Home About us Current issue Ahead of Print Back issues Submission Instructions Advertise Contact Login   

Search Article 
Advanced search 
Saudi Journal of Kidney Diseases and Transplantation
Users online: 4210 Home Bookmark this page Print this page Email this page Small font sizeDefault font size Increase font size 

Table of Contents   
Year : 2021  |  Volume : 32  |  Issue : 2  |  Page : 398-414
Comparison of various predictive equations for glomerular filtration rate in healthy individuals and chronic kidney disease patients from North India

Department of General Medicine, Government Medical College and Hospital, Chandigarh, India

Click here for correspondence address and email

Date of Web Publication11-Jan-2022


A noninvasive and accurate estimate of the glomerular filtration rate (GFR) is an essential prerequisite for medical professionals. In the absence of 24-h urinary creatinine clearance, various predictive equations can be utilized for estimating GFR. A cross-sectional observational study was conducted on healthy adults as well as adult chronic kidney disease (CKD) Indian population. In normal males and females, Modification of Diet in Renal Disease-4 (MDRD-4) and Cockcroft- Gault (CG) were the best equations respectively, which showed the best correlation and best precision in CKD stage 1 males and females, MDRD-4 and MDRD-6 were the best equations respectively In CKD stage 2 males and females, CKD-Epidemiology Collaboration (CKD-EPI) was adjudged the best equation, which showed the best correlation, best precision, and least bias. In CKD stage 3 males and females, CG and CKD-EPI were the best equation respectively, with the best correlation, best precision, least biased, and most accuracy. In CKD stage 4 males and females, MDRD-6 and MDRD-4 showed the best correlation, best precision and most accuracy respectively. In CKD stage 5, CKD-EPI demonstrated the best results in both sexes. We observed that all the predictive equations were good estimates of GFR in one or other stages of CKD, but no single predictive equation showed consistent results when compared among normal subjects and CKD sub-groups.

How to cite this article:
Janjirala SV, Gupta M, Cruz SD. Comparison of various predictive equations for glomerular filtration rate in healthy individuals and chronic kidney disease patients from North India. Saudi J Kidney Dis Transpl 2021;32:398-414

How to cite this URL:
Janjirala SV, Gupta M, Cruz SD. Comparison of various predictive equations for glomerular filtration rate in healthy individuals and chronic kidney disease patients from North India. Saudi J Kidney Dis Transpl [serial online] 2021 [cited 2022 Aug 11];32:398-414. Available from: https://www.sjkdt.org/text.asp?2021/32/2/398/335452

   Introduction Top

Chronic kidney disease (CKD) is a significant public health problem due to its numerous social, psychological, and financial implications on individual and health resources. There is progressive loss of renal function in CKD. Renal function measured by glomerular filtration rate (GFR) is the most widely accepted standard for the assessment of renal functions in health and disease. The normal value of GFR is ~130 mL/min/m.[1] GFR is more reliable than blood urea nitrogen and serum creatinine (SCr) which does not rise above the normal range until 60% of kidney function is lost. SCr also varies substantially across age, gender, ethnicity, and anthropometric measurements. Creatinine clearance (CrCl) rate is the volume of blood plasma that is cleared of creatinine per unit time. This method does not have drawbacks such as hospitalization, radiation exposure, repetitive sampling, and hourly intravenous infusions. The main concern is difficulty in assuring complete 24-h urine collection.

There are different techniques to calculate GFR precisely, typically by using various filtration markers like inulin, I-iothalamate, Cr-ethylenediaminetetraacetic acid, 99Tc-diethylenetriaminepentaacetic acid, and iohexol.[1] Unfortunately, these methods cannot be implemented in daily practice, because they are expensive and time consuming. Various equations have been developed and tested by research groups and the most validated among these are the Cockcroft-Gault formula (CG), Modification of Diet in Renal Disease (MDRD-4 and MDRD-6), and CKD-Epidemiology Collaboration (CKD-EPI) equation. Equations estimating GFR based on SCr adjusted for the above factors are more accurate and precise than estimates of GFR from SCr alone.[2] Although validated in large groups of patients and healthy individuals worldwide, these equations may have some inherent bias and therefore their applicability in the Indian population needs to be further verified. There is also paucity of data regarding the utility and optimality of these predictive equations in both healthy and CKD. The present study focused on comparing the precision and accuracy of GFR estimated by the current equations with the GFR measured by 24-h endogenous CrCl in the north Indian population.

   Aims and Objectives Top

We aimed to assess the GFR in healthy and CKD patients by different methods of estimating GFR; and to validate the presently available predictive equations for estimating GFR in the Indian population.

   Materials and Methods Top

Study design and setting

A cross-sectional observational study was conducted between January 2014 and June 2015 on participants enrolled from the Department of Medicine’s outpatient, inpatient areas, and renal clinic at Government Medical College and Hospital, Chandigarh.

Inclusion criteria

Included were patients with CKD stage 1–5 (as outlined below); healthy individuals with normal GFR (controls). Stages of CKD were defined as per guidelines of the National Kidney Foundation.[2]

Exclusion criteria

Those who were <20 or >0 years of age; immunocompromised patients or kidney transplant recipients; patients with single kidney; acutely ill patients; patients with body mass index <18 and >40 kg/m2; patients with amputated limbs; pregnancy; patient refusing consent were excluded in the study.


A convenient sampling technique was utilized, where patients who fulfilled the criteria were enrolled till data was saturated (i.e. till the sample size of 600 patients achieved).

Data collection

Written informed consent and Ethical clearance were duly obtained from Institutional Ethics Committee. Detailed demographic data, medical history, and meticulous physical examination were carried out. All participants underwent renal function tests and 24-h urinary estimations for creatinine and protein in the hospital laboratory using standard techniques. Further categorization was done after calculating measured GFR by using 24-h urinary creatinine clearance (UrCrCl) into six subsets [Table 1], as per guidelines of CKD National Kidney Foundation.[2] From laboratory data obtained, GFR was calculated using predictive equations of estimating GFR, namely MDRD-4, MDRD-6, CKD-EPI, and CG formula. SCr was measured by Jaffe modified kinetic Alkaline Picrate method. All physical measurements and laboratory investigations were carried out on the same day.
Table 1: Staging of chronic kidney disease and distribution of study population.

Click here to view

   Statistical Analysis Top

Data analysis was performed using IBM SPSS Statistics software version 21.0 (IBM Corp., Armonk, NY, USA). Quantitative variables were analyzed by using measures of central tendency along with their standard deviation. Independent sample t-test and ANOVA were used as test of significance. Qualitative variables were presented as frequency and proportions. Chi-square test was used as test of significance. A P <0.05 was considered as statistically significant. Data were analyzed for Pearson’s correlation coefficient (r), precision (R2), accuracy, and bias. Precision (R2) of the equations was assessed based on the degree of spread of the series of observations and is reflected by the amount of expected variation in the estimates. The R2 Statistic was derived by simple linear regression analysis. Bias is any systematic nonrandom deviation causing a prediction error and was defined by mean prediction error (ME): ME = ΣPEj/N, Where PEi = predicted value - true value and N = sample size. The accuracy at 50% was represented by the percentage of predicted GFR values within 50% of measured endogenous UrCrCl values.

   Results Top

Six hundred patients were enrolled in the study, of them 21 patients were excluded for improper urine collection. Of a total 579 patients, there were 300 males and 279 females. 69 parti-cipants were healthy subjects. Results were compared gender wise as estimating equations use different constants for male and female populations. Gender-wise distribution of participants is shown in [Table 1].

Anthropometric data of study population

Mean age of the study population was 47 ± 14.4 years (males: 47.12 ± 14.8 years, females: 46.8 ± 14 years). Detailed clinical and laboratory parameters including 24-h UrCrCl are represented in [Table 2]. All subsets were age- and sex-matched except CKD stage 4 in which females outnumbered males. [Figure 1] illustrates GFRs of the study groups measured by endogenous 24-h UrCrCl. Females had higher mean GFRs in all categories except in normal and CKD stage 5. Mean GFRs among various subsets are expressed in [Table 3].
Figure 1: Mean glomerular filtration rates of the study population by urinary creatinine clearance.
GFRs: Glomerular filtration rates, UrCrCl: Urinary creatinine clearance.

Click here to view
Table 2: Characteristics of the study population.

Click here to view
Table 3: Gender and chronic kidney disease stage-wise overall glomerular filtration rate values.

Click here to view

Comparison of four predictive equations with urinary creatinine clearance

Among all the male subjects (normal and CKD combined), all the equations showed significant correlation, however best was seen with CKD-EPI. CKD-EPI was also found to have the best precision. CG was documented to be the least biased. Details are shown in [Table 4].
Table 4: Performance of equations with urinary creatinine clearance.

Click here to view

In females, although all the equations exhibited significant correlation, best results were seen with MDRD-4 and MDRD-6. Both also showed best precision. As MDRD-4 was less biased than MDRD-6, we infer that MDRD-4 is better than MDRD-6 in females. Details are shown in [Table 4].

Normal healthy individuals; in males, MDRD-4 showed the best correlation, however, it did not reach significance. MDRD-4 also demonstrated maximum precision but underestimated GFR. CKD-EPI was the most biased which overestimated GFR. CG and CKD-EPI were found to be the most accurate equations at 50% of our standard, both having 91% accuracy. We infer that MDRD-4 is the best predictive equation in this subset, due to its greatest correlation and precision. In females, CG showed the best correlation and most precision. CKD-EPI was most biased which overestimates GFR and CG were least biased. However, CKD-EPI demonstrated the highest accuracy of 95%. We infer that CG is the best predictive equation in this subset due to its highest correlation, precision, and smallest bias. The relative performance of these equations is shown in [Table 5].
Table 5: Performance of predictive equations with urinary creatinine clearance in normal population and chronic kidney disease stage 1.

Click here to view

CKD stage 1: In males, CG, MDRD-4 and MDRD-6 showed the best correlation. MDRD-4 was most precise. Most biased equation was MDRD-6 which over-estimates GFR and the least biased was CKD-EPI. CG, CKD-EPI, and MDRD-4 were most accurate, each achieving 100% accuracy. We can conclude that MDRD-4 is the best equation in this subset, due to its maximum correlation and precision. In females, MDRD-6 showed the best correlation, however, it did not reach significance. MDRD-6 was found to be the most precise equation. Most biased equation was MDRD-4, which underestimated GFR and least biased was CG. CG, CKD-EPI, and MDRD-4 were most accurate each having 96% accuracy. The detailed evaluation is shown in [Table 5].

CKD stage 2: In CKD stage 2 males, CKD-EPI showed the best correlation. CKD-EPI was found to be the most precise equation. The most biased equation was MDRD-6 which overestimates GFR and least biased was CKD-EPI. CG and MDRD-4 were most accurate each having 98% accuracy. We infer that CKD-EPI is the best equation in this subset, which has maximum correlation, precision and least bias. In females, CKD-EPI showed maximum correlation and also demonstrated the highest precision (R2 = 0.236). Most biased equation was MDRD-6 and the least biased was CKD-EPI. CG, CKD-EPI, MDRD-4 are most accurate, each having 100% accuracy. We consider CKD-EPI to be the equation in this subset due to its greatest correlation, precision and least bias. The comparative performance is shown in [Table 6].
Table 6: Performance of predictive equations with urinary creatinine clearance in chronic kidney disease stage 2 and 3.

Click here to view

CKD stage 3: In males, Pearson correlation (r) was significant for all the equations, however the best correlation was with CG, which was also the most precise. The most biased estimate is MDRD-4, which over-estimates GFR, and least biased was CG. CG demonstrated 99% accuracy. CG was the best equation in this subset, which has the best correlation, best precision, least biased, and most accuracy. In females, the best correlation was with CKD-EPI which was also the most precise. The most biased equation was MDRD-6, which overestimates GFR and least biased estimate is CG. All the equations are equally accurate, each having 96% accuracy. CKD-EPI is the best equation in this subset, having best correlation and best precision. The comprehensive evaluation is shown in [Table 6].

CKD stage 4: In males, all the equations have shown a significant correlation, but the best correlation was with MDRD-6, also the most precise. Most biased equation was CG and least biased was MDRD-4. CKD-EPI, MDRD-4, MDRD-6 were most accurate with each equation having 99% accuracy. We infer that MDRD-6 is best, as it has the most correlation, precision, and accuracy. In females, all the equations showed significant correlation, but the best correlation and precision were with MDRD-4. MDRD-6 was most biased and CG was the least. CG was found to be the most accurate (99%). We pronounce MDRD-4 as the best equation in this subset due to its highest correlation and precision. The relative performance is shown in [Table 7].
Table 7: Performance of predictive equations with urinary creatinine clearance in chronic kidney disease stage-4 and 5.

Click here to view

CKD stage 5: In males, best correlation and precision were with CKD-EPI, although all equations uniformly correlated well. Maximum and minimum bias was noted with CG and MDRD-6, respectively. MDRD-6 exhibited the greatest accuracy at 100%. We infer that, CKD-EPI is the best equation in this group, with best correlation and best precision. In females again, best correlation and precision were with CKD-EPI. Maximum and minimum bias were noted with CG and MDRD-4 respectively. CKD-EPI, MDRD-4, MDRD-6 were the most accurate having 98% accuracy. CKD-EPI is deemed to be the best equation in this subset as it depicts the highest correlation, precision, and accuracy. The comparative assessment is shown in [Table 7].

The study population’s GFR estimated by the predictive equations is compared with measured 24-h endogenous UrCrCl. The correlation of the equations with measured GFR is explained in detail in each subset. The p-values of these correlations are depicted in [Table 8]. Each predictive equation is paired and compared with UrCrCl for Pearson correlation (r).
Table 8: Representation of correlation between predictive equations and urinary creatinine clearance.

Click here to view

   Discussion Top

Human kidney contains about 1 million glomeruli, each of which is approximately 150–200 microns in diameter. Approximately 180 L of tubular fluid is produced daily by the process of ultrafiltration at the rate of 125 mL/min. GFR is the key parameter to make diagnostic or therapeutic decisions as well as a tool to monitor the course of kidney disease. Both cross-sectional and longitudinal studies in normal men demonstrate a decline in GFR of ~10 mL/min/1.73 m2 per decade after the age of 30 years, however, equivalent research in women is unavailable.[3] GFR in young Southeast Asian population is slightly below that in Caucasians, though with similar age-related decline.[3] Most work on GFR estimation in healthy individuals is presented from North-American and European regions, so data on non-white races is insufficient. Related literature on healthy Indians is scant and is limited by diverse techniques in GFR measurement and income-plete ascertainment of protein intake.[4]

As the definition of classifying CKD becomes more dependent on the accurate calculation of GFR, it is imperative that a simple and reliable method to calculate GFR is obtained. In clinical practice GFR measurements are based on 24-h endogenous CrCl or SCr based equations. SCr may not rise above the upper limit of the reference range unless GFR is <60 mL/min/ 1.73 m2. Elevated creatinine has 100% specificity and 60% sensitivity.[5] Although SCr concentration can provide a rough index of GFR, numerous factors (age, gender, race, drugs, diet, body size, and laboratory and analytical methods) can lead to errors in the estimation of GFR from SCr concentration alone. Few cross-sectional studies have postulated serum cystatin C to be more precise test of kidney function than SCr since it was believed to be less influenced by muscle mass or diet, although it has not been shown to be superior to SCr based predictive equations.[6]

Development and validation of GFR estimating equations have been done with appropriate attention to epidemiologic and statistical techniques. Various such equations have been used by research groups worldwide and the most validated among these are CG formula, MDRD-4 and MDRD-6 variables, CKD-EPI and Mayo clinic quadratic equation. In choosing a prediction equation to estimate GFR, one should consider both bias and precision of equation-generated estimates. The computing of these equations is discussed in detail below: CG formula: eCreatinine clearance = (140-age × body weight) × 0.85 (if female)/SCr × 72 MDRD-4 variable: estimated GFR (mL/min/ 1.73 m2) = 186.3 × [SCr (mg/dL)/88.4]-1.154 × (age in years)-0.203 × (0.742 if female) × (1.210 if black).

MDRD-6 variable: estimated GFR (mL/min/ 1.73 m2) = 170 × (creatinine/88.4)-0.999 × (age)-0.176 × (SU/0.257)-0.170 × (serum albumin)+0.318 × (0.762 if female) × (1.180 if black).

Chronic Kidney Disease-Epidemiology Collaboration

eGFR = 141 × min (SCr/k, 1)α × max (SCr/k, 1)-1.209 × 0.993age × 1.1018 (if female) × 1.159 (if black); Where k = 0.7 if female, k = 0.9 if male; α= -0.329 if female, α= -0.411 if male, min = the minimum of SCr /k or 1 and max = the maximum of SCr/k or 1

MDRD Study Group has most recently advocated the formula for calculating GFR which is employed by most laboratories to estimate GFR and this forms the basis of CKD.[7] Adoption of automatic reporting of 4 or 6 variable MDRD-eGFR has been often criticized as it underestimates GFR in healthy people resulting in overestimating the prevalence of CKD and its accuracy may become questionable once GFR falls <60 mL/min.[8]

MDRD equation was initially developed in CKD patients who were predominantly white and had non-diabetic kidney disease, later it was validated in diabetics, kidney transplant recipients, and other races like African Americans.[9] MDRD generally provides unbiased and reasonably accurate estimates across a wide range of subgroups and has been shown to be more accurate than the CG equation. As stated above, GFR estimation may not be accurate in healthy individuals using abbreviated MDRD or CG equations. However, few authors claim MDRD-4 to be next to ideal in predicting GFR in the elderly in daily practice. In order to answer whether predicting GFR can substitute for a reference method in the Indian population, a study among renal donors (healthy individuals) revealed that MDRD-4 and MDRD-6 were the most precise and accurate while CG-GFR was the least biased.[10],[11] However, the authors concluded that poor correlation and level of error exhibited by these equations made them suboptimal for donor evaluation.[11]

CKD-EPI applies the same variables as MDRD but includes a nonlinear term for SCr that substantially reduces bias at higher GFR, enabling numeric eGFR. Most, but not all studies confirm the greater accuracy of the CKD-EPI equation compared to MDRD.[9],[10] In addition because of lesser bias, the use of CKD-EPI equation leads to lower prevalence estimates of decreased GFR in cross-sectional studies and more steep risk relationships of eGFR to adverse outcomes in longitudinal studies.[12] In the Chinese population with CKD, MDRD equation overestimated GFR in patients with CKD stages 4 to 5 and underestimated GFR in CKD stage 1.[13] Results on CKD-EPI or CG equations performing better compared to 4-variable MDRD have also been authenticated by other collaborators.[2] Another research group has deliberated that MDRD underestimates GFR by 0.99 mL/min/1.73 m2 and CG overestimates GFR by 1.99 mL/min/1.73m2.[14] In an additional group of Chinese with CKD, the CKD-EPI equation was adjudged as better than MDRD.[15] In a population-based study, MDRD and CG were believed to overestimate prevalence of CKD in the general population compared to MAYO quadratic equation.[16] Other workers have proven that the MDRD equation offers no advantage and is even less accurate in normal or increased GFR.[17]

Identifying and stratifying patients at risk of renal disease is now an integral part of clinical nephrology. This is a study in the Indian population, who has hitherto little or no representation in sub-populations from which these equations were derived. With the recent K/DOQI guidelines recommending GFR estimation even in patients having normal SCr for screening, appropriate classification, and stratification of patients with renal disease, this study has important clinical implications.[1] We reviewed the world literature on the subject using Medline search with keywords of comparison, predictive equations, GFR, CKD/healthy. We found an exhaustive number of heterogeneous studies on GFR in the normal population, but limited information on CKD patients.

Evaluation of predictive equations in the healthy population

Mahajan et al examined 122 normal Indians and pronounced MDRD-4 as the most accurate at 50% of DTPA scan, which was their gold standard, MDRD-4 and MDRD-6 most precise and CG-GFR as least biased.[11] This study however, did not include CKD-EPI-based estimations. A cross-sectional study by Vervoort et al in 92 patients in the Netherlands observed that MDRD is less accurate than corrected CrCl or CG in normal and diabetic patients.[17] Lin et al examined 117 healthy subjects and documented MDRD equations to be more precise and accurate than CG-CrCl (CG corrected for GFR) and CG-GFR (without correction) when compared to 125I-Iothalamate and 99mTc-DTPA whereas CG-GFR was least biased.[18] Jahan investigated 61 renal donors from Bangladesh and concluded that no predictive equation (CG- CCr, CG-GFR and MDRD) is equivalent to GFR calculated by 99mTc-DTPA scan.[8] Chung et al in 207 Korean healthy donors inferred that CKD-EPI is more accurate and least biased compared to MDRD4, MDRD6 and CG equations and MDRD underestimated GFR significantly when compared to 99mTcDTPA.[19] Kaitwatcharachai et al inferred that both CG- CrCl and MDRD significantly underestimate GFR in healthy Thai adults.[20] In our healthy population we infer that CG and CKD-EPI are more accurate in males, whereas only CKD-EPI is accurate in females. MDRD4 and CG are the most precise in normal male and female adults respectively.

Evaluation of predictive equations in chronic kidney disease patients

In a major study by Bostom et al in Germany, the authors analyzed 109 patients with known kidney disease (having serum Cr ≤1.5 mg/dL) and compared eight prediction equations including CG-CrCl, MDRD-4, and MDRD-6 with iohexol GFR values. The best accuracy was obtained for CG-CrCl and the most precise for the MDRD-4 equation.[21] Although, a landmark study, it focused predominantly on patients with mild renal dysfunction, so the data could not be extrapolated to severe stages of CKD. In contrast, we have evaluated all the parameters including correlation, precision, bias, and accuracy encompassing both genders and every stage of CKD. Kaun et al, in a study conducted in the United Kingdom on 29 end-stage renal diseases (ESRD) study population, compared MDRD-4, MDRD-5, MDRD-6, CG equations with inulin clearance. This study observed that all MDRD related equations were equally accurate among each other but more than CG.[22] Lui et al, conducted a study in 319 elderly Chinese with various stages of CKD, using 99mTc-DTPA as standard compared against CG, CKD-EPI, MDRD-4, MDRD-6, Jelliffe, Hull.[23] They documented that in CKD stages 1 and 2, CKD-EPI provided better results when compared with standard GFR. In CKD stage 3, six-variable MDRD, Hull equation, and CG equation fared better. In CKD stages 4 and 5, Jelliffe-1973 equation revealed higher accuracies.[23] Teo et al described CKD-EPI as more accurate than MDRD, particularly at higher GFRs. They recommend adopting the CKD-EPI equation without ethnic adjustment for estimating GFR in multiethnic Asian patients (Chinese, Malaysian, Indian, and Singapore) with CKD.[24] As can be noted from the above statistics, none of these studies have shown any consistency in the correlation of any one or more predictive equations with the measured GFR, making it difficult for clinical use. In our study, different equations have shown different correlations when compared with our standard. As the estimates failed to reach correlation of 1, (which is ideally impossible) we assessed the performance of these equations using other variables such as precision, bias, and accuracy in order to make a decision which equation is to be used at a particular stage of CKD and also for normal adults. In our patients of CKD stage 1, we observed that CG, CKD-EPI, and MDRD-4 were equivalent in terms of accuracy in both males and females. In CKD stage 2, CG and MDRD-4 demonstrated equal accuracy in males, whereas CG, CKD-EPI, and MDRD-4 proved equally accurate in females. In CKD stage 3, CG established itself to be most accurate in males, whereas all the equations were uniformly accurate in females. In CKD stage 4, CKD-EPI, MDRD-4, and MDRD-6 were equivalently accurate in males, while in females CG was adjudged most accurate. In CKD 5, MDRD-6 showed maximum accuracy, whereas in females CKD-EPI, MDRD-4 and MDRD-6 had higher accuracy.

   Limitations Top

We used 24-h urinary creatinine clearance as the standard method for GFR estimation. This method has few drawbacks, including errors with both under or over-collection and also errors due to tubular secretion of creatinine. We could not include a detailed 24-h dietary recall for the protein intake and we excluded the morbidly obese and very elderly subjects from our study which may limit the application of the present results in them.

   Study strengths Top

Our study has a relatively large sample size, which is unique. It has highlighted the aspect of gender differences in GFR values in health and disease which has not been emphasized in the past. We have compared the predicting equations with endogenous CrCl in all the classes of CKD including all four widely utilized predictive equations simultaneously. Our methodology did not require any exposure to radiation or repetitive sampling, hospitalization, or intravenous infusions.

   Conclusions Top

We conclude that these predictive equations which are derived from SCr and various demographic variables are good in estimating GFR in the CKD population. These predictive equations show poor correlation, less precision and more bias in the normal study population when compared to the CKD population. However, there is no single equation which is the most precise, most accurate and least biased in all the stages of CKD. The choice of a predictive equation in a given individual should be made keeping these observations in mind. We recommend further research on this subject using nuclear studies and community-based populations.

Conflict of interest: None declared.

   References Top

Stevens PE, Levin A; Kidney Disease: Improving Global Outcomes Chronic Kidney Disease Guideline Development Work Group Members. Evaluation and management of chronic kidney disease: Synopsis of the kidney disease: Improving global outcomes 2012 clinical practice guideline. Ann Intern Med 2013;158:825-30.  Back to cited text no. 1
National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: Evaluation, classification, and stratification. Am J Kidney Dis 2002;39:S1-266.  Back to cited text no. 2
Jafar TH, Islam M, Jessani S, et al. Level and determinants of kidney function in a South Asian population in Pakistan. Am J Kidney Dis 2011;58:764-72.  Back to cited text no. 3
Barai S, Gambhir S, Prasad N, et al. Levels of GFR and protein-induced hyperfiltration in kidney donors: A single-center experience in India. Am J Kidney Dis 2008;51:407-14.  Back to cited text no. 4
Levey AS. Measurement of renal function in chronic renal disease. Kidney Int 1990;38:167-84.  Back to cited text no. 5
Stevens LA, Coresh J, Schmid CH, et al. Estimating GFR using serum cystatin C alone and in combination with serum creatinine: A pooled analysis of 3,418 individuals with CKD. Am J Kidney Dis 2008;51:395-406.  Back to cited text no. 6
Mathew TH, Johnson DW, Jones GR; Australasian Creatinine Consensus Working Group. Chronic kidney disease and automatic reporting of estimated glomerular filtration rate: Revised recommendations. Med J Aust 2007; 187:459-63.  Back to cited text no. 7
Jahan F, Chowdhury MN, Mahbub T, et al. Assessing glomerular filtration rate in healthy adult potential kidney donors in Bangladesh: A comparison of various prediction equations with measured glomerular filtration rate by diethylentriamine pentaacetic acid renogram. Bangladesh Med Res Counc Bull 2013;39:74-9.  Back to cited text no. 8
Cirillo M, Lombardi C, Luciano MG, Bilancio G, Anastasio P, De Santo NG. Estimation of GFR: A comparison of new and established equations. Am J Kidney Dis 2010;56:802-4.  Back to cited text no. 9
Kitiyakara C, Yamwong S, Vathesatogkit P, et al. The impact of different GFR estimating equations on the prevalence of CKD and risk groups in a Southeast Asian cohort using the new KDIGO guidelines. BMC Nephrol 2012; 13:1.  Back to cited text no. 10
Mahajan S, Mukhiya GK, Singh R, et al. Assessing glomerular filtration rate in healthy Indian adults: A comparison of various prediction equations. J Nephrol 2005;18:257-61.  Back to cited text no. 11
Levey AS, Stevens LA. Estimating GFR using the CKD Epidemiology Collaboration (CKD-EPI) creatinine equation: More accurate GFR estimates, lower CKD prevalence estimates, and better risk predictions. Am J Kidney Dis 2010; 55:622-7.  Back to cited text no. 12
Zuo L, Ma YC, Zhou YH, Wang M, Xu GB, Wang HY. Application of GFR-estimating equations in Chinese patients with chronic kidney disease. Am J Kidney Dis 2005;45:463-72.  Back to cited text no. 13
Froissart M, Rossert J, Jacquot C, Paillard M, Houillier P. Predictive performance of the modification of diet in renal disease and Cockcroft-Gault equations for estimating renal function. J Am Soc Nephrol 2005;16:763-73.  Back to cited text no. 14
Li JT, Xun C, Cui CL, et al. Relative performance of two equations for estimation of glomerular filtration rate in a Chinese population having chronic kidney disease. Chin Med J (Engl) 2012;125:599-603.  Back to cited text no. 15
Shankar A, Lee KE, Klein BE, et al. Estimating glomerular filtration rate in a population-based study. Vasc Health Risk Manag 2010;6:619-27.  Back to cited text no. 16
Vervoort G, Willems HL, Wetzels JF. Assessment of glomerular filtration rate in healthy subjects and normoalbuminuric diabetic patients: Validity of a new (MDRD) prediction equation. Nephrol Dial Transplant 2002;17:1 909-13.  Back to cited text no. 17
Lin J, Knight EL, Hogan ML, Singh AK. A comparison of prediction equations for estimating glomerular filtration rate in adults without kidney disease. J Am Soc Nephrol 2003;14:2573-80.  Back to cited text no. 18
Chung BH, Yu JH, Cho HJ, et al. Comparison of estimating equations for the prediction of glomerular filtration rate in kidney donors before and after kidney donation. PLoS One 2013;8:e60720.  Back to cited text no. 19
Kaitwatcharachai C. Bedside renal assessment: A comparison of various prediction equations in Thai healthy adults. J Med Assoc Thai 2006;89 Suppl 2:S146-50.  Back to cited text no. 20
Bostom AG, Kronenberg F, Ritz E. Predictive performance of renal function equations for patients with chronic kidney disease and normal serum creatinine levels. J Am Soc Nephrol 2002;13:2140-4.  Back to cited text no. 21
Kuan Y, Hossain M, Surman J, El Nahas AM, Haylor J. GFR prediction using the MDRD and Cockcroft and Gault equations in patients with end-stage renal disease. Nephrol Dial Transplant 2005;20:2394-401.  Back to cited text no. 22
Liu X, Cheng MH, Shi CG, et al. Variability of glomerular filtration rate estimation equations in elderly Chinese patients with chronic kidney disease. Clin Interv Aging 2012;7:409-15.  Back to cited text no. 23
Teo BW, Koh YY, Toh QC, et al. Performance of the CKD-EPI creatinine-cystatin C glome-rular filtration rate estimation equations in a multiethnic Asian population. Singapore Med J 2014;55:656-9.  Back to cited text no. 24

Correspondence Address:
Monica Gupta
Department of General Medicine, Government Medical College and Hospital, Chandigarh
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/1319-2442.335452

Rights and Permissions


  [Figure 1]

  [Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7], [Table 8]


    Similar in PUBMED
    Search Pubmed for
    Search in Google Scholar for
    Email Alert *
    Add to My List *
* Registration required (free)  

   Aims and Objectives
    Materials and Me...
   Statistical Analysis
   Study strengths
    Article Figures
    Article Tables

 Article Access Statistics
    PDF Downloaded207    
    Comments [Add]    

Recommend this journal