Cover Image
close this bookMethods for the Evaluation of the Impact of Food and Nutrition Programmes (UNU, 1984, 287 p.)
close this folder4. Measuring impact using laboratory methodologies
View the document(introduction...)
View the documentIntroduction
View the documentDevelopment of a primary nutritional deficiency
View the documentChoice of tissue for laboratory assessment
View the documentSelection of laboratory methodologies for nutritional impact evaluation
View the documentLaboratory methods for assessment of nutritional impact
View the documentAnnex A. Laboratory evaluation of protein nutriture
View the documentAnnex B. Laboratory evaluation of vitamin A nutriture
View the documentAnnex C. Suggested methods for hematology
View the documentReferences

Annex A. Laboratory evaluation of protein nutriture

1. Determination of Serum Albumin

Serum albumin levels can be determined by standard electrophoresis, salt fractionation, or dye-binding techniques. Although electrophoretic analyses can provide precise, specific quantitative data, for this the technique requires an electrophoresis apparatus and densitometer or colorimeter and a separate determination of total protein. In contrast, the salt fractionation procedure is rather non-specific and suffers from lack of precision. On the other hand, specific dye-binding methods are rapid, involve few manipulations and specialized equipment, and do not require a separate determination of total protein. A widely used method of this type, based on bromcresol green (BCG) (12), is described here.


The addition of albumin to a solution of bromcresol green in a 0.075 M succinate buffer, pH 4.20, results in an increase in absorbance of 628 nm. The absorbance-concentration relationship is linear for samples containing up to 6 g/dl albumin. Bilirubin, moderate lipaemia, and salicylate do not interfere with the analysis. The use of a non-ionic surfactant (Brij-35) reduces the absorbance of the blank, prevents turbidity, and provides linearity. The results with this method agree very well with those obtained by electrophoresis and salt fractionation. The method is simple, it has excellent precision, and the reagents are stable.


Spectrophotometer or colorimeter equipped with 630 nm interference filter.


  • Succinate buffer 0.10 M, pH 4.0: Dissolve 11.9 9 of succinic acid in about 800 ml of water; adjust pH to 4.0 with NaOH and dilute to 1 litre with water. Store at 4° C.
  • Bromcresol green solution: Stock 0.60 mM. Dissolve 419 mg of BCG in 10 ml of 0.1 N NaOH in a 1 litre volumetric flask. Dilute to volume with water and store at 4° C.
  • Working dye solution: Dilute one volume of stock BCG solution with three volumes of 0.10 M succinate buffer, add 4.0 ml 30 per cent Brij-35 per litre, and carefully adjust pH to 4.20 + 0.05. Store at 4° C.
  • Brij-35 solution, 30 per cent. This solution is available from Technicon Instrument Company, Tarrytown, New York. Brij-35 is also obtainable from Fisher Scientific Company.
  • Albumin standard solution: Stock, 10.00 g/dl: Dissolve 10.00 g of HSA-V corrected for moisture content, and 50 mg of sodium azide in water in a 100 ml volumetric flask. Dilute to volume with water and store at 4° C.
  • Albumin standard solutions: Working; 2 0, 3.0, 4.0, 5.0 and 6.0 g/dl Appropriate dilutions of the stock albumin. Standard solutions are made with 50 mg/dl aqueous sodium azide solution. Store at 4° C


A sample of 25 µl of serum or working standard solutions are added to 5.0 ml of working dye solution. The solution is vigorously mixed and allowed to stand 10 minutes at 25° C. The absorbance is measured at 628 nm after adjusting the instrument to zero absorbance with the working dye solution. If a serum sample is extremely lipaemic, a serum blank is prepared by adding 25 µl of the sample to 5.0 ml of 0.075 M succinate buffer. Its absorbance with water as a reference is subtracted from the absorbance of the unknown. The albumin concentration of the serum is obtained from an absorbance-concentration plot, or, if the response of the instruments is linear, only a single standard solution (2.0 g/dl) is required and the serum albumin concentration is calculated in the usual manner. A typical standard curve is seen in figure 4.A.1 (see

FIG. 4.A.1. The Absorbance-Concentration Relationships of the BCG Method).


2. Determination of Serum Prealbumin by Radial Immunodiffusion (RID)

Procedure as Described by Manufacturer:

Although the antibody for human prealbumin can be developed by standard techniques, commercial kits are available from Calbiochem-Behring Corp. (10399 North Torrey Pines Rd., La Jolla, CA 92037, USA) and may be more economical if a limited number of assays are anticipated. Since prealbumin is synthesized in the liver, acute and chronic liver disease will reduce serum concentrations independent of protein nutriture. Total fasting for more than 48 hours also will reduce serum levels by limiting the substrate available for hepatic protein synthesis. Hence, in the absence of liver disease and prolonged absolute fasting, prealbumin is a sensitive parameter for improvement in protein nutritional status.


A protein (antigen) solution is applied to a cylindrical well cut in a gel matrix containing a uniform concentration of monospecific antibodies. Antigen placed in the well diffuses radially, producing a precipitin ring. Precipitin rings can be read any time after overnight incubation, or endpoint. Results are quantitated by comparing the diameter of the precipitin ring produced by the sample to the precipitin rings produced by standards of known concentrations.


  • M-Partigen(tm) prealbumin radial immunodiffusion plates.
  • 3 x 0.5 ml protein Standard Serum B solutions. Protein Standard Serum B, a stabilized pooled human serum, is supplied in three prediluted solutions, each containing amounts of prealbumin as indicated on the label, for construction of the reference curve.


Plate preparation: Carefully remove the plate from the container. Open the plate by pressing the thumbs firmly on the centre of the lid while holding the lid at edges. Allow the plate to stand open at room temperature to permit evaporation of any moisture that may have condensed in the wells.

The user can choose, according to workload requirements, to make determinations based either on an overnight readout from a reference curve or on an endpoint readout (after 48 hours). The initial steps of the procedure are the same for either method.


  • Overnight Method (after 18 hours of diffusion time): Measure the diameter (d) of the precipitin rings to 0.1 mm accuracy. Plot the diameter of the precipitin rings obtained from the Protein Standard Serum B solutions (abscissa) against their respective concentrations (ordinate) on semilog graph paper and draw a straight line. (See

    FIG. 4.A.2. Examples of Reference Curves. A: Overnight method-using semilog graph paper. B: Endpoint methodusing linear graph paper.). Values for the test samples are then determined from this reference curve. If the precipitin ring diameter (d) of the test sample exceeds that of the most concentrated standard, reassay the sample at a higher dilution; the reference curve cannot be extrapolated.
  • Endpoint Method (after 48 hours of diffusion time): Measure the diameter (d) of the precipitin rings to 0.1 mm accuracy, and calculate d2 (mm2). Plot the squared diameter (d2) of the precipitin rings obtained from the Protein Standard Serum B solution (abscissa) against their respective concentrations (abscissa) on linear graph paper. A straight line is drawn (which should intercept the ordinate at 11 + 3.5 mm2). Values for the test samples are then determined from this reference curve.

The assay range of M-Partigen(tm) prealbumin radial immunodiffusion plates is indicated on the label. If the sample concentration exceeds the upper limit of the assay range, the test should be repeated with appropriate dilutions of the specimen.

If the reference curve is not linear, or does not intercept the ordinate at 11 + 3.5 mm2 when the endpoint curve is plotted, the procedure should be repeated; errors in technique or product instability should be considered.