Dynamic airflow visualization studies, more commonly known as smoke pattern testing, have been a regulatory requirement for unidirectional airflow areas in sterile manufacturing for over 30 years. The changes to USP Chapter <797> published in 2008 finally brought that same expectation to unidirectional airflow areas involved in sterile compounding.

Chapter <797> states: 
“The airflow in the PEC shall be unidirectional (laminar flow) and because of the particle collection efficiency of the filter, the first air at the face of the filter is, for the purposes of aseptic compounding, free from airborne particulate contamination. HEPA-filtered air shall be supplied in critical areas (ISO Class 5) at a velocity sufficient to sweep particles away from the compounding area and maintain unidirectional airflow during operations. Proper design and control prevents turbulence and stagnant air in the critical area. In situ air pattern analysis via smoke studies shall be conducted at the critical area to demonstrate unidirectional airflow and sweeping action over and away from the product under dynamic conditions.”¹

This is essentially the same language and the same requirement that has been imposed by the FDA since the 1980s, yet many sterile compounding operations are only now beginning to embrace smoke testing as a useful tool, while others continue to resist the mandate with excuses, such as: The test introduces contamination into my critical site . . . It is a subjective test and, therefore, not appropriate . . . It’s too hard . . . I am not qualified to do the test . . . and It’s too disruptive to my process. However, in the final analysis, these arguments are irrelevant. The FDA has made their expectations clear through the public posting of warning letters and Form 483 observations.² A quick review of the 483 observations posted on the FDA Web site illustrates clearly the importance of performing this test to ensure appropriate unidirectional airflow; well over half of the 483s include either a citation for not performing the test or for performing the test incorrectly. In addition, many state boards of pharmacy are now citing facilities for failing to perform smoke studies.

Test Objectives 
Although the testing procedure is simple, achieving good results can be challenging without an understanding of the test’s objectives and the desired outcome. A smoke pattern test is used to ensure that first air (see SIDEBAR 1 Acronyms and Definitions) reaches the critical site, that the direct compounding area (DCA) is free from turbulence and updrafts, and that the inside of the primary engineering control (PEC) is effectively isolated from the room (see SIDEBAR 2 Objectives of Smoke Pattern Testing). Turbulent airflow or air rolling off IV bars, lips, or other obstructions interferes with first air. Air should exit the HEPA filter/diffuser in a smooth, non-turbulent manner and continue unobstructed to the air return grilles of a biological safety cabinet (BSC) or compounding aseptic isolator (CAI), and, in the case of a laminar airflow workstation (LAFW), to the room after passing over the critical site. 

A smoke pattern test relies on a visual medium (smoke) being generated in a manner that allows observation of airflow patterns. The visual medium must be nearly neutrally buoyant. Water-based fog generators, such as CO₂ generators (dry ice and water), create an effluent that is significantly heavier than air, thus causing fall-out during a test. Updrafts and turbulence may be missed when testing with water-based generation systems. 

Glycol-based, hand-held smoke generators and theatrical fog generators have been used for this type of testing for years with better results than those achieved with CO2-based generators. Even these, however, must be used with caution. Generators that produce too much smoke often mask airflow problems with sheer volume. Most sterile compounding applications benefit from small-volume generators. Air ventilation chemical smoke tubes provide controllable streams of visual medium, are easy to use, and can pinpoint exact locations to be observed without overwhelming the PEC with smoke. 

Sterile compounding personnel must perform airflow smoke pattern testing under dynamic operating conditions. Few, if any, perfect unidirectional airflow devices exist; every engineering control has its own idiosyncrasies. Even devices with common model numbers may demonstrate differences in airflow patterns, which means that airflow and its effects on the pharmacy’s processes must be determined for each PEC individually. Training for all compounding technicians must include training on the specific devices they will be using and outline the different considerations for horizontal airflow (LAFW) and for vertical airflow (BSC, CAI, and integrated vertical laminar flow [VLF]) environments. It is vital to remember that a technician cannot be effective in an environment in which unidirectional airflow is unreliable, and that even a perfectly functioning LAFW is useless if technicians have not been trained to use it properly. 

Conducting smoke pattern testing illustrates that proper training is a priority to the pharmacy. While a written report is important for demonstrating that all areas are appropriate for their intended tasks and that technicians are qualified for all of the areas in which they will work, recording a video of the process can be even more useful, not only as a documentation tool, but also for training purposes. In fact, many regulators, including the FDA and many state boards of pharmacy, now expect to see a brief video of smoke pattern testing conducted in each of the applicable areas by the personnel using the environments. 

Horizontal Airflow LAFWs
On the market for over 40 years, horizontal airflow LAFWs are the most common PEC for sterile compounding, but seldom are used properly. A smoke pattern test is required to establish where aseptic processes should occur and where sterile wrappings should be removed; these activities should not occur in the same location. When performing the smoke pattern test, remember that you are establishing best work practices, so put aside outdated paradigms that are no longer relevant. For example, the old adage that you must work at least six inches in from the outer edge of the work surface makes no sense when one of the objectives is to ensure that particles are removed from the work area and do not re-enter critical sites. Removal of sterile wrappings is often better performed at the outer edge of the worktable or directly in front of an air return, where the wrappings can be removed easily without fear of particle re-entrainment. Additionally, many LAFWs are equipped with a protective lip at the rear of the work surface to prevent spills from soiling the HEPA filter. These lips provide a vertical edge that interrupts the unidirectional horizontal airflow pattern on the work surface. In these devices, airflow on the work surface is better at the outer edges than it is closer to the HEPA filter. 

Air in a horizontal airflow application travels from the entrance plane, which is the HEPA filter positioned at the rear of the work area, to the exit plane, which is the room after the air passes over the operator. In these environments, the smoke must be generated directly downstream (ie, in front) of the HEPA filter/diffuser and followed over every critical site until it exits the LAFW to ensure that it does not re-enter the critical area. To properly operate LAFWs, ensure that de-wrapping occurs to the left or right of the DCA, not in front or behind. Verify that airflow within all critical locations, including de-wrapping and the DCA, is smooth and free from turbulence and updrafts, that all technicians are proficient in every PEC that they may be required to use, and that any discovered imperfections in airflow do not impact first air at the critical site (see PHOTOS 1-3).

Class II Biological Safety Cabinets
Airflow in all Class II BSCs used for sterile hazardous drug compounding is vertical flow. Air flows from the entrance plane, which is the HEPA filter/diffuser positioned above the work surface, to the exit plane, which is the work surface. The air return grilles positioned at the front and rear of the work tray are the most critical elements in establishing optimal positioning of the DCA and de-wrapping zones. Air pulled into these grilles is of higher velocity and is more robust than air pushed out of the HEPA filter. Prior to positioning the DCA, observe the airflow within the entire work area. Notice that the worst airflow in a vertical airflow device is at the center of the work tray. While air is pulled to the front and back, it is somewhat static at the center. As you move away from the center of the work surface, airflow improves. Positioning work toward the front of the work tray allows you to take advantage of the strong suction of the front air return grille (see PHOTO 4). Performance of every BSC is unique based on the return air configuration and airflow velocity for that cabinet. 



When testing these units, the smoke must be generated directly downstream (below) the HEPA filter/diffuser and followed over every critical site until it exits the work area into the return air grilles. In these environments, remember that de-wrapping should occur to the left or right of the DCA, not in front or behind. Verify that airflow within all critical locations, including de-wrapping and the DCA, is smooth and free from turbulence and updrafts, that all technicians are proficient at working in all of the PECs they may be required to use, and that any discovered airflow imperfections do not impact first air at the critical site. Note that air under an IV bar is in a turbulent zone created by the bar and, therefore, is not first air. When an IV bar is used, the bag must be pulled to the front and side to create access to undisturbed unidirectional airflow. In addition, ensure that all technicians understand that blocking the front intake grille will allow air to cross from the room directly into the DCA, potentially contaminating CSPs. Often, the best airflow is toward the front of the work area.

Compounding Aseptic Isolators
Airflow in CAIs is vertical flow. (Note that those isolators designed with HEPA filters and air returns in the same plane above the work surface have turbulent—not unidirectional—airflow and, therefore, are not appropriate for sterile compounding and are not discussed herein.) In the CAI, air flows from the entrance plane, which is the HEPA filter/diffuser positioned above the work surface, to the exit plane, which is the work surface. Air return grilles, positioned at the front and rear of the work tray, are the most critical elements in establishing the optimal positioning of the DCA and de-wrapping zones. Most CAIs employ lower airflow velocities than those in BSCs. Finding robust unidirectional airflow can be challenging. Similar to BSCs, the best airflow often is close to the return air grilles. Prior to positioning the DCA, observe the airflow within the entire work area. Notice that the worst airflow in a vertical airflow device is at the center of the work tray. Air is pulled to the front and back, but is somewhat static at the center. As you move away from the work surface, the airflow improves. Positioning work toward the front of the work tray allows you to take advantage of the strong suction of the front air return grille. Performance of every CAI is unique based on the return air configuration and airflow velocity for that isolator. 

Only a few isolators are equipped with three access ports. The remainder can be difficult to test without the aid of a continuous smoke generator. An inexpensive and effective solution is to attach a ventilation smoke tube to a simple aquarium pump. This allows for the creation of a continuous stream of smoke that can be affixed to a rigid stand, allowing the smoke to be positioned directly above the DCA for dynamic testing. The smoke must be generated directly downstream (below) the HEPA filter/diffuser and followed over every critical site until it exits the work area into the return air grilles. When testing these isolators, consider that de-wrapping should occur to the left or right of the DCA, not in front or behind. As with BSCs, verify that airflow within all critical locations, including de-wrapping and the DCA, is smooth and free from turbulence and updrafts, that all technicians are proficient at working in every PEC that they may be required to use, and that any discovered airflow imperfections do not impact first air at the critical site. Again, note that IV bars create a turbulent zone beneath them; air under the bar is not first air. When an IV bar is used, the bag must be pulled to the front and side to create access to undisturbed unidirectional airflow. Often, the most usable airflow is toward the front of the work area.

Site-Built VLF Areas
Airflow smoke pattern testing is particularly important in custom-made VLF areas, as open architecture design is more complex to build and use effectively than standard PECs (see PP&P’s May 2014 article, Open Architecture Design Considerations in the Cleanroom, at www.pppmag.com/opencleanroomdesign).³  In these areas, air flows from the entrance plane, which is the ceiling-mounted HEPA filter, to the exit plane—the strategically located air return grilles. Note that the FDA has cited many pharmacies for poorly performing site-built VLF areas. When testing these areas, consider that de-wrapping should occur to the left or right of the DCA, not in front or behind. As with other environments, verify that airflow within all critical locations, including de-wrapping and the DCA, is smooth and free from turbulence and updrafts, that all technicians are proficient at working in every PEC that they may be required to use, and that any discovered airflow imperfections do not impact first air at the critical site. Again, note that the air under an IV bar is in a turbulent zone created by the bar and, therefore, is not first air. When an IV bar is used, the bag needs to be pulled to the front and side to create access to undisturbed unidirectional airflow. 

Creating robust unidirectional airflow in a site-built installation can be challenging for even experienced contractors, making it critical that smoke testing be performed in these environments. 

A Valuable Training Tool
Smoke pattern testing is essential for verifying proper airflow performance, but it may be equally important in its role as a training tool to demonstrate effective use of unidirectional airflow. Although it is often associated with the certification process, the test does not need to be performed only by a certifier. The compounding supervisor, training specialist, or anyone responsible for the sterile compounding process can conduct smoke pattern testing. Certifiers are experts on airflow within PECs, but typically they are not experts in sterile compounding processes. Therefore, the best approach often is a team effort.

References

1. USP <797>: Guidebook to Pharmaceutical Compounding—Sterile Preparations. Rockville, MD: The United States Pharmacopeial Convention; February 2008:43. 
2. US Food and Drug Administration. 2014 Pharmacy Inspections and Related Records. http://www.fda.gov/AboutFDA/CentersOffices/OfficeofGlobalRegulatoryOperationsandPolicy/ORA/ORAElectronicReadingRoom/ucm384667.htm. Accessed September 16, 2014.
3. Diorio L, Thomas D. Open architecture design considerations in the cleanroom. Pharm Purch Prod. 2014;11(5):40-42.

James T. Wagner is president of Controlled Environment Consulting in Hellertown, Pennsylvania. Jim has over 30 years of experience designing and evaluating facilities used for aseptic processing and was a member of the 2005-2010 USP Sterile Compounding Committee responsible for Chapter <797>.

SIDEBAR 1
Acronyms and Definitions

• Biological Safety Cabinet (BSC): A ventilated cabinet having an open front with inward airflow for personnel protection, downward HEPA-filtered laminar airflow for product protection, and HEPA-filtered exhausted air for environmental protection
• Compounding Aseptic Isolator (CAI): For the purposes of this article, this term also encompasses Compounding Aseptic Containment Isolators (CACIs)
• Critical Area: An ISO Class 5 environment
• Critical Site: A location that includes any component or fluid pathway surfaces or openings exposed, and at risk of, direct contact with air
• Direct Compounding Area (DCA): The critical zone within the PEC where critical sites are exposed to first air
• First Air: The air exiting the HEPA filter/diffuser that is essentially particle-free
• Primary Engineering Control (PEC): Any unidirectional airflow ISO Class 5 device used for the critical aseptic process
• Unidirectional Airflow: Airflow moving in a single direction in a robust and uniform manner and at sufficient speed to reproducibly sweep particles away from the critical processing area
• Vertical Laminar Flow (VLF): A site-built, integrated vertical laminar flow ISO Class 5 area 

SIDEBAR 2 
Objectives of Smoke Pattern Testing

• Identify and differentiate the DCA from other areas within the PEC
• Demonstrate first air at all critical sites within the DCA
• Prove that the DCA is free from turbulence and updrafts
• Prove that outside, unfiltered air does not enter the PEC 
• Trace unidirectional airflow from the HEPA/diffuser to and over the product and out of the PEC