MR signal response to hyperoxic and hypercapnic respiratory challenges in the brain at 3 T

A. Müller; S. Remmele; I. Wenningmann; M. Soehle; S. Flacke; A. Kovacs; F. Träber; R. König; H. Clusmann; W. A. Willinek; H. Dahnke; J. Gieseke; H. H. Schild; P. Mürtz

doi:10.1055/s-0028-1124042

RöFo - Fortschritte auf dem Gebiet der Röntgenstrahlen und der bildgebenden Verfahren, Table of Contents

MR signal response to hyperoxic and hypercapnic respiratory challenges in the brain at 3 T

A Müller ¹, S Remmele ², I Wenningmann ³, M Soehle ³, S Flacke ¹, A Kovacs ¹, F Träber ¹, R König ¹, H Clusmann ⁴, WA Willinek ¹, H Dahnke ², J Gieseke ¹^,⁵, HH Schild ¹, P Mürtz ¹

¹Department of Radiology, University of Bonn, Germany
²Philips Research Europe, Hamburg, Germany
³Department of Anesthesiology and Intensive Care Medicine, University of Bonn, Germany
⁴Department of Neurosurgery, University of Bonn, Germany
⁵Philips Medical Systems, Best, The Netherlands

Abstract

Purpose: To measure and quantify changes of R2* relaxation rates in the brain during breathing of Carbogen and CO₂/air and to demonstrate clinical application of the proposed method.

Materials and Methods: In nine healthy volunteers and in two patients with a meningioma and a glioblastoma, respectively, cerebral tissue response to respiratory challenges using Carbogen and CO₂/air was monitored with a dynamic R2* weighted multi-gradient-echo sequence with high temporal and spatial resolution. The breathing protocol consisted of 3/4/3 minutes of breathing air/gas/air, respectively. The ΔR2* response was analysed voxel-by-voxel. From all voxels with significant changes a global ΔR2* response function was obtained by averaging all voxel time series.

Results: In volunteers, almost all voxels showed a negative ΔR2* response as physiologically expected, showing the high sensitivity of the method. The maximum change of the ΔR2* response function during Carbogen breathing was almost twice as high as during CO₂/air breathing (-2.4±0.3s^-1 vs. -1.5±0.3s^-1). For both gases, the ΔR2* response to the respiratory challenge was about three times slower (40–50s) than the response to normalization (14–15s). All measured ΔR2* response curves showed the same dynamic behavior.

In both patients, the examination was well tolerated and ΔR2* could be successfully determined, which allowed the assessment of tumor tissue responses.

Conclusion: The determination of ΔR2* in response to elevated levels of O₂ and CO₂ in blood provides a robust and reproducible method to assess oxygenation changes and vasoreactivity non-invasively. Initial examinations of tumor patients demonstrate the feasibility in clinical applications.