Due to the paucity of long-term observational data, it is still unclea
r whether the decreased glomerular filtration rate (GFR) noted in olde
r humans is a result of true changes in renal function over time. The
fact that a carefully characterized subpopulation of subjects showed n
o decrease in GFR over time suggests that the 'physiological decrease'
in GFR noted in the elderly is not inevitable. In studies in our pati
ent population, there was a significant correlation between creatinine
clearance and nutritional protein intake in elderly patients without
renal disease. In our studies, elderly subjects without renal disease
who ingested >1 g/kg day of protein had creatinine clearances in the r
ange of 90-100 ml/min/1.73 m(2), while those with lower protein intake
s had lower creatinine clearances. Our studies suggest that GFR is not
a fixed function, and that its value may change both over short perio
ds and over decades in humans, with these changes being associated wit
h changes in nutritional protein intake. Low levels of GFR are not nec
essarily equivalent to a diagnosis of renal disease. Diet must be cons
idered in the assessment of renal function in elderly patients before
a diagnosis of renal insufficiency can be made. Decreased GFR is not a
n inevitable consequence of ageing in humans.