Cerebral Hemodynamics: An Algorithm of its Estimation with Cross-Spectral Analysis of Spontaneous Oscillations of Blood Flow and Blood Pressure

Semenyutin V.*, Aliev V.*, Kozlov A.*, Nikitin P.*, Patzak .**

* Polenov Research Neurosurgical Institute, Saint Petersburg, Russia;
** Johannes-Mueller Institute of Physiology University Hospital Charite, Humboldt - University of Berlin, Germany


Cross-spectral analysis of both slow oscillations of linear blood flow velocity (LBFV) in cerebral vessels and systemic blood pressure (SBP) is a non-invasive method, which allows to estimate the system of cerebral circulation (SCC) in situ. This fact conditions its value for clinical studies. In spite of a rather great number of reports, there is no algorithm of their carrying out.

The research goal was development of an algorithm of assessing SCC with cross-spectral analysis and use of Statistica 6.0.

Material and Methods. We examined 30 healthy volunteers and 30 cases with impaired cerebral circulation. Continuous recording (5 min) of LBFV in the middle cerebral artery and SBP was carried out at rest with the help of Multi Dop X (DWL, Germany) and Finapres-2300 (Ohmeda, USA). The algorithm for estimation of the SCC state consisted in consecutive assessment of B-waves (0.008-0.05 Hz) in LBFV, a phase shift between LBFV and SBP in the range of M-waves (0.05-0.15 Hz). The obtained results were compared with autoregulation speed (RoR), estimated by a cuff test.

Results. As for volunteers, mean values of LBFV, SBP, spectral density and an amplitude of B-waves, a phase shift in the range of M-waves and RoR were as follows: 6.83±2.5 cm/s, 90.2±2.6 mm Hg, 557±70 (cm/s)2/Hz, 3.0±0.2 cm/s, 1.00±0.05 rad and 34.1±5.2%/s respectively. Spectral density of B-waves of less than 1000 (cm/s)2/Hz, estimated on the basis of classification trees, was an index of a normal state of SCC. A phase shift of less than 0.75 rad, which correlated with RoR reduction (less than 17%), was indicative of impaired autoregulation.

Conclusion. The algorithm of carrying out cross-spectral analysis of slow waves of LBFV and SBP with the help of Statistica 6.0 can be used for assessment of SCC in neurosurgical patients.