• Abstract and Figures
• Public Full-text

Prevalence of Abnormal Urinary Albumin Excretion Rate in Hypertensive Patients with Impaired Fasting Glucose and Its Association with Cardiovascular Disease ʈ Rafael Marı´n,* Pilar Rodrı´guez,† Salvador Tranche,‡ Jose´Redo´n,§ Francisco Morales-Olivas, Alberto Galgo,¶ Miguel A´ ngel Brito,** Javier Mediavilla,†† Jose´Vicente Lozano,‡‡ Claudia Filozof;§§ and the MAGAL group *Nephrology Service, Hospital Universitario Central de Asturias, and ‡El Cristo Health Centre, Oviedo, †Fingoi Health ʈ Centre, Lugo, §Internal Medicine Service, Hospital Clı´nico, Department of Pharmacology, University of Valencia, and ‡‡Valencia Health Centre, Valencia, ¶Espronceda Health Centre and **Endocrinology, Hospital Puerta de Hierro, Madrid, ††Pampliega Health Centre, Burgos, and §§Scientific Department, Cardiovascular Section, Bristol-Myers- Squibb, Madrid, Spain The prevalence and significance of microalbuminuria in hypertensive patients with impaired fasting glucose (IFG) has received very little attention. A total of 10,320 hypertensive patients who attended primary care centers were enrolled in this study, and the final analysis was done in 7625 patients: 1459 without IFG (plasma glucose <100 mg/dl), 3010 with IFG (plasma glucose >100 mg/dl and <126 mg/dl), and 3156 with type 2 diabetes (plasma glucose >126 mg/dl). Microalbuminuria was determined using the Micro Albustix reactive strip from Bayer (high urinary albumin excretion [UAE]: Albumin/creatinine ratio >3.4 mg/mmol). The proportion of patients with high UAE was 39.4, 48.3, and 65.6%, respectively, in the three groups (P < 0.01 for the trend). The differences in UAE between the group with IFG and the group with normal fasting glucose persisted after adjustment for age, gender, systolic BP, fasting plasma glucose, and cardiovascular comorbidity (odds ratio 1.74; 95% confidence interval 1.08 to 2.80). Hypertensive patients with IFG and high UAE showed a higher prevalence of ischemic heart disease, cardiac insufficiency, left ventricular hypertrophy, atrial fibrillation, and renal insufficiency than the group with normal UAE. Global prevalence of cardiovascular conditions was 30.4% in the group with high UAE compared with 21.4% in the group with normal UAE (odds ratio 1.60; 95% confidence interval 1.31 to 1.95). It is concluded that almost half of hypertensive patients with IFG have high UAE and a higher prevalence of associated cardiovascular involvement and renal insufficiency. J Am Soc Nephrol 17: S178–S188, 2006. doi: 10.1681/ASN.2006080912 uring the past 20 yr, epidemiologic evidence has in- Only a few studies have assessed the prevalence of mi- dicated that microalbuminuria is predictive of in- croalbuminuria in the general population (3,11,12), which D creased renal and cardiovascular morbidity and mor- ranges from 7.2 to 7.8% and can reach 16% in hypertensive tality not only in patients with diabetes but also in the general patients and even 29 to 31% in patients with diabetes. In the population (1–4). This association has proved to be indepen- Australian Diabetes, Obesity, and Lifestyle Study (AusDiab), dent of traditional risk factors such as age, BP, dyslipidemia, or the prevalence of microalbuminuria in the subgroup with dia- left ventricular hypertrophy (LVH) (5,6). These data led the betes was 21% (13). This study showed a prevalence of mi- latest report of the Joint National Committee in the United croalbuminuria of 5.1% in patients with normal fasting glucose States to accept microalbuminuria as a first-degree cardiovas- (NFG), which increased to 9.3% in those with impaired fasting cular risk factor (7). glucose (IFG; Ն110 and Ͻ126 mg/dl). This continuous and The positive association between the albumin/creatinine ra- direct relationship between urinary albumin excretion rate tio and cardiovascular and renal morbidity and mortality is (UAE) and glucose tolerance already had been reported many Ͼ evident from very low levels ( 5 mg/g creatinine), well below years before in a study that involved a very low number of the standard microalbuminuria threshold (30 mg/g creatinine) patients (14). (8,9). Recently, post hoc studies showed that the reduction in Cardiovascular risk that is associated with impaired carbo- microalbuminuria levels by therapy is associated with a reduc- hydrate metabolism (15) and microalbuminuria seems very tion in cardiovascular events (8,10). likely to increase in hypertensive patients, although this has not yet been assessed. The main objective of our study was to assess the prevalence of elevated UAE (eUAE) in hypertensive pa- Address correspondence to: Dr. Rafael Marı´n, Servicio de Nefrologı´a, Hospital Universitario Central de Asturias, C/Celestino Villamil s/n, 33006, Oviedo, tients who had IFG and attended primary care units and to Spain. Phone: ϩ34-985-108000; Fax: ϩ34-985-108015; E-mail: [email protected] compare it with the prevalence of eUAE in hypertensive pa- Copyright © 2006 by the American Society of Nephrology ISSN: 1046-6673/1712-0178 J Am Soc Nephrol 17: S178–S188, 2006 Microalbuminuria in Hypertensive Patients with IFG S179 tients with type 2 diabetes and with those with NFG. The (lower or higher) would be estimated with more accuracy. Under this secondary objectives were to analyze the association between assumption and taking into consideration a clinically relevant differ- eUAE and other cardiovascular risk factors and to evaluate its ence of 4 to 5 units in UAE between the IFG and type 2 diabetes groups, association with cardiovascular morbidity. a total of 1600 to 2500 patients were necessary in each group to estimate this prevalence with an accuracy of between 2 and 3%, accepting an ␣ Ͻ 0.05 and a ␤ Ͻ 0.10 errors in a two-tailed P evaluation. Nevertheless, Materials and Methods with the intention of analyzing subgroups and allowing for a 25% loss MicroAluminuria en pacientes con glucemia Basal ALterada (MA- in the number of cases, the total number of patients was increased to GAL) is a multicenter, cross-sectional, observational study that was that mentioned above. carried out by 1723 primary care physicians between September and December 2005. Each physician consecutively enrolled six patients with Statistical Analyses essential hypertension in the following proportions: Three with IFG The variables are presented either as frequency and 95% CI, mean Ն Ͻ ( 100 mg/dl and 126 mg/dl), two with type 2 diabetes, and one with (SD), or median (25th to 75th percentiles) according to the distribution Ͻ NFG ( 100 mg/dl). of the variable. The t test was used for two-group comparisons, and A total of 10,320 patients were enrolled. The inclusion criteria were ANOVA was used for comparisons with more than two groups. The Ͼ Յ age 18 and 80 yr, previous diagnosis of hypertension and BP Tukey test was used to establish the difference between pairs of means Ն 140/90 mmHg, or receiving antihypertensive medication. Patients when necessary. The association between qualitative variables was with type 1 diabetes were excluded, as were pregnant or breast-feeding 2 evaluated using the ␹ or Fisher exact test. Differences were considered women; patients with urinary infection, pyuria, hematuria, or fever; significant at P Ͻ 0.05. and those who had done strenuous physical exercise during the 24 h To evaluate the factors that explained the presence of a high UAE, we before determination of microalbuminuria. constructed a multiple logistic regression model using the variables The study was approved by the Ethics and Clinical Trials Committee that were statistically significant in the previously mentioned terms. of the Hospital Central de Asturias. Informed consent of the patient Traditional risk factors also were included. The odds ratio (OR) of these was obtained in all cases. variables was reported, together with their CI and level of significance. The analysis was carried out using SAS for Windows (version 9.1; SAS Study Protocol Institute, Cary, NC). For each patient, the following were obtained: Demographic data (age and gender); clinical data (family history of premature cardiovas- Results cular disease, personal history of smoking and dyslipidemia, and pres- The final analysis was made in 7625 patients: 1459 in the NFG ence of associated cardiovascular disease [ischemic heart disease, car- group, 3010 in the IFG group, and 3156 in the type 2 diabetes diac insufficiency, cerebrovascular disease, and peripheral artery group. A total of 2695 (26.1%) were excluded for one of the disease]); and a physical examination that included weight, height, following causes of erroneous classification: Not fulfilling the body mass index (BMI), abdominal waist circumference, and systolic ϭ (SBP) and diastolic BP (DBP). Plasma creatinine, glucose, HbA , total age condition (n 357), being classified in more than one 1c ϭ cholesterol, LDL cholesterol, HDL cholesterol, and triglycerides were metabolism group (n 400), having normal BP and not receiv- ϭ measured after at least8hoffasting. UAE was determined in an ing antihypertensive medication (n 721), being included in isolated sample of first-morning urine by the semiquantitative method the normal carbohydrate metabolism group or in the IFG group using the Micro Albustix reactive strip (Bayer, Milano, Italy). For cal- and receiving antidiabetes medication (n ϭ 6 and n ϭ 674, culation of renal function, the GFR was estimated (ml/min per 1.73 m2) respectively), being in the type 2 diabetes group and having according to the abbreviated equation of the Modification of Diet in NFG with no antidiabetes medication (n ϭ 30), being in the IFG Renal Disease (MDRD) study (16). An electrocardiogram was used to group and presenting fasting glucose Ն126 mg/dl (n ϭ 318), or examine the presence of LVH (Sokolow and/or Cornell criteria) and being in the normal carbohydrate metabolism group and hav- auricular fibrillation. Similarly, we recorded patient medication data: ing HbA Ͼ7.5% (n ϭ 189). Antihypertensive drugs, antidiabetics, and lipid-lowering and anti- 1c The characteristics of the patients who were included in the platelet agents.

Load more

  11

Copy Link