Slow Spin Exchange Explains the Effects of Diuresis on Gadolinium Enhancement in MR Imaging in the Kidney, not in Other Organs

Editor: Katzberg et al (1) described the increased intensity of the T1-weighted signal at magnetic resonance (MR) imaging of the kidneys and other abdominal organs after injection of gadolinium-based contrast material. They showed that this effect was greatly enhanced in patients who also received furosemide prior to imaging. Pointing to the confinement of gadolinium in extracellular space and to the furosemide-induced transport of water from intrato extracellular space, the authors attributed the additional increase in signal intensity to slow spin exchange between the two spatial compartments. In other words, they assumed that the signal intensity of intracellular space does not change in the presence of extracellular gadolinium, because the intracellular T1 remains constant. We doubt that this mechanism explains the results for organs other than the kidneys, for the following reasons: (a) The exchange of spins (ie, free tissue water) between intraand extracellular space should be intermediate to fast at the usual dose levels of the contrast agent (2), and (b) a considerable fraction of the furosemide-induced signal intensity increase was observed by Katzberg et al before the arrival of the contrast medium bolus in the organ under examination (1, figs 2–4). We believe that other effects may explain the results in these organs: For instance, the increase in signal intensity could be due to a furosemide-induced increase in T2, even when the spin exchange is fast. However, a furosemide-induced increase in T2 could not explain the kidney enhancement diagrammed in Figure 1 (1); according to the data presented there, furosemide affects only the level of signal intensity enhancement after the administration of contrast material, not the precontrast signal intensity. In our view, the increase in gadolinium-related enhancement of the kidney may be caused primarily by furosemide-induced alteration of the size of the nephronal lumen. The first pass of the arterial bolus of gadolinium-based contrast material through the kidneys brings a large molar mass of the contrast agent into the nephronal system by glomerular filtration (the filtered bolus). The injected dose in the experiment conducted by Katzberg et al was 0.1 mmol/kg. With a typical patient weight of 75 kg, the total dose would have been 7.5 mmol. Given that the fractional blood flow to each kidney is about 10% and the filtered fraction of the blood plasma is about 20%, the filtered bolus would have had a molar mass of 0.15 mmol. In a typical kidney weighing about 0.3 kg, this molar mass would have a concentration of 0.5 mmol/kg during its passage through the nephrons—five times the mean concentration in the whole body. The filtered bolus would reside in the nephrons for several minutes before being excreted, and during that time it would be an important contributor to renal signal intensity. It is plausible that the spin exchange in the kidney is slow during nephronal passage of the filtered bolus because of the compartmentalization of the bolus among the nephronal lumina and because the transport of water from the nephrons to the renal interstitium is regulated (eg, by osmotic gradients and by hormonal control). Furosemideinduced diuresis can be expected to lead to a decrease in water resorption from the nephrons and a corresponding increase in nephronal volume. In the presence of slow