The impact of diabetes on one-year health status outcomes following acute coronary syndromes

Acute coronary syndromes (ACS), including unstable angina, non-ST-elevation myocardial infarction (NSTEMI) and ST-elevation myocardial infarction (STEMI), affect approximately one million Americans each year [1]. As improved therapies have resulted in lower ACS mortality, patient health status has been increasingly recognized as a critical outcome, with greater emphasis being placed on the therapeutic goals of reducing symptoms, improving function, and optimizing health related quality of life (HRQoL). In addition to being an important patient-centered outcome, health status predicts subsequent mortality and morbidity in cardiac populations [2–4]. To date, few studies have examined the factors associated with health status outcomes following ACS.

Diabetes is an established and highly prevalent risk factor for cardiac morbidity and mortality, and therefore is an important comorbidity in patients with ischemic heart disease (IHD). However, little is known about the association between diabetes and health status outcomes after ACS. It is important to assess the impact of diabetes on health status outcomes in cardiovascular populations. First, there is a general perception that patients with diabetes are less likely to experience angina than patients without diabetes[5]. Yet, patients with diabetes often have a high burden of coronary atherosclerosis[6], suggesting increased potential for ischemic symptoms. To date, the severity of angina following ACS in patients with diabetes versus those without has not been described. Second, although diabetes has been associated with worse overall HRQoL in general populations [7, 8], the impact of diabetes on cardiac-specific health status after ACS compared to traditional cardiac factors (e.g. ejection fraction) is not known.

The goal of this study was to prospectively evaluate the association between diabetes and cardiac-specific health status (angina frequency, physical limitation and HRQoL) in patients one year after an ACS, independent of other known predictors. The results of this study are intended to inform clinicians about the importance of diabetes as a risk factor for health status outcomes after ACS, and potentially lead to interventions to improve health status outcomes in patients with diabetes.

Of the 1,199 patients in the cohort, 326 patients (27%) had diabetes. Baseline characteristics of patients with and without diabetes are listed in Table 1. Compared to those without diabetes, patients with diabetes were more likely to be female, non-Caucasian and to have a history of MI, CABG and heart failure. Hyperlipidemia, hypertension, stroke, peripheral vascular disease, renal dysfunction and chronic lung disease were also more common among those with diabetes. There was no significant difference among those with and without diabetes in left ventricular function.

ACS characteristics and treatments are presented in Table 2. Patients with diabetes were more likely to present with unstable angina (52% vs. 40%; p < 0.001) and less likely to present with STEMI (20% vs. 31%; p < 0.001). Patients with diabetes were less likely to have coronary angiography performed during index ACS hospitalization (68% vs. 82%; p < 0.001). The proportion of participants with and without diabetes receiving angiography was significantly different among those with unstable angina (65% vs. 74%), but similar among those with NSTEMI (84% vs. 88%) and STEMI (95% vs. 96%). Among patients who underwent coronary angiography, those with diabetes were more likely to have 3-vessel coronary artery disease (58% vs. 45%; p = 0.003). In the year following the index ACS hospitalization, there was no significant difference between those with and without diabetes in the number of percutaneous interventions performed (16.5% vs. 20%; p = 0.82). There was a trend toward more CABG procedures (12% vs. 6%; p = 0.06) in patients with diabetes.

At the time of the index hospitalization, patients with diabetes had lower mean SAQ scores for angina frequency (72 vs 77; p = 0.001) and physical limitation (78 vs 87; p < 0.001), but no significant difference in HRQoL scores (49 vs 50; p = 0.25). One year following ACS, patients with diabetes had significantly lower SAQ angina frequency (85 vs. 91; p = 0.0006), physical limitation (82 vs. 92; p < 0.0001) and HRQoL (76 vs. 84; p < 0.0001) scores compared to patients without diabetes.

In unadjusted analysis, patients with diabetes, when compared to those without diabetes, were significantly more likely to have angina of any degree (37% vs. 27%; p = 0.003), cardiac-related physical limitation (35% vs. 27%; p < 0.001), and HRQoL deficits (53% versus 42%; p = 0.035, Figure 1) one year following ACS. Furthermore, patients with diabetes had worse overall physical health status (mean PCS scores 37 vs. 44; p < 0.0001), and mental health status (mean MCS scores 52 vs. 55; p = 0.0013) compared to patients without diabetes.

In multivariable analyses, diabetes remained significantly associated with an increased presence of angina, cardiac-related physical limitation and HRQoL deficits one year after ACS. After adjustment for demographic, cardiac, non-cardiac, and treatment variables, patients with diabetes had significantly higher odds of having angina (OR 1.36; 95% CI 1.01–1.38), cardiac-related physical limitation (OR 1.94; 95% CI 1.57–3.24) and HRQoL deficits (OR 1.43; 95% CI 1.01–2.04) at one year compared to patients without diabetes.

In multivariable models, additional predictors of having angina were age (OR 1.08; 95% CI 1.00–1.15 per decade decrement), history of MI (OR 1.60; 95% CI 1.21–2.13), hypercholesterolemia (OR 0.75; 95% CI 0.57–0.98), history of alcohol or substance abuse (OR 2.0; 95% CI 1.20–3.38), chronic lung disease (OR 2.74; 95% CI 1.76–4.26), and history of peripheral vascular disease (OR 1.75; 95% CI 1.02–3.02). Additional predictors of physical limitation were female gender (OR 1.65; 95% CI 1.19–2.28), Caucasian race (OR 0.61; 95% CI 0.40–0.93), history of MI (OR 1.61; 95% CI 1.15–2.24), history of alcohol or substance abuse (OR 2.28; 95% CI 1.27–4.09), discharge prescription of an angiotensin converting enzyme inhibitor or an angiotensin receptor blocker (OR 1.52; 95% CI 1.07–2.16) and chronic lung disease (OR 2.15; 95% CI 1.15–4.00). The additional predictors of HRQoL deficits were age (OR 1.28; 95% CI 1.14–1.45 per decade decrement), ejection fraction (OR 1.12 (1.01–1.27 per 10% decrease), acute thrombolytic therapy (OR 1.92; 95% CI 1.01–3.66), and chronic lung disease (OR 2.36; 95% CI 1.33–4.20).

In multivariable linear regression analyses, diabetes remained significantly associated with worse overall physical function as measured by the SF-12. PCS scores were, on average, three points lower for those with diabetes compared to those without diabetes (β = 2.85 +/- 1.12, p = 0.01). Diabetes was not independently associated with differences in MCS scores (β = 1.26 +/- 0.93, p = 0.18).

The goal of this study was to examine the association between diabetes and patients' health status outcomes one year after an ACS. We found that patients with ACS and coexisting diabetes were more likely to present with unstable angina, and were less likely to undergo coronary angiography during ACS hospitalization. At one year following ACS, patients with diabetes were significantly more likely to have angina, cardiac-related physical limitation and HRQoL deficits compared to patients without diabetes. These associations remained significant after adjustment for a wide array of demographic, disease severity, comorbid condition, and treatment variables. Diabetes was also negatively associated with overall physical health status, but was not significantly associated with overall mental health status.

It may be surprising that patients with diabetes were more likely to have angina one-year post ACS. Previous studies have shown that patients with diabetes have a higher rate of silent myocardial ischemia [15–17], although this has not been a consistent finding [18, 19]. However, it is not appropriate to conclude that diabetics with coronary artery disease necessarily experience less angina. To our knowledge, no previous study has directly evaluated the angina burden of patients with diabetes after ACS. The results of this study support that patients with diabetes are more likely than non-diabetics to have angina one year after ACS.

The results of this study support the potential for more aggressive identification and treatment of angina in patients with diabetes following ACS in order to improve outcomes. Angina decreases patients' quality of life, is associated with worse patient satisfaction, and increases the risk for subsequent ACS and death[2, 3, 20–22]. There are a range of therapies for angina, including pharmacologic management, coronary revascularization, and other interventions such as enhanced external counterpulsation [23]. However, angina is often inadequately recognized and treated in routine clinical practice, including under-use and under-dosing of beta-blockers, long acting nitrates, and calcium channel blockers [24–27]. Future studies should assess whether the post-ACS outcomes of patients with diabetes can be improved by employing strategies to enhance post-discharge surveillance and treatment of angina such as disease management or cardiac rehabilitation. While evidence supports these interventions in general cardiac populations, [28, 29] studies have not specifically targeted patients with diabetes.

We also found that patients with diabetes and unstable angina were less likely to receive coronary angiography, raising the possibility that greater use of an early invasive strategy in diabetic patients could improve outcomes. The discrepancy in angiography rates did not appear to be wholly or directly attributable to clinician concerns over impaired renal function given that 94% of diabetic patients had admission creatinine levels <2.0 μg/ml. Current American Heart Association/American College of Cardiology guidelines for ACS state that in the absence of high-risk indicators such as new ST-segment depression, symptoms of heart failure or high-risk findings on non-invasive testing, an early conservative or invasive approach can be taken with similar morbidity and mortality outcomes [30]. Given the results of this study, in which diabetics are more prone to persistent angina and diminished quality of life after an ACS, future clinical trials should assess whether an early invasive approach in patients with diabetes presenting with ACS, even in the absence of other high-risk indicators, can improve health status outcomes.

The results of this study expand the previous literature on diabetes and health status in cardiac populations. Prior studies have suggested that diabetes is associated with greater overall physical limitation and worse quality of life following CABG surgery or PCI [31, 32]. Previous studies on the association between diabetes and post-ACS health status outcomes have been conflicting [33, 34]. Importantly, studies to date have not included evaluation of cardiac-specific health status, and the absence of baseline measures of health status in most of these studies precluded the ability to adjust for baseline health status differences. This study therefore adds to our knowledge by prospectively evaluating the influence of diabetes on the health status of patients with ACS, using both cardiac-specific and generic measures of health status in a large unselected population of patients across the spectrum of ACS, including adjustment for baseline health status.

Potential limitations to this study should be considered. First, missing follow-up health status assessments can bias quality of life studies [14]. However, there was no difference in the rate of missing surveys between patients with and without diabetes. Furthermore, statistical methods to assess for the impact of missing values did not suggest significant bias in the multivariable models. Second, as an observational study, residual unmeasured confounding may be present. However, a wide range of measured variables were considered in multivariable regression models. One potential confounder, severity of CAD, is difficult to quantify in patients with diabetes, who are known to be at risk for more diffuse CAD. In this study, severity of CAD was adjusted for as possible using the combination of 1, 2, or 3-vessel coronary artery disease and left ventricular ejection fraction. Finally, because the focus of the parent study was ACS and not diabetes per se, details on diabetic microvascular complications were not collected and could not be specifically considered in multivariable models. However, renal function, peripheral vascular disease and baseline health status were included, which are likely to reflect pre-existing diabetic complications influencing health status.

In conclusion, we found that diabetes was associated with more angina, greater cardiac-specific physical limitation, and worse HRQoL one year after an ACS. Future studies should evaluate interventions to reduce angina burden and improve the health status of patients with diabetes following ACS. Potential strategies include an early invasive approach for unstable angina in patients with diabetes and more aggressive surveillance and therapy for angina in the post-ACS period. It is hoped that such interventions will optimize the patient-centered outcomes of symptom burden, physical function, and quality of life for the growing population of patients with diabetes and CAD.