Another reason for measuring microalbuminuria: its possible association with heart failure

Background

Urinary albumin excretion is a predictor of renal, as well as cardiovascular morbidity and mortality in patients with 1) hypertension, 2) diabetes and in 3) the general population (1).
While an albumin/creatinine ratio of 30 to 300 µg albumin/mg defines microalbuminuria (2) any degree of albuminuria, according the Heart Outcomes Prevention Evaluation (HOPE) study, is a risk factor for cardiovascular events – regardless of whether diabetes is present or not (3).
However, in a study presented by Haller et al. on the awareness and behaviour of European physicians in relation to the value of microalbuminuria and its correlation with cardiovascular risk factors and organ damage, only 42.5% of diabetologists and 35.5% of cardiologists reported measuring microalbuminuria in their type 2 diabetic and hypertensive patients (4).
Moreover, heart failure is a clinical syndrome resulting from a variety of causes, including coronary heart disease hypertension, and diabetes. Microalbuminuria and heart failure have usually been mutually associated (1, 5-9). 

 1 – Microalbuminuria in nondiabetic and nonhypertensive subjects

Microalbuminuria can be present in nondiabetic and nonhypertensive subjects and is associated with cardiovascular disease. 
Hillege et al. reported the results of a cross-sectional cohort study conducted in 2001, in which albuminuria was determined in 40,619 subjects between the ages of 28 to 75 years in the city of Groningen in the Netherlands.
The study found that 1) microalbuminuria (defined as 20-200 mg/L) was present in 7.2% (n=2,918) of subjects and independently associated with cardiovascular risk factors and cardiovascular morbidity - a majority of subjects (74.9%) had no reported diabetes or hypertension 2) macroalbuminuria (albumin concentration >200 mg/L) was present in 0.7% (n=282)(10).  

Romundstad et al. reported the results of a cross-sectional study (HUNT, 1995-97), investigating the connection between microalbuminuria and known cardiovascular risk factors/markers and disease in 9,598 randomly selected nondiabetic/nonhypertensive subjects (≥20 years). Of these, 2,175 (44.1%) had neither diabetes nor treated hypertension. When adjusted for other cardiovascular risk factors/markers, microalbuminuria was associated with increased blood pressure (BP) - systolic and diastolic - and pulse pressure in both genders; and with age and coronary heart disease in men (11).  

2 – Microalbuminuria in diabetes, hypertension and coronary heart disease

Microalbuminuria is an early sign of diabetic nephropathy and a risk marker for premature death from cardiovascular disease in patients with Type 1 diabetes.
Hovind et al. prospectively evaluated 277 newly diagnosed Type 1 diabetic patients for a median follow-up of 18 years. The reported cumulative incidence of persistent microalbuminuria and persistent macroalbuminuria was respectively 33.6% and 14.6%. Significant predictors for the development of persistent microalbuminuria were a 10-fold increase in microalbuminuria, being male, a 10 mmHg increase in arterial blood pressure (BP) and a 1% increase in HbA1c (12).
Amin et al. prospectively evaluated 527 Type 1 diabetic patients who had been diagnosed at mean age of 8.8 years for 9.8 years. Authors reported a cumulative prevalence of microalbuminuria at 25.7% after 10 years of diabetes and 50.7% after 19 years. The only modifiable adjusted predictor for microalbuminuria was high HbA1c concentrations. Microalbuminuria was reported as persistent in 48% and cumulative prevalence of progression from microalbuminuria to macroalbuminuria was 13.9%. Modifiable predictors of macroalbuminuria were higher HbA1c levels and both persistent and intermittent microalbuminuria (13).
Numerous follow-up studies involving patients with Type 2 diabetes have also demonstrated the relationship of both microalbuminuria and macroalbuminuria with progression to end-stage renal disease and increased risk of death, which is largely due to cardiovascular disease.  

Valmadrid et al. conducted a prospective cohort study of 840 individuals with Type 2 diabetes, 50.8% of whom had normoalbuminuria, while 24.8% had microalbuminuria and 20.5% had macroalbuminuria. After a median follow-up of 12 years, those with microalbuminuria and macroalbuminuria independently had significant higher risks of cardiovascular mortality (14). Nevertheless, further studies show that regression of microalbuminuria to normoalbuminuria occurs in 31-58% of adults with microalbuminuria 6-8 years after the onset of microalbuminuria and is connected to modifiable factors (such as glycaemic control, BP, lipid concentrations), use of renin-angiotensin aldosterone system (RAAS) blockade drugs and statins and indicates that elevated microalbuminuria does not imply progressive nephropathy inexorably (12,14-18). 

The association of microalbuminuria with essential hypertension has been reported in several studies. The LIFE Study investigators have demonstrated, in a prospective study of 8206 patients with stage II or III essential hypertension, that increased urine albumine/creatinine ratio resulted in increasing risk for cardiovascular morbidity and mortality among hypertensive patients with left ventricular hypertrophy. Risk was reported to increase at much lower levels than what had previously been reported among diabetic patients. They found no thresholds or plateaus. They suggested instead that detecting microalbuminuria would help clinicians decide when to initiate antihypertensive therapy, since it represents target organ damage (19). The existence of microalbuminuria in patients with essential hypertension is a strong indicator of microvascular damage. Vascular endothelial growth factor is the most important regulator of pathological or physiological angiogenesis and it additionally leads to increased vascular permeability. Ebinç et al. have demonstrated that vascular endothelial growth factor levels are higher in essential hypertension in the presence of microalbuminuria, which may be important in the early diagnosis of vascular damage in essential hypertension. Vascular endothelial growth factor may also increase glomerular permeability and lead to microalbuminuria in essential hypertension (20).

Berton et al., prospectively studied 496 patients who had been admitted to hospital for suspected acute myocardial infarction in order to assess whether microalbuminuria increases during this event and also whether it predicts in-hospital mortality. Of these, 360 have shown evidence of acute myocardial infarction. Mortality rate progressively and significantly increased with increasing levels of microalbuminuria. In a hazard model, microalbuminuria was a better predictor of in-hospital mortality than Killip class or echocardiographic left ventricular ejection fraction (21).
In a prospective study of 2762 men and women 30-70 years of age, Klausen et al. reported that even at low levels (≥6.8 µg/min) microalbuminuria was strongly associated with increased and death, irrespective of age, gender, hypertension, diabetes, renal function and lipids (22). Rein et al., have investigated the connection of albuminuria with angiographically determined coronary atherosclerosis in 914 consecutive patients who had been referred for coronary angiography in order to evaluate established or suspected coronary heart disease. The prevalence of coronary stenosis of ≥ 50% has been reported as significantly greater in patients with albuminuria than in those with normoalbuminuria, in both patients with and without Type 2 diabetes, irrespective of conventional cardiovascular RF and the estimated glomerular filtration rate (eGFR) (23).

3 – Higher levels of microalbuminuria in heart failure patients than in the general population

Van de Wal et al. have evaluated 94 stable chronic heart failure patients, mean age 69 ± 12 years, all of whom had been receiving ACE inhibitors, for 3 months. Ischaemia was reported as the underlying cause of heart failure in 61 patients. Microalbuminuria was present in 32% of patients, which is significantly higher than in the general population; yet no association was found with either renal or neurohormonal parameters. Authors suggest that this could owe to endothelial dysfunction and vascular permeability (5). We compared the levels of microalbuminuria in 27 nonhypertensive, nondiabetic systolic HF patients, 74% of whom had been using ARBs, with 27 healthy controls, paired for age and sex. Sixteen patients (59.3%) had microalbuminuria, despite ARB use, 18.5% had macroalbuminuria and 22.2% normoalbuminuria. Urine albmumine to creatine ratio was significantly higher in systolic heart failure patients, irrespective of aetiology (Chagas disease, ischaemia, idiopathic, post-partum), as seen in Figure 1.

Figure 1 - Comparison of albumin creatinine ratio (ACR) and urinary albumin (Alb) values between systolic heart failure patients and control subjects. 

.
Figure legend: The values are expressed as µg albumin/mg creatinine. The symbol (*) indicates a statistically significant difference (P<0.05).
 

There was no significant difference regarding New York Heart Association functional class (6). Jackson et al. measured Urine albmumine to creatine ratio at baseline and during follow-up of 2,310 subjects who had been enrolled in the CHARM study; 30% were reported as having had microalbuminuria and 11% macroalbuminuria. 
The prevalence of increased microalbuminuria was similar in patients with reduced and preserved left ventricular ejection fraction and was associated with increased risk of death from cardiovascular disease or admission to hospital with worsening heart failure and death from any cause. This occured irrespectively of renal function, diabetes or hypertension. Treatment with candesartan did not reduced or prevented microalbuminuria (7). Masson et al., by using data from the GISSI-HF trial, evaluated microalbuminuria in 2,131 patients. Nineteen point nine % had microalbuminuria and 5.4% had macroalbuminuria. Irrespective of diabetes, essential hypertension or renal function, there was significant progressive increase in the adjusted rate of mortality associated with elevated microalbuminuria, which remained unaffected by n-3 polyunsaturated fatty acids or rosuvastatin (8).

Jackson et al. evaluated urine albumin to creatinine ratio in 190 patients randomised in the ALOFT trial6  56% had normal albumin excretions, 33% microalbuminuria and 11% macroalbuminuria. Patients with microalbuminuria had a greater prevalence of diabetes, and a lower eGFR. Of the non-diabetic patients, 28% had microalbuminuria and 7% macroalbuminuria. Microalbuminuria was reported as independently associated with HbA1c, N-terminal pro- brain natriuretic peptide (NT-pro-BNP) and left ventricular diastolic dimension. Microalbuminuria was not associated with markers of inflammation or RAAS activation and was not reduced by aliskiren (9). In all these studies, the suggested mechanisms for the association of microalbuminuria with heart failure are related to renal endothelial dysfunction and neurohormonal alterations. Since most patients had albuminuria despite the difference in RAAS blockade, it was suggested that the albuminuric response to these drugs in heart failure is different than in diabetic nephropathy, which owes to as yet unknown mechanisms.

Conclusions

Microalbuminuria can be observed in healthy individuals, although it is more prevalent in diabetes and essential hypertension, which are relevant risk factors for coronary heart disease and heart failure. Presence of microalbuminuria might be related to endothelial damage, which may anticipate heart failure.
Irrespective of its mechanisms, the presence of microalbuminuria generally indicates worse prognosis. Its regular determination should be encouraged in patients with heart failure and its risk factors.
Although there are controversies in the literature, it seems clear that, before the start of heart failure, the use of RAAS blockers delays or even contributes to the normalisation of microalbuminuria. Once heart failure develops, these drugs may restore cardiac function, but not necessarily affect microalbuminuria. Therefore, in our opinion, microalbuminuria probably precedes heart failure.