Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

9. Some rat and human bioassay procedures

Standardization of Animal Assays
Method for Determination of Protein Efficiency Ratio and Net Protein Ratio*
A Procedure for Determination of Net Protein Utilization Using Rats - Body Nitrogen Technique*
Nitrogen Growth Index in Rats
Procedure for the Estimation of Relative Protein Value in Rats
Procedure for the Estimation of Relative Net Protein Ratio in Rats
Human Studies

The advantages, disadvantages, potential role, and current status of a whole series of bioassay procedures for protein quality have been discussed in chapters 4 and 5. In chapter 6 some of the criteria involved for the selection of the most suitable procedure for particular purposes were also discussed. Within this chapter many procedures are described, not only as to how they should be performed but also how the results obtained should be calculated and presented. The techniques described range from the determination of PER, a widely used, simple, but limited assay procedure, to human multi-level feeding trials of considerable expense and complexity. Also included are many assays of moderate complexity such as NPU, both by carcass analysis and nitrogen balance, RNPR, and the rat slope-ratio procedures NGI and RPV. As in chapter 8, inclusion of techniques does not necessarily imply endorsement of these proceudres but, again, a desire to make available detailed methodology for a wide range of techniques for workers in the field.

In the discussion of clinical methods in chapter 5 detailed consideration was given to the precautions necessary in any human study. This is essential reading in conjunction with the more detailed procedures listed later in this chapter. Some general precautions necessary for rat bioassays follow.

Standardization of Animal Assays

To be able to detect small differences in protein quality, a strict standardization of the experimental procedures is required.

The climatic conditions are very important. Room temperature between 22 and 24 C can be recommended for rats. A relative humidity between 50 and 65 per cent seems to give optimal conditions. Furthermore, the air shift should be adjusted to prevent odour.

The weight, strain, and sex of the experimental animals should be standardized. Futhermore, healthy animals are a prerequisite. These can only be obrained from a well nourished breeding colony under veterinary supervision.

Male rats seem to be the sex preferred, and Wistar and Sprague Dawley are the strains most frequently used. The initial weight of the experimental animals is of importance in some assays, and a body weight of 64-68 9 at the beginning of the experiment can be recommended for all of the assay procedures discussed.

In all animal assays an accurate measurement of feed consumption is required. Animals are not always cooperative and this can lead to considerable spillage problems. Nevertheless, these problems cannot be ignored or food losses neglected. Accurate determination of food spillage and separation of food from faecal and urinary contamination is a tedious but essential part of the assay. Therefore, care must be taken when constructing cages and diet containers to minimize this problem. A heavy metal net placed on top of the diet can successfully be used. Separation of urine and faeces is another problem for balance procedures. However, a nylon net placed at an angle of approximately 60 underneath metabolism cages seems to solve this problem (1).

Method for Determination of Protein Efficiency Ratio and Net Protein Ratio*

Animals. Use weanling male rats of a single strain, 20-23 days of age, ten for each diet.

Diets. Use a basal diet of the following composition on an air-dried basis: cornstarch, 80 per cent; corn oil or cottonseed oil, 10 per cent; non-nutritive cellulose, 5 per cent; salts, USP XIV, 4 per cent; and vitamin mixture, 1 per cent. Incorporate the protein food under test into the diet at the expense of cornstarch to give 10 per cent (9.710.3) protein (N x 6.25). Prepare a vitamin supplement containing in 1 9 the following amounts of vitamins: vitamin A, 1,000 IU; vitamin D, 100 IU; vitamin E, 10 IU; vitamin K (menadione), 0.5 mg; thiamine, 0.5 mg; riboflavin, 1.0 mg; pyridoxine, 0.4 mg; pantothenic acid, 4.0 mg; niacin, 4.0 mg; choline, 200 mg; inositol, 25 mg; para-aminobenzoic acid, 10 mg; vitamin B1 2, 2 microgrammes biotin, 0.02 mg; and folic acid, 0.2 mg. Add sufficient cellulose to make 1 9. Supply the diet and water ad libitum.

Assay period. Use a four-week period.

Cages. Use individual cages provided with feeders that will reduce food spillage to a minimum.

Randomization. Use a randomized block design in which blocks represent variations in initial weight. Randomize the rats in each block for diet and cage. If variation between litters is significant, use design to permit removal of this variation.

Records. Maintain a weekly and ten-day record of food consumption and body weight.

For PER determination. In addition to the test group, maintain a reference standard group of rats on a diet consisting of the basal ration with casein at the level of 10 per cent protein. The test diet high nitrogen casein approved by the Animal Nutrition Research Council (ANRC) and prepared by the Sheffield Chemical Company, Norwich, New York, USA, has been found to be satisfactory.

At four weeks, calculate the PE R for each food and for the reference standard casein as:

PER gain of test animal protein consumed

If a correction to casein = 2.5 is to be made, proceed as follows:

Corrected PER = PER x 2.5 determined PER for reference standard casein

For NPR determination. In addition to the test group, maintain a control group of rats, matched with the test animals with respect to weights, on a diet consisting of the unmodified basal ration. At ten days, calculate the NPR for each food as follows:

NPR = gain of test animal + loss of control animal protein consumed by test animal

Note: It is important that weights of animals in test and control groups be matched to within 1 9 for the NPR determination. See also the procedure for RNPR (p. 112).

A Procedure for Determination of Net Protein Utilization Using Rats - Body Nitrogen Technique*

Wean four litters of eight hooded ratsa at 23 days; introduce into the experimental room, and feed on a stock diet for one week. The temperature of the animal room should be 28 +1 C b The animals should be 30 +1 days at the commencement of the trial. The range of rat weights within a litter should not exceed 5 9, and the range of the mean weights of the litter should also not exceed 5 9.

Divide the four litters into eight groups in such a way that each group contains one rat from each litter, and the range of the total weights of the groups does not exceed 2 9; ignore sex differences.C Place each group of four rats in a single cage d with a mesh bottom (approximately 1 cm square). Underneath each cage place a metal tray (3 cm larger than the cage) with a layer of sawdust 2 cm deep covered with a sheet of filter paper. Feed one of the groups a non-protein diet (see below) and the remaining seven groups the test diets. Water and food should be given ad libibim; the dry powdered diets sould be dispensed from weighed tins, and presented in pots designed to minimize spillage. Record daily the total weights of each group.

At the end of ten days, kill the animals with chloroform, make incisions into the skull, thoracic, and body cavities, and lay each group in a tray 20 x 20 x 4 cm) of known weight. Weigh before and after drying in a hot-air oven at 105 C for 48 hours; calculate body water and dry weight. Run the dried carcasses three times through a domestic mincer, and retain for nitrogen determination. Brush the spilt food and faeces from the filter paper and separate, using a 10-mesh sieve (2.5 mm). Dry the spilt food together with the uneaten food left in the pot at 105C. Calculate the dry food intake from the loss of weight of the food tin multiplied by the percentage dry weight of the diet, minus the dried spilt and uneaten food.

Determine the nitrogen contents of foods and carcasses by macro-digestion with 10 ml concentrated H2SO4/g dry matter plus an extra 10 ml, and 20 9 of a mixture of 1 part SeO2, 1 part CuSO4. 5HzO, and 8 parts K2SO4; the samples in duplicate should be 1-2 9 for foods and 2-3 9 for carcasses. Maintain the digestion mixture just below boiling point two hours after clearing: cool and make up to an appropriate volume for the estimation of ammonia.. Calculate the nitrogen intake of each group (I) and their total carcass nitrogen (B), including the non-protein group (IK and BK). and hence NPUh as

B - BK + IK 100

All test diets should be assayed in duplicate with a different set of rats at different times. If the duplicates are not within four NPU units, a third replicate should be carried out. All assay data should be inspected to ensure that the caloric intake has been adequate and that the group weight curves have been smooth and rectilinear; outlying results may be rejected for either of these two reasons.

The non-protein diet is composed of 15 per cent cooking fat, 10 per cent potato starch,, 15 per cent glucose, 5 per cent vitaminized carbohydrate (see below), 5 per cent salt mixture (see below), and 50 per cent maize (corn) starch. The vitaminized carbohydrate,/ containing 0.06 9 thiamine hydrochloride, 1.2 9 calcium pantothenate, 4.0 9 nicotinic acid, 4.0 9 of inositol, 12.0 9 para-aminobenzoic acid, 0.04 9 biotin, 0.04 9 folic acid, 0.001 9 cyanocobalamin, and 12.0 9 choline chloride, is made up to 1 kg with maize (corn) starch. The salt mixturem is made by mixing 60 per cent Ca3(PO4)2, 25 per cent NaCI and 15 per cent KCI; to this mixture is added 2 per cent of minor salts consisting of 30 per cent iron citrate 3H20, 30 per cent magnesium carbonate levis, 30 per cent MnCI24H2O, 7 per cent basic copper carbonate, 3 per cent ZnC03, 0.1 per cent NalO3, and 0.1 per cent Na F. The complete non-protein diet should contain less than 0.1 per cent N.

Notes a. Strain of rat: In our experience [i.e., Miller, in ref. 2] three different strains of rats gave similar results. b. Temperature: This is within the range of thermal neutrality for the rat. At lower temperatures, rats may burn protein for energy purposes, and lower values for NPU would be obtained. c. Sex differences: There is no significant difference between values obtained using all males and those using all females. d. Rats per cage: Animals housed individually tend to eat more food, but the values obtained for NPU are unaffected. e. Food pots: Various designs are available, but they should be suitable for feeding dry powders, and the caloric intake of the animals should not be restricted. f Intestinal contents: Calculations including and excluding the contents of the gut give substantially the same results. g. N/HzO ratio: In the method described by Miller and Bender [3, 4], the nitrogen content of the carcasses is calculated from their water content knowing the N/H20 ratio at 40 days of age. Variations in this ratio from colony to colony are of great interest, and may be calculated from the data; the mean N/H20 ratio should be retained for future use when the determination of carcass nitrogen would be unnecessary except in exceptional circumstances. h. Digestibility: A nitrogen determination on the faeces enables the calculation of digestibility and hence biological value. i. Standard deviation: The SD of the method is 3.7 NPU units. Thus, two-thirds of the results should fall within four units. Where these results are reported, the standard error should be less than four units. j. Potato starch: Raw potato starch is not digested by the rat and acts as "bulk." We have no experience with solkafloc, celluflour, etc. k. Maize (corn) starch: Rice starch is a good substitute.

Vitamins: A well-tried alternative may be used. m. Salts: A well tried alternative may be used.

Methodology for Net Protein Utilization (TD x BV/100) Determinations in Rats

Experience with the original nitrogen-balance procedure has led to the following approach. The experimental procedure has been described by Eggum (1).

Groups of five Wistar male rats weighing 65 - 68 9 are used. The preliminary period lasts for four days and the balance period for five days. The rats are weighed at the beginning of the experiments and divided into groups of five so that the average weights of the groups differ by no more than +0.5 9. Weighings are repeated at the end of the preliminary and balance periods; access to feed and water are restricted three hours before weighing. Each animal receives 150 mg nitrogen and 10 9 dry matter daily throughout the preliminary and balance periods. The nitrogen-free mixture (for adjusting protein level) consists of autoclaved potato starch, sucrose, cellulose powder, soy bean oil, minerals, and vitamins (1).

The experimental diets are weighed out into plastic boxes with tightly fitting lids for each of the preliminary and balance periods. The feed is weighed each day from these boxes in four daily allowances in the preliminary period and in five during the balance period. Any remaining feed is weighed and taken into consideration in the calculation of the experimental results. As an internal standard, casein fortified with 1 per cent DL-methionine is used. This diet has an NPU of 88.

The determination of metabolic and endogenous protein is determined by feeding a separate group a diet cntaining 4 per cent freeze-dried ether-extracted egg protein. Egg protein at this level is completely utilized by rats. Consequently, nitrogen in the urine and faeces must be of endogenous origin.

Calculations NPU = NNretained = I - (F - FK) - (U - UK) D N absorbed I - (F ~ FK) N intake I BV N retained = I - (FK) - (U - UK) N absorbed I - IF - FK)

Note: Abbreviations as defined in the Glossary (p. 123).

Nitrogen Growth Index in Rats

This method is a multiple-point assay that includes a nitrogen-free diet group as part of the experimental design. The method of procedure is as follows:

  1. Protein foods are first analysed for nitrogen content and, if possible, for their amino acid composition.
  2. Diets are then made to contain graded levels of protein as indicated below for various food protein groups: a. cereal grains: 0, 3, 6, and 9 per cent; or 0, 2, 4, 6, and 8 per cent; b. food legumes: 0, 3, 6, 9, and 12 per cent; c. oilseed meals: 0, 3, 6, 9, and 12 per cent; d. animal food proteins: 0, 2, 4, 6, and 8 per cent; e. standard protein: casein at 0, 2, 4, 6, and 8 per cent.
  3. In addition to the protein source, the diets contain a mineral and vitamin mixture equal to that used for other rat assays and 6 per cent oil so that all diets are isocaloric. To adjust to 100 per cent, corn starch is utilized. Fibre content should be 5 per cent of the diet.
  4. All diets are analysed for nitrogen content and stored at 4C during the experimental period.
  5. Weanling rats 22-23 days of age are distributed by weight among experimental groups, assigning ten animals per group. Statistical analysis has suggested that the number of rats per group can be decreased to six.
  6. The animals are fed ad libitum for a 14-day experimental period, obtaining weight changes every seven days and food intake every two days.
  7. From the food intake and protein content of the diet, the protein intake for each rat is then determined.
  8. Using regression analysis (see Glossary), Y = mX + C, where Y is weight gain and X is protein intake, the quality of the protein is estimated to be equal to m, the coefficient of regression. In this calculation, all values, including those in the nitrogen free group, are utilized.
  9. If faeces are collected during the first 5 days of the 14-day assay, protein digestibility is also obtained. In such a case it is important to determine daily food intake for the same period of time.
  10. The experiment provides the following information: a. The absolute protein quality given by the coefficient of regression. b. The relative protein quality when the above value is expressed as a percentage of the standard protein value used for reference. c. Protein intake for maintenance (weight gain equals zero) for the protein under test, or as a percentage of the reference protein. d. Protein digestibility, if faecal collection and analysis are performed.

A worked example of the calculation of NGI from the regression equations is shown in table 16.

Procedure for the Estimation of Relative Protein Value in Rats

This assay is based upon the slope of the dose response curve of animals fed the test protein compared to the slope of a similar curve from animals fed the standard protein. It should be noted that this test is strictly valid only if the dose-response curve is essentially linear in each instance. It should be further noted that the accuracy of

TABLE 16. Calculation of NGI and RPV from Rat Growth and Protein Intake Data: 14-Day Experiment

  Level of Individual protein intake g/day (X)           Individual weight gain g/day (Y)          
  protein fed 1 2 3 4 5 6 1 2 3 4 5 6
  30 g/kg 0.35 0.29 0.31 0.30 0.36 0.30 0.82 1.06 1.00 0.93 0.95 1.03
  50 g/kg 0.60 0.50 0.52 0.53 0.51 0.56 1.92 1.70 2.25 1.87 2.00 2.10
  80 91kg 1.01 0.93 0.90 0.83 0.89 0.88 3.50 3.30 3.00 3.75 3.82 3.53
  Test protein                        
  50g/kg 0.17 0.17 0.20 0.21 0.23 0.19 0.01 0.03 0.08 0.12 0.12 0.07
  70 g/kg 0.40 0.36 0.35 0.33 0.38 0.35 0.47 0.40 0.51 0.60 0.59 0.50
  90 g/kg 0.58 0.55 0.56 0.60 0.62 0.57 0.93 1.12 1.04 1.00 0.95 0.98
  Non-protein - - - - - - - - - - - - - - - - - - - - - - - - -0 57 -0.63 -0.50 4.67 -0.51 0.49


Regression equations relating weight gain (Y) to protein consumed (X):

A. Non-protein data not included (i.e., RPV) i. Lactalbumin Y = 4.12X - 0.28: R = 0.96 N = 18 ii. Test protein Y = 2.35X - 0.36: R = 0.97 N = 18 RPV = 2.35/4.12 = 0.57

B. Non-protein data included (i.eNGI)

i. Lactalbumin Y = 4.40X - 0.47: R = 0.98 N = 24 NGI (lactalbumin) = 4.40

ii. Test protein Y = 2.66X - 0.50: R = 0.98 N = 24 NGI (test protein) = 2.66

Relative NGI = 2.66/4 40 = 0.60

The estimate is a function of the numbers of animals utilized, the variance around the regression lines, and the range of protein intakes included in the test. The wider the range tested, the more accurate the result will be, provided the regression is linear. Departure from linearity may be expected if too high or too low levels are tested. Appropriate tests for curvature should be included. The actual( levels appropriate for any given protein depend upon the nutritional quality of the protein and the most limiting amino acid. For many foods, the upper level is determined by the protein content.

The basal diet contains:

g/kg d iet

Cornstarch 840-860

0il* 100

Salt mix** 30-50

Vitamin mix*** 10

One group of animals receives the basal diet. Three groups of animals receive diets containing three levels of standard lactalbumin, and three groups receive diets containing selected levels of the test material. Protein sources are incorporated into the basal diet at the expense of the starch. The most appropriate levels of protein to be used depend upon the quality of the protein and practical considerations. The suggested range of levels for various materials are:

Lactalbumin 20-80
Casein, meat 40-100
Soy protein 40-100
Corn 40
Wheat 50
Wheat gluten 60-240
Bread 50-120
Rice 40

As is indicated above, for many cereals, the upper limit will be determined by the protein content of the test material. The midlevel should be approximately equally spaced between the highest and lowest levels,

Weanling rats of either sex, or balanced for sex distribution, 21-23 days of age, are placed in individual cages and fed the 80 g/kg protein reference standard diet for two days. Six rats are then assigned to each group. The weight of each group is equalized; the range in body weight should not be greater than 15 G.

The animals are weighed twice weekly and food consumption is determined at least weekly. Food containers designed to minimize food spillage should be used, and papers or cardboard placed under cages with daily weighings of spilled food. The assay period is two weeks. A worked-out example for RPV of a test protein as compared to lactalbumin is shown in table 16.

The slope of the dose-response curve is determined by calculating the regression of weight gain on protein eaten for the animals consuming each protein source. A good approximation can usually be obtained by plotting mean values on graph paper and drawing the line by inspection; however, the capability of linear regression analysis is now so widespread on small calculators that the slope and intercept should be readily available.

The relative protein value is calculated:

RPV =( slope of the test material ) / ( slope of the standard lactalbumin )

Hypothetical growth and food intake data are shown in table 16 where regression equations are calculated and the slopes compared to give RPV.

As has been indicated, aberrant results will be obtained if the levels of protein selected are either too high or too low. Downward curvature at the high levels of intake suggests that the level selected was too high, and will yield an underestimate of protein quality. The slope of the regression line obtained after e)iminating the animals consuming the highest protein level should be re-examined.

Procedure for the Estimation of Relative Net Protein Ratio in Rats

When a one-dose assay is necessary due to a small amount of test material, or when it is desirable as a screening method, the NPR method is probably the most appropriate.

The instructions regarding animals and diets described for RPV are appropriate. A single level of protein close to the upper limit of the expected linear range is used. One group of ten animals is fed the basal diet, and five to ten animals, depending upon the amount of test material available, are given the test diet. A reference protein is fed at the 80 g/kg protein level. Weight changes and food consumption, adjusted for spillage, are calculated twice weekly for two weeks.

NPR for each animal is calculated according to the formula:

NPR = [ ( weight gain of test animal ) + (average weight loss of animals fed the basal diet)] / (protein consumed by the test animal )

and these are averaged to give a mean va)ue and a standard error.

The RNPR is the NPR of the test material expressed relative to the NPR of the reference standard, lactalbumin. The ratio may also be expressed as a percentage.

RNPR NPR test_ and the standard error of this ratio is calculated according to the formula in chapter 10. It should be noted that the NPR assay frequently overestimates the value of poor quality proteins. (NPU by weight gain is similar to NPR). Examples of NPR and RNPR calculations are shown in table 17.

TABLE 17. Calculation of Net Protein Ratio and Relative

Net Protein Ratio {RNPR) from Rat Data


  Mean Mean Diet NPR* RNPR**
  weight food N    
  gain intake %    
  (g) (g)      
Lactalbumin 69.7 249.2 1.30 4.04 1.00
Test protein A 63.2 243.6 1.57 3.15 0.78
Test protein B 11.7 125.3 1.61 1.89 0.47
Non-protein - 12.1 90.3 0.04    


* Net protein ratio is weight gain of test group of animals plus mean weight loss of non-protein control group per gram protein IN x 6.25) consumed; e.g., for lactalbumin

NPR = 69.7 + 12.1 = 4.04

(249.2 x 1.30 x 6.25)/100

* Since lactalbumin = 1.00 by definition, for test protein A

RNPR = 3.15/4.04 = 0.78

and for test protein B

RNPR = 1.89/4.04 = 0.47

The example shows calculation of NPR from group mean data. Calculations can also be made for individual animals and the result expressed as a mean.

Human Studies

In many countries it is now the case that any study involving human subjects must first be found acceptable and approved by an institutional committee concerned with ethics and human experimentation. For the protection of human subjects, it is recommended that if such a committee is not already in existence, one should be formed and its approval sought before any human studies are commenced (5).

Procedure for Nitrogen Balance Index, Relative Nitrogen Retention, and Relative Protein Value in Adults

Because there has been no significant attempt to standardize procedures for protein quality evaluation studies in humans, the following are proposed as a series of guidelines. While strict adherence to these guidelines may not always be possible, reasons should be given for significant departures from them.

  1. Selection of subjects - eight to ten - on the basis of (a) previous dietary intake; (b) freedom from disease or metabolic abnormalities by history and physical examination.
  2. The subjects are then given, for at least five days, a high-quality protein diet (0.6 g/kg/day), with energy intake carefully adjusted to their requirement.
  3. A basal diet must be established that is acceptable in taste and energy density, and complete in all other nutrients (vitamins, minerals). Each period is preceded by one day of the nitrogen-free diet at the same energy level. A standard multi-vitamin supplement should be given daily, plus an adequate supplement of essential minerals.
  4. The basal diet should be analysed for nitrogen content, which should be very low. Prior to the actual study, an acceptable liquid formula or diet should be developed to incorporate the protein, and each subject should sample this diet in advance and agree to consume it throughout the study. An example of a diet utilizing whole dried egg powder is given in table 18. If limitaton of cholesterol intake is desirable, high[-quality egg albumin can be substituted.
  5. All subjects are then given the basal diet and a level of energy appropriate to the individual.
  6. The above is then followed by consecutive protein feeding for 10 days, per protein level of 0.3, 0.4, 0.5, and 0.6 g/kg/day. Either half of the subjects should receive the test levels in a descending order and half in an ascending order, or the periods should be randomized in a Latin square design. If the quality of the protein is expected to be substantially less than that of milk, meat, or egg, the above levels should be increased by 0.1 9 protein/kg/day.

TABLE 18. Example of the Composition of an Experimental Diet

    Diet formula  
Non-protein     Level of
Ingredient Level of energy sources2 intake
Whole, dried egg powder3 66.0 g Carbonated beverages variable
Dextrose-maltose4 180.0 g Sucrose soft drink variable
Celluloses 0.4 g Corn starch dessert variable
Pectin 0.4 g Protein-free cookies variable
Sodium chloride 1.0 g Vitamin supplement6  
Ca1O (OH)2 (PO4)6 1.6 g Mineral supplement7  
K2HPO4 4 7 g    
Corn oil 118.0 g    
Water 480.0 ml    
Total formula intake 852.1 g    


  1. Intake is given for a 70-kg subject.
  2. Intake is varied among subjects and between diet periods. Fat and carbohydrates each provide approximately 40-50% of total energy. One package of sugarless chewing gum [Trident, American Chickle Co., Long Island, N.Y., USA) is also allowed.
  3. Spray-dried, pasteurized whole egg powder obtained from Oskaloosa Food Products Corp., Oskaloosa, lowa, USA.
  4. Compound obtained from Mead Johnson Laboratories, Evansville, Ind., USA. Approximate analysis: maltose 56%, dextrins 42%.
  5. Avicel micro-crystalline cellulose obtained from F.M.C. Corp., Philadelphia, Pa., USA.
  6. Each subject receives two Unicap Plus Iron vitamin capsules daily (Upjohn Co., Kalamazoo, Mich., USA). Each capsule contains: vitamin A, 5,000 units; vitamin D, 400 units; thiamine mononitrate (B1 ), 2.5 mg; riboflavin (B2 ), 2.5 mg; ascorbic acid (C), 50 mg; niacinamide, 20 mg; pyridoxine HCI (B6 i, 0.5 mg; calcium pantothenate, 5 mg; cyanocobalamin (B1 2 ), 2 microg; vitamin E, 10 IU; iron (50 mg ferrous sulfate), 10 mg; iodine (as potassium iodate), 0.15 mg; copper (as sulfate), 1 mg; manganese (as sulfate), 1 mg; magnesium (as sulfate), 6 mg; potassium (as sulfate), 5 mg; calcium (as lactate and carbonate), 35 mg. Subjects are also given one folic acid tablet daily (Moore Drug Exchange, New Britain, Conn., USA} containing 1 mg folic acid, and two choline tablets per day (PIUS Products, Los Angeles, Calif., USA), each providing 250 mg choline.
  7. Trace mineral supplement contains: 659.9 mg MgO; 1.79 mg CuCI2.2H20; 14.6 mg ZnSO4 7H20; 5.12 mg MnSO4 H20;0.21 mg Na2MoO4~2H20; 1.07 mg Cr2(8o4)2 15H2o; 28.3 mg KAl(SO4)212H20;trace (~0.01 mg) Na2SeO2. Each subject receives one or two NaCI tablets (1 9) per day.
  8. A faecal marker substance is fed at the start and end of each balance period (e.g., charcoal or autoclaved carmine), and all faeces are collected and pooled until the second marker appears in the stool. This stool is then discarded.
  9. All urine must be collected between the same 24-hour periods each day and refrigerated as soon as convenient. Five ml of 1 N HCI should be added, in advance, to each 1- to 2-litre specimen bottle.
  10. Representative samples of the food consumed and pooled faeces for each period, and daily urine samples are analysed for nitrogen. (Sweat samples may be necessary in addition under hot climate conditions or with heavy exercise.)
  11. From the intake and excretions of nitrogen, the following can be calculated: a. NB at each level of protein intake, b. protein digestibility, c. the regression of Nl and NA on NB.
  12. The above calculations provide: a. protein digestibility, b. protein intake for nitrogen equilibrium, c. relative protein quality when a reference value is obtained or available.

Procedure for Children

The methods of procedure for children follow the steps indicated for adults, except that they are not fed a protein-free diet (the protein-free period has proven unnecessary because children adapt more rapidly), and the levels of protein intake may go from 0.8 to 2.0 9 protein/kg/day. Each protein level is fed for a nine-day period, of which the first three are adaptation to the level and the last six, divided into two three day periods, are for the quantitative collection of faeces and urine. It is possible to get adequate three-day faecal collection in children, while in adults less frequent stools make necessary a longer collection. Therefore, at each level of protein intake, two balances are obtained, which introduces a replication in the test per protein level. Calculations follow those already described. The background to these calculations has already been discussed in chapter 5, and worked examples were shown (pp. 66 - 67).

Short-Term Nitrogen-Balance Assay

On the basis of the experience of the group working at the Institute of Nutrition of Central America and Panama (6-9), it is justifiable to propose a short nitrogen-balance-index assay for both adult and young human subjects for wider testing. The same conditions, analysis, and calculations as those for the conventional method are followed, except that protein feeding at each level of intake is carried out for two days with no period for adaptation to each protein level. Equilibrium is thus not achieved at each level, but the sequentia( sequentialsamples appear to give comparable results if ascending and descending patterns are equally represented and averaged.

The assay levels used are 0.3, 0.4, 0.5, and 0.6 9 of protein/kg/day. With poor quality proteins, the levels are 0.4, 0.5, 0.6, and 0.7 g/kg/day. Half of the subjects receive the highest level for eight days and the other half the lowest level. Each level is then tested for two days in either descending or ascending order. Daily 24-hour urine samples are collected throughout. Faeces can be pooled for the entire period and total faecal nitrogen compared with total nitrogen intake to give a digestibility figure from which daily absorption can be calculated.

The method is probably equally applicable to children, utilizing the same protein levels as recommended for the "long-term" assay in this age group, but has not yet been evaluated directly.


1. B.O. Eggum, A Study of Certain Factors Influencing Protein Utilization in Fats and Pigs, publ. no. 406 (National Institute of Animai Science, Copenhagen, 1973).

2. Committee on Protein Malnutrition, Food and Nutrition Board, National Research Council, Evaluation of Protein Quality, ed. P.L. Pellett, NAS - NRC Publ. 1100 (National Academy of Sciences,Washington, D.C.,1963).

3. A.E. Bender and D.S. Miller, "A New Brief Method of Estimating Net Protein Value," Biochem. J., 53: vii (1953).

4. D.S. Miller and A.E. Bender, "The Determination of the Net Utilization of Proteins by a Shortened Method," Brit. J. Nutr., 9: 382-388 (19551.

5. J.H.U. Brown, "Functions of an Institutional Review Board and the Protection of Human Subjects," Fed. Proc., 80: 2049-2050 (1979).

6. D. Navarrete, V.A.L. de Daqui, L.G. Elias, P.A. Lachance, and R. Bressani, "The Nutritive Value of Egg Protein as Determined by the Nitrogen Balance Index (NBI)," Nutr. Repts. Internat, 16: 695-704 (1977).

7. R. Bressani, D.A. Navarrete, V.A.L. de Daqui, L.G. Elias, J. Olivares, and P.A. Lachance, "Protein Quality of Spray-Dried Whole Milk and of Casein in Young Adult Humans Using a Short Term Nitrogen Balance Index Assay," J. Fd. Sci.. 44: 1136-1149 (1979).

8. R. Bressani, B. Torun, L.G. Elias D.A. Navarrete, and E.A. Vargas, "A Short Term Procedure to Evaluate Protein Quality in Young and Adult Human Subjects," in C.E. Bodwell, J.S. Adkins, and D.T. Hopkins, eds., Protein Quality in Humans: Assessment and In Vitro Estimation (Avi Publishing Co., Westport, Conn., USA, 1980 [in press] ).

9. D.A. Navarrete, L.G. Elias, J.E. graham, and R. Bressani, "The Evaluation of the Protein Quality of Soybean Products by Short-Term Bioassays in Adult Human Subjects," Arch. Latinoamer. Nutr., 29: 386-401 (1979).

Contents - Previous - Next