Prediction and Analysis of Heart Failure Decompensation Events Based on Telemonitored Data and Artificial Intelligence Methods.

Cardiovascular diseases are the leading cause of death globally, taking an estimated 17.9 million lives each year. Heart failure (HF) occurs when the heart is not able to pump enough blood to satisfy metabolic needs. People diagnosed with chronic HF may suffer from cardiac decompensation events (CDEs), which cause patients' worsening. Being able to intervene before decompensation occurs is the major challenge addressed in this study. The aim of this study is to exploit available patient data to develop an artificial intelligence (AI) model capable of predicting the risk of CDEs timely and accurately. Materials and Methods: The vital variables of patients (n = 488) diagnosed with chronic heart failure were monitored between 2014 and 2022. Several supervised classification models were trained with these monitoring data to predict CDEs, using clinicians' annotations as the gold standard. Feature extraction methods were applied to identify significant variables. Results: The XGBoost classifier achieved an AUC of 0.72 in the cross-validation process and 0.69 in the testing set. The most predictive physiological variables for CAE decompensations are weight gain, oxygen saturation in the final days, and heart rate. Additionally, the answers to questionnaires on wellbeing, orthopnoea, and ankles are strongly significant predictors.

Intermittent inotropic support with levosimendan in advanced heart failure as destination therapy: The LEVO-D registry.

AIM: Patients with advanced heart failure (AHF) who are not candidates to advanced therapies have poor prognosis. Some trials have shown that intermittent levosimendan can reduce HF hospitalizations in AHF in the short term. In this real-life registry, we describe the patterns of use, safety and factors related to the response to intermittent levosimendan infusions in AHF patients not candidates to advanced therapies. METHODS AND RESULTS: Multicentre retrospective study of patients diagnosed with advanced heart failure, not HT or LVAD candidates. Patients needed to be on the optimal medical therapy according to their treating physician. Patients with de novo heart failure or who underwent any procedure that could improve prognosis were not included in the registry. Four hundred three patients were included; 77.9% needed at least one admission the year before levosimendan was first administered because of heart failure. Death rate at 1 year was 26.8% and median survival was 24.7 [95% CI: 20.4-26.9] months, and 43.7% of patients fulfilled the criteria for being considered a responder lo levosimendan (no death, heart failure admission or unplanned HF visit at 1 year after first levosimendan administration). Compared with the year before there was a significant reduction in HF admissions (38.7% vs. 77.9%; P < 0.0001), unplanned HF visits (22.7% vs. 43.7%; P < 0.0001) or the combined event including deaths (56.3% vs. 81.4%; P < 0.0001) during the year after. We created a score that helps predicting the responder status at 1 year after levosimendan, resulting in a score summatory of five variables: TEER (+2), treatment with beta-blockers (+1.5), Haemoglobin >12 g/dL (+1.5), amiodarone use (-1.5) HF visit 1 year before levosimendan (-1.5) and heart rate >70 b.p.m. (-2). Patients with a score less than -1 had a very low probability of response (21.5% free of death or HF event at 1 year) meanwhile those with a score over 1.5 had the better chance of response (68.4% free of death or HF event at 1 year). LEVO-D score performed well in the ROC analysis. CONCLUSION: In this large real-life series of AHF patients treated with levosimendan as destination therapy, we show a significant decrease of heart failure events during the year after the first administration. The simple LEVO-D Score could be of help when deciding about futile therapy in this population.

Prediction of heart failure decompensations using artificial intelligence and machine learning techniques / Predicción de las descompensaciones de la insuficiencia cardíaca mediante técnicas de inteligencia artificial y aprendizaje automático

Abstract Introduction: Heart failure (HF) is a major concern in public health. We have used artificial intelligence to analyze information and improve patient outcomes. Method: An Observational, retrospective, and non-randomized study with patients enrolled in our telemonitoring program (May 2014-February 2018). We collected patients’ clinical data, telemonitoring transmissions, and HF decompensations. Results: A total of 240 patients were enrolled with a follow-up of 13.44 ± 8.65 months. During this interval, 527 HF decompensations in 148 different patients were detected. Significant weight increases, desaturation below 90% and perception of clinical worsening are good predictors of HF decompensation. We have built a predictive model applying machine learning (ML) techniques, obtaining the best results with the combination of "Weight + Ankle + well-being plus alerts of systolic and diastolic blood pressure, oxygen saturation, and heart rate." Conclusions: ML techniques are useful tools for the analysis of HF datasets and the creation of predictive models that improve the accuracy of the actual remote patient telemonitoring programs.

Resumen Introducción: La insuficiencia cardíaca (IC) es un motivo de gran preocupación en la salud pública. Hemos utilizado técnicas de aprendizaje automático para analizar información y mejorar los resultados. Métodos: Estudio observacional, retrospectivo y no aleatorizado, con los pacientes incluidos en el programa de telemonitorización de IC de nuestro centro desde mayo 2014 hasta febrero 2018. Se han analizado datos clínicos, transmisiones de telemonitorización y descompensaciones de IC. Resultados: 240 pacientes incluidos con un seguimiento de 13.44 ± 8.65 meses. En este intervalo se han detectado 527 descompensaciones de IC en 148 pacientes diferentes. Los aumentos significativos de peso, la desaturación inferior al 90% y la percepción de empeoramiento clínico, han resultado buenos predictores de la descompensación de IC. Hemos construido un modelo predictivo aplicando técnicas de aprendizaje automático obteniendo los mejores resultados con la combinación de "Peso + Edemas en EEII + empeoramiento clínico + alertas de tensión arterial sistólica y diastólica, saturación de oxígeno y frecuencia cardiaca". Conclusiones: Las técnicas de inteligencia artificial son herramientas útiles para el análisis de las bases de datos de IC y para la creación de modelos predictivos que mejoran la precisión de los programas de telemonitorización actuales.

Impact of environmental factors on heart failure decompensations.

AIMS: Heart failure (HF) is a clinical syndrome caused by a structural and/or functional cardiac abnormality, resulting in a reduced cardiac output and/or elevated intracardiac pressures at rest or during stress. This disease often causes decompensations, which may lead to hospital admissions, deteriorating patients' quality of life and causing an increment on the healthcare cost. Environmental exposure is an important but underappreciated risk factor contributing to the development and severity of cardiovascular diseases, such as HF. METHODS AND RESULTS: We used two different sets of data (January 2012 to August 2017): one related to the number of hospital admissions and the other one related to the environmental factors (weather and air quality). Admissions related data were grouped in weeks, and then two different studies were performed: (i) a univariate regression to determine whether the admissions may influence future hospitalizations prediction and (ii) a multivariate regression to determine the impact of environmental factors on admission rates. A total number of 8338 hospitalizations of 5343 different patients are available in this dataset, with a mean of 4.02 admissions per day. In European warm period (from June to October), there are significant less admissions than that in the cold period (from December to March), with a clear seasonality of admissions, because there is a similar pattern every year. Air temperature is the most significant environmental factor (r = -0.3794, P < 0.001) related to HF hospital admissions, showing an inversed correlation. Some other attributes, such as precipitation (r = 0.0795, P = 0.05), along with SO2 (precursor of acid rain) (r = 0.2692, P < 0.001) and NOX air (major air pollutant formed by combustion systems and motor vehicles) (r = 0.2196, P < 0.001) quality parameters, are also relevant. Humidity and PM10 parameters do not have significant correlations in this study (r = 0.0469 and r = -0.0485 respectively), neither relevant P-values (P = 0.238 and P = 0.324, respectively). CONCLUSIONS: Several environmental factors, such as weather temperature and precipitation, and major air pollutants, such as SO2 and NOX air, have an impact on the HF-related hospital admissions rate and, hence, on HF decompensations and patient's quality of life.