Author(s): ESCAJAL, B., CASTILLO, M., PARODI-FEYE, A., MAGALLANES, C. , Institution: INSTITUTO SUPERIOR DE EDUCACIÓN FÍSICA - UNIVERSIDAD DE LA REPÚBLICA DE URUGUAY., Country: URUGUAY, Abstract-ID: 2729

INTRODUCTION:
Central cardiovascular response plays a key role in the analysis of responses and adaptations to aerobic training. Several methods are available to estimate variables such as cardiac output (CO) and stroke volume (SV), with different levels of validity, applicability, and portability. Among non-invasive techniques, reference methods such as Doppler echocardiography provide high accuracy but present limitations during exercise. In contrast, other devices enable estimation of these variables under exercise conditions through bioelectrical models or physiological estimates derived from gas exchange. In this context, the aim of the present study was to compare CO and SV values during track-based high-intensity interval training (HIIT) in highly trained endurance runners, obtained using impedance cardiography and estimated from a portable spirometer.

METHODS:
Ten trained endurance runners in the competitive period were included (5 men, 5 women; 22–35 years). Participants completed five 1200-m intervals on a track (3:30 min passive recovery) at 95–100% of VO₂max. During exercise, physiological variables were measured using a portable spirometer (MetaMax 3B, Cortex, Germany) and impedance cardiography (PhysioFlow PF-07 Enduro, France). For analysis, the final 60 s of each interval were used, representing a phase of relative physiological stability. CO, SV, and heart rate (HR) were analysed, the latter used as a reference to verify signal consistency between devices. Analyses were conducted at the interval level. Sampling frequency was matched between devices, and agreement was evaluated using Bland–Altman analysis. Data were processed in R.
RESULTS:
Bland–Altman analysis showed good mean agreement for HR (bias: −0.1 bpm; limits of agreement: −20.9 to 20.7 bpm). For CO, a bias of 4.9 L·min⁻¹ was observed (limits: −52.5 to 62.4 L·min⁻¹), indicating higher values with impedance cardiography and wide individual variability. For SV, the bias was −29.0 mL (limits: −117.1 to 59.1 mL), showing greater discrepancy between methods.
CONCLUSION:
Good agreement in HR supports comparability between devices. Although systematic differences were observed in CO and, to a greater extent, in SV, values remained within expected physiological ranges. These findings indicate that the methods are not interchangeable; however, spirometry-derived estimates may represent a reasonable alternative for estimating CO and, to a lesser extent, SV under field conditions, particularly when portable and non-invasive assessment is required.