Infective Endocarditis: How Bacteria Colonize a Heart Valve and Why the Decision Between Antibiotics and Surgery Is Not Simple

2. What It Is

Infective endocarditis (IE) is an infection of the endocardial surface of the heart, predominantly affecting heart valves. The infection causes formation of vegetations: masses of bacteria, fibrin, platelets, and inflammatory cells that adhere to the valve surface, disrupt valve function, and serve as a source of systemic septic emboli.

The clinical criteria for diagnosis are formalized in the modified Duke criteria (Li JS, et al., Clin Infect Dis 2000; 10.1086/313899), which combine pathological and clinical criteria into major and minor elements.

Major Duke criteria:

1. Positive blood cultures for typical IE organisms: Staphylococcus aureus; viridans streptococci; Streptococcus bovis; HACEK group organisms (Haemophilus, Aggregatibacter, Cardiobacterium, Eikenella, Kingella); community-acquired Enterococcus without primary focus; persistently positive blood cultures (two positive cultures at least 12 hours apart, or three of three positive, or majority of four or more positive cultures) for other organisms
2. Evidence of endocardial involvement: positive echocardiogram (vegetation, abscess, new partial dehiscence of prosthetic valve) or new valvular regurgitation

Minor Duke criteria:

1. Predisposing heart condition or IV drug use
2. Fever above 38 degrees Celsius
3. Vascular phenomena: septic arterial emboli, septic pulmonary infarcts, mycotic aneurysm, intracranial hemorrhage, conjunctival hemorrhages, Janeway lesions (painless hemorrhagic lesions on palms and soles)
4. Immunological phenomena: glomerulonephritis, Osler nodes (painful subcutaneous nodules on fingertip or toe pads), Roth spots (retinal hemorrhages), positive rheumatoid factor
5. Positive blood cultures not meeting major criteria
6. Positive echocardiogram findings not meeting major criteria

Definite IE: 2 major criteria; 1 major + 3 minor; 5 minor. Possible IE: 1 major + 1 minor; 3 minor.

Microbiology

Staphylococcus aureus (including methicillin-resistant S. aureus, MRSA) is now the most common cause of IE in the United States, responsible for approximately 30-40% of cases 5 / Solid ). S. aureus IE carries the highest mortality among common IE organisms (in-hospital mortality 25-35%) and the highest rates of complications (embolic events, intracardiac abscess, heart failure).

Viridans streptococci (S. sanguinis, S. mitis, S. mutans, S. oralis): normal oral flora. Enter the bloodstream from dental procedures or spontaneous bacteremia from poor dental hygiene. Classic cause of left-sided IE on native valves with underlying valve disease. Generally lower virulence than S. aureus; better prognosis if valvular destruction is not severe.

Streptococcus bovis (gallolyticus): Bacteremia with S. bovis/gallolyticus should trigger colonoscopy regardless of IE status, because it is associated with colorectal cancer or polyps in approximately 25-50% of cases 5 / Solid 90023-7).

Enterococcus: Second most common gram-positive IE organism. E. faecalis (which is susceptible to ampicillin) is the predominant species; E. faecium (which is more resistant) is emerging. Enterococcal IE in older patients is often healthcare-associated.

HACEK organisms: Slow-growing gram-negative organisms that require prolonged blood culture incubation (previously missed on 5-day cultures). Modern blood culture systems now detect them reliably.

Culture-negative IE: Approximately 5-10% of IE cases have no organism identified on standard blood cultures. Causes: prior antibiotic treatment (the most common cause), fastidious organisms (Bartonella, Coxiella burnetii/Q fever, Tropheryma whipplei), cell-dependent organisms not growing in standard blood culture media. Serologic testing and 16S rRNA PCR of valve tissue at surgery can identify the organism in culture-negative IE 5 / Solid .

3. The Mechanism

The sequence of valve infection

Normal valve endothelium is non-thrombogenic and resistant to infection. IE typically requires a breach:

1. Non-bacterial thrombotic endocarditis (NBTE): Minor valve trauma from turbulent flow (prior valve disease, bicuspid valve, congenital defect, prosthetic valve) or systemic hypercoagulable states creates sterile platelet-fibrin thrombi on the valve surface.
2. Bacteremia: Even transient bacteremia (from dental procedures, skin infection, IV drug injection, GI procedures) introduces organisms into the bloodstream.
3. Bacterial adherence: S. aureus and viridans streptococci express adhesins that bind to the fibronectin and fibrin in NBTE thrombi, initiating valve colonization.
4. Vegetation growth: The colonizing bacteria trigger further fibrin deposition, platelet aggregation, and inflammatory cell recruitment, building a vegetation mass over days to weeks.

Vegetation consequences

Valve destruction: Vegetations are not passive. The proteolytic enzymes released by bacteria (particularly S. aureus) destroy valve leaflet tissue, causing progressive valvular regurgitation. Perivalvular extension produces ring abscess, fistulas, and conduction abnormalities (new heart block from extension into the AV node is an ominous sign requiring urgent surgery).

Systemic embolization: Vegetations fragment and embolize. Left-sided IE (aortic or mitral valve) produces systemic emboli: stroke (cerebral embolism), splenic infarct, renal infarct, septic arterial embolism to the extremities. Right-sided IE (tricuspid valve, pulmonic valve) produces pulmonary emboli, which may cavitate and become septic pulmonary abscesses. Vegetation size above 10 mm is an independent predictor of embolic events 5 / Solid 90756-4).

4. How We Diagnose

Blood cultures

The single most important diagnostic test. Three sets of blood cultures (each set consisting of an aerobic and an anaerobic bottle) drawn from different venipuncture sites over at least one hour should be obtained before antibiotics are started. This maximizes detection sensitivity: bacteremia in IE is typically continuous, so three sets from different sites provide near 98% sensitivity for the major pathogens 5 / Solid . If the patient is clinically unstable, two sets are drawn immediately and antibiotics started; a third set can follow.

Never start antibiotics in a febrile patient with a murmur, prosthetic valve, or cardiac implant without obtaining at least two sets of blood cultures first. A treated bacteremia without microbiological identification leaves the diagnosis, the organism, the sensitivity pattern, and the antibiotic duration undefined.

Echocardiography

TTE (transthoracic echo): Sensitivity for IE vegetations is approximately 75% in native valve IE, lower for prosthetic valves (45-65%) and small vegetations. TTE should be performed in all patients with suspected IE. It provides valve morphology, vegetation characteristics, and hemodynamic consequences.

TEE (transesophageal echo): Sensitivity above 90% for vegetations, with superior visualization of the posterior aortic root, mitral valve, prosthetic valves, and perivalvular structures. TEE is recommended when:

• TTE is non-diagnostic or technically limited
• Prosthetic valve IE is suspected
• Intracardiac abscess or fistula is suspected
• S. aureus bacteremia is present without an obvious extracardiac source (high IE probability)

TEE should be repeated within 5-7 days if initial TEE is negative but clinical suspicion remains high, because early vegetations below the resolution threshold may evolve into visible structures.

CT and nuclear imaging

Cardiac CT is increasingly used for:

• Aortic root abscess and perivalvular fistula detection
• Pseudoaneurysm identification
• Surgical planning

18F-FDG PET/CT is useful for suspected prosthetic valve endocarditis (PVE), where echocardiography has limited sensitivity. Increased FDG uptake around the prosthesis ring suggests active infection. The modified Duke criteria for PVE now include PET/CT positivity as a minor criterion 5 / Solid .

5. The Evidence

Antibiotic regimens

IE treatment requires prolonged bactericidal antibiotic therapy, because vegetations are avascular and densely populated with bacteria in stationary growth phase that resist standard bacteriostatic dosing.

S. aureus native valve IE:

• Methicillin-susceptible S. aureus (MSSA): nafcillin 2g IV q4h OR oxacillin for 6 weeks (or possibly 4 weeks for uncomplicated right-sided IE). Gentamicin is NOT added (adds nephrotoxicity without benefit in the current data; Solid; Fowler VG 2006 JAMA STAPH-IE trial).
• MRSA native valve IE: vancomycin 25-30 mg/kg/day divided q8-12h, targeting AUC/MIC of 400-600 mg·h/L, for 6 weeks. Daptomycin is an alternative for MRSA IE at 8-10 mg/kg/day.

Viridans streptococcal native valve IE:

• Penicillin-susceptible (MIC below 0.12 mcg/mL): IV penicillin G or ceftriaxone 2g IV once daily for 4 weeks (or 2 weeks with ceftriaxone plus gentamicin in uncomplicated cases) 5 / Solid .

Enterococcal native valve IE:

• E. faecalis: ampicillin 2g IV q4h plus ceftriaxone 2g IV q12h for 6 weeks (the GAMES registry-based dual beta-lactam approach, preferred to avoid aminoglycoside nephrotoxicity; Solid; Fernandez-Hidalgo N, et al., Clin Infect Dis 2013; 10.1093/cid/cit399).

Oral step-down: the POET trial (2019)

The POET trial (Partial Oral Treatment of Endocarditis) enrolled 400 patients with left-sided IE (native or prosthetic) caused by streptococci, Enterococcus faecalis, S. aureus, or coagulase-negative staphylococci, who were stable after 10 or more days of IV antibiotics and randomized them to switch to oral therapy versus continue IV therapy for the remaining treatment duration 5 / Solid . The oral switch was non-inferior: all-cause mortality, unplanned cardiac surgery, embolic events, or bacteremia at 6 months was 12.1% for oral vs 12.6% for IV (non-inferiority met). This changed practice: early oral step-down is now an accepted approach for stable patients who meet eligibility criteria, reducing hospitalization duration and IV line-related complications.

Surgical indications

Surgery during active IE is associated with higher operative risk than elective valve surgery, but the risks of continued infection and valve destruction are even higher in specific scenarios. The 2015 AHA IE guidelines define surgical indications by urgency 5 / Solid :

Emergency/Urgent (within 24-48 hours):

• Severe acute AR or MR causing cardiogenic shock or refractory pulmonary edema

Urgent (within days):

• Moderate-to-severe valve dysfunction causing refractory heart failure
• Perivalvular extension: abscess, fistula, or new heart block
• Persistent bacteremia or fever after 5-7 days of appropriate antibiotics despite adequate source control
• Large mobile vegetation above 10 mm with an embolic event while on antibiotic therapy
• Large vegetation above 15 mm (even without embolism) in left-sided IE

Non-urgent:

• Recurrent embolic events despite therapy
• Relapsing prosthetic valve endocarditis

The EASE trial and the Kang NEJM 2012 trial (Kang DH, et al., NEJM 2012; 10.1056/NEJMoa1112073) showed that early surgery versus conventional therapy reduced the composite of embolic events, recurrent IE, or death by 41% (HR 0.08 for early surgery primary endpoint, but the trial was small) 4 / Promising .

The IVDU endocarditis surgical debate

IV drug use-related right-sided IE has a specific complexity: the risk of reinfection after valve surgery is substantial if the patient continues to use IV drugs post-operatively. Re-surgery on a replacement valve in an active user is associated with high mortality and raises ethical questions about resource allocation. Multiple center-specific policies exist; increasingly, harm reduction frameworks (medication-assisted treatment for opioid use disorder with buprenorphine or methadone, combined with addiction medicine consultation) are integrated into the surgical decision pathway 4 / Promising . The Society of Thoracic Surgeons has published a statement discouraging policies that categorically exclude IV drug users from surgery, noting that such policies disproportionately harm a marginalized population and conflict with the ethical principle of equal access to life-saving treatment.

6. The Patient Experience

The patient experience of IE divides into two distinct populations: the patient who was previously healthy and develops acute IE (typically S. aureus, from a skin infection, IV drug use, or healthcare exposure), and the patient with underlying cardiac disease (prosthetic valve, prior valve repair, congenital heart disease) who develops a more subacute course (typically viridans streptococcal or enterococcal).

The acute S. aureus IE patient is in a genuine emergency from presentation: high fever, bacteremia, rapidly evolving valve dysfunction, and potentially early systemic emboli. The family of this patient will witness a multi-week hospital stay, possibly an operation on an actively infected heart, and the uncertainty about whether IV antibiotics will hold the infection long enough for surgical timing to be favorable.

The subacute viridans IE patient may have had weeks of fever, malaise, and weight loss before diagnosis. The valve destruction progresses more slowly, providing a longer treatment window but also a longer period of diagnostic uncertainty.

For both groups, the antibiotic course is a minimum of 4-6 weeks (8 weeks for prosthetic valve). This is a profound disruption of normal life. PICC lines, IV antibiotic infusions, weekly laboratory monitoring, and restricted activity. The POET trial’s demonstration of oral step-down eligibility has shortened inpatient requirements for some patients, but the total antibiotic duration remains unchanged.

Prophylaxis: the 2007 AHA guideline revision dramatically narrowed dental prophylaxis recommendations. Antibiotics before dental procedures are now recommended ONLY for patients with:

• Prosthetic cardiac valves (mechanical or bioprosthetic)
• Prior IE
• Congenital heart disease (unrepaired cyanotic CHD, repaired CHD with residual defect adjacent to prosthetic material, completely repaired CHD with prosthetic material in the first 6 months after repair)
• Cardiac transplant with valve disease

The pre-2007 practice of prophylaxis for any underlying cardiac condition (including MVP with MR, bicuspid aortic valve, mitral annular calcification) was abandoned because no RCT ever showed prophylaxis prevented IE from dental procedures, and the antibiotics carried their own risk 5 / Solid .

7. Decisions and Trade-Offs

Timing of surgery

The decision of when to operate in active IE is one of the most complex in all of cardiac surgery. Earlier surgery prevents further cardiac destruction and embolism but operates on an actively infected field with inflamed tissue. Later surgery allows antibiotic therapy to sterilize the vegetation but risks progressive valve damage or embolic events.

The current consensus: operate as soon as possible when any of the urgent/emergency indications are present. Do not wait for arbitrary “antibiotic sterilization” periods that have no evidence basis. The one exception: in patients who have had a recent large stroke (above 2 cm on CT), waiting 2-3 weeks after the stroke to operate reduces the risk of hemorrhagic conversion of the infarcted brain during cardiopulmonary bypass 5 / Solid .

MRSA IE: antifungal approach parallels for resistant organisms

For MRSA IE, vancomycin remains standard but is not ideal: it is bacteriostatic, requires AUC-guided dosing, and has significant nephrotoxicity. Daptomycin at 8-10 mg/kg/day is an alternative, but daptomycin is inactivated by pulmonary surfactant: it should NOT be used for right-sided IE with septic pulmonary emboli (the organism is in the lung parenchyma where it will not be reached). For MRSA right-sided IE, vancomycin is preferred.

The S. bovis colonoscopy imperative

Any patient with S. bovis bacteremia, with or without IE, needs colonoscopy within 3 months if not done recently. The association with colorectal neoplasia is strong enough to be a clinical mandate. This detail is frequently overlooked in the acute management of the IE diagnosis: the infectious disease and cardiology teams manage the heart, and the colonoscopy recommendation is not made.

Clinical Synthesis

Infective endocarditis is a condition where this clinical framework intersects at multiple levels: upstream prevention (protecting the at-risk patient from bacteremia triggers), recognition (patients with prosthetic valves, prior IE, and congenital defects understanding their higher risk), and access to the expert multidisciplinary team that improves outcomes.

The endocarditis team model is now endorsed by major guidelines: centers with dedicated IE teams (cardiology, cardiac surgery, infectious disease, ideally neurology for embolic complications) have better outcomes than those without 4 / Promising . Major endocarditis programs in Illinois: Northwestern Memorial Hospital (Chicago), University of Chicago Medical Center, Rush University Medical Center (Chicago), and Loyola University Medical Center (Maywood).

For patients at increased risk (prosthetic valves, prior IE, specific congenital heart disease), structured remote monitoring includes explicit education about:

• The specific dental prophylaxis indications that apply to them (most do qualify; the general public does not)
• The importance of urgent blood culture collection before antibiotics when unexplained fever persists beyond 48-72 hours
• The red flags that should prompt emergency department evaluation: new murmur with fever, stroke symptoms with fever, splenic or renal infarct symptoms

The patient on the west side of Chicago with S. aureus tricuspid valve IE and IV drug use needs not just six weeks of daptomycin or vancomycin. He needs addiction medicine consultation, housing assistance, naloxone prescription for his household, and ideally buprenorphine initiation before discharge. Without these, the IE is treated and the patient is discharged to the same circumstances that produced it.

That is not a medical failure. That is a system failure. This clinical framework cannot fix opioid addiction policy in Chicago. But it can insist that the clinical system offers the full scope of what medicine can do when it works well.

The Signal Check is fifteen questions mapping the cardiovascular risk pattern across the physiological domains most commonly missed in standard screenings. It produces a specific starting point for your next clinical conversation.