Microbial contamination poses enormous risks to consumers of pharmaceuticals. There is also the associated financial liabilities and potential for damage to the pharmaceutical manufacturer’s reputation. To guard against these risks, pharmaceutical manufacturers have typically collected hundreds of samples per week, incubated them on agar plates for 7 to 14 days, and then counted the colonies to judge for the presence or absence of bacteria. This approach is very time-consuming and runs the risk that by the time a problem is discovered, a large amount of money may have been invested in manufacturing the product. An even worse scenario is that the product may have already been shipped, creating a potential risk to consumers of the drug.

These challenges help to explain the increasing interest by pharmaceutical manufacturers in rapid microbiological methods (RMMs) that provide the ability to detect microbial contamination in a fraction of the time of traditional methods. RMM instruments have been on the market for a number of years, but recent developments, such as performance improvements and cost reductions in the technology, have made them more attractive than in the past. The U.S. Food and Drug Administration (FDA) has also helped to spur the introduction of RMMs through its Process Analytical Technology (PAT) initiative which encourages real-time process monitoring. The result is that a recent survey shows that more than 70% of biopharmaceutical manufacturers either are currently using or plan to introduce RMM technology within three years.1

The risk of microbial contamination is confirmed by the consistent number of recalls ordered by the FDA due to microbial contamination. From 1991 to 1998, for example, the FDA ordered 46 such recalls, an average of 6 per year. Amongthese recalls, 6 were in the Class I category, 25 were in the Class II category, and 15 were in the Class III category.2 As an example, a recall was recently issued for four lots of cefazolin for injection. Cefazolin is an injectable antibiotic used in a hospital environment to treat skin and respiratory infections. Certain lots of the active ingredient used to manufacture the product were shown to contain microbial contamination (Bacillus pumilus, Staphylococcus homin-is, Propionibacterium acnes, or Micrococcus luteus). The contaminated products posed a serious or life-threatening risk for some patients. Therecalled lots contained 379,975 vials.3

The most common microbial testing applications in the biopharmaceutical manufacturing environment as identified by the survey cited above are environmental monitoring, including class A and B cleanrooms cited by 65% of respondents, non-sterile products, such as water for injection (WFI), purified water (PW), and buffers, cited by 57%, raw materials cited by 46%, and sterile products cited by 43%. The survey respondents identified qualitative analysis (30%), quantitative analysis (72%), and identification (35%) as key applications. The numbers add to greaterthan 100% because survey respondents felt that more than one application was important so they were allowed to select more than one answer.1The most common conventional microbial detection method involves overfilling an agar plate with media and then pressing the open side of the plate against the surface to be monitored. This approach typically provides about 60% recovery, but has the disadvantage that the potential exists to contaminate the surface being measured. An alternate approach involves rubbing the surface with a cotton swab. This approach greatly reduces the risk of contamination, but recovery is typically only 20%. The traditional methods take about 15 to 20 minutes to prepare the sample, a few minutes to plate the sample, 7 to 14 days forincubation, and then 15 to 20 minutes for analysis.

An advantage of the conventional method is that it does not require any capital investment. Its greatest drawback is the amount of time required to obtain results.It often does not make sense to hold up the manufacturing process until the microbiologicaltests have been completed. The result is that millions of dollars may be investedin materials and processing expenses that will be wasted if the results showthat microbial contamination was present. The dangers to consumers of the productas well as monetary costs and risks to the manufacturer’s reputation willof course be far greater if the product should be shipped before the contaminationis discovered. Another disadvantage is that conventional methods depend on theoperator to follow proper procedures in collecting the sample, preparing theplate, and analyzing the results.

RMMs have arisen in response to these weaknesses of the conventional methods. Rapid testing is the fastest growing segment of the microbial testing market and is expected to exceed $90 million by 2007, according to Strategic Consulting, Inc. The basic idea is to reduce the microbial assay time from one to two weeks to one day. A variety of different technologies have been developed in the effort to achieve this goal, although not all have demonstrated the ability to work in the real-world pharmaceutical manufacturing environment. Growth-based technologies, such as adenosine triphosphate (ATP) bioluminescence,are based on the measurement of biochemical or physiological parameters that reflect the growth of the microorganisms. Viability-based technologies, such as solid-phase cytometry, do not require growth for detection but rather use a variety of methods to detect viable cells. Cellular component-based technologies, such as mass spectroscopy, search for a specific cellular component within the cell. Nucleic acid-based technologies, such as DNA probes, monitorthe presence or absence of nucleic acid.

Despite the fact that this technology has been available for a number of years, the implementation of RMM has been delayed by a number of factors. One of the most important has been the perception that the regulatory authorities may have concerns about the technology and may not support its introduction. Companies were worried that they might face a difficult time in obtaining approval for a new drug application (NDA) that incorporated RMMs. But evidence suggests that key regulatory groups do in fact support RMMs. In recent years, the regulatory authorities and industry associations have issued a variety of documents that assist pharmaceutical manufacturers in implementary and preparing regulatory submissions of RMMs.

The Parenteral Drug Association (PDA) issued Technical Report Number 33 whichprovides validation criteria for RMM chemistry methods.4 The UnitedStates Pharmacopeia(USP) has published a chapter <1223> that defines validation criteria forRMM methods.5 In 2004, the FDA published a guidance document on aseptic processingof pharmaceutical products. It includes a provision for use of alternative microbiologicaltest methods. The UK’s Medicines and Healthcare Products Regulatory Authority(MHRA) of the United Kingdom stated in June 2003 in its publication, Mail: “Rapidmicrobiological methods offer substantial advantages over conventional methodsfor speed of the test and should be an integral part of process analytical technology(PAT) and process understanding.”

The FDA’s Center for Drug Evaluation and Research (CDER) said in its document guideline, Sterile Drug Products Produced by Aseptic Processing (September 2004), that “Other suitable microbiological test methods can be considered for environmental monitoring, in process control testing, and finished product release testing after it is demonstrated that the methods are equivalent or better than traditional methods (e.g., USP).” Dr. David Hus-song, Director, New Drug Microbiology Staff for the CDER, said in a presentation at a PDA meeting recently that “Rapid microbiological methods are encouraged by the FDA for improved process control. Rapid methods are encouraged for product release.”

Furthermore, the FDA’s PAT initiative is designed “to encourage real-time process monitoring…to facilitate introduction of new technology”. The FDA defines PAT as “systems for analysis and control of manufacturing processes based on timely measurements, during processing, of critical quality parameters and performance attributes of raw and in-process materials and processes to assure acceptable end product quality at the completion of the process.” PAT applications often include the collection of real-time data frequently using in-line methods to capture information about the quality of a pharmaceutical product earlier in the production process. Current RMMs are still conducted off-line on a laboratory bench so they are not yet capable of providing results in real-time. While this is not as advantageous as realtime measurements, it offers a substantial improvement over traditional microbiological methods.

There are several key guidance documents that serve as a guide to validating RMM. The first is PDA Technical Report 33: Evaluation and Implementation of New Microbiological Testing Methods which was published in June 2000.4This is the most detailed and informative description of validation of alternative microbiological methods that includes RMM. Report 33 includes a recommendation that an Equipment Qualification Model Approach be followed for the adoption of new methods. This is a holistic approach based on a User Requirement Specification (URS). The URS helps the user select a technology most appropriate for the applications. The other key steps in the URS approach recommended in Report 33 are performance of Installation Qualification (IQ), Operational Qualification (OQ), and Performance Qualification (PQ). Report 33 identifies four key aspects of PQ, the most important component of RMM validation: accuracy, sensitivity, linearity, and reproducibility.

Chapter <1223> in the USP is another important document that covers the topic of PQ for alternative microbiological technologies.5 The chapter defines validation criteria for qualitative and quantitative test methods. Among the most important parts is the definition of a comparability protocol approach in which current and new methods are tested simultaneously. USP <1223> lists the following recommended validation parameters for alternative methods:

  • Accuracy
  • Precision
  • Specificity
  • Detection Limit
  • Quantification Limit
  • Linearity
  • Range
  • Ruggedness
  • Repeatability

The FDA granted its first RMM global approval to Glax-oSmithKline (GSK) to use the Pallchek™ Luminometer as part of the quality control process for its prescription nasal spray product. This technology, based on ATP bioluminescence, identifies the presence or absence of microbiological contamination in less than 24 hours, enabling GSK to release product to market up to four days earlier than before. The process approval received by GSK covers the use of RMM for specific non-sterile products as well as pharmaceutical grade water. The following section explains how RMM was validated based on Report 33 and USP <1223>.

The RMM technology approved by the FDA is the ATP bioluminescence based on the luciferin-luceriferase substrate-enzyme which contains proteins extracted from the firefly, Photinus pyralis. Firefly luciferase (luciferin 4-monooxygenase, EC is a 62-kDa protein. A detailed structure is given in Conti et al.6 Bioluminescence based on the firefly luciferin-luciferase enzyme substrate system is an oxidative decarboxylation reaction in which D-luciferin is oxidized to oxyluciferin by the enzyme luciferase in the presence of ATP and oxygen. The conversion of D-luciferin to oxyluciferin releases photons with a maximum energy at a wavelength of 562 nm. During the first reaction step, D-luciferin and ATP react to form luciferin adenylate complex. This chemical complex binds strongly to the catalytic site of the luciferase enzyme.

In the presence of molecular oxygen, the luciferin adeny-late complex is oxidized to oxyluciferin. During this metabolic process, photons are emitted.

One of the key attributes that USP <1223> Report 33 recommends should be validated is the detection limit. For the ATP bioluminescence system used by GSK, the detection is a function of the photomultiplier in the instrument, the activity of the reagent kit, and the background light level associated with the sample. The photon count measured by instruments based on ATP bioluminescence is expressed as Relative Light Units (RLUs), which can be directly related to ATP content which in turn is correlated to the level of microorganisms present in the sample. For this particular system, the theoretical detection limit is around 100 attomoles/L (10-17moles/L) ATP after allowing for background in a typical sample. A critical question to address in validation is how this limit of detection applies to micro-organisms.

The limit of detection for microorganisms is influenced by the growth phase and the mode of growth, such as culture media, temperature, stress factors in the environment, etc. The typical limit of detection without an enrichment step is around 100 colony forming units (cfu). With an enrichment step, typically incubating the sample in tryptone soya broth for up to 24 hours, this increases to 1 cfu. For this study, the organisms used are those cited in the USP and the European Pharmacopeia (EP) for tests designed to show the absence of so called objectionable bacteria, together with some microorganisms that represent important potential contaminants of the pharmaceutical manufacturing environment.

The organisms used for this study were:

  • Staphylococcus aureus (ATCC 6538)
  • Escherichia coli (ATCC 25922)
  • Pseudomona aeruginosa (ATCC 9027)
  • Salmonella albony (NCTC 6017)
  • Staphylococcus epidermidis (ATCC 122228)
  • Micrococcus luteus (ATCC 9341)
  • Ralstonia picketii (ATCC 49129)
  • Bacillus subtilis (ATCC 6633)
  • Aspergillus niger (ATCC 16404)
  • Candida albicans (ATCC 10231)
  • Penicillium notatum (ATCC 9179)

These organisms were inoculated into tryptone soya broth to give samples containing between 10 and 100, between 1 and 10, and less than 1 cfu. Inoculated samples were incubated overnight then used for conventional plate counting. The remaining broth was filtered and RLUs were measured on the filter using the bioluminescence reagent. Five replicate experiments were performed for each organism tested. In all cases, organisms were detected (down to a level of as little as 1 cfu) with RLU readings measured at typically 104 to 105 times (based on typical readings being 106to 107versus background at typically 102) in the inoculated samples, compared to uninoculated controls.

The data obtained demonstrates that RMM is as capable of detecting the organism as the traditional microbiological method. The key difference is of course that detection with RMM is possible in 24 hours instead of a week. The specificity requirement is met by performing testing using 12 different microorganisms including Gram-positive bacteria, Gram-negative bacteria, yeasts, and molds that represent microorganisms cited in the USP and the EP as well as other recognized and important contaminants found in bioburden in pharmaceutical manufacturing facilities.

In addition to using ATP bioluminescence to perform presence/absence testing, there are certain specific applications in which it can be used to save significant time and, thus, avoid costly and problematic repeat testing. This can be done by estimating accurately and in real time the number of organisms in a pure culture. Examples of important test methods that use pure cultures include Disinfectant Efficacy Testing and Antimicrobial Effectiveness Testing.

The basic concept behind this approach is that for pure cultures of microorganisms grown under standardized conditions there is a reproducible and close correlation between microbial count and ATP content. Thus, by preparing a pure culture serially diluted in 10-fold steps across a relevant concentration range and performing side-by-side measurements of ATP bioluminescence (expressed as RLUs) and microbial count (expressed as cfu) it is possible to correlate RLU readings to cfu measurements. Having produced a correlation curve as just described, it is then possible to determine cfu readings for other samples as follows:

  • Take small sample of a diluted pure culture to be used for a given test
  • Use ATP bioluminescence to make an RLU reading
  • Use the correlation curve just described to convert the RLU reading just obtained to cfu

This value determines whether that dilution is suitable for the test to be performed.

RMMs provide the potential for pharmaceutical manufacturers to obtain more accurate and reliable data in less time at a lower cost. Regulatory authorities are now supporting the implementation of RMMs and most major pharmaceutical manufacturers have begun their implementation. The key factor driving the implementation of RMMs is their potential to avoid expending millions in raw materials and processing costs on batch that later must be destroyed due to microbial contamination.


  1. Workshop presented by P. Newby at the European Compliance Academy (ECA) Masterclass 'PAT in Microbial QC' 25-27, Berlin, January 2006 (P.R. Ball, personal communication).
  2. Swarbrick, J. and Boylan, J.C. Encyclopedia of Pharmaceutical Technology, Second Edition, Volume 3. New York: Dekker Encyclopedias.
  3. Wachter, K. “Contamination prompts recall of cefazolin.” Skin & Allergy News, April, 2006.
  4. Parenteral Drug Association. Technical Report #33: Evaluation, Validation, and Implementation of New Microbiological Testing Methods. Bethesda, MD: PDA, (2000).
  5. United States Pharmacopeia, “Validation of Alternative Microbiological Methods,” Pharmaceutical Forum, Jan.-Feb. 2003.
  6. Conti, E., Franks, N.P., and Brick, P. (1996). “Crystal structure of firefly lu-ciferase throws light on a superfamily of adenylate-forming enzymes. Structure 4: 287-298.

Peter R. Ball, Loris Arbizzani, and Christopher J. Mach all work at Pall Life Sciences. Please address all correspondence to Christopher J. Mach at