Aims/hypothesis: We sought to characterise the effect of the age-related decline of GFR on hyperfiltration in type 2 diabetes and to identify clinical characteristics associated with hyperfiltration. Materials and methods: GFR was measured in 662 type 2 diabetic patients by plasma disappearance of 99 m-technetium-diethylene-triaminepenta-acetic acid. The prevalence of hyperfiltration was calculated using both an age-unadjusted GFR threshold of >130 ml min −1 1.73 m −2 and an age-adjusted threshold incorporating a decline of 1 ml min −1 year −1 after the age of 40. The hyperfiltering patients were compared with type 2 diabetic subjects who had a GFR between 90 and 130 ml min −1 1.73 m −2 and were matched for age, sex and disease duration to allow for identification of modifiable factors associated with hyperfiltration. Results: The prevalence of hyperfiltration was 7.4% when age-unadjusted and 16.6% when age-adjusted definitions were used. The age-unadjusted vs -adjusted prevalence rates for hyperfiltration were 50 vs 50%, 12.9 vs 23.4% and 0.3 vs 9.0% for patients aged <40 years, 40 to 65 years and >65 years, respectively. Both the age-unadjusted and -adjusted hyperfiltration groups had lower mean diastolic blood pressure and lower serum creatinine levels than the control groups. Although the ageunadjusted hyperfiltration group had larger kidneys compared to the control group, this difference was no longer significant when the age-adjusted definition was used. There were no differences in HbA 1 c, mean arterial pressure, antihypertensive use, insulin therapy, dyslipidaemia, frequency of macro-or microvascular complications, BMI, urinary sodium, urea and albumin excretion between the groups. Conclusions/interpretation: Hyperfiltration was still more common among younger patients with type 2 diabetes even after adjusting for the expected age-related decline in GFR. Hyperfiltration was associated with a lower mean diastolic blood pressure independent of age.