Use of a Cardiac Scale to Predict Heart Failure Events: Design of SCALE-HF 1

Figures

• 

  Figure 1.. Pathophysiology of congestion and decompensation. Clinical symptoms such as weight gain are detectable only a few days before a heart failure (HF) event. Fluid buildup is detectable about 7 days prior and changes in hemodynamics occur 3 weeks before an HF event. Adapted from Adamson4 with permission. Copyright ©2023, Springer Nature.

• 

  Figure 2.. Bodyport cardiac scale and data collection for various biomarkers . BCG indicates ballistocardiography; and IPG, impedance plethysmography.

• 

  Figure 3.. An example of a 66-year-old female patient with heart failure with preserved ejection fraction who was discharged from a hospitalization for heart failure within the prior week. A, The patient’s fluid vector (impedance based). There is a gradual increase in the fluid vector exceeding 2 SDs compared with the mean of a healthy control population on day 14 (blue dotted line). A 3-lbs weight gain occurred on day 35 (black dotted line) and HF symptoms on day 41 (yellow dotted line). The patient was hospitalized on day 88 (red dotted line) for an HF event, and oral diuretics were increased in the outpatient setting. B, The corresponding weight trend during the same time period. HF indicates heart failure; and IV, intravenous.

• 

  Figure 4.. The Bodyport cardiac scale shows high correlation with the gold standard direct Fick method. Scatter plots with regression line for cardiac output (left) and stroke volume (right) measured with the scale and direct Fick method. Cardiac output (64 data pairs, r=0.85) and stroke volume (64 data pairs, r=0.81). Preexercise data are denoted with black circles and postexercise data with white circles.

• 

  Figure 5.. Central mean arterial pressure (cMAP) assessments using the cardiac scale compared with a reference device, the SphygmoCor XCEL™ brachial blood pressure cuff. Left, Scatter plot with a linear regression line for the scale-derived and reference device cMAP measurements (n=252 data pairs from 88 participants; r=0.75; mean absolute percent error, 9.6%). Right, Bland-Altman analysis plotting the bias (mean error of cardiac scale minus reference device) of −0.14 mm Hg and 95% limits of agreement (LOA) of −26.1 to 25.8 mm Hg (dotted lines).

• 

  Figure 6.. Assessment of left ventricular ejection time using the cardiac scale. A, The LVET interval (blue) is calculated as the time interval between the B-point from the IPG signal and the X-point from the IPG derivative (ΔIPG). B, The PEP interval (pink) is the time interval between the R-peak from the ECG and the I-peak from the BCG signal. BCG indicates ballistocardiography; and IPG, impedance plethysmography.