Validity and Reliability of a Clinical Method for Assessment of Cardiorespiratory Fitness Using Seismocardiography

20 healthy participants were recruited for the study. The study included three identical visit to the laboratory within 14 days.

The participants received both written and oral information about the study and the possible risks associated hereof, before volunteering and signing a written informed consent form.

The participants arrived at the laboratory after an overnight fast from latest 10pm the day before (water excepted). Furthermore, the participants were asked to avoid strenuous exercise 48 hours prior to the test day and eat a habitual diet, which had to be replicated before the second and third visit.

Testing at each visit to the laboratory included:

Completion of a short version of the International Physical Activity Questionaire, about leisure time physical activities.

Measurements of height, weight and waist and hip circumference.

Measurement of body composition using a Dual-energy X-ray absorptiometry (DXA) scan on day 1.

Measurement of body composition using a bioelectrical impedance analysis scale on day 1, 2 and 3.

Measurement of blood pressure in the supine position. 3 measures with 2 minute resting in between.

Measurement of VO2max using a non-exercise testing method based on seismocardiography.

Collection of a 15 ml resting venous blood sample for measurements of resting levels of hormones, metabolites and lipids in the blood plasma.

Measurement of lung function in terms of FEV1 and FVC (Vyntus SPIRO spirometer, Vyaire Medical).

Determination of VO2max applying a graded exercise test on a bicycle ergometry until voluntary exhaustion with continuous pulmonary gas exchange measurements.

Procedure and analysis:

Bioelectrical impedance analysis scale (Tanita MC-780MA, Tanita Corporation of America Inc): The participants was analysed on the scale after voiding and wearing minimal clothing.

DXA scanning (Lunar iDXA, GE Healthcare): The participants was placed in the supine position in the DXA scanner after voiding and while wearing minimal clothing.

Blood samples: The blood samples was collected from the vein cubiti medialis in the forearm. The blood was sampled in iced BD vacutainers (5 mL Aprotinin, 4 mL EDTA and 2 mL Lithium Heparin) and immediately centrifuged at 4000 g for 10 minutes at 4⁰C (Rotina 380R, Hettich). The plasma was collected and stored at -80⁰C for later analyzes of resting levels of metabolites, lipids and hormones.

The concentrations of hemoglobin, hematocrit, glucose and lactate was immediately analyzed though sampling of 2 mL blood in a specialized syringe (safePICO Aspirator, Radiometer) and using an automatic analyzer (ABL800 Flex, Radiometer).

SCG VO2max prediction:

The SCG VO2max prediction was performed as previously described with the same type of equipment. Brief, the SCG recording was performed using an iWorx IX-228/s (IWORX, Dover, New Hampshire) connected to a PC. LabScribe was used as recording software (version 3. Dover, New Hampshire). The accelerometer (Silicon Design 1521-002) had a resolution of ± 2 g, a low noise at 7 μg / √ Hz and a frequency 70 response of 0-300 Hz and was placed on sternum 2 cm proximal of processus xiphoideus. The recording lasted 5 minutes and was performed in the supine position. VentriJect Aps later performed the signal processing and VO2max estimation. VentriJect Aps was informed of the age, height and weight of the participants for inclusion in the prediction model, but was blinded of the measured VO2max.

Hereafter a new devolved wireless device (Seismofit, VentriJect Aps) was used to determine SCG VO2max. The Seismofit contains a 3v battery and a three-axis digital output accelerometer (LIS3DHH, ST Microelectronics). The Seismofit device is connected to a phone application where the participants demographic data (age, height and weight) was entered before performing the recording. For the recording, the participants was lying in the supine position and the Seismofit device was placed on sternum 2 cm proximal of processus Xiphoideus with double adhesive tape. The predicted SCG VO2max value appeared on the phone screen within 5 minutes.

The Seismofit VO2max prediction was performed twice with 5 minutes rest in between. This was done to assess the test-retest reliability of the device.

The iWorx VO2max was done to compare the previous equipment with the new wireless equipment with automatic signal processing.

The gold standard VO2max determination:

A graded exercise test (GET) until voluntary exhaustion on a bicycle ergometer was performed. The protocol consisted of a five-minute warm-up at 80 W for women and 100 W for men and increased hereafter with 20 W for women and 25 W for men every minute until exhaustion.

Participants was wearing a chest strapped heart rate monitor (Polar Vantage XL, Polar) with continuous heart rate measures during the test. Furthermore the participants, was asked about perceived exhaustion on the 6-20 Borg scale directly following exhaustion.

Statistical analysis:

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

A Mixed-effect model, Bland-Altman analysis with 95% limits of agreement, Pearson correlations and Mean Absolute Percentage Error was used for assessment of validity between the Seismofit VO2max and GET VO2max, Seismofit VO2max measure 1 and 2 and between Seismofit VO2max and iWorx VO2max.

Coefficient of Variation (CV), Standard Error of Estimate (SEE) and Intraclass Correlation Coefficient (ICC) was used to assess reliability of VO2max measurements.

The project was approved by the Science Ethical commitee of the greater region of Copenhagen (H-19081375) the 3rd of February 2020. The protocol of the study adhered to the principles of the Helsinki declaration.