The added value of transesophageal echocardiography (TEE) over transthoracic echocardiography in the assessment of left-sided infective endocarditis has been extensively validated in the literature. Little research has dealt with the role of echocardiography in right-sided infective endocarditis (RSE), however. In this review, the differences between RSE and left-sided endocarditis and the different types of RSE according to the types of patients who have the disease are described. Both issues have important implications for echocardiographic workup. Moreover, a systematic echocardiographic protocol to avoid missing right-sided vegetations and several specific morphologic aspects of RSE are reviewed. Normal right-sided structures, which may mimic vegetations, particularly when the clinical picture is compatible, are described. Finally, the value of transthoracic echocardiography and TEE in RSE is reviewed according to the publications available. The diagnostic yield of transthoracic echocardiography is comparable with that of TEE in intravenous drug users. On the contrary, TEE is mandatory in patients with cardiac devices. A Bayesian-based diagnostic approach is proposed for a third poorly characterized group of patients with RSE who are not drug addicts, have no cardiac devices, and have no left-sided endocarditis (the “three no’s” endocarditis group).
Attention ASE Members:
ASE has gone green! Visit www.aseuniversity.org to earn free CME through an online activity related to this article. Certificates are available for immediate access upon successful completion of the activity. Non-members will need to join ASE to access this great member benefit!
Differences Between Right-Sided and Left-Sided Infective Endocarditis
RSE and LSE have not been systematically compared. However, descriptive series on RSE and LSE give us enough information to assert that RSE has unique features. Table 1 , obtained with data from our institution, summarizes those differences and supports the concept that RSE and LSE are quite different with regard to epidemiology, microbiology, morphology, and outcomes. Patients with RSE are younger, more frequently have antecedents of intravenous drug use, less often have diabetes mellitus, have a specific microbiologic pattern (with Staphylococcus aureus being the dominant organism), have larger vegetations, have fewer PACs, and have better outcomes. Moreover, pulmonary thromboembolism is more frequent, as expected. Older series of intravenous drug users with RSE showed similar results : S. aureus was found to be the causative microorganism in 82% of cases, 80% had vegetations > 1 cm, and mortality was 7%. No PACs were reported. More recently, a series of 220 intravenous drug users with endocarditis, 88% of them with RSE, showed 77% S. aureus infection and 6% short-term mortality. One of the patients who died had a tricuspid vegetation and an interventricular abscess.
| Variable | RSE ( n = 135) | LSE ( n = 692) | P |
|---|---|---|---|
| Age (y) | 54 ± 19 | 62 ± 14 | <.001 |
| Men | 74% | 63% | .01 |
| Intravenous drug users | 29% | 3% | <.001 |
| Diabetes mellitus | 10% | 21% | <.01 |
| Positive blood cultures | 83% | 71% | <.01 |
| S. aureus | 40% | 16% | <.001 |
| Coagulase negative Staphylococci | 30% | 15% | <.001 |
| Pulmonary thromboembolism | 25% | 0.5% | <.001 |
| Maximal vegetation diameter (mm) | 16.4 ± 8 | 13.7 ± 7 | <.01 |
| Periannular complications | 7% | 26% | <.001 |
| In-hospital mortality | 7% | 26% | <.001 |
Differences Between Right-Sided and Left-Sided Infective Endocarditis
RSE and LSE have not been systematically compared. However, descriptive series on RSE and LSE give us enough information to assert that RSE has unique features. Table 1 , obtained with data from our institution, summarizes those differences and supports the concept that RSE and LSE are quite different with regard to epidemiology, microbiology, morphology, and outcomes. Patients with RSE are younger, more frequently have antecedents of intravenous drug use, less often have diabetes mellitus, have a specific microbiologic pattern (with Staphylococcus aureus being the dominant organism), have larger vegetations, have fewer PACs, and have better outcomes. Moreover, pulmonary thromboembolism is more frequent, as expected. Older series of intravenous drug users with RSE showed similar results : S. aureus was found to be the causative microorganism in 82% of cases, 80% had vegetations > 1 cm, and mortality was 7%. No PACs were reported. More recently, a series of 220 intravenous drug users with endocarditis, 88% of them with RSE, showed 77% S. aureus infection and 6% short-term mortality. One of the patients who died had a tricuspid vegetation and an interventricular abscess.
| Variable | RSE ( n = 135) | LSE ( n = 692) | P |
|---|---|---|---|
| Age (y) | 54 ± 19 | 62 ± 14 | <.001 |
| Men | 74% | 63% | .01 |
| Intravenous drug users | 29% | 3% | <.001 |
| Diabetes mellitus | 10% | 21% | <.01 |
| Positive blood cultures | 83% | 71% | <.01 |
| S. aureus | 40% | 16% | <.001 |
| Coagulase negative Staphylococci | 30% | 15% | <.001 |
| Pulmonary thromboembolism | 25% | 0.5% | <.001 |
| Maximal vegetation diameter (mm) | 16.4 ± 8 | 13.7 ± 7 | <.01 |
| Periannular complications | 7% | 26% | <.001 |
| In-hospital mortality | 7% | 26% | <.001 |
Classification of Right-Sided Infective Endocarditis
According to the type of patient who hosts the infective process, we can distinguish four different types of RSE. As discussed later, this classification is closely related to the echocardiographic approach.
RSE in intravenous drug users is more frequent in young human immunodeficiency virus seropositive and immunosuppressed patients, and S. aureus is the predominant organism, with methicillin-resistant strains becoming more prevalent. The site most frequently affected is the tricuspid valve, and mortality is in the range of 7%. Remarkably, recurrent infective endocarditis is more common in intravenous drug users, and the median interval between episodes is shorter in addicts than in nonaddicts. Pathophysiology has yet to be proven. Damage to the valves from injected matter, large injected bacterial inocula, and abnormalities of immune function have been incriminated.
RSE in patients with implantable cardiac devices is an infection not limited to the pocket, which would be called local device infection, but one that extends to the leads. Its frequency is increasing given the rising number of patients with implantable cardioverter defibrillators. Staphylococci are also the most frequent microorganisms found, predominantly coagulase-negative Staphylococci . The period from the initiation of symptoms to diagnosis may be very long. Pathophysiology is well known: the process begins after contamination by local bacteria at the time of implantation or during late manipulation. The therapeutic strategy must include extraction of the device, preferably by a percutaneous approach. A recent descriptive study of 100 patients with RSE in cardiac devices demonstrated that percutaneous lead extraction is feasible and safe. No patient required surgical extraction, and only five patients had complications associated with the procedure: embolization of the vegetation in two, embolization of a lead fragment to the pulmonary artery in one, severe tricuspid regurgitation secondary to a flail posterior leaflet in one, and hypotension that required vasopressor support in one. Finally, reported short-term mortality is in the range of 5% to 11%, with 1-year mortality of 27%.
RSE in patients who are not intravenous drug users, who have no implanted cardiac devices, and who have no LSE (the “three no’s” endocarditis group) is an unsettled entity. Most published reports consist of old descriptions of single cases. The largest series, described by Nandakumar and Raju, summarizes the characteristics of 29 cases of tricuspid valve endocarditis reported on by different groups during the last decades of the previous century. A study by Naidoo analyzed 15 cases, but it was a retrospective analysis that also included cases of LSE. Our group described a prospective series of 17 patients and found that they were young patients, one third of them had intravascular catheters, S. aureus was the most frequent causative agent, surgery was needed in 29%, and mortality was 12%. Interestingly, PACs were seen in 17% of patients (all three were fistulae; no abscesses or pseudoaneurysms were seen).
Last, RSE can affect patients with right-heart congenital abnormalities. This condition is rare in adults and is beyond the scope of this review. We refer the reader to a comprehensive review on this topic.
Echocardiographic Protocol for Right-Sided Structures
A complete transthoracic examination of the right heart must include standard-plane views as well as modified views. The standard long-axis view from the parasternal position excludes the right-sided heart, except a small part of the right ventricular outflow tract. This view does not visualize right-sided structures to which vegetations attach and rarely helps in ruling out endocarditis. Tilting the head of the transducer toward the left shoulder, which angles the central ray slightly rightward, and rotating it 15° to 30° clockwise provides the parasternal long axis of the right ventricular inflow tract ( Figure 1 A). Tilting it to the opposite direction, which angles the central ray slightly leftward, provides the parasternal long axis of the pulmonary artery, the anterior leaflet on the upper right, and the posterior on the bottom left ( Figure 1 B). The former is crucial to interrogate the tricuspid and the Eustachian valves. When the transducer is tilted enough, the image shows the posterior leaflet of the tricuspid valve on the left and the anterior leaflet on the right. This is the only view in which the posterior leaflet, the smallest one, is seen. If the tilting is less pronounced, the valve seen on the left is the septal. The remaining projections show the anterior leaflet, the largest one and the most frequent site for vegetations to attach, and the septal leaflet. The parasternal long-axis view of the right ventricular inflow tract, or right ventricular inflow view, is adequate for visualizing the coronary sinus as it enters the right atrium adjacent to the tricuspid annulus. The inferior vena cava is seen entering the right atrium inferior to the coronary sinus. The Eustachian valve is usually found at the junction of the inferior vena cava and the right atrium. Exceptionally, a tiny Thebesian valve can be visualized within the coronary sinus ( Figure 2 ). Finally, the parasternal short-axis view shows the entrance of the inferior vena cava, the Eustachian valve, and the lateral and anterior leaflets of the tricuspid valve and the pulmonary valve.
The apical views are useful for assessment of the morphology and function of the tricuspid valve. Finally, a subxiphoid examination must be always performed. By slowly rotating the transducer clockwise, one can go from a long-axis view of the heart, in which the Eustachian and tricuspid valves are seen, to a long-axis view of the right ventricular outflow tract.
The transesophageal examination must include the four-chamber view, the two-chamber view, the frontal long-axis view of the coronary sinus, the gastric short-axis view, and the basal short-axis view, the latter to specifically interrogate the pulmonary valve. The main advantage of TEE over TTE lies in patients with implanted cardiac devices. Vegetations are frequently located at the level of the upper part of the right atrium and even into the superior vena cava, and both are areas nicely targeted with TEE but poorly evaluated with TTE.
Specific Morphologic Aspects of Right-Sided Infective Endocarditis
Right-sided vegetations can be attached to normal structures (tricuspid, Eustachian, or pulmonary valves; Figures 3–5 , Video 1 ), to cardiac devices (pacemakers [ Video 2 ] and implantable cardioverter-defibrillators), or to central lines. Even the right atrial wall can be a site for a vegetation to settle. Irrespective of the location of the infection, the presence of an endothelial lesion, pannus, or fibrin thrombotic material should precede the development of a vegetation. Vegetations are larger than those found in LSE, which is probably related to the lower right chamber pressures, permitting rapid growth of the vegetation. This is in keeping with the fact that vegetations tend to adhere to the surface of the leaflets facing the lower pressure chamber (the atrium in the mitral valve, the ventricle in the aortic valve). Moreover, PACs are very rarely encountered in RSE. Most series do not report the rate of PACs. However, as mentioned previously, PACs have been found in patients with “three no’s” endocarditis. It is worth mentioning that although PACs are rare in RSE, right-sided structures should be interrogated in patients with LSE. Aortic abscesses can extend through the membranous or ventricular septum into the right side, involving the tricuspid apparatus near the septal leaflet or even the pulmonary valve.
