Content uploaded by Basil Athanassiadis
Author content
All content in this area was uploaded by Basil Athanassiadis on Feb 17, 2020
Content may be subject to copyright.
REVIEW
A review of the effects of formaldehyde release
from endodontic materials
B. Athanassiadis
1
, G. A. George
2
, P. V. Abbott
3
& L. J. Wash
4
1
Private Practice, Brisbane, Qld;
2
School of Chemistry, Physics and Mechanical Engineering, Queensland University of
Technology, Brisbane, Qld;
3
School of Dentistry, University of Western Australia, Perth, WA; and
4
School of Dentistry,
University of Queensland, Brisbane, Qld, Australia
Abstract
Athanassiadis B, George GA, Abbott PV, Wash LJ.
A review of the effects of formaldehyde release from
endodontic materials. International Endodontic Journal,48,
829–838, 2015.
Formaldehyde is present in most living cells and the
environment. In dentistry, patients may be exposed to
formaldehyde through the use of several endodontic
materials (e.g. AH 26) and during formocresol pulpot-
omies. This review outlines how the human body
reacts to formaldehyde exposure, how recent data has
relooked at the issue of carcinogenicity and leukaemia
associated with formaldehyde, and whether it is possi-
ble to quantify the amount of formaldehyde produced
by endodontic cements. The review analyses the way
formaldehyde is produced from epoxy resins and
addresses the question of whether the amount of form-
aldehyde from endodontic cements is large enough to
override the body’s ability to deal with its own endoge-
nous levels of formaldehyde and should the amount of
formaldehyde produced be a concern.
Keywords: AH 26, carcinogen, endogenous levels,
formaldehyde release, safety.
Received 29 May 2014; accepted 30 September 2014
Introduction
The release of formaldehyde is a widely known effect
that occurs with many materials used in dentistry.
Some typical examples are methacrylate and urethane
dimethacrylate (base plates used in orthodontics and
prosthetics), composite resin for restorations, epoxy
resins used in root canal treatment and formocresol
in pulpotomies (Santerre et al. 2001, Kopperud et al.
2011). Formaldehyde has been used in the manufac-
ture of particle board, plywood, glues and foam insu-
lation. Approximately 80% of its use is for plastic and
resin manufacture. The remaining 20% is used in
agriculture (seed treatment), reagents in laboratories
and preservatives in cosmetics (Restani & Gali 1991).
It is important to understand the difference between
formaldehyde gas and formaldehyde solution in order
to avoid confusion. Formaldehyde is a flammable gas
with a pungent, strong odour (at greater than 0.3
parts per million [ppm]). It is highly soluble in water
(up to 55%), acetone, benzene, chloroform and etha-
nol. Most formaldehyde is sold as aqueous solutions,
known as formalin, containing 30–50% formaldehyde
with methanol as a stabilizer to prevent it polymeriz-
ing into a solid form. Formaldehyde solution is a clear
colourless liquid also with a pungent and irritating
odour (Lewis & Chestner 1981, Budavari 2001,
Sweetman 2011). The chemical formula of formalde-
hyde is HCHO (or CH
2
O), and it is the simplest
aldehyde being a mono-aldehyde compared to
the di-aldehyde of glutaraldehyde. It is a gas that dis-
solves easily in water to form methylene hydrate
(HO-CH
2
-OH). Methylene hydrate molecules react with
one another, combining to form polymers with
most of the formaldehyde existing as low polymers
(n=2–8intheformulaHO-[CH
2
O]
n
-H. Higher polymers
Correspondence: Basil Athanassiadis, 268 Ipswich Rd cnr
Cornwall Street, Annerley, 4103 Brisbane, Qld, Australia
(e-mail: basildent@bigpond.com).
International Endodontic Journal,48, 829–838, 2015©2014 International Endodontic Journal. Published by John Wiley & Sons Ltd
doi:10.1111/iej.12389
829
(nup to 100), which are insoluble, are white powders,
and these are called paraformaldehyde (Kiernan
2000). Paraformaldehyde has a formaldehyde compo-
nent of 91–99%, and in liquid phase, the powder and
the water vaporize at room temperature to release
formaldehyde (Lewis & Chestner 1981). Formocresol
was introduced into dentistry as a pulpotomy agent by
Buckley in 1904 and consists of 35% tricresol, 19%
formaldehyde, 15% glycerine and water. Variations
include the use of a diluted formula (1 : 5) and of the
mixing of the formocresol into a zinc-oxide eugenol
base instead of applying the solution to the pulp on a
cotton pellet (Kahl et al. 2008, Balaprasanna 2011).
In endodontic practice, there are materials that
contain formaldehyde and/or paraformaldehyde –
such as N2 paste (Indrag-Agsa, Losone, Switzerland),
Endomethasone (Septodont, Paris, France), Riebler’s
paste (Amubarut; Wera Karl, Biesingen, Germany)
and SPAD (Trailement, Quetigny, France) –which
have paraformaldehyde levels between 4 and 8%
(Bonsor & Pearson 2013). Other materials such as
the epoxy resin cements, for example AH26 (De Trey
Dentsply, Konstanz, Germany) and AH Plus (De Trey
Dentsply), do not contain formaldehyde as an ingredi-
ent, but they release minimal levels of formaldehyde
during their setting reaction. The formaldehyde
release peaks after 2 days and then slowly decreases
for a maximum period of 2 weeks (Lewis & Chestner
1981, Koch 1999).
The American Association of Endodontists issued a
position paper on the use of formaldehyde- and para-
formaldehyde-containing materials in which they rec-
ommended that they should not be used during
endodontic treatment due to their toxicity and carcin-
ogenicity (AAE 2013).
An electronic search was conducted using PubMed
and Google Scholar search engines to identify appro-
priate articles written in the English language. The
following keywords were used: formaldehyde, carcino-
gen, toxicity, formocresol, endodontics, epoxy resins,
AH26, AHPlus. Textbooks such as Merck Index and
Martindales were also examined for relevant informa-
tion. Government agency websites in the USA, Austra-
lia and United Kingdom related to formaldehyde were
examined. The reference lists of the identified publica-
tions were also examined for additional articles.
Sources of formaldehyde
The possible routes of exposure to formaldehyde are
by ingestion, inhalation, dermal absorption and blood
exchange. Once absorbed, formaldehyde is very
quickly broken down. Almost every tissue in the body
has the ability to breakdown formaldehyde. It is usu-
ally converted to a nontoxic chemical called formate,
which is excreted in the urine. Formaldehyde can also
be converted to carbon dioxide and breathed out of
the body. Formaldehyde is irritating to tissues when it
comes into direct contact with them. The most com-
mon symptoms include irritation of the eyes, nose
and throat, along with increased tearing, which
occurs at air concentrations of 0.4–3 ppm (ATSDR
1999, Golden 2011).
Humans are exposed to formaldehyde on a daily
basis from various sources related to lifestyle and
diet. Some of these sources are foods such as shiitake
mushrooms (40–380 ppm), fresh seafoods (mackerel,
squid, scallop, octopus at 2 ppm), fruit and vegeta-
bles (3–22 ppm). Inhalation of trace amounts of
formaldehyde can easily occur from multiple sources,
such as the decomposition of plant residues, automo-
tive exhaust, cigarette smoke, outgassing of furniture
and joinery items manufactured from chipboard or
plywood, insulating materials used in construction,
workplace use of various synthetic resins and glues,
fabrics, cosmetics and hair straightening products.
Combining these various sources, the World Health
Organization (WHO) estimated daily intake of
formaldehyde for an adult is about 10.55 mg day
1
,
comprising 9.4 mg day
1
from food, 1 mg day
1
from inhalation and 0.15 mg day
1
from water
(Restani & Gali 1991, WHO 2001, Tang et al.
2009).
Formaldehyde is present in virtually all cells in the
human body as a by-product of the metabolism of ser-
ine, glycine, methionine and various other amino
acids. Endogenous levels of metabolically produced
formaldehyde range from 3 to 12 ng g
1
of tissue
(WHO 2001, Kahl et al. 2008). Formaldehyde can be
readily detected in human plasma with typical con-
centrations of 2.5 ppm (Restani & Gali 1991, Leh-
man-McKeeman 2010, Golden 2011, Checkoway
et al. 2012). No increase in formaldehyde concentra-
tion was seen in the blood of humans, rats and mon-
keys following exposure to concentrations of 1.9 ppm
(2.3 mg m
3
), 6 ppm (7.2 mg m
3
) and 14.4 ppm
(17.3 mg m
3
) gaseous formaldehyde, respectively
(IPCS 2002). This has been attributed to the deposi-
tion of formaldehyde principally in the respiratory
tract and its rapid metabolism (Heck et al. 1985,
Casanova et al. 1988). Exogenous formaldehyde does
not accumulate in the body as it has a biological
Formaldehyde, controversy, risks, quantification Athanassiadis et al.
©2014 International Endodontic Journal. Published by John Wiley & Sons LtdInternational Endodontic Journal,48, 829–838, 2015
830
half-life of only 1–1.5 min, and it is quickly cleared
from human plasma. Such rapid metabolism would
inhibit the systemic distribution of formaldehyde
(Restani & Gali 1991, NICNAS 2006).
Formaldehyde is present in low concentrations
(<0.2%) in a wide variety of consumer products. It is
used as a preservative for the control of bacteria and
fungae in water-based solutions in both industrial
and consumer products including dishwashing liq-
uids, disinfectants, fabric conditioners, shampoos,
conditioners and shower gels. Many of these products
are released directly into wastewater streams during
their use and hence, they are a source of formalde-
hyde levels in water (NICNAS 2006). Some common
formaldehyde-releasing preservatives include DMDM
Hydantoin (1,3-dihydroxymethyl-5, 5-dimethyl hyd-
antoin) and imidazolidinyl urea. The free formalde-
hyde content in DMDM Hydantoin is usually up to
2%. Any bacterial activity consumes the free formal-
dehyde which is then replenished from the parent
compound. Over a period of time, all formaldehyde
from the donor molecule is used up in preserving the
product against microbes (NICNAS 2006).
Antimicrobial actions
Formaldehyde solution is bactericidal, sporicidal and
virucidal, but it works more slowly than glutaralde-
hyde. Formaldehyde is an extremely reactive chemical
that interacts with protein, DNA and RNA. When
applied to unbroken skin, formaldehyde solution hard-
ens the epidermis, renders it tough and white, and
produces a local anaesthetic effect. Formaldehyde
prevents tissue autolysis as it binds to protein and
prevents enzymatic degradation of proteins (Kurji
2009). Diluted formaldehyde solution containing
0.75% formaldehyde w/w has been used to treat
warts on hands and feet (McDonnell & Russell 1999,
Kiernan 2000, Sweetman 2011).
Root canal cements that produce formaldehyde on
setting may allow the material to exert some anti-
microbial action to counter the effects of any resid-
ual bacteria left in the root canal system at the
time of root filling. Formaldehyde is a relatively
nonspecific bactericidal agent, affecting the growth
and viability of most gram-positive and gram-nega-
tive bacteria as well as fungi (Sweetman 2011).
Formaldehyde gas has very little penetrative power
as it readily condenses on surfaces and polymerizes.
Its effectiveness as an antimicrobial agent requires a
high relative humidity of 80–90%. Formaldehyde
gas is used for disinfection of rooms and cabinets,
and it can be used with low temperature steam for
sterilization of heat sensitive items (Sweetman
2011).
Some epoxy resin endodontic cements (e.g. AH 26)
contain hexamethylenetetramine (HMT- also known
as hexamine, methenamine or urotropine), which
itself exerts antibacterial actions, and was first used
as an antiseptic agent (urotropine) in 1894 (Grayson
2010). HMT is a white powder which is freely solu-
ble in water and soluble in alcohol with an alkaline
pH, but the hippurate and mandelate salts of HMT
have a pH of 4 when in solution. HMT owes its anti-
bacterial action to the release of formaldehyde which
is slowly liberated by hydrolysis at acidic pH (<5.5).
The hippurate or mandelate salts are also used for
long-term suppression of chronic or recurrent lower
urinary tract infections. It is only active in acidic
urine when formaldehyde is released (Sweetman
2011). Almost no hydrolysis of HMT occurs at physi-
ological Ph, and it is therefore virtually inactive in
the body at neutral pH (Scott & Wolf 1962, Sweet-
man 2011). The half-life is approximately 4 h, and it
is rapidly and almost completely eliminated in the
urine (Sweetman 2011). In humans, no harmful
reactions or complications have been observed in
patients receiving HMT as an antiseptic at dose levels
of 4–6 g day
1
for weeks (Restani & Gali 1991).
HMT continues to be used as an antibacterial agent,
most commonly as a food preservative because of its
antimicrobial activity and its lack of taste and odour
(Restani & Gali 1991). HMT is also used widely in
the manufacturing of particle board, plywood and
foam insulation.
Hexamine, N
4
(CH
2
)
6
, liberates formaldehyde under
acidic hydrolysis. This hydrolysis is accelerated by
heating and decreasing the pH (Dreyfors et al. 1989,
Grayson 2010). Such conditions would be uncommon
following the use of root canal medicaments, but it
may occur if instrumentation and root canal filling
were done in one appointment and periapical inflam-
mation was present. Therefore, the amount of formal-
dehyde produced when hexamine is used in a root
canal filling and following a period of medicament
use would most likely be inconsequential.
The chemical formula for the breakdown of HMT is
as follows:
N
4
(CH
2
)
6
(HMT) +6 H2O ?4NH
3
(ammo-
nia) +6CH
2
O (formaldehyde)
The key ingredient is the requirement of an acidic
pH (<pH 5.5) (Scott & Wolf 1962, Sweetman 2011).
Athanassiadis et al. Formaldehyde, controversy, risks, quantification
International Endodontic Journal,48, 829–838, 2015©2014 International Endodontic Journal. Published by John Wiley & Sons Ltd 831
The controversy of formaldehyde being a
carcinogen
Concerns regarding formaldehyde release are based
on its known properties as an irritant as well as con-
cerns that it may be a carcinogen. There is contro-
versy as to the risk that formaldehyde presents as a
carcinogen, and the possibility that it is a human car-
cinogen is impossible to exclude formally (Sweetman
2011) even though formaldehyde is not a direct
genotoxic agent at sites distant to the portal of entry
(nose, oral cavity) (Checkoway et al. 2012, Gentry
et al. 2013). A substantial body of new evidence has
appeared in the literature between 2010 and 2012,
which shows no increased incidence of nasopharyn-
geal cancer in humans who have a mean formalde-
hyde exposure level of <1 ppm, up to peak levels of
4 ppm (Bolt & Morfeld 2013).
A major issue with assessing the possible carcinoge-
nicity of formaldehyde is that it is present at relatively
constant levels in the blood of humans. These back-
ground levels of formaldehyde create an analytical
problem in differentiating altered or damaged DNA
(adducts) that result from endogenously generated
formaldehyde from those that are directly related to
any exogenous chemical exposure. Various regulatory
bodies have set exposure limits to formaldehyde in air
(Table 1). The International Agency for Research in
Cancer (IARC) has classified formaldehyde as ‘carcino-
genic to humans’ (Cogliano et al. 2005), although
Marsh et al. (2010) later showed that some of the
studies on which this IARC classification was based
had incomplete data sets and striking discrepancies
and presented misleading evidence. The US Occupa-
tional Safety and Health Administration (OSHA) and
the Australian National Industrial Chemicals Notifica-
tion and Assessment Scheme (NICNAS) both regard
formaldehyde as a possible human carcinogen and
consider that it can cause cancer in animals at high
levels that are not found in the majority of workplaces.
Duhayon et al. (2008) stated that the most recent
epidemiological studies indicate that the statement
‘formaldehyde is carcinogenic to humans’ is probably
too strongly worded. In a review of all cohort studies
published to February 2007 (Bosetti et al. 2008)
noted that industry workers and professionals exposed
to formaldehyde showed no appreciable excess risk of
cancers of the oral cavity, pharynx, sinus, nasal cav-
ity and lungs.
Logically, one would expect a site-exposure rela-
tionship with malignancies related to the tissues most
exposed to formaldehyde from inhalation or ingestion.
In cases of high exogenous exposure to formaldehyde,
DNA effects are limited to the respiratory tract, and
lesions have not been observed beyond the point of
contact from inhalation exposure to formaldehyde
(Lehman-McKeeman 2010) nor are levels of adducts
different in remote sites. In other words, exogenous
formaldehyde can cause DNA adducts in nasal epithe-
lial DNA from direct inhalation exposure but not in
bone marrow and other distant sites. The concept
that inhaled formaldehyde could cause leukaemia or
influence myeloid progenitor cells or other bone
marrow cells have been formally excluded (Bolt &
Morfeld 2013). A recent review (Gentry et al. 2013)
concluded that there is no association between form-
aldehyde exposure and myeloid or lymphoid malig-
nancies. Likewise, there is no consistent evidence of
genotoxicity in the bone marrow following exogenous
formaldehyde exposure (Checkoway et al. 2012). The
Table 1 Formaldehyde gas exposure limits from different regulatory bodies
Legal exposure limits
US-OHSA,
NIOSH, HESIS
EU-SCOEL
(2008)
UK-HSE
(2011)
AUS-NICNAS
(2006)
Threshold limit value (TLV) 0.3 ppm 0.2 ppm N/A N/A
Permissible exposure level (PEL)
Limit (PEL) over an 8-h workshift
0.75–1.00 ppm 0.2 ppm 2.0 ppm 0.3 ppm
Short-term exposure limit (STEL)-over 15-min period 2.0 ppm 0.4 ppm 2.0 ppm 0.6 ppm
Immediately dangerous to life and health (IDLH) >20 ppm N/A N/A N/A
Data derived from Bolt & Morfeld 2013, Duhayon et al. 2008, and regulatory body websites. OHSA, US Occupational Safety and
Health Administration; NIOSH, US National Institute of Occupational Safety and Health; NICNAS, Australian National Industrial
Chemicals Notification and Assessment Scheme; HESIS, Californian Hazard Evaluation System and Information Service;
EU-SCOEL, European Union-Scientific Committee on Occupational Exposure Limits for formaldehyde; UK-HSE, United Kingdom-
Health and Safety Executive; TLV, FA concentration should not exceed this value at any time; N/A, not available. 1 ppm =1 part
of formaldehyde gas per million parts of air. Conversion factors (in air): 1 ppm =1.25 mg m
3
, 1 mg m
3
=0.8 ppm (at 20 °C and
1013 hPa) (WHO 2001, Arts et al. 2006).
Formaldehyde, controversy, risks, quantification Athanassiadis et al.
©2014 International Endodontic Journal. Published by John Wiley & Sons LtdInternational Endodontic Journal,48, 829–838, 2015
832
extensive metabolic capability of humans and the
findings that no inhaled formaldehyde gets past nasal
epithelium into the systemic circulation strongly sug-
gest that formaldehyde should be more appropriately
characterized as a chemical with adverse effects
rather than as a carcinogen (Golden 2011).
Exposure limits
Formaldehyde affects humans when breathing its
vapours or touching the liquid. It reacts quickly with
body tissues and affects the site of direct contact (e.g.
eyes, nose, throat and skin). Formaldehyde can
destroy the skin’s protective oils causing dryness,
cracking and dermatitis. High levels (5–30 ppm) can
severely irritate the lungs causing chest pain and
shortness of breath (HESIS 2011). In humans, odour
perception of formaldehyde (0.5–1.0 ppm) precedes
sensory irritation (>2.0 ppm) of the nose, throat and
eyes, with eye irritation accepted as the most sensitive
end-point. Short-term exposure (<1 h) to formalde-
hyde below 2 ppm produces no toxicological effects
on the eyes or upper respiratory tract. Moderate eye,
nose and throat irritation occurs at 2–3 ppm. A form-
aldehyde concentration of 0.1 ppm is unlikely to
result in any irritant effects for individuals, including
children, asthmatics and the elderly. Adverse effects
occur only at the point of contact after the concentra-
tion achieved is in excess of endogenous levels, and it
exceeds the body’s ability to maintain homoeostasis
(Golden 2011). Formaldehyde vapour is irritant to
the nose, eyes and upper respiratory tract and may
cause coughing, spasm and oedema of the larynx,
bronchitis and pneumonia with asthma-like symp-
toms with repeated exposure (Sweetman 2011).
The minimum risk level (MRL) is an estimate of the
daily human exposure to a hazardous substance that
is likely to be without appreciable risk of adverse, non-
cancer health effects over a specified duration of expo-
sure. For oral exposure to formaldehyde, an MRL of
0.3 mg kg
1
day
1
has been derived for intermediate-
duration exposure, and an MRL of 0.2 mg kg
1
day
1
has been derived for chronic-duration exposure
(ATSDR 1999).
In the blood, the mean formaldehyde concentra-
tion is reported as 2.24 mg 0.07 mg kg
1
(Restani & Gali 1991). WHO estimates that an adult
is exposed to 10.55 mg formaldehyde day
1
from
food, air and water. There are fairly constant endog-
enous levels of 2.5 ppm formaldehyde in blood
(Golden 2011). Szarvas et al. (1986) determined the
endogenous level of formaldehyde in blood at an
average of 0.5 lgmL
1
of blood. With approxi-
mately 5 L of blood volume in humans, this equates
to 2.5 mg of endogenous formaldehyde circulating at
any one time. Endogenous turnover of formaldehyde
was estimated to be approximately 878–
1310 mg kg
1
body weight per day, assuming a
half-life of 1–1.5 min (EFSA 2014).
Quantification of formaldehyde
There are several methods used to assess formalde-
hyde release from endodontic materials, and each has
its advantages and disadvantages (Table 2). The
direct instrumental methods of GC-MS (Sp
angberg
et al. 1993) and direct spectrophotometry in the
infrared or ultraviolet ranges (Leonardo et al.1999)
can detect formaldehyde but cannot give absolute
concentrations. The remaining methods involve an
initial reaction with colorimetric reagents to form
formaldehyde adducts which are then analyzed by
their ultraviolet absorption spectra, generally after
removing other components by HPLC (Cohen et al.
1998, Koch 1999, Koch et al. 2001). Current meth-
ods include direct instrumental analysis (e.g. gas
phase spectroscopy using a tunable diode laser) and
derivatization methods involving subsequent chroma-
tography and ultraviolet detection, as well as more
sensitive methods using fluorescence (Li et al. 2005).
The biggest challenge with fluorescence assay meth-
ods is to avoid false positives due to the presence of
contaminants or nonspecific reactions (Compton &
Purdy 1980). Nevertheless, the fluorescence method
when compared with the derivatization/HPLC method
has been found to give statistically comparable results
as well as having high sensitivity (Pinheiro et al.
2004).
In order to determine the amount of formaldehyde
produced in a typical root canal filling, the mass of
resin deposited has to be calculated. Results from
three studies were used to determine the average sur-
face area found in a prepared molar root canal (Peters
et al. 2001, 2003, H€
ubscher et al. 2003). The aver-
age from the three studies for the total surface area of
a prepared maxillary molar was 74.68 mm
2
. Average
film thickness determined from two studies (Weis
et al. 2004, Jung et al. 2005), was used to determine
the volume of resin and found the average film
thickness was 18.0 lm. This was used in the calcula-
tion below. AH 26 was used as an example in the
calculations below as this material produces the high-
Athanassiadis et al. Formaldehyde, controversy, risks, quantification
International Endodontic Journal,48, 829–838, 2015©2014 International Endodontic Journal. Published by John Wiley & Sons Ltd 833
est formaldehyde release from the epoxy resin
cements. Epoxy resins such as AH 26 and AH Plus
release their maximum amount of formaldehyde after
2 days. There is then a gradual decline to zero with
no further release after 2 weeks.
The density of AH 26 powder, based on the above
composition is 7.06 g cm
3
(taking the density of
bismuth trioxide as 8.5 g cm
3
and that of hexa-
methylenetetramine as 1.33 g cm
3
). This powder
is then mixed with AH26 epoxy resin (density
1.16 g cm
3
) at a recommended ratio of 2 : 1. The
manufacturer’s instructions imply measurement by
volume as this is the usual mode of measurement in
the dental surgery setting. Mixing to this specifica-
tion gives a composition of 75% powder and 25%
resin by weight and thus an average density of
5.59 g cm
3
.
The volume of cement will be equal to the surface
area of the root canal multiplied by the thickness of the
cement. Therefore, based on the above findings, the
volume of cement used in a typical molar root canal
filling is 74.68 mm
2
90.018 mm =1.34 mm
3
.
The density is equal to the mass divided by the vol-
ume. Hence, the mass of sealer used is 7.5 milligrams
(mg) in a typical root canal filling.
Cohen et al. (1998) used 5 g of AH 26, previously
cured for 1 h and mixed it with 100 mL distilled
water. This was adjusted to a pH of 5.0 to maximize
the amount of formaldehyde released from the sealer.
They reported that 1347 ppm was released. Unfortu-
nately, this result cannot be converted to an amount
that could be used to work out the formaldehyde
levels per mg of material.
Koch et al. (2001) reported that the mean formal-
dehyde release from AH 26 was 6.6 lgmg
1
. This
material was stored dry for 6 months before ground
samples were taken and analyzed. Therefore, for a
typical root canal filling, the amount of formaldehyde
released would be 7.5 mg 96.6 lg formalde-
hyde mg
1
cement =49.5 lg formaldehyde. Unfortu-
nately, the length of time and the storage conditions
mean the result is of limited value.
In another study, Koch (1999) determined that
8 mg of formaldehyde was produced per gram of
Table 2 Laboratory studies of formaldehyde release from endodontic materials
Reference Method and materials Results Comment
Sp
angberg
et al. (1993)
GC-MS after mixing and after 100%
RH exposure of AH 26 and N2
Relative FA levels from MS ion
intensities. No [FA] possible
but N2 level was ~1000 times that
of AH 26
Method has limited quantification
and easily saturated at high
[FA]. Cannot be converted
to ppm
Cohen et al.
(1998)
DNPH reaction with FA then HPLC
following method EPA8315. AH
26, AH Plus and EZ-Fill sonicated
in pH 5 buffer at 40 °C for 60 min
with DNPH
AH 26: 1347 ppm
AH Plus: 3.9 ppm
EZ-Fill: 540 ppm
(based on mass of resin)
No standard deviation given
No indication of cure time before
buffer immersion. Extensive
workup required before HPLC
separation and detection.
Sensitivity 0.25 ppm
Leonardo
et al. (1999)
Direct UV–vis and IR
spectrophotometry of AH 26,
AH Plus, Top Seal and
Endomethasone. Cured 72 h
FA detected in AH 26 and
Endomethasone but not
quantified. No FA detected
from AH Plus and Top Seal
Simple UV spectrophotometry
without formaldehyde
derivatization is subject to
interferences. IR
spectrophotometry was not
quantitative
Koch (1999) Colorimetric analysis of FA with
acetylacetone and ammonia using
Visible spectra at 412 nm, for
AH 26, Amubarut and N2.
Specificity confirmed with HPLC.
Effect of mix ratio, storage time and
surface to volume ratio on
[FA] reported
Highest [FA] initially after mixing:
AH 26: 8000 1800 ppm
Amubarut: 70000 5000 ppm
N2: 17 000 2700 ppm
After 48 h reduced by 94%
(AH 26); 61% (Amubarut) and
74% (N2)
Accepted analytical method
(Hantzsch reaction).
Sensitivity ~20 ppm
Koch et al. (2001) FA reaction with dimedone
followed by HPLC of adduct
detected at 260 nm. AH 26,
Amubarut and N2 analysed after
6 months storage, with grinding
before analysis
AH 26: 6600 2600 ppm
Amubarut: 8300 1000 ppm
N2: 300 100 ppm
Method previously used by
Ruyter (1980). Variability in
results may reflect sample more
than method
Formaldehyde, controversy, risks, quantification Athanassiadis et al.
©2014 International Endodontic Journal. Published by John Wiley & Sons LtdInternational Endodontic Journal,48, 829–838, 2015
834
AH 26 immediately after mixing and this then
reduced to <1mgg
1
after 48 h. Utilizing the high-
est formaldehyde release (8 mg formaldehyde g
1
),
the amount of formaldehyde released in a typical root
canal filling with 7.5 mg sealer would be 0.06 mg of
formaldehyde, which is 1/40 of the normal endoge-
nous levels in all humans and 1/175 of the WHO
(2001) daily intake value. The formaldehyde amount
falls to zero after 2 weeks as discussed earlier. The
amount of 0.06 mg of formaldehyde is comparable to
the daily formaldehyde intake for individuals in
homes with people who smoke, which has been esti-
mated to be 0.03–0.067 mg (Nazaroff & Singer
2004).
When performing a pulptomy, the mean dose of
formocresol has been determined to be 0.013 mg per
pellet, but the actual dose that interacts with the pulp
is probably smaller than this (Kahl et al. 2008, Milnes
2008). This amount is 1/810 of the 10.55 mg day
1
of formaldehyde that occurs in our daily intake from
food, water and air (WHO 2001).
How formaldehyde is generated in the
curing of epoxy resins
Epoxy resins are thermosetting polymers formed by a
step polymerization with a suitable cross-linking agent
such as a diamine. Because of their industrial impor-
tance, the mechanism of curing for epoxy resins has
been studied in great detail to achieve optimum
mechanical properties and environmental durability
(Halley & George 2009). The properties achieved
depend on the chemical structure of the original resin
and curing agent, and the number of crosslinks per
unit volume (the crosslink density) achieved in the
curing process.
The curing process is a chemical reaction in which
the terminal epoxide groups in epoxy resin react with
a curing agent to form a cross-linked three-dimen-
sional network. The epoxy resin cement AH26 was
introduced to dentistry for root canal fillings over
50 years ago, and an unusual feature of this resin was
the choice of HMT as the curing agent. HMT is a ter-
tiary amine that, to the authors’ knowledge, had not
previously been in use for epoxy resin curing, but it
was a key ingredient in the curing of phenolic resin
precursors and in rubber vulcanization (Dreyfors et al.
1989). In phenolic resins, the curing reaction relies on
the liberation of formaldehyde in the presence of acid.
In contrast to its use in phenolic resins, the curing
reaction of hexamine with epoxy resins does not
require the liberation of formaldehyde but rather the
attribute of the HMT molecule as a tertiary amine.
There is no other industrial application of epoxy res-
ins that uses HMT as a curing agent (Dreyfors
et al.1989), possibly because of the slow curing reac-
tion and the likely liberation of formaldehyde if expo-
sure to acid subsequently occurs.
It has been reported (Oliver & Abbott 1991) that
moisture is required for the polymerization of AH26
(i.e. DGEBA epoxy resin cured with HMT). This is an
interesting result as epoxy resin cure does not usually
require the presence of moisture (Halley & George
2009) although trace amounts could potentially cata-
lyze the reaction through a process of hydrogen bond-
ing. The liberation of formaldehyde and ammonia by
the hydrolysis of HMT is not required for it to function
as a tertiary amine for catalysis of the polymerization
of DGEBA epoxy resin in AH26. It is possible that an
intermediate in the hydrolysis of hexamine is a pri-
mary amine, and this may accelerate the reaction so
that the effect is a secondary one and not a require-
ment for the reaction to occur. What is important is
that the HMT is a catalyst, so it is not consumed in
the anhydrous curing reaction. If the cure is truly cat-
alytic, then all HMT added to the resin is expected to
be present at the end of the curing reaction.
In summary, there should be no production of
formaldehyde from epoxy resins when cured with
agents other than HMT. The production of formalde-
hyde from HMT is a consequence of subsequent
hydrolysis and not the curing process of the epoxy
resin in AH26.
Evaluation of the risks posed by
formaldehyde release from root canal
cements
Block et al. (1980) filled dogs teeth with N2 paste
containing 6.5% paraformaldehyde and measured the
amount of
14
C-labelled paraformaldehyde released
into the circulation after 28 days. They showed that
the maximum level was reached after 1 day and it
was spread systemically, but there was no quantifica-
tion of the amounts. In another study (
‘
s-Gravenmade
et al. 1981), using 15% w/v formaldehyde placed in
extracted human teeth, the amount of formaldehyde
released through dentine and cementum into distilled
water was determined. Formaldehyde penetrated read-
ily through the apical third of roots within 60 s, par-
ticularly in roots derived from young patients. Values
of 80–120 lg per 3 h were obtained after insertion of
Athanassiadis et al. Formaldehyde, controversy, risks, quantification
International Endodontic Journal,48, 829–838, 2015©2014 International Endodontic Journal. Published by John Wiley & Sons Ltd 835
between 5 and 10 lL into the root canal (
‘
s-Graven-
made et al. 1981). Taken together, these studies
showed that formaldehyde permeates through dentine
and cementum quickly, and formaldehyde could enter
adjacent tissues and the systemic circulation.
However, these materials are no longer used in con-
temporary endodontic practice, and their high con-
centrations of formaldehyde do not reflect the amount
of formaldehyde that would be present in the systemic
circulation when formaldehyde is released from epoxy
resins (e.g. AH26, AH Plus). It is known that formal-
dehyde is rapidly metabolized in the human body,
and this was highlighted in a study by (Kahl et al.
2008) involving 30 children aged 2–6 years age who
underwent 85 pulpotomies (1–5 per child). A cotton
pellet with 0.013 mg of formocresol was placed in the
pulp chamber for 5 min for each pulpotomy per-
formed. Blood samples were taken before and after
the pulpotomies, yielding a total of 312 blood sam-
ples. There was no detectable formaldehyde above
normal baseline physiologic concentrations in any of
the blood samples.
In pulpotomies, formocresol is positioned in the
base of the pulp chamber in a fixed area which is in
direct contact with pulp tissue (for a period of 5 min).
In a root canal filling, the path for formaldehyde
release is more complex as the formaldehyde is pro-
duced from the epoxy resin cement which is spread
throughout the entire root canal (coronal to apical).
The formaldehyde has to be released and it then has
to diffuse through the dentine and cementum or via
the apical foramen and into the blood supply.
Systemic absorption is of no importance since after
formocresol is placed and formaldehyde is produced, it
vaporizes quickly and is rapidly eliminated in the
urine. Vaporization and dissolution of formaldehyde
in exudate minimizes the potential toxicity of root
canal medication (Gutierrez et al.1991).
Appraisals of risk from formaldehyde
The authors of two older articles (Lewis & Chestner
1981, Lewis 1998) have presented a view on formal-
dehyde which is not consistent with the more recent
reports of measured formaldehyde release from dental
materials used in vitro. These earlier articles discussed
chronic exposure of high levels of formaldehyde
rather than the short-term exposures used for formoc-
resol pulpotomies (i.e. minutes), and the doses from
epoxy resins in root canal fillings that are at least
40-fold lower than those normally ingested or present
in the circulation. There is no current evidence of
harm in humans from the latter (Sue Seale 2010).
This same confusion is evident in a paper from the
same author (Lewis 2010) which claims that
‘recently formaldehyde was strongly associated with
leukaemia whilst generally accepted as a direct cause
of nasopharyngeal cancer’. The research on formalde-
hyde referred to in that paper has for the most part
been discredited or updated, primarily because newer
and more rigorous methodologies have been used to
investigate formaldehyde (Milnes 2008).
In conclusion, it appears that the amount of form-
aldehyde released during pulpotomies with formocre-
sol and from resin-based root canal fillings are at
least 1/40 less than the normal endogenous levels in
humans, and they do not pose any health risks.
References
American Association of Endodontists (AAE) Revised April
(2013) Position Statement -Paraformaldehyde-Containing
Endodontic Filling and Sealing Materials. URL http://www.
aae.org/uploadedfiles/publications_and_research/guidelines_
and position statements/paraformaldehydefillingmaterials.
pdf (accessed August 2014).
Arts JHE, Rennen MAJ, De Heer C (2006) Inhaled formalde-
hyde: evaluation of sensory irritation in relation to carcin-
ogenicity. Regulatory Toxicology and Pharmacology 44,
144–60.
Agency for Toxic Substances and Disease Registry (ATSDR)
(1999) Toxicological Profile for Formaldehyde.pdf; pp4.
URL http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=220
&tid=39 (accessed August 2014).
Balaprasanna KC (2011) Pulpotomy in primary teeth –a
review. Journal of Indian Academy of Dental Specialists 2,
29–31.
Block RM, Lewis RD, Hirsch J, Coffrey J, Langeland K (1980)
Systemic distribution of N2 paste containing
14
C parafor-
maldehyde following root canal therapy in dogs. Oral Sur-
gery, Oral Medicine and Oral Pathology 50, 350–60.
Bolt HM, Morfeld P (2013) New results on formaldehyde:
the 2nd International Formaldehyde Science Conference
(Madrid, 19–20 April 2012). Archives of Toxicology 87,
217–22.
Bonsor SJ, Pearson G (2013) A Clinical Guide to Applied
Dental Materials. London: Churchill Livingstone Elsevier,
pp. 207.
Bosetti C, McLaughlin JK, Tarone RE, Pira E, La Vecchia C
(2008) Formaldehyde and cancer risk: a quantitative
review of cohort studies through. Annals of Oncology 19,
29–43.
Budavari S (2001) An Encyclopedia of Chemicals, Drugs and
Biologicals, The Merck Index, 13th edn. Rahway, NJ: Merck
Publishing Group, pp. 751.
Formaldehyde, controversy, risks, quantification Athanassiadis et al.
©2014 International Endodontic Journal. Published by John Wiley & Sons LtdInternational Endodontic Journal,48, 829–838, 2015
836
Casanova M, Heck HD, Everitt JI, Harrington WW Jr, Popp
JA (1988) Formaldehyde concentrations in the blood of
rhesus monkeys after inhalation exposure. Food and Chemi-
cal Toxicology 26, 715–6.
Checkoway H, Boffetta P, Mundt DJ, Mund KA (2012) Criti-
cal review and synthesis of the epidemiologic evidence on
formaldehyde exposure and risk of leukemia and other
lymphohematopoietic malignancies. Cancer Causes and Con-
trol 23, 1747–66.
Cogliano VJ, Grosse Y, Baan RA, Straif K, Secretan MB,
Ghissassi FE (2005) Meeting report: summary of IARC
monographs on formaldehyde, 2-butoxyethanol, and
1- tert-butoxy-2-propanol. Environmental Health Perspec-
tives 113, 1205–8.
Cohen BI, Pagnillo MK, Musikant BL, Deutsch AS (1998)
Formaldehyde evaluation from endodontic materials. Oral
Health 88,37–9.
Compton BJ, Purdy WC (1980) Fluoral-P, a member of a
selective family of reagents for aldehydes. Analytica Chimi-
ca Acta 119, 349–57.
Dreyfors JM, Jones SB, Sayed Y (1989) Hexamethylenetetra-
mine: a review. American Industrial Hygiene Association 50,
579–85.
Duhayon S, Hoet P, Van Maele-Fabry G, Lison D (2008)
Carcinogenic potential of formaldehyde in occupational
settings: a critical assessment and possible impact on occu-
pational exposure levels. International Archives of Occupa-
tional and Environmental Health 81, 695–710.
European Food Safety Authority (EFSA) (2014) Endogenous
formaldehyde turnover in humans compared with exoge-
nous contribution from food sources- scientific report.
EFSA Journal 12, 3550.
European Union –Scientific Committee on Occupational
Exposure Limits for formaldehyde EU-SCOEL (2008) Rec-
ommendation from the Scientific Committee on Occupa-
tional Exposure Limits for Formaldehyde. URL http://
www.google.com.au/url?sa=t&rct=j&q=&esrc=s&source=
web&cd=1&cad=rja&uact=8&ved=0CCoQFjAA&url=http%
3A%2F%2Fec.europa.eu%2Fsocial%2FBlobServlet%3FdocId
%3D3863%26langId%3Den&ei=k_t5U-6EEs3LlAXsj4CgBw
&usg=AFQjCNFvHupN_iufs7MkoMZiaEZK6fjnRg&sig2=
HwV9IkHtAASTbGc38j6Q3g&bvm=bv.66917471,d.dGI
(accessed March 2014).
Gentry PR, Rodricks JV, Turnbull D et al. (2013) Formalde-
hyde exposure and leukemia: critical review and reevalua-
tion of the results from a study that is the focus for
evidence of biological plausibility. Critical Reviews in Toxi-
cology 43, 661–70.
Golden R (2011) Identifying an indoor air exposure limit for
formaldehyde considering both irritation and cancer haz-
ards. Critical Reviews in Toxicology 41, 672–721.
s-Gravenmade EJ, Wemes JC, Dankert J (1981) Quantitative
measurements of the diffusion in vitro of some aldehydes
in root canals of human teeth. Oral Surgery, Oral Medicine
and Oral Pathology 52,97–100.
Grayson MKL (2010) Kucers’ The Use of Antibiotics, 6th edn,
Vol. 1. Florida: CRC Press, pp. 1205.
Gutierrez JH, Donoso E, Villena F, Jofre A (1991) Diffusion
of medicaments within root canal dentin - a scanning
electron microscopic study. Oral Surgery, Oral Medicine,
Oral Pathology, Oral Radiology, and Endodontics 72,
351–8.
Halley PJ, George GA (2009) Chemorheology of Polymers.
Cambridge: Cambridge University Press, pp. 52–6.
Health and Safety Executive (HSE) (2011) United Kingdom
EH40/2005 Workplace exposure limits. URL http://www.
bristol.ac.uk/nsqi-centre/media/Policies/WELs%202011.pdf
(accessed March 2014).
Heck Hd’A, Casanova-Schmitz M, Dodd PB, Schachter EN,
Witek TJ, Tosun T (1985) Formaldehyde (CH2O)
concentrations in the blood of humans and Fischer-
344 rats exposed to CH2O under controlled condi-
tions. American Industrial Hygiene Association Journal 46,
1–3.
Hazard Evaluation System & Information Service (HESIS)
(2011) California Department of Public Health Occupa-
tional Health Branch Formaldehyde fact sheet. URL http://
www.cdph.ca.gov/programs/hesis/Documents/formaldehyde.
pdf (accessed March 2014).
H€
ubscher W, Barbakow F, Peters OA (2003) Root-canal
preparation with Flex Master: canal shapes analysed by
micro-computed tomography. International Endodontic Jour-
nal 36, 740–7.
IPCS (International Programme on Chemical Safety) (2002)
Concise International Chemical Assessment Document 40:
Formaldehyde. Geneva: World Health Organization.
Jung M, Lommel D, Klimek J (2005) The imaging of root
canal obturation using micro-CT. International Endodontic
Journal 38, 617–26.
Kahl J, Easton J, Johnson G, Zuk J, Wilson S, Galinkin J
(2008) Formocresol blood levels in children receiving den-
tal treatment under general anesthesia. Pediatric Dentistry
30, 393–9.
Kiernan JA (2000) Formaldehyde, formalin, paraformalde-
hyde and glutaraldehyde: what they are and what they
do. Microscopy Today 8,8–12.
Koch MJ (1999) Formaldehyde release from root-canal seal-
ers: influence of method. International Endodontic Journal
32,10–6.
Koch MJ, Wunstel E, Stein G (2001) Formaldehyde release
from ground root canal sealer in vitro. Journal of Endodon-
tics 27, 396–7.
Kopperud HM, Kleven IS, Wellendorf H (2011) Identification
and quantification of leachable substances from polymer-
based orthodontic base-plate materials. European Journal of
Orthodontics 33,26–31.
Kurji ZA (2009) Outcomes of a Modified Pulpotomy Tech-
nique. Thesis submitted in for the degree of Master of Sci-
ence in Pediatric Dentistry, Graduate Department of
Dentistry, University of Toronto, Canada.
Athanassiadis et al. Formaldehyde, controversy, risks, quantification
International Endodontic Journal,48, 829–838, 2015©2014 International Endodontic Journal. Published by John Wiley & Sons Ltd 837
Lehman-McKeeman L (2010) Paracelsus and formaldehyde
2010. The dose to the target organ makes the poison. Tox-
icological Sciences 116, 361–3.
Leonardo MR, Bezerra da Silva LA, Filho MT, Santana da
Silva R (1999) Release of formaldehyde by 4 endodontic
sealers. Oral Surgery, Oral Medicine, Oral Pathology, Oral
Radiology, and Endodontics 88, 221–5.
Lewis B (1998) Formaldehyde in dentistry: a review for the
millennium. The Journal of Clinical Pediatric Dentistry 22,
167–77.
Lewis B (2010) The obsolence of formocresol. Journal of the
California Dental Association 38, 103–7.
Lewis BB, Chestner SB (1981) Formaldehyde in dentistry: a
review of mutagenic and carcinogenic potential. Journal of
the American Dental Association 103, 429–34.
Li J, Dasgupta PK, Luke W (2005) Measurement of gaseous
and aqueous trace formaldehyde. Revisiting the pentanedi-
one reaction and field applications. Analytica Chimica Acta
531,51–68.
Marsh GM, Youk AO, Morfeld P, Collins JJ, Symons JM
(2010) Incomplete follow-up in the national cancer insti-
tute’s formaldehyde worker study and the impact on sub-
sequent re analyses and casual evaluations. Regulatory
Toxicology and Pharmacology 58, 233–6.
McDonnell G, Russell AD (1999) Antiseptics and disinfec-
tants: activity, action, and resistance. Clinical Microbiology
Reviews 12, 147–79.
Milnes AR (2008) Is formocresol obselete? A fresh look at
the evidence concerning safety issues. Journal of Endodon-
tics 34, S40–6.
National Institute for Occupational Safety and Health (NI-
OSH) (1988) Occupational Safety and Health Guideline for
Formaldehyde. URL http://www.cdc.gov/niosh/docs/81-
123/pdfs/0293.pdf (accessed March 2014).
Nazaroff WW, Singer BC (2004) Inhalation of hazardous air
pollutants from environmental tobacco smoke in US resi-
dences. Journal of Exposure Analysis and Environmental Epi-
demiology 14(Suppl), S71–7.
National Industrial Chemicals Notification & Assessment
Scheme (NICNAS) (2006) Priority Existing Chemical
Assessment Report No. 28 Formaldehyde pp; 41,42,
54. URL http://www.nicnas.gov.au/data/assets/pdf_file/
0006/4389/PEC_28_Formaldehyde_Full_ReportPDF.pdf
(accessed March 2014).
Occupational Safety and Health Administration (OHSA)
(2002) Formaldehyde Fact sheet pdf. U.S. Department of
Labor. URL http://www.osha.gov/OshDoc/data_General_
Facts/formaldehyde-Factsheet.pdf (accessed September
2010).
Oliver C, Abbott P (1991) AH26 re-examined. Australian
Endodontic Newsletter 17,24–7.
Peters OA, Sch€
onenberger K, Laib A (2001) Effects of four
Ni–Ti preparation techniques on root canal geometry
assessed by micro computed tomography. International
Endodontic Journal 34, 221–30.
Peters OA, Peters CI, Sch€
onenberger K, Barbakow F (2003)
ProTaper rotary root canal preparation: effects of canal
anatomy on final shape analysed by micro CT. Interna-
tional Endodontic Journal 36,86–92.
Pinheiro HLC, de Andrade MV, Pereira PA, de Andrade JB
(2004) Spectrofluorimetric determination of formaldehyde
in air after collection onto silica cartridges coated with
Fluoral P. Microchemical Journal 78,15–20.
Restani P, Gali CL (1991) Oral toxicity of formaldehyde and
its derivatives. Critical Reviews in Toxicology 21, 315–28.
Ruyter IE (1980) Release of formaldehyde from denture base
polymers. Acta Odontologica Scandinavica 38,17–27.
Santerre JP, Shajii L, Leung BW (2001) Relation of dental
composite formulations to their degradation and the
release of hydrolyzed polymeric-resin-derived products.
Critical Reviews in Oral Biology & Medicine 12, 136–51.
Scott CR, Wolf PA (1962) The antibacterial activity of a
series of quaternaries prepared from hexamethylenetetra-
mine and halohydrocarbons. Applied Microbiology 10,
211–6.
Seale NS (2010) Article on formocresol is misleading. Journal
of the California Dental Association 38, 309.
Sp
angberg LS, Barbosa SV, Lavigne GD (1993) AH 26
releases formaldehyde. Journal of Endodontics 19, 596–8.
Sweetman SC (2011) The Complete Drug Reference. Martindale,
37th edn. London: Pharmaceutical Press, pp. 325, 1788.
Szarvas T, Szatloczky E, Volford J, Trezl L, Tyihak E, Rusznak
I (1986) Determination of endogenous formaldehyde level
in human blood and urine by dimedone -
14
C radiometric
method. Journal of Radioanalytical and Nuclear Chemistry
Letters 106, 357–67.
Tang X, Bai Y, Duong A, Smith MT, Li L, Zhang L (2009)
Formaldehyde in China: production, consumption, expo-
sure levels, and health effects. Environment International
35, 1210–24.
Weis MV, Parashos P, Messer HH (2004) Effect of obturation
technique on sealer cement thickness and dentinal
tubule penetration. International Endodontic Journal 37,
653–63.
WHO (2001) Formaldehyde. Air Quality Guidelines, 2nd
edn, Chapter 5.8. Copenhagen: WHO, pp. 3. URL https://
www.google.com.au/#q=WHO+foramdehyde+air+quality+
guidlines (accessed February 2014).
Formaldehyde, controversy, risks, quantification Athanassiadis et al.
©2014 International Endodontic Journal. Published by John Wiley & Sons LtdInternational Endodontic Journal,48, 829–838, 2015
838
