Data were taken from the ENRICA (Study on Nutrition and Cardiovascular Risk in Spain), whose methods have been reported previously.14–17 Briefly, this is a cross-sectional study conducted from June 2008 to October 2010 with 12,948 individuals representative of the non-institutionalized population of Spain aged ≥18 years. Participants were selected by multistage clustered random sampling. The sample was first stratified by province and by size of municipality. Clusters were then selected randomly in 2 stages, municipalities and census sections. Finally, the households within each section were selected by random landline telephone dialing. Subjects in the households were selected proportionally to sex- and age-distribution of the population of Spain. Information was collected by telephone interview, face-to-face interview, physical examination, and by taking blood and urine samples in the households.

The study protocol was approved by the clinical research ethics committees of the Hospital Universitario La Paz in Madrid and the Hospital Clínic in Barcelona. All study participants gave written informed consent for the study.

Weight, height, and waist circumference were measured twice in each subject under standardized conditions using electronic scales (model Seca 841, precision to 0.1kg), portable extendable stadiometers (model Ka We 44 444Seca), and flexible, inelastic belt-type tapes. Body mass index (BMI) was calculated as weight in kg divided by squared height in m.

Blood pressure (BP) measurement was performed by certified trained personnel, using standardized procedures with validated automatic devices (model Omron M6) and cuffs of three sizes according to arm circumference. Two sets of BP readings were made separated by 90min. In each set, BP was measured 3 times at 1–2min intervals, after a 3–5min resting period. For analyses, BP was calculated as the mean of the last five readings. Study participants reported their smoking status. Physical activity was assessed with the questionnaire from the EPIC (European Prospective Investigation into Cancer) study in Spain,18 which combines physical activity at work and during leisure time into 4 levels (very active, moderately active, moderately inactive, and inactive). Individuals also reported physician-diagnosed CVD. All variables were defined according to international guidelines.19,20

Participants provided 12-hour fasting blood and spot urine samples at their homes. Laboratory determinations were performed centrally at the Center of Biological Diagnosis of the Hospital Clínic in Barcelona, using standard procedures and appropriate quality controls.14–17 Serum glucose, hemoglobin A1c, total cholesterol, high-density lipoprotein (HDL) cholesterol, and triglycerides were determined by enzymatic methods, and low-density lipoprotein (LDL) cholesterol by the Friedewald formula. Creatinine was determined by Jaffé, alcalin picrate by kinetic reaction, and microalbumin by polyethylene glycol-enhanced immunoturbidimetry.

CKD was defined according to current Kidney Disease: Improving Global Outcomes (KDIGO) guidelines using estimated glomerular filtration rate (eGFR) calculated with the CKD – Epidemiology Collaboration (CKD-EPI) equation, and albuminuria measured by the albumin-to-creatinine ratio (ACR).1 CKD was defined as either a low eGFR (<60mL/min/1.73m2) or an increased ACR (≥30mg/g). Subjects were classified according the KDIGO GFR 6-category staging (grades G1, G2, G3a, G3b, G4, and G5 corresponding to eGFR ≥90, 60–89, 45–59, 30–44, 15–29, and <15mL/min/1.73m2, respectively), and albuminuria 3-category staging (A1, A2, and A3 corresponding to ACR <30, 30–300, and >300mg/g, respectively). Then, 6 CKD grades were established (G1 CKD for G1 eGFR with increased albuminuria A2 or A3, G2 CKD for G2 eGFR with increased albuminuria A2 or A3, G3a CKD for G3a eGFR with either normoalbuminuria A1 or increased albuminuria A2 or A3, G3b CKD for G3b eGFR with either normoalbuminuria A1 or increased albuminuria A2 or A3, G4 CKD for G4 eGFR with either normoalbuminuria A1 or increased albuminuria A2 or A3, and G5 CKD for G5 eGFR with either normoalbuminuria A1 or increased albuminuria A2 or A3. Subjects were also classified according the prognostic KDIGO 6 × 3 table (6 eGFR categories × 3 albuminuria categories) to estimate percentages of subjects at moderately increased risk, high risk, and very high risk of complications from CKD.

A total of 10 classic CVD risk factors were assessed: age, hypertension, general obesity, abdominal obesity, current smoking, diabetes, increased LDL cholesterol, low HDL-cholesterol, hypertriglyceridemia, and sedentary habit. Risk age was considered >65 years for women and >55 years for men. Hypertension was defined as systolic BP ≥140mmHg, and/or diastolic BP ≥90mmHg, and/or on current antihypertensive drug treatment.19 Obesity was defined as BMI was ≥30kg/m2, and abdominal obesity as waist circumference was >88cm in women and >102cm in men. Diabetes was defined as glucose ≥126mg/dL or glycosylated hemoglobin ≥6.5%, or using oral antidiabetic drugs or insulin.20 Increased LDL cholesterol was >115mg/dL, low HDL-cholesterol was <46mg/dL in women and <40mg/dL in men, and hypertriglyceridemia was >150mg/dL, and hypercholesterolemia was total cholesterol >190mg/dL, or receiving lipid-lowering drugs.19 Sedentary habit was considered when physical activity was less than moderate. In participants with no previous CVD, the 10-year risk of CVD death was estimated using the SCORE (Systematic COronary Risk Evaluation) equation for low-risk European regions, which includes Spain, being high risk considered when risk was ≥5%.19

Of the 12,948 study participants, the following were excluded because lacking data: 1251 on BP, sociodemographic or clinical variables, or laboratory values other than kidney-related parameters, and 192 on serum creatinine or albuminuria. Thus, 11,505 individuals were valid for the analyses (88.9% of the total sample).

Analyses accounted for the complex sampling design, so individual observations were weighted to reconstruct the Spanish population, and the variances were corrected to obtain appropriate 95% confidence intervals (CI) for the main results. The Student t test and ANOVA were used to compare means and the χ2 test was used to compare proportions. Trends in prevalence of CKD across stage, age, and number of CV risk factors were also calculated (P for trend).

Finally, we performed multivariate logistic regression of factors associated with CKD. Sociodemographic variables and cardiovascular risk factors that were statistically significant in univariate analyses or considered a priori clinically relevant were selected for multivariable analyses. Results were summarized with odds ratios and their 95% confidence intervals [CI].

Statistical significance was set at 2-sided P<0.05. The analyses were performed with the SPSS package 18.0 (IBM, Armonk, NY).

The prevalence of CKD worldwide is 11.8% in women and 0.4% in men.7 There are huge geographic differences (from 4.1% in Saudi-Arabian women to 25.7% in men from El Salvador), and by income status (8.6% in men and 9.6% in women in high-income countries vs. 10.6% in men and 12.5% in women in low- and middle-income countries). In China the estimated CKD prevalence in adults was 10.8%.8

In Europe CKD prevalence ranges from 3.3% in Norway to 17.3% in northeast Germany.10 In Spain, the prevalence of CKD in the present study was higher than in a previous and much smaller study conducted in 42 municipalities (9.2%),21 probably reflecting changes over time and methodological differences between studies. In the United States, stage 1–4 CKD prevalence in adults >20 years was 13.1% in the early 2000s, and 13.7% in US Hispanics/Latinos.5,9 From the data available, we can derive that, methodological differences aside, the prevalence of CKD in Spain is comparable to that in the United States and higher than in many European countries.10

CVD is common in patients with CKD, and frequently underdiagnosed and undertreated.20 Current European guidelines on CVD prevention22 indicate that ‘microalbuminuria’ and/or a moderately reduced eGFR (30–60mL/min/1.73m2) confers high risk for CVD, and proteinuria and/or a severely reduced eGFR (<30mL/min/1.73m2) should to be considered as very high risk situations. CKD is associated with an increased risk of CVD, independent of conventional CVD risk factors, which are known to be more common in patients with CKD. In addition, inflammatory mediators and promoters of calcification induce vascular damage and may explain the association CKD–CVD even after adjustment for conventional risk factors.13,23

In agreement with previous population-based studies,5,8,9 prevalence of situations at high or very high risk according to KDIGO score where low. In our study the prevalence of an eGFR <30mL/min/1.73m2 was 0.20%, and that of proteinuria was 0.30%. Thus, most of adults showing CKD had situations at moderate risk of cardio-renal complications.

The prevalence of CKD was higher as the number of CVD risk factors increased. Most subjects had 2–6 CVD risk factors and a relatively moderate risk of having CKD. That's where lies the potential for prevention of CKD and its complications using both high-risk individual and moderate-risk population strategies. In particular, as in some other studies,5,8,9,24 people who are older, men, with hypertension, diabetes, low HDL-cholesterol, sedentary, or with CVD are at high risk and should be screened for CKD.

Main strengths of this study are the representativeness of adults at the countrywide level, large sample size, and the use of international methodological standards. This study has also some limitations. As most population-based surveys, we did not follow patients for 3 months to confirm CKD. This is necessary to diagnose CKD in the individual patient, but the implications of using a single determination for eGFR and albuminuria in epidemiological studies remain to be assessed. In addition, we did not include institutionalized population, who is usually old and sick; it may have underestimate the prevalence of CKD. Also, some selection bias cannot be ruled out, as the main reason for not participating in the study was to avoid a blood draw. Lastly, the cross-sectional design impeded to obtain causal conclusions on CKD correlates.