 

	UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

WASHINGTON, D.C.  20460

OFFICE OF           

PREVENTION, PESTICIDES

AND TOXIC SUBSTANCES

Date: March 27, 2008

MEMORANDUM

SUBJECT:	Formaldehyde/paraformadehyde  - Report of the Antimicrobials
Division Toxicity Endpoint Selection Committee (ADTC). 

FROM:	Timothy F. McMahon, Ph.D., Chair, ADTC

Antimicrobials  Division (7510C)

 

TO:		Sharon Carlisle, Chemical Review Manager 

		Mark Hartman, Branch Chief

		Regulatory Management Branch II

		Antimicrobials Division (7510C)

		

		         And

		Timothy F. McMahon, Ph.D., Risk Assessor

		Antimicrobials Division  (7510C)

 		PC Code:  043001; 043002 

On  January 29, 2008,  the Antimicrobials Division's Toxicity Endpoint
Selection Committee (ADTC)  evaluated the toxicology data base of
formaldehyde,  and selected  toxicological endpoints appropriate for the
currently registered uses of formaldehyde and paraformaldehyde. The
conclusions of this meeting are presented in this report.



Committee Members in Attendance

Members present were:  John Redden;  Stephen Dapson, Ph.D.;  Jonathan
Chen, Ph.D., Michelle Centra, Najm Shamim, Ph.D; Timothy F. McMahon,
Ph.D.; Roger Gardner; Jenny Tao.

 



I. INTRODUCTION

Formaldehyde  is used primarily as a fumigant in agricultural premises
such as poultry and swine farms and processing plants as well as in
citrus and mushroom houses.  It is used as a hard surface disinfectant
in commercial premises, industrial premises and veterinary clinics. 
Formaldehyde containing products are also used in oil drilling wells for
preservation of processing waters. There are no dietary uses of
formaldehyde.  

™Fumigant is used as a bacteriostat, fungicide, and sanitizer in
hair/beauty salons and barber shops, and Sun Pac is used as mildewcide
for closets, cupboards, dresser drawers, trunks, suitcases, lockers,
golf bags, trailers, bathroom and kitchens, and mobile homes.



II.	HAZARD IDENTIFICATION

A1. Acute Reference Dose (aRfD) [general population including infants
and children]

There are no expected dietary exposures to formaldehyde or
paraformaldehyde from the currently registered uses.  Therefore an 
acute reference dose value is not needed. 

 

 

A2. Acute Reference Dose (females 13-49) 

 There are no expected dietary exposures to formaldehyde or
paraformaldehyde from the currently registered uses.  Therefore an 
acute reference dose value is not needed. 

 	B. Chronic Reference Dose (cPAD)

There are no expected dietary exposures to formaldehyde or
paraformaldehyde from the currently registered uses.  Therefore a
chronic reference dose value is not needed.  There is an existing
chronic reference dose value in the current EPA IRIS assessment for
formaldehyde, but for registered antimicrobial uses this value is not
needed. 

C. Incidental Oral Exposure  

 The committee determined that an incidental oral endpoint was not
needed for formaldehyde. Formaldehyde is  highly volatile with a low
percentage of active ingredient in those products with residential
exposures (laundry detergents, general household cleaners) and residues
available for incidental oral exposure are not expected to occur.  An
accidental ingestion is considered a misuse and is not a regulatory
endpoint. Therefore, no incidental oral endpoint was selected.  



 

D. Dermal Exposure  

There are available studies on the dermal irritancy and dermal
sensitization potential of formaldehyde (Krivanek et al., Flyvholm et
al, 1997). These data demonstrate irritancy and sensitization potential
of formaldehyde, which is well known. However, the committee determined
that dermal endpoints are not needed for formaldehyde for the registered
antimicrobial uses. Residential uses do not involve purposeful contact
with the skin.  Use in laundry detergents and household cleaners is not
expected to result in any significant dermal exposure based on the high
water solubility of formaldehyde and the volatility of the active
ingredient. 

 

E. Inhalation Exposure (all durations) 

The committee considered the data set forth in the 2001 ACGIH
publication on formaldehyde as relevant for selection of a non-cancer
inhalation endpoint for use in risk assessment. Inhalation is the major
route of exposure to formaldehyde and thus hazard and risk need to be
addressed from inhalation exposures. 

As noted in the ACGIH document, “A TLV-Ceiling of 0.3 ppm (0.37 mg/m3)
is recommended for occupational exposure to formaldehyde. This value is
recommended to minimize the potential for sensory irritation, chiefly
eye and upper respiratory tract. Although the recommended TLV is
intended to protect nearly all workers, ACGIH recognizes that the value
may not safeguard that portion of the worker population (10-20%)
reported to be responsive to low ambient concentrations (< 0.25 ppm), of
the chemical…”

Additional published literature on human exposure to formaldehyde via
inhalation (Ballarin, C. at al., Mutat. Res. 280(1), 1992;
Akbar-Khanzadeh F., Am. J. Ind. Med. 26(1), 1994; Dally, KA et al.,
Arch. Environ. Health 36(6), 1981; Alexandersson and Hedenstierna, Arch.
Environ. Health 43(3), 1988; 	Alexandersson and Hedenstierna, Arch.
Environ. Health 44(1), 1989) show effects of formaldehyde on pulmonary
function and irritation of the eye and nasal passages at similar
concentrations.  The level of formaldehyde causing such symptoms does
not appear to vary regardless of exposure duration (i.e. hours to
years). 

In 2005, the Registration Division, in conjunction with the
Antimicrobials Division, issued an emergency exemption for use of
paraformaldehyde to decontaminate microbiological containment areas and
equipment. In this assessment, the NOAEL of 0.1 ppm was selected  from
the epidemiology study of Horvath et al. (JAMA 259(5), 1988), who
recorded complaints of eye, nose, and throat irritation in particle
board workers at concentrations of formaldehyde from 0.4 – 1.0 ppm. 
Four additional studies were cited in the 2005 assessment as co-critical
in support of the selected NOAEL value. 

For occupational exposures, the committee felt that the 0.1 ppm
inhalation endpoint was appropriate. No uncertainty factor was applied
for occupational assessments.  However, for residential and children’s
inhalation exposure, an uncertainty factor of 10x was applied to the 01
ppm endpoint.  The common effects of formaldehyde exposure are various
symptoms as a result of irritation of the mucosa in the eyes and upper
airways. In the non-industrial indoor environment, sensory reactions are
typical effects, but there are large individual differences in the
normal population and between hyperreactive and sensitized people. 
Thus, the committee concluded that in order to protect sensitive
subpopulations (children, older people and/or sensitized persons), an
extra safety factor of 10X is applied to address this concern. 

									



F. Recommendation for Aggregate Risk Assessments

 Inhalation exposures are appropriate for aggregate risk assessment. 

CLASSIFICATION OF CARCINOGENIC POTENTIAL

Formaldehyde has been extensively investigated for carcinogenic
potential and several regulatory organizations including the government
of Australia and IARC have reviewed the available carcinogenicity data
in animals and humans with respect to formaldehyde.  A detailed review
is not necessary here but links to the documentation are:   HYPERLINK
"http://www.nicnas.gov.au"  www.nicnas.gov.au  for the Australian
review, and   HYPERLINK "http://monographs.iarc.fr/" 
http://monographs.iarc.fr/  for the IARC publication.  The U.S. EPA
National Center for Environmental Assessment (NCEA) is also in the
process of updating the carcinogenic assessment for formaldehyde. 
Information from the draft document is reproduced here with permission. 

The Agency is currently reevaluating the carcinogenic potential of
formaldehyde. The historical and ongoing development of an inhalation
unit risk value to assess the carcinogenic potential of formaldehyde is
briefly summarized below. Contributors to this summary included
scientists from several EPA program offices (OPP, OPPT, ORD, OAR, and
NHEERL)

e human carcinogen with a potency factor of 1.3 E-5 per (μg/m3) ) on
the basis of squamous cell nasal tumors observed in a two-year study in
rats (Kerns et al., 1983).  

In 1999 the Chemical Industry Institute of Toxicology (CIIT) developed a
health risk assessment for formaldehyde based upon the animal toxicology
data (CIIT, 1999).  This document presented the dose-response modeling
of these data in two distinct parts: 1). based upon a biologically-based
dose response (BBDR) model , 2) benchmark dose models that were based
upon point of departures at various response levels of the tumor and
precursor data.  Both these approaches made extensive use of the
available time-to-tumor and mechanistic information. The 1999 assessment
was subsequently published in various articles in peer-reviewed journals
(2001, 2002, 2003, 2004).

In 1999, the U.S. EPA’s Office of Air and Radiation and Office of
Research and Development, in conjunction with Health Canada, conducted
an external peer review workshop for the CIIT BDDR model as well as an
external written peer review and public comment period for their
assessments. While the review was largely positive on the overall
approach in the assessment, reviewers also pointed to the potential for
significant uncertainty due to model mis-specification and uncertainties
in key parameters involved in the BBDR model

Based on the peer review of the CIIT model, OAR determined in 2004 that
the CIIT model was the most appropriate tool for risk assessment for
formaldehyde.   OAR has subsequently used the formaldehyde cancer
potency derived using the CIIT model for a number of risk assessments
involving formaldehyde emissions to the atmosphere such as the Plywood
and Composite Wood Products National Emission Standard for Hazardous Air
Pollutants (final rule 2004, reconsidered final rule 2006, remanded to
EPA by court 2007); Control of Hazardous Air Pollutants from Mobile
Sources (Final Rule 2007); and Proposed Rule for National Emission
Standard for Combustion Turbines (2004). Health Canada, Australia, the
World Health Organization, and the German MAK Commission have also used
the CIIT model. Model strengths include consideration of the mode of
action data for formaldehyde and a conservative approach to account for
potential direct DNA interaction and mutation induction.  Model
uncertainties include variability for some of the parameters of the
model (e.g., cell proliferation) which can affect predictions of risk
(Subramanian et al 2007;   2008 [in press]).

In 2004, NCEA convened a panel of experts, including scientists from
CIIT, to provide advice on these and other critical biological and
statistical uncertainties.  The strength of the CIIT model is its
consideration of mode of action and extensive mechanistic information.

Although current OAR assessments still use the CIIT model, these
assessments now acknowledge previously unknown uncertainties with the
CIIT model when characterizing the risk results.   

In 2004, the International Agency for Research on Cancer (IARC)
characterized formaldehyde as a human carcinogen based on their review
of the current literature (IARC, 2004), including data in humans on 
nasopharyngeal cancer,  cancer of the nasal cavity and paranasal
sinuses,and  leukemia.  It should be noted that some epidemiology
studies did not find a reported association between formaldehyde
exposure and carcinogenicity. For example, Coggon et al, 2003 studied
over 14,000 workers exposed to formaldehyde in industrial workplaces and
reported no excesses of either leukemia or nasal and nasopharyngeal
cancer.

In 2005, the Scientific Review Panel (SRP) of the California Office of
Environmental Health Hazard Assessment responded to the CA Air Resources
Board request to reevaluate the carcinogenic potential of formaldehyde. 
The Panel noted in this 2005 review that OEHHA’s November 2002
evaluation of a petition had included the 1999 report on the CIIT model
and other information, and that California’s OEHHA had concluded that
“the evidence…(1) did not change the determination that formaldehyde
is a carcinogen; (2) presented information that considered the
possibility of non-linear dose response relationships, but presented no
clear grounds to review the original “no threshold” determination;
and (3) did not provide any new epidemiology or bioassays supporting a
change in potency.   In addition, there was insufficient information to
fully evaluate the CIIT model, issues such as model uncertainty were not
adequately addressed….”   The Scientific Review Panel’s overall
conclusion in 2005 was,  “The Panel concluded that there was not
sufficient new data to support the petition to review the [OEHHA’s
earlier 1992] formaldehyde risk assessment.  In addition, the newly
published studies represented relevant new information, but they did not
allow determination of a causal relationship between formaldehyde
exposure and leukemia.  These studies deserve further evaluation over
time given their potential importance.”  Froines (2005).

 

EPA is currently completing a new IRIS assessment and unit risk value
for formaldehyde; the reassessment is scheduled to start internal peer
review in May 2008 and begin independent external peer review in January
2009
(http://cfpub.epa.gov/ncea/iristrac/index.cfm?fuseaction=viewChemical.sh
owChemical&sw_id=1031).  EPA anticipates that the peer review of the
formaldehyde assessment will be a longer process then that of EPA’s
reregistration process scheduled to conclude in September 2008. 

	

Based of the on going development of the science to predict carcinogenic
potential of formaldehyde, OPP has decided to present the formaldehyde
cancer risks for the pesticidal uses using both the existing 1991 IRIS
cancer unit risk of 1.3 E-5 per (µg/m3) and the CIIT BBDR model until
any new cancer estimates are fully peer reviewed.  OPP also acknowledges
the wide range in cancer risks using these approaches and will
coordinate with other offices in EPA on the outcome of the upcoming peer
review process on the carcinogenicity of formaldehyde.  Because
formaldehyde air concentrations approach those associated with ocular
and respiratory tract irritation, the risk mitigation measures to be
implemented in the meantime for the pesticidal uses will be based on
mitigating the non-cancer effects at a limit of 0.01 ppm.  It is
believed that this level will reduce exposures sufficiently such that
the cancer risks would not be of concern.  The EPA  process of
regulating pesticides allows for reevaluation at any time if new
information from the peer review process of the carcinogenic potential
of formaldehyde warrants.

MUTAGENICITY

 Formaldehyde’s mutagenicity has been examined in a variety of in
vitro and in vivo test systems. In a bacterial reverse mutation test
(MRID 00132156), formaldehyde (2%) was tested at concentrations of
0.001, 0.01, 0.10, 1.0, or 5.0 µL and found to be negative. In a second
submitted study (MRID 00132157), formaldehyde (2%) was tested at
concentrations of 3.0, 15.0, 75.0, 150, or 300 µg/plate and found to be
positive in the bacterial reverse mutation assay. Formaldehyde caused a
positive response (3.2-fold increase) on tester strain TA98 without
metabolic activation. A 1.9-fold increase was observed on TA98 with
metabolic activation. Also, increases of 2.2-fold and 1.7-fold were
observed on tester strain TA100 with and without activation,
respectively. In an in vitro mammalian chromosome aberration test (MRID
00132168), formaldehyde (37% formalin), was tested on Chinese hamster
ovary cells at concentrations of 28.43, 37.91, or 50.55 nL/mL. The test
article caused a significant dose-dependant increase in the frequencies
of chromosome aberrations in the Chinese Hamster Ovary cells, both with
and without S-9 activation. One submitted study (MRID 00132169), tested
formaldehyde (37%) for Unscheduled DNA synthesis (UDS) in Primary rat
liver hepatocytes. The test material was tested at concentrations of
0.0005, 0.001, 0.005, 0.01, 0.02, or 0.04 µL/mL and found to cause no
significant increase in UDS in rat hepatocytes. 

In published studies, formaldehyde has shown both positive and negative
results in the Ames Salmonella assay (Donovan et al., 1983; Connor et
al., 1983, 1985;  Frei et al., 1984; Fiddler et al., 1984; Oerstavik and
Hongslo, 1985; Takahashi et al., 1985; Schmid et al., 1986; Zielenska
and Guttenplan, 1988;  Le Curieux et al., 1993; O’Donovan and Mee
(1993) Watanabe et al., 1996; Dillon et al., 1998; Ryden et al., 2000;
Wilcox et al., 1990; Jung et al., 1992; Marnett et al., 1985; Mueller et
al., 1993).

	Temcharoen and Thilly (1983) examined the capacity of formaldehyde to
induce forward mutations to 8-azaguanine resistance in S. typhimurium TM
677, a his+ revertant of TA 1535. Both toxicity and mutagenicity were
obtained at formaldehyde concentrations of 0.17 mM in the absence of S9
and 0.33 mM in the presence of S9 Dillon et al. (1998) employed
Salmonella strains TA102 and TA104 because they are more sensitive to
oxidative mutagens. Formaldehyde was mutagenic in both strains, as well
as in TA100. However, the authors reported that the mutagenic activity
was not reduced in TA104 in the presence of S9 from either
Aroclor-induced male Fischer F 344 rats or male B6C3F1 mice. 

	In another study, formaldehyde induced forward mutations to
trifluorothymidine resistance in mouse lymphoma L5178Y tk+/- cells both
in the absence and presence of rat liver S9 (higher concentrations
required for effect with S9). Both toxicity and mutagenicity were
abolished when formaldehyde dehydrogenase was incorporated in the
exposure medium (Blackburn et al., 1991).

d by formaldehyde, with SSB at concentrations greater than 200 M and
a reduction of radiation-induced breaks (indirect measure of DPX) at 50
M. Formaldehyde-induced DPX were repaired 24 hours after the compound
was removed from the culture.

	

	In vivo, no treatment-related increase in either micronuclei or
chromosome aberrations were observed following  intraperitoneal exposure
to formaldehyde at 0, 6.25, 12.5, or 25 mg/kg. (Natarajan et al. (1983)
).  Similarly,  chromosomal analysis of spermatocytes at metaphase I did
not reveal any chromosomal lesions in Q strain mice injected
intraperitoneally with 50 mg/kg of the compound (Fontignie-Houbrechts,
1981).   Exposure of male and female Fischer F-344 rats to 0.5, 6, or 15
ppm (0.6, 7.4, 18.5 mg/m3) formaldehyde by inhalation for 6 hours/day
for 5 days showed no increases in either SCE or chromosome aberrations
at any dose level (Kligerman et al. (1984) )  . 

FQPA CONSIDERATIONS		

There are no tolerances for formaldehyde or paraformaldehyde and the use
patterns considered for the reregistration eligibility decision do not
involve dietary exposure.  As a result, a FQPA safety finding is not
applicable.

DATA GAPS / REQUIREMENTS

There are no acceptable repeat dose dermal toxicity studies for
formaldehyde. Although currently there are no dermal endpoints selected,
a dermal toxicity study is considered a data gap with respect to the
dermal concentration of formaldehyde that causes systemic effects and
whether any systemic effects are precluded by the irritancy and/or
sensitization potential of the chemical. 

 

VII. 	ACUTE TOXICITY  

  Acute Toxicity data for Formaldehyde technical a.i.

Guideline Number	Study Type/ Test substance (% a.i.)	MRID

Number/ Citation	Results	Toxicity Category



870.1100

(§81-1)	

Acute Oral – Guinea Pig 

Purity 37.3% - Formaldehyde

	

00058054	

LD50 = 260 mg/kg	

II



870.1200

(§81-2)

	

Acute Dermal – Rat

Purity  37.3% - Formaldehyde

	

00058054	

  SEQ CHAPTER \h \r 1 LD50 = 300 mg/kg

	

II



870.1200

(§81-2)	

Acute Dermal – Rabbit

Purity  37.3% - Formaldehyde	

00058054	

  SEQ CHAPTER \h \r 1 LD50 = 240 mg/kg

	

II



870.1200

(§81-2)

	

Acute Dermal – Dog

Purity  37.3% - Formaldehyde

	

00058054	

  SEQ CHAPTER \h \r 1 LD50 = 550 mg/kg

	

II



870.1300

(§81-3)

	

Acute Inhalation – Mouse and Rat	

See Open Literature studies in Toxicity Profile for Formaldehyde



870.2400

(§81-4)

	

Primary Eye Irritation - 

Purity  37.3% - Formaldehyde

	

00058054	

Severe eye irritant	

I



870.2500

(§81-5)	

Primary Dermal Irritation 

Purity  37.3% - Formaldehyde

	

00058054

	

Formation of vesicles with superficial necrosis or nodules.

	

I



870.2600

(§81-6)	

Dermal Sensitization – Guinea pigs

Purity 40.0% - Formaldehyde

	

40161103	

Extreme Sensitizer	

NA



VIII. 	  SEQ CHAPTER \h \r 1 SUMMARY OF TOXICOLOGY ENDPOINT SELECTION
FOR OF FORMALDEHYDE

  SEQ CHAPTER \h \r 1 Exposure

Scenario	Dose Used in Risk Assessment

(mg/kg/day) 	Target MOE, UF, 

Special FQPA SF* for Risk Assessment	Study and Toxicological Effects

Dietary Risk Assessments

Acute Dietary

(general population including infants and children) 	An acute dietary
assessment is not needed for the registered antimicrobial uses of
formaldehyde. 

Chronic Dietary

(all populations)	 A chronic dietary assessment is not needed for the
registered antimicrobial uses of formaldehyde. 

Non-Dietary Risk Assessments

Incidental Oral  

	 An incidental oral risk assessment is not required for the registered
antimicrobial uses of formaldehyde. 

Dermal  (all durations)	A dermal risk assessment is not
r煥極敲⁤潦⁲桴⁥敲楧瑳牥摥愠瑮浩捩潲楢污甠敳⁳
景映牯慭摬桥摹⹥ܠ䤇桮污瑡潩൮愨汬搠牵瑡潩獮 万
䅏䱅⠠畨慭⥮㴠〠ㄮ瀠浰ഠഠ܍䙕㴠ㄠ⠠捯畣慰楴湯污
ഩ䙕㴠ㄠ‰爨獥摩湥楴污ഩܠ䄠䝃䡉㈠〰‱異汢捩瑡潩
⁮湯映牯慭摬桥摹൥

â

ä

å

æ

ç

'

(

7

P

¤

¨

Ã

á

ä

å

ç

  hk

 hk

hª

愀Ĥ

愀Ĥ

 Horvath, E.P. et al. (1986): JAMA 259(5): 701-707.  Based on complaints
of eye, nose, and throat irritation in particle board workers at
concentrations of formaldehyde from 0.4 – 1.0 ppm.  

Redden, J. (2005): Section 18 Emergency Exemption for the use of
Paraformaldehyde: U.S. Army Medical Research Institute of Infectious
Diseases.

Cancer	 Formaldehyde is currently classified as a  B1 (probable human
carcinogen) in EPA’s IRIS assessment. IARC has classified formaldehyde
as “carcinogenic to humans.” The Agency has decided to present the
formaldehyde cancer risks for the pesticidal uses using both the
existing 1991 IRIS cancer unit risk of 1.3 E-5 per (µg/m3) and the CIIT
BBDR model until any new cancer estimates are fully peer reviewed

 



 PAGE   

 PAGE   2 

 PAGE  4 

 PAGE  4 

		     	 

 PAGE  14 

