European Study Group on Cardiovascular Oscillations

ESGCO 2026: Characterization of complex cardiovascular and cerebrovascular interactions in postural syncope

Background of the Challenge: 

Maintaining blood pressure and cerebral perfusion during upright posture relies on tightly coordinated regulation among cardiovascular, respiratory, and cerebrovascular systems. In individuals prone to postural syncope, this coordination can break down under orthostatic stress. The head-up tilt (HUT) test provides a controlled framework to study these mechanisms through continuous recordings of heart rate, blood pressure, respiration, and cerebral blood flow. Traditional analyses focus on single variables or pairwise relationships, but these approaches overlook the complex, dynamic interdependence among physiological systems. Recent advances in network physiology enable the characterization of multiorgan interactions as evolving physiological networks, offering new insights into autonomic regulation and instability.

Aim of the Challenge: 

The challenge aims to explore multivariate time series collected during prolonged HUT in healthy controls and patients prone to syncope, to identify physiological and network-based markers of impaired autonomic regulation. Participants are invited to apply methods from time series analysis, causality estimation, and network modelling to uncover the dynamic organization and reconfiguration of physiological interactions preceding syncope. The goal is to promote quantitative, integrative approaches that advance understanding of cardiovascular and autonomic dysfunction under orthostatic stress.

Description of the Database: 

The database contains a group of 13 healthy subjects, and an age- and sex-matched group of 13 patients prone to postural syncope. All participants underwent resting recordings in the supine position to eliminate the effects of physical activity, and two phases of head-up tilt. The data set includes exported values of R to R intervals, systolic and mean blood pressure values, mean cerebral blood flow velocity and respiratory amplitude values. No pharmacological or non-pharmacological interventions were administered during the recordings.

Beat-to-beat time series were extracted from cardiac, vascular, respiratory and cerebral biosignals acquired at the Neurology Division of Sacro Cuore Hospital, Negrar, Italy. The project was approved by the local bioethics committee.

The database is available at:  https://esgco2026.unipa.it/challenge/

NEGRAR_DATABASE.zip containing 78 text files with the extension .csv. Each .csv file is from a different individual in the NEGRAR population (c: control; p: patient) undergoing the three phases of the protocol (01-02-03). The last two to three digits of the file name refer to the subject (01 to 013); for example, c02010 contains the data of the 10th control subject acquired during the 2nd phase of the protocol.

Each file contains five columns with synchronised beat-to-beat values of:

  • RR [s] – duration of RR intervals in seconds
  • SAP [mmHg] – finger systolic arterial pressure in mmHg
  • MAP [mmHg] – finger mean arterial pressure in mmHg
  • CBFV [cm/s] – cerebral blood flow velocity in cm/s
  • RESP [L] – respiratory amplitude in L.
Real Data: 
NEGRAR Database
Real data - attachment: 
Package icon NEGRAR_database.zip
Description of the experimental protocol: 

The database comprises 13 subjects prone to syncope (SYNC, age: 28 ± 9 yrs, 5 males; > 3 unexplained events of syncope in the previous 2 years) and 13 age and gender-matched control individuals (nonSYNC, age: 27 ± 8 yrs; 5 males) enrolled at the Neurology Division of Sacro Cuore Hospital, Negrar, Italy. Signal acquisition was carried out during the resting supine position and prolonged 60° HUT.

Electrocardiogram (ECG), continuous photoplethysmographic arterial pressure (AP) measured using a volume-clamp device from the middle finger of the right hand, cerebral blood flow velocity (CBFV) measured from the middle cerebral artery through a transcranial Doppler device, and respiratory amplitude signal measured through a thoracic impedance belt were acquired synchronously at a sampling rate of 1 kHz. Cardiovascular, respiratory and cerebrovascular beat-to-beat variability time series were extracted during the supine rest, and two phases of HUT, i.e., early and late tilt. Heart period (HP) was computed as the temporal distance between two consecutive R peaks on the ECG. The nth systolic AP (SAP) value was measured as the maximum of the AP signal inside the nth HP. The nth diastolic AP (DAP) was taken as the minimum AP value following the nth SAP. The mean AP (MAP) values were computed by integrating the AP signal between the occurrences of (n-1)th DAP and nth DAP and then by dividing the result by the duration of the nth diastolic interval (i.e., the time distance between the occurrences of (n-1)th DAP and nth DAP). The mean CBFV (MCBFV) values were computed by integrating the CBFV signal between the diastolic values (i.e., the minima of the CBFV close to the occurrences of (n-1)th DAP and nth DAP) and then by dividing the result by the time distance between the two diastolic values. The nth RESP value was computed sampling the respiration signal on the nth R peak of the ECG.

References to quote when using the database

[1] Gelpi, F., Bari, V., Cairo, B., De Maria, B., Tonon, D., Rossato, G., Faes, L. and Porta, A., 2022. Dynamic cerebrovascular autoregulation in patients prone to postural syncope: comparison of techniques assessing the autoregulation index from spontaneous variability series. Autonomic Neuroscience, 237, p.102920. https://doi.org/10.1016/j.autneu.2021.102920

[2] Pernice, R., Sparacino, L., Bari, V., Gelpi, F., Cairo, B., Mijatovic, G., Antonacci, Y., Tonon, D., Rossato, G., Javorka, M. and Porta, A., 2022. Spectral decomposition of cerebrovascular and cardiovascular interactions in patients prone to postural syncope and healthy controls. Autonomic Neuroscience, 242, p.103021. https://doi.org/10.1016/j.autneu.2022.103021