Abstract
Objective
The aim of this study was to explore the possible influences of age and gender on urinary Alpha 1‐Microglobulin (α1‐MG) concentrations and establish specific reference values of urinary α1‐MG in a group of healthy adults.
Methods
Two hundred and ninety‐nine adults (141 males and 158 females) aged 20‐60 were selected and grouped by gender and age. Urinary levels of α1‐MG were detected in morning spot‐urine samples, and statistical analysis was performed to explore the association between urinary α1‐MG and clinical parameters, and the differences between groups were compared. The 95th percentile of distribution was used as the normal upper limit.
Results
The value of urinary α1‐MG was sex‐dependent (P<0.001), the 95th percentile of urinary α1‐MG of males was 26.4 mg/L, whereas females was 8.6 mg/L. Whereas urinary α1‐microglobulin‐to‐creatinine ratio (α1‐MG/Cr) was both gender‐and age‐dependent (both P<0.001), and age 30 may be the cut‐off point in this study. The 95th percentile of males and females aged 20‐30 was 16.9 mg/g, of males aged 31‐60 was 19.8 mg/g, of females aged 31‐60 was 28.5 mg/g.
Discussions
The relationship between urinary α1‐MG concentrations and gender, age is different from previous studies. It is recommended that hospitals and laboratories should develop their own α1‐MG reference intervals based on their experimental conditions.
Keywords: α1‐microglobulin‐to‐creatinine ratio, Alpha 1‐microglobulin, reference value
1. INTRODUCTION
Alpha 1‐Microglobulin (α1‐MG) is a low molecular weight glycoprotein mostly produced by liver and exists in blood in free or bound (complexes with IgA or albumin) forms.1, 2 Under normal conditions, approximately 99.9% of the free α1‐MG filtered freely through the renal glomerular basement membrane is reabsorbed by the proximal tubular cells. Any proximal tubular cell dysfunction could result in increased quantities of α1‐MG in the urine, hence urinary α1‐MG has been recognized as a sensitive marker for the early detection of tubular injury in many diseases,3, 4 especially in diabetic kidney disease in recent years.5, 6 Urinary α1‐MG has been recommended by expert consensus on clinical diagnosis of diabetic nephropathy as an indicator to screen early tubular involvement in Chinese adults.7
Klaus Jung and his fellow8 found the normal reference interval of urinary α1‐microglobulin‐to‐creatinine ratio (α1‐MG/Cr) was as follows: 1.27 g/mol (11.2 mg/g) creatinine for people aged 18‐40 and 2.20 g/mol (19.4 mg/g) creatinine for people aged >40.8 Then Lothar Thomas9 agreed the previous reference value range of α1‐MG/Cr, and gave a new one for urinary α1‐MG: 12 mg/L or 20 mg/24 h for all ages. However, there are few detailed and comprehensive studies on the reference interval of urinary α1‐MG concentrations in Chinese people, and we usually refer to the western results. But our recent studies found that, the incidence of elevated urinary α1‐MG/Cr was higher in men than in women,10 and male was the most important risk factor associated with tubular injury.11 Similarly, a study about urinary α1‐MG in Singapore also found the urinary α1‐MG excretion increased with age, and was higher in men (1.41 mg/mmol urine creatinine) than in women (1.14 mg/mmol urine creatinine).5 Therefore, whether there is gender difference in urinary α1‐MG/Cr in healthy individuals is not clear so far.
Here, we investigated the possible influences of age and gender on urinary α1‐MG concentrations in a group of Chinese healthy adults, and tried to establish reference values for further clinical applications.
2. MATERIALS AND METHODS
2.1. Study population
People who visited the Medical examination center of Tianjin Union Medical Center between October 2016 and January 2017 were recruited into our study. They were screened by self‐reporting, professional doctor's examination and laboratory examination. Inclusion criteria were as follows: (1) aged between 20‐60 years old; (2) no self‐reported hypertension, diabetes, cardiovascular diseases, or kidney diseases, no pregnancy, no cancer, infections, or inflammatory conditions; (3) body mass index (BMI) between 18 and 28 kg/m2; (4) systolic blood pressure (SBP) <140 mm Hg and diastolic blood pressure (DBP)<90 mmHg; (5) no dyslipidemia, and total cholesterol (TC) <6.22 mmol/L, triglyceride (TG) <2.26 mmol/L; (6) white cells(−), blood cells(−), and protein(−) of urinalysis; (7) estimated glomerular filtration rate (eGFR) ≥90 mL/min/1.73 m2. Finally, a total of 299 persons (141 males and 158 females)were selected, and grouped according to gender and age.
Our study conforms to the provisions of the ethical standards of the Health bureau of Tianjin, and all participants provided written informed consents.
2.2. Measurements
Anthropometric measurements and laboratory examinations were recorded. Morning spot‐urine samples and blood samples were obtained from subjects at physical examination, then stored at −80°C, thawed at room temperature, and centrifuged (400 × g, 15 min) before use. Blood and urine samples were tested in our clinical laboratory department. Urinary α1‐MG were assessed by nephelometry immunoassay on a Dade‐Behring BNII special protein analyzer, with reagents provided by manufacturer. We assigned 5.56 mg/L to α1‐MG concentration lower than 5.56 mg/L as the detection limit was 5.56 mg/L. Urinary creatinine was measured by an enzymatic method on AbbottIC16000. α1‐MG/Cr were calculated. We calculated eGFR by the 2009 Chronic Kidney Disease Epidemiology Collaboration (2009 CKD‐EPI) equation: (1) for females: if serum creatinine (Scr) ≤62 μmol/L, eGFR=144×(Scr/62)−0.329×0.993age; if Scr>62 μmol/L, eGFR=144×(Scr/62)−1.209×0.993age; (2) for males: if Scr ≤80 μmol/L, eGFR=141×(Scr/80)−0.411×0.993age; if Scr>80 μmol/L, eGFR=141×(Scr/80)−1.209×0.993age.
2.3. Statistical analysis
Analyses were performed with SPSS 20.0 (IBM Corporation, Armonk, NY, USA). Normal distribution data (like age, SBP and DBP, et al.) were expressed as mean±SD, skew distributional data (α1‐MG and α1‐MG/Cr) were expressed as median and inter‐quartile range. Logarithmic transformation of α1‐MG and α1‐MG/Cr was used for further analysis. Univariate linear regression analysis with Log(urinary α1‐MG/Cr) or Log(urinary α1‐MG) as the dependent variable, respectively, whereas age, gender, BMI, SBP, DBP, fasting glucose(FG), TG, TC, eGFR as the independent variable was used to evaluate the association between them. Then the related variables were further analyzed by multivariate stepwise regression analysis to explore the influencing factors of urinary α1‐MG and α1‐MG/Cr. Differences among groups were tested by T test for normal distribution variables and Mann‐Whitney U test for non‐normal distribution variables (α1‐MG and α1‐MG/Cr). Finally, for different groups, the 95th percentile was used to determine the upper normal boundary of urinary α1‐MG and α1‐MG/Cr. P<0.05 was considered statistically significant.
3. RESULTS
Table 1 shows the baseline characteristics of all participants including 141 males and 158 females. The average age of males was lower than females, but they had higher BMI, DBP, SBP, FG, and TG values. Even though there was no significant difference in renal function, the males had significantly higher urinary α1‐MG values, but apparently lower urinary α1‐MG/Cr values, compared with females.
Table 1.
Characteristics of participants
| Male | Female | P | |
|---|---|---|---|
| N | 141 | 158 | |
| Age (year) | 37.6±10.1 | 40.9±10.7 | 0.007 |
| BMI (kg/m2) | 23.4±2.4 | 22.4±2.5 | 0.001 |
| SBP (mmHg) | 118.7±11.1 | 113.5±12.5 | <0.001 |
| DBP (mmHg) | 78.2±7.3 | 75.0±7.6 | <0.001 |
| FG (mmol/L) | 5.2±0.4 | 5.1±0.4 | 0.047 |
| TC (mmol/L) | 4.5±0.6 | 4.5±0.7 | 0.815 |
| TG (mmol/L) | 1.1±0.5 | 0.9±0.4 | <0.001 |
| eGFR (mL/min/1.73 m2) | 113.0±9.8 | 111.5±9.9 | 0.192 |
| α1‐MG (mg/L) | 7.8 (5.4, 12.5) | 5.4 (5.4, 5.4) | <0.001 |
| α1‐MG/Cr (mg/g) | 7.2 (4.5,9.5) | 10.4 (5.9, 14.7) | <0.001 |
P<0.05 was considered statistically significant.
The results of the linear regression analysis with Log(urinary α1‐MG/Cr) or Log(urinary α1‐MG) as the dependent variable, respectively, were shown in Tables 2 and 3. In univariate linear regression analysis, the relationship between gender (β=0.198, P<0.001), DBP (β=0.003, P=0.022), TG (β=0.106, P<0.001), eGFR (β=−0.003, P=0.006), and Log(urinary α1‐MG) were significant, but only gender (β=0.2, P<0.001) and eGFR (β=−0.004, P<0.001) still maintained a significant correlation with Log(urinary α1‐MG) after multivariate analysis. In contrast, the relationship between Log(urinary α1‐MG/Cr) and gender, age were significant in both univariate and multivariate analysis, and the significant correlation of eGFR with Log(urinary α1‐MG/Cr) lost after multivariate analysis (P=0.005 vs P=0.409).
Table 2.
Regression analyses with Log(urinary α1‐MG) as dependent variable
| Univariate | Multivariate | |||
|---|---|---|---|---|
| Variables | Standardβ | P | Standardβ | P |
| Gender | 0.198 | <0.001 | 0.2 | <0.001 |
| Age | 0.001 | 0.512 | – | – |
| BMI | 0.003 | 0.497 | – | – |
| SBP | 0.002 | 0.082 | – | – |
| DBP | 0.003 | 0.022 | 0.02 | 0.956 |
| FG | 0.053 | 0.052 | – | – |
| TC | 0.027 | 0.119 | – | – |
| TG | 0.106 | <0.001 | 0.053 | 0.826 |
| eGFR | −0.003 | 0.006 | −0.004 | <0.001 |
P<0.05 was considered statistically significant.
Table 3.
Regression analyses with Log(urinary α1‐MG/Cr) as dependent variable
| Univariate | Multivariate | |||
|---|---|---|---|---|
| Variables | Standardβ | P | Standardβ | P |
| Gender | −0.147 | <0.001 | −0.125 | <0.001 |
| Age | 0.008 | <0.001 | 0.007 | <0.001 |
| BMI | 0.001 | 0.883 | – | – |
| SBP | 0.002 | 0.168 | – | – |
| DBP | 0.002 | 0.354 | – | – |
| FG | 0.001 | 0.988 | – | – |
| TC | 0.012 | 0.651 | – | – |
| TG | −0.044 | 0.263 | – | – |
| eGFR | −0.005 | 0.005 | 0.127 | 0.409 |
P<0.05 was considered statistically significant.
All participants were grouped according to gender and age, and the differences in urinary α1‐MG (no differences between groups, data not displayed) or α1‐MG/Cr (as shown in Tables 4 and 5) between groups were also comparised. In males, the urinary α1‐MG/Cr of the 31‐40 or 41‐50 years old groups were similar to the other groups, but the 20‐30 group was significantly lower than that of 51‐60 group. While in females, only the 20‐30 group had significantly lower urinary α1‐MG/Cr than the other three groups. Then we compared the urinary α1‐MG/Cr between males and females before and after the age of 30, and found the urinary α1‐MG/Cr was similar between males and females in the age group less than 30 years old, but in the other group, the value of males was significantly lower than that of females. As the average age in the 31‐60 group was different between men and women, linear regression analysis was chosen and the association between gender but not age with urinary α1‐MG/Cr was found (data not displayed).
Table 4.
Urinary α1‐MG/Cr after grouping by age and gender
| Males | Females | |||
|---|---|---|---|---|
| Age (year) | N | α1‐MG/Cr | N | α1‐MG/Cr |
| 20‐30 | 40 | 5.4 (3.7, 8.5) | 33 | 5.4 (2.6, 9.8) |
| 31‐40 | 51 | 7.2 (4.5, 9.0) | 44 | 12.1 (7.6, 16.7) |
| 41‐50 | 25 | 8.8 (4.5, 12.7) | 51 | 11.5 (7.3, 18.3) |
| 51‐60 | 25 | 8.7 (5.8, 11.8) | 30 | 10.9 (7.0, 14.7) |
Table 5.
Comparison of urinary α1‐MG/Cr after grouping by age and gender
| Age (N) | 20‐30 | 31‐40 | 41‐50 | 51‐60 |
|---|---|---|---|---|
| For males | ||||
| 20‐30 (40) | – | 0.137a | 0.056a | 0.016a |
| 31‐40 (51) | 0.137a | – | 0.23a | 0.088a |
| 41‐50 (25) | 0.056a | 0.23a | – | 0.884a |
| 51‐60 (25) | 0.016a | 0.088a | 0.844a | – |
| For females | ||||
| 20‐30 (33) | – | <0.001b | <0.001b | <0.001b |
| 31‐40 (44) | <0.001b | – | 0.715b | 0.538b |
| 41‐50 (51) | <0.001b | 0.715b | – | 0.739b |
| 51‐60 (30) | <0.001b | 0.538b | 0.739b | – |
P values of Mann‐Whitney U test between two groups.
P values of Mann‐Whitney U test between two groups.
The value of α1‐MG was sex‐dependent, the 95th percentile of urinary α1‐MG for males was 26.4 mg/L, for females was 8.6 mg/L. While urinary α1‐MG/Cr was both gender‐and age‐dependent, the 95th percentile for males and females aged 20‐30 was 16.9 mg/g, for males aged 31‐60 was 19.8 mg/g, for females aged 31‐60 was 28.5 mg/g.
4. DISCUSSION
In the current clinical application of urinary protein as a sensitive marker of renal tubular injury, we tended to use the reference value recommended by Lothar Thomas,9 as there is no specific definition for Chinese people. Some people also use different reference value for clinical research (0.07‐5 mg/g).12 Then this study investigated the possible influences of age and gender on urinary α1‐MG concentrations and tried to establish specific reference values in a group of healthy Chinese adults.
The differences in baseline characteristics between females and males in this study were similar to a cross‐sectional study including more than three thousand people from China.13 The value of urinary α1‐MG was associated with gender, but did not related to age in this study, different from Klaus Jung's findings.8 A study of children with type 1 diabetes also found no correlation between urinary α1‐MG and age.6 The reason for this phenomenon can be explained from three aspects. First, the detection method for urinary α1‐MG we used here is nephelometry on a Dade‐Behring BNII special protein analyzer, different from the single radial immunodiffusion used by Klaus Jung. So the detection results of these two methods inevitably exist a certain differences. Second, the eGFR of all subjects included in this study is not less than 90 mL/min/1.73 m2, ensuring the renal function within the normal range,14 and no participants were older than 60 years. As we all know, the kidney function of people more than 60 years old is obviously lower than that of young people.15, 16 Because of the younger age and relatively normal tubular function, there may be no difference in urinary protein between age groups. Finally, the number of samples we selected was not sufficient, and there was no differences of urinary α1‐MG in women among all age groups, but an increasing trend with age was found in males, even though the difference was not significant enough (data not displayed). So the lack of older people may also be one of the reasons for this result.
As the value of urinary α1‐MG was gender‐dependent and a marked within‐day variations was noted in urinary α1‐MG,17 using the ratio of urinary α1‐MG to creatinine is an improved practical estimate of the excretion rates,18 can reduce the difference between men and women to a certain extent. So the reference interval established by Lothar Thomas for urinary α1‐MG/Cr took into account the age difference, but did not consider gender difference. Our results indicates in healthy adults aged 20‐30 years old, there is indeed no difference of urinary α1‐MG/Cr between men and women, whereas remarkable difference is still existed in 31‐60 groups. This may be due to a decline in renal function after 30 years of age.16 Then age 30 maybe a possible cut‐off point for the value of urinary α1‐MG/Cr in this study, lower than the cut‐off age of 40 raised by Lothar Thomas.9 Based on these information, the establish of new reference ranges for urinary α1‐MG/Cr should consider both age and gender in Chinese healthy people.
As the values of urinary α1‐MG and α1‐MG/Cr are skew distributional, we use the upper 95th percentile as reference ranges in this study. Then for urinary α1‐MG, the reference range for males is 26.4 mg/L, for females is 8.6 mg/L. Although Timo Kouri17 proposed that as a clinical consequence, higher cut‐off values for positive results must be applied to maintain specificity for the detection of disease, however, the value of urinary α1‐MG in females is far less than 12 mg/L, mostly even less than 5.4 mg/L. So we think 8.6 mg/L maybe a more suitable upper limit for women. Subsequently, for urinary α1‐MG/Cr, the reference range for males aged 31‐60 is 19.8 mg/g, similar to that raised by Lothar Thomas9 for people aged more than 40 years old (19.4 mg/g); but for females aged 31‐60 is 20.5 mg/g, higher than that of males. This result is different from a cross‐sectional study5 performed in Singapore about diabetic patients where the urinary α1‐MG/Cr is higher in men than in women. The difference of the population composition between these two researches (normal vs diabetes) may account for this phenomenon.
In addition to gender and age, there is a weak but significant negative correlation between eGFR and urinary α1‐MG in both univariate and multivariate analysis. This finding further supports the use of urinary proteins as markers of renal tubular injury.19, 20 In addition, the significant correlation of eGFR with urinary α1‐MG/Cr in univariate analysis lost after multivariate analysis, it is different from our previous study in diabetic patients.11 Similarly, the relatively normal kidney function and younger age may interpret this result, too.
Inevitably, there are some shortcomings in this study. First, the type of specimen is morning or random urine, there may be a certain degree of variation; the relatively fewer specimens may also affect the accuracy and reliability of the results. Second, the level of serum α1‐MG was not detected, so the high level of urinary α1‐MG induced by serum protein could not be excluded. Finally, the eGFR of all subjects is not less than 90 mL/min/1.73 m2, and we did not screened for people with high filtration risk. Consequently, there is a risk of renal insufficiency in this population. But other than that, this study also has advantages. We investigate the reference value of urinary α1‐MG and α1‐MG/Cr in Chinese adults for the first time. With the increasing need of clinical applications of urinary α1‐MG, the reference value may play a more and more important role. So the next work of our group is to explore more people to establish a better reference value for use.
In conclusion, we find the value of urinary α1‐MG was sex‐dependent, whereas urinary α1‐MG/Cr was both gender‐ and age‐dependent. Then we use the 95th percentile as reference value, find that of urinary α1‐MG for males was 26.4 mg/L, for females was 8.6 mg/L; of urinary α1‐MG/Cr for males and females aged 20‐30 was 16.9 mg/g, for males aged 31‐60 was 19.8 mg/g, for females aged 31‐60 was 28.5 mg/g. Only 299 normal subjects were enrolled in this study, we need a larger sample size to establish a better reference value for Chinese people to better serve the clinical applications.
Zhang Q, Jiang X, Cui X‐F, Liu R. A study on the biological reference interval of urinary alpha 1‐microglobulin in a group of Chinese people. J Clin Lab Anal. 2018;32:e22305 10.1002/jcla.22305
REFERENCES
- 1. Moresco R. N., Sangoi M. B., De Carvalho J. A. M., et al. Diabetic nephropathy: traditional to proteomic markers. Clin Chim Acta, 2013;421:17‐30. [DOI] [PubMed] [Google Scholar]
- 2. Penders J, Delanghe JR. Alpha 1‐microglobulin: clinical laboratory aspects and applications. Clin Chim Acta. 2004;346:107‐118. [DOI] [PubMed] [Google Scholar]
- 3. Kang J, Liu J, Ding H, et al. Urine alpha1‐microglobulin is a better marker for early tubular dysfunction than beta2‐microglobulin among tenofovir‐exposed human immunodeficiency virus‐infected men who have sex with men. Br J Infect Dis. 2015;19:410‐416. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Ohashi N, Ishigaki S, Kitajima K, et al. The level of urinary α1 microglobulin excretion is a useful marker of peritubular capillaritis in antineutrophil cytoplasmic antibody associated vasculitis. Clin Exp Nephrol. 2014;19:851‐858. [DOI] [PubMed] [Google Scholar]
- 5. Ching‐Ye Hong, Vivian NG, Kenneth Hughes, et al. Urinary α1‐Microglobulin as a marker of nephropathy in type 2 Diabetic Asian subjects in Singapore. Diabetes Care 2003;26:338‐342. [DOI] [PubMed] [Google Scholar]
- 6. Saif A, Soliman N. Urinary α1‐microglobulin and albumin excretion in children and adolescents with type 1 diabetes. J Diabetes. 2017;9:61‐64. [DOI] [PubMed] [Google Scholar]
- 7. Chinese Society of Endocrinology, Chinese Medical Association . Expert consensus on clinical diagnosis of diabetic nephropathy in Chinese adults. Chin J Endocrinol Metab, 2015;31:379‐385. [Google Scholar]
- 8. Klaus Jung, Monika Pergande, Gottfried Schreiber, et al. Reference intervals for a1‐microglobulin in urine. Clinica Chimica Arm, 1992;206:245‐247. [DOI] [PubMed] [Google Scholar]
- 9. Thomas L. Clinical Laboratory Diagnostics: use and Assessment of Clinical Laboratory Results. TH‐Books, Frankfurt am Main; 1998: 362. [Google Scholar]
- 10. Wang H‐B, Yang Q‐H, Jiang X, et al. Tubular proteinuria is the dominant type of proteinuria in an elderly community population in China. Int Urol Nephrol. 2015;47:1541‐1546. [DOI] [PubMed] [Google Scholar]
- 11. Jiang X., Zhang Q., Wang H.‐B., et al. Associations of urinary, glomerular, and tubular markers with the development of diabetic kidney disease in type 2 diabetes patients. J Clin Lab Anal, 2017;00:e22191 10.1002/jcla.22191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Petrica Ligia, Vlad Adrian, Gluhovschi Gheorghe, et al. Proximal tubule dysfunction is associated with podocyte damage biomarkers nephrin and vascular endothelial growth factor in type 2 Diabetes Mellitus patients: a cross‐sectional study. PLoS ONE. 2014;9:8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Xu R, Zhang L, Zhang P, et al. Gender‐specific reference value of urine albumin–creatinine ratio in healthy Chinese adults: results of the Beijing CKD survey. Clin Chim Acta. 2008;398:125‐129. [DOI] [PubMed] [Google Scholar]
- 14. KDOQI . KDOQI clinical practice guidelines and clinical practice recommendations for diabetes and chronic kidney disease. Am J Kidney Dis 2007;2(suppl 2):S12‐S154. [DOI] [PubMed] [Google Scholar]
- 15. Glassock RJ, Oreopoulos D. G. Aging and Chronic Kidney Disease. Nephron Clin Pract, 2011;119(Supp 1):C1. [DOI] [PubMed] [Google Scholar]
- 16. Ye CY, Cui LL. Kidney variation and function evaluation in elderly patients. Chin J Kidney Dis Invest (Electronic Edition), 2015;4:182‐186. [Google Scholar]
- 17. Kouri Timo, Harmoinen Aimo, Laurila Kaija, et al. Reference Intervals for the Markers of Proteinuria with a Standardised Bed‐Rest Collection of Urine. Clin Chem Lab Med. 2001;39:418‐425. [DOI] [PubMed] [Google Scholar]
- 18. Shidham G, Hebert LA. Timed urine collections are not needed to measure urine protein excretion in clinical practice. Am J Kidney Dis. 2006;47:8‐14. [DOI] [PubMed] [Google Scholar]
- 19. Gluhovschi C., Gluhovschi G., Petrica L., et al. Urinary biomarkers in the assessment of early diabetic nephropathy. J Diabetes Res, 2016; 10.1155/2016/4626125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. MacIsaac RJ, Ekinci EI, Jerums G. Markers of and risk factors for the development and progression of diabetic kidney disease. Am J Kidney Dis. 2014;63:S39‐S62. [DOI] [PubMed] [Google Scholar]