Author + information
- Barry Greenberg, MD* ()
- ↵*Reprint requests and correspondence:
Dr. Barry Greenberg, Advanced Heart Failure Treatment Program, University of California, San Diego, 200 West Arbor Street, San Diego, California 92103-8411
Heart failure (HF) prevalence is increasing in the U.S. and throughout the rest of the world (1,2). Although therapeutic advances have improved outcomes over the past several decades, HF patients continue to experience diminished quality of life as well as a substantial reduction in their quantity of life. The cost of providing medical care (which in the U.S. is driven largely by hospitalizations) for an expanding HF population imposes an increasingly heavy burden on health care systems throughout the world. Although novel experimental drugs and devices hold promise for improving the clinical course of HF patients, the overall impact of these therapies is likely to be modest at best. Clearly, new strategies are needed for dealing with this growing pandemic.
HF is the consequence of a variety of diseases that affect the cardiovascular system. Diabetes is a well-recognized risk factor for the development of HF, and in this issue of the Journal, From et al. (3), from the Mayo Clinic in Rochester, Minnesota, describe the progression from pre-clinical diastolic dysfunction to overt HF in a large cohort of diabetic patients. To accomplish this, they reviewed the medical records of Olmsted County residents over a slightly less than 6-year period beginning in 2001. They identified 12,014 diabetic patients, of whom 2,770 had tissue Doppler echocardiographic studies performed, and included 1,760 patients who were without either existing HF or left-sided valvular lesions for subsequent follow-up analysis. The investigators wisely excluded patients who were diagnosed with HF within 6 months of the echocardiogram based on the premise that the test likely resulted in the diagnosis/recognition of HF in these cases. They found that 23% of the study population had evidence of “pre-clinical diastolic dysfunction” defined as an E/e′ (passive transmitral left ventricular inflow velocity to tissue Doppler imaging velocity of the medial mitral annulus during passive filling ratio) >15. These patients tended to be older, less likely to be male, and more likely to have hypertension and/or coronary disease than diabetic patients without evidence of diastolic dysfunction. Not surprisingly, diabetic patients with diastolic dysfunction had larger left atrial volumes and greater left ventricular (LV) muscle mass. Over 5 years of follow-up, the cumulative probability of HF was 36.9% in the patients with pre-clinical diastolic dysfunction compared with only 16.8% of patients without diastolic dysfunction. Multivariate analysis confirmed that even after adjustment for age, sex, hypertension, coronary artery disease, left atrial size, and LV muscle mass, diastolic dysfunction remained an independent predictor of heart failure risk and that every unit of increase in E/e′ above 15 was associated with a 3% greater likelihood of the future development of heart failure. Secondary analysis indicated that diastolic dysfunction was an independent risk factor for death but not for atrial fibrillation.
Although the study was generally well done, there are some concerns about the accuracy of diagnostic coding in defining HF and (to a lesser degree) diabetes. The investigators provide no information about the rate of progression or incidence of diastolic dysfunction in the population nor do they indicate whether systolic abnormalities appeared during the course of follow-up. Although the Olmsted County population may be representative of white Americans, it contains few Hispanics, African Americans, or Asians, so that the ability to generalize results to more diverse populations is limited. There are also concerns about inherent bias related to the clinical decision of obtaining an echocardiogram in these patients in the first place. Nonetheless, the finding of an association between the presence of diastolic abnormalities in a diabetic population and progression to overt HF is almost certainly correct, and it has important implications.
What do these results tell us about the presence of pre-clinical diastolic dysfunction in diabetic patients and subsequent risk? The report confirms previous studies showing that asymptomatic diabetic patients often manifest abnormalities in LV structure as well as abnormalities in both diastolic and systolic function (4,5). Of greater importance is that these results alert us to the fact that the presence of an increased E/e′ predicts in a linear fashion the future risk of the development of overt HF. Using American College of Cardiology/American Heart Association criteria, the patients from Olmsted County with pre-clinical diastolic dysfunction should probably be classified as having stage B HF, with their condition considered analogous to that of patients with asymptomatic LV systolic dysfunction. Although the present report does not define the clinical scenarios that lead to the diagnosis of HF, it is worth noting that the diabetic patients with pre-clinical diastolic dysfunction from Olmsted County developed HF and/or who died at a rate that was similar to that reported for patients with asymptomatic LV systolic dysfunction in the prevention arm of the SOLVD (Studies of Left Ventricular Dysfunction) trial (6). So much for the benign nature of HF with preserved ejection fraction, even in its pre-clinical phase.
The molecular and signaling mechanisms through which diastolic dysfunction develops and then progresses to overt HF in diabetic patients have not yet been fully elucidated. Identification of the pathways involved, however, is fundamental for providing direction for novel therapies that can prevent or delay this process. Was progression to HF due to increased deposition of fibrous tissue and/or alterations in the structure of the interstitial matrix, changes in myocardial energetics, abnormalities in cardiomyocyte calcium handling, myocardial ischemia due to microvascular disease, or some yet unidentified process that was taking place in the hearts of these diabetic patients? What was the role of advanced glycation end products and increased oxidative stress? Why are diastolic abnormalities (and HF, for that matter) more common in female diabetic patients than in their male counterparts? Animal models and methods for addressing at least some of these questions in human patients are currently available, and it is hoped that the results from this study will further stimulate researchers to define the underlying causes responsible for the progression to HF in diabetic patients. The importance of better defining these mechanisms cannot, in my opinion, be overstated. Witness only the disappointing results of recent clinical trials conducted in patients with HF and preserved ejection fraction (7,8) to understand why targeting therapies to clearly defined pathologic processes is essential if we hope to succeed in the future.
Absent definitive information about underlying pathophysiology, do the findings of the present study have implications for the current management of diabetic patients? For instance, should all diabetic patients undergo screening echocardiograms to identify those who are at increased risk of the development of HF? Given the high and growing prevalence of diabetes in the U.S., this approach is not likely to be cost-effective. The presence of pre-clinical systolic and diastolic dysfunction by tissue Doppler imaging studies, however, has been associated with elevated natriuretic peptide levels (9,10), and measurement of natriuretic peptides in blood seems to be an effective tool for screening in a population that is at high risk of the development of heart failure (11) Moreover, measurement of natriuretic peptides can be used to identify American College of Cardiology/American Heart Association stage A and B patients who are at increased risk of cardiovascular events (12). Biomarkers that detect interstitial matrix turnover in patients also seem to be useful in identifying patients with diastolic dysfunction who are at risk (13). The cost of these tests is relatively low, and one could envision biomarker screening of diabetic patients to decide whether to proceed with an echocardiogram. Carefully designed algorithms using 1 or more of these biomarkers to determine the likelihood of pre-clinical diastolic dysfunction in asymptomatic diabetic patients will need to be designed and validated in diverse populations for this purpose.
Assuming that diabetic patients with pre-clinical diastolic dysfunction have been identified, what should be the next step? Specifically, is there something that can be done to reduce HF risk in this population? Whether tight control of diabetes will be effective in this regard is uncertain, and recent trials assessing the effects of diabetic therapy on clinical outcomes suggest that this may not be the case (14,15). Given the strong association between diabetes and other HF risk factors such as hypertension and coronary artery disease (which was confirmed in the present study), I believe that identifying pre-clinical diastolic dysfunction in diabetic patients should provide the impetus for greatly intensifying efforts to aggressively treat the myriad of risk factors that are found in this population. In particular, treatment of dyslipidemia and hypertension to achieve published target goals should be emphasized. The approach should also ideally include strategies to modify exercise and dietary patterns. Finally, the diabetic population with pre-clinical diastolic dysfunction provides an excellent focus for clinical trials evaluating novel treatment approaches, particularly if they target underlying causes of cardiac dysfunction. Given the present information about HF risk in diabetic patients and the medical and societal implications of the ongoing HF pandemic, can we afford not to aggressively treat these patients or to conduct studies that define the best approaches for prevention in this population? I think not. If we fail in this regard, the cost to our patients and to society will be high and the effects long-lasting.
↵* Editorials published in the Journal of the American College of Cardiologyreflect the views of the authors and do not necessarily represent the views of JACCor the American College of Cardiology.
- American College of Cardiology Foundation
- Rosamond W.,
- Flegal K.,
- Friday G.,
- et al.
- From A.M.,
- Scott C.G.,
- Chen H.H.
- Lubien E.,
- DeMaria A.,
- Krishnaswamy P.,
- et al.
- Daniels L.,
- Clopton P.,
- Jiang K.,
- Greenberg B.,
- Maisel A.
- Martos R.,
- Baugh J.,
- Ledwidge M.,
- et al.