美国制药评论刊文质疑环境监测与无菌保证的关联并主张重新评估其定位

Introduction

The global healthcare industry supplies parenteral products providing patient treatment. These products require rigorous control of contaminants. The tolerance for contamination is such that total reliance on end-product testing is recognized as statistically inadequate.^1-3 Current Good Manufacturing Practices provide a framework for the preparation of sterile products defining the facility, personnel, material, procedures, and other practices necessary to produce safe parenteral products. A prevalent expectation within CGMP operations is the conduct of environmental monitoring (EM) within the production facility.

The history of EM in healthcare is obscure. There are anecdotal references to the conduct of surface sampling and settle plates in the 1950s.^4,5 It was performed on an as-needed basis in clinical settings, but implementation of hygienic controls was determined to be the more effective means to prevent patient infection.^6,7 However, over time EM gradually took on greater importance in the global healthcare industry and is more intensively practiced in parenteral manufacturing than in other health-related disciplines.^{1, 8-10} Global expenditures for aseptic processing EM underscore the intensity of this activity. Expenditures reached approximately $1B in 2024 with an estimated growth rate of 5.7% annually.¹¹

The purpose of environmental monitoring is to demonstrate an effective state of control relative to the production environment and signal when that situation no longer exists. This approach emphasizes control charts demonstrating normality. However, EM has evolved into a major activity with dedicated departments, specialty software, sophisticated sampling systems, and a regulatory obsession with respect to the results.

However, the correlation between environmental monitoring results and product microbiological quality is extremely difficult to establish.^12-15 This is due in part to the fact that 99% or greater of the EM results reported for critical ISO 5/Grade A environment are contamination free. That being the case, it is reasonable to ask if the massive cost and human effort required to support environmental monitoring at the current level of intensity is warranted.

More precisely, have industry, and most critically patients, benefited from the massive expenditures made on EM? When EM was introduced, there were no defined requirements because it was implemented only as a general assessment of facility hygiene. EM was implemented to provide a qualitative assessment regarding the usefulness of disinfection procedures, ventilation systems, and operator gowning. General recovery targets may have existed, but alert and action levels did not. Additionally, it is worth remembering that the word aseptic literally means without infection rather than sterility.

When process validation was introduced in the late 1970s, EM and media fill testing took on added significance.¹⁰ By the mid-1980s regulators began to consider EM a process control activity which served to amplify its perceived value.^16-18 Alert and action levels were implemented, and industry accepted that without challenge. Although EM practices had never been correlated to sterility, it seemed a logical extension of the assessments already being performed. Initially EM alert and action levels were not considered to be product release criteria.

Beginning in the early 1990s, some regulators suggested that alert and action levels should be considered control points and the data arising from EM should be plotted and trends analyzed.¹⁸ Alert and action levels were gradually interpreted as an integral part of process control and EM gradually became a product release criterion. As it was asserted that EM was a means to demonstrate sterility assurance, it morphed into a secondary sterility test. Our industry accepted this evolution of EM without protest considering it emblematic of the effort that we as a regulated industry must undertake given the criticality of patient safety.

The increased criticality of EM in quality control and assurance was not accompanied by a micro-analytical validation effort intended to assess the actual capabilities of EM. Had EM been a compendial release test it would have been necessary to evaluate, at a minimum, its limit of detection, its sensitivity, and statistical reliability. This was not done, and EM has no standard method, and is executed without positive or negative controls. It requires the participation of a sampler who must execute an aseptic handling step that is not required to produce a parenteral product. It is accepted that this added aseptic intervention contributes no risk to the EM data and none to patient risk.

Since it has never been subject to analytical method validation there is no supportive evidence that EM has the accuracy, precision, reproducibility, or sensitivity to perform any role in sterility assurance determination. The diversity of equipment, methods, and media employed across the industry has never been considered. The microbiological facts regarding statistics of microbial enumeration have been known for over a century, but the scientific realities have been ignored, although many degreed microbiologists have asked relevant questions.

There are many different types of air samplers, with different sample volumes, recovery rates, and aseptic intervention requirements. Surface sampling can be performed by different device configurations from wet or dry swabs to contact plates. Yet somehow, regardless of the multiplicity of practices, systems, lack of statistical data and analytical suitability there is a widespread belief that EM can provide evidence of sterility or the lack thereof on a lot-by-lot basis.

Our Core Premise

The purpose of environmental monitoring is to demonstrate the general hygiene of the production environment. Given its history and the absence of any significant qualification as a method, an objective reconsideration of current EM practices is appropriate. The following list identifies the assumptions, omissions, and errors in its application that lead us to question its current prominence.

The distribution of contaminants (viable and non-viable) suspended in air systems is non-uniform regardless of whether the air supply is turbulent or laminar as contamination is not inherently present in the air supply, rather it is derived from personnel, materials and/or equipment.^12,13
There is no data that correlates airborne and/or surface microbial populations with their introduction into the product.^12-15
The various EM sampling systems for viables and non-viables are not equivalent and cannot be used interchangeably.
Global regulatory standards for viable contamination are invalid as there are no defined/referenced sampling methods.^2,3
A less than 1 CFU/sample limit of detection of EM is impossible because no current media can recover all bacteria and mold species. This alone rules out its use as an adjunct sterility test.^19,20 Zero recovery in analytical microbiology means nothing grew, it does not mean sterile.
There is no EM data that correlates directly to media fill outcomes as measured by contamination rates. In fact, there is documented evidence to the contrary.²¹
That EM methods cannot differentiate between one (1) CFU and three (3) CFU making the distinction between alert and action level analytically and statistically invalid.^19,20
EM results cannot be used to accurately determine the presence (or absence) of contamination in aseptically produced materials because they are not tests of the actual product.¹⁹
Surface samples are subject to the same limitations with different test equipment, sampling technique, and sample surfaces and lack any standardization. A singular expectation is inappropriate.¹⁹
EM results from aseptically gowned personnel cannot be consistently linked to product contamination. We know from many peer-reviewed scientific articles that gowned clean room staff release 1000 cfu/hour or more into their environment.^22,23
EM is not suitable as a means for microbiological classification of classified environments given the diversity of system designs and operating practices as well as its analytical and statistical limitations.²⁴
Microbiology is a science based upon large populations. It cannot differentiate among small populations. The analytical methods do not have a limit of detection of one and the true limit of quantitation is higher (estimated >10).¹⁸
One cell cannot cause an infection, and even pathogenic bacteria are not going to cause serious health issues unless they possess a genetic mutation that makes them resistant to antibiotics.^25-27
There is no such thing as an INDIVIDUAL objectionable organism. Microorganisms do damage via infections in instances where they can proliferate. They do not proliferate on a floor, wall or stainless-steel surface unless conditions are present to provide them with nutrition.²⁸
In microbiological analysis, zero does not mean zero. It only means nothing was recovered because it failed to grow. Thus, any standard that suggests zero is acceptable and that any recovery requires an investigation is inappropriate.^18,19

These constraints are significant enough that a return to the original tenets behind EM usage is in order. EM is an analytically limited, statistically variable general indication of microbial control that does not allow direct correlation of the results obtained and the microbiological safety of a manufactured parenteral product.

None of the core EM precepts have been experimentally confirmed, although some have been assessed indirectly and disproven.²¹ The capabilities of EM should have been comprehensively qualified and validated experimentally before implementation as a regulatory requirement. We have collectively reviewed thousands of EM results for media fill tests and have never been able to establish any kind of correlation between EM findings and media fill test success or failure. We have reviewed media filling results in which multiple alert or even action level EM events occurred while no containers were contaminated. Conversely, we have reviewed media fills in which no contamination was recovered at all by EM and one or more media fill containers in the test were contaminated.

There is no scientific evidence that EM has sufficient precision to reliably detect a difference between an alert level of one CFU and action level of two or three CFU. This requirement mandates that extensive excursion reports, deviation reports, or investigative risk analyses are conducted. These results have been known to be statistically inconclusive for more than a century. Microbiology is a logarithmic science. As Scott Sutton pointed out, “Virtually all clean room action levels as set by regulations are deep in the noise range of the plate count method.”¹⁸ Russell Madsen and David Hussong pointed out “a mean count of one CFU at the % standard deviation is 100%; at a count of three CFU it is 58%; at 10 CFU it is 32%.”¹⁹ There are no analytical or statistical means to make a qualitative or quantitative differentiation between any EM result in the order of magnitude between one and ten CFU. Nothing brings EM’s limitations into stark relief better than the fact 99% of the Grade A samples show no growth at all and hence produce only null data.

Many of today’s most capable aseptic processing systems are so effective as to render contamination undetectable. We must question the utility of EM in monitoring these at all. Instead, absence of detection in these superior designs has prompted some to increase the level of EM in efforts to detect what they presume must be present. This approach has two major flaws. First, despite the known limitations of EM in detection of microorganisms, more monitoring will somehow ‘prove’ absence. Second, EM is benign and its execution does not adversely impact the very conditions it is trying to demonstrate. The principle that the safest intervention is the one that is not performed is ignored.²⁹ In fact, what we’ve seen is that even with more intensive microbiological monitoring the rates of recovery are unchanged. Thus, the upshot of more EM sampling is more plates yielding null results, it is not greater insight into product safety.

What is the real take-away from the evaluation of EM presented in this article? EM was never subject to analytical evaluation and fitness for purpose before it was introduced as a regulatory expectation. The expansion in the scope of EM has occurred without rigorous scientific review. EM practices have been stitched together into a patchwork of prescriptive cGMP expectations based on regulatory inspectional opinions and nothing else. We have even seen EM applied to non-aseptic processes without any evaluation of their scientific value or appropriateness. No single assumption in the myriad of CGMP practices which fall under the EM umbrella was ever subject to scientific discussion, analysis, or validation either before or after implementation. Yet, there have been sweeping conclusions made from limited, unvalidated, imprecise, and highly variable data. We find these conclusions unpersuasive. Sadly, some of the defining concepts of EM go against common microbiological as well as general scientific and engineering knowledge. This should, of course, be rectified because, as CDC observed regarding EM in the clinical sphere,“ (Environmental monitoring)… …is an expensive and time-consuming process that is complicated by many variables in protocol, analysis and interpretation.”³⁰ We have learned from the expansive conduct of EM or related microbiological studies done in clean rooms:

It has been confirmed that the only significant source of microbial contamination in aseptic cleanrooms is personnel.
Microorganisms detectable by EM do not enter aseptic environments through HEPA filters that are certified to be integral.
Minor HEPA filter leaks are not detectable by EM, and the pattern of detected contamination is not impacted by minor filter defects.
Sample locations using a grid, as is often done in certifying clean rooms for particulate air quality, are unwarranted in microbiological EM as microbial contamination comes from the personnel inside the clean room, not the ventilation system.
Repeating a sample at a location at which an alert or action level result was observed is of little or no value because thousands of changes of fresh air will have occurred before the result is obtained.

We are on the threshold of a new generation of aseptic processing practices reliant heavily on closed systems, isolators, automation, robotics and advances in process diagnostics in which the underlying weakness in manned environments, the gowned operator, is no longer relevant. Absent the primary contamination source and the EM practices that evolved from it, our industry faces a conundrum. Are we to be constrained to force fit, and potentially adversely impact, advanced technologies designed to eliminate contamination by imposing the same unproven EM practices on them? Respect that many of the newer EM practices being introduced suffer the same limitations as those of the past. The negative absolute conundrum will persist regardless of the sensitivity of the means of measurement.

Conclusion

Considering the advances in aseptic processing and their effectiveness in eliminating the primary sources of contamination (i.e., human activity), it is not our intent to entirely disparage the value of monitoring. Environmental monitoring has some value in trend analysis within an overall contamination control strategy. However, it is not a control mechanism. It was originally implemented as and remains a feedback tool to ensure that the actual controls are effective. Employed strategically and scientifically, monitoring can provide a warning when primary systems drift off track. It is not, and has never been, an indicator of product sterility.

We assert that proposed environmental monitoring standards and regulations incorporating them must be subjected to rigorous and unbiased scientific evaluation. Only if they pass that analysis should they be embraced. Until that is done, let us employ environmental monitoring for its inherent value, as one measure of contamination control, but not the final judge of product quality.

Given the rapid advancement of aseptic processing technology are we going to continue to conduct EM at a high level of intensity when the contamination rate is 1 in 10,000 or 1 in 1,000,000?³¹ We believe that, like the land-line telephone, EM is on its way to extinction. It is time to start the conversation, and that conversation must emphasize science and science alone. This is not the first time that a multi-million-dollar industry has been disrupted, and it will not be the last.

References

USP <1211> Sterility Assurance, USP 41, 2019.
FDA, Guidance for Industry, Sterile Drug Products Produced by Aseptic Processing Current Good Manufacturing Process, 2004.
EMA, Annex 1, Manufacture of Sterile Medicinal Products, 2022.
A. Andon, B., Active Air vs. Passive Air (Settle Plate) Monitoring in Routine Environmental Monitoring Programs, PDA J of PST, Vol. 60, No. 6., pp 350-355, 2006.
Whyte, W. Bailey, P.V. et al, “Evaluation of the routes of bacterial contamination occurring during aseptic pharmaceutical manufacturing, JPST, vol. 36, No. 3, pp 102-107, 1982.
Roberts, C.S., Comments on Darwinism, in Proceedings of the Baylor University Medical Center, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3246855/ 2012.
Van den Ende, M. & Spooner, M., Reduction of dust-borne bacteria in a ward by treating floor and bedclothes, Lancet 241: 400-401, 1941.
Smith, P., Watkins, K. & Hewlett, A., Infection control through the ages; American Journal of Infection Control (40) pp 35-42, 2012.
Sutton, S., The Environmental Monitoring Program in a GMP Environment; Journal of GXP Compliance Vol 14 (3) pp 22-29, 2010.
PDA, TR #22, Process Simulation Testing for Aseptically Filled Products, 1997
https://grandviewresearch.com/industry-analysis/environmental-monitoring-market (accessed July 2026).
Kremer, T. & Patel, R., “Correlation between Environmental Monitoring and Product Bioburden”. Biomedical Instrumentation & Technology, Vol.53 s2, pp 32-37, 2019.
Wang, J., et al, “Spatial ecology meets quality control: a GIS-integrated strategy for visualizing and managing microbial contamination in sterile pharmaceutical cleanrooms”, ASM, Microbiology Spectrum, Vol. 14, No.7,2026.
Salaman-Byron, A., “Limitations of Microbial Environmental Monitoring Methods in Cleanrooms, American Pharmaceutical Review, Vol. 21, 2018.
Marsit, N., Saadawi, S. & Alennabi, K., “Challenges of growth-based microbiological methods in sterility assurance of pharmaceutical product manufacturing”, Discover Pharmaceutical Sciences, 2025.
FDA, Guidance for Industry, Sterile Drug Products Produced by Aseptic Processing Current Good Manufacturing Process, 1987.
PDA TR #13, Fundamentals of an Environmental Monitoring Program, 2001.
Sutton, S., Accuracy of Plate Counts in Microbiology Topics, Journal of Validation Technology pp 42-46, 2011.
Hussong, D., & Madsen, R., Analysis of Environmental Microbiology Data from Cleanroom Samples; Pharm. Tech. Aseptic Processing Issue, pp 10-15, 2004.
Ljungqvist, B., & Reinmüller, B., (2021) People as a Contamination Source in Pharmaceutical Cleanrooms- Source Strengths and Calculated Concentrations of Airborne Contaminants; PDA Journal of Pharma Sci and Technol. 75(2): pp 119-127, 2021.
Lindsay, J., Environmental Impact on Media Fills, biopharm-guides.com, 2012.
Whyte, W., Cleanroom Design, 2nd Edition, John Wiley & Sons, Chichester, 1999.
Ljundvist, B., and Rheinmuller, B., “Modern Cleanroom Clothing Systems: People as a Contamination Source,” PDA Journal of Pharmaceutical Science & Technology, Vol. 57, No 2, pp. 114-125, 2003.
Agalloco, J., Madsen R., and Akers, J., “Classification and Monitoring of Controlled Environments”, published in Pharmaceutical Technology Advancing Development and Manufacturing e-book, pp 10-13, May 2019.
https://microbeonline.com/infective-dose-and-lethal-dose/, accessed July 20,2026.
Leggett, H., Cornwallis, C, & West. S., “Mechanisms of Pathogenesis, Infective Dose and Virulence in Human Parasites”. PLoS Pathogens 8(2): e1002512. doi:10.1371/journal.ppat 1002512, 2012.
Schmid-Hempel, P., & Frank, S., “Pathogenesis, Virulence, and Infective Dose”, PLoS Pathogens, Vol. 3, No.10, pp.1372-1373, 2007.
Akers, J., Agalloco, J, & Madsen, R., “The Microbial Attribute of Objectionableness”, published on-line on LinkedIn.com, June 2021.
Agalloco, J. “Management of Aseptic Interventions”, Pharmaceutical Technology, Vol. 29, No. 3, p. 56-66, 2005.
Centers for Disease Control and Prevention (CDC), Guidelines for Environmental Infection Control in Healthcare Facilities, Environmental Infection Guidelines, Part 1, 2026 https://www.cdc.gov/infectioncontrol/hcp/environmental-control/environmental-sampling.html., accessed July 20, 2026.
Agalloco, J. “EMA’s Annex 1: Lead, Follow or Get Out of the Way”. American Pharmaceutical Review, Vol. 28, No. 5, pp. 18-26, 2025.

Subscribe to our e-Newsletters
Stay up to date with the latest news, articles, and events. Plus, get special
offers from American Pharmaceutical Review delivered to your inbox!
Sign up now!