Acta Neurochir (2002) 144: 811–816 DOI 10.1007/s00701-002-0935-4 Acta Neurochirurgica Printed in Austria Experimental Research FastinVivoWaterQuantificationinRatBrainOedemaBasedonT 1 Measurement at High Magnetic Field A. Schwarcz1,Z.Berente2,3,E.O ¨ sz2, and T.Do ´czi1 1 Department of Neurosurgery, University of Pe ´cs, Hungary 2 Department of Biochemistry and Medical Chemistry, University of Pe ´cs, Hungary 3 PTE-MTA Research Group for Chemical Sensors, Hungarian Academy of Sciences, Hungary Published online August 12, 2002 6 Springer-Verlag 2002 Summary Background. In vivo water content determination based on mag- netic resonance (MR) method is of importance in clinical practice as well as in animal studies to follow up the treatment given in order to reduce brain oedema. The methods proposed in the literature so far are largely time consuming. The aim of this study was to find a fast in vivo water quantification method having real advantage for patients su¤ering from critical conditions. Method. Cold injury was applied to provoke brain oedema in fourteen rats. T 1 values of both the oedematous area and the con- tralateral normal cortex were determined by two independent meth- ods 24 hours after the cold impact. First, from a series of images re- corded by inversion recovery spin echo (IRSE) sequence and then by progressive saturation experiment performed by localised MR spec- troscopy using stimulated echo acquisition mode (STEAM). To re- duce the acquisition time, a two-element repetition time array was optimised for the STEAM experiment, whereas four inversion times were used for T 1 mapping. Both methods were validated against gel phantoms with known T 1 values. After the MR measurements the animals were sacrificed and the water contents of the regions of in- terest were determined by gravimetric wet-dry method. Findings. The reciprocals of the in vivo measured T 1 values were correlated with the reciprocals of the brain water contents. STEAM experiment showed stronger correlation ðr ¼ 0:96Þ than IRSE ðr ¼ 0:93Þ. In addition, STEAM provided more accurate T 1 values in the phantom study. Determination of brain water content based on T1 measurement does work also at high magnetic field. Determi- nation of brain water content by Magnetic Resonance Spectroscopy is feasible within 2 minutes. Interpretation. Using the presented fast method, water content can be determined within a couple of minutes in animal experiments as well as in the daily clinical practice. Keywords: Brain oedema; magnetic resonance; rat; water. Introduction Accurate in vivo estimation of brain water content in di¤erent pathological conditions, such as brain trauma, tumour, stroke, is essential for monitoring the e¤ect of treatment aimed at reducing brain swelling. In the pre-CT/MRI era no direct method was available for the measurement of acute changes in brain water content. Magnetic resonance (MR) techniques have o¤ered a new, direct, non-invasive, quantitative approach for the determination of brain tissue water content [19]. These methods were based either on magnetic resonance imaging (MRI) or on localised magnetic resonance spectroscopy (MRS). MRI techniques provide an overall water image ex- tracted from calibrated longitudinal relaxation time (T 1 ) map [1, 2, 8, 9, 20] or spin density (M 0 ) map [17, 18, 20, 21] of a whole slice. The water quantification based on M 0 map has limitations caused mainly by the B 1 inhomogeneity. Obtaining a water map necessitates keeping the subject in the magnet for extensive periods of time [9], which, in critical conditions, may not be feasible. The application of fast T 1 mapping techniques [4] is often limited in clinical practice due to hardware (relatively long gradient rise times) and software (pulse sequence manipulation and fractional k-space filling