March 7, 2002
Mr. Sam Pulcrano
Manager – Safety Performance Management
United States Postal Service
475 L’Enfant Plaza, SW Room 9912
Washington, D.C. 20260–4231
This interim letter reports the currently available results of a health hazard evaluation (HHE) conducted by investigators from the National Institute for Occupational Safety and Health (NIOSH) at the Trenton Processing and Distribution Center (TPDC) located in Trenton, New Jersey. This HHE was conducted at the request of the United States Postal Service for collaborative research to evaluate air sampling methods for Bacillus anthracis spores.
NIOSH investigators visited the TPDC on February 4–8, 2002. The visit began with an opening conference and proceeded to environmental sampling for the remainder of the week. Environmental sampling consisted of area air samples and surface wipe samples for B. anthracis, collected around a delivery bar code sorter (DBCS #70) before and after the machine was operated. This machine had undergone initial spot cleaning, and was enclosed in a polyethylene structure.
INITIAL MACHINE CLEANING
The IT Group, Incorporated (IT) cleaned DBCS #70 following closure of the TPDC in October 2001, using the following four techniques:
# Initial spot cleaning using a high–efficiency particulate air (HEPA) vacuum cleaner
# Application of a 10% bleach solution with a contact time of 10 minutes
# Application of a neutralizing solution (sodium thiosulfate)
# Final rinse with clean water
Following cleaning, the DBCS machine was covered in plastic. In January 2002 a plastic–walled enclosure (approximately 70' x 15' x 7') was constructed around DBCS #70. The plastic covering over the machine was removed (the cover was cleaned prior to its removal).
Surface wipe and air samples were collected before, during, and after DBCS #70 was operated. A “test deck” of sample mail was run through the machine for 45 minutes. The DBCS #70 computer program indicated that 6,521 “test deck” sheets had run through it. Six area air sampling locations were established around DBCS #70 at a 2–3' height. Sample preparation and processing was conducted immediately outside of the DBCS enclosure, but within the controlled access area of the TPDC.
Composite wipe samples were collected with sterile polyester/rayon pads moistened with sterile water according to Centers for Disease Control and Prevention (CDC) guidelines.[i] Wipe samples were collected on the Jogger, Feeder, and Reader sections of the DBCS. In the final stacker (bin) area, each composite wipe sample consisted of a column of four bins.
Filter Cassette Samples
Air samples for B. anthracis were collected using:
37–millimeter (mm) diameter, 0.8–micrometer (Fm)
pore size mixed cellulose ester (MCE) membrane filters at a calibrated
sampling rate of 2.0 liters per minute (lpm);
37–mm diameter, 0.8 Fm pore size polytetrafloroethylene (PTFE) membrane
filters at a calibrated sampling rate of 2.0 lpm;
37–mm diameter, 3.0 Fm nominal pore size gelatin–coated (GEL) filters
at a calibrated sampling rate of 2.0 lpm;
4.8–centimeter diameter, 1.0 Fm
pore size polyester felt filter disk (PFFD) at a sampling rate of
approximately 400 lpm. Samples were collected using a single filter, dry filter unit
samples for B. anthracis were
collected using Tryptic Soy Agar (TSA) with 5% sheep blood in Anderson
single–stage cascade impactors at an approximate sampling rate of 28 lpm.
Samples were collected using aseptic techniques (e.g., the cleaning and
sanitization of working surfaces, the wiping of Andersen sampler collection
surfaces with alcohol between samples, and the inversion of sample plates
before and after sampling).
wipe sampling results were reviewed to determine if the machine was
contaminated before and after DBCS #70 was turned on. Air sampling results were compared between the various
instruments to determine if there was one that more appropriately
characterized the airborne environment for B.
1 presents the locations where MCE, PTFE, GEL, and agar plates were collected
as well as the three DFU locations.
wipe samples were collected around the Jogger, Feeder, and Reader areas and
the remaining 48 samples were collected in the stacker (bin) area.
duration was approximately eight hours each day. For MCE and PTFE filter samples, one sample was collected for
a 2–hour session and one sample for an 8–hour session. GEL filter samples were collected for 1– and 2–hour
sessions. Andersen samples were
collected approximately every ten minutes, in succession, during the first two
hours. One 10–minute sample was
then collected every 2–hours at each sample location over the remaining
6–hour period. DFU filters were
collected for either 2– or 8– hour sessions near sample locations 4, 5,
and 6. One of the DFU instruments
was a modified unit provided by URS.
PROTOCOL FOR LABORATORY ANALYSIS OF SAMPLES
samples were analyzed by a NIOSH contract laboratory (Microbiological Services
Incorporated, Houston, Texas) according to the CDC document entitled, “Basic
Laboratory Protocols for the Presumptive Identification of Bacillus anthracis”.[ii]
The confirmation of presumptive positive samples was accomplished by
Gamma Phage Lysis (GPL) and Direct Fluorescence Antibody (DFA) analyses.
results for the various air samples before DBCS #70 was in operation are
presented in Table 1; results after DBCS #70 was in operation are presented in
Table 2. In Table 2, the air
samples were started at the same time the machine was started and ran for
their allotted time periods. Samples
are referred to as positive or negative for B.
anthracis. Positive samples
were confirmed by GPL and DFA. Table
3 presents the “test deck” information from the DBCS machine operation
total of 106 wipe samples were collected on the DBCS machine (53 samples were
collected before and 53 after the machine was in operation).
All stackers (bins) were included in the wipe sampling.
All of the wipe samples were positive.
MCE air samples were collected before the machine was in operation; these were
all negative. Twelve MCE air
samples were collected after the machine was in operation. Stations 4, 5, and 6 each had positive samples during the
2–hour period. Station 5 had a
positive sample for the 8–hour period.
PTFE air samples were collected before the machine was in operation; these
were all negative. Twelve PTFE
air samples were collected after the machine was in operation. Stations 3, 4, and 5 each had positive samples during the
2–hour period. Station 2, 3,
and 4 each had positive samples for the 8–hour period.
GEL air samples were collected before the machine was in operation; these were
all negative. Thirty–six GEL
air samples were collected after the machine was in operation.
Stations 2, 4, 5, and 6 each had one positive sample of the two
1–hour samples collected during the 2–hour period.
Both 1–hour samples at Station 3 were positive.
During the 8–hour period, Stations 1, 2, and 6 each had one positive
sample. During the 8–hour
period, Stations 3 and 4 each had two positive samples.
DFU air samples were collected before the machine was in operation.
All filter samples were negative.
Three DFU air samples were collected after the machine was in
operation. All stations had
positive samples during the 2–hour and 8–hour period.
Sixty–five Andersen samples were collected before the machine was in operation. During the first 2–hour session, stations 1 and 2 each had two positive samples, station 4 had one positive sample, and station 5 had three positive samples. During the 6–hour period, stations 1 and 5 each had one positive sample. Sixty–four agar plate air samples were collected after the machine was in operation. During the first 2–hour session, station 1 had eight positive samples and stations 2–5 each had ten positive samples. During the 6–hour period, station 1 had one positive sample, station 2 had three positive samples, and stations 3–5 had two positive samples.
sheets of the “test deck” that went through the machine for the 45 minute
operation time were sent to the lab for analysis. One of the sheets was positive.
The following discussion is based on the analysis of the results currently available. Although the samples have been fully confirmed positive or negative, further laboratory analysis (quantitative results) and discussion of these results is warranted and will be presented in the final report.
Before DBCS #70 Operating
All wipe samples collected were positive.
As the bin numbers increased, the amount of contamination increased,
with wipe samples collected at approximately bin #100 to the end of the
machine having confluent (heavy) growth.
# The wipe samples seem to indicate that this particular DBCS machine was still highly contaminated with B. anthracis spores.
# Since positive Andersen samples were collected at the beginning, middle, and end of the 8–hour time period at various sample locations, it does not appear that a single event generated spores.
DBCS #70 Operating
# Again, all wipe samples collected were positive and as the bin numbers increased, the amount of contamination increased, with wipe samples collected at approximately bin #100 to the end of the machine having confluent (heavy) growth.
# Almost all of the collected Andersen samples were positive. An initial analysis of the samples seems to indicate a declining number of colonies per sample as the time passed during the 8–hour session. Only one sample was negative during the first two hours of sampling. This seems to indicate that the DBCS machine released a large amount of spores when in operation.
# 33%, 50%, and 36% of the MCE, PTFE, and GEL samples were positive, respectively. The MCE and PTFE positive samples did not seem to be consistently located at any one particular sampling location or time. The GEL filters seemed to be more consistently positive at locations 3 and 4 (between the Jogger, Feeder, and Reader sections of the machine). Also, all of the positive GEL samples were collected in the first four hours of sampling.
All DFU filters were positive during this testing period.
For the 2–hour sampling period, the total air volume sampled by the
DFU equates to approximately ¼ of the enclosure air volume.
All air sampling media used in this evaluation were capable of
collecting B. anthracis spores.
However, the initial analysis of the results seems to indicate that the
Andersen sampler collected B. anthracis
spores more consistently than other sampling media.
The Andersen sampling method appears to be more sensitive than the
other methods since the only positive air samples collected before the machine
was operated, was with these samplers. However,
further analysis of the samples is in progress which will provide quantitative
The positive Andersen samples before the machine was operated suggest
that even walking and light work in the enclosure may be sufficient to
reaerosolize anthrax spores.
intent of this interim letter is to summarize the currently available
analytical results from environmental samples collected during the NIOSH
investigation conducted on February 4–7, 2002.
If you have any questions, please do not hesitate to contact me at
Robert E. McCleery, MSPH
Industrial Hygiene Section
Hazard Evaluations and Technical
Division of Surveillance, Hazard
Evaluations and Field Studies
Gus Baffa, NRLCA
Dick Collins, Mail Handlers
Al Farranto, NALC
Corey Thompson, APW
Kevin McMahon, The IT Group
Don Schill, New Jersey Department of Health and Senior Services
John Frietas, URS