Hyperkalemia is common in patients with chronic kidney disease, often asymptomatic, and associated with fatal arrhythmias. Galloway and coauthors developed and validated a deep convolutional neural network for detection of hyperkalemia using 1 576 581 electrocardiograms (ECGs) from 449 380 patients and validated in 61 965 patients with stage 3 or greater chronic kidney disease. Using only 2 ECG leads (leads I and II), a deep-learning model detected hyperkalemia in patients with renal disease with an area under the receiver operating characteristic curve of 0.853 to 0.883. The application of artificial intelligence to the ECG may enable screening for hyperkalemia. Prospective studies of this deep-learning model are warranted.
The mechanisms underlying the association of high blood pressure variability (BPV) with cardiovascular events are not well understood. Clark and coauthors performed a post hoc patient-level analysis of 7 randomized clinical trials involving 3912 patients who underwent serial intravascular ultrasonography over a 24-month period to assess percent atheroma volume progression. Increases in percent atheroma volume were significantly associated with systolic BPV, diastolic BPV, and pulse pressure variability. Systolic BPV was associated with cumulative ischemic events, including death, myocardial infarction, and acute coronary syndrome.
Reducing 30-day readmission rates has been emphasized in the United States but not Canada. Samsky and coauthors examined trends in readmission rates in patients hospitalized for heart failure (HF) in both countries from 2005 through 2015. All-cause 30-day readmission rates declined in both Canada (19.7% to 17.6%) and the United States (21.2% to 18.5%), as did heart failure–specific readmissions (8.4% to 6.9% and 7.6% to 5.7%, respectively). There was no significant difference in either country for all-cause readmission rates after implementation of the US Hospital Readmission Reduction Program. In an Invited Commentary, Jha notes that this well-intentioned policy may not be as helpful to patients as intended and that incentives are needed that truly promote integrated, higher-quality care.
Invited Commentary
Patterns of hypertension risk development and lifetime risks of hypertension under the American Heart Association/American College of Cardiology 2017 thresholds for hypertension (≥130/80 mm Hg) compared with the Seventh Joint National Commission hypertension thresholds are unknown. Chen and coauthors assessed individual-level pooled data in 13 160 participants in 3 contemporary cohorts in the Cardiovascular Lifetime Risk Pooling Project: the Framingham Offspring Study, the Coronary Artery Risk Development in Young Adults study, and Atherosclerosis Risk in Communities study. White men, African American men, white women, and African American women had baseline prevalences of hypertension of 30.7%, 23.1%, 10.2%, and 12.3%, respectively, with lifetime risks of 83.8%, 86.1%, 69.3%, and 85.7%. These were significantly greater than corresponding lifetime risks under the Seventh Joint National Commission threshold for hypertension.