Diabetes influences the performance of creatinine-based equations for estimating glomerular filtration rate in the elderly population

Diabetic status and the performances of creatinine- and cystatin C-based eGFR equations

BACKGROUND

The estimation of glomerular filtration rate (GFR) is one tool to detect renal disease. The most used biomarker remains serum creatinine and the European Kidney Function Consortium (EKFCcrea) equation is the most validated in Europe. More recently, cystatin C has been proposed as a biomarker. We studied the performances of the EKFC equations in a large cohort of subjects according to their diabetic status.

METHODS

Four cohorts from the EKFC dataset were retrospectively considered in which the diabetic status was available. GFR was measured by plasma clearances (mGFR; iohexol or chromium 51-ethylenediaminetetraacetic acid). The performance of the equations was assessed by calculating bias, precision [interquartile range (IQR)] and P30 (percentage of eGFR values within ±30% of mGFR).

RESULTS

In the whole population (N = 6158), the median age was 61 years (IQR 47-72) and 45.8% were women. The mean mGFR was 60 ml/min/1.73 m2 (IQR 39-82). Compared with non-diabetic individuals (n = 5124), diabetic patients (n = 1034) were older, more frequently male, heavier and had lower mGFR. The performance of the EKFCcys equation was similar to that of the EKFCcrea equation, but the EKFCcrea+cys equation had a better P30 than the single-biomarker equations. P30 values were substantially lower in diabetic patients than in non-diabetic patients, but according to a matched analysis, this is mainly explained by the difference in GFR levels between the two populations, not by diabetic status.

CONCLUSION

We showed that the equation combining creatinine and cystatin C performed better. If the accuracy of equations seems better in non-diabetic than in diabetic individuals, it is more likely due to differences in GFR levels rather than diabetic status.

Assessment of the 2023 European Kidney Function Consortium (EKFC) equations in a Chinese adult population

OBJECTIVES

The European Kidney Function Consortium (EKFC) developed two novel equations in 2023 for estimating glomerular filtration rate (GFR): one sex-free cystatin C-based equation (EKFCCys) and one creatinine-cystatin C combined equation (EKFCCr-Cys). This study compared their performance with the previous creatinine-based EKFC equation (EKFCCr) and commonly used Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) and Berlin Initiative Study (BIS) equations in Chinese adults.

METHODS

A total of 2,438 Chinese adults (mean age=53.04 years) who underwent the 99mTc-DTPA renal dynamic imaging for reference GFR (rGFR) were included. Diagnostic value was evaluated using correlation coefficients, sensitivity, specificity, and area under the receiver operating characteristic curve (ROCAUC). Performance was assessed in terms of bias, precision (interquartile range of the median difference [IQR]), accuracy (percentage of estimates ±30 % of rGFR [P30], and root-mean-square error [RMSE]) across age, sex, and rGFR subgroups. Gender differences in bias and P30 were also analyzed.

RESULTS

Average rGFR was 73.37 mL/min/1.73 m2. EKFC equations showed stronger correlations and larger AUCs compared to the parallel CKD-EPI equations, with EKFCCr-Cys demonstrating the greatest improvement (R=0.771, ROCAUC=0.913). Concerning bias, precision, and accuracy, EKFC equations consistently outperformed CKD-EPI equations. EKFCCr-Cys and EKFCCr performed acceptably well in the entire population and were equivalent to BIS equations in the elderly. All equations, including EKFCCys, showed similar P30 accuracy across sexes.

CONCLUSIONS

EKFC equations provided a reasonable alternative for estimating GFR in the Chinese adult population. While EKFCCys did not outperform EKFCCr, EKFCCr-Cys improved the accuracy of single-marker equations.

Comparison of glomerular filtration rate estimating equations in older adults: A systematic review and meta-analysis

BACKGROUND

Debates persist regarding the performance of existing glomerular filtration rate (GFR) estimating equations in older individuals. We performed this meta-analysis to assess the accuracy and bias of six commonly used equations, including the Chronic Kidney Disease Epidemiology Collaboration creatinine equation (CKD-EPICr) and its combination with cystatin C (CKD-EPICr-Cys), with the corresponding pair of the Berlin Initiative Study equations (BIS1 and BIS2) and the Full Age Spectrum equations (FASCr and FASCr-Cys).

METHODS

PubMed and the Cochrane Library were searched for studies comparing estimated GFR (eGFR) with measured GFR (mGFR). We analyzed the difference in P30 and bias among the six equations and investigated subgroups based on the area (Asian and non-Asian), mean age (60-74 years and ≥75 years), and levels of mean mGFR (<45 mL/min/1.73m2 and ≥45 mL/min/1.73m2).

RESULTS

27 studies with 18,112 participants were included, all reporting P30 and bias. BIS1 and FASCr exhibited significantly higher P30 than CKD-EPICr. While no significant differences were observed between FASCr and BIS1, or among the three combined equations in terms of either P30 or bias. Subgroup analyses revealed FASCr and FASCr-Cys achieved better results in most situations. However, in the subgroup of mGFR<45 mL/min/1.73m2, CKD-EPICr-Cys had relatively higher P30 and significantly smaller bias.

CONCLUSIONS

Overall, BIS and FAS provided relatively more accurate estimates of GFR than CKD-EPI in older adults. FASCr and FASCr-Cys may be better suited for various conditions, while CKD-EPICr-Cys would be a better option for older individuals with impaired renal function.

A back propagation neural network approach to estimate the glomerular filtration rate in an older population

BACKGROUND

The use of creatinine-based glomerular filtration rate (GFR)-estimating equations to evaluate kidney function in elderly individuals does not appear to offer any performance advantages. We therefore aimed to develop an accurate GFR-estimating tool for this age group.

METHODS

Adults aged ≥ 65 years who underwent GFR measurement by technetium-99 m-diethylene triamine pentaacetic acid (^99mTc-DTPA) renal dynamic imaging were included. Data were randomly split into a training set containing 80% of the participants and a test set containing the remaining 20% of the subjects. The Back propagation neural network (BPNN) approach was used to derive a novel GFR estimation tool; then we compared the performance of the BPNN tool with six creatinine-based equations (Chronic Kidney Disease-Epidemiology Collaboration [CKD-EPI], European Kidney Function Consortium [EKFC], Berlin Initiative Study-1 [BIS1], Lund-Malmö Revised [LMR], Asian modified CKD-EPI, and Modification of Diet in Renal Disease [MDRD]) in the test cohort. Three equation performance criteria were considered: bias (difference between measured GFR and estimated GFR), precision (interquartile range [IQR] of the median difference), and accuracy P30 (percentage of GFR estimates that are within 30% of measured GFR).

RESULTS

The study included 1,222 older adults. The mean age of both the training cohort (n = 978) and the test cohort (n = 244) was 72 ± 6 years, with 544 (55.6%) and 129 (52.9%) males, respectively. The median bias of BPNN was 2.06 ml/min/1.73 m², which was smaller than that of LMR (4.59 ml/min/1.73 m²; p = 0.03), and higher than that of the Asian modified CKD-EPI (-1.43 ml/min/1.73 m²; p = 0.02). The median bias between BPNN and each of CKD-EPI (2.19 ml/min/1.73 m²; p = 0.31), EKFC (-1.41 ml/min/1.73 m²; p = 0.26), BIS1 (0.64 ml/min/1.73 m²; p = 0.99), and MDRD (1.11 ml/min/1.73 m²; p = 0.45) was not significant. However, the BPNN had the highest precision IQR (14.31 ml/min/1.73 m²) and the greatest accuracy P30 among all equations (78.28%). At measured GFR < 45 ml/min/1.73 m², the BPNN has highest accuracy P30 (70.69%), and highest precision IQR (12.46 ml/min/1.73 m²). The biases of BPNN and BIS1 equations were similar (0.74 [-1.55-2.78] and 0.24 [-2.58-1.61], respectively), smaller than any other equation.

CONCLUSIONS

The novel BPNN tool is more accurate than the currently available creatinine-based GFR estimation equations in an older population and could be recommended for routine clinical use.

The Chronic Kidney Disease Epidemiology Collaboration equations perform less well in an older population with type 2 diabetes than their non-diabetic counterparts

Objectives

The Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations are based on creatinine alone (CKD-EPIcr), cystatin C alone (CKD-EPIcys) and combined creatinine and cystatin C (CKD-EPIcr-cys). It remains unclear whether these equations perform differently in older adults with type 2 diabetes than they do in non-diabetic older individuals.

Methods

This single-center cross-sectional study was performed in adults aged ≥ 65 years between January 2019 and December 2021. Glomerular filtration rate (GFR) was measured by technetium-99m-diethylene triamine pentaacetic acid (^99mTc-DTPA) renal dynamic imaging. The bias (difference between measured and estimated GFR), precision [interquartile range (IQR) of the median difference between measured GFR and estimated GFR] and accuracy P30 (percentage of estimated GFR within 30% of measured GFR) were considered the criteria of equation performance.

Results

Finally, 476 participants were enrolled, including 243 adults with type 2 diabetes and 233 non-diabetic adults. The mean age of the included participants was 71.69 ± 6.4 years and 262 (55%) were male. The mean measured GFR was 49.02 ± 22.45 ml/min/1.73 m². The CKD-EPIcr-cys equation showed significantly greater bias and lower accuracy (P30) in individuals with diabetes than in the non-diabetic group (median bias, 4.08 vs. 0.41 ml/min/1.73 m², respectively, p < 0.05; P30, 63.78% vs. 78.54%, respectively, p < 0.05). The precision IQR indicated that CKD-EPIcr-cys had also lower precision in individuals with diabetes than in the non-diabetic controls (17.27 vs. 15.49 ml/min/1.73 m², respectively). Similar results were observed for CKD-EPIcr and CKD-EPIcys equations. The P30 of all three equations failed to reach 80% in diabetic and non-diabetic groups.

Conclusions

The performance of the CKD-EPI equations was lower in a group of patients aged ≥ 65 years with type 2 diabetes than in non-diabetic counterparts. However, each equation still had limitations regarding accuracy in older adults with or without diabetes.