scholarly journals Decompensation in Critical Care: Early Prediction of Acute Heart Failure Onset

10.2196/19892 ◽  
2020 ◽  
Vol 8 (8) ◽  
pp. e19892
Author(s):  
Patrick Essay ◽  
Baran Balkan ◽  
Vignesh Subbian
Keyword(s):  
Risk Factors ◽  
Heart Failure ◽  
Risk Factor ◽  
Random Forest ◽  
Acute Heart Failure ◽  
Vital Signs ◽  
The United States ◽  
Data Set ◽  
Failure Risk ◽  
Timeline Analysis

Background Heart failure is a leading cause of mortality and morbidity worldwide. Acute heart failure, broadly defined as rapid onset of new or worsening signs and symptoms of heart failure, often requires hospitalization and admission to the intensive care unit (ICU). This acute condition is highly heterogeneous and less well-understood as compared to chronic heart failure. The ICU, through detailed and continuously monitored patient data, provides an opportunity to retrospectively analyze decompensation and heart failure to evaluate physiological states and patient outcomes. Objective The goal of this study is to examine the prevalence of cardiovascular risk factors among those admitted to ICUs and to evaluate combinations of clinical features that are predictive of decompensation events, such as the onset of acute heart failure, using machine learning techniques. To accomplish this objective, we leveraged tele-ICU data from over 200 hospitals across the United States. Methods We evaluated the feasibility of predicting decompensation soon after ICU admission for 26,534 patients admitted without a history of heart failure with specific heart failure risk factors (ie, coronary artery disease, hypertension, and myocardial infarction) and 96,350 patients admitted without risk factors using remotely monitored laboratory, vital signs, and discrete physiological measurements. Multivariate logistic regression and random forest models were applied to predict decompensation and highlight important features from combinations of model inputs from dissimilar data. Results The most prevalent risk factor in our data set was hypertension, although most patients diagnosed with heart failure were admitted to the ICU without a risk factor. The highest heart failure prediction accuracy was 0.951, and the highest area under the receiver operating characteristic curve was 0.9503 with random forest and combined vital signs, laboratory values, and discrete physiological measurements. Random forest feature importance also highlighted combinations of several discrete physiological features and laboratory measures as most indicative of decompensation. Timeline analysis of aggregate vital signs revealed a point of diminishing returns where additional vital signs data did not continue to improve results. Conclusions Heart failure risk factors are common in tele-ICU data, although most patients that are diagnosed with heart failure later in an ICU stay presented without risk factors making a prediction of decompensation critical. Decompensation was predicted with reasonable accuracy using tele-ICU data, and optimal data extraction for time series vital signs data was identified near a 200-minute window size. Overall, results suggest combinations of laboratory measurements and vital signs are viable for early and continuous prediction of patient decompensation.

scholarly journals Decompensation in Critical Care: Early Prediction of Acute Heart Failure Onset (Preprint)

2020 ◽  
Author(s):  
Patrick Essay ◽  
Baran Balkan ◽  
Vignesh Subbian
Keyword(s):  
Risk Factors ◽  
Heart Failure ◽  
Risk Factor ◽  
Random Forest ◽  
Acute Heart Failure ◽  
Vital Signs ◽  
The United States ◽  
Data Set ◽  
Failure Risk ◽  
Timeline Analysis

BACKGROUND Heart failure is a leading cause of mortality and morbidity worldwide. Acute heart failure, broadly defined as rapid onset of new or worsening signs and symptoms of heart failure, often requires hospitalization and admission to the intensive care unit (ICU). This acute condition is highly heterogeneous and less well-understood as compared to chronic heart failure. The ICU, through detailed and continuously monitored patient data, provides an opportunity to retrospectively analyze decompensation and heart failure to evaluate physiological states and patient outcomes. OBJECTIVE The goal of this study is to examine the prevalence of cardiovascular risk factors among those admitted to ICUs and to evaluate combinations of clinical features that are predictive of decompensation events, such as the onset of acute heart failure, using machine learning techniques. To accomplish this objective, we leveraged tele-ICU data from over 200 hospitals across the United States. METHODS We evaluated the feasibility of predicting decompensation soon after ICU admission for 26,534 patients admitted without a history of heart failure with specific heart failure risk factors (ie, coronary artery disease, hypertension, and myocardial infarction) and 96,350 patients admitted without risk factors using remotely monitored laboratory, vital signs, and discrete physiological measurements. Multivariate logistic regression and random forest models were applied to predict decompensation and highlight important features from combinations of model inputs from dissimilar data. RESULTS The most prevalent risk factor in our data set was hypertension, although most patients diagnosed with heart failure were admitted to the ICU without a risk factor. The highest heart failure prediction accuracy was 0.951, and the highest area under the receiver operating characteristic curve was 0.9503 with random forest and combined vital signs, laboratory values, and discrete physiological measurements. Random forest feature importance also highlighted combinations of several discrete physiological features and laboratory measures as most indicative of decompensation. Timeline analysis of aggregate vital signs revealed a point of diminishing returns where additional vital signs data did not continue to improve results. CONCLUSIONS Heart failure risk factors are common in tele-ICU data, although most patients that are diagnosed with heart failure later in an ICU stay presented without risk factors making a prediction of decompensation critical. Decompensation was predicted with reasonable accuracy using tele-ICU data, and optimal data extraction for time series vital signs data was identified near a 200-minute window size. Overall, results suggest combinations of laboratory measurements and vital signs are viable for early and continuous prediction of patient decompensation.

Is hypocapnia a risk factor for non-invasive ventilation failure in acute heart failure?

2020 ◽  
Vol 41 (Supplement_2) ◽  
Author(s):  
L Carrillo-Aleman ◽  
A Montenegto Moure ◽  
L Lopez Gomez ◽  
P.S Bayoumi Delis ◽  
A.A Agamez Luengas ◽  
...  
Keyword(s):  
Risk Factors ◽  
Heart Failure ◽  
Risk Factor ◽  
Respiratory Failure ◽  
Respiratory Rate ◽  
Acute Heart Failure ◽  
Invasive Ventilation ◽  
Coronary Syndrome ◽  
Non Invasive ◽  
Non Invasive Ventilation

Abstract Introduction Non-invasive ventilation (NIV) has become a standard treatment for acute respiratory failure. Multiple factors associated with failure of this ventilatory technique have been described. Some authors postulate that the presence of hypocapnia at the onset of NIV increases the mortality of patients with acute heart failure (AHF). Purpose To analyse whether the presence of hypocapnia is a risk factor for failure of NIV in the patient with AHF. Methods Observational, retrospective study on a prospective database. All patients with AHF admitted to Intensitive Care Unit (ICU) between January 1997 and December 2017 for respiratory failure and requiring NIV are included. The inclusion criteria were the presence of dyspnea, respiratory rate ≥30 and PaO2/FiO2 <250 mmHg. The exclusion criteria were the presence of cardiogenic shock and AHF due to involvement of the right ventricle. Hypocapnia is defined as the presence of PaCO2 <35 mmHg) in basal gasometry prior to NIV, normocapnia as PaCO2 between 35 and 45 mmHg and PaCO2 hypercapnia greater than 45 mmHg. NIV failure is defined as the need for endotracheal intubation or death in ICU. Quantitative variables are expressed as means ± standard deviation, and qualitative variables as percentages. Comparison between variables has been made using the Ji2 linear trend test and single factor ANOVA. Multivariate analysis was performed using logistic regression with the calculation of odds ratios (OR) and their 95% confidence intervals (CI-95%). Results A total of 1009 patients with AHF, 158 (15.7%) normocapnic, 361 (35.8%) hypocapnic and 490 (48.5%) hypercapnic were analyzed. The age in the 3 groups was 73.3±10.4, 73.3±11.2 and 75.6±8.9 years (p=0.001), respectively. In the normocapnic group the respiratory rate was 36±4, PaCO2 40±3 and PaO2/FiO2 125±31. In the hypocapnic group 37±3, 28±3 and 134±30; and in the hypercapnic group 37±6, 65±16 and 126±36, respectively. NIV failure was observed in 15 (9.5%) of normocapnic patients, 56 (15.5%) of hypocapnic patients and 54 (11%) of hypercapnic patients (p=0.070). Independent risk factors for NIV failure were SAPS II (OR=1.07, CI-95%=1.04–1.09), order of non-intubation (OR=2.88, CI-95%=1.45–1.81), baseline SOFA (OR=1.76, CI-95%=1.48–2.08), HACOR index at 1 hour NIV (OR=1,62, CI-95%=1.45–1.08), the presence of acute coronary syndrome (OR=2.18, CI-95%=1.18–4.01), the presence of NIV-related complication (OR=6.42, CI-95%=3.47–11.89) and hypocapnia at the onset of NIV (OR=3.842, CI-95%=2.02–7.27). Conclusions Hypocapnia at the beginning of NIV in the patient with AHF is a frequent finding. Among the risk factors for poor prognosis, the presence of hypocapnia is a strong predictor of NIV failure. Funding Acknowledgement Type of funding source: None

scholarly journals ASSOCIATION OF HEART FAILURE RISK FACTORS WITH LEFT VENTRICLE LONGITUDINAL CHANGE IN HISPANICS: THE ECHOCARDIOGRAPHIC STUDY OF LATINOS (ECHO-SOL)

2021 ◽  
Vol 77 (18) ◽  
pp. 3380
Author(s):  
Nestor Vasquez ◽  
Ayana April-Sanders ◽  
Katrina Swett ◽  
Jorge Kizer ◽  
Bharat Thyagarajan ◽  
...  
Keyword(s):  
Risk Factors ◽  
Heart Failure ◽  
Left Ventricle ◽  
Longitudinal Change ◽  
Failure Risk ◽  
Echocardiographic Study

scholarly journals THU0442 GOUT AND HEART FAILURE IN THE US: A NATIONAL PERSPECTIVE

2020 ◽  
Vol 79 (Suppl 1) ◽  
pp. 458.2-458
Author(s):  
G. Singh ◽  
M. Sehgal ◽  
A. Mithal
Keyword(s):  
Risk Factors ◽  
Heart Failure ◽  
Confidence Intervals ◽  
Burden Of Illness ◽  
Average Length ◽  
Private Insurance ◽  
Significant Risk ◽  
Case Fatality ◽  
The United States ◽  
The Us

Background:Heart failure (HF) is the eighth leading cause of death in the US, with a 38% increase in the number of deaths due to HF from 2011 to 2017 (1). Gout and hyperuricemia have previously been recognized as significant risk factors for heart failure (2), but there is little nationwide data on the clinical and economic consequences of these comorbidities.Objectives:To study heart failure hospitalizations in patients with gout in the United States (US) and estimate their clinical and economic impact.Methods:The Nationwide Inpatient Sample (NIS) is a stratified random sample of all US community hospitals. It is the only US national hospital database with information on all patients, regardless of payer, including persons covered by Medicare, Medicaid, private insurance, and the uninsured. We examined all inpatient hospitalizations in the NIS in 2017, the most recent year of available data, with a primary or secondary diagnosis of gout and heart failure. Over 69,800 ICD 10 diagnoses were collapsed into a smaller number of clinically meaningful categories, consistent with the CDC Clinical Classification Software.Results:There were 35.8 million all-cause hospitalizations in patients in the US in 2017. Of these, 351,735 hospitalizations occurred for acute and/or chronic heart failure in patients with gout. These patients had a mean age of 73.3 years (95% confidence intervals 73.1 – 73.5 years) and were more likely to be male (63.4%). The average length of hospitalization was 6.1 days (95% confidence intervals 6.0 to 6.2 days) with a case fatality rate of 3.5% (95% confidence intervals 3.4% – 3.7%). The average cost of each hospitalization was $63,992 (95% confidence intervals $61,908 - $66,075), with a total annual national cost estimate of $22.8 billion (95% confidence intervals $21.7 billion - $24.0 billion).Conclusion:While gout and hyperuricemia have long been recognized as potential risk factors for heart failure, the aging of the US population is projected to significantly increase the burden of illness and costs of care of these comorbidities (1). This calls for an increased awareness and management of serious co-morbid conditions in patients with gout.References:[1]Sidney, S., Go, A. S., Jaffe, M. G., Solomon, M. D., Ambrosy, A. P., & Rana, J. S. (2019). Association Between Aging of the US Population and Heart Disease Mortality From 2011 to 2017. JAMA Cardiology. doi:10.1001/jamacardio.2019.4187[2]Krishnan E. Gout and the risk for incident heart failure and systolic dysfunction. BMJ Open 2012;2:e000282.doi:10.1136/bmjopen-2011-000282Disclosure of Interests: :Gurkirpal Singh Grant/research support from: Horizon Therapeutics, Maanek Sehgal: None declared, Alka Mithal: None declared

scholarly journals AGE-SPECIFIC TRENDS AND OUTCOMES OF PATIENTS HOSPITALIZED WITH ACUTE HEART FAILURE IN THE UNITED STATES

2021 ◽  
Vol 77 (18) ◽  
pp. 540
Author(s):  
Ayman Elbadawi ◽  
Alexander Dang ◽  
Islam Elgendy ◽  
Ravi Thakker ◽  
Aiham Albaeni ◽  
...  
Keyword(s):  
United States ◽  
Heart Failure ◽  
Acute Heart Failure ◽  
The United States

Characteristics of Hepatitis B virus induced Hepatocellular Carcinoma in Egyptian patients

QJM ◽  
2021 ◽  
Vol 114 (Supplement_1) ◽  
Author(s):  
Eman Mahmoud Fathy Barakat ◽  
Khalid Mahmoud AbdAlaziz ◽  
Mohamed Mahmoud Mahmoud El Tabbakh ◽  
Mohamed Kamal Alden Ali
Keyword(s):  
United States ◽  
Risk Factors ◽  
Hepatocellular Carcinoma ◽  
Risk Factor ◽  
Hepatitis C ◽  
Hepatitis B ◽  
The United States ◽  
Age Difference ◽  
Co Morbidity ◽  
Incidence And Mortality

Abstract Background Hepatocellular carcinoma (HCC) is the most common primary liver malignancy and is a leading cause of cancer-related death worldwide. In the United States, HCC is the ninth leading cause of cancer deaths. Despite advances in prevention techniques, screening, and new technologies in both diagnosis and treatment, incidence and mortality continue to rise. Cirrhosis remains the most important risk factor for the development of HCC regardless of etiology. Hepatitis B and C are independent risk factors for the development of cirrhosis. Alcohol consumption remains an important additional risk factor in the United States as alcohol abuse is five times higher than hepatitis C. Diagnosis is confirmed without pathologic confirmation. Screening includes both radiologic tests, such as ultrasound, computerized tomography, and magnetic resonance imaging, and serological markers such as αfetoprotein at 6-month interval. Aim To compare characteristics and behavior of Hepatocellular carcinoma (HCC) in chronic HCV patients and HVB patients Patients and Methods The current study was conducted on patients with de HCC presented at HCC clinic, Tropical medicine department Ain Shams University Hospitals between December 2017 and D ecember 2018, aged (18-70 years old) . Results eline characteristics of study population shown in Table 1 at enrolment, including gender, Education status, co-morbidity, underlying presence or absence of cirrhosis, Child-Pugh class of patients infected with viral hepatitis, and alpha-fetoprotein levels. Male proportion observed to be predominant in both HCV (62%) and HBV (75.4%) infected HCC population. Overall prevalence of HCV and HBV in patients having HCC was 65.95% and 34.04%, respectively. Presence of underlying liver cirrhosis was more significantly associated with HCV seropositives as compared to HBV seropositive patients (p0.05). Table 2 shows comparison of means between HCV and HBV seropositive patients with HCC. In univariate analysis, mean age difference (11.6 years), and total bilirubin levels (-1.91mg/dl) were the only statistically significant observations noted among HCV-HCC group (p = 0.05) Conclusion Hepatocellular carcinoma is mainly caused by Hepatitis C and Hepatitis B viruses, but latter showed predominance, comparatively worldwide and correlated HBV directly as a cause of HCC rather than HCV whose relation with HCC is still unclear (Shepard et al., 2006; Di Bisceglie, 2009). Because of the geographical differences and risk factors, the epidemiological burden of HCV and HBV has been observed different in different areas of the world. In developing countries due to high burden of HCV infection as compared to HBV such as in Taiwan (HCV 17.0%, HBV 13.8%) (Kao et al., 2011), Guam (HCV 19.6%, HBV 18%) (Haddock et al., 2013), and Pakistan (HCV 4.8%, HBV 2.5%) (Rehman et al., 1996; Raza et al., 2007; Qureshi et al., 2010; Butt et al., 2012;) will possibly

Abstract 1122‐000035: Is Age Critical in Determining Stroke Severity Within the Telestroke Network?

2021 ◽  
Vol 1 (S1) ◽  
Author(s):  
Cassie A Simmons ◽  
Nicolas Poupore ◽  
Fernando Gonzalez ◽  
Thomas I Nathaniel
Keyword(s):  
Risk Factors ◽  
Heart Failure ◽  
Risk Factor ◽  
Univariate Analysis ◽  
Baseline Risk ◽  
Stroke Severity ◽  
Stroke Patients ◽  
Stroke Risk Factor ◽  
Patients Âgés ◽  
Increased Risk

Introduction : Age is the single most important risk factor for stroke and an estimated 75% of all strokes occur in people >65 years of age. In addition, adults >75 years’ experience more hospitalization stays and higher mortality rates with an estimated 50% in the occurrence of all strokes. Several comorbidities have been linked to an increased risk and severity of acute ischemic stroke (AIS). How these factors differentially contribute to the severity of stroke in patients ages >65 and <75 as well as those ≥75 is not known. In this study, we aim to investigate how age, coupled with various clinical risk factors, affects AIS severity within these two age categories. Methods : This retrospective data analysis study was conducted using the data collected from the PRISMA Health Stroke Registry between 2010 and 2016. Baseline clinical and demographic data for patients ages >65 and <75 as well as those ≥75 was analyzed using univariate analysis. Receiver operating characteristic (ROC) curve analysis and multivariate regression models were used to examine the association of specific baseline risk factors or comorbidities associated with worsening or improving neurologic functions. The primary functions were risk factors associated with improving or worsening neurologic outcome in each age category. Results : Adjusted multivariate analysis showed that AIS population of patients >65 and <75 experiencing heart failure (OR = 4.398, 95% CI, 3.912 – 494.613, P = 0.002) and elevated HDL levels (OR = 1.066, 95% CI, 1.009 – 1.126, P = 0.024) trended towards worsening neurologic functions while patients experiencing obesity (OR = 0.177, 95% CI, 0.041 – 0.760, P = 0.020) exhibited improving neurologic functions. For the patients ≥75 years of age, direct admission (OR = 0.270, 95% CI, 0.085 – 0.856, P = 0.026) was associated with improvement of patients treated in the telestroke. Conclusions : Age is a strong risk factor for AIS, and aged stroke patients have higher morbidity and worsening functional recovery than younger patients. In this study, we observed differences in stroke risk factor profiles for >65 and <75 and ≥75 age categories. Heart failure and elevated HDL levels were significantly associated with worsening neurologic functions among AIS for patients aged >65 and <75. Obese patients and individuals ≥75 years who were directly admitted were most likely to exhibit improving neurologic functions. Most importantly, findings from this study reveal specific risk factors that can be managed to improve the care in older stroke patients treated in the telestroke network.

Abstract 5921: Medicare Beneficiaries with Mild to Severe Heart Failure See 15–23 Different Providers Annually

Circulation ◽  
2008 ◽  
Vol 118 (suppl_18) ◽  
Author(s):  
Robert L Page ◽  
Christopher Hogan ◽  
Kara Strongin ◽  
Roger Mills ◽  
JoAnn Lindenfeld
Keyword(s):  
Heart Failure ◽  
Severe Heart Failure ◽  
The United States ◽  
Outpatient Setting ◽  
Fiscal Year ◽  
Hospital Inpatient ◽  
Medicare Beneficiaries ◽  
Improvement Program ◽  
Data Set ◽  
Mild Disease

In fiscal year 2003, Medicare beneficiaries with heart failure (HF) accounted for 37% of all Medicare spending and nearly 50% of all hospital inpatient costs. On average, each beneficiary had 10.3 outpatient and 2 inpatient visits specifically for HF. Despite significant improvements in medical care for HF, mortality and hospital admissions remain high. No data exist regarding the number of providers ordering and providing care for this population. An analysis of fiscal year 2005 Medicare claims was conducted, using a 5% sample standard analytic and denominator file, limited data set version to extrapolate the 34,150,200 Medicare beneficiaries. Three cohorts were defined according to mild, moderate, severe HF employing the Centers for Medicare and Medicaid Services Hierarchical Condition Categories Model and Chronic Care Improvement Program definitions. HMO enrollees, persons without Part A and Part B coverage, and those outside the United States were excluded. We identified physicians by using the unique physician identification number of performing physicians. Based on inclusion criteria, 173,863 beneficiaries were identified. The average number of providers providing care in all sites were 15.9, 18.6, 23.1 for beneficiaries with mild, moderate, and severe HF, respectively; and 10.1, 11.5, and 12.1 in the outpatient setting, respectively. The average number of providers ordering care in all sites consisted of 8.3, 9.6, and 11.2 for beneficiaries with mild, moderate, and severe HF, respectively; and 6.5,7.3, and 7.8 in the outpatient setting, respectively. For beneficiaries with mild disease, only 10% of all office visits were specifically for HF, while those with moderate or severe disease, only 20% were specifically for HF. Medicare beneficiaries with HF, even those with mild disease, have a large number of providers ordering and providing care. These data highlight the importance for developing systems and processes of coordinated care for this population.

Abstract 13354: Event-Free Survival in African American Patients with Heart Failure

Circulation ◽  
2014 ◽  
Vol 130 (suppl_2) ◽  
Author(s):  
Kelly L Adams ◽  
Rebecca L Dekker ◽  
Terry A Lennie ◽  
Misook L Chung ◽  
Kathleen Dracup ◽  
...  
Keyword(s):  
Risk Factors ◽  
Heart Failure ◽  
African American ◽  
Ace Inhibitors ◽  
Nyha Class ◽  
Future Research ◽  
Event Free Survival ◽  
Data Set ◽  
Free Survival ◽  
The Relationship

Introduction: Health outcomes such as event-free survival (cumulative end-point in time to first health event) in heart failure (HF) patients is worse in African American than Caucasians. While the direct impact of traditional risk factors on outcomes are recognized, it is unknown how sociodemographic and psychosocial variables, disease, and treatment factors may alter the relationship between race and event-free survival. Hypothesis: Sociodemographics (age, gender, economic status), psychosocial factors (anxiety, depression), disease factors (smoking, functional status, diabetes) and treatments (beta blockers, ACE inhibitors) moderate the relationship between race and shorter event-free survival among patients with HF. Methods: Data were analyzed from 993 outpatients in a multicenter HF registry who were followed for a median of 1.9 years (37% female, 11.3% African American, 64±13 years, 44% NYHA Class III/IV). Data were collected via chart review and interview. Potential proposed moderators were analyzed with race as the predictor and the outcome event-free survival. Regressions were conducted on event-free survival using race and each proposed moderator, and the product of race and each moderator. Results: A primary analysis showed that African American patients are 1.54 times more likely to experience a cardiac event within this data set (p=.003). Further regression analyses indicate event-free survival in African American patients with HF is not moderated by the proposed moderators (all p>.05). Although an incomplete moderation, interactions with medication and race demonstrated better outcomes in African Americans than Caucasians not on ACE inhibitors, but Caucasians on prescribed ACE inhibitors have better comparative outcomes. Conclusions: Although many modifiable and non-modifiable risk factors may be associated with event free survival in African American HF patients, sociodemographic, psychosocial, disease, and treatment factors do not moderate the relationship between race and event-free survival. Future research is needed to better understand what factors contribute to and moderate evident disparities in the event-free survival of African American patients with HF.

Sign in / Sign up

Export Citation Format

Share Document