Machine Learning-Based Mortality Prediction in Chronic Kidney Disease among Heart Failure Patients: Insights and Outcomes from the Jordanian Heart Failure Registry.

Background and Objectives: Heart failure (HF) is a prevalent and debilitating condition that imposes a significant burden on healthcare systems and adversely affects the quality of life of patients worldwide. Comorbidities such as chronic kidney disease (CKD), arterial hypertension, and diabetes mellitus (DM) are common among HF patients, as they share similar risk factors. This study aimed to identify the prognostic significance of multiple factors and their correlation with disease prognosis and outcomes in a Jordanian cohort. Materials and Methods: Data from the Jordanian Heart Failure Registry (JoHFR) were analyzed, encompassing medical records from acute and chronic HF patients attending public and private cardiology clinics and hospitals across Jordan. An online form was utilized for data collection, focusing on three kidney function tests, estimated glomerular filtration rate (eGFR), blood urea nitrogen (BUN), and creatinine levels, with the eGFR calculated using the Cockcroft-Gault formula. We also built six machine learning models to predict mortality in our cohort. Results: From the JoHFR, 2151 HF patients were included, with 644, 1799, and 1927 records analyzed for eGFR, BUN, and creatinine levels, respectively. Age negatively impacted all measures (p ≤ 0.001), while smokers surprisingly showed better results than non-smokers (p ≤ 0.001). Males had more normal eGFR levels compared to females (p = 0.002). Comorbidities such as hypertension, diabetes, arrhythmias, and implanted devices were inversely related to eGFR (all with p-values <0.05). Higher BUN levels were associated with chronic HF, dyslipidemia, and ASCVD (p ≤ 0.001). Higher creatinine levels were linked to hypertension, diabetes, dyslipidemia, arrhythmias, and previous HF history (all with p-values <0.05). Low eGFR levels were associated with increased mechanical ventilation needs (p = 0.049) and mortality (p ≤ 0.001), while BUN levels did not significantly affect these outcomes. Machine learning analysis employing the Random Forest Classifier revealed that length of hospital stay and creatinine >115 were the most significant predictors of mortality. The classifier achieved an accuracy of 90.02% with an AUC of 80.51%, indicating its efficacy in predictive modeling. Conclusions: This study reveals the intricate relationship among kidney function tests, comorbidities, and clinical outcomes in HF patients in Jordan, highlighting the importance of kidney function as a predictive tool. Integrating machine learning models into clinical practice may enhance the predictive accuracy of patient outcomes, thereby supporting a more personalized approach to managing HF and related kidney dysfunction. Further research is necessary to validate these findings and to develop innovative treatment strategies for the CKD population within the HF cohort.

Improved recovery of cardiac auscultation sounds using modified cosine transform and LSTM-based masking.

Obtaining accurate cardiac auscultation signals, including basic heart sounds (S1 and S2) and subtle signs of disease, is crucial for improving cardiac diagnoses and making the most of telehealth. This research paper introduces an innovative approach that utilizes a modified cosine transform (MCT) and a masking strategy based on long short-term memory (LSTM) to effectively distinguish heart sounds and murmurs from background noise and interfering sounds. The MCT is used to capture the repeated pattern of the heart sounds, while the LSTMs are trained to construct masking based on the repeated MCT spectrum. The proposed strategy's performance in maintaining the clinical relevance of heart sounds continues to demonstrate effectiveness, even in environments marked by increased noise and complex disruptions. The present work highlights the clinical significance and reliability of the suggested methodology through in-depth signal visualization and rigorous statistical performance evaluations. In comparative assessments, the proposed approach has demonstrated superior performance compared to recent algorithms, such as LU-Net and PC-DAE. Furthermore, the system's adaptability to various datasets enhances its reliability and practicality. The suggested method is a potential way to improve the accuracy of cardiovascular diagnostics in an era of rapid advancement in medical signal processing. The proposed approach showed an enhancement in the average signal-to-noise ratio (SNR) by 9.6 dB at an input SNR of - 6 dB and by 3.3 dB at an input SNR of 10 dB. The average signal distortion ratio (SDR) achieved across a variety of input SNR values was 8.56 dB.

Validity of TIMI Risk Score and HEART Score for Risk Assessment of Patients with Unstable Angina/Non-ST Elevation Myocardial Infarction Presented to an Emergency Department in Jordan.

Purpose: To examine the validity and predictability of thrombolysis in myocardial infarction (TIMI) risk and HEART scores in patients presenting to the emergency department (ED) with chest pain in Jordan (representative of the Middle East and North Africa Region, MENA). Patients and Methods: Risk scores were calculated for 237 patients presenting to the ED with chest pain. Patients were followed-up prospectively for the need for percutaneous coronary intervention, major adverse cardiovascular events, and all-cause mortality, looking for correlation and accuracy between the predicted cardiovascular risk from TIMI risk score and HEART score and the clinical outcome. Results: Of the 237 patients, approximately 77% were diagnosed with unstable angina and 23% diagnosed with non-ST elevation myocardial infarction (NSTEMI). about two thirds of the study population were smokers and known to have hypertension and dyslipidaemia. In 50 patients, the primary outcome (need for percutaneous coronary intervention (PCI) and/or major adverse cardiovascular events (MACE) at days 14 and 40, all-cause mortality) was observed. Regarding the predictability of the TIMI score, a larger number of events were observed in the study population than predicted. Patients with TIMI scores of 3 to 5 have about a 5-8% higher event rate than predicted. Conclusion: Both TIMI and HEART risk scores were able to predict an elevated risk of major cardiovascular adverse events (MACE). The overall impression was that the TIMI risk score tended to underestimate risk in the study population.

Clinical Profiles and One-Year Outcome in Middle Eastern Patients With Atrial Fibrillation and Hypertension: Analysis From the Jordan Atrial Fibrillation Study.

Studies on the impact of hypertension (HTN) on the outcome of patients with atrial fibrillation (AF) in the Middle East are scarce. The aim of this contemporary multicenter study is to evaluate the effect of the coexisting HTN on the baseline clinical profiles and 1-year prognosis in a cohort of Middle Eastern patients with AF. Consecutive AF patients in 29 hospitals and cardiology clinics were enrolled in the Jordan AF study (May 2019-December 2020). Patients were prospectively followed up for 1 year, and the study had no influence on their treatment, which was at the discretion of the treating physician. We compared clinical features, use of medications, and 1-year prognosis in patients with AF/HTN compared with AF/no HTN. Among 1849 non-valvular AF patients, 76.4% had HTN, with higher prevalence of diabetes, dyslipidemia, coronary heart disease, stroke, and left ventricular hypertrophy in HTN patients. There was a higher thromboembolic and bleeding risk among HTN patients. At 1 year, HTN patients had significantly higher rates of stroke and systemic embolism (SSE) (4.5%), acute coronary syndrome (ACS) (2.4%), rehospitalization (27.9%), and major bleeding events (3.0%) compared with non-HTN patients. In this cohort, the coexistence of HTN was associated with worse baseline clinical profile and 1-year outcomes.

Mechanical Aortic Valve Thrombosis with Heart Failure Successfully Treated with Oral Anticoagulation: A Case Report.

Background: The use of anticoagulation is mandatory for prevention of prosthetic valve thrombosis (PVT) worldwide, regardless of the valve type or position in the heart. In case a thrombosis causes symptomatic dysfunction, treatment usually includes the use of thrombolytic therapy or surgery. We report a case of PVT involving a patient with a mechanical aortic valve which was treated entirely with the use of anticoagulation therapy (warfarin). Case Presentation: A 58-year-old man had an aortic valve replacement using a Carbomedics® mechanical valve due to severe aortic stenosis as a result of a calcific bicuspid native aortic valve. He was commenced on warfarin after surgery which was continued thereafter. He presented to our hospital after three years with shortness of breath at rest. On clinical examination, his condition was poor with a New York Heart Association functional classification of IV. He was in sinus rhythm and had an enlarged heart shadow on chest X-ray. Transesophageal echocardiography (TEE) revealed aortic valve regurgitation with vegetations on the anterior valve leaflet causing reduced hemi leaflet motility and a mean pressure gradient of 50 mmHg. Cinefluoroscopy revealed a dysfunctional mechanical valve leaflet. Surgery was at high risk of mortality due to the patient's clinical status and he was continued on warfarin therapy with close monitoring. Cinefluoroscopy and echocardiography done six months later revealed complete dissolution of thrombus and a normally functioning mechanical aortic valve. Conclusion: Only a few cases of symptomatic, thrombotic mechanical aortic valve were entirely treated with anticoagulation only. Our patient is one such case who had resolution of symptoms and improvement on NYHA functional classification (IV to I).

The Impact of COVID-19 National Lockdown on Myocardial Infarction (MI) Hospitalizations in Northern Jordan.

Background: During COVID-19 lockdown periods, several studies reported decreased numbers of myocardial infarction (MI) admissions. The lockdown impact has not yet been determined in developing countries. The aim of this study was to investigate the impact that of the lockdown measures might have had on the mean number of MI hospital admissions in Northern Jordan. Methodology: A single-center study examined consecutive admissions of MI patients during COVID-19 outbreak. Participants' data was abstracted from the medical records of King Abdullah University Hospital between 2018 and 2020. Mean and percentages of monthly admissions were compared by year and by lockdown status (pre-lockdown, lockdown, and post-lockdown time intervals). Results: A total of 1380 participants were admitted with acute MI symptoms: 59.2% of which were STEMI. A decrease in number of MI admissions was observed in 2020, from 43.1 (SD: 8.017) cases per month in 2019 to 40.59 (SD: 10.763) in 2020 (P < 0.0001) while an increase in the numbers during the lockdown was observed. The mean number during the pre-lockdown period was 40.51 (SD: 8.883), the lockdown period was 44.74 (SD: 5.689) and the post-lockdown was 34.66 (SD: 6.026) (P < 0.0001 for all comparisons). Similar patterns were observed when percentages of admissions were used. Conclusion: Upon comparing the lockdown period both to the pre- and post-lockdown periods separately, we found a significant increase in MI admissions during the lockdown period. This suggests that lockdown-related stress may have increased the risk of myocardial infarction.

Myocardial injury and its correlation to mortality in hospitalized COVID-19 patients: A retrospective cohort study.

Purpose: This retrospective observational study was conducted to assess the clinical characteristics and outcomes of hospitalized COVID-19 patients with positive cardiac enzymes in the King Abdullah University Hospital (KAUH) in Irbid, Jordan. Patients and methods: A total of 676 patients admitted to KAUH for moderate-to-severe COVID-19 were included in the study. Clinical and mortality data were collected from patients' electronic medical records. Results: A significant association was found between myocardial injury and In-hospital mortality. Seven comorbidities were identified as risk factors for myocardial injury: Hypertension, diabetes mellitus (DM), previous cerebrovascular accident (CVA), ischemic heart disease (IHD), heart failure, chronic kidney disease (CKD), and cardiac arrhythmias. The need for intensive care unit (ICU) for invasive ventilation was also associated with myocardial injury. Acute kidney injury (AKI) during hospitalization had a significantly higher incidence of myocardial injury and mortality. Acute myocardial infarction (MI) and acute peripheral vascular disease (PVD) were also associated with higher mortality. Conclusion: Myocardial injury is an important predictor of mortality in patients with moderate-to-severe COVID-19 disease. Patients with a history of hypertension, diabetes mellitus, any vascular diseases, cardiac arrhythmias or heart failure are considered high-risk for adverse outcome. Additionally, COVID-19 patients with myocardial injury and acute kidney injury were recognized with the highest mortality rate.

Clinical and electrocardiogram presentations of patients with high serum potassium concentrations within emergency settings: a prospective study.

BACKGROUND: Elevated potassium level is a common and reversible peri-arrest condition. Diagnosis and management of hyperkalemia in a short time is critical, where electrocardiogram (ECG) alterations might be helpful. We aimed to investigate the role of clinical features and ECGs in early diagnosing and treating hyperkalemia. METHODS: Prospectively, adult patients who presented to the emergency department (ED) from July 2019 to March 2020 with hyperkalemia (serum potassium ≥5.5mmol/L) were included. History was obtained, and laboratory investigations and ECGs were performed at the presentation and before initiating hyperkalemia therapy. Hyperkalemia severity was divided into mild (5.5-5.9mmol/L), moderate (6.0-6.4mmol/L), and severe (≥6.5mmol/L). A cardiologist and emergency physician blinded to laboratory values, study design, and patients' diagnoses interpreted ECGs and presenting symptoms independently to predict hyperkalemia. RESULTS: Sixty-seven hyperkalemic patients with a mean (±SD) serum potassium level of 6.5±0.7mmol/L were included in this study. The mean age was 63.9±15.1, and 58.2% were females. Hyperkalemia was mild in 10.4%, moderate in 40.3%, and severe in 49.3%. Almost two thirds of patients (71.6%) had hypertension, 67.2% diabetes, and 64.2% chronic kidney disease. About one-quarter of patients (22.4%) were asymptomatic, while fatigue (46.3%), dyspnea (28.4%), and nausea/vomiting (20.9%) were the most common presenting symptoms. Normal ECGs were observed in 25.4% of patients, while alterations in 74.6%. Atrial fibrillation (13.4%), peaked T wave (11.9%), widened QRS (11.9%), prolonged PR interval (10.5%), and flattening P wave (10.5%) were the most common. Peaked T wave was significantly more common in severe hyperkalemia (87.5%) than in mild and moderate hyperkalemia (12.5%, 0.0%, respectively) (p=0.041). The physicians' sensitivities for predicting hyperkalemia were 35.8% and 28.4%, improved to 51.5% and 42.4%, respectively, when limiting the analyses to severe hyperkalemia. The mean (±SD) time to initial hyperkalemia treatment was 63.8±31.5 min. Potassium levels were positively correlated with PR interval (r=0.283, p=0.038), QRS duration (r=0.361, p=0.003), peaked T wave (r=0.242, p=0.041), and serum levels of creatinine (r=0.347, p=0.004), BUN (r=0.312, p=0.008), and CK (r=0.373, p=0.039). CONCLUSIONS: The physicians' abilities to predict hyperkalemia based on ECG and symptoms were poor. ECG could not be solely relied on, and serum potassium tests should be conducted for accurate diagnosis.

Statin eligibility based on the ACC/AHA guidelines among Middle Eastern patients with diabetes mellitus presenting with acute myocardial infarction.

BACKGROUND: Statin eligibility based on the American College of Cardiology/American Heart Association cholesterol guidelines among patients with diabetes admitted with first time acute myocardial infarction has not been evaluated in the Middle East. PURPOSE: To assess statin eligibility for diabetic patients admitted with first time myocardial infarction in Jordan according to ACC/AHA guidelines. METHODS: Consecutive patients admitted with a first acute myocardial infarction who were not taking statins, and had their serum lipoproteins measured upon hospital admission were enrolled in the study. Statin eligibility among patients with diabetes admitted with first time myocardial infarction was determined based on the ACC/AHA guidelines. RESULTS: Of 774 patients enrolled, 292 (37.30%) had diabetes. Compared with non-diabetic patients, those with diabetes were females, older, more hypertension, more hypercholesterolemia, more triglycerides, more diastolic blood pressure, less smokers and less low density lipoprotein. Among patients with diabetes, 242 diabetic patients (82.9%) were statin eligible, including 20 (6.90%) for having high serum levels of low density lipoprotein cholesterol (LDL-C) >190 mg/dL, and 222 (76%) for being aged 40-75 years with LDL-C 70-189 mg/dL. No patient had a calculated atherosclerotic cardiovascular risk score ≥7.5%. On the other hand, 393 non-diabetic patients (81.3%) were statin eligible, including 41 (8.50%) for having high serum levels of low density lipoprotein cholesterol (LDL-C) >190 mg/dL, and 351 (72.80%) for being aged 40-75 years with LDL-C 70-189 mg/dL. CONCLUSIONS: Based on the ACC/AHA guidelines, the majority of patients with diabetes admitted with first acute myocardial infarction would have been eligible for statin treatment if they have LDL-c >190 mg/dl or aged 40-75 years old and they have their LDL 70-189 mg/gl. More efforts should be taken for patients who are female, older than 50 years, hypertensive, elevated diastolic blood pressure have hypercholesterolemia, and elevated triglycerides because of their significant association with diabetes.