Validity of SeismoFit VO2max Estimation in Patients With Heart Failure or Ischemic Heart Disease

Validity of SeismoFit VO2max Estimation in Patients With Heart Failure or Ischemic Heart Disease

Measurement of cardiorespiratory fitness (VO2max) is considered an important tool in risk prediction of cardiovascular disease and overall patient management. The gold standard method for determining VO2max is a maximal cardiopulmonary exercise test (CPET). This requires time, maximal exercise until voluntary exhaustion and expensive equipment and are therefor not always suitable. A non-exercise VO2max prediction model using seismocardiography (SCG) at rest in combination with demographic data has been proposed as an possible alternative. SCG is a non-invasive three-dimensional measurement technique of precordial vibrations caused by the beating heart and can provide information on cardiac performance. New advances in low-weight three-axis accelerometer, signal processing and feature selection has made this methodology attractive in the recent years. VentriJect Aps has develop a medical device for measuring SCG (SeismoFit) together with an cloud solution for signal processing and prediction of VO2max. The validity of the SeismoFit device has previously been assesses in healthy subjects, but not yet in patients.

The aim of this study is therefore to investigate the validity of the SeismoFit VO2max estimation in patients with heart failure (HF) or ischemic heart disease (IHD).

The study is composed of three parts in order to investigate the validity and reliability of the SeimsoFit VO2max estimation.

1) A cross-sectional part: 80 patients with HF and 60 patients with IHD are recruited for this part. An interim data analysis is planed after 50% of the patients have been through testing, in order to allow for algorithm adjustment, before the data analysis of the last 50% are performed. This part requires one test day.

2) A longitudinal part: 60 patients with HF and 40 patients with IHD are recruited to perform a follow-up test (from part 1) after being medical customized to the treatment doses. This is typically within 3-4 month, depending on the patient.

This part requires one follow-up test day in addition to the test in part 1.

3) A test-retest part: 20 HF and 20 IHD patients are recruited for this part. This part requires one test day within 2-7 days after the last test in part 2.

The testing procedures are identical for the different study parts and includes:

• Measurement of weight and height.
• Estimation of VO2max with the SeismoFit device at supine rest. The SeismoFit device is patented by the company VentriJect Aps and it contains a three-axis accelerometer and a 3-v battery. The SeismoFit is attached to the lower part of the sternum with double adhesive tape. The SCG recording takes 40 seconds and signal process takes approximately one to two minutes depending on the internet connection. The algorithm uses SCG information and weight, height, age and sex.
• A direct measurement of VO2max applying a graded exercise test until voluntary exhaustion with continuous pulmonary gas exchange measurements. The protocol will be customized for the different patients. However, a starting workload of 50 Watt with 25 Watt increments every 2 minutes are the standard protocol used. The patients will be monitored with ECG and blood pressure measurements during the test. Directly following termination of the test, the patients will rate their perceived exertion on the Borg 6-20 scale.

Data from echocardiography measurement and blood samples performed during the standard care treatment at the hospital will be available in this study.

Statistical analysis:

All data will be checked for normal distribution (Shapiro-Wilk test) and equal variance (Brown-Forsythe test). The significant level is p<0.05.

Validity assessment of the SeismoFit VO2max estimation compared with the CPET measured VO2max for HF and IHD patients will be analysed with; A Student's paired t-test or a mixed effect model (part 2), Pearson correlation coefficient r, Standard Error of Estimate (SEE), Mean Absolute Percentage Error (MAPE), Coefficient of Variation (CV%) and Bland-Altman analysis with 95% limits of agreement (BA-plot). Pearson r interpretation is: very high > 0.90, high 0.70-0.90, moderate 0.50-0.70, low 0.30-0.50 and little if any 0.00-0.30. SEE is calculated using: =√(Y-Y')/n-2, with Y representing CPET VO2max and Y' representing SeismoFit VO2max. SEE interpretation is: very good < 4.0 ml/min/kg, good 4.0-6.0 ml/min/kg, fair 6.0-8.0 ml/min/kg, poor 8.0-10.0 ml/min/kg and > 10.0 ml/min/kg very poor. A MAPE ≤ 10% will be used as a criteria level for the SeismoFit VO2max estimation to be considered clinically relevant.

Reliability assessment is analysed with a mixed-effect model, within-subject standard deviation, CV% and Pearson r. The within-subject standard deviation is calculated taking the squared root of the within-subject variance. Statistical analyses will be performed, and figures constructed in R Studio, GraphPad Prism and Microsoft Excel.

Inclusion Criteria:

• Fulfilling the guidelines for rehabilitation sat out by the Danish Society of Cardiology.
• Approved for exercise testing on a cycle ergometer by a specialized doctor in cardiology.

Exclusion Criteria:

• Absolute contraindications with training
• Conditions that prevent maximal exercise testing assess by a specialized doctor in cardiology.