Patients with type 2 diabetes mellitus are well-known to

Transcription

1 Ideas and Opinions Annals of Internal Medicine Individualizing Glycemic Targets in Type 2 Diabetes Mellitus: Implications of Recent Clinical Trials Faramarz Ismail-Beigi, MD, PhD; Etie Moghissi, MD; Margaret Tiktin, NP; Irl B. Hirsch, MD; Silvio E. Inzucchi, MD; and Saul Genuth, MD One of the first steps in the management of patients with type 2 diabetes mellitus is setting glycemic goals. Professional organizations advise setting specific hemoglobin A 1c (HbA 1c ) targets for patients, and individualization of these goals has more recently been emphasized. However, the operational meaning of glycemic goals, and specific methods for individualizing them, have not been well-described. Choosing a specific HbA 1c target range for a given patient requires taking several factors into consideration, including an assessment of the patient s risk for hyperglycemia-related complications versus the risks of therapy, all in the context of the overall clinical setting. Comorbid conditions, psychological status, capacity for self-care, economic considerations, and family and social support systems also play a key role in the intensity of therapy. The individualization of HbA 1c targets has gained more traction after recent clinical trials in older patients with established type 2 diabetes mellitus failed to show a benefit from intensive glucose-lowering therapy on cardiovascular disease (CVD) outcomes. The limited available evidence suggests that near-normal glycemic targets should be the standard for younger patients with relatively recent onset of type 2 diabetes mellitus and little or no micro- or macrovascular complications, with the aim of preventing complications over the many years of life. However, somewhat higher targets should be considered for older patients with long-standing type 2 diabetes mellitus and evidence of CVD (or multiple CVD risk factors). This review explores these issues further and proposes a framework for considering an appropriate and safe HbA 1c target range for each patient. Ann Intern Med. 2011;154: For author affiliations, see end of text. Patients with type 2 diabetes mellitus are well-known to have an increased risk for atherosclerotic cardiovascular disease (CVD) and microvascular complications. The results of recent large multicenter trials in patients with established type 2 diabetes mellitus (1 3) have confounded the optimization of therapeutic approaches to prevent these sequelae. Our purpose is to present a framework for setting a glycemic target range for an individual patient in the outpatient setting. This article represents our perspective and should not be taken as the view of any entity or organization. Table 1 (4 7) summarizes the major outcomes of 4 large trials conducted in patients with type 2 diabetes mellitus. The UKPDS (United Kingdom Prospective Diabetes Study) (4) tested the effect of more versus less intensive glycemic control in middle-aged patients (mean age, 53 years) with newly diagnosed type 2 diabetes mellitus. The results, similar to those of the DCCT (Diabetes Control and Complications Trial) (8) in patients with type 1 diabetes mellitus, demonstrated that intensive therapy delayed the onset and reduced the progression of albuminuria and retinopathy but did not decrease the risk for myocardial infarction (MI). A 10-year observational follow-up of the DCCT (9) found a reduced rate of MI, despite hemoglobin A 1c (HbA 1c ) values that converged at approximately 8.0% after 2 years. See also: Web-Only Conversion of graphics into slides The findings of the UKPDS contrast with epidemiologic data (10) that show an exponential increase in CVD events as a function of increasing HbA 1c. Three large, multicenter trials (1 3), conducted in older patients (aged 60 to 66 years) with well-established type 2 diabetes mellitus and either multiple risk factors or a previous CVD event, tested whether even more intensive glucose control would reduce CVD risk. The ACCORD (Action to Control Cardiovascular Risk in Diabetes) trial (1) tested the effect of intensive versus standard glycemic control (HbA 1c 6.0% vs. 7.0% to 7.9%) on CVD events. The intensive glycemic control in this 5-year trial was stopped after an average of 3.5 years of follow-up because of increased all-cause and CVD-related mortality with no reduction in the primary outcome (a composite of nonfatal myocardial infarction, nonfatal stroke, and CVD death). The ADVANCE (Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified-release Control Evaluation) trial (2) assessed the effect of decreasing HbA 1c to 6.5% or less on vascular outcomes. The VADT (Veterans Affairs Diabetes Trial) (3) enrolled participants (97% men) with poorly controlled type 2 diabetes mellitus. It assessed CVD outcomes from reducing HbA 1c to less than 6.0% versus 8% to 9%. Although the ADVANCE trial and the VADT found no increase in mortality, they also reported no beneficial effect of intensive glucose control on their composite macrovascular outcomes. In addition, weight gain and 2- to 3-fold higher rates of severe hypoglycemia occurred with intensive glycemic treatment. In contrast with the CVD findings, all 4 trials reported that intensive glycemic control had positive effects on various microvascular outcomes, with prevention of albuminuria being common to all. More intensive control also improved retinopathy and American College of Physicians

2 Setting Glycemic Targets in Type 2 Diabetes Mellitus Ideas and Opinions Table 1. Selected Characteristics and Outcomes of the UKPDS, ACCORD Trial, ADVANCE Trial, and VADT Variable UKPDS ACCORD Trial ADVANCE Trial VADT References Duration of study, y Goal 1 4 Intensive therapy FPG 6.0 mmol/l HbA 1c 6.0% HbA 1c 6.5% HbA 1c 6.0% ( mg/dl) Standard therapy FPG 15.0 mmol/l ( mg/dl) HbA 1c, 7.0% 7.9% Standard HbA 1c values HbA 1c, 8.0% 9.0% HbA 1c achieved, % 1 4 Intensive therapy Standard therapy Change in achieved HbA 1c,% Annual event rates per 100 patients Severe hypoglycemia* 1 4 Intensive therapy Standard therapy Primary outcome 1 4 Intensive therapy Standard therapy All-cause mortality 5 Intensive therapy Standard therapy Cardiovascular mortality 5 Intensive therapy Standard therapy Retinopathy 1 4, 6 Intensive therapy Standard therapy Visual deterioration 1 4, 6 Intensive therapy Standard therapy Neuropathy 1 4, 7 Intensive therapy Standard therapy Renal failure** 1 4, 7 Intensive therapy Standard therapy Renal failure and retinopathy 4, 7 Intensive therapy Standard therapy Microalbuminuria 1, 3, 4, 7 Intensive therapy Standard therapy Macroalbuminuria 1, 3, 4, 7 Intensive therapy Standard therapy ACCORD Action to Control Cardiovascular Risk in Diabetes; ADVANCE Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified-release Control Evaluation; FPG fasting plasma glucose; HbA 1c hemoglobin A 1c ; UKPDS United Kingdom Prospective Diabetes Study; VADT Veterans Affairs Diabetes Trial. * An episode that requires third-party assistance. P 0.05 compared with standard glycemic therapy. An aggregate end point of any diabetes-related end point (UKPDS); a composite of nonfatal myocardial infarction, nonfatal stroke, and fatal myocardial infarction and stroke (ACCORD); combined microvascular and macrovascular disease (ADVANCE); and time to occurrence of a composite of major cardiovascular events (VADT). Nine-year results (UKPDS), worsening of 3 steps in fundus photographs (ACCORD), new or worsening retinopathy (defined as development of proliferative retinopathy, macular edema, or photocoagulation therapy or diabetes-related blindness) (ADVANCE), or progression to proliferative retinopathy (VADT). Two-line (UKPDS, ADVANCE, and VADT) or 3-line (ACCORD) change in visual acuity. Nine-year results (UKPDS), loss of sensation to light touch (10-g monofilament) (ACCORD), new or worsening neuropathy (ADVANCE), or peripheral neuropathy (VADT). ** Dialysis or plasma creatinine level 250 mol/l ( 2.8 mg/dl) not related to acute intercurrent illness (UKPDS); initiation of dialysis, end-stage renal disease, renal transplantation, or increase in serum creatinine level to 292 mol/l ( 3.3 mg/dl) in the absence of an acute reversible cause (ACCORD); need for renal replacement therapy or death from renal causes (ADVANCE); or creatinine level 265 mol/l ( 3.0 mg/dl) (VADT). A composite of retinopathy, photocoagulation, vitreous hemorrhage, and fatal or nonfatal renal failure (UKPDS and ACCORD). Nine-year results (UKPDS). neuropathy outcomes in both the UKPDS and the AC- CORD trial (4, 6, 7). These findings have clinicians pondering the clinical implications of these trials, the role of intensive glucose control in prevention of complications, how to balance the potential benefits of intensive glycemic control against its risks in an individual patient, and how to arrive at an appropriate glucose target for a specific patient. These questions are posed in the context of published guidelines from professional groups with different HbA 1c goals, which are largely based on the results of older clinical trials and epidemiologic data. Each organization discusses the 19 April 2011 Annals of Internal Medicine Volume 154 Number 8 555

3 Ideas and Opinions Setting Glycemic Targets in Type 2 Diabetes Mellitus importance of individualizing glycemic goals but provides little guidance on how this should be done. Establishing an individualized goal requires the consideration of 2 sets of variables, clinical characteristics and the psychosocioeconomic setting both of which are specific to each patient. Although we discuss clinical characteristics first, the psychosocioeconomic setting often plays a determining role in setting glycemic targets (11, 12). Of note, some of our recommendations are based on clinical judgment, rather than strong evidence; these may be viewed as being overly conservative, but we prefer to err on the side of safety. CLINICAL CHARACTERISTICS Comorbid Conditions Significant comorbid conditions may limit survival or be debilitating, and they may also interfere with the implementation of glycemic control strategies. In addition, a greater disease burden may increase the number of medications that a patient receives. This can lead to confusion, errors, poor adherence, increased side effects and cost, drug drug interactions, and frustration. Patients with other medical conditions that are expected to reduce life expectancy, and thereby shorten the period in which diabetic complications can develop, should be targeted to a higher glycemic range. Age Patient age is also an important consideration. The younger the age at the onset of diabetes mellitus, the greater the cumulative exposure to hyperglycemia and subsequent risk for complications; the older the age at onset, the higher the probability of existing comorbid conditions and the shorter the expected life span. These issues are increasing in importance, because more patients develop diabetes mellitus at a younger age and older patients live longer. The ACCORD trial reported a trend toward lower all-cause mortality among participants younger than 65 years at baseline who were randomly assigned to receive standard treatment (13), whereas the ADVANCE trial reported lower rates of combined major macro- and microvascular events in younger participants in the intensivetherapy group (2); however, neither finding was statistically significant. In a VADT substudy (14), younger patients (aged 57 vs. 64 years) with lower coronary calcium scores (by computed tomography) at baseline had fewer CVD events with intensive control. No trial data on patients younger than 45 years with newly diagnosed type 2 diabetes mellitus are available. However, a near-normal HbA 1c target range seems appropriate for younger patients, who are unlikely to have established atherosclerosis. Duration of Diabetes Mellitus Participants in the UKPDS had newly diagnosed type 2 diabetes mellitus, whereas those in the ACCORD and ADVANCE trials and the VADT had type 2 diabetes mellitus for 8.0 to 11.5 years. The long-term follow-up of the UKPDS showed fewer micro- and macrovascular events as a result of previous intensive control (9). Participants in the VADT with established CVD had type 2 diabetes mellitus for 2 years longer than those without established CVD (3). These findings imply (but do not prove) that intensive treatment is more likely to have benefits the earlier it is begun. This may be especially true in patients with a family history of premature coronary artery disease. Presence of Cardiovascular (Macrovascular) Disease Patients with type 2 diabetes mellitus and a history of myocardial infarction are at high risk for recurrent events (15). Among the participants enrolled in the 3 recent trials, 32% to 40% had a history of a CVD event; these participants had greater mortality. However, intensive treatment led to no reduction in new CVD events or mortality (1 3). In the VADT substudy (14), patients with higher coronary calcium scores (associated with more advanced CVD) showed no reduction in CVD events with intensive glycemic treatment. These findings imply that a higher HbA 1c target may be more appropriate for such patients. Presence of Microvascular Disease The Kumamoto study (16) tested the effect of multiple daily insulin injections versus standard insulin therapy in Japanese patients with type 2 diabetes mellitus. The investigators reported that intensive glycemic control decreased the progression of albuminuria and retinopathy that were present at enrollment (secondary prevention). The recent trials showed modest reductions in progression of albuminuria (Table 1), and the ACCORD trial found that intensive control decreased the progression of retinopathy (6) and neuropathy but not the composite outcome of advanced microvascular disease (7). A microvascular complication in an organ often suggests that other microvascular (and macrovascular) complications may be present or evolving; more aggressive glycemic control may therefore be considered in the presence of microalbuminuria without elevated serum creatinine levels or in patients with early retinopathy (6, 7, 16). However, few data indicate that strict glucose control alters the progression of renal disease once serum creatinine level is elevated ( 221 mol/l [ 2.5 mg/dl]). Special precautions are advised in patients with autonomic neuropathy, who are predisposed to hypoglycemia unawareness and have increased cardiac mortality. History of Severe Hypoglycemia The biggest impediment to achieving near-normal glycemic control is the increased occurrence of severe hypoglycemia (that requiring third-party assistance). Severe hypoglycemia occurred approximately 2 to 3 times more often with intensive therapy in the trials (Table 1) and is more common in persons with low cognitive function (17). In older patients with type 2 diabetes mellitus, dementia is associated with episodes of severe hypoglycemia April 2011 Annals of Internal Medicine Volume 154 Number 8

4 Setting Glycemic Targets in Type 2 Diabetes Mellitus Ideas and Opinions (18), and such episodes in patients with CVD may also lead to myocardial ischemia or arrhythmia (19). The ACCORD and ADVANCE trials both reported higher mortality rates in participants with 1 or more episodes of severe hypoglycemia (20, 21), although no direct causeand-effect relationship was established. For these and other pragmatic reasons, less intensive HbA 1c targets are widely accepted as appropriate for patients with recent severe hypoglycemia. PSYCHOSOCIOECONOMIC CONTEXT Patient-specific psychological, social, and economic conditions and underlying capacities for self-management play a critical role in setting targets. These issues can be explored during a structured and detailed interview with the patient and discussion with family members, as appropriate (22). Although many of the topics we discuss seem self-evident, health care providers who are pressed for time may not address these critical issues properly. These topics are not all-inclusive and often overlap. It is important to remember that the patient must implement the strategy, and their full understanding and acceptance of the means of controlling their blood glucose level safely is critical (22 24). Safety Concerns and Support Systems Safety is of paramount importance with the use of any glycemic control strategy, particularly when drugs with a higher risk for severe hypoglycemia are being used. Safety is often related to living conditions and family support systems (25); for example, a highly intensive target would be inappropriate for an insulin-treated patient who lives alone and has no routine daily check made by family, friends, or neighbors. Patient education and health coaching may also have positive effects on patient empowerment, self-care, and outcomes (23, 24). Adverse Effects of Medications Adverse effects can include weight gain and hypoglycemia associated with the use of insulin or sulfonylureas; weight gain, edema, heart failure, and fractures associated with thiazolidinediones; or gastrointestinal side effects of metformin and certain incretin-based drugs. Drug reactions increase in frequency with the use of multiple medications. Because intensive glycemic targets usually require polypharmacy, the risk benefit ratio of adding another medication to reduce blood glucose level requires careful consideration. Figure. Framework to assist in determining glycemic treatment targets in patients with type 2 diabetes. Most Intensive 6.0% Less Intensive Least Intensive Psychosocioeconomic considerations Highly motivated, adherent, Less motivated, nonadherent, knowledgeable, excellent limited insight, poor self-care self-care capacities, and capacities, and weak comprehensive support systems support systems Low 7.0% 8.0% Hypoglycemia risk Moderate High Patient age, y Disease duration, y Other comorbid conditions None Few or mild Multiple or severe Established vascular complications None Cardiovascular disease None Early microvascular Advanced microvascular Glycemic goals and treatment intensities are shown in terms of increasing severity or magnitude of clinical variables, as well as with limitations set by the psychosocioeconomic context. Greater height of a triangle indicates increased clinical concern about the considered variable. If a patient s position on the various triangles is widely disparate, the treatment target should be determined by the farthest-right position. As always, sound clinical judgment should prevail in these circumstances. The location of the triangles in the figure is not meant to represent their relative importance in setting glycemic targets. The depicted targets assume stable outpatient treatment protocols. Depending on the set glycemic target range for any given patient, the target range may have to be decreased (for example, for a patient in the intensive care unit with an acute infection) or increased (for example, for a patient admitted for acute renal injury) for various periods. Note that although hemoglobin A 1c and mean blood glucose levels have a strong positive correlation in populations, this relationship varies substantially at an individual level and across certain populations (for various medical, nonmedical, and unknown reasons) among both glucose levels at a given hemoglobin A 1c value and hemoglobin A 1c values at a given average blood glucose level (30). A hemoglobin A 1c value represents the mean effect of glycation reaction on hemoglobin over 2 to 3 months, whereas blood glucose levels obtained by fingersticks give a more accurate picture of glycemic control on a day-to-day basis. Psychological and Cognitive Status Diabetes mellitus is a chronic disease that requires lifelong lifestyle modification and ongoing medical management. Depression, which is often present in patients with type 2 diabetes mellitus, limits the successful attainment of goals (26). Loss of cognitive function is amplified in patients with type 2 diabetes mellitus who have mild clinical or subclinical cerebrovascular disease or concomitant Alzheimer disease. A Mini-Mental Status Examination to assess cognitive function may be an important component of the clinical evaluation of patients with type 2 diabetes mellitus (27). Economic Considerations The cost of certain treatment methods, especially newly marketed medications and glucose test strips, may be prohibitive for many patients. Glycemic control can often be achieved with older, commonly available, and less expensive medications, but patients may feel that they are receiving inferior treatment. In addition, all-cause mortality rates for patients with type 2 diabetes mellitus are higher among lower socioeconomic groups (28). A careful discussion of this psychologically and socially sensitive issue is important before finalizing goals April 2011 Annals of Internal Medicine Volume 154 Number 8 557

5 Ideas and Opinions Setting Glycemic Targets in Type 2 Diabetes Mellitus Table 2. Proposed Approximate HbA 1c Targets Determined by Clinical Characteristics (in the Absence of Severe Hypoglycemia)* Age Duration of Diabetes Mellitus Complications Treatment Intensity (HbA 1c Target) Macrovascular Microvascular 45 y Any None and None or early Most intensive ( 6.5%) Any Established and/or Advanced Less intensive ( 7.0%) y Short None and None or early Intensive (6.5% 7.0%) Long None and None or early Less intensive ( 7.0%) Any Established and/or Advanced Not intensive (7.0% 8.0%) 65 y Short None and None or early Less intensive ( 7.0%) Long None and None or early Not intensive (7.0% 8.0%) Any Established and/or Advanced Moderated ( 8.0%) 75 y or infirm at any age Any Any and/or Any Moderated ( 8.0%) HbA 1c hemoglobin A 1c. * This is a summarized and highly simplified version of the text, with each of the depicted categories encompassing large segments of the population with type 2 diabetes mellitus. Patient characteristics should be considered sequentially from left to right. For example, for a person aged 43 years who has had diabetes for 6 years and had a myocardial infarction 3 years ago, we recommend a less intensive target of approximately 7%, regardless of microvascular complications. Alternatively, the same patient with no history of cardiovascular disease and few or no microvascular complications could be treated most intensively, if it was otherwise safe. Some of the proposed glycemic targets, especially the HbA 1c target 6.5%, should be viewed as suggestions and are not based on direct evidence from randomized, controlled trials. Treatment intensity should be applied after consideration of all clinical characteristics and the psychosocioeconomic context, as discussed. Not all patients fit precisely in these categories, and clinical judgment must be exercised in setting an appropriate glycemic target range for each patient. Target blood glucose level should be increased immediately after a severe hypoglycemic episode. In the absence of evidence from controlled trials, we recommend relaxation of targets by an average of mmol/l (45 60 mg/dl), or approximately 1.5% to 2.0% of HbA 1c, for at least several weeks. A more prolonged relaxation of goals is indicated after 2 or more episodes, especially if the episodes occur within weeks of each other. The presence of hypoglycemia unawareness necessitates further relaxation of targets for more prolonged periods, and perhaps permanently. Other serious diseases also necessitate reassessment of the optimal glucose levels. Note that the relationship between average blood glucose level and HbA 1c varies (30). Approximately 5 to 10 years or less. Numbers are inexact because of the uncertainty of onset of type 2 diabetes mellitus. Our choice of a 5-year threshold is somewhat arbitrary and is meant to express disease of relatively short duration. Approximately 10 to 20 years or more. Numbers are inexact because of uncertainty of onset of type 2 diabetes mellitus. We chose the 10-year threshold because the average known duration of type 2 diabetes mellitus among participants in the 3 large, recent trials was approximately 10 years. The goal is to lessen the risk for hypoglycemia while reducing the possibility of large glycosuria, water and electrolyte loss, infections, and nonketotic hyperosmolar coma. We suggest this target because it represents a mean blood glucose level of approximately 10.0 mmol/l (180 mg/dl), at which point, most patients will not be consistently or significantly glycosuric (30). Quality of Life The ultimate goal of treatment is to enhance both the short- and long-term quality of life. The variables that play into this imprecise phrase include those previously discussed and many others, which vary from patient to patient and reflect the patient s needs, desires, and beliefs. Of note, type 2 diabetes mellitus is associated with a 2- to 3- times higher prevalence of functional disabilities and comorbid conditions, mostly related to CVD and obesity (29). Sensitivity to these concerns is essential to setting a target range that is safe, acceptable, and attainable and ensures patient satisfaction. SETTING THE GLYCEMIC TARGET RANGE Setting the target range is a 2-step process, in which the clinical characteristics are considered and the range is then amended or modified on the basis of psychosocioeconomic factors. The Figure and Table 2 suggest glycemic target ranges based on patient clinical characteristics. These represent oversimplifications of complex medical decisionmaking processes. Although most patients fit into the described categories, not all are satisfactory matches. Some of the proposed glycemic goals, especially the target HbA 1c range of 6.5% or less, are not supported by evidence from randomized, controlled trials. As detailed in Table 2, suggested targets should be relaxed for the short or longer term if severe hypoglycemia or serious comorbid conditions are present. CONCLUSION On the basis of our interpretation of these landmark trials, we advance the following general recommendations for setting glycemic target ranges. First, in younger persons with relatively recent onset of type 2 diabetes mellitus, strict glycemic control aimed at a near-normal (or normal) glycemic target range should be implemented whenever safe and possible, with the goal of preventing microvascular and macrovascular complications over the life span (1 4, 6, 16). Second, in older persons with established type 2 diabetes mellitus of long duration and evidence of or risk factors for CVD, a more relaxed target range can be considered (1 3). Finally, the proposed glycemic target range should then be adjusted on the basis of the patient s psychosocioeconomic context. In summary, the glycemic target range for patients with type 2 diabetes mellitus should be individualized according to age; stage of disease, both in terms of duration and presence of macro- and microvascular complications; and propensity for hypoglycemia. In addition, careful consideration must be given to each patient s capacities, desires, and values, as well as their living situation, support systems, cognitive status, overall prognosis, and life expectancy. In nearly all circumstances, the patient should be an active participant in setting goals. Finally, glycemic targets should not be viewed as fixed goals; they should be flexible and be adapted to changes in the patient s health and living conditions April 2011 Annals of Internal Medicine Volume 154 Number 8

6 Setting Glycemic Targets in Type 2 Diabetes Mellitus Ideas and Opinions From Case Western Reserve University, Cleveland, Ohio; University of California, Los Angeles, Los Angeles, California; University of Washington, Seattle, Washington; and Yale University, New Haven, Connecticut. Potential Conflicts of Interest: Disclosures can be viewed at M Requests for Single Reprints: Faramarz Ismail-Beigi, MD, PhD, Department of Medicine, Case Western Reserve University, Euclid Avenue, Cleveland, OH ; , fxi2@case.edu. Current author addresses and author contributions are available at References 1. Gerstein HC, Miller ME, Byington RP, Goff DC Jr, Bigger JT, Buse JB, et al; Action to Control Cardiovascular Risk in Diabetes Study Group. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med. 2008;358: [PMID: ] 2. Patel A, MacMahon S, Chalmers J, Neal B, Billot L, Woodward M, et al; ADVANCE Collaborative Group. Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes. N Engl J Med. 2008;358: [PMID: ] 3. Duckworth W, Abraira C, Moritz T, Reda D, Emanuele N, Reaven PD, et al; VADT Investigators. Glucose control and vascular complications in veterans with type 2 diabetes. N Engl J Med. 2009;360: [PMID: ] 4. UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet. 1998;352: [PMID: ] 5. Turnbull FM, Abraira C, Anderson RJ, Byington RP, Chalmers JP, Duckworth WC, et al; Control Group. Intensive glucose control and macrovascular outcomes in type 2 diabetes. Diabetologia. 2009;52: [PMID: ] 6. Chew EY, Ambrosius WT, Davis MD, Danis RP, Gangaputra S, Greven CM; ACCORD Study Group, ACCORD Eye Study Group. Effects of medical therapies on retinopathy progression in type 2 diabetes. N Engl J Med. 2010; 363: [PMID: ] 7. Ismail-Beigi F, Craven T, Banerji MA, Basile J, Calles J, Cohen RM, et al; ACCORD trial group. Effect of intensive treatment of hyperglycaemia on microvascular outcomes in type 2 diabetes: an analysis of the ACCORD randomised trial. Lancet. 2010;376: [PMID: ] 8. The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of longterm complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993; 329: [PMID: ] 9. Holman RR, Paul SK, Bethel MA, Matthews DR, Neil HA. 10-year follow-up of intensive glucose control in type 2 diabetes. N Engl J Med. 2008; 359: [PMID: ] 10. Khaw KT, Wareham N, Bingham S, Luben R, Welch A, Day N. Association of hemoglobin A 1c with cardiovascular disease and mortality in adults: the European prospective investigation into cancer in Norfolk. Ann Intern Med. 2004;141: [PMID: ] 11. U.S. Department of Veterans Affairs. Management of Diabetes Mellitus in Primary Care. Washington, DC: U.S. Department of Veterans Affairs; Accessed at on 7 February Qaseem A, Vijan S, Snow V, Cross JT, Weiss KB, Owens DK; Clinical Efficacy Assessment Subcommittee of the American College of Physicians. Glycemic control and type 2 diabetes mellitus: the optimal hemoglobin A1c targets. A guidance statement from the American College of Physicians. Ann Intern Med. 2007;147: [PMID: ] 13. Calles-Escandón J, Lovato LC, Simons-Morton DG, Kendall DM, Pop- Busui R, Cohen RM, et al. Effect of intensive compared with standard glycemia treatment strategies on mortality by baseline subgroup characteristics: the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial. Diabetes Care. 2010;33: [PMID: ] 14. Reaven PD, Moritz TE, Schwenke DC, Anderson RJ, Criqui M, Detrano R, et al; Veterans Affairs Diabetes Trial. Intensive glucose-lowering therapy reduces cardiovascular disease events in Veterans Affairs Diabetes Trial participants with lower calcified coronary atherosclerosis. Diabetes. 2009;58: [PMID: ] 15. Saydah S, Tao M, Imperatore G, Gregg E. GHb level and subsequent mortality among adults in the U.S. Diabetes Care. 2009;32: [PMID: ] 16. Ohkubo Y, Kishikawa H, Araki E, Miyata T, Isami S, Motoyoshi S, et al. Intensive insulin therapy prevents the progression of diabetic microvascular complications in Japanese patients with non-insulin-dependent diabetes mellitus: a randomized prospective 6-year study. Diabetes Res Clin Pract. 1995;28: [PMID: ] 17. de Galan BE, Zoungas S, Chalmers J, Anderson C, Dufouil C, Pillai A, et al; ADVANCE Collaborative Group. Cognitive function and risks of cardiovascular disease and hypoglycaemia in patients with type 2 diabetes: the Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation (ADVANCE) trial. Diabetologia. 2009;52: [PMID: ] 18. Whitmer RA, Karter AJ, Yaffe K, Quesenberry CP Jr, Selby JV. Hypoglycemic episodes and risk of dementia in older patients with type 2 diabetes mellitus. JAMA. 2009;301: [PMID: ] 19. Desouza C, Salazar H, Cheong B, Murgo J, Fonseca V. Association of hypoglycemia and cardiac ischemia: a study based on continuous monitoring. Diabetes Care. 2003;26: [PMID: ] 20. Bonds DE, Miller ME, Bergenstal RM, Buse JB, Byington RP, Cutler JA, et al. The association between symptomatic, severe hypoglycaemia and mortality in type 2 diabetes: retrospective epidemiological analysis of the ACCORD study. BMJ. 2010;340:b4909. [PMID: ] 21. Zoungas S, Patel A, Chalmers J, de Galan BE, Li Q, Billot L; ADVANCE Collaborative Group. Severe hypoglycemia and risks of vascular events and death. N Engl J Med. 2010;363: [PMID: ] 22. Peyrot M, Rubin RR, Lauritzen T, Skovlund SE, Snoek FJ, Matthews DR, et al. Patient and provider perceptions of care for diabetes: results of the crossnational DAWN Study. Diabetologia. 2006;49: [PMID: ] 23. Blickem C, Bower P, Protheroe J, Kennedy A, Vassilev I, Sanders C, et al. The role of information in supporting self-care in vascular conditions: a conceptual and empirical review. Health Soc Care Community [PMID: ] 24. Wolever RQ, Dreusicke M, Fikkan J, Hawkins TV, Yeung S, Wakefield J, et al. Integrative health coaching for patients with type 2 diabetes: a randomized clinical trial. Diabetes Educ. 2010;36: [PMID: ] 25. August KJ, Sorkin DH. Marital status and gender differences in managing a chronic illness: the function of health-related social control. Soc Sci Med. 2010; 71: [PMID: ] 26. Katon W, Fan MY, Unützer J, Taylor J, Pincus H, Schoenbaum M. Depression and diabetes: a potentially lethal combination. J Gen Intern Med. 2008; 23: [PMID: ] 27. Alencar RC, Cobas RA, Gomes MB. Assessment of cognitive status in patients with type 2 diabetes through the Mini-Mental Status Examination: a crosssectional study. Diabetol Metab Syndr. 2010;2:10. [PMID: ] 28. Lipscombe LL, Austin PC, Manuel DG, Shah BR, Hux JE, Booth GL. Income-related differences in mortality among people with diabetes mellitus. CMAJ. 2010;182:E1-E17. [PMID: ] 29. Kalyani RR, Saudek CD, Brancati FL, Selvin E. Association of diabetes, comorbidities, and A1C with functional disability in older adults: results from the National Health and Nutrition Examination Survey (NHANES), Diabetes Care. 2010;33: [PMID: ] 30. Nathan DA, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ; A 1c - Derived Average Glucose Study Group. Translating the A 1c assay into estimated average glucose values. Diabetes Care. 2008;31: [PMID: ] 19 April 2011 Annals of Internal Medicine Volume 154 Number 8 559

7 Annals of Internal Medicine Current Author Addresses: Drs. Ismail-Beigi and Genuth and Ms. Tiktin: Department of Medicine, Case Western Reserve University, Euclid Avenue, Cleveland, OH Dr. Moghissi: 4644 Lincoln Boulevard, Suite 409, Marina Del Rey, CA Dr. Hirsch: University of Washington Medical Center, 1959 Northeast Pacific Street, Seattle, WA Dr. Inzucchi: Yale University School of Medicine, 333 Cedar Street, New Haven, CT Author Contributions: Conception and design: F. Ismail-Beigi, E. Moghissi, M. Tiktin, S.E. Inzucchi. Analysis and interpretation of the data: F. Ismail-Beigi, E. Moghissi, I.B. Hirsch, S.E. Inzucchi, S. Genuth. Drafting of the article: F. Ismail-Beigi, E. Moghissi, M. Tiktin, I.B. Hirsch, S. Genuth. Critical revision of the article for important intellectual content: F. Ismail-Beigi, E. Moghissi, M. Tiktin, I.B. Hirsch, S.E. Inzucchi, S. Genuth. Final approval of the article: F. Ismail-Beigi, E. Moghissi, M. Tiktin, I.B. Hirsch, S.E. Inzucchi, S. Genuth. Collection and assembly of data: F. Ismail-Beigi, E. Moghissi. W April 2011 Annals of Internal Medicine Volume 154 Number 8