CA1235050A

CA1235050A - Methods of and/or materials for determining glucose levels in blood samples

Info

Publication number: CA1235050A
Application number: CA000418374A
Authority: CA
Inventors: John R. Baker
Original assignee: Individual
Current assignee: Individual
Priority date: 1981-12-23
Filing date: 1982-12-22
Publication date: 1988-04-12
Also published as: EP0085263B1; DE3273255D1; DK567782A; ZA829252B; NZ199380A; JPS58154660A; DK158482C; ATE22181T1; AU9162682A; EP0085263A1; JPH02263163A; HK16990A; US4645742A; US4642295A; JPH0113062B2; DK158482B; AU554029B2; IN156045B

Abstract

ABSTRACT

The determination of glucose levels in at least part of a blood sample for use in detecting diabetes in patients or deciding treatment levels is effected by controlling the temperature and pH of the sample adding a colouring agent such as nitro blue tetrazolium and after a first delay taking a first colour measurement then after a second delay taking a further colour measurement and then conforming the colour change with the colour change in a standard solution, the selected colouring agent and pH being such that the glucose in the sample is still reacted associated with an amine group of protein in the blood and when the glucose has undergone a molecular rearrangement and during colour measurement is in the staple fructosamine form.

Description

This invention relates to methods of and/or means for determining glucose levels in blood samples and has been devised particularly though not solely for use in detecting diabetes in patients and/or deciding treatment levels for diabetes patients.
It is an object of the present invention to provide methods of and/or materials for determining the glucose levels in blood samples which will at least provide the public with a useful choice.
I Accordingly in one aspect the invention consists in a method of determining serum glucose levels in a blood sample or a sample derived from blood, said method comprising the steps of maintaining the sample at a controlled temperature, controlling the pi of the sample to a suitable value, between 10 and 11, adding a coloring agent to the sample and after a first delay in taking a first color measurement and after a second delay in time taking a second color measurement and comparing the resultant color change with the color change in a standard solution, the selected coloring agent and pi conditions being such that a change of color in the coloring event is caused by the glucose in the sample while that glucose is still reacted or associated with an amine group of protein in that blood, and when the glucose has undergone a molecular rearrangement and is present in the form of fructosaTnineO
In a further aspect the invention consists in a method of detecting diabetes Maltese in human beings comprising the steps of taking a blood sample or a sample derived from blood from the patient adding a buffer to adjust the pi of the sample to a suitable value, between 10 and 11, and aye _ adding a coloring agent which changes color J the selected coloring agent and pi conditions being such that the change of color in the coloring agent is caused by the glucose in the sample while the glucose is still reacted or associated with amine groups of protein in that sample, and the glucose has undergone a molecular rearrangement and is present in the form of fructosamine, reading first and second color levels after a first and a second timed interval from adding the coloring agent, obtaining I corrected readings and comparing such corrected readings with standards representative of known stages of diabetes Maltese.
Background to the invention and a description of the preferred and alternative embodiments of the invention will now be given with reference to the accompanying drawings in which:
Figure 1 is a graphs of serum fructosamlne levels in patients with normal glucose tolerance, impaired glucose tolerance and diabetes Maltese respectively applying WOO
criteria for the interpretation of 75gm glucose tolerance test.
Figure 2 is a graph of serum fructosamlne levels in pregnant patients at 28-32 weeks gestation; and Figure 3 is a flow diagram of a continuous testing method using the invention.
GRENADA TO THE INVENTION
Glycosylation of proteins can occur as a non-enzymic post-translational modification directly dependent upon prevailing glucose concentration. Consequently diabetics I
tend to have elevated concentrations of glycosylproteins and the degree of glycoslyation of hemoglobin and serum proteins has been correlated with indices of glycemia.
Because glycosylprotein concentrations reflect an average of serum glucose levels over a period of time their determination provides an attractive means of monitoring diabetic control and the present invention provides a convenient and practical system for effecting such monitoring.
The present invention makes use of a characteristic of blood that blood glucose reacts continuously with proteins in the circulation of in particular human beings in that glucose binds to amino groups of protein to form an aldimine, a Showoff base, which undergoes an molecular (Nader) rearrangement to form a stable ketoamine (herein generically termed "fructosaminen~. The chemistry of this is discussed later in this specification. The present invention at least in the preferred form relates to a calorimetric assay based on the property of fructosamines to act as reducing agents in alkaline solution.
In recent years the major effort in managing diabetes has concentrated on the prevention or amelioration of the chronic complications of diabetes. These complications retinopathyr nephropathy, neuropathy and atherosclerosis -are derived from prolonged hyperglycemia, and can be prevented if the blood glucose is maintained as close as possible to normal.

sty of the existing methods for determining the degree of diabetic control are unreliable because they require patient co-operation (ego fractional urine collections by the patient at home) or are cumbersome and unreliable (e.g.
24 hour urinary glucose excretion). The blood glucose level fluctuates considerably throughout the day influenced by diet, activity, and treatment. Random blood glucose levels have been found to be an inadequate and even misleading index of diabetic control. In the management of established diabetes there is a need for a long term indicator of blood glucose levels. One of the first tests investigated in this context was glycosylated hemoglobin.
This test has been useful in the management of diabetes but the test has proved expensive and difficult to reproduce.
The need for a more reliable test has led to the present investigation of serum fructosamine. Fructosamine, which is a product of the interaction of serum glucose with serum proteins over days to weeks, meets some of these objections, since it is stable, and does not change with short term fluctuations in blood glucose levels. This test also has the advantage that it is inexpensive, may be automated, and is very reproducible. It can be done on approximately 2 mls of blood Fructosamine provides an index of diabetic control that is dependent of the patient's diet, activity or treatment.
Its major clinical use is the regulation of treatment for known diabetics and assessment of new treatments. It can also be used to screen populations with an increased risk I
of developing diabetes - pregnant women, the obese and the ethnic groups with known tendency to diabetes (e.g.
Polynesians. The distinction between the diabetic patient and the patient with normal glucose tolerance test is good but is less impressive for patients with impaired glucose tolerance.
Over the last year investigations have been done into the levels of serum fructosamine in normal pregnant patients, in pregnant mothers with abnormal glucose tolerance, and those with established diabetes In normal patients (76 patients with normal glucose tolerance tests evaluated using O'Sullivan's data O'Sullivan JOB Mason CAM., Charles D., Dandier R. (1973) Screening criteria for high risk gestation Al diabetes patients. Amer. J. Obstet.
11~: 895.) the mean serum fructosamine level at 28-32 weeks gestation was 1.51 molehill with a standard deviation of 0.24.
Recent fructosamine levels collected at 28~32 weeks gestation in patients with gestation Al diabetes" and those with established diabetes have shown that there is a significant difference between the three groups (figure

2). The mean fr~ctosamine levels in patients with diabetes diagnosed in pregnancy is 1.91 molehill with a standard deviation of 0.23 and in established diabetes is 2.17 with I a standard deviation of 0.36. Using a level of 1.8 molehill as the upper limit of normal for serum fructosamlne it is possible that its use in a screening test would lead to a detection rate of 80~ of diabetes in pregnancy.

DETAILED DISSUASION OF THE INVENTION:
Fructosamine bound to serum proteins is measured according to the invention by conversion to its active enol form in alkali, that is to say in a pi which lies between 10 and 14.
We have found that different pi levels give different absorbency in the coloring agent used in the method of the invention and accordingly the preferred pi lies between 10.0 and 11.0 and is preferably between 10.5 and 10.8.
The enol form of fructosamine is a chemically active Lo substance and causes a suitable coloring agent to change color the coloring agent being preferably a dye selected from tetrazolium salts e.g. twitter vitro blue or preferably nitro-blue tetrazolium which changes to a highly colored blue (purple) formazan dye having a broad absorbency peak at about 535nm.
It is part of the invention that to avoid interference from other serum reducing substances the absorbency change is measured at a suitable time aster adding the coloring agent and for example 10 to 15 minutes is allowed to reduce or avoid this interference from other serum reducing substances The sample is one derived from blood and preferably is a sample of whole blood or of whole blood serum.
Some precautions are necessary in connection with the blood sample in that moderate to severe haemolysis may give false low results Although whole blood serum and plasma may be used as the sample material, plasma is much less suitable for the procedure since anticoagulants may interfere with the results.

I
Samples should preferably be sofa with no added preservative.
To provide a suitable pi value in the blood sample, preferably a buffer is added to the blood sample in the form of a solution. 'rho buffer is selected and proportioned to give a pi in the blood sample of between 10.0 and 11.0 preferably between 10.5 and 10.8. This buffer preferably comprises sodium carbonate and sodium bicarbonate in suitable proportions and we have found that it is preferable to provide the buffer in association with the coloring agent as a single reagent. The buffer desirably contains sodium carbonate in the proportion of 0.795gm sodium carbonate to 0.210gm of sodium bicarbonate.
Other buffers giving the desired pi may be used. As above stated, the coloring agent is preferably nitro-blue tetrazolium (NUT).
The reagent is mixed with the blood or serum samples and left to stand for a suitable period of time, for example 10 to 15 minutes, before the color change is measured, for example the absorbency change is measured using a suitable spectrophotometer.
The measuring of the color change is preferably checked by using a procedure in which standard solutions are checked for color readings. The preferred standard I solutions comprise a protein such as albumin to which is added a synthetic fructosamine such as l-deoxy,l-mor~holinofructose (DMF). Such standard solutions are preferably prepared by adding allocates of a O~Olm aqueous solution of l-deoxy,l-morpho]inofructose (DMF) to an albumin solution. The albumin may be Albumin glumly it gloater) diluted to ogle for used S These standard solutions are made up in albumin as its presence is necessary to move the peak absorbency of DMF.
However, the albumin itself also has some activity and an albumin solution suitably calibrated against the preferred (DMF) standards or calibrated against a protein solution LO with known 14C-glucose or glucose incorporated my also be used as standard solutions.
Experiments suggest that the measurements are influenced by reductants of low molecular mass to only a minor degree and are consistent with the notion that NUT
reduction reflects mainly the concentration of high molecular mass ketoamines or glycosylproteins, inn fructosamine.
When estimated as the change in absorbency at 530nm over successive five minute intervals between 5 and 20 ; 20 minutes, NUT reduction is linearly related to DMF
concentrations at each time interval, linearity being maintained to Molly of DMF/l, the highest concentration tested. It should be noted that, if plotted, the lines do not pass through the origin as the standard contains 40g of albumin/l, which itself has some activity.
The methods of the present invention may be performed manually or in automated systems. The assay is simple to do manually and is reproducible. It is also fast.

23~ D
Although in general only 12 samples per hour can be measured where the full time course is recorded in each case, by reading samples every 20 seconds at 10 and 15 minutes in a discontinuous assay 40 samples per hour can be accommodated.
The reagent comprising the buffer and the coloring event and the standard solutions comprising a protein such as albumin and a synthetic fructosamine such as l~deoxy,l-morpholinofructose (DMF) are both novel compositions which form part of the present invention.
Accordingly, the invention also consists in a reagent for use in the methods of the present invention, said reagent comprising a buffer and a coloring agent which changes color; This reagent when added to a blood sample or a sample derived from blood provides coloring agent and pi conditions such that a change of color in the coloring event is caused by the glucose in the sample while the glucose is still reacted or associated with amine groups or protein in that sample and when the glucose has undergone a molecular rearrangement and is in the form of fructosamine.
The invention further consists in a standard solution comprising a protein such as albumin and a synthetic frl]ctosamine such as DMF, preferably a O.Olm aqueous solution of DMF in an albumin solution of ogle concentration.
Changes in the method and materials set forth above may be made while still utilizing the principle of the invention namely the enol form of fructosamlne is a chemically active substance which converts a suitable coloring agent to give a color or absorbency change which can be measured and compared with a standard to given an indication of the level of glucose in blood samples even though that glucose remains reacted or associated with protein or amine in the blood and has undergone a molecular rearrangement to form fructosamine.
To those skilled in the art to which this invention relates, many changes in construction and widely differing embodiments and applications of the invention will suggest themselves without departing from the scope of the invention as defined in the appended claims The disclosures and the descriptions herein are purely illustrative and are not intended to be in any sense limiting.
Example l ; manual Procedure Melissa Albumin is human albumin from the Commonwealth Serum Laboratories, Melbourne, Australia and the NUT is from Sigma Chemical Coy, St. Louis, My., United States of America.
To give a color reagent: Old Knickknack pi 10.8), weigh out successively -Nikko 7.95g Nikko 2 1 g Nitro-blue tetrazolium(sigma) 0.2 g Ethanol 40ml Deionized water up to 1 lithe Flush each reagent into a lamely volumetric flask with a little distilled (deionized water. Dissolve by gently swirling and making up to lamely with distilled water. The pi should be 10~5 to 11.00 and is preferably 10.8 The reagent is not stable at room temperature and should be made up fresh each week or alternatively and preferably it made up in larger quantities, divided into Alcott parts, stored frozen and then thawed before use. The albumin standard comprises albumin 40g in 1 lithe of saline (0.14m Nail).
To give l-deoxy,l-morpholinofructose (DMF) Molly LO standard solutions, 124.5mgm of DMF (MOE) is weighed out and made up to loom with albumin 40g/1. The standard is made up in albumin as its presence is necessary to move the peak absorbency of DMF. However, the albumin itself also has some activity and this is subtracted from the standard activity in the calculation shown below Standards of concentrations of loo 200; 3.0 and Molly are similarly prepared. The standards are divided into 0.5ml allocates and stored at -20C.
The E~cedure adopted for measuring fructosamine concentration in blood samples is as follows:
Blood samples are collected in plain tubes, allowed to clot and separated promptly. When not assayed immediately serum may be kept at -20C at which temperature fructosamine is stable indefinitely.
O.lml of serum is added to lml of carbonate buffer (O.lmol/l,pFI 10.8) containing Molly of nitro-blue tetrazolium (NBT)/l at 37C; the absorbency at 530nm is measured at 10 and 15 minutes after mixing and compared with that of standards of l-deoxy/l-morpholinofructose (DMF) plus albumin (40g/1) treated in identical fashion.

Absorbency is measured in a recording spectrophotometer such as Pyre Unicam SPY or Gil ford 250. Reaction rates may, if desired, be followed on a recording spectrophotometer which is thermostatically controlled.
So that timing can be accurately maintained, the serum samples are taken at selected timed intervals e.g. 15 seconds and O.lml of each serum sample is added to lamely color reagent. The absorbency is then read at 530nm at 10 minutes and 15 minutes from commencement of the test. The albumin and DMF standards are treated to similarly to give a check. Corrected readings of ructosamine activity can be calculated as follows (Absorbency):-(monomania) Test Jo _ 5 (A15mins-AlOmins) Std-(A15mins-AlOmins)Alb.=Fructosamine Cone in Units/L

~ef~Range-1~40-1.96 Units/L
votes.
1. Plasma is less suitable for this test, as anticoagulants may interfere.
2. Specimens are stable at least three months when stored at -20C.

3. Controls - Commercially available lyophilised sofa both bovine and human based provide suitable low, medium and high controls.

, En aye The Determination I Serum or Plasma Fructosamine ~ncentration by a continuous flow analyzer eke a Techniçon Aut~-Analyser As already noted, albumin bound fructosamlne reduces nitro-blue tetrazolium dye at alkaline phi Non specific reduction by other substances goes to completion in a few minutes while fructosamine reduction is ongoing for a much longer time.
As shown in figure 3, where a continuous flow apparatus it used, such as a Technic on Auto-Analyser r compensation for nonspecific reduction by other substances can be made by using a blank channel lo The color change on this channel is read after 5.5 minutes of reaction time while the test channel 2 is read after 8.5 minutes of reaction time. The reaction streams are heated to 45C to shorten the peak delay time and the color reagent contains ethanol in order to linearize the response. The ruler is read at 550nm.
Serum albumin concentration is determined in parallel as the fructosamine level in albumin deficiency states is invalidated.
yummy description Referring to figure 3 the sample admitted through line 3 is prediluted (1:15) then resample into the reagent streams 4 and 5. The blank channel incorporates a delay coil 6 with a shorter heating coil so that the reaction time is three minutes less than the test channel t but the total dwell time it the same. The reduction takes place at 45C, the peaks arrive at the calorimetry simultaneously where the blank reading is automatically subtracted. Exact phasing is achieved with a phasing coil in the blank line, just prior to the calorimetry.
Method l. Select a sampling rate of 75/hr, sample time 40 sea, wash time Seiko.
2. Connect delineate and NUT reagent to reagent bottles.
I Turn on pump.

4. Take standards from deep freeze and allow to stand at room temperature.

5. After 10 minutes turn on recorder and set baselines at 10%.
5. Load and run standards, followed by samples. A control is placed every 10 samples.
7. Rinse reagent lines with Noah (lo) for 5 minutes after last peak has come through, followed by 10 minutes rinsing with water.
itch Structure a) Standard Batch Cup Number 1. 0 standard 2. 1.0 molar std.
3. I molar std.
4. 3.0 molar std.
5. 4.0 molar std.

6. 5.0 molar std.

I
b) Patient Batch Cups 5, 25 LOW CONTROL SERF
Cups 15, 35 HIGH CONTROL SERF
Cups 10~ 20, 30, 40 REV CONTROL
I Lo Ox The standards are made up in albumin solution (40g/1) end need to be corrected for albumin activity before the standard curve is plotted. The "O" standard is subtracted prom the other standards and the corrected standards used to prepare the standard curve with the peak heights (y axis) plotted against concentration (x axis).
The sample peaks are read off directly from the standard curve.
e~gellts 15 Delineate Sodium Chloride (BDH? 0.9g Bridge%) 20ml Water to 1 lithe Color Reagent Nitro-blue tetrazolium (Sigma) 0.20g Sodom Carbonate(Riedel-de Ann) 7.95g . Sodium Bicarbonate ) 2.10g Water to 1 lithe aye&
A stock solution of deoxy-morpholino fructose (DYE') is prepared by weighing 1.245g of DMF and dissolving in 1 lithe of albumin solution. The concentration is ' milliliter.
The albumin solution is prepared by dissolving albumin, 40g in 1 lithe of saline (0.14m Nikolai I
Working standards are prepared from this at concentrations of 1.0; 2.0, 3.0; 4~0 and 5.0 mm~l/litre. A
blank is prepared using only the albumin delineate with no added DMF.
POINTS OF SHEA
1. The sampling rate depends on the time taken for the peaks to come Jo steady state. Some plateau time is required on both test and blank so that they can be phased. Sampling time as little as 30 seconds is sometimes enough, but 40 seconds is better for routine running 2. Standards contain albumin and need to be refrigerated.
3. The system requires 5 minutes rinse with sodium hydroxide (lo) after each run.
sample 3 The DeteLm~nation of Serum Fr-uç-osamine~by Becker Discrete Analy~rZ~BBOTTl A. Half 1) Buffer and color reagent:~O.lM Knickknack pi 10,57) Weight out successively -Nikko 0~795g Nikko 0.210g Nitro-blue tetrazolium(NBT) 0.029 2) Standard solutions are prepared by adding allocates of a O.Olm aqueous solution of deoxymorpholinofructose (DMF) to an albumin solution DMF Albumin 0.14M Nail (O.OlM) (2509/1) O 0 18ml 82ml lamely loll 18ml 72ml 2.0Tnmol/L 20ml 18ml 62ml molehill 30ml Moe 52ml molehill 40ml 18ml 42ml I
The standards are divided into 0.5ml allocates and stored at 20C.
Flush each reagent into a loom volumetric flask with a little distilled water. Dissolve by gentle swirling and make up to loom with distilled water. The pi should be 10.55 to 10.60 and is preferably 10.57. The reagent is not stable indefinitely and should be made up fresh each week or stored at -20C.
B. Procedure:
Conducting a series of tests on an Abbott 200 Bichromatic Analyzer is effected as follows:
1. Set up the instruments as follows:

Test code 78 Sample volume 25uL (07) - Reagent volume 2501L ~06) Filter 550/650 Rx Type Rate RX Direction ` up Analysis time 5 minutes Revolutions 4 Temperature 37C

2. Use a new clean cuvette.

3. Activate the WASH cycle to flush water through the system and remove bubbles. Working reagent is then substituted for water and PRIME cycle commenced Discard the first 3 jets then recycle back into the bottle at least ten times.
4. Insert the required number of sample cups in the carousel and add 50uL of sample to each cup.
01 ~120 02 err standard 03 -- lamely 04 2.Ommol/L

-lo-05 3.Ommol/L
06 molehill 32 NORMAL control 09 to 30 are patient specimens. Use a Noel positive displacement syringe.
5. A calibration curve is included with each run. Rotate the carousel to position 00 and press the RUN button.

6. The BYWAY 200 will make 4 revolutions, the first dispensing sample and reagent, the second a delay cycle, the third and fourth initial and final absorbency readings respectively. The difference between the initial and final absorbency readings is used to compute the sample concentration using the standard curve provided.

7. Remove the multicurvette and rinse with water. Leave to soak overnight in water. Flush the pipette with "Countrywide cleaning solution by activating the WASH cycle, and repeat the wash procedure with water

8. The normal range is 1.14 to 1 molehill. Diabetic specimens will usually be greater than molehill.

9. The temperature is controlled, preferably to 37C.
~ointfi technique 1. The BYWAY 200 will automatically reject incorrect RX type Rx direction, temperature, revolution, and time, but I incorrect filters or sample/reagent volumes settings.

2. Color reagent with sodium carbonate/bicarbonate buffer is stable if stored at 4C for up to 1 week. Check the pi immediately prior to use and ensure that it is 10.55 to 10.60. Ensure that all NUT is dissolved before using the color reagent 3. Control specimens should give values within the 2 x SD
range in accordance with preferred quality control charts. It is preferable to document all assays where excursion outside the 2 x SD range occurs 4. Samples should be sofa with no added preservative.

The methods of the present invention utilize the following chemistry:
Proteins can be glycosylated in viva by a non-enzymatic reaction between glucose and available amino groups.

Protein Ho COO Protein -N = OH Protein -NH -SHEA
OWE OWE) I
~4H904 C4H904 Shag Glucose Glycosylamine Fructosamine lo Glycosylation is dependent upon glucose concentration so that diabetics have higher fructosamine concentrations than normal. Fructosamines in Alkaline solution form eneaminols which are reducing substances.

OH-R - NH -f~2 -I NH - OH
= o foe 4H904 C4HgO4 Ret enol(Reducing) In the preferred form ox the invention, the reducing activity of albumin bound fructosamine can be detected with Vitro blue tetrazolium at alkaline pi and is measured as a rate of change of absorbency.
From the foregoing it will be seen that the invention at least in the preferred form provides advantages compared with or overcomes disadvantages of present methods.

In particular, the clinical significance of the present invention can be summarized as follow:
1. Fructosamine is an index of metabolic control in patients with diabetes Maltese. It reflects patient compliance, quality of care, and the efficacy of insulin therapy. By use of the present invention fructosamine can be measured at weekly intervals, and it is believed that such measurement accurately and reliably detects improvement/deterioration in diabetic I control consequent to a change in management.
2. E'ructosamine reflects the glycosylation of serum proteins. The main contribution, greater than By% is from albumin. The kinetics of its decay approximate the behavior of albumin. ~Tl/2 = 20 days).
3. the value of the test is severely impaired in individuals with hypoalbuminaemia albumin less than glue).
4. Fructosamine correlates directly to the fasting blood glucose levels and is relatively less affected by post prandial hyperglycemia. Sample time and its relation to diet is unimportant.
5. Fructosamine may be useful as a screening test to detect individuals with diabetes Maltese.
I By conducting the reaCtiOJI in mild alkaline conditions (pi less than 11.0) and at controlled temperatures (T
less than 50C) the method according to the invention does not measure free glucose or glucose in the labile aldimine form but measures only glucose in the stable ~fructosamine form.

I

~3~5~
7. sty deferring the measurement of color change for a few minutes after commencing the test the contribution of interfering substances is minimized.

Claims

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A method of determining the level of fructosa-mine in a blood sample or sample derived from blood, wherein the level of fructosamine reflects an average serum glucose level in the sample over a period of time said method comprising the steps of maintaining the sample at a controlled temperature below 50 C, controll-ing the pH of the sample to a value between 10 and 11, adding a colouring agent to the sample and after a first delay in time taking a first colour measurement at a predetermined wavelength and after a second delay in time taking a second colour measurement at the predetermined wavelength and determining the fructosamine level in said sample by comparing any resultant change between said first and second colour measurements with those of stand-ard solutions wherein the colouring agent, timing of the delays, wavelength and pH conditions are selected such that any change of colour in the colouring agent between said first and second colour measurements is caused pre-dominantly by glucose in the sample that is reacted or associated with an amine group of protein and has under-gone a molecular re-arrangement to form fructosamine and not materially by any non-specific reducing substances which may be present in the sample.

2. The method as claimed in Claim 1 wherein said pH
level is controlled by adding to the sample a buffer which includes said colouring agent.

3. The method as claimed in Claim 2 in which said buffer comprises sodium carbonate and sodium bicarbonate.

4. The method as claimed in Claim 1 wherein said colouring agent comprises a tetrazoliun salt which changes colour on being reduced.

5. The method as claimed in Claim 4 wherein said colouring agent is nitro-blue tetrazolium.

6. The method as claimed in Claim 1 wherein the method steps are carried out in a automatic testing ap-paratus on a series of samples.

7. The method as claimed in Claim 6 wherein said series of samples are maintained at a temperature of about 45 C.

8. The method as claimed in Claim 1 which includes the step of calibrating a testing instrument used during testing of said sample using at least one testing stand-ard which gives a known reading.

9. The method as claimed in Claim 1 wherein said standard solutions comprise aqueous solutions of a pro-tein and 1-deoxy-1-morpholinofructoes.

10. The method as claimed in Claim 9 wherein said protein is albumin.

11. The method as claimed in Claim 1 wherein said first and second colour measurements are obtained by measuring colour absorbances at 530 nm 5-10 minutes and 8-15 minutes respectively after adding the colouring agent to the sample.

12. The method as claimed in Claim 11 wherein said colour absorbances are measured about 10 minutes and 15 minutes respectively after adding the colouring agent to the sample.

13. The method as claimed in Claim 11 wherein the method steps are carried out at about 37 C.

14. A method of detecting diabetes mellitus by de-termining the level of fructosamine in a blood sample comprising the steps of maintaining the blood sample at a controlled temperature below 50 C, controlling the pH
of the sample to a value between 10 and 11, adding a colouring agent to the sample and after a first delay in time taking a first colour measurement at a predetermined wavelength and after a second delay in time taking a second colour measurement at a predetermined wavelength, obtaining corrected readings utilizing said colour mea-surements and comparing the corrected readings with corrected readings of standards representative of known stages of diabetes mellitus, wherein the colouring agent, timing of the delays, wavelength and pH conditions are selected such that any change of colour in the colouring agent between said first and second colour measurements is caused predominantly by glucose in the sample that is reacted or associated with an amine group of protein and has undergone a molecular re-arrangement to form fruc-tosamine and not materially by any non-specific reducing substances which may be present in the sample.

15. The method as claimed in Claim 14 wherein said first and second colour measurements are taken at about S
and 10 mintues respectively after adding said colouring agent to the sample.

16. The method as claimed in Claim 14 wherein said colouring agent is a tetrazolium salt.

17. The method as claimed in Claim 16 wherein said colouring agent is nitro-blue tetrazolium.

18. A standard solution for use in determination of serum fructosamine levels in a blood sample or a sample derived from blood, said standard solution comprising an aqueous solution of a protein and 1-deoxy-1-morpholinofructose.

19. A reagent for use in determining serum fructosamine levels in a blood sample or a sample derived from blood, comprising a sufficient amount of an alkaline buffer and a tetrazolium dye to convert any fructosamine present in a sample to its active enol form and react over time after the buffer and tetrazolium dye have been added to the sample such that any colour change after an initial period of time is caused predominantly by glucose in the sample that is reacted or associated with an amine group of protein and has undergone a molecular re-arrangement to form fructosamine and not materially by any non-specific reducing substances which may be present in the sample.

20. The standard solution as claimed in Claim 18 wherein said protein is albumin.

21. The reagent as claimed in Claim 19 wherein said tetrazolium dye is nitro-blue tetrazolium.

22. A method of determining serum or plasma glucose levels in a blood sample or a sample derived from blood which comprises reacting the sample containing fructosamine with a colouring oxidant agent under alkaline conditions and comparing the resultant colour with the colour in a standard solution.

23. The method of Claim 22 further comprising eliminating reducing substances other than fructosamine in said sample to avoid interference from such other reducing-substances.

24. The method of Claim 22 further comprising oxidizing reducing substances other than fructosamine in said sample.

25. A method of determining serum or plasma glucose levels in a blood sample or a sample derived from blood which comprise reacting the sample containing fructosamine with a colouring oxidant agent under alkaline conditions and measuring the resultant colour.