Evaluation of the Usefulness of Echocardiography in Patients With Staphylococcus Aureus Bacteremia (ET-AUREUS Study).

2- Hypothesis: Prospective evaluation of a cohort of patients with OAB, who undergo TEE, TTE, will provide insight into the actual risk of IE in those patients at low risk.

3- Objectives:

• Main objective:
• To estimate the prevalence of IE, defined by the modified Duke criteria, in patients with low risk according to the VIRSTA scale.
• To determine whether the risk of IE in patients with OAB identified as low risk using the VIRSTA scale is low enough to safely omit transthoracic or transesophageal echocardiography using this scale.
• Secondary objectives
• To determine whether the risk of IE in patients with SAAB identified as low risk using the PREDICT or POSITIVE scales is low enough to safely omit transthoracic or transesophageal echocardiography.
• To identify risk factors for IE in patients with SAAB.
• Determine whether the risk of IE in patients with no risk factors is low enough to safely omit transthoracic or transesophageal echocardiography.
• Estimate the number of echocardiographies that could be avoided by these approaches.
• Estimate the cost-benefit of omitting echocardiographic testing in patients identified as low risk by the above methods. 3- Material and methods: *Design: Multicenter prospective cohort study. *Project execution sites: Hospitals with the capacity to detect AFB, contact the patient within 72-96h and the possibility of performing transthoracic echocardiography (TTE) and transesophageal echocardiography (TEE) in the center itself. Currently 13 collaborating centers plus the sponsoring center (Hospital Puerta de Hierro).
• Sample size calculation: Sample size calculation was performed in order to estimate the prevalence of IE in patients with OAB identified as low risk using the VIRSTA scale. For a confidence level of 95%, with a loss rate of 10%, precision of 2% and an expected proportion of 2% of IE in low-risk patients, a sample size of 210 VIRSTA-negative patients was calculated. Given that at inclusion it will be unknown whether the patient is VIRSTA negative and that, among the objectives of the study is the identification of other subgroups of low-risk patients, taking into account that approximately 45% of patients with AFB are VIRSTA negative, a total sample size of 470 patients with AFB is estimated.
• Intervention:

Systematic performance of TTE and TEE in all included patients. TTE will be performed and reported first, followed by TEE (same or different days). Both tests should be performed within 10 days of the patient's inclusion in the study.

The studies will be performed in each of the centers participating in the project with the latest generation echocardiographic equipment commonly used in each center. All echocardiograms will be anonymized and assigned a correlative patient number.

The acquisition of the images of the vegetation or infectious complication will be zoomed and 3 consecutive cardiac cycles will be recorded and sent in DICOM format (both cine and image) to the centralized imaging laboratory, where interpretation will be performed by at least two cardiologists with expertise in echocardiography.

The cardiologists interpreting the echocardiograms centrally will be blinded to any clinical, microbiological and other complementary test data of the patient, as well as to the local interpretation of the images. The cardiologist interpreting the TTE will be blind to the TEE and vice versa.

• Definitions of positive echocardiographic findings for IE:

Echocardiograms with at least one of the following findings will be considered positive or suggestive of IE:

• Vegetation: echogenic mass, usually mobile attached to an endocardial surface or intracardiac device. Most frequently attached to a valve and with independent movement. They are usually located on the ventricular side of ventriculo-arterial valves and on the atrial side of atrioventricular valves.
• Perivalvular abscesses/infection: areas of inhomogeneous perivalvular thickening, may be echodense or echolucent, usually located in the valve annuli.
• Fistula: communication between two adjacent cavities, the Doppler-color passes from one cavity to another depending on the pressure in both.
• Pseudoaneurysm: pulsatile perivalvular cavity, without echoes inside, into which flow enters, detected by doppler-color.
• Perforation: Destruction of the valvular surface that causes acute valvular insufficiency; the valvular defect is usually seen by 2D and the insufficiency that passes through it, which is usually eccentric. Multiple perforations may coexist in the same valve.
• Endocarditis on prosthetic valve: in addition to all of the above, images compatible with paravalvular leak of new appearance, detected with color Doppler, or valve dehiscence will be considered.
• Predefined cutoff points to reasonably rule out IE:

IE will be considered reasonably ruled out when the estimated risk of IE is low enough that the costs and risks of the tests needed for diagnosis (TTE or TEE) outweigh the benefits.

In defining cutoff points for this low risk of IE, we consider several recent publications that estimate the risk of IE above which a patient benefits from such evaluations (8,24), as well as the usefulness of the negative likelihood ratio (LR(-) or likelihood ratio (-)) in predicting this risk, as has been reported by other authors (25). In line with these studies, the following cut-off points are predefined:

1. If the risk of IE was less than 0.5% and the CP(-) of the scale less than 0.05, IE would be considered reasonably ruled out without the need for any echocardiographic assessment;
2. If the risk of IE was between 0.5-2% or the CP(-) of the scale between 0.05-0.10, IE would be considered reasonably ruled out without the need for TEE when a negative TTE with good visualization of the heart valves is available.
3. If the risk of IE was between 2-5% or the CP(-) of the scale between 0.10-0.20, IE could be considered reasonably ruled out with a negative TTE (without the need for TEE) only if the prevalence of IE after TTE was less than 1%. Otherwise, the use of TEE would be considered necessary.
4. If the risk of IE was greater than 5% or the CP(-) of the scale was greater than 0.20, TEE would be considered necessary to safely rule out IE.

• Variables to be studied:

Demographic data, comorbidities, mode of acquisition and primary focus of bacteremia, clinical, microbiological, echocardiographic and evolution data from patients' medical records, as well as echocardiographic findings will be collected. Data will be collected from the electronic history of each center and included in an anonymized form in an online medical database platform (eCDR).

*Statistical analysis: Quantitative variables will be presented as median and 25th and 75th percentiles; and qualitative variables by absolute and relative frequencies.

An interval estimation (95%) of the prevalence of IE in patients identified as low risk will be performed using the VIRSTA, PREDICT and POSITIVE scales. The prevalence of IE in patients without any of the identified factors will also be estimated using secondary analyses (see below). The diagnostic validity indices sensitivity, specificity, negative (PC-) and positive (PC+) likelihood ratios, and negative (NPV) and positive predictive values (PPV) vs. the gold standard (modified Duke criteria) will be estimated. With these data, we will estimate the risk of IE in an identified low-risk patient from a population with a prevalence similar to ours. Echocardiography (TTE and/or TEE) will be considered dispensable in patients with SAAB when the estimated risk of IE in patients identified as low risk and the CP(-) of the scale is below the aforementioned predefined cut-off points.

For the study of IE risk factors, univariate inferential analysis will be performed using the Mann-Whitney U test for quantitative variables and with the χ2 test or Fisher's exact test when necessary for qualitative variables. Those variables that are clinically and statistically significant will be included in a multivariate logistic regression model.

Finally, the number of echocardiographic tests, both TTE and TEE that could have been safely omitted using the different approaches will be estimated. For the cost-benefit study, the following risks will be taken into account: increased mortality due to nondiagnosis of IE: 15% (2), increased mortality due to TEE: 0.1% (3), increased mortality due to TTE: 0%. The costs of performing tests are estimated following the catalog of public prices of the Community of Madrid (Order 727/2017): cost TEE: €80.

Bilateral p-values of less than 0.05 will be considered statistically significant. All statistical analyses will be performed using the SPSS version 25 software package.