Discrepancies between the decisions of an endocarditis team and the modified Duke's Criteria for the diagnosis of infective endocarditis

Abstract

Background/Introduction The 2015 ESC endocarditis guidelines recommend that a multi-disciplinary endocarditis team should be involved in caring for patients with endocarditis. This approach recognises the heterogeneity of infective endocarditis, and the difficulty in distinguishing endocarditis from other infective and non-infective diseases. Purpose This study looks at suspected cases of infective endocarditis discussed at the endocarditis MDT, discrepancies between MDT outcome and the modified Duke's criteria in diagnosing endocarditis, and number of referrals to MDT over time. Methods Patients were identified by review of MDT outcome forms. Demographic data, predisposing conditions, imaging findings, microbiology results and final diagnosis were identified by reviewing MDT forms and electronic charts. Results 234 patients were identified from MDT outcome forms. 118 (50.4%) patients were over 65, 165 (70.5%) were male, and 65 (27.8%) had previous valve surgery. Crosstabulation of MDT decision against outcome by modified Duke's criteria are seen in Table 1. The endocarditis team identified 7 patients with definite or possible endocarditis who were excluded by applying the modified Duke's criteria. 1 of these patients required aortic valve replacement. 2 patients with disseminated Staphylococcal infections without endocarditis were classed as definite endocarditis by the modified Duke's criteria. Referral data and outcomes are included in Figure 1. Conclusions The endocarditis team can be used to improve the sensitivity and specificity of the modified Duke's criteria. As the endocarditis team became established the volume of referrals has increased, along with the number of patients who have alternate diagnoses. Practitioners should bear this in mind when establishing endocarditis teams. Figure 1. Referrals over time by outcome Funding Acknowledgement Type of funding source: None