Application of transcranial Doppler sonography to evaluate cerebral hemodynamics in carotid artery disease. Comparative analysis of different hemodynamic variables.

BACKGROUND AND PURPOSE

Transcranial Doppler sonography in combination with manipulation of cerebral resistance vessels is widely used to screen patients with suspected intracranial hemodynamic disturbances. Maximal flow velocity (Vmax), mean flow velocity (Vmean), cerebral pulsatility index (CPi), and cerebral resistance index (CRi) have all been used to describe cerebral hemodynamics. The present study examined CO2 reactivity of the above hemodynamic variables with respect to its variability between different age groups and its capability to discriminate between normal and abnormal findings.

METHODS

Absolute and relative CO2 reactivity of Vmax, Vmean, CRi, and CPi were determined in both hemispheres in 30 young and 37 elderly control subjects and in 245 consecutive patients with strictly unilateral symptomatic (n = 101) or asymptomatic (n = 144) carotid artery disease (> 80% stenosis or occlusion).

RESULTS

Hemispheric reactivities of Vmean, CRi, and CPi were significantly age dependent. Hemispheric Vmax reactivity and interhemispheric differences of individual reactivities (except absolute CPi reactivity) did not vary with age and could therefore be used to define normal values. Patient classification according to these values revealed different frequencies of subjects with pathological findings (3% for hemispheric Vmax reactivity, 5% to 7% for interhemispheric differences of Vmax or Vmean reactivity, 39% and 45% for interhemispheric differences of relative CRi and CPi reactivity, respectively).

CONCLUSIONS

Hemispheric reactivities are less suitable to evaluate cerebral hemodynamics than interhemispheric differences, since most of the latter do not vary with age. However, interhemispheric differences vary with respect to their discriminatory power. Power is low for interhemispheric differences of Vmax and Vmean reactivity, since the corresponding frequencies of abnormal findings do not differ from the 5% frequency expected in the reference population (reference range defined as mean +/- 2 SD). With respect to the discriminatory power, interhemispheric differences of relative CRi and CPi reactivity may be superior to other parameters.