Find best premium and Free Joomla templates at GetJoomlaTemplatesFree.com

β2-Microglobulin – reappraisal of an alternative to creatinine in assessment of kidney function in children with urinary tract malformations

Kolja Kvist¹, Henrik Gutte², Jorgen Thorup¹,³
¹Department of Paediatric Surgery , Rigshospitalet, Copenhagen, Denmark
²Department of Clinical Physiology, Nuclear Medicine and PET, Rigshospitalet, Copenhagen, Denmark.
³Department of Paediatric Surgery, Rigshospitalet and Faculty of Health Science, University of Copenhagen, Denmark

 

Correspondence 

Prof Jorgen Thorup
Department of Paediatric Surgery
4072, Rigshospitalet
DK-2100 Copenhagen, Denmark
E-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.  

 

Abstract

Objective: To evaluate serum β -microglobulin as an alternative estimator of the glomerular filtration rate (GFR) in children.

Patients and methods: 111 children between the age of 1 and 178 months (mean 43 months) with congenital urinary tract abnormalities were included. They were divided into 3 age-groups, each analysed separately. 51Cr-ethylenediaminetetraacetic acid (51Cr-EDTA) plasma clearance and simultaneous measurements of serum creatinine and serum β 2 -microglobulin were done. 51Cr-EDTA plasma clearance served as gold standard. The results were analysed using receiver operating curves (ROCcurves), kappa statistics, and McNermars β test, as well as sensitivity, specificity, and positive and negative predictive value.

Results: Serum β 2 2 -microglobulin was more sensitive (0.75 – 1.00) than serum creatinine (0.11–0.47), and the difference became more attenuated the younger the child. Also the kappa values and ROC-curves indicated a closer approximation between β 2 -microglobulin and 51Cr-EDTA. However, serum creatinine had higher specificity, and McNermars β test confirmed that the observed differences were significant.

Conclusion: β -microglobulin is a better estimator of GFR in children than serum creatinine, and the difference is enhanced the younger the child. The higher sensitivity and negative predictive value makes it useful as a screening parameter, since a normal value with high accuracy attests a normal GFR. 

 

 

Introduction

Antenatally diagnosed urinary tract abnormalities are seen in up to 4% of foetuses. Further evaluation after birth often warrants assessment of glomerular filtration rate (GFR). Plasma clearance measurements with radio-GFR). Plasma clearance measurements with radioisotopes ( 51 Crom-ethylenediaminetetraacetic acid ( CrEDTA), I-iothalamat, or technetium-99m diethyltriaminepentaaceticacid ( 125 Tc-DTPA)) are the methods of choice. However the complexity of the invasive procedures involved as well as the – albeit minimal – radiation exposure are undesirable, and a simpler and less toxic screening parameter would be preferable.

Ideally GFR could be determined by measuring the serum concentration of an endogenously produced marker. This marker should be produced at a constant rate, be freely filtered across the glomeruli, be neither secreted nor reabsorbed by the tubuli, and not be subject to extra-renal breakdown. Traditionally the parameter of choice has been serum creatinine (crea). Unfortunately the use hereof in paediatric cases is particularly problematic, because the concentration is affected by factors other than GFR. These include tubular secretion, extra-renal breakdown, and dependence on muscle-mass [1 – 4].

Table 1: Creatinine and B2M vs. GFR measured by EDTA, all ages

Creatinine \ GFR

SB2M \ n = 111

Severely reduced

Moderately reduced

Mildly reduced

Normal

Creatinine

High

5

5

7

5

Normal

1

0

18

70

SB2M

> 215

5

5

17

17

< 215

1

0

8

58


Consequently the search for a marker less afflicted by these factors continues, and several low-molecular-weight proteins have been investigated, among them B2M.

It represents the light chain of the MHC I complex, and as such is found on all nucleated cells. It was first isolated and described by Berggård and Bearn in 1968 [5]. It consists of 99 amino acids and has a molecular weight of 11,815 Da [5, 6]. Unlike its alpha-counterpart, it is not subject to degradation in plasma [6, 7]. Rather it is freely filtered across the glomeruli, whereupon it is completely reabsorbed and metabolized in the proximal tubules [6, 7], indicating an inverse relationship between B2M and GFR. This, and the fact that the concentration is age and gender independent [5 – 8], suggest that it may be a more sensitive marker of GFR than crea, especially regarding children. 

Patients and methods

 The design was a retrospective cross-sectional study. The medical records of children treated in the Department of Paediatric Surgery Rigshospitalet, Denmark, during a five year period were reviewed. A total of 111 children, who had simultaneously undergone Cr-EDTA clearance measurements as well as determination of crea and B2M, were identified and included. The cohort consisted of 76 boys and 35 girls, ages 1 – 178 months (mean 43 months), with congenital urinary tract malformations – the two most common being congenital hydronephrosis and posterior urethral valves, each present in 25 cases – necessitating evaluation of GFR. The children were divided into 3 age-groups (all ages, ≤ 24 months, and ≤ 6 months), which were analyzed separately. The method used for measuring the GFR in children is determined from a one-plasma sample. Following the intravenous injection of 2-3 MBq Cr-EDTA, blood were drawn at 90-120 minutes. Plasma samples, standards and blank tubes for background were then assayed together 51 in a gamma counter, and sample counts were compared with standards to calculate the plasma concentration of Cr-EDTA. A single-exponential curve was fitted to the plasma data by fitting a straight line to the plot of the log concentration of 51 Cr-EDTA against time. From this line, the half-life T 1/2 51 and the volume of distribution V were determined, and the slope–intercept GFR was calculated. The distribution space was estimated from the body surface area [9, 10].

51Cr-EDTA plasma clearance served as gold standard for GFR and was designated normal, mildly, moderately, or severely decreased after age- and surface-correction was performed. 51 Crea was measured according to the Jaffé method on a Hitachi 917 (Boehringer Mannheim, Germany) with reagents from Boehringer Mannheim, Germany, and designated high or normal for that age according to the following reference-intervals:

0 - < 1 year: 0.020 – 0.050 mmol/l
1  - < 12 years: 0.035 – 0.075 mmol/l
12  - < 16 years: 0.040 – 0.090 mmol/l

B2M was measured on an AutoDELFIATH (Wallac, OY) using solid phase multisite fluorometry. The result was designated as either normal or abnormal, according to the upper reference value derived from the receiver operating curves (ROC-curves) (≥215 nmol/l).

Table 2: Creatinine and B2M vs. GFR measured by EDTA, age ≤ 24 months

Creatinine \ GFR

SB2M \ n = 62

Severely reduced

Moderately reduced

Mildly reduced

Normal

Creatinine

High

2

2

2

3

Normal

0

0

12

41

SB2M

> 215

2

2

13

16

< 215

0

0

1

28


Table 3: Creatinine and B2M vs. GFR measured by EDTA, age ≤ 6 months

 

Creatinine \ GFR

SB2M \ n = 28

Severely reduced

Moderately reduced

Mildly reduced

Normal

Creatinine

High

1

0

0

3

Normal

0

0

8

16

SB2M

> 215

1

0

8

8

< 215

0

0

0

11


Both crea and B2M was compared to 51Cr-EDTA, and level of agreement evaluated using kappa statistics (k), sensitivity (sens), specificity (spec), positive predictive value (pp), and negative predictive value (np). B2M and crea was compared using ROC-curves and McNemars χ2 test. All statistical computations were done using windows based SPSS 11.0.

 4 JPSS 8 1 2010-4-1  4 JPSS 8 1 2010-4-2  4 JPSS 8 1 2010-4-3
Figure 1: ROC Curve for Creatinine and b2m, all ages Figure 2: ROC Curve for Creatinine and b2m, age ≤ 24 months Figure 3: ROC Curve for Creatinine and b2m, age ≤ 6 months

 

Results

The results of comparing GFR according to crea or B2M versus 51Cr-EDTA are shown in tables 1 – 3. The ROC-curves from which the upper normal cut-off for B2M was derived is illustrated in fig. 1 – 3. The same figures also hints, that B2M is a better estimator of GFR than crea, since the area under the curves reflect the goodness of agreement between the two parameters and 51Cr-EDTA (table 4). Table 5 summarises the results of the statistical computations. As is evident B2M is a more sensitive marker of GFR than crea, and the difference becomes more attenuated the younger the child. The calculated McNermars χ2-test (table 6) confirms that the observed difference is significant. The kappa-values reflect that both parameters show fair to moderate agreement with 51Cr-EDTA, except for the youngest age-group, where the likelihood of obtaining the result seen for crea is no better than guessing.

Table 4: areas under the ROC-curves

Fig 1

Fig 2

Fig 3

Creatinine

802

787

708

SB2M

814

797

778


Table 5: Sensitivity (sens.), specificity (spec.), positive predictive value (pp.), negative predictive value (np.), and kappa (κ)

 

SENS.

SPEC.

PP

NP

K

Age≤ 178 m n =111

Creatinine

0.47

0.93

0.77

0.79

0.45

B2M

0.75

0.77

0.61

0.87

0.50

Age ≤ 24 m n = 62

Creatinine

0.33

0.93

0.67

0.77

0.31

B2M

0.94

0.64

0.52

0.97

0.47

Age ≤ 6 m

n = 28

Creatinine

0.11

0.84

0.25

0.67

-0.06

B2M

1.0

0.58

0.53

1.00

0.47


Table 6: McNernars ß² test

 

Age \ χ² (GFR)

Group \

χ² (normal)

 

χ² (reduced)

 

Age ≤ 178 m

n = 111

7.56

5.72

Age ≤ 24 m

N = 62

11.08

9.09

Values above 3.82 correlate to p < 0.05. The last group (age≤ 6 months) was too small for the test.

Discussion

The present study was motivated by the distressing insufficiency of serum creatinine to reliably distinguish between normal and mildly decreased GFR in children, prompting the need for plasma clearance measurements using radioisotopes [3]. This insufficiency of crea is in part due to extra-renal breakdown and tubular secretion, both of which mask a decline in GFR – and both of which increases as GFR declines [1 – 4]. Also its dependence on gender, age and dietary intake – even the manner in which meat is cooked – further confounds it as an accurate estimator of renal function [1 – 4]. Neonatally the use of crea is hampered by its ability to cross the placenta, rendering the concentration shortly after birth a reflection of the maternal kidney function, rather than the child’s.

B2M is not confounded by these factors. However increased concentrations are seen in conjunction with febrile episodes [12] – particularly when caused by lymphocytic viruses – although B2M was still more sensitive than crea in one study aiming to detect acute kidney injury in critically ill children of whom most were admitted due to infections [13]. Certain haematological diseases such as leukaemia and myelomatoses, as well as some autoimmune disorders like SLE, Rheumatic Arthritis and Sjögrens syndrome also stimulate the production [6, 7]. In such cases the concentration of B2M does not reflect the GFR, but for the most part these diseases are recognizable, and – apart from the viral infections and leukaemia – uncommon in the paediatric population.

Our results are in accordance with other investigations showing B2M’s superiority over crea [11, 13 – 18], although some have found that the height/creatinine ratio (e.g., Schwartz formula) performs equally well or better [11, 17, 18]. Most studies have been performed on adults and there are differences in design and choice of statistical parameters. Most – among them three included here [15, 16] - have relied on linear regression analysis and coefficients of correlation. Hereby the inverse relationship between B2M and GFR has been documented, which is why we chose not to repeat this.

Common to those stating similar statistical parameters is a higher specificity (0.95 – 0.98) [11, 14, 16, 17], two also have a higher sensitivity (0.94) [14, 16], and one had lower sensitivity (0.59) [17]. These differences could be due to a difference in defining a decreased GFR. In our study the 51Cr-EDTA measurements were corrected for body-surface area as well as age, allowing for the maturation GFR undergoes during the first years.

Another intriguing explanation for the lower specificity for B2M seen in our study could be an overestimation of GFR by 51Cr-EDTA. This overestimation is partly due to the slope clearance method itself independently of marker and partly to extra-renal breakdown of 51Cr-EDTA specifically [2]. These two factors could account for the number of false positive results seen with B2M, thus making it a more precise indicator of a reduced GFR than 51Cr-EDTA. As we used MAG-3 renography for the patients in the present study we have no data on DMSA/DTPA renal scans, that might have shown if both kidneys were abnormal when 51Cr-EDTA clearance was normal, supporting the hypothesis of renal hyperfiltration. MAG-3 renography can not specify if both kidneys were abnormal in cases where 51Cr-EDTA clearance was normal but B2M was elevated.

In the present study we have used a simple plasma sample technique obtained at 90 – 120 minutes following the injection of 51Cr-EDTA. In young children this technique is not as accurate as a multiple sample method and a minimum of 2 samples post injection gives more reliable results particularly in children. However, two times blood sampling in children is usually cumbersome for the patient, parents and laboratory technicians.

Owing to the shortcomings, albeit recognizable, on behalf of B2M, other low molecular weight proteins have been investigated (ex. Beta-Trace Protein, Alpha1-microglobulin, and Cystatin C). Of these Cystatin C is the most extensively studied and perhaps the most promising [8, 13, 17, 18 – 20]. It is produced by all nucleated cells, comprised of 120 aminoacids, and has a mass of 13,343 Da. Similar to B2M the production is gender and age independent, and the serum concentration is inversely correlated to GFR. To date the only factors known to affect the concentration regardless of kidney function is thyroid dysfunction, and possibly pregnancy, corticosteroid therapy and cyclosporine treatment [19].

Most studies comparing B2M and Cystatin C in children show the latter to be marginally better [8, 13, 17, 18]. However, one drawback to Cystatin C is that its sole use thus far is the evaluation of kidney function. It has not been related to any known pathological disease states [19]. This is contrasted by the widespread use of B2M in oncology and autoimmune disorders, making it a routine measurement in most laboratories.

All of the children in the present study had their GFR evaluated by 51Cr-EDTA due to clinical circumstances. Some of these evaluations could be avoided if there was a sufficiently sensitive endogenously produced marker with a high negative predictive index. In this regard specificity is not as important. The above results show that B2M has these qualities. We therefore recommend it as a screening parameter in patients with congenital urinary tract abnormalities, because a normal concentration with high accuracy indicates a normal GFR. Especially in the very young, where serum creatinine is unreliable and minimal testing is desirable

 

 

 

References

  1. Perrone RD, Madias NE, Levey AS (1992) Serum creatinine as an index of renal function: New insights into old concepts. Clin Chem 38: 1933 – 1953
  2. Rahn KH, Heidenrieich S, Brückner D (1999) How to assess glomerular function and damage in humans. J Hypertens 17: 309 – 317
  3. Skinner R, Cole M, Pearson ADJ, Keir MJ, Price L, Wyllie RA (1994) Inaccuracy of glomerular filtration rate estimation from height/plasma creatinine ratio. Arch Dis Child 70: 387 – 390
  4. Wetzels JFM, Huysmans FThM, Koene RAP (1988) Creatinine, a marker of glomerular filtration rate. Neth J Med 33: 144 – 153
  5. Berggard I, Bearn AG (1968) Isolation and properties of a low molecular weight b2-globulin occurring in human biological fluids. J Biol Chem 243: 4095 – 4113
  6. Revillard JP, Vincent C (1988) Structure and metabolism of beta-2-microgobulin. Contr Nephrol 62: 44 – 53
  7. Schadijn GHC, Statius van ELW (1987) b2-Microglobulin: Its significance in the evaluation of renal function. Kidney Intl 32: 635 – 641
  8. Filler G, Witt I, Priem F, Ehrich JHH, Jung K (1997) Are Cystatin C and b2-Microglobulin better markers than serum creatinine for prediction of a normal glomerular filtration rate in pediatric subjects? Clin Chem 43: 1077 – 1078
  9. Rehling M, Moller ML, Thamdrup B, Lund JO, Trap-Jensen J (1984) Simultaneous measurement of renal clearance and plasma clearance of 99mTc-labelled diethylenetriaminepentaacetate, 51Cr-labelled ethylenediaminetetraacetate and inulin in man. Clin Sci 66: 613 – 619
  10. Fjeldborg P, Brochner-Mortensen J (1986) Determination of 51Cr-EDTA clearance in infants by a single capillary blood sample. Scand J Clin Invest 46: 335 – 340
  11. Parkin A, Smith HC, Brocklebank JT (1989) Which routine test for kidney function. Arch Dis Child 64: 1261 – 1263
  12. Donaldson MDC, Chambers RE, Woolridge MW, Whicher JT (1990) Alpha1-microglobulin, beta2-microglobulin and retinol binding protein in childhood febrile illness and renal disease. Pediatr Nephrol 4: 314 – 318
  13. Herrero-Morin JD, Malaga S, Fernandez N, Rey C, Dieguez MA, Solis G (2007) Cystatin C and beta2-microglobulin: markers of glomerular filtration in critically ill children. Critical Care 11: 1 – 7
  14. Braren V, Goddard J, Brill AB, Touya JJ (1979) Beta-2-microglobulin as a renal diagnostic agent. Urology 13: 624 – 628
  15. Van Acker KJ, Vlietinck RF, Steels PM (1984) Estimation of glomerular filtration rate from b2-microglobulin serum levels in children. Int J Pediatr Nephrol 5: 59 – 62
  16. Jung K, Schultze BD, Sydow K, Pergande M, Precht K, Schreiber G (1987) Diagnostic value of low-molecular mass proteins in serum for the detection of reduced glomerular filtration rate. J Clin Chem Clin Biochem 25: 499 – 503
  17. Filler G, Priem F, Vollmer I, Gellermann J, Jung K (1999) Diagnostic sensitivity of serum cystatin for impaired glomerular filtration rate. Pediatr Nephrol 13: 501 – 505
  18. Filler G, Priem F, Lepage N, Sinha P, Vollmer I, Clark H (2002) b-Trace Protein, Cystatin C, b2-Microgloulin, and Creatinine Compared for Detecting Impaired Glomerular Filtration Rates in Children. Clin Chem 48: 729 – 736
  19. Randers E, Kristensen JH, Erlandsen EJ, Danielsen H (1998) Serum cystatinC as a marker of the renal function. Scand J Clin Lab Invest 58: 585 – 592
  20. Corrao AM, Lisi G, Di Pasqua G, Guizzardi M, Marino N, Ballone E, Lelli Chiesa P (2006) Serum Cystatin C as a reliable marker of changes in glomerular filtration rate in children with urinary tract malformations. J Urol 175: 303 – 309