Limited echocardiography and focused hand-held ultrasound by non-experts with decisionsupport for diagnosis and follow-up of heart failure patients.

Abstract

This thesis focused on the use of limited echocardiography and focused hand-held ultrasound examinations performed by non-experts to improve the diagnostics and follow-up of patients with heart failure. Increasing age in the population leads to a higher number of patients suffering from heart failure. The burden for the patients and the health care system includes frequent hospital admissions, increasing costs, as well as a high morbidity and high mortality. There is a need to improve the diagnostic and follow-up workflow, and utilizing non-expert health care personnel supported by technological advances may increase accessibility. The long-term goal is to provide high quality diagnostics at the patients’ point-of-care and reduce hospital admissions, patient transportation, and costs. In the study I, the agreement of limited echocardiographic ultrasound protocols performed by cardiac nurses in combination with telemedical interpretation by an external cardiologist compared to reference exams by an in-hospital cardiologists were evaluated on patients followed up due to heart failure. In the study II-IV, different aspects of focused cardiac ultrasound examinations by general practitioners and cardiac nurses using hand-held ultrasound devices combined with decisionsupport software was explored. The latter studies evaluate the performance and clinical influence of hand-held ultrasound examinations by inexperienced users supported by automatic decision support software assessing left ventricular ejection fraction (autoEF) and the mitral annular plane systolic excursion (autoMAPSE) as well as telemedical image evaluation by an external cardiologist. Study I showed that limited echocardiography by nurses supported by telemedical interpretation could provide comparable results to the reference examination and was considered feasible with reasonable time spent on selected patients. Study II-IV concludes that the feasibility of hand-held ultrasound examinations and automated decision-support software was user dependent. The general practitioners can benefit from adding a focused ultrasound examination and from telemedical support from an expert when evaluating patients with potential heart failure. However, the automatic decision-support tools failed to give reliable outputs and was only to a small degree explained by image quality and the experience of the users. Even experienced cardiologist struggled to achieve reliable results. The results constitute a step in the right direction to increase accessibility to more precise and reliable heart failure diagnostics. The findings of modest feasibility, agreement, reliability, and clinical influence highlight the need for clinical validation of all novel technology before implementation into clinical practice, as well as refinement of the automated decision-support tools studied in this work.