Improving diabetes diagnosis and management in myocardial infarction patients: overcoming clinical inertia

Abstract

Diabetes is underdiagnosed and undertreated among acute myocardial infarction (AMI) patients. The early diagnosis and treatment of diabetes during AMI facilitates improved risk stratification, use of appropriate revascularization strategies and secondary prevention medications, and timely initiation of glycemic therapy. Accurate diagnostic methods, such as hemoglobin A1c, should be evaluated for hospitalized AMI patients. In addition, efforts to improve the uptake of diabetes screening and management in the hospitalized setting should occur. Possible actions include the use of clinical information systems to generate physician reminders for diabetes detection and management, audit and feedback programs, and professional society initiatives to address diabetes screening and therapy initiation through clinical guidelines and performance measures. Through the application of both these and other efforts listed in the manuscript, the rates of undiagnosed and undertreated diabetes among AMI patients can be significantly reduced, which would lead to an improvement in both diabetic and cardiovascular outcomes.

Keywords:

• diabetes
• diagnosis
• education
• myocardial infarction
• treatment

Diabetes confers significant risk to patients with an acute myocardial infarction (AMI), but is often neglected during the acute hospitalization. For example, among AMI patients with known diabetes, up to two-thirds are sub-optimally treated for their diabetes [1,2]. Even more alarmingly, although approximately 20% of AMI patients without known diabetes have evidence of the condition during hospitalization, only 40% receive a formal diagnosis and treatment [3,4]. These sub-optimal diagnosis and treatment rates likely reflect focus on AMI management over chronic conditions during hospitalization. However, diabetes can have an important influence on management of these patients, and, as such, measures to improve the detection and/or management of diabetes during the AMI hospitalization are needed.

It is well established that diabetes has significant impact on prognosis among AMI patients, with a 10-year mortality rate of 39% compared with 29% among non-diabetic patients [5]. In addition, prior studies have reported increased clinical severity of AMI in diabetic patients, including greater burden of atherosclerosis [6], angina [7] and left ventricular dysfunction [8], suggesting increased morbidity in this population post-AMI. Accordingly, timely diabetes diagnosis assists with risk stratification. In addition, diabetes influences many treatment decisions during the AMI hospitalization, including revascularization decisions [9] and use of therapies like angiotensin antagonists [10]. Early detection also allows for early initiation of glucose-lowering therapies, diabetic education and more intensive outpatient follow-up for diabetic care. Furthermore, therapy initiation during hospitalization – a time period during which patients are a ‘captive audience’ – has been shown to promote outpatient medication adherence and improved long-term glycemic control [11]. Although improved long-term glycemic control has not yet been shown to directly translate into improved cardiovascular outcomes, it does reduce microvascular complications – a valuable goal in and of itself [12]. We believe that these results can be extrapolated to diabetic AMI patients, although further research is needed to test this hypothesis.

Although the European Society of Cardiology recommends diabetes screening of AMI patients [13] and a scientific statement from the American Heart Association suggests screening hyperglycemic MI patients for diabetes [14], definitive clinical guidelines for screening for diabetes during AMI hospitalization by other major professional societies, such as the American College of Cardiology, are currently unavailable. Traditional methods of diabetes diagnosis – fasting blood glucose assessment and oral glucose tolerance testing (OGTT) – are problematic in AMI. The transient stress-induced glycemic dysregulation that occurs during AMI can affect the reliability of blood glucose measurements, irrespective of their timing. Although OGTT has excellent reliability for detecting diabetes, it is time consuming and requires multiple glucose measurements, which is difficult given the short stays that are typical of most AMI hospitalizations. Ideally, OGTT should be done several days after the AMI to avoid transient glucose abnormalities [15].

One potential solution for diabetes screening in AMI patients is universal HbA1c testing. HbA1c is a widely available and easily performed test that is less susceptible to the short-term glucose derangements that accompany AMI. In fact, a recent scientific statement released by the American Heart Association suggests screening for diabetes using HbA1c in all AMI patients with admission hyperglycemia [14]. Because of its ease of administration and potential for accuracy, we believe that HbA1c testing may represent the optimal screening strategy in all hospitalized AMI patients. In addition to identifying patients with previously undiagnosed diabetes, HbA1c can also identify patients with pre-diabetes – a group that represents nearly one-third of AMI patients who may benefit from nutritional counseling and lifestyle changes [16]. Accordingly, we believe that the best modality for diagnosing diabetes in the setting of an AMI during hospitalization is HbA1c measurement. Patients whose HbA1c value is >6.5% should have the value confirmed with a repeat test at or shortly following discharge. Following discharge after an AMI, either OGTT or HbA1c can be performed. OGTT has a higher sensitivity in diagnosis of AMI, but is more logistically complicated than HbA1c measurement. Currently, there is no consensus on using one test over the other.

Beyond the value of using HbA1c to screen for glycemic abnormalities among patients without known diabetes, it is also important for guiding optimal glucose-lowering treatment of patients with known diabetes. Among AMI patients with known diabetes, assessment of HbA1c levels is currently recommended by the American Diabetes Association [12]. However, studies have shown that over 30% of diabetic patients do not receive HbA1c testing during AMI hospitalization [1,2]. This is a particularly relevant finding as patients with known diabetes who had HbA1c tested during their AMI hospitalization were far more likely to have their glucose-lowering medications intensified at discharge, thereby increasing the likelihood of improved long-term glycemic control [1]. These findings further support the concept that chronic diabetes management is often relegated to the outpatient setting, where it frequently remains unaddressed [17]. Future studies should evaluate both the clinical and economic impact of HbA1c screening during AMI in improving diabetic care and reducing its attendant adverse outcomes.

At present, optimal glucose target in diabetic AMI patients during hospitalization is unknown. Major society guidelines recommend maintaining blood glucose levels below 180 mg/dl in hospitalized AMI patients, while avoiding hypoglycemic episodes. These recommendations arise from randomized trials that have not shown a benefit with intensive glycemic therapy (defined as blood glucose <110 mg/dl) compared with standard glucose values (blood glucose <180 mg/dl). However, these studies have been small, with some failing to attain glucose values that would be considered ‘intensive’. Hence, future studies should address this important question to provide better guidance for physicians treating diabetic AMI patients. Nonetheless, apart from glycemic therapy, appropriate care provided to diabetic AMI patients should also include multifactorial risk reduction involving optimal blood pressure and lipid lowering therapies. This frequently includes use of angiotensin antagonists for hypertension and high-dose, high-intensity statins.

Given the importance of optimal diabetes recognition and management among AMI patients, efforts are needed to improve its execution in current clinical care. A variety of health care delivery tools can be deployed to achieve this. One example is use of clinical information systems. Automated reminders for assessing HbA1c could be generated for all hospitalized patients with elevated cardiac enzymes. Not only would such universal screening increase detection rates of previously unrecognized diabetes, it would also ensure assessment of glycemic control in patients with known diabetes. Furthermore, electronic information systems can provide clinical decision support by guiding physicians in the best way to initiate diabetes management. In addition to clinical reminders, audit and feedback programs for physicians could be implemented to assist with their efforts to screen AMI patients for diabetes.

Finally, major professional societies in cardiology and diabetes need to provide explicit recommendations on diabetes screening and therapy to improve the management of AMI patients with diabetes. Although the current US AMI guidelines and performance measures focus on use of evidence-based medications and secondary prevention strategies, there are no specific recommendations on diabetes screening or management [18]. Such recommendations in the form of guidelines and performance measures would have a significant influence on physician performance. Performance measures for diabetes management can serve as vehicles to accelerate the appropriate translation of scientific evidence into clinical practice, measure the quality of care provided and identify opportunities for improvement. These societies should consider diabetes assessment with HbA1c among AMI patients and among those with high HbA1c levels, diabetes education, in-hospital initiation or titration of glucose-lowering therapies before discharge and arrangement for outpatient follow-up in diabetic clinics. Through these and the other efforts listed above, the rates of undiagnosed and undertreated diabetes among AMI patients can be significantly reduced, ultimately leading to an improvement in both diabetic and cardiovascular outcomes.

Financial & competing interests disclosure

TM Maddox is supported with a VA Health Services Research and Development career development award. The other authors report no relevant disclosures. No external sponsors were involved in any part of this study. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.