CLB-08764; No. of pages: 6; 4C: Clinical Biochemistry xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
3Q1
Xuejing Wang a,1, Kiyoshi Ichihara b,⁎, Guobin Xu a,2, Yoshihisa Itoh c a
7
a r t i c l e
8 9 10 11 12
Article history: Received 12 February 2014 Received in revised form 14 May 2014 Accepted 19 May 2014 Available online xxxx
13 14 15 16 17 18
Keywords: Creatinine Ethnicity Glomerular filtration rate Chronic kidney disease Equation
i n f o
R O
Clinical Laboratory, Peking University First Hospital, Beijing, China Faculty of Health Sciences, Yamaguchi University Graduate School of Medicine, Ube, Japan Clinical Laboratory, Asahikawa Medical College, Asahikawa, Japan
a b s t r a c t
E
D
P
Background: Estimated glomerular filtration rate (eGFR) is currently calculated using various equations and serum creatinine (Scr) value measured by different assays. Differences among these eGFRs deserve further study. Methods: Volunteers from eight Asian regions (n = 3283; age 20–65 years, 1454 men, 1829 women) were recruited. The Chronic Kidney Disease Epidemiology Collaboration equation (EPI), Modification of Diet in Renal Disease Study equation (MDRD) for Japanese (MDRDJap) and MDRD for Chinese (MDRDChi) were selected. Jaffe and enzymatic assays were used to measure Scr. Six eGFRs were obtained for each volunteer: EPI equation using Scr value of enzymatic assay (EPI/E) and Jaffe assay (EPI/J); MDRDJap equation using Scr value of the two assays (MDRDJap/E, MDRDJap/J); and MDRDChi equation using Scr value of the two assays (MDRDChi/E, MDRDChi/J). Results: Neither Scr nor eGFR showed significant regional difference. We compared eGFR calculated using the same equation but with different assays. The medians (2.5%, 97.5%) of eGFR difference were 2.0 (−7, 14) mL/min/ 1.73 m2 for EPI, 3.0 (−12.0, 18.0) mL/min/1.73 m2 for MDRDJap, and 5.0 (−18, 30) mL/min/1.73 m2 for MDRDChi. We also compared eGFR calculated using different equations but with the same assay. The medians (2.5%, 97.5%) of eGFR difference were 11 (−6, 56) mL/min/1.73 m2 between MDRDChi/E and EPI/E; 26 (9, 35) mL/min/1.73 m2 between EPI/E and MDRDJap/E; and 39 (22, 65) mL/min/1.73 m2 between MDRDChi/E and MDRDJap/E, respectively. Conclusions: eGFR difference caused by using different equations is much larger than that caused by using different Scr assays. A common equation for GFR estimation is encouraged for use in Asians. © 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
E
C
c
T
b
O
4 5 6
F
2
Call for the use of a common equation for glomerular filtration rate estimation in East and South-East Asia
1
R
39 37 38
Introduction
41 42
In general, a glomerular filtration rate (GFR) that is lower than 60 mL/min/1.73 m2 or a urine albumin to creatinine ratio that is higher than 30 mg/g is considered to be an indicator of chronic kidney disease (CKD) [1]. Early diagnosis and treatment of CKD can prevent or slow progression of the disease, which could reduce the incidences of endstage kidney and cardiovascular diseases [2,3]. When such benefit is accepted by the clinical community, physicians are more likely to send a patient to a specialist if they identify such a sign on laboratory reports. It is suspected that the automated laboratory reporting of estimated GFR (eGFR) is associated with a significant increase in the rate of referral of patients to nephrologists [4]. At present, many equations to estimate GFR have been developed, but they were established on the basis of various populations and various serum creatinine (Scr) measurement assays, and referred to various “gold standards” for GFR measurement.
47 48 49 50 51 52 53 54
N C O
45 46
U
43 44
R
40
⁎ Corresponding author at: Department of Clinical Laboratory Sciences, Faculty of Health Sciences, Yamaguchi University Graduate School of Medicine, Minami-Kogushi 1-1-1, Ube 755-8505, Japan. Fax: +81 836 35 5213. E-mail address: [email protected] (K. Ichihara). 1 Present address: Clinical Laboratory, Civil Aviation General Hospital, Beijing, China. 2 Present address: Clinical Laboratory, Beijing Cancer Hospital, Beijing, China.
19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
A fundamental aspect related to GFR estimation deserves thorough investigation: that is, to what extend do the equation and Scr measurement assay affect values of eGFR. A multicenter study titled “The Asian Project for Collaborative Derivation of Reference Intervals” was conducted in East and South-East Asia from 2009 to 2010, aiming at collaborative derivation of reference intervals. The initial report of the study [5] showed that no regional difference was found in Scr level in East and South-East Asians. The present paper reports our follow-up work on this previous study [5]. The objectives of this study were to investigate differences of eGFR among various equation–assay combinations and to study the feasibility of using a common equation in Asian populations. Our hypothesis of using a common equation among all Asians is motivated by the accepted understanding that the Modification of Diet in Renal Disease Study equation (MDRD) is suitable to be used in all Caucasians. In this paper, because Japanese accounted for 46% and Chinese accounted for 28% of all volunteers, we selected 3 equations that contain ethnic considerations: the first is the MDRD equation for Japanese (MDRDJap) [6], which is recommended for use in Japanese by the Japan Society of Nephrology; the second is the MDRD equation for Chinese (MDRDChi), which had been used to investigate the prevalence of CKD in China [7]; and the third is the two-level Chronic Kidney
http://dx.doi.org/10.1016/j.clinbiochem.2014.05.058 0009-9120/© 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Please cite this article as: Wang X, et al, Call for the use of a common equation for glomerular filtration rate estimation in East and South-East Asia, Clin Biochem (2014), http://dx.doi.org/10.1016/j.clinbiochem.2014.05.058
55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76
2
77 78
X. Wang et al. / Clinical Biochemistry xxx (2014) xxx–xxx
For each volunteer, the following six eGFRs were obtained: EPI using the Scr value of the enzymatic assay (EPI/E) and Jaffe assay (EPI/J); MDRDJap using the Scr value of the enzymatic assay (MDRDJap/E) and Jaffe assay (MDRDJap/J); and MDRDChi using the Scr value of the enzymatic assay (MDRDChi/E) and Jaffe assay (MDRDChi/J).
82
Materials and methods
Statistical analysis
134
83
Participant enrollment, sample collection and measurement
135
84
93
This study further explored the data of an existing database created in a comprehensive survey among many cities in eight Asian regions. Participant recruitment, sample collection, specimen transportation, the assay's performance validation and target analyte measurement of the survey were described in a previous report [5]. The study was approved by the Ethical Committee of Yamaguchi University Graduate School of Medical Sciences in December 2008. All the volunteers were duly informed of the medical treatment and procedures associated with this study, and all gave their written consent to participate in this research.
The same statistical methods as reported in the previous study [5] were used. The magnitudes of variation due to sex, age and region were analyzed by three-level nested ANOVA. In brief, the magnitude of variation due to each factor (sex, age and region) was expressed as standard deviations (SD) including between-region SD (SDregion), between-sex SD (SDsex) and between-age SD (SDage). The relative magnitude of each factor to that of the residual SD representing a net between-individual SD (SDnet-btw-indiv) was computed as the SD ratio (SDR) by the following formula:
94
Performance of serum creatinine assays
95 96
107 108
Scr was measured by UniCell DxC 800 analyzer (Beckman-Coulter, USA) using a kinetic Jaffe assay. The enzymatic assay was performed on a Hitachi 917 analyzer, P module, and the reagent was provided by Shino, Japan. All measurement procedures followed the manufacturer's specifications. The accuracy of the two assays was validated by measuring a certified reference material (JCCRM521-10, serum matrix; see the Supplementary material for its certification). The biases of the Jaffe assay and enzymatic assay were − 2.7 μmol/L and − 2.7 μmol/L, respectively, when the Scr level is 78.7 μmol/L, and −5.3 μmol/L and − 3.5 μmol/L, respectively, when the Scr level is 194.5 μmol/L. The within-run coefficient of variation was 2.41% for the Jaffe assay and 1.41% for the enzymatic assay, and the day-to-day coefficient of variation was 2.97% for the Jaffe assay and 1.58% for the enzymatic assay during the study period.
109
Estimation of glomerular filtration rate
110 111
Three equations, the EPI, MDRDJap and MDRDChi, were used. When EPI is used to calculate eGFR, the following equations are used for males and females. For males:
103 104 105 106
112 113 115 116
−0:411
Scr≤0:9 : eGFR ¼ 141 ðScr=0:9Þ ScrN0:9 : eGFR ¼ 141 ðScr=0:9Þ
U
For females: 120 121
0:993
Age
:
N
118
−1:209
Age
0:993
−0:329
Scr≤0:7 : eGFR ¼ 144 ðScr=0:7Þ ScrN0:7 : eGFR ¼ 144 ðScr=0:7Þ
−1:209
Age
0:993 0:993
Age
:
123 124
When the MDRDJap is used to calculate eGFR, the following equation is used for males and females. −1:094
eGFR ¼ 194 ðScrÞ
−0:287
ðAgeÞ
127 129
When the MDRDChi is used to calculate eGFR, the following equation is used for males and females. −1:234
eGFR ¼ 175 ðScrÞ
−0:179
ðAgeÞ
½if female; 0:79
O
133
136 137 138 139 140 141 142 143
R O
P
D
146 147 148 149 150 151 152 153 154
Results Sex, age and regional difference of Scr and eGFR
155
Sex difference was significant for the Scr value because the SDR for Scr was approximately 1.83. However, the sex difference was relatively subdued for the eGFR value because the SDR for eGFR was approximately 0.32. Age difference was significant for eGFR because the SDRs for eGFR were 1.24 for males and 1.42 for females. In contrast, age difference showed no significant effect for the Scr value because the SDRs for Scr were only 0.08 for males and 0.00 for females. Both Scr and eGFR showed no noticeable regional differences by three-level nested ANOVA. The SDRregion for Scr and eGFR was all below 0.25 and was considered not significant (Table 1).
156 157
Scr distribution (by enzymatic assay) in eight Asian regions
167
The Scr value at each percentile in the eight Asian regions studied was shown in Table 2. The Scr value was obtained from the enzymatic assay. The Harris–Boyd method was applied, and an az value greater than 3.0 was deemed to represent a significant difference. Taking Japan as the comparison region, the az values, which ranged from 0.0 to 2.5, suggested no significant regional differences among the eight regions.
168
SDRsex
Scr eGFR by EPI/E
1.83 0.32
158 159 160 161 162 163 164 165 166
169 170 171 172 173 174 t1:1
Table 1 SDratio of Scr and eGFR (EPI/E) by nested ANOVA.
½if female; 0:739
126
131 132
145
The SDR of each factor indicates its degree of influence on the target analyte. An SDR of ≥0.3 was regarded as high, an SDR between 0.26 and 0.29 was considered moderate, and an SDR of ≤ 0.25 was considered low. The Harris and Boyd algorithm [11] was used to determine whether regional groups could be merged. Distribution of eGFR was analyzed based on nonparametric descriptive statistics, by use of a generalpurpose statistical software, Statflex Version 6.0 (Artech Co., Osaka, Japan).
E
101 102
130
SDRfactor ¼ SDfactor =SDnet‐btw‐indiv :
T
99 100
C
97 98
E
91 92
R
89 90
R
87 88
O
85 86
C
79 80
F
81
Disease Epidemiology Collaboration equation (EPI) [8], which has been proved not to require additional ethnic factors and which can be used directly in Asians [9,10]. These three equations all use Scr in their calculation. Two common types of assays were used to measure Scr: the Jaffe assay and enzymatic assay.
Q2t1:2
SDRage
t1:3
SDRregion
Male
Female
Male
Female
t1:4
0.08 1.24
0.00 1.42
0.24 0.00
0.22 0.00
t1:5 t1:6
Standard deviation ratio (SDR) of each factor indicates the degree of influence on the target analyte. SDR ≥0.3 was regarded as high, SDR between 0.26 and 0.29 was considered moderate, and SDR ≤0.25 was considered low [5].
Please cite this article as: Wang X, et al, Call for the use of a common equation for glomerular filtration rate estimation in East and South-East Asia, Clin Biochem (2014), http://dx.doi.org/10.1016/j.clinbiochem.2014.05.058
t1:7 t1:8 t1:9
X. Wang et al. / Clinical Biochemistry xxx (2014) xxx–xxx Table 2 Distribution of Scr (enzymatic assay) of healthy populations in eight Asian regions. Scr μmol/L
N
t2:5 t2:6 t2:7 t2:8 t2:9 t2:10 t2:11 t2:12 t2:13 t2:14 t2:15 t2:16 t2:17 t2:18 t2:19 t2:20 t2:21 t2:22
Male
Female
Japan Seoul Beijing Taiwan Hong Kong HCMC KL Jakarta All Japan Seoul Beijing Taiwan Hong Kong HCMC KL Jakarta All
877 55 51 134 103 106 50 54 1430 1092 70 78 172 121 164 79 78 1854
2.50%
25%
50%
75%
97.50%
57.5 54.4 60.8 60.9 61.0 59.5 64.3 57.1 57.5 39.8 40.0 42.8 40.7 41.6 39.1 40.6 30.9 39.8
67.2 67.4 68.5 69.8 71.8 69.8 73.4 66.3 68.1 49.5 46.9 49.5 46.9 50.4 46.9 47.7 42.4 47.7
72.5 74.3 73.4 77.8 78.7 75.1 78.2 74.7 74.3 53.9 52.2 53.9 51.3 54.8 51.3 53.0 48.6 53.0
79.6 78.5 78.7 84.9 84.9 83.1 86.6 81.3 80.4 58.3 55.7 59.2 55.7 60.1 55.7 58.3 55.7 58.3
93.7 92.3 104.2 102.9 97.2 94.5 101.9 97.5 97.2 69.8 67.4 69.4 65.9 65.8 64.5 70.9 67.7 69.0
F
Region
O
Sex t2:3 t2:4
R O
t2:1 t2:2 Q3
3
Harris–Boyd method all compared with Japan
az az az az az az az
= = = = = = =
0.0 0.7 2.0 2.5 1.2 2.3 0.4
az az az az az az az
= = = = = = =
1.3 0.3 1.9 0.6 1.9 0.3 2.2
t2:23 t2:24
The Harris–Boyd method was used to test for the need of partitioning reference intervals by region. The z score representing differences in two means is adjusted to az by multiplying a coefficient (k), which depends on the sample sizes of the two groups. For partition of reference interval, az ≥3.0 is usually considered as significant [11].
175
eGFR comparison in eight Asian regions
176 177
180 181
Low levels of biases in both the Jaffe assay (2.7 μmol/L) and enzymatic assay (2.7 μmol/L) were identified, but they were both within the permitted range (b4.4 μmol/L) [12]. Therefore, the measured values of Scr were used without further correction to calculate eGFR. Sex and age differences relating to eGFR were investigated by using the Scr measured by enzymatic assay. Within the same age group, the
t3:1 t3:2
Table 3 Age-dependent distribution of eGFR by EPI/E. Age
Region
n
t3:37 t3:38
40–49
50–65
E
P
2.50%
25%
50%
75%
97.50%
103.3 91.1 104.3 98.5 109.4 109.6 105.3 92.8 95.7 80.8 88.9 95.3 92.7 91.3 91.2 103.0 87.8 84.9 86.5 92.3 82.0 85.8 89.1 87.5 79.2 66.9 69.8 78.0 74.0 73.9 78.2 66.2
121.6 117.4 118.2 119.4 122.8 118.8 121.1 119.0 113.1 111.9 112.4 113.1 112.4 109.6 113.7 114.5 104.7 105.2 104.5 105.3 104.4 100.5 106.2 105.4 95.8 96.5 92.6 97.7 94.5 92.8 96.4 96.8
125.3 123.1 127.6 125.3 126.6 122.6 125.4 124.0 117.0 119.3 116.8 117.6 118.0 114.7 118.4 117.2 108.6 108.7 109.4 109.8 113.4 106.9 109.7 109.2 100.9 101.5 101.4 103.1 103.1 99.9 100.4 100.0
129.8 128.6 130.3 131.2 131.5 126.7 129.7 127.9 121.4 124.3 121.6 121.4 124.2 120.4 125.3 119.6 113.1 114.9 114.9 113.8 119.2 110.7 114.0 111.3 104.2 104.2 107.0 105.3 109.4 104.4 107.4 106.6
138.1 138.7 136.6 142.3 152.8 133.8 140.4 135.5 131.1 132.4 130.0 132.5 132.6 128.1 130.4 125.3 121.9 123.7 120.6 121.6 128.0 116.2 123.5 114.1 114.7 114.0 115.4 113.8 123.6 111.9 111.6 112.8
R R
568 26 31 76 69 106 61 35 538 41 40 74 56 81 29 31 444 35 31 94 52 51 27 31 419 23 27 62 47 32 12 35
N C O
30–39
Japan Seoul Beijing Taiwan Hong Kong HCMC KL Jakarta Japan Seoul Beijing Taiwan Hong Kong HCMC KL Jakarta Japan Seoul Beijing Taiwan Hong Kong HCMC KL Jakarta Japan Seoul Beijing Taiwan Hong Kong HCMC KL Jakarta
U
20–29
D
eGFR by EPI/E
t3:3 t3:4 t3:5 t3:6 t3:7 t3:8 t3:9 t3:10 t3:11 t3:12 t3:13 t3:14 t3:15 t3:16 t3:17 t3:18 t3:19 t3:20 t3:21 t3:22 t3:23 t3:24 t3:25 t3:26 t3:27 t3:28 t3:29 t3:30 t3:31 t3:32 t3:33 t3:34 t3:35 t3:36
E
T C
178 179
sex differences relating to eGFR calculated by the three equations were all statistically significant. However, the degree of sex difference, such as 4.1 ± 0.3 mL/min/1.73 m2 for EPI, and 3.3 ± 0.5 mL/min/1.73 m2 for MDRDJap, was not clinically significant, except for MDRDChi, which showed a relatively larger sex difference (15 ± 0.8 mL/min/1.73 m2). Because age difference relating to eGFR was significant, the participants were divided into 4 age groups. Table 3, as represented by EPI/E, showed that within the same age group, the distributions of eGFR were very
Harris–Boyd method all compared with Japan
az az az az az az az
= = = = = = =
0.2 1.1 0.7 1.5 0.2 1.3 1.2
az az az az az az az
= = = = = = =
0.4 0.1 0.4 1.3 0.2 0.5 0.3
az az az az az az az
= = = = = = =
0.9 0.4 0.4 1.7 0.9 0.2 0.9
az az az az az az az
= = = = = = =
0.4 0.5 0.2 0.8 0.1 0.1 0.6
The Harris–Boyd method was used to test for the need of partitioning reference intervals by region. The z score representing differences in two means is adjusted to az by multiplying a coefficient (k), which depends on the sample sizes of the two groups. For partition of reference intervals, az ≥3.0 is usually considered as significant [11].
Please cite this article as: Wang X, et al, Call for the use of a common equation for glomerular filtration rate estimation in East and South-East Asia, Clin Biochem (2014), http://dx.doi.org/10.1016/j.clinbiochem.2014.05.058
182 183 184 185 186 187 188 189
208
Differences in eGFRs
209 210
The differences in eGFRs caused both by using the same equation but with different assays, and by using the same assay but with different equations were listed in Table 4. In the first three records of Table 4, which show eGFR differences caused by using different assays, it can be seen that the equation which produces the least difference is the EPI equation. Between EPI/E and EPI/J, it can be seen that a median of the difference is about 2 mL/min/1.73 m2, whereas the range of the difference is from −6 to 14 mL/min/1.73 m2 (2.5% to 97.5%). A moderate difference was found with the MDRDJap equation. Between MDRDJap/E and MDRDJap/J, the difference has a median of 3 mL/min/1.73 m2 and a range from − 12 to 18 mL/min/1.73 m2 (2.5% to 97.5%). The highest difference was between MDRDChi/E and MDRDChi/J, which has a median of 5 mL/min/1.73 m2 and a range from −21 to 30 mL/min/1.73 m2 (2.5% to 97.5%). In the last three records of Table 4, it can be seen that eGFR difference caused by using different equations was much larger than that caused by using different assays. For example, when enzymatic assays were used for estimating eGFR, the medians (2.5%, 97.5%) of the eGFR crossequation difference were 11 (− 6, 56) mL/min/1.73 m2 between
215 216 217 218 219 220 221 222 223 224 225 226 227
C
213 214
E
211 212
R
205 206
R
203 204
0 −1 −1
2 3 5
4 7 12
14 18 30
Caused by equations MDRDChi/E − EPI/E EPI/E − MDRDJap/E MDRDChi/E − MDRDJap/E
−6 9 22
3 22 31
11 26 39
24 30 48
56 35 65
t4:3 Q4 t4:4 t4:5 t4:6 t4:7 t4:8 t4:9 t4:10 t4:11 t4:12
MDRDChi/E and EPI/E, 26 (9, 35) mL/min/1.73 m2 between EPI/E and MDRDJap/E, and 39 (22, 65) mL/min/1.73 m2 between MDRDChi and MDRDJap. eGFR difference caused by using different equations was much larger than that caused by using different assays. The difference in eGFR was further evaluated in two eGFR categories, eGFR b 90 mL/min/1.73 m2 and eGFR N 90 mL/min/1.73 m2, where eGFR was calculated using EPI/E (Fig. 2). As a whole, eGFR difference in EPI/E b 90 mL/min/1.73 m 2 was smaller than that in EPI/E N 90 mL/min/1.73 m2. When EPI/E was lower than 90 mL/min/1.73 m2, the eGFR difference caused by assays was mostly within ±10 mL/min/1.73 m2. However, eGFR difference caused by equations has substantial variations. The eGFR difference between MDRDChi/E and EPI/E was found to be the smallest, with a range from − 5.0 to 7.6 mL/min/1.73 m2 (2.5% to 97.5%), whereas the eGFR difference between EPI/E and MDRDJap/E ranged from 17.2 to 27 mL/min/1.73 m2, and that between MDRDChi/E and MDRDJap/E ranged from 17.0 to 33.6 mL/min/1.73 m2. When EPI/E was higher than 90 mL/min/1.73 m2, the eGFR difference caused by assays became slightly larger, but mostly within ± 20 mL/min/1.73 m2. The smallest difference existed between EPI/E and EPI/J. However, the eGFR difference caused by equations was much larger, ranging from − 6 to 56 mL/min/1.73 m2 between MDRDChi/E and EPI/E, 9 to 35 mL/min/1.73 m2 between EPI/E and MDRD Jap /E, and 23 to 66 mL/min/1.73 m 2 between MDRD Chi /E and MDRDJap/E.
228 229
Discussion
253
230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252
One of our findings is that, similar to Scr, Scr-based eGFR does not re- 254 sult in a significant difference among populations in the eight Asian 255
O
201 202
−6 −12 −21
C
199 200
97.5%
N
198
75%
U
196 197
50%
F
207
Because no regional difference was found, eGFRs from subjects of all eight regions were collectively analyzed to reveal any distribution patterns. The distribution patterns of six eGFRs were shown in Fig. 1. Overall, the distribution patterns of eGFR by EPI (including EPI/E and EPI/J) and by MDRDJap (including MDRDJap/E and MDRDJap/J) were similar and showed a relatively smaller between-individual biological variation than that of MDRDChi (including MDRDChi/E and MDRDChi/J). Taking the entire distribution pattern of eGFR values by EPI as a whole, it represents a right-shifted distribution, by approximately 20 mL/min/1.73 m2, of the MDRDJap, indicating a systematic difference. The distribution pattern of the segment between the 2.5 percentile and the median by EPI was similar to that of MDRDChi, whereas the distribution pattern of the segment between the median and the 97.5 percentile by EPI was shorter than that of MDRDChi equation.
25%
O
194 195
2.5%
Caused by assays EPI/E − EPI/J MDRDJap/E − MDRDJap/J MDRDChi/E − MDRDChi/J
R O
eGFR distribution patterns by various equation–assay combinations
eGFR difference (mL/min/1.73 m2)
T
193
t4:1 t4:2
Table 4 Distribution of eGFR differences caused by assays and equations.
P
192
similar to each other among the eight regions. When using other equation–assay combinations to calculate GFR, similar results were obtained, and no regional difference of eGFR was found (data not shown).
D
190 191
X. Wang et al. / Clinical Biochemistry xxx (2014) xxx–xxx
E
4
Fig. 1. Distribution patterns of six eGFRs. The unit of the horizontal axis is mL/min/1.73 m2.
Please cite this article as: Wang X, et al, Call for the use of a common equation for glomerular filtration rate estimation in East and South-East Asia, Clin Biochem (2014), http://dx.doi.org/10.1016/j.clinbiochem.2014.05.058
5
R O
O
F
X. Wang et al. / Clinical Biochemistry xxx (2014) xxx–xxx
EPI/E < 90ml/min/1.73m2
EPI/E > 90ml/min/1.73m2
266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296
297
Conclusions
320
This study revealed that creatinine levels were similar independent of the assay, and the eGFRs determined by country-specific MDRD and EPI were also similar, in healthy populations from eight Asian regions. However, the difference in eGFR caused by using different eGFR equations was much greater than that caused by using different Scr assays. Nowadays, people frequently move across regions or even countries, and the world is in a process of rapid globalization. To survey the eGFR value in a continuous and comparable way, we call for a common version of the eGFR equation to be used in East and South-East Asia, or globally, for the automated reporting of eGFR by clinical laboratories.
321 322
Conflicts of interest
332
D
The EPI exploration was based on multiple ethnic populations, including Asians. A recent study [9] showed that the EPI did not require an ethnic factor when used with CKD patients in Asia. An Australia guideline [10] stated that EPI has been validated as a tool to estimate GFR in some non-European populations, including South-East Asian, African, Indian and Chinese individuals living in Western countries. Kong et al. [17] compared the MDRDChi with EPI and demonstrated that the two equations performed equally in CKD staging in the general population. In our study, when EPI/E was b 90 mL/min/1.73 m2, the difference between EPI/E and MDRD Chi was mostly below 10 mL/min/1.73 m 2 . This small difference between the two equations is consistent with the result of Kong et al. Although the difference between EPI/E and MDRDChi was much larger when EPI/E was N90 mL/min/1.73 m2, this will not influence CKD staging. This result indicated that, at a certain level of GFR, EPI/E and MDRDChi have equal performance and that the EPI might be used directly in Chinese without an ethnic factor. Our research has certain limitations as a practical matter of selecting and treating subjects. First, we did not measure GFR, so we could not evaluate the accuracy of the equations. Second, the enrolled subjects included an elderly (age N60 years) population of only 1.9%. It would be desirable to enroll more elderly volunteers with different levels of GFR to better compare the performance of the equations.
E
T
C
265
E
263 264
R
261 262
R
259 260
regions studied, no matter which equation–assay combination is used. The distribution corresponding to EPI/E in this study (Supplementary Table 1), in which half of the volunteers were healthy Japanese, is similar to the result in another study, in which the subjects were healthy Chinese [13]. The similarity of eGFR distribution between these two studies strongly suggested that a common equation could be used to estimate eGFR across different Asian regions, at least in Chinese and Japanese. Another finding is that different equations cause much larger differences in eGFR than those by different assays. When using the same equation with different Scr assays, the eGFR difference was derived from the difference in Scr values. Guided by the recommendations to improve Scr measurement [12], almost all IVD manufacturers have implemented procedures for standardization of Scr assays [14]. Thus, the difference in Scr among assays is mainly caused by the interference materials in blood. Although different assays respond to non-creatinine substances in different ways [15], the contribution of non-specificity to a Scr value is supposed to be in a limit range. For example, in our study, about 95% of the assay difference was below 10 μmol/L and about 70% of the difference was below 5 μmol/L, with the result that most of the eGFR difference caused by assays was within the range of −15 to 15 mL/min/1.73 m2. However, the same Scr value results in obviously different eGFRs when different equations are used, especially in subjects with a normal to higher GFR as shown in Fig. 2. One possible reason is that the mean GFR for equation development was different in each study. For example, the mean ± SD of GFR in the EPI development was 68 ± 40 mL/min/1.73 m2 , it was higher than that in MDRDChi development, which was 55.1 ± 35.1 mL/min/1.73 m 2, and it was also higher than that in MDRDJap development, which was 59.1 ± 35.4 mL/min/1.73 m2. This might explain the large difference in normal to higher GFR values (such as that of N90 mL/min/1.73 m2) among the equations. Another possible reason that may explain the large effect created by the equations is that the “gold standard” GFR was obtained through different protocols. For example, MDRDJap used inulin clearance to obtain the standard GFR; MDRDChi used 99mTcDTPA plasma clearance; and the EPI equation used iothalamate clearance. There exists certain difference among the various protocols. For example, 99mTc-DTPA usually result in a higher “gold standard” GFRs than that by inulin [16], which might explain why higher eGFR values are calculated with the MDRDChi than with the MDRDJap.
N C O
257 258
U
256
P
Fig. 2. Distribution of eGFR difference categorized by EPI/E. When EPI/E was lower than 90 mL/min/1.73 m2, the difference between MDRDChi and EPI was the smallest, but when EPI/E was higher than 90 mL/min/1.73 m2, the difference between the two equations was much larger.
None.
Please cite this article as: Wang X, et al, Call for the use of a common equation for glomerular filtration rate estimation in East and South-East Asia, Clin Biochem (2014), http://dx.doi.org/10.1016/j.clinbiochem.2014.05.058
298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319
323 324 325 326 327 328 329 330 331
333
X. Wang et al. / Clinical Biochemistry xxx (2014) xxx–xxx
352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
C
407
F
Appendix A. Supplementary data
O
350 351
E
348 349
R
346 347
R
344 345
O
342 343
C
340 341
N
339
U
337 338
395 396 397 398 399 400 401 402 403 404 405 406
Supplementary data to this article can be found online at http://dx. 408 doi.org/10.1016/j.clinbiochem.2014.05.058. 409 References
R O
This research was conducted collaboratively among 1) the committee of plasma protein and committee of reference intervals and decision limits in the International Federation of Clinical Chemistry and Laboratory Medician (IFCC), 2) the Scientific Committee in the Asian and Pacific Federation of Clinical Biochemistry, 3) the Working Group on the Guideline for Common Reference Interval in the Japan Society of Laboratory Medicine, and 4) the Committee on Plasma Protein of the Japan Society of Clinical Chemistry. This study was supported by the Committee on Reference Interval and Decision Limit of the IFCC. Research funds include those from a Scientific Research Fund (No. 21406015: 2009–2011) provided by the Ministry of Education, Culture, Sports, Science and Technology of Japan, a Research Promotion Project Fund of the JSLM (2008–2009), and a Scientific Research Fund of the APFCB, and a research fund (No. Z121107005112006, 2012–2015) provided by Beijing Municipal Science and Technology Commission. Most of the reagents and labor required for testing the large number of analytes were generously offered by Beckman-Coulter, Inc. Additional reagents were offered by Nittobo Co., Eiken Chemical Co. and Otsuka Pharmaceutical Co. The sampling equipment (vacuum sample tubes, needles and the holders) was supplied by Becton-Dickinson (BD). Other supplies, such as serum container tubes and boxes and ID labels were purchased by use of the Scientific Research Funds offered by the APFCB. We are grateful to all of the laboratories and staff that have made contributions to the study. The collaborating laboratories outside Japan included 1) Gangnam Severance Hospital, Yonsei University, Seoul; 2) Kangbuk Samsung Hospital, Sung Kyun Kwan University, Seoul; 3) Myongji Hospital, Kwandong University College of Medicine, Seoul; 4) Prince of Wales Hospital, Chinese University of Hong Kong, Hong Kong; 5) Macau Institute for Applied Research in Medicine and Health, Macau University of Science and Technology, Macau; 6) Prodia Clinical Laboratory, Jakarta; 7) Mackay Memorial Hospital, Taipei; 8) Cathay General Hospital, Taipei; 9) Yuan Ching Clin Laboratory, Taipei; 10) National Cheng Kung University, Tainan; 11) Chi-Mei Med Center, Tainan; 12) National Heart Institute, Kuala Lumpur; 13) Gleneagles Intan Medical Centre, Kuala Lumpur; and 14) Medic-Lab, Ho Chi Minh City. The collaborating laboratories in Japan included 15) Yamaguchi University, Ube; 16) Yoshida Hospital, Asahikawa; 17) Hokkaido University, Sapporo; 18) Kishimoto Clinical Laboratory, Tomakomai; 19) Iwate Medical School, Morioka; 20) Hirosaki University, Hirosaki; 21) Hachinohe Red-Cross Hospital, Hachinohe; 22) Hachinohe Central Hospital, Hachinohe; 23) Chiba Cardiovascular Center, Ichihara; 24) Chiba University, Chiba; 25) Tokyo University, Tokyo; 26) Toho University, Tokyo; 27) Shinshu University, Matsumoto; 28) Matsumoto JAM Medical Center, Matsumoto; 29) Ohmachi Municipal Hospital, Omachi; 30) Yamanashi University, Kofu; 31) Fukui University, Fukui; 32) Kanazawa Medical School, Kanazawa; 33) SRL-Kanazawa, Kanazawa; 34) Mie University, Tsu; 35) Nagoya University, Nagoya; 36) Anjo Kosei Hospital, Anjo; 37) Fujita Health University, Toyoake; 38) Osaka University, Suita; 39) Tenri Hospital, Tenri; 40) Osaka Municipal University, Osaka; 41) National Cardiovascular Center, Suita; 42) Kawasaki Medical School, Kurashiki; 43) Kurashiki Central Hospital, Kurashiki; 44) Okayama University, Okayama; 45) Okayama Medical Laboratory, Kurashiki; 46) Matsuda Hospital, Kurashiki; 47) Hiroshima University, Hiroshima; 48) Yamaguchi University, Ube; 49) Tokuyama Central Hospital, Shunan; 50) Yamaguchi Prefectural Medical Center, Hofu;
393 394
[1] Kidney Disease: Improving Global Outcomes (KDIGO) Work Group. KDIGO clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int Suppl 2013;3:1–150. [2] Lewis EJ, Hunsicker LG, Clarke WR, Berl T, Pohl MA, Lewis JB, et al. Renoprotective effect of the angiotensin-receptor antagonist irbesartan in patients with nephropathy due to type 2 diabetes. N Engl J Med 2001;345:851–60. [3] Wright Jr JT, Bakris G, Greene T, Agodoa LY, Appel LJ, Charleston J, et al, African American Study of Kidney Disease and Hypertension Study Group. Effect of blood pressure lowering and antihypertensive drug class on progression of hypertensive kidney disease: results from the AASK trial. JAMA 2002;288:2421–31. [4] Hemmelgarn BR, Zhang J, Manns BJ, James MT, Quinn RR, Ravani P, et al. Nephrology visits and health care resource use before and after reporting estimated glomerular filtration rate. JAMA 2010;303:1151–8. [5] Ichihara K, Ceriotti F, Tam TH, Sueyoshi S, Poon PM, Thong ML, et al. The Asian project for collaborative derivation of reference intervals: (1) strategy and major results of standardized analytes. Clin Chem Lab Med 2013;51:1429–42. [6] Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, et al. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis 2009;53(6):982–92. [7] Zhang LX, Wang F, Wang L, Wang W, Liu B, Liu J, et al. Prevalence of chronic kidney disease in China: a cross-sectional survey. Lancet 2012;379:815–22. [8] Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro 3rd AF, Feldman HI, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009;150:604–12. [9] Teo BW, Xu H, Wang DH, Li J, Sinha AK, Shuter B, et al. Estimating glomerular filtration rates by use of both cystatin c and standardized serum creatinine avoids ethnicity coefficients in Asian patients with chronic kidney disease. Clin Chem 2012;58:450–7. [10] Johnson DW, Jones GR, Mathew TH, Ludlow MJ, Doogue MP, Jose MD, et al. Chronic kidney disease and automatic reporting of estimated glomerular filtration rate: new developments and revised recommendations. Med J Aust 2012;197:222–3. [11] Harris EK, Boyd JC. Statistical basis of reference values in laboratory medicine. New York: Marcel Dekker; 1995. [12] Myers GL, Miller WG, Coresh J, Fleming J, Greenberg N, Greene T, et al. Recommendations for improving serum creatinine measurement: a report from the Laboratory Working Group of the National Kidney Disease Education Program. Clin Chem 2006;52:5–18. [13] Wang X, Xu G, Li H, Liu Y, Wang F. Reference intervals for serum creatinine with enzymatic assay and evaluation of four equations to estimate glomerular filtration rate in a healthy Chinese adult population. Clin Chim Acta 2011;412:1793–7. [14] http://nkdep.nih.gov/lab-evaluation/gfr/creatinine-standardization/recommendations.shtml#ivd; July 18, 2013 . Accessed on. [15] Greenberg N, Roberts WL, Bachmann LM, Wright EC, Dalton RN, Zakowski JJ, et al. Specificity characteristics of 7 commercial creatinine measurement procedures by enzymatic and Jaffe method principles. Clin Chem 2012;58(2):391–401. [16] Dai SS, Yasuda Y, Zhang CL, Horio M, Zuo L, Wang HY. Evaluation of GFR measurement method as an explanation for differences among GFR estimation equations. Am J Kidney Dis 2011;58(3):496–8. [17] Kong X, Ma Y, Chen J, Luo Q, Yu X, Li Y, et al. Evaluation of the Chronic Kidney Disease Epidemiology Collaboration equation for estimating glomerular filtration rate in the Chinese population. Nephrol Dial Transplant 2013;28(3):641–51.
P
335 336
51) Saiseikai Yamaguchi Hospital, Yamaguchi; 52) Saint Hill Hospital, Ube; 53) Kochi University, Nangoku; 54) Kochi Red-Cross Hospital, Kochi; 55) Kumamoto University, Kumamoto; 56) Ryukyu University, Naha; and 57) Adventist Medical Center, Naha. This study is also indebted to Professor Christopher W. K. Lam (Macau Institute for Applied Research in Medicine and Health, Macau University of Science and Technology, Macau), Emeritus Professor Takeshi Kanno (Hamamatsu University School of Medicine), Emeritus Professor Tadashi Kawai (Jichi University School of Medicine), Dr. Katsuhiko Kuwa (National Metrology Institute of Japan), Prof. Susumu Osawa (Kyushu University School of Medicine), and Prof. Shigemi Hosogaya (Kagawa Prefectural College of Health Sciences), who gave invaluable advice and encouragement to promote and carry out this study.
D
Acknowledgments
T
334
E
6
Please cite this article as: Wang X, et al, Call for the use of a common equation for glomerular filtration rate estimation in East and South-East Asia, Clin Biochem (2014), http://dx.doi.org/10.1016/j.clinbiochem.2014.05.058
410 411 Q6 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
3Q1
Xuejing Wang a,1, Kiyoshi Ichihara b,⁎, Guobin Xu a,2, Yoshihisa Itoh c a
7
a r t i c l e
8 9 10 11 12
Article history: Received 12 February 2014 Received in revised form 14 May 2014 Accepted 19 May 2014 Available online xxxx
13 14 15 16 17 18
Keywords: Creatinine Ethnicity Glomerular filtration rate Chronic kidney disease Equation
i n f o
R O
Clinical Laboratory, Peking University First Hospital, Beijing, China Faculty of Health Sciences, Yamaguchi University Graduate School of Medicine, Ube, Japan Clinical Laboratory, Asahikawa Medical College, Asahikawa, Japan
a b s t r a c t
E
D
P
Background: Estimated glomerular filtration rate (eGFR) is currently calculated using various equations and serum creatinine (Scr) value measured by different assays. Differences among these eGFRs deserve further study. Methods: Volunteers from eight Asian regions (n = 3283; age 20–65 years, 1454 men, 1829 women) were recruited. The Chronic Kidney Disease Epidemiology Collaboration equation (EPI), Modification of Diet in Renal Disease Study equation (MDRD) for Japanese (MDRDJap) and MDRD for Chinese (MDRDChi) were selected. Jaffe and enzymatic assays were used to measure Scr. Six eGFRs were obtained for each volunteer: EPI equation using Scr value of enzymatic assay (EPI/E) and Jaffe assay (EPI/J); MDRDJap equation using Scr value of the two assays (MDRDJap/E, MDRDJap/J); and MDRDChi equation using Scr value of the two assays (MDRDChi/E, MDRDChi/J). Results: Neither Scr nor eGFR showed significant regional difference. We compared eGFR calculated using the same equation but with different assays. The medians (2.5%, 97.5%) of eGFR difference were 2.0 (−7, 14) mL/min/ 1.73 m2 for EPI, 3.0 (−12.0, 18.0) mL/min/1.73 m2 for MDRDJap, and 5.0 (−18, 30) mL/min/1.73 m2 for MDRDChi. We also compared eGFR calculated using different equations but with the same assay. The medians (2.5%, 97.5%) of eGFR difference were 11 (−6, 56) mL/min/1.73 m2 between MDRDChi/E and EPI/E; 26 (9, 35) mL/min/1.73 m2 between EPI/E and MDRDJap/E; and 39 (22, 65) mL/min/1.73 m2 between MDRDChi/E and MDRDJap/E, respectively. Conclusions: eGFR difference caused by using different equations is much larger than that caused by using different Scr assays. A common equation for GFR estimation is encouraged for use in Asians. © 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
E
C
c
T
b
O
4 5 6
F
2
Call for the use of a common equation for glomerular filtration rate estimation in East and South-East Asia
1
R
39 37 38
Introduction
41 42
In general, a glomerular filtration rate (GFR) that is lower than 60 mL/min/1.73 m2 or a urine albumin to creatinine ratio that is higher than 30 mg/g is considered to be an indicator of chronic kidney disease (CKD) [1]. Early diagnosis and treatment of CKD can prevent or slow progression of the disease, which could reduce the incidences of endstage kidney and cardiovascular diseases [2,3]. When such benefit is accepted by the clinical community, physicians are more likely to send a patient to a specialist if they identify such a sign on laboratory reports. It is suspected that the automated laboratory reporting of estimated GFR (eGFR) is associated with a significant increase in the rate of referral of patients to nephrologists [4]. At present, many equations to estimate GFR have been developed, but they were established on the basis of various populations and various serum creatinine (Scr) measurement assays, and referred to various “gold standards” for GFR measurement.
47 48 49 50 51 52 53 54
N C O
45 46
U
43 44
R
40
⁎ Corresponding author at: Department of Clinical Laboratory Sciences, Faculty of Health Sciences, Yamaguchi University Graduate School of Medicine, Minami-Kogushi 1-1-1, Ube 755-8505, Japan. Fax: +81 836 35 5213. E-mail address: [email protected] (K. Ichihara). 1 Present address: Clinical Laboratory, Civil Aviation General Hospital, Beijing, China. 2 Present address: Clinical Laboratory, Beijing Cancer Hospital, Beijing, China.
19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
A fundamental aspect related to GFR estimation deserves thorough investigation: that is, to what extend do the equation and Scr measurement assay affect values of eGFR. A multicenter study titled “The Asian Project for Collaborative Derivation of Reference Intervals” was conducted in East and South-East Asia from 2009 to 2010, aiming at collaborative derivation of reference intervals. The initial report of the study [5] showed that no regional difference was found in Scr level in East and South-East Asians. The present paper reports our follow-up work on this previous study [5]. The objectives of this study were to investigate differences of eGFR among various equation–assay combinations and to study the feasibility of using a common equation in Asian populations. Our hypothesis of using a common equation among all Asians is motivated by the accepted understanding that the Modification of Diet in Renal Disease Study equation (MDRD) is suitable to be used in all Caucasians. In this paper, because Japanese accounted for 46% and Chinese accounted for 28% of all volunteers, we selected 3 equations that contain ethnic considerations: the first is the MDRD equation for Japanese (MDRDJap) [6], which is recommended for use in Japanese by the Japan Society of Nephrology; the second is the MDRD equation for Chinese (MDRDChi), which had been used to investigate the prevalence of CKD in China [7]; and the third is the two-level Chronic Kidney
http://dx.doi.org/10.1016/j.clinbiochem.2014.05.058 0009-9120/© 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Please cite this article as: Wang X, et al, Call for the use of a common equation for glomerular filtration rate estimation in East and South-East Asia, Clin Biochem (2014), http://dx.doi.org/10.1016/j.clinbiochem.2014.05.058
55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76
2
77 78
X. Wang et al. / Clinical Biochemistry xxx (2014) xxx–xxx
For each volunteer, the following six eGFRs were obtained: EPI using the Scr value of the enzymatic assay (EPI/E) and Jaffe assay (EPI/J); MDRDJap using the Scr value of the enzymatic assay (MDRDJap/E) and Jaffe assay (MDRDJap/J); and MDRDChi using the Scr value of the enzymatic assay (MDRDChi/E) and Jaffe assay (MDRDChi/J).
82
Materials and methods
Statistical analysis
134
83
Participant enrollment, sample collection and measurement
135
84
93
This study further explored the data of an existing database created in a comprehensive survey among many cities in eight Asian regions. Participant recruitment, sample collection, specimen transportation, the assay's performance validation and target analyte measurement of the survey were described in a previous report [5]. The study was approved by the Ethical Committee of Yamaguchi University Graduate School of Medical Sciences in December 2008. All the volunteers were duly informed of the medical treatment and procedures associated with this study, and all gave their written consent to participate in this research.
The same statistical methods as reported in the previous study [5] were used. The magnitudes of variation due to sex, age and region were analyzed by three-level nested ANOVA. In brief, the magnitude of variation due to each factor (sex, age and region) was expressed as standard deviations (SD) including between-region SD (SDregion), between-sex SD (SDsex) and between-age SD (SDage). The relative magnitude of each factor to that of the residual SD representing a net between-individual SD (SDnet-btw-indiv) was computed as the SD ratio (SDR) by the following formula:
94
Performance of serum creatinine assays
95 96
107 108
Scr was measured by UniCell DxC 800 analyzer (Beckman-Coulter, USA) using a kinetic Jaffe assay. The enzymatic assay was performed on a Hitachi 917 analyzer, P module, and the reagent was provided by Shino, Japan. All measurement procedures followed the manufacturer's specifications. The accuracy of the two assays was validated by measuring a certified reference material (JCCRM521-10, serum matrix; see the Supplementary material for its certification). The biases of the Jaffe assay and enzymatic assay were − 2.7 μmol/L and − 2.7 μmol/L, respectively, when the Scr level is 78.7 μmol/L, and −5.3 μmol/L and − 3.5 μmol/L, respectively, when the Scr level is 194.5 μmol/L. The within-run coefficient of variation was 2.41% for the Jaffe assay and 1.41% for the enzymatic assay, and the day-to-day coefficient of variation was 2.97% for the Jaffe assay and 1.58% for the enzymatic assay during the study period.
109
Estimation of glomerular filtration rate
110 111
Three equations, the EPI, MDRDJap and MDRDChi, were used. When EPI is used to calculate eGFR, the following equations are used for males and females. For males:
103 104 105 106
112 113 115 116
−0:411
Scr≤0:9 : eGFR ¼ 141 ðScr=0:9Þ ScrN0:9 : eGFR ¼ 141 ðScr=0:9Þ
U
For females: 120 121
0:993
Age
:
N
118
−1:209
Age
0:993
−0:329
Scr≤0:7 : eGFR ¼ 144 ðScr=0:7Þ ScrN0:7 : eGFR ¼ 144 ðScr=0:7Þ
−1:209
Age
0:993 0:993
Age
:
123 124
When the MDRDJap is used to calculate eGFR, the following equation is used for males and females. −1:094
eGFR ¼ 194 ðScrÞ
−0:287
ðAgeÞ
127 129
When the MDRDChi is used to calculate eGFR, the following equation is used for males and females. −1:234
eGFR ¼ 175 ðScrÞ
−0:179
ðAgeÞ
½if female; 0:79
O
133
136 137 138 139 140 141 142 143
R O
P
D
146 147 148 149 150 151 152 153 154
Results Sex, age and regional difference of Scr and eGFR
155
Sex difference was significant for the Scr value because the SDR for Scr was approximately 1.83. However, the sex difference was relatively subdued for the eGFR value because the SDR for eGFR was approximately 0.32. Age difference was significant for eGFR because the SDRs for eGFR were 1.24 for males and 1.42 for females. In contrast, age difference showed no significant effect for the Scr value because the SDRs for Scr were only 0.08 for males and 0.00 for females. Both Scr and eGFR showed no noticeable regional differences by three-level nested ANOVA. The SDRregion for Scr and eGFR was all below 0.25 and was considered not significant (Table 1).
156 157
Scr distribution (by enzymatic assay) in eight Asian regions
167
The Scr value at each percentile in the eight Asian regions studied was shown in Table 2. The Scr value was obtained from the enzymatic assay. The Harris–Boyd method was applied, and an az value greater than 3.0 was deemed to represent a significant difference. Taking Japan as the comparison region, the az values, which ranged from 0.0 to 2.5, suggested no significant regional differences among the eight regions.
168
SDRsex
Scr eGFR by EPI/E
1.83 0.32
158 159 160 161 162 163 164 165 166
169 170 171 172 173 174 t1:1
Table 1 SDratio of Scr and eGFR (EPI/E) by nested ANOVA.
½if female; 0:739
126
131 132
145
The SDR of each factor indicates its degree of influence on the target analyte. An SDR of ≥0.3 was regarded as high, an SDR between 0.26 and 0.29 was considered moderate, and an SDR of ≤ 0.25 was considered low. The Harris and Boyd algorithm [11] was used to determine whether regional groups could be merged. Distribution of eGFR was analyzed based on nonparametric descriptive statistics, by use of a generalpurpose statistical software, Statflex Version 6.0 (Artech Co., Osaka, Japan).
E
101 102
130
SDRfactor ¼ SDfactor =SDnet‐btw‐indiv :
T
99 100
C
97 98
E
91 92
R
89 90
R
87 88
O
85 86
C
79 80
F
81
Disease Epidemiology Collaboration equation (EPI) [8], which has been proved not to require additional ethnic factors and which can be used directly in Asians [9,10]. These three equations all use Scr in their calculation. Two common types of assays were used to measure Scr: the Jaffe assay and enzymatic assay.
Q2t1:2
SDRage
t1:3
SDRregion
Male
Female
Male
Female
t1:4
0.08 1.24
0.00 1.42
0.24 0.00
0.22 0.00
t1:5 t1:6
Standard deviation ratio (SDR) of each factor indicates the degree of influence on the target analyte. SDR ≥0.3 was regarded as high, SDR between 0.26 and 0.29 was considered moderate, and SDR ≤0.25 was considered low [5].
Please cite this article as: Wang X, et al, Call for the use of a common equation for glomerular filtration rate estimation in East and South-East Asia, Clin Biochem (2014), http://dx.doi.org/10.1016/j.clinbiochem.2014.05.058
t1:7 t1:8 t1:9
X. Wang et al. / Clinical Biochemistry xxx (2014) xxx–xxx Table 2 Distribution of Scr (enzymatic assay) of healthy populations in eight Asian regions. Scr μmol/L
N
t2:5 t2:6 t2:7 t2:8 t2:9 t2:10 t2:11 t2:12 t2:13 t2:14 t2:15 t2:16 t2:17 t2:18 t2:19 t2:20 t2:21 t2:22
Male
Female
Japan Seoul Beijing Taiwan Hong Kong HCMC KL Jakarta All Japan Seoul Beijing Taiwan Hong Kong HCMC KL Jakarta All
877 55 51 134 103 106 50 54 1430 1092 70 78 172 121 164 79 78 1854
2.50%
25%
50%
75%
97.50%
57.5 54.4 60.8 60.9 61.0 59.5 64.3 57.1 57.5 39.8 40.0 42.8 40.7 41.6 39.1 40.6 30.9 39.8
67.2 67.4 68.5 69.8 71.8 69.8 73.4 66.3 68.1 49.5 46.9 49.5 46.9 50.4 46.9 47.7 42.4 47.7
72.5 74.3 73.4 77.8 78.7 75.1 78.2 74.7 74.3 53.9 52.2 53.9 51.3 54.8 51.3 53.0 48.6 53.0
79.6 78.5 78.7 84.9 84.9 83.1 86.6 81.3 80.4 58.3 55.7 59.2 55.7 60.1 55.7 58.3 55.7 58.3
93.7 92.3 104.2 102.9 97.2 94.5 101.9 97.5 97.2 69.8 67.4 69.4 65.9 65.8 64.5 70.9 67.7 69.0
F
Region
O
Sex t2:3 t2:4
R O
t2:1 t2:2 Q3
3
Harris–Boyd method all compared with Japan
az az az az az az az
= = = = = = =
0.0 0.7 2.0 2.5 1.2 2.3 0.4
az az az az az az az
= = = = = = =
1.3 0.3 1.9 0.6 1.9 0.3 2.2
t2:23 t2:24
The Harris–Boyd method was used to test for the need of partitioning reference intervals by region. The z score representing differences in two means is adjusted to az by multiplying a coefficient (k), which depends on the sample sizes of the two groups. For partition of reference interval, az ≥3.0 is usually considered as significant [11].
175
eGFR comparison in eight Asian regions
176 177
180 181
Low levels of biases in both the Jaffe assay (2.7 μmol/L) and enzymatic assay (2.7 μmol/L) were identified, but they were both within the permitted range (b4.4 μmol/L) [12]. Therefore, the measured values of Scr were used without further correction to calculate eGFR. Sex and age differences relating to eGFR were investigated by using the Scr measured by enzymatic assay. Within the same age group, the
t3:1 t3:2
Table 3 Age-dependent distribution of eGFR by EPI/E. Age
Region
n
t3:37 t3:38
40–49
50–65
E
P
2.50%
25%
50%
75%
97.50%
103.3 91.1 104.3 98.5 109.4 109.6 105.3 92.8 95.7 80.8 88.9 95.3 92.7 91.3 91.2 103.0 87.8 84.9 86.5 92.3 82.0 85.8 89.1 87.5 79.2 66.9 69.8 78.0 74.0 73.9 78.2 66.2
121.6 117.4 118.2 119.4 122.8 118.8 121.1 119.0 113.1 111.9 112.4 113.1 112.4 109.6 113.7 114.5 104.7 105.2 104.5 105.3 104.4 100.5 106.2 105.4 95.8 96.5 92.6 97.7 94.5 92.8 96.4 96.8
125.3 123.1 127.6 125.3 126.6 122.6 125.4 124.0 117.0 119.3 116.8 117.6 118.0 114.7 118.4 117.2 108.6 108.7 109.4 109.8 113.4 106.9 109.7 109.2 100.9 101.5 101.4 103.1 103.1 99.9 100.4 100.0
129.8 128.6 130.3 131.2 131.5 126.7 129.7 127.9 121.4 124.3 121.6 121.4 124.2 120.4 125.3 119.6 113.1 114.9 114.9 113.8 119.2 110.7 114.0 111.3 104.2 104.2 107.0 105.3 109.4 104.4 107.4 106.6
138.1 138.7 136.6 142.3 152.8 133.8 140.4 135.5 131.1 132.4 130.0 132.5 132.6 128.1 130.4 125.3 121.9 123.7 120.6 121.6 128.0 116.2 123.5 114.1 114.7 114.0 115.4 113.8 123.6 111.9 111.6 112.8
R R
568 26 31 76 69 106 61 35 538 41 40 74 56 81 29 31 444 35 31 94 52 51 27 31 419 23 27 62 47 32 12 35
N C O
30–39
Japan Seoul Beijing Taiwan Hong Kong HCMC KL Jakarta Japan Seoul Beijing Taiwan Hong Kong HCMC KL Jakarta Japan Seoul Beijing Taiwan Hong Kong HCMC KL Jakarta Japan Seoul Beijing Taiwan Hong Kong HCMC KL Jakarta
U
20–29
D
eGFR by EPI/E
t3:3 t3:4 t3:5 t3:6 t3:7 t3:8 t3:9 t3:10 t3:11 t3:12 t3:13 t3:14 t3:15 t3:16 t3:17 t3:18 t3:19 t3:20 t3:21 t3:22 t3:23 t3:24 t3:25 t3:26 t3:27 t3:28 t3:29 t3:30 t3:31 t3:32 t3:33 t3:34 t3:35 t3:36
E
T C
178 179
sex differences relating to eGFR calculated by the three equations were all statistically significant. However, the degree of sex difference, such as 4.1 ± 0.3 mL/min/1.73 m2 for EPI, and 3.3 ± 0.5 mL/min/1.73 m2 for MDRDJap, was not clinically significant, except for MDRDChi, which showed a relatively larger sex difference (15 ± 0.8 mL/min/1.73 m2). Because age difference relating to eGFR was significant, the participants were divided into 4 age groups. Table 3, as represented by EPI/E, showed that within the same age group, the distributions of eGFR were very
Harris–Boyd method all compared with Japan
az az az az az az az
= = = = = = =
0.2 1.1 0.7 1.5 0.2 1.3 1.2
az az az az az az az
= = = = = = =
0.4 0.1 0.4 1.3 0.2 0.5 0.3
az az az az az az az
= = = = = = =
0.9 0.4 0.4 1.7 0.9 0.2 0.9
az az az az az az az
= = = = = = =
0.4 0.5 0.2 0.8 0.1 0.1 0.6
The Harris–Boyd method was used to test for the need of partitioning reference intervals by region. The z score representing differences in two means is adjusted to az by multiplying a coefficient (k), which depends on the sample sizes of the two groups. For partition of reference intervals, az ≥3.0 is usually considered as significant [11].
Please cite this article as: Wang X, et al, Call for the use of a common equation for glomerular filtration rate estimation in East and South-East Asia, Clin Biochem (2014), http://dx.doi.org/10.1016/j.clinbiochem.2014.05.058
182 183 184 185 186 187 188 189
208
Differences in eGFRs
209 210
The differences in eGFRs caused both by using the same equation but with different assays, and by using the same assay but with different equations were listed in Table 4. In the first three records of Table 4, which show eGFR differences caused by using different assays, it can be seen that the equation which produces the least difference is the EPI equation. Between EPI/E and EPI/J, it can be seen that a median of the difference is about 2 mL/min/1.73 m2, whereas the range of the difference is from −6 to 14 mL/min/1.73 m2 (2.5% to 97.5%). A moderate difference was found with the MDRDJap equation. Between MDRDJap/E and MDRDJap/J, the difference has a median of 3 mL/min/1.73 m2 and a range from − 12 to 18 mL/min/1.73 m2 (2.5% to 97.5%). The highest difference was between MDRDChi/E and MDRDChi/J, which has a median of 5 mL/min/1.73 m2 and a range from −21 to 30 mL/min/1.73 m2 (2.5% to 97.5%). In the last three records of Table 4, it can be seen that eGFR difference caused by using different equations was much larger than that caused by using different assays. For example, when enzymatic assays were used for estimating eGFR, the medians (2.5%, 97.5%) of the eGFR crossequation difference were 11 (− 6, 56) mL/min/1.73 m2 between
215 216 217 218 219 220 221 222 223 224 225 226 227
C
213 214
E
211 212
R
205 206
R
203 204
0 −1 −1
2 3 5
4 7 12
14 18 30
Caused by equations MDRDChi/E − EPI/E EPI/E − MDRDJap/E MDRDChi/E − MDRDJap/E
−6 9 22
3 22 31
11 26 39
24 30 48
56 35 65
t4:3 Q4 t4:4 t4:5 t4:6 t4:7 t4:8 t4:9 t4:10 t4:11 t4:12
MDRDChi/E and EPI/E, 26 (9, 35) mL/min/1.73 m2 between EPI/E and MDRDJap/E, and 39 (22, 65) mL/min/1.73 m2 between MDRDChi and MDRDJap. eGFR difference caused by using different equations was much larger than that caused by using different assays. The difference in eGFR was further evaluated in two eGFR categories, eGFR b 90 mL/min/1.73 m2 and eGFR N 90 mL/min/1.73 m2, where eGFR was calculated using EPI/E (Fig. 2). As a whole, eGFR difference in EPI/E b 90 mL/min/1.73 m 2 was smaller than that in EPI/E N 90 mL/min/1.73 m2. When EPI/E was lower than 90 mL/min/1.73 m2, the eGFR difference caused by assays was mostly within ±10 mL/min/1.73 m2. However, eGFR difference caused by equations has substantial variations. The eGFR difference between MDRDChi/E and EPI/E was found to be the smallest, with a range from − 5.0 to 7.6 mL/min/1.73 m2 (2.5% to 97.5%), whereas the eGFR difference between EPI/E and MDRDJap/E ranged from 17.2 to 27 mL/min/1.73 m2, and that between MDRDChi/E and MDRDJap/E ranged from 17.0 to 33.6 mL/min/1.73 m2. When EPI/E was higher than 90 mL/min/1.73 m2, the eGFR difference caused by assays became slightly larger, but mostly within ± 20 mL/min/1.73 m2. The smallest difference existed between EPI/E and EPI/J. However, the eGFR difference caused by equations was much larger, ranging from − 6 to 56 mL/min/1.73 m2 between MDRDChi/E and EPI/E, 9 to 35 mL/min/1.73 m2 between EPI/E and MDRD Jap /E, and 23 to 66 mL/min/1.73 m 2 between MDRD Chi /E and MDRDJap/E.
228 229
Discussion
253
230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252
One of our findings is that, similar to Scr, Scr-based eGFR does not re- 254 sult in a significant difference among populations in the eight Asian 255
O
201 202
−6 −12 −21
C
199 200
97.5%
N
198
75%
U
196 197
50%
F
207
Because no regional difference was found, eGFRs from subjects of all eight regions were collectively analyzed to reveal any distribution patterns. The distribution patterns of six eGFRs were shown in Fig. 1. Overall, the distribution patterns of eGFR by EPI (including EPI/E and EPI/J) and by MDRDJap (including MDRDJap/E and MDRDJap/J) were similar and showed a relatively smaller between-individual biological variation than that of MDRDChi (including MDRDChi/E and MDRDChi/J). Taking the entire distribution pattern of eGFR values by EPI as a whole, it represents a right-shifted distribution, by approximately 20 mL/min/1.73 m2, of the MDRDJap, indicating a systematic difference. The distribution pattern of the segment between the 2.5 percentile and the median by EPI was similar to that of MDRDChi, whereas the distribution pattern of the segment between the median and the 97.5 percentile by EPI was shorter than that of MDRDChi equation.
25%
O
194 195
2.5%
Caused by assays EPI/E − EPI/J MDRDJap/E − MDRDJap/J MDRDChi/E − MDRDChi/J
R O
eGFR distribution patterns by various equation–assay combinations
eGFR difference (mL/min/1.73 m2)
T
193
t4:1 t4:2
Table 4 Distribution of eGFR differences caused by assays and equations.
P
192
similar to each other among the eight regions. When using other equation–assay combinations to calculate GFR, similar results were obtained, and no regional difference of eGFR was found (data not shown).
D
190 191
X. Wang et al. / Clinical Biochemistry xxx (2014) xxx–xxx
E
4
Fig. 1. Distribution patterns of six eGFRs. The unit of the horizontal axis is mL/min/1.73 m2.
Please cite this article as: Wang X, et al, Call for the use of a common equation for glomerular filtration rate estimation in East and South-East Asia, Clin Biochem (2014), http://dx.doi.org/10.1016/j.clinbiochem.2014.05.058
5
R O
O
F
X. Wang et al. / Clinical Biochemistry xxx (2014) xxx–xxx
EPI/E < 90ml/min/1.73m2
EPI/E > 90ml/min/1.73m2
266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296
297
Conclusions
320
This study revealed that creatinine levels were similar independent of the assay, and the eGFRs determined by country-specific MDRD and EPI were also similar, in healthy populations from eight Asian regions. However, the difference in eGFR caused by using different eGFR equations was much greater than that caused by using different Scr assays. Nowadays, people frequently move across regions or even countries, and the world is in a process of rapid globalization. To survey the eGFR value in a continuous and comparable way, we call for a common version of the eGFR equation to be used in East and South-East Asia, or globally, for the automated reporting of eGFR by clinical laboratories.
321 322
Conflicts of interest
332
D
The EPI exploration was based on multiple ethnic populations, including Asians. A recent study [9] showed that the EPI did not require an ethnic factor when used with CKD patients in Asia. An Australia guideline [10] stated that EPI has been validated as a tool to estimate GFR in some non-European populations, including South-East Asian, African, Indian and Chinese individuals living in Western countries. Kong et al. [17] compared the MDRDChi with EPI and demonstrated that the two equations performed equally in CKD staging in the general population. In our study, when EPI/E was b 90 mL/min/1.73 m2, the difference between EPI/E and MDRD Chi was mostly below 10 mL/min/1.73 m 2 . This small difference between the two equations is consistent with the result of Kong et al. Although the difference between EPI/E and MDRDChi was much larger when EPI/E was N90 mL/min/1.73 m2, this will not influence CKD staging. This result indicated that, at a certain level of GFR, EPI/E and MDRDChi have equal performance and that the EPI might be used directly in Chinese without an ethnic factor. Our research has certain limitations as a practical matter of selecting and treating subjects. First, we did not measure GFR, so we could not evaluate the accuracy of the equations. Second, the enrolled subjects included an elderly (age N60 years) population of only 1.9%. It would be desirable to enroll more elderly volunteers with different levels of GFR to better compare the performance of the equations.
E
T
C
265
E
263 264
R
261 262
R
259 260
regions studied, no matter which equation–assay combination is used. The distribution corresponding to EPI/E in this study (Supplementary Table 1), in which half of the volunteers were healthy Japanese, is similar to the result in another study, in which the subjects were healthy Chinese [13]. The similarity of eGFR distribution between these two studies strongly suggested that a common equation could be used to estimate eGFR across different Asian regions, at least in Chinese and Japanese. Another finding is that different equations cause much larger differences in eGFR than those by different assays. When using the same equation with different Scr assays, the eGFR difference was derived from the difference in Scr values. Guided by the recommendations to improve Scr measurement [12], almost all IVD manufacturers have implemented procedures for standardization of Scr assays [14]. Thus, the difference in Scr among assays is mainly caused by the interference materials in blood. Although different assays respond to non-creatinine substances in different ways [15], the contribution of non-specificity to a Scr value is supposed to be in a limit range. For example, in our study, about 95% of the assay difference was below 10 μmol/L and about 70% of the difference was below 5 μmol/L, with the result that most of the eGFR difference caused by assays was within the range of −15 to 15 mL/min/1.73 m2. However, the same Scr value results in obviously different eGFRs when different equations are used, especially in subjects with a normal to higher GFR as shown in Fig. 2. One possible reason is that the mean GFR for equation development was different in each study. For example, the mean ± SD of GFR in the EPI development was 68 ± 40 mL/min/1.73 m2 , it was higher than that in MDRDChi development, which was 55.1 ± 35.1 mL/min/1.73 m 2, and it was also higher than that in MDRDJap development, which was 59.1 ± 35.4 mL/min/1.73 m2. This might explain the large difference in normal to higher GFR values (such as that of N90 mL/min/1.73 m2) among the equations. Another possible reason that may explain the large effect created by the equations is that the “gold standard” GFR was obtained through different protocols. For example, MDRDJap used inulin clearance to obtain the standard GFR; MDRDChi used 99mTcDTPA plasma clearance; and the EPI equation used iothalamate clearance. There exists certain difference among the various protocols. For example, 99mTc-DTPA usually result in a higher “gold standard” GFRs than that by inulin [16], which might explain why higher eGFR values are calculated with the MDRDChi than with the MDRDJap.
N C O
257 258
U
256
P
Fig. 2. Distribution of eGFR difference categorized by EPI/E. When EPI/E was lower than 90 mL/min/1.73 m2, the difference between MDRDChi and EPI was the smallest, but when EPI/E was higher than 90 mL/min/1.73 m2, the difference between the two equations was much larger.
None.
Please cite this article as: Wang X, et al, Call for the use of a common equation for glomerular filtration rate estimation in East and South-East Asia, Clin Biochem (2014), http://dx.doi.org/10.1016/j.clinbiochem.2014.05.058
298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319
323 324 325 326 327 328 329 330 331
333
X. Wang et al. / Clinical Biochemistry xxx (2014) xxx–xxx
352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
C
407
F
Appendix A. Supplementary data
O
350 351
E
348 349
R
346 347
R
344 345
O
342 343
C
340 341
N
339
U
337 338
395 396 397 398 399 400 401 402 403 404 405 406
Supplementary data to this article can be found online at http://dx. 408 doi.org/10.1016/j.clinbiochem.2014.05.058. 409 References
R O
This research was conducted collaboratively among 1) the committee of plasma protein and committee of reference intervals and decision limits in the International Federation of Clinical Chemistry and Laboratory Medician (IFCC), 2) the Scientific Committee in the Asian and Pacific Federation of Clinical Biochemistry, 3) the Working Group on the Guideline for Common Reference Interval in the Japan Society of Laboratory Medicine, and 4) the Committee on Plasma Protein of the Japan Society of Clinical Chemistry. This study was supported by the Committee on Reference Interval and Decision Limit of the IFCC. Research funds include those from a Scientific Research Fund (No. 21406015: 2009–2011) provided by the Ministry of Education, Culture, Sports, Science and Technology of Japan, a Research Promotion Project Fund of the JSLM (2008–2009), and a Scientific Research Fund of the APFCB, and a research fund (No. Z121107005112006, 2012–2015) provided by Beijing Municipal Science and Technology Commission. Most of the reagents and labor required for testing the large number of analytes were generously offered by Beckman-Coulter, Inc. Additional reagents were offered by Nittobo Co., Eiken Chemical Co. and Otsuka Pharmaceutical Co. The sampling equipment (vacuum sample tubes, needles and the holders) was supplied by Becton-Dickinson (BD). Other supplies, such as serum container tubes and boxes and ID labels were purchased by use of the Scientific Research Funds offered by the APFCB. We are grateful to all of the laboratories and staff that have made contributions to the study. The collaborating laboratories outside Japan included 1) Gangnam Severance Hospital, Yonsei University, Seoul; 2) Kangbuk Samsung Hospital, Sung Kyun Kwan University, Seoul; 3) Myongji Hospital, Kwandong University College of Medicine, Seoul; 4) Prince of Wales Hospital, Chinese University of Hong Kong, Hong Kong; 5) Macau Institute for Applied Research in Medicine and Health, Macau University of Science and Technology, Macau; 6) Prodia Clinical Laboratory, Jakarta; 7) Mackay Memorial Hospital, Taipei; 8) Cathay General Hospital, Taipei; 9) Yuan Ching Clin Laboratory, Taipei; 10) National Cheng Kung University, Tainan; 11) Chi-Mei Med Center, Tainan; 12) National Heart Institute, Kuala Lumpur; 13) Gleneagles Intan Medical Centre, Kuala Lumpur; and 14) Medic-Lab, Ho Chi Minh City. The collaborating laboratories in Japan included 15) Yamaguchi University, Ube; 16) Yoshida Hospital, Asahikawa; 17) Hokkaido University, Sapporo; 18) Kishimoto Clinical Laboratory, Tomakomai; 19) Iwate Medical School, Morioka; 20) Hirosaki University, Hirosaki; 21) Hachinohe Red-Cross Hospital, Hachinohe; 22) Hachinohe Central Hospital, Hachinohe; 23) Chiba Cardiovascular Center, Ichihara; 24) Chiba University, Chiba; 25) Tokyo University, Tokyo; 26) Toho University, Tokyo; 27) Shinshu University, Matsumoto; 28) Matsumoto JAM Medical Center, Matsumoto; 29) Ohmachi Municipal Hospital, Omachi; 30) Yamanashi University, Kofu; 31) Fukui University, Fukui; 32) Kanazawa Medical School, Kanazawa; 33) SRL-Kanazawa, Kanazawa; 34) Mie University, Tsu; 35) Nagoya University, Nagoya; 36) Anjo Kosei Hospital, Anjo; 37) Fujita Health University, Toyoake; 38) Osaka University, Suita; 39) Tenri Hospital, Tenri; 40) Osaka Municipal University, Osaka; 41) National Cardiovascular Center, Suita; 42) Kawasaki Medical School, Kurashiki; 43) Kurashiki Central Hospital, Kurashiki; 44) Okayama University, Okayama; 45) Okayama Medical Laboratory, Kurashiki; 46) Matsuda Hospital, Kurashiki; 47) Hiroshima University, Hiroshima; 48) Yamaguchi University, Ube; 49) Tokuyama Central Hospital, Shunan; 50) Yamaguchi Prefectural Medical Center, Hofu;
393 394
[1] Kidney Disease: Improving Global Outcomes (KDIGO) Work Group. KDIGO clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int Suppl 2013;3:1–150. [2] Lewis EJ, Hunsicker LG, Clarke WR, Berl T, Pohl MA, Lewis JB, et al. Renoprotective effect of the angiotensin-receptor antagonist irbesartan in patients with nephropathy due to type 2 diabetes. N Engl J Med 2001;345:851–60. [3] Wright Jr JT, Bakris G, Greene T, Agodoa LY, Appel LJ, Charleston J, et al, African American Study of Kidney Disease and Hypertension Study Group. Effect of blood pressure lowering and antihypertensive drug class on progression of hypertensive kidney disease: results from the AASK trial. JAMA 2002;288:2421–31. [4] Hemmelgarn BR, Zhang J, Manns BJ, James MT, Quinn RR, Ravani P, et al. Nephrology visits and health care resource use before and after reporting estimated glomerular filtration rate. JAMA 2010;303:1151–8. [5] Ichihara K, Ceriotti F, Tam TH, Sueyoshi S, Poon PM, Thong ML, et al. The Asian project for collaborative derivation of reference intervals: (1) strategy and major results of standardized analytes. Clin Chem Lab Med 2013;51:1429–42. [6] Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, et al. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis 2009;53(6):982–92. [7] Zhang LX, Wang F, Wang L, Wang W, Liu B, Liu J, et al. Prevalence of chronic kidney disease in China: a cross-sectional survey. Lancet 2012;379:815–22. [8] Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro 3rd AF, Feldman HI, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009;150:604–12. [9] Teo BW, Xu H, Wang DH, Li J, Sinha AK, Shuter B, et al. Estimating glomerular filtration rates by use of both cystatin c and standardized serum creatinine avoids ethnicity coefficients in Asian patients with chronic kidney disease. Clin Chem 2012;58:450–7. [10] Johnson DW, Jones GR, Mathew TH, Ludlow MJ, Doogue MP, Jose MD, et al. Chronic kidney disease and automatic reporting of estimated glomerular filtration rate: new developments and revised recommendations. Med J Aust 2012;197:222–3. [11] Harris EK, Boyd JC. Statistical basis of reference values in laboratory medicine. New York: Marcel Dekker; 1995. [12] Myers GL, Miller WG, Coresh J, Fleming J, Greenberg N, Greene T, et al. Recommendations for improving serum creatinine measurement: a report from the Laboratory Working Group of the National Kidney Disease Education Program. Clin Chem 2006;52:5–18. [13] Wang X, Xu G, Li H, Liu Y, Wang F. Reference intervals for serum creatinine with enzymatic assay and evaluation of four equations to estimate glomerular filtration rate in a healthy Chinese adult population. Clin Chim Acta 2011;412:1793–7. [14] http://nkdep.nih.gov/lab-evaluation/gfr/creatinine-standardization/recommendations.shtml#ivd; July 18, 2013 . Accessed on. [15] Greenberg N, Roberts WL, Bachmann LM, Wright EC, Dalton RN, Zakowski JJ, et al. Specificity characteristics of 7 commercial creatinine measurement procedures by enzymatic and Jaffe method principles. Clin Chem 2012;58(2):391–401. [16] Dai SS, Yasuda Y, Zhang CL, Horio M, Zuo L, Wang HY. Evaluation of GFR measurement method as an explanation for differences among GFR estimation equations. Am J Kidney Dis 2011;58(3):496–8. [17] Kong X, Ma Y, Chen J, Luo Q, Yu X, Li Y, et al. Evaluation of the Chronic Kidney Disease Epidemiology Collaboration equation for estimating glomerular filtration rate in the Chinese population. Nephrol Dial Transplant 2013;28(3):641–51.
P
335 336
51) Saiseikai Yamaguchi Hospital, Yamaguchi; 52) Saint Hill Hospital, Ube; 53) Kochi University, Nangoku; 54) Kochi Red-Cross Hospital, Kochi; 55) Kumamoto University, Kumamoto; 56) Ryukyu University, Naha; and 57) Adventist Medical Center, Naha. This study is also indebted to Professor Christopher W. K. Lam (Macau Institute for Applied Research in Medicine and Health, Macau University of Science and Technology, Macau), Emeritus Professor Takeshi Kanno (Hamamatsu University School of Medicine), Emeritus Professor Tadashi Kawai (Jichi University School of Medicine), Dr. Katsuhiko Kuwa (National Metrology Institute of Japan), Prof. Susumu Osawa (Kyushu University School of Medicine), and Prof. Shigemi Hosogaya (Kagawa Prefectural College of Health Sciences), who gave invaluable advice and encouragement to promote and carry out this study.
D
Acknowledgments
T
334
E
6
Please cite this article as: Wang X, et al, Call for the use of a common equation for glomerular filtration rate estimation in East and South-East Asia, Clin Biochem (2014), http://dx.doi.org/10.1016/j.clinbiochem.2014.05.058
410 411 Q6 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456
