The term “microalbuminuria” was first proposed in the early 1960s, when Professor Harry Keen’s Group at Guy’s Hospital developed a radioimmunoassay technique for measuring in the urine of patients with type 1 diabetes, very low concentrations of albumin, well below the detection threshold of commonly used methods (Keen and Chlouverakis 1963). However, it was not until the 1980s, that it became an official part of the medical lexicon, when Svendsen and coll (Svendsen et al. 1981) and Viberti and coll (Viberti et al. 1982) described MAU as the presence of albuminuria below the detection limit of a standard dipstick, but at a level, revealed by using sensitive immunological methods, that was highly predictive of future overt proteinuria in diabetic patients.

It was initially defined as an albumin excretion rate between 20 and 200 μg/min. Although the lower bound was chosen because 95 % of ‘normal’ individuals had excretion rates below that limit, it was recognized that risk of progression to nephropathy was elevated among diabetics in the ‘high normal’ range (The Chronic Kidney Disease Prognosis Consortium 2011; Svendsen et al. 1981; Viberti et al. 1982). Mogensen was the first to describe the importance of MAU not only as a renal risk factor but also as a powerful predictor of CV mortality in patients with type 2 diabetes (Mogensen and Christensen 1984; Mogensen 1984).

In recent years, however, it has received increased attention as a prognostic marker for CV and/or renal risk even in non-diabetic subjects (Ruilope 2002; Matsushita et al. 2010; Gerstein et al. 2001; Hillege et al. 2001; 2002; Arnlov et al. 2005; Ruggenenti and Remuzzi 2006; Yuyun et al. 2004; Cirillo et al. 1998; Pontremoli et al. 1997; Pedrinelli et al. 2002; Bramlage et al. 2007; Coresh et al. 2007; Agrawal et al. 1996; Cerasola et al. 2008; 2010; Leoncini et al. 2010).

Similar to the relationship between BP and risk of CV events, mounting evidence indicates a continuous relationship between albumin excretion and risk (Matsushita et al. 2010; Gerstein et al. 2001; Hillege et al. 2001; 2002; Arnlov et al. 2005; Ruggenenti and Remuzzi 2006).

European studies report a 2.2–11.8 % prevalence of MAU in the general population (Hillege et al. 2001; Yuyun et al. 2004; Cirillo et al. 1998; Pontremoli et al. 1997; Pedrinelli et al. 2002; Bramlage et al. 2007; Coresh et al. 2007). Among the European epidemiological investigations performed in this field deserves to be mentioned the PREVEND (Prevention of Renal and Vascular End-stage Disease) study (Hillege et al. 2001) which involved 40,856 inhabitants of the city of Groningen (The Netherlands), aged 28–75 years. Microalbuminuria, defined as urinary albumin concentration 20–200 mg/L, was present in 7.2 % of population. After excluding the diabetic and hypertensive subjects MAU was still prevalent in 6.6 % of the individuals, and it was independently associated with age, gender, hypertension, diabetes, smoking, previous myocardial infarction and stroke. Cardiovascular risk factors were already elevated at levels of urinary albumin currently considered to be normal (10–20 mg/L or 15–30 mg per 24 h) (Hillege et al. 2001).

These percentages indicate that MAU is more often present in subjects with CV risk factors; however, in apparently healthy subjects MAU can frequently be encountered.

Data from the US National Health and Nutrition Examination Surveys (NHANES) showed an increase in prevalence of MAU (defined as a urinary albumin– creatinine ratio [ACR] of 30–300 mg/g) from 7.1 to 8.2 % during the survey periods 1988–1994 and 1999–2004. The increase was attributed to older age of the population, greater proportion of minority groups, prevalence of hypertension and diabetes and higher body mass index (Coresh et al. 2007).

In arterial hypertension, prevalence of MAU ranging from 5 to 60 % has been reported (Agrawal et al. 1996; Cerasola et al. 2008; 2010; Leoncini et al. 2010; Böhm et al. 2007; Meccariello et al. 2016). This wide range may be due to differences in ethnic groups, specimen collection, cut-off level of albumin excretion, analytical methods and influence of anti-hypertensive medications. The distribution of demographic and coexisting diseases may also contribute.

In the abovementioned large-scale population surveys, the NHANES III (Coresh et al. 2007) and the PREVEND study (Hillege et al. 2001), MAU was detected respectively in 16 % and 11.5 % of people with hypertension.

The international, observational, practice based study i-SEARCH (Survey for Evaluating Microalbuminuria Routinely by Cardiologists in patients with Hypertension) was designed to assess the frequency with which MAU occurred in a very large group of hypertensive outpatients attending a cardiologist or internist. A total of 21,050 patients from 26 countries were included in the primary analysis. Overall, this study demonstrated a very high worldwide prevalence (58.4 %) of MAU in high-risk cardiovascular patients, but with a considerable variation across countries (Böhm et al. 2007).

The use of a semi-quantitative test, which tend to overestimate urine albumin concentration, may explain to some extent the unusually high prevalence of microalbuminuria reported in this large-scale study (Böhm et al. 2007), as well as in other ones (Bramlage et al. 2007; Agrawal et al. 1996). In a very recent study conducted in 1024 unselected hypertensive patients followed by 13 Italian general practitioners MAU was detected in 35 % of the overall population (Dworkin et al. 1983).

A lower frequency of MAU (22.7 %) was observed in the REDHY (REnal Dysfunction in HYpertension) study that was conducted in 1856 non-diabetic middle-aged subjects with arterial hypertension and without cardiovascular complications and known renal diseases (Cerasola et al. 2008; 2010). Moreover, in the I-DEMAND (Italy Developing Education and awareness on Microalbuminuria in patients with hyperteNsive Disease) study, an observational, cross-sectional investigation performed in 87 centers of specialized care (Internal Medicine, Cardiology, Nephrology, Diabetology) MAU was found in 27 % of the entire population, including 3534 patients, 37 % of whom had diabetes mellitus (Leoncini et al. 2010).

The presence of microalbuminuria implies dysfunction of the glomerular filtration barrier. It may result from haemodynamic-mediated mechanisms and/or functional or structural impairment of the glomerular barrier.

Microalbuminuria in essential hypertensive patients is the consequence of an increased transglomerular passage of albumin rather than the result of a decrease in the proximal tubule reabsorption of albumin. At least two mechanisms have been proposed for the greater albumin excretion rate (AER) in some patients with essential hypertension: increased glomerular hydrostatic pressure or increased permselectivity of the glomerular basement membrane (Mountokalakis 1997). Glomerular hydrostatic pressure is regulated by the relative vasoconstriction-vasodilatation of the afferent and efferent glomerular arterioles. The tone of these arterioles is regulated by different mechanisms, and their sensitivity to pressor/depressor substances also varies substantially (Dworkin et al. 1983).

A variety of endocrine, paracrine, and autocrine substances, as well as pharmacological agents, may influence intraglomerular hemodynamic independently of actions on systemic blood pressure (BP).

Normally, an elevation of systemic arterial pressure is associated with vasoconstriction of the glomerular afferent arterioles, which prevents transmission of the elevated hydrostatic pressure to the glomerulus and maintains the glomerular hydrostatic pressure unaltered (Hostetter et al. 1981). This protects the glomeruli from the potential damages of hypertension. If the autoregulatory adaptation of the glomerular afferent arterioles is defective, increased glomerular hydrostatic pressure may ensue. Alternatively, an exaggerated vasoconstriction of the efferent arterioles may increase intraglomerular hydrostatic pressure, even in the presence of normal systemic pressure (Hostetter et al. 1981).

A large body of experimental and clinical evidence supports the notion that derangements of these adaptive mechanisms are important determinants for the susceptibility to develop progressive renal disease (Mountokalakis 1997).

In 1992 our group, in order to verify if in essential hypertension (EH) MAU increase could be due to hemodynamic modifications or to glomerular structural changes, in a very small group of newly diagnosed essential hypertensives (n = 30; EHs) having 24-h AER > 16 μg/min (n =15) and in 15 EHs with 24-h AER < 16 μg/min, the day- and night-time behaviour of creatinine clearance (Ccr), as well as AER clearance (AER-C) and fractional clearance (AER-FC), and behaviour of BP were evaluated (Cottone and Cerasola 1992). Patients with 24-h AER > 16 μg/min showed hyper-filtering values of both 24-h and daytime creatinine clearance than the other group of EHs, while during the night period, there were no significant differences between the two groups. On the contrary, AER and both AER-C and AER-FC resulted markedly and significantly higher in the EHs with 24-h AER > 16 μg/min not only in the 24-h evaluation, but also during the night-time study notwithstanding the significant decrease in BP and in Ccr observed during the night. We concluded that these data, in the absence of correlations between BP and AER-FC seemed to demonstrate that among newly diagnosed essential hypertensives a subgroup of them could have early renal hemodynamic changes (Cottone and Cerasola 1992). These hemodynamic modifications, along with defects of the glomerular membrane permselectivity, led to increased microalbuminuria.

Hyperfiltration is probably mediated by abnormal transmission of systemic hypertension to the glomerulus through a disturbance in glomerular autoregulation and/or from progressive loss of functioning nephrons. Of the non-haemodynamics, functional abnormalities of the glomerular basal membrane have been claimed, although some evidence has been against this in hypertension.

More widely accepted, however, is that MAU reflects the kidney expression of a more generalised state of endothelial dysfunction (Deckert et al. 1989; Pedrinelli et al. 1994; Cottone et al. 2000; 2007).

With regard to systemic endothelial dysfunction, our group hypothesized that in EHs, plasma levels of pro-atherogenic adhesion molecules would be increased and related with AER. Thus, we studied biochemical markers of endothelial activation ICAM-1 and VCAM-1, and their relationship with AER in a group of individuals with uncomplicated EH (Cottone et al. 2007). One hundred patients with essential hypertension and no diabetes or ultrasonographic evidence of atherosclerosis were included in the study. EHs were first studied overall, than were divided into two subgroups: those with AER >20 μg/min (MAUs) and those with AER <20 μg/min (non-MAUs). Microalbuminuric hypertensives had greater levels of adhesion molecules than non- MAUs. In multiple regression models in hypertensive persons AER was independently associated with ICAM-1, and VCAM-1. These findings showed that in EH there is a very early activation of endothelial adhesion molecules favouring atherosclerosis.

A further interesting data emerging from that study was the significant difference of plasma concentrations of adhesion molecules when comparing non-MAU hypertensives with healthy controls. Indeed, it seemed that endothelial activation expressed by adhesion molecules would be earlier than microalbuminuria, confirming that microalbuminuria could be considered a marker of systemic endothelial dysfunction (Cottone et al. 2007). A study by Klausen (Klausen et al. 2004) demonstrated that in the general population urinary albumin excretion, below the MAU definition, was associated with increased coronary heart disease risk, independently of hypertension. Thus, the Authors hypothesize that MAU emerges later in the atherosclerotic process. Our findings, showing a pro-atherogenic endothelial activation in the presence of values of AER currently considered as ‘normal’ seemed to be in line with this finding.

Considering the role that inflammation plays in the development of endothelial changes that lead to atherosclerosis, studies on this issue were performed (Pedrinelli et al. 2004; Festa et al. 2000; Jager et al. 2002; Kshirsagar et al. 2008).

C-Reactive Protein (CRP), a well-known marker of inflammation, was positively associated with microalbuminuria in the large data set compiled from the National Health and Nutrition Examination Surveys (NHANES) 1999 through 2004 (Kshirsagar et al. 2008). In this study, including 12,831 US men and women, the multivariate analysis showed that an increase of one milligram per liter in CRP concentration was significantly associated with a 2 % increased odds of microalbuminuria (p = 0.0003) (Kshirsagar et al. 2008).

Microalbuminuria can be revealed by several methods based on immunologic detection (immunonephelometry, immunoturbidimetry, radioimmunoassay, enzyme-linked immunosorbent assay) (Miller et al. 2009). Among these there are also a variety of semiquantitative dipsticks, such as Clinitek Microalbumin Dipsticks and Micral-Test II test strips, which can be used for MAU screening. The reported sensitivity and specificity of these tests range from 80 to 97 % and 33 to 80 %, respectively (Miller et al. 2009).

Microalbuminuria has been defined as an AER higher than the threshold value obtained from studies assessing the risk for developing nephropathy in diabetes, that is an AER between 20 and 200 μg/min. It is now clear that its significance extends beyond nephropathy and it likely mirrors a more widespread vascular injury.

It should be noted that AER may also be expressed in terms of milligrams per day (mg/day), in which case the range for microalbuminuria is 30–300 mg/day (Levin et al. 2013; Mancia et al. 2013).

Indeed, there is growing evidence, arising from several prospective studies that urinary albumin excretion levels well below the current microalbuminuria threshold (“low-grade albuminuria”) are also associated with an increased risk of incident cardiovascular disease and all-cause mortality (Matsushita et al. 2010; van der Velde et al. 2011; Astor et al. 2011; Nitsch et al. 2013; Mahmoodi et al. 2012; Fox et al. 2012; Hallan et al. 2012; Klausen et al. 2004; Redon and Williams 2002). Even in apparently healthy individuals (without diabetes or hypertension), such an association has been shown (Hillege et al. 2001). These epidemiological data prompted some authors to propose the adoption of a lower AER cut-off point for the detection of subjects with an enhanced cardiovascular risk (Redon and Williams 2002) and other ones (Ruggenenti and Remuzzi 2006) to abandon the terms of microalbuminuria and macroalbuminuria and replaced with ‘urine albumin’, because the use of arbitrary dichotomous categorisation does not reflect the continuously increasing risk associated with progression of urine albumin concentrations. Moreover, the term microalbuminuria may be confusing, since it should reflect small albumin molecules, and not small amounts of albumin (Ruggenenti and Remuzzi 2006).

Despite this criticism, the term microalbuminuria has become widely accepted in clinical practice.

Although 24-h urine collection is the gold standard for the detection of microalbuminuria, it has been suggested that screening can be more simply achieved by a timed urine collection or by untimed spot urine sample. In this latter case the confounding effect of variations in urine volume on the urine albumin concentration can be avoided normalizing the urinary albumin concentration to the urinary creatinine concentration (since creatinine excretion rate is considered constant) (Miller et al. 2009; Levey et al. 2009).

Indeed, the albumin/creatinine ratio (ACR) from spot urine, preferably that first voided in the morning, may be considered equivalent to the values during a 24-h urine collection (Miller et al. 2009; Levey et al. 2009).

Even if the ACR corrects for unknown urine volumes, it needs theoretically differentiation of males from females in whom creatininuria is lower because of reduced muscle mass (Miller et al. 2009; Cirillo et al. 2006; Mogensen et al. 1995), a fact not taken into account, by the KDIGO guidelines for evaluation of CKD (Levin et al. 2013) and by the 2013 ESH-ESC guidelines for management of arterial hypertension (Mancia et al. 2013), that for the definition of microalbuminuria do not recommend the use of gender-specific ACR thresholds, but a single cut-off value, that for simplicity was arbitrarily rounded to 30 mg/g (Levin et al. 2013; Mancia et al. 2013) (Fig. 1).

For the same reasons described above, albumin excretion will be underestimated in a muscular man with a high rate of creatinine excretion and overestimated in a cachectic patient in whom muscle mass and creatinine excretion are markedly reduced (Miller et al. 2009; Cirillo et al. 2006).

A number of physiologic and pathologic factors must be taken into account when interpreting AER and ACR results. Albumin excretion is normally about 25 % higher during the day, and it can vary by 10–25 or more in day-to-day measurements. In addition to age and sex, body mass index and a high-protein meal can all affect the AER. Vigorous exercise can cause a transient increase in albumin excretion. As a result, patients should refrain from vigorous exercise in the 24 h prior to the test. Measurement can be further confounded by fever, congestive heart failure, urinary tract infection, and by some drugs (Mogensen et al. 1995). Because the limited reproducibility of AER and ACR measurements, most expert committees recommend that a presumptive indication of microalbuminuria should be confirmed by quantitative measurement of urinary albumin in at least two of three, preferably nonconsecutive, specimens (Mogensen et al. 1995). Even a single determination of elevated albumin concentration, however, can predict (albeit with reduced precision) renal and cardiovascular diseases (Gerstein et al. 2001; Hillege et al. 2002).

There is a strong evidence of a close relationship of microalbuminuria with a variety of cardiovascular risk factors, such as hypertension, diabetes, aging, smoking, hyperlipidemia and metabolic syndrome.

The association of microalbuminuria with all these factors is so relevant that MAU may legitimately be regarded as an integrated marker of cardiovascular risk (Pedrinelli et al. 2002).

Several studies have shown significant correlations between blood pressure values and AER (Cirillo et al. 1998; Bramlage et al. 2007; Coresh et al. 2007; Agrawal et al. 1996; Cerasola et al. 2008; 2010; Leoncini et al. 2010; Böhm et al. 2007; Cerasola et al. 1989; 1996; Hsu et al. 2009). There is also evidence that the relationship of BP with albuminuria is relatively continuous and graded, with even high-normal levels of BP associated with albuminuria (Hsu et al. 2009).

In general, the association of BP values with albuminuria becomes even closer when BP is recorded through ambulatory blood pressure monitoring (Cerasola et al. 1989; 1996; Palatini et al. 1995) which provides a more precise estimation of the real BP status. Ambulatory BP monitoring also allowed to show that there are no significant differences in AER between the white coat hypertensive and the normotensive subjects (Cerasola et al. 1995; Palatini et al. 1998) and that urinary albumin levels are higher in hypertensives in whom a blunted or absent nocturnal fall of BP occurs (non dippers) (Bianchi et al. 1994; Redon et al. 1994). In this context it is interesting to note that a clinical condition characterized by a non-dipping BP pattern, such as the obstructive sleep apnea syndrome, has been associated with microalbuminuria in patients with arterial hypertension (Tsioufis et al. 2008).

Very recently, in a group of more than 300 untreated hypertensive subjects, we reported a positive association of AER with average real variability (ARV) of 24-h systolic blood pressure (Fig. 2), a measure of short-term blood pressure variability endowed with prognostic implications (Mulè et al. 2016). This association was weakened, but still significant, taking into account the effect of the mean level of 24-h SBP and other potential confounders. Moreover, in the subset of patients with MAU and inverse relationship between ARV of 24-h SBP and eGFR was also found (Mulè et al. 2015a).

Accumulating data indicate that MAU clusters with several metabolic abnormalities (Cirillo et al. 1998; Pontremoli et al. 1997; Cerasola et al. 2008; 2010; Leoncini et al. 2010; Campese et al. 1999; Cerasola and Cottone 1997; Pinto-Sietsma et al. 2003; Klausen et al. 2009; Mulè et al. 2006; Palaniappan et al. 2003; Andronico et al. 1998; Srinivasan et al. 2000; Alberti and Zimmet 1998; Jager et al. 1998; Chen et al. 2004; Mulè et al. 2005; Cuspidi et al. 2004; Klausen et al. 2007; Parving et al. 2006), including some phenotypes of the metabolic syndrome (MetS) and may be indeed a part of this syndrome (Alberti and Zimmet 1998).

We have previously shown, in a group of 353 essential hypertensive subjects that the prevalence of microalbuminuria and the levels of AER were higher in patients with MetS than in those without it (Mulè et al. 2005) (Fig. 3). Furthermore, although researchers have reported mixed results (Andronico et al. 1998; Srinivasan et al. 2000; Jager et al. 1998) on the association between MAU and hyperinsulinemia and insulin-resistance, multiple studies confirmed this relation (Andronico et al. 1998; Srinivasan et al. 2000). An investigation of 5659 men and women from the NHANES III, confirmed the association between MAU and the MetS, with the strongest association being with high fasting serum glucose and high BP (Palaniappan et al. 2003). A further analysis of the same study documented that the multivariate-adjusted odds ratios of microalbuminuria increased progressively with a higher number of components of the MetS, defined by the ATP III guidelines (Chen et al. 2004).

Similar results were obtained in patients with arterial hypertension (Mulè et al. 2005; Cuspidi et al. 2004).

In the general population of the Copenhagen City Heart Study not only the strong association between microalbuminuria and the MetS was confirmed, but interestingly it was also observed that MAU (even when defined by a very low cut-off value, that is > 5 μg/min) confers an increased risk of death and CV disease to a similar extent as the MetS and independently of it and of other confounding factors (Klausen et al. 2007).

A relationship between higher AER and cigarette smoking has been described in diabetic individuals (Parving et al. 2006; Gerstein et al. 2000) and in hypertensive people (Andronico et al. 2005), in subjects with increased CV risk (Gerstein et al. 2000) as well as in the general population (Pinto-Sietsma et al. 2000).

In the PREVEND study, current smokers had a higher median albumin excretion than nonsmokers and were more likely to have microalbuminuria. After adjustment for several potential confounding factors, persons who smoked 20 or fewer cigarettes/day and persons who smoked more than 20 cigarettes/day, when compared to nonsmokers, showed a relative risk of microalbuminuria of 1.92 [CI, 1.54–2.39] and 2.15 [CI, 1.52–3.03], respectively (Pinto-Sietsma et al. 2000).

However, overall, in the various studies, the link between smoking and MAU is not very strong and it seems unlikely that smoking may explain much of the excess CV risk associated with MAU.

Besides the associations reported between microalbuminuria and various conventional cardiovascular risk factors, significant correlations have been observed between increased AER and nontraditional risk factors for cardiovascular diseases.

For example, during the last decade, several cross-sectional investigations have documented that microalbuminuria is related to various inflammatory markers (Pedrinelli et al. 2004; Festa et al. 2000; Jager et al. 2002; Kshirsagar et al. 2008; Mulè et al. 2009) and to some markers of endothelial damage and dysfunction, including von Willebrand factor and adhesion molecules (sVCAM1 and sICAM1 and e-selectin) (Pedrinelli et al. 1994; Cottone et al. 2000; 2007).

The influence of glomerular filtration rate (GFR) on the microalbuminuria of hypertension merits a comment. The prevalence of microalbuminuria increases as the GFR decreases, although not always in parallel. Moreover, when GFR is < 60 ml/min/1.73 m², the probability of AER normalisation during antihypertensive treatment is clearly reduced (Pascual et al. 2006).

Changes in proteinuria have been suggested as a surrogate outcome for kidney disease progression to facilitate the conduct of clinical trials (Levey et al. 2009). The progression of CKD is often slow, and until late stages, it is often asymptomatic. Thus, end points for clinical trials may be long delayed from disease onset and the time that interventions may be effective. Surrogate end points may provide an opportunity to detect early evidence of effectiveness. Proteinuria is an accepted marker for kidney damage; is related to diagnosis, prognosis, and treatment in kidney disease; and has been suggested as a surrogate outcome for clinical trials of kidney disease progression (Pascual et al. 2006). A biomarker is a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes, or pharmacological responses to a therapeutic intervention (Biomarkers Definitions Working Group 2001). By definition, proteinuria and decreased GFR are biomarkers for CKD and potentially could be surrogate end points for kidney failure because in general, they precede the development of kidney failure.

An intermediate end point is a biomarker that is intermediate in the causal pathway between an intervention and a clinical end point. Decreased GFR also is an intermediate end point because it is on the causal pathway to kidney failure. Doubling of serum creatinine level is accepted as a surrogate end point because it reflects a large decrease in GFR and predicts the development of kidney failure. The increase in albumin excretion rate potentially could be a surrogate outcome for a large decrease in GFR in clinical trials.

Indeed, among the impressive number of data the PREVEND Study offered, the role of albumin excretion as a better marker than estimated GFR to identify individuals at risk for accelerated GFR loss is relevant. The 8592 patients who were included in this study were followed for a 4-year period. Among them, 134 patients with macroalbuminuria, 128 with erythrocyturia, and 103 with impaired renal function were identified. In the general population the prevalence of macroalbuminuria, erythrocyturia, and impaired renal function was calculated to be 0.6, 1.3, and 0.9 %, respectively (Halbesma et al. 2006). After a mean follow-up of 4.2 year, the macroalbuminuria group showed a −7.2 ml/min/1.73 m² estimated GFR loss compared with −2.3 ml/min/1.73 m² in the control group (difference p < 0.001). After exclusion of individuals with diabetes, the observed renal function decline in the macroalbuminuria group was 7.1 ml/min/1.73 m². It is interesting that eGFR fell by only 0.2 ml/min/1.73 m² (−0.4 %) in participants with impaired renal function. Participants who at baseline had both macroalbuminuria and impaired renal function (n =18) experienced a rate of renal function decline of 9.0 ml/min/1.73 m² (Halbesma et al. 2006).

More recently, it was analyzed whether screening for albuminuria in the general population identifies individuals at increased risk for renal replacement therapy or accelerated loss of renal function. In a general population-based cohort of 40,854 individuals aged 28–75 year, a first morning void for measurement of urinary albumin was collected. In a subset of 6, 879 individuals, 24-h urinary albumin excretion and estimated GFR at baseline and during 6 years of follow-up were measured. Linkage with the national renal replacement therapy registry identified 45 individuals who started renal replacement therapy during 9 year of follow-up. The quantity of albuminuria was associated with increased renal risk: the higher the level of albuminuria, the higher the risk of need for renal replacement therapy and the more rapid renal function decline. A urinary albumin concentration ≥ 20 mg/L identified individuals who started renal replacement therapy during follow-up with 58 % sensitivity and 92 % specificity. Of the identified individuals, 39 % were previously unknown to have impaired renal function. Restricting screening to high-risk groups (e.g., known hypertension, diabetes, cardiovascular disease, and older age) reduced the sensitivity of the test only marginally but failed to identify 45 % of individuals with micro- and macroalbuminuria. Therefore, individuals with elevated levels of urinary albumin are at increased risk for RRT and accelerated loss of renal function. Screening for albuminuria identifies patients at increased risk for progressive renal disease, 40–50 % of whom were previously undiagnosed or untreated (van der Velde et al. 2009).

In arterial hypertensive subjects a retrospective cohort analysis of 141 hypertensive individuals followed up for approximately 7 years was carried out several years ago (Bigazzi et al. 1998). During follow-up, the rate of clearance of creatinine from patients with microalbuminuria decreased more than did that from those with normal urinary albumin excretion (Bigazzi et al. 1998).

Similar associations between albuminuria and renal function decline have been described by Viazzi et al in a larger cohort of patients with essential hypertension. Subjects who developed a renal event had higher baseline albumin-to-creatinine ratio compared with subjects who did not develop a renal event (5.12 vs 4.42 mg/g; p < 0.001) (Viazzi et al. 2010).

According to several studies microalbuminuria correlate with various cardiac abnormalities and diseases, including left ventricular (LV) hypertrophy and dysfunction, electrocardiographic abnormalities, and coronary atherosclerosis.

There is an extensive and highly consistent body of evidence showing that microalbuminuric patients exhibited a higher prevalence of left ventricular hypertrophy (LVH), assessed either by electrocardiography or echocardiography, compared to normoalbuminurics (Cerasola et al. 1989; 1996; 2004; Palatini et al. 1995; Pontremoli et al. 1999; Wachtell et al. 2002a; b; Tsioufis et al. 2002; Ratto et al. 2008; Smilde et al. 2005; Lieb et al. 2006).

Since the first description of our group in 1989 of a close relationship between LV mass and albumin excretion rate in hypertensive patients (Cerasola et al. 1989), the vast majority of the following reports supported the view that hypertensives with elevated AER had higher cardiac mass, indicating that early renal damage and LVH occur in a parallel fashion.

It is important to note that the association between left ventricular mass and AER not only reflects an abnormal pressor overload, but remains statistically significant after accounting for blood pressure values (Cerasola et al. 1996; 2004; Palatini et al. 1995; Pontremoli et al. 1999; Wachtell et al. 2002a; 2002b; Tsioufis et al. 2002; Ratto et al. 2008; Smilde et al. 2005). Further support to the blood pressure independent relationship of microalbuminuria with LVH arises from the observation that inappropriate left ventricular mass, that is the LV mass exceeding the compensatory needs for cardiac workload, is more strongly associated with microalbuminuria than do appropriate LV mass (Ratto et al. 2008).

Even if albumin excretion rate and LV mass are significantly and independently correlated, AER determination may add information on cardiovascular risk stratification beyond those provided by ultrasonographic detected LVH. Indeed, in a group of 312 essential hypertensive patients, we observed that a more intensive investigation for target organ damage, including ultrasound examination of the heart to detect LVH and microalbuminuria determination, beyond routine work-up alone, increases the proportion of hypertensive patients who should be classified as having a high absolute risk of cardiovascular morbidity and mortality. Overall, 26 % of patients changed risk category (mostly shifting from the medium- to high-risk stratum), a proportion that was significantly different from the percentage of patients reclassified after the addition to the routine work-up of either microalbuminuria (14 %) or echocardiography alone (16 %) (Cerasola et al. 2004). In some (Pontremoli et al. 1999), but not all studies (Wachtell et al. 2002a) microalbuminuric subjects showed a higher prevalence of concentric than eccentric LVH, being the former geometric pattern associated with a worse outcome than the latter.

Furthermore, patients with microalbuminuria showed subclinical impairment of systolic and diastolic LV (Pontremoli et al. 1999; Wachtell et al. 2002a). In the LIFE study, patients with microalbuminuria had significantly lower endocardial and midwall fractional shortening. On the other hand, patients with abnormal diastolic LV filling parameters had a significantly increased prevalence of microalbuminuria (Wachtell et al. 2002a).

Further data supporting the close association between elevated AER and cardiac abnormalities derive also from the cross-sectional relationship observed between MAU and silent myocardial ischemia, which can be evidenced by ST segment and T wave changes on an electrocardiogram (Diercks et al. 2000). Moreover, a significant and independent association between MAU and various ECG abnormalities (arrhythmias, intraventricular conduction defects, ventricular repolarization alterations and left-axis deviation) in the large observational I-DEMAND study, including 4121 hypertensive patients without overt cardiovascular disease, was found (Sciarretta et al. 2009).

Elevated AER was also directly associated with angiographic evidence of CAD. A study of 308 patients who underwent elective coronary angiography revealed that patients with angiographic evidence of CAD had significantly higher urinary albumin levels than disease-free individuals and that AER correlated with the severity of coronary atherosclerosis at angiography (Tuttle et al. 1999).

A significant association between microalbuminuria and several functional and structural changes of the arterial tree, beyond the coronary bed, has been described

Despite some conflicting results, several cross-sectional studies (Yokoyama et al. 2004; Bigazzi et al. 1995; Rodondi et al. 2007; Furtner et al. 2005; Geraci et al. 2016; Jørgensen et al. 2007), found that MAU was associated with higher thickness of the intima and media (IMT) layers of the carotid artery. In a wide population of hypertensive subjects with (n = 183) and without CKD (n = 280), we recently found greater values of carotid IMT in microalbuminuric patients when compared to normoalbuminuric ones (Geraci et al. 2016) (Fig. 4).

Moreover, in the Bruneck Study, a prospective population-based survey including 684 Caucasians adults, ACR was significantly and independently associated with the presence and severity of carotid and femoral atherosclerosis (Furtner et al. 2005). In addition, microalbuminuria predicts the development and progression of carotid atherosclerosis (Jørgensen et al. 2007).

Hence, it is not unexpected that microalbuminuria in several studies has been associated with a greater incidence of stroke (Gerstein et al. 2001; Hillege et al. 2002; Yuyun et al. 2004), and with cerebral small vessel disease (Ravera et al. 2002; Wada et al. 2007).

In addition, albumin excretion rate correlates with functional abnormalities of the vasculature, such as alterations of flow- and nitroglycerin-mediated brachial artery dilatation (Stehouwer et al. 2004; Malik et al. 2007) and impaired large artery elastic properties (Mulè et al. 2004; 2009; 2010; Smith et al. 2005; Hermans et al. 2007; Upadhyay et al. 2009; Munakata et al. 2009).

Large artery stiffness, especially aortic stiffness, assessed by pulse wave velocity (PWV) measurement is now well accepted as an independent predictor of cardiovascular morbidity and mortality. We demonstrated in a sample of 140 untreated nondiabetic essential hypertensive patients that microalbuminuria, was significantly associated with an augmented aortic stiffness, independently of low-grade inflammation, expressed by increased plasma level of high-sensitivity C-reactive, and of other potential confounding factors (Mulè et al. 2009). Our findings, which were replicated in a wider group of hypertensive patients (Mulè et al. 2010), are in line with previous observations of our group (Mulè et al. 2004) and of other authors (Yokoyama et al. 2004; Smith et al. 2005; Hermans et al. 2007; Upadhyay et al. 2009; Munakata et al. 2009) that reported significant relations of microalbuminuria with different indices of reduced arterial distensibility in a variety of populations. At the level of renal vasculature, microalbuminuria has been associated with increased intrarenal resistive index (RRI), a sonographic parameter, which is defined as the dimensionless ratio of the difference between maximum and minimum (end-diastolic) flow velocity to maximum flow velocity (Mulè et al. 2015b; Viazzi et al. 2015) (Fig. 5).