Key words: Elderly, 80 and over, estimating equations, kidney function tests. Correspondence: Claudio Pedone, MD, MPH, Area di Geriatria, Università Campus Biomedico, Via dei Compositori 130, 00128 Roma, Italy. E-mail: c.pedone@unicampus.it Received June 14, 2007; accepted in revised form January 16, 2008. Reliability of equations to estimate glomerular filtration rate in the very old Aging Clinical and Experimental Research Claudio Pedone 1 , Roberto Semeraro 2 , Domenica Chiurco 1 , Francesco D’Andria 2 , Maurizio Gigante 2 , Anna Coppola 2 , Andrea Corsonello 3 and Raffaele Antonelli-Incalzi 1,2 1 Area di Geriatria, Università Campus Biomedico, Roma, 2 Fondazione S. Raffaele - Cittadella della Carità, Taranto, 3 Istituto Nazionale di Ricovero e Cura per Anziani (INRCA), Cosenza, Italy ABSTRACT. Background and aims: Few studies have investigated the reliability of formulas estimating renal function in very old people. Methods: We studied 154 el- derly people (mean age: 82 yrs). Serum creatinine (SC) was measured by the Jaffé method, and creatinine clear- ance (CrCl) with 24-h urine collection. Agreement was measured with the average ratio estimated/measured CrCl, and precision with the 95% agreement intervals (95% AI). We calculated the proportion of residents correctly classified as having renal insufficiency (accuracy). Results: The Cockcroft-Gault (CG) and Modification of Diet in Renal Disease 1 (MDRD1) formulas showed good average agreement with measured CrCl (0.95 and 1.016, respectively); the MDRD2 formula was more bi- ased. Results were consistent in women, whereas the MDRD1 was more biased in men (average ratio: 1.196). The 95% AI showed that all formulas can yield results as low as 50% or as high as 200% of measured CrCl. The proportion of people with CrCl<60 ml/min misclassified by the CG, MDRD1, and MDRD2 formulas as having normal renal function was 21.4%, 27.0%, and 38.8%, re- spectively. These results were consistent across the var- ious subgroups, especially in subjects with normal SC. Conclusions: The clinical usefulness of formulas com- monly used to estimate CrCl was limited, regardless of subjects’ characteristics. (Aging Clin Exp Res 2008; 20: 496-502) © 2008, Editrice Kurtis in the general population (2) and the elderly (3), as well as in selected populations such as those with previous my- ocardial infarction, congestive heart failure or chronic obstructive pulmonary disease (4-6). CRF also affects the kinetics of several drugs and then requires dose adjustment to prevent overdosage and adverse drug reactions. Diagnosing CRF is crucial for optimal management of elderly patients, but the most easily available marker of this disease, serum creatinine, is not sufficiently sensitive. The age-related loss of muscle mass and perhaps also the age-related decrease in protein intake can maintain serum creatinine normal in spite of depressed clearance (7). Various formulas, such as the Cockcroft-Gault (CG) (8) or Modification of Diet in Renal Disease (MDRD) (9, 10) for- mulas, have been proposed to estimate creatinine clear- ance or the glomerular filtration rate at the bedside. However, elderly people have been underrepresented in the populations used to develop these predictive equa- tions, and studies carried out to test them in elderly pop- ulations have shown disappointing results. In one series of 1246 home-dwelling subjects over 64, the CG and MDRD formulas yielded 58.6% and 35.7% prevalence of renal failure, but their accuracy could not be estimated be- cause the GFR was not directly measured (11). Similarly, the CG formula gave a higher estimate of renal failure than the MDRD in a population of 9931 Canadians over 64 years old living in long-term facilities (12). A study conducted in 154 octagenarian in-patients showed that the CG, MDRD and Jeliffe formulas all overestimated measured creatinine clearance (13). Even in young adult subjects with chronic kidney disease and normal serum creatinine, none out of 8 predictive equations achieved a combination of accuracy and precision sufficient to justi- fy its routine use (14). Lastly, in healthy adult subjects, the MDRD formula underestimates GFR by 29% (15). ORIGINAL ARTICLES INTRODUCTION Chronic renal failure (CRF), defined as a glomerular fil- tration rate (GFR) less than 60 ml/min/1.73 m 2 , is com- mon in the elderly, with an estimated prevalence of 18.4% after the age of 70, compared with 4.0 in people aged 50- 69 years (1). It is an independent risk factor for mortality Aging Clin Exp Res, Vol. 20, No. 6 496 pprop rop bbyy tth no norr 0.95 0.95 an andd a wass more more bi bi-- women, men, wwhereas hereas the th iinn mmen (a ( verage rage ratio: ratio: 1.196 11 ) 6 dd tthat h all all fform or f f f ulas u cann yield yie r % or or aass hhigh igh as a 20 2000%% of of mea por orti tion on of of peo eopp oo le le wwit ith CCrC Cl<60 6 hee CCG, G, MDR MDRD1, D and all ree r r r nal nal fun cti ely T dj d ustm ug re reacti actions ons.. optimmaall manag anagemment ent mos mostt eaassillyy l l aava va a ila labl b a ee mmaarrke ker of th creattiiniine ne, iss nnot ot ssuf uffi ficiently elated loss off mus muscl clee mmass a and e-r e elat ated ed ddec ecrrea ease s in pr prootein in ccre reaati t nine nine no normal rma in s Va Vari rious us ffor ormula MModdifficati m Cl. Cl. Th Thee min min mi misc sclassified lassified D2 2 formulas formulas as as hav hav 21.4% 4%,, 227.0 7.0%, % and and resul esults ts wwere e ccon onsiste t ups, ps, esp especially ecially in in : The The clin clin n main main dep epre ressse sedd cle clear s thee CCoock ckcr crof oftt f --GGaa ll iiet et iinn RRenal enal Di Disseas eas avvee been been pr proopo possed e t nce e or thhee gglome lomer Howe Howeve ver, r elde ld in in tthe he