Interobserver and intraobserver reliability of venous transcranial color-coded flow velocity measurements

Background and Purpose. Venous transcranial color-coded duplex sonography i s a new technique for noninvasive evaluation of the intracranial venous sys tem. However, the interobserver and intraobserver reliability of this metho d is unclear. Methods. In 23 healthy volunteers (30 +/- 7.3 years of age), the deep middle cerebral vein (dMCV), basal vein (BV), vein of Galen (VG), and straight (SRS), transverse (TS), and superior sagittal (SSS) sinuses in addition to the arterial segments of the circle of Willis were insonated t hrough the temporal bone window on 2 consecutive days by 2 experienced exam iners. The examiners were blinded to each other's results. The interobserve r and intraobserver reliability was calculated using a method described by Bland and Altman, resulting in 2-SD confidence intervals. Results. Non-angl e-corrected and angle-corrected systolic and end diastolic venous flow velo cities (FV) were in good accordance with published normal values, ranging b etween 8.6 and 19.2 cm/s. The interobserver reliabilities for non-angle-cor rected systolic FVs in the dMCV, BV, VG, SRS, and TS were +/-1.8, 2.4, 2.6, 3.3, and 4.6 cm/s; for angle-corrected systolic FVs, the interobserver rel iabilities were +/-2.5, 3.1, 13.9, 11.6, and 7.7 cm/s. The intraobserver re liabilities for non-angle-corrected systolic FVs in the dMCV, BV, VG, SRS, and TS were +/-2.9, 3.2, 2.6, 3.2, and 6.1 cm/s; for angle-corrected systol ic FVs, the intraobserver reliabilities were 3.2, 3.7, 13.9, 11.6, and 7.5 cm/s. Angle correction was not attempted for the SSS. The interobserver and intraobserver reliabilities for systolic FVs in the SSS were 3.3 and 3.3 c m/s, respectively. Conclusions. Intracranial venous FVs can be measured wit h a high interobserver and intraobserver reliability in healthy human subje cts. Intraobserver reliability was higher for cerebral veins than for dural sinuses, predisposing them for follow-up examinations; however, angle corr ection for venous FVs in the VG and the SRS is not advisable.