The dissertation

D 1 De Groot AC Thesis Adverse Reactions To Cosmetics
PDF – 6,4 MB 145 downloads

The seed is sown

The saga begins in the years 1976-1979, when I was training as a dermatologist in Groningen. The allergy department was managed by staff member Dr. Johan P. Nater. He had first practiced in The Hague, but had come to Groningen when my trainer, Prof. dr. A.H Klokke, became professor and head of the department in 1971. Nater's great wish was to get a special lectorate or professorship in dermatoallergology. Apparently Klokke had promised him that, but it never happened, possibly due to opposition from Klokke, who, it was rumored, did not want any competition. Johan and I got along very well during my internship in the allergy department and afterwards. He invited me to co-write his chapter 'Drugs used on the skin' in the book Meyler's side effects of drugs. The collaboration  has led to a large number of joint publications, including a book with the somewhat pompous title Unwanted effects of cosmetics and drugs used in dermatology, which has had 3 editions.

At the end of my training, in the autumn of 1979, Klokke asked me to stay and succeed Dr. Kruizinga as chef de policlinique. Johan really wanted me to stay, but he was honest enough to say 'but then you have to be able to get your PhD, Ton, otherwise you'd better start for yourself'. However, Klokke didn't give me that chance and so I left. Johan and I kept in touch afterwards. He was very lonely in the clinic and called me regularly, when I practiced in 's-Hertogenbosch, during my consultation hours to have a chat. That was sometimes difficult, because my practice was terribly busy, but I was grateful to him and I have never cut off the conversations.

In 1982 Johan suggested that we also discuss the side effects of cosmetics in our chapter 'Drugs used on the skin'. I didn't feel like it at first, but agreed and in 1983 our chapter 'Drugs used on the skin and cosmetics' appeared in Side effects of drugs Annual 7. It turned out to be a golden move. I used to test all my patients who came for eczema with 'patch tests' (epicutaneous tests, epicutaneous allergy tests), not only with the European standard series, but also with substances that they had skin contact with. For women, these were usually cosmetics and I routinely patch tested a 'cosmetic tray' that contained known allergens from cosmetics. I regularly found positive patch test reactions to one or more of those products. The patient was of course then advised not to use these cosmetics anymore. However, a major problem was that in most cases the ingredient responsible for the allergic reaction was unknown. The ingredients back then did not have to be stated on the packaging or a leaflet, as is currently the case ('ingredient labelling').

Sometimes it was clear on the basis of the results of the European standard series and the cosmetics series which component could be the allergen. The standard series contained some perfume mixtures, lanolin alcohol (wool alcohols, a base ingredient), p-phenylenediamine (hair dye), parabens, formaldehyde and quaternium-15 (all three preservatives), and my cosmetics series contained tosylamide/formaldehyde resin, a component of and major allergen in nail polish. In the case of a positive reaction to one of those substances, it was sometimes plausible that this was the allergen (or one of the allergens) in the cosmetic, for example the perfume in perfumed products.

Positive patch test reactions to cosmetic creams                 

Sometimes, however, there was a positive patch test only to a cosmetic product and none in the standard series or cosmetic series. Then it was unclear what the cause was. In such cases, I would send the cosmetic product to the Keuringsdienst van Waren ('Goods Inspection') (Head: Dr. Dhiam H. Liem, later Dr. Jan Willem Weijland, who would become one of my paranymphs). He would then contact the manufacturer or importer of the product and request all ingredients. When the manufacturer/importer cooperated (which was far from certain) these chemicals were diluted to the correct concentration with the appropriate solvent (as far as they were known, otherwise we made an educated guess) and sent to me after which  

we patch tested these ingredients on the patient. Sometimes - bummer - no positive reactions were observed; this can have many causes, but it goes too far to explain this here. Sometimes one or more reactions were observed, mostly to a 'known allergen', an allergen that had already been described as such in literature. Occasionally, however, we found a new allergen, not previously described as the cause of an allergic reaction. That was great fun, of course, which would  result in a new publication ('As far as we are aware, ... has not previously been described as an allergen in cosmetics').  I discovered as new allergens the UV-filter bornelone, the emulsifier oleamidopropyl dimethylamine, the antibacterial agent benzoxonium chloride, Persea gratissima oil (avocado oil) and later a few more, 7 or 8 in total.

Two trumps for a doctoral research

Because of my interest in cosmetic allergy, I gradually rolled into a run-up project to a PhD on the subject of 'Cosmetic side effects' and I started making plans with Johan, who would become my co-supervisor. Of course I couldn't possibly collect enough data in my own private practice for a full and high-quality dissertation. However, we had two important trumps. The first was a long-running research project in Sellingen, Groningen, and the second asset was the Committee Contact Dermatoses of the Dutch Association for Dermatology and Venereology.

The first trump card: in the village of Sellingen, a cohort of people from the general population was regularly examined for lung problems by the Academic Hospital Groningen; the dermatology department sometimes participatred in this project with a dermatological research subject. After the lung examinations in 1985, the participants were asked whether they had ever experienced a side effect of cosmetics in the previous 5 years. Soap, shampoo and bath foam were also counted as cosmetics. Redness, 'rash' or eczema, but also mild symptoms such as itching, chafing, dry skin   

I've never been to Sellingen myself, two students from Groningen did the work for me

and burning were considered side effects. Those who answered affirmative were then interviewed by two medical students to obtain details. Just over 12% indicated that they had indeed had such reactions, of which 2/3 were women (while there were more male  participants). Despite the fact that most reactions were described as mild , yet 30% had gone to the GP because of their complaint(s). We concluded that side effects of cosmetics are far from rare in the general population, especially among women.


Then came the idea of conducting a similar study among clients of beauticians. We assumed that more side effects would be found in this population, on the one hand because it almost exclusively concerns women and on the other hand because they come into contact with cosmetics above average. But how was I supposed to handle that? I approached a large number of beauticians in 's-Hertogenbosch and the surrounding area by telephone in the evenings. I explained who I was, what my goal was, that I would like to enlist their help and of course what their task would be. They would have to ask their clients whether they had ever experienced side effects from cosmetics in the past five years. Those who answered 'yes' to this question were then further interviewed using a questionnaire. The results would  later be verified by me or one of the students and then the clients were invited for patch tests. We assumed that only a small minority of all side effects of cosmetics is due to contact allergy, and that most reactions are caused by irritation from soap, bath foam, etc. However, to verify this, allergy tests needed to be performed.

Exude enthusiasm!

What I have learned from this project is that in order to get people's cooperation (it was extra work for the beauticians), one has to be enthusiastic or at least appear enthusiastic. There were evenings when I was very tired after a long day of practice or dreaded explaining the whole story on the phone for the umpteenth time. I soon noticed that just then the number of beauticians who were willing to cooperate was small, most of them refused. So from that moment on I forced myself, no matter how tired I was (I had 2 very busy practices in Carolus and Willem-Alexander hospital and a baby and a 3-year-old at home), to appear enthusiastic in language and tone and that worked wonderfully. Eventually, 25 beauticians participated in the study. In a period of 5 months they interviewed 982 of their clients and 254 (26%) gave a positive response of having had a reaction to cosmetics. Of these 254, I invited 172 for patch tests (the rest lived too far away), 160 agreed of who 150 showed up. In only 7% of them, contact allergy was proven or considered plausible based on the allergy tests. This confirmed what we already had hypothesized: irritation is much more important than allergy in cosmetic product side effects. Incidentally, it was a huge logistic job, because regular patient care had to continue and everyone had to be approached personally. Hospital employees wondered what I did with all those carts with bunches of flowers that I drove to my outpatient clinic on Thursdays (when the allergy tests were read) for several months: a gift to the participants for their willingness to participate in the study! All beauticians received a thesis with a personal message of thanks from the author!

Allergic contact eczema due to Balsam of Peru. Balsam of Peru is the balsam obtained from the bark of the tree Myroxylon balsamum (L.) Harms var. pereirae (Royle) Harms. This tree grows in the coastal areas of El Salvador as well as Nicaragua, Honduras, Guatemala, Cuba, Mexico, Costa Rica and Panama at altitudes between 300 and 700 meters above sea level. Contrary to what the name suggests, Balsam of Peru does not come from Peru (from where it used to be exported by the Spaniards) but from El Salvador. The substance has been used for all kinds of medicinal purposes since the Middle Ages. Locally it is (was) used for the treatment of, among other things, chronic ulcers, poorly healing wounds, decubitus, eczema, itching, hemorrhoids, scabies, frostbite, and diaper rash. In 2016, the European Medicines Agency concluded that there is no sufficient evidence of effectiveness for any medical indication. Due to the great risk of contact allergy, Balsam of Peru has long been banned from use in cosmetics.


You can read all about Balsam of Peru in the review article I wrote about it in 2019:

De Groot AC. Myroxylon pereirae resin (Balsam of Peru) – A critical review of the literature and assessment of the significance of positive patch test reactions and the usefulness of restrictive diets. Contact Dermatitis 2019;80:335-353.

Perubalsem Overzichtsartikel
PDF – 2,2 MB 133 downloads

Committee Contact Dermatoses

Next, the second asset: the Contact Dermatoses Committee. This was a committee of the Dutch Society for Dermatology and Venereology, in which all academic centers were represented and which also included a small number of dermatologists in private practice, including me. After Prof. Suurmond had stepped down from its Board, I became chairman of the committee in 1984 (and continued to be until 1999). With the help of a number of committee members, I was now able to examine large numbers of patients with suspected allergy to cosmetics. With Johan Nater I wrote the research protocols and, through the cosmetics section of the Keuringdienst van Waren in Enschede, I provided the necessary test materials, which were sent to the participants for patch testing in their patients with cosmetic allergy. In two other studies, patients suspected of contact allergy/cosmetic allergy were tested with preservatives and perfume ingredients, which at the time were considered (and proved to be) the main causes of cosmetic allergy. From these studies, a new major allergen emerged: Kathon CG, a preservative, which would become Chapter 4 in my dissertation.

The allergens in cosmetics

An important part of my PhD research focused on what exactly are the allergens in cosmetics, i.e. the ingredients that  cause allergic reactions. First I performed a retrospective study of the results of allergy testing in my own patient population from the period 1981-1985. It showed that the most important allergens were perfume raw materials, preservatives (especially Kathon CG), the emulsifier oleamidopropyl dimethylamine (in baby body lotion of a well-known brand), tosylamide/formaldehyde resin (in nail polish), lanolin and p-phenylenediamine in hair dye. We then conducted a prospective study with the members of the Committee Contact Dermatoses, in which 119 patients with proven cosmetic allergies were tested with all ingredients of the cosmetics that had caused allergic contact dermatitis. This research broadly confirmed my own data. 53 causative allergens were identified, of which Kathon  CG  (n=33), various perfume ingredients (n=39), tosylamide/formaldehyde resin (n=15) and oleamidopropyl dimethylamine (n=13) were by far the most important.

Kathon CG is a 3:1 mixture of methylchloroisothiazolinone and methylisothiazolinone. In the second half of the 1980s and the first half of the 1990s it became a major cause of cosmetic allergy; is was Chapter 4 in my dissertation

Allergic contact eczema caused by Kathon CG in cleansing milk

In the elaboration of the thesis, chapter 1 became an extensive introduction with a literature review; chapter 2 contained the studies performed in the population in Sellingen and the clients of beauticians; chapter 3 was about the studies of the allergens in cosmetics (with a very extensive literature review of what was already known on the subject); 4 and 5 were Kathon CG and oleamidopropyl dimethylamine; and 6 contained the summaries, conclusions and recommendations. This was followed by a few appendices and an extensive register of all substances, and the book ended with the conclusions and recommendations in Dutch and my Curriculum vitae. The thesis had 231 + ix pages with a hard cover, sewn binding.

Conclusions of the doctoral research

The main conclusions of the PhD research were:

1. Side effects of cosmetics and toiletries (soap, shampoo, bath and shower foam) are not rare: in a period of 1-5 years         they can occur in 10% of the adult population;
2.  Women report side effects almost twice as often; this difference is largely caused by cosmetics applied to the face;
3.  The majority of side effects are mild (although 30% of patients visit their GP) and are caused by irritation, not allergy;       this mainly results from using products for hygiene such as soap, bath foam, shower foam and shampoo;
4.  Almost half of all allergic reactions to cosmetics are caused by skin care products (creams, lotions);
5. The main categories of allergens in cosmetics are preservatives, perfume ingredients and emulsifiers; the main                   allergens are Kathon CG, perfume raw materials, tosylamide/formaldehyde resin and oleamidopropyl dimethylamine.

Recommendations based on the doctoral research

The main recommendations resulting from the PhD research were:

1. Since most of the side effects of cosmetics are based on irritation, more research needs to be done on the irritation           potential of cosmetics and their ingredients
2. Kathon CG must not be used in stay-on cosmetics (products that are not immediately washed off the skin) in a                    concentration of 7 ppm (parts per million) or higher (later it would be completely banned in the European Union in            cosmetic products that remain on the skin)
3. Oleamidopropyl dimethylamine should not be used in cosmetic products in a concentration of 0.3% or higher
4. Research should be done into alternatives for tosylamide/formaldehyde resin in nail polishes (good alternatives were        indeed found later and currently this substance is probably not used anymore)
5. Kathon CG should be added to the European standard series, which is patch tested in all patients with suspected                contact dermatitis.

In addition, 14 substances were shown that – to suit local conditions – could be included in a cosmetic range that is routinely tested in all patients suspected of having a cosmetic allergy.


The cover of the thesis Adverse reactions to cosmetics. I wouldn't choose the color green now. The woman on the left has applied make-up earlier, looks in the (broken) mirror and sees that she has developed a side effect (allergic contact dermatitis) from the applied cosmetics around the eyes. These are drawings made based on a photo of my wife Janny.

Suggested next chapter: