European Journal of Nuclear Medicine Vol. 29, No. 7, July 2002 Abstract. Effects of oxygen-derived free radicals are sug- gested to be a potential pathogenic factor for endothelial dysfunction. In this study we sought to evaluate the effect of hydroxyl radicals on the human coronary vascular bed in type I diabetes mellitus using positron emission tomog- raphy (PET). Thirteen patients with type 1 diabetes under- went PET using nitrogen-13 ammonia at rest and during sympathetic stimulation with the cold pressor test (CPT). The rest-stress study protocol was repeated twice (on dif- ferent days) using pre-stress infusion of either saline as placebo or deferoxamine, an iron chelator which inhibits generation of hydroxyl radicals. At rest, global MBF was higher in diabetics than in normal controls (78.1±17.5 vs 63.2±14.9 mg 100 g –1 min –1 , P<0.05) and myocardial vas- cular resistance (MVR) showed a trend towards lower val- ues (patients, 1.28±0.35; controls, 1.55±0.32, P=NS). CPT increased MBF in all controls while 7/13 diabetics responded normally. CPT decreased MVR in 10/13 con- trols but in only 4/13 diabetics. There was no significant difference in the duration of diabetes, HbA1c, daily insu- lin dose, body mass index, or lipid profiles between pa- tients with and patients without abnormal MBF or MVR responses. Pre-stress infusion of deferoxamine normalized MBF response in all six patients, and MVR response in six of the nine patients. Another group consisting of seven patients underwent a rest-rest protocol after infusion of deferoxamine and saline to investigate the effect of defer- oxamine on resting MBF. Deferoxamine did not change the resting MBF (deferoxamine, 81±17 ml 100 g –1 min –1 ; saline, 75±19 ml 100 g –1 min –1 , P=NS) or MVR (deferox- amine, 1.0±0.5 mmHg ml –1 100 g –1 min –1 ; saline, 1.2±0.6 mmHg ml –1 100 g –1 min –1 , P=NS). In conclusion, inhibition of hydroxyl radical formation using deferoxam- ine significantly improved the responses of coronary mi- crovasculature to sympathetic stimulation. Hydroxyl radi- cals may play a role in the pathogenesis of flow abnormal- ities in type 1 diabetes. Keywords: Endothelium – Free radicals – Type 1 diabe- tes mellitus – Positron emission tomography Eur J Nucl Med (2002) 29:891–898 DOI 10.1007/s00259-002-0799-0 Introduction Diabetic coronary arteries are known to suffer impaired vasodilator function, an abnormality preceding macro- vascular arteriosclerosis [1, 2, 3, 4, 5]. Recent animal studies have shown significant endothelial dysfunction in diabetic vessels, which may explain the pathogenesis of vascular dysfunction [6, 7, 8]. Since inhibition of free radicals restores endothelial dysfunction, it has been hy- pothesized that increased oxidative stress, particularly due to hydroxyl radicals, deactivates the vasodilatory ef- fect of nitric oxide (NO) in diabetic vessels [6]. In hu- mans, Nitenberg et al. supported the hypothesis by mea- suring the relative response of coronary arterial flow to the cold pressor test (CPT) [9]. However, no previous studies have tried to quantitatively measure absolute myocardial blood flow and the effects of free radicals on human coronary arteries in diabetes mellitus. Positron emission tomography (PET) is clinically available for the quantitative measurement of myocardial Dr. Hattori is recipient of a fellowship from Alexander von Hum- boldt Stiftung, Bonn, Germany Naoya Hattori ( ) Nuklearmedizinische Klinik und Poliklinik, Klinikum rechts der Isar, Technischen Universität München, Munich, Germany Original article Deferoxamine improves coronary vascular responses to sympathetic stimulation in patients with type 1 diabetes mellitus Naoya Hattori 1, 3 , Oliver Schnell 2 , Frank M. Bengel 1 , Julian Rihl 2 , Stephan G. Nekolla 1 , Alexander E. Drzezga 1 , Eberhard Standl 2 , Markus Schwaiger 1 1 Nuklearmedizinische Klinik und Poliklinik, Klinikum rechts der Isar, Technischen Universität München, Munich, Germany 2 Diabetes Research Center, Schwabing City Hospital, Munich, Germany 3 Correspondence to: Department of Molecular and Medical Pharmacology, UCLA School of Medicine, Box 956948, Los Angeles, CA 90095-6948, USA, e-mail: hattori@kuhp.kyoto-u.ac.jp, Tel.: +1-310-7944472, Fax: +1-310-8254517 Received 1 December 2001 and in revised form 10 February 2002 / Published online: 24 April 2002 © Springer-Verlag 2002