Stratification in Determining Bias in GFR Estimating Equations

Determining bias is important when assessing glomerular filtration rate (GFR), and stratification by subgroups can show where equations perform differently. Stratifying on the level of only eGFR and measured glomerular filtration rate (mGFR) is widespread but can lead to erroneous conclusions.

Therefore, Derek K. Ng and other researchers compared and contrasted biases (eGFR relative to mGFR) with 95% CI within strata of mGFR only, eGFR only, and the average of mGFR and eGFR using data from the Chronic Kidney Disease in Children study.

A total of 304 participants provided 843 GFR studies with a mean mGFR of 48.46 (SD=22.72) and mean eGFR of 48.67 (SD=22.32). Correlation was 0.904. Although there was strong agreement, eGFR greatly overestimated mGFR when mGFR was <30 (+6.2%; 95% CI, +2.9% to +9.7%); eGFR vastly underestimated mGFR when mGFR was >90 (–12.2%; 95% CI, −17.3% to −7.0%).

There were also significant biases in the opposite direction when stratifying by eGFR only. Conversely, when stratifying by the average of eGFR and mGFR, biases were not significant (+1.3% and −1.0%, respectively), consistent with strong agreement.

The study authors concluded that, although it is common, stratifying only by mGFR or eGFR to assess eGFR biases can lead to inappropriate inference because of intrinsic statistical issues highlighted by their research. Instead, they recommend using the average of eGFR and mGFR for valid inferences in evaluations of eGFR biases.

Source: Pediatric Nephrology