Measured GFR does not outperform estimated GFR in predicting CKD-related complications

Abstract

Although many assume that measurement of glomerular filtration rate (GFR) using a marker such as iothalamate (iGFR) is superior to equation-estimated GFR (eGFR), each of these methods has distinct disadvantages. Because physicians often use renal function to guide the screening for various CKD-associated complications, one method to compare the clinical utility of iGFR and eGFR is to determine the strength of their association with CKD-associated comorbidities. Using a subset of 1214 participants in the Chronic Renal Insufficiency Cohort (CRIC) Study, we determined the cross-sectional associations between known complications of CKD and iGFR, eGFR estimated from serum creatinine (eGFR_Cr), and eGFR estimated from cystatin C (eGFR_cysC). We found that none of the measures of renal function strongly associated with CKD complications and that the relative strengths of associations varied according to the outcome of interest. For example, iGFR demonstrated better discrimination than eGFR_Cr and eGFR_cysC for outcomes of anemia and hemoglobin concentration; however, both eGFR_Cr and eGFR_cysC demonstrated better discrimination than iGFR for outcomes of hyperphosphatemia and phosphorus level. iGFR and eGFR had similar strengths of association with hyperkalemia/potassium level and with metabolic acidosis/bicarbonate level. In conclusion, iothalamate measurement of GFR is not consistently superior to equation-based estimations of GFR in explaining CKD-related comorbidities. These results raise questions regarding the conventional view that iGFR is the "gold standard" measure of kidney function.

Figure 1.

Scatterplot showing levels of hemoglobin, potassium, bicarbonate, and phosphate versus GFR directly measured using urinary clearance of iothalamate (iGFR), serum creatinine estimated GFR (eGFR_Cr), and serum cystatin C estimated GFR (eGFR_cysC).