Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Estimated Glomerular Filtration Rate, All-Cause Mortality and Cardiovascular Diseases Incidence in a Low Risk Population: The MATISS Study

  • Chiara Donfrancesco ,

    Contributed equally to this work with: Chiara Donfrancesco, Simonetta Palleschi

    chiara.donfrancesco@iss.it

    Affiliation National Centre of Epidemiology, Surveillance and Health Promotion, Istituto Superiore di Sanità, Rome, Italy

  • Simonetta Palleschi ,

    Contributed equally to this work with: Chiara Donfrancesco, Simonetta Palleschi

    Affiliation Department of Hematology, Oncology and Molecular Medicine, Istituto Superiore di Sanità, Rome, Italy

  • Luigi Palmieri,

    Affiliation National Centre of Epidemiology, Surveillance and Health Promotion, Istituto Superiore di Sanità, Rome, Italy

  • Barbara Rossi,

    Affiliation Department of Hematology, Oncology and Molecular Medicine, Istituto Superiore di Sanità, Rome, Italy

  • Cinzia Lo Noce,

    Affiliation National Centre of Epidemiology, Surveillance and Health Promotion, Istituto Superiore di Sanità, Rome, Italy

  • Fabio Pannozzo,

    Affiliation Azienda Unità Sanitaria Locale (AUSL), Latina, Italy

  • Belinda Spoto,

    Affiliation Epidemiologia Clinica e Patofisiologia delle Malattie Renali e dell’Ipertensione Arteriosa, CNT, IBIM, Reggio Calabria, Italy

  • Giovanni Tripepi,

    Affiliation Epidemiologia Clinica e Patofisiologia delle Malattie Renali e dell’Ipertensione Arteriosa, CNT, IBIM, Reggio Calabria, Italy

  • Carmine Zoccali,

    Affiliation Azienda Unità Sanitaria Locale (AUSL), Latina, Italy

  • Simona Giampaoli

    Affiliation National Centre of Epidemiology, Surveillance and Health Promotion, Istituto Superiore di Sanità, Rome, Italy

Correction

15 Jan 2014: Donfrancesco C, Palleschi S, Palmieri L, Rossi B, Lo Noce C, et al. (2014) Correction: Estimated Glomerular Filtration Rate, All-Cause Mortality and Cardiovascular Diseases Incidence in a Low Risk Population: The MATISS Study. PLOS ONE 9(1): 10.1371/annotation/1f5e18af-4a68-4419-9f3f-7e8bff410b48. https://doi.org/10.1371/annotation/1f5e18af-4a68-4419-9f3f-7e8bff410b48 View correction

Abstract

Background

Chronic kidney disease (CKD) independently increases the risk of death and cardiovascular disease (CVD) in the general population. However, the relationship between estimated glomerular filtration rate (eGFR) and CVD/death risk in a general population at low risk of CVD has not been explored so far.

Design

Baseline and longitudinal data of 1465 men and 1459 women aged 35-74 years participating to the MATISS study, an Italian general population cohort, were used to evaluate the role of eGFR in the prediction of all-cause mortality and incident CVD.

Methods

Bio-bank stored sera were used to evaluate eGFR at baseline. Serum creatinine was measured on thawed samples by means of an IDMS-calibrated enzymatic method. eGFR was calculated by the CKD-EPI formula.

Results

At baseline, less than 2% of enrolled persons had eGFR < 60 mL/min/1.73m2 and more than 70% had a 10-year cardiovascular risk score < 10%. In people 60 or more years old, the first and the last eGFR quintiles (<90 and ≥109 mL/min/1.73m2, respectively) were associated to an increased risk for both all-cause mortality (hazard ratio 1.6, 95% confidence interval 1.2-2.1 and 4.3, 1.6-11.7, respectively) and incident CVD (1.6, 1.0-2.4 and 7.0, 2.2-22.9, respectively), even if adjusted for classical risk factors.

Conclusions

These findings strongly suggest that in an elderly, general population at low risk of CVD and low prevalence of reduced renal filtration, even a modest eGFR reduction is related to all-cause mortality and CVD incidence, underlying the potential benefit to this population of considering eGFR for their risk prediction.

Introduction

Chronic kidney disease (CKD) is recognized as a global public health problem [1,2]. Patients with CKD are far more likely to die, mainly from cardiovascular diseases (CVD), than to develop kidney failure [3]. CKD is also a risk factor for adverse outcomes in other chronic diseases [4]. The incorporation of simple renal biomarkers in the clinical practice for CVD prevention could be accomplished quickly and the question whether screening the general adult population for renal dysfunction would be useful to prevent incident CVD as well as all-causes mortality appears to be a relevant public health issue [5].

Several studies investigated whether impaired renal function is associated with increased risk of death or incident cardiovascular events beyond traditional risk factors [6-13]. The majority of these studies found that overt renal dysfunction independently and significantly associates with an increased risk of both mortality and CVD events. However, whether milder forms of renal dysfunction increases CVD and/or death risk remains a much debated issue [14]. In this regard it has to be noted that studies performed so far were conducted in countries at high risk of CVD and/or high prevalence of CKD while, to our knowledge, no studies have been carried out in populations at both low CKD and CVD risk. Countries at low CVD risk could even take greater advantage of investigating the additive value of novel biomarkers to improve the traditional risk models and better stratify persons with low or moderate CVD risk [15].

The Italian population has a substantially lower risk for cardiovascular events than Northern European populations and previous observations in the Malattie Cardiovascolari Aterosclerotiche, Istituto Superiore di Sanità (MATISS) cohort fully confirmed the low risk for CVD in a community of central Italy [16]. The present study assesses the relationship between estimated glomerular filtration rate (eGFR) and risk of all-cause mortality and incident CVD of a population at low risk of CVD and low prevalence of reduced renal filtration using longitudinal data and stored biological samples of the MATISS study [17].

Methods

Study population

Base-line data of the MATISS Study were collected from 1993 to 1996 in 2592 men and 2866 women, aged 20-79 years, randomly selected from the Italian general population stratified by age and sex. The cohort was extracted from the electoral rolls of a geographical area located about 100 km southeast of Rome (central Italy) and comprising four adjacent rural towns (Sezze, Roccagorga, Bassiano, and Priverno). Participation rate was 60%. MATISS study was approved as part of the CUORE Project by Ethical Committee of the Istituto Superiore di Sanità (ISS, the Italian National Institute of Health) on 15 March 2006; the Ethical Committee approved the use of pooled samples for research activity in epidemiological and genetic/genomic studies. Personal data are stored and managed according to the Italian legislative Decree 196/03 and the Code of Conduct to manage personal data for statistic and scientific purposes (2004). Personal data are excluded from the database used for statistical analyses that, instead, includes an unique code for each person. The same codes were used to label the serum samples stored in the biological bank.

Among MATISS participants, 1465 men and 1459 women with available stored serum samples, free of cardiovascular events and 35-74 years old at baseline were included in the statistical analysis.

Cardiovascular risk factors and eGFR at baseline

People were invited by letter at the screening centre. Risk factors were collected using standardized procedures published elsewhere [18,19].

Biological specimens collected at study baseline were preserved in nitrogen vapours in the biological bank of the National Centre of Epidemiology, Surveillance and Health Promotion of the ISS. For the purpose of this study, stored serum samples were thawed and assayed for creatinine concentration by the Renal Unit of the department of Hematology, Oncology and Molecular Medicine of the ISS. Serum creatinine was measured by means of an enzymatic IDMS-calibrated method (Creatinine Enzymatic Test, Instrumentation Laboratory SpA, Milano, Italy) performed on the automatic analyzer ILAB300plus (Instrumentation Laboratory SpA). The trueness was checked by analyzing the Certified Reference Material from the National Institute of Standards and Technology (Creatinine in Frozen Human Serum, SRM 967a, target value = 0,847 mg/dL). Method imprecision (coefficient of variation, CV), evaluated as pooled standard deviation (SD) of duplicate measurements of real samples, was 2.5% (mean creatinine value = 0.76 mg/dL). The total error of the method, calculated as bias + 1.96* CV, was less than 5%, thus fulfilling the recommendations of the Laboratory Working Group of the National Kidney Disease Education Program [20]. GFR was estimated from serum creatinine concentration according to the CKD-EPI formula [21].

End-points

All-cause mortality and fatal and non-fatal coronary and cerebrovascular events were considered as end-points and followed-up until December 2004. Major cardiovascular events, in particular myocardial infarction, stroke, revascularization treatment (bypass or angioplasty), sudden death, were identified with record linkage with mortality and hospital discharge registers. Fatal and non-fatal coronary events were validated using MONICA diagnostic criteria [22,23]. Mortality registers and death certificates were used to assess all-cause mortality [24]. Median time of follow-up was 10.3 years both for all-cause mortality and incident major CVD. Persons lost at follow-up were 21 (0.72% of the baseline cohort).

Statistical analyses

ANOVA models and chi-squared test were used to compare mean values of cardiovascular risk factors and risk conditions prevalence respectively. Pearson correlation analyses were performed to asses correlation among traditional risk factors and eGFR. The 10-years fatal and non-fatal cardiovascular risk score at baseline was based on the CUORE Project risk equation for person aged 35-69 years [25,26]. According to KDOQI guidelines, eGFR values less than 60 mL/min/1.73m2 were considered suggestive of CKD [2].

Cox proportional hazard models were performed to assess the prediction role of eGFR in relation to risk of all-cause mortality and incident CVD. eGFR was considered as either continuous or categorical variable and models were adjusted for age, gender and traditional risk factors. Systolic and diastolic blood pressure, total and HDL cholesterol, triglycerides, glycemia, body mass index (BMI), hypertension treatment and smoking habit were considered as possible covariates; only significant and non-collinear risk factors were included in models.

Due to the U shape of the association between eGFR and risk of death or CVD, Cox models including eGFR as continuous variable were performed with the exclusion of persons with eGFR ≥ 109 mL/min/1.73m2.

Given the very low number of persons with CKD (less than 2%), subjects were grouped in three classes based on eGFR quintiles (second, third and fourth quintiles were considered the reference group). Due to the interaction between eGFR and age, Cox proportional hazard models were also performed by age class: 35-59 years old and 60-74 years.

P-value ≤0.05 were considered statistically significant. Statistical analyses were performed using SAS software, release 9.2 (SAS Institute Inc., Cary, North Carolina, USA).

Results

Baseline associations

Table 1 shows main baseline characteristics of 1465 men and 1459 women. Mean age was 53 years both in men and women, only 1.0% of men and 1.6% of women had eGFR less than 60 mL/min/1.73m2 (range: 37-137 mL/min/1.73m2) and only 0.3% had elevated level of serum creatinine (>1.5 mg/dL for men and >1.3 mg/dL for women, range: 0.4-1.8 mg/dL). The CUORE Project 10-year cardiovascular risk score resulted less than 10% for 73% of men and 92% of women. The eGFR was highly correlated with age (r= -0.544 for men and -0.671 for women, p<0.0001). The prevalence of persons with eGFR<60 mL/min/1.73m2 was 3% in those 60 years of age or older (eGFR mean±SD: 88.4±11.7 mL/min/1.73m2, range: 37-113 mL/min/1.73m2) and 0.5% in those with age less than 60 years (eGFR, mean±SD: 102.7±10.9 mL/min/1.73m2, range: 39-137 mL/min/1.73m2).

VariablesMen Women
nmeansdnmeansd
Age (years)1465531114595311
Serum creatinine (mg/dL)14650.90.114590.70.1
eGFR by CKD-EPI (mL/min/1.73 m2)^1465981214599914
Systolic Blood Pressure (SBP) (mm Hg)1464140.621.91457143.924.5
Diastolic Blood Pressure (DBP) (mm Hg)1464861214578713
Serum Total Cholesterol (mg/dL)1463223.643.51457222.441.1
Serum HDL-Cholesterol (mg/dL)1463481314575312
Serum Triglycerides (mg/dL)1463173.8130.71457130.272.0
Serum Fasting Glucose (mg/dL)1463932314578825
Body Mass Index (BMI) (kg/m²)146027.53.7145529.14.9
Cigs°/Day (Current Smokers only) 5411712158107
10-year cardiovascular risk (%)°°13738.28.713373.24.4
npercentnpercent
eGFR by CKD-EPI^ <60 mL/min/1.73 m2141.0231.6
Elevated serum creatinine §50.350.3
Cigarette Smoking 57139.016411.3
Diabetes906.2805.5
Blood Pressure -- SBP/DBP §§
Normal17311.818312.6
Prehypertension50134.238926.7
Hypertension-Stage I38726.432522.3
Hypertension-Stage II or treated40327.556038.4
Hypertension Treatment15510.634623.7
Obesity (BMI>=30.0 kg/m2)30821.156739.0
Primary or secondary school118981.2131890.4
10-year CVR of the CUORE Project
CVR < 10.0%99672.5123292.2
CVR 10.0-19.9%24718.0886.6
CVR >=20.0%1309.5171.3

Table 1. Baseline descriptive statistics of the MATISS study, men and women, aged 35-74 years, free of previous cardiovascular diseases.

Sd: standard deviation;
^Estimated glomerular filtration rate (eGFR) by Chronic Kidney Disease Epidemiology Collaboration;
° Cigs is for cigarettes;
°° 10-year cardiovascular risk of the CUORE Project based on age, systolic blood pressure, total and HDL cholesterol, diabetes, smoking habit, treatment for hypertension. It is for men and women 35-69 years old;
§For men, serum creatinine ≥1.5 mg/dl; for women ≥ 1.3 mg/dl;
§§Normal: Systolic Blood Pressure (SBP) < 120 AND Diastolic Blood pressure (DBP) < 80 mmHg, no antihypertensive drug treatment; Prehypertension: SBP 121-139 OR DBP 81-89 mmHg, no antihypertensive drug treatment; Hypertension Stage I: SBP 140-159 OR DBP 90-99 mmHg, no antihypertensive drug treatment; Hypertension Stage II: SBP > 160 OR DBP > 100 mmHg OR antihypertensive drug treatment
CSV
Download CSV

The boundaries of eGFR quintiles were 39-89 mL/min/1.73m2, 90-96 mL/min/1.73m2, 97-101 mL/min/1.73m2, 102-108 mL/min/1.73m2 and 109-137 mL/min/1.73m2. Based on quintiles, eGFR was stratified into three classes: low eGFR (<90 mL/min/1.73m2), normal eGFR (90-108 mL/min/1.73m2) and high eGFR (≥109 mL/min/1.73m2). Individuals in the third group were younger and with a more favourable CVD risk profile at baseline, with the exception of the smoking habit (Table 2). In the high eGFR group, only 5.1% of men and 0.6% of women had a 10-year cardiovascular risk (CVR) ≥ 10% whereas, in the low eGFR group, the percentages of men and women with a CVR ≥ 10% were 43.0% and 19.2%, respectively.

VariablesMen Women
eGFR ^ <90 mL/min/1.73 m290≤eGFR ^ <109 mL/min/1.73 m2eGFR ^ ≥109 mL/min/1.73 m2eGFR ^ <90 mL/min/1.73 m290≤eGFR ^ <109 mL/min/1.73 m2eGFR ^ ≥109 mL/min/1.73 m2
nmeansdnmeansdnmeansdp-valuenmeansdnmeansdnmeansdp-value
Age (years)2965911893539276426<0.0001301629804549354425<0.0001
Serum creatinine (mg/dL)2961.00.18930.80.12760.80.1<0.00013010.80.18040.70.13540.60.1<0.0001
eGFR by CKD-EPI (mL/min/1.73 m2)^2968098939952761135<0.0001301781180410053541144<0.0001
Systolic Blood Pressure (SBP) (mm Hg)296145248921412227613316<0.0001301154268021452435413220<0.0001
Diastolic Blood Pressure (DBP) (mm Hg)2968613892861227686121301901480287133548313<0.0001
Serum Total Cholesterol (mg/dL)2962264589122342276223450.4552301228418022264135421039<0.0001
Serum HDL-Cholesterol (mg/dL)29648138914813276451103015313802541335453110
Serum Triglycerides (mg/dL)2961751398911671122761931700.01643011397780213270354118710.000
Serum Fasting Glucose (mg/dL)29692198919423276902703019025802882535484240
Body Mass Index (BMI) (kg/m²)2962848912842732740.0006300305801295354285<0.0001
Cigs°/Day (Current Smokers only) 7614113291713136201101910774107651081
10-year cardiovascular risk (%)°°23312118649927644<0.0001224667593435411<0.0001
n%n%n%n%n%n%
eGFR by CKD-EPI^ <60 mL/min/1.73 m21450000-2380000-
Elevated serum creatinine §51.7893100.0276100.0<0.000151.7804100.0354100.0<0.0001
Cigarette Smoking 87293473913750<0.000119678106719<0.0001
Diabetes196.4596.6124.40.380217.0435.4164.50.377
Blood Pressure -- SBP/DBP §§
Normal237.811012.34014.5144.78510.68423.7
Prehypertension9231294331154254181992513638
Hypertension-Stage I7726.023025.88029.06220.619424.26919.5
Hypertension-Stage II or treated10435258294115<0.000117157324406518<0.0001
Hypertension Treatment6120.6849.4103.6<0.000111939.519524.3329.0<0.0001
Obesity (BMI>=30.0 kg/m2)70241952243160137463274110329<0.0001
Primary or secondary school25084.573982.920072.50.00028293.774492.729282.5<0.0001
10-year CVR of the CUORE Project
CVR < 10.0%13357.160169.626294.918180.869992.135299.4
CVR 10.0-19.9%602617620114351651721
CVR >=20.0%4017.28710.131.1<0.000183.691.200.0<0.0001

Table 2. Baseline descriptive statistics by classes of estimated glomerular filtration rate of the MATISS study, men and women, aged 35-74 years, free of previous cardiovascular diseases.

Sd: standard deviation;p-values refer to ANOVA for continuous variables and to chi-squared test for categoricalvariables;
^Estimated glomerular filtration rate (eGFR) by Chronic Kidney Disease Epidemiology Collaboration;
° Cigs is for cigarettes;
°° 10-year cardiovascular risk of the CUORE Project based on age, systolic blood pressure, total and HDL cholesterol, diabetes, smoking habit, treatment for hypertension. It is for men and women 35-69 years old;
§For men, serum creatinine ≥1.5 mg/dL; for women ≥ 1.3 mg/dL;
§§Normal: Systolic Blood Pressure (SBP) < 120 AND Diastolic Blood pressure (DBP) < 80 mmHg, no antihypertensive drug treatment; Prehypertension: SBP 121-139 OR DBP 81-89 mmHg, no antihypertensive drug treatment; Hypertension Stage I: SBP 140-159 OR DBP 90-99 mmHg, no antihypertensive drug treatment; Hypertension Stage II: SBP > 160 OR DBP > 100 mmHg OR antihypertensive drug treatment
CSV
Download CSV

All-cause mortality and incident cardiovascular diseases associations

During the follow-up, 163 and 83 deaths, and 100 and 46 cardiovascular events occurred in men and women, respectively (Table 3). Among the 14 men with CKD only 2 died (one for unspecified chronic heart disease and one for cancer) during follow up and no one experienced major coronary or stroke events; 1 death (unspecified chronic heart disease) and 1 major coronary or stroke event occurred among the 23 women with CKD. Crude mortality rate per 1,000 persons decreased from the first to the third eGFR group: 206, 99 and 51 for men and 116, 51 and 20 for women, as well as crude CVD: 91, 68 and 43 for men and 66 , 30 and 6 for women respectively.

MenWomen
All cause mortality
Number of participants1,4651,459
Events16383
Persons-years (py)14,57114,338
Event/10,000 py111.957.9
Incident cardiovascular disease
Number of participants1,4651,459
Events10046
Persons-years (py)14,30914,241
Event/10,000 py69.932.3
Incident coronary disease
Number of participants1,4651,459
Events6718
Persons-years (py)14,31214,410
Event/10,000 py46.812.5
Incident cerebrovascular disease
Number of participants1,4651,459
Events3728
Persons-years (py)14,46714,267
Event/10,000 py25.619.6

Table 3. Follow-up information by gender of the MATISS study, men and women, aged 35-74 years, free of previous cardiovascular diseases.

CSV
Download CSV

Since a strong interaction between eGFR and age was found, Cox proportional hazard model analyses were carried out by age classes (34-59 and 60-74 years) (Table 4). In the whole sample, only people with eGFR ≥ 109 mL/min/1.73m2 showed a significantly high hazard ratio (HR) in relation to all-cause mortality. Among elderly persons, those with an eGFR value between 90 and 108 mL/min/1.73m2 (reference group) resulted having a risk of death 40% lower than those with an eGFR less than 90 mL/min/1.73m2 (corresponding to HR 1.6), even when the analysis was adjusted for classical risk factors; in the same age group, an eGFR ≥ 109 mL/min/1.73m2 was associated to a risk of death 4 times higher than that of the reference group, and more than 2 times higher than that of the group with eGFR < 90 mL/min/1.73m2. No significant association was found for 35-59 years old people.

AllAge 35-59 years Age 60-74 years
Age and sex adjustedAdjusted for classical risk factors^Sex adjustedAdjusted for classical risk factors^^Sex adjustedAdjusted for classical risk factors^^
HR95% confidence intervalHR95% confidence intervalHR95% confidence intervalHR95% confidence intervalHR95% confidence intervalHR95% confidence interval
All-cause mortality
eGFR<90 (mL/min/1.73 m2)1.00.81.41.10.81.50.90.42.00.90.42.11.51.12.01.61.22.1
90≤eGFR<109(mL/min/1.73 m2)1.0--1.0--1.0--1.0--1.0--1.0--
eGFR≥109 (mL/min/1.73 m2)2.61.54.42.31.34.00.90.51.60.90.51.63.71.310.04.31.611.7
Number of persons2924291720392033885884
Numbers of deaths2462456262184183
Incident cardiovascular disease
eGFR<90 (mL/min/1.73 m2)0.90.61.31.00.71.50.20.11.00.30.11.11.41.02.21.61.02.4
90≤eGFR<109(mL/min/1.73 m2)1.0--1.0--1.0--1.0--1.0--1.0--
eGFR≥109 (mL/min/1.73 m2)1.60.83.11.50.82.90.60.31.20.60.31.35.61.718.17.02.222.9
Numebr of persons2924291720392033885884
Numbers of cardiovascular events14614451509594

Table 4. Cox model hazard ratios for classes of estimated glomerular filtration rate of the MATISS study, men and women, aged 35-74 years, free of previous cardiovascular diseases.

Estimated glomerular filtration rate (eGFR) by Chronic Kidney Disease Epidemiology Collaboration;
HR: hazard ratios; ^All cause mortality models were adjusted for age, gender, current smoking , fasting glycemia, systolic blood pressure.
Incident cardiovascular diseases models were adjusted for age, gender, current smoking , fasting glycemia, systolic blood pressure
^^All cause mortality models were adjusted for gender, current smoking , fasting glycemia, systolic blood pressure.
Incident cardiovascular diseases models were adjusted for gender, current smoking , fasting glycemia, systolic blood pressure.
CSV
Download CSV

Similar results were found for incident of major CVD, although the statistical significance was reached only for the group of elderly people with eGFR ≥ 109 mL/min/1.73m2.

By further stratifying the first eGFR class (<90 mL/min/1.73m2,) into two groups based on the conventional threshold for CKD (60 mL/min/1.73m2,), the individuals with a modestly reduced eGFR ( ≥60 and <90 mL/min/1.73m2) still showed a 67% higher risk to die (p= 0.0008) or suffer a CV event (p=0.0165).

These results were confirmed when eGFR was tested as a continuous variable: a significant, independent of classical risk factors, inverse relationship between eGFR and all-cause mortality was found; a similar trend was also found for incident CVD (Table 5).

All Age 35-59 years Age 60-74 years
eGFR (mL/min/1.73 m2) eGFR (mL/min/1.73 m2) eGFR (mL/min/1.73 m2)
Coeff.HR at more adverse level*p-valueCoeff.HR at more adverse level*p-valueCoeff.HR at more adverse level*p-value
All cause mortality
Classical risk factors adjusted-0.00250.970.68290.01811.180.3290-0.01780.810.0020
Incident cardiovascular disease
Classical risk factors adjusted-0.00200.980.80300.02621.280.2063-0.01620.830.0451

Table 5. Relation of estimated glomerular filtration rate to prediction of all-cause mortality and cardiovascular events of the men and women, aged 35-74 years, free of previous cardiovascular diseases.

Estimated glomerular filtration rate (eGFR) by Chronic Kidney Disease Epidemiology Collaboration;
Due to the U shape of the association between eGFR and risk of death or CVD, Cox models were performed with the exclusion of persons with eGFR ≥ 109 mL/min/1.73m2;
*Hazard ratio with level 1 standard deviation (std) higher: all eGFR standard deviation (SD) 11.2, age 35-59 eGFR SD 9.3, age 60-74 eGFR SD 11.6;
All cause mortality models were adjusted for age (except those stratified by age classes), gender, current smoking , fasting glycemia, systolic blood pressure.
Incident cardiovascular diseases models were adjusted for age (except those stratified by age classes), gender, current smoking , fasting glycemia, systolic blood pressure and total cholesterol.
CSV
Download CSV

Discussion

This study shows that, in a general population at low risk to develop CVD and with a low prevalence of reduced renal filtration (only 1.0% of men and 1.6% of women had eGFR less than 60 mL/min/1.73m2), elderly individuals with either high or low eGFR show higher mortality rates as compared to those with a normal eGFR. No such difference was found in young people (less than 60 years old). This result is consistent when considering eGFR as a variable either categorical or continuous and, more important, is independent on major traditional risk factors such as smoking habit, glycemia and systolic blood pressure. A similar association emerged also with incident CVD, although the statistical significance was lower.

The association of kidney dysfunction to all cause-mortality in elderly persons is in line with other studies in Scandinavian countries. The second Nord-Trøndelag Health Study (HUNT II), a Norwegian community-based health study, found that reduced kidney function and microalbuminuria are independent risk factors for cardiovascular death particularly in elderly persons among which, by adding renal function indicators, a strong improvement of cardiovascular risk models was found [6]. Data from the Uppsala Longitudinal Study of Adult Men (ULSAM), a Sweden community-based cohort, indicated that in elderly men the simultaneous addition of several biomarkers of cardiovascular and renal abnormalities substantially improves the risk stratification for death from cardiovascular causes [27]. An interpretation of these findings is that, due to the higher prevalence of subclinical cardiovascular damage, the predictive power of traditional risk factors is attenuated in the elderly, thus increasing the relative weight of other risk factors such as eGFR. Our data specifically confirm this hypothesis in a Southern European population at low risk of CVD with a CV mortality about 40% to 80% lower than that in Scandinavian countries.

In our analyses, the association between eGFR and incidence of major CVD, even if present in elderly persons, resulted less significant than for all-cause mortality. Such an evidence could be plausibly explained by the lower statistical power due to the lower number of major cardiovascular events as compared to the number of deaths.

In line with previous reports [9,10,28-30], we found that the risk of death and incidence of major CVD increased at high eGFR values. This finding may appear at odds with the observation that, as expected, in the baseline analyses people with high levels of eGFR were younger, with lower blood pressure, and with a lower 10-year cardiovascular risk at baseline. Importantly, in our population high eGFR were unrelated to obesity (Table 2). The increased risk associated with high eGFR values is generally ascribed to the fall in serum creatinine due to muscle wasting secondary to ill health. Since a positive association between current smoking and increased eGFR has been also found in the present study as well as in previous reports [31-33], we hypothesize that the smoking habit prime some pathogenic processes which underlie, though on a different time scale, both the increase of eGFR and death/CVD risk. In support of this hypothesis is the evidence that current smoking is truly associated with an increase in GFR when the latter was measured by iohexol clearance in a sample from the general population [34].

Some limitations in our study should be mentioned. First, the lack of kidney damage biomarkers: the assessment of micro- and macro-albuminuria together with the calculation of eGFR gives much more information than the assessment of eGFR alone, especially for the interpretation of high eGFR values [35] and a possible underestimation of the prevalence of CKD in the study population. In addition, creatinine clearance measurements using 24-hour urine samples could provide useful information for the assessment of eGFR in persons with exceptional dietary intake (vegetarian diet, creatine supplements) or muscle mass (amputation, malnutrition, muscle wasting), and for checking the need to start dialysis in renal patients [2]. However, the above conditions would concern only a small percentage of persons in a general population study such as the MATISS study.

Second, we assayed creatinine concentration of serum specimens that had been kept frozen for about 20 years. However, previous literature data show that creatinine levels in serum are stable for at least 25 years at -25°C [36]; moreover, as for our experience, the sample storing time had no apparent effect on the analytical perfomances of the creatinine assay adopted.

Finally, due to the low cardiovascular risk and low prevalence of reduced renal filtration in our cohort, the generalizability of our results to other races and ethnic groups may be limited. However this point represents at the same time one of the strengths of this study since surveys conducted in this kind of populations, while having the unique capability to evidence subtle interactions between early renal dysfunctions and risk to develop CVD or to die, are still lacking in the literature.

Another strength of the present study is that we used an enzymatic, IDMS-calibrated method to measure serum creatinine, preferred to the widely used Jaffe method because of the fewer assay interferences [37].

Finally, a main strength of this study is the soundness of the methods applied to construct and update the MATISS database. Actually, the MATISS cohort is one of the first, longer-studied Italian cohorts used to build the cardiovascular risk chart and scores of the CUORE Project, two tools nowadays used in clinical practice to assess cardiovascular risk in primary prevention of adults in Italy [25,38].

Conclusions

The assessment of CKD is potentially practicable on a population-wide basis as it involves relatively simple blood and urine tests. The present analysis shows for the first time that in an elderly, general population with low cardiovascular risk and low prevalence of reduced renal filtration (<60 mL/min/1.73m2), even a modest eGFR reduction is related to all-cause mortality and CVD incidence, underlying the potential benefit to this population of considering eGFR for their risk prediction. Preventive treatment is increasingly offered to elderly persons, but applying current guidelines for primary prevention to this population is problematic because one should treat nearly everyone. Thus, more accurate and complex risk models, including new risk factors such as kidney function indicators, should be developed for this population.

The present data provide a solid basis for future, larger studies aimed to definitely assess whether, in countries with CVD and CKD low risk profiles, screening elderly people for eGFR might significantly improve CVD and mortality risk prediction to justify the screening cost and effort.

Acknowledgments

The MATISS Study Research Group include Simona Giampaoli, Luigi Palmieri, Francesco Dima, Cinzia Lo Noce, Chiara Donfrancesco, Patrizia Caiola De Sanctis, Fabio Pannozzo, Agata Poce, Antonio Menditto, Anna Minoprio, Massimo Pasquali, Augusto Santaquilani, Paolo Lombari. National Centre for Epidemiology, Surveillance and Health Promotion, Istituto Superiore di Sanità, Rome, Italy. SYSKID investigators (Italy) include Carmine Zoccali, Belinda Spoto and Giovanni Tripepi.

Author Contributions

Conceived and designed the experiments: SG FP. Performed the experiments: SG CL SP BR. Analyzed the data: CD LP. Contributed reagents/materials/analysis tools: SP CL BR CZ BS GT SG. Wrote the manuscript: CD SP SG CZ. Interpretation of data: CD SP SG CZ BS GT.

References

  1. 1. Levey AS, Atkins R, Coresh J, Cohen EP, Collins AJ et al. (2007) Chronic kidney disease as a global public health problem: approaches and initiatives—a position statement from Kidney Disease Improving Global Outcomes. Kidney Int; 72: 247–259. doi:https://doi.org/10.1038/sj.ki.5002343. PubMed: 17568785.
  2. 2. National Kidney Foundation (2002) K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classifi cation, and stratification. Am J Kidney Dis 39: S1–266. doi:https://doi.org/10.1053/ajkd.2002.30571. PubMed: 11904577.
  3. 3. Keith DS, Nichols GA, Gullion CM, Brown JB, Smith DH (2004) Longitudinal follow-up and outcomes among a population with chronic kidney disease in a large managed care organization. Arch Intern Med 164: 659–663. doi:https://doi.org/10.1001/archinte.164.6.659. PubMed: 15037495.
  4. 4. Fried LF, Katz R, Sarnak MJ, Shlipak MG, Chaves PH et al. (2005) Kidney function as a predictor of noncardiovascular mortality. J Am Soc Nephrol 16: 3728–3735. doi:https://doi.org/10.1681/ASN.2005040384. PubMed: 16251239.
  5. 5. Perk J, De Backer G, Gohlke H, Graham I, Reiner Z et al. (2012) European guidelines on cardiovascular disease prevention in clinical practice (Version 2012). Eur Heart J 33: 1635–1701. doi:https://doi.org/10.1093/eurheartj/ehs092. PubMed: 22555213.
  6. 6. Hallan S, Astor B, Romundstad S, Aasarød K, Kvenild K et al. (2007) Association of Kidney Function and Albuminuria With Cardiovascular Mortality in Older vs Younger Individuals: The HUNT II Study. Arch Intern Med 167: 2490-2496. doi:https://doi.org/10.1001/archinte.167.22.2490. PubMed: 18071172.
  7. 7. Di Angelantonio E, Chowdhury R, Sarwar N, Aspelund T, Danesh J et al. (2010) Chronic kidney disease and risk of major cardiovascular disease and non-vascular mortality: prospective population based cohort study. BMJ 341: c4986. doi:https://doi.org/10.1136/bmj.c4986. PubMed: 20884698.
  8. 8. van der Velde M, Bakker SJ, de Jong PE, Gansevoort RT (2010) Influence of Age and Measure of eGFR on the Association between Renal Function and Cardiovascular Events. Clin J Am Soc Nephrol 5: 2053–2059. doi:https://doi.org/10.2215/CJN.08851209. PubMed: 20705964.
  9. 9. Chronic Kidney Disease Prognosis Consortium (2010) Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: a collaborative meta-analysis. Lancet 375: 2073–2081. doi:https://doi.org/10.1016/S0140-6736(10)60674-5. PubMed: 20483451.
  10. 10. Matsushita K, Selvin E, Bash LD, Astor BC, Coresh J (2010) Risk implications of the new CKD-EPI equation as compared to the MDRD Study equation for estimated glomerular filtration rate: the Atherosclerosis Risk in Communities (ARIC) Study. Am J Kidney Dis 55: 648–659. doi:https://doi.org/10.1053/j.ajkd.2009.12.016. PubMed: 20189275.
  11. 11. Di Angelantonio E, Danesh J, Eiriksdottir G, Gudnason V (2007) Renal function and risk of coronary heart disease in general populations: New prospective study and systematic review. PLOS Med 4: e270. doi:https://doi.org/10.1371/journal.pmed.0040270. PubMed: 17803353.
  12. 12. Chang A, Kramer H (2011) Should eGFR and Albuminuria be added to the Framingham risk score? Chronic Kidney Disease and Cardiovascular risk prediction. Nephron Clin Pract 119: c171-c178. doi:https://doi.org/10.1159/000325669. PubMed: 21811078.
  13. 13. Ito H, Pacold IV, Durazo-Arvizu R, Liu K, Shilipak MG et al. (2011) The effect of including cystatin C or creatinine in a cardiovascular risk model for asymptomatic individuals: the multi-ethnic study of atherosclerosis. Am J Epidemiol 174: 949-957. doi:https://doi.org/10.1093/aje/kwr185. PubMed: 21880578.
  14. 14. Glassock RJ, Winearls C (2008) An epidemic of chronic kidney disease: fact or fiction? Nephrol Dial Transplant 23: 1117–1121. PubMed: 18359870.
  15. 15. De Backer G, Ambrosioni E, Borch-Johnsen K, Brotons C, Cifkova R et al. (2003) European guidelines on cardiovascular disease prevention in clinical practice: Third Joint Task Force of European and other Societies on Eur. Heart: J 24:1601-10.
  16. 16. Menotti A, Lanti M, Puddu PE, Kromhout D (2000) Coronary heart disease incidence in northern and southern European populations: a reanalysis of the seven countries study for a European coronary risk chart. Heart 84: 238-244. doi:https://doi.org/10.1136/heart.84.3.238. PubMed: 10956281.
  17. 17. Giampaoli S, Poce A, Sciarra F, Lo Noce C, Dima F et al. (1997) Change in cardiovascular risk factors during 10-year community intervention program. Acta Cardiol 52: 411-422. PubMed: 9428939.
  18. 18. Ferrario M, Chiodini P, Chambless LE, Cesana G, Vanuzzo D et al. for the CUORE Project Research Group (2005) Prediction of coronary events in a low incidence population. Assessing accuracy of the CUORE Cohort Study prediction equation. Int J Epidemiol 34: 413–421. doi:https://doi.org/10.1093/ije/dyh405. PubMed: 15659467.
  19. 19. World Health Organization, Office of Cardiovascular Diseases, WHO MONICA Project (1999) MONICA Manual [online resource]. Available: http://www.ktl.fi/publications/monica/manual/index.htm. URN; NBN:fi-fe19981146. Accessed 9 September 2013.
  20. 20. Myers GL, Miller WG, Coresh J, Fleming J, Greenberg N et al. (2006) Recommendations for improving serum creatinine measurement: a report from the Laboratory Working Group of the National Kidney Disease Education Program. Clin Chem 52: 5-18. doi:https://doi.org/10.1373/clinchem.2005.0525144. PubMed: 16332993.
  21. 21. Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd et al. (2009) A new equation to estimate glomerular filtration rate. Ann Intern Med 150: 604-612. doi:https://doi.org/10.7326/0003-4819-150-9-200905050-00006. PubMed: 19414839.
  22. 22. Giampaoli S, Palmieri L, Panico S, Vanuzzo D, Ferrario M et al. (2006) Favorable cardiovascular risk profile (low risk) and 10-year stroke incidence in women and men: findings on twelve Italian population samples. Am J Epidemiol 163: 893-902. doi:https://doi.org/10.1093/aje/kwj110. PubMed: 16554350.
  23. 23. Palmieri L, Donfrancesco C, Giampaoli S, Trojani M, Panico S et al. (2006) Favorable cardiovascular risk profile and 10-year coronary heart disease incidence in women and men: results from the Progetto CUORE. Eur J Cardiovasc Prev Rehabil 13: 562–570. doi:https://doi.org/10.1097/01.hjr.0000221866.27039.4b. PubMed: 16874146.
  24. 24. World Health Organization (1977) International classification of diseases. Manual of the international statistical classification of diseases, injuries, and causes of death. Ninth Revision. Geneva, Switzerland: World Health Organization.
  25. 25. Palmieri L, Panico S, Vannuzzo D, Ferrario M, Pilotto L et al. (2004) La valutazione del rischio cardiovascolare globale assoluto: il punteggio individuale del progetto CUORE. Ann Ist Super Sanità [Annals of the National Health Institute] 40: 393–399.
  26. 26. Giampaoli S, Palmieri L, Donfrancesco C, Panico S, Vannuzzo D et al. (2007) Cardiovascular risk assessment in Italy: the CUORE project risk score and risk chart. It J Public Health 5: 102–109.
  27. 27. Zethelius B, Berglund L, Sundström J, Ingelsson E, Basu S et al. (2008) Use of Multiple Biomarkers to Improve the Prediction of Death from Cardiovascular Causes. N Engl J Med 358: 2107-2116. doi:https://doi.org/10.1056/NEJMoa0707064. PubMed: 18480203.
  28. 28. Matsushita K, Mahmoodi BK, Woodward M, Emberson JR, Jafar TH et al. (2012) Comparison of risk prediction using the CKD-EPI equation and the MDRD study equation for estimated glomerular filtration rate. JAMA 307: 1941–1951. doi:https://doi.org/10.1001/jama.2012.3954. PubMed: 22570462.
  29. 29. Cirillo M, Lombardi C, Mele AA, Marcarelli F, Bilancio G (2012) A population-based approach for the definition of chronic kidney disease: the CKD Prognosis Consortium. J Nephrol 25: 7-12. doi:https://doi.org/10.5301/jn.5000045. PubMed: 22038337.
  30. 30. Mahmoodi BK, Matsushita K, Woodward M, Blankestijn PJ, Cirillo M et al. for the Chronic Kidney Disease Prognosis Consortium (2012) Associations of kidney disease measures with mortality and end-stage renal disease in individuals with and without hypertension: a meta-analysis. Lancet 380: 1-13. doi:https://doi.org/10.1016/S0140-6736(11)61925-9. PubMed: 22770441.
  31. 31. Pinto-Sietsma SJ, Mulder J, Janssen WM, Hillege HL, de Zeeuw D et al. (2000) Smoking is related to albuminuria and abnormal renal function in nondiabetic persons. Ann Intern Med 133: 585-591. doi:https://doi.org/10.7326/0003-4819-133-8-200010170-00008. PubMed: 11033585.
  32. 32. Nitsch D, Felber Dietrich D, von Eckardstein A, Gaspoz JM, Downs SH et al. (2006) Prevalence of renal impairment and its association with cardiovascular risk factors in a general population: Results of the Swiss SAPALDIA study. Nephrol Dial Transplant 21: 935–944. PubMed: 16390852.
  33. 33. Ishizaka N, Ishizaka Y, Toda E, Shimomura H, Koike K et al. (2008) Association between cigarette smoking and chronic kidney disease in Japanese men. Hypertens Res 31: 485- 492. doi:https://doi.org/10.1291/hypres.31.485. PubMed: 18497468.
  34. 34. Mathisen UD, Melsom T, Ingebretsen OC, Jenssen T, Njølstad I et al. (2011) Estimated GFR Associates with Cardiovascular Risk Factors Independently of Measured GFR. J Am Soc Nephrol 22: 927–937. doi:https://doi.org/10.1681/ASN.2010050479. PubMed: 21454717.
  35. 35. Pinto-Sietsma SJ, Janssen WM, Hillege HL, Navis G, De Zeeuw D et al. (2000) Urinary albumin excretion is associated with renal functional abnormalities in a nondiabetic population. J Am Soc Nephrol 11: 1882-1888. PubMed: 11004219.
  36. 36. Gislefoss RE, Grimsrud TK, Mørkrid L (2008) Long-term stability of serum components in the Janus Serum Bank. Scand_J Clin Lab Invest 68: 402-409. doi:https://doi.org/10.1080/00365510701809235. PubMed: 18752145.
  37. 37. Greenberg N, Roberts WL, Bachmann LM, Wright EC, Dalton RN et al. (2012) Specificity characteristics of 7 commercial creatinine measurement procedures by enzymatic and Jaffe method principles. Clin_Chem 58: 391-401. doi:https://doi.org/10.1373/clinchem.2011.172288. PubMed: 22166253.
  38. 38. Donfrancesco C, Palmieri L, Vanuzzo D, Panico S, Ferrario MM et al. on behalf of Progetto CUORE and Osservatorio Epidemiologico Cardiovascolare research groups (2010) Il Progetto CUORE: analisi preliminari per l'aggiornamento delle carte del rischio e del punteggio individuale. G Ital Cardiol 11 Suppl 3(5): 20S-24S.