Effects of endotoxin lung injury on NMR T2 relaxation

Determining the effect of water temperature on the T1 and T2 relaxation times of the lung tissue at 9.4 T MRI: A drowning mouse model

Japanese individuals have a unique culture of soaking in a bathtub, and forensic pathologists have experienced fatal cases due to drowning. However, T1 and T2 relaxation times of a drowning lung are poorly documented. In the present study, we investigated the relationship between drowning water temperature and T1 and T2 relaxation times of drowning lung tissues at 9.4 T MRI (Bruker, BioSpec94/20USR). The mice used as animal drowning models were directly submerged in freshwater. Water temperature was set to 8 °C-10 °C (cold), 20 °C-22 °C (normal), 30 °C, and 45 °C. The regions of interest (ROIs) on the axial section of the third slice were set at the central and peripheral areas of each-the left and the right-lung. T1 relaxation times measured immediately after death differed by the presence or absence of soaking water, except in case of cold water temperature. In the drowning groups, T1 relaxation time showed a linear dependency on water temperature. By contrast, T2 relaxation time was almost constant regardless of the presence of drowning under the same temperature condition; when compared in the lung areas of the same individuals, the times were uniformly reduced in drowning models. To minimize the effects of hypostasis and decomposition, we performed measurements immediately after death and were able to determine the noticeable difference in drowning water temperature. These results may be useful for qualitative assessments of a drowning lung and may serve as a basis when imaging the human body during forensic autopsy cases.

In Vivo Measurements of T2 Relaxation Time of Mouse Lungs during Inspiration and Expiration

Purpose

The interest in measurements of magnetic resonance imaging relaxation times, T1, T2, T2*, with intention to characterize healthy and diseased lungs has increased recently. Animal studies play an important role in this context providing models for understanding and linking the measured relaxation time changes to the underlying physiology or disease. The aim of this work was to study how the measured transversal relaxation time (T2) in healthy lungs is affected by normal respiration in mouse.

Method

T2 of lung was measured in anaesthetized freely breathing mice. Image acquisition was performed on a 4.7 T, Bruker BioSpec with a multi spin-echo sequence (Car-Purcell-Meiboom-Gill) in both end-expiration and end-inspiration. The echo trains consisted of ten echoes of inter echo time 3.5 ms or 4.0 ms. The proton density, T2 and noise floor were fitted to the measured signals of the lung parenchyma with a Levenberg-Marquardt least-squares three-parameter fit.

Results

T2 in the lungs was longer (p<0.01) at end-expiration (9.7±0.7 ms) than at end-inspiration (9.0±0.8 ms) measured with inter-echo time 3.5 ms. The corresponding relative proton density (lung/muscle tissue) was higher (p<0.001) during end-expiration, (0.61±0.06) than during end-inspiration (0.48±0.05). The ratio of relative proton density at end-inspiration to that at end-expiration was 0.78±0.09. Similar results were found for inter-echo time 4.0 ms and there was no significant difference between the T2 values or proton densities acquired with different interecho times. The T2 value increased linearly (p< 0.001) with proton density.

Conclusion

The measured T2 in-vivo is affected by diffusion across internal magnetic susceptibility gradients. In the lungs these gradients are modulated by respiration, as verified by calculations. In conclusion the measured T2 was found to be dependent on the size of the alveoli.

A continuous-flow perfusion system for the maintenance and NMR study of small tissue samples in vitro

To describe and evaluate a novel perfusion system developed to maintain excised tissue in a flowing, oxygenated bathing solution during acquisition of nuclear magnetic resonance (NMR) data, and in addition allow precise data to be acquired continuously while altering the composition of the bathing solution surrounding the tissue. A chamber to house the tissue sample was constructed of interlocking sections of polyethylene tubing, and had approximate internal dimensions of 4 mm in diameter and 4 mm in height. Temperature-controlled, physiologically appropriate buffer solution was pumped via an infusion pump through the chamber, entering and exiting by way of small openings on either end. Immediately surrounding the polyethylene chamber was a tight-fitting four-loop solenoid RF coil. Measured proton NMR parameters were found to be fairly insensitive to the flow rate of the buffer if this coil was used only for reception and a larger-volume transmit-only coil was used for excitation. Temperature control of the sample was successfully implemented between 25 and 40 degrees C. The perfusion system was found to be resistant to the effects of flow rate, as well as a useful tool for the administration of drugs or agents to the tissue. Changes in buffer composition could be performed on the fly without the need to reposition the sample each time a change was made. This avoidance of repositioning was found to yield a fivefold improvement in the precision of T(2) spectral parameters (using frog sciatic nerve as a sample).

Noninvasive detection of endotoxin-induced mucus hypersecretion in rat lung by MRI

Using magnetic resonance imaging (MRI), we detected a signal in the lungs of Brown Norway rats after intratracheal administration of endotoxin [lipopolysaccharide (LPS)]. The signal had two components: one, of diffuse appearance and higher intensity, was particularly prominent up to 48 h after LPS; the second, showing an irregular appearance and weaker intensity, was predominant later. Bronchoalveolar lavage fluid analysis indicated that generalized granulocytic (especially neutrophilic) inflammation was a major contributor to the signal at the early time points, with mucus being a major factor contributing at the later time points. The facts that animals can breathe freely during data acquisition and that neither respiration nor cardiac triggering is applied render this MRI approach attractive for the routine testing of anti-inflammatory drugs. In particular, the prospect of noninvasively detecting a sustained mucus hypersecretory phenotype in the lung brings an important new perspective to models of chronic obstructive pulmonary diseases in animals.

Imaging the changes in renal T1 induced by the inhalation of pure oxygen: a feasibility study

The effect of the inhalation of pure oxygen on the kidney was evaluated by measuring monoexponential T1 and T2* relaxation times in nine volunteers using a multiple-shot turbo spin echo and multiple echo gradient echo sequences, respectively. The T1 of the renal cortex decreased significantly when breathing pure oxygen as compared to normoxia (from 882 +/- 59 to 829 +/- 70 msec, P < 0.05), while that of the renal medulla was unchanged. No significant changes were seen in the T2* of either compartment. Dynamic imaging using an inversion recovery sequence with an optimized inversion time typically produced signal changes of 20% in the renal cortex. Studies to assess if oxygen-induced changes in flow contributed to this effect showed that the flow contribution was not significant. Although longer inversion times (880 ms) produced optimal contrast, acceptable contrast was also obtained at shorter inversion times (450 msec) in the renal cortex, spleen, and lung, with the latter being of opposite polarity to the other two tissues, implying a shorter parenchymal T1 than previously reported in the literature. The results are consistent with oxygen acting as an intravascular contrast agent which induces a shortening of T1 in the arterial blood volume.

Characterization of bleomycin lung injury by nuclear magnetic resonance: correlation between NMR relaxation times and lung water and collagen content

The response of the NMR relaxation times (T(1), CPMG T(2), and Hahn T(2)) to bleomycin-induced lung injury was studied in excised, unperfused rat lungs. NMR, histologic, and biochemical (collagen content measurement) analyses were performed 1, 2, 4, and 8 weeks after intratracheal instillation of saline (control lungs) or 10 U/kg bleomycin sulfate. The control lungs showed no important NMR, water content, histologic, or collagen content changes. The spin-spin relaxation times for the fast and intermediate components of the CPMG decay (T(2f) and T(2i), respectively) increased 1 week after bleomycin injury (acute inflammatory stage) and then progressively decreased during the following 2-8 weeks (i.e., with the development of the chronic, fibrotic stage of the injury). The slow component (T(2s)) showed no significant changes. The response of T(1) and the slow component of the Hahn T(2) was, on the whole, similar to that of CPMG T(2f) and T(2i). T(1) changes were very small. Lung water content increased 1 week after injury. Histologic and biochemical assessment of collagen showed that collagen content was close to control at 1 week, but markedly increased at 2, 4, and 8 weeks. T(1) and T(2) data were directly correlated with lung water content and inversely correlated with collagen content. Our results indicate that NMR relaxation time measurements (particularly T(2)) reflect the structural changes associated with bleomycin injury. The prolonged T(2) relaxation times observed in the acute stage are related to the presence of edema, whereas the subsequent decrease in these values marks the stage of the collagen deposition (fibrotic stage). CPMG-T(2) and Hahn-T(2) measurements can be valuable as a potentially noninvasive method for characterizing bleomycin-induced lung injury and pathologically related lung disorders.

MRI of lung parenchyma in rats and mice using a gradient-echo sequence

Signal of lung parenchymal tissue from the living rat and mouse lung was detected at 4.7 T with a good signal-to-noise ratio and motion-suppressed artifacts using a short TE gradient-echo sequence. Neither cardiac nor respiratory gating were applied, and animals respired freely during data collection. Mean T(2)* relaxation times of parenchyma in the anterior, middle and posterior regions of both lungs ranged between 403 and 657 micros and 397 and 751 micros, respectively for the rat and mouse. For the rat in the prone position, there was a gradient in T(2)* values, from the posterior to the anterior regions of both lungs. In the supine position, however, T(2)* values were larger in the posterior and in the anterior portions. For the mouse in both prone and supine positions, there was a tendential gradient in T(2)* from the anterior to the posterior portions. The robustness of the approach renders it well suited for routine applications, e.g. in pharmacological studies concerning asthma models in small rodents. The method was applied to lung inflammation models involving challenge with ovalbumin or lipopolysaccharide.

Magnetic resonance T2* measurements of the normal human lung in vivo with ultra-short echo times

The objective of this study was to measure T2* values of the normal human lung in vivo during breathhold using a rapid gradient-echo sequence with ultra-short echo times (TE). A sagittal slice of the right lung was imaged in six volunteers with various TE ranging from 0.5 ms to 5 ms using a clinical 1.5 Tesla MR scanner. T2* values were calculated in a region of interest in the dependent and non-dependent lung. In the dependent lung, T2* values of 1.1 ms+/-0.15 ms were measured, and in the non-dependent lung, 0.86 ms+/-0.11 (p < 0.01). T2* measurements of the normal human lung during breathhold are feasible with a clinical MR unit. The short T2* values require the use of very short TE times (< 2.5 ms) in gradient-echo sequences to obtain adequate signal intensity from lung tissue.