You are on page 1of 8

Practice report  Particulate matter

PRACtice RePORT

Generation of particulate matter during handling


of needle and syringe packaging
Eric S. Kastango, James T. Wagner, Kari B. Kastango,
Nicholas E. Kastango, and Terry J. Wagner

R
ecently, Trissel et al.1 demon-
strated that the use of proper Purpose. Generation of airborne par- generated visible particles and raised air-
aseptic work practices by phar- ticulate matter during the handling and borne particle counts. Peeling open plastic
opening of various syringe and needle film packages and opening hard plastic
macists and technicians can have a
packages in a laminar-airflow workbench packages generated fewer airborne par-
favorable effect on the rate of con- (LAFW) and in a biological safety cabinet ticulates than did pushing devices through
tamination of compounded sterile (BSC) was measured to compare the effects the packaging. For all methods of package
preparations (CSPs). These practices on air cleanliness conditions (International opening, average counts downstream from
include wiping down components Organization for Standardization [ISO] class the direct compounding area exceeded ISO
before placing them in the primary 5 within the LAFW or BSC and ISO class 7 in class 5 conditions. Counts in the LAFW buf-
engineering control (i.e., laminar- buffer areas). fer area did not exceed ISO class 7.
Methods. Twenty-five to 50 packages Conclusion. All methods of separating
airflow workbench [LAFW] or bio-
of each of 12 needle or syringe products and opening the packaging of needles and
logical safety cabinet [BSC]) and were opened. Probes were configured to syringes generated particles. The peel-and-
routinely disinfecting gloved hands count airborne particles during the separa- present technique generated the lowest
with wipes or pads wetted with 70% tion of strip packages and the opening of particulate volume. The LAFW and BSC
isopropyl alcohol. packages by peeling back the top web or were equally effective in maintaining low
Almost 30% of respondents to a pushing the device through the packaging particle counts.
recent survey of the impact of United (for soft packages) or by twisting apart hard
packages. Index terms: Air; Contamination; Control,
States Pharmacopeia (USP) chapter
Results. The numbers of particles were not quality; Equipment; Needles; Packaging;
797 reported that they do not wipe significantly different between the LAFW Syringes
down components and equipment and BSC. The separation of strip packages Am J Health-Syst Pharm. 2008; 65:1443-50
prior to placement in the buffer area,
even though chapter 797 mandates
this practice.2 Chapter 797 contains
critically important minimum prac-
tice and quality standards intended variability in the intended strength cal contaminants, or (5) ingredients
to prevent patient harm, including of correct ingredients that exceeds ei- of inappropriate quality in CSPs.3
death, as a result of (1) microbial ther monograph limits for official ar- We conducted a study to examine
contamination (nonsterility), (2) ticles or 10% for nonofficial articles, and reinforce the aseptic compound-
excessive bacterial endotoxins, (3) (4) unintended chemical and physi- ing principles underlying chapter

Eric S. Kastango, M.B.A., B.S.Pharm., FASHP, is President, Clinical 184 Columbia Turnpike, Suite 282, Florham Park, NJ 07932-1314
IQ, LLC, Florham Park, NJ. James T. Wagner is President, Controlled (eric.kastango@clinicaliq.com).
Environment Consulting, Hellertown, PA. Kari B. Kastango, M.S., Supported by a grant from BD Medical—Medical Surgical Sys-
Ph.D., is Senior Biostatistician, Averion International Corp., South- tems, Franklin Lakes, NJ.
borough, MA. Nicholas E. Kastango is pursuing a B.S. degree in
integrated business and engineering, Lehigh University, Bethlehem, Copyright © 2008, American Society of Health-System Pharma-
PA. Terry J. Wagner is pursuing a B.S. degree in food sciences, Penn cists, Inc. All rights reserved. 1079-2082/08/0801-1443$06.00.
State University, College Station, PA. DOI 10.2146/ajhp070444
Address correspondence to Eric S. Kastango at Clinical IQ, LLC,

Am J Health-Syst Pharm—Vol 65 Aug 1, 2008 1443


Practice Report  Particulate matter

797. Our study had three objectives: tion are currently exempt from these tional airflow, sweeping air from the
(1) to statistically determine if the placement requirements. HEPA filter over the work area to the
number of particles generated dur- Buffer areas and ante areas are front and rear return grilles.
ing the opening of packages is dif- clean, but they are not sterile or The LAFW and BSC were cleaned
ferent between two types of primary 100% particle-free environments. with 70% isopropyl alcohol before
engineering controls (LAFWs and When properly designed, engineered, testing. During the testing, all person-
BSCs), (2) to determine if the open- built, and maintained, these areas nel present in the buffer area were
ing of syringe and needle packaging will maintain particle counts that do gowned in accordance with USP
overcomes the ability of the primary not exceed 10,000 ppcf (ISO class 7) chapter 797 requirements: hairnet,
engineering control to maintain or 100,000 ppcf (ISO class 8) of air beard cover if applicable, facemask,
International Organization for Stan- under dynamic conditions. LAFWs, gown, and shoe covers. The technician
dardization (ISO) class 5 air cleanli- BSCs, and isolators maintain particle opening the packages wore powder-
ness conditions (<3,520 particles counts that do not exceed 100 ppcf free nonsterile gloves that were peri-
per cubic meter [ppcm] or <100 (ISO class 5). odically disinfected with a 70% iso-
particles per cubic foot [ppcf] of 0.5 Needles and syringes come in a propyl alcohol spray throughout the
mm and larger), and (3) to determine variety of sizes and packaging con- testing. All syringes and needles were
the ability of the LAFW’s buffer area figurations. Product packages range removed from their outer cardboard
to maintain ISO class 7 (<352,000 from blister packages (with a paper shipping box and placed into plastic
ppcm or <10,000 ppcf) conditions or plastic top film and plastic bot- containers before entering the buffer
in dynamic operating situations. tom film) to solid plastic cylinders area.
We collected data on airborne par- and plastic trays. The manner in All packaging debris generated
ticulate matter generated during the which packages are introduced to the during the testing was discarded in a
handling and opening of various sy- compounding area and primary en- trash can in the buffer area. All opened
ringe and needle packages in LAFWs gineering control, wiped down, and syringes and needles were classified as
and BSCs. removed from packaging will, along “clean” sharps and were placed into
To date, only one other published with the engineering control design, sharps containers that were disposed
article has examined the genera- determine the process-generated of by an approved waste disposal
tion of particles from supplies and contamination in the DCA (the por- company according to state medical
components used in sterile product tion within the primary engineer- waste regulations.
preparation, and it did not directly ing control where “first air”—the Packaging configuration. Several
address particle counts.4 unobstructed air exiting the high- different syringe and needle packag-
efficiency particulate air [HEPA] ing configurations from three vendors
Background filter in a unidirectional stream— were identified and used in the study
Pharmacists and technicians use interacts with the critical site). (Table 1). Some products, including
approximately 1 billion syringes and many needles and some of the syring-
needles annually in aseptic com- Methods es, came in single packages attached
pounding. These supplies must be All testing was conducted at a tech- to each other as perforated strip pack-
brought into an ISO class 5 primary nical training center.a The training ages. Twelve different products were
engineering control while still inside center has a sterile compounding suite tested in this study.
their primary packaging and removed consisting of an ISO class 8 ante area Method of opening. Each product
from this packaging just before use. and an ISO class 7 buffer area. A 4-ft packaging configuration was opened
Such engineering controls include LAFWb and a 4-ft BSCc were located in by at least two different methods
LAFWs, BSCs, compounding aseptic the buffer area and used for the test- (when feasible) in both LAFW and
isolators (CAIs), and compound- ing. All engineering controls (produc- BSC environments. For the needles
ing aseptic containment isolators ing ISO class 5, 7, and 8 areas) were and syringes that came in single pack-
(CACIs). LAFWs and BSCs are to be tested before the study and certified ages attached to each other with a
located in controlled environments by a qualified techniciand according to perforated top web, the generation of
that include an ISO class 7 buffer area performance standards for each area airborne particulates was evaluated
served by an ISO class 8 (or cleaner) or device5-8 and in-process revisions during separation of the packaging
“ante area.” Certain isolator designs to USP chapter 797.3 The LAFW uses as well as during the opening of indi-
(including most CAIs and CACIs) horizontal unidirectional airflow, vidual packages.
that isolate the direct compounding sweeping air from the HEPA filter Four methods of opening or sepa-
area (DCA) from the buffer area dur- across the work surface to the buffer rating a product’s packaging were
ing material transfer and manipula- area. The BSC uses vertical unidirec- considered:

1444 Am J Health-Syst Pharm—Vol 65 Aug 1, 2008


Practice report  Particulate matter

ticle counter.f Each particle counter


Table 1.
was connected to a laptop com-
Packaged Needles and Syringes Tested
puter via category-5 Ethernet cable
Manufacturer and through an Intel five-port network
Catalog Numbera Package Typeb adapter. The data were collected on
Needle, 18 gauge p a r t i c l e - m e a s u re m e n t - s y s t e m
  BD 305196 Soft pack, strip software. g Each particle counter
  Tyco 1188818112 Soft pack, strip, all film was tested, calibrated, and certified
  Tyco 8881250016 Hard pack, singles before the test; each one sampled at
  Terumo 3NN-1838R Soft pack, strip 1 ft3/min with a minimum sensitiv-
Syringe, 1 mL ity of 0.3-mm particles. All particle
  BD 309628 Soft pack, strip
counts reported are for particles
  Tyco 1180100777 Soft pack, strip
of 0.5 mm or larger, and the testing
Syringe, 10 mL
  BD 309604 Soft pack, singles
was performed under dynamic op-
  Terumo 3SS-10L Soft pack, singles, all film erating conditions, unless otherwise
Syringe, 12 mL stated. The 0.5- m m particle size
  Tyco 1181200777 Soft pack, strip threshold was chosen on the basis of
  Tyco 8881512878 Hard pack, singles USP chapter 797 and current good
Syringe, 60 mL manufacturing practices as defined
  BD 309653 Soft pack, singles, all film by the Food and Drug Administra-
  Tyco 1186000077 Soft pack, strip, all film tion. The size range used is generally
a
Manufacturers were BD, Franklin Lakes, NJ; Tyco (also known as Kendall or Sherwood), Mansfield, MA; and considered the minimum size of vi-
Terumo, Somerset, NJ.
b
Soft pack = needle or syringe encased between paper and a soft plastic film or between two pieces of plastic
able particles—those that contain
film; strip = soft packs of five individually encased needles or syringes joined side by side and separated from living microorganisms—that travel
each other by tearing a perforation; singles = needle or syringe packages available separately (not joined in a
strip); all film = soft pack in which the needle or syringe is packaged between two pieces of plastic film; hard pack
individually or are the carriers for
= needle or syringe packaged in rigid plastic cylinder. smaller viable particles. Before the
testing, all surfaces (floors, walls,
and ceiling) of the ante area and
buffer area were mop cleaned and
1. Peel-and-present: This is the manu- are then opened with either the peel- disinfected with a germicidal de-
facturers’ recommended method and-present or pop-through method. tergent followed by a 1% sodium
of opening packages. The operator hypochlorite rinse.
opens the device packaging by peel- For each method of opening, 25 Statistical methods. Outcome
ing the top web from the bottom web, individual packages of each product variable. The outcome variable was
attempting to not tear any part of the configuration were opened. The the total number of particles released
packaging. burst method was used to separate 10 from a given product packaging con-
2. Pop-through: This method is not a strips of each product that came in figuration opened with a given meth-
recommended technique and is not this configuration. More than 1500 od. This was calculated by adding the
supported by manufacturers; never- individual packages were opened counts for the sampling period from
theless, it is frequently used. The op- during the two-day study. probe #1 (immediately downstream
erator opens the device packaging by Particle-count probes. A trained of the DCA) when the LAFW envi-
pushing the device out of the packag- technician performed all package ronment was used and the counts
ing through the paper side of the web, manipulations at a fixed location (the from probe #3 (immediately down-
causing a tear in the paper. DCA) within the ISO class 5 primary stream of the DCA) when the BSC
3. Twist-off: This is the manufacturers’ engineering controls (the LAFW and environment was used. For example,
recommended method of opening BSC). Four or five particle-count in the BSC it took 2:11 minutes to
certain rigid packages. The operator probes were placed at locations with- open 25 packages of the 18-gauge BD
opens the device by twisting the base in and around the LAFW and within needle by the pop-through method.
cap off the rigid plastic cylinder that the BSC environments (Figures 1 and Probe #3 counted 987 particles, 2026
contains the device. 2). For a given engineering control, particles, and 1 particle at times
4. Burst: The operator takes up strips the DCA remained in a fixed, con- 1:00, 2:00, and 2:11 minutes, respec-
of product packaging (five units per stant location for all samples tested. tively. Therefore, the total number
strip) and quickly separates them into Instrumentation. Each probe of particles for this product and this
individually packaged devices, which was attached to an individual par- method of opening in the BSC envi-

Am J Health-Syst Pharm—Vol 65 Aug 1, 2008 1445


Practice Report  Particulate matter

Figure 1. Placement of particle-counting probes within the laminar airflow workbench. Probe #1 was placed approximately 7 inches
directly downstream of the direct compounding area (DCA), with downstream flow determined by a visual smoke test employing venti-
lation smoke tubes.e This position served as a positive control point, where particles were detected during the smoke test. Probe #2 was
placed approximately 7 inches from the face of the high-efficiency particulate air (HEPA) filter in the direct line of airflow, as predeter-
mined by a smoke test, between the filter and the DCA. This position served as a negative control point with a particle count of 5 particles
per cubic foot or less during the smoke test. Probes #3 and #4 were placed approximately 12 inches to the left and right, respectively, of
the DCA and were as close as possible to the left and right of the operator’s hands without particles being detected during a smoke test.
Probe #5 was in the buffer area within 4 feet behind and above the operator, where the particle detector captured the accumulation of
particles within the area during the smoke test. Arrows indicate direction of airflow.

HEPA Filter

Probe location #3
Probe
location #2
A
DC
Probe location #1
Probe location #4

Probe location #5

ronment was 3014 (the sum of the distributions (test for overdisper- equal to that of the operational BSC,
three counts). sion). There was evidence of overdis- and the alternative hypothesis was
Sample size. The sample chosen persion (the true variance is bigger that the environmental impact of the
contained 12 different product pack- than the mean).9,10 Negative binomial operational LAFW was not equal to
ing configurations and 25 individual regression was used to test the study’s that of the operational BSC. Negative
packages of each configuration as hypothesis. binomial regression was used to test
a matter of convenience. No power Hypothesis. The purpose of the the null hypothesis.
calculations were done to determine study was to determine whether
sample size, since the analysis was there was a difference between the Results and discussion
exploratory in nature. operational LAFW and BSC in total The particles detected during the
Distributional assumption of out- number of particles released into the four different syringe-opening exer-
come variable. The assumption that environment from a given product. cises fell in two distinct categories:
the total number of particles (0.5 In mathematical terms, the loga- small, nonvisible airborne particles
mm and larger) released from a given rithm of the mean particle count was detected only with the particle coun-
product was a Poisson-distributed modeled as a linear function of the ters and larger particles visually de-
random variable was checked by environmental setting. The null hy- tected on the work surface. The larger
using a likelihood ratio test based pothesis was that the environmental particles were not analyzed as part of
on Poisson and negative binomial impact of the operational LAFW is this study.

1446 Am J Health-Syst Pharm—Vol 65 Aug 1, 2008


Practice report  Particulate matter

Figure 2. Placement of particle-counting probes within the biological safety cabinet (BSC) as viewed from the side (left panel) and front
(right panel) of the BSC. Probe #1 was placed directly under the direct compounding area (DCA) and as close to the work surface of the
BSC as the height of the probe and sample tubing would allow. Probe #2 was placed 6 inches below the high-efficiency particulate air
(HEPA) filter diffuser and directly above the DCA. This position served as a negative control point with a particle count of 5 particles per
cubic foot or less during the smoke test. Probe #3 was placed in the direct route of airflow, as determined by the smoke test, between
the DCA and the BSC’s intake grille, approximately 1 inch from the grille. Probes #4 and #5 were placed as close as possible to the left and
right, respectively, of the operator’s hands without particles being detected during a smoke test and as close to the work surface of the
BSC as the height of the probe and sample tubing would allow. Arrows indicate direction of airflow.

Side View Front View

HEPA Filter HEPA Filter

Probe location #2 Probe location #2

Smoke split

Sash Height Sash Height


A
DC
A
DC

Probe Probe Probe


Probe location #1 location #4 location #1 location #5

Probe location #3

There was no statistically sig- Some of the products tested had the paper-based packages averaged
nificant difference between the LAFW a top and bottom web plastic film 1934 ppcf when opened by the same
and BSC in the numbers of particles package (e.g., BD 60-mL syringe and method.
generated during package opening. Terumo 10-mL syringe). Since these The hard-pack sterile barrier
Therefore, reporting here focuses on packages had no paper component, packages of the Tyco 18-gauge
the following: (1) characterizing the there did not lend themselves to the needle and Tyco 12-mL syringe can
particles generated in the LAFW, (2) pop-through method. When the be opened in only one way (twist-
describing the numbers of down- syringe or needle is pushed through, off), and that method generated a
stream particles detected in the LAFW these packages tend to either not very low concentration of airborne
by probe #1 and in its buffer area by tear at all or tear with extreme dif- particles for both products. The Tyco
probe #5, and (3) stating whether or ficulty. These all-film packages 18-gauge needle hard pack averaged
not the opening of syringe and needle were opened only with the peel- 42 ppcf, and the Tyco 12-mL syringe
packaging overcame the ability of the and-present method. Their opening hard pack averaged 181 ppcf down-
LAFW and the BSC to maintain ISO generated very low concentrations stream of the DCA. These packages,
class 5 or the buffer area to maintain of particles downstream of the DCA. however, did generate a small volume
ISO class 7 air cleanliness under dy- The BD 60-mL syringe averaged of visible plastic “nibs” that were de-
namic operating conditions. Mean 174 ppcf and the Terumo 10-mL tected on the work surface after the
particle counts appear in Table 2. syringe averaged 260 ppcf, whereas cap was twisted off.

Am J Health-Syst Pharm—Vol 65 Aug 1, 2008 1447


Practice Report  Particulate matter

Table 2.
Mean Particle Counts for Products Opened by All Methods (Particles/m3)a

Method and Probe Location


Product Downstream b
Upstream c
Leftd Righte Bufferf
Peel-and-presentg
  Needle, 18 gauge
   BD 305196 471 0 1 0 191
   Terumo 3NN-1838R 938 0 0 3 833
   Tyco 1188818112 520 0 0 0 359
  Syringe, 1 mL
   BD 309628 269 0 0 3 649
   Tyco 1180100777 6,230 0 0 0 609
  Syringe, 10 or 12 mL
   BD 309604 582 0 0 0 488
   Terumo 3SS-10L 260 0 0 1 218
   Tyco 1181200777 5,713 2 1 7 255
  Syringe, 60 mL
   BD 309653 174 0 0 10 223
   Tyco 1186000077 4,182 0 17 14 282
Pop-throughh
  Needle, 18 gauge
   BD 305196 2,472 0 0 0 281
   Terumo 3NN-1838R 17,937 0 0 0 545
  Syringe, 1 mL
   BD 309628 2,397 0 0 0 403
   Tyco 1180100777 34,243 0 0 0 549
  Syringe, 10 or 12 mL
   BD 309604 4,512 0 0 0 420
   Tyco 1181200777 17,222 0 0 4 418
Twist-offi
  Needle, 18 gauge
   Tyco 8881250016 42 0 0 0 466
  Syringe, 12 mL
   Tyco 8881512878 181 0 0 0 749
Burstj
  Needle, 18 gauge
   BD 305196 1,304 0 0 0 214
   Terumo 3NN-1838R 3,761 0 0 8 666
   Tyco 1188818112 1,803 0 0 0 447
  Syringe, 1 mL
   BD 309628 1,494 1 4 60 447
   Tyco 1180100777 1,925 4 0 0 525
  Syringe, 12 mL
   Tyco 1181200777 4,659 28 4 128 293
  Syringe, 60 mL
   Tyco 1186000077 4,196 37 91 53 567
a
Unless otherwise noted, results are combined for the laminar airflow workbench (LAFW) and biological safety cabinet (BSC).
b
Probe #1 in the LAFW and #3 in the BSC.
c
Probe #2 in the LAFW and BSC.
d
Probe #3 in the LAFW and #4 in the BSC.
e
Probe #4 in the LAFW and #5 in the BSC.
f
Probe #5 outside the LAFW. There was no corresponding probe outside the BSC.
g
Used for needles and syringes in soft packs.
h
Used for needles and syringes in soft packs that were not all film.
i
Used for needles and syringes in hard packs.
j
Used for needles and syringes in soft pack strips.

1448 Am J Health-Syst Pharm—Vol 65 Aug 1, 2008


Practice report  Particulate matter

Upstream versus downstream syringe) with a right probe count the primary engineering control, was
particle counts. Sterile compound- of 128 ppcf. After strip-packaged 81 air changes per hour (ACPH),
ing should be performed in a uni- products were separated, opening which exceeds the minimum ACPH
directional device, using the first-air with the peel-and-present and pop- specified in USP chapter 797. Chap-
concept to facilitate aseptic tech- through methods resulted in particle ter 797 requires at least 30 ACPH
nique. First air is virtually free of counts of 0 ppcf at the upstream, left, (with no less than 15 ACPH from the
particulate contaminants. All aseptic and right probes, except for the Tyco room if using HEPA-filtered air from
manipulations must be carried out 60-mL and 10- or 12-mL syringe the primary engineering controls in
in the unobstructed first-air zone. packages. the air change calculation) for the
As such, the counts upstream of the Given the data and the large visi- ISO class 7 buffer area.
DCA (first-air) should not exceed ble particles, strip packages should be Implications. Primary engineer-
100 ppcf at any time and should not separated and wiped down outside ing controls used in sterile com-
be affected by any process, including the buffer area to prevent the release pounding typically employ unidirec-
package opening within the DCA. of large particles within the primary tional airflow designed to sweep the
For all samples opened with the engineering control, buffer area, and DCA with particulate-free, HEPA-
peel-and-present, pop-through, and storage bins. filtered air (first air); this eliminates
twist-off methods, the number of Pop-through versus peel-and- outside contamination and removes
counted particles measured up- present Method. For the peel-and- process-generated contamination. A
stream in the LAFW and BSC aver- present method, which is recom- comparison of the particle counts re-
aged 0 ppcf; for the burst method, mended by manufacturers, the corded at the various probe locations
the upstream count averaged only average downstream particle counts demonstrated the effectiveness of
10 ppcf. Average downstream counts were much lower than for the pop- the unidirectional airflow in particu-
for all methods exceeded ISO class 5 through method. For example, the late removal. Regardless of particle
conditions and were as high as 13,131 particle count for the 18-gauge counts downstream of the DCA, the
ppcf for the pop-through method. Terumo needle was only 938 ppcf particle counts upstream of the DCA
Package separation by burst for peel-and-present but was 17,937 were within ISO class 5 air cleanliness
method. During and after the burst- ppcf for pop-through. standards. Performing all package
ing of 50 packages for all of the The pop-through method of manipulation downstream of the
paper-backed syringe or needle strip handling the device packaging gener- DCA and keeping a clear zone 12
packs, large white paper particles ates a very high level of particulates inches on both sides will ensure that
were visually observed and were re- within the ISO class 5 engineering this required level of air cleanliness is
corded with a digital camera. A dark control and should be avoided. In- maintained. Packages should never
plastic sheet was placed on the work stead, the peel-and-present method be manipulated directly in front of or
surface of the engineering control to should be used. over opened vials, ampuls, or other
facilitate visual inspection. The num- Impact on the ISO class 7 buffer sterile compounding supplies.
ber and size of these particles were area. The impact on the buffer area Before any supplies are placed in
not quantified, but their presence is particle counts of opening pack- the primary engineering control, all
undesirable and should be avoided. ages in the LAFW was determined by items must be wiped down with a
In addition to these large par- comparing the background counts 70% isopropyl alcohol-wetted wipe
ticles that were observed visually on with the levels measured by probe #5 and the supplies’ outer packaging
the work surface, the airborne par- during testing. Background counts in must be removed. The separation
ticulate levels recorded at the left and the buffer area in the vicinity of the of packages (burst method) creates
right probes during the burst method testing ranged from 411 to 608 ppcf a significant burden of large par-
were higher than those during the before testing began. The average ticulates that is not easily controlled
peel-and-present and pop-through particle count in the buffer area dur- by the unidirectional airflow. An
methods. The burst method particle ing the package opening procedures efficient way to control the particu-
counts for the left and right probes was 482 ppcf. The highest count at late burden from this process is to
averaged 14 and 36 ppcf, respectively, any time in the buffer area during separate packages outside the aseptic
while peel-and-present averaged only the package opening procedures was compounding environment before
2 and 4 ppcf, respectively. 1906 ppcf. At no time did the counts wiping down the separated packages
In addition, particle counts during in the buffer area exceed ISO class 7 and transferring them into the buffer
the burst method exceeded ISO class conditions. area.
5 (100 ppcf) air cleanliness for one The buffer area air exchange rate, Both horizontal and vertical uni-
of the packages (Tyco 10- or 12-mL including the HEPA-filtered air from directional airflow in the devices

Am J Health-Syst Pharm—Vol 65 Aug 1, 2008 1449


Practice Report  Particulate matter

used for this study effectively con- e


Ventilation smoke tubes, Drager Safety AG is revised and finalized. www.USP.org
& Co., Germany. The filling layer of the airflow (accessed 2007 Dec 3).
trolled the particulate burden cre- tube is impregnated with fuming sulfuric acid. 4. Stubbs S. How to minimize contamina-
ated by opening the packages. When air is pumped into the tube by means of tion when transferring items into hos-
Well-ventilated buffer areas with the rubber bulb, sulfuric acid aerosol emerges pital cleanrooms. Control Environments.
in the form of white smoke. 2006; Jun:11-4.
HEPA-filtered air do not appear to be f
Lasair-II 310-A particle counters (serial 5. ISO 14644-1:1999 cleanrooms and as-
affected by particulate release during nos. 49348, 49373, 49375, 49376, and 39242), sociated controlled environments—part
the opening of syringe and needle Particle Measuring Systems, Boulder, CO. 1: classification of air cleanliness. Geneva,
Each particle counter was certified by the Switzerland: International Organization
packages in the LAFW. manufacturer to meet or exceed all published for Standardization; 1999.
specifications and was calibrated within the 6. IEST-RP-CC002.2. Unidirectional flow
Conclusion previous six months using equipment and clean-air devices. Rolling Meadows, IL:
standards of accuracy specified by the Nation- Institute of Environmental Sciences and
All methods of separating and al Institute of Standards and Technology. Technology; 1999.
opening the packaging of needles g
Pharm Net version 3.0, Particle Measuring 7. NSF/ANSI 49-2004. Class II (laminar
and syringes generated particles. The Systems, Inc., Boulder, CO. flow) biosafety cabinetry. Ann Arbor, MI:
NSF International; 2004.
peel-and-present technique gener- References 8. Controlled Environment Testing As-
ated the lowest particulate volume. 1. Trissel LA, Gentempo JA, Saenz LM et sociation. CETA certification guide
The LAFW and BSC were equally al. Effect of two work practice changes for sterile compounding facilities
on the microbial contamination rates of CAG-003-2006, June 7, 2006. www.
effective in maintaining low particle pharmacy-compounded sterile prepara- cetainternational.org/reference/
counts. tions. Am J Health-Syst Pharm. 2007; C E TA A s e p t i c C o m p o u n d i n g
64:837-41. CertificationGuide.pdf (accessed 2007
a
Micro-Clean, Inc. Bethlehem, PA. 2. Candy TA, Schneider JA, Pedersen CA. Apr 2).
b
Horizontal LAFW (model NU-201-430), Impact of United States Pharmaco- 9. Cameron AC, Trivedi PK. Regression
serial no. 14854SV, NuAire, Plymouth, peia chapter 797: results of a national analysis of count data. Cambridge,
MN. survey. Am J Health-Syst Pharm. 2006; United Kingdom: Cambridge Univ. Press;
c
Forma class II, type A1 BSC (model 1200), 63:1336-43. 1998.
serial no. MCI04605, Thermo Scientific, 3. United States Pharmacopeial Convention, 10. Stokes ME, Davis CS, Koch GG. Categori-
Waltham, MA. Inc. General chapter 797 pharmaceutical cal data analysis using the SAS system.
d
An employee of Micro-Clean, Inc. compounding—sterile preparations 2nd ed. Cary, NC: SAS Institute; 2000.

1450 Am J Health-Syst Pharm—Vol 65 Aug 1, 2008