Albumin Methodology Change

From Tuesday 6 March the laboratories supporting Sydney Local Health District at Concord, Canterbury and Royal Prince Alfred Hospitals, and in South West Sydney Local Health District at Liverpool, Bankstown, Campbelltown, Fairfield and Bowral Hospitals, will be changing the methodology used in measuring albumin. This is part of a national (and global) effort to harmonise assays and reference intervals to bring greater consistency between laboratories.

The new method is more specific, resulting in a more accurate measure of albumin

Results will be on average lower by 4-5g/L. Greater differences may be seen in patients with low (<30g/L) albumin levels. (e.g. on average 6-8g/L for dialysis and ICU patients).

A new reference interval for serum/plasma albumin will apply.

A note will be appended to albumin results regarding new method and reference intervals.

A new formula will be applied for the calculation of corrected (adjusted) calcium.  Corrected calcium results should not change on average.

This change has already occurred across NSW Health Pathology laboratories, together with other centres in Australia and globally.

Serum / Plasma Reference Intervals



Reference Interval



0day to <4weeks




4weeks to <110y







For more information, contact the Biochemistry laboratory at Royal Prince Alfred Hospital, p. 9515 5244, Concord Hospital p. 9767 6663 or Liverpool Hospital p. 8738 5088

Current status of albumin measurement

Serum/Plasma Albumin is currently measured using a dye-binding (colorimetic) method. Traditionally, our laboratories have been using Bromcresol green (BCG) as the dye. We are now changing methodology and moving to the internationally used, more specific and accurate measure of albumin using the dye, Bromcresol purple (BCP).

Differences (bias) between BCP and BCG

Studies of BCG suggest an overestimation of albumin concentration with a mean positive bias when compared with ‘gold standard’ (immunochemical) methods (1)

  • Studies comparing BCG and BCP have demonstrated the relationship AlbBCG=5.5+AlbBCP performed well in both renal patients and unselected outpatients (2)
  • When comparing the BCP and BCG methods, Canterbury Health Laboratories, New Zealand (Appendix 1) and Royal North Shore Hospital (Appendix 2) both reported  4-5 g/L bias at higher albumin values and greater bias (6-8 g/L)  at lower albumin levels values
  • This trend of increased bias at lower values is also observed in the regression lines of Royal Prince Alfred Hospital (RPAH), Concord Hospital, Liverpool Hospital and Fairfield Hospital. {(RPA: y = -6.78+1.03x (Deming); Concord: y = -5.353 + 1.027x (Passing-Bablock); Liverpool y = -5.81+1.04x (Passing-Bablock); Fairfield y = - 5.33 +1.07x (Passing-Bablock)}.
  • RPA observed an average bias of -5.5 g/L between BCG and BCP methods on the Roche Cobas e602 analyser.
  • When RPA Quality Assurance Program (QAP) results using BCG citrate method were compared to median results for the laboratories using the BCP method in the RCPAQAP  (Cycle 106 – 2017), there was an average bias of -5.1 g/L.
  • This is consistent within confidence limits of the average bias observed at RPA with their internal regression analysis (-5.5 g/L)
Clinical significance


BCP results will be on average lower by 4-5 g/L and a new reference interval will be implemented in line with these changes.

Corrected (adjusted) calcium results

Corrected (adjusted) calcium concentrations are calculated to compensate for albumin concentration, allowing a single reference interval to be used irrespective of albumin level

A minor adjustment to the formula for adjusted calcium will be made on implementation of the new albumin assay. The new adjusted calcium formula will be: Adjusted Ca = calcium + ((40-BCP albumin) * 0.02. (3)


Globulin is a calculated value (Globulin = total protein – serum albumin).

With lower albumin results, the calculated globulin result will be higher than previously.

This will not affect the values of measured immunoglobulins (IgG, IgA, IgM etc) as these are measured separately.

Renal Failure

BCP can underestimate albumin in a small proportion of patients due to the presence of a uraemic toxin. (4-6)

In general, the overestimation of the true serum albumin using the current method (bromocresol green (BCG)) is a more significant problem and bromcresol purple (BCP) albumin results are still more accurate overall in this patient group; especially in the low range of most clinical importance. (7,8)

Multiple myeloma staging

The International Staging System and Revised International Staging System for multiple myeloma do not specify a preferred method for albumin determination.  (9,10)

Amongst other parameters, different stages are based upon the β2 microglobulin levels and  an albumin level (cut off of 35g/L).

Australasian Working Party on Standardized Reporting of Protein Electrophoresis recommends using BCP (or capillary zone electrophoresis) to measure serum albumin but BCG is still acceptable. (11)


In the initial Child’s classification from 50 years ago, serum albumin was measured by a colour reaction that shows poor specificity for albumin. The reference interval at the time was 46 - 67 g/L, which is higher than that used at present, (12,13) and the cut off (35 g/L ) was considered significantly low at the time.

It is recognised that some clinicians prefer the alternative MELD score, which does not include serum albumin.

Other body fluids

Measurement of albumin in urine (albumin:creatinine ratio) and in CSF will still be performed using alternate methods (e.g. immunochemical) and will not change.

In other fluids (e.g. ascitic fluid), lower albumin levels may be seen with the new BCP method.  The difference should be small as these fluids would contain less interfering proteins and early studies have suggested minimal different between the BCP and BCG methods in ascitic fluid. (14)


Duly EB, Grimason S, Grimason P, Barnes G, Trinick TR. Measurement of serum albumin by capillary zone electrophoresis, bromocresol green, bromocresol purple, and immunoassay methods. J Clin Pathol 2003;56:780-2

Clase CM, St Pierre MW, Churchill DN. Conversion between bromcresol green- and bromcresol purple-measured albumin in renal disease. Nephrology Dialysis Transplantation, 16 (9) 2001,  1925-9

 Tietz Textbook of Clinical Chemistry and Molecular Diagnostics, 6th Edition, Elsevier,  (2017). Edited by Nader Rifai ; senior editors, Andrea R. Horvath, Carl T. Wittwer, p1433

Mabuchi H, Nakahashi H. Underestimation of serum albumin by the bromcresol purple method and a major endogenous ligand in uremia. Clin Chim Acta 1987;167:89-96.

Beyer C, Boekhout M, van Iperen H. Bromcresol purple dye-binding and immunoturbidimetry for albumin measurement in plasma or serum of patients with renal failure. Clin Chem 1994;40:844-5.

Parikh C, Yalavarthy R, Gurevich A, Robinson A, Teitelbaum I. Discrepancies in serum albumin measurements vary by dialysis modality. Ren Fail 2003;25:787-96.

Michael Peak BR. Are routine serum albumin assays suitable for haemodialysis patients? In: AACB Annual Scientific Conference; 2008; 2008.

Carfray A, Patel K, Whitaker P, Garrick P, Griffiths GJ, Warwick GL. Albumin as an outcome measure in haemodialysis in patients: the effect of variation in assay method. Nephrol Dial Transplant 2000;15:1819-22.

Greipp PR, San Miguel J, Durie BG, et al. International staging system for multiple myeloma. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 2005;23:3412-20.

Palumbo A, Avet-Loiseau H, Oliva S, et al. Revised International Staging System for Multiple Myeloma: A Report From International Myeloma Working Group. Journal of Clinical Oncology. 2015;33(26):2863-2869.

Tate J, Caldwell G, Daly J, et al. Recommendations for standardized reporting of protein electrophoresis in Australia and New Zealand. Ann Clin Biochem 2012;49:242-56.

Child CG, Turcotte JG. Surgery and portal hypertension. Major problems in clinical surgery 1964;1:1-85.

Rodkey FL. Direct Spectrophotometric Determination of Albumin in Human Serum. Clin Chem 1965;11:478- 87.

Thompson S, Ding L.T he Bromocresol Purple Method of Albumin Measurement Significantly Underestimates the Serum Ascites Albumin Gradient , RCPA Update Pathology 2-4th March, Sydney, 2018