Journal of Human Hypertension (2002) 16, 399–404
 2002 Nature Publishing Group All rights reserved 0950-9240/02 $25.00
www.nature.com/jhh
ORIGINAL ARTICLE
Microalbuminuria is an integrated marker
of subclinical organ damage in primary
hypertension
G Leoncini1, G Sacchi2, M Ravera1, F Viazzi1, E Ratto1, S Vettoretti1, D Parodi1,
GP Bezante3, M Del Sette4, G Deferrari1 and R Pontremoli1
1
Department of Internal Medicine, Section of Nephrology, University of Genoa, Italy; 2Department of
Experimental Medicine, University of Genoa, Italy; 3Department of Internal Medicine, Section of
Cardiology, University of Genoa, Italy; 4Department of Neurological Science and Neuro-rehabilitation,
University of Genoa, Italy
Increased urine albumin excretion is associated with an
unfavourable cardiovascular risk profile and prognosis
in primary hypertension, even though its pathogenesis
is currently unknown. Microalbuminuria (Mi) has been
proposed as an integrated marker to identify patients
with subclinical organ damage, but its routine use is still
too often neglected in clinical practice. The aim of our
study was to evaluate the relationship between urinary
albumin excretion and early signs of subclinical target
organ damage (TOD), namely left ventricular hypertrophy and carotid atherosclerosis in a large group of
non diabetic hypertensive patients. A group of 346
never treated patients with primary hypertension (212
men, 134 women, mean age 47 ± 9 years) referred to our
clinic were included in the study. They underwent the
following procedures: (1) family and personal medical
history and physical examination; (2) clinical blood
pressure measurement; (3) routine blood chemistry and
urine analysis including determination of urinary albumin excretion (ACR); (4) electrocardiogram; (5) ultrasound evaluation of left ventricular mass (LVMI) and
carotid artery thickness (IMT). The overall prevalence of
Mi, left ventricular hypertrophy, and carotid plaque was
13, 51, and 24% respectively. Mi was significantly correlated with LVMI (P ⬍ 0.0001), IMT (P ⬍ 0.0001) and several metabolic and non-metabolic risk factors (blood
pressure, body mass index, serum lipids). Cluster analysis identified three subgroups of patients who differ significantly with regards to TOD and albuminuria (P ⭐
0.001 for each of the examined variables). Patients with
higher IMT and LVMI values also showed increased ACR
levels. Furthermore, patients with microalbuminuria
were more likely to have both LVH and IMT values above
the median for the study population (OR 21, C.I. 4.6–
99.97, P ⬍ 0.0001). Mi is an integrated marker of subclinical organ damage in patients with primary hypertension. Evaluation of urinary albumin excretion is a
specific, cost-effective way to identify patients at higher
risk for whom additional preventive and therapeutic
measures are advisable.
Journal of Human Hypertension (2002) 16, 399–404. DOI:
10.1038/sj/jhh/1001408
Keywords: microalbuminuria; target organ damage; essential hypertension; cardiovascular risk
Introduction
Assessment of subclinical end-organ damage is a
key element in the evaluation of patients with primary hypertension. In fact, detecting left ventricular
hypertrophy (LVH) or peripheral atherosclerosis
markedly increases the overall cardiovascular risk
profile and may be helpful both for deciding
whether to begin treatment and for identifying optimal target blood pressure.1,2
Correspondence: R Pontremoli, MD, PhD, Department of Internal
Medicine, University of Genoa, Viale Benedetto XV, 6-16132
Genoa, Italy. E-mail: rpontrem얀libero.it
Received 6 November 2001; revised 14 February 2002; accepted
18 February 2002
Microalbuminuria (Mi) has been associated with
a number of unfavourable biohumoral risk factors as
well as with subclinical organ damage in nondiabetic patients with primary hypertension.3,4
Recently, it was proven to be a predictor of cardiovascular morbidity and mortality in long term longitudinal studies.5,6 Even though the pathogenesis of
increased urinary albumin excretion (UAE) is at
present still unclear, this abnormality has been proposed as a useful integrated marker of target organ
damage (TOD) and increased cardiovascular risk.7
However, routine evaluation of Mi is not recommended by international guidelines as part of the
diagnostic work-up of every hypertensive patient
and its potential diagnostic power has not yet been
fully exploited in clinical practice.
Microalbuminuria and organ damage
G Leoncini et al
400
The aim of the present study was to evaluate the
relationship between UAE and other signs of TOD,
namely LVH and carotid atherosclerosis in a large
group of untreated hypertensive patients. We tested
the hypothesis that microalbuminuria is associated
with the presence of target organ damage and therefore may be helpful in identifying the subgroup of
hypertensive patients at higher cardiovascular risk.
Patients and methods
Patients and design
From January 1996 to January 1999 all previously
untreated patients with primary hypertension
attending the out-patient hypertension clinic of the
Nephrology Unit at the Department of Internal Medicine (University of Genoa, Italy) were asked to participate in this study which was part of a larger clinical trial approved by the Ethical Committee of our
Institution. Altogether 518 hypertensive patients
were seen at our clinic within that period of time.
The selection criteria for the study led to the
exclusion of 125 patients as follows: diabetes mellitus (32 patients), severe obesity (defined as body
weight ⬎150% of ideal body weight) (22 patients),
renal disease (21 patients), chronic heart failure
(NYHA class III and IV) (17 patients), neoplastic disease (seven patients), hepatic disease (10 patients),
positive history or clinical signs of ischaemic heart
disease (14 patients), disabling diseases such as
dementia or inability to cooperate (two patients).
Hypertension was defined according to the criteria
in the fifth report of the Joint National Committee
(JNC V) as an average blood pressure of 140/90
mm Hg on at least two different occasions. Of the
remaining 393 patients eligible to participate in the
study on the basis of available clinical and laboratory data, 15 refused and 32 did not meet study criteria based on the results of additional tests prescribed for clinical reasons during their first
examination. Data obtained from the remaining 346
patients (all Caucasian Europeans) form the basis of
the present report. Of the participating patients, 246
(71%) had never been treated for hypertension,
while 100 (29%) had received antihypertensive
treatment in the past, albeit intermittently and not
during the 6 months prior to the study.
After informed consent had been obtained, all
patients were subjected to the following procedures
within 2 weeks of the initial visit: (1) clinic blood
pressure measurement, (2) blood and urine sampling
for routine blood chemistry and urine examination,
(3) evaluation of UAE, (4) standard 12-lead electrocardiogram (ECG), (5) echocardiogram, and (6)
carotid ultrasonogram. On the study day, height and
weight were measured, then venous blood was
drawn after an overnight fast in order to measure
haematochemical parameters. Blood pressure was
measured by a trained nurse, with the patient in a
sitting position after a 5-min rest, with a mercury
Journal of Human Hypertension
sphygmomanometer (cuff size 12.5 × 40 cm). The
systolic and diastolic blood pressures were read to
the nearest 2 mm Hg. Disappearance of Korotkoff’s
sounds (phase V) was the criterion for diastolic
blood pressure. The lowest of three consecutive
readings was recorded. Body mass index (BMI) was
calculated by the formula: BMI = weight (kg)/height
(m2). Low-density cholesterol (LDL)-cholesterol was
calculated using Friedewald’s formula.8 Creatinine
clearance was calculated using Cockcroft-Gault’s
formula9 and is expressed in ml/min. Family history
and lifestyle habits were assessed by means of a
standard questionnaire. Smoking was graded using
a three-point scale: non-smoker, ex-smoker, and
active smoker.
Albuminuria
The presence of microalbuminuria was evaluated in
each patient by measuring the albumin-to-creatinine
ratio (ACR) on three non-consecutive, first morning
urine samples. The ACR was calculated as follows:
urine albumin concentration (milligrams per
litre)/urine creatinine concentration (mmol per litre)
and expressed in mg/mmol. Urine albumin concentration was measured by a commercially available
radio-immunoassay kit (Sclavo, Cinisello Balsamo,
Italy). The intra- and inter-assay coefficients of variation of ACR in our laboratory were 4.5% and 6.1%
respectively. The average (arithmetic mean) of the
three ACRs from each patient was calculated in
order to categorise patients. In order to account for
differences in basal creatinine excretion rates and
BMI, different values were used to define microalbuminuria in males (ACR between 2.38 and 19)
and in females (ACR between 2.96 and 20) respectively. These criteria proved to have good sensitivity
and specificity in detecting albumin excretion rates
between 20 and 200 ␮g/min.10,11
Electrocardiography
Standard ECGs were evaluated to detect LVH. LVH
was defined by the presence of at least one of two
electrocardiographic criteria: the Sokolow-Lyon
voltage and the gender-specific Cornell voltage.12,13
Echocardiography
All echocardiographic studies were performed using
an Acuson XP-128 ultrasound machine. Echocardiograms were obtained at rest with patients supine
in the left lateral position, using standard parasternal and apical views. The overall monodimensional
left ventricular measurements and the bidimensional (apical four- and two-chamber) views were
obtained according to the recommendations of the
American Society of Echocardiography. All tracings
were obtained and read by a single observer blinded
to the clinical characteristics of the patients under
Microalbuminuria and organ damage
G Leoncini et al
observation. LV mass was derived using the formula
described by Devereux and associates:14,15
LV mass (grams) = 0.80 × 1.04 [(VSTd + LVIDd +
PWTd)3 − (LVIDd)3] + 0.6
where VSTd is ventricular septal thickness at end
diastole, LVIDd is LV internal dimension at enddiastole, and PWTd is LV posterior wall thickness
at end-diastole. Left ventricular mass was corrected
for height2.7 (LVMI), and expressed in units of
grams/meter (g/m2.7). The presence of LVH was
defined for LVMI ⬎51 g/m2.7 in either gender.16–17
Carotid ultrasonography
The intima plus media thickness (IMT) of both
carotid arteries was evaluated by high resolution US
scan as described by Kawagishi18 using a 10-MHz
in-line duplex Diasonic Spectra System. The carotid
artery was scanned at the bifurcation and at the common carotid artery (CCA). At each longitudinal projection the far-wall IMT, as defined by Weldelhag,19
was measured at the distal end of the CCA, 10 mm
caudally to the point where the near and far walls
lose their parallel configuration. Carotid plaque was
defined as IMT ⬎1.3 mm. IMT was always measured
on the CCA outside the plaque, if any was present.
Each measurement was calculated using the average
of three readings.
Statistical analysis
All data are expressed as mean ± s.e.m. Comparison
of proportion between groups was performed using
the ␹2 test. Variables found to deviate from normality were log-transformed (log10) before statistical
analysis. The Pearson correlation test was used to
study the linear relationship between ACR and other
continuous variables.
Multivariate (k-mean cluster) analysis was performed on the entire cohort of patients to identify subgroups (clusters) of patients on the basis of the severity of target organ damage (ie, IMT and LVMI values)
and urine albumin excretion (ACR). The aim of the
k-mean clustering procedure is to classify patients
in clusters with the lowest within-group variability
and the highest among group variability for any
given set of variables. Patients within the same cluster show similar values for the variables taken into
consideration, while differing from patients in the
other clusters for those same variables. The algorithm starts with k random clusters, and then moves
cases in and out of clusters in an attempt to (a) minimise variability within clusters and (b) maximise
variability among clusters. This is analogous to
‘ANOVA in reverse’ in the sense that the significance test in ANOVA measures the among group
variability against the within-group variability when
computing the significance test for the hypothesis
that the means in the groups are different from each
other. To evaluate the appropriateness of classi-
fication, the within-cluster variability (low if the
classification is good) is compared to the betweencluster variability (high if the classification is good),
ie, a standard among-group analysis of variance for
each variable is performed.20 Cluster analysis,
applied to our study sample of 346 patients, allowed
us to detect a difference of 0.8 mg/mmol in ACR,
3.5 g/m2.7 in LVMI, and 0.01 mm in IMT respectively, with the two-sided significance test at a 1%
level and a ␤ probability of 0.1.
All statistical analyses were performed using Statview for Windows, SAS Institute Inc, version 5.0.1,
Cary, NC, USA. A P value of ⬍0.05 was considered
statistically significant.
401
Results
Study population
Clinical characteristics of the study patients are
reported in Table 1. The overall prevalences of
microalbuminuria, LVH, and carotid plaque were
13, 51, and 24% respectively. Seventy-five patients
(22%) had electrocardiographic signs of LVH
(according to one of the two study criteria). Overall,
ECGs were available for 339 out of 346 patients. ACR
was significantly correlated with blood pressure,
BMI, serum lipids, as well as the severity of endorgan damage, namely LVMI and carotid IMT
(Table 2).
To further investigate the relationship between
albuminuria and TOD, k-mean cluster analysis was
applied to data from the entire study population.
Three subgroups of patients were identified (cluster
no 1, n = 58, cluster no 2, n = 129 and cluster no 3,
n = 123), differing significantly from each other by
Table 1 Clinical characteristics of study patients (n = 346)
Age (years)
Gender (% males)
Body mass index (kg/m2)
Systolic blood pressure (mm Hg)
Diastolic blood pressure (mm Hg)
Heart rate (beats/min)
Reported duration of hypertension (months)
Prevalence of active smokers (%)
Serum glucose (mmol/L)
Uric acid (␮mol/L)
Calc. creatinine clearance (ml/min)
Triglycerides (mmol/L)
HDL-Cholesterol (mmol/L)
Total serum cholesterol (mmol/L)
Serum creatinine (mg/dl)
ACR (mg/mmol)
Prevalence of microalbuminuria (%)
Common carotid IMT (mm)
Prevalence of carotid plaque (%)
LVMI (g/m2.7)
Prevalence of LVH (%)
47 ± 9.2
61
26.4 ± 3.4
158 ± 15.4
102 ± 7.8
75 ± 3.1
36 (3–270)
23
5 ± 0.6
303 ± 86.8
90 ± 22.8
0.40 (0.29– 4.1)
1.34 ± 0.4
5.53 ± 1.1
0.9 ± 0.2
0.6 (0.1–19)
13
0.7 ± 0.02
24
52 ± 0.9
51
Data are mean ± s.d., except for duration of hypertension, triglycerides and ACR mean reported as median (range). HDL = highdensity lipoprotein; ACR = urinary albumin to creatinine ratio;
IMT = intimamedia thickness; LVMI = left ventricular mass index;
LVH = left ventricular hypertrophy.
Journal of Human Hypertension
Microalbuminuria and organ damage
G Leoncini et al
402
Table 2 Univariate correlations between Log ACR (n = 346) and
selected clinical variables
Variable
Correlation
SBP (mm Hg)
DBP (mm Hg)
MBP (mm Hg)
BMI (kg/m2)
Log triglycerides
(mmol/L)
1/HDL/LDL
IMT (mm)
LVMI (g/m2.7)
P
Confidence
intervals
95%
lower
95%
upper
0.210
0.178
0.226
0.137
0.175
0.002
0.002
0.0001
0.02
0.003
0.100
0.068
0.112
0.022
0.61
0.314
0.284
0.333
0.248
0.284
0.141
0.355
0.342
0.03
⬍0.0001
⬍0.0001
0.014
0.220
0.206
0.264
0.477
0.466
SBP = systolic blood pressure; DBP = diastolic blood pressure;
MBP = mean blood pressure; BMI = body mass index; HDL = highdensity lipoprotein; LDL = low-density lipoprotein; IMT = carotid
intima plus media thickness; LVMI = left ventricular mass index.
the severity of target organ damage (ie, carotid IMT
and LVMI) as well as by ACR (P ⬍ 0.0001 for each
of the variables examined by multimethod, between
group ANOVA, and contrast analysis) (Figure 1 and
Table 3). There was no difference regarding the percentage of active smokers in the three subgroups of
patients. Patients with more severe involvement of
end organs had a higher prevalence of ECG-detected
LVH (25 vs 13%, OR 4.59 P = 0.03), higher blood
pressure (P ⬍ 0.01) and worse lipid profiles (P ⬍
0.03). Moreover, the subgroup of patients with TOD
also showed higher BMI (P = 0.04), age (P = 0.03),
and BP levels (P = 0.01) as compared with the subgroup without TOD (data not shown). Furthermore,
patients with Mi were more likely to have both LVH
and thicker IMT (above the median) (OR 21, 95% CI
4.6–99.97, P ⬍ 0.0001) (Figure 2). Mi showed 96%
specificity and 45% sensitivity for detecting LVH or
the concomitant presence of LVH and increased
IMT. These values led to 88% and 90% positive predictive power (PPP) and 59% and 69% negative pre-
Figure 1 Plot of urinary albumin excretion (ACR), left ventricular
mass index (LVMI), and carotid intima-media thickness (IMT)
values in patients grouped according to cluster analysis. k-mean
cluster analysis was performed on the entire cohort of patients.
Three subgroups were identified on the basis of both the degree
of microalbuminuria and the severity of target organ damage (ie,
LVMI and carotid IMT). The three clusters differ significantly for
each variable we examined. Data are mean ± s.e.m. Actual values
for LVMI, ACR and IMT can be obtained by multiplying the
reported data by 50, 5 and 2 respectively.
dictive power for the presence of LVH or both
TOD respectively.
Discussion
The present study shows that Mi is a marker of subclinical end-organ damage, ie LVH and peripheral
atherosclerosis. In fact, UAE was strongly correlated
with LVMI and carotid IMT in our study population
of non-diabetic hypertensive patients (Table 2). By
using cluster analysis we were able to demonstrate
in a new and original way that different types of
TOD tend to aggregate with increased UAE in the
same subgroup of patients (Figure 1) and therefore,
that Mi is an excellent integrated marker of high risk
cardiovascular status. The three subgroups of
patients identified by cluster analysis differ significantly as for the presence and degree of TOD, as well
as for UAE, while patients within the same cluster
show similar values as indicated by ANOVA (Table
Table 3 Cluster analysis of ACR and target organ damage
Variable
Descriptive statistics
Cluster no 1 (n = 58)
Mean
ACR
IMT
LVMI
4.5
0.9
68.6
Cluster no 2 (n = 129)
Analysis of variance
Cluster no 3 (n = 123)
St. Variance Mean
dev.
St. Variance Mean
dev.
St. Variance
dev.
5.5
0.3
8.5
2.9
0.2
3.5
1.4
0.2
5.9
30.7
0.1
72.7
1.8
0.7
53.1
8.7
0.04
12.2
0.8
0.6
40.0
2.0
0.03
34.6
Between
Sum of
Squares
Within
Sum of
Squares
F
P
183.2
0.65
11508.9
1038
4.527
3323.9
9.1
7.4
178.3
0.0002
0.001
⬍0.0001
Among the entire study population, three subgroups of patients were identified (cluster no 1, no 2, and no 3), differing significantly
from each other by the presence and severity of several features of target organ damage (ie, carotid IMT and LVMI) as well as by urinary
albumin excretion (ACR) (P ⬍ 0.001 for each of the variables examined by multimethod, between group ANOVA, and contrast analysis).
Degrees of freedom 1–307. ACR = albumin to creatinine ratio, mg/nmol; IMT = intima plus media thickness, mm; LVMI = left ventricular
mass index, g/m2,7.
Journal of Human Hypertension
Microalbuminuria and organ damage
G Leoncini et al
Figure 2 Carotid intima plus media thickness (IMT) is plotted
against the left ventricular mass index (LVMI) according to each
patient’s urinary albumin excretion status (normoalbuminurics
open circles, microalbuminurics filled circles). Orthogonal lines
mark the cut-off values for left ventricular hypertrophy and IMT
above the median of the study population. The odds ratio for a
microalbuminuric patient of having both LVH and increased IMT
is 21 (95% CI 4.6–99.97, P ⬍ 0.0001).
3). Thus, it appears that patients with higher LVMI
and IMT (ie those in cluster no 1) also have higher
UAE values. Moreover, the clustering of UAE with
higher values of LVMI and IMT is in agreement with
previous reports by our group and by others,10,21–23
and indicate an association among microalbuminuria, LVH and carotid atherosclerosis. Therefore,
measuring urinary ACR, a simple, low cost and readily available test, can be regarded as a cost-effective
way to identify, among the large number of hypertensive patients, those who need additional preventive measures and/or more aggressive antihypertensive treatment. In fact, when the degree of
subclinical organ damage is observed on the basis of
UAE (Figure 2), it appears that patients with Mi are
more likely to have both LVH and higher carotid
IMT (P ⬍ 0.0001). The probability that a patient with
increased UAE also shows these two signs of TOD
is 21 times higher as compared with patients with
normal urine albumin excretion (PPP 88% for LVH
and 90% for both LVH and increased IMT). Thus,
the high specificity of Mi for identifying TOD makes
it an attractive screening test for all hypertensive
patients.
The
pathogenetic
mechanisms
underlying
increased UAE in non-diabetic hypertensive
patients are not fully understood. The severity of
blood pressure load and the increased systemic permeability to albumin, possibly due to early endothelial dysfunction, seem to play a major role in the
development of this abnormality.24 Several data,
however, suggest an interplay among additional
factors/mechanisms such as abnormalities in the
lipid metabolism,25,26 prothrombotic factors,3,27 and
increased activity of the renin-angiotensin system.28
Finally, a functional haemodynamic abnormality (ie
an increased filtration fraction due to higher angiot-
ensin II levels) and/or the presence of structural
changes within the kidney cannot be ruled out as
causes of Mi in essential hypertension, at least until
these issues have been addressed by properly
devised studies, and the renal prognostic value of
mild increases in UAE established. Our data show
significant linear correlations between UAE and
blood pressure, BMI, and serum lipids. These findings, together with the clustering of hypertensive
TOD and differences among clusters as for several
known biochemical and clinical risk factors, suggest
multifactorial pathogenesis although they do not
prove it.
In conclusion the present study confirms and
extends previous work on the role of Mi as an integrated marker of increased cardiovascular risk in
essential hypertension. On the basis of these results
we propose searching for Mi as part of the initial
work-up of every hypertensive patient.
403
Acknowledgements
This work was supported in part by grant no
RF99.52 from Ministero della Sanità, Ricerca Finalizzata 1999. The excellent technical help of Cinzia
Tomolillo, BS is gratefully acknowledged.
References
1 The sixth report of the Joint National Committee on
prevention, detection, evaluation, and treatment of
high blood pressure. Arch Intern Med 1997; 157:
2413–2446.
2 Chalmers J et al. World Health Organization-International Society of Hypertension Guidelines for the
management of hypertension. Guidelines sub-committee of the World Health Organization. Clin Exp Hypertens 1999; 21: 1009–1060.
3 Pontremoli R. Microalbuminuria in essential hypertension – its relation to cardiovascular risk factors.
Nephrol Dial Transplant 1996; 11: 2113–2115.
4 Bianchi S, Bigazzi R, Campese VM. Microalbuminuria
in essential hypertension: significance, pathophysiology, and therapeutic implications. Am J Kidney Dis
1999; 34: 973–995.
5 Borch-Johnsen K et al. Urinary albumin excretion. An
independent predictor of ischemic heart disease. Arterioscler Thromb Vasc Biol 1999; 19: 1992–1997.
6 Jensen JS et al. Arterial hypertension, microalbuminuria, and risk of ischemic heart disease. Hypertension
2000; 35: 898–903.
7 Viberti G, Thomas S. Should we screen for microalbuminuria in essential hypertension? Am J Kidney
Dis 1999; 34: 1139–1141.
8 Friedewald WT, Levy RI, Fredrickson DS. Estimation
of the concentration of low density lipoprotein cholesterol in plasma without use of the preparative ultracentrifuge. Clin Chem 1972; 18: 499–502.
9 Cockcroft WD, Gault MH. Prediction of creatinine
clearance from serum creatinine. Nephron 1976; 16:
31–34.
10 Pontremoli R et al. Prevalence and clinical correlates
of microalbuminuria in essential hypertension. The
MAGIC study. Hypertension 1997; 30: 1135–1143.
Journal of Human Hypertension
Microalbuminuria and organ damage
G Leoncini et al
404
11 Mangili R et al for the Italian Microalbuminuria Study
Group. Prevalence of hypertension and microalbuminuria in adult type 1 (insulin dependent) diabetic
patients without renal failure in Italy. Validation of
screening techniques to detect albuminuria. Acta
Diabetol 1992; 29: 156–166.
12 Sokolow M, Lyon TP. The ventricular complex in left
ventricular hypertrophy as obtained by unipolar precordial and limb reads. Am Heart J 1949; 37: 161–186.
13 Casale PN et al. Improved sex-specific detection of left
ventricular hypertrophy: development and prospective validation of improved criteria. J Am Coll Card
1985; 6: 572–580.
14 Schiller NB et al. Recommendations for quantitation
of the left ventricle by two dimensional echocardiography: American Society of Echocardiography Committee on standards, subcommittee on quantitation of
two-dimensional echocardiograms. J Am Soc Echocardiogr 1989; 2: 358–367.
15 Devereux RB et al. Echocardiographic assessment of
left ventricular hypertrophy: comparison to necropsy
findings. Am J Cardiol 1986; 57: 450– 458.
16 De Simone G et al. Effect of growth on variability of
left ventricular mass: assessment of allometric signals
in adults and children and their capacity to predict
cardiovascular risk. J Am Coll Cardiol 1995; 25:
1056–1062.
17 Savage DD et al. The spectrum of left ventricular
hypertrophy in a general population sample: the Framingham study. Circulation 1987; 75 (Suppl I): I-26–1–
33.
18 Kawagishi T et al. High-resolution B-Mode ultrasonography in evaluation of atherosclerosis in uremia. Kidney Intern 1995; 48: 820–826.
Journal of Human Hypertension
19 Weldelhag I, Wiklund O, Wikstrand J. Atherosclerotic
changes in the femoral and carotid arteries in familial
hypercholesterolemia: ultrasonographic assessment of
intima-media thickness and plaque occurrence. Arterioscler Thromb 1993; 13: 1404 –1411.
20 Tryon RC, Bailey D. Cluster Analysis. McGraw-Hill
Book Co: New York, 1970, pp 1–347.
21 Pontremoli R et al. Left ventricular geometry and function in patients with essential hypertension and microalbuminuria. J Hypertens 1999; 17: 993–1000.
22 Bigazzi R et al. Increased Thickness of carotid artery
in patients with essential hypertension and microalbuminuria. J Hum Hypertens 1995; 9: 827–833.
23 Pedrinelli R. Microalbuminuria in essential hypertension. Nephron 1996; 73: 499–505.
24 Pedrinelli R et al. Microalbuminuria and endothelial
dysfunction in essential hypertension. Lancet 1994;
344: 14 –18.
25 Pontremoli R et al. Prevalence of micro- and macroalbuminuria and their relationship with other cardiovascular risk factors in essential hypertension.
Nephrol Dial Transplant 1995; 10 (Suppl 6): 6–9.
26 Campese VM, Bianchi S, Bigazzi R. Association
between hyperlipidemia and microalbuminuria in
essential hypertension. Kidney Int 1999; 71: S10–S13.
27 Agewall S, Lindstedt G, Fagerberg B. Independent
relationship between microalbuminuria and plasminogen activator inhibitor-1 activity (PAI-1) in clinically
healthy 58-year-old men. Atherosclerosis 2001; 157:
197–202.
28 Baldoncini R et al. High plasma renin activity is combined with elevated urinary albumin excretion in
essential hypertensive patients. Kidney Int 1999; 56:
1499–1504.