Abstract
Renal insufficiency is a serious complication of lithium treatment. Therefore, regular monitoring of plasma (P) creatinine is always part of lithium treatment safety routines. Recently P-cystatin C-estimated glomerular filtration rate (cystatin C-eGFR) had been launched as a preferable alternative to P-creatinine. To find out which of the two alternatives to prefer in the safety routines for lithium-treated patients. All 201 patients on lithium treatment at the Department of Psychiatry in Lund, Sweden. During 14 months P-cystatin C was included in the safety routines besides routine P-creatinine every 4 months. At the end of the study period, 182 patients were eligible for analysis. With iohexol clearance as reference for GFR (performed in 111/182 patients) we calculated positive and negative predictive values (PPV, NPV) for P-creatinine and for creatinine-eGFR and cystatin C-eGFR, obtained by prediction equations. We also calculated the agreement between the measures of GFR (including repeatability). PPV for cystatin C-eGFR (65%) was better than for creatinine-eGFR (48%). Combining the two resulted in a PPV of 56% and marginally increased NPV to 95%. The average of cystatin C-eGFR and creatinine-eGFR yielded PPV 67% and NPV 92%. The agreement between creatinine-eGFR and GFR was better than the agreement between cystatin C-eGFR and GFR, but both were clinically unacceptable. The repeatability of P-creatinine was acceptable for psychiatric purposes. The repeatability of cystatin C-eGFR was inferior to that of P-creatinine. Our results do not justify replacing P-creatinine by cystatin C-eGFR in the lithium treatment safety routines.