Diabetes Care 1 Microvascular and Macrovascular Kamel Mohammedi,1 Mark Woodward,1,2,3 Yoichiro Hirakawa,1 Disease and Risk for Major Sophia Zoungas,1,4 Bryan Williams,5 Liu Lisheng,6 Anthony Rodgers,1 Peripheral Arterial Disease in Giuseppe Mancia,7 Bruce Neal,1 Stephen Harrap,8 Michel Marre,9,10,11 and Patients With Type 2 Diabetes John Chalmers,1 on behalf of the ADVANCE Collaborative Group DOI: 10.2337/dc16-0588 OBJECTIVE Peripheral arterial disease (PAD) is a common manifestation of atherosclerosis in type 2 diabetes, but the relationship between other vascular diseases and PAD has been poorly investigated. We examined the impact of previous microvascular and 1The George Institute for Global Health, Univer- macrovascular disease on the risk of major PAD in patients with type 2 diabetes. sity of Sydney, Sydney, NSW, Australia 2The George Institute for Global Health, Univer- RESEARCH DESIGN AND METHODS sity of Oxford, Oxford, U.K. 3 We analyzed 10,624 patients with type 2 diabetes free from baseline major PAD in Department of Epidemiology, Johns Hopkins University, Baltimore, MD the Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Con- 4Monash Centre for Health Research and Imple- trolled Evaluation (ADVANCE) clinical trial. The primary composite outcome was mentation, School of Public Health and Preven- major PAD defined as PAD-induced death, peripheral revascularization, lower- tive Medicine, Monash University, Clayton, PATHOPHYSIOLOGY/COMPLICATIONS limb amputation, or chronic ulceration. The secondary end points were the PAD Victoria, Australia 5Institute of Cardiovascular Sciences, University components considered separately. College London (UCL) and National Institute of Health Research UCL Hospitals Biomedical Re- RESULTS search Centre, London, U.K. Major PAD occurred in 620 (5.8%) participants during 5 years of follow-up. Base- 6The Chinese Hypertension League Institute, line microvascular and macrovascular disease were both associated with sub- Beijing, China 7The University of Milan-Bicocca and Istituto sequent risk of major PAD after adjustment for age, sex, region of origin, and Auxologico Italiano, Milan, Italy randomized treatments. However, only microvascular disease remained signifi- 8The University of Melbourne and Royal Mel- cantly associated with PAD after further adjustment for established risk factors. bourne Hospital, Melbourne, Victoria, Australia 9 The highest risk was observed in participants with a history of macroalbuminuria INSERM, UMRS 1138, Centre de Recherche des Cordeliers, Paris, France (hazard ratio 1.91 [95% CI 1.38–2.64], P < 0.0001) and retinal photocoagulation 10Assistance Publique Hopitauxˆ de Paris, Bichat therapy (1.60 [1.11–2.32], P = 0.01). Baseline microvascular disease was also Hospital, DHU FIRE, Department of Diabetology, associated with a higher risk of chronic lower-limb ulceration (2.07 [1.56–2.75], Endocrinology and Nutrition, Paris, France 11 ´ ´ P < 0.0001) and amputation (1.59 [1.15–2.22], P = 0.006), whereas baseline macro- Universite Paris Diderot, Sorbonne Paris Cite, UFR de Medecine,´ Paris, France vascular disease was associated with a higher rate of angioplasty procedures (1.75 Corresponding author: John Chalmers, chalmers@ [1.13–2.73], P =0.01). georgeinstitute.org.au. CONCLUSIONS Received 18 March 2016 and accepted 4 July 2016. Microvascular disease, particularly macroalbuminuria and retinal photocoagula- Clinical trial reg. no. NCT00145925, clinicaltrials tion therapy, strongly predicts major PAD in patients with type 2 diabetes, but .gov. macrovascular disease does not. This article contains Supplementary Data online at http://care.diabetesjournals.org/lookup/ suppl/doi:10.2337/dc16-0588/-/DC1. Type 2 diabetes is associated with an increased risk of premature death (1). Cardio- © 2016 by the American Diabetes Association. vascular disease is the leading cause of morbidity and mortality in patients with Readers may use this article as long as the work is type 2 diabetes, who have two to three times the risk of developing myocardial properly cited, the use is educational and not for infarction and stroke compared with people without diabetes (2). Peripheral arterial profit, and the work is not altered. Diabetes Care Publish Ahead of Print, published online July 25, 2016 2 Peripheral Arterial Disease in Type 2 Diabetes Diabetes Care disease (PAD) is a common and severe ethics committee of each participating Malaysia, and India), established market clinical manifestation of atherosclerosis center, and all participants provided writ- economies (Australia, Canada, France, (3,4) and is especially frequent in pa- ten informed consent. All participants in Germany, Ireland, Italy, The Netherlands, tients with type 2 diabetes, with an ap- ADVANCE were included in the current New Zealand, and U.K.), and Eastern proximately threefold increased risk study except 516 for whom a history of Europe (Czech Republic, Estonia, Hun- compared with a population without di- PAD was established at baseline. PAD gary, Lithuania, Poland, Russia, and abetes (5). In the Action in Diabetes and was defined at baseline as a lower-limb Slovakia). Asia was considered the ref- Vascular Disease: Preterax and Diamicron amputation of at least one digit, chronic erence group on the basis of a previous MR Controlled Evaluation (ADVANCE) ulceration of a lower limb ($6 weeks) report of low prevalence of PAD in clinical trial, the incidence of PAD was believed to be due to arterial insuffi- Asians (13). Estimated glomerular filtra- comparable to the incidence of major ciency, or a peripheral revasculari- tion rate (eGFR) was computed using coronary events and stroke (6). PAD is zation procedure (surgery, angioplasty, the Chronic Kidney Disease Epidemiol- associated with poor outcomes, leading or emergency thrombolysis). ogy Collaboration equation. Cognitive to a high rate of amputation and death function was estimated by the Mini- (7), and has been associated with an in- Primary and Secondary End Points Mental State Examination (MMSE) score creased risk of cardiovascular morbidity The primary composite outcome for this and considered as normal (MMSE score fi and mortality (8,9). PAD mainly affects analysis was major PAD, de ned as at $28) or reduced (MMSE score ,28). the infrapopliteal arteries and may in- baseline, or death as a result of PAD. Educational accomplishment was de- duce more damage in small than in Each PAD outcome was considered sep- fined as age at completion of the highest large vessels in patients with type 2 di- arately as a secondary end point. PAD level of formal education and catego- abetes (7,10). The impact of prevalent outcomes were collected systematically rized as basic ($16 years) or low (#15 macrovascular or microvascular disease for all participants during the scheduled years). History of microvascular disease on the risk of developing PAD has not study visits every 2 years from case re- was defined as the presence at baseline yet been reliably compared in a contem- port forms and from reports of serious of at least macroalbuminuria (urinary porary cohort of patients with type 2 adverse events, without adjudication. albumin-to-creatinine ratio [ACR] .300 diabetes. The aim of the current study Information about the occurrence of mg/mg), retinal photocoagulation ther- was to determine the impact of micro- study outcomes and of all serious ad- apy, proliferative retinopathy, macular vascular and macrovascular disease at verse events was reported at the time edema, or blindness. History of macro- baseline on the development of major of occurrence between visits. When vascular disease was defined as the PAD during follow-up in the ADVANCE study outcomes or serious adverse presence at baseline of at least myocar- study. events occurred, the responsible inves- dial infarction, stroke, coronary artery tigator of each center ensured that the bypass graft, percutaneous transluminal RESEARCH DESIGN AND METHODS event was reported immediately by coronary angioplasty, hospital admis- completing a serious adverse events Participants sion for unstable angina, or transient ADVANCE was a large, multicenter, in- form. The data and safety monitoring ischemic attack. ternational randomized trial in patients committee regularly reviewed all such with type 2 diabetes (11). Its objectives events for each center. Statistical Analyses were to test the effects of intensive glu- Quantitative variables were expressed Selection of Candidate Risk Factors for cose control by using a gliclazide modi- as mean (SD) or median (interquartile Major PAD fi – ed release basedregimenandblood The initial set of candidate risk factors range) for those with skewed distribu- fi pressure treatment by using a xed- for the development of major PAD col- tions. Categorical parameters were ex- dose combination of perindopril and lected in ADVANCE at baseline were all pressed as numbers and percentages. indapamide on the incidence of major demographic, anthropometric, and clini- Characteristics of participants accord- microvascular and macrovascular events. cal parameters; risk factors for cardiovas- ing to the incidence of major PAD were The design and clinical characteristics of cular diseases; renal function biomarkers; compared at baseline in each individual x2 participants in ADVANCE have been cognitive function; and educational ac- region of origin by using ,ANOVA,or fl published previously (6,11,12). Brie y, complishment. Candidate risk factors Wilcoxon test. Cox proportional hazards 11,140 patients with type