| Publication Type |
Journal Article |
| Authors |
Illapha Cuba Gyllensten, A. G. Bonomi, Harald Reiter, Oliver Amft |
| Title |
Telemonitoring prediction alarms for detecting upcoming decompensations in heart failure patients using daily weight and trans-thoracic impedance measures |
| Abstract |
Background: Detecting upcoming decompensation events in heart failure (HF) patients could, with the advent of novel sensors in telemonitoring systems, provide new strategies in treating HF remotely. Purpose: The aim of this study was to compare the predictive value for acute HF of trans-thoracic impedance measurements and published methods based on patient weight. Method: A retrospective analysis of a cohort of 91 HF patients using a telemonitoring system for daily weight and trans-thoracic impedance measurements. Average follow-up was 10 months during which 19 adjudicated hospitalizations for HF decompensation occurred. Prediction alarms found in literature were generated from data on weight change and trans-thoracic impedance according to thresholds (rule-of-thumb) and trends (moving average convergence divergence, MACD). Results: A weight increase of or above 2 kg from baseline (previous 28 days) before decompensation occurred in 39% of the patients but only 11% experienced this weight increase rapidly (≥2kg over 3 days). Rule-of-thumb methods on weight change showed poor sensitivity (6%) for high specificity ranges (95%-100%). When applying MACD the decompensation predictive value of weight change improved (36% sensitivity at 95% specificity). Impedance dropped by 10% in 56% of the patients before decompensation. Impedance thresholds showed a sensitivity of 50% at 95% specificity. Conclusion: Rapid weight increase is a specific but insensitive predictor of decompensation when compared to trans-thoracic impedance. A fusion of several sensor modalities could even further improve the prediction of upcoming events of acute heart failure. |
| Publication |
European Journal of Heart Failure Supplements |
| Volume |
11 |
| Issue |
S1 |
| Pages |
S149 |
| Date |
2012 |
