H. Terheyden et al., Acoustic rhinometry: validation by three-dimensionally reconstructed computer tomographic scans, J APP PHYSL, 89(3), 2000, pp. 1013-1021
The aim of the present study was a validation of acoustic rhinometry (AR) b
y computed tomography (CT). Six healthy subjects were examined by CT and AR
. The CT data were processed in a computer program (AutoCAD), and a virtual
three-dimensional model of each nasal cavity was constructed. This model p
ermitted an individual prediction of the center line of the sound wave prop
agation through the air volume of the nasal cavity with the cross-sectional
areas oriented perpendicularly to this line. The area-distance curves deri
ved from AR and CT were compared. Linear regression analysis revealed a rea
sonable agreement of AR and CT in the anterior nose below a mean of 6 cm di
stance from the nostrils [r = 0.839, P < 0.01, m = 1.123, b = -0.113 (AR =
m x CT + b)]. The measuring accuracy using CT as gold standard revealed a m
ean error at the nasal valve of <0.01 cm(2) (4.52%) and at the nasal isthmu
s of 0.02 cm(2) (1.87%). Beyond 6 cm, the correlation decreased (r = 0.419)
, and overestimation of the true area occurred (>100%). In conclusion, the
measurements were reasonably accurate for diagnostic use up to the turbinat
e head region. Certain factors induce an overestimation of the true areas b
eyond this region. However, these factors are constant and reproducible in
a single subject, and intraindividual comparative measurements are possible
beyond the turbinate head region.