Early syncope detection during head up tilt test by analyzing interactions between cardio-vascular signals

Head up tilt test is a well-known medical procedure used to diagnose vasovagal syncope. Several studies have focused on the early prediction of the outcomes of this test to predict syncope before its occurrence. Therefore, the evaluation of the dynamic interactions between several cardiovascular variables can play an important role to improve our understanding on the function of the baroreflex control of arterial blood pressure and can help to enhance the prediction performance. In this aim, the smoothed pseudo-Wigner-Ville distribution (SPWD) approach was used to characterize the time-frequency relationship between time series extracted from ECG (RR-interval) and blood pressure (amplitudes of the systolic blood pressure, the peak amplitude of the first derivative of the blood pressure, dP/dt and the variability of the pulse transit time, PTT). Two new indexes quantifying the relationship between time series were introduced. The first index computed the distribution of time-frequency coherence between 2 time series in different frequency bands and the second one computed the scalar product of the time-frequency spectra of these time series. This approach was applied to predict the outcome of a head up tilt test (HUTT) based on the first 15 min, to early detect subjects that would faint or not during this test. Our results demonstrate that the strongest parameter to predict HUTT result is the dynamic interaction between RR-interval and the amplitude of systolic blood pressure (AmpS), using the scalar product of the auto time-frequency spectra computed with SPWD (sensitivity: 87.5%; specificity: 93.8%). This study also shows that in high frequency band, the similarity between the RR-interval and the AmpS time series is higher during the first 15 min of HUTT in subjects that would faint. As a clinical viewpoint, these results could bring new perspectives to better understand physiological mechanisms during HUTT and syncope, which are still unclear and could lead also to a reduction of HUTT duration, which is currently between 30 and 45 min.

[1]  G Carrault,et al.  Time-frequency relationships between heart rate and respiration: A diagnosis tool for late onset sepsis in sick premature infants , 2009, 2009 36th Annual Computers in Cardiology Conference (CinC).

[2]  Rolf Vetter,et al.  Prediction of vasovagal syncope from heart rate and blood pressure trend and variability: experience in 1,155 patients. , 2007, Heart rhythm.

[3]  Paolo Rizzon,et al.  Shortened Head-Up Tilting Test Guided by Systolic Pressure Reductions in Neurocardiogenic Syncope , 2002, Circulation.

[4]  Mathieu Feuilloy,et al.  Early prediction of unexplained syncope by support vector machines , 2007, Physiological measurement.

[5]  H. Engels,et al.  Effect of Caffeine on LT, VT and HRVT , 2012, International Journal of Sports Medicine.

[6]  Nello Cristianini,et al.  Kernel Methods for Pattern Analysis , 2003, ICTAI.

[7]  Movahed Mr,et al.  Prediction of non-occurrence of syncope during a tilt-table test by early heart rate variations. , 2001 .

[8]  Gerald Matz,et al.  Time-frequency coherence analysis of nonstationary random processes , 2000, Proceedings of the Tenth IEEE Workshop on Statistical Signal and Array Processing (Cat. No.00TH8496).

[9]  Å. Björck.  Numerics of Gram-Schmidt orthogonalization , 1994 .

[10]  D Lacroix,et al.  Inadequate sympathovagal balance in response to orthostatism in patients with unexplained syncope and a positive head up tilt test , 1999, Heart.

[11]  J. Vesin,et al.  Autonomic Imbalance Assessed by Heart Rate Variability Analysis in Vasovagal Syncope , 1994, Pacing and clinical electrophysiology : PACE.

[12]  H Yamaguchi,et al.  Does an Early Increase in Heart Rate During Tilting Predict the Results of Passive Tilt Testing? , 2000, Pacing and clinical electrophysiology : PACE.

[13]  E Vicaut,et al.  Prediction of head-up tilt test result by analysis of early heart rate variations. , 1997, Circulation.

[14]  G Noll,et al.  Progressive withdrawal of muscle nerve sympathetic activity preceding vaso-vagal syncope during lower-body negative pressure. , 1996, Clinical science.

[15]  Gérard Dreyfus,et al.  Ranking a Random Feature for Variable and Feature Selection , 2003, J. Mach. Learn. Res..

[16]  A H Madrid,et al.  [Usefulness of the RR variability in the diagnosis of neurogenic syncope]. , 1994, Revista espanola de cardiologia.

[17]  Nasser M. Nasrabadi,et al.  Pattern Recognition and Machine Learning , 2006, Technometrics.

[18]  Boualem Boashash,et al.  Cross spectral analysis of nonstationary processes , 1990, IEEE Trans. Inf. Theory.

[19]  M. Frenneaux,et al.  Baroreflex sensitivity in patients with vasovagal syncope. , 1997, Circulation.

[20]  Josef Kittler,et al.  Floating search methods for feature selection with nonmonotonic criterion functions , 1994, Proceedings of the 12th IAPR International Conference on Pattern Recognition, Vol. 3 - Conference C: Signal Processing (Cat. No.94CH3440-5).

[21]  Gérard Dreyfus,et al.  Apprentissage statistique : Réseaux de neurones - Cartes topologiques - Machines à vecteurs supports , 2008 .

[22]  P. Vardas,et al.  Autonomic Nervous System Activity During Tilt Testing in Syncopal Patients, Estimated by Power Spectral Analysis of Heart Rate Variability , 1997, Pacing and clinical electrophysiology : PACE.

[23]  Pablo Laguna,et al.  Characterization of Dynamic Interactions Between Cardiovascular Signals by Time-Frequency Coherence , 2012, IEEE Transactions on Biomedical Engineering.

[24]  Richard Sutton,et al.  Key challenges in the current management of syncope , 2012, Nature Reviews Cardiology.

[25]  Sergio Cerutti,et al.  Time-frequency and time-varying analysis for assessing the dynamic responses of cardiovascular control. , 2002, Critical reviews in biomedical engineering.

[26]  T. Rector,et al.  Comparison of tilt angles and provocative agents (edrophonium and isoproterenol) to improve head-upright tilt-table testing. , 1998, The American journal of cardiology.

[27]  N Lippman,et al.  Failure to decrease parasympathetic tone during upright tilt predicts a positive tilt-table test. , 1995, The American journal of cardiology.

[28]  Richard O. Duda,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[29]  Lotfi Senhadji,et al.  Time-frequency characterization of interdependencies in nonstationary signals: application to epileptic EEG , 2005, IEEE Transactions on Biomedical Engineering.

[30]  Bernhard E. Boser,et al.  A training algorithm for optimal margin classifiers , 1992, COLT '92.

[31]  Larry A. Rendell,et al.  The Feature Selection Problem: Traditional Methods and a New Algorithm , 1992, AAAI.

[32]  Christopher M. Bishop,et al.  Pattern Recognition and Machine Learning (Information Science and Statistics) , 2006 .

[33]  L T Mainardi,et al.  Assessment of the dynamic interactions between heart rate and arterial pressure by the cross time–frequency analysis , 2012, Physiological measurement.

[34]  B Vielle,et al.  Early predictive indexes of head-up tilt table testing outcomes utilizing heart rate and arterial pressure changes. , 2001, The American journal of cardiology.

[35]  Guy Carrault,et al.  Improving ECG Beats Delineation With an Evolutionary Optimization Process , 2010, IEEE Transactions on Biomedical Engineering.

[36]  J. Kim,et al.  Decreased entropy of symbolic heart rate dynamics during daily activity as a predictor of positive head-up tilt test in patients with alleged neurocardiogenic syncope. , 2000, Physics in medicine and biology.

[37]  G Carrault,et al.  Kernel based support vector machine for the early detection of syncope during head-up tilt test , 2014, Physiological measurement.

[38]  José Luis Rojo-Álvarez,et al.  On the feasibility of tilt test outcome early prediction using ECG and pressure parameters , 2011, EURASIP J. Adv. Signal Process..