An Approach to Improve the Negative Predictive Value and Clinical Utility of Transthoracic Echocardiography in Suspected Native Valve Infective Endocarditis

Abstract

Background: In patients with suspected native valve infective endocarditis, current guidelines recommend initial transthoracic echocardiography (TTE) followed by transesophageal echocardiography (TEE) if clinical suspicion remains. The guidelines do not account for the quality of initial TTE or other findings that may alter the study's diagnostic characteristics. This may lead to unnecessary TEE when initial TTE was sufficient to rule out vegetation.

Methods: The objective of this study was to determine if the use of a strict definition of negative results on TTE would improve the performance characteristics of TTE sufficiently to exclude vegetation. A retrospective analysis of patients at a single institution with suspected native valve endocarditis who underwent TTE followed by TEE within 7 days between January 1, 2007, and February 28, 2014, was performed. Negative results on TTE for vegetation were defined by either the standard approach (no evidence of vegetation seen on TTE) or by applying a set of strict negative criteria incorporating other findings on TTE. Using TEE as the gold standard for the presence of vegetation, the diagnostic performance of the two transthoracic approaches was compared.

Results: In total, 790 pairs of TTE and TEE were identified. With the standard approach, 661 of the transthoracic studies had negative findings (no vegetation seen), compared with 104 studies with negative findings using the strict negative approach (meeting all strict negative criteria). The sensitivity and negative predictive value of TTE for detecting vegetation were substantially improved using the strict negative approach (sensitivity, 98% [95% CI, 95%-99%] vs 43% [95% CI, 36%-51%]; negative predictive value, 97% [95% CI, 92%-99%] vs 87% [95% CI, 84%-89%]).

Conclusions: The ability of TTE to exclude vegetation in patients is excellent when strict criteria for negative results are applied. In patients at low to intermediate risk with strict negative results on TTE, follow-up TEE may be unnecessary.

Keywords: Echocardiography; Infective endocarditis; Screening; Transesophageal echocardiography.

Figure 1

Study cohort derivation and strict negative criteria This figure outlines the derivation of the study cohort and shows the list of strict negative criteria Abbreviations: TTE, Transthoracic Echocardiogram; TEE, Transesophageal Echocardiogram; LVAD, Left Ventricular Assist Device; *Normal anatomy defined as tricuspid aortic, pulmonic, and tricuspid valves, mitral valve without mitral annular calcification, no mitral valve prolapse, no atrial septal defect or ventricular septal defect, repaired or unrepaired.

Figure 2

A. Flow diagram showing the results of using the standard approach. B. Flow diagram showing the results of using the strict negative approach. Abbreviations: TTE, Transthoracic Echocardiogram; TEE, Transesophageal Echocardiogram

Figure 3

Test indication curve comparing the negative likelihood ratios for the standard and strict negative approach. This figure uses the negative likelihood ratio to display the relationship between pre and post test probability for each approach. In order to focus on the improved negative predictive value when using the strict negative criteria, the positive likelihood ratio curves were not included. Abbreviations: TTE, Transthoracic Echocardiogram;