Right-sided endocarditis

Article written and reviewed by Dr Luca Moretti

Right-sided endocarditis is an infection of the endocardium that predominantly involves the tricuspid valve and, less often, the pulmonary valve, subvalvular apparatus or mural surfaces of the right-sided chambers. Vegetations associated with transvenous leads may coexist with valvular infection, but management of the implanted system requires the specific pathway for cardiac device infections. The right-sided location mainly changes the distribution of emboli and haemodynamic consequences: the lung becomes the principal target organ, while valvular insufficiency produces systemic venous congestion and right ventricular volume overload.
This form is not necessarily a mild version of left-sided endocarditis. In many patients medical therapy controls the infection, but persistent sepsis, tricuspid destruction, respiratory impairment and reinfections can produce a severely disabling course.

The most represented populations include people who inject drugs, patients with venous access or intracardiac devices, patients receiving hemodialysis and those with congenital heart disease or previous right-heart surgery. These groups are not equivalent: they differ in age, pathogens, implanted materials, resistance patterns and the feasibility of removing the portal of entry. Injection drug use does not justify automatically attributing every bacteremia to the tricuspid valve, nor does it exclude concomitant left-sided involvement. Similarly, an infected catheter can cause bacteremia without endocarditis: documentation of the cardiac site remains necessary.

Etiology, pathogenesis and pathophysiology

The process begins when bacteremia encounters an endocardial surface favorable to adhesion. Microtrauma, abnormal jets in congenital heart disease, contact with catheters and transvenous leads, and particles introduced into the venous circulation can promote platelet and fibrin deposits. These provide a colonizable substrate but are not essential for all organisms: Staphylococcus aureus has adhesive and invasive properties that allow infection even of previously normal valves. Repeated inocula and persistence of an intravascular source increase the probability that the focus will become established, regardless of the severity of the initial valvular abnormality.

S. aureus is the dominant organism in many series, with the proportion of methicillin resistance depending on local context. Coagulase-negative staphylococci become important in the presence of foreign material; streptococci, enterococci and Gram-negative organisms may be responsible under specific conditions. Identification of Pseudomonas, other difficult-to-treat pathogens or fungi changes prognosis and strategy and makes it inappropriate to extend short regimens studied for selected staphylococcal forms. The species should be identified, accompanied by susceptibility testing and linked to the plausible portal of entry, without deriving treatment from the right-sided location alone.

In devices, biofilm formation promotes persistence on the material even when susceptibility testing shows in vitro sensitivity. A transvenous lead can be colonized from the pocket or during bacteremia and subsequently involve the tricuspid valve through contact and propagation. However, a mass attached to a lead may also consist of fibrin or sterile thrombus: intracardiac material increases risk but does not make every mobile image infectious. This distinction has concrete consequences because definite infection generally requires complete system removal, whereas an incidental finding must be interpreted in context.

The tricuspid vegetation contains microorganisms, fibrin, platelets and inflammatory material. Fragmentation releases septic emboli into the pulmonary arterial bed, where vascular occlusion combines with bacterial inoculation and parenchymal injury. This produces peripheral nodules, infarct-like areas, cavitations and abscesses; extension to the pleura may cause infected effusion or empyema. Repeated emboli increase pulmonary vascular resistance and right ventricular workload, while impaired gas exchange worsens hypoxemia. Pulmonary injury may therefore reduce tolerance of tricuspid regurgitation even though the two components have different origins.

Destruction of leaflets or chordae compromises coaptation and produces tricuspid regurgitation. During systole, part of the ventricular volume flows back into the right atrium, increasing venous pressure and reducing the efficiency of forward output. Initially, the venous system and right ventricle may tolerate substantial regurgitation; with annular dilatation, worsening ventricular function or increased pulmonary resistance, a cycle of progressive congestion develops. The severity of regurgitation therefore does not coincide with the severity of heart failure at every stage and should be assessed together with right ventricular function and pulmonary load.

Isolated pulmonary-valve endocarditis is rare and may be favored by congenital heart disease, conduits or prostheses in the right ventricular outflow tract. The substrate may combine stenosis, regurgitation and foreign material, so the consequences do not match those of a native tricuspid valve. Increased right ventricular pressure may result from outflow obstruction, whereas severe pulmonary regurgitation causes volume overload. Vegetations may be difficult to visualize in standard views: knowledge of surgically modified anatomy and previous examinations guides imaging. These forms often require involvement of an adult congenital heart disease center and are not automatic candidates for short tricuspid regimens.

Systemic arterial embolism is not the usual pathway of a right-sided vegetation. A cerebral, splenic or peripheral event should prompt investigation for concomitant left-sided disease or a right-to-left communication through which infected material can reach the systemic circulation. A patent foramen ovale can become haemodynamically relevant when right-sided pressures rise. Renal manifestations from immune complexes or sepsis can also occur without systemic embolization. Distinguishing these mechanisms avoids interpreting every extracardiac lesion as proof of a second vegetation.

Clinical manifestations

The most characteristic picture combines fever and bacteremia with respiratory symptoms: cough, pleuritic pain, dyspnea and hemoptysis may dominate the presentation and initially suggest pneumonia. Subsequent appearance of infiltrates at different sites, cavitation of peripheral nodules and persistently positive blood cultures instead suggest repeated hematogenous seeding. A partial response to an antibiotic prescribed for a respiratory infection can mask the cardiac source without eliminating it. The absence of an obvious murmur does not reduce suspicion sufficiently when microbiology and pulmonary distribution are concordant.

Tricuspid regurgitation may produce a holosystolic murmur that increases with inspiration, but its intensity does not reliably measure severity, especially in acute disease or at low pressures. Jugular venous distension, prominent systolic venous waves, pulsatile hepatomegaly, edema and ascites indicate systemic congestion. A patient may have major valvular incompetence without cardiogenic pulmonary edema: in right-sided disease, hypoxemia more often results from embolic-infectious lung injury. Coexisting left-sided heart failure, fluid overload or sepsis-related lung injury nevertheless requires separate assessment.

Persistent fever may reflect still-active vegetations, an infected access site, septic thrombophlebitis, pulmonary abscesses or metastatic foci. Low back pain, joint pain, soft-tissue swelling and neurological abnormalities require targeted investigation. In people who inject drugs, it is important to examine injection sites and identify skin lesions, abscesses and thrombosis while maintaining a clinical relationship that permits a reliable history. Untreated pain and withdrawal can hinder investigations and favor premature discharge; they are not issues separate from management of the infection.

Investigations and diagnosis

At least three sets of blood cultures are collected before therapy, when possible without delaying treatment in an unstable patient. Adequate volume and correctly obtained peripheral samples matter more than coincidence with a fever peak. Identification of S. aureus requires investigation of the source, follow-up cultures and assessment for possible secondary foci. In patients with vascular access, catheter and peripheral cultures may help define device-related bacteremia but do not replace endocardial assessment. Previous antibiotics should be documented by molecule, dose and timing because they affect microbiological yield.

Transthoracic echocardiography is particularly useful for the tricuspid valve, which lies close to the chest wall and is often well visualized. The examination describes the attachment site, dimensions and mobility of vegetations, leaflet and chordal integrity, regurgitation, chamber dimensions and right ventricular function. Left-sided valves and, with dedicated views, the pulmonary valve should also be examined. An isolated vegetation measurement is vulnerable to plane-related error and changes after embolization: serial comparison should use comparable images and consider overall anatomical evolution.

Transesophageal echocardiography is indicated when transthoracic imaging is inadequate or discordant, in the presence of devices, suspected left-sided or pulmonary-valve involvement, and persistent unexplained bacteremia. In isolated native right-sided disease with good-quality transthoracic images and unequivocal findings, it may not be required in every case. An initially negative echocardiogram does not exclude early infection: if suspicion remains high, it is generally repeated within five to seven days, sooner if deterioration occurs. Intracardiac echocardiography is used in selected circumstances, mainly procedural, and does not automatically replace the conventional pathway.

Quantification of tricuspid regurgitation requires integration of anatomy, vena contracta, Doppler signals, hepatic venous flow and remodeling. Estimation of pulmonary pressure from the tricuspid jet can be unreliable in very severe regurgitation, when the gradient between the right ventricle and atrium equalizes rapidly. Indices of right ventricular function are also load-dependent: apparently preserved annular excursion does not guarantee effective forward output. The report should therefore explain the mechanism of regurgitation and its consequences, rather than limiting itself to a severity category.

A right-sided mass must be distinguished from normal anatomical structures or non-infected findings such as the Chiari network, a prominent Eustachian valve and fibrin sheaths on transvenous leads. Attachment point, continuity with venous structures, comparison with previous examinations and microbiology are more useful than mobility alone. A thrombus in transit may also appear highly mobile but has a different thromboembolic context and distribution. The simultaneous presence of systemic infection does not necessarily transform a pre-existing finding into a vegetation; conversely, a known anatomical structure can rarely become colonized. The diagnostic question should therefore concern the interaction between the finding and the infection, not merely the name of the mass.

Chest CT identifies peripheral nodules, cavitations, wedge-shaped opacities, collections and pleural complications. CT angiography is selected when vascular involvement must be defined or concomitant thromboembolism distinguished. A cavitary lesion is not specific for septic embolism: malignancies, primary pulmonary infections and vasculitis can produce similar images. Concordance among cultures, echocardiography and lesion distribution supports interpretation. PET/CT may help detect infected material or extracardiac foci, but a negative PET does not exclude a native-valve vegetation.

The 2023 Duke-ISCVID criteria organize microbiology, imaging and systemic manifestations. Septic pulmonary infarcts are included among vascular phenomena, whereas injection drug use is a recognized predisposition; neither element in isolation proves endocarditis. An equivocal image of thrombus on a lead should not be promoted to vegetation merely to achieve a numerical combination. Classification is updated with subsequent cultures and examinations, keeping diagnostic certainty separate from the clinical need to treat an unstable patient.

Clinical Duke-ISCVID 2023 combinations:

• Definite endocarditis: two major criteria; or one major and three minor criteria; or five minor criteria.
• Possible endocarditis: one major and one minor criterion; or three minor criteria, without meeting the requirements for definite or rejected disease.
• A pathological diagnosis can define endocarditis through tissue examination; failure to meet a clinical combination does not automatically make a lesion sterile.

Additional investigations complete the patient assessment: complete blood count, renal and liver function, urinalysis and symptom-guided imaging document organ injury and infectious foci. With appropriate exposures, testing for HIV, HBV and HCV is offered with information and consent because concomitant infections and their treatments influence the pathway. An abnormal neurological examination requires investigation for left-sided involvement or a shunt, in addition to other possible causes. Assessment of substance-use disorder should begin during hospitalization and contribute to concrete treatment planning.

Treatment and prognosis

Initial therapy should be active particularly against staphylococci, including MRSA coverage when epidemiology and severity require it, and broadened according to healthcare exposure, vascular access and previous microbiology. Once the organism is identified, therapy is changed to a targeted regimen. For MSSA, antistaphylococcal beta-lactams or cefazolin are preferred; for MRSA, vancomycin with exposure monitoring or daptomycin in a specialist regimen are used. The Fowler trial supported the use of daptomycin in bacteremia and right-sided endocarditis caused by S. aureus, but dose selection and any combination therapy should reflect subsequent recommendations and the individual case.

Daptomycin is inactivated by pulmonary surfactant and is not a treatment for alveolar pneumonia. This property does not mean that any septic pulmonary embolism automatically excludes its use for an endocardial focus: hematogenous embolization and primary pneumonia are different conditions. However, abscesses, substantial parenchymal involvement or lack of respiratory response require reassessment of coverage and penetration into the foci. This distinction prevents both inappropriate use for pneumonia and indiscriminate exclusion in otherwise treatable right-sided bacteremia.

The usual duration is four to six weeks, adjusted for pathogen, substrate and complications. A two-week regimen with oxacillin or cloxacillin, without an aminoglycoside, applies only to isolated native tricuspid endocarditis caused by MSSA, with rapid clinical and microbiological response within the first 96 hours, a vegetation no larger than 20 mm, and no left-sided infection, prosthesis, metastatic foci or other complications requiring prolonged therapy. Significant renal failure, empyema or severe immunosuppression make automatic extension of this strategy inappropriate. MRSA, vancomycin and device infection are not interchangeable variants of the short-course protocol.

For MSSA treated with a beta-lactam, usual adult doses include oxacillin or cloxacillin at approximately 12 g/day in divided doses, or cefazolin generally 6 g/day, with individual adjustments. Drug choice does not coincide with eligibility for short-course treatment: substituting one molecule does not automatically preserve the evidence supporting a two-week protocol. For MRSA, vancomycin requires exposure and renal-function monitoring; with daptomycin, contemporary recommendations consider high doses, often 10 mg/kg/day, and selected combinations, with creatine kinase monitoring. Weight, renal function and dialysis materially affect dose and interval.

When the organism is a streptococcus or an enterococcus, treatment follows the properties of the pathogen and substrate, not a generic “right-heart” regimen. For susceptible E. faecalis, ampicillin-ceftriaxone synergy may be appropriate, particularly when aminoglycoside toxicity is to be avoided; for fungi and difficult Gram-negative organisms, the team considers the probability of surgical source control early. A culture that identifies multiple organisms requires distinction among polymicrobial infection, contamination and concomitant sources. In all cases, fever persisting with sterile blood may result from embolic necrosis or collections and does not automatically demonstrate resistance of the isolated organism.

Clearance of bacteremia is documented with subsequent cultures. If positivity persists, it is verified that the drug is active and correctly dosed, and infected catheters, thrombophlebitis, collections and left-sided involvement are sought. An aminoglycoside should not be added automatically to compensate for a source that has not been removed, particularly because of renal toxicity. Accesses that are no longer necessary are removed; cardiac devices with definite infection require an extraction strategy. Pulmonary abscesses and pleural collections may prolong treatment and require drainage even while the valvular focus is improving.

Surgery is discussed in the presence of persistent bacteremia after about one week of appropriate therapy, severe acute tricuspid regurgitation with right-sided dysfunction not controlled by diuretics, recurrent pulmonary emboli with respiratory compromise, or concomitant left-sided involvement with a surgical indication. Residual vegetations larger than 20 mm after recurrent pulmonary emboli constitute an important indication; size alone, without context, does not mandate intervention. Instability and anatomical progression may require action before the time threshold: it is not a minimum period during which worsening sepsis or heart failure should be tolerated.

When technically feasible, tricuspid valve repair preserves the native apparatus and limits foreign material, provided it permits complete removal of infected tissue. An extensively destroyed valve may require replacement; prosthesis choice considers thrombogenicity in the right-sided position, anticoagulation, age and reinfection risk. Valvectomy without replacement leaves free regurgitation and may be tolerated only in selected circumstances: elevated pulmonary resistance or reduced right ventricular reserve markedly worsens the consequences. Avoiding a prosthesis is not always equivalent to preserving acceptable function.

Percutaneous aspiration of vegetations may reduce the infected mass in selected patients, often those at high surgical risk or in preparation for transvenous lead extraction. It does not reconstruct a destroyed leaflet, does not guarantee removal of biofilm and does not replace antimicrobial therapy. Evidence comes predominantly from observational series with substantial case selection; reducing echocardiographic size does not demonstrate improved survival. Bleeding, vascular injury, embolization and worsening regurgitation should be included in the procedural assessment.

Anticoagulants are not initiated to treat septic emboli as if they were ordinary venous thromboembolism. Hemoptysis, infected vascular lesions and possible cerebral complications may increase bleeding risk; a separate indication, such as documented venous thrombosis, requires an individualized balance. Right-sided congestion may benefit from diuretics and fluid-balance management, avoiding both excessive preload and reduced perfusion in a septic patient. Respiratory treatment and drainage of collections should accompany control of the cardiac source.

The choice of oral continuation therapy should be distinguished from the social need to discharge the patient. Specific experiences and regimens exist for selected right-sided forms, but they require a susceptible organism, source control, and verifiable absorption and adherence. Interactions and resistance make it inappropriate to automatically revive older combinations merely because they avoid venous access. OPAT assessment considers infusion feasibility, catheter care, blood monitoring and patient circumstances; a history of injection drug use alone guarantees neither infeasibility nor safety. The pathway is built together with treatment of substance-use disorder and community support.

In opioid use disorder, buprenorphine or methadone when indicated, withdrawal treatment, analgesia and support for continuity of care are parts of treatment. Access to services, housing availability, harm reduction and overdose prevention should be addressed, without automatically making an indicated procedure conditional on prior demonstrated abstinence. Imminent self-discharge requires a shared plan to reduce harm and complete as much therapy as possible. Assessment for OPAT or oral treatment is based on an appropriate regimen, stability and individual feasibility; the POET study of left-sided disease does not constitute direct validation for this population.

Prognosis depends on the organism, source control, right ventricular function, pulmonary injury and the ability to complete care. Relatively low early mortality in young patients with isolated tricuspid disease does not describe outcomes in older dialysis patients or device recipients. After treatment, residual regurgitation, ventricular function and respiratory status should be documented, distinguishing echocardiographic persistence of an organized vegetation from active infection. Relapse and reinfection require new cultures and microbiological comparison; reinfection may arise from a new exposure even after adequate initial cure.

Complications

Pleuropulmonary complications can progress from embolic nodules to cavitations, abscesses and empyema, with respiratory failure. Rupture of a peripheral cavity may cause pneumothorax or a bronchopleural communication; significant hemoptysis requires vascular assessment and urgent source control. Improvement in bacteremia does not imply immediate resolution of collections, in which antibiotics may be insufficient without drainage. Respiratory deterioration during therapy should therefore reopen anatomical assessment rather than being automatically attributed to unavoidable emboli or a generically slow response.

Right heart failure may persist after sterilization because of irreversible loss of valvular tissue and chamber remodeling. Venous congestion contributes to renal and hepatic dysfunction, while residual pulmonary vascular injury increases afterload. Late reassessment should determine whether the patient has tolerated regurgitation or progressive dysfunction requiring intervention, without confusing absence of fever with cardiovascular recovery. Timing of delayed correction depends on function, symptoms and documented control of infection.

Septic thrombophlebitis associated with vascular access or peripheral injections may constitute a second source of pulmonary emboli and bacteremia. In this setting, catheter removal or treatment of the valve does not necessarily eliminate the infected thrombus. Pain, limb edema and persistent positivity point toward vascular ultrasonography or appropriate imaging, with drainage of adjacent collections when necessary. The indication and duration of anticoagulation depend on thrombosis, site, bleeding and other complications; they are not inferred from the mere presence of septic emboli. Recognizing the venous source avoids attributing every new event to the tricuspid vegetation.

Sepsis, glomerulonephritis, drug toxicity and hypoperfusion may combine in multiorgan injury. Paradoxical embolism can also cause stroke and systemic ischemia, particularly when right-sided pressures are elevated. Closing a shunt during active infection is not an automatic response: introducing material into an infected field can create a new problem, and the strategy must be coordinated with source control. In the long term, reinfection and new exposures can reproduce disease on a valve that is already damaged, making cardiological follow-up and treatment of substance-use disorder closely interdependent.