IMC Journal
IMC Journal
IMCJMS


Abstract (887)  Download PDF (849) Print
Issue: Vol.8 No.2 - July 2014
Is Estimated Glomerular Filtration Rate (eGFR) a Better Predictor than Creatinine Cutoff to Detect Chronic Kidney Disease (CKD)?
Authors:
Parvin Akter Khanam
Parvin Akter Khanam
Affiliations

Department of Epidemiology and Biostatistics,BIRDEM,122, Kazi Nazrul Islam Avenue, Shahbagh, Dhaka-1000

,
Tanjima Begum
Tanjima Begum
Affiliations

Department of Epidemiology and Biostatistics,BIRDEM,122, Kazi Nazrul Islam Avenue, Shahbagh, Dhaka-1000

,
Md. Morshed Alam Khan
Md. Morshed Alam Khan
Affiliations

Department of Biochemistry,BIRDEM,122, Kazi Nazrul Islam Avenue, Shahbagh, Dhaka-1000

,
Sarwar Iqbal
Sarwar Iqbal
Affiliations

Department of Nephrology,BIRDEM,122, Kazi Nazrul Islam Avenue, Shahbagh, Dhaka-1000

,
Akhter Banu
Akhter Banu
Affiliations

Institution of Nutrition and Food Science,University of Dhaka,Dhaka

,
Mir Masudur Rhaman
Mir Masudur Rhaman
Affiliations

Department of Community Medicine,Ibrahim Medical College,122, Kazi Nazrul Islam Avenue, Shahbagh, Dhaka-1000

,
M Abu Sayeed
M Abu Sayeed
Affiliations

Department of Community Medicine,Ibrahim Medical College,122, Kazi Nazrul Islam Avenue, Shahbagh, Dhaka-1000

Abstract

Chronic kidney disease (CKD) with diabetes mellitus is one of the most common and major public health problems globally. In Bangladesh, several studies indicate an increasing prevalence of diabetes though very few studies are available on CKD. For CKD, diagnostic method, criteria or cutoffs still remained undecided. This study aimed to determine the prevalence of CKD among the hospitalized patients and to compare the diagnostic approach practiced in the hospital.

Methods: All patients admitted to the Department of Nephrology at BIRDEM from May 1 to July 31, 2012 were selected for investigation. An almost equal number of patients were also selected from other units of Medicine. The information included were age, sex, social class, blood pressure, height, weight, blood glucose, creatinine, triglycerides, total cholesterol, high-density lipoproteins and electrolytes. The CKDcreat was diagnosed based on creatinine (>1.2mg/dl) and the CKDgfr based on estimated GFR (<60 ml/min/1.73m2) following Kidney Disease Outcomes Quality Initiative (K/DOQI) guideline. The comparisons of characteristics were made between CKDcreat and non-CKDcreat (£ 1.2 vs.>1.2 mg/dl) groups. Similar comparisons were also made between CKDgfr and non-CKDgfr (>60 vs. £ 60 ml/min/1.732) groups.

Results: A total of 4172 patients got admitted in the study period of 90 days; and 442 patients (m / f = 256 / 186) were investigated. Of the total (n=4172), 241 (5.8%) had CKDcreat and 272 (6.5%) had CKDgfr. Of the investigated 442 patients, 241 (54.5%) had CKDcreat and 272 (61.5%) had CKDgfr. The differences of characteristics between CKDcreat and non-CKDcreat groups were almost similar to the differences between CKDgfr and non-CKDgfr groups. Higher age, higher social class and higher blood pressure showed significant (p<0.001) and similar associations with both CKDcreat and CKDgfr. Interestingly, if the cut-off of eGFR is taken at <90 ml/min/1.732, as suggested by K/DOQI, the prevalence of CKDgfr increases to 86.7%. This indicates a wide variation (32.2%) between the two criteria (CKDcreat: creat >1.2 mg/dl and CKDgfr: <90 ml/min/1.732). Thus, a large proportion remained either under- or over-diagnosed depending on the criterion used.

Conclusion: The prevalence of CKD among the hospitalized patients was found not negligible. The comparisons of two diagnostic criteria did differ and eGFR (K/DOQI) could detect higher proportion of CKD, which might be an over-diagnosis. Further study taking microalbuminuria, gross proteinuria, albumin-creatinine ratio and cystatin C may validate the method for the diagnostic accuracy of CKD, which my help assessing the prevalence of CKD accurately.

Ibrahim Med. Coll. J. 2014; 8(2): 50-55

Address for Correspondence:Dr. Parvin Akter Khanam, Assistant Professor, Department of Epidemiology and Biostatistics, BIRDEM General Hospital, 122 Kazi Nazrul Islam Avenue, Shahbag, Dhaka-1000

 

Introduction

Chronic kidney disease (CKD) is a growing public health problem both in the developing and developed world.1 Prevalence is estimated to be 8-16% worldwide. The complications of CKD related to and resulted in increased all-cause and cardiovascular mortality.2 More striking is the fact that diabetes mellitus is the most common cause of chronic kidney disease, but in some regions other causes, such as herbal and environmental toxins, are more common.2 About 5% of the adult populations have some form of kidney damage and every year millions of people die prematurely of cardiovascular diseases linked to CKD. The recent literatures indicate that diabetes and hypertension are becoming the most common causes of CKD, especially in older people both in developed and developing nations,3,4 CKD is estimated to effect 19 million people of US population and greater than 50 million people worldwide.5,6 In Bangladesh, a survey among the disadvantaged community in Dhaka City revealed that 13.1% had CKD.7 This indicates that the prevalence of CKD is not negligible. Early diagnosis of CKD and intervention are the imperative measures to prevent or retard life-threatening complications. The intervention measures initiating low-protein dietary changes, close monitoring of blood pressure, control of blood glucose levels, health related education, exercise, and so on.9 The aim of this study was to estimate the burden of CKD in hospitalized patients and to compare the two diagnostic criteria practiced in the hospital setting with a view to accept a cheaper and simpler diagnostic method.

 

Subjects and Methods

The data were collected from admitted patients at BIRDEM general hospital for 90 days from May 1 to July 31, 2012. All patients with the diagnosis of CKD admitted to the Department of Nephrology unit were selected for investigation irrespective of the clinical status except those undergoing dialysis of any form. An equal number of patients were also randomly selected from Department of Medicine only. The data related to socio-demography (age, sex, social class), blood pressure, anthropometry (height, weight for BMI), laboratory investigation (blood glucose, creatinine, triglycerides, total cholesterol, high-density lipoproteins, electrolytes) were collected. Usually, two diagnostic criteria are used at BIRDEM for the diagnosis of CKD. The CKDcreat group was diagnosed based on creatinine (>1.2mg/dl) and the CKDgfr group based on estimated GFR (eGFR: <60 ml/min/1.73m2)) following Kidney Disease Outcomes Quality Initiative (K/DOQI) guideline. The characteristics of CKDcreat group were compared with non-CKDcreat and CKDgfr was compared withthe non-CKDgfr group.3 The eGFR for isotope dilution mass spectrometry (IDMS) traceable serum creatinine values were as follows:8 eGFR(mL/min/1.73m2)=175x(SCr)–1.154x(Age)–0.203 (0.742 if female).3 The CKD stages (1 to 5) based on (K/DOQI) were analyzed in various combination with the CKDcreat and NCKDcreat.

Statistical analyses: Socio-demographic characteristics were given in percentages for qualitative and mean (SD) for quantitative variables. Independent t-tests were applied for comparisons of characteristics between CKDcreat and NCKDcreat group and between CKDgfr and NCKDgfr groups to see any difference observed between these two comparisons. The prevalence rates for CKDcreat and CKDgfr were givenin percentages. We also used c2-test to assess risk factors like sex, age, residence, social class, hypertension status, occupation and smoking for both types of CKD. The values for eGFR based on K/DOQI and the corresponding values for creatinine were shown in means with 95% confidence interval. A p< 0.05 was considered statistically significant. All statistical analyses were performed using SPSS 20.0.

 

Results

A total of 4172 patients got admitted to BIRDEM during the study period of 90 days from May 1 to July 31, 2012. Of them, 442 patients were selected for investigation. All patients (n = 250) who had CKD admitted to the Department of Nephrology were included in this study. Additionally, 192 (5%) patients, randomly selected from the rest (4172 – 250 =3922, not known to have CKD) were also included.

Of the study population (n = 442) the males were 256 and females were 186. Based on the two criteria (creat >1.2mg/dl) and eGFR (<60 ml/min/ 1.73m2), the prevalence of CKDcreat and CKDgfr was 5.8% and 6.5%, respectively, among the admitted (n=4172) patients. In other words, at any given period in a hospital, the prevalence of CKD ranges from 5 – 7%. In contrast, when only the investigated (n = 442) patients were considered the prevalence of CKDcreat was 54.5% and CKDgfr, was 61.5%. (table 1). If the cut-off of eGFR is taken at <90 ml/min/1.73m2, as suggested by K/DOQI, the prevalence of CKDgfr increased further to 86.7%. Thus, there was a wide variation (32.2%) between the two criteria (CKDcreat: creat >1.2 mg/dl and CKDgfr: <90 ml/min/1.732).

 

Table-1: Comparison of CKD prevalence based on creatinine (³1.2mg/dl) and K/DOQI* (ml/min/1.73m2)

 

 

The socio-demographic characteristics are shown in Table 2. The mean (SD) age was 56.1 (13.9) and BMI was 23.3 (4.3). Forty years and above comprised   almost 90%.


Table-2: Demographic characteristics of the study population (N=442).



The comparisons of characteristics between CKD and Non-CKD groups are shown in Table 3. The comparisons are shown separately (CKDcreat vs. NCKDcreat and CKDgfr vs. NCKDgfr). Age, BMI, Hemoglobin and SBP differed significantly in either comparison. The results of comparisons between CKDcreat and non-CKDcreat groups did not differ from the results of comparisons between CKDgfr and non-CKDgfr groups.

 

Table-3: Comparison of characteristics between non-CKD (NCKD) and CKD based on serum creatinine and eGFR (cut-off: creat 1.2mg/dl and eGFR 60 ml/min/1.73m2).3

 

 

 Regarding the risk factors higher age, higher social class and higher blood pressure showed significant (p<0.001) associations with both CKDcreat and CKDgfr (Table 4). It was observed that the levels of significance related to risks were almost similar for both types. Additionally, we estimated the means with 95% confidence interval for creatinine level with the corresponding values for eGFR based on K/DOQI (Table 5).

 

Table-4: Comparison of Prevalence of CKD (based on both criteria) according to sex, age, residence, social class, hypertension, occupation and smoking habit.

 

 

 Table-5: Kidney disease outcome quality initiative (K/DOQI) classification based on National kidney Foundation (NKF)3

 

 

Discussion

This study is the first of its kind in a Bangladeshi diabetic population that compared the prevalence of CKDcreat and CKDgfr. The comparisons were made between the characteristics between CKDcreat and NCKDcreat. Similar comparisons were made between CKDgfr and NCKDgfr. Usually, the creatinine level exceeding 1.2mg/dl has long been used, at BIRDEM or elsewhere in Bangladesh, as a diagnostic cutoff for the impaired renal function. After the introduction of eGFR staging (K/DOQI)3 most physicians are inclined to accept this staging. Possibly, this newer diagnostic staging criteria is more useful. But the estimation of eGFR needs body surface area (BSA: 1.73m2), which varies among populations. Thus, there remains a chance of usual variation of eGFR and may result differently in Bangladeshi population, and in particular, diabetic population. Diabetic nephropathy is the leading cause of chronic kidney disease in patients starting renal replacement therapy10 and is associated with increased cardiovascular mortality.11 So, accurate assessment of the prevalence of CKD is important and so is the importance of its correct diagnostic criteria.

Then which criteria should we follow? The prevalence of CKDgfr (eGFR <90 ml/min/1.73 m2) was found 86.7%, based on K/DOQI (table 3). In contrast, the prevalence of CKDcreat was found 54.5%. based on creatinine level (>1.2mg/dl). This means that about 38.5% remained undiagnosed by CKDcreat or over-diagnosed by CKDgfr criteria. The controversy remained still unsettled as reported from Pakistan.12 Had we included other variables like micro-albuminuria, gross proteinuria, albumin-creatinine ratio or evidence of other micro-angiopathic (retinopathy or neuropathy) and macro-angiopathic lesions like coronary artery disease or cardiovascular morbidity then we could have better assessment of diagnosis or grading of CKD. The recent suggestion is that serum cystatin C alone or creatinine plus cystatin C may predict better CKD.13 But, this recommendation was challenged by others.14 Considering the above mentioned studies it remained unsettled issue to recommend an accurate diagnostic tool for CKD.

As regards the risk factors CKD was found significantly associated with older age (table 4) which is consistent with the other studies.15,16 This study also suggests that urbanization, presence of hypertension, are major risk factors for the development of diabetes as well as CKD. As we know low socioeconomic status was associated with a greater risk for CKD, but in our study we observed that the rich socioeconomic group had greater risk for CKD. This may or may not be true because in Bangladesh hospitalized patients mostly comprised of rich class. Further study based on population may yield a reasonable assessment.

This study concludes that the prevalence of CKD among the hospitalized patients is almost comparable to other studies and the prevalence was found much higher if K/DOQI is used. Older age, hypertension, rich class and urbanization were found significantly associated CKD. The study suggests that inclusion of serum creatinine with eGFR, micro-albuminuria, gross proteinuria, albumin-creatinine ratio and cystatin C in a prospective cohort may determine more reliable and acceptable method for the staging of CKD, which in turn may help screening of CKD. We also propose that any population, free from diseases, should have the reference values (mean, median, deviation percentile) of creatinine and body surface area (for eGFR), any value exceeding 95th percentile may be considered abnormal for staging of CKD.

 

Acknowledgements

We are indebted to Ibrahim Medical College for the financial support. We acknowledge the active cooperation extended by the doctors, nurses and all other hospital staff of Nephrology and Medicine Units of BIRDEM. We are also grateful to the participants who volunteered the study.

 

References

1.    The World Health Report 2003. Shaping the future. World Health Organization; 2003.

2.    Jha V, Garcia-Garcia G, Iseki K, Li Z, Naicker S, Plantter B, Saran R, Wang AY, Yang CW. Chronic kidney disease: global dimension and perspective. Lancet 2013; 382: 260-72.

3.    Levey AS, Coresh J, Balk E, et al. National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Inter Med 2003; 139: 137-47.

4.    Thorp ML, Eastman L, Smith DH, Johnson ES. Managing the burden of chronic kidney disease. Dis Manag 2006; 9: 115-24.

5.    Coresh J, byrd-Holt D, Astor BC, Briggs JP, Eggers PW, Lacher DA, Hostetter TH: Chronic kidney disease awareness, prevalence, and trends among U.S adults, 1999 to 2000.  J Am Soc Nephrol 2005; 16: 180-88.

6.    Dirks JH, deZeeuw D, Agrwal SK, et al. Prevention of chronic kidney and vascular disease: toward global health equity: the Bellagio 2004 Declaration. Kidney Int 2005; 98(Suppl): S1-S6.

7.    Huda MN, Alam KS, Rashid HU, Alam MR, Rahman MH and Selim SI. Prevalence of Chronic Kidney Disease in adult Disadvantageous Population. Journal of Chittagong Medical College Teachers’ Association 2010; 21(2): 25-29.

8.    http://www.globalrph.com/multiple_crcl_idms.htm

9.    McClellan WM: Epidemiology and risk factors for chronic kidney disease. The Medical clinics of North America 2005; 89(3): 419-45.

10.  Bethesda, MD. US Renal Data System: USRDS 2003 Annual Data Report: Atlas of End-Stage Renal Disease in the United States. National Institute of Health, National Institute of Diabetes and Digestive and Kidney Diseases 2003.

11.  Valmadrid CT, Klein R, Moss SE, Klein BE: The risk of cardiovascular disease mortality associated with microalbuminuria and gross proteinuria in persons with older-onset diabetes mellitus. Arch Intern Med 2000; 160: 1093–1100.

12.  Jessani S, Levey AS, Bux R, Inker LA, Islam M, Chaturvedi N et.al. Estimation of GFR in South Asians: a study from the general population in Pakistan. Am J Kidney Dis. 2014; 63(1): 49-58. doi: 10.1053

13.  Shlipak MG, Matsushita K, Ärnlöv J, Inker LA, Katz R, Polkinghorne KR, Rothenbacher D, Sarnak MJ, Astor BC, Coresh J, Levey AS, Gansevoort RT; CKD Prognosis Consortium. Cystatin C versus creatinine in determining risk based on kidney function. N Engl J Med. 2013; 369(10): 932-43.

14.  Rule AD1, Bailey KR, Lieske JC, Peyser PA, Turner ST. Estimating the glomerular filtration rate from serum creatinine is better than from cystatin C for evaluating risk factors associated with chronic kidney disease. Kidney Int. 2013; 83(6): 1169-76. doi: 10.1038/ki.2013.7.

15.  Mulder WJ, Hillen HFP. Renal function and renal disease in the elderly. Eur J Int Med 2001; 12: 86-97.

16.  Rowe JW, Andres R, Tobin JD, Norris AH, Shock NW. The effect of aging on creatinine clearance in men: a cross sectional and longitudinal study. J Ger 1996; 31: 155-63.