of the underlying cardiac lesion. CHF may develop indolently during the prolonged period prior to diagnosis in some patients with subacute infective endocarditis, or also may develop dramatically in patients with acute S. aureus infective endocarditis when it is often accompanied by the sudden onset of a new murmur of reguritant blood flow secondary to destruction of the aortic valve or mitral valve or its supporting structures. CHF occurs more frequently with left-sided than right-sided infective endocarditis, and with aortic more frequently than mitral involvement, and CHF is more severe with sudden than insidious development.

Extracardiac manifestations include: 1) embolic events that result in infarction of numerous organs, such as the lung in right-sided infective endocarditis or the brain, heart, bowel, spleen, kidneys, or extremities in left-sided infective endocarditis; 2) suppurative complications that include abscesses, septic infarcts and mycotic aneurysms; and 3) immunologic reactions to the valvular infection that include glomerulonephritis, sterile meningitis, and polyarthritis, and a variety of vascular phenomena, such as mucocutaneous or conjunctival petechiae see photos, splinter hemorrhages see photos, Roth spots, Janeway lesions see photos, and Osler’s nodes. The development of clinically apparent splenomegaly and many of the various non-suppurative peripheral vascular phenomena is related to the duration of illness prior to presentation. The frequency of these clinical manifestations  (<50%) is currently less than in the past, as a result of shorter duration of illness prior to diagnosis.

Neurologic complications, which occur in about 20-40% of patients with left-sided infective endocarditis (108), are usually due to systemic embolization, often a devastating complication when involving the cerebral circulation. Up to 2/3 of emboli events involve the CNS, usually in the distribution of the middle cerebral artery. Systemic embolization occurs in 22 to 50% of patients with infective endocarditis; it is more frequent with S. aureus, HACEK organisms, Abiotrophia species, and fungal infective endocarditis, increasing or stable vegetation size on therapy, vegetations larger than 1 cm in diameter on echocardiography, and vegetations of the anterior leaflet of the mitral valve (105120). Frank cerebral abscess is rare, but occurs in patients with S. aureus endocarditis. Septic pulmonary emboli that appear as multiple round infiltrates commonly occur in patients with tricuspid valve S. aureus infective endocarditis and may cavitate or be complicated by empyema. The frequency of systemic embolization decreases dramatically during the first two weeks of successful antimicrobial therapy, as the vegetation heals (55). Mycotic aneurysms are an unusual, but important, complication of infective endocarditis. These aneurysms characteristically develop at arterial bifurcations, as a result of septic embolization or immune vasculitis involving the vasa vasorum, and involve more commonly cerebral arteries, followed by visceral arteries (e.g. , splenic, superior mesenteric) and arteries of the extremities, the abdominal aorta and sinus of Valsalva.

Mycotic aneurysms may be asymptomatic, but can become clinically evident quite suddenly without warning as a result of precipitous rupture or gradually as a result of a slow leak, even months or years after completion of successful therapy. Unremitting headache, visual disturbance, cranial nerve palsy, meningeal signs, change in mental status, or a focal neurologic deficit is suggestive of an impending rupture of an intracranial mycotic aneurysm. Cerebrospinal fluid can show microscopic white and red blood and an elevated protein, or it can be grossly bloody. Signs of blood loss at any site in a patient with infective endocarditis should suggest rupture of a mycotic aneurysm.

back to top

Nosocomial Endocarditis

Nosocomial infective endocarditis has been reported to account for about 10% of cases of infective endocarditis (10,40). The clinical presentation of nosocomial infective endocarditis is similar to that of community-acquired infective endocarditis. The source of bacteremia can be identified in >90% of cases of nosocomial infective endocarditis. The most important bacteremia-inducing event during hospitalization that results in infective endocarditis is use of an intravascular device, present in up to 50% of cases. A major predisposing cardiac lesion for nosocomial infective endocarditis is a prosthetic cardiac valve (present in up to 50% of cases). Although community-acquired S. aureus bacteremia, without the presence of a primary focus of infection was thought to be a major indicator of underlying infective endocarditis, recent studies have documented that nosocomial S. aureus bacteremia is frequently complicated by infective endocarditis (10—30% of the time) with or without a removable focus of infection (43). Similarly, the risk of developing infective endocarditis during an episode of hospital-acquired enterococcal bacteremia had been considered very low (<1%) (78), but nosocomial enterococcal infective endocarditis may be an emerging problem, especially if the patient has an underlying cardiac risk, such as a prosthetic valve, and the enterococcal species is E. faecalis (2,41). Sources for nosocomial enterococcal bacteremia include genitourinary or gastrointestinal procedures, intravascular catheter infection, or surgical wound infection. Although blood cultures are usually positive, the diagnosis of nosocomial infective endocarditis is frequently delayed due to failure to recognize the presence of early endocardial infection, which may be only evident on transesophageal echocardiography.


Echocardiography has become second in importance only to culture of blood in the investigation of patients who are clinically suspected to have infective endocarditis (32). Echocardiography can visualize valvular vegetations see photos , satellite vegetations, flail valves, ruptured chordae, perivalvular abscesses, fistulas, valvular perforations, and mycotic aneurysms. Echocardiography is also relied upon to identify predisposing cardiac lesions, and the causes and severity of congestive heart failure by assessment of ventricular size, wall motion, and dynamic function. Two-dimensional transthoracic echocardography (TTE) and transesophageal echocardiography (TEE), the two currently performed types of echocardiography, are usually safe and portable to the bedside. TTE is rapid, noninvasive, and relatively inexpensive. TEE is invasive, requires sedation, and is more expensive. TTE can give more general information as to cardiac structure and function. TEE is frequently helpful in situations where TTE is not, e.g. , in the presence of obesity, emphysema, small vegetations, e.g. , in early infective endocarditis, a prosthetic cardiac valve the structural components of which may impair the image of TTE, and infection in perivalvular tissue where PVE often starts. Hemodynamic complications, such as central or perivalvular regurgitant flow in the presence of a prosthetic valve can easily be detected and semiquantitated by additional color flow Doppler.

The sensitivity of TEE is greater (90-100%) than that of TTE (< 80%), allowing for detection of vegetations as small as 1 mm. TEE also has a specificity (i.e. , true negative rate), positive predictive accuracy (the probability of infective endocarditis when the echocardiogram is positive), and a negative predicative accuracy (the probability of no infective endocarditis when the echocardiogram is negative) of 95-100%. However, although low, the frequency of a falsely positive finding that can simulate a vegetation, due for example to nonspecific valvular thickening, rupture of a valve leaflet or mitral valve chordae, may be greater than the frequency of infective endocarditis in certain populations. Therefore TEE should only be done for diagnosis of endocarditis in patients in whom the diagnosis of infective endocarditis is at least “possible endocarditis” according to clinical or microbiologic criteria (see Duke criteria, Table 4). Use of TEE as the initial diagnostic test is recommended to detect small vegetations in patients with staphylococcal bacteremia and possible early endocarditis, suspected complicated infective endocarditis (e.g. , perivalvular abscess or conduction block), and suspected PVE.

Echocardiography may be falsely negative if vegetations are very small or have already embolized. If initial echocardiography fails to reveal evidence of infective endocarditis in a patient strongly suspected of having infective endocarditis or its intracardiac complications and another diagnosis is still not apparent, the TEE should be repeated in about one week. Sequential echocardiography during the course of antimicrobial therapy is used to assess healing of vegetations and detect development or progression of complications and guide decisions as to the need and timing of surgery. The finding of early closure of the mitral valve, as a consequence of elevated left ventricular end-diastolic pressure, associated with acute aortic valve infective endocarditis, has been used to predict the need for surgery; and the detection of an enlarging vegetation during the course of therapy or an especially a large vegetation (e.g. , >10 mm in its largest dimension) in some studies has been found to signify a poorer outcome, i.e. , the greater likelihood of embolization, congestive heart failure, the need for surgery and death. However, echocardiographic results must be taken in context of the specific clinical situation, especially the hemodynamic status, prior emboli events, and the type of pathogen involved, e.g. , antibiotic resistant gram-negative bacilli or fungi, for assessment of surgical intervention. Echocardiograph at the completion of antimicrobial therapy is needed to establish a new baseline as these patients remain at high risk for recurrent infective endocarditis and development of CHF that may require subsequent prosthetic valve placement if not done during period of antibiotic therapy.

back to top

Other Investigative Procedures

Cardiac catheterization can provide important information and should not be avoided when indicated in selected patients with endocarditis for fear of dislodging emboli. Coronary angiography is used to assess the presence of significant coronary artery disease prior to elective placement of prosthetic cardiac valves in patients who are over 40 years of age and have additional atherogenic risk factors. Contrast-enhanced CT or MRI can best evaluate abscesses or infarcts of the intra-abdominal organs (spleen or liver). Contrast-enhanced CT or MRI/MRA of the head may provide evidence of mycotic aneurysms. Magnetic resonance angiography for the detection of intracranial aneurysms is promising, but 4-vessel cerebral angiography remains the standard for evaluation for aneurysms smaller than 5 mm.

Electrocardiographic Manifestations

A baseline electrocardiogram (EKG) should be obtained to assess the presence of conduction abnormalities that develop in about 10-20% of patients with infective endocarditis as a consequence of burrowing valve ring abscesses. Prolongation of the PR interval may be the initial indication of the sudden development of more severe conduction abnormalities, such as complete heart block. Other abnormalities that can be detected by EKG include myocardial infarction and pericarditis.

Hematologic Manifestations

Progressive anemia of chronic disease with normochromic, normocytic indices routinely develops in subacute infective endocarditis with relatively normal platelet, white blood cell and differential counts. In acute infective endocarditis due to S. aureus, anemia may be initially absent although the white blood cell count is usually elevated with a shift to the left, and the platelet count is often low. PVE with an unstable prosthesis may cause acute hemolysis. The erythrocyte sedimentation rate is routinely elevated in infective endocarditis except when there is CHF or hypofibrinogenemia secondary to disseminated intravascular coagulation.

Renal Manifestations

Proteinuria and microscopic hematuria are common findings, occurring in up to 50% of patients. Renal emboli or focal glomerulonephritis can cause microscopic hematuria, but gross hematuria usually indicates renal infarction. Renal failure that develops in a patient with infective endocarditis is usually due to diffuse immune-complex glomerulonephritis.

Other Laboratory Manifestations

Serologic evidence of circulating immune complexes (CIC) may by found in endocarditis, the frequency of which is related to the duration of illness. Occasional false-positive non-treponemal serologic tests for syphilis occur. The cerebrospinal fluid may show polymorphonuclear leukocytes and moderately elevated protein concentration in up to 15% of patients, with a normal glucose concentration and negative culture. Frank bacterial meningitis, although unusual, can occur in S. aureus infective endocarditis.

back to top


Evidence for persistent bacteremia and cardiac valvular involvement, the sine qua non of infective endocarditis, should be sought in patients with suspicious clinical findings (e.g. , fever, risk factors, vascular or immune complex phenomena). Standardized criteria for the assessment of patients with suspected infective endocarditis were proposed in 1994 and subsequently modified in 2000 by the group at Duke University (3276). These so-called “Duke Criteria” use findings at surgery or autopsy, predisposing factors, blood culture and other laboratory data, and echocardiographic data to rank the probability of the diagnosis of infective endocarditis as definitive, possible or rejected. These diagnostic infective endocarditis criteria have been validated in several different patient populations. Definitive diagnosis of infective endocarditis depends on proof of cardiac valvular infection by histology or culture of vegetations obtained at the time of surgery or autopsy or by histology or culture of vegetations obtained at the time of surgical removal of an arterial embolus. In lieu of surgery or autopsy, definitive diagnosis can be established by having 2 of 3 of the following major criteria: 1) echocardiographic demonstration of endocardial involvement with characteristic vegetation (oscillating intracardiac mass), valve ring abscess, or new prosthetic valve dehiscence; 2) a new murmur of valvular regurgitation (worsening or changing or pre-existing murmur is not sufficient); or 3) demonstration intravascular infection with multiple blood cultures obtained over an extended period of time that are positive for a microorganism consistent with endocarditis or a single positive blood culture or serology for Coxiella (see Table 4). A definite diagnosis can also be made if the patient has one of the above major criteria plus 3 minor criteria (see Table 4) or 5 minor criteria. The diagnosis of possible infective endocarditis is made if the patient has 1 major plus 1 minor or 3 minor criteria (see Table 4). The diagnosis of infective endocarditis is rejected if findings do not meet criteria for possible infective endocarditis as above, or there is either no pathologic evidence of infective endocarditis at autopsy or surgery with less than 4 days of antibiotic therapy, rapid resolution of clinical findings with short-term antibiotic therapy, or a firm alternate diagnosis. The Duke criteria are intended to be guides and not a substitute for clinical judgment; for example, empiric therapy must often be initiated based on the clinical picture (acute findings) and cardiac (e.g. , presence of a prosthetic valve or history of prior infective endocarditis) or noncardiac risk factors (e.g. , intravenous drug users) before results of blood cultures or other diagnostic studies are available to apply the Duke criteria.

Microbiologic Investigation

Isolation of a pathogen from several blood cultures that are obtained over an extended period of time is important both to confirm the diagnosis of endocarditis and to enable determination of the optimal antibiotic regimen. Bacteremia in infective endocarditis is characterized by a constant number of organisms/ml of blood (usually 20-200 CFU/ml), the specific density being characteristic for that particular patient, unrelated to the height of the patient’s temperature or the site of blood sampling (e.g. , arterial versus venous blood), except for a slight fall in numbers across the hepatic or splenic circulation (9). Intermittently positive blood cultures are unusual in the absence of prior antimicrobial therapy. Less than 5-15% of patients with infective endocarditis have sterile blood cultures if adequate an adequate number of blood cultures are obtained prior to the start of antimicrobial therapy (132). The proper method for obtaining blood for culture includes: 1) Disinfection of the skin with 80-95% alcohol and then 2% iodine or iodophor solution, allowing the disinfectant to remain on the skin for at least 1 minute; 2) Withdrawal of at least 10 ml of blood per blood culture in an adult through the least contaminated site, preferably an antecubital vein, rather than, for example, the femoral vein. Two to three blood cultures should be obtained in this manner at least 1 hour apart to demonstrate that the bacteremia is continuous. In septic patients suspected of having infective endocarditis who are in immediate danger of death, two to three blood cultures should be obtained by separate venipuncture within one hour prior to initiation of empiric antibiotic therapy. In patients suspected of having coagulase-negative staphylococcal infective endocarditis, e.g. , patients with a prosthetic valve or other intracardiac foreign body, then it is advisable to draw 3 or more blood cultures initially to help distinguish coagulase-negative staphylococcal contamination of blood cultures from bacteremia.

The clinical microbiology laboratory should be advised of the suspected diagnosis of infective endocarditis, as some organisms require special media or more prolonged incubation of blood cultures (up to 3 weeks) for detection. In the absence of prior antibiotic therapy, the first three blood cultures are expected to be positive in over 95% of patients with positive cultures (132). Prior antibiotic therapy, fastidious bacteria (such as the Abiotrophia species, the HACEK group of organisms, NeisseriaBrucellaLegionellaBartonellaTropherymaChlamydiaCoxiella, and rickettsia), and fungi can result in negative cultures. Identification of these organisms will frequently require additional procedures, such as broad spectrum bacterial and fungal PCR and DNA sequencing on vegetations or emboli and histochemical stains and immunohistology of vegetations and emboli and serology, as well as Legionella urinary antigen testing, as outlined in Tables 2, 35) (12,38,52,106).

Gram stain of the blood cultures may identify some pathogens, which may not be otherwise apparent. For example, Abiotrophia species that grow in blood cultures as chains of gram-positive cocci will fail to grow when subcultured on blood agar plates; these organisms will require subculture on pyridoxal HCl- or l-cysteine-enriched agar or on plates streaked with S. aureus.

In the face of a preceding course of antibiotic, if clinical conditions permit, further antibiotic therapy should be held and blood cultures repeated until positive. The longer the duration since that last dose of antibiotic or the shorter the preceding course of antibiotic, the more likely the blood cultures will be positive. Bacteriuria with either enterococci or S. aureus occurs in infective endocarditis due to the respective organism.

A variety of in vitro tests must be done on the pathogen isolated from blood to assess susceptibility to potential bactericidal drugs (Table 5). Enterococci should be tested for beta-lactamase production (which predicts resistance to anti-enterococcal beta-lactam antibiotics such as penicillin and ampicillin), for high-level-gentamicin and streptomycin resistance (which predicts lack of synergy with a combination of the respective aminoglycoside plus a cell-wall active drug, such as vancomycin, penicillin, or ampicillin), in addition to susceptibility to penicillin and vancomycin (see below). Synergism (see below) between cell wall-active antibiotics and aminoglycosides has been demonstrated for other microorganisms, including viridans streptococci, staphylococci, P. aeruginosa and aerobic enteric gram-negative bacilli in addition to enterococci and its presence can be assessed by a special in vitro test, the so-called time-kill assay.

Assay of peak serum bactericidal activity early in the course of therapy against the patient’s pathogen had been recommended in the past for non-standard regimens or unusual pathogens and, if inadequate, the dose of the antibiotic was increased (although not at the cost of toxicity) and the serum retested. However, the methodology of the serum bactericidal assay in general has not been well standardized, and scant clinical data exists to validate these recommendations. Measurements of vancomycin and aminoglycoside serum levels are helpful to assure adequate, but nontoxic antibiotic levels.

The pathogen should be retained in the laboratory for susceptibility testing to additional antibiotics if the need arises later, and relapse microorganisms or microorganisms obtained from tissue at surgery later in the course of antibiotic therapy should be retested for antimicrobial susceptibility.

back to top


Effective antimicrobial therapy of infective endocarditis optimally requires identification of the specific pathogen and assessment of its susceptibility to various antimicrobial agents. Therefore every effort must be made to isolate the pathogen prior to initiation of antimicrobial therapy, if clinically feasible. In septic patients suspected of having infective endocarditis, empiric antibiotic therapy should be targeted at the most likely pathogens in that particular clinical setting and should include an anti-staphylococcal agent. Standardized regimens have been recommended by the American Heart Association (7) for the most common pathogens, which include viridans streptococci, enterococci, staphylococci and HACEK organisms on native and prosthetic valves (Tables 6a and 6b). The minimal requirements for an effective antimicrobial regimen include:

Bactericidal Activity

Antimicrobial drugs that only inhibit microbial growth (bacteriostatic agents) are not able to clear pathogens from infected tissues if unaided by host defenses, such as polymorphonuclear leukocytes, antibody and complement. Because these host defenses are thought not to operate within vegetations (except in tricuspid valve vegetations, where polymorphonuclear leukocytes may aid the effect of an antimicrobial agent) (45,138), clearance of bacteria from these vegetations requires a bactericidal antibiotic. In fact, eradication of all pathogens from the vegetation is thought to be essential to cure endocarditis. If any bacteria remain after completion of antibiotic therapy, the residual organisms will resume growth and result in relapse. If the pathogen cannot be eliminated completely by antimicrobial therapy, e.g. , if relapse occurs or the patient has persistent bacteremia, the infected vegetation may need to be excised surgically to obtain a cure. For microorganisms without predictable susceptibility, bactericidal activity of an antimicrobial agent for the particular patient’s pathogen must be assessed by determination of the minimal inhibitory (MIC) and minimal bactericidal concentrations (MBC) of the antimicrobial agents in vitro (Table 5).


Unlike streptococci, enterococci are inhibited but not killed by cell wall-active antibiotics such as vancomycin and the anti-enterococcal beta-lactam antibiotics (e.g. , penicillinampicillinamoxicillin, and piperacillin) when these drugs are used alone. Ticarcillinaztreonam, the anti-staphylococcal penicillins, e.g. , nafcillin and methicillin, the cephalosporins, and the carbapenem meropenem have no or limited activity against enterococci. In addition, enterococci are intrinsically resistant to low concentrations of all aminoglycosides, but usually will be killed by either very high concentrations of aminoglycoside alone (concentrations that are too high to achieve clinically without toxicity) or by a combination of a cell wall-active antibiotic plus low concentrations of the aminoglycoside that can be achieved in patients without excessive toxicity. The bactericidal activity of the cell wall-active antibiotic plus aminoglycoside combination (termed synergy) is due to enhanced intracellular penetration of the aminoglycoside caused by the cell wall–active agent (84). Either gentamicin or streptomycin may be used in combination with an anti-enterococcal beta-lactam or vancomycin to treat enterococcal infective endocarditis that exhibit synergism with these combinations. However, some strains of enterococci exhibit high-level resistance (HLR) in vitro to aminoglycosides (2000 mcg/ml of streptomycin or 500 mcg/ml of gentamicin) (15).

HLR is usually due to plasmid- or transposon-mediated aminoglycoside–modifying enzymes (63). Strains of enterococci that exhibit HLR to gentamicin or streptomycin will not exhibit synergy when exposed to the cell wall-active antibiotic combined with the corresponding aminoglycoside at low aminoglycoside concentrations (15). Therefore the strain of enterococcus causing infective endocarditis should be tested for HLR to both gentamicin and streptomycin. Strains with HLR to gentamicin will be resistant to all other aminoglycosides, but may still be sensitive to high levels of streptomycin and strains with HLR to streptomycin may still be sensitive to high levels of gentamicin. If the strain exhibits HLR to only one of these two aminoglycosides, only the particular aminoglycoside to which the strain is sensitive should be used in the cell wall-active antibiotic/aminoglycoside treatment regimen. If the strain is sensitive to high levels of both gentamicin and streptomycin, gentamicin is usually the preferred aminoglycoside for combination therapy because of the greater availability of laboratories that will perform assays of serum concentrations of gentamicin. Also, combinations of a beta-lactam/gentamicin are preferable to vancomycin/gentamicin because of the increased risk of nephrotoxicity with the vancomycin/gentamicin combination (46). E. faecium strains produce low levels of an aminoglycoside-modifying enzyme that inactivates kanamycin, tobramycin and netilmicin, which results in loss of synergy when these aminoglycosides are combined with cell wall-active antibiotics, even in the absence of HLR to these drugs (85). If there is HLR to both gentamicin and streptomycin, no reliable bactericidal antibiotic or combination of antibiotics is currently available with a successful clinical track record.

Based on studies in animal models of enterococcal endocarditis, once daily dosing of the aminoglycoside is not as efficacious as daily divided aminoglycoside dosing (7). Streptomycin dosing is divided into 2 and gentamicin into 3 equally divided doses every 24 hours for 4 to 6 weeks, the longer duration being reserved for those patients with duration of symptomatic illness > 3 months (135), or prosthetic valve endocarditis (see Tables 6a and 6b). Lower doses of gentamicin (3mg/kg daily) are equally efficacious as higher doses and nephrotoxicity is less common with lower gentamicin dosing regimens (17134135136).

Rare strains of enterococcus are beta-lactamase-positive (53,92). Routine MIC-testing of beta-lactamase-positive strains may fail to disclose this type of penicillin-resistance; use of a higher than standard bacterial inoculum for MIC testing or nitrocefin-disk may detect the presence beta-lactamase in these strains. Infective endocarditis caused by beta-lactamase-positive enterococci can be treated with ampicillin/sulbactam or vancomycin. If these strains are also high-level-aminoglycoside-resistant, as is commonly the case, ampicillin/sulbactam or vancomycin is used alone for 8-12 weeks (see Tables 6a and 6b), but relapse rate is high.

E. faecium (MIC 4-32 ug/ml) are more resistant to penicillin than E. faecalis MIC 1-4 ug/ml). At one center, one third of the E. faecium were highly resistant to penicillin [MIC > 200 micrograms/mol) but did not produce beta-lactamase; this high-level intrinsic penicillin resistance resulted in the loss synergism when an aminoglycoside antibiotic agent was combined with penicillin in vitro and in an experimental rat model of endocarditis (14) If the enterococcus has high-level intrinsic penicillin-resistance, vancomycin can be used in combination with an aminoglycoside, if the strain is high-level-aminoglycoside-susceptible (7).

Enterococci have developed vancomycin-resistance (MIC >4ug/ml) as a result of modification of the drug’s target. Of the five phenotypes described, vanA, vanB and vanC are most common: vanA (high level-vancomycin resistance, MIC >64 ug/ml) is more common in E. faecium than E. faecalis; vanB (intermediate to high level resistance, MIC 16-512 ug/ml) is found in both of these species; and vanC (low level resistance. MIC 2-32 ug/ml) is found intrinsically in E. casseliflavus and E. gallinarum. If E. faecium is vancomycin-resistant, infectious disease consultation should be sought, as these strains are usually multidrug-resistant. However, E. faecalis, E. casseliflavus and E. gallinarum are usually penicillin-susceptible. Linezolid and quinupristin/dalfopristin have only inhibitory activity against enterococci, and the activity of quinupristin/dalfopristin is limited to E. faecium (34), but success has been reported for these antibiotics in some patients with enterococcal infective endocarditis (3,5,6,49,61,82,100,127,130,142). Daptomycin, the only available lipopeptide antibiotic, exhibits concentration-dependent bactericidal activity against both E. faecalis and E. faecium (1). Although, daptomycin has been reported to be effective in experimental vancomycin-resistant enterococcal endocarditis (129), clinical data thus far are limited (111) Double beta-lactam combinations (imipenem/ampicillin or cephalosporin/ampicillin) have been successful in experimental models, but clinical experience is limited (7).

back to top


Synergistic combinations of aminoglycoside and beta-lactam are also used to shorten the course of therapy for viridans streptococcal infective endocarditis due to a penicillin-sensitive strain (MIC < 0.1 mcg/ml) from 4 weeks when the beta-lactam (ceftriaxone 2 g IV or IM once daily) is used alone to 2 weeks for the combination therapy (ceftriaxone 2 g IV or IM plus gentamicin 3 mg/kg IV each given as a single daily dose) with comparable results as a result of more rapid bactericidal activity from the combination therapy (112). A single daily dose of ceftriaxone is an attractive alternate to penicillin; because of its long half-life and good potency against these streptococci; serum levels of ceftriaxone unbound to serum protein remain well above the MIC and MBC for over 24 hour. Indeed, two weeks of combined therapy with ceftriaxone plus gentamicin once daily allows outpatient therapy for stable patients with uncomplicated disease due to penicillin-susceptible strains (MIC <0.1 mcg/ml) (118); however, use of the 2-week combination regimen that includes gentamicin is not intended for patients with renal or 8th cranial nerve impairment, and is not preferred for patients >65 years of age. Synergistic combination is also used for treatment of infective endocarditis due to penicillin-intermediate susceptible viridans streptococcal strains (MIC 0.2-0.5 mcg/ml) and infective endocarditis due Abiotrophia and Granulicatella (NVS), and Gemella species (see Tables 6a and 6b). Infective endocarditis due to NVS, Gemella species, and viridans streptococcal strains more resistant to penicillin (MIC >1.0 mcg/ml) are treated with the regimens recommended for enterococcal infective endocarditis (see Tables 6a and 6b).

Although there are limited clinical data evaluating therapeutic regimens for infective endocarditis caused by S. pneumoniae, S. pyogenes, and Groups B, C, and G streptococci, high-dose penicillin or ceftriaxone for 4 weeks is recommended, even for infective endocarditis caused by strains of S. pneumoniae with MIC 0.1 to 4 ug/ml (7). Vancomycin therapy should be administered only to patients with infective endocarditis due to these microorganisms who are unable to tolerate a ß-lactam antibiotic. In general, strains of group B, C, and G streptococci are slightly more resistant to penicillin than are strains of group A streptococci. Some authorities recommend the addition of gentamicin to penicillin for at least the first 2 weeks of a 4- to 6-week course of antimicrobial therapy for group B, C, and G streptococcal infective endocarditis (7). Early cardiac surgical intervention may improve survival rates of patients with S. pneumoniae or ß-hemolytic streptococcal endocarditis (7).

back to top


Staphylococci that are sensitive to methicillin and oxacillin can be treated with vancomycin or an anti-staphylococcal beta-lactam, such as nafcillin or cefazolin, but a beta-lactam is always preferable over vancomycin, as illustrated by the poor response to glycopeptides (teicoplanin or vancomycin) for S. aureus infective endocarditis (18,115). Glycopeptides may be less effective because of poor penetration into vegetations, more rapid clearance in certain patients, relatively slow bactericidal action of vancomycin (42) and the longer duration of S. aureus bacteremia on vancomycin therapy, i.e. , 7-9 days for vancomycin (70) versus 3 days for nafcillin. (62). Vancomycin is recommended for patients with methicillin (or oxacillin)-resistant staphylococci (MRSA and for patients with a history of immediate-type reaction to penicillin. A first generation cephalosporin, such as cefazolin, is recommended for patients with non-immediate-type reactions to penicillin. Ceftriaxone has relatively poor anti-staphylococcal activity (i.e. , the short duration that serum drug levels not protein-bound exceed the MICs that are higher for S. aureus than for other ceftriaxone-sensitive gram-positive pathogens (95) and should not be used for this indication, despite the ease of its once daily administration.

MRSA are cross-resistant to all currently available beta-lactams, including cephalosporins and carbapenems; MRSA that are hospital-acquired are also frequently resistant to other classes of antibiotics, except the glycopeptides, the lipopeptide daptomycin, the oxazolidinone linezolid, and the streptogramins (quinupristin/dalfopristin combination), although occasional emergence of resistance to these antibiotics has been documented (35). However, in the past decade MRSA have been noted to be increasingly community-acquired. These strains unlike nosocomial MRSA, have a distinctive staphylococcal chromosomal cassette (SCC type IV) that encodes methicillin-resistance, have a potent virulence factor (Panton-Valentine toxin), frequently cause severe skin and soft tissue infections, and are usually sensitive to other classes of antibiotics, including the macrolides and clindamycin, although resistance to these and other antibiotics (e.g. , the fluoroquinolones) is now emerging (72).

Because each class of antibiotics has a unique target in the bacterial cell, there will not be cross-resistance with other classes unless there is a common microbial target. For example, macrolides (e.g. , erythromycin), the lincosamide clindamycin and the streptogramin b quinupristin (the so-called macrolide-lincosamide-streptogramin b or MLSb group of antibiotics) share the same ribosomal target. Resistance to these antibiotics can be caused by inducible methylation of the ribosomal target (68). Only erythromycin induces MLSb resistance in S. aureus, so that strains with inducible MLSb resistance will be erythromycin-resistance and clindamycin-sensitive, and will be killed by quinupristin/dalfopristin, as will S. aureus strains that lack MLSb resistance. However, constitutive MLSb-type resistance is a frequent finding in MRSA; these strains are both erythromycin- and clindamycin-resistant, and resistant to the streptogramin b component quinupristin. Quinupristin/dalfopristin will not be bactericidal against these strains that are constitutive MLSb-resistant (140), and clinical experience with MRSA infective endocarditis treated with quinupristin/dalfopristin is limited (20,27,86) Even for clindamycin-sensitive S. aureus, clindamycin therapy of infective endocarditis has a high relapse rate (131).

Both MSSA and nosocomial or community-acquired MRSA are sensitive to daptomycin and linezolid. Daptomycin exhibits concentration-dependent bactericidal activity against S. aureus, and has been reported to be effective in right-sided S. aureus endocarditis (44), but emergence of daptomycin-resistance was noted on daptomycin therapy. In a prospective  study, patients with S. aureus bacteremia with or without endocarditis were randomized to receive monotherapy with 6 mg/kg of daptomycin intravenously per kilogram daily or combination therapy with initial low-dose gentamicin plus with an anti-staphylococcal penicillin or vacomycin. A successful outcome was documented for 44% of patients who received daptomycin as compared with 42% of patients who received standard therapy. The success rates were similar in subgroups of patients with complicated bacteremia, right-sided endocarditis, and MRSA. As compared with daptomycin therapy, standard therapy was associated with a non-significantly higher rate of adverse events that led to treatment failure due to the discontinuation of therapy. Clinically significant renal dysfunction occurred in 11% of patients who received daptomycin and in 26% of patients who received standard therapy (44). The authors concluded that daptomycin was non-inferior to comparator agents. Weekly CPK levels should be monitored as this is a potential side effect of therapy. In addition, daptomycin penetrates lung poorly (113), and may not be effective if there is hematogenous spread of S. aureus to the lung. Also, although linezolid penetrates the lung well, it has only limited bactericidal activity, and clinical experience is limited (37).

Clinical experience with doxycycline or minocycline for S. aureus infective endocarditis is limited (67). Many strains of MRSA are sensitive to trimethoprim-sulfamethoxazole, which has been shown to be effective in one series of cases (81), in the event that vancomycin is not tolerated. Strains should be tested for susceptibility to rifampin or gentamicin, if their use is planned as adjunctive therapy in an attempt to enhance bactericidal activity of vancomycin or an anti-staphylococcal beta-lactam to treat staphylococcal PVE; however, the effectiveness of gentamicin or rifampin used in this manner for staphylococcal native valve endocarditis has been questioned (62,70), although their use combined with vancomycin is substantiated in experimental endocarditis and clinical experience for prosthetic valve endocarditis (PVE) due to methicillin-resistant coagulase-negative staphylococci and by analogy combined with methicillin for PVE due to MSSA and methicillin-sensitive coagulase-negative staphylococci (56). The clinical efficacy of various treatment regimens (e.g. , daptomycin, linezolid, quinupristin/dalfopristin, etc) for infective endocarditis due to staphylococcal strains partially or fully resistant to vancomycin has not been established.

Six weeks of antibiotic therapy is usually recommended for S. aureus infective endocarditis, because of the common presence of purulent complications, such as perivalvular abscess or extracardiac metastatic abscesses, although 4 weeks may be sufficient for uncomplicated infection.

back to top

Other Pathogens

Therapy for endocarditis due to gram-negative bacilli other than the HACEK group, anaerobes, and diphtheroids should be developed in consultation with an infectious diseases specialist. Bactericidal activity for anaerobic gram-negative bacilli can frequently be achieved with metronidazole; for diphtheroids with vancomycin-aminoglycoside combination: and for aerobic enteric gram-negative bacilli or P. aeruginosa with a beta-lactam-aminoglycoside combination or a fluoroquinolone such as ciprofloxacin, but emergence of resistance during antimicrobial therapy may be a problem (141). Gram-negative bacilli known to be expressing AmpC beta-lactamase or an extended-spectrum beta-lactamase are best treated with a carbapenem or fluoroquinolone, although resistance to these drugs is also increasing (98). Bartonella endocarditis can be treated with doxycycline 100 mg every 12h for 6 weeks plus gentamicin 1 mg/kg every 8h for the first 2 weeks, or if gentamicin cannot be used plus rifampin 300 mg po or IV every 12h (7).

Blood Culture-Negative Infective Endocarditis

Blood culture-negative native valve endocarditis in patients who have received antibiotic before blood cultures are obtained is generally treated empirically with antibiotic regimens that will cover S. aureus, viridans streptococci, enterococci and HACEK microorganisms: ampicillin/sulbactam 12g/24h IV in 4 equally divided doses plus gentamicin 3mg/kg/24h in 3 equally divided doses for 4-6 weeks or in patients unable to tolerate penicillins vancomycin 15 mg/kg every 12h, ciprofloxacin 500 -750 mg po or 400 mg IV every 12h, and gentamicin 3mg/kg/24h in 3 equally divided doses for 4-6 weeks (7). Blood culture-negative early PVE is treated empirically with vancomycin 15 mg/kg every 12h IV plus cefepime 2g every 8h IV and rifampin 300 mg po or IV every 8h for at least 6 weeks and gentamicin 3mg/kg/24h in 3 equally divided doses for the first 2 weeks (7). Blood culture-negative late PVE is treated empirically with the same regimens recommended for blood culture-negative native valve endocarditis above for at least 6 weeks (7).

Fungal Infective Endocarditis

Aspergillus more commonly than Candida can cause blood culture-negative endocarditis (Candida species routinely grow on routine blood cultures) while Candida can cause endocarditis in patients with central venous catheters or prosthetic valves. When suspected, treatment includes amphotericin B or an echinocandin for 6 weeks and valve replacement. If Candida species are isolated, appropriate azole (fluconazole for C. albicans) or echinocandin (forC. glabrata) therapy should be the mainstay of therapy based on susceptibility testing or predicted susceptibilities based on species. After an initial clinical response, long-term suppression with an oral azole should be considered after completion of combined medical/surgical management or after medical therapy alone if the patient is not a surgical candidate (7).

back to top

High concentrations of the antimicrobial agent in the vegetation

Optimally the antimicrobial agent should have a minimal “inoculum effect,” i.e. , should exhibit the least reduction in potency when tested against high microbial densities of 108-11 CFU/g that exist in vegetations; these microbial densities are higher than the standard inocula of 105-6 CFU/ml used to perform the MIC and MBC test. Large doses of beta-lactam antibiotics must be used to overcome the inoculum effect, a characteristic of these antibiotics. Doses of the antimicrobial agent must also be large enough to achieve high concentrations of the antimicrobial agent in the blood that will facilitate passive diffusion of the antimicrobial agent into the depths of the vegetation where the microcolonies of the pathogen are located. Despite high concentrations in blood, some antimicrobial agents fail to penetrate vegetations deeply enough, e.g. , amphotericin B, or penetrate vegetations unevenly, e.g. , teicoplanin (24,107). Since amphotericin B is the only fungicidal agent, cure of fungal endocarditis usually requires surgery or chronic suppressive therapy after an initial response to an anti-fungal agent in patients who are too ill to undergo valve replacement.

Prolonged duration of antimicrobial therapy

Duration of therapy varies with the specific pathogen, the site of the infection, and type of antibiotic. For example, bacterial clearance is more rapid for viridans streptococci than staphylococci, in tricuspid versus aortic vegetations, with anti-staphylococcal beta-lactams than vancomycin or with combinations of cell wall-active agent plus aminoglycoside than cell wall-active agent alone. More rapid clearance in these special circumstances may permit a shorter course of therapy to achieve cure (see above).

Over 90% of the microbial population in the vegetation is non-growing and metabolically inactive once the infection has become well established (29). Non-growing organisms are more likely to be found in the central portions of the microcolonies in the deeper regions of the vegetation. Because the beta-lactams are only active against growing microorganisms, which express the targets of these antibiotics (penicillin-binding proteins) (124), each dose is only able to reduce the microbial count in that small portion (less than 10%) of the population that happens to be growing at the time of drug administration, which results in a slow rate of bactericidal action (54). The duration of drug therapy therefore must be prolonged in order to result in complete clearance of the pathogen from the vegetation.

The duration of antimicrobial therapy after valve replacement depends to some extent on evidence for active infection at the time of surgery. In patients with positive intraoperative cultures or gram stain, a full course of post-operative therapy is reasonable, otherwise an additional 2 weeks and for those operated near the end of the standard period of treatment, simply completing the time remaining on the regimen for that type endocarditis (90).

Dosing should be frequent enough to prevent resumption of microbial growth between doses

The microorganisms that remain after a brief in vitro exposure to an antibiotic frequently exhibit a post-exposure delay in further growth, the so-called post-antibiotic effect (73). Unfortunately, no such effect occurs with enterococci or P. aeruginosa in the rat model of endocarditis, despite in vitro demonstration of a post-antibiotic effect (50,51). Thus, even though a bactericidal effect can achieved in the vegetation in the early portions of a dosing interval when vegetation levels of the drug are high, if antibiotic vegetation levels are not maintained at least above the MIC during the rest of the dosing interval, resumption of growth of residual organisms may occur and efficacy may be compromised. This requires relatively short dosing intervals for drugs with short serum half lives, especially for microorganisms with high MICs (54).

back to top


Replacement of an infected native valve with a valvular prosthesis is indicated in the following situations (Table 7).

1) CHF secondary to valvular dysfunction. CHF in infective endocarditis indicates a grave prognosis. In those patients with CHF, cardiac valve replacement should not be delayed to allow further antibiotic therapy. Delayed surgery when associated with worsening of CHF increases operative mortality from 6-8% in patients with mild or no CHF to 17-33% with severe CHF. The incidence of reinfection of a newly placed cardiac valvular prosthesis is estimated to be 2-3%, which is far less than the mortality associated with uncontrolled CHF. Although operative mortality and PVE is higher when a prosthetic valve is implanted in the presence rather than absence of active infection, the overall outcome is better if the valve replacement is prompt, before severe CHF or spread of infection into the perivalvular tissues ensues.

2) Multiple clinically significant emboli, despite antibiotic therapy for two weeks. However the first or second embolic episode may so impair the patient, that prosthetic valve replacement at that point may be futile. Use of a variety of factors to predict significant embolization, such as the large size or continued enlargement of vegetations on medical therapy, location on anterior leaflet of the mitral valve, the pathogen being fungi or S. aureus, as indicators for valve replacement remains unresolved.

3) Endocarditis due to certain pathogens that rarely respond to medical therapy alone such as fungi, enterococci for which there is no synergistic bactericidal combination (e.g. , high-level aminoglycoside-resistant enterococci and beta-lactam or fluoroquinolone resistant gram-negative bacilli.

4) Uncontrolled bacteremia despite optimal antibiotic therapy. However, it should be remembered that the average duration of S. aureus bacteremia on vancomycin therapy is 8 days, with bacteremia persisting for several weeks in some patients before its ultimate resolution.

5) Valve ring abscess. Although valve ring abscess, especially if small (<2 cm) and not complicated by fistulous tracts, rarely may heal on antimicrobial therapy alone (64), surgery is indicated for progressive enlargement of a valve ring abscess on antibiotic therapy when monitored by transesophageal echocardiography (121). Because of extensive periannular tissue destruction in some patients, human aortic homograft may be required to replace the destroyed valve and supporting tissues. Patients with valve ring abscess must also be monitored closely for the development of conduction abnormalities, which may require placement of a transvenous pacemaker because of the risk of high-grade heart block.

To avoid the complications of prosthetic valve placement (e.g. , PVE, bleeding, thrombo-embolic events, and prosthetic valve deterioration over time), new surgical options for treatment of native valve infective endocarditis have been proposed as an alternative to placement of a prosthetic valve; these options include valve debridement, valvuloplasty, and repair or replacement of the paravalvular structure with pulmonary root autograft. Presence of a prosthetic valve in an intravenous drug users is problematic, because the prosthetic valve places these frequently noncompliant patients at continued high risk of PVE. Alternatively, tricuspid valve resection without placement of a prosthetic valve can be tolerated hemodynamically for extended periods of time in many of these patients.

The surgical indications for PVE are the same as those outlined for native valve endocarditis above. Some patients with PVE without evidence of infection at the valve annulus may be treated medically.

Tissue should be removed at surgery for microbiologic evaluation (e.g. , gram or other histochemical staining, culture and other studies, such as PCR when indicated. The presence of microorganisms in surgical specimens will determine duration of post-operative antibiotic therapy (see above), and retesting their antimicrobial susceptibility will evaluate the emergence of antibiotic resistance on therapy.

Intrathoracic, intra-abdominal or peripheral mycotic aneurysms usually require surgical excision or endovascular stenting. Intracranial mycotic aneurysms should be followed closely with serial angiograms or MRI/MRA (23,128,133). Although some small intracranial aneurysms may heal on medical therapy alone, most will usually require prompt surgical clipping or ligation or endovascular stenting as an alternative to surgery if accessible and enlarging or bleeding. Myocardial revascularization should be performed at the time of elective valve surgery if significant coronary artery disease is present. However, patients who require emergency placement of prosthetic valve for hemodynamic decompensation secondary to acute infective endocarditis may not tolerate the radiocontrast load necessary for coronary angiography and the additional bypass surgery.

Embolization of vegetations may result in bland splenic infarcts or septic infarcts/abscesses. Contrast-enhanced abdominal computerized tomography and MRI usually can distinguish a septic from bland process. Ultrasound may show sonolucent areas when abscesses are present. Septic infarcts and abscesses, which can cause persistent or recurrent fever and bacteremia despite appropriate antimicrobial therapy, will usually require splenectomy, with increasing use of percutaneous drainage as an alternative when feasible in patients who are poor surgical candidates (16). Pneumococcal vaccination should be done prior to splenectomy, and splenectomy should be planned prior to anticipated prosthetic valve placement to prevent hematogenous microbial seeding of the newly placed prosthetic valve from the splenic abscesses (114).

back to top


Although anticoagulant therapy may impede further enlargement of vegetations in experimental models of infective endocarditis, anticoagulant therapy is relatively contraindicated in infective endocarditis due to the increased risk of intracranial hemorrhage from either occult mycotic aneurysms, cerebral emboli, or cerebral immune vasculitis. Anticoagulation may be used for an over-riding indication that is separate from infective endocarditis, but for deep vein thrombophlebitis of the lower extremities, an inferior vena caval filter would be preferable to anticoagulation. Anticoagulation is particularly problematic when S. aureus is the pathogen because of the increased risk of cerebral embolism with this organism, and in patients with infective endocarditis who undergo prosthetic valve replacement within 2-4 weeks after a neurologic event. In these later patients, it has been suggested that valve surgery be delayed for a minimum of 2 weeks after either a cerebral embolus or bleed or repair of intracranial mycotic aneurysm and use of a bioprosthesis that will not require post-operative continuation of anticoagulation is preferable to use of a mechanical prosthesis that will require further anticoagulation. Similarly, anticoagulation should be discontinued in patients with S. aureus PVE for at least 2 weeks after a cerebral embolus to prevent acute hemorrhagic transformation of the cerebral infarct.

Although aspirin has shown a beneficial effect in experimental S. aureus endocarditis (65), in a randomized trial of aspirin in patients with infective endocarditis, aspirin had no effect on vegetation resolution and valvular dysfunction, did not reduce the risk of embolic events and was likely to be associated with an increased risk of bleeding (19).

back to top


Use of shorter courses of antibiotic therapy, administration of parenteral antibiotic therapy at home, and oral regimens have been investigated in selected groups of patients to shorten the length of hospitalization (87). Before considering outpatient therapy, most patients should first be evaluated and stabilized in the hospital; only very rarely can some patients be managed entirely as outpatients. Patients can be selected for administration of parenteral therapy at home by their being at low risk for complications of endocarditis: The presence of poorly controlled CHF, neurological findings that may result from systemic emboli or bleeding MAs, cardiac conduction abnormalities, valve ring abscesses (usually detected by TEE), persistent fever, positive blood cultures, and prosthetic valve endocarditis preclude home intravenous therapy. Having a focal infection that in itself would require more than 2 weeks of antimicrobial therapy, PVE, and renal or eighth nerve impairment would preclude use of short-course beta-lactam-aminoglycoside combination therapy for streptococcal or uncomplicated tricuspid valve S. aureus infective endocarditis. Patients with penicillin-susceptible viridans streptococcalendocarditis (penicillin MIC <0.1 µg/mol) whose disease is apparent for <3 months and is uncomplicated at the time of admission generally do well with outpatient intravenous therapy (i.e. , 2 weeks of ß-lactam plus an aminoglycoside or 4 weeks of a ß-lactam such as ceftriaxone) alone. Because of the unreliable absorption of orally administered agents, oral therapy is generally not recommended. However, use of central intravenous catheters in patients, especially intravenous drug users, with infective endocarditis places them at risk of IV line sepsis and superinfection infective endocarditis. While on home antimicrobial therapy, patients must be monitored closely by a home infusion team and patients must have ready access to experienced physicians to monitor periodically (e.g. , weekly) for development of complications of infective endocarditis and drug-related side effects of antimicrobial therapy (e.g. , vestibular, auditory, and nephrotoxicity from aminoglycosides, leukopenia and thrombocytopenia from ß-lactams and vancomycin, and nephrotoxicity from the combination of vancomycin and gentamicin). The standard regimens used to treat penicillin-sensitive streptococcal infective endocarditis require either continuous infusion of penicillin or frequent intravenous administration.

back to top


Once on appropriate antimicrobial therapy, most patients will note a sense of well-being, lessened fatigue, and improved appetite, and the temperature will usually fall to normal levels within 2-5 days; the ESR, anemia and renal function may take weeks to months to improve (74). Circulating immune complexes and related serologic findings that include hypocomplementemia, mixed cryoglobulinemia and rheumatoid factor, also tend to resolve gradually with effective antibiotic therapy (8). A variety of tests are done to monitor both the antimicrobial effects (see below) and the potential adverse reactions to the drugs used to treat the infection, which include neutropenia (beta-lactams, vancomycin, linezolid), thrombocytopenia (beta-lactams, vancomycin, linezolid), azotemia (aminoglycosides and much less commonly vancomycin, trimethoprim/sulfamethoxazole, and beta-lactams), elevated CPK (daptomycin) and hyperkalemia (trimethoprim/sulfamethoxazole). Drug interactions are also problematic for vancomycin with gentamicin (azotemia) and rifampin with multiple drugs, e.g. , diminished prothrombin time with warfarin.

After antimicrobial therapy is started, blood cultures should be repeated daily or every other day until sterile to assess for clearance of bacteremia. Blood cultures for streptococci and enterococci should become sterile after 1-2 days of appropriate therapy and for S. aureus, after 3-5 days; however, with vancomycin therapy as noted above, blood cultures for S. aureus may take 1 to 2 weeks to become sterile. If no organism is isolated from blood, but there is a good clinical response to the antimicrobial regimen, these antibiotics should be continued. If no organism is isolated and there is no clinical response to empiric therapy after 1-2 weeks, infective endocarditis due to a fastidious pathogen, e.g. , fungi, anaerobes, etc. , or a diagnosis other than infective endocarditis should be considered, such as antiphospholid antibody syndrome (APA); APA, which is frequently complicated by cerebral emboli, requires anticoagulant, rather than antibiotic therapy (48).

If the pathogen is initially isolated from blood, and appropriate antimicrobial therapy started, but fever persists or recurs, blood cultures should be repeated to assess persistent or relapsing infection; other possibilities (Table 8), which include most commonly pulmonary or systemic embolization. Blood cultures are repeated two and four weeks after therapy has been completed, because relapse is most common within one month. Relapse rate for native valve endocarditis caused by penicillin-susceptible viridans streptococci is <1-2%, for aminoglycoside-susceptible enterococci, 8-12%, and higher for S. aureus, other pathogens, and PVE treated by the recommended regimens (Tables 6a and 6b) (109).

back to top


Factors that affect mortality (47) include the infecting organism (the mortality of endocarditis due to fungi, Pseudomonas aeruginosa, and aerobic enteric gram-negative bacilli > staphylococci > enterococci > viridans streptococci) (83), the site of infection (aortic alone or aortic plus mitral > mitral alone > tricuspid infection), PVE vs. native valve endocarditis (early onset PVE > late onset PVE > native valve endocarditis), age (higher in the elderly and very young) and gender (men > women), and the presence of certain complications, such as heart or renal failure, rupture of a mycotic aneurysm, cardiac arrhythmias and conduction abnormalities, perivalvular extension, cerebral emboli, and perhaps severe immunosuppression due to HIV infection. Heart failure remains the leading cause of death. However, with increasing use of prosthetic valve replacement for heart failure, the leading cause of death may shift to neurologic complications due to embolic episodes or mycotic aneurysms (79), or uncontrolled infection due to antibiotic resistant microorganisms. Mortality is higher for patients with shorter durations of illness before initiation of antibiotic therapy, perhaps by the fact that acute infective endocarditis is frequently due to S. aureus (47,110). Following cure of one episode of endocarditis, patients remain at greatly increased risk for reinfection and PVE, and their increasing age contributes to the poor prognosis during the follow-up (80,122).


The effect of infective endocarditis prophylaxis with antimicrobial agents has been estimated to be very modest, i.e. , <10% of all cases are preventable by prophylaxis, assuming all patients with cardiac risk factors are given appropriate antibiotic prophylaxis and also assuming that the prophylaxis regimen is 100% effective (58). For example, only about 1/2 of cases have recognizable predisposing cardiac lesions, most cases of infective endocarditis do not occur following an invasive procedure, and less than about 2/3 of cases are due to microorganisms (viridans streptococci and enterococci) against which prophylactic regimens are directed. However in those patients who are known to have a risky cardiac lesion (Table 9) and are to undergo certain procedures (Table 10), with subsequent disruption of local bacteria that are capable of producing endocarditis and have predictable susceptibility to antibiotics with minimal inconvenience, toxicity and cost, the American Heart Association (AHA) has made recommendations shown in Tables 11a11b (25).The 2007 AHA Guidelines no longer recommend endocarditis prophylaxis for esophagogastroduodenoscopy, colonoscopy, previous CABG, congenital heart disease not listed in Table 9, pacemakers, or routine genitourinary procedures due to the risks of antibiotic therapy outweighing the benefits. If a patient is already receiving long-term antibiotic therapy with an antibiotic that is also recommended for IE prophylaxis for a dental procedure, it is prudent to select an antibiotic from a different class rather than to increase the dosage of the current antibiotic.

Additional preventive measures are avoid unnecessary use of intravascular catheters (a major predisposing event for nosocomial endocarditis), aggressively treat focal infections early with appropriate antibiotic therapy and surgical drainage, obtain at least 3 blood cultures for any febrile illness and before initiation of antibiotic therapy, and maintain good dental hygiene in patients at increased risk for infective endocarditis.

back to top


1. Akins RL, Rybak MJ. Bactericidal activities of two daptomycin regimens against clinical strains of glycopeptide intermediate-resistant Staphylococcus aureus, vancomycin-resistant Enterococcus faecium, and methicillin-resistant Staphylococcus aureus isolates in an in vitro pharmacodynamic model with simulated endocardial vegetations. Antimicrob Agents Chemother 2001;45:454-9. [PubMed] 

2. Anderson DJ, Murdoch DR, Sexton DJ, et al. Risk factors for infective endocarditis in patients with enterococcal bacteremia: a case-control study.Infection 2004;32:72-7. [PubMed]

3. Ang JY, Lua JL, Turner DR, Asmar BI. Vancomycin-resistant Enterococcus faecium endocarditis in a premature infant successfully treated with linezolid. Pediatr Infect Dis J 2003;22:1101-3. [PubMed]

4. Angrist AA, Oka M. Pathogenesis of bacterial endocarditis. JAMA 1063; 183: 249-252. [PubMed]

5. Archuleta S, Murphy B, Keller MJ. Successful treatment of vancomycin-resistant Enterococcus faecium endocarditis with linezolid in a renal transplant recipient with human immunodeficiency virus infection. Transpl Infect Dis 2004;6:117-9. [PubMed]

6. Babcock HM, Ritchie DJ, Christiansen E, et al. Successful treatment of vancomycin-resistant Enterococcus endocarditis with oral linezolid. Clin Infect Dis 2001;32:1373-5. [PubMed]

7. Baddour LM, Wilson WR, Bayer AS, et al. Infective endocarditis. Diagnosis, Antimicrobial therapy, and management of complications. Circ 2005;111:3167-84. [PubMed]

8. Bayer AS, Theofilopoulos AN, Eisenberg R, et al. Circulating immune complexes in infective endocarditis. N Engl J Med 1976;295:1500-5. [PubMed]

9. Beeson PB, Brannon ES, Warren JV. Observations on the sites of removal of bacteria   from the blood in patients with bacterial endocarditis. J Exper Med 1945; 81; 9-23. and Rev Infect Dis 1985;7:565-73. [PubMed]

10. Berlin JA, Abrutyn E, Strom BL, et al. Incidence of infective endocarditis in the Delaware Valley, 1988-1990. Am J Cardiol 1995;76:933-6. [PubMed]

11. Brouqui, P, Raoult D. Endocarditis due to rare and fastidious bacteria. Clin. Microbiol. Rev 2001;14:177-207. [PubMed]

12. Brouqui P, Raoult D. New insight into the diagnosis of fastidious bacterial endocarditis. FEMS Immunol Med Microbiol 2006;47:1-13. [PubMed]

13. Burnette-Curley D, Wells V, Viscount H, et al. FimA, a major virulence factor associated with Streptococcus parasanguis endocarditis. Infect Immun 1995; 63: 4669-79. [PubMed]

14. Bush LM, Calmon J, Cherney CL, et al. High-level penicillin resistance among isolates of enterococci. Implications for treatment of enterococcal infections. Ann Intern Med 1989;110:515-20. [PubMed]

15. Calderwood SA, Wennersten C, Moellering RC Jr, et al. Resistance to six aminoglycoside aminocyclitol antibiotics among enterococci: prevalence, evolution, and relationship to synergism with penicillin. Antimicrob Agents Chemother 1977;12:401-5. [PubMed]

16. Carbonell AM, Kercher KW, Matthews BD, et al. Laparoscopic splenectomy for splenic abscess. Surg Laparosc Endosc Percutan Tech 2004;14:289-91. [PubMed] 

17. Carrizosa J, Levison ME. Minimal concentrations of aminoglycoside that can synergize with penicillin in enterococcal endocarditis. Antimicrob Agents Chemother 1981;20:405-9. [PubMed]

18. Chambers HF, Miller RT, Newman MD. Right-sided Staphylococcus aureus endocarditis in intravenous drug users: two-week combination therapy. Ann Intern Med 1988;109:619-24. [PubMed]

19. Chan KL, Dumesnil JG, Cujec B, et al. Investigators of the Multicenter Aspirin Study in Infective Endocarditis. A randomized trial of aspirin on the risk of embolic events in patients with infective endocarditis. J Am Coll Cardiol 2003;42:775-80. [PubMed]

20. Chesi G, Colli A, Mestres CA, et al. Multiresistant-MRSA tricuspid valve infective endocarditis with ancient osteomyelitis locus. BMC Infect Dis 2006;6:124. [PubMed]

21. Choudhury R, Grover A, Varma J, et al. Active infective endocarditis observed in an Indian hospital 1981-1991.Am J Cardiol 1992;70:1453-8. [PubMed]  

22. Cohen PS, Maguire JH, Weinstein L. Infective endocarditis caused by gram-negative bacteria: A review of the literature, 1945-77. Prog Cardiovasc Dis 1980; 22: 205-42. [PubMed]

23. Corr P, Wright M, Handler LC. Endocarditis-related cerebral aneurysms: radiologic changes with treatment. Am J Neuroradiol 1995;16:745-8. [PubMed]

24. Cremieux AC, Maziere B, Vallois JM, et al. Evaluation of antibiotic diffusion into cardiac vegetations by quantitative autoradiography. J Infect Dis 1989;159:938-44. [PubMed]

25. Wilson W, Taubert KA, Gewitz M, Lockhart PB, Baddour LM, et al. Prevention of Infective Endocarditis. Guidelines from the American Heart Association. Circulation April 2007. Epub [PubMed]

26. Di Filippo S, Delahaye F, Semiond B, et al. Current patterns of infective endocarditis in congenital heart disease. Heart 2006;92:1490-5. [PubMed]

27. Drew RH, Perfect JR, Srinath L, Kurkimilis E, Dowzicky M, Talbot GH. Treatment of methicillin-resistant staphylococcus aureus infections with quinupristin-dalfopristin in patients intolerant of or failing prior therapy. For the Synercid Emergency-Use Study Group. J Antimicrob Chemother 2000;46:775-84. [PubMed]

28. Durack DT, Beeson PB. Experimental bacterial endocarditis I: Colonization of a sterile lesion. Br J Exp Pathol 1972; 53:44-9. [PubMed]

29. Durack DT, Beeson PB. Experimental bacterial endocarditis. II. Survival of a bacteria in endocardial vegetations. Br J Exp Pathol 1972;53:50-3. [PubMed]

30. Durack DT. Experimental bacterial endocarditis. IV. Structure and evolution of very early lesions. J Pathol 1975;115:81-9. [PubMed]

31. Durack DT, Beeson PB. Protective role of complement in experimental Escherichia coli endocarditis. Infect Immun 1977;16:213-7. [PubMed]

32. Durack DT, Lukes AS, Bright DK; Duke Endocarditis Service. New criteria for diagnosis of infective endocarditis: utilization of specific echocardiographic findings. Am J Med 1994;96:200–9. [PubMed]

33. Duval X, Selton-Suty C, Alla F, et al. Association pour l'Etude et la Prevention de l'Endocardite Infectieuse. Endocarditis in patients with a permanent pacemaker: a 1-year epidemiological survey on infective endocarditis due to valvular and/or pacemaker infection. Clin Infect Dis 2004;39:68-74. [PubMed]

34. Eliopoulos GM. Quinupristin-dalfopristin and linezolid: evidence and opinion. Clin Infect Dis 2003;36:473-81. [PubMed]

35. Eliopoulos GM. Antimicrobial agents for treatment of serious infections caused by resistant Staphylococcus aureus and enterococci. Eur J Clin Microbiol Infect Dis 2005;24:826-31. [PubMed]

36. Elliot SD. Bacteremia and oral sepsis. Proc R Soc Med 1939;32:747-54

37. Falagas ME, Manta KG, Ntziora F, Vardakas KZ. Linezolid for the treatment of patients with endocarditis: a systematic review of the published evidence. J Antimicrob Chemother 2006;58:273-80. [PubMed]

38. Fenollar F, Raoult D. Molecular genetic methods for the diagnosis of fastidious microorganisms. APMIS 2004;112:785-807. [PubMed]

39. Fernandez Guerrero ML, Aguado JM, Arribas A, et al. The spectrum of cardiovascular infections due to Salmonella enterica: a review of clinical features and factors determining outcome. Medicine (Baltimore) 2004;83:123-38. [PubMed]

40. Fernandez-Guerrero ML, Verdejo C, Azofra J, de Gorgolas M. Hospital-acquired infectious endocarditis not associated with cardiac surgery: an emerging problem. Clin Infect Dis 1995;20:16-23. [PubMed]

41. Fernandez-Guerrero ML, Herrero L, Bellver M, et al. Nosocomial enterococcal endocarditis: a serious hazard for hospitalized patients with enterococcal bacteraemia. J Intern Med 2002;252:510-5. [PubMed]

42. Flandrois JP, Fardel G, Carret G. Early stages of in vitro killing curve of LY146032 and vancomycin for Staphylococcus aureus. Antimicrob Agents Chemother 1988;32:454-7. [PubMed]

43. Fowler VG Jr, Miro JM, Hoen B, et al; ICE Investigators. Staphylococcus aureus endocarditis: a consequence of medical progress. JAMA 2005;293:3012-21. [PubMed]

44. Fowler VG Jr, Boucher HW, Corey GR, et al;  S. aureus Endocarditis and Bacteremia Study Group. Daptomycin versus standard therapy for bacteremia and endocarditis caused by Staphylococcus aureus. N Engl J Med 2006;355:653-65. [PubMed]

45. Francioli PB, Freedman LR. Streptococcal infection of endocardial and other intravascular vegetations in rabbits: natural history and effect of dexamethasone. Infect Immun 1979;24:483-91. [PubMed]

46. Goetz MB, Sayers J. Nephrotoxicity of vancomycin and aminoglycoside therapy separately and in combination. J Antimicrob Chemother 1993;32:325-34. [PubMed]

47. Gold MJ. Cure rates and long-term prognosis. In: Kaye D, ed. Infective Endocarditis, 2nd ed. New York: Raven Press. 1992: 455-64.

48. Gonzalez-Juanatey C, Gonzalez-Gay MA. Libman-sacks endocarditis and primary antiphospholipid syndrome. J Heart Valve Dis 2005;14:700-2. [PubMed]

49. Hamza N, Ortiz J, Bonomo RA. Isolated pulmonic valve infective endocarditis: a persistent challenge. Infection 2004;32:170-5. [PubMed]

50. Hessen MT, Pitsakis PG, Levison ME. Post Antibiotic Effect of Penicillin plus Gentamicin versus Enterococcus faecalis In Vitro and In Vivo. Antimicrob Agents Chemother 1989;33:608-ll. [PubMed] 

51. Hessen MT, Pitsakis PG, Levison ME. Absence of a Post Antibiotic Effect in Experimental Right and Left Sided Pseudomonas aeruginosa Endocarditis Treated with Imipenem, with or without Gentamicin. J Infect Dis 1988;l58:542-8. [PubMed]

52. Houpikian P, Raoult D. Diagnostic methods. Current best practices and guidelines for identification of difficult-to-culture pathogens in infective endocarditis. Cardiol Clin 2003;21:207-17. [PubMed]

53. Ingerman M, Pitsakis PG, Rosenberg A, Hessen MT, Abrutyn E, Murray BE, Levison ME. Beta-lactamase production in experimental endocarditis due to aminoglycoside-resistant Streptococcus faecalis. J Infect Dis 1987;155:1226-32. [PubMed]

54. Ingerman MJ, Pitsakis PG, Rosenberg AF, Levison ME. The importance of pharmacodynamic effects (Rate of Bactericidal Action and Post-Antibiotic Effect) in determining the dosing interval in therapy for experimental Pseudomonas endocarditis in the rat. J Infect Dis 1986; l53:707-7l4. [PubMed]

55. Johnson WD. The clinical syndrome. In: Kaye D ed. Infective endocarditis. Baltimore MD: University Park Press. 1976:87-100.

56. Karchmer AW, Archer GL, Dismukes WE. Staphylococcus epidermidis causing prosthetic valve endocarditis: microbiologic and clinical observations as guides to therapy. Ann Intern Med. 1983 Apr;98(4):447-55. [PubMed]

57. Kaye D. Definitions and Demographic Characteristics. In: Kaye D ed. Infective endocarditis. Baltimore MD: University Park Press. 1976:1-10.

58. Kaye D, Abrutyn E. Prevention of bacterial endocarditis: 1991. Ann Intern Med 1991;114:803-4. [PubMed]

59. Kerr A. Subacute bacterial endocarditis. Chapter II. Definitions and Incidence. Springfield Ill Charles C Thomas. 1955: 31-4.

60. Kerr A. Subacute bacterial endocarditis. Chapter II. Definitions and Incidence. Springfield Ill Charles C Thomas. 1955:31-4.

61. Konstantinov IE, Zehr KJ. Aortic root replacement in a patient with vancomycin-resistant Enterococcus faecium endocarditis and leukemia. Chest 2001;120:1744-6. [PubMed]

62. Korzeniowski O, Sande MA. Combination antimicrobial therapy for Staphylococcus aureus endocarditis in patients addicted to parenteral drugs and in nonaddicts: A prospective study. Ann Intern Med 1982;97:496-503. [PubMed]

63. Krogstad DJ, Korfhagen TR, Moellering RC Jr, et al. Aminoglycoside-inactivating enzymes in clinical isolates of Streptococcus faecalis. An explanation for resistance to antibiotic synergism. J Clin Invest 1978;62:480-6. [PubMed]

64. Kunis RI, Sherrid MV, McCabe JG, et al. Successful medical therapy of mitral annular abscess complicating infective endocarditis. J Am Coll Cardiol 1986; 7: 953-5.

65. Kupferwasser LI, Yeaman MR, Shapiro SM, et al. Acetylsalicylic acid reduces vegetation bacterial density, hematogenous bacterial dissemination, and frequency of embolic events in experimental Staphylococcus aureus endocarditis through antiplatelet and antibacterial effects. Circulation. 1999;99:2791-7. [PubMed]

66. Lacassin FHoen BLeport C, et al Procedure associated with infective endocarditis in adults. A case control study. Eur Heart J 1995;16:1968-7. [PubMed]

67. Lawlor MT, Sullivan MC, Levitz RE, et al. Treatment of prosthetic valve endocarditis due to methicillin-resistant Staphylococcus aureus with minocycline. J Infect Dis 100;161:812-14. [PubMed]

68. Leclercq R, Courvalin P. Bacterial resistance to macrolide, lincosamide, and streptogramin antibiotics by target modification. Antimicrob Agents Chemother 1991;35:1267-72. [PubMed]

69. Levine DP, Crane LR, Zervos MJ. Bacteremia in narcotic addicts at the Detroit Medical Center. II. Infectious Endocarditis: a prospective study. Rev Infect Dis. 1986; 8:374-96. [PubMed]

70. Levine DP, Fromm BS, Reddy BR. Slow response to vancomycin or vancomycin plus rifampin in methicillin-resistant Staphylococcus aureus endocarditis. Ann Intern Med 1991;115:674-80. [PubMed]

71. Levison ME. Pathogenesis of endocarditis In: Kaye D ed. Infective endocarditis. Baltimore MD: University Park Press. 1976: 29-41.

72. Levison ME, Fung S. Community-associated methicillin-resistant Staphylococcus aureus: Reconsideration of therapeutic options. Current Infect Dis Reports 2006; 8:23-30. [PubMed]

73. Levison M. Pharmacodynamic Considerations in the Medical Treatment of Bacterial Endocarditis. Rev Soc Cardiol Estado de Sao Paulo 1995;5:402-7.

74. Levison ME Response to therapy. In: Kaye D, ed, Infective Endocarditis. Baltimore MD: University Park Press. 1976:185-99.

75. Li W, Somerville J. Infective endocarditis in the grown-up congenital (GUCH) population. Eur Heart J 1998;19:166–73. [PubMed]

76. Li JS, Sexton DJ, Mick N, Nettles R, Fowler VG Jr, Ryan T, et al. Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis. Clin Infect Dis 2000;30:633–8. [PubMed]

77. Lowy FD, Fant J, Higgins LL, et al. Staphylococcus aureus-human endothelial cell interactions. J Ultrastruct Mol Struct Res 1988;98:137-46. [PubMed]

78. Maki DG, Agger WA. Enterococcal bacteremia: clinical features, the risk of endocarditis, and management. Medicine (Baltimore) 1988;67:248-69. [PubMed]

79. Mansur AJ, Grinberg M, da Luz PL, Bellotti G. The complications of infective endocarditis. A reappraisal in the 1980s. Arch Intern Med 1992;152:2428-32. [PubMed]

80. Mansur AJ, Dal Bo CM, Fukushima JT, Pomerantzeff PM. Relapses, recurrences, valve replacements, and mortality during the long-term follow-up after infective endocarditis.Am Heart J 2001;141:78-86. [PubMed]

81. Markowitz N, Quinn EL, Saravolatz LD. Trimethoprim-sulfamethoxazole compared with vancomycin for the treatment of Staphylococcus aureus infection Ann Intern Med 1992;117:390-398. [PubMed]

82. Matsumura S, Simor AE. Treatment of endocarditis due to vancomycin-resistant Enterococcus faecium with quinupristin/dalfopristin, doxycycline, and rifampin: a synergistic drug combination. Clin Infect Dis 1998;27:1554-6. [PubMed]

83. McDonald JR, Olaison L, Anderson DJ, et al. Enterococcal endocarditis: 107 cases from the international collaboration on endocarditis merged database. Am J Med 2005;118:759-66. [PubMed]

84. Moellering RC Jr, Weinberg AN. Studies on antibiotic synergism against enterococci. II. Effect of various antibiotics on the uptake of 14 C-labeled streptomycin by enterococci. J Clin Invest 1971;50:2580-4. [PubMed]

85. Moellering RC, Korzeniowski OM, Wennerstein C, et al. Species-specific resistance to antimicrobial synergism in Streptococcus faecium and Streptococcus faecalis J Infect Dis 1979;140:203-8. [PubMed]

86. Mohan SS, McDermott BP, Cunha BA. Methicillin-resistant Staphylococcus aureus prosthetic aortic valve endocarditis with paravalvular abscess treated with daptomycin. Heart Lung 2005;34:69-71. [PubMed]

87. Monteiro CA, Cobbs CG. Outpatient Management of Infective Endocarditis. Curr Infect Dis Rep 2001;3:319-27. [PubMed]

88. Moreillon P, Que YA, Bayer AS. Pathogenesis of streptococcal and staphylococcal endocarditis. Infect Dis Clin North Am 2002;16:297-318. [PubMed]

89. Morris CD, Reller MD, Menashe VD. Thirty-year incidence of infective endocarditis after surgery for congenital heart defect. JAMA 1998;179:599–603. [PubMed]

90. Morris AJ, Drinkovic D, Pottumathy S, et al. Bacteriologic outcome after valve surgery   for active infective endocarditis: implications for duration of treatment after surgery. Clin Infect Dis 2005;41:187-94. [PubMed]

91. Murray HW, Roberts RB. Streptococcus bovis bacteremia and underlying gastrointestinal disease. Arch Intern Med 1978;138:1097-9. [PubMed]

92. Murray BE, Church DA, Wanger A, et al. Comparison of two beta-lactamase-producing strains of Streptococcus faecalis. Antimicrob Agents Chemother 1986;30:861-4. [PubMed]

93. Mylonakis E, Calderwood SB. Infective endocarditis in adults. N Engl J Med 2001; 345: 1318-30. [PubMed]

94. Okell CC, Elliot SD. Bacteremia and oral sepsis with special reference to the etiology of subacute endocarditis. Lancet 1935; 2:869-74.

95. Palmer SMKang SLCappelletty DMRybak MJ. Bactericidal killing activities of cefepime, ceftazidime, cefotaxime, and ceftriaxone against Staphylococcus aureus and beta-lactamase-producing strains of Enterobacter aerogenes and Klebsiella pneumoniae in an in vitro infection model. Antimicrob Agents Chemother 1995;39:1764-71. [PubMed]

96. Parker MT, Ball LC. Streptococci and aerococci associated with systemic infection in man. J Med Microbiol. 1978; 9: 275-302. [PubMed]

97. Patel R, Piper KE, Rouse MS, et al. Frequency of isolation of Staphylococcus lugdunensis among staphylococcal isolates causing endocarditis: a 20-year experience. J Clin Microbiol 2000; 38: 4262-3. [PubMed]

98. Paterson DL. Resistance in gram-negative bacteria: enterobacteriaceae. Am J Med 2006;119(6 Suppl 1):S20-8.


99. Ramirez-Ronda CH. Adherence of glucan-positive and glucan-negative streptococcal strains to normal and damaged heart valves. J Clin Invest 1978; 62: 804-14. [PubMed]

100. Rao N, White GJ. Successful treatment of Enterococcus faecalis prosthetic valve endocarditis with linezolid. Clin Infect Dis 2002;35:902-4. [PubMed] 

101. Raoult D, Fournier PE, Drancourt M, et al. Diagnosis of 22 new cases of Bartonella endocarditis. Ann Intern Med 1996;125:646-52. [PubMed]

102. Roberts RB, Wilson WR, Bouvet A, Acar J. Nutritionally variant streptococcal (NVS) endocarditis. 22nd Interscience Conference on Antimicrobial Agents and Chemotherapy. Washington DC: American Society for Microbiology, 1982:130.

103. Roberts GJ. Dentists are innocent! "Everyday" bacteremia is the real culprit: a review and assessment of the evidence that dental surgical procedures are a principal cause of bacterial endocarditis in children. Pediatr Cardiol 1999;20:317–25. [PubMed]

104. Rodbard S. Blood velocity and endocarditis. Circ 1963; 27:18-28. [PubMed]

105. Rohmann S, Erbel R, Gorge G, et al. Clinical relevance of vegetation localization by transesophageal echocardiography in infective endocarditis. Eur Heart J 1992;13:446-52. [PubMed]

106. Rovery C, Greub G, Lepidi H, et al. PCR detection of bacteria on cardiac valves of patients with treated bacterial endocarditis. J Clin Microbiol 2005;43:163-7. [PubMed]

107. Saleh-Mghir A, Lefort A, Petegnief Y, et al. Activity and diffusion of LY333328 in experimental endocarditis due to vancomycin-resistant Enterococcus faecalis. Antimicrob Agents Chemother 1999;43:115-20. [PubMed]

108. Salgado AV, Furlan AJ, Keys TF, et al. Neurologic complications of endocarditis: a 12 year experience. Neurology 1989; 39: 173-78. [PubMed]

109. Santoro J, Ingerman M. Response to therapy: Relapse and Reinfection. In: Kaye D, ed. Infective Endocarditis, 2nd ed. New York: Raven Press. 1992: 423-34.

110. Sarli Issa V, Fabri J Jr, Pomerantzeff PM, et al. Duration of symptoms in patients with infective endocarditis. Int J Cardiol 2003;89:63-70. [PubMed]

111. Segreti JA, Crank CW, Finney MS. Daptomycin for the treatment of gram-positive bacteremia and infective endocarditis: a retrospective case series of 31 patients. Pharmacotherapy 2006;26:347-52. [PubMed]

112. Sexton DJ, Tenenbaum MJ, Wilson WR, et al. Ceftriaxone once daily for four weeks compared with ceftriaxone plus gentamicin once daily for two weeks for treatment of endocarditis due to penicillin-susceptible streptococci. Endocarditis Treatment Consortium Group. Clin Infect Dis 1998;27:1470-4. [PubMed]

113. Silverman JA, Mortin LI, Vanpraagh AD, et al. Inhibition of daptomycin by pulmonary surfactant: in vitro modeling and clinical impact. J Infect Dis 2005;191:2149-52. [PubMed]

114. Simsir SA, Cheeseman SH, Lancey RA, et al. Staged laparoscopic splenectomy and valve replacement in splenic abscess and infective endocarditis. Ann Thorac Surg 2003;75:1635-7. [PubMed]

115. Small PM, Chambers HF. Vancomycin for Staphylococcus aureus endocarditis in intravenous drug users. Antimicrobe Agents Chemother 1990;34:1227-31. [PubMed]

116. Starkebaum M, Durack D, Beeson P. The “incubation period” of subacute bacterial endocarditis. Yale J Biol Med 1977;50:49-58. [PubMed]

117. Scheld WM, Valone JA, Sande MA. Bacterial adherence in the pathogenesis of endocarditis. J Clin Invest 1978; 61: 1394- 1404. [PubMed]

118. Stamboulian D. Outpatient treatment of endocarditis in a clinic-based program in Argentina. Eur J Clin Microbiol Infect Dis 1995;14:648-54. [PubMed]

119. Steckelberg JM, Wilson WR. Risk factors for infective endocarditis. Infect Dis Clin North Am 1993;7:9-19. [PubMed]

120. Steckelberg JM, Murphy JG, Ballard D, et al. Emboli in infective endocarditis: the prognostic value of echocardiography. Ann Intern Med 1991;114:635-40. [PubMed]

121. Steckelberg JM, Murphy JG, Wilson WR. Management of complications of infective endocarditis. In: Kaye D ed. Infective Endocarditis 2nd ed New York: Raven Press. 1992:435-54.

122. Steckelberg JM, Wilson WR. Risk factors for infective endocarditis. Infect Dis Clin North Am 1993;7:9-19. [PubMed]

123. Stein DS, Nelson KE. Endocarditis due to nutritionally deficient streptococci: therapeutic dilemma. Rev Infect Dis 1987;9:908-16. [PubMed]

124. Stevens DL, Yan S, Bryant AE. Penicillin-binding protein expression at different growth stages determines penicillin efficacy in vitro and in vivo: an explanation for the inoculum effect. J Infect Dis 1993;167:1401-5. [PubMed]

125. Strom BL, Abrutyn E, Berlin JA, et al. Risk factors for infective endocarditis: oral hygiene and nondental exposures. Circ 2000;102:2842-8. [PubMed]

126. Strom BL, Abrutyn E, Berlin JA, et al. Dental and cardiac risk factors for infective endocarditis. A population-based, case-control study. Ann Intern Med 1998;129:761-9. [PubMed]

127. Thompson RL, Lavin B, Talbot GH. Endocarditis due to vancomycin-resistant Enterococcus faecium in an immunocompromised patient: cure by administering combination therapy with quinupristin/dalfopristin and high-dose ampicillin. South Med J 2003;96:818-20. [PubMed]

128. Vankatesh SK, Phadke RV, Kalode RR, et al. Intracranial infective aneurysms presenting with hemorrhage: an analysis of angiographic findings, management and outcome. Clin Radiol 2000;55:946-53. [PubMed]

129. Vouillamoz J, Moreillon P, Giddey M, Entenza JM. Efficacy of daptomycin in the treatment of experimental endocarditis due to susceptible and multidrug-resistant enterococci. J Antimicrob Chemother 2006 Oct 9. [PubMed]

130. Wareham DW, Abbas H, Karcher AM, Das SS. Treatment of prosthetic valve infective endocarditis due to multi-resistant Gram-positive bacteria with linezolid. J Infect 2006;52:300-4. [PubMed]

131. Watanakunakorn C. Clindamycin therapy of Staphylococcus aureus endocarditis: clinical relapse and development of resistance to clindamycin, lincomycin, and erythromycin. Am J Med 1976;60:419-25. [PubMed]

132. Werner AS, Cobbs CG, Kaye D. Studies on the bacteremia of bacterial endocarditis. JAMA 1967; 202: 199-203. [PubMed]

133. Wiebers DO, Whisnant JP, Huston J 3rd, et al. Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and cardiovascular treatment. Lancet 2003;362:103-10. [PubMed]

134. Wilson WR, Wilkowske CJ, Wright AJ, et al. Treatment of streptomycin-susceptible and streptomycin-resistant enterococcal endocarditis. Ann Intern Med 1984;100:816-23. [PubMed]

135. Wilson WR, Geraci JE. Treatment of streptococcal infective endocarditis. Am J Med 1985;78:128-37. [PubMed]

136. Wright AJ, Wilson WR, Matsumoto JY, et al. Influence of gentamicin dose size on the efficacies of combinations of gentamicin and penicillin in experimental streptomycin-resistant enterococcal endocarditis. Antimicrob Agents Chemother 1982;22:972-5. [PubMed]

137. Yeaman MR, Bayer AS. Staphylococcus aureus, Platelets, and the Heart. Curr Infect Dis Rep 2000;2:281-98. [PubMed]

138. Yersin BR, Glauser MP, Freedman LR. Effect of nitrogen mustard on natural history of right-sided streptococcal endocarditis in rabbits: role for cellular host defenses. Infect Immun 1982;35:320-25. [PubMed]

139. Yersin BGlauser MPGuze PAGuze LB. Freedman LR. Experimental Escherichia coli endocarditis in rats: roles of serum bactericidal activity and duration of catheter placement. Infect Immun 1988;56:1273-80. [PubMed]

140. Zarrouk V, Bozdogan B, Leclercq R, et al. Activities of the combination of quinupristin-dalfopristin with rifampin in vitro and in experimental endocarditis due to Staphylococcus aureus strains with various phenotypes of resistance to macrolide-lincosamide-streptogramin antibiotics. Antimicrob Agents Chemother 2001;45:1244-8. [PubMed]

141. Zimhony O, Chmelnitsky I, Bardenstein R et al. Endocarditis caused by extended-spectrum-beta-lactamase-producing Klebsiella pneumoniae: emergence of resistance to ciprofloxacin and piperacillin-tazobactam during treatment despite initial susceptibility. Antimicrob Agents Chemother 2006;50:3179-62[PubMed]

142. Zimmer SM, Caliendo AM, Thigpen MC, Somani J. Failure of linezolid treatment for enterococcal endocarditis. Clin Infect Dis 2003;37:e29-30. [PubMed]

back to top


Table 1: Risk Factors for Infective Endocarditis



  • IV drug abuse
  • Male
  • Advancing age
  • Recent dental surgery or other invasive procedures
  • Nosocomial bacteremia
  • Permanent venous access lines
  • Surgically constructed pulmonary shunts
  • Degenerative valvular lesions
  • Congenital heart disease
  • Prosthetic valves
  • Mitral valve prolapse with insufficiency
  • Rheumatic heart disease
  • Previous infective endocarditis
  • Hypertrophic cardiomyopathy

Table 2: Considerations for Testing in Culture Negative IE

Special Culture Requirements Serologies Other Testing
Histoplasma capsulatum  (fungal cultures) H. capsulatum Tropheryma whippelii(PCR of tissue)
Aspergillus (fungal cultures) Coxiella burnetii
Blastomycosis dermatidis (fungal culture) Chlamydia psittaci
Bartonella species (prolonged incubation) Legionella species
Erysipelothrix sp. (fungal cultures) Brucella species
Bartonella quintana or henselae

Table 3. Epidemiological Clues in Etiological Diagnosis of Culture-Negative Endocarditis

Epidemiological Feature

Common Microorganism(s)

Injection drug use

S aureus, including community-acquired oxacillin-resistant strains Coagulase-negative staphylococci, ß-Hemolytic streptococci; Fungi; Aerobic Gram-negative bacilli, includingPseudomonas aeruginosa, Polymicrobial

Indwelling cardiovascular medical devices

S aureus, Coagulase-negative staphylococci; Fungi; Aerobic Gram-negative bacilli; Corynebacterium sp

Genitourinary disorders, infection, manipulation, including pregnancy, delivery, and abortion

Enterococcus sp, Group B streptococci (S agalactiae), Listeria monocytogene, Aerobic Gram-negative bacilli,Neisseria gonorrhoeae

Chronic skin disorders, including recurrent infections

S aureus, ß-Hemolytic streptococci

Poor dental health, dental procedures

Viridans group streptococci, "Nutritionally variant streptococci", Abiotrophia defectiva, Granulicatella sp,Gemella sp, HACEK* organisms

Alcoholism, cirrhosis

Bartonella sp, Aeromonas sp, Listeria sp, S pneumoniae,ß-Hemolytic streptococci

Burn patients

S aureus, Aerobic Gram-negative bacilli, including P aeruginosa; Fungi

Diabetes mellitus

S aureus, ß-Hemolytic streptococci ,S pneumoniae

Early (≤1 y) prosthetic valve placement

Coagulase-negative staphylococci, S aureus, Aerobic Gram-negative bacilli; Fungi Corynebacterium sp Legionella sp

Late (>1 y) prosthetic valve placement

Coagulase-negative staphylococci, S aureus, Viridans group streptococci, Enterococcus species, Fungi,Corynebacterium sp

Dog–cat exposure

Bartonella sp, Pasteurella sp, Capnocytophaga sp

Contact with contaminated milk or infected farm animals

Brucella sp, Coxiella burnetii, Erysipelothrix sp


Homeless, body lice

Bartonella sp


Salmonella sp, S pneumoniae, S aureus

Pneumonia, meningitis

S pneumoniae

Solid organ transplant

S aureus, Aspergillus fumigatus, Enterococcus sp, Candidasp

Gastrointestinal lesions

S bovis, Enterococcus sp, Clostridium septicum

                       *Haemophilus parainfluenzae, H. aphrophilus, Actinobacillus actinomycetemcomitans, Cardiobacterium hominis, Eikenella                         corrodens, and Kingella kingae

                       Adapted from Baddour LM, Wilson WR, Bayer AS, et al. Infective endocarditis. Diagnosis, Antimicrobial therapy, and management of                       complications. Circ 2005;111:3167-84.

Table 4. Modified Duke Criteria for the Diagnosis of Infective Endocarditis.

1. Definite diagnosis of infective endocarditis
  • Pathologic criteria
    a. Histological and/or microbiologic evidence of infection at surgery or autopsy 
  • Clinical criteria
    a. 2 major criteria; OR
    b. 1 major/3 minor; OR
    c. 5 minor

2. Possible diagnosis:
     a. 1 major criteria and major criteria; OR
     b. 3 minor;

3. No endocarditis:
     a. Firm alternate diagnosis
     b. Clinical resolution with ≤4 days of antimicrobial therapy
     c. No evidence of infective endocarditis at surgery or autopsy with ≤4 days of antimicrobial therapy;
     d. Failure to meet criteria for possible infective endocarditis, as above

Major Criteria
1. Blood culture
  • 2 separate blood cultures positive for:
    a. viridans streptococci, Streptococcus bovis, HACEK*, Staphylococcus aureus.
    b. Community-acquired enterococci, in absence of primary focus.
  • Microorganisms consistent with endocarditis isolated from:
    a. at least 2 blood cultures drawn >12 hours apart.
    b. 3 of 3, or a majority of 4 or more with 1st and last obtained >1hour apart.
  • Single positive blood culture for Coxiella burnetii or antiphase 1 IgG antibody >1:800

2. Evidence of endocardial involvement

  • Echocardiography: Positive for oscillating intracardiac mass on valve or supporting structure, in path of regurgitant jet, or on implanted material in the absence of an alternative anatomic explanation; or valve ring abscess; or new partial dehiscence of valvular prosthesis.

  • New valvular regurgitant murmur (increasing or changing or pre-existing murmur not sufficient).
Minor Criteria
  • Predisposing heart condition or injection drug use.
  • Fever, >380C.
  • Major arterial emboli, septic pulmonary infarcts, mycotic aneurysm, intracranial hemorrhage, conjunctival hemorrhage, and Janeway lesions.
  • Immunologic phenomena: glomerulonephritis, Roth’s spot, Osler’s node, and rheumatoid factor.
  • Positive blood culture that does not meet major criterion (as noted above) or serologic evidence of active infection with organism consistent with infective endocarditis.
*Haemophilus parainfluenzae, H. aphrophilus, Actinobacillus actinomycetemcomitans, Cardiobacterium hominis, Eikenella corrodens, and Kingella kingae
 Modified from Li JS, Sexton DJ, Mick N, et al. Proposed modification to the Duke criteria for the diagnosis of infective endocarditis. Clin Infect Dis 2000;30:633-8.

Table 5. In Vitro Assays




  Viridans streptococcus   Broth dilution test   Penicillin MIC
  Enterococcus   Broth dilution test
  Growth in:
     • 500 ug/ml of Gentamicin
     • 1000 ug/ml Streptomycin

  Nitrocephin degradation

  Penicillin MIC
  MICHigh-level resistancea:              • Gentamicin
     • Streptomycin

  Beta-lactamase production

  S. aureus and Coagulase-  Negative Staphylococci   Nitrocephin degradation
  Oxacillin/methicillin sensitivity
  Broth dilution test
  Beta-lactamase production
  Vancomycin MIC
  Rifampin MIC
  Gentamicin MIC
  Other pathogens   Broth dilution tests   Antibiotic MIC/MBCb
  All pathogens   Serum antibiotic concentrations Peak and trough vancomycinc   and aminoglycosided concentrations
  No pathogen isolated Histochemical stains of   vegetations/emboli
Immunohistology of vegetations/emboli
Broad-spectrum bacterial and fungal
PCR and DNA sequencing on Vegetation/emboli
Legionella urinary antigen assay


a. The infecting strain of enterococcus recovered from patients with endocarditis should be tested for susceptibility to high levels of both gentamicin and streptomycin but not other aminoglycosides. Strains that are resistant to high levels of gentamicin are resistant to other aminoglycosides, except some of these strains may be susceptible to high levels of streptomycin.
Choice of an aminoglycoside for synergy should be based on in vitro high-level aminoglycoside susceptibility testing. If the strain is susceptible to high levels of both gentamicin and streptomycin, gentamicin is preferred because determination of gentamicin serum levels is more generally available. If the strain exhibits high-level resistance to one of these aminoglycosides, use only the aminoglycoside to which the strain is sensitive. If the strain is resistant to high levels of both gentamicin and streptomycin, no aminoglycoside is available to synergize with a cell wall-active antibiotic.
b. MIC/MBC testing may be useful for nonstandard antimicrobial regimens or unusual pathogens
c. Vancomycin “peak” serum levels should be obtained 1 h after completion of a 1-2h infusion and should be in the range of 30-45 ug/ml. Vancomycin trough levels obtained just before the next dose should be 10-15 ug/ml.
d. Gentamicin “peak” serum levels obtained 1 h after start of a 20-30 min IV infusion or IM injection of 1 mg/kg should be about 3-4 ug/ml and trough level should be <1 ug/ml. Streptomycin peak serum level 1h after IM administration of 7.5 mg/kg is about 15-20 ug/ml and trough should be about 5 ug/ml.
MIC, minimal inhibitory concentration; MBC, minimal bactericidal concentration; MRSA, methicillin-resistant S. aureus; MRCNS, methicillin-resistant coagulase-negative staphylococci; TMP/SMX ( trimethoprim/sulfamethoxazole);
------ Modified from Levison ME: In vitro assays. In Kaye D (ed. ): Infective Endocarditis. 2nd Ed. New York, Raven Press, 1992:151-167.

Table 6a. Standard Antibiotic Therapy for Native Valve Endocarditis due to Common Pathogens (Doses are for Adults with Normal Renal Function)


Primary Regimen(s)


Alternative Regimen(s)


Viridans group Streptococcus
Streptococcus bovis (PCN MIC <0.12 ug/ml)
PCN G 12-18 million units IV daily0
Ceftriaxone 2 grams IV daily
Vancomycin2 (trough goal ~15-20)
4 weeks PCN G 12-18 million daily0 + gentamicin1 3 mg/kg IV daily
Ceftriaxone 2 grams IV daily+ gentamicin1 3 mg/kg IV daily
Daptomycin8 6 mg/kg IV daily
2 weeks (only if uncomplicated right sided endocarditis6 and 2 weeks of gentamicin given)
4 weeks (for daptomycin)
Viridans group Streptococcus
Streptococcus bovis (PCN MIC>0.12-<0.5 ug/ml)
PCN G 24 million IV daily0 + gentamicin1 3mg/kg IV daily
Ceftriaxone 2 grams IV daily + gentamicin1 3 mg/kg IV daily
4 weeks of PCN or Ceftriaxone + 2 weeks of gentamicin



Vancomycin2 (trough goal ~15-20)
Daptomycin8 6 mg/kg IV daily
4 weeks
Penicillin-resistant Streptococci (MIC >0.5 ug/ml),
Abiotrophia species4(nutritionally variant streptococci)
Ampicillin 2 grams IV every 4 hours + gentamicin1 1 mg/kg IV every 8 hrs
PCN G 18-30 million units IV daily0 + gentamicin1 1 mg/kg IV every 8 hrs
6 weeks Vancomycin2 (trough goal 15-20) +/- gentamicin1 1 mg/kg IV every 8 hrs
Daptomycin8 6 mg/kg IV daily +/- gentamicin1 1 mg/kg IV every 8 hrs
6 weeks
Methicillin-sensitive Staphylococcus aureus Oxacillin 2 grams IV every 4 hours +/- gentamicin 1 mg/kg IV every 8 hrs (for up to 2 weeks)
Cefazolin5 1.5 grams IV every 8 hours +/- gentamicin1 1 mg/kg IV every 8 hrs (for up to 2 weeks)
2 weeks (uncomplicated6 right sided endocarditis only and treatment mustinclude 2 weeks of gentamicin)
6 weeks (all others)
Daptomycin8 6 mg/kg IV daily +/- gentamicin1 1 mg/kg IV every 8 hrs
Daptomycin8 6 mg/kg IV daily + rifampin 600 mg PO or IV daily
Vancomycin2 (trough goal ~15-20)
6 weeks
Methicillin-resistant Staphylococcus aureus Daptomycin8 6 mg/kg IV daily +/- gentamicin1 1 mg/kg IV every 8-12 hrs
Vancomycin (trough goal ~15-20) +/- gentamicin1 1 mg/kg IV every 8-12 hrs
6 weeks (duration of gentamicin will vary by clinical scenario) Daptomycin8 6 mg/kg IV daily +/- Rifampin 600 mg PO or IV daily +/- gentamicin1 1 mg/kg IV every 8-12 hrs
Quinupriston-Dalfopristin 7.5 mg/kg IV every 8 hrs
6 weeks
HACEK7 organisms Ceftriaxone 2 grams IV daily 4 weeks Ampicillin/sulbactam 3 g IV every 6h
Ciprofloxacin 400 mg IV every 12 hours or 500 mg PO every 12 hours
4 weeks
All other bacteria or fungal species Please refer to Antimicrobial Therapy and Vaccines Volume I: Microbes for further guidance. Varies



0PCN can be dosed via continuous infusion or every 4 hrs
1Gentamicin should not be used in patients with creatinine clearance <30 ml/min, or patients with impaired 8th cranial nerve function. Other potentially nephrotoxic drugs, e.g. , non-steroidal anti-inflammatory drugs, should be used with caution in patients receiving gentamicin. Caution when using gentamicin with vancomycin due to increased risk of nephrotoxicity. Adjust dose of gentamicin based on renal function.
2Vancomycin is used only for patients with immediate type penicillin-allergic reactions, i.e. , urticaria, angioedema, or anaphylaxis, to penicillin. Caution when using gentamicin with vancomycin due to risk of nephrotoxicity. Vancomycin is inferior to beta-lactams for the treatment of methicillin-sensitive Staphylococcus aureus.
3Enterococcus sensitive to penicillin, vancomycin and aminoglycosides.
4Because of technical difficulties in susceptibility testing of Abiotrophia species and virulence of these organisms, many experts recommend treating endocarditis due to these strains with the standard regimen recommended for enterococci. Strongly consider dual antibiotic therapy.
5Cefazolin is used for patients with a non-immediate-type penicillin allergy
6Uncomplicated right-sided IE: normal renal function, no extra-pulmonary metastatic infection, no left-sided valvular IE.
7HACEK: Haemophilus parainfluenzae, H. aphrophilus, Actinobacillus actinomycetemcomitans, Cardiobacterium hominis, Eikenella corrodens, and Kingella kingae
8Daptomycin is equivalent to beta-lactam antibiotics for the treatment of methicillin-sensitive Staphylococcus aureus. Endocarditis data is for Staphylococcal aureus only.
*Adapted and modified from Baddour LM, Wilson WR, Bayer AS, et al. Infective endocarditis. Diagnosis, Antimicrobial therapy, and management of complications. Circ 2005;111:3167-84. Utility of daptomycin, linezolid, quinupriston/dalfopriston and treatment of VRE are not part of the original guidelines.

Table 6b. Standard Antibiotic Therapy for Prosthetic Valve Endocarditis due to Common Pathogens (Doses are for Adults with Normal Renal Function)


Primary Regimen(s)


Alternative Regimen(s)


Viridans group Streptococcus 
Streptococcus bovis(PCN MIC <0.12ug/ml)
PCN G 12-18 million units IV daily0 +/- gentamicin 3 mg/kg IV daily
Ceftriaxone 2 grams IV daily +/- gentamicin 3 mg/kg IV daily
6 weeks (gentamicin for 2 weeks only) Vancomycin2 (trough goal ~15-20)
Daptomycin8 6mg/kg IV daily
6 weeks
Viridans group Streptococcus
Streptococcus bovis(PCN MIC >0.12-<0.5 ug/ml)
PCN G 24 million IV daily0 + gentamicin1 3 mg/kg IV daily
Ceftriaxone 2 grams IV daily + gentamicin1 3 mg/kg IV daily
6 weeks (gentamicin for 2 weeks only)



Vancomycin2 (trough goal ~15-20)
Daptomycin8 6 mg/kg IV daily


6 weeks
Penicillin-resistant Streptococci (MIC >0.5 ug/ml),
Abiotrophia species4(nutritionally variant streptococci)
Ampicillin 2 grams IV every 4 hours + gentamicin1 1 mg/kg IV every 8 hrs
PCN G 18-30 million units IV daily0 + gentamicin1 1 mg/kg IV every 8 hrs
6 weeks Vancomycin2 (trough goal 15-20) +/- gentamicin1 1 mg/kg IV every 8 hrs
Daptomycin8 6 mg/kg IV daily +/- gentamicin1 1 mg/kg IV every 8 hrs
6 weeks
Methicillin-sensitive Staphylococcus aureus Oxacillin9 2 grams IV every 4 hours + gentamicin 1 mg/kg IV every 8 hrs + Rifampin600 mg PO/IV daily


Cefazolin5 1.5 grams IV every 8 hours + gentamicin1 1 mg/kg IV every 8 hrs + rifampin 600 mg IV/PO daily
6+ weeks (2 weeks of gentamicin) Vancomycin2 (trough goal ~15-20) + gentamicin 1 mg/kg IV every 8 hrs + Rifampin 600 mg IV/PO daily
Daptomycin8 6 mg/kg IV daily + gentamicin1 1 mg/kg IV every 8 hrs + rifampin 600 mg PO/IV daily
6+ weeks (2 weeks of gentamicin)
Methicillin-resistant Staphylococcus aureus Daptomycin8 6 mg/kg IV daily + Rifampin 600 mg PO or IV daily + gentamicin1 1 mg/kg IV every 8 hrs
Vancomycin (trough goal ~15-20) + gentamicin1 1 mg/kg IV every 8 hrs +/- rifampin 600 mg PO/IV daily
6+ weeks (2 weeks of gentamicin) Please refer to Antimicrobial Therapy and Vaccines Volume I: Microbes for further guidance. 6+ weeks
HACEK7 organisms Ceftriaxone 2 grams IV daily 6 weeks Ampicillin/sulbactam 3 g IV every 6h
Ciprofloxacin 400 mg IV every 12 hours or 500 mg PO every 12 hours
6 weeks
All other bacteria or fungal species Please refer to Antimicrobial Therapy


Vaccines Volume I: Microbes for further guidance.



0PCN can be dosed via continuous infusion or every 4 hrs
1Gentamicin should not be used in patients with creatinine clearance <30 ml/min, or patients with impaired 8th cranial nerve function. Other potentially nephrotoxic drugs, e.g. , non-steroidal anti-inflammatory drugs, should be used with caution in patients receiving gentamicin. Caution when using gentamicin with vancomycin due to increased  risk of nephrotoxicity. Adjust dose of gentamicin based on renal function.
2Vancomycin is used only for patients with immediate type penicillin-allergic reactions, i.e. , urticaria, angioedema, or anaphylaxis, to penicillin. Caution when using gentamicin with vancomycin due to risk of nephrotoxicity. Vancomycin is inferior to beta-lactams for the treatment of methicillin-sensitive Staphylococcus aureus.
3Enterococcus sensitive to penicillin, vancomycin and aminoglycosides.
4Because of technical difficulties in susceptibility testing of Abiotrophia species and virulence of these organisms, many experts recommend treating endocarditis due to these strains with the standard regimen recommended for enterococci. Strongly consider dual antibiotic therapy.
5Cefazolin is used for patients with a non-immediate-type penicillin allergy
6Uncomplicated right-sided IE: normal renal function, no extra-pulmonary metastatic infection, no left-sided valvular IE.
7HACEK: Haemophilus parainfluenzae, H. aphrophilus, Actinobacillus actinomycetemcomitans, Cardiobacterium hominis, Eikenella corrodens, and Kingella kingae
8Daptomycin is equivalent to beta-lactam antibiotics for the treatment of methicillin-sensitive Staphylococcus aureus. Only data for daptomycin in prosthetic valve infections is anecdotal case reports. Endocarditis data is for Staphylococcal aureus only.
9Oxacillin and nafcillin can be used interchangeably.
*Adapted and modified from Baddour LM, Wilson WR, Bayer AS, et al. Infective endocarditis. Diagnosis, Antimicrobial therapy, and management of complications. Circ 2005;111:3167-84. Utility of daptomycin, linezolid, quinupriston/dalfopriston and treatment of VRE are not part of the original guidelines.

Table 7: Indications for Surgical Intervention with Infective Endocarditis

Clinical Situations

Echocardiography Findings

Refractory CHF secondary to valvular dysfunction Persistent vegetations after a major systemic embolic episode
>2 serious systemic embolic episode Large (>1cm diameter) anterior mitral valve vegetation
Uncontrolled infection (persistent bacteremia) Acute mitral insufficiency
Endocarditis caused by certain pathogens (Candida sp, Fungi, Enterococci with synergistic treatment options, left sided MRSA endocarditis) Increase in vegetation size 4 weeks after antibiotic therapy
Ineffective antimicrobial therapy Valve perforation or rupture
Most cases of prosthetic valve IE (if no perivalvular involvement, medical treatment may be considered) Periannular extension of infection including perivalvular or myocardial abscess
Resection of mycotic aneurysms Physiologically significant valve dysfunction as demonstrated by echo

Table 8. Reasons For Inadequate Clinical Response

  •   Inadequate therapy: wrong drug, wrong dose
  •  Infarcts secondary to emboli
  • Metastatic abscesses of the spleen, kidney, brain, etc. , which may require surgical drainage
  • Suppurative thrombophlebitis at site of an IV catheter, with or without superinfecting endocarditis.
  • ·Other superinfections: e.g. , C. difficile colitis, urinary tract infection
  • Febrile reaction to the antimicrobial agent or other another drug
  • Another unrelated febrile illness, e.g. , deep vein thrombophlebitis

Table 9: Cardiac Conditions Associated with Highest Risk of Adverse Outcome from Endocarditis for which Prophylaxis with Dental Procedures is Recommended.

1. Prosthetic cardiac valve
2. Previous endocarditis

3. Congenital heart disease*

  • Unrepaired cyanotic CHD, including palliative shunts and conduits
  • Completely repaired congenital heart defect with prosthetic material or     device, whether placed by surgery or by catheter intervention, during the     first 6 months after the procedure†
  • Repaired CHD with residual defects at the site or adjacent to the site of a     prosthetic patch or prosthetic device (which inhibit endothelialization)
4. Cardiac transplantation recipients who develop cardiac valvulopathy
                            *Except for the conditions listed above, antibiotic prophylaxis is no longer recommended for any other form of CHD.

                            †Prophylaxis is recommended because endothelialization of prosthetic material occurs within 6 months after the procedure.

Table 10. Procedures for Which Antibiotic Prophylaxis is Recommended for Patients in Table 9.

  • Dental/oral surgery likely to cause bleeding (manipulation of gingival tissue or periapical region of the teeth or perforation of the oral mucosa).
  • Tonsillectomy and/or adenoidectomy
  • Invasive procedure of respiratory tract which involves biopsy or incision of mucosa
  • Surgical procedures involving infected soft tissue or musculoskeletal tissue

Table 11a. Regimens for Dental Procedures (give 30-60 minutes prior to procedure)

Amoxicillin 2 g PO
Unable to take PO
Ampicillin 2 g IV/IM
Cefazolin 1 g IV/IM
Ceftriaxone 1 g IV/IM
Penicillin allergy-oral
Cephalexin 2 g PO
Clindamycin 600 mg PO or IV/IM
Azithromycin 500 mg PO
Penicillin allergy-unable to take PO
Cefazolin 1 g IV/IM
Ceftriaxone 1 g IV/IM
Clindamycin 600 mg IV/IM


Table 11b: Regimens Involving Non-Dental Procedures of Patients Listed in Table 9.

Respiratory Procedures
  •  Antibiotics as recommended in table 5
  •  If Staphylococcal infection is known or suspected, then substitution with an anti-staphylococcal penicillin (Oxacillin, nafcillin) should be utilized
  • If MRSA infection known or suspected, then substitution with Vancomycin should be utilized.
Genitourinary Procedures
  • Antibiotics only indicated if infection is present
  • Ampicillin, amoxicillin, vancomycin should be considered to ensure enterococcal coverage
Skin and Soft Tissue Procedures
  • Prophylactic antibiotics to prevent endocarditis indicated if infection is present
  • Oxacillin, cefazolin, clindamycin should be considered. Utilize vancomycin if MRSA known or suspected.

Guided Medline Search For Historical Aspects

Infective Endocarditis