RAPID COMMUNICATION Noninvasive Assessment of Early Kidney Allograft Dysfunction by Blood Oxygen Level-Dependent Magnetic Resonance Imaging Arjang Djamali, 1,3 Elizabeth A. Sadowski, 2 Millie Samaniego-Picota, 1 Sean B. Fain, 2 Rebecca J. Muehrer, 1 Sara K. Alford, 1 Thomas M. Grist, 2 and Bryan N. Becker 1 Background. Blood oxygen level-dependent (BOLD) magnetic resonance imaging (MRI) is a noninvasive method to assess tissue oxygen bioavailability, using deoxyhemoglobin as an endogenous contrast agent. We hypothesized that BOLD-MRI could accurately discriminate different types of rejection early after kidney transplantation. Methods. Twenty-three patients underwent imaging in the first four months posttransplant. Five had normal func- tioning transplants and 18 had biopsy-proven acute allograft dysfunction (acute tubular necrosis [ATN, n=5] and acute rejection [n=13] including borderline rejection: n=3; IA rejection: n=4; IIA rejection: n=6: C4d(+) rejection: n=9). Results. Mean medullary R2* (MR2*) levels (a measure directly proportional to tissue deoxyhemoglobin levels) were significantly higher in normal functioning allografts (R2*=24.3/s2.3) versus acute rejection (R2*=16.6/s2.1) and ATN (R2*=20.9/s1.8) (P0.05). The lowest MR2* levels were observed in acute rejection episodes with vascular injury i.e. IIA and C4d (+). Similarly, the lowest medullary to cortical R2* ratios (MCR2*) were present in allografts with IIA (1.240.05) and C4d(+) rejection (1.260.06). ROC curve analyses suggested that MR2* and MCR2* values could accurately discriminate acute rejection in the early posttransplant period. Conclusions. BOLD-MRI demonstrated significant changes in medullary oxygen bioavailability in allografts with biopsy-proven ATN and acute rejection, suggesting that there may be a role for this noninvasive tool to evaluate kidney function early after transplantation. Keywords: Blood oxygen level-dependent, Magnetic resonance imaging, Rejection, Transplantation, Kidney, Oxygen- ation, Acute tubular necrosis, Hypoxia, Medulla, Cortex. (Transplantation 2006;82: 621–628) B lood oxygen level-dependent (BOLD) magnetic reso- nance imaging (MRI) is a noninvasive method to assess tissue oxygen bioavailability, using deoxyhemoglobin as an endogenous contrast agent (1–3). The technique is based on the biophysical properties of hemoglobin (Hb). Oxyhemo- globin is a diamagnetic molecule that creates no magnetic moment (oxygen bound to iron), whereas deoxyhemoglobin is a paramagnetic molecule that generates magnetic moments by its unpaired electrons (iron) (4–6). Increased blood con- centrations of deoxyhemoglobin result in increased magnetic spin dephasing of blood water protons and decreased signal in- tensity on T2* (apparent spin-spin relaxation time)-weighted MR imaging sequences (1–6). The apparent relaxation rate or R2* (=1/T2*) may be calculated as the slope of log e (inten- sity) versus echo time and is directly proportional to the tissue content of deoxyhemoglobin (2). Increased R2* levels imply increased deoxyhemoglobin (decreased oxyhemoglobin) and decreased partial pressures of oxygen (PaO2) in tissues (1–3). BOLD-MRI uses deoxyhemoglobin as an endogenous con- trast molecule and thus, can be used to determine intrarenal oxygen bioavailability (4). BOLD-MRI has been used to investigate human and experimental models of kidney disease (1–3, 7–13). Early re- ports using BOLD-MRI to image kidneys demonstrated that furosemide increased medullary PaO2 in healthy human subjects, consistent with the drug’s property to inhibit ac- tive sodium reabsorption and oxygen consumption in the medullary thick ascending limb of Henle (mTAL) (2). In contrast, acetazolamide had no effect on medullary PaO2, confirming the findings from previous experimental stud- ies using microelectrodes (14) and providing evidence that BOLD-MRI could accurately detect changes in intrarenal ox- Parts of this work were supported by an AST/Fujisawa Award and National Institutes of Health grant DK 067981-02 (A.D.), AHA SDG 0235290N (M.S.), National Institutes of Health grant 5K24 DK616962-03 (B.N.B.), and by the M01RR03186 grant from the General Clinical Research Cen- ters Program of the National Center for Research Resources, National Institutes of Health. 1 Departments of Medicine, University of Wisconsin Madison, Madison, WI. 2 Departments of Radiology, University of Wisconsin Madison, Madison, WI. 3 Address correspondence to: Arjang Djamali, M.D., 3034 Fish Hatchery Road, Suite B, Madison, WI 53713. E-mail: axd@medicine.wisc.edu Received 27 February 2006. Revision requested 4 April 2006. Accepted 27 April 2006. Copyright © 2006 by Lippincott Williams & Wilkins ISSN 0041-1337/06/8205-621 DOI: 10.1097/01.tp.0000234815.23630.4a Transplantation • Volume 82, Number 5, September 15, 2006 621