DOI: 10.1111/jdv.

15946 JEADV

REVIEW ARTICLE

Mineral oils and waxes in cosmetics: an overview mainly

based on the current European regulations and the safety
profile of these compounds
B. Chuberre,1 E. Araviiskaia,2 T. Bieber,3 A. Barbaud4,*
1

al Cosmetique Active International, Levallois-Perret, France
L’Ore
2
Department of Dermatology and Venereal Diseases, First Pavlov State Medical University of St. Petersburg, St. Petersburg, Russia
3
Department of Dermatology and Allergy, University of Bonn, Bonn, Germany
4
AP-HP.Sorbonne Universite
, Tenon Hospital, Department of Dermatology and Allergology, Sorbonne University, Paris, France
*Correspondence: A. Barbaud. E-mail: annick.barbaud@aphp.fr

Abstract
Mineral oils and waxes are mixtures of predominantly saturated hydrocarbons consisting of straight-chain, branched
and ring structures with carbon chain lengths greater than C14. They have been used for many decades in skin and lip
care cosmetic products due to their excellent skin tolerance as well as their high protecting and cleansing performance
and broad viscosity options. In contrast to vegetable oils, mineral oils are non-allergenic since they are highly stable and
not susceptible to oxidation or rancidity. They have a long history of safe use which is confirmed by clinical and epidemi-
ological data. In Europe, mineral oils are only permitted in cosmetics if compliant with purity specifications on polycyclic
aromatic hydrocarbons and safety requirements laid down in the European pharmacopoeia and the EU cosmetics regu-
lation EC/1223/2009. The high quality of these mineral oils is assured by robust quality assurance and a refining/purifica-
tion process designed to exclude substances with carcinogenic potential and to minimize the presence of mineral oil
aromatic hydrocarbons. Given their highly lipophilic properties, mineral oils do not penetrate human skin and, thus, are
not systemically bioavailable in the body. Moreover, no significant changes in the skin and no effects on any internal
organ system have been reported and attributed to the topical application of refined mineral oils. Regarding potential oral
exposure from cosmetic lip care products, Cosmetics Europe, the European trade association for the cosmetics and per-
sonal care industry, has advised cosmetic manufacturers to only use mineral oil fractions for which recognized food
acceptable daily intake (ADI) values apply. The estimated dose of mineral oils ingested via lip care products contributes
to <10% of the ADI value and should therefore be considered of no toxicological concern.
Received: 17 July 2019; Accepted: 3 September 2019

Conflict of interest
BC is employed by Cosmetique Active International. EA and TB are members of the Scientific Advisory Board of
Cosmetique Active International. AB was a member of an international advisory board for La Roche-Posay in
2017. The authors declare they have no conflicts of interest that might be relevant to the contents of this
manuscript.

Funding source
Medical writing was funded by Cosmetique Active International.

Introduction genotoxic substances. To date, there is no alert on the car-

cinogenic effect of MOSH. The refining process of crude oil
Background consists in various steps, such as distillation, extraction, crys-
Mineral oils and waxes are chemical substances prepared from tallization and purification by acid treatment, hydrotreatment
naturally occurring crude petroleum oil. They mainly consist and/or solvent extraction. The successive purification steps
of mineral oil saturated hydrocarbons (MOSH) and mineral allow to remove impurities and to reduce PAH levels to trace
oil aromatic hydrocarbons (MOAH). The latter could contain amounts to meet specific standards of purity and quality for
polycyclic aromatic compounds, such as polycyclic aromatic pharmaceutical-grade mineral oils as stipulated by interna-
hydrocarbons (PAH), which are potentially carcinogenic and tional and European pharmacopoeia monographs. Moreover,

JEADV 2019, 33 (Suppl. 7), 5–14 © 2019 European Academy of Dermatology and Venereology
6 B. Chuberre et al.

the production processes of mineral oils follow good manufac- ointments and lotions to sunscreens, lip care products and hair
turing practices and quality assurance controls (EU-GMP, gels. Concentrations of mineral oils in cosmetics range from 1%
Directive 2003/94/EC and ISO 9001 or ISO 14001). The final to 99% depending on the product.
refined mineral oils also named ‘white oils’ are complex and In Europe, where the regulation on cosmetics is stringent,
strongly lipophilic mixtures of hydrocarbons of different struc- only highly refined mineral oils and microcrystalline waxes of
tures and sizes (from 14 to more than 90 carbon molecules) pharmaceutical grade, that is complying with the specifications
that condition their viscosity and melting point. Higher of the European Union cosmetics regulation EC/1223/20096 and
molecular mass hydrocarbons increase the melting point of the European Pharmacopoeia, are used in cosmetic products for
the oil until it becomes waxy at room temperature. In general, skin application. Accordingly, the full refining history has to be
highly refined mineral oils and waxes are colourless, odourless known and the starting material should not be carcinogenic
and tasteless liquids or opaque waxes. Due to their various (IP346 method) to prevent exposure to mineral oils with a
physicochemical properties, they are used in many industrial potential concern to health. Residual levels of aromatic com-
applications, including food, pharmaceutical and cosmetic pounds (MOAH and PAH) have to be minimized through the
products. They also have medicinal usage, such as in oily corresponding technical refinement, and levels of PAH, such as
tulles, patch tests and as laxatives (paraffin oil). benzo[a]pyrene to most common PAH, should not exceed
0.005% w/w.6 More specifically, for lip care products which have
Classification of mineral oils The Joint FAO/WHO Expert a high oral exposure, Cosmetics Europe, the European trade
Committee on Food Additives (JECFA) have classified highly association for the cosmetics and personal care industry, has
refined mineral oils and waxes based on criteria related to their advised manufacturers to only use food-grade mineral oils and
viscosity, average molecular mass and carbon number at 5% waxes with an ADI value that meets the associated specifications
boiling point (Table 1). Moreover, the JECFA1 and the Euro- for physicochemical properties and the most stringent purity
pean Food Safety Authority (EFSA)2,3 have evaluated their safety requirements on PAH (i.e. trace levels in the range of
profile and allocated them a specific acceptable daily intake 0.0000001% w/w or 10 ppb) to ensure a safe use for the con-
(ADI) value (Table 1). Thus, mineral oil classes with an ADI sumer.7 Of note, some manufacturers also use food-grade min-
(i.e. microcrystalline waxes and high and medium viscosity Class eral oils for topical products. However, this recommendation
I mineral oils), referred to as food-grade mineral oils, are autho- does not support the use of medium and low viscosity Class II
rized as food additives in compliance with this limit. For med- and III mineral oils in lip care products, as they can be easily
ium and low viscosity Class II and Class III mineral oils, an ADI absorbed orally, and an ADI has not yet been established for
has not yet been established due to the lack of safety data about those classes.7,8
them (the temporary ADI estimated in 2002 for these classes was On cosmetic labels, mineral oil-based ingredients are
withdrawn by the JECFA in 20121). These estimated ADI limits described by International Nomenclature of Cosmetic Ingredi-
are currently under debate.4,5 ents names presented in the following non-exhaustive list: Paraf-
finum liquidum; Paraffin; Isoparaffin; Synthetic wax;
Mineral oils in cosmetics In the cosmetic field, mineral oils and Microcrystalline wax and hydrogenated microcrystalline wax;
waxes are used in various products as they offer broad viscosity Cera microcrystalline; Petrolatum; Ceresin; and Ozokerite. In
options enabling viscosity regulation of a formulation, they have addition to MOSH, cosmetics may also contain synthetic hydro-
protective and lubricating properties avoiding dehydration of carbons, such as polybutene or polyethylene waxes with poly-
the skin, and they are stable and dermatologically well tolerated. olefin oligomeric saturated hydrocarbons (POSH), which are
Those excellent moisturizers and non-allergenic ingredients are structurally similar to MOSH without having any impurities
thus incorporated in many cosmetic products, from skin creams, such as PAH. Such mixtures of MOSH and POSH used as raw

Table 1 Classes of mineral waxes and oils†

Classes of mineral waxes and oils Kinematic viscosity Average molecular Carbon number at Acceptable daily
at 100°C (mm²/s) mass (g/mol) 5% boiling point‡ intake (ADI, mg/kg)
Microcrystalline waxes ≥11 ≥500 ≥C25 0–20
High viscosity mineral oils ≥11 ≥500 ≥C28 0–12
Medium viscosity Class I 8.5–11 480–500 ≥C25 0–12
Medium-low viscosity Class II 7.0–8.5 400–480 ≥C22 None
Medium-low viscosity Class III 3.0–7.0 300–400 ≥C17 None

†Adapted from WHO Expert Committee on Food Additives report in 2012 and the latest European recommendations on ‘Mineral hydrocarbons in cos-
1

metic lip care products’ in 2018.7 ‡This specification means there is no more than 5% of hydrocarbons with a carbon chain length less than the indicated
number.

JEADV 2019, 33 (Suppl. 7), 5–14 © 2019 European Academy of Dermatology and Venereology
Safety of mineral oils used in cosmetics 7

material for lip care products should also comply with the Euro- expected from oral intake of highly refined mineral oils, if lip
pean specifications for either microcrystalline waxes, medium care products comply with the recommendation of Cosmetics
viscosity Class I or high viscosity mineral oils.7 Europe.7,8 However, as shown by random testing, not all prod-
ucts available on the European market strictly comply with this
Rationale of the present article regulation. Thus, such products, probably imported from coun-
Despite the fact that mineral oils have been safely used in cos- tries out of Europe with less stringent or no regulation on cos-
metic products for over 100 years, recent discussions on the metics, should be further scrutinized by health authorities.
safety of mineral hydrocarbons in food and cosmetic lip care The objective of the present document is to give an overview
products have been raised by consumer organizations, media of the current safety data on mineral oils and waxes in cosmetics,
and some national health institutions. Indeed, the EFSA9 and for skin and lip use, based on data available in the recent BfR
the French Agency for food, environmental and occupational opinion, other European and international official authorities’
health & safety (ANSES, Agence nationale de s
ecurit
e sanitaire reports and additional information available in the medical liter-
de l’alimentation, de l’environnement et du travail)10 identified ature.
exposure to MOSH and MOAH, via food and contamination of
food by extensive migration from recycled paper and board Method of data search
packaging, as being of potential concern for human health. In The BfR conducted a comprehensive literature search in various
view of this concern and a possible oral exposure via other scientific and chemistry databases.8 These data are used in the
sources than food, several European consumer organizations current article as well as information from the EFSA, the JECFA
called into question the safety profile of mineral oils in lip and the Canadian health authority. In addition, an updated
balms.11 Following a cooperative testing of a range of lip balms search of safety data on mineral oils and waxes available from
representing the main national and international brands, the publications in peer-reviewed journals has been performed using
European Consumer Organisation Bureau Europ
een des Unions PubMed on 17 January 2019 with the following search terms:
de Consommateurs (BEUC) considered that daily application of ‘Mineral Oil/adverse effects’ OR ‘Mineral Oil/pharmacokinetics’
lip care products represents a major contribution to the overall OR ‘Mineral Oil/pharmacology’ OR ‘Mineral Oil/poisoning’ OR
consumer exposure to MOSH and POSH. Hence, they asked the ‘Mineral Oil/toxicity’ OR ‘MOSH’ AND ‘Cosmetics’ OR ‘Sun-
European Commission to establish health-based guidance values screening Agents’ OR ‘Sunscreen’. The articles were screened by
(ADI or tolerable dose intake, TDI) for mineral oils in lip care two reviewers based on titles and abstracts to select relevant arti-
products, in particular MOSH and MOAH, to reduce the cles. An updated search of data available in grey literature was
amount of potentially problematic mineral oil hydrocarbons in also conducted.
these cosmetics.11 By association and likely because of confusion,
this warning led the lay public and the media to feed the polemic Pharmacological data
of using mineral oils in all cosmetics.
Along this line, the safety profile of highly refined mineral oils Dermal penetration
in cosmetic products has been recently assessed by the German In a recent review summarizing 13 in vitro and in vivo (human
Federal Institute for Risk Assessment (BfR)8 that conducted their and animal) studies on the dermal penetration of mineral oils
own analyses of MOSH and MOAH in various cosmetic prod- and waxes used in cosmetic applications, all studies demon-
ucts. Similarly to the BEUC,11 they found MOAH levels up to strated that the majority of the investigated substances were
the single digit percentage range in petrolatum and several cos- mostly restricted to the stratum corneum of the skin (Table 2).13
metic products for skin and lip application.8 Moreover, many lip In human volunteer studies, petrolatum and paraffin oils were
care products contained >5% or >10% of MOSH with a chain not detected in the epidermis and the depth of penetration into
length <C25,8,11 which is higher than the 5% recommended by the stratum corneum was estimated to be between 6 and 30 lm
Cosmetics Europe for lip care products,7 thus potentially expos- (Table 2). Moreover, the swelling potential of mineral oils did
ing consumers to these more strongly absorbed MOSH. Another not lead to an increased penetration beyond stratum corneum
independent testing also demonstrated that a high number of lip layers. So, there was no evidence from these various studies that
care products (8/57, 14% of tested products) from the German mineral oils and waxes are percutaneously absorbed and become
retail trade contained high MOAH levels (0.4% to 1.6% w/w), systemically available. Consequently, given the high viscosity
similar to levels in mineral oils of technical quality.12 Overall, and hydrophobicity of the mineral oil matrix restricting diffu-
the German BfR concluded from their global risk assessment sion, negligible uptake of MOSH, MOAH and residual PAH
based on a systematic literature review on mineral oils that no components is expected after topical application. Current regu-
health risk for the consumer can arise from the cutaneous appli- latory positions generally agree to consider hydrocarbons with
cation of cosmetic products containing pharmaceutical- and carbon chain lengths >C20 as non-systemically available via the
food-grade mineral oils.8 Moreover, health risks are not to be cutaneous route of exposure.8,14

JEADV 2019, 33 (Suppl. 7), 5–14 © 2019 European Academy of Dermatology and Venereology
 8

Table 2 Human studies analysing the skin penetration of mineral oils and waxes†
Mineral oil type Subject type Mineral oil application Bioanalytical Detection zone‡ Reference
method
Dose Zone Duration Replicate Stratum corneum Epidermis (living

JEADV 2019, 33 (Suppl. 7), 5–14

(mg/cm2) part)
Petrolatum A (5–15 years)§ 17 Thigh 15 and 4 at least Light microscopy Upper layer (at all Sebaceous gland Strakosch
B (20–50 years) Forearms 30 min (Sudan IV to time points) orifice (2–24 h) et al.30
C (55–70 years) Back 1, 2, 4, 6, 12 follow the oil) Upper hairshaft
Abdomen and 24 h (6 to 24 h)
Sebaceous
gland (24 h)
Petrolatum 9 adults 4.5 Volar forearm 30 min 12 per Confocal Raman Outermost layers¶ Minor Stamatas
Paraffin oil 1M/8F (30–60 years) 90 min time point Microscopy (up to 30 lm et al.31
depth)
Paraffin oil 7 infants 30 min
5M/2F (6–10 months)
Petrolatum 6 adults 2 Volar forearm 30 min 3 Laser scanning Skin surface only ND Patzelt et al.32
Paraffin oil 3M/3F (25–50 years) microscopy Uppermost layers
(curcumin follow
the oils)
Petrolatum 6 adults 2 Inner forearm 1h 10 at least Confocal Raman Upper layers (about ND Choe et al.33
3M/3F (av 44 years) Microscopy 6.3  1.2 lm
depth)
Petrolatum 6 adults 2 Inner forearm 1h 10 at least Confocal Raman Upper layers Minor (did not Choe et al.34
3M/3F (23–62 years, (2 areas) Microscopy 7.0  0.8 and reach stratum
av 37 years) (4 methods ††) 6.3  1.2 lm spinosum)
(methods 2 and 4)
20.7  4.9 lm
(method 3)
Paraffin oil 6.8  0.9 and
6.5  1.6 lm
(methods 2 and 4)
15.3  5.3 lm
(method 3)

†Adapted from the review of Petry et al.13 ‡These studies mentioned that petrolatum and paraffin oil were never detected in the dermis or the systemic circulation. §All 3 groups were constituted of
an equivalent number of male and female subjects (exact number not specified in original article). The author noticed that the penetration was faster in the skin of young than older subjects. ¶The
authors mentioned that adult and infant skin reacted similarly regarding lipid uptake. ††The authors mentioned that method 1 was not applicable to mineral oils because of the absence of Raman
spectra marker peaks.
av, average; F, female subject; h, hour; m, months; M, male subject; min, minute; ND, not detected; y, years.
B. Chuberre et al.

© 2019 European Academy of Dermatology and Venereology

Safety of mineral oils used in cosmetics 9

Oral absorption oils in multiple species.17 Cutaneous irritation representing

For cosmetic products, oral intake by a direct (lip care products) treatment-related toxicity was only found in mice, which have
or indirect (hand-to-mouth contact, hand-to-food contact) a more permeable and susceptible skin than the human skin.17
route is also relevant. Mineral oils ingested orally follow the Overall, in animals, no significant changes in the skin and no
route of intestinal resorption of dietary fats. MOSH and PAH effects on any internal organ system have been reported and
are both resorbed by the small intestine primarily into the lym- attributed to the subchronic topical application of refined
phatic system and to a lower extent into the liver portal vein.8 white mineral oils.
Experimental studies in animals showed that the absorption of In humans, patch-test studies with medicinal white oils or
MOSH varies with the carbon chain length, from 90% for C14– waxes showed no skin sensitization (in almost 80 000 patients18)
C18 to 25% for C26–C29, and further decreases with increasing or only mild erythema in sporadic cases even on damaged
carbon number, until above C35 for which absorption is negligi- skin.8,18–20 These cases were never related to cosmetic allergy.
ble.9 MOSH are then metabolized to fatty alcohols and fatty White petrolatum and waxes are hence considered non-sensiti-
acids by both the small intestine and the liver. Due to low bio- zers and rarely allergen, the sensitizing components being proba-
transformation rates, MOSH having carbon number between 16 bly PAH.18,21 Other human studies pointed out that
and 35 may accumulate in different tissues (see below section moisturizers containing 10% of mineral oil have a slight reduc-
Toxicity and cutaneous side-effects). Although there is no exper- ing effect on the minimal erythemal dose (MED) of skin for
imental data on gastrointestinal resorption for MOAH, it can be UVB irradiation (decreased by 5–16%), making the skin slightly
assumed, from their structural and physicochemical properties more sensitive to UV, but this sensitivity effect to the sun was
similar to MOSH and PAH, that they can be resorbed via the similar to that of seasonal changes (average MED increase of
gastrointestinal tract in a comparable manner. The data also 14% between January and April).22 Thus, according to current
indicate that MOAH are well absorbed and extensively metabo- data in humans, there is no robust evidence on dermatotoxicity,
lized before being excreted, thus resulting in the absence of detrimental sun sensitivity and toxicological effects of mineral
bioaccumulation.9,15 oils via the cutaneous route.8,22
In nine healthy female subjects aged 20–31 years and receiv-
ing a single oral dose (1 mg/kg) of a low viscosity Class III min- Oral exposure
eral oil, blood concentration of hydrocarbons (C19–C24) was Acute oral toxicity is not relevant in the context of exposure
below the detection limit of 0.16 lg/mL at all time points (1– via cosmetic products, and besides, MOSH and MOAH have
168 h).16 This toxicokinetic study thus concluded to a negligible low acute oral toxicity.8,9 Regarding subchronic and chronic
absorption of a low viscosity white mineral oil at a dietary expo- toxicity, a number of studies carried out in rodents with dif-
sure of 1 mg/kg, which can reflect daily oral exposure to lip care ferent food-grade mineral oils indicated that repeated oral
products (see below section Exposure estimation to MOSH/ exposure to most of those grades caused bioaccumulation of
MOAH via lipstick use). MOSH in rodent tissues, including liver, mesenteric lymph
In conclusion, mineral oils and waxes are hardly absorbed by nodes and fat.9 The tissue concentrations in rats were consid-
the skin and do not therefore present a systemic health risk to erably lower with microcrystalline waxes and high viscosity
the consumer even after repeated and long-term cutaneous mineral oils than with paraffin waxes, medium and low vis-
exposure.8 Considering oral route, intestinal absorption of cosity mineral oils.5,8,9,23 Moreover, the MOSH accumulation
MOSH in rats was found to be inversely proportional to the in the rat liver was shown to be reversible.5,8,23 While high
number of carbon atoms, and a negligible absorption has been doses of low viscosity mineral oils and paraffin waxes led to
shown in humans with a 1 mg/kg dose of a low viscosity white granuloma or microgranuloma formation associated with
oil.16 MOSH are metabolized to fatty alcohols and fatty acids by inflammatory responses in rat livers, no changes were
the small intestine and the liver. MOAH are well absorbed and observed with the most highly refined mineral oils and waxes.
extensively metabolized by a similar route and do not bioaccu- In the absence of overt clinical toxicity, the JECFA1 and the
mulate. EFSA9 considered those granulomas of little concern from a
toxicological point of view. The EFSA also estimated that no-
Toxicity and cutaneous side-effects observed-adverse-effect levels of 1200 mg/kg/day could be
used for establishing a 100-fold lower ADI guidance value of
Cutaneous exposure 12 mg/kg/day for high and medium Class I viscosity mineral
At least seven rodent studies on subchronic toxicity after skin oils.3,9 A further toxicological study in rats confirmed liver
application of mineral oils consistently showed no adverse weight increase and granuloma induction in a dose- and
effects.8 Indeed, there were no indications of histopathological time-dependent manner, following feeding with different
(mice, rats, rabbits), haematological or cutaneous changes MOSH mixtures representative of the whole MOSH range in
(rabbits) after repeated, long-term, topical application of white human diet.24 Low and medium Class II and III mineral oils

JEADV 2019, 33 (Suppl. 7), 5–14 © 2019 European Academy of Dermatology and Venereology
10 B. Chuberre et al.

(fraction <C25) were more involved in the formation of hep- administration, it is important to note that animal studies did
atic granuloma, whereas high and medium Class I mineral oils not show any indication of granuloma formation in the liver for
(fraction >C25) were more critical to liver weight increase. medium and high viscosity mineral oils and microcrystalline
Following findings in animals,25 MOSH accumulation in tis- waxes, which are the qualities recommended by the Cosmetics
sues has been thoroughly examined in humans (N = 37; Europe for lip care products.7 Additionally, high doses used in
Table 3).8,15,25 Lipogranulomas were detected in samples of animal studies are not relevant for mineral oils used in cosmetic
human fat, as well as of liver, spleen and mesenteric lymph nodes, ingredients. Health authorities also considered lipogranulomas
in correlation with deposits of MOSH15 that were mainly attribu- observed in human tissues as non-adverse effects.1,9 For all those
ted to exposure to mineral oils via food. Both Class I and Class II reasons, no toxic risk can be expected from the oral uptake of
and III mineral oils were detected in human liver autopsy sam- MOSH through lip care products which comply with the Cos-
ples, indicating their accumulation.15 Of note, the level of MOSH metics Europe recommendations.7,8
in fatty tissue was found to increase over time.8,26 However, the
incidence of lipogranulomas in human tissues is low and not Carcinogenicity/mutagenicity
associated with inflammatory reactions or other adverse conse- Mineral oil saturated hydrocarbons are not carcinogens by
quences, leading health authorities to consider these lesions as dif- themselves.9 The carcinogenicity of a mineral oil mixture is
ferent from inflammatory granulomatous alterations in rat livers, related to its content in MOAH, specifically the PAH which are
and thus of no toxicological relevance for humans.1,9 Regarding classified as carcinogenic and mutagenic by the international
the increase in organ weight, it has not yet been investigated in and European health agencies (EFSA, JECFA and IARC) and in
humans. It should also be noted that in pregnant women accordance with the European Classification, Labelling, Packag-
(N = 142), significant (P < 0.05, univariate analysis) but weak ing (CLP) regulation (Regulation EC No 1272/2008). For
correlations (Spearman 0.17 < r < 0.26) were found between MOAH mixtures, there are no dose-response data on their car-
increased MOSH levels in adipose tissue and the use of sunscreen cinogenicity. However, in end-product mineral oils of pharma-
during pregnancy, as well as the use of hand cream and lipstick in ceutical and food grades, PAH are almost completely eliminated
daily life before pregnancy.26 More detailed analyses of these data- through extensive refining. Health Canada confirmed by their
set by the BfR indicated that levels of MOSH in fat tissue were on own analyses that no PAH or only traces (<0.00001%, w/w) were
average 2.2 times higher in women who used these cosmetic detected in a set of topical products containing petrolatum.14
products than in women who did not.8 These correlations could Furthermore, for extracts with a PAH content < 0.7%, the
be explained by direct and indirect oral uptake, but the cause-to- mutagenic index was found to be null.9 Thus, the carcinogenic
effect relationship has not yet been demonstrated. The accumula- potential of the entire mineral oil as a mixture is classified
tion of foreign substances in the human body is undesired, but it according to the valid CLP regulation (Regulation EC No 1272/
is not considered per se as an adverse effect. 2008), stating that mineral oils containing <3% w/w of sub-
Overall, there are no toxic concerns for mineral oils and waxes stances extractable with the solvent DMSO (IP346 method)
through the cutaneous route. Moreover, there is no evidence of should not be classified and labelled as carcinogenic. The weight
a relevant contribution of cosmetic products via skin uptake to percentage of substances extractable with DMSO hereby repre-
the accumulation of MOSH in the human body. Regarding oral sents an indirect measure of the content of PAH in mineral oil.

Table 3 Overall levels of MOSH in various human tissues obtained from autopsies†
Tissue type Measured MOSH concentrations (mg/kg) Tissue mass (mg) Calculated MOSH levels (mg)

Min-max Average Median Min-max Average Median

Fat tissue‡ 17–493 130 87 12 000–66 000 325–12 400 3390 2200
Liver‡ 14–901 131 71 1560 23–1400 205 110
Spleen‡ 6–1400 93 28 140 1–194 13 4
Lung‡ <2–91 12 7 840 0–77 8 4
MLN‡ 21–1390 223 166 nd nd nd nd
Heart§ <2–41 9 6 280 0–12 2 1
Kidney§ <2–12 6 6 300 0–4 1 1
Brain§ <2 nd nd nd nd nd nd

†Adapted from Barp et al.15 ‡Tissue samples of about 7 g were collected from 37 subjects (26M/11F) aged 25–91 years. §Tissue samples of about 7 g
were collected from 14 subjects (9M/5F).
Max, maximal; Min, minimal; MLN, mesenteric lymph nodes; MOSH, mineral oil hydrocarbons; nd, not determined.

JEADV 2019, 33 (Suppl. 7), 5–14 © 2019 European Academy of Dermatology and Venereology
Safety of mineral oils used in cosmetics 11

The threshold of 3% is considered highly conservative and ade- evidence for skin tumour development in humans despite dec-
quately protective for human health regarding carcinogenicity. ades of long-term usage of moisturizers.22
As an example, a low viscosity white oil used in cosmetic prod-
ucts exhibits values <0.2% w/w for DMSO extracts and a muta- Oral exposure
genic index of 0.02, supporting no indication of carcinogenicity The EFSA evaluated the carcinogenicity of the approved white
or mutagenicity.8 These assessments are underpinned by the oils with oral consumption based on one feeding study in rats.2
findings of animal studies summarized below. Medium and high viscosity white oils of food grade were admin-
istered to rats up to a dose of 1200 mg/kg/day for 2 years, and
Cutaneous exposure none of them were found to be carcinogenic.23 For microcrys-
In at least five separate carcinogenicity studies, lifetime topical talline waxes, there are no oral carcinogenicity studies. Neverthe-
exposure of mice to white mineral oils (average dose of less, according to the JECFA, no carcinogenic effects could be
296 mg/kg/day) for up to 2 years revealed no evidence of detected in a 2-year feeding study in rats with paraffin waxes of
tumorigenicity, either at the site of application or in any of similar viscosity.1 Taking into account results for paraffin waxes
the internal organs examined (Table 4).14,17 Moreover, a and for medium and high viscosity white oils, the EFSA con-
recent comprehensive review conducted by the Concawe con- cluded that there is no indication of carcinogenic potential for
firmed that highly refined mineral oils were non-carcinogenic food-grade microcrystalline waxes.3
(three white oils samples tested in CF1 mice) and that the In conclusion, carcinogenicity and mutagenicity of mineral
IP346 method is suitable for the prediction of dermal carcino- oils are caused mainly by some MOAH, including PAH. Cumu-
genic potential of mineral oils.27 Additional animal studies lative evidence demonstrated that highly refined white mineral
have suggested that emollients including mineral oil may con- oils with a very low or undetectable content of PAH are neither
tribute to UV-induced photocarcinogenesis.22 However, these carcinogenic nor mutagenic in vivo. These findings, together
studies were of a very limited size and conducted in a mutant with the restricted penetration of mineral oils to the outermost
mouse that is highly sensitive to UV, which is considered layers of the skin, support the non-carcinogenic potential of
insufficient for a risk assessment. The relevance of these stud- food-grade mineral oils and waxes used in lip care products,
ies to human use of mineral oils is thus limited as there is no both by cutaneous and oral routes.

Table 4 Animal studies analysing carcinogenicity following topical exposure to highly refined white oils†
Species Number sex Mineral oil application Summary of results ‡ Reference

Duration Volume (lL) Dose §

(mg/kg/day)
C3H/HeJ mice¶ 50 M 39/week 25 296 There was no evidence of any related McKee et al.36
lifetime histopathological changes in any of
the organs examined
There was no evidence of tumour
formation at either the site or
application or in any of the internal
organs
C3H/HeJ mice 50 M 39/week 25 296 There were no apparent abnormalities McKee and Lewis37
lifetime in visceral organs evaluated during
autopsy and gross examination
C3H/HeJ mice 140 M 39/week 25 296 There were no apparent abnormalities Biles et al.35
lifetime in visceral organs evaluated during
autopsy and gross examination
C3H/HeJ mice 40 M 29/week 37.5 296 There were no apparent abnormalities McKee et al.38
24 months in visceral organs evaluated during
autopsy and gross examination.
C3H/HeJ mice 30 M 39/week 20 238 There were no apparent abnormalities McKee et al.39
50 M lifetime 25 296 in visceral organs evaluated during
autopsy and gross examination

†Adapted from the review of Nash et al.17 The mineral oil in these studies was described as ‘highly refined white oil’ (e.g. Mineral Oil, USP) and used at
the maximum tolerable dose. ‡A complete autopsy was performed. Tissues prepared for microscopic examination included the following: brain, heart,
lung, spleen, kidney, liver and lymph nodes (cervical and mesenteric). §The dose was calculated assuming that density of mineral oil was 0.83 g/mL and
the average weight of a mouse 0.03 kg. ¶This mouse strain is particularly sensitive to tumour formation mediated by polycyclic aromatics.
M, male; ND, not detected; min, minute; w, week.

JEADV 2019, 33 (Suppl. 7), 5–14 © 2019 European Academy of Dermatology and Venereology
12 B. Chuberre et al.

Reproductive toxicity levels would have similar consequences is unknown. Thus,

There are currently limited reproductive toxicology data in results of these studies do not allow to draw any conclusion on
experimental animals on mineral oil MOSH and MOAH frac- the relevance of such effects after chronic exposure to mineral
tions by cutaneous and oral routes. oils via the oral route.28 Of note, the lack of immune effects in
rats presenting strong inflammatory granuloma in the liver indi-
Cutaneous exposure cates that the two reactions may have independent mecha-
The developmental toxicity of a non-food grade MOAH mixture nisms.24
(a heavy paraffinic distillate aromatic extract) was tested in one In humans, a few epidemiological studies and case reports
study by skin exposure in pregnant rats (125–1000 mg/kg/day), (N = 49) have suggested an association between exposure to
and results indicated signs of maternal and reproductive toxicity, high doses of different types of mineral oils (i.e. cutting fluid,
with red vaginal discharge, haematological changes, decreased motor oil, hydraulic oil, form oil, asphalt, any mineral oils,
bodyweight gain and lower litter size with a higher number of mainly via skin and unintended inhalation by aerosols) and
non-viable offsprings.9 Foetal toxicity including a significant increased risk to develop autoimmune diseases, such as rheuma-
decrease in foetal bodyweight and incompletely ossified skull toid arthritis.9,28 Moreover, subcutaneous exposure to high
bones was also observed. There are no reports of reproductive levels of mineral oils in aesthetic surgery may be associated with
toxicity potential for food-grade mineral oils in animals by this local and systemic autoimmunity reactions.28,29 In particular,
route. autoimmune/inflammatory syndrome induced by adjuvant min-
eral oils, defined as the infiltration of oily type modelling sub-
Oral exposure stances for cosmetic purposes (illegal practice), is characterized
In two developmental toxicity studies conducted in female rats, by chronic granulomatous inflammation with increase of proin-
oral exposure to white mineral oils (mostly MOSH) at 1000– flammatory cytokines.29 The systemic manifestations of this syn-
5000 mg/kg/day before or during gestation until lactation did drome can be non-specific and specific, meeting criteria for any
not induce maternal or fetal toxicity.9 There were no other treat- autoimmune disease in 40% of patients (systemic lupus erythe-
ment-related effects. Moreover, no effects were detected with matosus, scleroderma and systemic sclerosis, rheumatoid arthri-
regard to reproductive ability in rats (100–1000 mg/kg/day). tis, dermatomyositis and overlap syndrome). Despite being
Similarly, oral administration of MOAH to gestating female rats prohibited in the world, this cosmetic practice remains common
(15–135 mg/kg/day from day 6 to day 19 post-coitum) induced in many countries thus representing a serious health problem.29
no treatment-related death in any of the dams and no clinical However, in those instances, actual levels of exposure to mineral
signs. Only a decreased terminal bodyweight was observed in oils (not all of food grade and by routes other than cutaneous)
females for the high-dose group.9 in affected individuals have not usually been measured and other
Overall, there was no reproductive toxicity by oral route in genetic and environmental factors might also be involved in
experimental animals. A moderate developmental toxicity was those autoimmune reactions (including infection, vaccination
observed only after skin application of a non-food grade mineral and tobacco smoke).28
oil containing MOAH to gestating female rats. The EFSA thus So far, there are no indications of altered immune function or
concluded that an effect on reproduction cannot be fully autoimmunity after per oral or epidermal exposure to mineral
excluded for MOAH when using non-food grade mineral oil, oils in animal models and in humans. Where presumptive asso-
i.e., non-highly refined mineral oils which are not used in cos- ciations have been found, high levels of exposure to mineral oils
metics.9 (of all types) mainly occurred via inhalation or intradermal
route.
Autoimmunity
Hydrocarbons are known to induce inflammation and act as Exposure estimation to MOSH/MOAH via lipstick
adjuvants by enhancing the immune responsiveness, so that oil- use
based materials, such as Freund’s adjuvants, can be used in vac- In 2012, the dietary exposure of the European population to
cines. In arthritis-prone rodent models, intradermal and MOSH was estimated to range from 0.03 to 0.3 mg/kg/day, that
intraperitoneal injections of high doses of certain MOSH can is a 60 kg person may consume 1.8–18 mg/day of MOSH via
induce autoimmune responses with autoantibodies and associ- food.9 Moreover, lip care products may participate in the total
ated clinical signs.9,28 Weaker effects were observed following oral exposure and accumulation of MOSH, as they were detected
short-term percutaneous exposure on abraded skin, yet there in some of these products8,11 at levels significantly higher than
was no information about a dose-response relationship. Further- the 5% recommended by Cosmetics Europe.7
more, a few short-term studies found no clear immune modula- Nevertheless, these findings have to be considered in view of
tion or no arthritic reaction in rats after oral exposure to MOSH the real oral exposure to lip care products and their non-carci-
(5–120 days),9,24 but whether long-term oral exposure to low nogenic potential. Indeed, the amount of lip balm ingested by

JEADV 2019, 33 (Suppl. 7), 5–14 © 2019 European Academy of Dermatology and Venereology
Safety of mineral oils used in cosmetics 13

consumers was estimated reaching up to 20 g per year through Delphine Blanchet (L’Oreal Research and Development) for
daily use – that is the equivalent of 4 typical lip balms.11 Based contributing to critically review the manuscript.
on MOSH contents up to 74% in lip balms, their contribution
to MOSH exposure can be estimated up to 0.7 mg/kg/day for a References
60 kg person, thus representing <10% of the ADI value of 1 JECFA. Evaluation of certain food additives - Seventy-sixth report of the
Joint FAO-WHO Expert Committee on Food Additives. WHO Technical
12 mg/kg/day set for the highly refined mineral oils used in these
Report Series 974, 2012.
products.8 As a reminder, a similar oral dose (1 mg/kg) of white 2 EFSA. Scientific opinion on the safety assessment of medium viscosity
mineral oils administered to volunteers led to undetectable white mineral oils with a kinematic viscosity between 8.5 – 11 mm²/s at
hydrocarbons (C19–C24) in blood.16 Moreover, the presence of 100 C for the proposed uses as a food additive. EFSA J 2013; 11: 3073.
3 EFSA. Scientific opinion on the re-evaluation of microcrystalline wax
MOSH and MOAH in itself is not indicative of any carcinogenic
(E905) as a food additive. EFSA J 2013; 11: 3146.
potential and highly refined mineral oils are almost devoid of 4 Grob K. Toxicological assessment of mineral hydrocarbons in foods: state
potentially harmful PAH. Thus, this moderate oral intake of of present discussions. J Agric Food Chem 2018; 66: 6968–6974.
hydrocarbons through lip care products is not expected to repre- 5 Barp L, Biedermann M, Grob K et al. Mineral oil saturated hydrocarbons
(MOSH) in female Fischer 344 rats; accumulation of wax components;
sent a significant safety risk for the consumer.8 However, con- implications for risk assessment. Sci Total Environ 2017; 583: 319–333.
sumers should be aware that not all lip care products available 6 EU No 1223/2009. Commission Regulation (EU) on cosmetic products
on the European market or websites are fully compliant with the No 1223/2009 Official Journal of the European Union (30 November
European regulations and some may contain higher levels of 2009).
7 Cosmetics Europe. Mineral hydrocarbons in cosmetic lip care products.
MOSH and MOAH than recommended. As shown in this arti- Recommendation No 14 (17-09-2018), 2018.
cle, following the European regulations6,7 on cosmetics is a safety 8 BfR. Highly refined mineral oils in cosmetics: Health risks are not to be
guarantee for consumers, who are encouraged to purchase lip expected according to current knowledge. Bundesinstitut f€ ur risikobewer-
tung, Updated BfR opinion No 008/2018 (Feb 19, 2018), 2018.
care products from reliable vendors or official websites and to
9 EFSA. Panel on contaminants in the food chain (CONTAM); scientific
control the country of origin mentioned on their labelling. opinion on mineral oil hydrocarbons in food. EFSA J 2012; 10: 2704.
10 ANSES. Proces-verbal du GT “Evaluation des substances et proc
ed
es sou-
Conclusion mis a autorisation en alimentation humaine - ESPA”– 19 janvier 2017 sur
la SAISINE No 2017-SA-0070 relative a la migration des compos
es
On the basis of the information presented in this overview,
d’huiles min
erales dans les denr
ees alimentaires a partir des emballages en
highly refined mineral oils and waxes used in cosmetic products papiers et cartons recycl
es, 2017.
can be considered as safe in human for: 11 BEUC. EU consumer organisations call for action on mineral oils in lip
• Skin application: no health risk was identified in humans balms. Bureau Europ
een des Unions de Consommateurs; BEUC-X-2017-
128, 2017.
with respect to systemic toxicity, dermatotoxicity, repro-
12 Weber S, Schmidt T, Schumacher P et al. Quantification of mineral oil
ductive toxicity and the development of autoimmune aromatic hydrocarbons (MOAH) in anhydrous cosmetics using 1H
response and disease. Moreover, there is no evidence of NMR. J Chem 2019; 2019: 10.
tumorigenicity or genotoxicity via the skin. 13 Petry T, Bury D, Fautz R et al. Review of data on the dermal penetration
of mineral oils and waxes used in cosmetic applications. Toxicol Lett 2017;
• Oral exposure: although repeated oral exposure to mineral 280: 70–78.
oils may cause bioaccumulation of MOSH in human tis- 14 Health Canada. Screening assessment petroleum sector stream approach
sues, it was considered of little concern from a toxicological petrolatum and waxes [Stream 4]. Chemical abstracts service registry
point of view. There is also no indication of carcinogenicity, numbers: 8009-03-8, 64742-61-6, 64743-01-7. Cat. No.: En14-255/2016E-
PDF, ISBN 978-0-660-05468-1, 2016.
mutagenicity, reproductive toxicity or autoimmunity with 15 Barp L, Kornauth C, Wuerger T et al. Mineral oil in human tissues, Part
mineral oils of pharmaceutical and food grades. I: concentrations and molecular mass distributions. Food Chem Toxicol
• Lip care products: although they may participate in the total 2014; 72: 312–321.
16 Boogaard PJ, Goyak KO, Biles RW et al. Comparative toxicokinetics of
oral exposure and accumulation of MOSH in humans, no
low-viscosity mineral oil in Fischer 344 rats, Sprague-Dawley rats, and
significant health risk can be expected from lip care prod- humans–implications for an Acceptable Daily Intake (ADI). Regul Toxicol
ucts that comply with the recommendation of Cosmetics Pharmacol 2012; 63: 69–77.
Europe, that is highly refined mineral oils with undetectable 17 Nash JF, Gettings SD, Diembeck W, Chudowski M, Kraus AL. A toxico-
logical review of topical exposure to white mineral oils. Food Chem Toxi-
or trace levels of PAH (<3% w/w of substances extractable
col 1996; 34: 213–225.
with the solvent DMSO), <5% of low molecular hydrocar- 18 Schnuch A, Lessmann H, Geier J, Uter W. White petrolatum (Ph. Eur.) is
bons (<C25)7 and a daily ingestion below the ADI value of virtually non-sensitizing. Analysis of IVDK data on 80 000 patients tested
12 mg/kg/day (set for highly refined mineral oils).8 between 1992 and 2004 and short discussion of identification and desig-
nation of allergens. Contact Dermatitis 2006; 54: 338–343.
19 Ulrich G, Schmutz JL, Trechot P, Commun N, Barbaud A. Sensitization
Acknowledgements to petrolatum: an unusual cause of false-positive drug patch-tests. Allergy
The authors thank C
ecile Desjobert, C
eline Zimmer and Marielle 2004; 59: 1006–1009.
Romet (Sant
e Active Edition) for medical writing assistance. We 20 Tam CC, Elston DM. Allergic contact dermatitis caused by white petrola-
tum on damaged skin. Dermatitis 2006; 17: 201–203.
also gratefully acknowledge Dagmar Bury, Maya Krasteva and

JEADV 2019, 33 (Suppl. 7), 5–14 © 2019 European Academy of Dermatology and Venereology
14 B. Chuberre et al.

21 Dooms-Goossens A, Degreef H. Contact allergy to petrolatums. (I). Sensi- 31 Stamatas GN, de Sterke J, Hauser M, von Stetten O, van der Pol A. Lipid
tizing capacity of different brands of yellow and white petrolatums. Con- uptake and skin occlusion following topical application of oils on adult
tact Dermatitis 1983; 9: 175–185. and infant skin. J Dermatol Sci 2008; 50: 135–142.
22 Rawlings AV, Lombard KJ. A review on the extensive skin benefits of 32 Patzelt A, Lademann J, Richter H et al. In vivo investigations on the pene-
mineral oil. Int J Cosmet Sci 2012; 34: 511–518. tration of various oils and their influence on the skin barrier. Skin Res
23 Trimmer GW, Freeman JJ, Priston RA, Urbanus J. Results of chronic diet- Technol 2012; 18: 364–369.
ary toxicity studies of high viscosity (P70H and P100H) white mineral oils 33 Choe C-S, Lademann J, Darvin ME. Gaussian-function-based deconvolu-
in Fischer 344 rats. Toxicol Pathol 2004; 32: 439–447. tion method to determine the penetration ability of petrolatum oil into
24 Nygaard UC, Vege A, Rognum T et al. Toxic effects of mineral oil satu- in vivo human skin using confocal Raman microscopy. Laser Phys 2014;
rated hydrocarbons (MOSH) and relation to accumulation in rat liver. 24: 105601.
Food Chem Toxicol 2019; 123: 431–442. 34 Choe C, Lademann J, Darvin M. Analysis of human and porcine skin
25 Biedermann M, Barp L, Kornauth C et al. Mineral oil in human tissues, in vivo/ex vivo for penetration of selected oils by confocal Raman micro-
part II: characterization of the accumulated hydrocarbons by comprehen- scopy. Skin Pharmacol Physiol 2015; 28: 318–330.
sive two-dimensional gas chromatography. Sci Total Environ 2015; 506– 35 Biles RW, McKee RH, Lewis SC, Scala RA, DePass LR. Dermal carcino-
507: 644–655. genic activity of petroleum-derived middle distillate fuels. Toxicology
26 Concin N, Hofstetter G, Plattner B et al. Evidence for cosmetics as a 1988; 53: 301–314.
source of mineral oil contamination in women. J Womens Health 36 McKee RH, Stubblefield WA, Lewis SC et al. Evaluation of the dermal
(Larchmt) 2011; 20: 1713–1719. carcinogenic potential of tar sands bitumen-derived liquids. Fundam Appl
27 CONCAWE. Critical review of the relationship between IP346 and dermal Toxicol 1986; 7: 228–235.
carcinogenic activity. Health and Toxicology Subgroup-Report No 6/16, 37 McKee RH, Lewis SC. Evaluation of the dermal carcinogenic potential of
2016. liquids produced from the Cold Lake heavy oil deposits of northeast
28 Kimber I, Carrillo JC. Oral exposure to mineral oils: is there an associa- Alberta. Can J Physiol Pharmacol 1987; 65: 1793–1797.
tion with immune perturbation and autoimmunity? Toxicology 2016; 38 McKee RH, Daughtrey WC, Freeman JJ et al. The dermal carcinogenic
344–346: 19–25. potential of unrefined and hydrotreated lubricating oils. J Appl Toxicol
29 Vera-Lastra O, Medina G, Cruz-Dominguez MP et al. Autoimmune/in- 1989; 9: 265–270.
flammatory syndrome induced by mineral oil: a health problem. Clin 39 McKee RH, Scala RA, Chauzy C. An evaluation of the epidermal carcino-
Rheumatol 2018; 37: 1441–1448. genic potential of cutting fluids. J Appl Toxicol 1990; 10: 251–256.
30 Strakosch EA. Studies on ointments. J Pharmacol Exp Ther 1943; 78: 65.

JEADV 2019, 33 (Suppl. 7), 5–14 © 2019 European Academy of Dermatology and Venereology