Author(s): HANSEN, M.T., HUSTED, K., MODVIG, J.L., LANGE, K.K., WEINREICH, C.M., TRANBERG, C., RØMER, T., DELA, F., HELGE, J.W., Institution: UNIVERSITY OF COPENHAGEN, Country: DENMARK, Abstract-ID: 1300

INTRODUCTION:
Directly measured pulmonary gas exchange rates during graded exercise testing (CPET) until exhaustion using online equipment is the gold standard method for the determination of peak oxygen consumption (VO2peak). This methodology requires time, maximal exercise, and expensive equipment and is therefore not always possible. A non-exercise VO2peak estimation model using seismocardiography (SCG) at rest in combination with known demographic determinants of VO2peak has been developed and validated in healthy subjects. SCG adds an objective measure of cardiac function to the model through the measurement of precordial vibrations using an accelerometer. The aim was to investigate the validity and applicability of the SCG VO2peak method in overweight and obese subjects undergoing a 14-week lifestyle intervention. It was hypothesised that the SCG method would be accurate and able to detect changes in VO2peak.
METHODS:
The study was carried out at a privately run grown-up school (Ubberup Hoejskole, UBH) that offers 14-week courses on lifestyle changes where students live at the school and voluntarily participate in daily lectures and activities. Men and women with age and BMI between 18-70 years and 25-50 kg/m2 were tested at baseline (n=68) and after 14 weeks (n=53) at UBH. Testing included the determination of demographic variables, an SCG VO2peak estimation performed at supine rest with the Seismofit® recording device placed on the lower part of the sternum, and a CPET VO2peak on a cycle ergometer until voluntary exhaustion. Parametric statistics and an alfa of 0.05 was applied.
RESULTS:
Agreement in VO2peak at baseline (n=68, SCG: 26.8±1.9 ml/min/kg, CPET: 26.4±1.7 ml/min/kg, mean±95% confidence interval) showed a bias of 0.5±1.2 ml/min/kg with 95% limits of agreement (LoA) ranging ±10.1 ml/min/kg (Bland-Altman plot). Furthermore, a Pearson’s correlation of r=0.77 (p<0.0001) and a standard error of estimate (SEE) of 5.2 ml/min/kg was found between methods. At follow-up (n=53) weight was reduced by 6.5±1.0 kg (Student parried t-test, p<0.0001). CPET VO2peak was increased by 3.2±0.68 ml/min/kg and 168±58 ml/min and SCG VO2peak by 2.5±0.58 ml/min/kg and 87±56 ml/min (two-way ANOVA repeated measure: intervention p<0.0001, method p=0.939 and interaction p=0.125, relative VO2peak). A Pearson’s correlation of r=0.40 (p<0.05) was found between the delta change in relative VO2peak but not for absolute VO2peak r=0.11 (p=0.446).
CONCLUSION:
The SCG method is accurate for the estimation of VO2peak in an overweight and obese population and the method is appropriate for detecting group changes in both relative and absolute VO2peak following a 14-week lifestyle intervention. Furthermore, the method can detect individual changes in VO2peak but not independently of weight changes. The latter is a strong finding as prediction models often cannot detect changes over time, however, the applicability is still limited by the relatively large variation in LoA and SEE which is inherent to prediction models.