Author + information
- Jalal K. Ghali, MD∗ ()
- Department of Internal Medicine, Division of Cardiology, Mercer University School of Medicine, Macon, Georgia
- ↵∗Reprint requests and correspondence:
Dr. Jalal K. Ghali, Department of Internal Medicine, Division of Cardiology, Mercer University School of Medicine, 707 Pine Street, Macon, Georgia 31201.
The past 2 decades have witnessed a rising interest in biomarkers, as reflected by close to 6,500 publications in the past decade alone (1). Several definitions of biomarkers have been suggested; I prefer to use the following: “A characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention” (2).
A complex process that requires multiple phases of investigation is required to validate the clinical use of a biomarker (3). These include a hypothesis that links the biomarker with the disease, assessment of its sensitivity and specificity, and correlation of the levels of the biomarkers with disease severity and or progression. If all of the above is fulfilled, then the logical next phase would be a clinical assessment of targeting the pathophysiological mechanism(s) leading to the rise of the biomarker.
There are a variety of ways that biomarkers could provide useful information and assist clinicians in managing patients with heart failure (HF). These include a better understanding of the pathophysiology, help in establishing or confirming the diagnosis, risk stratification, monitoring of progression, assessment of the impact of therapy, identification of specific markers of disease progression that demonstrate the potential to be important targets of therapy, and finally, management guidance (3).
It is striking that the vast majority of studies, reviews, and recommendations regarding biomarkers in HF have addressed predominantly or exclusively the diagnostic or prognostic role of biomarkers and their potential impact on guiding management (1). Less attention has been given to the potential role of biomarkers in understanding the pathophysiology of HF, and their usefulness in this field has not been emphasized or fully explored.
Albuminuria is a unique biomarker, because contrary to many other biomarkers that reflect a single biological process, albuminuria is thought to reflect multiple pathophysiological processes (4), including endothelial dysfunction, diffuse vascular damage and increased transvascular leakage of macromolecules, systemic inflammation, activation of the renin-angiotensin aldosterone system, impaired oxidative metabolism, insulin resistance, renal damage related to altered glomerular hemodynamics, and abnormal tubular function. Albuminuria has been associated with homocysteine levels, plasminogen activator inhibitor-1 activity, fibrinogen, the presence of comorbidities, and atherosclerosis.
The association of albuminuria with these numerous processes underlines its biological complexity and could explain its profound prognostic role. The association of albuminuria with cardiovascular disease is well established in the general population and in patients with diabetes mellitus, hypertension, or kidney disease. What is striking is that this association is continuous across the entire spectrum of albumin in the urine (5). Thus, the classic classification of albuminuria as normal (<30 mg/24 h), microalbuminuria (defined as 30 to 300 mg/24 h), and macroalbuminuria (>300 mg/24 h) is obsolete. Any degree of albuminuria is a risk factor for cardiovascular disease, and this relationship is graded and continuous.
The association of albuminuria with left ventricular hypertrophy (LVH) and increased left ventricular (LV) mass has been well documented in the general population (6) in patients with hypertension (7–11) or diabetes (12,13). In fact, albuminuria has been associated with relative wall thickness and with greater concentric remodeling, LV end-systolic stiffness, and worse diastolic function (14–16). Previous studies in patients with diabetes have strongly suggested the association of albuminuria with systolic and diastolic LV function (12–14).
In the HyperGEN (Hypertension Genetic Epidemiology Network) study (15), albuminuria was associated with increased LV mass in normotensive people with preserved EF, normal LV dimensions, and fractional shortening, which suggests that albuminuria is associated with early structural changes in the myocardium. In a recent elegant report from the HyperGEN study (16), assessment of myocardial function in 1,894 participants with normal LV ejection fraction and wall motion demonstrated an inverse relationship between albuminuria and longitudinal strain in all participants and between albuminuria and E/e′ ratio in those with hypertension or LVH, which indicates a continuous relationship between albuminuria and subclinical myocardial dysfunction.
The association of albuminuria with the development of HF is firmly established by several cohort studies in the general population, as well as in patients with established cardiovascular risk factors, including diabetes mellitus, hypertension, renal disease, and LVH. It is interesting to note that in the community-based PREVEND study (Prevention of Renal and Vascular End-stage Disease), urinary albuminuria was associated with the development of HF with preserved (HFpEF) but not reduced LV ejection fraction (17).
In this issue of JACC: Heart Failure, Katz et al. (18) report on data collected prospectively in 144 patients with HFpEF who underwent tissue Doppler imaging and right ventricular (RV) evaluation and in whom urinary albumin excretion was measured. The authors found that higher urinary albumin was associated with higher LV mass and lower global longitudinal strain. More importantly, the authors reported the novel finding that a higher concentration of albumin in the urine was associated with greater RV wall thickness and worse systolic function as measured by RV fractional area change.
Before highlighting the novel findings, a few limitations need to be mentioned. The generalizability of the study could have been better achieved by consecutive enrollment. Loop diuretic agents were used by only 43% of patients in quartile 1 and 56% of patients in quartile 2, a finding that raises concern about the accuracy of diagnosis of HF in some patients. Finally, it would have been desirable to confirm the hemodynamic findings in at least a subset of the patients. These limitations, however, do not undermine the importance of the study. The authors cleverly hypothesized that albuminuria would be associated with measures of cardiac structure and function, including the RV. Contrary to the multitude of previous studies performed to delineate the diagnostic and prognostic roles of biomarkers, the authors chose to use albuminuria to further our understanding of the pathophysiology of HFpEF, a novel approach in the use of biomarkers. Their findings challenge the traditional view that RV remodeling and dysfunction are simply secondary to pulmonary venous hypertension. The fact that the association between albuminuria and RV size and function parameters persisted after adjustment for estimated pulmonary artery systolic pressures strongly suggests a pathophysiological link between 1 or more of the multiple mechanisms of albuminuria and RV remodeling. Building on previous observations (12–16), the findings of this study lend support to the designation of albuminuria as a marker for myocardial involvement. Therefore, myocardial involvement should be added to the list of biological processes that albuminuria may reflect.
Moreover, there is compelling evidence to suggest that albuminuria should be considered a therapeutic target. In 8,206 hypertensive patients with LVH randomized to losartan or atenolol (19) and 1,513 patients with diabetic nephropathy randomized to losartan or placebo (20), changes in albuminuria were predictive of cardiovascular outcomes. Admittedly, the few encouraging reports on the reduction of cardiovascular risk associated with reduction of albuminuria do not provide convincing evidence to influence clinical practice; these data, however, combined with findings from 23,480 high-risk patients randomized to telmisartan, ramipril, or both that indicate that changes in albuminuria were predictive of mortality and cardiovascular and renal outcomes independent of baseline albuminuria (21), form a solid base on which a case could be built for conducting clinical trials to assess the role of albuminuria in selecting responders.
It is reasonable to speculate that at least for interventions proven to lower albuminuria, the presence of albuminuria at baseline and changes in albuminuria may influence outcomes. Therefore, in future HF trials, stratification of randomization by albuminuria and monitoring of albuminuria should be implemented.
The prevalence of HFpEF was recognized more than 2 decades ago (22). Only in the past decade, however, have intense efforts been devoted to identifying treatment that would improve outcomes. Likewise, the first evidence linking albuminuria to LV function was identified more than a decade ago (14); one would hope that the findings of Katz et al (18), which further our understanding of HFpEF, will not have to wait long before they find a practical application for albuminuria in HF clinical trials.
↵∗ Editorials published in JACC: Heart Failure reflect the views of the authors and do not necessarily represent the views of JACC: Heart Failure or the American College of Cardiology.
Dr. Ghali has reported that he has no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Ahmad T.,
- Fiuzat M.,
- et al.
- Mostafa A.,
- El-Haddad M.,
- Ghali J.K.
- Stehouwer C.D.,
- Smulders Y.M.
- Leib W.,
- Mayer B.,
- Stritzke J.,
- et al.
- Wachtell K.,
- Palmieri V.,
- Olsen M.H.,
- et al.
- Liu J.E.,
- Robbins D.C.,
- Palmieri V.,
- et al.
- Katz D.H.,
- Selvaraj S.,
- Aguilar F.G.,
- et al.
- Brouwers F.P.,
- de Boer R.A.,
- van der Harst P.
- Katz D.H.,
- Burns J.A.,
- Aguilar F.G.,
- Beussink L.,
- Shah S.J.
- De Zeeuw D.,
- Remuzzi G.,
- Parvin H.H.,
- et al.
- Schmieder R.E.,
- Mann J.F.E.,
- Schumacher H.,
- et al.,
- on behalf of the ONTARGET Investigators