Medical Device Link.

IVD Technology Magazine
IVDT Article Index

Originally Published September 2000

Effects of adhesive migration in lateral-flow assays

Kevin D. Jones, Anne K. Hopkins

The materials used for lateral-flow diagnostic tests have two common characteristics: they are porous and relatively weak. To compensate for their lack of physical strength, such materials have traditionally been mounted onto support cards, usually by means of an adhesive (see Figure 1). In the past, product developers have used almost anything sticky to mount their materials, including thermal setting adhesives.

Today, advances in adhesive technology have made available a wide range of adhesive products that are appropriate for use in lateral-flow test applications. By selecting an adhesive that functions compatibly with the porous materials used in lateral-flow test strips, product developers can help to ensure the proper operation of such tests throughout their planned shelf life.

For the vast majority of lateral-flow applications, most product developers now specify pressure-sensitive adhesives (PSAs), which present relatively few complications for test design, processing, or use. When an appropriate adhesive formulation is used, PSAs have proven to be a reliable means of bonding together the component materials of lateral-flow tests.

Figure 1. Diagram of a typical lateral-flow assay showing the location of adhesive layer used to bond porous membrane materials to a supporting substrate.

In a typical lateral-flow application, bonding of the component materials is accomplished when the adhesive penetrates into the porous materials, thereby linking them together. This process of adhesive migration under normal conditions is known as cold flow. Because no heat is applied during the process of laminating a PSA to other components of a lateral-flow test, some degree of cold flow is essential for the formation of a bond between the materials. However, an adhesive that exhibits either inadequate or excessive cold flow can present challenges for the developers of such diagnostic products.

To ensure proper manufacturing and performance of their lateral-flow tests, product developers should therefore study what type of adhesive to use and select one that offers the best compromise between bond strength and the degree of adhesive migration. When product developers study, test, and allow for the influences of adhesive migration, the effects resulting from this phenomenon are usually minimal.

The Scope of Migration Effects

If left unexamined, however, excessive adhesive migration can affect many of the components of a lateral-flow assay. If the level of cold flow is too low, the result can be a low initial bond strength that may result in inadequate bonding of the component materials. To obtain a higher bond strength, the product developer must therefore specify an adhesive that offers a higher degree of cold flow.

On the other hand, a level of cold flow that is too high can result in blocked pores, hydrophobic patches, and material rewetting problems that may interfere with the performance of the test. These symptoms can result from the migration of the adhesive into the porous materials of the test after they have been bonded together, especially during long periods of shelf storage.

In many cases, problems associated with cold flow can be resolved through the use of direct-cast membranes. For instance, such membranes can eliminate bottom-up adhesive migration during storage because the supporting plastic sheet prevents adhesives from entering the pores of the membrane. Nevertheless, there are cases when the differences in processing or performance characteristics between supported and unsupported membranes are so great that developers must use unsupported membranes.

Even when direct-cast membranes are used, however, excessive adhesive migration can interfere with the proper functioning of other test components—including sample wicks, conjugate pads, and absorbents. If a sample wick is used to control sample volume within a test, for instance, adhesive migration into the wick can affect the total volume of sample available. Similarly, when a test requires a filtration step (as in the case of tests requiring blood separation), migrating adhesive can block a membrane filter, making less sample available for the test.

Adhesive migration can also occur in tests where the manufacturer has specified a plastic film to be used as a protective layer on top of the test membrane—a practice that is becoming increasingly common. Any top-down adhesive migration occurring in tests with these protective laminates will have a more significant effect than in unprotected tests. Excessive adhesive migration can interfere with the development of the test line that must be viewed through the laminate, resulting in a visible reduction in the intensity of the line.

Manufacturers that are developing membrane-based quantitative tests can also encounter challenges in dealing with cold flow. If migrating adhesive blocks part of the porous structure it can reduce the bed volume of the material, thus affecting the calibration of the test. The current generation of quantitative instruments read tests by optical means (either reflectance or transmittance). In either case, obstruction of part of the membrane can change the amount of sample and conjugate flowing through the capture zone, and thus alter the sensitivity of the test. Migrating adhesive that blocks a part of the membrane structure can also change the amount of material available to absorb or reflect light, thereby affecting test readings.

Similar effects can occur in magnetic particle-based systems, where test readings are based on the total amount of conjugate bound in the capture zone. If the available area of the capture zone is blocked with migrating adhesive, the test reading will decrease accordingly. Moreover, such readings can deteriorate further with continued adhesive migration during storage, so that over time the same concentration of sample will result in different readings.

The presence of hydrophobic spots on test membranes can have a significant effect on test operation regardless of the method used for quantification. Patches in the membrane that do not wet out will lead to variations in results, especially if the hydrophobic patch is within the region of the capture zone that is being measured.

Adhesive Migration Studies

Pressure-sensitive adhesives are categorized according to their relative degree of hardness. An adhesive with high hardness possesses a low level of cold flow and low initial bond strength. An adhesive with low hardness possesses a high level of cold flow and higher initial bond strength.

To study the effects of different adhesive formulations on lateral-flow materials, three adhesives were produced with differing levels of hardness (see Table I). A 23-µm layer of each adhesive under investigation was applied to a plastic support card, onto which typical diagnostic materials were then laminated. These materials included nitrocellulose membranes in a range of pore sizes, cellulose membranes, and glass fiber products (see Table II).

The diagnostic materials were laminated at a constant pressure (20 kPa) and placed in sealed foil pouches. Sets of the prepared pouches were then stored in a temperature-controlled room (at 20°C) or placed in an incubator (at 37°C). Samples were removed at regular intervals and tested for the range of properties under investigation.

A model lateral-flow immunoassay was used to assess the effects of the various adhesive formulations. Lateral wicking rate was assessed by measuring the time required for deionized water to migrate a known distance (7.5 cm for glass fiber and cellulose membranes; 2 cm for nitrocellulose membranes).

The performance of the nitrocellulose membranes was also tested for protein application, capture-line intensity, and hydrophobic patches by using a model lateral-flow assay (monoclonal anti-ßhCG capture and conjugate with samples of urine spiked with hCG). All tests were repeated 10 times on two separate pieces of the material.

The laminated materials were also analyzed using a scanning electron microscope (SEM). For this examination, cross sections of the laminated materials were cut to permit measurement of the thickness of the adhesive layer.

Study Results

When compared with the results of wicking tests conducted on unlaminated materials, the results of this study indicated that the lateral wicking rate increased over time for all the materials that were laminated onto an adhesive card (see Table III). The amount of the increase depended upon the hardness of the adhesive used in laminating the materials. Following is a summary of the test results for wicking time, inspection for hydrophobic spots, capture-line intensity, and depth of adhesive migration.

Figure 2. The effect of adhesive type on lateral wicking of PuraBind membranes.

Figure 3. The number of hydrophobic patches seen on PuraBind membranes after three months storage at 37°C.

Wicking Time. The time required for a sample to wick a predetermined distance was measured for all samples after storage for three months at 37°C (see Figure 2). In this study, the most serious effects on wicking time were those involving nitrocellulose membranes. When an 8-µm nitrocellulose membrane was laminated using the adhesive with the lowest hardness rating (RHC), the time required for the sample to wick 2 cm increased by as much as 45%. Use of the intermediate adhesive (RHB) resulted in an increase of 14%, while use of the hardest adhesive (RHA) resulted in an increase of only 6%.

Similar effects were observed for all the other laminated membranes evaluated in this study. In all cases, use of the adhesive with the lowest hardness rating (RHC) resulted in significantly longer wicking times.

Inspected after three months at 37°C, the laminated cellulose membranes (3MM) showed very little change in wicking time. There was no significant observed increase in wicking time for the samples that used either the hardest (RHA) or intermediate (RHB) adhesives. The samples that used RHC adhesive showed only a 5% increase in lateral wicking time.

The laminated glass fiber membranes exhibited an intermediate level of performance. In the samples laminated using RHC, the observed increases in wicking time were significant (GFD membranes incurred a 20% increase; F075-174 membranes a 27% increase).

Hydrophobic Spots. When a lateral-flow test is performed, users occasionally observe small spots that do not wet out. This occurrence commonly results from imperfections in the way that the liquid runs through the membrane, which can be caused by either pore occlusion or an increase in the hydrophobic character of the membrane.

Figure 4. SEM showing migration of RHA adhesive into nitrocellulose membrane (PuraBind AF) after one month at 37°C.

Figure 5. SEM showing migration of RHB adhesive into nitrocellulose membrane (PuraBind AF) after one month at 37°C.

Figure 6. SEM showing migration of RHC adhesive into nitrocellulose membrane (PuraBind AF) after one month at 37°C.

The occurrence of hydrophobic spots can typically be prevented by blocking the membrane after the capture line has been applied. In this study, however, the laminated membranes were left unblocked in order to permit comparisons to be made. For evaluation purposes, 20 5 x 25-mm test strips were cut for each membrane. The test was then performed, and the hydrophobic spots were counted (see Figure 3 and Table IV).

The fewest number of hydrophobic spots appeared on the control sample and on those membranes laminated using the hardest adhesive (RHA). The membranes laminated with RHB adhesive showed a slightly greater number of spots. The greatest number of spots appeared on the membranes laminated using the adhesive with the lowest hardness rating (RHC). The hydrophobic character of those laminated membranes also continued to change during storage.

Capture-Line Intensity. In this study, performance of the model assay as measured by the intensity of the capture line was also found to be directly related to the type of adhesive used to laminate the test components. Laminates formed using the hardest adhesive (RHA) exhibited no capture-line problems. However, laminates formed using the adhesive with the lowest hardness rating (RHC) exhibited capture lines that appeared uneven and patchy. These membranes also showed areas of inconsistent rewetting and in some cases white spots that had not rewetted when the sample was applied.

Depth of Migration. After the prepared samples had been stored for one month at 37°C, SEM analyses were performed to evaluate the degree of adhesive migration (see Figures 4–6 and Table V). In all cases the original coating of the adhesive was 23 µm thick.

The SEM examinations showed that the depth of the adhesive layer increased for all of the samples during the month of storage. Although it was initially hypothesized that the degree of adhesive migration would be determined by the matrix through which the adhesive was moving, the SEM examinations revealed that the extent of adhesive migration was largely independent of the substrate or its characteristics.

Instead, the degree of adhesive migration turned out to depend mostly upon the nature of the adhesive under investigation. The hardest adhesive (RHA) showed the least migration, with the average depth of the adhesive layer increasing from 23 to 25 µm. With decreasing hardness, the depth of adhesive migration increased: for materials laminated with RHB the layer increased to approximately 30 µm; for materials laminated with RHC the layer increased to approximately 45 µm.

Study Implications

The results of this study indicate that porous materials can be affected by cold flow when product developers specify the use of an adhesive with a relative hardness that is too low. As the relative hardness of the adhesive increases, the degree of migration is reduced.

Similarly, the results suggest that adhesives with the highest levels of migration also cause the largest number of related effects. For instance, the adhesive with the lowest hardness level (RHC) not only had the highest level of migration, but also caused the greatest increase in wick time and produced the greatest number of hydrophobic spots.

The most significant real-time effects of adhesive migration are due to the increase in lateral wicking rate, the development of hydrophobic spots, and the reduction of the bed volume of the membrane. In line with the third of these causes, all of the effects observed were greatest when the membrane material was thinnest (e.g., nitrocellulose membranes were affected more than either the cellulose or the glass fiber grades that were investigated).

Although the membrane and conjugate-release pads are the most critical components for the performance of a lateral-flow test—and the ones studied for this article—it should be noted that adhesive migration occurs in all directions.^1,2 Thus, adhesives may also migrate sideways to the cut edges of the laminate, where they can then come into contact with and potentially adhere to other materials.

When assays are stored at elevated temperatures, the effects associated with selection of an inappropriate adhesive can be seen relatively rapidly (approximately one month). Under real-time conditions, however, such effects take longer to develop. And some effects are more readily observed than others. For laminates produced with RHC adhesives, wicking and hydrophobic effects can often be observed only after approximately five months (at 20°C), and it can take even longer for laminates produced with RHB adhesives.

Importantly, the effects observed in this study all appear within the period commonly accepted as a reasonable shelf life for a lateral-flow test. Most lateral-flow tests have a shelf life in excess of 12 months. But if it is going to cause any problems at all, excessive adhesive migration will likely begin to show its effects within that time.

Such migration will also have a significant impact upon semiquantitative or quantitative assays. In qualitative tests the bed volume of the membrane and change in wicking rate may have limited significance. But in quantitative assays any change in bed volume will also change the amount of sample passing through the capture zone, and this factor can affect the calibration of the test. Any change in wicking rate will also have a direct impact on the sensitivity of such tests. Obviously, these are issues that will need to be addressed by those who are developing the new generation of quantitative lateral-flow assays.

Conclusion

Product developers have many options to consider when they are designing a lateral-flow assay.³ They must select not only the conjugates and reagents that ultimately provide the test's diagnostic information, but also the substrates and other components that enable the chemistry to work. Whenever necessary, developers should undertake design testing to ensure that all components of their test systems are compatible with one another and are optimized for factors such as manufacturability, stability, and reliability.

By investigating the relative hardness of adhesives chosen for use in diagnostic assays, researchers should be able to predict the ways in which such adhesives will interact with attached substrates. With such information in hand, product developers can then select an adhesive that will provide the bond strength required by their test without incurring excessive migration or its related effects.

Selecting an appropriate adhesive can significantly increase the shelf life of finished diagnostic tests. Accelerated aging tests can provide useful information about the potential compatibility and functioning of test components over extended periods. To firmly establish the shelf life of a particular lateral-flow test, however, researchers should perform adequate shelf-life testing. As suggested by this study, such testing should include examination of the test device for effects related to adhesive migration.

References

1. Kevin D. Jones, "Troubleshooting Protein Binding in Nitrocellulose Membranes, Part 1: Principles," IVD Technology 5, no. 2 (1999): 32–41.

2. Kevin D. Jones, "Troubleshooting Protein Binding in Nitrocellulose Membranes, Part 2: Common Problems," IVD Technology 5, no. 3 (1999): 26–35.

3. Alan Weiss, "Concurrent Engineering for Lateral-Flow Diagnostics," IVD Technology 5, no. 7 (1999): 48–57.

Kevin D. Jones is manager for diagnostic technology and Anne K. Hopkins is a senior technologist at Whatman International (Maidstone, Kent, UK). This article originated in a poster presentation at the American Association for Clinical Chemistry Oak Ridge Conference (Boston, MA, May 2000), which was cosponsored by IVD Technology.