Synopsis

In this study the temporal resolution of ¹⁷O-MRI protocols for quantification of the cerebral metabolic rate of oxygen consumption (CMRO₂) in human brain were optimized using the profile likelihood (PL) method. For this, retrospective analysis of the influence of temporal resolution on the CMRO₂ values and their PL-based confidence intervals (CI) was performed. Dynamic ¹⁷O-MRI was implemented with a 3D golden-angle radial acquisition during and after inhalation of ¹⁷O gas. The results showed that ¹⁷O-MRI data acquired at a temporal resolution 120≤Δt≤165 seconds gave identifiable CMRO₂ values of 0.82‑1.18/0.92‑1.42 µmol/g_tissue/min in WM/GM brain regions.

Introduction

The cerebral metabolic rate of oxygen consumption (CMRO₂) is a useful biomarker of oxygen metabolism, which can be changed in diseased tissues such as brain tumors. CMRO₂ can be measured with nuclear medicine (¹⁵O‑PET)[1], or by direct ¹⁷O-MRI with administration of ¹⁷O-enriched gas [2–6]. Recently, we have applied the method of profile likelihood (PL) analysis to demonstrate that reliable measurements of CMRO₂ are possible with ¹⁷O‑MRI at 3T and proposed a flexible pharmacokinetic model for CMRO₂ quantification [6].

The purpose of this study was to optimize the temporal resolution of ¹⁷O-MRI protocols for CMRO₂ quantification to make it applicable in clinical trials. Therefore, a 3D golden-angle radial acquisition [7] was utilized for optimization of temporal resolution, and PL analysis was performed to find the optimal temporal resolution.

Material and Methods

A dynamic 4D ¹⁷O-MR data set of a male volunteer (50 years) was acquired in ¹⁷O‑MRI experiment [6] at a clinical 3T MR system (Siemens Magnetom TIM Trio) with a custom-built Tx/Rx ¹⁷O quadrature head coil. For imaging, a 3D ultrashort-TE density-adapted radial sequence [9] with a spiral phyllotaxis pattern [7] was used (T_RF=0.8ms; TR=7.57ms; TE=0.52ms; BW=175Hz/pixel; TR=5.7ms; nominal resolution Δx=8mm; total number of unique projections covering the whole k-space was 15844 with 466 interleaves, 34 projections per interleave and 128 sampling points per projection).

The imaging experiment was divided into a baseline phase (11.85min), wherein the natural abundance ¹⁷O‑MR signal was acquired, followed by two phases with a closed rebreathing (RB) system: the ¹⁷O‑phase (4.12min), during which 1.5L of 70%-enriched ¹⁷O gas was delivered, and a subsequent ¹⁶O‑phase phase (5min), during which the volunteer was breathing the gas stored in the RB system (which contains the exhaled ¹⁷O₂ gas that is then reused to increase the ¹⁷O‑MR signal) and 1.35L of fresh ¹⁶O gas was delivered. In the final washout phase (10.06min), the breathing system was opened and the volunteer was breathing room air.

Dynamic series of 3D ¹⁷O‑MR images were reconstructed with Kaiser-Bessel gridding at 15 different temporal resolutions Δt=[30,45,…,240]s. Segmented ¹H MPRAGE image was coregistered to 3D ¹⁷O‑MR images to extract signal dynamics in gray (GM) and white matter (WM) whole brain regions and in a small GM region (V=20 ml). In this small GM region a pharmacokinetic model [6] was modified to account for switching between ¹⁷O and ¹⁶O gas and was used to perform PL analysis. In particular, the identifiability of CMRO₂ and the confidence intervals calculation at various were assessed with the PL method: CMRO₂ is identifiable if PL‑based CIs are finite, but can be practically non-identifiable if CIs are infinite in one dimension [6,8]. The temporal resolution, which provided identifiable CMRO₂ values with the smallest CIs was used to calculate CMRO₂ values in whole GM and WM regions. 67% CIs were used to present calculated CMRO₂ values.

Results and Discussion

In ¹⁷O-MR images averaged over the total measurement time regions with high water content (e.g., eyes and ventricles) can be clearly identified (Fig.1). In the analysis of the dynamic ¹⁷O-MR data, the pharmacokinetic model for CMRO₂ quantification [6] was modified to incorporate both ¹⁶O and ¹⁷O gas administration, which allows to design more efficient ¹⁷O gas handling, which reduces the amount of inhaled rare and costly ¹⁷O gas and improves patient comfort.

Figure 2 shows the dependence of CMRO₂ and its asymmetric PL‑based CIs in a small GM region on the temporal resolution of the reconstructed ¹⁷O-MR images. Reliable CMRO₂ quantification with CIs less than 0.56 µmol/g_tissue/min is possible at 120≤Δt≤165 s. Exploiting the PL of the CMRO₂ for Δt=135s and Δt=195s are shown in Fig. 3. Finite CI for Δt=135s means that CMRO₂ is identifiable, whereas an infinite upper boundary of CI at Δt=195s represents practical non-identifiability caused by high noise and small amount of measurement points. Fitting results to the pharmacokinetic model at a temporal resolution of 135s are shown in Fig. 4. CMRO₂ values of 0.82-1.18/0.92-1.42 µmol/g_tissue/min found in whole brain WM/GM regions are in a good correspondence with the results of ¹⁵O‑PET [1] and ¹⁷O-MRI [2–6] studies.

Conclusions and Outlook

Acknowledgements

References

[1] Leenders KL, Perani D, Lammertsma AA, Heather JD, Buckingham P, Healy MJ, Gibbs JM, Wise RJ, et al. Cerebral blood flow, blood volume and oxygen utilization. Normal values and effect of age. Brain 1990;113:27–47.

[2] Atkinson IC, Thulborn KR. Feasibility of mapping the tissue mass corrected bioscale of cerebral metabolic rate of oxygen consumption using 17-oxygen and 23-sodium MR imaging in a human brain at 9.4 T. Neuroimage 2010;51:723–733.

[3] Hoffmann SH, Begovatz P, Nagel AM, Umathum R, Schommer K, Bachert P, Bock M. A measurement setup for direct ¹⁷O MRI at 7 T. Magn. Reson. Med. 2011;66:1109–1115.

[4] Hoffmann SH, Radbruch A, Bock M, Semmler W, Nagel AM. Direct ¹⁷O MRI with partial volume correction: first experiences in a glioblastoma patient. MAGMA 2014;27:579–87.

[5] Borowiak R, Kurzhunov D, Wagner P, Reisert M, Bock M. Dynamic ¹⁷O-MRI at 3 Tesla for in vivo CMRO₂ Quantification. In: In proceedings of 23rd Annual Meeting of ISMRM; Toronto, Canada. ; 2015. p. 4633.

[6] Kurzhunov D, Borowiak R, Hass H, Wagner P, Krafft AJ, Timmer J, Bock M. Quantification of oxygen metabolic rates in Human brain with dynamic ¹⁷O MRI: Profile likelihood analysis. Magn. Reson. Med. 2016;00:1–11.

[7] Piccini D, Littmann A, Nielles-Vallespin S, Zenge MO. Spiral phyllotaxis: The natural way to construct a 3D radial trajectory in MRI. Magn. Reson. Med. 2011;66:1049–1056.

[8] Nagel AM, Laun FB, Weber M-A, Matthies C, Semmler W, Schad LR. Sodium MRI using a density-adapted 3D radial acquisition technique. Magn. Reson. Med. 2009;62:1565–1573.

[9] Raue A, Kreutz C, Maiwald T, Bachmann J, Schilling M, Klingmüller U, Timmer J. Structural and practical identifiability analysis of partially observed dynamical models by exploiting the profile likelihood. Bioinformatics 2009;25:1923–1929.