Clinical Care/Education/Nutrition/Psychosocial Research BRIEF REPORT Coenzyme Q10 Improves Endothelial Dysfunction in Statin-Treated Type 2 Diabetic Patients SANDRA J. HAMILTON, PGDIPHEALSC weeks. After a 4-week washout, partici- GERARD T. CHEW, MD pants crossed over to the alternate treat- GERALD F. WATTS, DSC ment. Brachial artery ultrasonography was performed, and fasting blood and 24-h urine samples were collected at the OBJECTIVE — The vascular benefits of statins might be attenuated by inhibition of coen- start and end of each treatment period. zyme Q10 (CoQ10) synthesis. We investigated whether oral CoQ10 supplementation improves The Royal Perth Hospital Ethics Commit- endothelial dysfunction in statin-treated type 2 diabetic patients. tee approved the study. The brachial artery was imaged using RESEARCH DESIGN AND METHODS — In a double-blind crossover study, 23 statin- a 12-MHz transducer connected to an treated type 2 diabetic patients with LDL cholesterol Ͻ2.5mmol/l and endothelial dysfunction Ͻ Acuson Aspen ultrasound system (Sie- (brachial artery flow-mediated dilatation [FMD] 5.5%) were randomized to oral CoQ10 (200 mg/day) or placebo for 12 weeks. We measured brachial artery FMD and nitrate-mediated mens Medical Solutions, Malvern, PA), and FMD was measured as previously de- dilatation (NMD) by ultrasonography. Plasma F2-isoprostane and 24-h urinary 20- hydroxyeicosatetraenoic acid (HETE) levels were measured as systemic oxidative stress markers. scribed (4). Endothelium-independent nitrate-mediated dilatation was measured RESULTS — Compared with placebo, CoQ10 supplementation increased brachial artery FMD following sublingual administration of Ϯ ϭ ϭ ␮ by 1.0 0.5% (P 0.04), but did not alter NMD (P 0.66). CoQ10 supplementation also did glyceryl trinitrate (400 g). Ultrasound ϭ ϭ not alter plasma F2-isoprostane (P 0.58) or urinary 20-HETE levels (P 0.28). images were analyzed using semiauto- mated edge-detection software (5). CONCLUSIONS — CoQ10 supplementation improved endothelial dysfunction in statin- Total cholesterol, triglycerides, and treated type 2 diabetic patients, possibly by altering local vascular oxidative stress. HDL cholesterol were determined by en- Diabetes Care 32:810–812, 2009 zymatic methods, and LDL cholesterol was calculated using the Friedewald equation. GHb was measured using high- ndothelial dysfunction portends di- supplementation would improve endo- performance liquid chromatography. abetic vasculopathy. Endothelial thelial dysfunction in statin-treated type 2 Plasma CoQ10 was measured by reverse- E dysfunction reflects increased vascu- diabetic patients. phase high-performance liquid chroma- lar oxidative stress, whereby uncoupling tography using electrochemical detection of endothelial nitric oxide synthase activ- RESEARCH DESIGN AND (interassay coefficient of variation 14%). ity and mitochondrial oxidative phos- METHODS — We recruited type 2 di- Plasma F -isoprostane and 24-h urinary phorylation impairs the bioavailability abetic patients aged 40–79 years on sta- 2 Ն 20-hydroxyeicosatetraenoic acid levels and action of nitric oxide (1). ble-dose statin therapy for 6 weeks. (markers of systemic oxidative stress) Statins are widely used in diabetes Inclusion criteria were serum LDL choles- Ͻ were measured by gas chromatography- management and can reduce cardiovas- terol 2.5 mmol/l and endothelial dys- mass spectrometry (interassay coefficients cular events (2). However, a proportion of function, defined as brachial artery flow- Ͻ of variation 5.6 and 10%, respectively) statin-treated patients remain at risk of mediated dilatation (FMD) 5.5%. (6–8). cardiovascular disease. Statins inhibit Exclusions included use of antioxidant Data were analyzed using SPSS 15.0 conversion of 3-hydroxy-3-methylglu- supplements or other lipid-regulating (Chicago, IL) and SAS 9.1 (Cary, NC). taryl-CoA to mevalonate, but may thereby medications, GHb Ͼ8.5%, and blood Ͼ Plasma CoQ10 data (skewed distribution) also decrease production of other inter- pressure 150/90 mmHg. were logarithmically transformed for mediates in the cholesterol biosynthetic Eligible subjects were assigned in a parametric analysis. Treatment effects pathway, such as coenzyme Q10 (CoQ10) double-blind and randomized manner to were compared using mixed-effects mod- (3), an important intracellular antioxi- oral CoQ10 (200 mg/day) (Blackmores, els. Carryover effects were examined for dant. We hypothesized that oral CoQ10 Balgowlah, Australia) or placebo for 12 and excluded. ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● From the School of Medicine and Pharmacology, Royal Perth Hospital Unit, University of Western Australia, RESULTS — Participants were typi- Perth, Australia. cally middle-aged (mean Ϯ SD age 68 Ϯ 6 Corresponding author: Gerald F. Watts, [email protected]. years) and overweight (BMI 29 Ϯ 4 kg/ Received 22 September 2008 and accepted 6 February 2009. 2 Published ahead of print at http://care.diabetesjournals.org on 19 February 2009. DOI: 10.2337/dc08-1736. m ), with satisfactory control of glycemia Clinical trial reg. no. ACTRN12605000250639, anzctr.org.au. (GHb 6.9 Ϯ 0.7%), blood pressure (sys- © 2009 by the American Diabetes Association. Readers may use this article as long as the work is properly tolic 123 Ϯ 14 mmHg and diastolic 65 Ϯ cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons. 7), and lipids (LDL cholesterol 1.8 Ϯ 0.3 org/licenses/by-nc-nd/3.0/ for details. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby mmol/l). Median duration of diabetes was marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 8 years. Seventy-eight percent of subjects 810 DIABETES CARE, VOLUME 32, NUMBER 5, MAY 2009 Hamilton, Chew, and Watts had a history of hypertension, 48% had a Table 1—Effect of placebo and oral CoQ10 on arterial function, biochemical variables, and history of stroke or coronary disease, and blood pressure 26% had microvascular complications. Eighty-three percent of subjects were tak- Placebo Oral CoQ P ing antihyperglycemic medication, most 10 commonly metformin (78%); 52% were Baseline brachial artery diameter (mm) taking ACE inhibitors; 21% were taking Pretreatment 3.9 Ϯ 0.1 3.9 Ϯ 0.1 angiotensin receptor blockers; and 65% Treatment end 3.9 Ϯ 0.1 3.9 Ϯ 0.1 were taking aspirin. Atorvastatin was the Change Ϫ0.1 Ϯ 0.0 0.0 Ϯ 0.0 0.69 most commonly prescribed statin (52%), Brachial artery FMD (%) followed by simvastatin (35%) and prav- Pretreatment 2.2 Ϯ 0.6 2.2 Ϯ 0.7 astatin (13%). Treatment end 2.1 Ϯ 0.7 3.2 Ϯ 0.5 Baseline brachial artery diameter was Change 0.0 Ϯ 0.5 1.0 Ϯ 0.6 0.04 similar at all assessments and unaltered by Brachial artery NMD (%) Ϯ Ϯ CoQ10 supplementation (Table 1). CoQ10 Pretreatment 16.9 1.1 17.3 0.9 increased brachial artery FMD by mean Ϯ Treatment end 17.8 Ϯ 1.0 17.5 Ϯ 1.0 SEM 1.0 Ϯ 0.5% (P ϭ 0.04) compared with Change 0.9 Ϯ 0.9 0.2 Ϯ 0.8 0.66 ␮ placebo but did not alter nitrate-mediated Plasma CoQ10 ( mol/l) dilatation (P ϭ 0.66). Absolute percent Pretreatment 0.9 (0.2) 0.8 (0.3) FMD pre- and postplacebo, pre- and post- Treatment end 0.8 (0.2) 2.2 (1.5) Ͻ CoQ10, and change in percent FMD with Change 0.0 (0.1) 1.2 (1.5) 0.001 placebo and CoQ10 are shown in supple- Plasma F2-isoprostanes (pmol/l) mental Figures A1, A2, and A3 (available in Pretreatment 1,302 Ϯ 68 1,284 Ϯ 70 an online appendix at http://care.diabetes Treatment end 1,275 Ϯ 86 1,298 Ϯ 69 journals.org/cgi/content/full/dc08-1736/ Change Ϫ27 Ϯ 55 14 Ϯ 42 0.58 DC1), re- spectively. Despite increasing Urinary 20-HETE (pmol/24 h) plasma CoQ levels 2.7-fold (P Ͻ 0.001), Pretreatment 828 Ϯ 102 831 Ϯ 109 10 Ϯ Ϯ CoQ10 supplementation did not alter Treatment end 775 104 888 126 ϭ Ϫ Ϯ Ϯ plasma F2-isoprostane (P 0.58) or uri- Change 53 80 57 117 0.28 nary 20-hydroxyeicosatetraenoic acid levels GHb (%) (P ϭ 0.28) or influence glycemia, blood Pretreatment 7.0 Ϯ 0.1 7.0 Ϯ 0.2 pressure, or lipids (P Ͼ 0.05). Treatment end 6.9 Ϯ 0.2 7.0 Ϯ 0.2 Change Ϫ0.1 Ϯ 0.1 Ϫ0.1 Ϯ 0.1 0.58 LDL cholesterol (mmol/l) CONCLUSIONS — The new finding Pretreatment 1.9 Ϯ 0.1 1.7 Ϯ 0.1 Ϯ Ϯ was that CoQ10 supplementation im- Treatment end 1.9 0.1 2.0 0.1 proved endothelial dysfunction in statin- Change 0.1 Ϯ 0.1 0.2 Ϯ 0.1 0.41 treated type 2 diabetic patients, with no Systolic blood pressure (mmHg) alteration in two markers of systemic ox- Pretreatment 126 Ϯ 4 122 Ϯ 3 idative stress. This is consistent with our Treatment end 121 Ϯ 3 121 Ϯ 4 previous study in statin-naive dyslipide- Change Ϫ4 Ϯ 3 Ϫ1 Ϯ 2 0.38 mic type 2 diabetic patients in whom Diastolic blood pressure (mmHg) oral CoQ also improved brachial ar- Pretreatment 67 Ϯ 164Ϯ 2 10 Ϯ Ϯ tery FMD but did not alter plasma F2- Treatment end 65 1661 isoprostane levels (4). However, a study Change Ϫ2 Ϯ 11Ϯ 1 0.09 in coronary heart disease patients (20% Data are means Ϯ SEM or medians (interquartile range). Treatment effects compared using mixed-effects with diabetes and 80% statin-treated) models, with adjustment for baseline, treatment sequence, and period. 20-HETE, 20-hydroxyeicosatetra- enoic acid; NMD, nitrate-mediated dilatation. showed that oral CoQ10 increased both brachial artery FMD and endothelium- bound extracellular superoxide dis- 122 pmol/l, respectively; P ϭ 0.80) (4) with those in the statin-naive dyslipide- mutase activity, suggesting that the probably reflects their satisfactory glyce- mic type 2 diabetic patients in our previ- beneficial effects on endothelial function mic control; our results might have dif- ous study (0.8 [0.2] vs.