Background: We examined associations between cardiovascular diseases and risk factors with pathological levels of and significant changes in serum creatinine (SCr) in a large prevalence phase and longitudinal phase community-based sample of an elderly Italian population (ILSA Study) showing no clinical evidence of renal impairment.
Methods: The prevalence phase was performed on 2981 subjects, aged 65-84 years, who were negative for renal diseases, had available SCr values and had complete clinical information on their cardiovascular risk factors. Of these, 371 were considered 'healthy' since they were not affected by cardiovascular diseases or diabetes, whereas 2610 tested positive for cardiovascular diseases and were considered 'diseased'. The sex-specific 95th percentiles for SCr (cut-off points) were calculated in the healthy reference sample to define the upper limit for normal SCr values. The distribution and prevalence of diseased subjects having values over the cut-off point values were then estimated. Associations between values over the cut-off point levels and pathological or clinical conditions were analysed from the diseased sample. The longitudinal phase was carried out on 1906 subjects who had SCr values and sufficient clinical information for our investigation. The incidence of an increase of >26.5 micromol/l of SCr was evaluated in the longitudinal cohort.
Results: In healthy subjects, the 95th SCr percentiles (cut-off points) were 123.8 micromol/l in men and 97.2 micromol/l in women. In diseased subjects, the prevalence of SCr values over the cut-off point was 4.6% in men and 9.3% in women. In logistic regression analysis, independent variables that correlated with over the cut-off point SCr values were: age >75 years [odds ratio (OR) = 2.2; 95% confidence interval (CI) = 1.5-3.4], atherosclerosis of the lower limbs (OR = 2.0; 95% CI = 1.2-3.3), cerebrovascular disease (OR = 1.9; 95% CI = 1.2-3.3), angiotensin-converting enzyme (ACE) inhibitor medication (OR = 1.8; 95% CI = 1.2-2.8), fibrinogen values >3.5 g/l (OR = 1.2; 95% CI = 1.2-2.7) and diuretic treatment (OR = 1.6; 95% CI = 1.1-2.4). After a mean 3.6 years follow-up, multiple logistic regression analysis showed that risk factors for pathological loss of renal function (rise of SCr >26.5 micromol/l) were: current smokers >20 cigarettes/day (OR = 2.3; 95% CI = 1.0-5.3), fibrinogen values >3.5 g/l (OR = 2.2; 95% CI = 1.6-3.3), diabetes (OR = 1.8; 95% CI = 1.1-2.8), age >75 years (OR = 1.7; 95% CI = 1.2-2.4) and isolated systolic hypertension (OR = 1.6; 95% CI = 1.0-2.6). The loss of renal function examined during the longitudinal phase appeared to be independent of baseline SCr levels.
Conclusion: The present prevalence and longitudinal studies show that age-associated decline in renal function in elderly subjects is associated with co-existing cardiovascular diseases and risk factors. These observations should be incorporated into clinical practice since some of the factors detrimental to kidney function, such as smoking, altered fibrinogen levels and elevated systolic blood pressure, can be prevented and/or modified when appropriate measures are taken.