Urinary Albumin Excretion Rate Is Associated with Increased Ambulatory Blood Pressure in Normoalbuminuric Type 2 Diabetic Patients

Pathophysiology/Complications ORIGINAL ARTICLE

Urinary Albumin Excretion Rate Is Associated With Increased Ambulatory Blood Pressure in Normoalbuminuric Type 2 Diabetic Patients

1 1 CRISTIANE B. LEITAO˜ , MD MARCEL P. MOLON emerging evidence that patients with dia- 1 2 LUÍS H. CANANI, MD ANTONIOˆ F. PINOTTI, MD betes in the high-normal range of UAER 1 1 PATRÍCIA B. POLSON JORGE L. GROSS, MD are already at high risk for progressing to microalbuminuria or even more advanced stages of renal disease (6–9). Arterial hypertension follows the establishment of microalbuminuria in pa- OBJECTIVE — To evaluate the 24-h blood pressure profile in normoalbuminuric type 2 tients with type 1 diabetes (10). In type 2 diabetic patients. diabetes, this relationship is not that clear because hypertension is a common fea- RESEARCH DESIGN AND METHODS — A cross-sectional study was conducted in 90 type 2 diabetic patients with a urinary albumin excretion rate (UAER) Ͻ20 ␮g/min on two ture in these patients, regardless of renal occasions, 6 months apart (immunoturbidimetry). Patients underwent clinical and laboratory status (11). Ambulatory 24-h blood pres- evaluations. Ambulatory blood pressure monitoring and echocardiograms were also performed. sure monitoring has a better correlation with target organ damage than doctor’s RESULTS — UAER was found to correlate positively with systolic doctor’s office blood pres- office blood pressure measurements (12) sure measurements (r ϭ 0.243, P ϭ 0.021) and ambulatory blood pressure (24 h: r ϭ 0.280, P ϭ and allows the evaluation of blood pres- 0.008) and left ventricular posterior wall thickness (r ϭ 0.359, P ϭ 0.010). Patients were divided sure parameters, such as circadian blood Ͻ Ն Ն Ն ␮ into four groups according to UAER ( 5, 5–10, 10–15, and 15–20 g/min). Systolic pressure rhythm and blood pressure blood pressure parameters for the 1st, 2nd, 3rd, and 4th groups, respectively, were 123.0 Ϯ Ϯ Ϯ Ϯ loads. Nondiabetic healthy subjects in the 10.6, 132.5 15.0, 139.0 23.4, and 130.7 8.0 mmHg for 24-h blood pressure (ANOVA high-normal range of UAER (15–20 P ϭ 0.004) and 48.4 Ϯ 6.0, 54.5 Ϯ 11.2, 58.8 Ϯ 15.6, and 57.6 Ϯ 8.0 mmHg for 24-h pulse pressure (ANOVA P ϭ 0.003). A progressive increase in the prevalence of diabetic retinopathy mg/24 h) have higher blood pressure lev- was observed from the 1st to the 4th UAER group: 27.3, 43.8, 45.5, and 66.7% (P ϭ 0.029 for els than nondiabetic individuals in the trend). lower ranges of UAER (13). The same seems to be true for patients with type 1 CONCLUSIONS — In type 2 diabetic patients, UAER in the normoalbuminuric range is diabetes and UAER above the median (4.2 positively associated with systolic ambulatory blood pressure indexes, left ventricular posterior ␮g/min) (14), suggesting that high- wall thickness, and diabetic retinopathy, suggesting that intensive blood pressure treatment may normal albuminuria is associated with an prevent diabetes complications in these patients. increase in blood pressure values. Never- theless, no information is available con- Diabetes Care 28:1724–1729, 2005 cerning the relationship between UAER and 24-h ambulatory blood pressure in type 2 diabetes. Therefore, we hypothe- icroalbuminuria is a known risk fine microalbuminuria (urinary albumin sized that patients with type 2 diabetes in factor for the development of clin- excretion rate [UAER] Ͼ20 ␮g/min or the high-normal range of UAER would M ical nephropathy in type 1 and Ͼ30 mg/24 h) (5) was defined by consen- have higher blood pressure levels than pa- type 2 diabetes (1–4), and it is also an sus, based on studies performed in patients in the lower ranges. The aim of this independent risk factor for cardiovascular tients with type 1 and type 2 diabetes in study was to evaluate the blood pressure disease (4). The cutoff value used to de- the 1980s (1–4). However, there is patterns and the clinical and laboratory ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● profile of normoalbuminuric patients From the 1Endocrine Division, Hospital de Clıńicas de Porto Alegre, Universidade Federal do Rio Grande do with type 2 diabetes according to their Sul, Porto Alegre, Brazil; and the 2Cardiology Division, Hospital de Clıńicas de Porto Alegre, Universidade UAER levels. Federal do Rio Grande do Sul, Porto Alegre, Brazil. Address correspondence and reprint requests to Jorge L. Gross, Serviço de Endocrinologia do Hospital de Clıńicas de Porto Alegre, Rua Ramiro Barcelos 2350, Pre´dio 12, 4° andar, 90035-003, Porto Alegre, RS, Brazil. E-mail: [email protected]. RESEARCH DESIGN AND Received for publication 10 December 2004 and accepted in revised form 4 April 2005. METHODS — A cross-sectional study Abbreviations: UAER, urinary albumin excretion rate. was performed in normoalbuminuric A table elsewhere in this issue shows conventional and Syste`me International (SI) units and conversion type 2 diabetic patients regularly attend- factors for many substances. ing the diabetes outpatient clinic at Hos- © 2005 by the American Diabetes Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby pital de Clıńicas de Porto Alegre. Type 2 marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. diabetes was defined based on World

1724 DIABETES CARE, VOLUME 28, NUMBER 7, JULY 2005 Leita˜o and Associates

Health Organization criteria (i.e., Ͼ30 years of age at onset of diabetes, no previous episode of ketoacidosis or docu- mented ketonuria, and treatment with insulin only after 5 years of diagnosis) (15). Patients with other renal diseases, cardiac arrhythmia, or postural hypoten- sion were excluded. Normoalbuminuric patients were selected based on UAER values Ͻ20 ␮g/min on at least two occasions over the preceding 6 months while on their usual antihypertensive drugs. Pa- tients using ACE inhibitors had these medications suspended for 1 week, after which a 3rd UAER measurement was performed. Of the 92 patients recruited, 2 had UAER Ͼ20 ␮g/min and were not included in the study. Therefore, UAER was consistently within the normoalbuminuric range for all of the patients included. The study protocol was approved by the hospital’s research ethics committee, and Figure 1—Correlation between UAER (logarithmic scale) and 24-h systolic blood pressure. informed consent was obtained from all patients. the percentage of 24-h and daytime blood colorimetric method. The glomerular fil- pressure Ն140/90 mmHg and of night- tration rate was determined in 67 patients Clinical, blood pressure, and time blood pressure Ն120/80 mmHg. by the single-injection 51Cr-EDTA tech- echocardiographic evaluation Pulse pressure was the difference between nique (18). Demographic and anthropometric data the systolic and diastolic blood pressure were collected by means of an interview means. Patients with a night/day blood Statistical analysis and clinical examination, as previously pressure ratio Ͼ0.9 were considered to be One-way ANOVA or the ␹2 test were used described (16). Indirect ophthalmoscopy nondippers. to compare clinical and laboratory data. was performed through dilated pupils by Echocardiograms (n ϭ 60) were ob- The Bonferroni test was used to determine an ophthalmologist. For the purpose of tained according to the recommendations differences between groups. Quantitative this study, patients were classified only of the American Society of Echocardiog- variables without normal distribution according to the presence or absence of raphy (17), using standard parasternal were submitted to logarithmic transfor- any degree of diabetic retinopathy. and apical views with subjects in the par- mation. Data are expressed as the blood pressure evaluations were per- tial left decubitus position and using a means Ϯ SD, except for triglycerides, se- formed 1 week after withdrawal from all commercially available instrument (So- rum creatinine, and blood pressure loads, antihypertensive medications. Office nus 1000; Hewellet Packard). Left ven- which are expressed as the median blood pressure was measured with a mer- tricular mass was calculated based on wall (range). Correlations were performed cury sphygmomanometer, using the left thickness (end-diastolic ventricular inter- with the Pearson’s ␹2 (log-UAER versus arm and with the patient in a sitting posi- nal diameter, end-diastolic interventricu- blood pressure means and left ventricular tion, after a 5-min rest. The mean of two lar septum, and posterior wall) and was posterior wall thickness) or Spearman’s measurements was considered. The pa- adjusted to body surface area. The cardi- rank correlation (log-UAER versus blood tient was classified as hypertensive based ologist who performed the echocardio- pressure loads) tests, depending on the on use of antihypertensive drugs and/or grams was unaware of the subjects’ distribution of variables. Multiple linear office blood pressure Ն140/90 mmHg. clinical and laboratory characteristics. regression was performed with log-UAER Ambulatory 24-h blood pressure moni- as the dependent variable. P values Ͻ0.05 toring was performed by oscillometry Laboratory methods (two tailed) in the univariate analysis were (Spacelabs 90207), with a 15-min inter- UAER was measured in sterile 24-h timed considered to be significant. val during the daytime and a 20-min in- urine samples by immunoturbidimetry terval during the nighttime. Patients were (Microlab; Ames, Tarrytown, NY). HbA1c RESULTS — There was a positive and advised to maintain their usual daily ac- (AlC) was measured by a high-perfor- significant correlation found between tivities. Sleep time was recorded as the mance liquid chromatography system log-UAER and systolic office (r ϭ 0.243, period between the time when the patient (normal range 2.7–4.3%, Merck-Hitachi P ϭ 0.021), systolic 24-h (r ϭ 0.280, P ϭ went to bed and the time when the patient 9100). Fasting plasma glucose was mea- 0.008) (Fig. 1), systolic daytime (r ϭ woke up the next morning. The means of sured by the glucose-peroxidase colori- 0.264, P ϭ 0.013) and systolic nighttime 24-h, daytime, and nighttime systolic and metric enzymatic method (Biodiag- (r ϭ 0.261, P ϭ 0.013) blood pressure, as diastolic blood pressure were recorded. nostica). Creatinine was measured by the well as for log-UAER and systolic blood blood pressure load was defined as Jaffe´ method and the lipid profile by a pressure loads (24-h: r ϭ 0.353, P ϭ

DIABETES CARE, VOLUME 28, NUMBER 7, JULY 2005 1725 Albuminuria and ambulatory blood pressure

Table 1—Clinical and laboratory characteristics of type 2 diabetic patients according to UAER

1st group 2nd group 3rd group 4th group (Ͻ5 ␮g/min) (Ն5–10 ␮g/min) (Ն10–15 ␮g/min) (Ն15–20 ␮g/min) P n 37 32 11 10 Age (years) 56.7 Ϯ 10.1 57.1 Ϯ 10.5 57.8 Ϯ 7.5 58.5 Ϯ 11.3 0.962 Diabetes duration (years) 9.0 Ϯ 6.2 11.3 Ϯ 7.0 11.8 Ϯ 5.3 16.8 Ϯ 8.8 0.021* Male subjects 35.1 25.0 45.5 50.0 0.402 Caucasians 83.3 80.6 72.7 100.0 0.395 Smokers 16.2 9.7 9.1 0.0 0.686 Hypertension 56.8 62.5 72.7 80.0 0.510 Office blood pressure (mmHg) Systolic 136.4 Ϯ 17.5 139.7 Ϯ 21.7 143.6 Ϯ 20.6 145.9 Ϯ 8.6 0.446 Diastolic 83.8 Ϯ 10.9 86.5 Ϯ 11.2 93.2 Ϯ 14.8 82.3 Ϯ 9.4 0.084 BMI (kg/m2) 28.6 Ϯ 3.8 28.1 Ϯ 4.3 28.5 Ϯ 4.9 27.6 Ϯ 6.2 0.927 Retinopathy 27.3 43.8 45.5 66.7 0.029† Fasting plasma glucose (mg/dl) 152.5 Ϯ 57.5 167.3 Ϯ 65.6 170.2 Ϯ 62.8 138.1 Ϯ 39.0 0.477 A1C (%) 6.04 Ϯ 2.40 6.31 Ϯ 1.92 6.94 Ϯ 1.54 5.82 Ϯ 1.21 0.631 Total cholesterol (mg/dl) 181.6 Ϯ 44.9 218.6 Ϯ 48.8 218.8 Ϯ 52.4 199.8 Ϯ 55.1 0.011‡ HDL cholesterol (mg/dl) 44.6 Ϯ 9.2 48.3 Ϯ 13.7 46.9 Ϯ 17.4 46.1 Ϯ 13.3 0.686 Triglycerides (mg/dl) 120 (54–421) 131 (39–1115) 197 (50–392) 115 (68–609) 0.152 Serum creatinine (mg/dl) 0.8 (0.5–1.2) 0.8 (0.3–1.3) 0.9 (0.7–1.0) 0.9 (0.5–1.1) 0.204 GFR (ml minϪ1 1.73 mϪ2)§ 107.5 Ϯ 33.8 109.7 Ϯ 34.3 81.5 Ϯ 25.6 120.3 Ϯ 38.9 0.472 Data are means Ϯ SD, %, or median (range). *P Ͻ 0.05 for 1st vs. 4th group; †P for trend; ‡P Ͻ 0.05 for 1st vs. 2nd group; §glomerular filtration rate data available for 27, 23, 9, and 8 patients in the 1st, 2nd, 3rd, and 4th groups, respectively. GFR, glomerular filtration rate.

0.001; daytime: r ϭ 0.351, P ϭ 0.001; the four groups, as was the ethnic group 11.3 mmHg for nighttime values and nighttime: r ϭ 0.229, P ϭ 0.031). distribution. (ANOVA P ϭ 0.049), reflecting decreased A positive and significant correlation Office blood pressure was similar in artery compliance. The same pattern was between the log-UAER and left ventricu- all groups (Table 1). In general, patients observed for the systolic 24-h blood pres- lar posterior wall thickness (r ϭ 0.359, belonging to the 1st UAER group (UAER sure load values: 14.8% (range 0–95.1), P ϭ 0.010) was also observed. This sup- Ͻ5 ␮g/min) had lower ambulatory sys- 38.8% (0–96.4), 33.7% (0–100.0), and ports the hypothesis that UAER is associ- tolic blood pressure levels than patients 34.9% (6.2–81.8); ANOVA P ϭ 0.001. It ated with target-organ injury, even if it is with higher UAER levels. Systolic blood was also observed for daytime blood pres- within the normal range, probably be- pressure values for the 1st, 2nd, 3rd, and sure loads: 7.1% (0–92.5), 35.2% (0– cause of higher blood pressure levels. 4th UAER groups, respectively, were as 94.6), 29.6% (2.3–100.0), and 22.8% There was no association between log- follows: 123.0 Ϯ 10.6, 132.5 Ϯ 15.0, (5.5–73.1); ANOVA P ϭ 0.011. There UAER and the other echocardiographic 139.0 Ϯ 23.4, and 130.7 Ϯ 8.0 mmHg was a borderline association between parameters analyzed, such as left ventric- for 24-h values (ANOVA P ϭ 0.004); UAER groups and nighttime blood pres- ular mass, septum thickness, and left 126.09 Ϯ 10.8, 135.0 Ϯ 15.1, 142.5 Ϯ sure load: 23.8% (0–100.0), 66.3% (0– atrium size. 23.2, and 133.3 Ϯ 9.23 mmHg for day- 100.0), 53.3% (13.5–100.0), and 49.5% Patients were divided into four time values (ANOVA P ϭ 0.004); and (4.8–100.0); ANOVA P ϭ 0.056. groups according to UAER (1st group: Ͻ5 116.2 Ϯ 12.6, 127.2 Ϯ 16.9, 131.9 Ϯ Nighttime diastolic blood pressure ␮g/min; 2nd group: Ն5–10 ␮g/min; 3rd 22.0, and 123.4 Ϯ 14.1 mmHg for night- levels were lower in the 1st group (UAER group: Ն10–15 ␮g/min; and 4th group: time values (ANOVA P ϭ 0.008). The Ͻ5 ␮g/min) compared with the 3rd Ն15–20 ␮g/min) (Table 1). Patients in post hoc analyses showing the differences group (Ն10–15 ␮g/min): 67.5 Ϯ 9.4 vs. the 4th UAER group had longer diabetes between specific groups are depicted in 75.9 Ϯ 14.4 mmHg (P ϭ 0.05). However, duration than patients in the 1st group. In Fig. 2. The 24-h, daytime, and nighttime no difference was observed for the 2nd terms of laboratory characteristics, pa- pulse pressures also increased across the and 4th groups: 72.7 Ϯ 7.9 and 67.2 Ϯ tients in the 1st group of UAER had the 1st, 2nd, 3rd, and 4th UAER groups, re- 10.4 mmHg for 24-h values (P Ͼ 0.05). lowest levels of total cholesterol. The spectively: 48.4 Ϯ 6.0, 54.5 Ϯ 11.2, There were no differences among the prevalence of diabetic retinopathy in- 58.8 Ϯ 15.6, and 57.6 Ϯ 8.0 mmHg for groups, respectively, regarding the other creased progressively from the 1st to the 24-h values (ANOVA P ϭ 0.003); 48.1 Ϯ diastolic blood pressure parameters: 4th UAER group (27.3, 43.8, 45.5, and 6.2, 54.7 Ϯ 11.2, 59.9 Ϯ 16.1, and 74.5 Ϯ 8.5, 77.9 Ϯ 7.8, 80.2 Ϯ 13.4, and 66.7%; P ϭ 0.029 for trend). Mean age, 57.8 Ϯ 7.7 mmHg for daytime values 73.1 Ϯ 7.6 mmHg for 24-h values BMI, proportion of male sex, smokers, (ANOVA P ϭ 0.001); and 48.6 Ϯ 7.6, (ANOVA P ϭ 0.122); 77.9 Ϯ 9.2, 80.3 Ϯ and hypertensive patients were similar in 54.5 Ϯ 12.3, 56.0 Ϯ 14.3 and 56.2 Ϯ 8.4, 82.6 Ϯ 13.1, and 75.5 Ϯ 8.9 mmHg

1726 DIABETES CARE, VOLUME 28, NUMBER 7, JULY 2005 Leita˜o and Associates

Figure 2—Ambulatory systolic blood pressure levels (A), pulse pressure (B), and systolic blood pressure loads (C) according to UAER. *P Ͻ 0.05 for 1st vs. 2nd and 3rd groups; †P Ͻ 0.05 for 1st vs. 2nd, 3rd, and 4th groups. I, 1st group (UAER Ͻ5 ␮g/min); II, 2nd group (UAER Ն5–10 ␮g/min); III, 3rd group (UAER Ն10–15 ␮g/min); IV, 4th group (UAER Ն15–20 ␮g/min). for daytime blood pressure (ANOVA 24-h systolic blood pressure (R ϭ 0.45, UAER level disclosed that patients with ϭ 2 ϭ ϭ Ն ␮ P 0.263); 7.4% (range 0–73.2), 16.6% Ra 0.16; P 0.004), daytime systolic UAER 5 g/min already had a worse ϭ 2 ϭ (0–89.1), 17.1% (0–94.4), and 7.4% blood pressure (R 0.45, Ra 0.15; cardiovascular risk profile. However, be- (0–58.4) for 24-h blood pressure loads P ϭ 0.003), nighttime systolic blood pres- cause of the limited number of patients in ϭ ϭ 2 ϭ ϭ Ն ␮ (ANOVA P 0.076); 6.8% (0–75.5), sure (R 0.44, Ra 0.15; P 0.017), the 4th group (UAER 15–20 g/min, 14.0% (0–56.1), 10.0% (0–96.4), and 24-h systolic blood pressure loads (R ϭ n ϭ 10), some of blood pressure differ- 2 ϭ ϭ 3.6% (0–57.4) for daytime blood pres- 0.47, Ra 0.17; P 0.001), daytime ences between this and the 1st group sure loads (ANOVA P ϭ 0.399); and systolic blood pressure loads (R ϭ 0.47, did not reach conventional statistical 2 ϭ ϭ 8.3% (0–86.7), 14.5% (0–96.7), 17.9% Ra 0.17; P 0.001), nighttime systolic significance. ϭ 2 ϭ (0–100.0), and 4.7% (0–60) for night- blood pressure loads (R 0.42, Ra In a previous study using office blood time blood pressure loads (ANOVA P ϭ 0.13; P ϭ 0.014), 24-h pulse pressure pressure measurements and only one ϭ 2 ϭ ϭ 0.356). Analyzing the nocturnal blood (R 0.44, Ra 0.15; P 0.031), and morning UAER sample, higher blood ϭ 2 ϭ pressure descent, there was no difference daytime pulse pressure (R 0.45, Ra pressure levels, as well as an increased among the groups, respectively, in the 0.15; P ϭ 0.017). This was not true for prevalence of diabetic retinopathy, were prevalence of nondippers for systolic mean nighttime pulse pressure (R ϭ 0.40, observed in type 2 diabetic patients with 2 ϭ ϭ blood pressure (64.9, 71.9, 54.5, and Ra 0.11; P 0.079). Similar results high-normal albuminuria (12.5–30 mg/l) 40.0%; P ϭ 0.289) and diastolic blood were obtained with alternative models in (19). A similar association between UAER pressure (35.1, 43.8, 54.5, and 30.0%; which cholesterol levels were replaced by and blood pressure levels was reported in P ϭ 0.585). triglycerides (data not shown). healthy subjects (13). Previous studies in Multivariate regression analyses were nondiabetic hypertensive (20) and nor- performed with log-UAER as the depen- CONCLUSIONS — In this sample of moalbuminuric type 1 diabetic (14) pa- dent variable. Diabetes duration, choles- normoalbuminuric type 2 diabetic patients showed high-normal levels of terol, fasting plasma glucose and serum tients, a positive correlation was observed albuminuria were associated with in- creatinine were included in the model as between UAER and systolic blood pres- creased ambulatory blood pressure levels. independent variables. In each regression sure indexes. Furthermore, these patients In nondiabetic hypertensive patients, model, only one blood pressure parame- had more retinopathy and increased left high-normal UAER was found to be re- ter was included as an independent vari- ventricular posterior wall thickness. lated to left ventricular hypertrophy (20). able. UAER remained correlated with Stratification of the patients according Furthermore, high-normal levels of albu-

DIABETES CARE, VOLUME 28, NUMBER 7, JULY 2005 1727 Albuminuria and ambulatory blood pressure minuria (Ն4.8 ␮g/min) have been re- logia (PRONEX), Conselho Nacional de Des- Kidney Dis 13:14–16, 1989 ported to be an independent risk factor envolvimento Cientı´fico e Tecnolo´gico 11. Mogensen CE: Microalbuminuria and hy- for coronary artery disease and all causes (CNPq), and Fundo de Incentivo a Pesquisa pertension with focus on type 1 and type of mortality in a 9-year cohort study of (FIPE) do Hospital de Clıńicas de Porto Alegre. 2 diabetes. J Intern Med 254:45–66, 2003 subjects from Copenhagen (21). C.B.L. was the recipient of a scholarship from 12. Clement DL, De Buyzere ML, De Bacquer the Fundaçaõ de Coordenaçaõ de Aperfeiçoa- DA, de Leeuw PW, Duprez DA, Fagard The association of UAER with dia- mento de Pessoal de Nı´vel Superior (CAPES), RH, Gheeraert PJ, Missault LH, Braun JJ, betic retinopathy and left ventricular wall and L.H.C. was the recipient of a scholarship Six RO, Van Der Niepen P, O’Brien E: thickness was probably related to in- from the Programa de Apoio a` Instalaçaõ de Prognostic value of ambulatory blood- creased systolic blood pressure in these Doutores–ProDoc (CAPES). pressure recordings in patients with patients. The increased systolic blood pres- The authors thank Dr. Sandra P. Silveiro for treated hypertension. N Engl J Med 348: sure observed could be caused by decreased revising this manuscript. 2407–2415, 2003 compliance in the major arteries. In fact, we 13. Cubeddu LX, Hoffmann IS, Aponte LM, observed higher pulse pressure levels in Nunez-Bogesits R, Medina-Suniaga H, this sample of patients. Higher pulse pres- References Roa M, Garcia RS: Role of salt sensitivity, sure was also reported in type 2 diabetic 1. Parving HH, Oxenboll B, Svendsen PA, blood pressure, and hyperinsulinemia in Christiansen JS, Andersen AR: Early de- determining high upper normal levels of patients with more advanced diabetic ne- tection of patients at risk of developing urinary albumin excretion in a healthy phropathy (micro- and macroalbumin- diabetic nephropathy: a longitudinal adult population. Am J Hypertens 16:343– uria) and proliferative retinopathy (22). study of urinary albumin excretion. Acta 349, 2003 The data concerning the current sam- Endocrinol (Copenh) 100:550–555, 1982 14. Poulsen PL, Ebbehoj E, Hansen KW, Mo- ple of type 2 diabetic patients, along with 2. Viberti GC, Hill RD, Jarrett RJ, Argyro- gensen CE: 24-h blood pressure and au- previous data on type 1 diabetic patients poulos A, Mahmud U, Keen H: Mi- tonomic function is related to albumin and nondiabetic healthy and hyperten- croalbuminuria as a predictor of clinical excretion within the normoalbuminuric sive subjects, suggest that albuminuria nephropathy in insulin-dependent diabe- range in IDDM patients. Diabetologia 40: seems to be a continuous risk marker for tes mellitus. Lancet 1:1430–1432, 1982 718–725, 1997 the development of target organ damage 3. Mogensen CE, Christensen CK: Predict- 15. World Health Organization: Diabetes Mel- ing diabetic nephropathy in insulin-de- litus: Report of a WHO Study Group. Ge- (such as retinopathy, left ventricular hy- pendent patients. N Engl J Med 311:89– neva, World Health Org., 1985 (Tech. pertrophy, and coronary artery disease) 93, 1984 Rep. Ser., no. 727), p. 1–113 and death. However, clinicians need a 4. Mogensen CE: Microalbuminuria pre- 16. Canani LH, Gerchman F, Gross JL: In- precise reference value to adequately dicts clinical proteinuria and early mortal- creased familial history of arterial hyper- guide patients’ treatment. UAER Ͼ10 ␮g/ ity in maturity-onset diabetes. N Engl J tension, coronary heart disease, and renal min has been associated with micro- and Med 310:356–360, 1984 disease in Brazilian type 2 diabetic pa- macroalbuminuria on Cox regression 5. Molitch ME, DeFronzo RA, Franz MJ, tients with diabetic nephropathy. Diabetes analysis in type 1 and type 2 diabetic pa- Keane WF, Mogensen CE, Parving HH, Care 21:1545–1550, 1998 tients (8,9). Our data indicate that pa- Steffes MW, American Diabetes Associa- 17. Quinones MA, Douglas PS, Foster E, tients with even lower levels of UAER tion: Nephropathy in diabetes (Position Gorcsan J 3rd, Lewis JF, Pearlman AS, Statement). Diabetes Care 27 (Suppl. 1): Rychik J, Salcedo EE, Seward JB, Steven- are at risk. S79–S83, 2004 son JG, Thys DM, Weitz HH, Zoghbi WA, It is important to point out that our 6. Forsblom CM, Groop PH, Ekstrand A, Tot- Creager MA, Winters WL Jr, Elnicki M, results are based on a cross-sectional terman KJ, Sane T, Saloranta C, Groop L: Hirshfeld JW Jr, Lorell BH, Rodgers GP, study and that our conclusions are limited Predictors of progression from normoalbu- Tracy CM, Weitz HH, the American Soci- to the observation of an association be- minuria to microalbuminuria in NIDDM. ety of Echocardiography, the Society of tween cardiovascular risk factors and Diabetes Care 21:1932–1938, 1998 Cardiovascular Anesthesiologists, the UAER. The establishment of a cause and 7. Gall MA, Hougaard P, Borch-Johnsen K, Society of Pediatric Echocardiography: effect relationship will require prospec- Parving HH: Risk factors for development of ACC/AHA clinical competence statement tive cohort studies. incipient and overt diabetic nephropathy in on echocardiography: a report of the In conclusion, normoalbuminuric patients with non-insulin dependent diabe- American College of Cardiology/Ameri- tes mellitus: prospective, observational can Heart Association/American College type 2 diabetic patients with UAER Ն5 ␮ study. BMJ 314:783–788, 1997 of Physicians-American Society of Inter- g/min should be considered to have a 8. Murussi M, Baglio P, Gross JL, Silveiro SP: nal Medicine Task Force on Clinical Com- high cardiovascular risk profile for which Risk factors for microalbuminuria and petence. J Am Soc Echocardiogr 16:379– intense treatment is indicated, especially macroalbuminuria in type 2 diabetic pa- 402, 2003 with regard to blood pressure control, so tients: a 9-year follow-up study. Diabetes 18. Chantler C, Barratt TM: Estimation of glo- as to reduce both cardiovascular mortality Care 25:1101–1103, 2002 merular filtration rate from plasma clear- and progression to more advanced stages 9. The Microalbuminuria Collaborative ance of 51-chromium edetic acid. Arch of diabetic nephropathy. Prospective in- Study Group: Predictors of the develop- Dis Child 47:613–617, 1972 tervention studies are needed to confirm ment of microalbuminuria in patients 19. Torffvit O, Agardh E, Agardh CD: Albu- whether these patients will in fact benefit with type 1 diabetes mellitus: a seven-year minuria and associated medical risk fac- prospective study. Diabet Med 16:918– tors: a cross-sectional study in 451 type from such an aggressive approach. 925, 1999 II (noninsulin-dependent) diabetic pa- 10. Jarrett RJ: Hypertension in diabetic patients. Part 2. J Diabet Complications 5:29– Acknowledgments— This study was par- tients and differences between insulin-de- 34, 1991 tially supported by the Projeto de Nu´ cleos de pendent diabetes mellitus and non- 20. Dell’Omo G, Penno G, Giorgi D, Di Bello Exceleˆncia do Ministe´rio de Cieˆncia e Tecno- insulin-dependent diabetes mellitus. Am J V, Mariani M, Pedrinelli R: Association

1728 DIABETES CARE, VOLUME 28, NUMBER 7, JULY 2005 Leita˜o and Associates

between high-normal albuminuria and H, Appleyard M, Jensen JS: Very low lev- 22. Knudsen ST, Poulsen PL, Hansen KW, risk factors for cardiovascular and renal els of microalbuminuria are associated Ebbehoj E, Bek T, Mogensen CE: Pulse disease in essential hypertensive men. with increased risk of coronary heart dis- pressure and diurnal blood pressure vari- Am J Kidney Dis 40:1–8, 2002 ease and death independently of renal ation: association with micro- and macro- 21. Klausen K, Borch-Johnsen K, Feldt-Ras- function, hypertension, and diabetes. Cir- vascular complications in type 2 diabetes. mussen B, Jensen G, Clausen P, Scharling culation 110:32–35, 2004 Am J Hypertens 15:244–250, 2002

DIABETES CARE, VOLUME 28, NUMBER 7, JULY 2005 1729