Temporal characteristics of diagnostic parameters measured by a subcutaneously insertable cardiac monitor during heart failure decompensation

Background Early detection and effective management of worsening heart failure (HF) are crucial to preventing disease progression and avoiding hospitalization. Insertable cardiac monitor (ICM) can measure physiological signals that may identify impending HF events. Purpose To evaluate the temporal characteristics of ICM-measured parameters in patients prior to and following a HF decompensation event (HFE). Methods We conducted an observational study involving NYHA class III HF patients implanted with ICMs. The analysis considered a total of 120 days for both the control group (no HFE) and the event group (with HFE). Parameters measured by the device such as activity duration, sleep angle, subcutaneous impedance based fluid volume, and R wave amplitude were analyzed 60 days prior to and 60 days following HFE. For the control group, non-overlapping windows of 120 days were examined for the entire follow-up. The Wilcoxon rank-sum test was used to compare the median differences between the event and control groups. Results There was a total of 104 patients enrolled in the study (mean age:66 ±11 years, 38% females, mean LVEF:44±14%). The event group included 94 HFEs from 39 patients, while the control group comprised 241 non-overlapping windows from 61 patients. Four patients were excluded as they lacked sufficient follow-up. Ensemble average for the different diagnostics parameters across all patients for the event and control groups are shown in figure a. Significant differences were observed in ICM measured parameters between the control and event groups: activity was lower (median difference: 0.018, 95% CI [0.017, 0.019], p < 0.001), sleep angle was higher (median difference: 5.10, 95% CI [4.48, 5.73], p < 0.001), subcutaneous impedance was lower indicating higher fluid (median difference: 54.62, 95% CI [49.21, 60.04], p < 0.001), and R wave amplitude was lower(median difference: 6.25, 95% CI [3.35, 9.87], p < 0.001) for patients with HFE. Conclusion ICM-measured diagnostic parameters exhibited temporal patterns and significant differences in patients with HFEs compared to controls. Integrating multiple ICM parameters into a single score could assist clinicians in identifying patients at a higher risk of HF events.