2216

Pulmonary circulation time measurements and lung blood volume estimation in mice by magnetic particle imaging and magnetic resonance imaging

Michael Gerhard Kaul¹, Johannes Salamon¹, Caroline Jung¹, Matthias Graeser^2,3, Harald Ittrich¹, Gerhard Adam¹, and Kersten Peldschus¹

¹Department for Interventional and Diagnistic Radiology and Nuclear Medicine, University Medical Center Hamburg, Hamburg, Germany, ²Section for Biomedical Imaging, University Medical Center Hamburg, Hamburg, Germany, ³Institute for Biomedical Imaging, Hamburg University of Technology, Hamburg, Germany

Synopsis

Pulmonary circulation time and/or the blood volume of the lungs can be used to depict pathologic changes. The purpose of our study was to determine the circulation time of a bolus of superparamagnetic iron oxide nanoparticles through the lungs by MPI and to estimate the blood volume of the lungs supplemented by cardiac MRI. The pulmonary circulation times were 907 ± 141 ms. The determination of the blood volume revealed a volume of 233 ± 48 µL.

Purpose

Magnetic particle imaging (MPI) is a new imaging modality scanning magnetic particle distributions in 3D with a high temporal resolution [1] while magnetic resonance imaging (MRI) is providing data with high spatial resolution. The purpose of our study was to determine the circulation time of a bolus of superparamagnetic iron oxide nanoparticles (SPIO) through the lungs by MPI and to estimate the blood volume of the lungs supplemented by cardiac MRI.

Methods and Materials

To prepare MPI reconstruction a robot integrated in the MPI scanner (Philips/Bruker, Germany) was measuring system functions of a point probe consistent of 2 µL of 100 mM SPIO (Perimag, Micromod, Germany) within 36 hours at 28 x 26 x 16 positions covering a volume of 28 x 26 x 16 mm³. All MPI measurements were carried out with a drive field amplitude of 12 mT/µ0 at 25 kHz and a gradient strength in z-axis of 2 T/m/µ0. The Lissajous trajectory of the field free point covered a field of view of 24x24x12 mm³ within 21.5 ms. An own developed receive coil with an inner diameter of 38 mm for mice was used [2]. Nine mice were anaesthetized and imaged by 7 T MRI (ClinScan, Bruker, Germany) using a 8-channel surface coil (Rapid Biomedical, Rimpar, Germany). For cardiac triggering a pulse oximeter was used (SA Instruments, New York, USA). A series of cardiac triggered 2D short axis cine scans were applied resulting in a voxel size of 167 x 167 x 1000 µm³ (see figure 1). Scan parameters were TR 5ms, TE 2 ms, 29 phases, matrix 192x192, FOV 32 mm, thickness 1mm, bandwidth 330 Hz/pixel. The cardiac stroke volume was estimated by measuring the inner volume of the left ventricle in diastole and systole. Image data processing was performed with ImageJ (NIH, USA) extended by in-house developed plugins (qMapIt). Several days later, the mice could recover and were examined again. The anaesthetized mice were positioned on a MR-MPI-compatible couch (Minervé, Esternay, France). Firstly, MRI was performed to acquire anatomic images of the mice that were later used for image fusion with MPI. Then the mice were transferred to the MPI scanner. During of a dynamic measurement a bolus of 1 µL of a 1 mM SPIO (Perimag, Micromod, Germany) following by saline was injected into the tail vein. Images were reconstructed using a reconstruction framework (MPILib, based on Julia Language). Hereby, an iterative Kacmarz algorithm applying a Tikhonov regularization was used. Reconstruction parameters were as follows: regularization factor 0.5, iterations 5, averages 1, SNR threshold 1.8, frequencies 80-125 kHz, bandwidth 125 Mhz, and spectral cleaning. The temporal resolution for a 3D dataset was 21.5 ms with a voxel size of 1 x 1 x 1 mm³. MRI and MPI were co-registered and fused to identify the left and right ventricle in MPI-data (see figure 2). Each pixel time series of the measured MPI-data was fitted to a sigmoid model function extracting the time of the bolus arrival [3]. Next, regions of interest were drawn in the left and right ventricle and the time difference of the arrival times of left and right ventricle was determined (see figure 4). The length of the RR-interval was determined by fourier analyzing the signal modulation in blood of the left ventricle (see figure 3).

The lung blood volume was calculated by:

V_lungBlood = V_{cardiacOutput} * τ_circulation = (V_distole - V_systole) / RR * τ_circulation

All values are given in means with standard deviation.

Results

The analysis of cardiac function revealed a mean inner volume of 33.75 ± 3.27 mm³ and an ejection fraction of 69 ± 6 %. In all nine mice the injection was successful generating short bolus profiles. The pulmonary circulation times were 907 ± 141 ms. The mean RR length by analyzing the signal modulation was 132 ± 11 ms. The RR length by measured in the initial MRI session was 131 ± 8 ms. The determination of the blood volume revealed a volume of 233 ± 48 µL.

Conclusion

Due to its high temporal resolution MPI is able to image the passage of a bolus from the right ventricle through the lungs to the left ventricle. The spatial resolution by MPI was poor in comparison to MRI. Combing results from MPI and MRI, in detail the cardiac output and the pulmonary circulation time, the blood volume of the lungs can be measured. Pulmonary circulation time and/or the blood volume of the lungs can be used to depict pathologic changes.

Acknowledgements

Thanks to the Free and Hanseatic City of Hamburg for funding this work.

References

1. Weizenecker J, Gleich B, Rahmer J, Dahnke H, Borgert J. Three-dimensional real-time in vivo magnetic particle imaging. Phys Med Biol. 2009;54: L1–L10.

2. Graeser M, Knopp T, Szwargulski P, Friedrich T, von Gladiss A, Kaul M, Krishnan KM, Ittrich H, Adam G, Buzug TM. Towards Picogram Detection of Superparamagnetic Iron-Oxide Particles Using a Gradiometric Receive Coil. Sci Rep. 2017;7: 6872.

3. Kaul MG, Salamon J, Knopp T, Ittrich H, Adam G, Weller H, Jung C. Magnetic particle imaging for in vivo blood flow velocity measurements in mice. Phys Med Biol. IOP Publishing; 2018;63: 064001.

Figures

Figure 1. Three cine sequences of the heart of a mice, showing four chamber view, long axis and series of short axis of the left ventricle used for volumetric measurements.

Figure 2. MRI and fused MRI-MPI-data at three distinct time points showing the infow of tracer into the right ventricle, the passage through the lungs and the arrival at the left ventricle.

Figure 3. Single-to-time course shows the signal change caused by the inflow of tracer into the right and left ventricle of a mice. Both processes are temporal separated due to the sharp bolus profile of around 600 µs. The observed signal modulation frequency of 7.2 Hz was caused by the cardiac cycle with a RR-interval length of 139 ms.

Figure 4. Bolus arrival time map fused with MRI. The presents of the bolus at 0.9 seconds in the right ventricle is shown in blue colors, and another 0.9 seconds later in the left ventricle in red. In between the lung tissue was perfused showing green and yellow colors.

Proc. Intl. Soc. Mag. Reson. Med. 27 (2019)

2216