Remote monitoring in heart failure with a focus on thoracic bioimpedance: challenges and opportunities

Abstract

Heart failure is a prevalent disease with high morbidity and mortality rates and a high number of (re)hospitalizations. The major reason for heart failure (re)hospitalizations is related to congestion or fluid overload. Patients with heart failure benefit from regular follow-up and monitoring of biomedical parameters for optimization of treatment strategies and early detection of disease progression. This can be done by invasive and non-invasive remote monitoring strategies. In this thesis, we added novel insight to the organization of a structured remote monitoring program for patients with cardiovascular implantable electronic devices, with a focus on the disease management strategies. First, we demonstrated a substantial amount of remote monitoring alerts with the largest amount being disease-related. Frequent phone contacts were triggered by these alerts and in most cases, general heart failure education sufficed. Next, we found that bioimpedance alerts constitute a substantial amount of incoming alerts when turned on during remote follow up. Although very promising for the early detection of impeding congestion, they did not show an influence on survival outcome. Though, a possible influence on heart failure-related hospitalizations, driven by the high number of heart failure education coupled to bioimpedance alerts, was observed. Next, we studied the use of non-invasive bioimpedance monitoring devices in congestive heart failure patients. Non-invasive bioimpedance measures correlated well with a patient's fluid balance on the individual level. On the population level, correlations are lower due to the high individual variability and different locations where the excessive fluid is coming from. Interestingly, the wearable bioimpedance device was capable of tracking peripheral edema in the case of a patient with right-sided heart failure. Furthermore, it has become clear that a lot of external factors exert an influence on bioimpedance measures, especially body posture. Most importantly, we observed that patients who do not show an improvement in thoracic impedance have a worse clinical outcome, which indicates its use as a prognostic parameter for clinical outcome. Finally, we demonstrated that the combination of remote monitoring with decision support algorithms is feasible and safe to use and leads to excellent patient satisfaction and adherence. However, due to the high quality of the usual care in our center, this tool might be more suitable in centers without intensive heart failure disease management programs.