Influence of muscle mass on the estimation of glomerular filtration rate in Japanese terminal cancer patients

Background Estimated glomerular filtration rate (eGFR) based on serum cystatin C (Scys) is useful for patients with decreased muscle mass, but has been also reported to be affected by cancer. The usefulness of Scys in eGFR in terminal cancer patients with decreased muscle mass is unknown. Therefore, we analyzed appropriate eGFR formulae for terminal cancer patients. Methods Study design was a retrospective observational study. Based on creatinine height index (CHI), 184 terminal cancer patients were stratified into CHI >= 90% (normal muscle mass, 59 patients); CHI 60-89% (mildly to moderately decreased muscle mass, 64 patients); and CHI < 60% (severely decreased muscle mass, 61 patients) groups. Twenty-four-hour creatinine clearance was measured and converted to the glomerular filtration rate (GFR) as a renal function measure. To estimate GFR, various eGFR formulae for Japanese were used: eGFR(Scys), eGFR(Scr5)and eGFR(Scr3), eGFR(average)and eGFR(Scys-Scr), and eGFR(CG), based on Scys, serum creatinine (Scr), Scys and Scr combined, and Cockcroft-Gault formula (CG), respectively. Errors between measured and estimated values of renal function were verified using mean prediction errors (ME). When a 95% confidence interval (CI) of ME included 0, the accuracy of the eGFR formula was graded as good. Results eGFR(Scys)ME was 0.2 (95% CI lower limit - 3.7, upper limit 4.0) mL/min/1.73 m(2)in CHI 60-89% group and 9.2 (6.1, 12.9) mL/min/1.73 m(2)in CHI < 60% group. eGFR(Scys)was most accurate among the eGFR formulae. Conclusions eGFR based on Scys was demonstrated as useful in terminal cancer patients with decreased muscle mass.