Lorenzo Fácila, MD, PhD Cardiology Department, Consorcio Hospitalario Provincial de Castellón, Avda, Dr Clara 19 12002, Castellón de la Plana, Spain E-mail: email@example.com
The association between renal dysfunction and cardiovascular risk in patients with hypertension and left ventricular hypertrophy (LVH) has not been specifically studied. The aim of this study was to analyze the association between renal function and the presence of cardiovascular disease in this group of patients. Hypertensive patients with electrocardiographic criteria for LVH were recruited in cardiology outpatient clinics from April 2003 until November 2004. Epidemiologic variables were determined, together with an estimation of the glomerular filtration rate by means of the Modification of Diet in Renal Disease (MDRD) and Crockoft-Gault equations. The population was classified according to the kidney disease stages of the National Kidney Foundation. A total of 3962 patients were included in the study, 47.6% of which were female, with a mean age of 67.2 years. The prevalence of established cardiovascular disease was higher in patients with a depressed glomerular filtration rate (68.3% vs 54.9%; P<.001). After adjusting for age, sex, body mass index, diabetes, smoking habits, and systolic and diastolic blood pressures, the stage of renal function was an independent predictor of the presence of cardiovascular disease (odds ratio, 1.5 [confidence interval, 1.19–2.02]; 2.1 [1.55–2.89]; and 2.6 [1.52–4.42], respectively, for stages 2, 3, 4–5, compared with stage 1). In hypertensive patients with electrocardiographic criteria for LVH, the determination of the glomerular filtration rate by the MDRD or Crockoft equations is easy and identifies a progressive and independent increase in cardiovascular risk.
Detection of silent cardiac damage in the form of left ventricular hypertrophy (LVH) is a key point in the evaluation of the hypertensive patient, combined with blood pressure (BP) values and other cardiovascular (CV) risk factors and markers.1,2
Recent studies have shown an increase in CV risk with the impairment of renal function in hypertension.3–6 As a result of this evidence, renal function has become one of the more relevant risk markers in hypertensive patients and is estimated to be associated with an increase in risk similar to that attributed to diabetes.
Renal function can be assessed by different methods. Isolated serum creatinine levels, although the more extended method, may incorporate data biased by the patient’s age, sex, and muscle mass. The development and validation of formulas to calculate the glomerular filtration rate (GFR) has outlined the importance of mild and moderate renal dysfunction as an independent CV risk factor. The latest version of the National Kidney Foundation (NKF) guidelines recommends the assessment of renal function by means of indirect calculation of the GFR with formulas accounting for serum creatinine, sex, weight, and age.7 The more extended formulas include the Crockoft-Gault8 and the simplified version of the Modification of Diet in Renal Disease (MDRD) equation.9 The latter equation, although more complex to calculate, allows for GFR estimation in more situations because it does not require knowledge of the patient’s weight. The MDRD equation was validated in a population with a high rate of renal dysfunction, thus it is especially recommended for patients with a GFR <60 mL/min.9
There are few studies of hypertension that evaluate the association between renal function and CV risk,10,11 and none have demonstrated this association in a wide cohort of hypertensive patients with LVH.
The aim of this study, therefore, was to assess the relationship between GFR, estimated by means of validated formulas, and the presence of CV disease (CVD) in hypertensive patients with LVH.
The VIIDA study is a cross-sectional, multicenter, epidemiologic trial, conducted in outpatient cardiology clinics across Spain. It was designed by the Hypertension and Outpatient Cardiology sections of the Spanish Society of Cardiology and approved by an independent clinical research ethics committee. Two hundred cardiologists participated in the study. Patients were enrolled from April 2003 through November 2004 (on 3 different days during the period, at 6-month intervals). After obtaining written consent, we included patients with arterial hypertension, regardless of their age or CVD history. We collected data from clinical history records and, when necessary, patients underwent complementary examinations.
We consecutively screened hypertensive patients attending the clinics. An initial questionnaire elicited data on history of hypertension, age, sex, weight, build, CV risk factors, and the presence of CVD. Presence of LVH was defined by Sokolow-Lyon electrocardiographic criteria (R wave in V5 through V6+S wave in V1 >35 mm), Cornell voltage criteria (R wave in aVL+S wave in V3 >20 mm in women or >28 mm in men), or both. Patients presenting electrocardiographic LVH underwent a complete examination to obtain demographic and anthropometric data, CV risk factors, and information about the presence or history of CVD (myocardial infarction, angina pectoris, intermittent claudication, heart failure, and stroke). Biochemical data were obtained from samples taken in the 6 months prior to data collection and, if unavailable, analyses were performed at the time of data collection. BP was measured according to standardized norms, with a mercury sphygmomanometer. Patients were seated and, after 5 minutes of rest, BP was measured 3 times at 2-minute intervals. We calculated the mean of the last 2 measurements and this was considered to be the patient’s BP. The presence of atrial fibrillation or flutter was recorded.
Renal Function and Stages
Serum creatinine was obtained from the most recent analysis performed in the previous 6 months, and the GFR was calculated with the simplified MDRD equation9 and the Cockroft-Gault equation,8 corrected for body surface area. The population was divided into 5 groups according to the GFR stages of the NKF12 using the values of the Crockoft-Gault formula. Due to the small number of patients included in groups 4 and 5, they were combined into 1 category (<30 mL/min/1.73 m2) for subsequent analyses.
Continuous variables were presented as mean (standard deviation) or median (interquartile range) as appropriate. Their comparison among the groups with and without renal insufficiency was performed using a Student t test or Mann–Whitney U test for skewed variables. Discrete variables were presented as percentages and compared by chi-square test. The contribution of each outcome variable was assessed using analysis of variance, with P values corrected according to the Bonferroni method. In all cases, the variable distribution was assessed by means of theoretical methods, and Leven’s test was employed to assess the homogeneity of variances.
Logistic regression analysis was used to select those variables that were significantly associated with the renal function distribution and the presence of CVD. The threshold for retaining a variable in the multivariate model was set at P<.1. A 2-sided P value of <.05 was considered to be statistically significant for all analyses. All statistical analyses were performed using SPSS version 11.0 (Chicago, IL).
During the recruitment period, 16,123 patients were screened and 4037 had electrocardiographic criteria for LVH (25.04%). Of those, 3962 (98.1%) were analyzed. The mean age was 67.2 years and 52.4% were males. The mean body mass index (BMI) was 27.8 kg/m2; 28.5% were overweight (BMI >25 kg/m2 and <30 kg/m2) and 23.5% were obese (BMI >30 kg/m2). A total of 25.4% were diabetic with 91.6% having type 2 diabetes, 56.8% had dyslipidemia, and 13.8% were active smokers. The mean number of antihypertensive agents used was 2.26±1.13 with 55.4% of the people taking diuretics, 47.8%β-blockers, 37.5% calcium channel blockers, 43.7% angiotensin-converting enzyme inhibitors (ACEIs), 39.9% angiotensin receptor blockers (ARBs), and 0.4%α-blockers.
The distribution of the study population among the GFR categories was 12% with stage 1 (GFR ≥90 mL/min/1.73 m2), 43% with stage 2 (60–89 mL/min/1.73 m2), 41% with stage 3 (30–59 mL/min/1.73 m2), 3% with stage 4 (15–29 mL/min/1.73 m2), and 1% with stage 5 (<15 mL/min/1.73 m2).
The CVDs and the rest of the analytical data are depicted in Table I.
Table I. Baseline Characteristics of the Study Population
Continuous variables are expressed as mean ± standard deviation. Discrete variables are expressed as number (percentage). aIncludes acute myocardial infarction, angina pectoris, stroke, peripheral artery disease, and heart failure.
Age >60 y
Acute myocardial infarction
Peripheral artery disease
Systolic blood pressure
Diastolic blood pressure
Glomerular filtration rate
Association Between GFR and Baseline Characteristics in Hypertensive Patients With LVH
Patients with lower GFRs were older, more frequently female, and more likely to have diabetes. The association between these factors and GFR was linear, ie, the lower the GFR, the more frequently these factors appeared (Table II). On the other hand, lower rates of obesity, overweight, and smoking were found among patients with renal dysfunction.
Table II. Baseline Characteristics According to the National Kidney Foundation Kidney Disease Stages
GFR, mL/min/1.73 m2 (No.)
P for Trend
Abbreviations: BMI, body mass index; CVD, cardiovascular disease (includes acute myocardial infarction, angina pectoris, stroke, peripheral artery disease, and heart failure); GFR, glomerular filtration rate; NS, not significant; SD, standard deviation.
Age, mean ± SD, y
Male sex, %
Active smoker, %
Obesity (BMI >30 kg/m2), %
Overweight (BMI >25 kg/m2), %
Calcium channel blockers, %
Angiotensin-converting enzyme inhibitor, %
Angiotensin receptor blocker, %
Few differences were found in the antihypertensive treatments among the GFR groups. There was greater use of diuretics and ACEIs in the groups with lower GFRs, with no differences in the prescription of β-blockers, calcium channel blockers, ARBs, or α-blockers (Table II).
Association Between Renal Function and CVD
The prevalence of established CVD was higher among patients with lower GFRs. Heart failure, peripheral artery disease, atrial fibrillation, and cerebrovascular disease were all more frequent in patients with lower renal function status (Figure 1), with this association being linear, whereas there were no differences in patients with regard to coronary artery disease (Table III).
Table III. Prevalence of Cardiovascular Disease According to the National Kidney Foundation Kidney Disease Stages
GFR (mL/min/1.73 m2)
P for Trend
Abbreviation: GFR, glomerular filtration rate.
Cardiovascular disease, %
Coronary artery disease, %
Peripheral artery disease, %
Atrial fibrillation, %
Heart failure, %
In the multivariate analysis, after adjusting for age, sex, BMI, diabetes, systolic BP, diastolic BP, and smoking, renal function stage was an independent predictor of the presence of CVD for stage 2 vs the reference group (stage 1) (adjusted odds ratio [OR], 1.55; 95% confidence interval [CI], 1.19–2.02); stage 3 vs the reference group (adjusted OR, 2.11; 95% CI, 1.55–2.89); and groups 4 and 5 (GFR <30 mL/min/1.73 m2) vs the reference group (adjusted OR, 2.59; 95% CI, 1.52–4.42) (Table IV).
Table IV. Multivariate Analysis–Independent Predictors of Cardiovascular Disease
Abbreviations: CI 95%, 95% confidence interval; OR, odds ratio.
Age (per year)
Body mass index
Stage 4 or 5
Systolic blood pressure
To better assess the linear relationship between renal dysfunction and CVD, we further divided stages of mild and moderate renal function impairment, stages 2 and 3, into 4 categories. Figure 2 shows the linear risk gradient associated with GFR deterioration.
In this analysis of the observational study VIIDA, which included 3962 hypertensive patients with LVH, the association of renal dysfunction with CVD was examined. This association was continuous and progressive throughout the stages of renal function impairment and was independent of other predictors of CVD such as BP values, BMI, diabetes, sex, or age. The progressive effect was observed, even in early stages of renal dysfunction, and increased with the deterioration of the renal function.
There are few studies assessing the relationship between renal function and CVD in hypertensive patients with LVH. The observational study Evaluación del Riesgo de un Primer Ictus en la Población Hipertensa Española en Atención Primaria (ERIC-HTA), carried out in hypertensive patients from general practice clinics, describes independent correlations among renal dysfunction, LVH, and established CVD.11 However, unlike our study, this study’s population had a lower prevalence of established CVD (26.4%), which is expected from a sample of patients recruited from general practice clinics.
In the Kaiser Registry,6 which included more than 1 million US patients, those with low estimated GFR at baseline were older than those with an estimated GFR of at least 60 mL/min/1.73 m2 (MDRD formula), and there was a wider representation of persons from minority groups among the population with low estimated GFR. Compared with the group with an estimated GFR of at least 60 mL/min/1.73 m2, the groups with a reduced estimated GFR also had a higher prevalence of prior CVD, proteinuria, diabetes, hypertension, hypoalbuminemia, prior hospitalizations, and other coexisting illnesses.
The association between CVD and renal dysfunction, the so-called cardiorenal syndrome, was described many years ago,13 and thereafter a narrow bidirectional relationship between the 2 organs has been recognized. Renal dysfunction, on the one hand, promotes CV damage. For example, a recent analysis from the Framingham study14 shows that patients with GFR <60 mL/min/1.73 m2 had a higher prevalence of CV risk factors (ie, obesity, dyslipidemia, hypertension, and diabetes) and worse control of BP values and glycosylated hemoglobin. In a meta-analysis with more than 22,000 patients, with a follow-up of 99 months, renal dysfunction was associated with a 2-fold risk of coronary artery disease and a 3-fold risk of stroke and all-cause mortality.15 In secondary prevention, renal function impairment also behaves as a strong independent predictor of mortality or CV events after an acute myocardial infarction,16 an episode of acute coronary syndrome without ST17 or with ST-segment elevation,18 and in patients with heart failure.19
On the other hand, CVD is also associated with a deterioration of renal function. Patients who develop heart failure after an acute myocardial infarction experience a decrease of 12 mL/min of creatinine clearance unless treated.20 Both diagnostic and therapeutic interventions have a deleterious effect on renal function including iodated contrasts for coronary artery disease diagnosis, high doses of diuretics in heart failure, and concomitant use of ACEI and nonsteroidal anti-inflammatory drugs. In our study, 39.2% of patients were taking diuretics, 30.5% were taking ACEIs, and 27.4% were undergoing ARB treatment.
Renal and CVDs share a number of risk factors and often have parallel evolution. Patients with renal disease, apart from having high rates of CV risk factors, experience alterations in calcium–phosphorus metabolism,21 endothelial function, activation of inflammation, increased oxidative stress,22 activation of the renin-angiotensin system, and high homocysteine levels.23 All of these contribute to the development of atherosclerosis and worsen the CV prognosis.24
The present study assessed renal function in patients with CVD using 2 different equations to calculate the GFR. These equations, the simplified version of the MDRD,9 and the Cockroft-Gault,8 account for age, creatinine, and sex,7 and had an intraclass correlation of 0.89 (data not shown). The good correlation between the formulas allowed for the utilization of either, and we decided to use the latter because it was designed in a population similar to ours and because we had the necessary weight and height data. The simplified version of the MDRD is more reliable in populations with a higher degree of renal dysfunction for whom there are no anthropometric data.7
The main limitations of this study are those inherent to any observational nonrandomized study, which does not allow for the assessment of prevalence. The lack of data from patients with hypertension but without LVH does not allow for comparisons between these groups. However, the large size of the sample with a high prevalence of LVH and associated comorbidity renders the present results reliable and valuable for our purposes.
In hypertensive patients with LVH attending cardiology outpatient clinics, the GFR determination by means of either the Crockoft-Gault or the simplified MDRD equations allows for the identification of independent and progressive CV risk.
The VIIDA study was designed and endorsed by the Hypertension and Outpatient Cardiology sections of the Spanish Society of Cardiology and financed by an unrestricted investigational grant from Merck Sharp & Dohme, Spain.