Introduction: The purpose of the study was to compare the performance of the simplified Modification of Diet in Renal Disease (MDRD) and Cockcroft-Gault (CG) equations with 24-hour urinary creatinine clearance (Ccr) in patients with spinal cord injury (SCI) and chronic kidney disease.
Methods: A retrospective diagnostic accuracy study of 116 patients with chronic SCI followed at the Memphis Veterans Affairs Medical Center Spinal Cord Injury Unit with measured Ccr (mCcr) <90 mL/min/1.73 m.
Results: Linear regression analysis relating estimated glomerular filtration rate (eGFR) to mCcr showed a highly significant correlation between mCcr and eGFR (n = 116; r = 0.81, r = 0.65, P < 0.0001); however, the relationship was more variable in the quadriplegic subjects (n = 52; r = 0.74, r = 0.54, P < 0.0001) than in the paraplegic subjects (n = 64; r = 0.86, r = 0.73, P < 0.0001). Both eGFR equations overestimated glomerular filtration rate (GFR) at all ranges of GFR in both subgroups of paraplegic subjects and quadriplegic subjects, with an MDRD fractional prediction error of 49% and 62%, respectively. Addition of a correction factor of 0.7 for MDRD and 0.8 for CG equations resulted in clinically acceptable fractional prediction error (below 20%) in both subgroups, especially in paraplegics with 3.9% and 3.6%, respectively. There was marked improvement in the performances of both eGFR equations, with better accuracy and precision after application of the correction factors.
Conclusions: Both MDRD and CG equations overestimate GFR in patients with chronic SCI at all stages of chronic kidney disease, particularly in quadriplegic subjects. An empirically derived correction factor markedly improved the performance and accuracy of both prediction equations.