[Federal Register Volume 67, Number 5 (Tuesday, January 8, 2002)]
[Notices]
[Pages 894-901]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 02-396]


=======================================================================
-----------------------------------------------------------------------

DEPARTMENT OF HEALTH AND HUMAN SERVICES

Centers for Disease Control and Prevention


Airborne Exposure Limits for Chemical Warfare Agents GA (Tabun), 
GB (Sarin), and VX

AGENCY: Centers for Disease Control and Prevention (CDC), Department of 
Health and Human Services.

ACTION: Notice of proposed revisions to CDC recommendations for 
protection of public health and safety during disposal or transport of 
chemical warfare agents GA (tabun), GB (sarin), and VX through revision 
of worker and general population airborne exposure limits.

-----------------------------------------------------------------------

Purpose

    CDC presents proposed recommendations for airborne exposure limits 
for the chemical warfare agents GA (tabun or ethyl N,N-dimethyl-
phosphoramidocyanidate, CAS 77-81-6); GB (sarin or O-isopropyl-
methylphosphonofluoridate, CAS 107-44-8); and VX (O-ethyl-S-(2-
diisopropylaminoethyl)-

[[Page 895]]

methylphosphonothiolate, CAS 50782-69-9).
    Before these recommendations are finalized, CDC requests comments 
from the public, all interested parties, environmental and health 
regulators, the Department of Defense (DOD), and other organizations 
involved in handling, transporting, or demilitarizing chemical warfare 
agents. More specifically, CDC seeks scientifically and professionally 
defensible data or information that would persuade CDC to alter its 
recommendations to be more or less conservative.
SUMMARY: CDC's recommendations are based on comments by scientific 
experts at a public meeting convened by CDC on August 23-24, 2000, in 
Atlanta, Georgia; the latest available technical reviews; and the 
latest available risk assessment approach frequently used by regulatory 
agencies and other organizations (1). Airborne exposure limits for 
chemical warfare agents GA (tabun), GB (sarin), and VX were re-
evaluated by using a conventional risk assessment methodology for 
developing airborne exposure limits described by the U.S. Environmental 
Protection Agency (EPA). This methodology is considered conservative; 
however, the calculated exposure limits are not numerically precise and 
do not define precise thresholds of potential human toxicity.

    Note: There is no indication that the current exposure limits, 
as implemented by the U.S. Army Program Manager for Chemical 
Demilitarization, have been less than fully protective of human 
health. This may be due to rigorous exposure prevention efforts in 
recent years as well as the conservative implementation of the 
existing limits.

    Proposed Airborne Exposure Limits for GB: CDC proposes a worker 
population limit (WPL) value of 3  x  10-5 mg/m\3\, 
expressed as an 8-hour time-weighted average (TWA). Additionally, CDC 
recommends a short-term excursion limit (STEL) of 1  x  10-4 
mg/m\3\ to be used in conjunction with the WPL. Exposures above the WPL 
up to the STEL should not be longer than 15 minutes and should not 
occur more than 4 times per day, and there should be at least 60 
minutes between successive exposures in this range. The STEL should not 
be exceeded during the work day, even if the cumulative exposure over 
the 8-hour TWA is not exceeded. CDC proposes a decrease in the general 
population limit (GPL) to 1  x  10-6 mg/m\3\. These WPL, 
STEL, and GPL values are approximately threefold lower than the values 
recently recommended by the U.S. Army. An immediately dangerous to life 
or health (IDLH) value of 0.1 mg/m\3\ is proposed for GB.
    Proposed Airborne Exposure Limits for GA: Although not as well-
studied as GB, GA is approximately equal in potency to GB. Therefore, 
CDC proposes the same exposure limits for GA as for GB.
    Proposed Airborne Exposure Limits for VX: CDC proposes that the VX 
WPL, expressed as an 8-hour time-weighted average, should be decreased 
to 1  x  10-6 mg/m\3\. Additionally, CDC proposes a VX STEL 
of 4  x  10-6 mg/m\3\. These proposed WPL and STEL exposure 
limits are a factor of 10 lower than the U.S. Army's recommendation. 
CDC proposes that the GPL for VX should be decreased to 6  x  
10-7 mg/m\3\ (a factor of 2 higher than the Army's 
recommendation). An IDLH value of 0.003 mg/m\3\ is proposed for VX. 
Acknowledging the gaps in the data base for this agent, CDC considers 
the proposed VX exposure limits subject to re-evaluation in 3 years. 
New VX toxicity studies, which are anticipated to be completed within 3 
years, have been recommended recently by the EPA National Advisory 
Committee for Acute Exposure Guideline Levels for Hazardous Substances 
(NAC/AEGL Committee). CDC agrees that new toxicity studies may be 
helpful in setting VX exposure limits.

DATES: Submit comments within 60 days following the date of this 
publication in the Federal Register.

ADDRESSES: Comments may be sent to the following:
    1. By mail. Submit your comments to Dr. Paul Joe, CDC, 4770 Buford 
Highway, Mail Stop F-16, Atlanta, Georgia 30341.
    2. In person or by courier. Deliver your comments to Dr. Paul Joe, 
CDC, 4770 Buford Highway, Mail Stop F-16, Atlanta, Georgia 30341.
    3. Electronically. You may submit your comments electronically by 
e-mail to Dr. Paul Joe at [email protected], or you can submit a computer 
disk to Dr. Paul Joe, CDC, 4770 Buford Highway, Mail Stop F-16, 
Atlanta, Georgia 30341. Electronic documents will be accepted in 
WordPerfect or Microsoft Word.

FOR FURTHER INFORMATION CONTACT: Dr. Paul Joe, Centers for Disease 
Control and Prevention, 4770 Buford Highway, Mail Stop F-16, Atlanta, 
Georgia 30341, Telephone number: 770-488-7091, E-mail address: 
[email protected].

SUPPLEMENTARY INFORMATION:

I. Background

    The nerve agents GA, GB, and VX are organophosphate esters that 
were designed specifically to cause incapacitation or death in military 
use. These agents are among the most potent of all chemical warfare 
agents and have extraordinarily high levels of acute toxicity. In vapor 
or aerosol form, the nerve agents can be inhaled or absorbed through 
the skin or the eyes. As a liquid, they can be absorbed through the 
skin, conjunctiva, and upper gastrointestinal tract. The agents' 
toxicity is related primarily to their ability to inhibit 
acetylcholinesterase, which is a critical enzyme needed for nerve 
function (2). Health symptoms can include runny nose, tightness in the 
chest, dimness of vision and pinpointing of eye pupils, difficulty in 
breathing, drooling and excessive sweating, nausea, vomiting, cramps, 
involuntary defecation and urination, twitching, staggering, headache, 
confusion, drowsiness, coma, and convulsions. The signs and symptoms 
can be followed by cessation of breathing and death (3). At superlethal 
doses, GB caused delayed neuropathy in antidote-protected chickens. VX 
has not been shown to cause delayed neuropathy in animals or humans. 
The health effects from low-dose chronic (long-term) exposure have not 
been demonstrated clearly.
    Studies of genotoxicity, carcinogenicity, developmental, and 
reproductive toxicity associated with GB and VX have been primarily 
negative (2,4).
    GA, GB, and VX no longer are manufactured in the United States; 
however, they are stored currently at eight locations in the 
continental United States by the DOD. Section 1412 of Public Law 99-145 
[50 U.S.C. 1521] mandates that the present stockpile of chemical 
warfare agents be destroyed. Public Law 91-121 and Public Law 91-441 
[50 U.S.C. 1512] mandate that the Department of Health and Human 
Services (DHHS) review DOD plans for transporting and/or disposing of 
chemical warfare agents and make recommendations for protecting human 
health and safety. DHHS delegated this authority to CDC.
    In 1987, CDC requested public comments on recommendations for 
protecting human health and the environment from potential adverse 
effects of long-term exposure to low-airborne doses of agents GA, GB, 
VX, mustard, and lewisite (L). CDC incorporated public comments, 
including comments from scientific experts outside CDC, and in 1988 
recommended worker and general population airborne exposure limits for 
GA, GB, VX, mustard (H, HD, HT), and L. (See Table 1.) The U.S. Army 
adopted these airborne exposure limits in 1990.

[[Page 896]]



                          Table 1.--Current CDC-Recommended Airborne Exposure Limits\*\
                       [All values expressed as milligrams per cubic meter air [mg/m\3\]]
----------------------------------------------------------------------------------------------------------------
                                     General population
              Agent                     limit (GPL)                    Worker population limit (WPL)
----------------------------------------------------------------------------------------------------------------
GA, GB..........................  0.000003 (3  x  10-6)..  0.0001 (1  x  10-4).
VX..............................  0.000003 (3  x  10-6)..  0.00001 (1  x  10-5).
H, HD, HT.......................  0.0001 (1  x  10-4)....  0.003 (3  x  10-3).
L...............................  0.003 (3  x  10-3).....  0.003 (3  x  10-3).
Averaging Time..................  72 hours...............  8 hours.
----------------------------------------------------------------------------------------------------------------
\*\ Referred to as ``Control Limits'' in Federal Register, Volume 53, No. 50, March 15, 1988, pp. 8504-07.

    The GPL is the maximum concentration to which members of the 
general population may be continually exposed 24 hours per day, 7 days 
per week. The GPL is intended for application to the entire general 
population, including all ages and medical conditions (e.g., infants, 
elderly, infirm, and healthy). Historically, the GPL for VX did not 
reflect a tenfold reduction from GB GPL as was the case for the WPLs. 
The primary reason for the difference was the technical inability to 
conduct VX air monitoring at such low concentrations. The analytical 
limitations are reflected further in the 72-hour averaging period 
rather than a more conventional 24-hour period.
    The WPL is intended to be assessed as a time-weighted average for a 
conventional 8-hour workday and a 40-hour week. This WPL represents a 
concentration to which it is believed that virtually all workers may be 
repeatedly exposed, day after day, without adverse effect. CDC 
recommends that the WPLs be implemented in conjunction with the medical 
surveillance provisions and other requirements defined in Department of 
Army Pamphlets 40-173 and 40-8 or successive documents (5,6).

    Note: The proposed risk assessment methodology derives exposure 
limits below concentrations where any acute or chronic effects would 
be expected to occur. The existing and proposed exposure limits are 
intended to protect workers and the public from potentially adverse 
effects from short-term or long-term exposure to GA, GB, and VX. The 
existence of potential adverse health effects resulting from long-
term, low-dose exposure to these agents has not been demonstrated 
clearly.

    Now, 13 years later, CDC is re-evaluating the limits for GA, GB, 
VX, and mustard based on the latest risk assessment models and any 
updated scientific data. On August 23-24, 2000, CDC convened a public 
meeting in Atlanta, Georgia, where outside scientists joined CDC 
scientists to discuss the exposure limits for GA, GB, and VX. The re-
evaluation consisted of lengthy review of all available information 
about the agents, including some information previously classified by 
allied nations, and therefore, are unavailable for the open review 
process used by CDC in the past. A public meeting to discuss the 
exposure limits for mustard was held on September 11-12, 2001. The 
proposed mustard exposure limits will be presented in a separate 
Federal Register announcement. The L stockpile is relatively small and 
located at only one storage site; therefore, revisions to the exposure 
limits for L are not being considered at the present time.

II. Approach and Methodology

A. Purpose of the Public Meeting

    The purpose of the public meeting was to discuss the airborne 
exposure limits for GA, GB, and VX recently proposed by the US Army. 
Attendees at the August 23-24, 2000, public meeting convened by CDC 
included risk assessors, toxicologists, physicians, a veterinarian, and 
several chemists. These experts were from universities, state 
environmental agencies, and non-CDC federal agencies. The scientific 
experts were asked whether or not there was a need to modify exposure 
limits to reflect current risk assessment methodologies and any newly 
available data. The meeting agenda included the following:
     Presentations on risk assessment models and scientific 
data and recommended modifications to existing exposure limits based on 
comments from individual scientific experts,
     Panel discussions by scientific experts, and
     Discussions of the technical feasibility to monitor at 
proposed modified exposure limits by air monitoring experts.
    The meeting was not held as a federal advisory committee; 
therefore, CDC did not seek unanimity or consensus; take votes; or rely 
solely on the attendees to formulate federal policy. Statements by 
members of the working group, which are included in this Federal 
Register notice, represent only one part of the information considered 
by CDC. The experts attended the meeting solely to provide their 
individual expert advice to CDC and the public for consideration.

B. Method for Deriving Exposure Limit Criteria

    The EPA risk assessment approach, which was used in this 
assessment, is used to extrapolate potential biological effects in 
humans at low-level exposures where such epidemiologic or toxicologic 
data are not directly available. This method for deriving exposure 
criteria has evolved over 30 years. This evaluation's approach was 
based on guidance described in an EPA publication (7). The derivation 
of a non-cancer exposure criteria involve the following:
     Defining the critical adverse effect (which is assumed 
protective for all other, often more serious, effects);
     Selecting the most appropriate animal or human study or 
studies, if more than one yields the same end point, to serve as the 
basis for a limit;
     Establishing a threshold dose below which adverse health 
effects are not expected to occur or are extremely unlikely; and
     Defining appropriate uncertainty factors (UFs) to apply to 
the threshold dose.
    In selecting a study, a no-observed-effect-level (NOEL)--a product 
of concentration and time (Ct) at which subjects showed no detectable 
effects-or a no-observed-adverse-effect-level (NOAEL)--a Ct at which 
subjects showed no detectable harmful effects--is preferred over a 
lowest-observed-effect-level (LOEL) or a lowest-observed-adverse-
effect-level (LOAEL)--the lowest Ct at which an effect or adverse 
effect was seen. Studies of human responses generally are preferred 
over studies on laboratory animals. Such preferences are not rigid; the 
number of subjects and technical aspects of how the study was conducted 
play an important role.
    As many as five uncertainty factors and one modifying factor may be 
applied to the selected exposure dose, which is usually a NOAEL or 
LOAEL (7,8). They are UFH (heterogeneity of susceptibility 
within human

[[Page 897]]

populations), UFA (animal to human extrapolation), 
UFS (subchronic to chronic exposure extrapolation), 
UFL (LOAEL to NOAEL extrapolation), UFD (database 
incomplete), and MF (modifying factors). By convention, these factors 
assume values of 1, 3, or 10 and are multiplied together to yield an 
overall uncertainty factor. The observed Ct product (LOAEL or NOAEL) is 
divided by the overall UF and then adjusted for duration of exposure 
(40 hours per week for workers and 168 hours per week for the general 
population). This basic method was applied by EPA to many common 
toxicants to establish chronic reference doses for human exposure. In 
summary, the general equation for deriving the WPL would be as follows:
[GRAPHIC] [TIFF OMITTED] TN08JA02.075

WPL = Worker population limit
LOAELinhal= Lowest observed adverse effect level (if 
available, use no observed adverse effect level)
Respexptl = Experimental subject minute volume
Respoccup = Occupational minute volume
Expoccup = Occupational exposure time (480 minutes/day x 5 
days)
Expexptl = Experimental exposure time
UFs = Uncertainty factors
MF = Modifying factor

    The GPL would be derived in an analogous manner, adjusting for 
continuous exposure, differences in assumed respiratory rates, and 
possible differences in application of certain UFs.
    The exposure criteria resulting from this risk assessment approach 
should be evaluated in context with the uncertainties and default 
assumptions used in the risk assessment approach. One of the 
uncertainties that needs to be considered is the ``order-of-magnitude'' 
imprecision associated with the exposure criteria estimate (8). From a 
purely mathematical standpoint, this refers to a log10 
interval around the exposure criteria estimate (i.e., approximately 
threefold above and below). It is important to recognize that this 
imprecision includes only the statistical uncertainty in interpreting 
the underlying data. Uncertainties inherent in the choice of the model 
to conduct the extrapolation are potentially far larger and cannot be 
quantified easily. Research into specific areas of uncertainty 
associated with the EPA methodology has been reported. Most studies 
support the belief that the uncertainty factors described above provide 
estimates that are protective or err toward lower limits (9). That is, 
the composite uncertainty factor tends to result in an estimate of the 
dose (or exposure limit) that is likely not to cause adverse health 
effects.

III. Presentations at the Public Meeting

A. U.S. Army Proposal

    The U.S. Army completed reviews of exposure limits for G-agents and 
VX and suggested lowering the GPL for one of the agents (10,11). (See 
Table 2.) The Army's proposals decreased the GPL for VX by one order of 
magnitude, from 3  x  10-6 milligrams per cubic meter air 
(mg/m3) to 3  x  10-7 mg/m3, and 
decreased the averaging time from 72 hours to 24 hours.

                                  Table 2.--U.S. Army-Proposed Exposure Limits
                        [All values expressed as milligrams per cubic meter air [mg/m3]]
----------------------------------------------------------------------------------------------------------------
                    Agent                                                                          Immediately
---------------------------------------------     General           Worker         Short-term      dangerous to
                                                 population       population     exposure limit  life and Health
                                                limit (GPL)      limit (WPL)        (STEL)           (IDLH)
 
----------------------------------------------------------------------------------------------------------------
GA, GB                 Proposed.............  0.000003 (3x10 - 0.0001 (1x10 -   0.0004 (4x10 -   0.1 (1x10 -1).
                                               6) .     4).              4)**.
VX                     Proposed.............  0.0000003 (3x10  0.00001 (1x10 -  0.00004 (4x10 -  0.01 (1x10 -2).
                                               -7).     5).              5).
GD                     Proposed.............  0.000001 (1x10 - 0.00003 (3x10 -  0.001 (1x10 -    0.05 (5x10 -2).
                                               6).      5).              3)**.
GF                     Proposed.............  0.000001 (1x10 - 0.00003 (3x10 -  0.001 (1x10 -    0.05 (5x10 -2).
                                               6).      5).              3)**.
Averaging time         24 hours.............  8 hours........  15 min, 4x/day.  30 miyn. ......
----------------------------------------------------------------------------------------------------------------
 24-hour time-weighted average.
 ** 8-hour time-weighted average worker limit may not be exceeded.

    The U.S. Army proposed exposure limits for agents GD (Soman, O-
pinacolyl-methylphosphonofluoridate, CAS 96-64-0) and GF (O-cyclohexyl-
methylphosphonofluoridate, CAS 329-99-7). These agents are not part of 
the U.S. stockpile, and neither transportation nor open-air testing is 
being considered for these agents. Therefore, they fall outside the 
scope of the DHHS/CDC mandate and were not considered in this process.
    The U.S. Army-proposed WPL for GB, expressed as an 8-hour time-
weighted average, is identical to the existing WPL but was derived from 
a different source. The proposed WPL was based on a human study 
conducted in 1949 by McKee and Woolcott, which yielded a LOAEL of 0.06 
mg/m3, 20 minutes/day for 4 days per week (12). Proposed 
uncertainty factors were UFH = 1, UFA = 1, 
UFS = 10, UFL = 3, UFD = 1, MF = 1 for 
an overall uncertainty factor of 30. Adjusting for differences in 
breathing rates and exposure durations yielded 3.3  x  10-5 
mg/m3, expressed as an 8-hour time-weighted average. This 
differs from the existing limit, 1  x  10-4 mg/m3 
by a factor of 3. The U.S. Army authors concluded that the methodology 
was not sufficiently precise to warrant a change from the existing 
limit to the newly calculated limit and proposed leaving the current 
limit unchanged. The same study was used as a basis for a GPL of 1.1 
x  10-6 mg/m3, which differed from the present 
GPL (3  x  10-6 mg/m3) by a factor of 3 and was 
deemed within an acceptable uncertainty range.
    The Army proposed a STEL of 0.0004 mg/m3 for GB. The 
STEL is defined as a 15-minute time-weighted average exposure that 
should not be exceeded during the workday, even if the 8-hour WPL is 
not exceeded. Exposures up to the STEL should not be longer than 15 
minutes and should not occur more than 4 times a day, and there should 
be at least 60 minutes between successive exposures in this range. The 
proposed STEL would have the effect of permitting four, 15-minute 
exposures per day up to 0.0004 mg/m3 of GB or GA

[[Page 898]]

with the added requirement that the 8-hour WPL may not be exceeded.
    The Army proposed a value of 0.1 mg/m3 as the 
immediately dangerous to life or health (IDLH) concentration for GB. 
The GB IDLH was based on an acute human toxicity study, and the value 
was calculated in accordance with National Institute for Occupational 
Safety and Health (NIOSH) guidance (10). The Army adjusted the IDLH 
down by a factor of 2 to address the female occupational worker 
population, which is potentially more sensitive than the male 
occupational worker population to GB vapor.
    There are limited data on VX compared to some of the other G-
agents, but the WPL recommendation was based on a relative potency 
estimate of 10 for pupillary constriction; so the Army-proposed WPL for 
VX is \1/10\ of the corresponding value for GB or 1  x  10-5 
mg/m3, and the Army-proposed GPL is 3  x  10-7 
mg/m3. A STEL of 4  x  10-5 mg/m3 for 
VX was proposed. The value was based ultimately on the WPL, such that 
four exposures per day at the STEL would not cause the WPL to be 
exceeded. The Army-proposed VX IDLH of 0.01 mg/m3 was also 
determined using a relative potency estimate of 10.

B. Airborne Exposure Limits for GB

1. WPL for GB
    The U.S. Army document served as the starting point for discussion 
at the public meeting (10). The expert panel members differed in their 
assessment of how best to derive limits from the available data. Most 
of the members thought that limits for GB should be based on the McKee 
and Woolcott study, which yielded a LOAEL of 0.06 mg/m3 
(12). However, one member was concerned about deriving long-term 
exposure limits from short-term experimental data, particularly when 
little long-term toxicity data are available.
    One member noted that application of an interspecies uncertainty 
factor greater than 1 is unjustified when evidence suggests that the 
species studied is as sensitive or more sensitive than man. A 
calculation based on the Weimer animal data using an interspecies 
uncertainty factor of 1 would yield a GB WPL of 5 x 10-\5\ 
mg/m\3\ (13). However, this member also thought that human studies 
should be given more weight and joined others in recommending a limit 
based on the McKee and Woolcott report (12). Another member argued for 
a limit of 4 x 10-\6\ mg/m\3\ based on uncertainty factors 
of 10 for short-term to long-term extrapolation and 3 for 
interindividual variability. Yet another member argued that studies by 
Harvey and Johns would be better critical studies to utilize (14, 15). 
Working from these data would yield a 15-minute STEL of 0.008 mg/m\3\ 
and a WPL for GB of 2.5 x 10-\6\ mg/m\3\ after an eightfold 
adjustment for time of exposure and a tenfold adjustment for cumulative 
effect.
    Four members recommended that if the U.S. Army-proposed derivation 
were used, CDC should accept the calculated exposure limit value 
(3.3 x 10-\5\ mg/m\3\) rather than utilizing the rounded-up 
value 1 x 10-\4\ mg/m\3\ that was recommended by the Army.
    Several members speculated that information concerning human 
exposures during manufacture and disposal of GB could be more relevant 
than the studies cited. Unfortunately, records of environmental 
conditions from the time period GB was manufactured are not adequate to 
support such analysis. Conversely, worker and environmental monitoring 
records for recent GB demilitarization activities are well documented. 
However, engineering controls to prevent exposure have been rigorous; 
therefore, GB exposures have been very rare, have occurred primarily 
during maintenance operations, and have been minimal. Consequently, 
these data are not useful for developing exposure limits.
2. STEL for GB
    The U.S. Army-proposed STEL was based on the WPL, such that four 
exposures per day at the STEL would not result in the WPL being 
exceeded. At the public meeting, the proposed STEL elicited 
considerable discussion. Several members of the expert panel thought 
that the Army-proposed STEL was too low numerically because of the 
method used to calculate it. Using the critical effect LOAEL, several 
experts recalculated a new value--a Temporary Excursion Limit (TEL). 
The TEL for GB was calculated to be 0.01 mg/m\3\ for a 5-minute 
exposure not more than once per day.
3. General Population Limit for GB
    The Army-proposed GPL for GB was based on the same study and the 
same method used for deriving the WPL. The GPL was calculated by 
adjusting for the longer time of exposure and greater population 
variability. The uncertainty factors were as follows: 10 for short-term 
to long-term extrapolation, 10 for variability among individuals, and 3 
for low-effect to no-effect extrapolation. Three members of the working 
group thought the Army-proposed GPL was adequate. One thought that the 
proposed limits were probably at least tenfold lower than needed to 
protect public health. That is, the GPL could be at least tenfold 
greater and still be protective. The member who proposed a WPL of 
2.5 x 10-\5\ mg/m\3\ advised adding an uncertainty factor of 
3 for variation within the population and an uncertainty factor of 3 
for extrapolating from low-effect to no-effect yielding the proposed 
value 3 x 10-\6\ mg/m\3\ but by a different line of 
reasoning. One member argued for a GPL of 1 x 10-\6\ mg/
m\3\, noting that using the lower value incorporated an uncertainty 
factor of 3 for variability within population. It was noted that the 
Johns data indicate that doses causing a given degree of pupillary 
constriction generally range over a factor of less than 2.0 from the 
geometric mean (that is, from about half the geometric mean to about 
twice the geometric mean), providing at least some evidence for small 
variability within human populations to this particular low-dose effect 
(15).

C. Airborne Exposure Limits for VX

    Exposure limits for VX were more difficult for the experts to 
address because the experimental VX data were considered inadequate and 
do not form a good basis for VX exposure limits. Nonetheless, one of 
the working group members noted that the VX studies by Bramwell and 
Crook argue for a VX WPL of 4 x 10-\7\ mg/m\3\ and 
3 x 10-\9\ mg/m\3\, respectively (16, 17). However, several 
panel members had scientific concerns about these studies. Regarding 
the Bramwell study, some panel members were concerned that benzene, 
which was used as a solvent in the VX generation, could alter the 
exposure characteristic of VX. As for the Crook study, the accuracy of 
the VX vapor concentrations was questioned.
    Because the available experimental VX data were considered 
inadequate, the derivation of the exposure limit was based on the 
relative potency of VX as compared to GB. The exposure limits proposed 
by the Army are based on a tenfold difference (relative potency) in the 
ability of VX to cause miosis compared to GB. This tenfold potency 
difference was questioned because some publications stated that the 
potency difference may be twelvefold to thirty-threefold or higher, 
especially at low concentrations (18,19).
    The Army's publication proposing VX exposure limits included little 
detail used in deriving the tenfold potency factor (11). However, 
discussion in a previous U.S. Army study indicated that recovery from 
the miosis effects of VX is about four times as fast as recovery from 
the effects of GB (19). According to

[[Page 899]]

this publication, if VX potency is about twenty-five times greater than 
GB but the effective recovery is four times greater, the relative VX 
potency for continuous exposure would be approximately 25/4 that of GB. 
The overall factor of 10 is a approximation (rounding) of 25/4. 
Compared to GB, VX does not undergo a second, irreversible, reaction 
known as aging as quickly when it reacts with acetylcholinesterase; 
this may be one reason that the biological effects of VX exposure 
recover more quickly when compared to GB.
    CDC notes that the EPA NAC/AEGL committee for VX recently proposed 
a relative potency of 12 with application of a modification factor of 3 
for the incomplete VX data set. The potency factor of 12 was based on a 
1971 British study that measured the ability of VX to cause 90% pupil 
constriction in rabbits (18). The application of a relative potency of 
12 with a modifying factor of 3 for the inadequate VX data base 
effectively resulted in a relative potency of 36 (3 x 12) (20).
    All experts would have preferred better, VX-specific data and would 
have preferred avoiding the use of relative potency, but four of the 
experts concurred with the use of an overall tenfold difference in 
relative potency for extrapolating from GB to VX. Given the preference 
of several experts for a GB WPL of 3 x 10-\5\ mg/m\3\, that 
would call for a WPL of 3 x 10-\6\ mg/m\3\ for VX rather 
than the 1 x 10-\5\ mg/m\3\ that was proposed by the Army. 
Accordingly, preference by several experts for a GB GPL of 
1 x 10-\6\ mg/m\3\ would suggest a VX GPL of 
1 x 10-\7\ mg/m\3\.

D. CDC's Proposed Recommendations

1. Airborne Exposure Limits for GB
    Noting the opinions of the experts at the public meeting, CDC 
proposes a change from the present exposure limit to the lower exposure 
limit derived from the McKee and Woolcott studies (12). CDC believes 
that the calculated WPL value of 3 x 10-\5\ mg/m\3\, 
expressed as an 8-hour time weighted average, will protect workers from 
short-term or long-term health effect exposures for a working lifetime. 
The CDC-proposed WPL value, consistent with the calculation from this 
risk assessment, is approximately threefold lower than the Army-
recommended value of 1  x  10-\4\ mg/m\3\. CDC recognizes 
that the risk assessment methodology is imprecise, and quantitative 
differences in risk between exposure concentrations of less than an 
order of magnitude generally are not discernable. However, CDC could 
not identify relevant examples from other risk assessments where such 
rounding-up had been conducted. Additionally, since the ``calculated'' 
WPL was thought to be technically feasible and four experts at the 
public meeting recommended using the ``calculated'' values from the 
risk assessment, CDC proposes the 3 x 10-\5\ mg/m\3\ as the 
WPL for GB.
    In addition to the WPL, CDC recommends a STEL of 
1 x 10-\4\ mg/m\3\ to be used in conjunction with the WPL. 
The STEL is defined as a 15-minute time-weighted average exposure that 
should not be exceeded during the workday, even if the 8-hour WPL is 
not exceeded. Exposures up to the STEL should not be longer than 15 
minutes and should not occur more than 4 times a day, and there should 
be at least 60 minutes between successive exposures in this range. The 
purpose of this value is to provide a reasonable limit to excursions 
over the WPL. The value associated with the STEL is numerically 
identical to the existing 8-hour time-weighted worker exposure limit.
    CDC proposes 1 x 10-\6\ mg/m\3\ as the GB GPL, expressed 
as a 24-hour time-weighted average. This GPL value, which is consistent 
with the calculation from the risk assessment, is threefold lower than 
the Army-recommended value and the current GPL. CDC believes current 
analytical methods can be modified to monitor at this new 
concentration.
    The expert panel members did not focus on, or object to, the Army-
proposed immediately dangerous to life or health (IDLH) value of 0.1 
mg/m\3\ for GB (10). Accordingly, CDC proposes an IDLH of 0.1 mg/m\3\.
2. Airborne Exposure Limits for GA
    Although not as well-studied, GA is approximately equal in potency 
to GB. The Army proposed, and members of the expert groups agreed, that 
it would be reasonable to use the same exposure limits for both. CDC 
proposes the same exposure limits (WPL, STEL, GPL, and IDLH) for GA as 
those recommended for GB.
3. Airborne Exposure Limits for VX
    Since the toxicity data for VX are inadequate, CDC proposes 
derivation of the VX WPL, STEL, and GPL from the calculated exposure 
limits for GB, using a relative potency of 12 compared to GB and 
application of a modifying factor of 3 for the incomplete VX data set. 
This approach, which effectively results in a relative potency of 36, 
is the same as that recently proposed by the EPA NAC/AEGL committee 
(20). CDC proposes that the WPL for VX should be decreased to 
1 x 10-\6\ mg/m\3\ (a factor of 10 lower compared to the 
current value and the U.S. Army's recommendation). Additionally, CDC 
proposes VX STEL of 4 x 10-\6\ mg/m\3\.
    CDC proposes a VX GPL of 6 x 10-\7\ mg/m\3\, expressed 
as a 24-hour time-weighted average. The VX GPL, derived from the GB GPL 
to which the relative potency of 12 and a modifying factor of 3 was 
applied, was initially calculated as 3 x 10-\8\ mg/m\3\. 
However, currently available monitoring methods are unable to reliably 
detect VX at this concentration. CDC believes that reliable monitoring 
is a crucial aspect for implementing the exposure limits and therefore 
proposes to increase the GPL to a concentration that can reliably be 
monitored. The CDC proposes 6 x 10-\7\ mg/m\3\ for the VX 
GPL, a value that is both protective and technically feasible to 
monitor.
    The proposed VX GPL of 6 x 10-\7\ mg/m\3\, used in 
conjunction with the existing perimeter montoring programs, will be 
protective because long-term releases of VX are unlikely. Routine 
maintenance and monitoring procedures implemented for worker safety 
near the potential sources of releases (where concentrations 
potentially would be higher than at the perimeter) prevent long-term 
releases. At demilitarization sites, perimeter monitoring results for 
12-hour samples are typically available within 72 hours. Detections of 
chemical agent above the action level result in (1) an investigation to 
determine the source of the vapor and (2) corrective action to 
eliminate the source. In the derivation of the GPL in accordance with 
EPA methodology, the exposure period of the critical study is adjusted 
for a continuous 7-day exposure for the general population. The 
perimeter monitoring results at demilitarization sites are obtained 
within 72 hours (3 days) following sampling. To correct the assumption 
of continuous exposure for 7 days, a factor of 3 days potential 
exposure per 7 days was applied to the calculated VX GPL of 
3 x 10-\8\ mg/m\3\. Additionally, in the derivation of the 
GPL, an uncertainty factor of 10 was applied to extrapolate from sub-
chronic to chronic exposures. Since a chronic exposure is unlikely, 
this extrapolation would not be needed. These calculations result in 
adjusting the initially calculated VX GPL of 3 x 10-\10\ mg/
m\3\ to 6 x 10-\7\ and support the conclusion that the 
proposed GPL of 6 x 10-\7\ is protective of human health. 
This adjustment of the VX GPL was made in acknowledgment of the 
technical limitations of current air monitoring methods, while assuring 
that the GPL would be protective of public health.

[[Page 900]]

    The expert group members did not object to the Army-proposed IDLH 
values for VX (0.01 mg/m\3\), although there was little specific 
discussion among the panel. In accordance with relative potency 
approach used for WPL and GPL (potency factor of 12 with a modifying 
factor of 3), CDC proposes a VX IDLH of 0.003 mg/m\3\.
4. Proposed Implementation of the VX GPL
    Current data suggest that air monitoring at the proposed VX GPL 
concentration is on the fringe of technical feasibility for current 
methods. CDC investigated this issue with representatives from NIOSH, 
the U.S. Army, and other independent consultants. CDC representatives 
heard compelling evidence that current VX air monitoring methods may 
need further development. At the proposed VX GPL, the mass of other 
ambient organic materials normally found in the air (background 
chemicals) will greatly exceed the mass of VX to be measured. These 
background materials cause analytical problems in discerning and 
quantifying VX.
    Halting disposal until improved monitoring methodology can be 
developed presents at least three grave problems:
    a. There is greater cumulative risk from continued storage compared 
to continued disposal under the existing exposure limits.
    b. The desired level of sensitivity and selectivity may not be 
easily attainable.
    c. The United States has treaty obligations to complete the 
disposal within a specified time.
    Inasmuch as delay in disposal presents an unacceptable risk to 
public health and safety, CDC proposes the following interim measures 
regarding monitoring at the proposed VX GPL:
    a. CDC proposes a multifaceted research program to look at 
commercially available systems that have the potential to improve air 
monitoring at the proposed VX GPL. Further, CDC recommends that the 
Army use one or more Ph.D.-level analytical chemist(s) who have air 
monitoring experience to direct this program.
    b. CDC proposes suspension of the 20% action level for the VX GPL 
until the monitoring methodology can be improved.
    c. For all demilitarization sites handling VX, CDC proposes that 
all qualitative responses above a 3:1 signal-to-noise ratio for VX from 
perimeter stations be evaluated (i.e., those that are below the limit 
of quantification for VX). When VX is qualitatively detected, action 
should be taken to investigate the possible sources of these responses.

E. Summary of Proposed Recommendations

    CDC's foremost concern is protecting human health and safety. This 
concern requires a carefully considered balance of utilizing best 
possible risk analysis while considering technical feasibility and 
avoiding unintended consequences of recommendations that could increase 
total risk. CDC's recommendations are made with this balance in mind.
    CDC proposes adjustments to the exposure limits for GA, GB, and VX 
to the values shown in Table 3.

                           Table 3.--DC Current and Proposed Airborne Exposure Limits
                       [All values expressed as milligrams per cubic meter air [mg/m \3\]
----------------------------------------------------------------------------------------------------------------
                             General population limit* (GPL)                                        Immediately
                       ------------------------------------------     Worker       Worker short-   dangerous to
         Agent                                                      population     term exposure      life or
                                                                   limit** (WPL)     limit***        health***
                                                                                      (STEL)          (IDLH)
----------------------------------------------------------------------------------------------------------------
GA, GB                  Current proposed........        3 x 10-6        1 x 10-4
                                                        1 x 10-6  3 x 10        1 x 10-4             0.1
                                                                              -5
VX                      Current proposed........        3 x 10-6        1 x 10-5
                                                        6 x 10-7        1 x 10-6        4 x 10-6           0.003
----------------------------------------------------------------------------------------------------------------
* 24-hour time-weighted average. For the VX GPL, analyses of sample results within 72 hours is required.
** 8-hour time-weighted average.
 To be implemented in conjunction with the GB STEL.
 To be implemented in conjunction with a VX STEL.
*** Not previously considered by CDC.

    Acknowledging the gaps in the database for VX, CDC considers the 
proposed VX exposure limits subject to re-evaluation in 3 years. New VX 
toxicity studies, which are anticipated to be completed within 3 years, 
have been recommended recently by the EPA NAC/AEGL Committee. CDC 
agrees that additional toxicity studies could be helpful in the 
derivation of exposure limits for VX.
    CDC does not specifically recommend the use of these airborne 
exposure limits for uses other than transportation, demilitarization, 
or general population protection. For example, the 8-hour WPL value 
historically has been used for the Army-designated 3X decontamination, 
surveillance activities of leakers in storage, and charcoal unit mid-
beds. CDC believes that the WPL is not necessarily applicable to all 
these activities, and the specific technical and safety requirements 
for each activity need to be considered individually.
    The allowable limits for stack emissions were not discussed at the 
meeting. The allowable stack concentration (ASC) is a ceiling value 
that serves as a source emission limit and not as a health standard. It 
is used for monitoring the furnace ducts and common stack. The ASC 
provides an early indication of an upset condition and must be 
measurable in a timely manner. Modeling of worst-case credible events 
and conditions at each installation should confirm that the GPL 
monitoring level is not exceeded at the installation boundary as a 
consequence of a release at or below the ASC. Lowering the GPL might 
have the effect of lowering the stack concentration limit; therefore, 
modeling will need to be conducted to determine if the existing ASCs 
continue to be appropriate.

    Dated: January 2, 2002.
Joseph R. Carter,
Associate Director for Management and Operations, Centers for Disease 
Control and Prevention (CDC).

References

    1. 65 FR 120, June 21, 2000, pp. 38559-60.
    2. Munro NB, Ambrose KR, Watson AP. Toxicity of the 
organophosphate chemical warfare agents GA, GB, and VX: implications 
for public protection. Environ Health Perspect 1994;102:18-38.
    3. U.S. Army. Detailed and general facts about chemical agents--
TB 218. Aberdeen Proving Ground, MD: U.S. Army Center for

[[Page 901]]

Health Promotion and Preventive Medicine, October 1996.
    4. Perrotta DM. Long-term health effects associated with 
subclinical exposures to GB and mustard. Armed Forces 
Epidemiological Board, Environmental Board. July 18, 1996.
    5. U.S. Army. Occupational health guidelines for the evaluation 
and control of occupational exposure to mustard agents H, HD, and 
HT. Washington, DC: Headquarters, Department of the Army, August 
1991; pamphlet 40-173.
    6. U.S. Army. Occupational health guidelines for the evaluation 
and control of occupational exposure to nerve agents GA, GB, GD, and 
VX. Washington, DC: Headquarters, Department of the Army, December 
1990; pamphlet 40-8.
    7. U.S. Environmental Protection Agency. Methods for derivation 
of inhalation reference concentrations and application of inhalation 
dosimetry. Washington, DC: U.S. Environmental Protection Agency, 
October 1994; Report No. EPA/600/6-90/066F.
    8. U.S. Environmental Protection Agency. Risk assessment 
guidelines for Superfund. Washington, DC: U.S. Environmental 
Protection Agency, Office of Emergency and Remedial Response. Human 
Health Evaluation Manual, Vol 1, Part A, Interim Final, 1989. 
Publication No. EPA/540/1-89/002.
    9. Dourson ML, Felter SP, Robinson D. Evolution of science-based 
uncertainty factors in non-cancer risk assessment. Regul Toxicol 
Pharmacol 1996;24:108-20.
    10. Mioduszewski RJ, Reutter SA, Miller LL, Olajos EJ, Thomson 
SA. Evaluation of airborne exposure limits for G-agents: 
occupational and general population exposure criteria. Aberdeen 
Proving Ground, MD: Edgewood Research, Development, and Engineering 
Center, U.S. Army Chemical and Biological Defense Command, April 
1998; Publication No. ERDEC-TR-489.
    11. Reutter SA, Mioduszewski RJ, Thomson SA. Evaluation of 
airborne exposure limits for VX: worker and general population 
exposure criteria. Aberdeen Proving Ground, MD: Edgewood Research, 
Development, and Engineering Center, U.S. Army Chemical and 
Biological Defense Command, February 2000; Publication No. ECBC-TR-
074.
    12. McKee WHE, Woolcott R. Report on exposures of unprotected 
men and rabbits to low concentrations of nerve gas vapour. PRP-143. 
Porton Down, 1949.
    13. Weimer JT, McNamara BP, Owens EJ, Cooper JG, van der Wal A. 
Proposed revision of limits for human exposure to GB vapor in 
nonmilitary operations based on one-year exposures of laboratory 
animals to low airborne concentrations. Aberdeen Proving Ground, MD: 
Edgewood Arsenal Technical Report, Project 728012.21, February 
1972--July 1979. U.S. Army 1979.
    14. Harvey JC. Clinical observations on volunteers exposed to 
concentrations of GB. MLRR 114. 1952.
    15. Johns RJ. Effect of low concentrations of GB on the human 
eye. MLRR 100. 1952.
    16. Bramwell ECB, Ladell WSS, Shepard RJ. Human exposure to VX 
vapour, PTP 830. Ministry of Defense, Chemical and Biological 
Defense Establishment, Salisbury, Wiltshire. January 1963. Report 
AD335612L.
    17. Crook JW, Hott P, Owens EJ, Cummings EG, Farrand RL, Cooper 
AE. The effects of subacute exposures of the mouse, guinea pig, and 
rabbit to low-level VX Concentrations. ARCSL-TR-82038. Aberdeen 
Proving Ground, MD: Chemical Systems Laboratory, June 1983. Report 
AD29615.
    18. Callaway S, Dirnhuber P. Estimation of the concentrations of 
nerve agent vapour required to produce measured degrees of miosis in 
rabbit and human eyes. Chemical Defense Establishment, Porton Down, 
Salisbury, Wiltshire. July 1971. Technical Paper No. 65.
    19. McNamara BP, Leitnaker FC, Vocci FJ. Proposed limits for 
human exposure to VX vapor in nonmilitary operations. Edgewood 
Arsenal, Aberdeen Proving Ground, MD: Headquarters, Department of 
the Army, July 1973; Edgewood Arsenal Special Publication EASP 1100-
1 (R-1).
    20. 66 FR 85, May 2, 2001, pp. 21940-21964.

[FR Doc. 02-396 Filed 1-7-02; 8:45 am]
BILLING CODE 4163-18-P