Extended-spectrum beta-lactamase-producing Enterobacteriacae: unexpected low prevalence of carriage in elderly French residents.

limited number of patients. Using these limited albuminuria and proteinuria data would have caused substantial selection bias. Second, not all creatinine values were measured by the same laboratory or by the same creatinine assay due to the design of the database, which collects data from practices throughout Flanders. However, all Belgian laboratories are subject to quality control measures (http://www.iph.fgov.be/ ClinBiol/bckb33/activities/external_quality/rapports/_down/ klinische_ chemie/2003/2_CHIMIE.pdf.2009), which diminished the analytical differences among the laboratories. In the period between 1994 and 2008, most laboratories in Belgium used a kinetic Jaffe method without IDMS standardisation. Finally, we used the four variable MDRD equations for the estimation of the GFR. This equation is currently the most used equation but is originally [3] constructed in a population with renal diseases aged younger than 70 years and the number of studies validating this formula in patients aged 70 and over remains limited until now [8]. To conclude the prevalence of CKD in general practice increases with ageing but the prevalence of stages 4 and 5 CKD is limited, even in the oldest of the old. In the subgroup of patients aged 90 years and older more than 40% of the patients had an eGFR ≥60 ml/min. So a large group of patients show no decline or no clinically relevant decline in eGFR with ageing.