Heart failure (HF) is a major clinical and public health challenge because of its high prevalence, mortality and morbidity, readmission rates, and direct and indirect costs. HF hospitalization rates are particularly high among recently discharged HF patients with more than half of the hospitalized HF patients being readmitted within 6 months of discharge. Accurate risk assessment and identifying the HF patients at high risk of hospitalization could help with guiding remote care.

Bioelectrical phase angle has the potential to help identify the HF patients that are at high risk for adverse outcomes such as HF hospitalizations. Phase angle is measured by applying alternating current to a body segment (eg, to a limb) and measuring the resultant voltage difference across the body segment. Cell membranes and tissue interfaces, acting as electrical capacitors, introduce a time delay between the current and the voltage. This time delay, known as the phase angle, is among the parameters that bioelectrical impedance analysis can measure. As cell membrane integrity and cell mass throughout the body are lost (eg, due to chronic disease, frailty, malnutrition, cachexia, or sarcopenia), the electrical capacitance decreases, shortening the time delay and reducing the phase angle.

The relationship between phase angle and cell membrane integrity has led to studies exploring phase angle’s association with chronic disease, mortality, frailty, malnutrition, cachexia, and sarcopenia. ^,,,, In a study with 389 HF participants and 3 years follow-up, it was observed that participants in the lowest quartile by phase angle at baseline (suggesting worse cell membrane integrity) were at increased risk of mortality compared to participants in the highest quartile by phase angle at baseline when adjusted for age, diabetes, and hemoglobin (Hazard ratio: 3.08, 95% CI: 1.06 to 8.99). In another study with 4667 participants aged 60 and older, frailty was defined as a combination of factors such as low BMI, slow walking, weakness, exhaustion and low physical activity, and it was observed that participants in the lowest quintile by phase angle were at increased risk of frailty compared to participants in the highest quintile by phase angle when adjusted for age, race, ethnicity, and comorbidity (For women: odds ratio: 4.4, 95% CI: 2.6 to 7.7. For men: odds ratio: 3.1, 95% CI: 1.2 to 7.9). In a study with 105 male participants with COPD, low phase angle was independently associated with sarcopenia (AUC: 0.78, 95% CI: 0.68 to 0.78) and with malnutrition (AUC: 0.75, 95% CI: 0.63 to 0.86). Frailty, malnutrition, cachexia, and sarcopenia are associated with adverse outcomes among HF patients. ^,,, Yet, these conditions are challenging to diagnose in a clinical setting. For instance, frailty diagnosis is typically based on a combination of anthropometry and physical function testing.

In this analysis, phase angle, measured by the Bodyport Cardiac Scale during the Surveillance and Alert-Based Multiparameter Monitoring to Reduce Worsening Heart Failure Events (SCALE-HF 1) study, ^, was evaluated for its association with HF hospitalization risk. The Cardiac Scale is an FDA-cleared device for measuring body weight, peripheral impedance, phase angle, pulse rate, and center of pressure in patients with fluid management-related health conditions ( Figure 1 ). The Cardiac Scale, with the familiar form factor of a weight scale, provides longitudinal monitoring of fluid volume status, and estimates a daily Congestion Index based on peripheral volume, volume change, and weight change. Worsening HF alerts are generated by the Cardiac Scale if the daily Congestion Index exceeds a preset threshold, which helps clinicians with timely interventions (eg, diuretic adjustments). With the Cardiac Scale, phase angle can also be measured at home in the same step that a patient takes to measure their weight. The approach is noninvasive and does not require an implant. SCALE-HF 1 was a prospective, multicenter study designed to use the Cardiac Scale biomarker data to derive and validate multivariate algorithms for predicting worsening HF events. The details have been previously described. ^,

The Bodyport Cardiac Scale device used during the study. The device captures weight, an electrocardiogram (ECG), an impedance plethysmograph (IPG) and a ballistocardiograph (BCG) signal.

Sarcopenia, cachexia, and malnutrition are highly prevalent among HF patients. ^,, Prior work has associated these conditions with adverse outcomes for HF patients. ^,, Further, it was observed that interventions targeting these conditions can improve outcomes for HF patients. For instance, the PICNIC study (Program of Nutritional Intervention in Chronic Heart Failure Patients) investigated whether nutritional intervention could improve outcomes in malnourished patients hospitalized with acute HF. In the PICNIC study, 120 HF patients were divided into two groups. The intervention group received standard HF treatment plus a personalized nutritional program designed by dietitians. The program included dietary recommendations and nutritional supplements if needed. The control group received only the conventional HF management. Patients in the intervention group experienced a significantly reduced risk of the primary endpoint, which was a composite of readmission for worsening HF or all-cause mortality, compared to the control group. The findings indicated that a nutritional intervention is an effective strategy for improving the prognosis of malnourished patients admitted for acute HF.

Detecting and treating malnutrition, cachexia, or sarcopenia in HF patients is vital for improving clinical outcomes. These interconnected conditions exacerbate the patient’s deteriorating state, but they are often difficult to diagnose in routine clinical practice. The evaluation is complicated by the presence of edema, which can mask underlying muscle and weight loss, leading to a stable body weight or body mass index (BMI) despite the patient’s severe malnutrition. In addition to clinical assessment, diagnosing these conditions requires a combination of specific tests. These include measuring muscle strength using a handgrip dynamometer, assessing body composition with bioelectrical impedance analysis, and evaluating physical performance through functional tests like the timed up-and-go test. Traditional methods, often conducted during infrequent clinic visits, can miss the early, subtle changes indicative of disease progression. Phase angle, as a proxy to loss of cell mass and compromised cell membrane integrity observed in sarcopenia, cachexia, and malnutrition, can help detect these conditions with a Cardiac Scale measurement taken in the same step that the patient takes to measure their weight.

More importantly, for HF patients with conditions such as sarcopenia, cachexia, and malnutrition, daily phase angle measurements can help with monitoring the effectiveness of the nutritional intervention. Continuous phase angle and fluid data from the Cardiac Scale can provide real-time, objective information, allowing clinicians to intervene promptly with targeted nutritional support and exercise therapy. Clinicians can further monitor if the dietary recommendations to treat these conditions are exacerbating fluid retention. This proactive approach may help prevent a decline in functional status, reduce hospital readmissions, and significantly improve quality of life and prognosis for HF patients. Further, remote monitoring programs typically have resource constraints, limiting the number of patients enrolled in the programs. Phase angle can help decide which patients to enroll in (eg, patients with lower phase angles) and which patients to graduate from (eg, patients whose phase angles have improved over time) the remote nutritional intervention and fluid status monitoring program.

Methods

Data collection

Patients with HF, irrespective of left ventricular ejection fraction (LVEF), were enrolled in the clinic or before hospital discharge following an HF hospitalization. The participants were provided a Cardiac Scale at enrollment, and they were expected to take a measurement once daily at home by standing on the scale with bare feet for about 20 seconds. When the participant stepped on the scale, the measurement started automatically and the device acquired the data in approximately 20 seconds, notifying the participant that the measurement was complete through a message on the integrated display and gentle haptic vibration. Data collected by the Cardiac Scale was automatically transmitted via cellular communication to the Bodyport cloud. Participants were shown their body weight and their change in body weight from their previous measurement on the integrated scale display. Site investigators and treating clinicians were blinded to all data collected from the scale unless the participant reported the available weight data. There was no attempt to influence clinical practice with the collected data. The baseline phase angle was defined as the median of the participant’s first week of phase angle measurements. Follow-up visits occurred remotely at 6 weeks, 3 months, and then every 3 months until the end of the study. The Duke Clinical Research Institute (DCRI) was the coordinating center (ClinicalTrials.gov Identifier: NCT04882449). The study was funded by Bodyport Inc. The authors were solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the paper and its final contents.

Population

Eligible participants met the following key inclusion criteria: Age ≥ 18 years with a diagnosis of symptomatic HF, including a worsening HF event (HF admission or unplanned outpatient administration of IV diuretics) in the preceding 12 months. Key exclusion criteria were a weight > 170 kg, chronic inotropic therapy, or chronic kidney disease requiring chronic dialysis. All patients signed informed consent.

Endpoint definition

The endpoint was the participant’s first HF event during the follow-up period. An HF event was defined as a hospital hospitalization with a primary diagnosis of HF in which the participant exhibited:

• (a)

  new or worsening symptoms of HF on presentation (eg, dyspnea, fatigue) or

• (b)

  objective evidence of new or worsening HF (eg, elevated jugular venous pressure, peripheral edema, pulmonary congestion on chest x-ray, elevated natriuretic peptide levels), and

• (c)

  received initiation or intensification of treatment for HF (eg, intravenous diuretics, vasodilators, or inotropes).

The date of an HF event was defined as the date of hospitalization associated with the exacerbation. An independent and central Clinical Events Committee (CEC) at the DCRI performed adjudication of clinical events. Two independent physicians reviewed site-reported events; any events with disagreements were escalated to the full panel for final adjudication.

Statistical analysis

Univariable and multivariable Cox proportional hazards models were utilized to examine the effect of the baseline phase angle on the hazard rate, where the time to event was the time until the participant’s first HF event (ie, first HF hospitalization). The model covariates were selected from baseline phase angle, age, sex, race, BMI, LVEF, inpatient status at study enrollment, coronary artery disease, comorbidities (anemia, atrial fibrillation, COPD, diabetes, hyperlipidemia, hypertension, history of tobacco use, renal disease) and laboratory tests (albumin, BUN, hemoglobin, eGFR, K, Na, NT-proBNP, and serum creatinine). The covariates’ hazard ratios, 95% CIs, and P -values (the Wald test), and the models’ P -values (log-likelihood ratio test) were estimated. The analyses were performed using the CoxPHFitter class within the lifelines package of the Python software. Subgroup analyses with the Cox proportional hazards model were conducted based on age, sex, race, BMI, LVEF, inpatient status at enrollment, and comorbidities. Kaplan–Meier analysis was conducted with the log-rank test to evaluate the difference in HF hospitalization-free rates between the phase angle quartiles. Any missing NT-proBNP measurement was imputed using a BNP to NT-proBNP conversion formula presented in literature.

The participants were partitioned into four groups based on baseline phase angle quartile. For each covariate, it was tested whether the groups were different either by the Mood’s median test (if the covariate was continuous) or by the chi-square test (if the covariate was binary). Eight separate Cox regression models (Models 1 through 8) were constructed where time to event was the time until the participant’s first HF event. Model 1 used univariable analysis with only the phase angle quartiles in the model. Model 2 included the covariates associated with phase angle and HF mortality risk in literature. Models 3 through 8 used the covariates in Table I by varying the imputation strategy and by varying the initial set of covariates. Models 3 and 6 were initialized with all of the covariates in Table I and utilized backward elimination until all remaining covariates had P <.10 (Wald test of the hazard ratio of the covariate). Models 4 and 7 were initialized with covariates with P <.20 (Mood’s median test or chi-square test of the difference between the four groups in Table I ) and utilized backward elimination until all remaining covariates had P <.10 (Wald test of the hazard ratio of the covariate). Models 5 and 8 were initialized with covariates with P <.10 (Mood’s median test or chi-square test of the difference between the four groups in Table I ) and utilized backward elimination until all remaining covariates had P <.10 (Wald test of the hazard ratio of the covariate). Models 3, 4, and 5 utilized complete-case analysis, while models 6, 7, and 8 utilized multivariate iterative imputation. Kaplan–Meier curves were constructed for the HF hospitalization-free rate, and the log-rank test was used to test the rate difference corresponding to the lowest phase angle quartile to the rates of the other quartiles. Subgroup analyses were conducted through univariable Cox proportional hazards models with the covariate being the phase angle and the subgroups defined based on age, sex, race, BMI, LVEF, enrollment pathway, and comorbidities.

Table I

Baseline characteristics grouped by phase angle

	Group 1 1st quartile Median (IQR) or n (%)	Group 2 2nd quartile Median (IQR) or n (%)	Group 3 3rd quartile Median (IQR) or n (%)	Group 4 4th quartile Median (IQR) or n (%)	P value
Age	72 (13)	67 (13)	62 (15)	53 (16)	<.001 *
Sex (female)	43 (55)	51 (65)	44 (56)	39 (51)	.34
Race (black)	18 (23)	23 (29)	22 (28)	33 (42)	.052
BMI	28 (10)	30 (8)	31 (11)	35 (10)	.0098 *
LVEF	40 (30)	45 (25)	35 (30)	30 (35)	.22
Inpatient at enrollment	28 (35)	29 (37)	31 (40)	28 (36)	.95
Coronary artery disease ⁎⁎	49 (62)	49 (62)	41 (52)	27 (35)	.0015 *
Comorbidities
Anemia	45 (57)	39 (50)	36 (46)	32 (42)	.25
Atrial Fibrillation	45 (57)	45 (57)	29 (37)	25 (32)	.0010 *
COPD	22 (28)	17 (21)	21 (26)	8 (10)	.033 *
Diabetes	43 (55)	38 (48)	28 (35)	24 (32)	.012 *
Hyperlipidemia	67 (85)	63 (80)	53 (67)	44 (57)	<.001 *
Hypertension	73 (93)	72 (92)	69 (88)	62 (81)	.076
History of tobacco use	50 (65)	44 (57)	48 (64)	39 (54)	.38
Renal disease	12 (15)	16 (20)	12 (15)	11 (14)	.73
Laboratory tests
Albumin (g/dl)	3.5 (0.78)	3.5 (0.70)	3.5 (0.70)	3.7 (0.70)	.17
BUN (mg/dl)	29 (19)	23 (16)	24 (18)	19 (9.0)	<.001 *
eGFR (mL/min/1.73 m 2)	47 (27)	53 (27)	49 (30)	63 (30)	.033 *
Hemoglobin (g/dl)	12 (2.9)	12.(3.2)	13 (2.4)	13 (2.8)	.018 *
K (mEq/l)	4.1 (0.63)	4.1 (0.80)	4.1 (0.70)	4.1 (0.60)	.98
Na (mEq/l)	139.0 (5.0)	138.0 (3.0)	138.0 (3.0)	138.0 (4.0)	.34
NT-proBNP (pg/mL)	3182 (6857)	1794 (5010)	1217 (4000)	1489 (3296)	.039 *
Serum Creatinine (mg/dl)	1.3 (0.71)	1.3 (0.60)	1.4 (0.73)	1.2 (0.55)	.12

Only gold members can continue reading. Log In or Register to continue

Share this:

• X
• Facebook

Related

Related posts:

1. Design and rationale of aspirin versus aspirin and fondaparinux prior to early invasive strategy in patients with NSTEMI: The FOXY trial
2. Anatomical, physiological and inflammatory characterization of nonculprit vessels in patients undergoing primary PCI for ST-elevation myocardial infarction in the presence of multivessel disease: Rationale and design of the PICNIC study
3. Self-expanding versus balloon-expandable transcatheter heart valves in patients with excessive aortic valve cusp calcification
4. Corrigendum to “Medical treatment of hypercortisolism with relacorilant: Final results of the phase 3 GRACE study” [American Heart Journal 278S (2024) Pages 10-11]

Stay updated, free articles. Join our Telegram channel

Join