Baseline characteristics of patients with chronic kidney disease stage 3 and stage 4 in spain: the MERENA observational cohort study

The MERENA study is a prospective, observational, non-intervening, multi-center study designed by the Spanish Study Group for Diabetic Nephropathy Grupo Español de Estudio de la n efropatía DIABética (GEENDIAB) of the Spanish Society of Nephrology to characterize the clinical management and outcomes after a 5-year follow-up of two cohorts, diabetic and non-diabetic, of patients with CKD stages 3-4 in a clinical practice setting. Secondary objectives included the assessment of cardiovascular comorbidities, blood pressure (BP), glycemic and lipid control, use of antiplatelet therapy, management of anemia and CKD-MBD, and degree of adherence to clinical practice guidelines by attending nephrologists. The study was conducted at the outpatient nephrology clinics of 55 hospitals evenly distributed through the Spanish territory from the Spanish National Healthcare system with free and equal access to all citizens.

All patients older than 18 years seen at participating nephrology units with CKD stages 3 and 4 (using MDRD-4 equation) according to K/DOQI guidelines [14] from December 1, 2003 to April 30, 2004 were eligible for inclusion in the study, whether they were incident (newly referred) or prevalent patients. Exclusion criteria were as follows: age < 18 yr, CKD stages other than 3 or 4, estimated time to progression to ESRD < 12 months, evidence of consumptive or disabling disease, malignancy, and active infection, inflammatory disorders, or expected death within 12 months. All participants gave written informed consent. The Ethics Committee of the Hospital Universitario Bellvitge in Barcelona approved the study protocol.

All consecutive patients who met the inclusion criteria of the study during the enrolment period were included. Standardized data collection was ensured and conducted by nephrologists using internet-based electronic case report forms. Baseline data included: sex, age, weight, height, body mass index (BMI), specialty of the physician who referred the patient to the nephrologist, K/DOQI CKD stage, and associated co morbid conditions, such as cardiovascular disease, diabetic status, hyperlipidemia, hypertension, or smoking status. The Charlson co-morbidity index was calculated. Renal function was estimated from the 4-variable Modification of Diet in Renal Disease (MDRD) equation, CKD-EPI and Cockroft-Gault equations were also used.

All the participating centers used the same criteria to define existing co morbidities and pathologies. These had been defined in the study's protocol, and together with the recommendations for adherence to the clinical practice guidelines, they had been published on the researchers' website.

Criteria for the diagnosis of coronary artery disease included history of acute myocardial infarction, angina or revascularization procedure. Cerebrovascular disease included transient ischemic attack, stroke, and hemiplegia. Peripheral vascular disease included lower limb intermittent claudication, revascularization (surgical or percutaneous) or amputation. Cardiovascular disease was defined as the history of at least one of the following: ischemic heart disease, cerebrovascular disease and peripheral vascular disease. Cardiac disease was defined as the history of one of the following: coronary artery disease, myocardial infarction, congestive heart failure or history of coronary revascularization. All diabetic patients met the classification criteria established by the American Diabetes Association. The diagnosis of diabetic nephropathy as a likely cause of CKD was made by each investigator. This was generally based on the presence of retinopathy, proteinuria diabetes mellitus duration and all relevant clinical data. Hypertension was considered if the patient had a BP ≥ 140/90 mm Hg or need for antihypertensive drugs. Dyslipidemia included total serum cholesterol > 5.2 mmol/L (200 mg/dL), or triglycerides > 1.7 mmol/L (150 mg/dL), or high-density lipoprotein (HDL) cholesterol < 1.0 mmol/L (40 mg/dL) in males or 1.3 mmol/L (48 mg/dL) in females or low-density lipoprotein (LDL) cholesterol > 2.6 mmol/L (100 mg/dL) or need of lipid-lowering drugs. Drugs prescribed were also recorded. Anemia was defined according to the K/DOQI 2006 Guidelines (serum hemoglobin < 12 g/dL in females and 13.5 g/dL in males) and by the 2004 revised European Best Practice Guidelines (EBPG) [18] (serum hemoglobin < 11.5 g/dL in females and 13.5 g/dL in males, and 12.0 g/dL in men aged > 70 years) or use of erythropoiesis-stimulating agents (ESA) at any Hb level.

Blood samples were analyzed at the respective hospital laboratories of the participating centers. Complete blood cell count, glycated hemoglobin (HbA1c), serum albumin and proteins, glucose, urea, creatinine, uric acid, lipid profile, iron status, potassium, bicarbonate, calcium, phosphorus, intact parathyroid hormone (i-PTH), and 24-hour urinalysis (proteinuria, creatinine, and urea nitrogen) were determined.

Missing data of the main variables at baseline are noted in brackets as follows: MDRD equation (0%), blood pressure (0%), triglycerides (3%), HDL-c (7.1%), LDL-c (7.1%), HbA1c in diabetics (15%), hemoglobin (0%), serum ferritin (7.4%), TSAT (31%), calcium (2%), phosphorus (2%), and i-PTH (12.7%).

The referring physicians for the entire study included: primary care physicians in 46.2% of cases, internal medicine in 12%, endocrinology in 8.9%, urology in 6.9%, and cardiology in 4.8%; the remaining patients were referred by other specialists or after a hospital admission.

Each investigator was advised to adhere strictly to the best practice guidelines and clinical recommendations concerning: lifestyle practices, target BP, HbA1c in diabetics and lipid levels, management of anemia and bone mineral disorders or use of antiplatelet agents. Recommendations were uploaded onto the study's researchers' website. The STROBE statement checklist has been considered in data analysis and in the preparation of the manuscript.

Data are presented as numbers and percentages, or as mean and standard deviation (SD). Differences between diabetic and non-diabetic patients were assessed with Student's t test for quantitative variables and the chi-square (χ²) test for categorical variables. Statistical significance was set at P < 0.05. Variables significantly associated with cardiovascular disease were assessed using binary logistic regression with forward stepwise selection procedure. Categorical variables were also adjusted by age when indicated. Analysis was performed using SPSS version 15.0 for Windows.

The mean age of the total cohort of 1,129 patients was 68 ± 13 years, (median age 70.9, range 19-94 years). There were no significant differences in age between CKD stages 3 and 4 (Table 1). There was a predominance of male subjects (64%) and the ethnicity was all Caucasian. The subjects had been follow up in Nephrology clinics for 24 months (24 ± 12) at the point of entry into the study although it varied widely within the cohort (range: 0-46 months). Only 70 patients (6.2%) were truly incidents (new patients). The mean BMI was 28.4 kg/m² and 31% of the patients had BMI exceeding 30 kg/m². BMI was significantly higher in stage 4 than stage 3 CKD (Table 1) BMI in male was lower than in female (28.0 ± 4.3 vs. 29.4 ± 5.8. p < 0.001).

There was also a high percentage of individuals with history of diabetes mellitus (40.8%) (n = 461) (91.7% type 2 diabetes mellitus and 8.2% type 1 diabetes mellitus), but diabetes as the cause of kidney disease was only attributed to 26.1% of the total cohort based on clinical assessment (See methods). The other causes of CKD were: vascular disease (in 28% of patients), glomerular disease (11.4%), interstitial renal disease (10.8%), polycystic kidney disease (3.5%), and others (5.7%), and the cause of the remaining 14.5% of cases were unknown.

The mean creatinine of the cohort was 2.4 mg/dL. The estimated creatinine clearance was 30 ± 8 ml/min when calculated by the Cockcroft-Gault formula. The MDRD estimated GFR was 28.8 ml/min/1.73 m². Using the MDRD formula for CKD staging, we found 38.5% of subjects with stage 3 CKD and 61.5% of subjects with stage 4 CKD.

The percentage of CKD stage 4 patients with BMI > 30 kg/m² was significantly higher than in CKD stage 3 patients (36.1 vs. 24.5%) despite the percentage of individuals with diabetes mellitus being lower in CKD stage 4 as compared with CKD stage 3 (37.6 vs. 46.1%).

History of cardiovascular disease (CVD) was reported in 39.1% of the cohort (Table 2). The percentage of patients with history of congestive heart failure was 17.5% of the total cohort. History of coronary artery disease was present in 21.6% with a prevalence of history of myocardial infarction in 11.1% of the patients. Cerebrovascular disease was documented in 12.2% and peripheral vascular disease in 19.7% of the patients. History of current smoking was elicited in 9.6% of the patient cohort while 36.2% of the patients were former smokers.

CKD stage 4 patients had a significantly higher percentage of CVD than CKD stage 3 patients (42.2 vs. 35.6%) (Table 2). The percentage of CVD was much higher in older subjects (age ≥ 60yrs) than in younger subjects (age < 60 yrs) (45.6 vs. 19.8%, p = 0.001).

The history of cardiac disease was also significantly higher in Stage 4 CKD than in stage 3 CKD (38.5 vs. 31.3% p < 0.01). There was also a significantly higher history of congestive heart failure (CHF) in stage 4 and 3 CKD patients (19.7% vs. 15.1%, p < 0.045) (Table 2). There were no significant differences in coronary artery disease, myocardial infarction, peripheral vascular disease and cerebrovascular disease between stage 4 CKD and stage 3 CKD patients (table 2).

To better examine CVD as a function of estimated GFR and given the wide rage of GFR in stage 3, we used the staging of 3a and 3b for this analysis. After adjusting for age, CVD increased with declining GFR with the hierarchy (stage 3a < stage 3b < stage 4) when calculated by CKD-EPI (31.8, 35.4, 42.1%, p 0.039) and Cockcroft-Gault formula (30.9, 35.6, 43.4%, p 0.010) and MDRD formula (32.5, 36.2, 42.2%,) but with the latter, it did not reach statistical significance (p 0.882) (Figure 1).

After subdividing stage 3 CKD into stages 3a and 3b, there were also significant differences in age adjusted cardiac disease, congestive heart failure and cerebrovascular disease but not peripheral vascular disease (Figure 2).

A majority of subjects had history of hypertension (92.7%) and there was no significant difference in the prevalence of hypertension between CKD stages 3 and 4 (Table 3). The definition of hypertension included subjects on antihypertensive treatment (See methods).

The percentage of patients with a systolic BP < 130 mm Hg and a diastolic BP < 80 mm Hg was 22.1% and 51.4%, respectively. Only 17.4% of the total cohort had systolic BP < 130 mm Hg and diastolic BP < 80 mm Hg (Table 3).

For the cohort as a group, the mean systolic BP was 141 ± 19 mm Hg and mean diastolic BP was 76 ± 11 mm Hg. CKD Stage 3 and 4 patients had a mean systolic BP which was not significantly different from each other. The mean diastolic BP, however, was slightly but significantly higher in stage 3 than stage 4 patients.

The percentage of patients on 3 or more anti-hypertensive drugs for the total cohort was 47.7% and there were no significant differences between stage 3 and 4 (Table 3). A slightly higher percentage of patients were on 3 or more anti-hypertensive medications in stage 4 than stage 3 CKD but the difference was not significant. The percentage of patients on angiotensin-converting enzyme (ACE) inhibitors was 43.3% and on Angiotensin II receptor blockers (ARB) was 44.8%. The patients on combination therapy with both ACE inhibitors and ARB were 10.8% of the total cohort.

There were a significantly lower percentage of subjects on ACE inhibitors and ACE inhibitors combined with ARBs in stage 4 than in stage 3 CKD patients (Table 3). No significant differences were noted among the percentage of patients on ARBs and combination therapy with ACEI and ARBs between stages 3 and 4 CKD. A slightly higher percentage of patients were on diuretics in stage 4 than stage 3. No significant differences were found in the percentage of patients on calcium channel blockers, beta blockers and alpha blockers between both the groups.

Hyperkalemia defined as potassium levels > 5.4 mEq/dl was more frequently observed in stage 4 than in stage 3 CKD (Table 3).

Proteinuria (> 300 mg/day) was present in 62.8% of the subjects with a mean of 1.2 ± 1.8 g/24 hrs) despite of the widespread use of ACE-inhibitors and ARB drugs. The mean protein excretion was not significantly different between stages 3 and 4 CKD groups (1.2 ± 1.8 vs. 1.3 ± 1.8 g/24 hrs). The percentage of patients with proteinuria however, was significantly higher in subjects with CKD stage 4 than in CKD stage 3 (66.8 vs 56.6%, p = 0.001).

The plasma lipid levels are given in Table 4. There were no significant differences in the LDL-cholesterol, HDL-cholesterol and the triglyceride levels in CKD stage 3 and 4 patients. The percentage of patients treated with statins was 54.7% and there were no differences in stage 3 and stage 4 patients (Table 4). Only one third of the patients had LDL-c below 100 mg/dl, and this percentage was lower for LDL-c < 70 mg/dl (8.3%). The percentage of patients treated with aspirin or other antiplatelet agents was 46.2% and no differences were found between stages 3 and 4 CKD patients.

According to the K-DOQI and the European Best Practices Guidelines (EBPG), anemia was present in 51.3% and 30.5% of the total subjects, respectively. The mean hemoglobin for the cohort as a whole was 12.8 g/dL and the majority of patients (87.8%) had hemoglobin greater than 11 g/dL (Table 5). The percentage of patients with severe anemia (hemoglobin < 9 g/dL) was 1.6% and the percentage of patients with hemoglobin between 9 and 11 g/dL was 11.5%. About 25% of the total cohort was receiving ESA and the mean hemoglobin in this group while receiving this therapy was 12 ± 1.5 g/dL and the range was wide (Table 5). The mean ferritin level was 150 ng/mL (range 5-1500) and the percentage of patients with transferrin saturation < 20% was 26.3%. The percentage of patients on oral and IV iron supplementation were 31.9% and 3.0%, respectively.

The mean hemoglobin was significantly lower in stage 4 when compared to stage 3 CKD patients (12.4 ± 1.6 vs. 13.2 ± 1.7 g/dL). The percentage of patients with hemoglobin greater than 11 g/dL was significantly higher in stage 3 than stage 4 CKD patients (91.1 vs 85.5% p < 0.001). Conversely, the percentage of patients with hemoglobin < 9 g/dL was significantly lower in stage 3 than stage 4 CKD patients (0.5 vs 2.3% p < 0.001). No significant differences were noted in the percentage of patients with hemoglobin between 9 and 11 g/dL between CKD stage 3 and 4 patients (Table 5).

A significantly higher proportion of CKD stage 4 subjects were receiving ESA as compared to stage 3 CKD (34.1% vs. 16%, respectively). No significant differences were found in the mean hemoglobin between CKD stage 4 and 3 patients treated with ESA. The percentage of patients receiving ESA with hemoglobin levels < 9 g/dL, between 9-11 g/dL and > 11 g/dL were not significantly different in stages 3 and 4 CKD (Table 5). No significant differences were reported in the mean weekly dose of rHu-EPO and Darbepoetin in both the CKD groups (data not shown). No significant differences were noted among the patients with transferrin levels < 20% in both the groups. A significantly higher percentage of patients in CKD stage 4 were receiving oral and IV iron supplementation when compared to stage 3 CKD patients (39.5% vs. 23.6% and 4.4% vs. 1.6%, respectively) (Table 5).

Figure 3 shows the distribution of anemia arbitrarily defined as Hb levels < 11 as a function of declining GFR. With declining GFR the % of patients with Hb < 11 g/dL increases progressively.

The mean phosphorus level was 3.7 mg/dL, the mean calcium-phosphorus product was 35 mg²/dL² and the mean intact PTH level was 145 pg/mL. The serum phosphorus was significantly higher in stage 4 than stage 3 CKD while the serum calcium was not significantly different (Table 6). The Ca-P product was slightly but significantly higher in stage 4 than stage 3. Intact PTH was significantly higher in stage 4 than stage 3 (166 vs. 121 pg/mL, p = 0.001).

85.1% of the patients had phosphorus levels within the K-DOQI target (Table 6) and 57.6% of the patients had mean calcium values within the K-DOQI target. 98.6% of the patients had the Ca-P product within the K-DOQI target whereas only 24.4% of the patients had mean intact PTH values within the K-DOQI target. The composite target for all the four parameters as per the K-DOQI guidelines was met in only 10.5% of the total cohort and 12.2 and 9.0% for stages 3 and 4, respectively.

The percentage of patients being treated with vitamin D supplementation and phosphate binders were 15% and 19.7%, respectively. A significantly higher percentage of patients were on vitamin D and phosphate binders in stage 4 than in stage 3 CKD (Table 6).