Irritancy Testing
Skin and Eye Irritancy Tests
Many products are tested to see if they will cause damage to the skin or eyes, including cosmetics and toiletries, medical products, industrial and agricultural chemicals, and cleaning products.
This document is a more detailed version of "Cosmetic testing". If you want a simple version, go to Cosmetic testing .
Toxicity testing
Tests for chemicals that cause cancer or birth defects
Skin Irritancy Test
To test whether a substance irritates the skin, sections of the animals' backs are shaved and sometimes abraded. The substance is applied to the skin and covered with gauze patches, usually for 4 hours. Researchers then look for signs of redness, inflammation, weeping or scabs to determine how irritating the substance is. An irritating product can make the skin red raw, which is obviously very painful.
Rabbits are usually used for this test. At the end of the test they are killed.
Draize Eye Irritancy Test
The outer layer of the eye, the cornea, is one of the most sensitive tissues in the body. It is richly supplied with nerve endings, which is why any irritation or damage is extremely painful. Everyone knows how uncomfortable it is to get something like shampoo or onion in the eye. We quickly wash it out. In comparison, the suffering of rabbits is greater, firstly because some the substances tested are more irritating, and secondly because the rabbits can't wash their eyes.
The first problem with these tests is obviously that they are very cruel. Irritation to the skin and especially the eye can be excruciatingly painful.
However, the tests are also inaccurate. In one study the same 12 substances were tested for eye irritancy in 24 well-established laboratories (1). Since the same substances were being tested, using the same method, and the same species of animal, you would expect scores for the degree of eye injury to be similar. On the contrary, scores varied between rabbits in the same laboratory, and varied widely between laboratories. Some substances that were rated as most irritating by some laboratories were rated as least irritating by others.
There are a number of differences between rabbit and human eyes (2):
What do these differences mean in terms of test results? One study compared information about accidental human exposure to products with the results of animal tests using rabbits and monkeys (3). All animal tests, especially the standard Draize test, overestimated how irritating a product was to the human eye. In other words, rabbit eyes were more sensitive than human eyes.
An experienced toxicologist has concluded: "
No single animal species has been found to model exactly for the human eye, either in anatomical terms or in response to irritation
" (4).
There are also considerable differences between human and rabbit skin. When 12 substances were tested on human and rabbit skin, results were similar only for the 2 most irritating products. The remaining 10 products were irritating to the rabbits but not the humans. Rabbits were again more sensitive (5).
In another study, a range of household products and industrial chemicals were tested on the skin of rabbits, guinea pigs and humans (6). Only 4 of the products were non-irritating in all 3 species. However, 12 products were more irritating in one or both of the animal species than in humans. A further 3 products were less irritating in one or both animal species than in humans.
The researchers concluded: "
Neither the rabbit nor the guinea pig provides an accurate model for human skin. The skin responses of these animals differ in both degree and in kind from those of human skin.
"
Animal tests for irritancy are far from perfect, as has also been acknowledged in court. In 1974 a woman in the USA took a shampoo company to court because shampoo that had accidentally splashed into her eye had caused her pain and damage to the cornea. The shampoo was shown to be an irritant in rabbit tests. The judge ruled in favour of the company, saying: "...
rabbit studies, standing alone, do not warrant condemnation of this product.
" (7). If the results of rabbit tests are ignored in this way, why are they done at all?
Many different in vitro (test tube) systems have been suggested as alternatives to animal irritancy tests. The following are only a few examples.
Why is a vegetable protein anything like the eye? The cells of a healthy cornea are organised around fine collagen fibres in a tight mesh structure. When the cornea is damaged by an irritant and swelling occurs, the spaces in the mesh increase and the mesh itself may be disorganised. This changes the way light passes through the cornea and makes it look cloudy rather than clear (8).
The molecules in the Eytex protein gel also have a highly organised structure which is changed by irritants. The more irritating a substance is, the more the structure of the molecule groups is changed, and the cloudier the gel appears.
The Eytex test agrees well with Draize test results, although ideally it should be compared to information about human eye irritation, since the Draize test itself is not accurate. Eytex is particularly good at identifying irritants, more so than at identifying non-irritants.
The second column shows how closely Eytex results agreed with other results, the third column shows what percentage of irritants were correctly identified by Eytex, and the final column shows how many cosmetic ingredients or products were tested.
Unlike in rabbit tests, there is also good agreement between laboratories.
Updated versions of the Eytex system are commercially available from
InVitro International.
There are various ways of measuring damage when an irritating product is applied to this test skin. For example, cells can be examined under the microscope, membrane damage can assessed by leakage of enzymes, or inflammation can be determined by release of interleukins. Whatever method is used, the result can be measured accurately, unlike in animal studies where observers estimate the degree of swelling or redness.
Results from this test so far have agreed well with animal studies, although ideally they should be compared to human information. Reconstructed human epidermis also has other uses, such as burns research, studying healing processes, and understanding mechanisms by which a chemical causes irritation at the cellular level (13).
However, to avoid species differences it would obviously be preferable to use human cells. One problem with using cells from human corneas has been that these cells don't live for very long. Now researchers have found a way of not only increasing the number of these cells, but also extending their life span so that they can be studied in more detail (15). Researchers used human corneas from an eye bank to grow the cells. This cell culture can be used not only to study eye irritation, but also wound healing, parasite infection, and radiation damage in the eye.
In one study there was an 83% agreement between results from the Neutral Red test and the Draize test (17).
Neutral Red test kits are commercially available. For example, you can buy human cells
Clonetics Corporation
.
The cells come in a flask with nutrient fluid. The test substance is added to the fluid, usually for 24 hours. Then the cells are removed and dye is added, usually for 3 hours. Finally the amount of dye absorbed by the cells is measured.
The irritancy of the substance is assessed by measuring the area, in millimetres, of dead cells under the filter paper, that is, cells that have lost their neutral red dye.
In one study using this test there was an 81% agreement with Draize results for 16 substances (18). In a further study of 22 substances there was 100% agreement (19). This test has already been used for some time to ensure that plastics used in medical devices will not be irritating.
As the table shows, the Microphysiometer test rated the irritancy of the 8 chemicals in the same order as the animal tests, with the same kind of increase. This test can also be used to measure the effects of chemicals over time, and whether any recovery from injury takes place.
The molecular structure of known substances is entered into a computer database. Particular chemical structures are linked to particular kinds of chemical activity, in this case irritancy. When a new substance is entered, the expert system tries to match its molecular structure to others in the database. If it finds a close similarity, it predicts that the new substance has the same level of irritancy.
A test using the DEREK expert system with 46 known substances was very successful in predicting their irritancy (21). A New York company called Health Designs has shown that such a system distinguished severe irritants from others in 91.5% of cases. It distinguished non-irritants from others in 93% of cases (2). The Multi-CASE system has successfully predicted the irritancy of 21 chemicals (22).
There is no test which is completely reliable, including animal tests. The best approach is to combine a number of different tests.
In the case of cosmetics, the manufacturer could choose ingredients that have been used for a long time, and so are likely to be safe. If the chemical is a new one, the first test could be the QSAR computer analysis to predict its likely irritancy.
In the next stage, a number of in vitro (test tube) tests could be used. Some tests are better than others for substances of a certain form, for example liquid rather than solid, and for substances of a certain chemical class, for example alcohols rather than oils. The aim would be to chose the best set of tests for a particular substance.
Doing more than one test is not a problem because in vitro tests are so much faster and cheaper than animal tests. For example, an Agarose Diffusion Test costs $50-$100 and takes 24 hours per product, whereas a Draize test costs $500-$700 and takes at least 3 days per product (18).
If the product is shown to be safe by this set of tests, it can then be trialed by human volunteers. This is the final and best test of all. By going through this series of steps products can be guaranteed to be safe without the suffering that is currently inflicted on animals.
Some states have laws limiting the use of the Draize test. Victoria has banned its use for product testing. In NSW it may only be used with the approval of the Minister to test products that are intended for application to the eye, for example, eye drops or ointments. In South Australia it may be used in research which benefits humans or animals when there is no alternative that is less painful.
I would like to see
References
for this document on skin and eye irritancy tests.
Problems with irritancy tests
Alternatives to animal tests
Eytex(TM)
This test for eye irritancy uses a vegetable protein extracted from jack beans. Like the cornea of the eye, this clear protein gel becomes cloudy when in contact with an irritating substance. In the Draize test, people have to estimate the degree of damage caused, that is, how swollen or red part of the rabbit's eye is. This system isn't very accurate. In the Eytex test, the degree of cloudiness ("damage") can be measured by a machine, a spectrophotometer, which is much more reliable.
Study
% Agreement
% Irritants
Substances
(9)
85% (Draize)
89%
101
(9)
92% (human)
12
(10)
78% (EEC labelling)
94%
142
(11)
80% (Draize)
100%
465
Reconstructed human epidermis
This is a multi-layered human skin grown in the laboratory, which can be used to test skin irritancy. It is sold commercially under trade names such as Skin Squared(TM) and Episkin(TM).
Corneal cell lines
The SIRC is an continuous cell line of rabbit corneal cells. These are cells that are now grown in the laboratory, and no further rabbits are killed. When 6 shampoos were tested on these cells, there was very good agreement with Draize results (14). The test assessed how much of a substance was needed to kill half the cells. Obviously the less of a substance that is needed to produce this result, the more damaging it is.
Neutral Red Uptake Test
Normal cells in culture readily absorb and hold this neutral red dye. When the cell membrane, or the lysosomes inside the cell are damaged by an irritating chemical, dye will be lost through the leaky membranes. Less dye will remain in the cell. A spectrophotometer is used to accurately measure how much has been lost (16).
Agarose Diffusion Test
The problem with cell cultures such as those in the Neutral Red Uptake Test is that the cells are in fluid, so only soluble substances can be tested. In the Agarose Diffusion Test a small amount of agarose (a seaweed extract) is added to form a gel layer. Some of the test substance is placed on a small piece of filter paper, which is then placed on the agarose. The substance diffuses through the agarose into the cell culture below.
Microphysiometer
An irritating product will produce changes in the functioning of cells. The microphysiometer is an instrument that detects very small changes in cell metabolism by measuring changes in the pH of the cell culture nutrient fluid (changes in lactate, CO2 production). When the microphysiometer measured how much of a product it took to depress the metabolic rate of human skin cells by 50%, there was very good agreement with animal tests, as shown in the table below (20).
Chemical
Animal irritancy
Microphysiometer
1
Mild
0.1
2
Mild
0.5
3
Moderate - Mild
0.7
4
Moderate - Mild
0.8
5
Moderate - Mild
0.9
6
Moderate
1.7
7
Severe - Moderate
3.9
8
Severe
4.1
Computer modelling
Expert computer systems can be used to predict the irritancy of new substances on the basis of what is already known about the irritancy of substances with a similar chemical structure. This approach is known as Quantitative Structure-Activity Relationship, or QSAR for short.
Human studies
Some cosmetics companies already use human volunteers to test new formulations (for an example visit the
Body Shop
web site). This is the most reliable test of all. Human irritancy can assessed through patch testing, where test substances are placed on small areas of the upper back and covered with a patch for 2 days (23).
Testing without animals
Draize tests in Australia
Toxicity tests
Tests for chemicals that cause cancer or birth defects