Measurement of microalbuminuria – what the nephrologist should know

During the last few years, a subtle increase in urinary albumin excretion (UAE) not detectable by routine methods, so called microalbuminuria, has been identified as a prognostic marker for renal and/or cardiovascular risk in diabetic and non-diabetic subjects [ 1 ]. Consequently, assessment of microalbuminuria is now recommended as a risk stratification strategy not only in diabetic subjects, but also in the management of hypertensive patients [ 2–5 ]. In order to make the best clinical use of UAE, the physician who measures UAE should know several facts: what kind of albumin molecules are present in the urine, and which methods are most suitable for assessing each of them; what method of urine sampling is recommended and how should one interpret the UAE values; how can one reduce the variability of the UAE estimate and how should one evaluate the results and manage the patient based on the results of UAE determination. Albumin is an electronegative serum protein with a molecular mass of 66 349 Da. After glomerular filtration, part of the albumin is reabsorbed by tubular epithelial cells. Proteases split the albumin molecule into fragments, some of which back-leak into the tubular fluid [ 6 ]. In addition, albumin can reach the urine from an inflammatory lesion at any site from the renal pelvis to the urethra. In the absence of inflammation in the urinary tract, intact albumin of glomerular origin is the major source of albumin in the urine and only a small amount of small albumin fragments are present. Albumin can be detected by several methods based on precipitation (boiling, sulphosalicylic acid), dye-binding (biuret, tetrabromphenol, albumin blue 580) or immunologic detection (radioimmunoassay, nephelometry, test-strip) ( Table 1 ). While the immunoreactive methods estimate only complete albumin molecules recognized by antibodies, peptide fragments of albumin can be assessed by dye tests and specific spectrophotometry [ 7 , 8 ]. The immunologic methods are most frequently used for clinical purposes, not only because they are easy to use at relatively low cost, but also because they are able to detect small amounts of albumin in the range defined as microalbuminuria, i.e. <200 mg/l. Methods to assess albumin in urine Methods to assess albumin in urine Recently it has been emphasized that the amount of complete molecules of albumin in the urine that is not recognized by the immunologic methods is not negligible. These molecules are detected, however, by high performance liquid chromatography (HPLC) [ 7 ]. Although its clinical significance is not fully understood, the proportion of non-immunoreactive over the immunoreactive albumin depends on the total amount of albumin in the urine: the lower the amount, the higher the proportion of non-immunoreactive albumin [ 9 ]. Whether or not the albumin which is not immunnoreactive represents an early stage in the natural history and indicates a high risk of progression to the stage of microalbuminuria must be assessed in future prospective studies [ 10 ]. Whatever the case, even when non-immunoreactive albumin can uncover subjects at risk of increasing UAE over time, considering cost and accessibility of the HPLC technique, its widespread application in large populations will be difficult unless more readily available methods are developed. The concentration of urinary albumin depends on the amount of albumin excreted and on the urine concentration. A precise assessment is only possible if these two factors are considered. Methods based on timed urine collection or on simultaneous assessment of urine creatinine concentration have been used to avoid these confounders of the urinary albumin concentration. Collecting and measuring albumin in urine samples during 24 h (mg/24 h) or timed-overnight (µg/min) is frequently used in specialized settings. In contrast, for primary care and epidemiological studies, a correction is performed by simultaneously assessing creatinine excretion (mg/g Cr or mg/mmol Cr) in spot samples, collected from the first voided urine or at the time of the clinic visit. The correspondence between different UAE values at the level of the classical threshold to define microalbuminuria, i.e. >30 mg/24 h, and at the lowest UAE that points to an increased cardiovascular risk, i.e. 5 µg/min at night, is shown in Table 2 . UAE threshold to define risk for nephropathy and for cardiovascular risk according urine samples and units UAE threshold to define risk for nephropathy and for cardiovascular risk according urine samples and units The variability of UAE is one of the most important limitations of assessment for clinical purposes. This variability is affected by a large number of factors which must be kept in mind when interpreting the results. Fever, vigorous exercise, heart failure, haematuria and urinary tract infection produce a transitory increase in UAE which is more or less persistent depending on the cause, the intensity and the duration of the respective condition [ 11 ]. Urinary albumin excretion follows a circadian pattern. In normal volunteers, UAE decreases at nighttime in recumbent position to approximately 70% of the values during the period of activity [ 12 ] ( Figure 1 ). A similar circadian variability is also observed in hypertensive subjects, although their UAE is significantly higher than in healthy normotensive individuals. In patients with type 1 diabetes, however, the differences in UAE between periods of activity and sleep are quite large when compared to non-diabetic subjects ( Figure 2 ). This may reflect a greater damage to and permeability of the glomerular basal membrane in diabetic subjects, coupled with higher variability in glomerular filtration as a consequence of transitory hyperglycaemic periods. Circadian variation of UAE in subjects with essential hypertension. UAE is lower in resting conditions at night than during daytime activity. Relationship between UAE assessed on two different days 1 week apart in patients with type 1 diabetes (open circles) and essential hypertension (closed circles). Besides transitory factors and circadian variability, the reproducibility of UAE is limited and several studies have reported coefficients varying from of 10 to 25% [ 12 ]. This variability depends not only on the collection time of the urine sample, but also on the type of disease and the absolute amount of urinary albumin excreted. From the above, it is clear that the reproducibility is lower in collections obtained during awake periods than for samples obtained during night time or for the first voided urine. Likewise, the reproducibility is poorer in type 1 diabetic as compared to hypertensive patients, even when the UAE is in the same range. Finally, the reproducibility is better in the low range than in the high range of UAE. UAE is largely skewed towards high values. From a practical point of view, assessing taking all these factors into consideration, we suggest UAE by determination of albumin in at least two urine samples obtained in the same fashion within a period of several days. The best samples are those collected from night time or first morning urine. Although it is advisable to test the urine shortly after its collection, if necessary, urine can be kept at 4°C and without exposure to light for a week before the measurement is made [ 13 ]. Conventionally, UAE values have been categorized according to the presence or the absence of microalbuminuria. The presence of microalbuminuria points to a high cardiovascular and/or renal risk which requires intensification of treatment: further reduction of blood pressure by more intensified antihypertensive treatment, improvement of glucose control and blockade of the renin–angiotensin system according to current guidelines. Microalbuminuria has classically been defined as UAE from 30–300 mg/24 h or equivalent amounts when timed overnight or spot urine samples are used (see Table 2 ). The definition is based on the results of studies in diabetic patients which established microalbuminuria as a marker of the risk to develop nephropathy [ 14 ]. Several pieces of evidence indicate that one misses valuable information when one uses this categorization of UAE. Treating UAE values as a continuous variable provides advantages when used at baseline to assess the risk and during the follow-up to assess the effect of treatment. Although the natural history of microalbuminuria is not well defined, subjects with UAE below the outlined threshold, but with values in the highest range of ‘normality’, are at risk of advancing toward the stage of microalbuminuria. In type 1 diabetes, a progressive increment of UAE in a relatively short period of time is observed in subjects who are prone to develop diabetic nephropathy, while subjects not prone maintain low stable UAE for years [ 15–17 ]. No data are available for hypertensive patients or for the general population, but it is logical to assume that a progressive increase of albuminuria will occur over time in these individuals as well [ 18 , 19 ]. The rate of change in this case is slower, however, than that observed in patients with type 1 diabetes. If the subject develops type 2 diabetes, the rate of progression is accelerated [ 20 ]. In one study of our group, hypertensive subjects with an initial UAE level in the high normal range, from 15–29 mg/24 h, had an increased risk of progressing towards microalbuminuria despite antihypertensive treatment [ 18 ]. Consequently, the importance of UAE values in the highest range of normality is not negligible with respect to the risk of progression of UAE when assessed at baseline. Even more important is the prognostic value of subtle increments in UAE. In studies on the potential prognostic value of microalbuminuria for cardiovascular events, the threshold value indicating an increased risk was well below the UAE values defining microalbuminuria regardless of the population included [ 21 ]. The progressive increase of risk at values below the threshold of microalbuminuria was also confirmed by Klausen who noted that the risk of cardiovascular events was increased in hypertensive patients across a wide range of UAE values [ 22 ]. In hypertensive patients with a timed overnight UAE above 5 µg/min, total mortality and coronary events were increased independent of blood pressure values and other potential confounders. The risk of coronary heart disease and death increased significantly by 70 and 50%, respectively, in patients with UAE values between 5 and 10 µg/min, and even by 100% for both outcomes in patients with UAE values higher than 10 µg/min. These findings illustrate that even such relatively low values of UAE have prognostic value [ 22 ]. Furthermore, reduction of UAE is a well-established treatment target in diabetic subjects. It protects not only against progression towards renal damage, but also reduces the risk of cardiovascular disease in this high risk group. In non-diabetic subjects the potential of microalbuminuria as an intermediate endpoint during antihypertensive treatment is still unclear, but recent information shows that the in-treatment levels of urinary albumin are closely related to the risk of a subsequent cardiovascular event [ 19 ]: the greater the reduction of UAE, the lower the cardiovascular risk. Monitoring UAE can therefore be clinically useful to assess the success of interventions. The assessment of subtle increases in UAE is a powerful way to identify individuals at high cardiovascular risk in need of multiple cardiovascular risk factor intervention. Changes in UAE seem to run in parallel to changes in cardiovascular risk. Prompt intervention to avoid a progressive increase of UAE may provide better protection. Assessment of albuminuria during treatment is an easy way to monitor the success of the therapy. Values of UAE lower than considered in the past are associated with increased cardiovascular risk and a higher risk of progression towards higher values of UAE. These considerations illustrate the necessity to avoid categorization of UAE according to whether the threshold of microalbuminuria is transgressed or not. We therefore encourage to treat UAE as a continuous variable and to base evaluation on the absolute values of UAE [ 21 ]. Conflict of interest statement . This is to confirm that no conflict of interest exist concerning to the above manuscript entitled ‘Measurement of microalbuminuria – what the nephrologist should know’.