Method and Device for the Semiquantitative Determination of Glucose in Aqueous Fluids

The detection of the concentration of glucose in urine is of great importance to diabetic individuals who must control their blood sugar level. Because continued control of blood sugar level is so important, a test device for the determination of glucose concentration in urine must be rapid and simple enough for the individual to use with ease and it must be quantitative enough to accurately measure increments of glucose concentration and thereby differentiate between various medical conditions.

A novel and very effective device for the quick and convenient determination of urine glucose concentration is disclosed in U.S. Pat. No. 3,814,668 (the '668 patent), issued June 4, 1974. This patent discloses a test strip which comprises a carrier member impregnated with glucose oxidase, a peroxidative active substance and a water soluble iodide salt which is coated with a hydrophobic, semi-permeable membrane such as ethyl cellulose.

The semi-permeable membrane coating is designed to prevent the retention of excess liquid sample on the test device and therefore provide more reliable semi-quantititive results. The patent teaches that a reasonably wide range of concentrations of materials can be used and more specifically discloses the application of the semi-permeable membrane from a solution containing 0.5 to 1 g of film forming polymer, e.g., ethyl cellulose, per 100 ml of organic solvent.

Test strips made by the process in the '668 patent are highly effective for the intended purpose. However, a phenomenon known as "windowframing" or "edgewetting" has been observed with these strips. Windowframing may reveal itself through the development of very small isolated wet spots around the margin of the strip after dipping. In some cases, a continuous wet margin around the reacted strip may be observed. The windowframing problem is strictly a phenomenon occuring at the edge of the strips, since the strips are prepared by cutting a sheet of the bibulous carrier after it is coated. Since the wet spots or wet margin are invariably of a different color from the rest of the reagent strip, visual matching of the reacted strips to color charts becomes difficult and may possibly lead to incorrect results. One solution to the problem has been to carefully control the ethoxyl content and viscosity of the ethyl cellulose. This technique introduces an additional parameter which increases quality control and raw material procurement problems.

It would be desirable, and it is an object of this invention, to provide a novel process for the preparation of the previously described glucose test strips which eliminates the problem of windowframing.

An additional object is to provide such a process which eliminates the problem of windowframing without slowing down the test strip's reactivity to an unacceptable rate.

A further object is to provide a process which permits the use of a wider range of ethyl cellulose formulations than was possible using prior art coating techniques.

An additional object is to provide an improved test strip prepared by the process described herein.

The present invention is an improvement in the process of preparing a test device for the semi-quantitative determination of glucose in urine which process comprises impregnating a sheet of a bibulous carrier with a solution containing glucose oxidase, a substance having peroxidative activity and a water soluble iodide salt, drying the treated carrier and then saturating the carrier with an organic solution of a film-forming material, drying the carrier to leave a dry residue of the film forming material and cutting the sheet into test strips. The improvement involves the use of a solution containing from about 1.75 to about 4.55 g/100 ml of ethyl cellulose in an appropriate solvent as the organic solution of film-forming material.

The method of preparing the test device of the present invention and the improved performance of such a device are illustrated by the following examples in which all percentages are by weight:

The following solution was prepared:

84.0 ml. stock ethylene diamine tetraacitic acid (EDTA) buffer.

1.8 g potassium iodide.

60.0 ml glucose oxidase (1054.5 International Units/ml.).

9.0 ml 10% polyvinylpyrrolidone (PVP) in water.

18.0 ml 10% Gantrez in water.

0.005 g FD&C Blue #1.

0.2 g peroxidase.

The stock EDTA buffer was prepared by mixing 19.2 g of citric acid, 93.6 g of sodium citrate and 72.0 g of Versene (tetrasodium EDTA) in 1008 ml of distilled water. The purpose of the buffer is to maintain the pH at a level of from 3.4 to 7.2 when the device is contacted with the fluid to be tested.

In operation, the enzyme glucose oxidase catalyzes the aerobic oxidation of glucose to produce gluconic acid and hydrogen peroxide. In addition to the enzyme glucose oxidase the system must include a material having peroxidative activity hence the use of peroxidase.

In addition to glucose oxidase and peroxidase, an iodide salt, in this case potassium iodide, is necessary as a chromogen. The color forming reaction is represented by the following equations:

Further to the utilization of an iodide salt as chromogen, a second indicator substance is preferably used to provide a gradient response to the presence of various concentration of glucose in the urine to be tested. In the above formulation FD&C Blue #1 has been chosen as the second indicator.

Polyvinylpyrrolidone and Gantrez (a copolymer of methylvinylether and maleic anhydride) are used as thickening agents which also function as enzyme stabilizers.

Sheets of Whatman 3 MM paper (approximately 7.6 cm.times.17.8 cm) were dipped into the solution and dried at 90.degree.-95.degree. C. for 10 to 15 minutes. The dried pieces of paper were then dipped into solutions of ethyl cellulose in 20/80 ethanol/toluene (v/v) in which the concentration of ethyl cellulose ranged from 0.75 to 4.55 g/100 ml. The selection of solvent is not critical provided that the material will solubilize ethyl cellulose and will evaporate from the device after application to leave a uniform residual layer of ethyl cellulose. Other suitable solvents include toluene, benzene and various combinations of toluene and ethanol.

After treatment with the ethyl cellulose solution the paper sheets were out into strips and the test strips were evaluated for their windowframing properties and reactivities with the results of this evaluation being set out in Table I.

From Table I, it can be determined that an ethyl cellulose concentration below 1.55 g/100 ml is unsatisfactory because of windowframing. At a concentration of 1.75 g/100 ml the windowframing phenomenon was observed to diminish and completely disappear as the concentration approached 3.35 g/100 ml. Thus, a minimum concentration of about 1.75 g/100 ml is necessary to provide test strips which are entirely suitable for their intended purpose. At higher concentrations of ethyl cellulose windowframing was not a problem, but as the concentration was increased the reaction slowed so that the use of a solution containing greater than about 4.55 g/100 ml would be impracticable.

The most desirable concentration range is from 2.35 to 3.35 g/100 ml since this will leave an ethyl cellulose residue thick enough to inhibit windowframing without significantly slowing down the reaction.

Pieces of Whatman 3 MM paper were impregnated in the first solution and dried as described in Example I. A second solution was prepared by dissolving 4.0 g/100 ml of ethyl cellulose in a 20/80 ethanol/toluene solution. The following types of ethyl cellulose obtained from Hercules, Inc. were used:

1. K-type: ethoxyl content (46.7%)

2. N-type: ethoxyl content (47.7%)

3. T-type: ethoxyl content (49.5%)

Pieces of the paper impregnated with the first solution were again impregnated in the above solutions and dried. Strip devices were prepared from these papers and evaluated for windowframing and reactivity. All three formulations showed identical reactivity and no windowframing. It can be concluded, therefore, that at an ethyl cellulose concentration of 4.0 g/100 ml, the ethoxyl content may range from 46.7% to 49.5% without fear of windowframing.

The experiment of Example II was repeated using the following types of ethyl cellulose obtained from Hercules, Inc.

1. N-22: viscosity* (20); ethoxyl content (47.8%)

2. N-50: viscosity (50); ethoxyl content (47.7%)

3. N-100: viscosity (95); ethoxyl content (47.7%)

4. N-200: viscosity (206); ethoxyl content (47.6%)

Test strips prepared from these materials indicated no reaction differences between them and no windowframing. This result indicates that the molecular weight of the ethyl cellulose selected for use is not a limiting factor.

The following samples of ethyl cellulose were used at the 3.0 g/100 ml level in 80/20 toluene/ethanol without observing any windowframing.

From the above examples it can be determined that strip devices can be prepared according to the method of the present invention which do not present windowframing problems. In addition ethoxyl content and viscosity of ethyl cellulose are no longer parameters controlling windowframing and performance of strips.

Quality product can be manufactured on a consistent basis thereby easing quality control problems. By eliminating the need to control ethoxyl content and viscosity, the types of ethyl cellulose available for use become more extensive. This enables one to select an ethyl cellulose material which possesses other beneficial properties. For example, ethyl cellulose T-10 from Hercules, Inc. is preferred for its high solubility and low viscosity which makes it easy to handle. These properties will also promote uniform coatings due to better impregnation of the filter paper to thereby effect better performance.